Accepted Manuscript

Removal of oil from water using polyurethane foam modified with nanoclay

Amir Ahmad Nikkhah, Hamid Zilouei, Ahmad Asadinezhad, Alireza Keshavarz

PII: S1385-8947(14)01273-XDOI: http://dx.doi.org/10.1016/j.cej.2014.09.077Reference: CEJ 12695

To appear in: Chemical Engineering Journal

Received Date: 28 June 2014Revised Date: 21 September 2014Accepted Date: 22 September 2014

Please cite this article as: A.A. Nikkhah, H. Zilouei, A. Asadinezhad, A. Keshavarz, Removal of oil from waterusing polyurethane foam modified with nanoclay, Chemical Engineering Journal (2014), doi: http://dx.doi.org/10.1016/j.cej.2014.09.077

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting proof before it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Removal of oil from water using polyurethane foam modified with nanoclay

Amir Ahmad Nikkhah, Hamid Zilouei*, Ahmad Asadinezhad, Alireza Keshavarz

Department of Chemical Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran

Corresponding author: Hamid Zilouei

Tel: +98 31 33913178

Fax: +98 31 33912677

Email: hzilouei@cc.iut.ac.ir

2

Abstract

To enhance the removal of oil contaminants from water, polyurethane foam structure was modified by

integrating cloisite 20A nanoclay into it. Pure and modified polyurethane foams (nanocomposite

adsorbents) were then characterized using scanning electron microscopy, X-Ray diffraction, and Fourier

transform infrared spectroscopy tests. Optimum weight fraction of the added cloisite 20A to the foam

structure was 3wt%, improving the sorption capacity up to 16% and oil removal efficiency up to 56% in

water-oil system. The reusability feature of blank polyurethane and nanocomposites with 3 wt% and 4

wt% of cloisite 20A nanoclay was studied through chemical regeneration by toluene and petroleum

ether. In the case of structurally modified polyurethane foams with nanoclay (nanocomposites),

chemical regeneration reduced the oil removal efficiency, but improved the adsorption capacity in the

range of low to medium oil initial concentration and reduced it in high oil initial concentrations. A

comparison between the obtained adsorption data and adsorption isotherm models, including Langmuir,

Freundlich and Redlich-Peterson, showed a good agreement with Langmuir and Redlich-Peterson

models.

Keywords: Oil pollution, Oil sorbent, Polyurethane foam, nanoclay, cloisite 20A

3

1- Introduction

Oil discharge into the natural environment and aquatic ecosystems can cause serious global,

ecological, and environmental problems [1]. Industrial development has increased oily wastewater

discharge to the environment [2]. Petroleum transportation with a yearly average of about 5 million tons

across seas poses a great risk of pollution to the marine ecosystem [3]. Oil industry, especially oil

refining processes and transport, has a major role in this problem. However, increasing oil consumption

and establishing more refineries near the cities and populated regions have caused severe pollution in the

underground and surface waters [4]. Oily contaminants in polluted water may be detected in different

forms like fats, lubricants, cutting liquids, heavy hydrocarbons (tars, grease, crude oils and diesel oil),

and light hydrocarbons (kerosene, jet fuel and gasoline) [2]. So, the removal of in situ oil and other types

of organic pollutants removal is crucial to prevent them from migrating and to reduce their disastrous

effects on the ecosystem [1].

Different techniques have been developed for the removal of oil contaminants from water. They are

classified into chemical, biological and physical methods [5]. These include different types of filters [6],

chemical dosing, reverse osmosis [7], gravity separation [8], ultra-filtration [9], micro-filtration [10],

biological processes [10], air flotation [11], membrane bioreactor [12], chemical coagulation,

electrocoagulation and electroflotation [13]. Adsorption is among the most profitable methods for the

removal of oil contaminants as it can effectively remove or recover oil from the water [14].

Oil sorbents are divided into three basic types: natural organic, natural inorganic and synthetic

adsorbents [15]. Some examples of natural sorbents used for the adsorption of oil are sugar cane bagasse

[16], vegetable fibers [17], sawdust bed [18], bentonite, chitosan, activated carbon [19], vermiculite

[20], chrome shavings [21] and peat [22]. Synthetic sorbents used for oil contaminants adsorption

include rubber powder [23], expanded perlite [24], polymeric material based on butyl rubber [25], high-

silica zeolites [26], carbonized pith bagasse [27], Wool-Based Nonwoven [28], hydrophobic aerogels

4

[29], Acetylated rice straw [30], exfoliated graphite [31], inorgano clays [32], Polypropylene [33] and

oleophilic polyurethane foams [34].

Polyurethane (PU) in the form of foam provides a large specific area and enough space for

adsorption. Polyurethane foams (PUFs) have shown the noticeable capability of oil adsorption due to

their special features such as low density, open-cell, high porosity, and industrial production. Nanoclay

is one of the possible materials that can modify the polyurethane foam structure. The presence of

cloisite 20A nanoclay in the structure of PUF has been found to enhance the foam strength [35] and

open its cells [36]. Furthermore, nanoclay itself is an oil adsorbent, but it needs structural modification.

