JOURNAL - dl.uctm.edu(bio)pharmaceutical, chemical and cosmetic industry, etc. Nanotechnology promises to exceed the impact of the “industrial revolution” with a projection of

VOLUME 50 2015 ISSUE 2

http://dl.uctm.edu/journal/ ISSN 1314-7471 (print)

ISSN 1314-7978 (on line)

JOURNAL

OF CHEMICAL TECHNOLOGY

AND METALLURGY

SOFIA

Journal of Chemical Technology and Metallurgy

The Journal of Chemical Technology and Metallurgy started originally in 1954 as Annual Journal of the former Higher

Institute of Chemical Technology. It ran in Bulgarian. In 2000 its name was changed to Journal of the University of

Chemical Technology and Metallurgy. It was published quarterly in English. Since 2013 it has run bimonthly as

Journal of Chemical Technology and Metallurgy.

Honorary Editor

R. Dimitrov

Editor-in-Chief

B. Koumanova

University of Chemical Technology and Metallurgy,

8 Kl. Ohridski blvd., 1756 Sofia, Bulgaria

Tel: (+ 359 2) 8163 302, E-mail: [email protected]

7BEditorial Board

S. J. Allen

Queens University of Belfast, UK

D. Angelova

University of Chemical Technology and Metallurgy, Bulgaria

M. Bojinov University of Chemical Technology and Metallurgy, Bulgaria

J. Carda

University Jaume I, Castellon, Spain

G. Cholakov University of Chemical Technology and Metallurgy, Bulgaria

V. Dimitrov Bulgarian Academy of Sciences

N. Dishovsky


S.J.C. Feyo de Azevedo Universidade do Porto, Portugal

M. Jitaru

University “Babeş -Bolyai”, Cluj-Napoca, Romania S. Kalcheva


F. Keil Hamburg University of Technology, Germany

T. Koinov


T. Konstantinova


M. Kucharski AGH University of Science and Technology, Krakow, Poland

J.M. Le Lann

Institut National Polytechnique, Ecole Nationale Supérieure des Ingénieurs en Arts Chimiques et Technologiques,

Toulouse, France

A. Mavrova


D. Mehandjiev Bulgarian Academy of Sciences

V. Meško

International Balkan University, Skopje, Macedonia

L. Mörl University “Otto-von-Guericke”, Magdeburg, Germany B. Nath

European Centre for Pollution Research, London, UK

D. Pavlov

Bulgarian Academy of Sciences

L. Petrov Bulgarian Academy of Sciences

D.C. Shallcross

The University of Melbourne, Australia

M. Simeonova


V. Stefanova


D. Stoilova

Bulgarian Academy of Sciences

N. Tsarevsky

Southern Methodist University, Dallas, Texas, USA

I. Turunen

Lappeenranta University of Technology, Finland

S. Vassileva


S. Veleva


L.Vezenkov


Ž. Živković

University of Belgrade, Technical Faculty, Bor, Serbia

Technical secretary: S. Georgieva

The annual subscription (for 6 issues) is 350 €, including postage, handling and packaging charge.

©UCTM, 2005. All rights reserved. Reproduction in part or whole without permission is strictly prohibited.

Prepress and design: A. Staneva; Printing: S Print Ltd., Sofia

The articles of this journal are indexed and abstracted in EBSCO, SCOPUS, Chemical Abstracts and Referativnii Journal Himii (VINITI).

http://dl.uctm.edu/journal/

Table of conTenTs

Risks and health effects from exposure to engineered nanostructures: A critical reviewVioleta Vasilevska Nikodinovska, Kristina Mladenovska, Anita Grozdanov............................................................117

Synthesis and characterization of diblock copolymers based on poly (ethylene glycol) and poly (4-vinyl pyridine) via atom transfer radical polymerizationStefani Vircheva, Rayna Bryaskova, Stoyan Miloshev, Nikolay Dishovsky.......................................................................135

Influence of substrate thickness on the quality attributes of polymer materials reverse printed by screen printing techniqueMladen Stančić, Branka Ružičić, Nemanja Kašiković,Dragoljub Novaković, Rastko Milošević......................................141

Study of the application in the brewing of enzyme preparation, containing laminarinase and lichenase, produced from Trichoderma sp. 405Nadejda Daskalova, Gabriela Marinova.........................................................................................................................149

Examination of the influence of the grinding degree and stabilizing agent on the rheological properties of aqua-coal fuel suspensions Izzat Eshmetov, Dilnoza Salihanova, Anvar Agzamhodjaev...................................................................................157

Specificities of the recirculating method for central Kyzyl Kum phosphorites processing to qualified phosphoric fertilizersUmarbek K. Alimov, Shafoat S. Namazov, Ahmed R. Reymov..............................................................................163

Investigation of porosity effects in the emission of pollutants in porous burnersFatemeh Bahadori, Khadijeh Mirza, Alireza Behroozsarand, Sima Rezvantalab........................................................171

Predicting the effect of cell geometry and fluid velocity on peM fuel cell performance by cfD simulationBahman ZareNezhad, Mohammad M. Sabzemeidani...............................................................................................176

Decision making for control of combustion process of pulverized coalKosta Boshnakov, Venko petkov, Metodi Nikolov.....................................................................................................183

current oscillations in epoxy-coated steel at anodic polarizationYordanka Marcheva....................................................................................................................................................193

A study on the kinetics of the electrodeposition of Ni, co and Ni-co alloy in citrate electrolyte part 1. The kinetic study of the independent electrodeposition of Ni and co Katya Ignatova, Daniela Lilova..................................................................................................................................199

Kinetic study on electrodeposited Ni-co alloy in citrate electrolyte Part 2. The kinetics of joint electrodeposition of Ni and Co in Ni-co alloy systemKatya Ignatova, Daniela Lilova..................................................................................................................................207

Journal of Chemical Technology and Metallurgy, 50, 1, 2015

Microstructure of cast iron after plasma bleachingSergey p. Nefedyev, Roman R. Dema, Svetlana A. Nefedyeva, Aleksey V. Yaroslavtcev.................................................213

Study of axial forces with the purpose to realize a combined process «helical rolling-pressing»Abdrakhman Naizabekov, Vitaly Talmazan, Alexander Arbuz, Toncho Koinov, Sergey Lezhnev.......................................217

Breakdown behaviour during reanodization of thick porous anodic alumina in aqueous and non-aqueous solutionsAlexander Zahariev....................................................................................................................................................223

Guide for Authors........................................................................................................................................................225


The present issue of Journal of chemical Technology and Metallurgy was published with the financial support of the Bulgarian Fund “Scientific Research”, Contract DNP 04/0044/12.12.2014.

Violeta Vasilevska Nikodinovska, Kristina Mladenovska, Anita Grozdanov

117

Journal of Chemical Technology and Metallurgy, 50, 2, 2015, 117-134

RISKS AND HEALTH EFFECTS FROM EXPOSURE TO ENgINEERED NANOSTRUCTURES: A CRITICAL REvIEw

Violeta Vasilevska Nikodinovska1, Kristina Mladenovska2, Anita Grozdanov3

1 Faculty of Medicine, University “Ss Cyril and Methodius” Skopje, Republic of Macedonia2 Faculty of Pharmacy, Department of Biopharmacy, University “Ss Cyril and Methodius” Skopje, Republic of Macedonia3 Faculty of Technology and Metallurgy, University “Ss Cyril and Methodius” Skopje, Republic of Macedonia

ABSTRACT

Nanotechnology and engineered nanostructures (ENSs) are becoming part of everyday life, starting from industrial application, even in food products, to gene therapy. Thus, tons and tons of nanoparticles (NPs) enter the environment and indirectly or directly - into the biological systems, including the human body. There are many con-troversial papers that describe interactions of the ENSs with biological systems and raise concern that intentional or unintentional human exposure to certain types of ENSs, may lead to significant health, i.e. toxicological effects. Because of our insufficient and contradictory knowledge about the health effects associated with the ENSs exposure, the aim of this paper is to summarize and systemize the already confirmed data and the latest found facts about ENSs and their health effects and to discuss the future opportunities and tasks in the field of nanotoxicology.

Keywords: engineered nanosized structures, nanotoxicology.

Received 24 July 2014Accepted 20 December 2014

INTRODUCTION

Development of the industrial sector based on appli-cation of ENSs and nanotechnologies is a fast-growing activity. Scientific achievements in many areas of na-notechnology have been successfully transferred into numerous practical applications in the electronic, food, (bio)pharmaceutical, chemical and cosmetic industry, etc. Nanotechnology promises to exceed the impact of the “industrial revolution” with a projection of $1 trillion market by 2015 [1 - 3].

Nanomaterials and/or nanosized structures were studied in many products such as composites, sensors and chemical catalysts, nutraceutical, cosmeceutical, drug, cell and gene delivery systems, medical and

electronic devices. However, due to their unique size and other physicochemical characteristics, they have raised a concern for their influence on human health and ecological systems. Research in this field is given a high priority and several reports on the toxicological properties of the ENSs with many documented deleteri-ous effects, particularly in animals, already exist [4 - 6].

The specific physicochemical properties (Fig. 1) that make the nanosized structures useful are the same that may hazard the human body and the environment. Size distribution, surface area, porosity, shape, charge density, electrophoretic mobility, strength, flexibility, crystallinity/ solubility, stability/(bio)degradability, bio(muco)adhesivity, biocompatibility, etc., can all af-fect the biodistribution of the ENSs and human health.


118

Human exposure to ENSs can be direct or indirect - through contamination of the environment, with the main entries in the human body via the intestinal tract, lungs and skin. Generation of reactive oxygen and nitrogen species (ROs) and action by dissolved ions (in a case of metal and metal oxide NPs) were accepted as common mechanisms of cytotoxicity. Cellular plasma membrane, lysosomes, mitochondria, nucleus, and intracellular structural (cytoskeleton) and globular proteins were identified as important targets (Fig. 2) [7].

NANOPARTICLES AND ENgINEERED NANOSTRUCTURES

NPs include all engineered and ambient spherical particles smaller than 100 nm. They are variably called ultrafine particles (UFPs) by toxicologists from the U.S. Environmental Protection Agency, Aitken mode and nucleation mode particles by atmospheric scientists, and engineered nanostructured materials - by material scientists [8 - 11]. UFPs include ambient and laboratory-generated NPs that are not produced in a controlled, engineered way. Engineered NPs include only spherical NPs specifically engineered in the laboratory, while other ENSs are labelled according to their shape as nanotubes,

nanofibres, nanowires, nanorings, etc. (Fig. 1).ENSs generated in working conditions can reach

high exposure concentrations, up to several hundred micrograms per cubic meter. Even very low concen-trations of nanosized materials in the air present very high particle number concentrations. For this reason and because of their different physicochemical proper-ties as compared to larger sized particles of the same components, they have been considered as the most dangerous fraction for the human health. ENSs are so tiny that small quantities (expressed in terms of mass) could have major toxic effects because of their large surface which is often correlated with higher biological reactivity (e.g. formation of free radicals). Insoluble and low soluble ENSs usually end up in the bloodstream after crossing all the respiratory or gastrointestinal protective mechanisms and then are distributed in the various organs and accumulated at specific sites. They can travel along the olfactory nerves and penetrate directly into the brain, just as they can pass through the cell barriers [12 - 14].

In general, the toxicological data specific to ENSs remain insufficient due to the closely specialized-limited number of studies, the short exposure period, the dif-

Fig. 1. Nanosized structures: physicochemical properties, surface modifications and (dispersion) state.

Nanosized structures: • Quantum dots • Carbon nanotubes • Graphene • Inorganic nanoparticles • Organic nanoparticles

Dimensions: • Number • Agregates • Agglomerates • Granulometric

distribution

(Dispersion) state: • Airborne aerosols • Gels, colloids • Liquids • Solid state

Surface modifications: • Pure • Coated • Chemically modified

Physicochemical properties: properties:

• Size • Shape • Surface area • Surface charge • Solubility • Crystallinity • Flexibility • Biopersistance

Shape: • Cube • Sphere • Needle • Tube • Shell • Fiber • Wire • Ring


119

ferent properties of the nanostructures tested and the often-unusual exposure route in the working environ-ment, among other factors. Depending on the type of nanostructure, nephrotoxicity, effects on reproduction, genotoxicity and cytotoxic effects, mainly in animal studies, have been reported so far. Some nanostructures caused granulomas, fibrosis and tumorous reactions in the lungs [9 - 11].Therefore, the population potentially exposed to ENSs should be prudent and apply safety measures of source elimination, exposure control and individual protection, both during the production and use of nanotechnology-based products. The most common materials in the descriptions of 580 of them are silver (139), followed by carbon (including fullerenes) (44), zinc (including zinc oxide) (28), titanium (including titanium dioxide) (28), silica (27) and gold (13) [15].

Classes of manufactured nanostructures and toxic effects

Carbon nanotubes (CNTs): Multi-walled (Fig. 3a) and single-walled carbon nanotubes (MWCNTs and SWCNTs) are constituted of a unique graphitic sheet or of two or more sheets nested together in a tubular mul-tilayer structure. They possess extraordinary properties: high electrical and thermal conductivity, great strength and rigidity, energy storage and field emission [1, 2]. CNTs can be synthesized by three main methods: laser ablation, arc discharge (both these methods are based on the evaporation of graphite due to high temperatures) and chemical vapor deposition; evaporated carbon condenses successively in the form of CNTs and other

structures, producing longer and less graphitized forms than the material obtained by the other methods. The arc discharge method has proved to be an interesting method for the mass production of CNTs. Also, other alternative methods were reported such as pyrolysis and electrolysis of molten lithium salts [2].

Fullerenes belong to a special carbon class of materi-als with spherical cages of 28 to more than 100 carbon atoms. They can be subjected to extreme pressures and regain their original shape when the pressure is released. These molecules are not modified and do not combine with each other. However, during manufacturing, cer-tain carbon atoms can be replaced with other atoms and form bondable molecules, thus producing a hard but elastic material. The surface chemical composition can be modified and different organic chains can be added or incorporated into these CNTs. Since fullerenes are empty structures with dimensions similar to several biologically active molecules, they can be filled with different substances and find medical applications [16].

The last member of the series of fused polycyclic aromatic hydrocarbons is the graphene (Fig. 3b) [17]. For several years scientists thought that planar graphene could not exist in a free state since it was unstable, compared to curved structures like CNTs and fuller-enes. However, Novoselov et al. have experimentally isolated a single graphene layer generating wide interest in many areas of material science and engineering [18]. Researchers have been looking for possible applications of graphene in bio-sensing devices and biomedical aims, such as drug delivery and cancer therapy. Graphene provides larger specific surface area than other CNTs and forms strong interactions with drugs being a good candidate for high drug loading.

The toxicity of these nanostructures is related to the potential presence of metal impurities (i.e. by-products of the synthesis) as well as to the surfactants used to disperse them in aqueous media. Harmful and patho-genic effects can arise due to their ability to enter the respiratory tract, deposit in the lungs, redistribute from the site of deposition, escape from the normal phagocytic defenses and modify the protein structure. Thus, they can potentially activate inflammatory and immunological responses, affecting normal organ functions. Several studies report on the toxicological effects of the CNTs

Fig. 2. Mechanisms of toxicity of ENSs.


120

and fullerene-based nanostructures and they are focused, mainly, on pulmonary and skin toxicity [19].

Scarce literature data describing in vitro toxicity of graphenes suggest that the latter can be either benign or cytotoxic depending on the layer number, lateral size, stiffness, hydrophobicity, surface functionalization and dose. Potential mechanisms of toxicity considered are the generation of reactive oxygen species (ROs) in tar-get cells, direct physical toxicity due to the interactions with the membrane lipids or indirect toxicity due to the adsorption of biological molecules [20]. The interac-tions between the dispersed graphene or graphene oxide sheets and target cells have been studied in monolayer cultures of lung epithelial cells [21], fibroblasts [22] and neuronal cells [23], where dose-dependent oxidative stress and slight loss of cell viability at high concentra-tions were observed. In addition, in vivo studies (in mice) demonstrated systemic biodistribution and biop-ersistence, following intravenous delivery of graphene oxides. At higher doses, granulomas in the lungs, liver, spleen and kidney were observed as well as deposition and retention in the lungs with production of lung injury and inflammation (Table 1) [24, 25]. It is still unknown whether graphene-family materials have the potential to induce sarcomas since they can have large surface areas, smooth topography and may be biopersistent similarly to tumorigenic solid-state implants.

Inorganic NPs of metals and their oxides: Insolu-ble inorganic NPs can be composed of pure metals or their oxides, or various inorganic products and alloys (Au, Ag, Co, Cr, ZnO, TiO2, CeO2, SiO2, CrO2). Only their nanometric dimensions distinguish them from the same products normally found on a larger scale. Because of their unique properties related to their nanometric scale, these particles are precisely produced. At this scale, they display mechanical, electrical and other properties that do not exist when in larger dimensions [6]. For ex-ample, titanium dioxide UFPs (Fig. 3c) ultrafine particles that fall within a nanosize range are commonly used as photocatalysts to clean air and water, as antibacterial agents on glass and steel, and as components of many cosmetics and sunscreens [26, 27], while silicon dioxide (SiO2) NPs (Fig. 3d) are widely applied in the fields, such as the chemical industry, medicine, cosmetics and agriculture [28].

Research results showed consistency in the sensi-tivity of aquatic, fish cell and human cell models to Ag and CeO2 particles of different size; with the observed sensitivity sequence from highest to lowest is: nano-Ag>microAg>nano CeO2 = micro CeO [29]. For some inorganic nanostructures, when the particle size is further reduced and approaches the Fermi wavelength of electrons (i.e., the electron de Broglie wavelength at the Fermi level 0.5 nm for Au and Ag), the continuous density of states breaks up into discrete energy levels leading to dramatically different optical, electrical and chemical properties. These particles, smaller than 2 nm, are usually called nanoclusters and they become molecu-lar species from which size-dependent strong fluorescent emission can often be observed upon photoexcitation in the UV-visible range. For this reason, they have been widely used for bioimaging and chemical sensing [30]. For example, 1.9 nm-gold NPs (gold core diameter), with a highly water soluble organic shell making the particles both useable at high concentrations (up to 1.5 g Au/cc) and well tolerated by animals even at high concentra-tions (LD50 > 1.4g Au/kg), are used for production of a novel nanotechnology X-ray contrast agent for in vivo use, AuroVist™ (www.nanoprobes.com).

Toxicity of these NPs has been evaluated by several research groups indicating that apoptosis can be induced by generation of ROs, which oxidative and inflammatory

a) b)

c) d)

Fig. 3. SEM of MWCNTs (a), nanostructured graphene (b), TiO2 NPs (c) and TEM of SiO2 NPs (d).


121

nature may then drive genotoxic and cytotoxic outcomes [28, 31 - 34]. This was confirmed when interactions of Ag NPs with human fibrosarcoma and human skin/carcinoma cells were studied [35]. The cytotoxicity of Ag NPs was also evaluated by MTT metabolic activ-

ity assay and membrane integrity assay in rat alveolar macrophages cell culture [36], while concentration-dependent genotoxicity was confirmed with the human lymphoblast TK6 cell micronucleus assay [37]. How-ever, no genetic toxicity in bone marrow was observed

Nanostructure Exposure route

Size [nm] Deposition Adverse health effects Reference

Polystyrene NPs

Intradermal 25-100 Skin Exacerbation of atopic dermatitis-like skin lesions, skin inflammation

62

TiO2 NPs

Oral Intravenous Intraperitoneal injection Dermal Intranasal instillation Inhalation Intratracheal instillation

25-155 36-48 < 100 4, 60 80, 155 < 100 nm 20-30

Liver, Kidneys, Spleen, Lungs Brain, Lungs, Spleen, Liver, Kidneys Spleen, Liver, Kidneys, Lungs Deep layer of epidermis, wider tissue distribution Brain (hippocampus) Lungs Lungs

Hepatic injury (spotty necrosis of hepatocytes, increased ALT/AST), myocardial damage (increased serum LDH, α-HBDH), nephrotoxicity (increased BUN) Liver (vacuolar degeneration, spotty necrosis, hydropic degeneration around the central vein, multifocal lesions at higher doses); Kidneys (swelling in the renal glomerulus, increase in blood URCA); Spleen (increased proliferation of local macrophages) ; Brain (neuronal cell degeneration, fatty degeneration in the hippocampus); Lung (granulomatous lesions) Thrombosis, hepatocelular necrosis and apoptosis, hepatic fibrosis Severe pathological change of skin and liver Oxidative damage, inflammation (increased TNF-α, IL-1β) Bronchoalveolar hyperplasia, metaplasia, fibrosis and tumorogenesis, impaired arteriolar dilatation Pulmonary inflammation

72

106

107

108

94

86, 91

109

SiO2 NPs Intranasal instillation

< 100 Brain Negative impact on striatum and dopaminergic neurons 71

Ag NPs

Oral Inhalation

56-60 18

Liver, Kidneys Blood, Stomach, Brain, Liver, Kidneys, Lungs, Testes

Slight liver damage (dose-dependent, more than 125 mg/kg), bile-duct hyperplasia, with necrosis, fibrosis No systemic adverse effects, incl. no genetic toxicity, alveolar inflammation, incl. thickened alveolar walls and small granuloma lesions

70, 75

38, 96

Au NPs Intratracheal instillation

2, 20, 200 Lungs No local and systematic adverse effects, incl. no genotoxicity

40, 78

PEG-Graphene sheets Graphene oxide sheets

Intravenous Intravenous Intratracheal instillation

10-30 < 100 < 100

Liver, Spleen (RES) Lungs Lungs

No appreciable toxicity at tested dose (20 mg/kg) after 3 month-treatment Inflammation cell infiltration, pulmonary edema, granuloma formation at high dose (10 mg/kg) Acute lung injury, chronic pulmonary fibrosis

25

24

110

f-CNTs (f-SWNTs, f-MWCNTs) SWCNTs

Intravenous Intratracheal instillation

30-38 < 100

Muscle, Skin, Kidney, Blood (after 30 min); Rapid renal clearance as intact NTs Lungs

No signs of renal or other severe acute toxicity responses Epitheloid granulomas (dose-dependent), interstitial inflammation, peribronchial inflammation and necrosis

85

4, 5, 111, 112

Polyalkyl-cyanoacrylate NPs

Intravenous < 100 Liver Modification in hepatic function (increased alpha 1-acid glycoprotein, decreased albumin secretion, lowered glucose production)

56

Table 1. Adverse health effects of various ENSs depending on the route of exposure in animal models.


122

when rats were exposed to Ag NPs by inhalation (90 days) (Table 1) [38]. When secreted NO levels were measured to investigate the correlation of nitrosative-oxidative stress and cytotoxicity induced by Ag NPs, it was confirmed that ultimately, the phagocytosis of Ag NPs stimulated inflammatory signaling through ROs generation in macrophages, followed by induced secre-tion of TNF-α, that may cause damage of cell membrane and apoptosis [39]. It has been reported also that Au NPs can catalyze nitric oxide, which reacts rapidly with O2 producing harmful peroxynitrite species that can disrupt lipids, DNA and proteins [40]. Ability of Au NPs to con-tribute in genotoxicity is a subject of intensive research [41], while with ZnO NPs, significant DNA damage in cells has been observed [42]. Penetration of nanosized TiO2 (5 - 20 nm) into the skin and its interaction with the immune system has been also demonstrated [43]. Other studies report that exposure to TiO2 NPs resulted in pulmonary inflammation and edema, accumulation of macrophages and pneumonocyte apoptosis [44, 45]. Genotoxicity in cultured cell lines for TiO2 NPs was also observed [46, 47], but there were also studies showing no DNA damage [48, 49].

Quantum dots and semiconductors: The quantum dots, called semiconductor nanocrystals or artificial at-oms, represent a special form of spherical nanocrystals from 1 to 10 nm in diameter, with unique size-dependent optical and electrical properties. Most of the research has been focused on semiconductor quantum dots, which display distinctive quantal effects, depending on the dimensions. They are used as fluorescent probes in diagnostic medical imaging and in therapeutics, because of their optical properties and their capacity to form covalent bonds with peptides, antibodies, nucleic acids or other low-weight molecules [6]. The first authors who demonstrated in vivo that CdSe/ZnS quantum dots coated with mercaptoacetic acid, could bond to blood transferrin were Chan and Nie in 1998, cited by Smith et al. [50]. This fluorescent complex was absorbed se-lectively by cancer cells. According to some authors, the toxicity of certain quantum dots could be linked to the release of cytotoxic ions, oxidative mechanisms and other less well-elucidated phenomena [51, 52]. Quan-tum dots contain toxic components, such as Cd (from Cd chalcogenide-based quantum dots) or Pb (from Pb

chalcogenide-based quantum dots). Cd2+ and Pb2+ could be released from quantum dots and then kill the cells. Thus, a direct way to avoid the possible toxicity of quantum dots is to make them well coated to become biologically inert. The coating materials can be low or nontoxic organic molecules/polymers (e.g. PEG) or inorganic layers (e.g. ZnS and silica). Few studies show direct, extracellular cytotoxicity of water-soluble CdSe and CdSe/ZnS quantum dots, because of Cd2+ release from the NPs due to poor purifications and/or simple surface cappings/coatings [51]. Quantum dots coated by simple molecules, such as mercaptoacetic acid, mercaptopropionic acid, 11-mercapto-undecanoic acid, 2-aminoethanethiol, are more toxic than the ones coated with silica. The silica layer is different from the amphi-philic polymer layers or small molecular ligands; it can be very thick (up to several micrometers) and therefore reduces the possible leaking of interior toxic Cd or Pb under physiological environments. There is no clear toxic effect observed if additional PEG molecules were attached to the silica exterior surface [51, 53].

Zero-valent metals (Fe(II)salts): The usually used ferrite (Fe3O4) NPs were synthesized from a reaction solution of FeCl2 and FeCl3, and an adjusting solution of NH4OH. They have been utilized as supermagnetic NPs for in vivo or in vitro biomedical applications, such as the contrast agents for magnetic resonance imaging, hyperthermic media for tumors treatment, magneto-targeting carriers for chemotherapeutic drugs and im-munoassay [54]. Not much is known about the toxicity of ferrite NPs at the cellular and molecular levels. Research of Ahamed et al. [55] showed that nickel ferrite NPs induced oxidative stress in A549 cells by induction of ROs and depletion of glutathione.

Insoluble organic NPs: Insoluble organic NPs are usually composed of various organic substances, often insoluble polymers (e.g. polystyrene, polycianoacr-ylate, polyacrylamide, polyacetamide, polymethacrylic acid, etc.), to which different organic (alkyl) radicals can be grafted. Under specific conditions, some of the substances can also be made soluble. Often, only their nanometric dimensions distinguish organic NPs from the same products normally found on a larger scale. Like all other NPs, they display catalytic, chemical or other properties that do not exist in larger dimensions. To this


123

group, nanocapsules, nanospheres and nanoshells, based on a wide variety of insoluble organic polymers, belong as well. Some of these structures are developed with an aim of integration with other substances, often medica-tions. The surface of these NPs can also be modified to interact specifically with certain sites of the body [6].

The (cyto)toxicity of these NPs is usually correlated with the rate of biodegradation of the polymer backbone and main products of biodegradation. Adhesion to the cell membrane can also increase cytotoxicity by local release of biodegradation products in high concentra-tions. For example, investigation of the liver toxicity of poly(alkyl cyanoacrylate) NPs showed slight modifica-tions in the hepatic function after chronic administration of NPs and these effects have been found reversible when the treatment was stopped. Because the mononu-clear phagocyte system is concerned, there is a potential danger for physical blockade of this tissue (Table 1) [51].

BIODISTRIBUTION AND ADvERSE HEALTH EFFECTS OF ENSs AFTER DIFFERENT ROUTES OF EXPOSURE

Skin exposure (Fig. 4a): Penetration of the skin barrier is size-dependent with NPs more likely to enter more deeply into the skin. However, in a broken skin, entry of even larger (0.5-7 µm) particles was observed and there is a hypothesis that skin when flexed can make the epidermis more permeable to NPs [57, 58]. No direct relation between other physicochemical properties and biodistribution of NPs into and via systemic circulation through the skin has been found, so far [59]. There are literature data, however, pointing to mechanical skin irritation and sensitisation with involvement of the mac-rophages, Langerhans cells or other cells. On exposure to the skin, particles of different surface characteristics attract different arrays of opsonins, which travel together with the NPs to different places at different times through the lipid layers within the cells of the stratum corneum and the appendage route (hair follicles, sweat glands). Internalization inside the cells may easily trigger the inflammatory cell response [59 - 61]. For example, in a study of Yanagisawa et al., exposure to polystyrene NPs under skin barrier dysfunction exacerbated atopic dermatitis-like skin lesions in NC/Nga mice, in a size-dependent manner [62]. In addition, skin inflammation

via over expression of CC-chemokines was observed, even in the absence of allergens in atopic subjects. However, in a study of Park et al., in which human skin equivalent model was used to deduce the toxicity of polystyrene and TiO2 NPs, no phototoxicity, acute cuta-neous irritation or skin sensitization was observed [63].

Oral ingestion (Fig. 4b): From the respiratory tract via the mucociliary escalator, ENSs can be subsequently ingested into the gastrointestinal tract or they can be direct-ly ingested with water, food or as drug delivery systems. After oral ingestion, depending dominantly on the size, surface charge and mucoadhesivity, particles could reach the colon or translocate from the lumen of the intestinal tract via the M cells in the Peyer’s patches and isolated follicles of the gut-associated lymphoid tissue, and/or via the normal intestinal enterocytes (Fig. 1) [64, 65]. The literature data suggest that the smaller the particle diam-eter, the faster they can permutate the mucus to reach the colonic enterocytes, whereas positively charged particles exhibit poor oral bioavailability through electrostatic repulsion and mucus entrapment [66].

Nanostructures entering the lymphatic may induce a secretory immune response, while those entering the sys-temic circulation can be distributed to different organs (e.g. kidney, lung, liver, spleen, brain, etc.). For example, in rats dosed orally with radiolabelled functionalized C60 fullerenes, 98 % were cleared in the faeces within 48 hours, while the rest was eliminated via urine, indi-cating uptake of the particles into the blood [67]. Also, uptake of larger TiO2 particles (150 - 500 nm) into the blood and distribution to the liver was observed [68].

Those particles, reaching the colon, can disrupt the epithelial barrier function by apoptosis of enterocytes, thus triggering mucosal inflammation and inducing in-flammatory bowel diseases [69]. Others, reaching organs other than the gut or crossing the blood-brain barrier, may induce oxidative stress locally (e.g., depletion of reduced and oxidized glutathione, inhibition of super-oxide dismutase activity, increase in catalase activity, etc.) and provoke toxic effects, including liver damage, bile-duct hyperplasia, nephrotoxicity, neurotoxicity, spermatotoxicity, etc [70, 71]. For example, after oral administration of TiO2 in mice, slight brain lesion with fatty degeneration in the hippocampus was observed. In the kidneys, serious swelling in the renal glomerulus


124

was detected, while in the liver tissue, the hydropic de-generation around the central vein was prominent and a spotty necrosis of the hepatocytes was found (Table 1) [72]. In studies, in which acute toxicity of SiC NPs was evaluated after their oral administration in Wistar rats, stases and haemorrhages in the heart, liver, kidney, lungs and spleen were observed [73,74]. Significant dose-dependent changes in the serum alkaline phosphatase and the cholesterol values, and moderate liver damage were also reported, when silver NPs in Sprague-Dawley rats were orally administered. In addition, significant de-crease in the total protein was observed when high dose was administered, while in the high and middle-dosed rats, red blood cells, haemoglobin and hematocrite were increased and coagulation time decreased. Interestingly, gender-related difference in the accumulation of the silver NPs in the kidneys was observed, with two fold higher concentration in the female kidneys (Table 1) [75]. Because of these effects, a more demanding regulation that includes both nonclinical and clinical studies exists for the NPs designed as drug or nutraceutical carriers.

Inhalation (Fig. 4c,d): Physicochemical properties of the inhaled particles and the materials themselves, especially size distribution, surface charges and surface chemistry, are important for the pathogenic effects into the lungs, as they influence both the deposition and the clearance rates [76, 77]. The main mechanism for deposition of inhaled ENSs in the respiratory tract is diffusion due to the displacement when the particles col-lide with the air molecules. The clearance of deposited particles in the respiratory tract is manifested by physical translocation of the particles by different mechanisms and chemical dissolution. The efficacy of this clearance mechanism depends highly on the efficiency of the alveolar macrophages to “sense” deposited particles, to move to the site of their deposition and to phagocytise them. Phagocytosis of the deposited particles takes place within a few hours, so by 6 - 12 hours after deposition, essentially all of the particles are phagocytised by the alveolar macrophages [78]. There is little evidence for paracellular transport of ENSs in the healthy lung, but it may be present when permeability is increased [79].

Fig. 4. Biodistribution of ENSs after skin exposure (a), oral ingestion (b), inhalation (c) and nose breathing (d).

a) b)

c) d)


125

The size was confirmed as the key parameter in the deposition of nanostructures into the lung. The smaller the particles, the deeper they can travel into the lung [80]. So, in the alveoli, particles smaller than 2.5 µm can be found, while inhaled UFPs (an aerodynamic diameter < 100 nm), deposited mainly in the alveolar region, are subject to easy and rapid translocation through the systemic circulation to other vital organs [59]. When fractional deposition of the NPs in the lung is up to 70 %, the larger of them are being recognized, phagocytised by alveolar macrophages and translocated into the intersticium [81, 82]. On the other hand, 5-nm particles showed an equal deposition of app. 30 % in all three regions (nasopharyngeal, tracheobronchial and alveolar region), 20-nm particles had the highest deposition efficiency in the alveolar region (~50 %), whereas in the tracheobronchial and nasopharyngeal regions, the deposition efficiency was app. 15 % [83]. Kim et al. suggested that 80 % of the 20-nm particles were deposited in the respiratory system [84].

As a result of macrophage phagocytosis of nano-structures in the deep lung and subsequent macrophage activation, chemokines, cytokines, ROs, C reactive protein and other mediators are released, resulting in sustained inflammation and eventually fibrotic changes. Those nanostructures, which are biopersistent for years in the alveoli (e.g. asbestos or silica) or contain mutagenic substances, are of high risk for developing cancer, due to their ability to interact with the epithelial cells and to enter the lung tissue. The same effects were observed with experimental instillation of TiO2 and carbon black [1, 85]. In the study of Heinrich et al., in which Wister rats were exposed to ultrafine TiO2 NPs for 24 months at an average concentration of 10 mg/m3, development of pulmonary tumours was observed [86]. With high chronic doses of particles with low solubility, bronchoalveolar hyperplasia, metaplasia, fibrosis and tumorogenesis were observed (Table 1).

The cascade of subsequent events and inflammation leads also to cardiovascular events and nervous system toxicity manifestations [87]. Exposure to ambient air pollution was shown to be associated with coagulation, vascular thrombosis, an increase in blood pressure and a decrease of heart rate, accelerated atherosclerosis, life-threatening arrhythmia and triggering of myocardial

infarction [88 - 90]. For example, exposure to nanosized titanium aerosols have been associated with arteriolar constriction, while inhalation of fine TiO2 NPs - with an impaired arteriolar dilatation [91, 92]. The loss of capacity for microvascular dilatation can decrease tis-sue perfusion and in this way, compromise its function. In addition, pulmonary exposure to diesel particles in the ultrafine range, potentiated myocardial ischemia in patients with pre-existing coronary heart disease and provoked acute increase in diastolic blood pressure within 2 h of exposure to the particles [93].

Considering the brain toxicity, extrapulmonary translocation of the NPs through the olfactory nerve was reported to be the most probable pathway for NPs to the brain [94 - 97]. The research data in which bio-distribution of elemental carbon and silver NPs were studied point to accumulation in the rat olfactory bulb after inhalation [95, 96]. In addition, the specific net-work of sensory nerve endings, between the nose and the olfactory bulb, is also involved in the transport of ENSs directly to the brain, via the olfactory nerve tract. Elder et al. suggested that the particle size plays the most important role in neuronal uptake and translocation of the ENSs [97]. For example, the solid particles deposited on the olfactory mucosa, have to be smaller than 200 nm in order to be transported through the cribriform plate. Using a predictive particle deposition rat model it was estimated that app. 11.5 % of the amount deposited on the olfactory mucosa was translocated to the olfactory bulb [98]. Research data point that intranasally instilled TiO2 NPs could be also translocated into the central nervous system of mice via the olfactory nerve tract and accumulated in the olfactory nerve layer, olfactory ventricle, cerebral cortex, thalamus and CA1 and CA3 regions of hippocampus. After 30 days of instillation of TiO2 suspension into the mice’s nasal cavity, the highest content of Ti in the hippocampus was observed when compared with the olfactory bulb and cerebellum, while in the cerebral cortex, the lowest quantity was detected. In addition, an increased neuronal cells count in olfactory bulb, and irregular arrangement of the neuronal cells in the olfactory nerve layers and hippocampus, were observed (Table 1) [94]. Additional neuronal transloca-tion pathways for ENSs via the trigeminus nerve and tracheobronchal sensory nerves, have been also reported


126

by Hunter and Day in their study in rat animal model [99]. They demonstrated translocation of the intranasally instilled rhodamine-labelled nanospheres (20 - 200 nm) to the trigeminal ganglion inside the cranium via up-take into the ophthalmic and maxillary branches of the trigeminus nerve, that supplies sensory nerve endings throughout the nasal mucosa. In another study, the same ENSs were intratracheally instilled in pigs. Neuronal translocation to the ganglion nodosum in the neck area that is networked into the vagal system was reported [100]. Determination of the effects of the translocated particles in the brain is important, also because some of them may develop parkinsonism 17 years earlier than the average population [101].

Air borne NPs have been also found to penetrate into the systemic circulation following inhalation, reaching other extrapulmonary organs, such as the liver, spleen, kidney, etc [57, 76, 102, 103]. For example, pulmonary exposure to diesel particles with ultrafine range, resulted in aggravation of the cisplatine-induced effects on serum urea and creatinine, urine N-acetyl-β-D-glucosaminidase activity and renal glutathione concentration in an acute renal failure animal model [104]. In addition, high ac-cumulation of silver NPs in the basal membranes of the medulla and cortex in the kidneys of female Sprague-Dawley rats, was reported without significant deterio-rations in the kidney function and histopathology. In a study of the whole body exposure, when rats were sub-jected to chronic inhalation of silver particles (90 day), lung and liver were reported as major targeted organs. Also, bile duct hyperplasia, dilatation of the central vein and increased foci were observed [96].

Parenteral injection (Fig. 1): ENSs easily reach liver after intravenous administration, accumulate and usually are retained there for a relatively long period of time. Using intravenous injection, Douglas et al. stud-ied the biodistribution of polymer-coated or uncoated poly(butyl 2-cyanoacrylate) NPs in rabbits [105]. About 60 % of the NPs were located in the liver and the spleen and app. 30 % remained in the bloodstream. The coat-ing had no significant influence on NPs biodistribution. When biodistribution of two types of radiolabelled (111In) functionalized SWCNTs after intravenous ad-ministration was followed, rapid first-order clearance from the blood compartment, through the renal excretion

route (3 - 3.5 hours), was observed, without any toxic side effects or mortality. Radioactivity in all examined organs after 30 min of administration was detected, with higher levels of radioactivity found in the liver, muscle, skin, kidney, lung and blood. Similarly, when TiO2 NPs (36 - 48 nm) were intravenously administered at high doses to mice, after 14 days of treatment, accumulation in the liver, lung, kidneys, spleen and heart was observed, with different degrees of damage in these organs, except in the heart (Table 1) [106].

EU RESEARCH ON EFFECTS OF NANOSTRUC-TURES ON HUMAN HEALTH AND ENvIRON-MENT

Since human health and environmental safety are of primary importance, the European Community pays special attention to the assessment of the risks and health effects of ENSs. In the frame of the NMP Theme of the FP7 program, each R&D project funded by the European Commission should contain a specific safety work package. One highly effective coordinative ac-tion, that brings together the best European experts in nanotoxicology, nanoecotoxicology and nano-risk as-sessment is NANOIMPACTNET (www.nanoimpactnet.eu). NANOIMPACTNET coordinates testing strategies and methods, screening tools, risk assessment tools and methodologies for more than 800 members of the research community, as well as to over 1000 companies developing or using nanomaterials. All this work is aim-ing towards the goal of implementing the EU’s Action Plan for Nanotechnology.

The main goal of another project, NANOINTER-ACT (www.nanointeract.net), was to develop a platform and a toolkit for understanding the interactions between the NPs and the living world. Research was performed for understanding how ENSs interact with the living systems, from the first contact, via absorption and locali-zation, to the final onset of functional impact. One of the main achievements of the NANOINTERACT has been the description of the nanoparticle-protein corona as a biological entity that interacts with the living system and cellular machinery. It also showed a first evidence of NPs modulating the rate of protein fibrillation in solution.

The NANOSAFE2 project (www.nanosafe.org)


127

was focused on a smaller number of ENSs for their safe industrial production and it has addressed mainly the exposure during manufacturing processes and nanomate-rials exposure in the environment. Screening techniques for CNTs, ZnO, carbon black, crystalline silica, with size of around 50 nm, have been developed. The project was related to development of advanced technologies for lim-iting exposition to ENSs in factories and leaks into the environment, with design of safe production equipment, dynamic confinement, individual protection devices and filtration. As an outcome of the NANOSAFE2 project, a database on ENSs toxicology was created.

DIPNA (www.dipna.eu) was an integrated platform focused on verification of the ENSs possible toxicity and ecotoxicity. The main goal of DIPNA was to provide information on in vitro assays for detection of ENSs effects on human cells, instruments and methods for environmental NPs detection, as well as recommenda-tions for nanotoxicologists and researchers. The impact of cerium, gold, cobalt and iron oxide NPs, in liquid suspension and in dry state, on various types of human cells was tested in vitro in order to identify biomarkers of nanotoxicity and design assays.

The NANOSH project (www.ttl.fi/partner/nanosh) included research on the toxic effects of NPs that were relevant to the occupational environment. Thirty five different nanomaterials have been characterized and published in a NANOATLAS book. Investigations of genotoxicity have provided evidence that both TiO2

NPs and CNTs have a potential to induce genotoxicity in vivo and in vitro. It was shown that TiO2 NPs induced pulmonary inflammation (both in in vivo and in vitro systems) when coated with amorphous silica, suggest-ing that the surface properties of the nanomaterials had important influence on the effects.

HINAMOX project (www.hinamox.eu) was focused on investigations of safety of the metal and metal oxide NPs (TiO2, ZnO, CuO, CeO2, Al2O3), with an aim to establish protocols for nanosafety testing of metal na-noparticles.