The main objective of this study is to investigate the effect of cloisite 20A nanoclay presence within

the PU foam structure on the removal efficiency of oil. The effect of chemical regeneration of the foam

and its performance are also investigated. The isotherm of the oil adsorption is also studied using this

sorbent.

2- Materials and Methods

2-1 Materials

NIXOL AM-313 polyether polyol was provided from KPX chemical CO. (South Korea)

and methylene diphenyl diisocyanate (MDI) was obtained from Daeyang International CO. (South

Korea). 1,1-dichoro-1-fluoroethane (HCFC 141b) was purchased from Lin´an E-COOL Refrigeration

Equipment CO. (China). Cloisite 20A nanoclay was PUFrchased from Sigma Aldrich(USA). Light

crude oil was obtained from Isfahan refinery feed stream (Iran). Toluene and Xylene were supplied by

Isfahan petrochemical complex (Iran) and petroleum ether with the boiling range of 30-60 ºC was

obtained from Pars Chemie Co. (Iran).

5

2-2 Synthesis of polyurethane foams

Deionized water was used as the chemical foaming agent while HCFC 141b was used as the

physical foaming agent. For the synthesis of open-cell PUF, at first, 10g of polyol was fully mixed with

0.1g of deionized water and 1g of HCFC 141b. Then, 4g of MDI was added to the homogeneous

mixture from the previous step, and it was completely mixed using a mixer at 1000 rpm. Immediately

after mixing, the mixer blade was taken out and the final mixture was left for 30 min to provide enough

time for the foaming reaction. It should be noted that the amount of added MDI to a definite quantity of

polyol depends on the stoichiometric reaction, but because of the net structure of polyol, this amount is

higher than that required by the stoichiometry. The whole procedure was performed under ambient

temperature (22±3ºC).

2-3 Synthesis of nanocomposite

For the synthesis of nanoclay-polyurethane foam (NCPUf) naonocomposite, nanoclay (NC) should

be completely dispersed in the polymeric structure in order to obtain nanostructured foam based on

nanoclay and polyurethane foam where polymeric chains diffuse into the silicate layers of nanoclay.

Nanoclay was dried in an oven under the temperature of 100 ºC for 24 hours, and then mixed with

polyol by means of a mixer at 1000 rpm for 24 hours. Then, the obtained mixture was mixed by an

ultrasonic for 25 min (by 10 min rest between each 5 min), and again the mixture was mixed by a mixer

at 1000 rpm for 2 hours. After that, the foaming agents including 1wt% deionized water and 5wt%

HCFC 141b were added while mixing to have a homogeneous mixture and then 35wt% MDI was

immediately added and was mixed for 20 seconds. At the end, the final mixture was left for 30 min to

provide enough time for the foaming reaction. Synthesized foam was cut into 1cm3 cubes to be used in

the sorption experiments.

6

2-4 Adsorption determination method

The method developed for the measurement of oil and water sorption capacity of the sorbent was

based on the Standard Test Method for sorbent performance of adsorbents (ASTM F726-99). All of the

sorption experiments were performed in water-oil system with different initial weight of oil. In water-oil

system test, crude oil was poured into a 600 ml beaker containing 250 ml of deionized water with the

thickness of oil layer being about 2-4 mm. Then, 1.0 g of the adsorbent cubes was added to the system

and the beaker was placed on a shaker at 100 rpm for 5 min ± 20 s. The content of the beaker was

allowed to settle for a period of 2 min. Then, adsorbent cubes were removed and put into glass beaker

using forceps and weighted accurately using balance to determine the total weight of adsorbed oil and

water. Weight of the adsorbed water was determined according to the Standard Test Method for Water

in Crude Oil by Distillation (ASTM D4006). Adsorbent cubes after the sorption stage, were completely

washed by 400 ml of xylene to extract the adsorbed oil and water. Then, the extracted mixture was

boiled in an azeothropic distillation apparatus, and distilled water was collected in a trap connected to

the distillation apparatus. Finally, the collected water was weighted to determine the adsorbed water.

Finally, the oil sorption capacity of the sorbent was calculated using the following equation:

S=��������

�� (1)

where S is the oil sorption capacity (g/g), Ss is the weight of saturated sorbent (water + oil + sorbent), Sw

is the weight of adsorbed water (g) and S0 is the initial dry weight of sorbent (g). Oil adsorption

percentage of sorbent was determined by the following equation:

�� =

�× 100 (2)

where Pa is the oil sorption percentage (oil removal percentage), Oa is the weight of adsorbed oil (g)

and Ot is the initial weight of oil (g).

7

Oil removal efficiency of the sorbent is defined as the ratio of adsorbed oil to the total weight of

adsorbed materials. It is calculated by the following equation:

R=��

�� (3)

where R is the oil removal efficiency, Mo is the weight of adsorbed oil (g) and Mt is the total weight of

adsorbed materials. All of the tests were performed at ambient temperature (22 ± 3ºC) and in duplicate.

If the value of any results deviated over 15% from the arithmetic mean of the two runs, the sample data

would be rejected and the test would be repeated.