PROSPECT (www.nanotechia-prospect.org) was a public/private partnership project supported by OECD and related to better understanding of the ENSs impact on humans and the environment. The programme pro-vided a direct contribution to the global safety assess-

ment of nanomaterials, described by the OECD Working Party on Manufactured Nanomaterials. The project contributed to the scientific foundations for adapting or creating test methods that enabled ecotoxicological hazard assessment of the nanomaterials, as well as to the establishment of quantitative relation between the structure and activity. The target particles (CeO2 and ZnO) are part of a list of 14 nanomaterials identified as commercially relevant for the global economic impact of nanotechnology.

The CELLNANOTOX project (www.fp6-cellna-notox.net) was focused on research of the interaction between NPs, and lung and intestine, as primary sites fol-lowing the penetration of NPs into the blood circulation, with the liver, kidney and the immunological system, as secondary sites of interaction. The project explored the mechanistic aspects of interactions, uptake and recycling of selected NPs by different cellular systems.

The ENPRA project (www.enpra.eu) aimed to de-velop a novel risk assessment approach, integrated with uncertainty analysis of ENSs. In vitro, in silico, animal and human models of exposure-dose-response relation-ships for five targeted organs (pulmonary, hepatic, renal, cardiovascular and developmental) and five end-points (oxidative stress, inflammation, immunotoxicity, fi-brogenecity and genotoxicity), have been used for risk assessment and high-throughput screening tests.

ENSSATOX (www.ennsatox.eu) was a project con-centrated on studying the hazard effects and activity of the nanostructures within a range of biological models of increasing complexity and organization in aquatic systems. The project has started from the hypothesis that the biological activity and environmental impact of NPs are directly dependent on their structure and functional-ity, and that the biological membrane is playing a very important role in the nanotoxicity, as a boundary of the living cells that nanostructures should cross and as a host of many physiological processes (respiration, nerve conduction), which can be desturbed by the structural disruption.

The main goal of the project NANOMMUNE (www.nanommune.eu) was, through a multidisciplinary approach, to establish an array of red-out systems for determination of toxicity, not only of current ENSs, but also for prediction of hazardous effects of new ENSs that


128

are being developed, thus enabling a sustainable growth of the nanotechnology-based-industry. This project was focused on the immune system, which has the capacity to respond to invasion by pathogens and foreign parti-cles. The core concept underpinning the project , was that the recognition versus non-recognition of ENSs by immune-competent cells will determine the distribution, as well as the toxic potential of the novel nanomaterials.

The aim of the project NanoReTox (www.nanoretox.eu) was to develop novel and robust protocols for safety testing of nanomaterials. It is especially related to metals, metal oxides and metalsulphides. The project intends to examine the molecular and cellular reactivity of well-characterized NPs by comprehensively addressing five key questions: how does the environment, into which nanoparticles are released, affect their physicochemical properties and bioreactivity; how does it affect their ability to interact with and/or penetrate mammalian and aquatic cells and organisms; is there a hierarchy of activity; which combinations of conditions are most likely to pose a risk to human health and the environ-ment; and how can this information be incorporated into a risk assessment model.

The project NEPHH (www.nephh-fp7.eu) was look-ing at human health and environmental implications of nanotechnology-based materials from holistic life-cycle perspectives. The nanomaterials selected by the NEPHH project, were polymer composites, reinforced with silicon-based laboratory-made and industrial NPs. NEPHH aimed to increase the understanding of the mechanisms of toxicity of nanosized dust generated during cutting, fracture, breakage or cracks of nanorein-forced end-use products and also to determine acceptable exposure limits.

The NEURONANO project (www.neuronano.eu) used a knowledge-based approach to address the ques-tions of whether NPs smaller than 40 nm, could poten-tially enter the brain by crossing the blood-brain barrier, and whether some ENSs may induce oxidative stress in living systems. NEURONANO aimed to establish a simple screening and risk assessment matrix for NPs in neurodegenerative diseases.

The INLIVETOX project (www.inlivetox.eu) devel-oped a novel modular microfluidics-based in vitro test system, modelling the response of cells and tissues to the

ingestion of NPs (gold, silver, TiO2). The project aimed to answer how the target tissues (the gastrointestinal tract, the liver) individually respond to NPs and how the interactions between the different tissues modulate their responses.

The aim of the NANEX project (www.nanex-project.eu) was to develop a catalogue of generic and specific (occupational, consumer and environmental release) exposure scenarios for ENSs, taking in account the entire lifecycle of nanomaterials. NANEX collected and reviewed available exposure information, with focus on three very relevant ENSs: high aspect ratio nanoma-terials (e.g. CNTs), mass-produced nanomaterials (e.g. ZnO, TiO2, carbon black) and specialized nanomaterials that were only produced on a small scale (e.g. Ag NPs). The exposure information included both quantitative (measurement results) and qualitative contextual expo-sure information (risk management measures).

The aim of the NANOTEST project (www.nanotest-fp7.eu) was to develop alternative testing strategies and high-throughput testing protocols, using in vitro and in silico methods essential for the risk assessment of NPs used in medical diagnostics.

The NANOPOLYTOX project (www.nanopolytox.eu) was designed to improve the understanding of the potential environmental/health impact of nanotechnol-ogy based products, at different stages of their life-cycle.

The ENRHES project (http://ihcp.jrc.ec.europa.eu/enrhes) performed a comprehensive and critical scien-tific review on the health and environmental safety of fullerenes, CNTs, metal and metal oxide nanomaterials. The review considered sources, pathways of exposure, the health and environmental outcomes of concern, in a context illustrating the state-of-the-art in the field, and informing the regulatory bodies on the potential risks of ENSs.

CONCLUSIONS

ENSs are becoming a part of everyday life improv-ing human health and life style. However, when enter-ing in the organism by various means and via different pathways, they agglomerate, dissolve and change their physicochemical properties as a consequence of the environmental impact on the body, thus affecting bio-


129

logical effects, which from beneficiary may turn into toxic. Therefore, the safety assessment of the ENSs is mandatory to recognize the risks and avoid the potential hazards.

We are witnesses that more and more data regard-ing toxicity of ENSs and methods of its assessment, are becoming available. This paper reports on many findings and achievements in this respect. Further stud-ies in which the impact of different structures and their physicochemical properties on toxicity after different exposure pathways has to be evaluated, are still needed. Well-summarized results are especially important for the industry and sectors, where ENSs are produced or used, for providing guidance on the assessment of exposure to these materials. An improved understanding of the risks for toxicity and health effects of the ENSs will aid in future development and exploitation of a variety of nanomaterials. It is of a paramount importance, na-notechnology and nanotoxicology to cooperate closely in order to provide useful information and to avoid misinformation about the toxicity potential of nano-structures. Material scientists, physicians, pharmacists and toxicologists should be included in nanotechnologi-cal developments and risk assessments in early stages, taking into consideration available and produced data, various structures, the dose-response relationship, in vivo and in vitro experiments and interactions between cells from different tissues and nanostructures. Projects that promote and coordinate research in the field of na-notoxicology, including development of alternative in vitro methods for toxicity evaluation, and that contribute to establishing/managing of databases and information systems of nanostructures toxicity should be continually sponsored, while the dialogue, critical discussions and networking between all stakeholders, industry, academia and regulatory bodies, must be continually promoted.

REFERENCES

1. S. Iijima, Helical microtubules of graphite carbon, Nature, 354, 1991, 56-58.

2. A. T. Dimitrov, Study of carbon nanotubes produc-tion by electrolysis in molten Li2CO3, MJCCE, 28, 1, 2009, 111–118.

3. D. Drobne, Nanotoxicology for safe and sustainable

nanotechnology., Arh. Hig. Rada Toksikol. 58, 2007, 471-478.

4. C.W. Lam, J.T. James, R. McCluskey, and R. L. Hunter, Pulmonary toxicity of single-wall carbon nanotubes in mice 7 and 90 days after intratracheal instillation., Toxicol. Sci., 77, 2004, 126-134.

5. A.A. Shvedova, E.R. Kisin, A.R. Murray, V.J. John-son, A.I. Potapovich, Y. Tyurina, O. Gorelik, Unusual inflammatory and fibrogenic pulmonary responses to single-walled carbon nanotubes in mice., Am. J. Physiol,.289, 2005, L698-L708.

6. C. Ostiquy, G. Lapointe, M. Troltier, L. Menard, Y. Cloutier, M. Boutin, M. Antoun, C. Normand, Health effect nanoparticles. Report R-451, IRSST, August, 2006.

7. E. Fröhlich, Cellular Targets and Mechanism in the Cytotoxic Action of non-biodegradable engineered nanoparticles, Cellular Targets and Mechanisms in the Cytotoxic Action of Non-biodegradable Engi-neered NanoparticlesCurr. Drug. Metab., 14, 9, 2013, 976-988.

8. SCENIHR: The scientific aspects of the existing and proposed definitions relating to products of nanosci-ence and nanotechnologies, 2007 .

9. J. Pekanen, M. Kulmala, Exposureassessment of ul-trafine particles in epidemiologic time-series studies, Scand. J. Work. Environ. Health, 30, 2, 2004, 9-18.

10. J.H. Seinfeld, S.N. Pandis, Atmospheric physics and chemistry: from air pollution to climate change, New York (NY): John Wiley & Sons, 1998.

11. M. Simko, M.O. Mattsson, Risks from accidental ex-posure to engineered nanoparticles and neurological health effects: a critical review, Part. FibreToxicol., 7, 2010, 42-57.

12. H.M. Kipen, D.L. Laskin, Smaller is not always better: nanotechnology yields nanotoxicology, Am. J. Physiol. Lung Cell Mol. Physiol., 289, 2005, L696-L697.

13. A.D. Maynard, R.J. Aitken, T. Butz, V. Colvin, K. Donaldson, G. Oberdorster, M.A. Philbert, J. Ryan, A. Seaton, V. Stone, S.S. Tinkle, L. Tran, N.J. Walk-er, D.B. Warheit, Safe handling of nanotechnology, Nature, 444, 2006, 267-269.

14. A. Grozdanov, Personal experience regarding na-noosis., Polimeri 3, 2008, 161-163.


130

15. Wilson International Center for Scholars, A nano-technology consumer products inventory; www.nanotechproject.org/conssumerproducts, 2007.

16. P. Holister, V. Roman, T. Harper, Fullerenes. Tech-nology White Papers #7. [S.l.] Cientifica, 2003, 12 p. Page d’accueil visionnée le 26/04/2004.

17. K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, M.I. Katsnelson, I.V. Grigorieva, S.V. Dubonos, A.A. Firsov, Two dimensional gas of massless dirac-fermions in graphene, Nature, 438, 2005, 197-200.

18. K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Ji-ang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films, Science, 306, 2004, 666-669.

19. G. Tejral, N.R. Panyala, J. Havel, Carbon nanotubes: toxicological impact on human health and environ-ment, J. Appl. Biomed., 7, 2009, 1-13.

20. A. Aydın, H. Sipahi, M. Charehsaz, Nanoparticles toxicity and their routes of exposures in recent advances in novel drug carrier systems, InTech, 2012, 483-500.

21. Y. Chang, S.T. Yang, J.H. Liu, E. Dong, Y. Wang, A. Cao, Y. Liu, H. Wang, In vitro toxicity of graphene oxide on A549 cells, Toxicol. Lett,.5, 3, 2011, 201-210.

22. K. Wang, J. Ruan, H. Song, J. Zhang, Y. Wo, S. Guo, D. Cui, Biocompatibility of grapheneoxide., NanoscaleRes. Lett., 6, 2011, 8.

23. Y. Zhang, S.F. Ali, E. Dervishi, Y. Xu, Z. Li, D. Casciano, A.S. Biris, Cytotoxicity effects of gra-phene and single-wall carbon nanotubes in neural phaeochromocytoma-derived PC12 cells, ACS Nano, 4, 2010, 3181-3186.

24. X. Zhang, J. Yin, C. Peng, W. Hu, Z. Zhu, W. Li, C. Fan, Q. Huang, Distribution and biocompatibility administration., Carbon, 49, 2011, 986-995.

25. K. Yang, J. Wan, S. Zhang, Y. Zhang, S. Lee, Z. Liu, In vivo pharmacokinetics, long-term biodistribution and toxicoilogy of PEG lyated graphene in mice, ACS Nano, 5, 2011, 516-522.

26. D. Sun, T.T. Meng, T.H. Loong, T.J. Hwa, Removal of natural organic matter from water using a nano-structured photocatalyst coupled with filtration membrane., Water Sci. Technol,. 49, 2004, 103-110.

27. K.J. Shieh, M. Li, Y.H. Lee, S.D. Sheu, Y.T. Liu, Y.C.

Wang, Antibacterial performance of photocatalyst thin film fabricated by defection effect in visible light., Nanomed., 2, 2006, 121-126.

28. Y-H. Park, H.C. Bae, Y. Jang, S.H. Jeong, H.N. Lee, W-I. Ryu, M.G. Yoo, Y-R. Kim, M-K. Kim, J.K. Lee, J. Jeong, S.W. Son, Effect of the size and surface charge of silica nanoparticles on cutaneous toxicity, Mol. Cell. Toxicol. 9, 2013, 67-74.

29. B.K. Gaiser, T.F. Fernandes, M. Jepson, J.R. Lead, C.R. Tyler, V. Stone, Assessing exposure, uptake and toxicity of silver and cerum dioxide nanopar-ticles from contaminated environments, Environ. Health, 21, 8, suppl.1, S2, 2009.

30. X. Hangxun, S.K. Suslick, Water-soluble fluorescent silver nanoclusters, Advanced Materials, 22, 2010, 1078-1082.

31. S.M. Hussain, K.L. Hess, J.M. Gearhart, K.T. Geiss, J.J. Schlager, In vitro toxicity of nanoparticles in BRL 3A rat liver cells, Toxicol. in vitro, 19, 2005, 975-983.

32. N. Miura, Y. Shinohara, Cytotoxic effect and apoptosis induction by silver nanoparticles in hela cells, Biochem. Biophys. Res. Comm., 390, 2009, 733-737.

33. K.M. Ramkumar, C. Manjula, G. Gnanakumar, M.A. Kanjwal, T.V. Sekar, R. Paulmurugan, P. Rajaguru, Oxidative stress-mediated cytotoxicity and apopto-sis induction by TiO2 nanofibers in hela cells, Eur. J. Pharm. Biopharm., 81, 2012, 324-333.

34. J.R. Gurr, A.S. Wang, C.H. Chen, K.Y. Jan, Ultrafine titanium dioxide particles in the absence of photo-activation can induce oxidative damage to human bronchial epithelial cells, Toxicol,. 213, 2005, 66-73.

35. S. Arora, J. Jain, J.M. Rajwade, K.M. Paknikar, Cellular responses induced by silver nanoparticles: in vitro studies, Toxicol. Lett., 179, 2008, 93-100.

36. C. Carlson, S.M. Hussain, A.M. Schrand, L.K. Bray-dich-Stolle, K.L. Hess, R.L. Jones, J.J. Schlager, Unique cellular interaction of silver nanoparticles: size-dependent generation of reactive oxygen spe-cies, J. Phys. Chem., B 112, 2008, 13608-13619.

37. Y. Li, D.H. Chen, J. Yan, Y. Chen, R.A. Mittelstaedt, Y. Zhang, A.S. Biris, R.H. Heflich, T. Chen, Geno-toxicity of silver nanoparticles evaluated using the ames test and in vitro micronucleus assay, Mutation


131

Res., 745, 2012, 4-10. 38. J.S. Kim, J.H. Sung, J.H. Ji, K.S. Song, J.H. Lee,

C.S. Kang, I.J. Yu, In vivo genotoxicity of silver nanoparticles after 90-day silver nanoparticles inhalation exposure, Safety and Health at Work, 2, 2011, 34-38.

39. E.J. Park, J. Yi, Y. Kim, K. Choi, K. Park, Silver nanoparticles induce cytotoxicity by a trojan-horse type mechanism, Toxicol. in vitro, 24, 2010, 872-878.

40. H.Y. Jia, Y. Liu, X.J. Zhang, L. Han, L.B. Du, Q. Tian, Y.C. Xu, Potential oxidative stress of gold nanoparticles by induced-no releasing in serum, J. Am. Chem. Soc., 131, 2009, 40-41.

41. M. Schulz, L.M. Hock, S. Brill, V. Strauss, S. Treu-mann, S. Groters, B.V. Ravenzwaay, R. Landsiedel, gold nanoparticles administered to the lungs of rats., Mutation Res.,745, 2012, 51-57.

42. V. Sharma, S.K. Singh, D. Anderson, D.J. Tobin, A. Dhawan, Zinc oxide nanoparticle induced genotox-icity in primary human epidermal keratinocytes, J. Nanosci.Nanotechnol.,11, 2011, 3782-3788.

43. M. Kreilgaard, Influence of microemulsions on cutaneous drug delivery, Adv. Drug Del.Rev. 54, Suppl.1, 2002, S77–S98.

44. D.B. Warheit, T.R. Webb, C.M. Sayes, V.L. Colvin, K.L. Reed, Pulmonary instillation studies with nanoscale TiO2 rods and dots in rats: toxicity is not dependent upon particle size and surface area, Toxicol. Sci., 91, 2006, 227-236.

45. D.B. Warheit, T.R. Webb, K.L. Reed, S. Frerichs, C.M. Sayes, Pulmonary toxicity study in rats with three forms of ultrafine-TiO2 particles: differential responses related to surface properties, Toxicol., 230, 2007, 90-104.

46. H. Türkez, F. Geyikoglu, An in vitro blood culture for evaluating the genotoxicity of titanium dioxide: the responses of antioxidant enzymes, Toxicol. Industr. Health, 23, 2007, 19-23.

47. J.J. Wang, B.J. Sanderson, H. Wang, Cyto- and genotoxicity of ultrafine TiO2 particles in cultured human lymphoblastoid cells, Mutation Res., 28, 2007, 99-106.

48. S. Hackenberg, G. Friehs, M. Kessler, K. Froelich, C. Ginzkey, C. Koehler, A. Scherzed, M. Burghartz, N.

Kleinsasser, Nanosized titanium dioxide particles do not induce DNA damage in human peripheral blood lymphocytes, Environ. Molecul. Mutagenesis, 52, 2010, 264-268.

49. M. Naya, N. Kobayashi, M. Ema, S. Kasamoto, M. Fukumuro, S. Takami, M. Nakajima, M. Hayashi, J. Nakanishi, In vivo genotoxicity study of titanium dioxide nanoparticles using comet assay following intratracheal instillation in rats, Regul. Toxicol. Pharmacol., 62, 2012, 1-6.

50. A.M. Smith, X. Gao, S. Nie, Quantum-dot nanocrys-tals for in-vivo molecular and cellular imaging, Photochem. Photobiol., 80, 2004, 377-385.

51. C. Kirchner, T. Liedl, S. Kudera, T. Pellegrino, A. Munoz Javier, H.E. Gaub, S. Stolzle, N. Fertig, W.J. Parak, Cytotoxicity of colloidal CdSe and CdSe/ZnS nanoparticles, Nano. Lett., 5, 2005, 331-338.

52. A. Shiohara, A. Hoshino, K. Hanaki, K. Suzuki, K. Yamamoto, On the cytotoxicity caused by quantum dots, Microbiol. Immunol., 48, 2004, 669-675.

53. W.Yu. William, C. Emmanuel, R. Drezek, V.L. Colvin, Water-soluble quantum dots for biomedical applications, Biochem. Biophys. Res. Comm., 348, 2006, 781-786.

54. C.M. Fu, Y.F. Wang, Y.F. Guo, T.Y. Lin, J.S. Chiu, In vivo bio-distribution of intravenously injected Tc-99 m labeled ferrite nanoparticles bounded with bio-compatible medicals, IEEE Transactions on magnetic, 41, 2005, 4120-4122.

55. M. Ahamed, M.J. Akhtar, M.A. Siddiqui, J. Ahmad, J. Musarrat, A.A. Al Khedhiry, M.S. Al Salhi, S.A. Alrokayan, Oxidative stress mediated apoptosis induced by nickel ferrite nanoparticles in cultured A549 cells, Toxicol., 283, 2011, 101-108.

56. R. Fernandez-Urrusuno, E. Fattal, D. Porquet, J. Feger, P. Couvreur, Evaluation of liver toxico-logical effects induced by polyalkylcyanoacrylate nanoparticles, Toxicol. Appl. Pharmacol., 130, 1995, 272-279.

57. G. Oberdöster, E. Oberdöster, J. Oberdöster, Na-notoxicology: An emerging discipline evolving from studies of utrafineparticles, Envion. Health. Perspect., 113,2005, 823-832.

58. S.S.Tinkle, J.M.Antonini, B.A.Rich, J.R.Roberts, R.Salmen, K.DePree, E.J. Adkins, Skin as a route


132

of exposure and sensitization in chronic beryllium disease, Environ.Health Perspec., 111, 2003,1202-1208.

59. P.H.M. Hoet, I. Bruske-Hohlfeld, O.V. Salata, Nanoparticles-known and unknown health risks, J. Nanobiotechnol., 2, 2004, 12.

60. M. Guix, C. Carbonell, J. Comenge, L. Garcнa-Ferndez, A. Alarcуn, E. Casals, Nanoparticles for cosmetics. How safe is safe? Contributions to sci-ence, 4, 2008, 213–217.

61. M. Schneider, F. Stracke, S. Hansen, U.F. Schaefer, Nanoparticles and their interactions with the dermal barrier, Dermatoendocrinol., 1, 2009, 197–206.

62. R. Yanagisawa, H. Takano, K.I. Inoue, E. Koike, K. Sadakane, T. Ichinose, Size effects of polystyrene nanoparticles on atopic dermatitis like skin lesions in NC/NGA mice, Int. J. Immunopathol. Pharmacol., 23, 2010, 131-141.

63. Y.H. Park, S.H. Jeong, S.M. Yi, B.H. Choi, Y.R. Kim, I.K. Kim, M.K. Kim, S. W. Son, Assessment of phototoxicity, skin irritation, and sensitization potential of polystyrene and TiO2 nanoparticles, J. Phys. Conf. Ser., 304 012050, 2011.

64. K. Mladenovska, E.I. Janevik, M.D. Glavas, E.F. Kumbaradzi, K. Goracinova, Biodistribution of 131I-BSA loaded gelatin microspheres after peroral application to BALB/c mice-particle size study, Acta Pharm., 53, 2003, 187-197.

65. J.F. Hillyer, R.M. Albrecht, Gastrointestinal per-sorption and tissue distribution of differently sized colloidal gold nanoparticles, J. Pharm. Sci., 90, 2001, 1927–1936.

66. L. Szentkuti, Light microscopical observations on luminally administered dyes, dextrans, nanospheres and microspheres in the pre-epithelial mucus gel layer of the rat distal colon., J.Control Rel., 46, 1997, 233-242.

67. S. Yamago, H. Tokuyama, E. Nakamura, K. Kikuchi, S. Kananishi, K. Sueki, In vivo biological behaviour of a water-miscible fullerene: 14C labeling, absorp-tion, distribution, excretion and acute toxicity, Chem. Biol., 2, 1995, 385-389.

68. J. Böckmann, T. Lahl, Eckert, B. Unterhalt, Blood titanium levels before and after oral administration titanium dioxide, Pharmazie, 55, 2000, 140-143.

69. M.C. Lomer, R.P. Thompson, J.J. Powell, Fine and ultrafine particles of the diet: influence on the mu-cosal immune response and association with Crohn’s disease, Proc. Nutr.Soc., 61, 2002, 123-130.

70. Y.S. Kim, M.Y. Song, J.D. Park, K.S. Song, H.R. Ryu, Y.H. Chung, H.K. Chang, J. H. Lee, K.H. Oh, B.J. Kelman, I.K. Hwang, I.J. Yu, Subchronic oral toxicity of silvernanoparticles, Part. Fibre Toxicol., 6, 2010, 20.

71. W. Jie, W. Chen, S. Jiao, X. Yang, Neurotoxicity of silica nanoparticles: brain localisation and dopa-minergic neurones damage pathways, CS Nano, 5, 2011, 4476–4489.

72. J. Wang, G. Zhou, C. Chen, H. Yu, T. Wang, Y. Ma, G. Jia, Y. Gao, B. Li, J. Sun, Y. Li, F. Jiao, Y. Zhao, Z. Chai, Acute toxicity and biodistribution of differ-ent sized titanium dioxide particles in mice after oral administration, Toxicol. Lett., 168, 2007, 176-185.

73. T. Ristoski, V. Vasilevska, K. Mladenovska, A. Grozdanov, Estimate of toxically influence of silicon carbide nonoparticles according to histopa-thologycal changes, Mac. Vet.Rev., 36, 2013, 25-31.

74. O. Lozano, J. Laloy, L. Alpan, J. Mejia, S. Rolin, O. Toussaint, J-M. Dogne, S. Lucas, B. Masereel, Effects of SiC nanoparticles orally administered in a rat model: biodistribution, toxicity and elemental composition gangs in feces and organs, Toxicol. Appl. Pharmacol., 264, 2012, 232-245.

75. Y.S. Kim, J.S. Kim, H.S. Cho, D.S. Rha, J.M. Kim, J.D. Park, B.S. Choi, R. Lim, H.K. Chang, Y.H. Chung, I.H. Kwon, J. Jeong, B.S. Han, I.J. Yu, Twenty eight-day oral toxicity, genotoxicity and gender-related tissue distribution of silver nanopar-ticles in Sprague-Dawley rats, Inhal. Toxicol., 20, 2008, 575-583.

76. W.G. Kreyling, M. Semmler, F. Erbe, P. Mayer, S. Takenaka, H. Schulz, G. Oberdörster, A. Ziesenis, Translocation of ultrafine insoluble iridium particles from lung epithelium to extra pulmonary organs is size dependent but very low, J. Toxicol. Environ. Health A., 65, 2002, 1513-1530.

77. E. Bermudez, J.B. Mangum, B.A. Wong, B. Asghar-ian, P.M. Hext, D.B. Warheit, J. Everitt, Pulmonary responses of mice, rats and hamsters to subchronic inhalation of ultrafine titanium dioxyde particles,


133

Toxicol. Sci., 77, 2004, 347-357.78. E. Sadauskas, N.R. Jacobsen, G. Dancher, M.

Stoltenberg, U. Vogel, A. Larsen, W. Kreyling, H. Wallin, Biodistribution of gold nanoparticles in mouse lung following intratechal instillation, Chem. Central Journal, 3, 2009, 16.

79. J.J. Meiring, P.J.A. Borm, K. Bagate, M. Semmler, J. Seitz, S. Takenaka, W.G. Kreyling, The influence of hydrogen peroxide and histamine on lung perme-ability and translocation of iridium nanoparticles in the isolated perfused rat lung, Part. Fibre. Toxicol., 2, 2005, 3.

80. W.G. Kreyling, M. Semmler-Behnke, J. Seitz, W. Scymczak, A. Wenk, P. Mayer, S. Takenaka, G. Oberdörster, Size dependence of the translocation of inhaled iridium and carbon nanoparticle aggregates from the lung of rats to the blood and secondary target organs, Inhal. Toxicol., 21, 2009, 55-60.

81. C. Mühlfeld, B. Rothen-Rutishauster, F. Blank, D. Vanhecke, M. Och, P. Gehr, Interactions of na-noparticles with pulmonary structures and cellular responses, Am. J. Physiol. Lung Cell Mol. Physiol., 294, 2008, L817-829.

82. C.C. Diagle, D.C. Chalupa, F.R. Gibb, P.E. Morrow, G. Oberdoster, M.J. Utell, M.W. Frampton, Ultrafine particle deposition in humans during rest and exer-cise, Inhal. Toxico1., 5, 2003, 539-552.

83. International Commission on Radiological Protec-tion. Human respiratory model for radiology protec-tion, Ann CRP., 24, 1994, 1-300.

84. C.S Kim, Method of calculating lung delivery and deposition of aerosol particles, Respir.Care, 45, 2000, 695-710.

85. R. Singh, D. Pantarotto, L. Lacerda, G. Pastorin, C. Klumpp, M. Prato, A. Bianco, K. Kostarelos, Tissue biodistribution and blood clearance rates of intrave-nously administered carbon nanotube radiotracers, PNAS, 103, 2006, 3357-3362.

86. U. Heinrich, R. Fuhst, S. Rittinghausen, O. Creutzen-berg, B. Bellmann, W. Koch, K. Levesen, Chronic inhalation exposure of Wistar rats and two differ-ent strains of mice to diesel engine exhaust, carbon black, and titanim dioxide, Inhal. Toxicol., 7, 1995, 533-556.

87. G.S. Kang, P.A. Gillespie, A. Gunnison, A.L.

Moreira, K-M. Tchou-Wong, L-C. Chen, Long-term inhalation exposure to nickel nanoparticles exacer-bated atherosclerosis in a susceptible mouse model, Environ. Health Perspect., 119, 2011.

88. D.R. Gold, A.A. Litonjua, J. Schwartz, M. Verrier, R. Milstein, A. Larson, Cardiovascular vulnerability to particulate pollution, Am. J. Respir. Care Med., 157, 1998, A261.

89. A. Peters, D.W. Dockery, J.E. Muller, M.A. Mit-tleman, Increased particulate air pollution and the triggering of myocardial infarction, Circulation, 103, 2001, 2810-2815.

90. A. Nemmar, M.F. Hoylaerts, P.H. Hoet, J. Vermylen, B. Nemery, Size effect of intratracheally instilled particles on pulmonary inflammation and vascular thrombosis, Toxicol. Appl. Pharmacol., 186, 2003, 38-45.

91. T.R. Nurkiewicz, D.W. Poter, A.F. Hubbs, J.L. Cumpston, B.T. Chen, D.G. Frazer, V. Castranova, Nanoparticle inhalation augments particle-depend-ent systemic microvascular dysfunction, Particle Fibre Toxicol., 12, 2008, 1.

92. J.C. Frisbee, Impared skeletal muscle perfusion in obese Zucker rats, Am. J. Physio. Regul. Integr. Comp Physiol., 285, 2003, R 1124-1134.

93. B. Urch, F. Silverman, P. Corey, J.R. Brook, K.Z. Lu-kic, S. Rajagopalan, R.D. Brook, Acute blood pres-sure responses in healthy adults during controlled air pollution exposures, Eviron. Health Perspect., 113, 2005, 1052-1055.

94. J. Wang, Y. Liu, F. Jiao, F. Lao, W. Li, Y. Gu, Y.F. Li, C.C. Ge, G.Q. Zhou, Y.L. Zhao, B. Li, Z. Chai, C. Chen, Time-dependent translocation and po-tential impairment on central nervous system by intranasaly instilled TiO2 nanoparticles, Toxicol., 254, 2008, 82-90.

95. G. Oberdörster, Z. Sharp, V. Atudorei, A. Elder, R. Gelein, W. Kreyling, C. Cox, Translocation of in-haled ultrafine particles to the brain, Inhal. Toxicol., 16, 2004, 437-445.

96. J.H. Sung, J.H. Ji, J.U. Yun, D.S. Kim, M.Y. Song, J. Jeong, B.S. Han, J.H. Han, Y. H. Chung, J. Kim, T.S. Kim, H.K. Chang, E.J. Lee, J.H. Lee, I.J. Yu, Lung function changes in Sprague-Dawley rats after prolonged inhalation exposure to silver nanoparti-


134

cles, Inhal. Toxicol., 20, 2008, 567-574.97. A. Elder, R. Gelein, V. Silva, T. Feikert, J. Carter,

R. Potter, A. Maynard, Y. Ilto, J. Finkelstein, G. Oberdörster, Translocation of inhaled ultrafine man-ganese oxide particles to the central nervous system, Environ. Health Perspect., 114, 2006, 1172-1178.

98. B. Asoharian, S. Anjilvel, A multi-path model of fiber deposition in the rat lung, Toxicol.Sci., 44, 1998, 80-86.

99. D.D. Hunter, R.D. Dey, Identification and neuropep-tide content of trigeminal neurons innervating the rat nasal epithelium, Neuroscience, 83, 1998, 591-599.

100. D.D. Hunter, B.J. Undem, Identification and substance P content of vagal afferent neurons innervating the epithelium of the guinea pig tra-chea, Am. J. Respir. Crit. Care Med., 159, 1999, 1943-1948.

101. B.A. Racette, L. McGee-Minnich, S.M. Moerlein, J.W. Mink, T.O. Videen, J.S. Perlmutter, Welding-related parkinsonism. Clinical features, treatment and pathophysiology, Neurology, 56, 2001, 8-13.

102. G. Oberdörster, Z. Sharp, V. Atudorei, A. Elder, R. Gelein, A. Lunts, W. Kreyling, C. Cox, Extrapul-monary translocation of ultrafine carbon particles following whole- body inhalation exposure of rats, J. Toxicol. Environ. Health A., 65, 2002, 1531-1543.

103. M.A. Akerman, W.C.W. Chan, P. Laakkonen, S.N. Bhatia, E. Ruoslahti, Nanocrystal targeting in vivo, Proc. Natl. Acad. Sci. USA, 99, 2002, 12617-12621.

104. A. Nemmar, S. Al-Salam, S. Zia, J.Yasin, I. Al Husseni, B.H. Ali, Diesel exhaust particles in the lung aggravate experimental acute renal failure, Toxicol. Sci., 13, 2010, 267- 277.

105. D.J. Douglas, S.S. Davis, L. Illum, Biodistribution of poly(butyl 2-cyanoacrylate) nanoparticules in rabbits, Int. J. Pharm., 34, 1996, 146-152.

106. J. Xu, H. Shi, M. Ruth, H. Yu, L. Lazar, B. Zou,

C.Yang, A.Wu, J. Zhao, Acute toxicity of intrave-nously administered titanium dioxide nanoparticles in mice, Plos One 8, 8, e70618, 2013, 1-6.

107. J. Chen, X. Dong, J. Zhao, G. Tang G, In vivo acute toxicity of titanium dioxide nanoparticles to mice after intraperitioneal injection, J. Appl. Toxicol., 29, 2009, 330-337.

108. J. Wu, W. Liu, C. Xue, S. Zhou, F. Lan, L. Bi, H. Xu, X. Yang, F-D. Zeng, Toxicity and penetration of TiO2 nanoparticles in hairless mice and porcine skin after subchronic dermal exposure, Toxicol. Lett., 191, 2009, 1-8.

109. D. Hohr, Y. Steinfartz, R.P. Schins, A.M. Knaapen, G. Martra, B. Fubini, P.J. Borm, The surface area rather than the surface coating determines the acute inflammatory response after instillation of fine and ultrafine TiO2 in the rat, Int. J. Hyg. Environ. Health., 205, 2002, 239-244.

110. B. Li, J. Yang, Q. Huang, Y. Zhang, C. Peng, Y. Zhang, Y. He, J. Shi, W. Li, J. Hu, C. Fan, Bio-distribution and pulmonary toxicity of intratrache-ally instilled graphene oxide in mice, NPG Asia Materials, 5, 2013, e44.

111. D.B. Warheit, B.R. Laurence, K.L. Reed, D.H. Roach, G.A.M. Reynolds, T.R. Webb, Compara-tive pulmonary toxicity assessment of single-wall carbon nanotubes in rats, Toxicol. Sci., 77, 2004, 117-125.

112. D.B. Warheit, B.R. Laurence, K.L. Reed, D.H. Roach, G.A.M. Reynolds, T.R. Webb, Pulmonary bioassay toxicity study in rats with single wall carbon nanotubes. Proceedings of the Comp. rend. du First International Symposium on Oc-cupational Health Implications of Nanomaterials, 2004, 12-14 October, Buxton, Great Britain, Ed. by the Health and Safety Executive, Great-Britain and the National Institute for Occupational Safety and Health, USA, 65-67, July, 2005.

Stefani Vircheva, Rayna Bryaskova, Stoyan Miloshev, Nikolay Dishovsky

135


SYNTHESIS AND CHARACTERIZATION OF DIBLOCK COPOLYMERS BASED ON POLY (ETHYLENE GLYCOL) AND POLY (4-VINYL PYRIDINE)

VIA ATOM TRANSFER RADICAL POLYMERIZATION


ABSTRACT

Atom transfer radical polymerization (ATRP) was applied for the preparation of block copolymers based on poly (eth-ylene glycol) monomethyl ether and poly (4-vinyl pyridine) (MeOPEG-b-P4VP). This was achieved using chloro-terminated PEG as a macroinitiator. Further, the ATRP was performed using 4-VP as a second forming block in the presence of CuCl/PMDETA as a catalyst system. The successful formation of MeOPEG-b-P4VP block copolymers was proved by 1H NMR with the presence of all characteristic signals arising from PEG and 4-VP units. The GPC results showed the preparation of well-defined block copolymers with low dispersity, which confirms the controlled character of radical polymerization. The thermal properties of thus prepared copolymers were characterized by TGA analysis.

Keywords: atom transfer radical polymerization, block copolymers, poly (ethylene glycol), poly (4-vinyl pyridine).

Received 15 December 2014Accepted 19 February 2015

INTRODUCTION

In the recent years, block copolymers have received considerable attention due to their ability to self assem-ble in solution forming a range of different morpholo-gies and sizes including lamellar, hexagonal-packed cylindrical, and body-centered cubic micellar structures, depending on the relative volume fractions of the blocks [1-3]. Therefore, the design of block copolymers with well-defined architecture, controlled molecular mass, polydispersity and chain composition is a very important research field. Controlled/living radical polymerization is a method which allows the synthesis of wide range of well-defined polymers/copolymers and this can be

achieved by three main methods: nitroxide-mediated polymerization (NMP), atom transfer radical polym-erization (ATRP) and reversible addition-fragmentation chain transfer polymerization (RAFT) [4]. Among them, atom transfer radical polymerization is the most effec-tive and used method since its introduction by Wang and Matyjaszewski [5-6]. The ATRP process uses an alkyl halide as an initiator and a metal in its lower oxidation state with complexing ligands. The process involves transfer of the halide from the dormant polymer chain to the metal complex, thus establishing a dynamic equilib-rium between active and dormant species. ATRP method is applied for the polymerization of large number of monomers as styrene and its derivatives, acrylates and

University of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected]


136

methacrylates [7]. The controlled chain growth with ATRP is very useful for the preparation of well-defined block copolymers because the growing polymer chain contains a stable halogen terminated ω-end that can act as an initiator for chain extension [8].

Recently, 4-vinylpyridine (4VP) was employed to prepare block copolymers via ATRP due to its ability to form self-assembly supramolecular structures [9 - 10] and complexes with metal ions [11]. The block copolymer of 4VP has been already prepared via ATRP method [12 - 15]. However, there are some limitations in the polymerization of 4VP by ATRP due to its strong coordination effect with the metal catalysts component in the polymerization system. Therefore, multidentate ligands with strong coordination ability and protic sol-vent have to be applied to minimize the contamination of the catalyst [16].

This work is focused on the synthesis of block co-polymers based on poly (ethylene glycol) monomethyl ether and poly (4 - vinyl pyridine) via atom transfer radical polymerization. For this purpose a novel chloro-terminated MeOPEG macroinitiator obtained by reaction of MeOPEG with thionyl chloride is applied for the polymerization of the second 4VP block.

EXPERIMENTAL

MaterialsPoly(ethylene glycol) monomethyl ether (MeOPEG,

Merck, Mn - 1900 g/mol) was dried under vacuum at 400C for 24 hour before use. Toluene (Sigma Aldrich, 99,8 %) was distilled. 4-vinylpyridine (Acros Organ-ics, 95 %) was passed though a neutral alumina (Al2O3)

column and distilled. Thionyl chloride (Acros Organics, 99 %), CuCl (Aldrich, 98 %), Pyridine (Sigma Aldrich, 99,8 %) and 1,1,4,7,7-pentamethyldiethylenetriamine (PMDTA, Across Organics, 98 %) were used as received.

Synthesis of MeOPEG-Cl macroinitiator20 g poly(ethylene glycol) monomethyl ether (Me-

OPEG, 0.025 mol) and pyridine (1.30 ml) were added to a three-neck flask equipped with a magnetic stir bar, a condenser and a calcium chloride drying tube and stirred. To this mixture, 20 ml toluene was added under stirring at 87ºC. Then, thionyl chloride 1.3 ml (0.0625 mol) was added slowly to the reaction mixture. The resulting mix-ture was stirred for 15 h at 87ºC. The solid obtained was

removed by filtration. The solution was concentrated by rotary evaporation in order to remove the residual thionyl chloride. The product was recovered by precipitation in cold diethyl ether, followed by filtration. Pale-yellow product was dried at 40ºC under vacuum.

Synthesis of MeOPEG-b-P4VP block copolymerThe MeOPEG-b-P4VP copolymer was synthesized

by atom transfer radical polymerization (ATRP) of 4-VP and MeOPEG-Cl macroinitiator using technique de-scribed by Matyaszewski at al. [6]. Polymerization was performed in a 50 ml flask, equipped with a magnetic stir bar and septum. MeOPEG-Cl (3g) was added to 17 ml i-propanol and heated at 60ºC in order to dissolve the macroinitiator and deoxygenated by passing N2 through the mixture. Then, the ligand (1,1,4,7,7-PMDTA, 0.330 ml) was added and the solution was stirred and deoxy-genated with N2 for 30 min, followed by addition of CuCl (0.155 g). The reaction mixture was stirred while become homogenous and green. At last, deoxygenated 4-VP (17 ml, preheated at 60ºC) was added. Polym-erization was performed at 60ºC for 48 h. The diblock copolymer MeOPEG-b-P4VP was purified by passing through Al2O3 column to remove the copper catalyst and then precipitated in cold diethyl ether, followed by filtration. The product was dried at 40ºC under vacuum.

MethodsSize exclusion chromatography (SEC) SEC analysis was performed on Waters 600 liquid

chromatograph equipped with a 410 refractive index detector and styragel HR columns HR 1, HR 2, and HR 4; and a UV detector-Waters 2487; in dimethylforma-mide (DMF). Polyethylene glycol standards were used for calibration.

The relative molecular mass (Mn, SEC) and polydisper-sity index (Mw/Mn) of the block copolymer were deter-mined in tetrahydrofuran (THF) (flow rate : 1mL min-1) at 40°C with a Waters liquid chromatrograph equipped with a 410 refractive index detector and Phenogel col-umns (three columns 50 Å, 100Å, 104Å). Polystyrene standards were used for calibration. Molecular mass was calculated on Clarity software.

Fourier Transform Infrared Spectroscopy (FTIR) was done on Perkin-Elmer Spectrum GX spectrometer.

1HNMR spectra of the MeOPEG macroinitiator and MeOPEG-b-P4VP copolymer were recorded with


137

a Brucker spectrometer (400 MHz) in CDCl3.

Thermogravimetric Analysis (TGA) TGA analyses were performed on TA Instruments Q

500. The temperature was increased from 30 to 700ºC at a heating rate of 20ºC/min under nitrogen atmosphere.

RESULTS AND DISCUSSION

Synthesis of MeOPEG-Cl macroinitiatorPoly (ethylene glycol) monomethyl ether (Me-

OPEG) (Mn = 1900 g mol-1), which has a hydroxyl end group, was converted to an appropriated ATRP macroinitiator for the polymerization of 4-VP. This was achieved by reaction of MeOPEG with thionyl chloride according to scheme 1 thus leading to the formation of suitable chloro-terminated ATRP macroinitiator.

vibration. The chlorination of MeOPEG-Cl macroinitia-tor is clearly evidenced by the appearance of a new band at 665 cm-1 which is characteristic for C-Cl bond (Fig. 1).

Further, MeOPEG-Cl was characterized by 1H NMR spectroscopy. 1H NMR spectrum of MeOPEG-Cl macroinitiator showed all characteristic for PEG units signals at 3.38 ppm (3H, -O-CH3) and 3.45-3.83 ppm (4H, -CH2-CH2-O-), (Fig. 2).

The number average molecular mass and molecular mass distribution of MeOPEG-Cl macrotinitator was determined by SEC. As seen from Fig. 3, SEC peak of MeOPEG-Cl macroinitiator was shifted toward higher molecular masses (Mn = 2050 g mol-1, PDI = 1.078) in comparison to the pure MeOPEG (Mn = 1949 g mol-1, PDI = 1.069) indicating the successful chlorination of MeOPEG macroinitiator, (Fig. 3).

Scheme 1. Chlorination of MeOPEG with thionyl chloride.

Thus prepared MeOPEG-Cl macroinitiator was characterized by FT-IR, 1H NMR and SEC analyses.

Initially, the MeOPEG-Cl macroinitiator was characterized by FT-IR analysis. IR spectra of pure MeOPEG and MeOPEG-Cl macroinitiator showed all characteristic signals for the MeOPEG units at: 2878 cm-1 (C-H stretching); 1095 cm-1 characteristic for -C-O-C- stretching vibration and 1469 cm-1 for C-H banding

Fig. 1. FT-IR spectra of MeOPEG and MeOPEG-Cl macroini-tiator.

Fig. 2. 1H NMR spectrum of MeOPEG-Cl macroinitiator.

Fig. 3. SEC analysis of pure MeOPEG and chlorinated MeOPEG-Cl.


138

Synthesis of MeOPEG-b-P4VP block copolymerP4VP is a pH sensitive polymer which is soluble

in water at pH lower than 4.7 and became insoluble in water at higher pH (> 4.7). Therefore, block copolymer of P4VP will respond to pH changes in the aqueous so-lution and self-assemble into different supramolecular structures [17].

The synthesis of MeOPEG-b-P4VP block copoly-mer by ATRP was performed in 2-propanol at 60ºC using MeOPEG-Cl macroinitiator in the presence of V4P and PMDETA/CuCl catalyst system according to Scheme 2.

The 1H NMR spectrum of the MeOPEG-b-P4VP block copolymer shows the peaks for MEOPEG’s termi-nal (a, 3H, -CH3) protons at 3.38 ppm and at 3.45-3.83 ppm (b, 4H, -CH2-CH2-O-) from the main PEG chain. The P4VP characteristic protons at 1.20 - 2.00 ppm (c, 3H), and its aromatic ring at 6.00 - 6.90 ppm (d, 2H)

and 7.70 - 8.90 ppm (e, 2H) assigned to the ortho- and meta-protons, respectively, were observed as well, (Fig. 4). This result demonstrated the successful synthesis of PEG-b-P4VP block copolymer.

The molecular masses and molecular mass distribu-tions of diblock copolymers were further investigated by SEC. Fig. 5 presents the SEC curve of the diblock copolymers PEG-b-P4VP block copolymer by using MeOPEG-Cl macroinitiator. SEC curve of diblock copolymer showed narrow and unimodal elution peak, which is clearly shifted toward higher molar mass (Mn = 6470 g mol-1, PD = 1.13) in comparison to the MeOPEG-Cl macroinitiator (Mn = 2050 g mol-1, PDI = 1.078), (Fig. 5).

Thermal stability of MeOPEG-Cl macroinitiator and MeOPEG-b-P4VP block copolymer were examined by thermal gravimetric analysis in the range from 30ºC

Scheme 2. Synthesis of MeOPEG-b-P4VP block copolymer by ATRP.

Fig. 4. 1H NMR spectrum of MeOPEG-b-P4VP block copolymers.

Fig. 5. SEC analysis of MeOPEG-b-P4VP block copolymer.


139

to 700ºC under nitrogen atmosphere.TGA analysis showed that MeOPEG-Cl macroini-

tiator and MeOPEG-b-P4VP block copolymer undergo thermal degradation above 300ºC, (Fig. 6). The maximal degradation temperature for the both were around 400ºC - 405ºC, which indicated that the second P4VP block did not impaire the thermal stability of the MeOPEG-b-P4VP block copolymer obtained.

CONCLUSIONS

MeOPEG-b-P4VP block copolymers were success-fully prepared by atom transfer radical polymerization. For the first time, MeOPEG-Cl macroinitiator, obtained by reaction of MeOPEG and thionyl chloride, was ap-plied for polymerization of 4VP by ATRP. The synthesis of well-defined MeOPEG-b-P4VP block copolymers was proven by 1H NMR spectroscopy with the presence of all typical for the both block signals. SEC analysis also confirms the formation of block copolymers by increasing of MeOPEG-b-P4VP molecular mass in comparison to the MeOPEG-Cl macroinitiator keeping low polydispersity index.

Acknowledgements

This study was supported by University of Chemical Technology and Metallurgy (UCTM), Sofia, Bulgaria, and the Scientific Research Center at UCTM.

REFERENCES

1. Sh-Ch. Chan, Sh.-W. Kuo, Ch.-H. Lu, H.-F. Lee, F.-Ch. Chang, Syntheses and characterizations of the multiple morphologies formed by the self-assembly of the semic-rystalline P4VP-b-PCL diblock copolymers, Polymer, 48, 2007, 5059-5068.

2. E.L. Thomas, D.M. Anderson, C.S. Henkee, D. Hoff-man, Periodic area-minimizing surfaces in block copolymers, Nature, 334, 1988, 598-601.

3. F.S. Bates, Polymer-polymer phase behavior, Science, 251, 1991, 898-905.

4. W.A. Braunecker, K. Matyjaszewski, Controlled/liv-ing radical polymerization: Features, developments, and perspectives, Progress in Polymer Science, 32, 2007, 93-146.

5. M. Jaymand, Synthesis and characterization of well-defined poly(4-chloromethyl styrene-g-4-vinylpyri-dine)/TiO2 nanocomposite via ATRP technique, J. Polym. Res., 18, 1617-1624.

6. J.S. Wang, K. Matyaszewski, Controlled/”Living” Radical Polymerization. Halogen Atom Transfer Radical Polymerization Promoted by a Cu(I)/Cu(II) Redox Process, Macromolecules, 28, 1995, 7901- 7910.

7. K. Matyjaszewski, J. Xia, Atom transfer radical polymerization, Chem. Rev., 101, 2001, 2921-2990.

8. K. Matyjaszewski, M.H. Acar, K.L. Beers, S. Coca, K.A. Davis, S.G. Gaynor, Block copolymer synthe-sis by atom transfer radical polymerization, Polym. Prepr., 40, 1999, 966-967.

9. R. Ma, B. Wang, Y. Xu, Y. An, W. Zhang, G. Li, L. Shi, Surface Phase Separation and Morphology of Stimuli Responsive Complex Micelles, Macromol. Rapid Commun.,28, 2007, 1062–1069.

10.W. Zhang, L. Shi, Z.-J. Miao, K. Wu, Y. An, Core–Shell–Corona Micellar Complexes between Poly(ethylene glycol)-block-poly(4-vinyl pyridine) and Polystyrene-block-poly(acrylic acid), Macro-mol. Chem. Phys., 206, 2005, 2354–2361.

11. S.Ch. Turmanova, Complexes of heavy metals with nitrogen containing copolymers – electrochemical and physicomechanical properties, eXPRESS Poly-mer Letters, 1, 2007, 585-593.

12. J Xia, X Zhang, K Matyjaszewski, Atom transfer radical polymerization of 4-vinylpyridine, Macro-

Fig. 6. TGA thermograms of MeOPEG-Cl macroinitiator and MeOPEG-b-P4VP block copolymer.


140

molecules, 32, 1999, 3531-3533.13. K.R.M. Vidts, F.E. Du Prez, Design of water-soluble

block copolymers containing poly(4-vinylpyridine) by atom transfer radical polymerization, European Polymer Journal, 42, 2006, 43-50.

14. Q. Bo, Y. Zhao, Double-hydrophilic block copolymer for encapsulation and two-way pH change-induced release of metalloporphyrins, Journal of Polymer Science, Part A: Polymer Chemistry, 44, 2006, 1734-1744.

15. L.M. Bronstein, S.N. Sidorov,V. Zhirov, D. Zhirov, Y.A. Kabachii, S.Y. Kochev, P.M. Valetsky, B. Stein, O.I. Kiseleva, S.N. Polyakov, E.V. Shtykova, E.V. Nikulina, D.I. Svergun, A.R. Khokhlov, Metalated

Diblock and Triblock Poly(ethylene oxide)-block-poly(4-vinylpyridine) Copolymers: Understanding of Micelle and Bulk Structure, J. Phys. Chem. B, 109, 2005, 18786-18798.

16. Q. Xiao, X. Zhang, J. Yi, X. Wang, H. Zhang, Synthesis of Poly(ethylene oxide)-b-polystyrene-b-poly(4-vinylpyridine) ABC Triblock Copolymers by Two-step Atom Transfer Radical Polymerization, Iranian Polymer Journal, 17, 2008, 781-790.

17. R. Bertani, P. Metrangolo, A. Moiana, E. Perez, T. Pilati, Supramolecular route to fluorinated coatings: self-assembly between poly (4-vinylpyridine) and haloperfluorocarbons, Adv. Mater, 14, 2002, 1197-1201.

Mladen Stančić, Branka Ružičić, Nemanja Kašiković, Dragoljub Novaković, Rastko Milošević

141


INFLUENCE OF SUBSTRATE THICKNESS ON THE QUALITY ATTRIBUTES OF POLYMER MATERIALS REVERSE PRINTED BY SCREEN PRINTING TECHNIQUE

Mladen Stančić1, Branka Ružičić1, Nemanja Kašiković2,

Dragoljub Novaković2, Rastko Milošević2

1 University of Banja Luka, Faculty of Technology, Graphic Engineering Vojvode Stepe Stepanovića 73, 78 000 Banja Luka Bosnia and Herzegovina E-mail: [email protected] University of Novi Sad Faculty of Technical Sciences, Graphic Engineering and Design Trg Dositeja Obradovića 6, 21 000 Novi Sad, Serbia

ABSTRACT

The printing substrate and its characteristics have a significant impact on print quality. Print quality includes de-sired colour reproduction and satisfactory reproduction of image elements. This study analyzes the transparent polymer material substrates, printed by a screen printing technique. The paper primary focus is on the analysis of colour differ-ences between obverse and reverse side of the printing substrate, and in addition, on line quality and the dot roundness analysis. Considering the results, it can be concluded that substrate thickness has influence on the colour reproduction in reverse printing when observing through a transparent substrate. Also, the differences between line area and perimeter in obverse and line area and perimeter – in the reverse printed side increases with the printing substrate thickness increase and dot roundness observed from the reverse side has a more irregular shape.

Keywords: polyethylene films, obverse and reverse printing, colour reproduction, line quality, dot roundness.

Received 18 September 2014Accepted 30 January 2015

INTRODUCTION

Polymer materials in the printing industry are mostly used in flexible packaging. As the market requires not only good quality but aesthetic appearance as well, manufacturers resort to attractive design and printing for guiding customers to faster recognition of their prod-ucts. Polymer substrates can be printed by flexographic, screen printing and, with technology advances, by ink jet printing technique. Nowadays, polymer substrates are mostly printed by flexographic technology, with low viscosity inks, but the disadvantage of this technique is the high implementation cost [1]. Screen printing has a great advantage in terms of cost when low-circulations are produced. Requirements for its use are very modest, and it can in essence be carried out anywhere in the world

with little prerequisites. Also, it is simple, fast, reproduc-ible, flexible and adaptable [2, 3]. On the other hand, polyethylene as a printing substrate is the most widely used plastic on earth with an annual production of ap-proximately 80 million tons [4]. Polyethylene is created by ethylene polymerization. The types are: PE-HD high density (> 0.941 g cm-3), PE-MD medium density (0.926 - 0.940 g cm-3), PE-LD low density (0.910 - 0.925 g cm-3) polyethylene. PE-LD is highly branched and contains relatively high amorphous content, which results in an outstanding clarity in the film for food packaging, which is a major application [5]. Polyethylene as a material has some characteristics that greatly affect the final result of the product. Beside transparency that shows the distor-tion of an object observed through the examined film, the haze has also a huge impact. It represents scattering


142

of light by a film that results in a cloudy appearance or poorer clarity of objects when viewed through the film. Light may be scattered by particles suspended in the substance, such as pigment particles or contaminants, or by an imperfect surface caused by dirt, oil, or a fine texture. Haze is an important appearance attribute which can be quantified and then used to assess the objects such as liquids, glass, plastics, painted panels, and even metals [6]. More technically, haze is the percentage of light transmitted through a film that is deflected more than 2.5° from the direction of the incoming beam. This property is used to describe transparent and translucent films, not opaque films. Haze is greatly influenced by material selection and product design. Additives and coatings usually contribute to increased haze. Also, thicker films will be hazier than thinner films. Additional variables, like process temperatures in the different stages of film-making, can further affect haze, so they are tightly controlled [7]. Reverse printing, i.e. printing on the underside of a transparent film, is mostly used in the production of packaging for food as snacks, juices, biscuits, ketchup, etc. In this technique the ink is printed on the inner layer of substrate and it does not become in direct contact with human skin, weather conditions and other substrates because in most of cases the ink is sandwiched between substrates by a lamination process. Reverse printed and laminated materials such as BOPP (biaxial oriented polypropylene), CPP (Cast Polypro-pylene), LDPE (Low density polyethylene) makes the perfect double, triple or quadruple layer films used for food packaging. Beside laminated products reverse printing is also used with an LDPE film for bottle label, overwrap film for books, bundling and pallets, towels and tissues, film for bakery goods, meat, coffee, frozen foods, liquid packaging (milk cartons and bag-in-box applications), bags and shoppers, packaged beverages industries and others. In case of a reverse printing sub-strate, characteristics can affect the image quality in many ways [8]. The colour reproduction can be charac-terized using the CIE L*a*b* colour coordinates. CIE L*a*b* coordinates enable colour differences estimation between samples, or between samples and standards, or in this research between reverse and obverse image of the same sample. Colour differences determination is based on calculations of the differences of colour space coordinates (ΔL, Δa, Δb) [9]. There are many systems that use different formulas (models) for colour distance

calculation, wherein the gain values can be classified into several groups: ΔEab < 0.2 (the difference is not visible), ΔEab between 0.2 and 1 (the difference is noticeable), ΔEab between 1 and 3 (the difference can be seen), ΔEab between 3 and 6 (the difference is easy to see) and ΔEab over 6 (obvious colour difference) [10]. Colour distance (ΔEab) is defined by the equation [11]:

(1)

In equation (1) ΔEab is the colour difference value, L* is the brightness value, a* is red/green colour value, b* is yellow/blue colour value.

Quantitative assessment of the colour reproduction is not enough for defining overall print quality. It is proven that print quality does not represent monotonous function of hue, saturation and value [12, 13]. Quality at-tributes as contrast, sharpness, macro non-uniformity are not connected to a colour reproduction but certainly have an affect on print quality. They are directly connected with line and dot quality that represent the component of any image [14]. Important print quality attributes are: contrast, sharpness, noise/graininess, edge ragged-ness, effective resolution, text quality, micro uniform-ity, macro non-uniformity, gloss uniformity, etc. [13]. Many researchers have questioned and acknowledged the importance of quality attributes. However, they did not reach an agreement which of them were the most important. Engeldrum in his work states that observers most likely will not be able to perceive more than five quality attributes simultaneously [15].

Line quality reproduction depends on numerous factors, including the printing plate (screen and the process of illumination), ink and substrate [16]. Some of the quality components are line width and its variations, raggedness and sharpness. Line raggedness indicates deviation of printed lines from ideal geometric shape and is the undesired line property that leads to print quality reduction. Line raggedness can be characterized by measuring area and perimeter of lines and comparing these values with ideal ones. Excessive line raggedness causes low sharpness of the prints and can cause text to become unclear or bold [11].

Printed dot fidelity is determined by evaluating dot area and its roundness [17]. Dot roundness represents the shape of the dot relative to a perfect circle. The ideal circular dot should be the one whose area is π/2 that of


143

the corresponding square pixel. If the dot is not perfectly round, its deviation can cause uneven ink coverage. Dot roundness is defined as [18]:Roundness = 4π (A/p2) (2)where A is the area of the dot and p is its perimeter. The roundness is equal to 1 for circle, and is less than one for any other closed figure. The closer the roundness is to 1, the better is the quality of the dot. Fleming et al. stated that their analyses and interpretation are general and applicable to any printing processes where image quality is governed by the smallest printable dot [19]. In screen printing, due to the technique itself, it is not expected that roundness reaches values as high as 1.

It should be pointed out that the aim of this study is to investigate the influence of transparent polyethylene film thickness on print quality. The analysis primaryly included the colour reproduction, but important quality attributes such as line quality and dot roundness, were also examined. Colour reproduction was analysed by spectrophotometric measurements, and secondary im-age quality parameters evaluation was performed by the method of digitally analysed scanned imprint images in specified programming languages.

EXPERIMENTAL

This research included four different thicknesses of polymer material of the same composition. The mate-rial used was PE-LD low density polyethylene (0.910 - 0.925 g cm-3). Used thicknesses were 30, 35, 38 and 110 mm. The samples were printed with the screen print-ing technique. Test image, used in research, was created using Adobe Illustrator CS5.5 software. The test image dimension was 210 x 297 mm and contained different elements used for print quality control. Elements used for obtaining the results were 2.54 x 2.54 mm solid tone patches, lines 2 pt width and dots with 0.6 mm diameter. Screen with thread count of 120 threads per cm was used for printing plates, which were made convention-ally using linearized positive films. As a photo sensitive coating Sericol Dirasol 22 emulsion was used. Exposure was performed 90 seconds with a RUVA lamp (40W), where the distance from the lamp to screen measured 20 cm. Printing speed amounted to 15 cm per second, squeegee hardness - 85 Shore Type A, while snap-off distance measured 4 mm. Ink used for printing was Seri-col Polyscreen yellow colour PS-043/1 (Mid Chrome).

Fixation of printed samples was carried out on room temperature in duration of 4 hours. After fixation, the printed sheets were digitalized by Mustek 1200 ub plus flat scanner for further analysis. The scanning resolution was set as 600 spi and all auto functions were turned off.

CIE L*a*b* colour coordinates on the obverse and reverse side of a printed sample were obtained using diffuse spectrophotometer Konica Minolta cm-2600d (Illuminant D65, Standard Observer 8º, geometric con-dition d/8, measurement area 11 mm).

Line reproduction was assessed by measuring area and perimeter of 2 points thick lines and comparing these values with area and perimeter values of an ideal image created using a computer in resolution of 600 ppi.

Dot roundness was obtained by measuring printed dots with 0.6 mm diameter. All values of importance were calculated using ImageJ software, implemented in Java for the creation, visualization, editing, processing, and image analysis [20].


Colour reproductionThe influence of substrate thickness on the change

of the sample features by screen printing technique was experimentally determined. First the CIE L*a*b* coordi-nates of the samples on the obverse printed side were de-termined with a spectrophotometer. Five measurements were performed for each sample, and was used the mean value as the result. The same process was performed for the reverse side of printing, and the obtained results are shown in Table 1, where the film thickness used in the printing process for every sample is specified.

On the basis of the obtained results the colour dif-ference values of the obverse side of the printed sample and the reverse side measured through the transparent film, are determined. The obtained results show that there is a change in colour. These changes are small and can be listed in two groups of values - a group of values for colour difference where the difference is noticeable (ΔEab between 0.2 and 1), and a group where the colour difference between obverse and reverse side results is visible (ΔEab between 1 and 3). An increase of colour dif-ference with an increasing of film thickness is noticeable which shows that film thickness affects colour, i.e. with the increase of film thickness the film has a greater haze and it brings changes to colour coordinates readings.


144

To determine the dependence of colour difference between two sides of the transparent polyethylene film sample on the thickness of the film, a mathematical mod-el, based on a simple linear regression, was developed (Bivariate Linear Regression). Simple linear regression focuses on explaining/predicting one of the variables on the basis of information for the other variable. The re-gression model thus examines changes in one variable as a function of changes or differences in values of the other variable. The relationship between dependent (criterion) variable and independent (predictor) variable explored by the linear regression is described by the general linear model. The mathematical equation for the general linear model using population parameters is [21, 22]:

εββ ++= XY 10 (3)where:Y - dependent variable,X - independent variable,

0β - an adjustment constant due to scale differences in measuring X and Y (the intercept or the place on the Y axis through which the straight line passes. It is the

value of Y when X = 0).1β - the change along the Y scale expected for every

unit changed in fixed values of X (represents the slope or degree of steepness).

ε - an error component for each individual meas-urement, the portion of Y score that cannot be accounted for by its systematic relationship with values of X.

Constants 0b and 1b that represent the theoretical evaluation of parameters 0β and 1β , are determined from the results of experimental research.

In Table 2 are shown the obtained results of the simple linear regression coefficients for the proposed mathematical model, equation (3) that determine the dependence of the colour difference ΔEab between two sides of the transparent film, on the film thickness L.

Statistical analysis of the colour difference results of printed transparent polyethylene film with a simple linear regression model analysis revealed the existence of a statistically reliable mathematical model of depend-ence of the colour difference from film thickness, with simple linear regression coefficient R2 = 0.966.

Table 1. Average values of CIE L*a*b* coordinates and ΔEab.

Table 2. Simple linear regression coefficients for the obtained mathematical model that deter-mine the dependence of the colour differences ΔEab between two sides of the transparent film print, on the film thickness L.

ΔEab = 0.43 + 0.01 ∙ L (mm) Simple linear

regression coefficient

Std. Error of the Estimate b0 = 0.43 b1 = 0.01

r2 s Std. Error t Std. Error t

0.966 0.08358 0.080 5.411 0.001 7.570

Film thickness

Sample side L* a* b* ∆Eab

30 mm Obverse 77.66 20.45 83.23 -

Reverse 77.39 20.91 83.60 0.65

35 mm Obverse 77.47 20.88 83.88 -

Reverse 77.25 21.41 83.40 0.75

38 mm Obverse 77.71 20.09 82.96 -

Reverse 77.50 20.38 83.78 0.89

110 mm Obverse 77.26 20.75 82.48 -

Reverse 76.46 21.77 81.75 1.48


145

Line reproductionLine analysis, as an essential attribute of print qual-

ity, was performed by measuring the surface area and perimeter lines. The results obtained by measuring the 2 pt thick lines from both sides of the printed sample are shown on Fig. 1. The smallest difference is with the thinnest film of 30 mm, and the difference increases with the increasing of film thickness, so the biggest difference is with the thickest film of 110 mm.

In this graph, the differences between the line areas from both sides of the film can be seen. It is necessary to examine the relations between differences of the line areas on the obverse and line areas on the reverse side of print, then between the line areas on the obverse side of the print and the ideal line as well as the difference between line areas on the reverse side and the ideal line. The results are shown in Fig. 2.

Results in Fig. 3 show 2 pt (0.7 mm) line perimeter values. The smallest difference between the line perim-

eters is on the 30 mm thick film sample and it gradually increases with increasing film thickness, so that the big-gest difference between the obverse and reverse side of a film is on the 110 mm thick film.

In this graph, differences between the line perimeters from both sides of the film can be seen. It is necessary to examine the relations between differences of line perimeters on the obverse and line perimeters on the reverse side of print, then between the line perimeter on the obverse side of the print and the ideal line, as well as the difference between line perimeters on the reverse side and the ideal line. The results are shown in Fig. 4.

Dot reproductionIn addition to analysis of line area and perimeter,

the quality of dot reproduction, which is determined on the basis of its roundness, is a more important factor for determining the print quality. Roundness of a dot was analyzed with ImageJ software. Dots with 0.6 mm in

Fig. 1. Graphical representation of line area values.

Fig. 2. Graphic representation of line area trend with increasing of film thickness.


146

Fig. 3. Graphic representation of line perimeter values.

Fig. 4. Graphic representation of line perimeter movements with increasing of film thickness.

Fig. 5. Graphic representation of dot roundness values.


147

diameter were measured. Measurements were performed with five dots from the same sample and average values are presented in Fig. 5.

Dot roundness, according to the results, decreases on the reverse compared to the obverse side of the print. There the dot gets a more irregular shape, far away from the ideal circle. The resulting changes, or distortion of dot roundness from the reverse, compared to the obverse side of print, are very small but they do exist and increase from the thinnest film, where the difference is minimum, to the thickest 110 mm film, at which point the dot has the most irregular shape. That growth is shown in Fig. 6.

CONCLUSIONS

The aim of this paper is to point out the influence of the substrate thickness on the reverse print on trans-parent polymeric materials. Low-density polyethylene films were printed with the screen printing technique, and the print was examined through the transparent substrate. With the aim of determining the influence of the substrate, the basic attributes of print quality were analyzed: reproduction of colour, lines and dots.

Light that is scattered upon passing through a film or sheet of a material can produce a hazy or smoky field when objects are viewed through the material, like with the reverse printing. Thin films make lower light scat-tering and with the increase of the film thickness, light scattering becomes higher. These changes of substrate thicknesses influence print quality attributes, i.e. the thicker the film, the greater the differences, between the

attributes measured from the obverse and the reverse side of print are.

Colour reproduction was analysed by spectropho-tometric measurements and a colour difference on the reverse side compared to the obverse print side, was noticed. Although the differences are small and unno-ticeable to the inexperienced eye, it is very important to note that they exist and that they are increasing with the increase of substrate thickness.

Assessment of quality attributes was performed by the analysis of digitalized print images in an appropri-ate programming language. Quality attributes that were analyzed are line quality and dot roundness.

Line quality was estimated on the basis of line area and line perimeter. Line area and line perimeter for all samples clearly deviate from the ideal. Comparison of the line areas and perimeters on the obverse and the reverse side, shows a difference that becomes greater with the film thickness increase.

Dot roundness, according to the results, decreases from the reverse of the print side, as compared to the printed side. Dots are getting an irregular shape, far away from the ideal circle. However, in this case the difference between the roundness of a dot from the printed side, and the roundness of a dot from the reverse side increases with the increase of the substrate thickness.

Considering all results, we can conclude that the substrate with its thickness affects the quality of print. With the aim of further knowledge, influence of transpar-ent film thickness on other important quality attributes such as text legibility and macro nonuniformity, should

Fig. 6. Graphic representation of dot roundness trend with increasing of film thickness.


148

be examined. It would also provided an opportunity to examine the influence of other material properties such as surface tension, surface roughness, dampening and contact angle on the print quality.

AcknowledgementsThe research is supported by the Ministry of educa-

tion, science and technology development of Republic of Serbia, project number: 35027 „Development of software model for scientific and production improve-ment in graphic industry“ and project CEEPUS III RS - 0704 - 03 - 1415, „Research and Education in the Field of Graphic Engineering and Design“.

REFERENCES

1. H. Kipphan, Handbook of print media: Technologies and Production Methods, Springer, Berlin, 2001

2. T.M. Lee, Y.J. Choi, S.Y. Nam, C.W. You, D.Y. Na, H.C. Choi, D.Y. Shin, K.Y. Kim, K.I. Jung, Color filter patterned by screen printing, Thin Solid Films, 516, 2008, 7875-7880.

3. F. Krebs, M. Jørgensen, K. Norrman, O. Hagemann, J. Alstrup, T. Nielsen, J. Fyenbo, K. Larsen, J. Kristensen, A complete process for production of flexible large area polymer solar cells entirely using screen printing - First public demonstration, Solar Energy Materials and Solar Cells, 93, 2009, 422-441.

4. R. Szentgyörgyvölgyi, E. Novotny, Investigation of flexographic printing on PE and BOPP foils, Proceedings of GRID ‘10, Novakovic, D. (Ed.), Novi Sad, Serbia, 2010, FTN, GRID, NOVI SAD (2010), pp. 337-342.

5. D.B. Malpass, Introduction to Industrial Polyethylene: Properties, Catalysts, and Processes, John Wiley & Sons, Inc., New Jersey, 2010

6. Technical Services Department Hunter Associates Laboratory, Inc., Insight on Color, Haze. HunterLab Application Note, 9, 6, 2008, 1-4

7. I. Vujković, K. Galić, M. Vereš, Ambalaža za paki-ranje namirnica, Tectus, Zagreb, 2007.

8. Y. Kipman, Image Quality Metrics for Printers and Media, IS&T’s Conference - Image Processing, Image Quality, Image Capture, Systems Conference, Portland, 1998, 183-187.

9. N.T. Kočevar, Kolorimetrijska analiza i vizuelna

ocjena boja na dvobojnoj tkanini. Tekstil, 55, 2006, 127-134.

10. K. Schläpfer, Farbmetrik in der grafischen Industrie. UGRA, St. Gallen, 2002.

11. M. Stančić, D. Novaković, N. Kašiković, V. Vukmirović, B. Ružičić, Influence of material composition on print quality digitally printed textile substrates, Tekstilna industrija, 1, 2013, 37-43.

12. E. A. Fedorovskaya, H. De Ridder, F. Blommaert, Chroma variations and perceived quality of colour images of natural scenes, Color research and ap-plication, 22, 2, 1997, 96-110.

13. M. Pedersen, N. Bonnie, J. Hardeberg, F. Albregtsen, Attributes of a new image quality model for color prints, Proceedings of Color Imaging Conference, Albuquerque, New Mexico, USA, 2009, 204-209.

14. A. Dhopade, Image quality assessemnet according to ISO 13660 and ISO 19751, Test Targets 9.0, Published by RIT School of Print Media, Rochester, NY, 2009, 43 - 50

15. P. G. Engeldrum, A Theory of Image Quality: The Image Quality Circle, Journal of Imaging Science and Technology, 2004, 446-456.

16. S. Ingram, Screen Printing Primer, Sec. Ed. Sewickley, GATFPress, 1999.

17. J. Sarafano, A. Pekarovicova, Factors Affecting Dot Fidelity in Solvent Based Publication Gravure, American Ink. Maker, 1999, 32-36.

18. M. Stančić, D. Novaković, I. Tomić, I. Karlović, Influence of Substrate and Screen Thread Count on Reproduction of Image Elements in Screenprinting, acta graphica, 23, 2012, 1-12.

19. P.D. Fleming, J.E. Cawthorne, F. Mehta, S. Halwawala, M.K. Joyce, Interpretation of Dot Fidelity of Ink Jet Dots Based on Image Analysis. The Journal of Imaging Science and Technology, 5, 2003, 394-399.

20. W.S. Rasband, ImageJ, U.S. National Institutes of Health, Maryland, USA, http://rsb.info.nih.gov/ij/, 1997-2006.

21. J. Komić, Metodi statističke analize kroz prim-jere - zbirka zadataka. Univerzitet u Banjoj Luci, Ekonomski fakultet, Banja Luka, 2000.

22. J. Pallant, SPSS - A Step by Step Guide to Data Analysis Using SPSS for Windows, Allan & Unwin, Australia, 2007.

Nadejda Daskalova, Gabriela Marinova

149


study of the application in the brewing of enzyme preparation, containing laminarinase and lichenase,

produced from TRICHODERMA SP. 405Nadejda Daskalova, Gabriela Marinova

Institute of Cryobiology and Food Technology53 Cherni Vrah Blvd.1407 Sofia, Bulgaria,E-mail: [email protected] [email protected] [email protected]

abstract

The β-glucanase enzymes find wide application in the brewing. It was obtained an enzyme preparation with laminarinase and lichenase activity, produced from Trichoderma sp. 405 and was investigated its application in the tech-nological process of brewing wort preparation. It was found that the studied enzyme preparation has a hydrolytic effect on the b-glucans and the hemicelluloses of the used ingredients. From economical point of view the obtained enzyme preparation, added at a dose of 0,04 % compared to the used ingredients, leads to receiving of brewing wort, containing b-glucans suitable for healthy lifestyle.

Keywords: laminarinase, lichenase, b-glucans, enzyme preparation, brewing, brewing wort.

Received 22 April 2014Accepted 30 January 2015

introduction

The b-glucanase enzymes pertain to the group of hemicellulases. They catalyze the degradation of b-D-glucans. Among the microorganisms, synthesizing b-glucanases, are the strains of genus Trichoderma - Trichoderma reesei [1], T. harzianum [2], T. viride [3], T. koningii T 199 [4], T. sp. GXC [5], T. asperellum [6]. The group of b-glucanase enzymes include also the laminarinase and the lichenase. The laminarinase [Е.С. 3.2.1.39] attacks 1,3-b-D-glycosidic bonds into 1,3-b-D-glucans. Its substrate is the polysaccharide laminarin [7]. The lichenase [Е.С. 3.2.1.73] hydrolyses b-D-glucans, containing 1,3- and 1,4-bonds. Its substrates are the li-chenin and the b-D-glucans in the cereals. The content of 1,3 and 1,4-b-glucans in the endosperm cell walls of the various cereals is different. Linear 1,3-;1,4-b-glucans have been isolated from seed endosperm of barley, oats, rye, wheat, millet, ryegrass. The content of β-glucans of different Bulgarian oats varieties ranges from 2,71 %

to 4,26 % [8], in different Bulgarian barley varieties is in the range of 3,15 % - 6,08 % [9], in traditional USA oats varieties is around 4,4 - 4,9 % and in some special selected USA oats varieties reaches to 7,5 - 7,8 % [10]. The content of b-glucans in barley ranges from 1,5 - 2,5 % [11] to 3,0 - 4,5 % [12]. The b-glucans are the main component of barley and malt cell walls [7, 11, 13]. It is considered that they contain approximately 75 % b-1,3-1,4-glucans, 20 % arabinoxylans (noted as pentosans) and 5 % proteins, with traces of ferulic acid. According to Bamforth et al. [14, 15] the internal layer of barley and malt cell walls to the endosperm is composed from dense layer of b-glucans, connected with proteins and the external layer contains xylans, arabinose residues and ester linked acetic and ferulic acid among them (Fig. 1).

Beer is traditionally made from malt, hops, yeasts and water, but the technology is based mainly on the ac-tion of the enzymes, activated during the brewing. In the process participate 4 main categories enzymes: a- and b-amylases, (carboxy)peptidases and b-glucanases. All


150

of them act synergistically. For brewing is necessary that the brewing wort con-

tains b-glucans as low as possible, i.e. the b-glucans have to be completely degradated. When they are not fully degradated they exert negative influence on the brewing wort and the beer- increased viscosity and deteriorated filtration [16 - 21]. The b-glucans are partially cleaved by the action of b-glucanases and as a result is observed the release of high molecular mass b-glucans from complex compounds with proteins and other substances, as well as a cleavage of the b-glucans with high molecular mass to low molecular mass b-glucans, including glucose [22].

Usually the b-glucanases, which are used in the brewing, are isolated from Trichoderma reesei, Aspergillus niger, Penicillium emersonii and Bacillus subtilis [18]. It was established better filtration velocity of the brewing wort when enzyme preparations with b-glucanase activity were added in the production process of beer and brewing wort [23]. More effective degradation of glucans and viscosity decreasing of brew-ing wort and beer were established when enzymes with endo-β-1,3(4)-glucanase and xylanase activities were used in the production process of beer [10].

The aim of this research is to create a technological scheme for receiving an enzyme preparation, containing b-glucanase (laminarinase Е.С.3.2.1.39 and lichenase Е.С.3.2.1.73), and to study its application in the brewing.

experimental It was used Trichoderma sp. 405, bought from the

National bank for industrial microorganisms and cell cultures - NBIMCC.

A spore inoculation product was obtained from the strain producer cultivated on potato-dextrose agar (PDA- Biokar Diagnostics) slants. Mandel’s sowing media was inoculated with the spore sowing material. Mandel’s fermentation media was inoculated with the sowing material [24]. The Mandel`s sowing media content is: NH4Cl 1.00 g l-1; urea 0.30 g l-1; KH2PO4 2.00 g l-1; (NH4)2SO4 1.40 g l-1; MgSO4.7H2O 0.30 g l-1; CaCl2.2H2O 0.40 g l-1; Corn extract 10.00 ml l-1; Glucose 20.00 g l-1. The Mandel’s fermentation media content is: NH4Cl 1.00 g l-1; urea 0.30 g l-1; KH2PO4 2.00 g l-1; (NH4)2SO4 1.40 g l-1; MgSO4.7H2O 0.30 g l-1; CaCl2.2H2O 0.40 g l-1; Corn extract 0.1 %; Wheat bran 1 %. The cultivation of Trichoderma sp. 405 was car-

ried out by the method of deep fermentation in 500 ml flasks on a shaker “Inkubations - Schüttelschrank BS-4 B.Braun” (220 rpm) for 144 hours at 28оС.

The activity of both enzymes was obtained spec-trophotometrically by measuring the reducing groups, which are released when hydrolyzing their substrates - laminarin and lichenin, respectively. An Unicam SP 1800 Ultraviolet spectrophotometer was used. For determina-tion of enzyme activities was applied Somogyi-Nelson’s method [25, 26]. One international unit (IU) of lamina-rinase activity is the amount of enzyme which catalyses the transformation of 1 mmol of reducing sugars in 1 cm3 per minute at 40oC in a citric phosphate buffer of pH 5.00. One international unit (IU) of lichenase activity is the amount of enzyme which catalyses the transforma-tion of 1 mmol of reducing sugars in 1 cm3 per minute at 40oC in an acetate buffer of pH 4.00.

The protein contents in the liquid culture in mg ml-1 was obtained by the spectrophotometric method (OD280/OD260) [27].

The lyophilization was carried out in a freeze-drying installation „HOCHVAKUUM” TG-16. The freezing process of the liquid culture was done at -35оС. The lyophilization’s duration was 28 hours. The cooling temperature of the desublimator was -75оС and the tem-perature of the plates was -35оС. The enzyme activities were recovered by adding 10 ml distilled water to the lyophilized liquid culture and then they were analysed by the above mentioned Somogyi-Nelson’s method [25, 26]. The residual moisture content of the lyophilized enzyme preparation was determined according to Bul-garian State Standard-1109-89- modified method by using an electronic scale “Sartorius thermo-control” with infrared heating of the samples.

The total mesophilic aerobic microorganisms (CFU g-1) of the lyophilized enzyme preparation is determined according to ISO 4833:2003 [28]; Escherichia coli and coliforms (CFU g-1) of the lyophilized enzyme preparation are analyzed according to ISO 21528-2: 2004 and ISO 4832: 2006 [29, 30]. The determination of Staphylococcus aureus (CFU g-1) of the lyophilized enzyme preparation is according ISO 6888-1: 1999 [31], and the determination of Salmonella sp. (CFU g-1) of the lyophilized enzyme preparation – according to ISO 6579: 2002 [32].

The gamma-irradiation of the samples lyophilized enzyme preparation was performed with the following


151

parameters: radionuclide Cs137; type of radiation - gam-ma rays; activity of Gamma-irradiation installation-„GAMA-1300”, dose rate 1,75 Gy min-1; radiation doses: 0, 5 and 10 kGy; operating temperature - 20оС - 22оС.

To study the application of the lyophilized enzyme preparation in obtaining of the brewing wort are con-ducted laboratory runs according to infusion regime in the production of 10 % bright beer when 30 % of the malt is replaced by 30 % unmalted barley. The enzyme preparation was dosed in the beginning of the process in amounts presented in Table 1 and the doses are con-formed with its activity and with the activities of other industrial preparations (from 776 to 12700 IU g-1 and recommendatory doses from 150-250 to 350-450 g t-1 grist) [23].

The brewing wort is obtained on laboratory au-tomatic apparatus of the German company “Bender & Hobein” by infusion method of the Department of Technology of Beer and Beverages of the Institute of Cryobiology and Food Technology. In this method the malt grist was mixed with water in a ratio 1:4 and after cytolytic, proteolytic and amylolytic degradation is adjusted to a certain volume and then is filtered. The obtained laboratory brewing worts are analysed by the following methods of the EBC Analysis Committee [33]: the time for saccharification is checked with io-dine solution [34]; the speed of draining is determined taking into account the past milliliters brewing wort for a certain period of time [34]; the yield of the extract is determined by calculating the amount of the extracted

substances from 100 g malt in the result of the mashing process [34]; the brewing wort extract is determined according to the measured relative mass of a certain quantity brewing wort and according to an official table is reported as g extract in 100 g brewing wort [33]; the determination of the brewing worts and beer рН val-ues is made with рН-meter [33]; the brewing wort and beer colour is determined spectrophotometrically (l = 430 nm) [33]; the brewing wort viscosity is measured with viscometer Höppler [33]; the soluble nitrogen is determined spectrophotometrically (l = 215 nm and l = 225 nm) [35]; the β-glucans determination is performed according to the method of Erdal [36].

results and discussion

technological scheme for receiving an enzyme prepa-ration with laminarinase and lichenase activities

It was developed a scheme for obtaining a lyo-philized enzyme preparation with laminarinase and lichenase activities, produced from Trichoderma sp. 405, which combines several technological stages, presented schematically in Fig. 2.

characterization of the received enzyme preparationThe total yield after vacuum-freeze drying of 1 l

liquid culture is 7,815 g lyophilized enzyme preparation. The obtained lyophilized enzyme preparation is a powder product with uniform consistency and yellowish-brown colour. The laminarinase activity of the filtered liquid

external layer

internal layer

Xylans Arabinose

b-Glucans Acetic acid

Proteins Ferulic acid

Fig. 1. Structure of the cell walls of barley. Fig. 1. Structure of the cell walls of barley.