2-5 Adsorbent regeneration method

Regeneration of the adsorbents was performed through chemical regeneration method. Toluene and

petroleum ether were used for this purpose. To do this, the used adsorbents were immersed in 150 ml of

toluene in a 250 ml beaker and then mixed by a mixer. After that, adsorbent cubes were taken out and

washed by petroleum ether for 3 times. Then, the cubes were dried in an oven for 1 hour at 65 ºC.

2-6 Adsorption Isotherms

The Langmuir model is based on monolayer adsorption onto a homogeneous structure without any

reaction between the adsorbed molecules. This model is commonly written as:

qe =qmax����

������ (4)

where qmax is the maximum sorption capacity (g/g), qe is the amount of adsorbate in the adsorbent

at equilibrium (g/g), Ce is the equilibrium concentration of oil (g/L) and KL is the Langmuir isotherm

constant (L/g). A dimensionless separation factor (RL) defined by Webber and Chakkravorti (1974) can

8

be used to determine the feasibility of the adsorption process using the Langmuir parameter KL [37].

Thus:

�� =�

������ (5)

where KL (L/g) refers to the Langmuir constant and Co is the adsorbate initial concentration (g/L).

The value of RL implies the adsorption to be unfavorable (RL>1), linear (RL = 1), favorable (0 < RL< 1)

or irreversible (RL =0) [38, 39].

The Freundlich isotherm can be applied to non-ideal adsorption on heterogeneous surfaces as well

as multilayer sorption. It is expressed by the following equation:

qe=KFCe1/n

(6)

where KF is the Freundlich equilibrium constant and indicates the adsorption capacity, n is the

Freundlich constant showing the affinity of the adsorbate for the surface of adsorbent, qe is the

equilibrium weight of adsorbate per unit weight of adsorbent (g/g), Ce is the concentration of adsorbate

in solution at equilibrium after the adsorption is complete (g/L) [39, 40].

Among three parameter isotherm models, Redlich-Peterson model is usually used in the case of

heavy metal and organic material adsorption. This model has advantages of both Langmuir and

Freundlich models. Redlich-Peterson isotherm model contains three parameters: A, B, and β. A and B are

Redlich-Peterson isotherm constants, as shown below:

qe=���

������ (7)

A is the L of solution per gram of adsorbent (L/g) and B is the L of solution per gram of adsorbate

(L/g). In the case of low values of (β), this equation will be linear and transforms to the Henry’s law.

Under the conditions where (A, B>>1 and β<1), this model resembles Freundlich model and A/B and (1-

9

β) are related to KF and n in Freundlich model, respectively. When β equals to 1, Redlich-Peterson

model resembles the Langmuir model [39, 41].

2-7 Characterization

FTIR spectra of blank PUF and structurally modified PUF were obtained using a FT-IR

spectrometer (JASCO, Model 6300, Japan). X-ray diffraction of the modified foam was obtained by an

XRD instrument (Bruker, Model D8 ADVANCE, Germany). The morphology of the synthesized foams

was studied by scanning electron microscopy (SEM) image (TESCAN - VEGA3, SBU – Easy Probe,

Czech).

3- Results and discussion

3-1 Fourier transform infrared spectroscopy (FTIR)

FTIR spectra of the synthesized PU foam are depicted in Fig. 1. The characteristic peaks of PU are

evident close to 3340 cm-1.These peaks belong to the corresponding vibration of hydroxyl functional

group (O-H), probably due to their existence in unreacted polyol (reaction of polyol with isocyanate).

The peaks near the wave number of 2950cm-1 are associated with the vibration of –CH2 and –CH

functional group in the carbonic chains. The sharp peak with the wave number of 1730cm-1 was related

to the urethane functional group and the carbonyl (C=O) functional group existing in urethane linkage.

The peak at 1175 cm-1 is ascribed to the vibration of etheric C-O group. The peaks at 1241 and 3335 cm-

1 are respectively due to the C-N and N-H groups. The signal of medium intensity observed around 1537

cm-1 belongs to the aromatic C=C vibration. Alkane C-H vibration with medium intensity is observed at

the wave number of 2960 cm-1. Broad peak of the alcoholic O-H is detected at 3335 cm-1. A peak due to

10

the esteric C=O of strong intensity is observed at 1730cm-1. Para arrangement of aromatic ring could be

seen with the wave number of 819cm-1.

3-2 X-ray diffraction analysis

Analysis of XRD was used to determine the dispersion state of cloisite 20A nanoclay within the PU

foam structure. The results of XRD for pure cloisite 20A nanoclay and its nanocomposites are presented

in Fig. 2. It is well established in the literature that when intercalation of nanoparticles takes place in the

foam structure, the characteristic XRD peak of the nanoclay shifts to lower angles due to the penetration

of the polymer chains into the interlayer space of the nanoclay (according to the Bragg's law in X- ray

subtraction). Also, disappearance of the XRD characteristic peak is ascribed to the exfoliation of the

nanoclays particles [42]. Based on the XRD pattern of the pure cloisite 20A, a major peak is evident at

2θ = 4˚ due to the scattering of the layered structure of the nanoclays. Nanocomposite adsorbent with

2%wt cloisite 20A shows no peak due to the occurrence of exfoliation where the layers are completely

dispersed within the foam. Nanocomposite adsorbent with 3%wt shows a shifted peak of much lower

intensity due to the existence of intercalated morphology. Regarding nanocomposite adsorbent with

4%wt cloisite 20A, the peak is evident at around 2θ = 1.7˚ which reveals the intercalation of the

nanoparticles in the latter sample.

3-3 SEM analysis

SEM images of the blank PUF cubes and nanocomposite with 3wt% of nanoclay are shown in Fig.

3. The SEM images reveal change of the pore shapes by the addition of nanoclay particles to the foam

structure. As it is seen, in the case of nanocomposite with 3wt% of nanoclay in comparison with blank

11

PUF, the radius of the pores has been decreased and the structure has become more homogeneous in

terms of the pores volume.

3-4 Oil adsorption capacity and oil removal efficiency in oil-water system

In adsorption experiment, the equilibrium time of 5 minutes was obtained based on primarily kinetic

experiment. Therefore, based on standard method, after 5 minutes adsorption experiment and settling

time of 2 minutes, the sorbents were withdrawn and weights of oil and water were measured.

NCPUF cubes showed better performance in adsorbing crude oil and maximum oil adsorption per unit

weight of sorbent was for NCPUF with 3 wt% of nanoclay. Compared with the blank PU foam, the

sorption capacity of NCPUF 2%, NCPUF 3% and NCPUF 4% was increased to 7%, 16%, and decreased

to 8% for crude oil, respectively (Fig. 4). A comparison of the maximum sorption capacity of modified

PUF used in this study with other previously reported sorbents is presented in Table 1.

3-5 The effect of oil initial concentration on the adsorption capability of nanocomposite

adsorbents

Addition of cloisite 20A to the polyurethane foam structure; as seen in SEM images, led to cell

opening and strengthened it [35, 36]. According to Fig. 5, it can be concluded that the optimum quantity

of added nanoclay to the structure of PUF for oil adsorption is 3wt%. Increasing the strength of foam

improves its ability to hold more oil in its structure, thereby increasing the adsorption capacity. But extra

cell opening increases the volume of pores, which, in turn, reduces the surface per volume of the

sorbent. Therefore, with the addition of nanoclay up to 3%wt, due to structure strengthening and

medium cell opening, the adsorption capacity is increased; however, the addition of nanoclay to more

than 3%wt, as it extremely open the foam cells, decreases the sorption capacity, which was more

12

sensible in high oil initial concentration. An extra increase of pores radius in nanocomposite sorbents

caused oil release during the sorbent removal from the polluted environment.

According to Fig. 5, the maximum oil removal percentage occurred in 40 g/L oil initial concentration. In

the low concentrations of oil, adsorption was increased due to the enhancement of toughness and

hydrophobicity of PU foam structure. Reduction of oil adsorption using nanocomposites observed in the

high concentrations of oil could be the result of the full opening of foam cells, causing the extra release

of oil from foam structure during adsorbent removal from the oily water.

Comparing the maximum light crude oil sorption capacity of the 3wt% NCPUF (21.5 g/g) with the

sorption capacity of commercial sorbents reported by Fingas et al [43], like polyester pads ( 9 g/g),

Polypropylene pads (8 g/g), vegetable fiber (4 g/g), peat moss (3 g/g), treated perlite (8 g/g) and clay

(kitty litter) (3 g/g), shows the acceptable capacity of the prepared sorbent for large scale applications.

3-6 Adsorption isotherms

Regression results of experimental data of crude oil adsorption of the samples, based on three

isotherm models, Langmuir, Freundlich and Redlich-Peterson, are illustrated in Fig.6 for blank PUF and

2wt% NCPUF. Adsorption isotherm parameters of these models are presented in Table 2 as well.

According to the R2 values of isotherm models, it is obvious that Redlich-Peterson and Langmuir

isotherms offer the best consistency with the obtained experimental data. The good consistency with

Langmuir isotherm suggests a monolayer and homogeneous adsorption on the surface of sorbent. By

considering the values of RL, it is clear that Langmuir model predicts the adsorption trend in low

concentrations better than in high concentrations of oil. There is a good agreement between

experimental data and Redlich-Peterson model with power bigger than 1, showing that besides surface

adsorption, there are also other reasons such as capillary properties which affect the adsorption process.

The presence of nanoclay in the nanocomposite sorbent structure increases KL in Langmuir isotherm and

13

demonstrates higher adsorption capacity in comparison with the PU foam sorbent. This decrease in β

parameter of Redlich-Peterson model is a result of enhanced structural homogeneity combined with

weak effects from other factors in the sorption process.

Comparing the curves of the experimental results with the classification proposed by Giles et al.

(1960) reveals close resemblance between the L curve and the obtained curve in this study [44]. This

type of system is considered to present the best curve and shows that the adsorbed molecules are likely

to be adsorbed flat, and the initial curvature shows that as more sites are occupied it becomes

increasingly difficult for a adsorbate molecule to find an empty site available (Fig. 6).