152

culture, determined immediately after the fermentation is 0,520 IU ml-1, and the lichenase activity - 0,107 IU ml-1. After vacuum-freeze drying the laminarinase and lichenase activities of the lyophilizate, determined im-mediately after the lyophilization, are respectively 65,26 IU g-1 and 13,59 IU g-1. It was found that the obtained lyophilizate is with average residual moisture content of 8,65 %. This ensures the structural and microbiological stability of the product during its storage.

The results from the microbiological analysis of the received lyophilized enzyme preparation show absence of pathogenic microorganisms - Escherichia coli and coliforms (CFU g-1 enzyme preparation), Staphylococcus aureus (CFU g-1 enzyme preparation) and Salmonella

sp. (CFU g-1 enzyme preparation) were not isolated. The total mesophilic aerobic microorganisms of the lyophilized enzyme preparation is 106 CFU g-1, i.e. this result is above the norm for total mesophilic aerobic mi-croorganisms according to the norms of Bulgarian State Standards and ISО. This can be explained by the fact that the filtering of the liquid culture and the subsequent preparation of the samples before the lyophilization process are performed in non-sterile conditions. In order to reduce the reported high number of total mesophilic aerobic microorganisms the lyophilized enzyme prepa-ration was subjected to a gamma-irradiation with low doses with 5 kGy and with 10 kGy, respectively.

After irradiation with 5 kGy and 10 kGy the lamina-

obtaining of a liquid culture after fermen-tation of Trichoderma sp. 405 for 144 hours

filtration of the liquid culture

placing the samples in freeze drying trays with thickness of the layer of

10 mm

freezing of the samples

at - 35оС ± 2оС

gamma-irradiation

microbiologically clean lyophilized enzyme

preparation

lyophilized enzyme preparation

lyophilization in freeze-

drying installation „hochVaKuum” tg-16

Fig. 2. Technological scheme for receiving an enzyme preparation with laminarinase and lichenase activities, produced from Trichoderma sp. 405.

Content of: Control sample Variant 1 Variant 2 Variant 3 Malt,

% of raw material 70 70 70 70

Barley, % of raw material 30 30 30 30

Enzyme preparation, % of raw material - 0,04 0,40 2,0

Table 1. Amounts enzyme preparation used in the receiving of brewing wort.


153

rinase and lichenase activities remain high. It was found that the irradiation with 5 kGy is more favorable, because the reducing of laminarinase and lichenase activities is less compared to the samples, irradiated with 10 kGy. The percentage of preservation of the enzyme activities is presented in Table 2. The results for total mesophilic aerobic microorganisms are shown in Table 3. From the analysis it can be concluded that the irradiation with 5 kGy is enough to ensure considerable preservation of the enzyme activities of the lyophilized enzyme preparation and its necessary microbiological purity, which makes possible to use it in contact with food products.

study of the application of the obtained enzyme in the brewing

It was studied the effect of the lyophilized enzyme (subjected to gamma-irradiation with 5 kGy) with lami-narinase and lichenase activities 60,13 IU g-1 and 12,43 IU g-1, respectively, in the wort preparation. Laboratory runs according to infusion regime in the production of 10 % bright beer, when 30 % of the malt is replaced by 30 % unmalted barley, were conducted.

In Table 4 are presented some physico-chemical parameters of the malt used in the laboratory runs and in Table 5 - some physico-chemical parameters of the barley used in the laboratory runs. With these ingredients it is expected to obtain beer with relatively good quality.

Table 6 shows the parameters of the brewing wort.

The used quantities of enzyme influence the filtration speed, the extract values and the viscosity. The time for saccharification for the first three samples is similar. The speed of draining of the brewing wort in the variants is respectively 0,77 %, 2,77 % and 3,41 % higher than that of the control sample. The extract of the obtained brewing worts is in the range of 10,20 - 10,31 % for the control sample and the first two variants. In Variant 3 the increasing of the extract is with 0,44 % relative to the control sample. The рН values meet the requirements for the technology of bright beers. They vary in a nar-row range 5,78 - 5,81. The brewing worts viscosity with dosed enzyme is the same 1,57 mPa s and it is 0,04 mPa s lower than that of the control sample.

In Fig. 3 are presented the received results for the yield of the extract. The values are in the range of 72,98 - 76,50 %. The yield in Variant 1 is 0,24 % higher than that of the control sample, in Variant 2 - 0,88 % higher than that of the control sample and in Variant 3 - 3,52 % higher than that of the control sample.

In Fig. 4 are shown the colour changes of the labora-tory brewing worts, measured spectrophotometrically. The control sample and Variant 1 have almost the same colour. Probably the enzyme colour and the doses in Variants 2 and 3 affect the colour of the brewing worts. In Variant 2 the increasing is 1,3 times and in Variant 3 - 2,7 times.

Analysis phase Laminarinase activity preservation, %

Lichenase activity preservation, %

Immediately after lyophilization of the liquid culture (0 kGy) 98,07 98,15

After gamma-irradiation of the lyophilizate with 5 kGy 90,37 89,81

After gamma-irradiation of the lyophilizate with 10 kGy 86,32 86,11

Table 2. Preservation of laminarinase and lichenase activities of the lyophilizate, determined immediately after lyophiliza-tion and after its gamma-irradiation with 5 kGy and with 10 kGy.

Total mesophilic aerobic microorganisms, determined of the lyophilized enzyme

preparation

Established result

Norm for Total mesophilic aerobic microorganisms according to Bulgarian

State Standards and ISО Before gamma-irradiation 106 CFU g-1 103 CFU g-1

After gamma-irradiation with 5 kGy 0 CFU g-1 103 CFU g-1

After gamma-irradiation with 10 kGy 0 CFU g-1 103 CFU g-1

Table 3. Total mesophilic aerobic microorganisms of the lyophilized enzyme preparation after gamma-irradiation.


154

168

41

14

15

0 20 40 60 80 100 120 140 160 180

Beta-glucans, mg/l

Control sample

Variant 1

Variant 2

Variant 3

Fig. 5. b-Glucans content in the obtained brewing worts.

Parameters Values

Moisture, % 4,3 Time for saccharification, min. 10 Extract, % ADM., - fine grist - roughly grist

79,4 78,0

Extract difference, % 1,4 Colour, ЕВС 2,5 Viscosity, mPa.s 1,47 Soluble nitrogen, - mg/100 ml - mg/100 g

76,6 681

Total protein, % ADM 10,4 Kolbach Index, % 40,9

Table 4. Analysis of the malt used for the laboratory runs.

Parameters Values

Moisture, % 12,4 Absolute mass, g 44,8 Uniformity over sieve, %: - 2,8 mm - 2,5 mm (2,5 + 2,8 mm) - 2,2 mm - below 2,2 mm

70,5 94,9 4,3 0,7

Impurities, % 0,1 Total protein, % ADM 11,3

Table 5. Analysis of the barley used for the laboratory runs.

Parameters Control sample Variant 1 Variant 2 Variant 3

Saccharification, min. 10 10 10 15 Filtration speed, l h-1.m-2 16,99 17,12 17,46 17,57

Extract, % 10,20 10,23 10,31 10,64 рН 5,80 5,78 5,81 5,78 Viscosity, mPa s 1,61 1,57 1,57 1,57

Table 6. Parameters of the laboratory brewing worts.

3.5 3.64.6

9.4

0

2

4

6

8

10

Co

lou

r, E

BC

Control sample Variant 1 Variant 2 Variant 3

Fig. 4. Colour of the obtained brewing worts.

72.98 73.22 73.86 76.5

0

10

20

30

40

50

60

70

80

Yie

ld o

f th

e ex

trac

t, %

Control sample Variant 1 Variant 2 Variant 3

Fig. 3. Yield of the extract in the obtained brewing worts.


155

Fig. 5 presents the effect of the input enzyme prepa-ration on the contents of b-glucans. In the control brew-ing wort they are in an amount of 168 mg l-1. The enzyme preparation has clearly underlined effect on them and breaks them down 4,1 times in variant 1, 2-12 times in variant 2 and 3-11 times in variant 3. Independently of the measured low laminarinase and lichenase activi-ties, the obtained enzyme, although it is not purified, can be considered suitable for use for the preparation of brewing wort.

conclusions

Based on the results for the brewing wort prepara-tion, it can be concluded that the enzyme prepared with laminarinase and lichenase activities, produced from Trichoderma sp. 405, has a hydrolytic effect on the b-glucans and the hemicelluloses of the used ingredients, more pronounced in decreasing of the b-glucans content and less in increasing of the brewing wort filtration speed.

references

1. T.M. Coenen, A.C. Schoenmakers, H. Verhagen, Safety evaluation of beta-glucanase derived from Trichoderma reesei: summary of tozicological data, Food Chem. Toxicol., 33, 10, 1995, 859-866.

2. E.F. Noronha, C.J. Ulhoa, Characterization of a 29-kDa beta-1,3-glucanase from Trichoderma harzianum, FEMS-Microbiol.-Lett., 183, 1, 2000, 119-123.

3. A.A. Kulminskaya, K.K. Thomsen, K.A. Shabalin, I.A. Sidorenko, E.V. Eneyskaya, A.N. Savel, K.N. Neustroev, Isolation, enzymatic properties and mode of action of an exo-1,3-b-glucanase from Trichoderma viride, Eur. J. Biochem., 268, 2001, 6123-6131.

4. J.Shi, F. Cui, Selection of beta-glucanase-producing Trichoderma koningii T199 and its fermentation conditions, Wei Sheng Wu Xue Bao, 41, 6, 2001, 750-752.

5. J.Sun, W. Li, S. Gu, Production and some properties of a beta-glucanase from Trichoderma sp. GXC, Wei Sheng Wu Xue Bao, 41, 4, 2001, 457-462.

6. M.T. Bara, A.L. Lima, C.J. Ulhoa, Purification and characterization of an exo-beta-1,3-glucanase produced by Trichoderma asperellum, FEMS-

Microbiol.-Lett., 219, 1, 2003, 81-85. 7. N.А. Rodionova, L.V. Kaprelyants, P.V. Serednitsky,

А.Yu. Кilimnik, Hemicelluloses of cereal grains and enzymes catalyzing their decomposition, Applied Bi-ochemistry and Microbiology, 28, 5, 1992, 645-665.

8. N. Мichalkova, Ivanka Petrova, Lubka Georgieva, Nadezda Antonova, Contents of beta-glucans in Bulgarian oat varieties, Food Processing Industry, 1, 2009, 50-53, ( in Bulgarian).

9. N. Мichalkova, Ivanka Petrova, Tsvetan Tsvetkov, Galina Mihova, Contents of beta-glucans in Bulgarian barley varieties, Food Processing Industry, 12, 2008, 49-51, ( in Bulgarian).

10. Yao, N., J., L. Jannink, P.J. White, Molecular weight distribution of (1→3) (1→4) β-D-glucan affects pasting properties of flour from oat lines with high and typical amounts of β-glucan”, Cereal Chem., 84, 2007, 471-479.

11. T.P. Lyons, Beta-glucan measurement and control, Technical Quarterly MBAA, 15, 2, 1978, 102-105.

12. De Sá M., R.G. Palmer, Assessment of Enzymatic Endosperm Modification of Malting Barley using individual Grain Analyses, Journal of The Institute of Brewing, 110, 1, 2004, 43-50.

13. M. Lišková, H. Frančáková, J. Mareček, Beta-glucan degradation during post harvest maturation of malting barley with emphasis on malt quality, Potravinarstvo, 4, 3, 2010.

14. C. Bamforth, M. Kanauchi, A Simple Model for the Cell Wall of the Starchy Endosperm in Barley, Journal of The Institute of Brewing, 107, 4, 2001, 235-240.

15. C.Bamforth, Does “Solubilase” exist?, Brewers` Guardian, 131, 7, 2002, 26-29.

16. C.W. Bamforth, H.L. Martin, The degradation of β-glucan during malting and mashing - the role of β-glucanase, Journal of the Institute of Brewing, 89, 1983, 303-307.

17. T. Benitez, C. Limon, J. Delgado-Jarana, M. Rey, Glucanolytic and other enzymes and their genes, 1998.

18. Y.M. Galante, A. de Conti, R. Monteverdi, Application of Trichoderma enzymes in the food and feed industries, Trichoderma and Gliocladium, v. 2 Enzymes, biological control and commercial applications., 1998, 327-342.

19. C.W. Bamforth, β-Glucan and β-glucanases in


156

malting and brewing: practical aspects, Brewers Digest, 69, 1994,12-16.

20. P. Gavlova, Beta-glucan and its importance in brewing, Brno, Beer and life, 16, Brno, Research Institute of brewing and malting affairs, Malting Institute, 2002.

21. J.M. Wang, G.P. Zhang, J.X. Chen, F.B. Wu, The changes of β-glucan content and β-glucanase activity in barley before and after malting and their relationships to malt qualities, Food Chemistry, 86, 2004, 223- 228.

22. L.T. Lusk, G.R. Duncombe, S.B. Kay, A. Navarro, D. Ryder, Barley β-glucan and beer foam stability, Journal of the American Society of Brewing Chemists, 59, 2001, 183-186.

23. G. Marinova, V. Bachvarov, Comparative study of the effect of enzyme preparations with beta-glu-canase activity in the production of wort and beer, Scientific works of University of Food Technologies, Plovdiv, v. LV, 2008, part 1, 163-168, ( in Bulgarian).

24. M. Mandels, I. Hontz, J. Nystrom, Enzymatic hydrolysis of waste cellulose, Biotechnology and Bioengineering, 16, 11, 1974, 1471-1493.

25. N. Nelson, A photometric adaptation of the Somogyi method for the determination of glucose, J. Biol. Chem., 153, 1944, 375-380.

26. M. Somogyi, Notes on sugar determination, J. Biol. Chem., 195, 1952, 19-23.

27. Lubov Iotova, Ivan Dobrev, Ivan Ivanov, Practicum in biochemistry, part 1, Sofia, 2000, p. 132-133, (in Bulgarian).

28. ISO 4833: 2003- Microbiology of food and animal feeding stuffs. Horizontal method for the enumeration of microorganisms. Colony-count

technique at 30°C. International Organization for Standardization, Geneva.

29. ISO 21528-2: 2004- Microbiology of food and animal feeding stuffs- Horizontal methods for the detection and enumeration of Enterobacteriaceae- Part 2: Colony-count method. International Organization for Standardization, Geneva.

30. ISO 4832: 2006- Microbiology of food and animal feeding stuffs - Horizontal method for the enumeration of coliforms- Colony-count technique. International Organization for Standardization, Geneva.

31. ISO 6888-1: 1999- Microbiology of food and animal feeding stuffs- Horizontal method for the enumeration of coagulase-positive staphylococci (Staphylococcus aureus and other species)- Part 1: Technique using Baird-Parker agar medium. International Organization for Standardization, Geneva.

32. ISO 6579: 2002- Microbiology of food and animal feeding stuffs- Horizontal method for the detection of Salmonella spp. International Organization for Standardization, Geneva.

33. EBC Analysis Committee, Analytica EBC, 5th Ed., AG Verlag und Druckerei, Zurich, 1998.

34. J. Moštek, Analytické metody ke cvičení z kvasné chemie a technologie”, SNTL, Praha, 1973.

35. G. Marinova, V. Bachvarov, A. Krasteva, Spectro-photometric determination of soluble nitrogen of malt and unhopped wort, Food Processing Industry, 1, 2005, 9-12, (in Bulgarian).

36. K. Erdal, P. Gjersten, Proceedings of the 11-th European Brewery Convention Congress, Madrid, 1967, 295.

Izzat Eshmetov, Dilnoza Salihanova, Anvar Agzamhodjaev

157


EXAMINATION OF THE INFLUENCE OF THE GRINDING DEGREE AND STABILIZING AGENT ON THE RHEOLOGICAL PROPERTIES

OF AQUA-COAL FUEL SUSPENSIONS Izzat Eshmetov, Dilnoza Salihanova, Anvar Agzamhodjaev

Institute of General and Inorganic Chemistry, Academy of Science of Uzbekistan, 77 a, st. Mirzo-Ulugbek, 100170, Tashkent, UzbekistanE-mail: [email protected]

ABSTRACT

It has been established that 50 - 60 minutes of Angren and Shargun coal’s grinding are sufficient to obtain particles of size not less than 50 µm, the amount of which is more than 80 % of the total mass. A considerable increase of the structural viscosity value and the stability of the aqua-coal fuel suspensions (ACFS) are observed under the studied conditions. It has been shown that the optimal concentrations of plasticizing reagents such as NaOH or Са(ОНE)2 are 1,0 % and 0,75 % from the total mass of ACFS based on Angren lignite. There are 38 - 40 % and 43 - 49 % of solid phases, respectively, in the ACFS.

For Shargun black coal the same plasticizing agents with similar concentrations have been applied. As a result there is 52 - 54 % of solid phase in the total mass of ACFS. The results show that the use of NaOH or Са(ОН)2 as the plasti-cizing agent reduced the structural viscosity of lignite and black coal ACFS from 2,52 to 1,28 Pа s and from 2,52 to1,34 Pа s, respectively. The value of viscosity of ACFS has thus been reduced practically two times in the studied conditions.

Keywords: aqua-coal fuel suspensions, stability, fluidity, humidity, viscosity, stabilizing, agent, grinding degree.

Received 03 October 2014Accepted 12 February 2015

INTRODUCTION It is known that the stabilization of a suspension is

determined by surface forces occurring in the thin layers, which round the particles of dispersed phase [1]. The role of the stabilizing agents (soda ash, sodium pyroph-osphate, iron chloride and aluminium chloride, alkaline salts of organic acid and sulfate alkaline solution, etc.) is that when agents are adsorbed on the particles of the dispersed phase they form films around them which reduce surface tension, block immediate contact and hinder the process of the particles coagulation. In the past some resins and their fractions, different soups, etc. have been applied as stabilizers but currently synthetics with complex composition are used as surface active sub-stances (SAS). Huge numbers of substances have been researched for stabilization of ACFS [2 - 5]. For lignites from deposits the necessity for individual selection of

chemical agent appears along the task for reduction of rheological properties and achievement of the necessary stability in static and dynamical conditions. Usually, the plasticizing agents are introduced into the suspension as water solution, so the chemical composition of water affects significantly their efficiency.

EXPERIMENTAL

The Angren lignite grade of 2BSC (brown, slab, coarse), grade of 2BNFSC-1 and 2BNFSC -2 (brown nut, fine, seed, chip) and the Shargun black coal grade of 1SCCNF (soft coking coarse, nut, fine) of Uzbekistan were the objects of our studies [6 - 12]. The presence of oxygen containing carboxyl and hydroxyl (phenolic) groups and humic acids are typical for the Angren lignite. It is known that humic acid in coal has a protective ef-fect on clay and coal particles showing their subsidence.


158

Their content is about 5 % in the Angren coal. NаОН and Са(ОН)2 were taken as stabilizing agent for ACFS. The structural viscosity of ACFS dependence on concentra-tion of alkali has an extremal character with a minimum.

The coal samples were ground in a ball mill model BML (ball mill laboratorial) to different contents of solid phase in ACFS - coal particles of size not less than 50 µm. The grinding time was determined by the yield of fine fraction of coal (less than 50 µm). The grain size composition was determined with sieves according to GOST 2093-82. Rheological characteristics were evalu-ated by determining structural viscosity on a BCH-3

rotary viscometer. The mass fraction of solid phase was defined by drying at 1050С (GOST 11014-84).


The aqua-coal suspension is a micro heterogeneous system that consists of two or more phases and it de-pends on the quantity of the dispersion medium. As a result one of the phases forms a medium divided into particles of dispersed phases. Usually the interval of particles size can be changed from some nanometers to 100 µm and more. Fundamental physicochemical signs

Grinding duration min.

Residue, % Rheological characteristics R50 less than

50 µm R200 less than

200 µm μс, Pа s τ0, Pа Stability, day

The Angren lignite grade of 2BSC 10 56-58 96-97 1,87 29 1 20 68-70 98-99 1,89 28 2 30 78-80 ~ 99 2,03 24 5 40 83-85 ~ 99,3 2,52 15 20 50 86-88 ~ 99,7 2,58 14 30 60 90-92 ~ 99,9 2,58 14 40

The Angren conditioned commodity grade of 2BNFSC -1 10 52-54 96-97 1,88 28 1 20 66-68 97-98 1,90 26 2 30 77-78 98-98,5 1,92 26 4 40 80-82 ~ 99,25 2,03 24 17 50 84-86 ~ 99,5 2,2 20 25 60 88-90 ~ 99,8 2,3 18 35

The Angren conditioned commodity grade of 2BNFSC-2 10 47-48 95-96 1,86 29 1 20 62-64 97-98 1,90 26 2 30 73-77 98-99 1,94 25 3 40 75-77 ~ 99 2,0 23 15 50 80-82 ~ 99,5 2,1 22 20 60 84-85 ~ 99,7 2,1 22 30

The Shargun black coal grade of 1SCCNF 10 55-57 96-97 1,75 36 1 20 65-67 98-98,5 1,78 34 2 30 78-80 98-98,5 1,82 32 4 40 81-84 ~ 99,25 1,84 27 18 50 87-88 ~ 99,6 1,89 28 26 60 89-91 ~ 99,9 1,90 26 36

*The content of solid phase of ACFS for Angren lignite grade of 2 BSC is 38,8 - 40,0 %, the grade of 2 BNFSC-1 is 43,6 - 45,0 %, the grade of 2 BNFSC-2 is 47,8 - 49,2 %, and for Shargun black coal grade of 1 SCCNF is 51,8 - 53,6 % from total mass, respectively.

Table 1. Influence of the grinding duration on the grinding degree and properties of ACFS* obtained with the Angren lignite (grade 2BSC, 2BNFSC -1, 2BNFSC-2) and the Shargun black coal (grade 1SCCNF ).


159

such as heterogeneity, a notable presence of a separating surface between the phases and dispersion (atomism) are general for all dispersed system. The role of the factors becomes more considerable when the particles size and their content in liquid medium are decreased in order to demonstrate the aggregative and sedimentation stability of fine system.

Table 1 shows the results for the grinding duration on the grinding degree and properties (structural viscosity, dynamic stress of shift and stability) of ACFS. It can be seen from Table 1, that with the increase of the grinding duration the yield of coal particles of fraction size less than 50 µm (R50) rises.

In this connection the structural viscosity of ACFS and its stability are increased. It is established that 50 - 60 minutes of grinding are sufficient for obtaining ACFS with an 80 % content of coal fraction less than 50 µm (R50). The plasticizing agents are added to elongate storage term of ACFS by increasing of solid phase in the suspension.

The plasticizing agents (lignosulphonates, CAR (coal alkali reagents), NaOH, Са(ОН)2, etc.) contribute mostly alkaline medium in the ACFS.

Increase of alkali content to 2 % leads to exfoliation of the flurry with formation of a hard residue. In the literature [3] for this fact the following explanation was given. The water-alkali reacts with lignite in the first stage and converts humate sodium into a water phase in the form of sodium salts:

NaOH + (HO – Humic)coal → (NaO – Humic)water + H2O

where: (HO – Humic)coal - humic acid in the coal, (NaO – Humic)water - dissolving of sodium humate.

The humate sodium solution is found in water solu-tion. In addition, this composition has water phase of ACFS in a large range of alkali concentration (more than 2.0 %) providing full transformation of humic acid into respective salts. The first stages of its interaction with coal with the low concentration of alkali can be described as equalization. Then the stage of humate hydrolysis and migration of sodium ions into the solid phase of coal followed:

(NaO – Humin)water + (HO – Humin) coal → → (HO – Humin)gel + (NaO – Humin)coal

It is known that humic acid and its sodium humate inserted as CAR reduce the rheological character of ACFS [13 - 16]. The sodium humate and the calcium humate are active substances, having plasticizing activ-ity. In this study, the possibility to use a CAR analog - alkaline agents of NaOH and Са(ОН)2 for ACFS has been studied. The resulting sodium humate and calcium humate have been formed directly at their adding into the ball mill and in the coal grinding process that allowed to simplify the process of introduction and batching of the agents. The process occurred exactly at little concentra-tions of alkali and leads to coagulation of humic acid, the passage of the true solution of the sodium humate into a colloid solution - a gel of humic acid, which stabilizes the aqua-coal fuel suspensions and reduce their viscosities.

Fig. 1. Dependence of the structural viscosity of ACFS obtained on the basis Angren lignite (grade of 2 BSC), from the amount of inserted sodium hydroxide (а) and calcium hydroxide (b).

Stru

ctur

al v

isco

sity

, Pа

* s 2,5

1,6 0,8 0,4 0

1,0

1,5

2,0

1,2 2,0 Amount of NаОН, %

2,5

1,6 0,8 0,4 0

1,0

1,5

2,0

1,2 Amount of Са(ОН)2, %

Stru

ctur

al v

isco

sity

, Pа*

s


160

Fig. 1 shows that the dependence of the structural viscosity of ACFS on concentration of inserted alkali has an extreme character. Results showed that the optimum amount of alkali plasticizing agents - NaOH was 1,0 % for obtaining an ACFS with Angren lignite, containing 38 - 40 % of solid phase in the suspension, as well as, a certain amount of Са(ОН)2 agent - 0,75 % including 43 - 49 % of solid phase in the suspension.

The application efficiency of alkali agents was

studied in the ACFS, obtained from the Angren lignite and the Shargun black coal (Table 2). The data on the influence of the stabilizing agents NaOH and Са(ОН)2 on the rheological characteristics of ACFS which were obtained from local types of coal are given in Table 2. They show that the application of NaOH or Са(ОН)2

as plasticizing agents for lignite and black coal reduces the structural viscosity of lignite and black coal’s ACFS from 2,52 - 1,28 Pа s tо 2,52 - 1,34 Pа s, respectively.

№

Coal samples

Content of solid

phase in the ACFS,

%

Chemical agent

Amount of agent

%

Rheological

characteristics

μс, Pа∙s

τ0, Pа

1 Angren lignite

grade of 2BSC 36 without

addition - 2,52 15

2 Angren lignite

grade of 2BNFSC -1 42 without

addition - 2,2 20

3 Angren lignite grade of 2BNFSC -2 45 without

addition - 2,1 22

4 Shargun black coal of 1SCCNF grade 52 without

addition - 1,84 30

5 Angren lignite of 2BSC grade 36 NaOH 1,0 1,28 42

6 Angren lignite of 2BNFSC -1 grade 42 NaOH 1,0 1,16 46

7 Angren lignite grade of 2BNFSC-2 45 NaOH 1,0 1,1 52

8 Shargun black coal grade of 1SCCNF 52 NaOH 1,0 1,06 56

9 Angren lignite grade of 2BSC 36 Са(ОН)2 0,75 1,34 38

10 Angren lignite grade of 2BNFSC -1 42 Са(ОН)2 0,75 1,22 44

11 Angren lignite grade of 2BNFSC -2 45 Са(ОН)2 0,75 1,18 45

12 Shargun black coal grade of 1SCCNF 52 Са(ОН)2 0,75 1,15 46

Table 2. Influence of the plasticizing agents - NaOH and Са(ОН)2 on the rheological properties and calorific value of ACFS, obtained from Angren lignite and Shargun black coal.


161

Notably the value of ACFS viscosity decreases practi-cally two times.

The initial ash of coal did not have a considerable influence on the reduction of the structural viscosity, which allowed to use the alkaline agents for the low-ash lignite and black coal. It should be noted that currently the use of an inexpensive and cost- effective agent with which to obtain a stable ACFS, is quite actual. That is why, usage of plasticizing agents as inexpensive and available as Са(ОН)2 for ACFS can be perspective and economically beneficial factor (the cost of Са(ОН)2 is 50 - 60 times lower than the cost of NaOH, etc.).

Thus, it is established that for production of lignite and black coal of ACFS the application of calcium hydroxide - Са(ОН)2 as plasticizing agent, simplifies and reduces the price of the production process with simultaneous increase of the slurry fuel stability and calorific value.

CONCLUSIONS

It has been established that 50 - 60 minutes of Uzbekistan Angren and Shargun coal grinding is suf-ficient to obtain particles of a size not less than 50 µm, which assemble more than 80 % from the total mass. A considerable increase of the structural viscosity and the stability of the aqua-coal fuel suspensions (ACFS) has been observed in these conditions.

It has been shown that in order to obtain ACFS on the basis of Angren and the Shargun coal the optimal concentration of plasticizing agents - NaOH or Са(ОН)2 should 1,0 % and 0,75 % from the total mass, with the content of solid phase 38 - 40 % and 43 - 49 %, respec-tively. Similar plasticizers with homothetic concentration were applied during the stabilization of Shargun black coal with amount of solid phase 52 - 54 % from the total mass of ACFS. The results testify that the application of NaOH or Са(ОН)2 as plasticizing agents for lignite and black coal, reduce the structural viscosity of lignite and black coal ACFS from 2,52 - 1,28 Pа s tо 2,52 - 1,34 Pа s, respectively. The value of ACFS viscosity has reduced practically two times.

REFERENCES

1. V.I. Murko, Influence of plasticizing reagents on rheological properties of water-coal fuel, Chemistry

of solid fuel, 2, 2001, 62-72, (in Russian). 2. G.S. Hodakov, Some singularity of preparation of

high-concentration water-coal slurry meant for hy-drotransportation, Collected articles of studies of VNIIPI hydropipeline, Мoscow, Russia, 1985, p.24-31. (in Russian).

3. М.P. Baranova, Obtainment and application of aqua-coal fuel suspensions from different degree meta-morphism of coal, PhD thesis, Krasnoyarsk, 2011, (in Russian).

4. Y.K. Seong, Rheology of brown coal - water suspen-sions, Rheological Act, 26, 3, 1987, 291-300.

5. Usui H. Ymasaki, Effect of temperature on the fluidity of a coal water mixture, Technology Reports of the Ymaduchi University, 3, 5, 1986, 393-398.

6. I.D. Eshmetov, Some factors influencing on rheo-logical properties of aqua-coal fuel suspensions of Uzbekistan, J. European Applied Sciences, 9, 2013, 102-107.

7. I.D. Eshmetov, А.А. Agzamkhodzhaev, Examination of viscosity of water-coal slurry fuel obtained on the basis of conditioned commodity Angren coal of Uzbekistan Republic, Journal of Chemical Industry, 91, 3, 2014, 118-122, (in Russian).

8. I.D. Eshmetov, D.S. Salihanova, А.А. Agzamk-hodzhaev, Obtaining of water-coal slurry fuel on the basis of conditioned commodity Angren coal by means their modification with mazut, Journal of Chemical Industry, 91, 3, 2014, 126-130, (in Russian).

9. I.D. Eshmetov, R.H. Gumarov, А.А. Agzamk-hodzhaev, Influence of calcium hydroxide on rheo-logical properties of water-coal slurry fuel obtained on the basis of Uzbekistan’s coal, Uzbek Chemical Journal, 1, 2014, 40-45, (in Russian).

10. I.D. Eshmetov, А.А. Agzamkhodzhaev, Water-coal slur-ry fuel obtained on the basis of Angren lignite, Journal Compositional materials, 1, 2014, 9-13. (in Russian).

11. I.D. Eshmetov, А.А. Agzamkhodzhaev, Z Is-kandarov, Examination of influence of calcium hydroxide on rheological properties of water-coal slurry fuel, Journal Chemical Technology, Control and management, 2, 2014, 42-46, (in Russian).

12. I.D. Eshmetov, А.А. Agzamkhodzhaev, Examination of water-coal slurry fuel viscosity obtained on the basis of conditioned commodity Angren coal, Uzbek Chemical Journal, 3, 2014, 31-36, (in Russian).

13. V.E. Gubin, V.V. Gubin, Piping transport of oil and


162

oil products, 1982, p. 296, (in Russian). 14. B.E. Dodge, A.B. Metzner, Turbulent Flow of non-

Newtonian Systems, AICHE Journal, 15, 12, 1959, 189-204, (in Russian).

15. M.P. Baranova, Theoretical and engineering founda-tion of obtaining of water-coal slurry fuel from low-metamorphosed coal, Energetics in global world, Collected reports of First International scientific and

technical congress, Krasnoyarsk, Russia, 12, 2010, 361-368, (in Russian).

16. V.A. Kulagin, M.P. Baranova, The technology of preparation and hydro transportation of water-coal slurry fuel for burning in industrial power-generat-ing boilers, Bulletin of KSTU graduating students association, Edition 12, Krasnoyarsk, Russia, 2006, 55-60, (in Russian).

Umarbek K. Alimov, Shafoat S. Namazov, Ahmed R. Reymov

163


SPECIFICITIES OF THE RECIRCULATING METHOD FOR CENTRAL KYZYL KUM PHOSPHORITES PROCESSING TO QUALIFIED PHOSPHORIC FERTILIZERS


Institute of General and Inorganic Chemistry, Academy of Sciences of Uzbekistan,77a, Mirzo-Ulugbek st., 100170, Tashkent, UzbekistanE-mail: [email protected].

ABSTRACT

With the purpose of obtaining concentrated grades of phosphoric fertilizers, the decomposition of phosphorus concen-trate from phosphorites of the Central Kyzyl Kum with evaporated wet processing with phosphoric acid by the recirculat-ing method in liquid phase, has been investigated. The results of the research have shown, that the expansion coefficient of the phosphorus concentrate is 83,34 - 90,29 %. In addition the filtration rate by liquid phase has been found. The degree of distribution of the phosphate constituent parts components has been given. The process of neutralization of the acid monocalcium phosphate, obtained from phosphorus concentrate with a 25 % water solution of ammonia (NH3) and limestone (СаСО3), has been studied. Products, containing 53,01 - 55,36 and 50,51 - 51,19 % Р2О5 with 4,13 and 1,92 МPа static strength, respectively, have been obtained. The latter can be recommended as a double superphosphate by the State Standard grade A.

Keywords: phosphorus concentrate, evaporated wet processing, expansion coefficient, filtration rate.

Received 01 September 2014Accepted 05 February 2015

INTRODUCTION

In the world’s practice some traditional technolo-gies based on acid treatment (nitric acid, sulfuric acid, phosphoric acid and hydrochloric acid) of phosphate raw materials to phosphorus containing fertilizers such as nitrophosphate, ammophos, simple superphosphate, double (triple) superphosphate and precipitate exist. Traditional acid decomposition is conducted by energy-,material- and capital intensive efforts with foundations of large tonnage - phosphogypsum, calcium chloride, and calcium nitrate wastes. Moreover, in the presented technologies there are significant and strict requirements for quality of the phosphate raw material. Their values have to be controlled for: content of Р2О5 from 24,5 to 39,4 %; СаО from 39,5 to 52,5 %; MgO from 0,1 to 3,5 %; total Fe2O3 and Al2O3 have to be from 1,5 to 3,3 %; СО2 from 1,3 to 1,8 %, residue should be 10 - 20 % on mesh sieve № 0071. Conditioned raw phosphates of Khibini, Florida, Moroccan, Syrian deposits meet the requirements, and are suitable for acid processing.

As opposed to them, the phosphorites of Central Kyzyl Kum are poor raw materials regarding content of Р2О5 (16 - 17 %), carbonate impurities (21 - 41 %), high values of calcium modulus (CaO:P2O5 > 2), chlorine (0,1 %) and organic impurities. In addition this raw material is characterized by thin intergrowth of phosphate minerals with calcite [1]. Therefore, attempts to enrich them with flotation have not lead to positive results [2, 3].

The most efficient method of enrichment can be selective removal of calcareous impurities out of the phosphorites by the nitric acid method [4] or thermal burning for CO2 removal [5]. The latter is implemented in the Kyzyl Kum phosphorite plant where ore reduc-tion, dry separation with obtaining phosphorite flour, deslimation and washing of raw chlorine, burning for the decomposition of carbonates and CO2 removal, are used.

As a result of this process, washed and burned con-centrate, that contains (mass %): 27.26 P2O5, 53.36 СаО, 1.3 AI2О3, 0.51 Fe2О3, 0.6 MgO, 1.91 F, 2.41 СО2, 0.04 CI, СаО:P2O5 = 1.96, is obtained. Since 2006, the Kyzyl Kum phosphorite plant has been producing annually up


164

to 400 thousand tons of washed and burned phosphorus concentrate. This concentrate is only supplied the Open Joint Stock company (OJSC) “Ammofos-Maxam” for production of ammonium phosphate (10 % N and 46 % P2O5, suprefos (8 - 15 % N and 20 - 24 % P2O5) and ammonium sulfate phosphate (15 - 19 % N and 4 - 23 % P2O5) [1]. These are products belonging to complexed fertilizers. They have to be applied before planting, with planting and into dressing. Phosphoric fertilizers are most efficient when they are applied with the autumn plough [6]. Simple superphosphate and double super-phosphate are ordinary phosphoric fertilizers. They are good quality phosphorous containing fertilizers, suit-able for application in any soil and under the all crops. However, simple superphosphate with content 16 - 19 % Р2О5 and high amount of gypsum (more than 45 - 50 % CaSO4) is limited for transportation to far distance. Double superphosphate has ballast of less concentrated phosphoric fertilizer in contents of 45 - 56 % Р2О5 and possesses agrochemical activity [7].

Studies of double superphosphate produced on the basis of phosphorites from Central Kyzyl Kum were reported [8, 9].

The deficiencies found in the mentioned studies are low-degree of efficiency of the phosphate raw material, application of high-priced ammonia and low content of Р2О5 in the product.

In our view one of the rational ways is decomposi-tion of phosphate raw material with evaporated wet processing phosphoric acid on recirculation in liquid phase. The present method includes phosphate raw material decomposition with a redundant norm of evaporated wet processing phosphoric acid, necessary for formation of monocalcium phosphate, a saturated solution of monocalcium phosphate, separation of in-soluble residues, monocalcium phosphate crystallization and separation from the mother solution, monocalcium phosphate neutralization, sulfuric acid extraction of mother solution, separation of gypsum and return of the phosphoric acid to decompose the next phosphate raw material. The main advantage is the recirculating processing of the phosphate occurrs in one cycle, which does not require the camera ripening and long storage after ripening for production of the traditional double superphosphate [1, 7].

There are data on processing of different kinds of phosphate raw material into phosphorus- containing

fertilizers with nontraditional cycle methods [10 - 12]. A possibility for double superphosphate production on the basis of decomposition of Apatite and Kingisepp phosphorus concentrate with evaporated wet processing phosphoric acid, containing 40 - 45 and 55 - 65 % Р2О5

has been shown. The norm of the phosphoric acid was varied within the ranges 400 - 600 % from a stoichiom-etry of monocalcium phosphate and at temperatures of 60 and 110 - 130°С for 1 - 2 hours. The crystallization process was carried out at temperature of 40 - 60°С for 90 minutes. After neutralization of the acid monocalcium phosphate with ammonia and limestone, granulated double superphosphate was obtained, containing (mass %): 51 - 54 Р2О5total; 50 - 52,5 Р2О5 acceptable; 39 - 50,7 Р2О5 water-soluble ; 2,0 - 5,9 N; 13 - 14 СаО and 55,5 - 57,55 Р2О5total, 55 - 56,3 Р2О5 acceptable, 49,2 - 52,7; Р2О5water-soluble. 2,5 Р2О5free, respectively. Their hygroscopicity was at level with the standard samples of superphosphate.

Phosphate ore of Chili say deposit (17,7 % Р2О5) was calcited apriory at 600 - 650°С for prevention of foaming and improvement of filtration rate [13]. The cycle method processing of this raw material was conducted with application of a 450 - 550 % norm of phosphoric acid (stoichiometric norm to monocalcium phosphate) concentration 40 - 41 % Р2О5. It was revealed that after 7 cycles process stabilization of the composi-tion of the circulating phosphoric acid was established. As a result a double superphosphate, containing 50,4 - 51,0 % Р2О5total; 46,7 - 47,0 % Р2О5 water-soluble; 0,75 - 1,2 % Р2О5free, 2,3 - 2,9 Н2О, that corresponds to the require-ment of State Standard 16306-80 for double (triple) superphosphate grade A, was obtained.

Research on the decomposition of unconditioned Karatau phosphorites such as deposits of Kokjon and Koksu showed [14], that the degree of decomposition of the raw material reached 99 - 99,5 %. However due to content of insoluble residue (SiO2, etc.) in the raw material, the latter was removed off the technologic process. The basic technological parameters were the following: norm of phosphoric acid to Ca(H2PO4)2 stoi-chiometry 450 - 500 %; concentration of acid - 40% Р2О5; decomposition temperature - 95°С; crystallization temperature – 40°С.

Currently there are no reports concerning the granular phosphorites of Central Kyzyl Kum by a cyclic method in order to treat them to phosphoric fertilizers. That is why this is the aim of the present research.


165

In this study results of laboratory research on the decomposition of washed and burned phosphorus concentrate from phosphorites of Central Kyzyl Kum, with redundant norm of evaporated wet processing phosphoric acid (stoichiometric norm to monocalcium phosphate), with subsequent separation into solid and liquid phase by filtration, have been given. The liquid phase has been cooled and removed to obtain commodity a monocalcium phosphate, which will neutralize with ammonia and limestone.

EXPERIMENTAL

Materials Washed and burned phosphorus concentrate (WBP),

containing (mass %): 25,77 Р2О5 total; 3,43 Р2О5 acceptable by 2 % of citric acid; 2,94 Р2О5 acceptable by 0,2 М of ethyl-ene diamine tetraacetate (EDTA); 52,70 СаОtotal; 17,34 СаО acceptable by 2 % of citric acid; 1,20 MgO; 0,63 Fe2O3; 1,15 Al2O3; 2,67 SO3; 3,43 CO2; 6,88 insoluble residue and evaporated wet processing phosphoric acid (WPA) with composition (mass %): 41,20 Р2О5; 0,20 СаО; 0,98 MgO; 0,79 Fe2O3; 1,76 Al2O3; 3,33 SO3 total 3,04 SO3 free

were studied.

Formulas for calculationsThe computational norm of WPA to stoichiometric

monocalcium phosphate was ranged within norms of 300, 350, 400 % calculated by:

WPA

2 5 2 5

3 3

0,3944 ( ) 100 0,3944 ( )[ ].

( ) 0,7 ( ) 0,7 ( )

.100%

in WPA WBP in WBP

WBP in WBP WBP total in WBP free in WPA

NP O m P O

m CaO m SO SOω ω

ω ω ω

=

⋅ ⋅ − ⋅ ⋅=

⋅ − ⋅ ⋅ − ⋅ (1)where 0,3944 is ratio of molar mass of СаО:Р2О5 in Са(Н2РО4)2; )( 52 WPAinOPω is the mass fraction of Р2О5

in wet processing phosphoric acid; WBPm is mass of the initial phosphorite; 2 5( )in WBPP Oω is the mass fraction of Р2О5 in initial phosphorite; )( WBPinCaOω

is the mass

fraction of СаО in initial phosphorite; 0,7 is ratio of molar mass of gypsum СаО:SО3 in initial phosphorite and wet processing phosphoric acid; 3( )total in WBPSOωis the mass fraction of total SO3 in initial phosphorite;

3( )freel in WPASOω is the mass fraction of free SO3 in wet

processing phosphoric acid.

The filtration rate was calculated according to: 2, /liquid phasem

kg m hourt S

ϑ = ⋅⋅ (2)

where, phase. liquidm - mass of liquid phase, кg; t - time from moment filling of funnel to flow of liquid phase off surface of solid phase, h; S - area of funnel surface ( 2rS ⋅= π ), m2.

The results for the expansion coefficient were cal-culated (Kexp) by the formula:

%100)(

100.22

1lub11exp

52

5252 ⋅⋅

⋅−−= −

WBPinitialintotalOP

cakefilterdriedinlesowaterOPtotalOP

mm

Kω

ωω

(3)where

totalOP 521ω – lesowaterOP lub521 −

ω - the mass fraction of

total and water-soluble form Р2О5 in dried filter cake; 1 in dried filtr cakem - mass of dried filter cake, g;

totalOP 522ω- the mass fraction of Р2О5 in initial phosphorite, g;

2 .in initial WBPm - mass of initial phosphorite, g.The degree of distribution of the components in the

acid undecomposed phosphate cake, acid monocalcium phosphate (MCP) and fugate (mother solution) were determined, according to:

%100))((

))((

22

..,.,.11 ⋅⋅

⋅=

total

solmothMCPcareundecin

commcomm

Rω

ω

(4)

where 1m - mass of acid undecomposed phosphate cake, acid monocalcium phosphate and fugate (mother solution), respectively, g; 2m - total mass of pulp, g;

1( )comω - mass fractions of Р2О5, СаО, MgO, Fe2O3, Al2O3 and SO3 in the acid undecomposed phosphate cake, acid monocalcium phosphate and fugate (mother solution), respectively; )(2 comω - total mass fraction of Р2О5, СаО, MgO, Fe2O3, Al2O3 and SO3, incoming with WPA and phosphorite.

The process neutralization of the acid monocalcium phosphate with limestone and ammonia was performed by the equations:

CaCO3 + 2H3PO4→Са(Н2РО4)2 + СО2 + H2O (5)

3NH3 + 2H3PO4→NH4H2PO4 + (NH4)2HPO4 (6)

EXPERIMENTAL PROCEDURE

Laboratory experiments were carried out in a temperature-controlled reactor (500 ml) equipped with a stirrer, with controlled speed. Temperature was kept at 95°С by the thermoregulator in the thermostat. For conduction of the experiment, a weighed amount of acid with reactor, were placed in the thermostat. On obtaining


166

of the predetermined temperature, we started to measure phosphate raw material at portions for 10 -15 min. After the feeding was finished, the time of the decomposition process was 60 min. Further, the obtained pulp was divided into solid and liquid phases by filtration under vacuum of 500 mm of mercury in a Buchner funnel. Determination of the degree of decomposition of the phosphorus concentrate was investigated simultane-ously in the process of filtration at the mentioned norm of evaporated wet processing phosphoric acid (300, 350 and 400 %). During the filtration, the acid solid residue was washed several times carefully with distilled hot water to remove any soluble substances. The washed and wet filter cake was dried in an electric oven at 90 - 95°С until reaching a constant mass. The dried filter cake was subjected to chemical analysis by conventional procedures [16, 17] for determination of total and water-soluble Р2О5.

In the separation process of pulp into solid and liquid phases, the filtrate was cooled within 2 hours at 40°С. As a result, the acid monocalcium phosphate crystallized. The monocalcium phosphate crystallization occurred as described in [15]. It should be noted, that the crystal-lization time with increasing the norm of the acid from 300 to 400 %, increased. For example, if crystals of monocalcium phosphate at 300 % were formed within 2 hours, the crystals of monocalcium phosphate at 350 % formed within 3 hours. Moreover, at norm of 400 % a seeding agent, which was finished product with size ≤1 mm, in quantity of 1 - 1,5 % from total mass of filtrate was added in order to speed up crystallization.

The described observation in all likelihood is ex-plained by unsaturation of the liquid phase (filtrate) for calcium ions (CaO+2). The obtained monocalcium phosphate product was separated from the filtrate with a laboratorial centrifuge CLS-5 (made in Russia). The rotating velocity was 3000 rpm. The centrifugation time was 15 min. Acid undecomposed phosphate cake, acid monocalcium phosphate (with some free form of Р2О5) and fugate (mother solution) were obtained dur-ing the research. These were named symbolically as phosphate constituent parts. Chemical analysis of the acid undecomposed phosphate cake, acid monocalcium phosphate and fugate (mother solution) was performed by conventional procedures [16-17]. Also a degree of distribution was defined in connection with the content of components in them.

The neutralization of the acid monocalcium phos-phate with 25 % water solution of ammonia (NH3) and limestone (СаСО3) was carried out in laboratory con-ditions. The composition of the limestone is (mass %): 54,45 СаО; 1,62 MgO; 0,24 Fe2O3; 0,30 Al2O3; 0,86 SO3; 43,83 СО2; 0,89 insoluble residue.

The reaction mixture at the process neutralization of the acid monocalcium phosphate with 25 % water solution of ammonia in chinaware, was heated up to 90 - 95°С. The ammoniated pulp had movability, which helped to exercise the processes of granulation and dry-ing without specific difficulties. A СО2 emission was observed when the acid monocalcium phosphate reacted with limestone, that suggest interaction of the limestone with free phosphoric acid.

Methods for analysis The obtained mixture of the, reaction was mois-

tened with water, and then it was granulated and dried in chinaware at 90 - 95°С.

Dried samples were analyzed for the content of various form of phosphorus, nitrogen and calcium by the respective procedures [16, 17]. The assimilated form of Р2О5 was determined with 2 % citric acid and 0,2 М of EDTA by the colorimetric method. The assimilated form of CaО was determined only with 2 % citric acid. Nitrogen content was obtained by sublimation (Kjeldal). The рН of the products was assessed in 10 % water suspension. In addition, the strength of the granules of 2 - 3 mm in diameter obtained as products - phosphorus-containing fertilizers, were determined using the special device МIP-10-1 according to Standard [18].


The laboratory results from the chemical analyses of the acid undecomposed phosphate cake, the acid mono-calcium phosphate and the fugate (mother solution) are shown in Table 1.

The results of the research show that at norms of 300, 350 and 400 % from stoichiometry for formation of monocalcium phosphate, the content of the total form of Р2О5 is 35,06, 32,45 and 31,78 %, in the acid undecomposed phosphate cakes, respectively. There are water-soluble and free forms of Р2О5 containing 33,36; 30,32; 28,06 and 25,14; 22,17; 19,90 %, respectively, at norms of 300; 350 and 400 %. The contents of SO3, СаО,


167

MgO, Fe2O3 and Al2O3 are changed from 8,41 tо 11,34 %; from 12,43 tо 14,04 %; from 0,99 tо 1,05 %; from 0,82 tо 0,98 % and from 1,51 tо 1,81 %, respectively, in the cake. All the same, the chemical composition of acid monocalcium phosphate is from 44,97 tо 45,79 %; from

6,74 tо 5,84 %; from 1,46 tо 0,77 %; from 0,88 tо 0,72 %; from 1,65 tо 1,23 %, for Р2О5, СаО, MgO, Fe2O3 and Al2O3 at norms of 300, 350 and 400 %, respectively. Be-sides, a considerable amount of 21,21 - 22,81 % Р2О5free is contained in the acid monocalcium phosphate. The

Table 1. Chemical composition of the phosphate constituent parts, obtained by basis decomposition of the phos-phorus concentrate from phosphorites of Central Kyzyl Kum with phosphoric acid.

Name of phosphate constituent parts

Chemical composition, % Р2О5total. Р2О5 water-

soluble Р2О5free. СаО MgO Fe2O3 Al2O3 SO3

Norm of H3PO4 300 % for Ca(H2PO4)2

Acid undecomposed phosphate cake

35,06 33,36 25,14 12,53 0,95 0,79 1,55 7,66 Norm of H3PO4 350 % for Ca(H2PO4)2



Acid monocalcium- phosphate




Mother solution

40,72 - - 2,10 0,80 0,66 1,84 0,012 Norm of H3PO4 350 % for Ca(H2PO4)2

42,46 - - 2,02 0,98 0,68 1,87 0,008 Norm of H3PO4 400 % for Ca(H2PO4)2

42,33 - - 1,97 1,11 0,71 1,89 0,01

Table 2. Degree of distribution of the chemical components.

Name of the phosphate constituent

Chemical composition, % Р2О5 СаО MgO Fe2O3 Al2O3 SO3

Norm of H3PO4 300 % for Ca(H2PO4)2

Acid undecomposed phosphate cake

37,34 69,86 39,55 42,70 38,66 98,41 Norm of H3PO4 350 % for Ca(H2PO4)2



Acid monocalcium- phosphate




Mother solution



54,10 16,09 55,76 45,98 56,02 0,17


168

mother solution has also total forms of Р2О5 and СаО in its composition in the range of 40,72 - 42,33 % and 1,97 - 2,10 %, respectively.

The degree of distribution was calculated in con-sideration of the content of the components in the acid undecomposed phosphate cake, acid monocalcium phosphate and mother solution. The results are shown in Table 2. Table 2 shows that the data for the degree of distribution change, depending on the norm of wet processing phosphoric acid - 300, 350 and 400 % from stoichiometry. So, with increasing the norm of phos-phoric acid, from 37,34 tо 23,15 % Р2О5; from 69,86 tо 65,37 % СаО; from 39,55 tо 29,20 % MgO; from 42,70 tо 35,63 % Fe2O3; from 38,66 tо 29,84 % Al2O3 pass into the acid undecomposed phosphate cake and from 36,92 tо 54,10 % Р2О5; from 9,93 tо 16,09 % СаО; from 28,25 tо 55,76 % MgO; from 30,33 tо 45,98 % Fe2O3; from 39,18 tо 56,02 % Al2O3 pass into the mother solution. The rest goes into the acid monocalcium phosphate.

As shown in Table 2, the whole amount of SO3 is

precipitated mainly in the acid undecomposed phosphate cake in the form of gypsum - CaSO4. The acid unde-composed phosphate cake and the mother solution are

subjected to sulfuric acid processing. Influence norm of phosphoric acid on degree of de-

composition of phosphate raw and filtration rate of liquid phase are studied. The results are shown in Table 3.

The data from the Table show that there is a depend-ence between the filtration rate and the expansion coef-ficient. This is explained by the fact that with increasing of the amount of liquid phase some active hydrogen goes, respectively, in the pulp.

Reasonably, the low degree of decomposition of the raw phosphate is explained by neutralization of the phosphoric acid of Central Kyzyl Kum phosphorite by impurities such as MgO, Fe2O3, Al2O3 , etc. The products of the neutralization of the acid monocalcium phosphate are shown in Table 4. As the data in the Table show, the products obtained by neutralization of the acid monoc-alcium phosphate by 25 % water solution of ammonia (NH3), present themself as concentrated phosphorus containing fertilizers, with a considerable amount of nitrogen. The basic components are varied from 53,01 tо 55,36 % Р2О5total, from 52,29 tо 54,52 % Р2О5acceptale

by citric acid, from 51,80 tо 53,87 % Р2О5acceptable by EDTA, from 44,73 tо 47,07 % Р2О5water-soluble, as well as

Table 3. Influence of the norm of phosphoric acid on the degree of decomposition of the raw phosphate and the filtration rate of the liquid phase.

Norm of WPA from

stoichiometry, %

Washed and dried filter

cake, %

Expansion coefficient,

%

filtration rate, кg/m2·h, by liquid phase

Р2О5total. Р2О5water-

soluble. Кexp ϑ

300 9,19 2,53 83,34 279,73 350 6,85 2,02 86,81 320,59 400 4,39 1,04 90,29 402,55

Table 4. Chemical composition of the concentrated phosphoric fertilizers, obtained by phosphoric acid processing of the phosphorus concentrate from phosphorites of Central Kyzyl Kum.

Norm of WPA from

stoichiometry, %

рН of 10 % suspen-sion

Chemical composition, mass %

Р2О5total. Р2О5accep.

by 2 % citric acid

Р2О5accep. by

0,2М EDTA

Р2О5water-

soluble. СаОtotal.

СаОaccep by 2 % citric acid.

СаОwater-

soluble. N

Acid Ca(H2PO4)2, neutralized with 25 % water solution of NH3 300 5,23 53,01 52,29 51,80 44,73 8,76 8,57 1,72 9,51 350 5,13 54,43 53,65 53,15 45,81 7,48 7,29 1,50 9,78 400 4,87 55,36 54,52 53,87 47,07 7,21 7,04 1,43 9,89

Acid Ca(H2PO4)2, neutralized with CaCO3 300 2,79 51,19 49,79 48,67 43,43 21,37 20,14 15,63 - 350 2,75 50,62 49,12 48,14 42,29 20,28 18,46 14,68 - 400 3,04 50,51 49,28 48,15 40,65 25,63 24,35 18,14 -


169

from 9,51 to 9,89 % nitrogen. This type of products, with high content of two components, can be named nitrogen-phosphorus fertilizers.

The products obtained by neutralization of acid monocalcium phosphate by limestone (mass %): 50,51 - 51,19 Р2О5total; 49,28 - 49,79 Р2О5accebtable by citric acid; 48,15 - 48,67 Р2О5acceptable by EDTA; 40,65 - 43,43 Р2О5water-soluble; where ratio of Р2О5acceptable by citric acid : Р2О5total = 97,26 - 97,56 %; Р2О5acceptable by EDTA : Р2О5total = 95,08 - 95,33 %; Р2О5water-soluble: Р2О5total = 80,48 - 84,84 %. There are water forms of P2O5 and CaO in the products. This indicates the presence of the product monocalcium phosphate, which is a basic product of double superphosphate. The difference between the total and the water form of P2O5 in the product, gives approximately the content of assimilated form of phos-phate - dicalcium phosphate. The difference between of the acceptable and water form of P2O5 in the product, is indicative of the content of tribasic calcium phosphate.

Therefore, this type of products can be applied as ordi-nary phosphoric fertilizers, that completely correspond to the requirements of State Standard for double (triple) superphosphate grade A [18].

With the purpose to investigate the mechanical prop-erties of the obtained products, we studied their static strength. The results are presented in Table 5.

A rise of strength scatter (3,48 - 4,13 МPа) in the products, obtained by neutralization of monocalcium phosphate with ammonia, in comparison to granulated fertilizer obtained with limestone (1,92 - 2,23 МPа), is observed. This is explained by the presence of Fe+3 and Al+3 in the evaporated wet processing phosphoric acid, which form a strong complex compound at the neutrali-zation with ammonia [19].

Therefore, the offered approach includes both processing with sulfuric acid for extraction of the undecomposed phosphate cake and mother solution, and neutralization of the acid monocalcium phosphate,

Table 5. Static strength of phosphorus containing fertilizers.

Norm of WPA from

stoichiometry, %

Р2О5total, %

Ntotal, %

Static strength of granules with size 2-3 mm

кg/granules кg/сm2 МPа

Acid Ca(H2PO4)2, neutralized with 25 % aqueous solution of NH3 300 53,01 9,51 1,76 35,48 3,48 350 54,43 9,78 1,95 39,31 3,85 400 55,36 9,89 2,09 42,13 4,13

Acid Ca(H2PO4)2, neutralized with CaCO3 300 51,19 - 0,97 19,55 1,92 350 50,62 - 1,07 21,57 2,11 400 50,51 - 1,13 22,78 2,23

Fig. 1. A block scheme for obtaining of double superphosphate by the cyclic method.

Decomposition

Filtration

Cooling and Crystallization

Centryfugation

Sulfuric acid extraction

WPA Phosphorus concentrate

Neutralization

H2SO4

Filtrate

Mother solution

Acid undecompoused phosphate cake

Pure H3PO4 Granulation and Dry

NH3 or CaCO3


170

to obtain nitrogen-phosphorus fertilizers and double superphosphate. More details can be found in Fig. 1.

CONCLUSIONS

The results of the carried research show the suit-ability for application of the liquid cyclic method for processing of low-grade phosphorites of Central Kyzyl Kum to highly concentrated nitrogen - phosphorus and double (triple) superphosphate fertilizers. It is proved, that under the treatment, the phosphorus concentrate from phosphorites of Central Kyzyl Kum with a redun-dant norm of evaporated wet processing phosphoric acid (300, 350, 400 % stoichiometric norm to monocalcium phosphate), the phosphate mineral is not uncovered completely. That is why we propose to mix it with mother solution, and to subject it to sulfuric acid processing, in order to return back produced phosphoric acid for the decomposition process. However, research on the sul-furic acid processing of the undecomposed phosphate cake and mother solution, would be the subject of our future work.

Therefore, a possibility for obtaining two types of phosphoric fertilizers on the basis of the suggested ap-proach, which is suitable for introduction into the soil in different agrochemical data, is created.

REFERENCES

1. B.M. Beglov, Sh.S. Namazov, Phosphorites of Central Kyzyl Kum and their processing, Tashkent, 2013, (in Russian).

2. S.A. Abdurahmanov, M.A. Askarov, N.A. Doniyarov, H.U. Amonov, Investigation of flotation enrichment of phosphate ore of Jeroy-Sardarinsky deposit, Uzb. Moun. Bull., 1, 2002, 44-45, (in Russian).

3. M.A. Askarov, N.A. Doniyarov, E.A. Nurkulova, Flotation of phosphorite ore, Uzb. Moun. Bull., 3, 2005, 87-89, (in Russian).

4. B.E. Sultonov, Sh.S. Namazov, B.S. Zokirov, Chemical enrichment of low-grade phosphorites of Central Kyzyl Kum, J. Chem. Technol. Metall., 49, 3, 2014, 288-292.

5. V. Cesar Norbert, Industrial removal of chlorides from phosphate rock at wet washing in extreme conditions of Kyzyl Kum’s desert, Uzb. Moun. Bull., 4, 2008, 54-58, (in Russian).

6. Practical recommendation on Agriculture: Soil, Water, Fertilizer, Таshkent, 1996, (in Russian).

7. M.А. Shapkin, Т.I. Zavertyaeve, R.Y. Zinuk, Double Superphosphate. Technology and application, Leningrad, “Chemistry”, 1987, (in Russian).

8. М.Т. Juraev, A.U. Erkaev, Sh.S. Namazov, B.M. Beglov, Production of double superphosphate on a basis phosphorites of Central Kyzyl Kum by flow line method, J. Chem. Ind., 5, 2002, 42-44 , (in Russian).

9. B.E. Shamaev, PhD thesis, Treatment of Kyzyl Kum phosphorites for phosphorus-containing fertilizers,Tashkent, 2007, (in Russian).

10. M.E. Pozin, B.A. Kopilev, A.V. Firskina, Van Lee Shen, Method of double superphosphate produsing, Patent for invention № 343969 USSR. М.Class. S 05 b 1/04. B.I. № 21, 1972, (in Russian).

11. R.Y. Zinyuk, T.I. Fomicheva, M.A. Shapkin, M.E. Pozin, Researches in sphere of cyclic method of double superphosphate producing, J. Appl. Chem., 52, 7, 1979, 1445-1450, (in Russian).

12. А.А. Кuznesov, PhD thesis, Development and Introduction of resource-saving technologies of phosphate fertilizers, Моscow, 1993, (in Russian).

13. S.M. Moldekov, K.T. Jantasov, J.K. Janboltaeva, J.M. Altibaev, O.S. Balabekov, O. Koblanova, Kinetics of decomposition of low-quality phospho-rites by phosphoric acid and production of double superphosphate by cyclic method, Modern High Technol., 11, 2013, 107-112.

14. B.B. Myrzahmetova, U. Besterbecov, I.A. Petropav-lovskiy, Production of double superphosphate from Kukjan and Koksu by liquid phase method, J. Join. Sci., 2, 2012, 60-64.

15. N.R. Yusupbekov, H.S Nurmuhammedov, S.G Zokirov, Chemical technology basic process and equipment, Тashkent, 2003, (in Russian).

16. State Standard 20851.2-75. Mineral fertilizers, The test method of phosphate, Publishing house of Standards, Moscow, IPK, 1997, 37, (in Russian).

17. Methodical instructions of performance of tests extraction pulps and extraction phosphoric acid, Open joint-stock company “Ammofos-Maxam”, Almalyk, 2010, 16, 22, (in Russian).

18. State standard 21560.2 - 82. Fertilizers. Test methods, Publishing Standards, Moscow, 2003, 13, (in Russian).

19. V.N. Kochetkov, Granulation of mineral fertilizers, Moscow, 1975, (in Russian).

Fatemeh Bahadori, Khadijeh Mirza, Alireza Behroozsarand, Sima Rezvantalab

171


INVESTIGATION OF POROSITY EFFECTS IN THE EMISSION OF POLLUTANTS IN POROUS BURNERS


Department of Chemical EngineeringUrmia University of TechnologyP.O. Box 57166-41Urmia, IranE-mail: [email protected]

ABSTRACT

Furnaces are the devices, providing heat to industrial operations such as chemical reactions, extraction of metals, production of ceramic, etc. The main challenge of furnaces is the emission of massive air pollutants. However, the porous burners produce less pollutants, compared to the rest of them. In this paper, the finite volume method is used to investigate the effects of porosity on the production of pollutants in the porous burner. The simulation results show that temperature distribution in the low porosity burner is more uniform than in the high porosity one. In addition, increasing the porosity the maximum temperature of the flame increases, and thus reduces the conversion of fuel to carbon monoxide and increases production of nitrogen oxides. Therefore, the average porosity leads to reduction of the total emission of pollutants.

The simulation results are in good agreement with experimental data.Keywords: burner, porous media, modeling, pollutants.

Received 01 September 2014Accepted 30 January 2015

INTRODUCTION

Environmental crisis is one of the main problems that the world is facing today. One of the strategies for controlling the air pollution is the increasing of combustion efficiency and reducing fuel consumption. So, a burner should control the combustion of fuel for producing heat.

Porous burners consume low value fuels for produc-ing heat. This may be a solution to reduce the production of carbon dioxide. Weinberg and Howell et al. studied the porous materials and their convective, conductive and radiative properties [1, 2]. They described the flame stabilization within a porous burner and showed that the porous matrixes have higher burning speeds and lower flammability limits, than open flames.

Yoshizawa et al. numerically simulated an one-di-mensional porous burner by considering one-step kinet-ics [3]. They studied the effects of the heat transfer on the burning rate and flame structure. Barra et al. numerically studied an one-dimensional two-layered porous burner

[4]. They showed that the first layer should have a low volumetric heat transfer coefficient, low conductivity, and high radiative coefficient. The second layer should have a high volumetric heat transfer coefficient, high conductivity, and medium radiative coefficient.

Yu et al. experimentally studied the thermal effi-ciency and the emission properties of porous burners [5]. They investigated three kinds of porous burners: metal fiber (MF), ceramic (CM) and stainless steel fin (SF). They showed that the MF and SF burners had the lowest and highest CO emissions between the three types of burners, respectively. However, the MF and SF burners had the highest and the lowest NOx emission among the burners, respectively.

Li et al. experimentally studied the liquid fuel com-bustion in a porous media burner [6]. They showed that the flame of porous burner is affected by the airflow velocity.

In this paper, the effects of porosity on the emission of pollutants and combustion quality in porous burners are investigated.


172

NUMERICAL SIMULATION OF POROUS BURNER

The porous burner system is composed of two parts: 1) a solid matrix and 2) void space, saturated by fluid. Therefore, the continuity and momentum equations are written for gas phase and the energy equations are written for both phases. In this paper, a steady state and a two-dimensional system are considered. The effect of gravity force is neglected. So, the governing equations are as follows:

The continuit y equation [7]: (1)

where ρ is fluid density (kg/m3), and v is gas velocity (m/s).The momentumequations[8]:

(2)

where P is pressure (Pa), is fluid viscosity (kg m-1 s-1), and ε is porosity of the porous matrix.Gas phase energy equation [7]:

(3)

where h′ is convective heat transfer coefficient, k is thermal conductivity (W m-1 K-1), Cp is heat capacity (W Kg-1 K-1), R is reaction rate (kg fuel m-3 s-1), h is specific enthalpy (J kg-1), and hcis heat of combustion (J kg-1).Solid phase energy equation [9]:

(4)where qr is radiative heat transfer per unit area (W m-1

K-1).Fuel diffusion [7]:

(5)

where D is diffusivity (m2 s-1), and yf is mass fraction of fuel.Product diffusion [7]:

(6)where yp is mass fraction of products.

A single-step global reaction mechanism is selected for the complete combustion of fuel.

EXPERIMENTAL

Geometry of porous burner was generated using Gambit 2.0 software. The dimensions of the simulated burner are 13 cm diameter × 110 cm length. In order to obtain grid independent results, different uniform grid numbers - 20 × 120, 40 × 240 and 50 × 300 were tested for porosity of 0.5. It was noticed that the dif-ference between the length of the concentration of the fully developed zone, predicted for 40 × 120 and 50 × 150 grids is insignificant (< 1 %). The length of the concentration of the fully developed zone for the tested grid points were 9.276, 9.301 and 9.305 for grids of 20 × 120, 40 × 240 and 50 × 300, respectively. Thus, 40 ×

Fig. 1. The effect of porosity on the temperature distribution in the center line of the porous burner.


173

Fig. 2. The effect of porosity on the temperature distribution in the porous medium burner.

Fig. 3. The effect of porosity of burner on carbon dioxide production.

Fig. 4. Fully developed length of concentration and temperature vs porosity.


174

240 grid was used for generating the results. The model equations were solved using a commercial flow simula-tion software, FLUENT 6.0.1.

The temperature of the inlet stream is 300K and the velocity is 0.2 m s-1. The studied reaction is combustion of methane with 20 % excess air as following:

The above equations were solved for 7 different porosities: 0.10, 0.25, 0.40, 0.55, 0.70, 0.85, and 1.00.


In order to verify the validity of the results, the temperature of the center line of the porous burner with porosity of 0.5 is compared with the experimental re-sults, presented in the literature. As it is seen in Fig. 1, the simulation results are in the line with experimental results [10].

In this paper, combustion of methane with 20 % excess air in the porous burner is studied. Fig. 2 shows the effects of porosity on the temperature distribution in the porous burner. As it is seen, an increase of porosity rises the maximum temperature of the flame. Whereas, in low porosities, due to faster heat conduction, tempera-ture throughout the porous medium is lower and more uniform than at high porosities.

Fig. 3 shows the amount of carbon dioxide produced in the length of the flame. As it is seen, increasing the porosity increases the production of carbon dioxide. In other words, combustion process performs better than in the low porosity burner. It is noticed that in the combustion process, pollutants such as CO and NOx are produced. If the maximum temperature of the flame increases, the amount of produced carbon monoxide is reduced. In addition, by increasing the temperature, the amount of produced nitrogen oxides increases. In order to reduce the emission of the burner, the operating temperature should be at an optimum in order to reduce the production of all pollutants. Comparison of Fig. 2 and Fig. 3 shows that an increase in the porosity causes increases of the temperature especially in the maximum point. So, at the high porosities, the quality of combus-tion is better than at low porosities, CO production decreases, but the production of NOx increases. There-fore, the optimum porosity for decreasing the pollutants

simultaneously according to Fig. 2 and Fig. 3 is 0.5. Fig. 4 shows the fully developed length of concen-

tration and temperature in the different porosities. As it is seen, the porosity increasing extends the combustion process forward and increases the concentration fully developed length. In the other word, combustion takes place in the greater part of the burner, so, the efficiency of the process increases. In addition, increasing of poros-ity increases the length of reaching to fully developed zone for temperature. In other words, in the low porosity, temperature is more uniform than in the high porosity. So, increasing the temperature increases the production of nitrogen oxides. Therefore, an average porosity should be considered for reducing CO and NOx production, simultaneously.

CONCLUSIONS

In this paper, combustion of methane in porous burn-ers is numerically studied. The finite volume method is used to investigate the variation of porosity effects on improving the combustion process and decreasing the production of CO and NOx. It is observed that flame temperature increases by increasing the porosity of the burner. Increasing the temperature of the burner de-creases the production of CO and increases the produc-tion of NOx. So, the use of an intermediate porosity for minimizing the emission of pollutants seems reasonable.

REFERENCES

1. F.J. Weinberg, The first half-million years of com-bustion research and today’s burning problems. In: Fifteenth symposium (international) on combustion, v. 15, 1. Pittsburgh, The Combustion Institute, 1974, p. 1-17.

2. J.R. Howell, M.J. Hall, J.L. Euzey, Combustion of hydrocarbon fuels within porous inert media, Prog. Energy Combust. Sci., 22, 1996, 121-145.

3. Y. Yoshizawa, K. Sasaki, R. Echigo, Analytical Study of the Structure of Radiation Rontrolled Flame, Int. J. Heat and Mass Transfer, 3l, 2, 1988, 311.

4. A.J. Barra, G. Diepvens, J.L. Ellzey, M.R. Henneke, Numerical study of the effects of material prop-erties on flame stabilization in a porous burner,


175

Combustion and Flame, 134, 4, 2003, 369-379.5. Yu Byeonghun, Kum Sung-Min, Lee Chang-Eon, Lee

Seungro, Combustion characteristics and thermal ef-ficiency for premixed porous-media types of burners, Energy, 53, 2013, 343-350.

6. Junwei Li, Jinghuai Huang, Mi Yan, Dan Zhao, Junying Zhao, Zhijun Wei, Ningfei Wang, Experimental study of n-heptane/air combustion in meso-scale burners with porous media, Experimental Thermal and Fluid Science, 52, 2014, 47-58.

7. C.L. Hackert, J.L. Ellzey, O.A. Ezekoye, Combustion and Heat Transfer in Model Two-Dimensional Porous

Burners, Combustion and Flame, 116, 1999, 177-191.8. P. Magnico, Analysis of permeability and effective viscos-

ity by CFD on isotropic and anisotropic metallic foams, Chemical Engineering Science, 64, 2009, 3564 - 3575.

9. P.F. Hsu, W.D. Evans, J.R. Howell, Experimental and numerical study of premixed combustion within nonhomogeneous porous ceramics, Combustion Science and Tech., 90, 1993, 149-172.

10. F. Durst, D. Trimis, Compact Porous Medium Burner and Heat Transfer Exchanger for Household Applications, EC project report, Contact no. JOEC-CT95-0019, 1996.


176


PREDICTING THE EFFECT OF CELL GEOMETRY AND FLUID VELOCITY ON PEM FUEL CELL PERFORMANCE BY CFD SIMULATION

Bahman ZareNezhad, Mohammad M. Sabzemeidani

Faculty of Chemical, Petroleum and Gas Engineering,Semnan University, PO Box 35195-363, Semnan, IranE-mail: [email protected]

ABSTRACT

The CFD simulation of a proton exchange membrane (PEM) fuel cell has been carried out to investigate the effects of different parameters influencing cell performance. The model considers the mass and momentum transfer, charge con-servation and electrochemical reactions in the presence of membrane and catalyst. The predicted polarization curve is in good agreement with the measured data in the range 0 - 1 A/cm2. An increase in flow velocity from 0.1 to 0.5 m/s leads to a decrease in the optimum channel width from 0.8 to 0.6 mm. Also there is a sharp decrease in current density as the channel width increases above 0.8 mm at all flow velocities. The presented CFD approach can be used for determination of the optimum fluid velocity and channel geometry of a PEM fuel cell.

Keywords: PEM fuel cell, CFD simulation, channel width, current density.

Received 04 November 2014Accepted 30 January 2015

INTRODUCTION

There has been a significant progress in the stud-ies of the proton exchange membrane (PEM) fuel cell technology in the last decade. Despite great successes, the automotive application of PEM fuel cell (PEMFC) faces many technical challenges [1, 2]. A PEM fuel cell consists of three essential compartments: polymer electrolyte membrane, an anode and a cathode [3]. At the anode the hydrogen is oxidized into electrons and protons, while at the cathode, the oxygen is reduced [4]. The wet reactant gas transfers through the gas dif-fusion layer (GDL) into the catalyst layer where the electrochemical reactions take place [5]. Electrodes are considered as porous media where reactant gases are distributed on the catalyst layers [6]. The PEM fuel cell is a renewable energy source and is important because of factors such as low operating temperatures, modular structure, quick start-up time, high power density, corro-sion resistance and environmental benefits. On the other hand, the fuel cells technologies are still more expensive

than the conventional fossile fule-based alternatives [7].The flow field design and scale of a bipolar plate

is one of the significant issues in PEM fuel cells [8]. Maharudrayya et al. studied the pressure drop and flow distribution in multiple parallel channel configurations, used in PEM fuel cell stacks [9]. Dutta et al. obtained velocity, density and pressure contours in the gas dif-fusion layers [10]. They used a finite volume technique for solving model equations. Their results showed that the current direction is drastically dependent on the mass transfer mechanism in the membrane-electrode assembly. Futerko and Hsing applied finite element method (FEM) for solving governing equations in the gas diffusion layers and flow channels [11].They studied the resistance of membranes in polymer electrolyte fuel cells. Their modeling findings illustrated the conclusion that the mole fraction of reactant gases, water content in the membrane and current density are dependent on pressure. More recently, Rodatz et al. carried out studies on the operational aspects of a PEM fuel cell stack under practical conditions [12]. Their work focused exclu-


177

sively on the pressure drop, two-phase flow and effect of bends. They noticed a decrease in the pressure drops at a reduced stack current. Ahmed and Sung performed a numerical model to investigate the effects of channel geometrical configuration [13].

In this work, the effects of channel geometry and fluid velocity on the performance of a single PEM fuel cell with a straight channel at co-current flow condi-tions are evaluated and the optimal operating conditions at different flow conditions are investigated by CFD simulation. The schematic diagram of simulated PEM fuel cell comprising a cathode, an anode, gas diffusion layers , catalyst layers and a proton exchange membrane is shown in Fig. 1.

Proposed CFD ModelingA 3-dimensional isothermal model is developed to

account for the mass, momentum and species balances in the flow channels, GDL and the catalyst layers. Elec-trochemical reactions in the catalyst layer and GDL parts of the fuel cell and membrane are also considered in the proposed model. Model assumptions include steady state condition, laminar flow regime, ideal gas behav-ior, gas phase, fixed proton conductivity of membrane, isotropic gas diffusion layer, activation over potential is constant, the membrane is not permeable for reactant gases, isothermal operation. All governing equations are summarized as follows:

Continuity equation:

. bru Qρ∇ =

(1)

Momentum balance equation:

2

2.[ (( ) ( ) ) ( )] ( ) 03

T br

p p br p

Qp u u u uk

µ µ µε ε ε

∇ − + ∇ + ∇ − ∇ − + =

2

2.[ (( ) ( ) ) ( )] ( ) 03

T br

p p br p

Qp u u u uk

µ µ µε ε ε

∇ − + ∇ + ∇ − ∇ − + =

(2)

Mass transfer equation:

. ( . ) i iij u Rρ ω∇ + ∇ =

(3)

Charge conservation equation:

,.( )effe e v totaliσ φ∇ ∇ = ± (4)

where brQ is source term in the catalyst layer due to reaction of reactant species and can be calculated from the following equation:

,br i m im i

Q R M=∑∑ (5)

,i mR are the electrochemical reactions of hydrogen and oxygen and water:

,,

i m vi m

m

iR

n Fυ

=

(6)

The electrochemical reactions of the sink term for water, hydrogen and oxygen are shown in Table 1.

The diffusional mass flux vector is calculated from the equation:

1( ( [( ) ]))i ik k k k Aik A

j D x x pp

ρω ω= − ∇ + − ∇∑

(7)

where ikD is diffusivity of mixture that can be calcu-lated using Maxwell-Stefan equations. The Maxwell-Stefan diffusion is a model for describing diffusion in multicomponent systems:

1.5effik ikD Dε= (8)

Table 1. Electrochemical reactions of the sink term for water, hydrogen and oxygen.

O2 H2 H2O

4viF

− 2

viF

− 2

viF

Fig. 1. Schematic diagram of a PEM fuel cell.


178

Within porous media, electrical conductivity of electrode and membrane are estimated from the modified Brueggemann equation:[14]

1.5effe e eσ σ ε= ×

(9)

The source term, ,v totali is given by:

,v total v locm

i a i=∑ (10)

The current densities, loci are calculated using the Butler-Volmer equation in anode and in cathode.

2

2

, ,0.50, ( ) (exp( ) exp( ))H a a act a c act

a a refH

Y F Fi i

Y RT RTα η α η−

= − (11)

2

2

, ,0, ( )(exp( ) exp( ))O c a act c c act

c c refO

Y F Fi i

Y RT RTα η α η−

= − (12)

where and are the exchange current density of the anode and the cathode at a reference temperature (25°C) and pressure (1 atm), respectively.

and are the anodic and the cathodic transfer coefficients for the reaction at the anode.

and are the anodic and the cathodic trans-fer coefficients for the reaction at the cathode. and are the mass fractions of hydrogen and oxygen at a reference temperature (25 °C) and pressure (1 atm), respectively. The values of the exchange cur-rent density and the transfer coefficients are listed from Siegel et al. [15]. R is the universal gas constant (8.314 J mol-1 K-1), F is the Faraday constant (C mol-1), and T is the temperature (K). The activation over potential,η is given by:

act s l eqEη φ φ= − − (13)

The relevant numerical and geometric parameters for this study are shown in Table 2 and Table 3.


The simulation results are compared with experi-mental data [16] in Fig. 2. The predicted cell potentialls are in excellent agreement with the measured data in the wide current density range of 0 - 1 A/cm2.

Fig. 3 shows the effect of the channel width on the

current density at a constant channel length and different flow velocities. Depending on the flow velocity, there is an optimum channel width for which a maximum current density is obtained. An increase in the flow velocity from 0.1 to 0.5 m/s leads to a decrease in the optimum channel width from 0.8 to 0.6 mm. There is a sharp decrease in

Parameters Value

Inlet anode/cathode temperature (°C) 80

Inlet anode/cathode pressure (atm) 3

Anode/cathode flow velocity (m/s) 2

Relative humidity of inlet gases (%) 100

Mole ratio of O2/N2 0.79/0.21

Mass fraction of H2O (anode) 0.7

Mass fraction of H2 (anode) 0.3

Mass fraction of H2O (cathode) 0.14

Mass fraction of O2 (cathode) 0.2

Mass fraction of N2 (cathode) 0.66

Inlet O2 concentration (kmol/m3 ) 0.00086

Inlet H2 concentration (kmol/m3 ) 0.04

Table 2. Operating conditions of the simulated PEM fuel cell.

Description Value

Channel length, mm 120

Channel width, mm 1.0

Channel depth, mm 1.2

Membrane thickness, mm 0.036

Catalyst layer thickness, mm 0.012

Electrode thickness, mm 0.21

Table 3. Geometric parameters of the simulated fuel cell.


179

current density as the channel width increases above 0.8 mm at all flow velocities, shown in Fig. 3.

Fig. 4 shows the effect of channel depth on the current density at a constant channel length and dif-ferent flow velocities. The numerical results confirm that the current density increases as the channel depth is increased at a constant flow velocity. Comparison of Figs. 3 and 4 suggests that regardless of channel width or depth, there is not a significant change in the current density at flow velocities higher than 0.2 m/s, which is quite interesting from engineering viewpoint.

For a cell potential of 0.6 V, the profile of the hydro-gen mole fraction at the anode channel and gas diffusion layer at the different flow velocities, is shown in Fig. 5. This figure testifies that the hydrogen mole fraction at the anode side decreases gradually from inlet to outlet of the channel due to the consumption of hydrogen at the anode catalyst layer. As the mole fraction of hydrogen in the gas diffusion layer decreases, the strong oxida-tion of hydrogen can be seen. Fig. 5 a,b shows that the

Fig. 2. Comparison between the predicted and the meas-ured cell potential.

Fig. 3. Effect of channel width on the PEM fuel cell cur-rent density at a cell potential of 0.6V.

Fig. 4. Effect of channel depth on the PEM fuel cell cur-rent density at a cell potential of 0.6V.

Fig. 5. Hydrogen mole fraction at the anode channel and gas diffusion layer: a) flow velocity 0.1 m/s; b) flow veloc-ity 0.5 m/s.

a)

b)


180

average outlet mole fraction of H2 increases, as the flow velocity is increased.

Fig. 6 displays the effect of channel width on the avarage outlet mole fraction of H2 at a constant chan-nel length for different flow velocities. The predicted results confirm that the avarage outlet mole fraction H2 is increased, as the channel width is increased at the different flow velocities. As observed, the maximum hydrogen consumption takes place at flow velocity of 0.1 m/s. At flow velocities higher than 0.2 m/s, there is no significant change in the hydrogen mole fraction at the cell outlet.

Fig. 7 displays the effect of channel depth on the avarage outlet mole fraction of H2 at a constant chan-nel length for different flow velocities. The numerical results confirm that the avrage outlet mole fraction H2 increases as the channel width is increased at the differ-ent flow velocities. As shown, the maximum hydrogen consumption takes place at a flow velocity as low as 0.1 m/s similar to what is observed in Fig. 6.

Fig. 8 compares the effect of channel depth and channel width at a fixed flow velocity of 0.1 m/s. As shown, the channel width is more effective than the channel depth, regarding the mole fraction of hydrogen at the channel outlet.

Fig. 9 displays the effect of channel width on the average outlet mole fraction of O2 at a constant channel length and a constant channel depth at different flow velocities. The predicted results confirm that the average outlet mole fraction of O2 increases as the channel width is increased at different flow velocities.

Fig. 10 shows the effect of channel depth on the average outlet mole fraction of O2 at a constant channel length and a constant channel width, at different flow velocities. The predicted results confirm that the average outlet mole fraction of O2 increases as the channel depth is increased at the different flow velocities.

Fig. 11 shows the relationship between the average outlet mole fraction of H2 and the inlet fluid velocity at a cell potential of 0.6 V. It is clearly seen that an

Fig. 8. Effect of channel dimensions on the mole fraction of H2 at outlet channel.

Fig. 6. Effect of channel width on the mole fraction of H2 at outlet channel.

Fig. 7. Effect of channel depth on the mole fraction of H2 at a outlet channel.

Fig. 9. Effect of channel width on the mole fraction of O2 at outlet channel.