3-7 Oil removal efficiency of the blank PUF and nanocomposite adsorbents

Oil removal efficiency of sorbents is one of the important factors in their application as it influences

the efficiency of the whole separation process. High oil removal efficiency is the result of the enhanced

use of adsorbent surface. As shown in Fig. 7, the oil removal efficiency is increased with the increase of

cloisite 20A in nanocomposite structure. For example, the oil removal efficiency of the nanocomposite

sorbents with 2wt%, 3wt% and 4wt% of cloisite 20A in initial oil concentration of 20 g/L is increased

up to 28%, 55%, and 62%, respectively. At the low initial oil concentration, the enhancement of oil

removal efficiency is due to both oil sorption enhancement and lower water adsorption. The higher

efficiency is due to the lower water adsorption and the high initial concentration of crude oil at which

the nanocomposite sorbent has lower adsorption in comparison with the pure foam.

3-8 Regeneration of nanocomposite adsorbents

Regeneration and reusability of a sorbent in separation processes can be regarded as one of the

most important factors in selecting a sorbent. Reusability experiments reveal the enhanced adsorption at

low to medium initial concentrations of oil and lower adsorption in high initial concentrations

14

(Fig.8). As already mentioned, nanoclay in the structure of PUF, as the structural modifier has major

effects on the foam, led to more hydrophobicity, homogeneous structure and strength. More adsorption

in low to medium initial concentration could be assigned to a layer of toluene which covers the surfaces

and improves the oleophilicity. On the other hand, the lower sorption in high initial concentrations could

be due to the penetration of toluene into the foam structure and the reaction with isocyanate or polyol

chains which reduces the strength of foam structure. This structure softening during the removal of the

sorbent from the polluted water causes some shape reforming by the gravity force, leading to more oil

leakage off the foam.

The experimental results showed that the recovery of nanocomposites through washing with

petroleum solvent reduces oil removal efficiency. This is due to the lower oil adsorption as a result of

structural strength weakening after being washed with toluene (Fig. 9).

3-9 Adsorption isotherms of regenerated sorbents

Regression results of the experimental data which belong to the regenerated NCPUFs with 3wt%

and 4wt% of cloisite 20A are evaluated with isotherm models of Langmuir, Freundlich and Redlich-

Peterson and their respective parameters are reported in Table 3. These results show satisfying fit with

Langmuir and Redlich-Peterson models.

With regards to the calculated Langmuir model, RL, which is in the range of 0 to 1, it is clear that

Langmuir model is an appropriate model for predicting the sorption behavior of the sorbent, especially

in low concentrations of the oil. Consistency of data with Langmuir model shows the homogeneity and

monolayer adsorption on the surface of the sorbent.

According to the determined KL and β in the case of pure and nanocomposite sorbents, chemical

recovery leads to more adsorption capacity and consistency with Langmuir model assumptions.

Therefore, Langmuir model parameters are more suitable for the recovered sorbents

15

4- Conclusions

It is concluded from the results of this research that the oil sorption capacity has increased in low

initial oil concentrations, while it has decreased in high initial oil concentrations in the case of

nanocomposites with 2%wt and 4%wt of cloisite 20A. The nanocomposite with 3%wt clay has shown

increase in adsorption at all initial oil concentrations. The adsorption efficiencies of nanocomposites

have increased in comparison with the pure PU foam. The chemical regeneration of the pure foam has

increased the adsorption capacity and efficiency; while in the case of the nanocomposites, the adsorption

capacity and efficiency have decreased except for 3%wt nanocomposite, which has shown an increase in

efficiency. Experimental data has revealed good fit with Langmuir and Redlich-Peterson model.

16

References

[1] D. Wang, T. Silbaugh, R. Pfeffer, Y.S. Lin, Removal of emulsified oil from water by inverse

fluidization of hydrophobic aerogels, Powder Technology, 203 (2010) 298-309.

[2] A. Srinivasan, T. Viraraghavan, Oil removal from water using biomaterials, Bioresource

Technology, 101 (2010) 6594-6600.

[3] A.A. Al-Majed, A.R. Adebayo, M.E. Hossain, A sustainable approach to controlling oil spills,

Journal of Environmental Management, 113 (2012) 213-227.

[4] A. Dongil, B. Bachiller-Baeza, A. Guerrero-Ruiz, I. Rodríguez-Ramos, A. Martínez-Alonso, J.

Tascón, Surface chemical modifications induced on high surface area graphite and carbon nanofibers

using different oxidation and functionalization treatments, Journal of colloid and interface science, 355

(2011) 179-189.

[5] Y.V. Pokonova, Carbon adsorbents from petroleum residues, Fuel science & technology

international, 11 (1993) 875-882.

[6] I.W. Cumming, R.G. Holdich, I.D. Smith, The rejection of oil using an asymmetric metal microfilter

to separate an oil in water dispersion, Water Research, 33 (1999) 3587-3594.

[7] S. Al-Jeshi, A. Neville, An experimental evaluation of reverse osmosis membrane performance in

oily water, Desalination, 228 (2008) 287-294.

[8] C. López-Vazquez, C. Fall, Improvement of a Gravity Oil Separator Using a Designed Experiment,

Water, Air, & Soil Pollution, 157 (2004) 33-52.

[9] B. Reed, W. Lin, R. Jr., J. Young, Treatment of Oily Wastes Using High-Shear Rotary

Ultrafiltration, Journal of Environmental Engineering, 123 (1997) 1234-1242.

[10] J.C. Campos, R.M.H. Borges, A.M. Oliveira Filho, R. Nobrega, G.L. Sant’Anna Jr, Oilfield

wastewater treatment by combined microfiltration and biological processes, Water Research, 36 (2002)

95-104.

17

[11] C. Teas, S. Kalligeros, F. Zanikos, S. Stournas, E. Lois, G. Anastopoulos, Investigation of the

effectiveness of absorbent materials in oil spills clean up, Desalination, 140 (2001) 259-264.

[12] W. Scholz, W. Fuchs, Treatment of oil contaminated wastewater in a membrane bioreactor, Water

Research, 34 (2000) 3621-3629.

[13] C.C. Ho, C.Y. Chan, The application of lead dioxide-coated titanium anode in the electroflotation

of palm oil mill effluent, Water Research, 20 (1986) 1523-1527.

[14] A. Cybulski, J. Trawczyński, Catalytic wet air oxidation of phenol over platinum and ruthenium

catalysts, Applied Catalysis B: Environmental, 47 (2004) 1-13.

[15] X. Qi, Z. Jia, Y. Yang, Sorption capacity of new type oil absorption felt for potential application to

ocean oil spill, Procedia Environmental Sciences, 10 (2011) 849-853.

[16] P.C. Brandão, T.C. Souza, C.A. Ferreira, C.E. Hori, L.L. Romanielo, Removal of petroleum

hydrocarbons from aqueous solution using sugarcane bagasse as adsorbent, Journal of Hazardous

Materials, 175 (2010) 1106-1112.

[17] T.R. Annunciado, T.H.D. Sydenstricker, S.C. Amico, Experimental investigation of various

vegetable fibers as sorbent materials for oil spills, Marine Pollution Bulletin, 50 (2005) 1340-1346.

[18] Á. Cambiella, E. Ortea, G. Ríos, J.M. Benito, C. Pazos, J. Coca, Treatment of oil-in-water

emulsions: Performance of a sawdust bed filter, Journal of Hazardous Materials, 131 (2006) 195-199.

[19] A.L. Ahmad, S. Sumathi, B.H. Hameed, Residual oil and suspended solid removal using natural

adsorbents chitosan, bentonite and activated carbon: A comparative study, Chemical Engineering

Journal, 108 (2005) 179-185.

[20] D. Mysore, T. Viraraghavan, Y.-C. Jin, Treatment of oily waters using vermiculite, Water

Research, 39 (2005) 2643-2653.

18

[21] A. Gammoun, S. Tahiri, A. Albizane, M. Azzi, J. Moros, S. Garrigues, M. de la Guardia,

Separation of motor oils, oily wastes and hydrocarbons from contaminated water by sorption on chrome

shavings, Journal of Hazardous Materials, 145 (2007) 148-153.

[22] T. Viraraghavan, G.N. Mathavan, Treatment of oil-in-water emulsions using peat, Oil and

Chemical Pollution, 4 (1988) 261-280.

[23] A.L. Ahmad, S. Bhatia, N. Ibrahim, S. Sumathi, Adsorption of residual oil from palm oil mill

effluent using rubber powder, Brazilian Journal of Chemical Engineering, 22 (2005) 371-379.

[24] C.-M. Seah, S.-P. Chai, A.R. Mohamed, Synthesis of aligned carbon nanotubes, Carbon, 49 (2011)

4613-4635.

[25] N.G. Sahoo, Y.C. Jung, H.H. So, J.W. Cho, Synthesis of Polyurethane Nanocomposites of

Functionalized Carbon Nanotubes by in-situ Polymerization Methods, Korean Physical Society, 51

(2007) S1-S6.

[26] H. Marsh, F.R. Reinoso, Activated carbon, Elsevier, 2006.

[27] S.-J. Park, S.-Y. Lee, K.-S. Kim, F.-L. Jin, A novel drying process for oil adsorption of expanded

graphite.

[28] A. Hirsch, O. Vostrowsky, Functionalization of carbon nanotubes, in: Functional molecular

nanostructures, Springer, 2005, pp. 193-237.

[29] L. Han, J. Zhu, J. Kang, Y. Liang, Y. Sun, Catalytic wet air oxidation of high-strength organic

coking wastewater, Asia-Pacific Journal of Chemical Engineering, 4 (2009) 624-627.

[30] X.-F. Sun, SunSun, J.-X. Sun, Acetylation of Rice Straw with or without Catalysts and Its

Characterization as a Natural Sorbent in Oil Spill Cleanup, Journal of Agricultural and Food Chemistry,

50 (2002) 6428-6433.