181

increase in the inlet fluid velocity leads to an increase in the the average outlet mole fraction of H2. It should be noted that an increase in inlet fluid velocity causes significant hydrogen loss at the channel outlet of PEM fuel cell. The presented CFD approach can be used for determination of the optimum fluid velocity and channel geometry of a PEM fuel cell. As shown in Fig. 12, an increase in flow velocity from 0.1 to 0.5 m/s leads to a

considerable increase in the current density. However, an increased flow velocity from 0.1 to 0.5 m/s leads to a significant increase in the mole fraction of H2 at the cell outlet according to Fig. 11. Therefore, the optimum design of the PEM fuel cell can be established by the proposed CFD model.

CONCLUSIONS

A 3-dimensional CFD simulation of a PEM fuel cell according to a rigorous finite element numerical method is presented. The simulation is conducted to investigate the effect of channel geometry and fluid velocity on the performance of a PEM fuel cell. The predicted polariza-tion curve is in good agreement with the measured data in the range 0 - 1 A/cm2. An increase in flow velocity from 0.1 to 0.5 m/s leads to a considerable increase in the current density. However, an increased flow velocity from 0.1 to 0.5 m/s leads to a significant increase in the mole fraction of H2 at the cell outlet. It is found that the channel width is more effective than the channel depth, regarding the mole fraction of hydrogen at the channel outlet. It is shown that depending on the flow velocity, there is an optimum channel width for which a maximum current density is obtained. The optimum design of the PEM fuel cell can be determined by using the proposed CFD model and economic considerations.

REFERENCES

1. S.G. Chalk, J.F. Miller, F.W. Wagner, Challenges for fuel cells in transport applications, J. Power Sources, 86, 2000, 86-94.

2. M.R. Ashraf Khorasani, S. Asghari, A. Mokmeli, M.H. Shahsamandi, B. Faghih Imani, A diagnosis method for identification of the defected cell(s) in the PEM fuel cells, International Journal of Hydrogen Energy, 35, 2010, 9269-9275.

3. J.H. Jang, W.M. Yan, C.C. Shih, Numerical study of reactant gas transport phenomena and cell per-formance of proton exchange membrane fuel cells, Journal of Power Sources, 156, 2006, 244-252.

4. B. Ramos-Alvarado, A. Hernandez-Guerrero, D. Juarez-Robles, P. Li, Numerical investigation of the performance of symmetric flow distributors as flow channels for PEM fuel cells, International journal of hydrogen energy, 37, 2012, 436-448.

Fig. 10. Effect of channel depth on the mole fraction of O2 at outlet channel.

Fig. 11. Effect of inlet fluid velocity on the average outlet mole fraction H2.

Fig. 12. Effect of inlet fluid velocity on the average cur-rent density.


182

5. H.S Chu, C. Yeh, F. Chen, Effects of porosity change of gas diffuser on performance of proton exchange membrane fuel cell, Journal of Power Sources, 123, 2007, 254-263.

6. F. Hashemi, S. Rowshanzamir, M. Rezakazemi, CFD simulation of PEM fuel cell performance: Effect of straight and serpentine flow fields, Mathematical and Computer Modelling, 55, 2012, 1540-1557.

7. N. Ahmadi, S. Rezazadeh, I. Mirzaee, N. Pour-mahmoud, Three-dimensional computational fluid dynamic analysis of the conventional PEM fuel cell and investigation of prominent gas diffusion layers effect, Journal of Mechanical Science and Technol-ogy, 26, 2012, 2247-2257.

8. X.D. Wang, Y.Y. Duan, W.M. Yan, Novel serpen-tine-baffle flow field design for proton exchange membrane fuel cells, Journal of Power Source, 173, 2007, 254-267.

9. S. Maharudrayya, S. Jayanti, A.P. Deshpande, Pres-sure losses in laminar flow through serpentine chan-nels in fuel cell stacks, J. Power Sources, 138, 2004, 1-13.

10. S. Dutta, S. Shimpalee, J.W. Van Zee, Numerical prediction of mass-exchange between cathode and

anode channels in a PEM fuel cell, International Journal of Heat and Mass Transfer, 44, 2001, 2029-2042.

11. P. Futerko, I.M. Hsing, Two-dimensional finite-ele-ment method study of the resistance of membranes in polymer electrolyte fuel cells, Electrochimical Acta, 45, 2000, 1741-1751.

12. P. Rodatz, F. Buechi, C. Onder, L. Guzzella, Opera-tional aspects of a large PEFC stack under practical conditions, J. Power Sources, 128, 2004, 208-217.

13. D.H. Ahmed, H.J. Sung, Effects of channel geometri-cal configuration and shoulder width on PEMFC per-formance at high current density, J. Power Sources, 162, 2006, 327-339.

14. P.T. Nguyen, T. Berning, N. Djilali, Computational model of a PEM fuel cell with serpentine gas flow channels, J Power Sources, 130, 2004, 149-57.

15. N.P. Siegel, M.W. Ellis, D.J. Nelson, M.R. von Spa-kovsky, A two-dimensional computational model of a PEMFC with liquid water transport, Journal of Power Sources, 128, 2004, 173-184.

16. L. Wang, A. Husar, T. Zhou, H. Liu, A parametric study of PEM fuel cell performances, International Journal of Hydrogen Energy, 28, 2003, 1263-1272.

Kosta Boshnakov, Venko Petkov, Metodi Nikolov

183


DECISION MAKING FOR CONTROL OF COMBUSTION PROCESS OF PULVERIZED COAL


University of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected]

ABSTRACT

A method and system for decision making and control (DMC) of combustion process of pulverizing fuel-air mixture in large multiburner furnace is proposed. Due to the lack of necessary information inference measurements are created on the basis on the first principle CFD model of the combustion process and a sequential image processing. Decision making procedure is based on Case Based Reasoning and Rule Based Reasoning which contains some elements of knowledge, acquired via a long time operator experience and off-line simulations on the CFD – model. Combination of off-line and on-line procedures is accepted in to two stage algorithm of DMC which improves the efficiency and safety of combustion process.

Keywords: Case Based Reasoning (CBR), combustion process, control, decision making, inference measurement, Rule Based Reasoning (RBR).

Received 05 January 2015Accepted 24 February 2015

INTRODUCTION

The combustion process (CP) is still important for a great number of plants in chemical, metallurgical and power industries. Despite of the large volume of long standing investigations, the control of the CP, especially in a big size combustion chamber with a multiburner systems, continues to be at least in part the problem of operator skill. The main reason is a lack of relevant in-formation for main characteristics of the turbulent flame. To cope with this obstacle into the last decade a lot of methods for flame measurement appear – laser, fiber-optical, video-camera, spectroscopical, colour-based [1]. This methods have a promising potential, but nowadays they are with limited commercial success because are costly, unreliable and with problematic maintenance in hard industrial conditions [1, 2].

In parallel a very extensive development of inference methods for measurement, monitoring, decision making and control, can be observed [3-5]. They are considered as an alternative or compliment solution in reengineer-ing initiative [6]. In the present paper a new approach is presented for decision making for improvement of combustion process based on softsensing and inference control [3, 5, 7]. A novel hybrid system is proposed based on mathematical modeling, image processing and incorporating some intelligent techniques containing knowledge elements – Case Based Reasoning (CBR) and Rule Based Reasoning (RBR) [8 - 11].

As a case study, a typical plant is chosen for decision making and control of combustion process of pulverized fuel-air mixture with direct feeding. The technological and economical considerations are combined with the

xNO minimization in flue gases.


184

STEAM BOILER PROPERTIES

A typical scheme of a steam drum-type boiler plant with mill-fans based pulverization system is shown in Fig. 1. The combustion process is carried out in furnace 1, where through a system of main (12) and auxiliary (13) burners a mixture of pulverized coal, air (primary and secondary) and flue gases is fed. The pulverization system comprises a mill-fan (8), electro-drive (9), iner-tial separator (10) and dust concentrator (11). The raw coal from the bunker 3 trough the feeder 4 is directed toward the drying shaft 5, where it is mixed with flue

gases intaked from the top of the furnace via the duct 2. The furnace crossection (Fig 1b) shows the placement of 8 mill-fans (A-H) with tangential organization of the combustion via central vortex.

The direct firing pulverizing system is a source for a lot of difficulties [1, 2, 4, 6]:

l The flame stability depends on the combination of many factors: mill-fans load, their technical condition, fuel/air flow ratio;

l Big number of disturbances such as changeable heating value of the coal, nonregular work of the row coal feeder, strongly variable temperature of the drying

Fig. 1. Typical scheme of a drum-type steam boiler.

3

4

6

Vac

8

9

11

12

13

θgs2

x

yz

MFH MFD

MFB

MFCMFG

MFF

MFEMFA

Vah

7

10

5

θgt

θaf

θgf

Ηf

14

15 θ’sh

Ο2

1


185

flue gases;l Risk of fire in the air-fuel mixture;l Problems with overloading and underloading of

the mill fans;l Relatively short operational life of the mill-fans

(less than 2000 h);l The conventional measurements in pulverizing

system are very poor. A large number of key variables are unmeasurable: fineness of the dust, flow rate of the air-fuel mixture, degree of the separator recirculation, flow rate of the additional cold air and hot air, moisture of the pulverized coal.

The number of control variables, only the throughput capacity of the fuel and secondary air, is very limited.

The pulverization system plays the main role in organization and maintaining the combustion process. Existing conventional monitoring and control systems are not able to cope with the listed above circumstance in order to ensure the flame stability, safety and effec-tiveness of the combustion process.

PROBLEM STATEMENT

The objective of the considered problem is to pro-pose a relevant procedure for Decision Making (DM) and control of dust-air mixture combustion process into a multiburner furnace in order to satisfy the next multicriteria requirements:

1 2 3 4 5, , , ,J J J J J J extr=< > − (1) where iJ are partial criteria as follows:

1J - Efficiency of the combustion process maximiza-tion getting up by: - Fuel/air ratio optimization, - Radiation/Convection heat balancing, - Fowling minimization;

2J - xNO concentration in flue gases restriction;3J - Minimization the risk of abnormal conditions

owing to:l Slagging,l Flame unstability (e.g. Lost Of Ignition - LOI),l Local overheating of constructive elements (e.g.

arousing the water tube ruptures),l Overloading or underloading of mill-fans,

l Fire into the pulverizing system;4J - Minimization of off/on and over switching of MF; 5J - Minimization of Kindle Oil Burner (KDB) switches

on/off.The problem can be solved via combination of

quantitative approaches (formal methods, CBR, RBR) and qualitative ones (operator expertise).

MATHEMATICAL MODELING OF THE COM-BUSTION PROCESS

In this work a first principal ε−k model of the combustion process is accepted. Following [12], the CFD model could be presented in the next generalized form:

( ) ϕϕϕϕΓϕρ p

iii

i

SSxx

ux

++

∂∂

∂∂

=∂∂

(2)

where ix is a direction of Cartesian coordinate system and ( 3,2,1i = ) for 3D space, iu are the velocity compo-nents in ix direction; ϕ is a generalized variable, which represents all main components of the thermodynamical state of the system – velocity, ε−k characteristics, fraction contents, entalphy (temperature) and masses;

ϕΓ is a generalized transfer index, corresponding to each particular acceptation of ϕ , ϕS and ϕpS are generalized sources terms for a gas and particle phases. More detailed description of this CFD model can be found in [12].

The next values for each discrete element of the mesh are determined: the trajectories of the dust parti-cles; the heat received from dust particles combustion; temperature of the gas and coal particles; density of gas-fuel mixture; sequential velocities of the gas and dust correction in dependence on the temperature and density in the mesh.

The input conditions for the model are: power unit load N,MW; total throughput capacity of the coal tB , t h-1; working lower heating value of the coal W

LQ , kJ kg-1; concentration of oxygen in flue gases 2O , %; ef-ficiency of the boiler bη , %; Temperature of the cold air acθ , C0 ; temperature of the hot secondary air ahθ , C0 ; temperature of the flue gases at the end of furnace gtθ , C0 ; working moisture of the coal wW , %; work-

ing mass of the ash in coal wA , % ; theoretical volume of air for combustion 0V , m3 kg-1 ; portion of regularly


186

blowed air g , %.In order to adapt the model parameters, the relevant

constructive data are used for combustion chamber, burners, mill-fans, inertial separators, etc.

Model tuning is carried out via iterative correction of model parameters (Fig. 2), using data from number of direct measurements, arranged in the following vectors:

( )1 2, , , , , , ,t i ac ac ah ahY N B B V V Oθ θ=

( )g,A,W,,QY wwB

wL

I2 η=

( )3 0, ,gt

I I I IgfY Vθ θ=

(3)( )'

1 , , , ,afi gsi sh sh fZ D Hθ θ θ=

( )1 , ,Mgsi gt gfZ θ θ θ=

The data for the temperature of intake drying gases gsθ

must be filtered because of significant measurements

errors due to fouling, slagging, aerodynamic shadows. This can be seen from the experimental data given in Table 1.

IMPUT DATA PROVIDING

Three types of input data for considered above first principle CFD model to compute the temperature, ve-locity and concentration fields of combustion process, are used:

From direct measurements: afiθ , gsiθ , 2O , acθ ,

Fig. 2. Scheme of model parameters correction.

Inferencemeasurement

SteamBojler

Modeltuning

Noise

z1M

_

Measurableinputs

Measurableoutputs

z1

z2

y1

y2Unmeasurableinputs

yI3

yI2

y1

ym

Unmeasurableoutputs

Model

Intelligentfiltration

e=z1- z2M

z2M

Table 1. Experimental data for temperatures of intake drying gases gsθ .

Experiment No

Mill - Fan A B C D E F G H

1 - 772 875 929 876 851 - 916 2 - 915 - 930 936 814 - 931 3 - - 953 923 871 872 877 - 4 - - 932 921 857 876 877 - 5 975 918 - 909 - 928 - 930 6 - 854 - 932 868 809 - 932 7 841 871 - 824 662 - 734 840 8 794 885 - 845 634 - 738 861 9 792 - - 793 549 - 693 803

10 978 906 886 812 1009 - 920 -


187

iB , ahiθ , aciV , atV , shG (i=1,8) From indirect measurements - fα , Bα , MFVFrom inference measurements:a. Based on information from direct and indirect

measurements: Bη , wLQ , wW , wA , dA ;

b. Based on established design models [13]: 0V ,gtθ , gtθ .

Here some of the inference measured values are considered, which contain new authors results.

Efficiency of the boiler can be found using two dif-ferent ways [4,13]:

Direct balance method

BAQw

L =η , ∑ ∆= jj iDA (4)

where jD - output steam flow rates, ji∆ - correspond-ing enthalpies differences.

Indirect (inverse) balance method6

21IN i

iqη

=

= −∑ (5)

where iq ( )62i ,= are respectively the relative heat losses with: waste gases, incomplete combustion, un-burned carbon to surrounding, and with physical heat of slag.

The heat balance can be presented additionally in the next form:

wL

qN qQB F

η = = , where NBF = (6)

where N is the current electrical power load, q - func-tion, which was derived using experimental data. It is presented in Fig. 3.

According to eq. (4) and eq. (6) we have two rela-tionships for the working lower heating value of fuel

wLW . As the right sides of the eqs. (4) and (6) are known,

and INη is defined from eq. (5), wLQ can be derived us-

ing the equation:

ˆ ( ) ( ) (1 ) ( )w w wL LD LFQ k Q k Q kβ β= + − (7)

where β is weighting coefficient.The relationship between the coal moisture wW

and the ash content wA have been found in the form

ˆ94,2 0,0139w w wLW A Q+ = − (8)

The values of 0V , gtθ , gfθ , which are of critical importance for the developed model parameters tuning, are derived following the recommendations in [13].

SIMULATION RESULTS

The established methods for combustion chamber design [e.g. 13] are based on using averaged values of temperatures in each crossection of the furnace. This design patterns do not give any data for the spatial displacement of the flame, no thermograms in zones of interests or hydrodynamic characteristics of the combus-tion process. As it was discussed above, the problem of local information of flow, temperature and concentra-tion fields are very important for avoiding unadmissible situations via proper control of the combustion process.

Realistic scenarios have been accepted to cover the representative operational situations with enough density. In order to obtain more detailed maps of the main characteristics of the combustion process a lot of simulation results using this different working structures, load, technical conditions and values of input variables have been received. As an illustration some of them are shown in Figs. 4-7. They are in accordance with Fig. 4, showing the horizontal and vertical cross-sections.

Some conclusions can be made from the analysis of the simulation results:

l The temperature and concentration fields are strongly nonregular, nonhomogeneous and with a lot of local peculiarities.

l The velocity field is more regular, but the corre-sponding maps are less informative than the temperature Fig. 3. Relative heat losses with waste gases.

q

150 180 210

2100

2000

2210 NbNbbq ++=

N


188

z

y

xC

AEZ=7

F

z=20

G

10

20

z=40

30

40

HD

30

B

z

xz=20

10

20z=4030

40

10

20

y=7

2010

Fig. 4. A scheme of horizontal (a) and vertical (b) cross-sections.

(a) (b)

X

Y

AE

F

G

H D

C

B

Fig. 5. Cross-section of the furnace at the burner level.

Fig. 6. Cross-section of the furnace at the top of the com-bustion chamber.

Fig. 7. Vertical cross-section distribution of the flame.

a)

b)

c)


189

distribution.l Nonregularities of the temperature field exist both

in radial and axial direction.l The concentration fields are closely correlated

with the temperature field.All simulation results show that the digital model is

very complex and can not be directly used for decision making and control.

The calculation time for each scenario is too long in comparison with the combustion time of a separate particle (3 - 4 s). Thus only after some kind of approxi-mation of the obtained 3D fields they can be acceptable for on-line implementation.

Image processing should be applied for appropriate features selection.

PROCEDURES PROVIDING COMPONENTS FOR DECISION MAKING AND CONTROL

Feature selectionNowadays the image processing [1, 2] allows the

problem of feature selection from sophisticated images to be solved successfully. In our study it is addressed in three most important cross-sections:

Horizontal cross-sections at the burner’s levelThe next features are selected in order to form the

attributes for CBR:l Deriving the center of combustion vortex from

velocity distribution (Fig. 5a),l Determination the degree of cross-section filing

by flame using the isoterms (Fig. 5b),Average temperature at the burner level,Estimation of the local burner stream flame length

for each working burner.b. Horizontal cross-sections at the top of the furnacel Deriving the center of the vortex,l Defining the zones with highest temperatures lean

against the water tube walls,l Determination the local average gas temperatures

into the inputs of intake drying shafts,l Average temperature of the flame at this level,l Finding the zones with the maximal temperatures,l Extraction the isotherms with temperature >

C10000 .c. Vertical cross-sections at the axis with the highest

temperature in the top of the furnace

l Defining the place of the maximum temperature in axial direction,

l Estimation of the equivalent diameter of the fireball.Case base for CBR formation CBR is promising approach for reasoning on the

base of analogy [8 - 10]. In this study the cases are pre-sented in established “problem (P) solution (S)” form:

,=< >C P S (9)

Here P is accepted to be considered as a “situa-tion” in a form

, , P=< >P I F T (10)

where I is the vector of input information, F is the vector of the selected features from the image process-ing, T is the vector of given thresholds as requirements. Each of vectors TFI ,, contains number of relevant attributes.S is solution for MF control actions and is presented as follow:

, , , ,atV=< >ac ah sS B V V T (11)

where , ,ac ahB V V are the vectors of the throughput capacity of the fuel, flow rate of the cold and hot air, respectively. atV is the total combustion air flow rate,

sT is the vector of the thresholds.As in this study a two stage procedure for DMC

is proposed, two corresponding case bases have been formed - simplified )S,P(C sss and full ),( FFF SPC .

Rule base for RBR formation On the base on a long time of operational experience,

existing technical instructions, and supplier’s require-ments a set of particular iRB have been derived. A small part of them is presented in Tables 2, 3 and 4.

Decision making approachThe basic acceptance in DM approach in this study are:l CBR is used in two different variants - simplified

and full;l CB is used to create CBR local regressions cor-

responding to the iKNN area of interest [8],l Simplified procedure at the first stage of DM


190

allows a faster retrieval. It is concentrated mainly on the most important aspect of the combustion process – availability ( 3J ),

l At the second stage of DM, partial criteria 543 JJJ ,, can be accepted as constraints,

l To faster the calculations for an optimal distribu-tion of the coal iB and cold air aciV between iMF the changes of control variables are carried out by actions over the opposite pairs of burners.

Software toolsThe established tools for CBR, myCBR and

Table 2. Regime checking.

1 IF 133fg hm10.1850V −< THEN there still exists reserve in exhausters

2 IF ( )wL

minff Q,Nαα ≥ THEN there exists reserve in combustion air

3 IF C180C100 0afi

0 ≤≤ θ (i=1,8) THEN there isn’t danger of fair in burner tract 4 IF 1

MFi1 th55Bth5 −− <≤ (i=1,8) THEN there exists grinding capacity in each mill-fan

5 IF OHmm2HOHmm10 2f2 << THEN the vacuum into the furnace is admissible

1 IF 15.0rr

max

≤=ρ THEN the flame deviation is OK

2 IF C300gsi <θ∆ THEN the radial asymmetry is OK

3 IF C9500maxgsi <θ THEN the temperature 9n intake shaft is OK

4 IF C11000gf <θ THEN the temperature of the flue gases at the end of furnace is OK

Table 3. Checking the slagging conditions.

Rule No IF THEN

1R 05,0<ρ No flame position correction

2R 15,005,0 ≤< ρ Flame position is corrected only via iMF loading with coal (

0Bi ≠∆ )

3R

3,015,0 ≤< ρ

Flame position is corrected with both loading ( 0B ≠∆ ) and flowrate of cold air 0Vac ≠∆ of corresponding pair Mill-Fan

4R 3,0>ρ with variations

Switching over another pair Mill-Fans

5R 3,0>ρ constantly Switching on new Mill-Fan … … …

iR Vibration amplitude of MF is mm5,0a ≥

Switching over new Mill-Fan

Table 4. Certain rules for decision making for separated Mill-Fan.

jColibry, have been used in accordance with the earlier accepted adaptation and modification [9, 11].

A SYSTEM FOR DECISION MAKING AND CON-TROL (DMC) OF COMBUSTION PROCESS

An iterative two level system for DMC is proposed (Fig. 8). At the first level 1L (left side of the scheme) the preliminary recommendations are deduced for:

a. Current structure of the pulverizing system (num-ber and indexing of the proposed for running mill-fans (MFs), state of the Kindl Oil burners (KOB);


191

Fig. 8. A system for decision making and control (DMC).

Creation of new complete situation description Pc

Check satisfying of all thresholds T

Necessity of change control variables

Logical block LB12

Finding the best relevant case C*21

The requirements for the level two are full tiled?

New indirect and inference measurements

QLw, Ww, Aw, Ad, ηb, αf, VMF, Nox, Qgt, Qgf, Vo,

New situation

Dispatching load demand change

New direct measurements

Θaf, Θgs, O2L,R, , , Vat, Θ′sh, hf, Bi, , , Gsh

Send received control variables Bi, , in upper level

Simplified case obtaining

Retrieval KNN11 in CBs

Finding the best relewant case C*11

RB query for rules and regime admissibility

Criteria and strategy for DM selection

Optimal distribution of coal between MFi

Secondary retrieval KNN12 in CBs

The requirements for the Level one are full tiled

No

Yes

Determination of control structure and variables Bi, ,

Logical block LB21

Retrieval KNN2 in CBF

Comparison solutions from C*21 and C*12

Finding the best relevant case C*12

Is distance //C*21 – C*12// > Δ?

Weighted new control decision for Bi, ,

Real accomplishment of control decision

Multicriteria estimation

No

Yes

Yes

No

New situation


192

b. Distribution of coal iB and cold air aciV for each i - indexed MF. To define these recommendations the subsistem 1L uses an information from direct, indirect and inference measurements. After retrieval of k-nearest neighbours ( 11KNN ) and adaptation according 4R cycle for CBR [8 - 10], a primary optimal ( , acB V ), distribution is carried out. Because the initial retrieved case is changed, an iterative procedure is started via

12KNN retrieval. When the requirements for level 1Lare covered, these recommendations are sent via logical block 21LB toward the functional level 2L (right side of Fig. 8). As the case base FCB addressed to the level 2Lcontains more attributes in comparison with sCB a new retrieval 2KNN is fulfilled. A comparison of both values

*12C and *

21C directs the procedure to be executed in

iMF , or to continue the improvement of the preliminary recommendation given from the first level 1L .

The proposed decision making and control system (DMC) is situation based. It adapts the pulverizing sys-tem toward the new conditions. In a relatively simple way the proposed system can maintains the thermal fireball in the center of the combustion chamber and will improve the efficiency of the whole combustion process accordingly the accepted multicriteria J (1) as well.

CONCLUSIONSA new method is proposed for a real time decision

making and control of flame position and combus-tion process improvement. Inference measurements are developed to add the lack values on the base of the conventional measurements. A hybrid system for decision making and control, based on first principle model of the combustion process, image processing, and building blocks containing knowledge elements - Case Based Reasoning and Rule Base Reasoning is considered. Experimental data from a multiburner com-bustion process are used to tune mathematical model and to acquire relevant production rules. The proposed hybrid system contains promising potential to assist the operators with different level of expertise to adjust the combustion process.

REFERENCES

1. R. Hansok, J. Jefries, X. Zhou, H. Liu, A. Klingbeit, Sensors for Advanced Combustion Control, GCEP Technical Report, Part II.4.3, 2004.

2. K. Sujatha, M. Venmathi, N. Pappa, Flame Monitoring in Power Station Boiler Using Image Processing, ICTACT Journal on Image and Video Processing, 2, 4, 2012, 427-434.

3. M. Hadjiski, Control of indirect measurable and para-metrically changeable technological plants, Thesis of Dissertation of Doctor of Science, Sofia, 1979, (in Bulgarian).

4. M. Hadjiski, V. Totev, R. Yusupov, Softsensing Based Flame Position Estimation in Steam Boiler Combustion Chamber, Proceedings of the International Workshop on Distributed Computer and Communication Networks, Sofia, Bulgaria, Technosphera, Moscow, 2005, 243-254.

5. P. Kadlec, B. Gabrys, Data-Driven Soft-Sensors in Process Industry, Computers in Chemical Industry, 35, 4, 2009, 795-814.

6. M. Hadjiski, S. Dukovski, Adaptive Control of 210 MW Boiler Milling System – Comparative Analysis of Several Approaches, Invited Session DYCOMANS/IFAC Symposium on Adaptive Control/Signal Processing, Budapest, Hungary, 1995.

7. C. Houn, B. Begley, D. Lable, T. Kinney, A. Morrow, A. Speziale, Smart Firing Control System, Proceedinds of the 55th Annual ISA Power Industry Division Symposium, Austin, USA, 2012.

8. S. Pal, S. Shin, Foundation of Soft Case-Based Reasoning, John Wiley, 2004.

9. M. Hadjiski, V. Boishina, Hybrid Supervisory Control of Complex Systems Incorporating Case Based Reasoning, Proceedings of the International Conference on Complex Systems, COSY 2011, Ohrid, Macedonia, 16-20 September 2011, 179-186.

10. M. Hadjiski, V. Boishina, Enhancing Functionality of Complex Plant Hybrid Control System Using Case-Based Reasoning, Proc. of 5th IEEE International Conference on Intelligent Systems, IS’2010, 7-9 July 2010, London, UK, 2010, 25-30

11. A. Atanasov, K. Boshnakov, Case-Based Reasoning Control of Biological Wastewater Treatment, J. Chem. Technol. Metall., 49, 2014, 503-510.

12. R.K. Boyd, J.H. Kent, Three-dimensional furnace computer modelling, Proceedings of the 21-st International Symposium on Combustion, The Combustion Institute, Pittsburgh, 1986, 265-274.

13. Heat Calculation of Steam Boilers. Normative meth-od (3rd ed.), S. Petersburg, NPOCKTI-VTI, 1998.

Yordanka Marcheva

193


CURRENT OSCILLATIONS IN EPOXY-COATED STEEL AT ANODIC POLARIZATION

Yordanka Marcheva

Technical University - Sofia 8 Kl. Ohridsky bld , 1756 Sofia, Bulgaria, E-mail: [email protected]

ABSTRACT

The current oscillatory behavior of epoxy-coated low carbon steel at anodic polarization was investigated in 3,5 % and 0,5 % solutions of NaCl. It was established, that the coated steel may show current oscillations at potentiostatic anodic polarization. The oscillations are of different type, period and amplitude and are more clearly expressed in 0,5 % solution. The presented results might be used in the field of electrochemistry, electronics and electrical engineering as well as in the field of oscillatory systems study.

Keywords: current oscillations, oscillatory systems, epoxy-coated steel, corrosion.

Received 11 December 2014Accepted 30 January 2015

INTRODUCTION

Electrochemical systems are complicated dynamic systems, determined by plenty of non-linear mecha-nisms. Oscillations of the current or potential can be observed in many systems and the investigation of these complex dynamic behaviors today became a new branch of the modern electrochemistry.

The presence of oscillations in some electrochemi-cal systems can be explained by coupling of chemical reactions and transport processes in multicomponent systems. While chemical reactions are time-dependent, the transport processes are related to change in location. By coupling of chemical reactions and surface diffusion chemical waves can be developed, which can create geometrical patterns (motifs) on the electrode surface, to induce mechanical stresses and convection and create periodical oscillations in time and space [1].

Many years ago a guess was made about a phenom-enological resemblance between electrophysiological and electrochemical oscillators and it was assumed that these two classes of nonlinear systems have a ki-

netic analogy in their behavior [2]. A hypothesis was proposed, that the instability as a result of coupling between the chemical reaction and diffusion could have an important role in the first biogenetic step. Studying the electrochemical oscillation systems, their coupling and network formation it would be possible to create a model about neurophysiologic networks and, in future, artificial neurons for the needs of robotics.

The study of oscillators will serve for understanding of highly organized molecular systems such as living bodies, and for understanding of mechanisms of the elec-trochemical reactions themselves as well. Therefore, the investigation of the electrochemical oscillatory systems, i.e. knowing of the conditions and factors, related to their oscillatory behavior and the type of the oscillations, as well as the reasons for this behavior, is very important in the investigation of the electrochemical networks.

Oscillations in current and potential in the electro-chemical systems have been observed and studied for a long time [1]. Nowadays it is assumed, that they are due to the cycling forming and dissolution of the passivating layers (oxide, hydroxide or salt) or porous film. Oscil-


194

lations are observed on the iron and cobalt electrodes in water solutions, containing aggressive ions; on Zn electrode at cathodic and anodic polarization; stainless steel in sulfuric acid solutions with chromic acid [3]. The onset of complex oscillations is explained by the development of pitting corrosion on the passive iron surface; pure iron and steel undergo a destruction of the passive state in neutral and alcaline solutions of NaCl [4 - 7]. The current oscillations are accompanied by a cyclic increase and a decrease in the amount of SiO2 covering the Si surface.

Recently, the classification of the electrochemical oscillators was vastly expanded and it is no limited only to the passivity phenomenon. In this category are included also: potential- depended adsorbtion of inhibi-tors; potential- depended desorbtion by the addition of halogenides (CI-, Br-, I-) or halogenide-like ions (e.g. SCN-); electrostatic effect at low ionic strength of the solution; electrodissolution of metals; electrodeposition of alloys; oscillatory behaviour in the oxidation of or-ganic fuels; anodic oxidation of small organic molecules and hydrogen etc.

Electrochemical processes are based on the interac-tion of chemical reactions, flow of reactant and products, and electrical effects. Because of the complex nature of the interactions many industrial applications require characterization of the electrochemical properties in the space and time [8]. For example, in the fuel cells, the interaction of current-generating chemical reactions with the fuel flow and external load on the multi-electrodes can produce complex responses; a proton exchange membrane fuel cell exhibits larger power in the oscil-latory state than in the corresponding stationary state [8], etc.

Usually, the electronic equipment and components are packed with polymers in order to be protected against mechanical and chemical damages, as well as against pollutions. Moreover, some of them need to be protected against electrostatic discharge or electromag-netic influence. However, all polymers are permeable to moisture. In the most metallic systems, protected by organic coatings, the absolute performance depends not so much on the quality of the coating but more on the response of the coating to defects that are either initially present or that arise in the course of service. Hence, the coating’s properties will have a direct influence on the device behavior [9, 10].

Studying the protective action against corrosion of the epoxy-coated steel with different content of quartz particles [11], it has been observed, that this system has an oscillatory properties at some conditions. In previous work [12] the influence of quartz particles in the coating on the oscillatory behavior of the system was examined. It would be interesting to investigate the behavior of the epoxy-coating, without quartz, as it would have wider application in the technics. For example the epoxy-resins are used widely in the electrical engineering as bonding compounds and as isolating and anticorrosion coatings [13]. This study would be interesting also from the point of view of a new oscillatory system, as up to now data about oscillators conductor- dielectric was not reported.

This paper presents the results of investigation of the oscillatory behavior of epoxy-coated steel in NaCl solutions during anodic polarization.

EXPERIMENTAL

The composition, the treatment of the steel speci-mens and samples preparation are described in details in [11]. The average dry thickness of the epoxy coating is about 150 mm.

The electrochemical measurements were performed using a potentiostat Princeton Applied Research (PAR), model 263A using the SoftCorr program. A cell for flat specimens (PAR) with a conventional three electrode arrangement was used. Saturated calomel electrode was used as a reference electrode, platinum mesh - as a counter and coated steel - as working electrode with exposed area 1 cm2. All experiments were performed within the first few hours of immersion in a solution of NaCl with concentration of 3,5 % and 0,5 %. The po-tentiodynamic polarization curves with low sweep rate 1 mV/s were performed in order to give the possibility to any surface reaction to take place.

The polarizations curves are related to charge-transfer reaction of steel surface, exposed in coating. In the case of coatings the polarization curves are more complicated, compared to these ones for metals, because they contain data not only for exposed steel but also for the bulk electrolyte resistance and the ionic resistance in the coating pores and defects. This ohmic drop can mask the answer of the metal substrate. The major problem with the use of polarization curves to characterize coat-ings is that a moderate amount of damage is induced

Yordanka Marcheva

195

to the coating and coating/metal interface, especially in cathodic polarization, so this surface cannot be used for further study. There are only some cases for anodic polarization curves (which are our object) with polariza-tion in very small potential range [14].

As the task of this work is to study the instability and oscillatory behavior of the system and not to esti-mate corrosion behavior of the coatings, correction of this curves is not necessary and calculation of corrosion parameters will not be presented here. For the needs of present work a comparison of the form and stability of curves is more important and will be presented and discussed. Obtaining the results about the eventually destroying or growing the defects of the coating during anodic polarization would give important information concerning the behavior of the coating under constant electrical tension.

RESULTS AND DISCUSSIONS

Polarization curves Figs. 1 and 2 show the polarization curves for epoxy

coated steel in 3,5 % NaCl and 0,5 % NaCl, respectively, for different time of immersion in solution.

In 3,5 % NaCl the corrosion potential and current are changed considerably during one hour, the current grows about twice and the potential shifts in negative direction. The curves for one and three hours contact of the samples with the solution are almost the same. The anodic branch of the curves at Fig.1 has not any pecu-liarity and only for one hour contact with the solution has a fluctuation at 0,2 V.

Usually new epoxy-coated steel with no any defects and good barrier properties shows in the first days a

positive potential. In the studied system, the starting corrosion potential of the coated metal is about 300 mV more positive compared to this one of the pure steel and in the first 1 - 2 hours is getting close to the potential of the pure steel, in particular for the 3.5 % solution [12]. According to Leidheiser [15] the potential decrease in the beginning (i.e. the shifting of the potential in negative direction) of the epoxy-coated conventional low-carbon steel can be related to the entering of the moisture to the steel/coating boundary, followed by a chemical rear-rangement in the border metal/oxide [15]. The fact, that the curves for one and three hours contact in solution, are almost the same (Fig.1) shows, that the processes of water penetration in the coating and the corrosion of steel surface become stable in one hour. This statement is confirmed from the dependence of the corrosion po-tential on time [12].

The anodic dissolution of iron in concentrated solu-tions of NaCl and FeCl2 was studied in [4]. It was shown the formation of anodic film, induced by the precipita-tion of ferrous chloride on the electrode surface and the initiation of the film formation requires a supersaturation [4]. The iron corrosion is initiated by pits and spreads out in the form of corrosion product layers, one of which is poorly adherent. Concerning these immersion times, it was shown that the role of the occluded pits is negligible by comparison with the porous film effect. At the corro-sion potential two diffusion processes are detected: the diffusion across the layer and the mass transport in the liquid phase. The degree of coupling between the two processes is dependent on the immersion time and the rotation rate [6]. The oxygen transport is produced not only in the liquid phase but also through porous layer of the corrosion products. The oxygen consumption

Fig. 2. Potentiodynamic polarization curves for epoxy-coated steel after 15 min (1) and 1 h (2) contact in 0,5 % NaCl.

Fig. 1. Potentiodynamic polarization curves for epoxy-coated steel after 15.min (1), 1.h (2) and 3.h (3) contact in 3,5.% NaCl.


196

occurs not only by electrochemical reduction but also by chemical oxidation of ferrous to ferric ions [7].

In 0,5 % NaCl solution (Fig. 2) the corrosion po-tential stays the same and the corrosion current grows slowly (compared to 3,5 % NaCl) with time after 1 hour immersion in the solution. As the all other conditions are the same, such difference in the corrosion parameters for both concentrations might be related to the difference in the concentration of the two solutions and especially to the amount of the chloride ions and their influence on the corrosion parameters and behavior of the steel. The anodic polarization diagrams in 0,5 % solution clearly show fluctuations after an hour in the solution, which suggest oscillation behavior of the current at potentio-static conditions.

Potentiostatic measurements The potentiostatic measurements were conducted

at each 100 mV in anodic direction, starting from the corrosion potential, according to the polarization curve. Some of the results are presented on Figs. 3 and 4. It can be seen, that at some conditions (depending on the defects, pores, solution concentration, etc.) the system shows periodical current oscillations at potentiostatic conditions at anodic potentials.

Current oscillations are seen in 3.5% solutions even without indications on the polarization curves (Fig.3, at 0,2 V). In 0.5 % solutions (Fig. 4) in the anodic range of potentials, at potentiostatic conditions, the current shows periodic oscillations with the time. The oscillations are of mixed type and are different in form, amplitude and period. At a given potential they change the amplitude while at the others they are stable. As the potential shifts in positive direction, the form and the period of the oscil-

lations are changed, which indicates a strong dependence of the processes, provoking them, on the potential.

The instability on the anodic polarization curve and oscillations shows a presence of defects or pores in the coating, allowing the electrolyte to reach the steel surface. The appearance of the oscillations after an hour contact with the solutions shows the necessity of time for the processes, which provoke them.

It is known, that after the immersion of the coated metal in the solution, the water is the first one, which enter in the volume of the coating, before the ions of the electrolyte [16]. Therefore, at the very beginning of the corrosion process on the steel the chloride ions are absent. The presence of oscillation after an hour, their change with the potential growth, as well as the depend-ence on the solution concentration might be explained with the entering of chloride ions and a relation between the oscillations and the chloride ions content in the coat-ing. The stabilization of the oscillations by amplitude and decrease of the period with the potential increase shows a strong influence of the electrical field on the rate of migration of the negative chloride ions in the coating and their influence on processes on the steel surface. The change in the character of the oscillations at very positive potentials may be related also with additional processes of anodic oxidation not only of the metal surface, but also of the components of the electrolyte.

In support of the assumption about the relation between the oscillations and the concentration of the chloride ions probably comes the results, published in [17, 18]. The authors found a close relation between the concentration of the chloride iron on the electrode surface at passive conditions and the form, frequency

Fig. 3. Current oscillations at potentiostatic conditions for epoxy-coated steel in 3,5 % NaCl, E = 0,2 V (after 3 h contact in solution).

Yordanka Marcheva

197

and the amplitude of the oscillations. It is suggested, that the chloride ions do not enter in the passive film, without an availability of a sufficient driving force as a result of concentration gradient of the chloride ions on the surface. At a sufficiently high concentration gradient only a part of these ions enters in the passive film and produce destruction of the film [17, 18].

Li et al. [23] investigate the oscillatory electrodisso-lution of iron in neutral solutions of NaCl. They suppose a presence of inner non-pore layer and external pore salt film, formed on the electrode surface during the oscil-latory processes. The oscillations with high frequency

and small amplitude may be related to formation and dissolution of the inner layer and those with low fre-quency and large amplitude to the growth and removal of the outer porous salt film [23].

The oscillatory electrodissolution of the metal materi-als is a result of a complicated process, including surface reactions, film formation and dissolution and mass transfer.

In the case of organic coatings these reactions are getting more complicated because of difficulties in the diffusion of the reagents or products through the pores and defects in the coating and require further extensive research.

Fig. 4. Current oscillations at potentiostatic conditions for epoxy-coated steel in 0,5 % NaCl, A) E = -0,25V; B) E = -0.15V; C) E = 0.6V.


198

CONCLUSIONS

Epoxy-coated low carbon steel may exhibit current oscillations at potentiostatic anodic polarization in NaCl solutions. The oscillations are of different type, period and amplitude and are clearly expressed in the 0,5 % solution.

An important result in this research is the possibil-ity an epoxy coating to show an oscillatory behavior, which is very important for the needs of electrochemical, electronic and electrical technologies. The reasons for the oscillatory behavior are complex and require further extensive research.

Acknowledgements The author was sponsored by the Greek Foundation

I.K.Y. in receiving the presented experimental results. The author thanks to Prof. N. Kouloumbi and Dr. A. Karantonis from Department of Materials Science, National Technical University of Athens, Greece, for the collaboration.

REFERENCES

1. J. Jorne, Oscillations and concentration patterns in electrochemical systems, Electrochimica Acta, 28, 1983, 1713-1717.

2. M. Pagitsas, M. Pavlidou, D. Sazou, Localized pas-sivity breakdown of iron in chlorate- and perchlorate-containing sulphuric acid solutions: A study based on current oscillations and a point defect model, Electrochimica Acta, 53, 2008, 4784-4795.

3. K. Ogura, W. Lou, M. Nakayama, T. Fukume, Potential oscillations of a stainless steel electrode during galvanostatic polarization in a mixed so-lution of sulfuric and chromic acids, Journal of Electroanalytical Chemistry, 441, 1998, 191-195.

4. H. Kuo, D. Landolt, Galvanostatic transient study of anodic film formation on iron in concentrated chloride media, Corrosion Science, 16, 1976, 915-922.

5. M.G. Alvarez, J.R. Galvele, The mechanism of pit-ting of high purity iron in NaCl solutions, Corrosion Science, 24, 1984, 27-48.

6. D. You, N. Pebere, F. Dabosi, An investigation of the corrosion of pure iron by electrochemical tech-niques and in situ observations, Corrosion Science, 34, 1993, 5-15.

7. A. Bonnel, F. Dabosi, C. Deslouis, M. Dupart, M. Keddam, B. Tribollet, Corrosion Study of a Carbon Steel in Neutral Chloride Solutions by Impedance Techniques, J. Electrochem. Soc., 130, 1983, 753-761.

8. I. Z. Kiss, N. Munjal, R. S. Martin, Synchronized current oscillations of formic acid electro-oxidation in a microchip-based dual-electrode flow cell, Electrochimica Acta, 55, 2009, 395-403.

9. M. Despotopoulou, M. T Burchill, Coatings for electrochemical applications, Progress in Organic Coatings, 45, 2002, 119-126.

10. R.P. Frankenthal , J.D. Sinclair, Corrosion of electronic materials and devices, Encyclopedia of Materials: Science and Technology”, Elsevier Science Ltd, 2001, pp. 2638-2644.

11. N.Kouloumbi, L.G.Ghivalos, P.Pantazopouluo, Effect of quartz filler on epoxy coatings behavior, Journal of Materials Engineering and Performance (JMEPEG), 12, 2003, 135-140.

12. A. Karantonis, Y. Marcheva, L. G. Ghivalos, N. Kouloumbi, Static and Dynamic phenomena during the electrodissolution of steel in aqueous NaCl solu-tions, Electrochemistry, 74, 2006, 744-751.

13. A. Todorova, G. Dustabanov, Materials in Electrical Engineering, Technical University - Sofia, 2008, (in Bulgarian).

14. J. N. Murray, Electrochemical test methods for evaluating organic coatings on metals: an update. Part III: Multiple test parameter measurements, Progress in Organic coatings, 31, 1997, 375-391.

15. H. Leidheiser, Electrochemical methods for apprais-ing corrosion protective coatings, J. Coat. Technol., 63, 1991, 21-31.

16. А. Amirudin, D. Thierry, Application of electro-chemical impedance spectroscopy to study the degradation of polymer-coated metals, Progress in Organic Coatings, 26, 1995, 1-28.

17. L. Li, C. Wang, S. Chen, X. Yang, Investigation of oscillatory electrodissolution processes by chlo-ride ion perturbation, Proc. of 16th International Corrosion Congress, 2005, Beijing, China.

18. L. Li, J. L. Luo, B. T. Lu, S. H. Chen, Effect of interface chloride ion perturbation on oscillatory electrodissolution, Electrochimica Acta, 50, 2005, 3524-3535.

Katya Ignatova, Daniela Lilova

199


A STUDY ON THE KINETICS OF THE ELECTRODEPOSITION OF Ni, Co AND Ni-Co ALLOY IN CITRATE ELECTROLYTE

PART 1. THE KINETIC STUDY OF THE INDEPENDENT ELECTRODEPOSITION OF Ni AND Co


Department of Inorganic and Electrochemical ManufacturingUniversity of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected]

ABSTRACT

The elecrodeposition of Ni and Co was studied by electrochemical techniques in varying compositions of slightly acidic citrate electrolyte. Cyclic voltammetry and current transient measurements were used to characterize the initial stages of nucleation, growth mechanism and electrodeposition of Ni and Co. The voltammetry analysis showed that the electrodeposition of both Ni and Co occurs under diffusion-kinetic control associated with a typical nucleation process. The deposition of Ni is in one act of charge transfer with exchange of two electrons, the deposition of Co is a two-stage process of reduction of free cobalt ions to cobalt, and reduction of complex ions type CoCit- to cobalt. The application of the method of chronoammetry proved that for cobalt the mechanism of nucleation is spontaneous, with three-dimensional nucleation and nuclei growth, while for nickel the slower progressive mechanism of nucleation is typical. The differences in the deposition mechanisms of the two metals is a possible explanation for their anomalous co-deposition in the elec-trolyte examined.

Keywords: electrodeposition kinetic, initial stages of nucleation, cyclic chronovoltammetry, chronoamperometry, nuclei growth.


INTRODUCTION

Over the past two decades there has been a growing interest in the applications of electrodeposited coat-ings of metals from the iron group (Ni, Co, and Fe) and their alloys for their unique magnetic, physical, mechanical,and thermophysical properties[1, 2].

The electrodeposition of Ni-Co alloys is classified as anomalous co-deposition which means that the less noble metal (Co) is preferentially deposited [10]. The disclosure of the reasons for anomalous co-deposition of Ni and Co in different electrolytes in many cases demands a combined application of several independent electrochemical methods and their diagnostic criteria [3, 5 - 10].

The electrochemical methods for preparation of metallic coatings and especially of the alloys are very attractive because of the high degree of control over the properties of the coatings while varying the experimental conditions [1 - 4]. To achieve this, a detailed knowledge on the mechanism of deposition of the metals both in-dividually and in alloy is needed.

There are not enough data for the kinetics of inde-pendent deposition of Ni and Co in slightly acidic citrate electrolyte. The deposition kinetics of nanostructured Ni coating in sulphate electrolyte in presence of various organic additives was studied [5, 6] as well as in the typical Watts-bath [2, 7]. There are data for the kinetics of electrodeposition of Co in sulphate solutions [8, 9]. While the cyclic chronovoltammetry can give anotion


200

about the overall mechanism and kinetics of the occur-ring electrode processes [12], the chronoamperometric method [11] provides information about the initial stages of nucleation and the rate of nuclei growth, which can be used when searching the reasons for the anomalous co-deposition of the two metals in an alloy [5]. In this method the potential changes gradually from the equi-librium potential to potentials at which the metal begins depositing at a noticeable speed. Under these conditions, it is possible to compare the characteristics current-time (I - t) for potentials at which the main reaction does not occur, with potentials at which three-dimensional nuclei are formed and grow on an extrinsic substrate. The nucleation of the metal phase and the increase of the number of nuclei is usually illustrated with a few maxi-mums in the I - t curves. The characteristic maximums indicate the mechanism of nucleation and the detention or the slow increase of the current up to a certain value suggest a three-dimensional growth of the nuclei [11].

The present study is aimed to obtaining some data about the general mechanism of independent deposition of Ni and Со in slightly acidic citrate electrolyte, as well as to obtain information about the initial stages of inde-pendent deposition of Ni and Co through application of the methods of cyclic voltammetry, chronoamperometry and chronopotentiometry.

EXPERIMENTAL

All electrochemical experiments were carried out in a thermostatic three-electrode cell without stirring. The working electrode was a copper disc with surface area 1 cm2 and the platinum plate anode was about 30 times larger. All potentials were referred to the saturated calomel electrode (SCE). The solutions were prepared solving reagents of chemically pure substances in double-distilled water. The composition of the work-

ing electrolytes is shown in Table 1.The slightly acidic citrate electrolytes were brought to рН = 5.5 through addition of relevant quantities of sodium hydroxide and citric acid and all experiments were carried out at temperature of 20°С.

All experiments were carried out using computer controlled potentioscan type Wenking (Germany). The cyclic voltammograms were recorded in potential area from 0.2 V to -1.9 V (SCE) at scanning rate from 30 to 120 mV s-1.The potentiostatic I - t dependencies and the galvanostatic E - t dependencies were recorded by rate equal to 1 s cm-1 with respect to time.


The method of cyclic voltammetry is used to define the area of potentials for reduction and to characterize the processes of deposition of Ni and Co. The characteristic voltammograms for deposition of Ni and Co recorded with scanning rates from 30 to 120 mV s-1 in solutions 3 and 6, i.e. with concentration of Ni and Co 0,3 gmol dm-3 are shown in Fig. 1 a, b.

There is a long initial polarization observed in the curves for independent deposition of Ni (Fig.1a) fol-lowed by a plateau of boundary current for Со (Fig.1b), whose value increases with the increase of the rate of potential scanning. With the increase of the rate of scanning, the plateau decreases and almost disappears at the highest value (Fig. 1a, curve 4).The kinetic of deposition of Ni corresponds to a combined control with predominantly activation nature of the polarization. The absence of anode branch is the proof for this finding. The plateau of boundary current is due to reaching the diffusion limitations at higher cathode potentials.

This conclusion corresponds to the data for the in-fluence of the concentration of Ni on the course of the cathode polarization curves (Fig. 2а) from which it fol-

Table 1. Composition of the working electrolytes in gmol dm-3.

Solution №

oCo2C + o

Ni2C + H3BO3 Na3citrat

1, 2 , 3 - resp.0,1; 0,2; 0,3

0,485 0,3

4, 5, 6 resp.0,1; 0,2;0,3

- 0,485 0,3


201

lows that the plateau of boundary current is more typical for the lower concentrations of Ni where obviously the diffusion limitations are achieved sooner.

The deposition curves of Со (Fig.1b) unlike those of Ni (Fig. 1a), have a pronounced cathode peak which increases with the increasing of potential scanning rate and the concentration of cobalt in the solution (Fig. 2b). The value of the peak current shifts to more cathodic potentials with the increase of the rate of scanning. There is not an identified oxidation peak in the working window of potentials when the curves turn scanning in anode direction. The appearance of the cyclic dependen-cies for deposition of Co evidences for the predominant diffusion limitations during its actual deposition (from -1.18 V to 1.30 V). This fact is supported by the observa-tion that the observed peak increases with the increase of the concentration of Со (Fig. 2b). There is a retention of the current observed in the initial area of potentials of the deposition curves of cobalt, but as our subsequent observations indicate, there is no formation of cobalt

Fig. 1. Cyclic voltammograms at different rates of potential scanning in electrolytes 3, with 0,3 gmol dm-3 Ni (а) and in electrolyte 6, with 0,3 gmol dm-3 Co (b).

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

4

32

1

0,3M Ni

i c, m

A cm

-2

E, V (SCE)

(1) 30 mV s-1; (2) 60 mV s-1;(3) 90 mV s-1; (4) 120 mV s-1

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

-------

432

1

0,3M Co

i c, m

A cm

-2

E,V (SCE)

(1) 30 mV s-1

(2) 60 mV s-1

(3) 90 mV s-1

(4)120 mV s-1

(a)

(b)

Fig. 2. Influence of the concentrations of Ni, solutions 1-3 (a) and of Co, solutions 4-6 (b) on the potentiodynamic polarization dependencies (at v=30 mV s-1).

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

-------

3

2

1

i c, m

A cm

-2

E,V (SCE)

(1) 0,1M Ni(2) 0,2M Ni(3) 0,3M Ni

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

-------

3

2

1

v= 30 mV s-1

i c, m

A cm

-2

E,V (SCE)

(1) 0,1M Co(2) 0,2M Co(3) 0,3M Co

(a)

(b)

coating within this area of potentials. The chronovoltammetric examination also showed

that the intersection of the cathode and anode branches of the curves is observed on the cathode sections that correspond to the potentials of nucleation: this is the area before the observed peaks, respectively plateaus for Ni. The voltammograms are analized based on Nicholson-Shain dependencies [12] and the developed diagnostic criteria of the method [13]. In Fig. 3а-d the dependencies 1/2

p v-i (Fig. 3a), ip/(v1/2)– v (Fig. 3b),

(Ep–Ep/2) – v1/2 (Fig. 3с), and Ep-lgv (Fig. 3d) obtained from the chronovoltammograms are shown that satisfy the following equations [5, 12]:

1/ 21/2 1/2

p onFi -0,496nFC D v

RTα∞ = (1)

p

2,3RTE K- logv2 nFα

= (2)

The linear character of dependencies ip – v1/2 (Fig. 3a),


202

and particularly the increase of the current in the peak with v1/2, evidence that the deposition of Ni, Сo is a diffusion controlled process. The application of the following diagnostic criteria specifies the degree of diffusion control and the interference on other stages in each case. Thus the practically weak dependence of ip/(v1/2) on v (Fig. 3b) in the deposition of Ni, as well as the absence of anode branch in the cyclic voltammograms for Ni (Fig. 1a) and the shift of the potential in the peak/semi-peak in cathode direction (Fig. 3с, curve 1) prove the quasi-reversible nature (diffusion-activation control) of the deposition of Ni [13]. The dependencies in Fig. 3c also indicate that the control for deposition of Co is different from those of Ni. It corresponds to diffusion-controlled process with interference on the chemical stage preceding the reaction of discharge [13] that could possibly be the dissociation of the adsorbed citrate complexes of Со. The shift of the potential in the peak towards more negative values (Fig. 3с, curve 2) is a proof supporting the above suggestion.

In order to further clarify the mechanism of inde-

pendent deposition and co-deposition of Ni and Co, the chronoammetric analysis was applied [11]. The typical time-current dependencies in compositions 3 and 6 in potentiostatic conditions are shown in Fig. 4a,b in a wide range of potentials applied from –0.850 V to -1.35 V. The same figure shows the processing of the curves in I – t-1/2 coordinates (Fig. 4a*,b*) and inI - t -3/2 coordinates (Fig. 5a,b) and this serves for the purpose of proving the growth mechanism of the nuclei formed, i.e. spontane-ous in the first case and progressive in the second one.

It was found that there is neither deposition of the two metals nor any reactions associated with the background electrolyte at potentials more positive than -0.600 V and only charging the capacity of the double electric layer occurs. There is a well defined and rec-ognizable maximum of current observed in the course of the chronoammograms recorded in electrolyte 1 for deposition of Ni (Fig. 4a), which is followed by an abrupt drop and subsequent retention, as the current follows the Cottrell equation that is valid in the case of diffusion control.

Fig. 3. Processing of the cyclic voltammograms and application of the diagnostic criteria of the method for Ni and Co.

5 6 7 8 9 10 118

16

24

32

402

1

i p, m

A cm

-2

v1/2

1 - 0,3M Ni2 - 0,3M Co

(а)

20 40 60 80 100 120

2

3

4

5

2

1

i p / v

1/2

v, mV s-1

1 - 0,3M Ni2 - 0,3M Co

(b)

5 6 7 8 9 10 110,08

0,12

0,16

0,20

0,24 2

1

E-E p/

2, V

v1/2

1 - 0,3M Ni2 - 0,3M Co

(c)

1,4 1,5 1,6 1,7 1,8 1,9 2,0 2,1

1,3

1,4

1,5

1,6

2

1E p,

V

log(v)

1 - 0,3M Ni2 - 0,3M Co

(d)


203

It can be said that this course proves the reduction of nickel ions occurring in one stage with transfer of two electrons under the reactions:

2Ni 2e Ni+ −+ ↔

3NiCit 2e Ni Cit− − −+ ↔ + The deposition of Ni initiates at potentials above

-0.950 V with the simultaneous evolution of hydrogen upon the reaction:

2 22H O 2e H 2OH− −+ ↔ + which are evidenced by the time-current dependencies recorded additionally only in the background electrolyte. The course of the chronoammograms recorded in elec-trolyte 6 for deposition of Со (Fig. 6b) is quite typical. In the case of Co, there are two subsequent peaks of the current in the potential area from -1.0 V to 1.2 V, instead of one peak. The first one, i.e. for smaller values of time, is sharper and the decrease of current follows the Cot-trell’s equation [14]. The second peak in the curves of Co evidences for a process of occurrence of both active nucleation and nuclei growth in conditions of diffusion limitations. The dependencies in Fig. 6b indicate that the deposition of Co occurs through a two-stage electron transfer as it has already been found for other electrolytes [8].The first peak corresponds to reduction of the nearest free cobalt ions under the reaction:

2adsCo 1e Co+ − ++ ↔

which is followed by a fast process of reduction of

the monovalent ions to cobalt:

adsCo 1e Co+ −+ ↔

As a result of depletion of the electroactive form

(Co2+) in the layer near to the electrode, the current decreases with time. The process is entirely diffusion controlled. With time the diffusing complex ions of cobalt (CoCit-) reach slower the electrode. Their de-layed reduction is also associated with purely chemical processes that precede the reduction itself, and namely dissociation of the adsorbed complexes, for which the criteria of cyclic chronovoltammetry applied above indicated. Upon reaching higher cathode values of po-tential, the reduction of the citrate complexes to cobalt takes place:

3CoCit 2e Co Cit− − −+ ↔ +

The decrease of current with time as a result of depletion of the complex ions of cobalt is smoother and is explained with the simultaneous growth mechanism of the nuclei formed. The suggestion that complex ions of Ni and Co are involved is supported by the previous investigations proving their shape and stability [15] that are also comparable with those in [16, 17]. It was found that in slightly acidic citrate electrolyte (рН = 5.5), nickel and cobalt present both as free metal ions (Ni2+and Co2+) and complex citrate ions mainly of the type NiCit- and CoCit- mainly with stability constants lgKst(Ni) = 5.379 and lgKst(Co) = 4.38.

The general conclusion suggested by the chronoam-metric analysis is that the behavior of the curves in Fig. 5(а, b) evidences for a typical process of nucleation with three-dimensional growth of the nuclei at predominantly diffusion control on the part of the electroactive forms [11]. The critical time for nucleation, i.e. the time in which the current decreases, decreases with the increase of the cathode potential applied.

The method of chronoammetry provides clear diag-nostic criteria to recognize the mechanism of nucleation. According to the theory developed by Scharifker-Hills [11], there are two boundary cases of nucleation with a diffusion controlled growth. At a high rate of nucleation all nuclei are formed spontaneously and the dependence time-current is described by:

[ ]1/ 2

1/ 2( ) . 1 exp( . )( )zFD ci t N kDt

tπ

π

= − −

(3)

where k = 8πCo∞ M/ρ; D is diffusion coefficient; Co

∞ is volumetric concentration of Ni2+ or Co2+; М is mollar mass; ρ is density of the metal; No is the number of spots for nucleation. In this case the chronoammograms are linearized in coordinates I – t -1/2.

In the second case. i.e. at lower rate of nucleation, when the nuclei are formed continuously within the entire time window before reaching the area of overlap-ping with the diffusion limitation around the growing nuclei, we can speak about progressive (continuous) nucleation. In this case, the chronoammograms are lin-earized in coordinates I - t -3/2. The course of the current for progressive three-dimensional growth mechanism is described by the following equation:

1/ 22

1/ 2( ) . 1- exp(- '. / 2)( ) ózFD ci t a N k Dt

tπ

π

=

(4)


204

where k` = 4/3(8πCo∞M/ρ)1/2.

The chronoammograms of Ni are linearized to a higher degree in coordinates I - t -3/2 (Fig. 5a), which

is indicative for a mechanism of progressive (continu-ous) nucleation. Unlike Ni, in the case of Co we can observe a better linearization of the chronoammograms

Fig. 4. Chronoammetric dependencies (a, b) and their derivative dependencies ic - t -1/2

(a*, b*) in solutions 3 for Ni (a, a*), 6 for Co (b, b*).

0 5 100

4

8

12

16

20

24

0,3M Ni543

21

i c, m

A cm

-2

t,s

(1) -0,90V;(2) -1,05V;(3) -1,20V;(4) -1,30V;(5) -1,35V

0,4 0,6 0,8 1,0 1,2 1,40

4

8

12

16

20 5

4

3

2

1

0,3M Nii c, m

A cm

-2

t-1/2

(1) -0,9V; (2) -1,05V; (3) -1,2V; (4) -1,30V;(5) -1,35V

(a) (a*)

0 5 100

10

20

30

40

50

0,3M Co

6543

21

i c, m

A cm

-2

t,sec.

(1) -0,85V; (2) -0,91V; (3) -1,01V; (4) -1,10V; (5) -1,20V; (6) -1,35V

0,4 0,6 0,8 1,0 1,2 1,40

10

20

30

40

50 5

4

3

21

0,3M Coi c,

mA

cm-2

t-1/2

(1) -0,91V; (2) -1,01V;(3) -1,10V; (4) -1,20V;(5) -1,35V

(b) (b*)

Fig. 5. Dependencies ic - t -1/2 (a*-c*), (a) - in solutions 3 (0,3M Ni), (b) - in solution 6

(0,3M Co).

0,4 0,6 0,8 1,0 1,2 1,40

10

20

30

40

50 5

4

3

21

0,3M Co

i c, m

A cm

-2

t-1/2

(1) -0,91V; (2) -1,01V;(3) -1,10V; (4) -1,20V;(5) -1,35V

0,0 0,5 1,0 1,5 2,0 2,50

10

20

30

40

505

4

3

2

1

0,3M Co

i c, m

A cm

-2

t-3/2

(1) -1,01V; (2) -1,1V;(3) -1,2V; (4) -1,35V

(a) (b)


205

in coordinates I - t -1/2 (Fig. 4b*), which evidences for a mechanism of spontaneous nucleation that is super imposed onto the general diffusion controlled process of deposition of Co.

It is possible that the mentioned difference in the mechanisms of nuclei growth is due to the two-stage transfer of charge in the case of Co since the first stage has already localized the needed spots for crystal growth on the surface. The higher speed of nucleation in the case of Co through the mechanism of spontaneous nucleation can be a possible reason for its identified preferential (anomalous) deposition together with Ni. It also follows from the curves in Figs. 4 and 5 that with the increase of the potential applied the slope of dependencies I - t -1/2

and I - t -3/2 increases for both metals, which corresponds to the increase of the rate of nucleation.

The determination of the transition time as a function of current can provide additional information about the mechanism of interaction [14]. The course of the cathode potentiograms for Ni and Со is shown in Fig. 6а,b for solutions 3 (0,3M Ni) and 6 (0,3M Co), correspondingly.

The curves in Fig. 6а indicate that the cathode potentials for deposition of Ni shift smoothly to more negative potentials. In the case of Co (Fig. 6b), in the intermediate area of cathode potentials applied the dependencies Е - t follow the equation of Karauglanov [14], which means that the transition time is achieved decreasing with the increase of the current density. Thus the conclusion is proved that although being combined, the control of the process of Co deposition is diffusion

to a higher extent than that of Ni. It is also proved that the mechanism of nucleation and nuclei growth of Co is faster than that of Ni.

CONCLUSIONS

The comparision of the dates from three electroana-lytical methods shows the difference of the deposition kinetics and of the mechanisms of nucleation and crystal growth for Ni and Co in citrate electrolyte. The deposition of Ni occurs in one act of charge transfer with exchange of two electrons, and in the case of Co the deposition is a two-stage process of reduction of free cobalt ions to cobalt, and reduction of complex ions type CoCit- to cobalt. In the case of Co, there is a mecha-nism of spontaneous nucleation and three-dimensional growth, while in the case of Ni it comes to slower process of progressive (continuous) nucleation. The differences in the deposition mechanisms of the two metals is a pos-sible explanation for their anomalous co-deposition in the electrolyte examined.

REFERENCES

1. J. Vazquez-Arenas, T. Treeratanaphitak, M. Pritzker, Formation of Co-Ni alloy coating under direct cur-rent, pulse current and pulse-reverse plating condi-tions, Electrochimica Acta, 62, 2012, 63-72.

2. R. Orinakova, A. Turonova, D. Kladekova, M. Galova, R. Smith, Recent developments in the electrodeposi-

0,4 0,6 0,8 1,0 1,2 1,40

10

20

30

40

50 5

4

3

21

0,3M Co

i c, m

A cm

-2

t-1/2

(1) -0,91V; (2) -1,01V;(3) -1,10V; (4) -1,20V;(5) -1,35V

0,0 0,5 1,0 1,5 2,0 2,50

10

20

30

40

505

4

3

2

1

0,3M Co

i c, m

A cm

-2

t-3/2

(1) -1,01V; (2) -1,1V;(3) -1,2V; (4) -1,35V

(a) (b)

Fig. 6. Chronopotentiograms in electrolytes 3 (a) and 6 (b) at different current densities.


206

tion of Ni and some Ni based alloys, J.Appl.Electro-chem., 36, 2006, 957-972.

3. E. Gomez, E. Valles, Electrodeposition of Ni-Co alloys on modified silicon substrates, J. Appl. Elec-trochemistry, 29, 1999, 805-812.

4. K.Y. Sasaki, B. Talbot, Electrodeposition of iron-group metals and binary alloys from sulfate baths, J. Elecrochem. Soc., 145, 3, 1998, 981-990.

5. M. Torabi, A. Dolati, A kinetic study on the electro-deposition of nickel nanostructure and its electrocata-lytic activity for hydrogen evolution reaction, J. Appl.Electrochem., 40, 2010, 1941-1947.

6. El-Feky Hesham, Hegem Mosaad, Roy Sudipta, H.Nadia, A.Baraka, Electrodeposition of Ni and NiCo alloys using cistein and conventional ultrasound waves, Science China Chemistry, 59, 10, 2013, 1446-1454.

7. S.H. Goods, J.J. Kelly, N.Y.K. Yang, Electrodeposited nickel-manganese, an alloy for microsystem applica-tions, Microsystem Technologies, 10, 2004, 498-505.

8. E. Chаssaing, Effect of organic additive on the elec-trocrystallization and the magnetoresistance of Cu-Co multilayers, J.Electrochem.Soc., 148, 10, 2001, C690-C694.

9. Min Gu, Initial stages of the electrocrystallization of Co-Cu alloys on GCE from the Co rich electrolytes, Electrochimica Acta, 52, 2007, 4443-4448.

10. S. Armyanov, Crystallographic structure and mag-netic properties of electrodeposited Co and Co alloys, Electrochim. Acta, 45, 2000, 3323-3335.

11. B. Scharifker, Gr. Hills, Theoretical and experimental studies of multiple nucleation, Electrochimica Acta, 28, 7, 1983, 879-889.

12. R.S. Nicholson, I. Shain, Theory of Stationary Elec-trode Polarography: Single Scan and Cyclic Methods Applied to Reversible, Irreversible and Kinetic Sys-tems, Analytical Chemistry, 36, 4, 1964, 706-723.

13. E.R. Brown, R.F. Large, CyclicVoltammetry, AC polarography and related techniques, 5th, Wiley, New York, 1971.

14. Z. Galus, Fundamentals of Electrochemical analysis, Ed. Elis Horwood, 1994, 606.

15. K. Ignatova, Effect of H3BO3 on complex formation and kinetics of electrodeposition of Ni-Co alloy in citrate electrolyte, Bulgarian Chem. Communica-tion, 2015, in press.

16. O.Yu. Zelenin, Interaction of the Ni2+ ion with citric acid in an aqueous solution, Russian Journal of Co-ordination Chem., 33, 5, 2007, 346-350, (in Russian).

17. N. Кotsakis, C.P. Raptopoulou, V. Tangoulis, A. Terzis, J. Giapintzakis, T. Jakusch, Correlation of synthetic, spectroscopic, structural and speciation studies in the biological relevant Co(II)- citrate system, Inor-ganic Chemistry, 42, 2003, 22-31.


207


KINETIC STUDY ON ELECTRODEPOSITED Ni-Co ALLOY IN CITRATE ELECTROLYTE

PART 2. THE KINETICS OF JOINT ELECTRODEPOSITION OF Ni AND Co IN Ni-Co ALLOY SYSTEM


Department of Inorganic and Electrochemical ManufacturingUniversity of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected]

ABSTRACT

In the paper the results from the study of the kinetic of joint electrodeposition of Ni and Co in Ni-Co alloy system are reported. It is found that the co-deposition of Ni and Co in Ni-Co alloy system in slightly acidic citrate electrolyte occurs under combined diffusion-kinetic control with a typical mechanism of nucleation similar to those in the case of Co. The application of the method of chronoammetry proved that the mechanism of nucleation for Ni-Co alloy is spontaneous with three-dimensional nucleation and nuclei growth. The difference in the independent deposition mechanisms of the two metals is a possible explanation for their anomalous co-deposition in the electrolyte examined. The anomalous co-deposition is more clearly noticeable in electrolytes with higher content of cobalt than nickel. The SEM-analysis proves that the alloys enriched in cobalt have crystallites with more even and rounded shape, which is expected in the case of spontaneous nucleation.

Keywords: electrodeposition mechanism, alloy system, cyclic chronovoltammetry, chronoammetry, nucleation, nuclei growth.


INTRODUCTION

The application of thin Ni-Co alloy films can be divided in three main categories: for decorative coatings purposes, for anti-corrosion protection in the industry [1, 2], and for highly functional coatings in various magnetic devices, especially in microsystems such as sensors, actuators, and memory devices [3, 4].

The electrodeposition of Ni-Co alloys is classified as anomalous co-deposition which means that the less noble metal (Co) is preferentially deposited [5 - 10]. This phenomenon is observed in sulphate [5, 10], sulfamate and chloride [6, 9], and citrateelectrolytes [4], both with application of stationary current modes [5, 9, 10], and various pulse modes [6 - 8]. To explain this phenomenon, different reasons are found and reported, e.g. increased рН close to the electrode surface with formation of co-balt- and nickel hydroxides on the surface [8]; two-stage

mechanism including adsorption of monovalent inter-mediate compounds and their reduction; underpotential deposition [9], and faster deposition kinetic of cobalt under diffusion control [10]. The most frequently cited model for anomalous co-deposition is based on the idea that the adsorbed forms of these metals include metallic mono-hydroxides Me(OH)ads

+ obtained as intermediate product during the deposition. Under this mechanism, the anomalously high content of cobalt in the alloyed Ni-Co coating results from the preferential adsorption of +

adsCo(OH) to +adsNi(OH) on the cathode surface [8].

This conclusion is based on the comparison between the equilibrium constants for formation of the metal hydroxide ions +Me(OH) . The decrease of this value is in direction >+Fe(OH) >+Co(OH) +Ni(OH) . On the other hand, the comparison between the thermodynamic data for these constant indicates that the difference be-tween their values is not large:


208

4,1logK4,3;logK4,5;logK NiOHCoOHFeOH === +++ [11]. The disclosure of the reasons for anomalous co-deposition of Ni and Co in different electrolytes in many cases demands a combined application of several independent electrochemical methods and their diagnos-tic criteria [5, 12 - 15].

The present study is aimed to obtaining some data about the mechanism of co-deposition of Ni and Со in slightly acidic citrate electrolyte through application of the method of cyclic chronovoltammetry, as well as to obtain information about the rate of nucleation on the initial stages of co-deposition of Ni and Co in Ni-Co alloy system through application of the methods of chronoammetry and chronopotentiometry. Our further goal is to clarify the specifics of the deposition of Ni and Co combining the data obtained with the results from the study on the composition and morphology of Ni-Co alloy with varying the content of cobalt and nickel in the electrolyte.

EXPERIMENTAL

All electrochemical experiments were carried out in a thermostatic three-electrode cell without stirring. The working electrode used was a copper disc with surface area 1 cm2, and the anode was a platinum plate with surface area almost 30 times larger than that of the cath-ode. All potentials were measured against a reference saturated calomel electrode (SCE). In order to compare the behavior of the two metals during electrodeposition with that of their co-deposition in Ni-Co alloy system, the solutions were prepared with different ratios Ni/Co (Table 1). The slightly acidic citrate electrolytes were brought to рН = 5.5 through addition of relevant quantities of sodium hydroxide and citric acid and all experiments were carried at temperature of 200С. The solutions were prepared solving reagents of chemically pure substances in double-distilled water.

All experiments were carried out using computer

controlled potentioscan type Wenking (Germany). The cyclic voltammograms were recorded in potential area from 0.2 V to -1.9 V (SCE) at scanning rate from 30 to 120 mV s-1. The potentiostatic I - t dependencies and the galvanostatic E - t dependencies were recorded at rate of potential scanning equal to 1 s cm-1 with respect to time. The examination of the morphology of Ni-Co alloys, as well as their composition (through EDSA analysis) was carried out using scanning electron microscope (SEM) with equipment of Oxford Instruments. JSM-6390-Jeol. The composition of the alloys was also examined using additional atomic-absorption analysis (ААА).


The characteristic voltammograms for deposition of Ni-Co recorded with scanning rates from 30 to 120 mV s-1 in solution 1 (Table 1) are shown in Fig. 1a and in Fig. 1b - the polarization dependencies at v = 30 mV s-1 in solutions with different ratios Ni/Co (Table 1).

In the deposition curves of Ni-Co alloy system (Fig. 1a) have a pronounced cathode peak which increases with the increase of potential scanning and the concentration of cobalt in the solution (Fig. 1b). The value of the peak current shifts to more cathode-like potentials with the increase of the rate of scanning. There is not an identified oxidation peak in the working window of potentials where the curves turn scanning in anode direction. The appear-ance of the cyclic dependencies evidences for the predomi-nant diffusion limitations during its actual deposition of Ni-Co (from -1.2 V to 1.35 V). This fact is supported by the observation that the observed peak increases with the increase of the concentration of Со (Fig. 1b).

The deposition potentials of Ni-Co alloy system are shifted in a strongly negative direction in comparision with that for Co (-1.18 V for Co and -1,4 V for Ni-Co), which evidences for additional difficulties in the pro-cesses of nucleation and crystal growth during the co-

Table 1. Composition of the working electrolytes in gmol dm-3. № of Solution (рН = 5,5)

oCo2C + o

Ni2C + H3BO3 Na3citrat oCo

oNi 22 /CC ++

1 0,3 0,3 0,485 0,3 1,0

2 0,2 0,3 0,485 0,3 1,5

3 0,1 0,3 0,485 0,3 3,0


209

deposition of the two metals. The chronovoltammetric examination also showed that the intersection of the cathode and anode branches of the curves is observed on the cathode sections that correspond to the potentials of nucleation. Fig. 2 a,b exhibits the comparison between the curves of independent deposition of Со and Ni and their co-deposition at two ratios in the solution (solu-tions 1 and 3 from Table 1), which indicates more clearly that in the total curve the total current is higher than the currents corresponding to the individual deposition of each metal.

Moreover, it is seen that in composition 1 (Fig. 2a), in which the content of the two metals is equal ((Ni/Co)sol = 1), the discharge of Co-ions occurs at more positive potentials compared to that of Ni. The so called anoma-lous deposition of the two metals is observed. With the increase of the relative content of Ni in the solution compared to that of Co (Fig. 2b, composition 3), the effect of the preferential deposition of Co obviously weakens thus creating conditions for enrichment of the Ni alloy. It can also be seen that with the increase of the relative content of Ni, the peak of reduction of the Ni-Co alloy shifts to more positive potentials (Fig. 2b) and thus the potential of nucleation decreases with the

increase of the ratio Ni/Co.The proven mechanism of anomalous deposition of

the two metals in compositions 1 and 3 is also confirmed by the identified influence of the potential and the ratio Ni/Co in the solution on the chemical content of Ni-Co alloy system, which is found through atomic absorption analysis (Fig. 3). It is seen that even when the relative content of Ni in the solution is three times higher (Fig. 3, curve 1) the content of Со in Ni-Co alloys is no less than 57 – 70 % mass, depending on the potential. The content of cobalt increases to 90 % mass when its content in the solution increases (Fig. 3, curve 3) and decreases to 78 % mass with the increase of the potential applied.

The voltammograms are analized based on Ni-cholson-Shain dependencies [17] and the developed diagnostic criteria of the method [16]. In Fig.4 а-d the dependencies 1/2

p v-i (Fig. 4a); vv/ip − (Fig. 4b), p p/2(E -E ) v− (Fig. 4с) и logv-E p (Fig. 4d). The linear character of dependencies ip – v1/2

(Fig. 4a), and particularly the increase of the current in the peak with v1/2, evidence that the deposition of Ni-Co system is a diffusion controlled process. According to the diagnostic criteria applied, the deposition kinetic of Ni-Co alloy is controlled to a greater extent by the

Fig. 1. Cyclic voltammograms at different rates of potential scanning in: (а) electrolytes 1 and (b) in electrolytes 1-3 (Table 1) at v = 30 mV s-1.

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

- - - - - - -

1

43

21

(Ni/Co)sol.=1CNi2+=0,3 Mi c,

mA

cm-2

E,V (SCE)

(1) 30 mV s-1

(2) 60 mV s-1

(3) 90 mV s-1

(4)120 mV s-1

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

- - - - - - -

CNi2+=0,3M3

2

1

i c, m

A cm

-2

-E,V (SCE)

Ni/Co in sol.:(1) - 3,0(2) - 1,5(3) - 1,0

(а)

(b)

Fig. 2. Comparison between the curves of independent deposition of Co и Ni and the curves for their co-deposition in compositions 1(а) and 3 (b) from Table 1.

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

60

-------

3

2

1

i c, m

A cm

-2

E,V (SCE)

(1) 0,3M Ni(2) 0,3M Co(3) (Ni/Co)sol.=1

0,6 0,8 1,0 1,2 1,4 1,6 1,80

10

20

30

40

50

3

2

1

i c, m

A cm

-2

E,V (SCE)

(1) 0,3M Ni(2) 0,1M Co(3) (Ni/Co)sol.=3

(а)

(b)


210

mum of current observed in the course of the chronoam-mograms recorded in electrolyte 1 (Fig. 6a), which is followed by an abrupt drop and subsequent retention, as the current follows the Cottrell equation that is valid in the case of diffusion control.

The mentioned mechanisms for Ni and Co deposi-tion [20] are combined at the different potentials in the chronoammograms of deposition of Ni-Co alloy system. The general conclusion suggested by the chronoam-metric analysis is that the behavior of the curves in Fig. 5а evidences for a typical process of nucleation with three-dimensional growth of the nuclei at predominantly diffusion control on the part of the electroactive forms [20]. The critical time for nucleation, i.e. the time in which the current decreases, decreases with the increase of the cathode potential applied.

The method of chronoammetry provides clear diagnos-tic criteria to recognize the mechanism of nucleation [19]. It was shown in [20] that the chronoammograms of Ni are linearized to a higher degree in coordinates I - t -3/2, which is indicative for a mechanism of progressive (continu-ous) nucleation. Unlike Ni, in the case of Co we can observe a better linearization of the chronoammograms in coordinates I - t -1/2, which evidences for a mechanism of spontaneous nucleation that is superimposed onto the general diffusion controlled process of deposition

Fig. 3. Content of Ni-Co alloys (% mass Co) in depend-ence of the applied potential in compositions 3 (curve 1); 2 (curve 2) и 1 (curve 3) of Table 1.

1,1 1,2 1,3 1,4 1,5

60

70

80

90 (0,3M Ni)

-----

3

2

1

% Co

E,V (SCE)

(Ni/Co)sol.:(1) 3,0; (2) 1,5; (3) 1,0

Fig. 4. Processing of the cyclic voltammograms and application of the diagnostic criteria of the method for compositions 1 [17, 18].

5 6 7 8 9 10 1135

40

45

50

55

60

65

i p, m

A cm

-2

v1/2

Ni/Co=1

(а)

20 40 60 80 100 120

5,6

6,0

6,4

6,8

7,2

i p / v

1/2

v, mV s-1

(b)

5 6 7 8 9 10 11

0,18

0,21

0,24

E p-Ep/

2, V

v1/2

(c)

1,4 1,5 1,6 1,7 1,8 1,9 2,0 2,1

1,5

1,6

1,7

1,8

E p, V

log(v)(d)

deposition kinetic of Co than that of Ni [20].In order to further clarify the mechanism of co-dep-

osition of Ni and Co, there the chronoammetric analysis was applied. The typical time-current dependencies in compositions 1 in potentiostatic conditions are shown in Fig. 5a in a wide range of potentials applied from – 0.850 V to -1.35 V. The same figure shows the processing of the curves in I - t-1/2 coordinates (Fig. 5b).

It was found that there is neither deposition of the two metals nor any reactions associated with the background electrolyte at potentials more positive than -0.600 V and only charging the capacity of the double electric layer occurs. There is a well defined and recognizable maxi-


211

of Co. It is possible that the mentioned difference in the mechanisms of nuclei growth is due to the two-stage transfer of charge in the case of Co since the first stage has already localized the needed spots for crystal growth on the surface. The higher speed of nucleation in the case of Co through the mechanism of spontaneous nucleation can be a possible reason for its identified preferential (anomalous) deposition together with Ni. It also follows from the curves in Fig. 5 that with the increase of the potential applied the slope of dependencies I - t-1/2 increases, which corresponds to the increase of the rate of nucleation.

Fig. 5. Chronoammetric dependencies (a) and their deriva-tive dependencies I - t -1/2 (b).

0 5 100

10

20

30

Ni/Co=1(0,3M Ni)

5

4

321

i c, m

A cm

-2

t,sec.

(1)-0,85V; (2)-0,92V;(3)-1,05V; (4)-1,21V;(5)-1,35V

0,4 0,6 0,8 1,0 1,2 1,4

10

20

30

404

3

2

1

(Ni/Co)sol.=10,3M Nii c,

mA

cm-2

t-1/2

(1) -0,85V; (2) -0,92V;(3) -1,21V; (4) -1,35V

(a) (b)

Fig. 6 a,b displays the SEM images of Ni-Co al-loy coating deposited at the same potential from elec-trolytes 3 (Fig. 6a) and 1 (Fig. 6b). The SEM images evidence that the alloys enriched in Ni (Fig. 6a) have larger crystals (about 500 nm) with idle-like shape of the crystallites. In the coatings enriched in Co (Fig. 6b), the crystallites formed are more rounded and smaller in dimension (about 300 nm), and their structure as a whole is smoother and more uniform in terms of morphology. This finding agrees with the theory [15, 16] proving that under the mechanism of spontaneous nucleation. which is more typical for Co and for Ni-Co, the crystallites shall be more uniform in shape and dimension.

CONCLUSIONS

The main conclusions that are imposed by this study are that the co-deposition of Ni and Co in Ni-Co alloy system becomes under combined diffusion-kinetic control with a typical mechanism of nucleation similar to those in the case of Co.

It was found that anomalous co-deposition of the metals occurs in slightly acidic citrate electrolytes, which can be explained with the different deposition kinetics and different mechanisms of nucleation and crystal growth for Ni and Co. With the increase of the relative content of Ni in the solution compared to that of Co, the effect of the preferential deposition of Co obviously weakens thus creating conditions for enrich-ment of the Ni alloy. The SEM-images prove that the crystallites in alloys enriched in Co have more even and rounded shape that agrees with the theory for the case of spontaneous nucleation.

Fig. 6. SEM images of Ni-Co alloys deposited in solutions 3(a) and 1 (b) at Е = -1.250 V (SCE).

a) 75 % mass Co - 25 % Ni (b) 86 % mass Co - 14 % Ni

0 5 100

10

20

30

Ni/Co=1(0,3M Ni)

5

4

321

i c, m

A cm

-2

t,sec.