[31] S. Keav, A. Martin, J. Barbier Jr, D. Duprez, Deactivation and reactivation of noble metal catalysts

tested in the Catalytic Wet Air Oxidation of phenol, Catalysis Today, 151 (2010) 143-147.

19

[32] H. Moazed, T. Viraraghavan, Use of Organo-Clay/Anthracite mixture in the separation of Oil from

Oily Waters, Energy Sources, 27 (2005) 101-112.

[33] R. Kenji, G. Takakiyo, G. Tomoki, I. Toru, U. Toru, H. Yoshiyuki, Oil- absorbent polymer and use

therefor, in, Google Patents, 1991.

[34] H. Li, L. Liu, F. Yang, Oleophilic Polyurethane Foams for Oil Spill Cleanup, Procedia

Environmental Sciences, 18 (2013) 528-533.

[35] M. Joulazadeh, A.H. Navarchian, Effect of process variables on mechanical properties of

polyurethane/clay nanocomposites, Polymers for Advanced Technologies, 21 (2010) 263-271.

[36] G. Harikrishnan, T.U. Patro, D.V. Khakhar, Polyurethane Foam−Clay Nanocomposites:  Nanoclays

as Cell Openers, Industrial & Engineering Chemistry Research, 45 (2006) 7126-7134.

[37] B.O. Ogunsile, A. Babarinde, K. Akinlolu, Adsorption of Malachite Green from Aqueous Solution

using Plantain Stalk (Musa paradisiaca).

[38] C. Aharoni, M. Ungarish, Kinetics of activated chemisorption. Part 2.-Theoretical models, Journal

of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases, 73 (1977)

456-464.

[39] K.Y. Foo, B.H. Hameed, Insights into the modeling of adsorption isotherm systems, Chemical

Engineering Journal, 156 (2010) 2-10.

[40] K. Okiel, M. El-Sayed, M.Y. El-Kady, Treatment of oil–water emulsions by adsorption onto

activated carbon, bentonite and deposited carbon, Egyptian Journal of Petroleum, 20 (2011) 9-15.

[41] F.-C. Wu, B.-L. Liu, K.-T. Wu, R.-L. Tseng, A new linear form analysis of Redlich–Peterson

isotherm equation for the adsorptions of dyes, Chemical Engineering Journal, 162 (2010) 21-27.

[42] J. Figueiredo, M. Pereira, M. Freitas, J. Orfao, Modification of the surface chemistry of activated

carbons, Carbon, 37 (1999) 1379-1389.

20

[43] M.F. Fingas, W.S. Duval, G.B. Stevenson, F.F. Slaney, Company, C.E.E. Branch, The Basics of Oil

Spill Cleanup: With Particular Reference to Southern Canada, Environmental Emergency Branch,

Environmental Protection Service, Environment Canada, 1979.

[44] C.H. Giles, T. MacEwan, S. Nakhwa, D. Smith, 786. Studies in adsorption. Part XI. A system of

classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in

measurement of specific surface areas of solids, J. Chem. Soc., (1960) 3973-3993.

[45] S. Suni, A.L. Kosunen, M. Hautala, A. Pasila, M. Romantschuk, Use of a by-product of peat

excavation, cotton grass fibre, as a sorbent for oil-spills, Marine Pollution Bulletin, 49 (2004) 916-921.

[46] D.C. Tuncaboylu, O. Okay, Preparation and characterization of single-hole macroporous organogel

particles of high toughness and superfast responsivity, European Polymer Journal, 45 (2009) 2033-2042.

[47] H. Li, L. Liu, F. Yang, Hydrophobic modification of polyurethane foam for oil spill cleanup,

Marine Pollution Bulletin, 64 (2012) 1648-1653.

21

Figures captions:

Fig. 1. FTIR spectra of blank polyurethane foam

Fig. 2. XRD pattern of cloisite 20A nanoclay (a) and nanocomposite adsorbent with 2wt% (b), 3wt% (c)

and 4wt% (d) of cloisite 20A nanoclay.

Fig. 3. SEM photographs of the blank PUF (a) and naonocomposite with 3wt% of nanoclay (b).

Fig. 4. The oil sorption capacity of blank PU foam cubes and NCPUF cubes in water–crude oil system.

Fig. 5. The effect of initial oil concentration on adsorption using PU foam and nanocomposite with

2wt%, 3wt% and 4wt% of nanoclay.

Fig. 6. Langmuir, Freundlich and Redlich-Peterson isotherms for the adsorption of crude oil onto the

blank PU foam (a) and NCPUF with 2 wt% of cloisite 20A nanoclay (b).

Fig. 7. Oil removal efficiency of nanocomposite polyurethane foam with 2wt%, 3wt% and 4wt% of cloisite

20A nanoclay.

Fig. 8. The effect of regeneration on the sorption performance of blank PU foam and naonocomposite

adsorbents with 3wt% and 4wt% of cloisite 20A nanoclay.

Fig. 9. Oil removal efficiency of fresh and recovered open-cell blank PU foam and nanocomposite with

3wt% and 4wt% of cloisite 20A nanoclay.

Fig. 1

600 1000 1400 1800 2200 2600 3000 3400

tra

nsm

itta

nce

(%

)

wavenumbers (cm-1)

Fig. 2

0 2 4 6 8 10

inte

nsi

ty (

a.u

)

2θ (°)

a

b

c

d

Fig.3.

Fig. 4

0

4

8

12

16

20

24

blank PUf NCPUf 2% NCPUf 3% NCPUf 4%

sorp

tion

cap

aci

ty (

g/g

)

Fig. 5

20

40

60

80

100

0 40 80 120 160 200

Pa

C0 (g/L)

A

A(2%)

A(3%)

A(4%)

(a) (b)

Figure6

Fig 7.

0

0.2

0.4

0.6

0.8

1

20(g/L) 40(g/L) 80(g/L) 120(g/L)

effi

cien

cy

C0(g/L)

A

A (2%20A)

A (3%20A)

A (4%20A)

Fig. 8

20

40

60

80

100

0 50 100 150 200

Pa

A

A"

20

40

60

80

100

0 50 100 150 200

Pa

A (3%20A)

A" (3%20A)

20

40

60

80

100

0 50 100 150 200

Pa

C0 (g/L)

A (4%20A)

A" (4%20A)

Fig. 9

0

0.2

0.4

0.6

0.8

1

20(g/L) 40(g/L) 80(g/L) 120(g/L)

effi

cien

cy

A

A"

0

0.2

0.4

0.6

0.8

1

1.2

20(g/L) 40(g/L) 80(g/L) 120(g/L)

eff

icie

ncy

A (4%20A)

A" (4%20A)

0

0.2

0.4

0.6

0.8

1

20(g/L) 40(g/L) 80(g/L) 120(g/L)

effi

cien

cy

Initial oil concentration

A (3%20A)

A" (3%20A)

22

Table 1 Oil sorption capacities of some sorbents

Oil sorbent Oil studied Sorption capacity

(g/g) Reference

Expanded perlite

Heavy crude

Light cycle

Up to 3.25

Up to 3.5

[11]

Cotton grass fiber

Diesel

Gasoline

20

19

[45]

Hydrophobic nano- silica

Diesel

Gasoline

14

15

[15]

Macroporousorganogel

Toluene

Crude oil

20.6

18.2

[46]

PUf

Kerosene

Diesel

10

23.45

[47]

PUf-LMA microshperes

Kerosene

Diesel

10.73

28.39

[47]

PUf-g-LMA

Kerosene

Diesel

20.97

37.64

[47]

23

PUf Light crude oil 18.5 This study

NCPUf 3% Light crude oil 21.5 This study

24

Table 2 Regression analysis for crude oil sorption by open-cell nanocomposites with 2wt%, 3wt% and

4wt%. of cloisite20A nanoclay and the parameters estimated using Langmuir, Freundlich and Redlich-

Peterson models.

Isotherms

adsorbent

PU foam

Blank

Nanocomposite

2%wt 3%wt 4%wt

Langmuir

qmax(g/g)

KL(L/g)

RL

R2

19.569

0.142

0.034 - 0.260

0.9895

20.325

0.146

0.033-0.255

0.9909

20.408

0.297

0.016 – 0.143

0.9918

17.699

0.166

0.029 – 0.232

0.9921

Freundlich

KF(g/g)/(g/L)n

n

R2

4.195

2.829

0.6662

0.632

1.451

0.9043

0.653

1.432

0.8779

0.744

1.574

0.8694

Redlich-Peterson

A(L/g)

B(L/g)β

β

R2

1.726

0.032

1.204

0.9939

1.843

0.034

1.200

0.9943

2.981

0.050

1.220

0.9956

1.744

0.040

1.180

0.9953

25

Table 3 Regression analysis for crude oil sorption by regenerated open-cell nanocomposites with 3wt%

and 4wt%. of 20A closite nanoclay and the parameters estimated using Langmuir, Freundlich and

Redlich-Peterson models.

Isotherm

Adsorbent

Blank PUF Nanocom 3%wt Nanocom4%wt

Langmuir

qmax(g/g)

KL(L/g)

RL

R2

21.692

0.232

0.021 – 0.177

0.997

20.284

0.259

0.019 – 0.162

0.9966

14.970

0.475

0.010 – 0.059

0.9986

Freundlich

KF(g/g)/(g/L)n

n

R2

5.767

3.149

0.6673

6.457

3.831

0.6215

7.649

7.342

0.397

Redlich-Peterson

A(L/g)

B(L/g)β

β

R2

3.682

0.110

1.090

0.9978

4.027

0.148

1.060

0.9972

5.368

0.309

1.03

0.9988

26

Highlights

• Nanocomposite of polyurethane foam with nanoclay was used as a novel oil sorbent.

• Different amounts of nanoclay in the foam were used to optimize its performance.

• Removal capacity and efficiency were improved up to 16% and 56%, respectively.

• Reusability feature of the prepared sorbent was investigated.