(1)-0,85V; (2)-0,92V;(3)-1,05V; (4)-1,21V;(5)-1,35V

0,4 0,6 0,8 1,0 1,2 1,4

10

20

30

404

3

2

1

(Ni/Co)sol.=10,3M Nii c,

mA

cm-2

t-1/2

(1) -0,85V; (2) -0,92V;(3) -1,21V; (4) -1,35V

(a) (b)

a)

b)


212

REFERENCES

1. R. Orinakova, A. Turonova, D. Kladekova, M. Ga-lova, R. Smith, Recent developments in the electro-deposition of Ni and some Ni based alloys, J. Appl. Electrochem., 36, 2006, 957-972.

2. S. Armyanov, Crystallographic structure and magnetic properties of electrodeposited Co and Co alloys, Electrochim. Acta, 45, 2000, 3323-3335.

3. L. Wang, Y. Gao, Q. Xue, L.H. Huiwen, T. Xu, Micro-structure and tribological properties of electrodepos-ited Ni-Co alloy deposit, Applied Surface Science, 242, 2005, 326-332.

4. E. Gomez, S. Pane, E. Valles, Electrodeposition of Co-Ni and Co-Ni-Cu systems in sulphate-citrate medium, ElectrochimActa, 51, 2005, 146-153.

5. A. Dolati, M. Sababi, E. Nouri, Ghorbani, A study on the kinetic of the elеctrodeposited Co-Ni alloy thin films in sulfate solution, Materials Chemistry and Physics, 102, 2007,118-124.

6. W. Hansal, B. Tury, M. Halmdienst, M. Versanyi, W. Kautek, Pulse plating of Ni-Co alloys: Deposition kinetic of Watts sulphamate and chloride electrolytes, Electrochim. Acta, 52, 2006, 1145-1151.

7. G. Chung, W. Chang, Effect of pulse frequency and current density on anomalous composition and nano-mechanical property of electrodeposited Ni-Co films, Thin Solid Films, 517, 2009, 4800-4804.

8. J. Vazquez-Arenas, T. Treeratanaphitak, M. Pritzker, Formation of Co-Ni alloy coating under direct cur-rent, pulse current and pulse-reverse plating condi-tions, Electrochimica Acta, 62, 2012, 63-72.

9. D. Golodnitsky, Yu. Rosenberg, A. Ulus, The role of anion additives in the electrodeposition of Ni-Co al-loys from sulfamate electrolyte, Electrochem. Acta, 47, 2002, 2707-2714.

10. K.Y. Sasaki, B. Talbot, Electrodeposition of iron-group metals and binary alloys from sulfate baths, J. Elecrochem. Soc., 145, 3, 1998, 981-990.

11. R.M. Smith, A.E. Martell, Critical Stability Con-stants, v. 4, Plenum Press, New York, 1989.

12. E. Gomez, E. Valles, Electrodeposition of Ni-Co alloys on modified silicon substrates, J.Appl.Elec-trochemistry, 29, 1999, 805-812.

13. E. Chаssaing, Effect of organic additive on the electrocrystallization and the magnetoresistance of Cu-Co multilayers, J.Electrochem.Soc.,148, 10, 2001, C690-C694.

14. Min Gu, Initial stages of the electrocrystallization of Co-Cu alloys on GCE from the Co rich electrolytes, Electrochimica Acta, 148, 10, 2007, 4443-4448.

15. M. Torabi, A. Dolati, A kinetic study on the electro-deposition of nickel nanostructure and its electro-catalytic activity for hydrogen evolution reaction, J. Appl.Electrochem., 40, 2010, 1941-1947.

16. B. Scharifker, Gr. Hills, Theoreticaland Experimental studies of multiple nucleation, Electrochimica Acta, 28, 7, 1983, 879-889.

17. R.S. Nicholson, I. Shain, Theory of Stationary Elec-trode Polarography: Single Scan and Cyclic Methods Applied to Reversible, Irreversible and Kinetic Sys-tems, Analytical Chemistry, 36, 4, 1964, 706-723.

18. E.R. Brown, R.F. Large, Cyclic Voltammetry, AC polarography and related techniques, 5th, Wiley, New York, 1971.

19. Z. Galus, Fundamentals of Electrochemical analysis, Ed. Elis Horwood, 1994, p. 606.

20. K. Ignatova, Effect of H3BO3 on complex formation and kinetics of electrodeposition of Ni-Co alloy in citrate electrolyte, Bulgarian Chem. Communica-tion, 2015, in press.

21. T.A. Green, A.E. Russel, S. Roy, The development of a stable citrate electrolyte for the electrodeposition of Cu-Ni alloys, J. Electrochem. Soc., 145, 3,1998, 875-880.

22. O.Yu. Zelenin, Interaction of the Ni2+ ion with citric acid in an aqueous solution, Russian Journal of Co-ordination Chem.,33, 5, 2007, 346-350, (in Russian).

23. N. Кotsakis, C.P. Raptopoulou, V. Tangoulis, A. Terzis, J. Giapintzakis, T. Jakusch, Correlation of synthetic, spectroscopic, structural and speciation studies in the biological relevant Co(II)- citrate system, Inorganic Chemistry, 42, 2003, 22-31.

20. K. Ignatova, D. Lilova, A study on the kinetic of the electrodeposition of Ni, Co and Ni-Co alloy in citrate electrolyte; Part 1: The kinetic study of the independent electrodeposition of Ni and Co, J. Chem. Technol. Metall., 50, 2, 2015.

Sergey P. Nefedyev, Roman R. Dema, Svetlana A. Nefedyeva, Aleksey V. Yaroslavtcev

213


MICROSTRUCTURE OF CAST IRON AFTER PLASMA BLEACHING


Magnitogorsk State Technical University named after G.I. NosovLenin Avenue, 38, 455045, Magnitogorsk, Сhelyabinsk region, RussiaE-mail: [email protected]

ABSTRACT

This work presents results of metallographic examination of the surface layers of grey cast iron, subjected to plasma bleaching. In the sintered layer are detected ledeburite structure, martensite, ferrite and graphite particles. It is shown that the pre-cast aluminum cladding surface increases the dispersion of carbide particles and the hardness of the melted layer.

Keywords: grey cast iron, microstructure, plasma fusion, bleaching, modifying aluminum, hardness.

Received 05 December 2014Accepted 16 February 2015

INTRODUCTION

Grey cast iron is widely used for the manufacture of many parts in a variety of industries. From cast irons are made the most vital parts of a car - engine blocks, brake discs [1], cylinder liners diesel engines [2], piston rings, etc. [3]. The most frequently used unalloyed or low-alloyed grades of grey cast iron are with a lamellar graphite form.

The physical and mechanical properties of service parts made of cast iron are determined primarily by features of its microstructure. The main feature of the microstructure of the cast iron is the existence of a lamel-lar graphite mold. Flake graphite gives greater continu-ity of the metallic matrix. Therefore, grey cast iron has relatively low resistance values for time at relatively low rupture ductility. In addition, even modified, or doped grey cast badly resists abrasion and sliding wear when applying high contact loads. On the other hand, due to the plate-like shape of graphite in gray cast iron, it has low sensitivity to surface stress concentrators. In addi-tion, grey cast iron has a high damping capacity (internal friction is 4 to 10 times higher than that of steel) and thus dampens vibrations well. So, in many cases it is expedi-

ent to produce grey cast iron core parts of automobiles, and the surface of the wear resistant materials to perform, for example, with a white cast iron alloy [4].

Widely reported in the literature is the possibility for hardening products for cast iron surface whitening dur-ing crystallization or in saturated surface diffusion with chromium, vanadium, or aluminum [5]. However, most of the known methods of whitening for the surface of cast irons require a number of additional technological meas-ures, that are not in a position to guarantee the quality of the bleached layer without pores, graphite inclusions and cracks. There is also a significantly complicated and costly process of hardening. Studies of the possibility for bleaching grey iron with plasma remelting surface [6], show the high efficiency of this method, however, data on the possibility for additional alloying the surface layer of iron in the literature is not enough.

Whitening grey cast iron by surface melting induc-tion, laser or other means, to ensure the formation of a surface layer of the structure, leads to a ledeburitic-martensitic composition with high hardness and wear resistance. The obtained continuous layer without large inclusions, cracks and roughness is satisfactory for dif-ficult tasks.


214

The aim of this work is to study the possibility of fusion plasma for surface chill cast iron with lamellar graphite, and to characterize the microstructure and properties of cast iron after application of an aluminum underlayer and subsequent fusion plasma.

EXPERIMENTAL

Hardened layers were obtained without the addi-tional cooling forced way, only relying on the heat sink into the metal. The starting temperature of the investi-gated irons was 20°C, after processing, it was raised to 80 - 120°C. Surface bleaching was made by remelting the cast iron surface by exposing it to the plasma jet. The influence of surface chill was assessed by measur-ing the hardness, microhardness and microstructure of the remelting zone, the transition zone and the heat affected zone.


Grey cast iron has in the initial state a ferritic-pearlitic metal based plate and a graphite mold (Fig. 1).

After the plasma iron surface remelting, its hardness increased from 24 HRC to 52 HRC. The depth of the surface layer bleached was about 2 mm. The bleached layer has a dense structure, with no visible cracks and chips. The microstructure of the bleached area of the iron is shown in Fig. 2. It can be seen that it is a hypoeutectic white cast iron, in which there are no graphite particles. The metal substrate is an austenite-martensite compo-sition in the axes of the dendrites. The microhardness

of the austenite-martensite composition is 550 - 590 HV10. The dendrites do not have a clearly defined gen-eral direction of the growth axes of the first order. This means that the heat in this layer during crystallization, was very effective and was carried out both in the direc-tion of the base metal and the atmosphere. The average dendritic parameter is 4 - 6 mm, which corresponds to approximately the crystallization rate in the direction of the axes of the dendrites of the first order of 1000°C s-1. In the interdendritic space there is an eutectic cementite type, with microhardness 722 - 810 HV10.

In the lower layers of the molten layer there is a transition zone where are present both the structure char-acteristic for the base metal, the remelting zone and the intermediate products transformations, resulting from the rapid crystallization of liquid iron (Fig. 3).

Clearly visible are also the graphite plates that have not dissolved in the iron melt during the plasma treat-ment, due to the short exposure time. For the graphite plates, on both sides an adjacent cementite eutectic type, in which the carbide particles are arranged perpendicular to the surface of the graphite plates, is observed. We can assume that the transition zone has been formed during accelerated cooling of molten iron, heated to tempera-tures of a solid-liquid state. When it is not soluble, the graphite plate is the substrate on which the crystals of the carbide phase are grown. The metal based eutectic structure formed martensite and austenite. At some distance from graphite plates is located an area of high structure, having martensite and austenite. The integral microhardness of these areas is 440 - 480 HV10. In the deeper layers of the samples, described above, the

Fig. 1. The microstructure of cast iron in the initial state, × 100.

Fig. 2. The structure of the refining zone, × 500.


215

composition is replaced by structures more typical for grey cast iron. The cooling rate at the surface remelting pig iron at these depths is relatively low, so martensite meets troost-martensitic structures.

It is known [7], that introducing iron into aluminum in an amount of 10 – 17 % leads to a significant increase in its hardness and wear resistance, due to the formation of carbides of the type (Fe, Al) 3C. The doped surface layers of aluminum on the workpiece whiten at plasma remelting only at its preliminary application to the surface of the reinforcement. Using the method of plat-ing, a flexible tool [8] for this purpose leads to obvious advantages, allowing to automate the process of apply-ing aluminum and provides a constant thickness of the aluminum layer on the surface of the product.

The plasma remelting of cast iron clad aluminum method leads to an even greater increase in hardness, than plasma remelting without cladding. Hardness in-creases from 52 HRC to 62 HRC, and remelting zone microhardness increases up to 1000 - 1050 HV. The general type of structure, formed in the zone remelting is unchanged (Fig. 4). The bleached layer has a typical eutectic structure. The number of the eutectic compo-nents decreases, and the volume fraction, occupied by the dendrite arm, increases. Based on the substantial solubility of aluminum in α-iron at a temperature rang-ing from the melting point to room temperature (from 30 % to 8 % by mass), suggests that, while most of the aluminum is contained in the base metal. The dendritic structure in the zone of remelting the plasma surface is coated with aluminum at 4 - 6 microns. Obviously, the

metal substrate is the refining zone and a minor amount of martensite austenite. Increasing the amount of mar-tensite due to the action of aluminum has led to a sharp increase in the hardness of hardened layer up to 62 HRC.

CONCLUSIONS

Thus, prior to application of grey cast aluminum surface by plating a flexible instrument and subsequent remelting of the surface of the plasma, can significantly improve the hardness of the hardened layer, compared with the plasma remelting without coating with the aluminum layer. The quality of the hardened layer is much higher than in the bleaching iron, obtained by induction heating. In the zone of remelting, there are no pores and soluble graphite inclusions, which gives reason to recommend this as a very effective treatment for grey iron parts, working in conditions of intense friction, temperature and contact stresses.

REFERENCES

1. D.A. Boldyrev, Made from ductile cast iron parts of the vehicle, Procuring production in mechanical engineering, 10, 2005, 3-6, (in Russian).

2. B.M. Astashkevich, S.S. Voinov, E.A. Shur, Laser hardening cylinder bushings of diesel engines, Met-allurgy and heat treatment of metals, 4, 1985, 12-15, (in Russian).

3. D.A. Boldyrev, V.M. Skantsev, A.A. Pichugin, In-crease basic and special properties of grey pearlite

Fig. 3. The microstructure of the transition zone from the refining zone to the base metal, × 500.

Fig. 4. Microstructure of the zone of remelting pre-wrought aluminum underlayer, × 500.


216

cast iron for cylinder blocks microalloying chromium and sulfur, VPO “Bryansk State Technical Univ.” (BGTU). Bryansk, 2012, (in Russian).

4. A.N. Emelyushin, S.P. Nefedyev, Investigation of the structure and impact-abrasive wear resistance of Fe-C-Cr-Mn-Si, additionally doped with nitrogen, Welding production, 10, 2011, 18-22, (in Russian).

5. A.A. Veselovskii, Improving the wear resistance of high-strength cast-iron gears by applying a thermodif-fusional vanadium coating, Russian Engineering Research, 32, 1, 2012, 45-47.

6. A.N. Emelyushin, S.P. Nefedyev, S.I. Platov, R.R.

Dema, Hardening of cast iron surface chill with microplasma reflow, Modern technologies of metal processing with the use of tools of superhard materi-als - new technologies and directions: Sat. Scien. tr. v. 3. Magnitogorsk: GOU VPO Magnitigorsk State Technical University, 2010, p. 80-87, (in Russian).

7. K.N. Vdovin, I.V. Ponurko, Inoculation of grey iron, Foundry, 10, 1996, 20-21.

8. V.I. Kadoshnikov, V.P. Antsupov, R.R. Dema, Expan-sion of technological possibilities cladding methods of flexible tools, Bulletin of Engineering, 10, 2003, 64-66.

Abdrakhman Naizabekov, Vitaly Talmazan, Alexander Arbuz, Toncho Koinov, Sergey Lezhnev

217


STUDY OF AXIAL FORCES WITH THE PURPOSE TO REALIZE A COMBINED PROCESS «HELICAL ROLLING-PRESSING»

Abdrakhman Naizabekov1, Vitaly Talmazan2, Alexander Arbuz3, Toncho Koinov4, Sergey Lezhnev2

1 Rudny Industrial Institute, Rudny, Republic of Kazakhstan2 Karaganda State Industrial University, Temirtau, Republic of Kazakhstan, E-mail: [email protected] Kazakh National University named after Al-Farabi, Almaty4 Department of Physical Metallurgy and Thermal Equipment, University of Chemical Technology and Metallurgy, 8 Kl. Ohridski, 1756 Sofia, Bulgaria E-mail: [email protected]

ABSTRACT

The key factor for practical realization of the combined process «helical rolling-pressing» is the axial force of the heli-cal rolling, which should provide a continuous pressing in the matrix after the rolls. For the measurement of the maximum axial rolling force, which is actually the reserve of friction forces, a special strain gauge has been made. The work was carried out at a three-roll helical rolling mill “10-30” for the case of hot rolling of steel bars with diameters of 16 - 25 mm at a reduction of 6 % of diameter and a temperature of 1000°C. The results shown are of 24 tests.

Keywords: helical rolling, axial force, strain gauge, combined process.

Received 19 January 2015 Accepted 27 February 2015

INTRODUCTION

The usage of ultrafine-grain (UFG) and nanostruc-tured (NS) metals and alloys as constructional and functional materials of the new generation promises huge advantages due to their properties. Different ways to obtain such materials with the use of intensive plastic deformation have been developed [1-2]. However, the in-dustrial production of UFG and NS materials is still often interlinked with high expenditures of time and energy, restrictions of sizes, and, as a whole, low adaptability to manufacturing that reflects in their cost. Therefore, the development of new principles for using severe plastic deformation (SPD) for production of volumetric NS metals with perspective properties is an actual target.

It is feasible to remove restrictions on the length of a blank piece and provide process continuity with a probable combination of a continuous process (for example, rolling) and an equal channel angular pressing (ECAP) into one process. The theoretical substantiation

and practical realization of the combined process on the basis of longitudinal rolling in smooth rolls and an equal channel angular echelon matrix [2] were made in works [3 - 6]. Results of these works have revealed the prospectivity for creation of such combined processes.

It is known that during the helical rolling, owing to trajectory and speed features of the metal flow, the formation of UFG structure is also possible. However, owing to the same features the central part of the bar can remain not treated [7]. The combination of helical rolling and ECA pressing can eliminate this problem, thus providing the continuity of the process and a po-tentially better treatment of the structure, during the pass in comparison to the above described combined process. This idea is described in details in [8], however its realization is interlinked with difficulties of practical nature, caused by the complexity of the helical rolling process. The key factor is the value of the maximum axial force (that is, essentially, the reserve of friction forces) during helical rolling, which should be sufficient


218

for maintenance of continuous pressing in the matrix standing after the mill. Thus, owing to complexity of the process, experimental research of axial force maximum values distribution have high value for designing of the combined installation, with the purpose of accounting for the total influence of random factors, that is the very aim of this work.

The object in view is reached by the consecutive solution of following issues: experiment definition, designing and manufacturing of the measuring equip-ment, carrying out of the experiment and processing of its results.

EXPERIMENTAL

Measurements of axial force during helical rolling were repeatedly made by various authors [9 - 10] for the case of a pipe blank piece piercing into the sleeve and had the target of the optimization of the mandrel form and the piercing process as a whole. However, measurements of the arising axial force for the case of full compulsory braking of the bar in mill rolls of this type, have not been carried out.

For assessment of the demanded data about the maximum axial effort, a decision was made regarding carrying out of the experiment in 3 sequences with 8 experiments in each, with one influencing factor - the ratio between rolls diameter (Dv) and the diameter of blank piece (D0), accordingly using different values of the factor for each sequence. Such assessment will allow to collect the statistical material for estimation of the maximum axial force, along all the product mix of the mill. Besides this, the use of a relative indicator (Dv/D0) as the factor will allow the expansion of the database of experimental data in the future at the expense of experi-ments for other mills with the same type of calibration and adjustment of rolls.

The experiment was carried out at the mill of radial-displacement rolling (RDR) “10-30”, designed by

National Research Technological University “Moscow Institute of Steel and Alloys” (Russia), as the one provid-ing the demanded structure of the bar after rolling [7, 11]. For the experiment initial profiles with diameters: 16 mm, 20 mm and 25 mm, as the most typical for mill product mix, have been chosen. The ratio between the diameter of the conical rolls (71 mm) and the diameter of the blank piece (Dv/D0), as well as the initial (D0) and the final dimensions (D1) of bars, by sequences of experiments, are given in Table 1.

The percentage reduction (ε, %) for all cases is assumed to be constant, and equal to 6 % of diameter, as convenient for mill adjustment, as well as due to the fear of possible damaging of rolls and camp structure under high load.

Round hot rolled bars (GOST 2590-88) with length of 300 mm were used as blank pieces for the experiment, with diameters according to Table 1. Steel grade St3 (0,14 - 0,22 % C) as one of the most globally widespread constructional material was chosen as the material for blank pieces. The temperature of the bars heating up was defined at the level of 1000°С, which was the average value of the temperature for heat treatment of this class of steel. The distance from the deformation region to the measuring plate corresponded to the prospective distance to the matrix during the combined process and was equal to 100 mm.

The scheme for carrying out of the experiment for measurement of the maximum axial force is shown on Fig. 1. The methodology of the experiment is as fol-lows. Bars for experiments with length of 300 mm, with diameters according to Table 1, in sets of 2 pieces are loaded into the tubular furnace, warmed up to 1000°С with staying of 16 - 30 minutes, depending on the sec-tion and arrangement of the blank pieces. After this, the bars in turns are put into the mill rolls. In the process of rolling the bar (1), moving forward, touches the measur-ing plate (3) with fixed edges and bends it elastically, under the influence of axial force (F). Plate deformation

Table 1. Ratio values by sequences of experiments.

Sequence of

experiments Dv/D0 D0, mm D1, mm Number of experiments

I 2,8 25 23,5 8

II 3,6 20 18,8 8

III 4,4 16 15 8


219

is perceived by resistive strain sensors (7) pasted on it and registered by a strain sensor station in the form of an effort schedule. The peak value of the loading is registered in the table of results.

In addition to the mill RDR “10-30”, the equipment involved in the experiment, includes the following: a tubular furnace Nabertherm R120/1000/13 (Nabertherm GmbH, Germany); a strain sensor station ZET-017-T8 (“ETMS” CJSC, Russia); a measuring plate with resis-tive strain gauge sensors TKFО1-2-200 (“ETMS” CJSC, Russia); a laptop for control of the strain sensor station and signals recording.

The use of a plate with fixed edges (beam scheme) as a measurement device, is caused by the features of the mill design, which is not intended for piercing of blank pieces, and therefore is complicated, unlike in the application of ready decisions (serial load cells) [9-11],. Besides this, such a scheme was chosen owing to the greatest linearity of measurements, less dependence on the point of force application [13], better security of the resistive strain sensors, both from temperature and from mechanical damages, simplicity and convenience of realization. The steel 5ХВ2С (alloyed spring steel) after tempering was chosen as the material of the measuring plate, as capable to endure considerable elastic deforma-tion. The dimensions of the plate were calculated with a condition for achievement of deformation in the places

of the resistive strain gauge sensors pasting, equal to the maximum admissible deformation of chosen resistive strain gauge sensors (2 %) after the force of 100 kN that is more than 2.5 times greater than the expected peak force. Plate edges, realising the beam scheme, lean against thick-wall brackets (4), connected to the front frame of the rolling mill adjusting bolts (6).

Thus, the measuring plate with four resistive strain gauge sensors, connected into the bridge scheme and providing thermal compensation, was developed and manufactured. Resistive strain gauge sensors were pasted using a special glue Z70 (Hottinger Baldwin Messtechnik GmbH, Germany). The scheme of pasting and connection of the resistive strain gauge sensors is shown on Fig. 2.

Resistive strain sensors are pasted symmetrically in the middle of the distance between the centre of plate and support points. Measuring (active) resistive strain sensors (RA) are pasted along the plate, compensatory sensors (RK) - across it, thus perceiving only the tempera-ture disturbance. During the connection into the bridge scheme, measuring (RA) and compensatory (RK) ele-ments alternate. Such connection provides the increase

Fig. 1. Scheme of maximum axial force measurement carrying out. (1 - bar; 2 – rolls, 3 - measuring plates, 4 – brackets, 5 – frame, 6 - adjusting bolts, 7 - resistive strain sensors, F - axial force).

Fig. 2. Scheme of resistive strain gauge sensors past-ing and connection to the measuring plate. a) Scheme of resistive strain gauge sensors pasting; b) Scheme of resistive strain gauge sensors connection (E - feeding of bridge; e0 - output voltage).

a)

b)


220

Fig. 3. Carrying out of the experiment. a) carrying out of the experiment (moment of rolls stoppage); b) bar after the test I-7.

a) b)

in sensitivity of the scheme during its protection against temperature distortions [13]. The scheme receives a feed of 5V direct current from the strain sensor station. Signal recording is made with frequency of digitization of 1 kHz. The measuring plate was calibrated at the tor-sion and tensile testing machine MI-40KU (Ukraine) in a mode of compression testing, as per methodology, minimizing the influence of the hysteresis. The essence of calibration is the composition of the dependence connecting the electric voltage in the scheme and the bending force, applied to the plate. This dependence should have a linear character. For carrying out of the calibration the plate was consistently loaded with force with the step of 5 kN in a range from zero to 35 kN. The corresponding values of the voltage in the scheme under loading and after its removal, were fixed. With the view of hysteresis reduction and accuracy increase, 3 passes on the specified range of forces – upwards, downwards and upwards were carried out, or 42 measurements in total (including measurements of zero values).

The data received in the process of calibration tests were statistically processed, and a regression equation was developed on their basis, connecting the force ap-plied to the plate (Fi, N) with the voltage (Ui, mV) in the scheme. The equation looks like: Fi = -3631.2 Ui + 10122. The ratio of determination R2 = 0.99998; the standard error of measurements attributed to the value of the working range of plate measurements (40 kN) based on results from 42 tests was less than 0.2 %. The received data were uploaded into the program of measurements registration and processing of the strain sensor station ZET-017-T8 for the possibility of signal recording in the form of the force trend. After the experiment little random control loadings were made, in which the devia-tion of values did not exceed the value specified above, which confirmed the accuracy and the stability of the work of the measuring plate.


All steps of the experiment have passed in the nor-mal mode. At the moment of the visible bend achieve-ment of the bar resting against the plate, the mill drive engines were synchronously stopped to make possible the taking out of the bar from the stand without interfer-ing with mill settings. The general view of the experi-mental installation and the bar, taken out after rolling, are shown on Fig. 3. On Fig 3a the bend of the measuring plate under the influence of the axial force is clearly seen. On Fig. 3b in the bottom part of the bar there are roll marks seen, characterizing the deformations region.

Force trends fixed by the strain sensor station are generally similar and have the same distinctive sections. As an illustration of this, Fig. 4 shows the trends of tests II-3 and II-6, fixed by strain sensor station. At the first section there is a sharp (during approximately 0.15 sec) increasing of the force, with some delay closer to the peak. At this stage, there is a bend of the plate under the influence of axial movement of the bar and a small

Table 2. Statistical characteristics of experiments.

Statistical parameter Experiments sequence/ (Dv/ D0)

I / 2,8 II / 3,6 III / 4,4

Fave, N 35 541 31 098 20 539

Fmax, N 39 394 34 484 21 753

Fmin, N 31 077 27 718 19 512

Fst, N 2 402 2 424 787


221

Fig. 4. Force trends for tests II-3 (a) and II-6 (b).

a)

b)

deformation of the head end of the bar. The shape of this section of the trend comes close to a parabola shape. Then there is a bend of the bar, accompanied by decreas-ing of the force by one third and a smooth alignment of the force decreasing, probably connected with the start of rolls sliding. Thus, it is important to notice, that at the last stage, the bar leans not only against the plate, but also against the part of the front frame, because at this stage it is generally strongly bent.

The moment of the stoppage of the drives is clearly visible at the trends in the form of a short negative spike to the right of the peak. As it was shown by the experiments, the moment of the drives stoppage does not influence the qualitative and the quantitative picture of the force changing in practice. The results of each sequence of experiments have been checked for presence of gross errors by Student’s t-test, and then statistical characteristics have been calculated for each sequence: arithmetic average (Fave), maximum (Fmax) and minimum (Fmin) values, standard deviation (Fst). The listed charac-teristics are shown in Table 2. Values of the force in all experiments are shown graphically on Fig. 5.

CONCLUSIONS

The equation ( ) ( )20 04,78 25,05 2,89v vF D D D D= − + + ,

characterizing the dependence of maximum axial force at helical rolling from the ratio between the rolls diameter and the diameter of blank piece with constant reduction ε = 6 % (shown in dotted line) has been obtained. The ratio of determination was equal to R2 = 0.92.

Fig. 5. Values of maximum axial force.


222

From the data it is possible to draw the conclusion that during the rolling of thicker profiles essentially a high disorder of values of forces is observed, that can be possibly explained by the increased influence of the following factors - features of internal and contact friction at the blank piece, its rheology, features of the deforming tool calibration.

The measured values of the axial force (up to 38 kN) evidence the principal possibility of the combined pro-cess of rolling-pressing realization with high values of the angles at the joint of the channels of the ECA-matrix (140° - 150°). It was established that after reaching of the force peak, there is a blank piece bend, thus, during the realization of the combined process, the main danger here is not in the lack of force, but in the possibility of a blank piece bend between the rolls and the matrix. Besides that, the value of the force can be raised a little by using of notched rolls at maximum reduction. The obtained results are comparable to results of research on similar profiles piercing forces on three-roll mills [9 - 10].

The results of this study can be used for the optimiza-tion of the process of continuous blank pieces piercing into the sleeve, and as results containing data on the re-serve of axial force (friction forces) of the helical rolling.

REFERENCES

1. R.Z. Valiev, I.V. Alexandrov, Volumetric nano-structural metal materials: receiving, structure and properties. Monography, Moskow, Akademkniga, 2007, (in Russian).

2. Zh.A. Ashkeev, A.B. Naizabekov, S.N. Lezhnev, A.R. Toleuova, Billet deformation in uniform-channel stepped die, Steel in Translation, 35, 2, 2005, 37-39.

3. S.N. Lezhnev, A.B. Naizabekov, Y.A. Panin, 20th Anniversary International Conference on Metallurgy and Materials, Metal 2011, May 18-20, 2011, Brno, Czech Republic, 272-277.

4. A.B. Naizabekov, S.N. Lezhnev, E.A. Panin, 17-th In-ternational Conference on Metallurgy and Materials, Metal 2008, May 13-15, 2008, Hradec nad Moravici, Czech Republic.

5. Patent of the Republic of Kazakhstan #25863. МПК B21J 5/00. The device for continuous metal pressing. A.B. Naizabekov, S.N. Lezhnev, E.A. Panin, 2013, Applicant and patent owner - Regional state enter-prise “Karaganda State Industrial University”. - № 2011/0762.1, applied 02.07.2011; published 15.02.2013, Bulletin of inventors, 2013, #7, (in Russian).

6. S. Lezhnev, E. Panin, I. Volokitina, Research of combined process “Rolling-pressing” influence on the microstructure and mechanical properties of aluminium, Advanced Materials Research, volume 814, 2013, 68-75.

7. S.P. Galkin, Trajectory and speed features of radial-displacement and screw rolling, Modern problems of metallurgy, Dnepropetrovsk, Sistemni technologii, 11, 2008, 26-33, (in Russian).

8. A. Naizabekov, S. Lezhnev, A. Arbuz, 22nd Inter-national Conference on Metallurgy and Materials, Metal 2013, May 15-17, 2013, Brno, Czech Republic, 422-426.

9. P.K. Teterin, The theory of cross-sectional and screw rolling, 2nd ed., Мoskow, Metallurgy, 1983, (in Russian).

10. D.A. Kovalev, PhD thesis, The research and develop-ment of the technology for helical rolling process for increase of plasticity hypereutectic silumin alloys, Мoskow, 2011, (in Russian).

11. Patent #2009737 of the Russian Federation, MPK5 В21В19/02. The three-roll mill for screw rolling and the technological tools for screw rolling mill, B.A. Ro-mantsev, V.K. Mihailov, S.P. Galkin, M.G. Degtyaryov, B.V. Karpov, A.P. Tchistova, № 5031365/27; applied 13.02.1992, published 30.03.1994, (in Russian).

12. Patent #2293619 of the Russian Federation, MPK В21В 19/00, The method of screw rolling, S.P. Galkin, Applicant and patent owner - National Re-search Technological University “Moscow Institute of Steel and Alloys” - #2006110612/02, applied 04.04.2006; published 20.02.2007. Bulletin of in-ventors, 2007, #5. page 46, (in Russian).

13. R.A. Makarov, L.B. Rensky, Tenzometry in mechani-cal engineering, Мoskow, Mechanical engineering, 1975, (in Russian).

Alexander Zahariev

223


Short Communication

BREAKDOWN BEHAVIOUR DURING REANODIZATION OF THICK POROUS ANODIC ALUMINA IN AQUEOUS

AND NON-AQUEOUS SOLUTIONSAlexander Zahariev

Department of Chemistry, Technical University,8 St. Kliment Ohridski blvd.,1000 Sofia, Bulgaria,E-mail: [email protected]

ABSTRACT

A comparative study towards reanodization of thick (22 µm) porous anodic alumina coatings in aqueous and non-aqueous electrolyte solutions is performed. The kinetics of pore-filling and breakdown characteristics dur-ing reanodization demonstrate unanticipated development such as some persistence in voltage growth in aqueous environment and extremely sharp voltage drop at breakdown threshold in non-aqueous one. These findings could be most probably assigned to the irregular structure and complicated phase composition of thick porous coatings.

Keywords: anodic aluminium oxide, reanodization, breakdown voltage.

Received 10 November 2014Accepted 12 February 2015

EXPERIMENTAL

The kinetics of pore-filling and breakdown behavior during re-anodization of porous anodic oxide films on aluminium are studied in two different (aqueous and non-aqueous) electrolytes - aqueous borate electrolyte (ABE) and ammonium salycilate in dimethylformamide (0.1M AS/DMF). Porous films are formed in 15 % sulfuric acid solution at current density 5x10-3 A cm-2 for 180 min. Under these conditions porous layers with a thickness of approximately 22 mm are formed [1]. The re-anodization is carried out at current density 1x10-3 A cm-2 for 1 hour. All studies were performed at room temperature. All these electrochemical procedures took place using two-electrode cell of working and auxiliary (Pt grid) electrodes, and a home-made high voltage galvanostat (600 V, 0.5 A). The formation voltage was controlled

and recorded on a precision multimeter (Mastech MS 8050) along with a PC-based data acquisition system.


As a result of these studies a series of re-anodization kinetic U(t)-curves are produced and are characterized with good reproducibility. An example of re-anodization curves derived using various electrolytes is presented in Fig. 1. It is interesting to note that in the kinetic curves shown, some peculiarities could be observed as well. In re-anodization in ABE, persistence in the values of voltage for a sizable period of time after reaching 150 V is observed. The emergence of breakdowns begins in about 3000 s which is quite unexpected. On the other hand, the curve recorded by pore-filling in 0.1M AS/DMF shows sharp voltage drop of approximately 600


224

Fig. 1. Kinetic curves of pore-filling of porous matrices formed in 15 % sulfuric acid solution at current density 5.0 mA cm-2 for 180 min. Reanodization was carried out at current density 1 mA cm-2 for 1 h.

volts at the first breakdown, compared to only 100 V in 1M AS/DMF [2]. Furthermore, comparing both curves a significant difference within the breakdown oscillations’ amplitude is obvious.

CONCLUSIONS

A series of re-anodization kinetic U(t)-curves is derived and they demonstrate very good reproducibility. The special features observed in the kinetic U(t)-curves for both electrolyte solutions are probably due to the large thickness of porous matrix, suggesting some peculiarities in the structure and phase composition of

the porous part of the film. Furthermore, tenfold lower concentration of AS/DMF solution reveals a significant effect on the first breakdown as well.

REFERENCES1. G. Patermarakis, P. Lenas, Ch. Karavassilis, G.

Papayiannis, Kinetics of growth of porous anodic Al2O3 films on Al metal, Electrochim. Acta, 36, 1991, 709-725.

2. Christian Girginov, Alexander Zahariev, Ivan Kanazirski, Breakdown phenomena during the for-mation of complex anodic oxide films on aluminium, Compt. Rend. Acad. Bulg. Sci., 64, 2011, 1117-1122.

225


Journal of Chemical Technology and Metallurgy is a specialized scientific edition presenting original research results in the field of chemical technology and metallurgy, chemical engineering, biotechnology, indus-trial automation, environmental protection and natural sciences. The articles published in Journal of Chemical Technology and Metallurgy refer to:

Inorganic Chemistry; Organic Chemistry; Analytical Chemistry; Physical Chemistry;

Organic Synthesis and Fuels; Polymer Engineering; Textile and Leather; Cellulose, Paper and Polygraphy; Inorganic and Electrochemical Productions;

Chemical Engineering; Industrial Automation; In-formation Technology; Biotechnology; Economics and Management of Chemical and Metallurgical Industry; Sustainable Development and Environmental Protection;

Physical Metallurgy; Metallurgy of Iron and Cast-ing; Metallurgy of Non-ferrous and Semiconducting Materials; Technology of Silicates; Nanomaterials.

Journal of Chemical Technology and Metallurgy publishes full-length research papers, critical and book reviews. Research papers are expected to be complete and authoritative accounts of work that has significance and interest, presented clearly and concisely. Critical reviews are commissioned by the Editor-in-Chief. Au-thors intending to offer critical reviews are invited first to contact the Editor-in-Chief.

Contributions will only be considered for publica-tion if they are relevant to the topics pointed above. Presentation and discussion should be at the level of the Journal status gained. The language can be a reason for rejection if below an acceptable level of clarity. De-tailed descriptions of the equipment used should only be given if the latter is new. Papers reporting experimental data without adequate interpretation are not acceptable. Contributions will be accepted for publication only on the recommendation of referees.

Submission of ManuscriptsManuscripts should be submitted to the Editor-in-

Chief at the following address:University of Chemical Technology and Metallurgy8 Kliment Ohridski blvd., Sofia 1756, Bulgaria, (for JCTM)Manuscripts may be submitted electronically to

e-mail: [email protected]

GUIDE FOR AUTHORS

Manuscript preparationIt is important that the file be saved in the native

format of the wordprocessor used. All contributions should be typed, double-spaced, 12-pt font. All pages must be numbered in sequence. Most formatting codes will be removed and replaced on processing the article. In particular, do not use the wordprocessor’s options to justify text or to hyphenate words. However, do use bold face, italics, subscripts, superscripts, etc. Do not embed graphically designed equations or tables, but prepare these using the wordprocessor’s facility. When preparing tables, if you are using a table grid, use only one grid for each individual table and not a grid for each row. If no grid is used, use tabs, not spaces, to align columns.

Essential title page informationTitle. It should be concise and informative. Titles

are often used in information-retrieval systems. Avoid abbreviations and formulae where possible.

Author names and affiliations. The names of the authors have to be clearly indicated. Present the authors’ affiliation addresses (where the actual work was done) below the names. Provide the full postal address of each affiliation, including the country name.

Corresponding author. Clearly indicate who will handle correspondence at all stages of refereeing and publication, also post-publication. Ensure that phone numbers (with country and area code) are provided in addition to the e-mail address and the complete postal address. Contact details must be kept up to date by the corresponding author.

Present/permanent address. If the author has moved since the work described in the article was done, or was visiting at the time, a “Present address” (or “Permanent address”) may be indicated as a footnote to that author’s name. The address at which the author actually did the work must be retained as the main, af-filiation address. Superscript Arabic numerals are used for such footnotes.

AbstractA concise and factual abstract is required. The ab-

stract should state briefly the purpose of the research, the principal results and major conclusions. An abstract is often presented separately from the article, so it must be able to stand alone. For this reason, References should be avoided, but if essential, then cite the author(s) and year(s). Also, nonstandard or uncommon abbreviations

226


should be avoided, but if essential they must be defined at their first mention in the abstract itself.

Article structureIntroductionState the objectives of the work and provide an

adequate background. Avoid a detailed literature survey or a summary of the results.

ExperimentalProvide sufficient details to allow the work to be

reproduced. Methods already published should be indi-cated by a reference: only relevant modifications should be described.

Results and DiscussionResults and the discussion provided should be clear

and concise.ConclusionsThe main conclusions of the study have to be pre-

sented in a short Conclusions section.AcknowledgementsAcknowledgements are given in a separate section

at the end of the article before the references.References should be cited in the text by an Arabic

numeral in square brackets, listed numerically and typed double-spaced on a separate sheet at the end of the paper. References should be arranged in the order in which they appear in the text. Journal titles should be abbreviated ac-cording to the latest Chemical Abstracts Service Source Index. Only articles that have been published or are in press should be included in the references. Unpublished results should be cited as such in the text. References in non-English language should be translated in English. The original language must be shown in brackets. In the reference list, the styling, punctuation and capitalization should be as follows:

For journals:S.K. Maji, A. Pal, T. Pal, A. Adak, Modeling and

fixed bed column adsorption of As(III) on laterite soil, Sep. Purif. Technol., 56, 2007, 284-290.

For books:A. Krueger, Carbon Materials and Nanotechnology,

Weinheim, WILEY-VCH Verlag GmbH & Co. KGaA, 2010.

For edited books:F.W. Cooke, J.E. Lemons, B.D. Ratner, in: B.D.

Ratner, A.S. Hoffman, F.J. Lemons (Eds.), Biomaterials Science, Academic Press, San Diego, 1996, p. 11.

For conference proceedings, symposia, etc.:E. Motta, D. Rajpathak, Z. Zdrahal, R. Roy, The

Epistemology of Scheduling Problems, F.V. Harmelen (ed.), Proceedings of the 15th European Conference on Artificial Intelligence, Lyon, France, 2002, 215-219.

When you are citing papers published in Journal of Chemical Technology and Metallurgy you should use the abbreviation of the title as follows:

J. Chem. Technol. Metall.

Formulae should be typewritten (Equation Editor) with ample space around them. The meanings of all symbols have to be given immediately after the equa-tion in which they are first used. Equations should be sequentially numbered (on the right side of the equation and in parentheses). Subscripts and superscripts should be clearly indicated. Greek letters and other non-Latin symbols should be typed using Times New Roman Gr or Symbols.

Units should be given in SI system. They should be in the form, e.g. g cm-1 rather than g/cm.

Figures should be numbered consecutively with Arabic numerals. All figures should be supplied on separate sheets. All illustrations must be readable when reduced to a width of 75 mm (single column figure) or 160 mm (double column figure). Photographs, charts and diagrams are all to be referred to as “Figures(s)” and should be numbered consecutively in the order to which they are referred. They should accompany the manuscript, but should not be included within the text. All illustrations should be clearly marked with the figure number. Figure captions should be typed double-spaced on a separate sheet. Figures should not include text. Graphic files should be in TIFF or JPG with resolution not less than 300 dpi.

Tables should be numbered consecutively with Arabic numerals in the order of appearance in the text. Each table should be given on separate sheet with a short descriptive title directly above it, with essential footnotes below.

One copy of the manuscript in English and one in Bulgarian (for Bulgarian authors) with a set of original illustrations and a disk must be submitted. The manu-script should be written with true-type font Times New Roman size 12 with double line spacing. Manuscripts of papers submitted to the journal will not be sent back to the authors.

Documents

JOURNAL - dl.uctm.edu(bio)pharmaceutical, chemical and cosmetic industry, etc. Nanotechnology promises to exceed the impact of the “industrial revolution” with a projection of