Chemical Engineering Journal 259 (2015) 611–619

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier .com/locate /cej

Organic–inorganic based nano-conjugate adsorbent for selectivepalladium(II) detection, separation and recovery

http://dx.doi.org/10.1016/j.cej.2014.08.0281385-8947/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel.: +81 791 58 2642; fax: +81 791 58 0311.E-mail addresses: rawual76@yahoo.com, awual.rabiul@jaea.go.jp (M.R. Awual).

Md. Rabiul Awual a,⇑, Md. Munjur Hasan b, Hussein Znad c

a Actinide Coordination Chemistry Group, Quantum Beam Science Center, Japan Atomic Energy Agency (SPring-8), Hyogo 679-5148, Japanb Shaheed Ziaur Rahman Medical College, Bogra 5800, Bangladeshc Department of Chemical Engineering, Curtin University, GPO Box U1987, Perth, WA 6845, Australia

h i g h l i g h t s

� Nano-conjugate adsorbent (NCA) isable to capture Pd(II) from acidicsolution.� The color formation indicates the

Pd(II) detection without using high-tech instruments.� The NCA has exhibited a high sorption

capacity due to spherical nanosizedcavities.� The NCA is reversible and reusable

without any significant deterioration.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 26 May 2014Received in revised form 8 August 2014Accepted 10 August 2014Available online 19 August 2014

Keywords:Nano-conjugate adsorbent (NCA)Palladium(II)Detection and separationSensitivity and selectivityHigh sorption and reusability

a b s t r a c t

The functional group containing organic ligand of N,N(octane-1,8-diylidene)di(2-hydroxy-3,5-dimethyl-aniline) (DHDM) was developed and then successfully anchored onto mesoporous silica for the prepara-tion of nano-conjugate adsorbent (NCA). After fabrication, the DHDM kept open functionality forcapturing palladium (Pd(II)) under optimum conditions. The NCA exhibited the distinct color formation(p–p transition) after adding the Pd(II) ions both in solid and liquid states. The solution pH played animportant role in the detection and sorption of Pd(II) but the prepared NCA worked well in the acidicpH region at 1.50. The data also clarified that the NCA did not form any color and signal intensity evenin the presence of diverse ions except Pd(II). The determined limit of detection to Pd(II) ions was lowas 0.14 lg/L. In Pd(II) sorption, the affecting factors such as solution pH, kinetics, isotherm models, com-peting ions and elution/regeneration were studied in detail. The NCA confirmed the rapid sorption prop-erty and the maximum sorption capacity was 213.67 mg/g due to spherical nanosized cavities with largesurface area and pore volume. The base metal of Cu(II) and Zn(II) did not hamper the Pd(II) sorption abil-ity of NCA in the acidic pH region. Therefore, it was expected that the Pd(II) could be separated from otherhard metal ions by the NCA. The data also clarified that the other competing metal ions did not decreasethe Pd(II) sorption capacity and NCA had almost no sorption capacity, which suggested the high selectiv-ity of Pd(II) ions by NCA. The adsorbed Pd(II) was eluted with 0.20 M HCl–0.20 M thiourea eluent andsimultaneously regenerated into the original form. The NCA was reversible and kept remaining function-ality for reuse in many cycles after an extraction/elution process without significant deterioration. There-fore the proposed NCA can be considered as a potential candidate for Pd(II) capturing from wastesamples.

� 2014 Elsevier B.V. All rights reserved.

612 M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619

1. Introduction

The precious metal consumption is increasing in recent yearsdue to the development of catalytic systems in automotive exhaustand the petrochemical industry. Of these precious metals,palladium (Pd(II)) is of great value in various fields with a widevariety of applications such as jewellery, fuel cells, automotive cat-alytic converters, pharmaceuticals, telecommunication, petroleumindustry, electronics, heat and corrosion resistance apparatus anddental medical devices [1–3]. Similarly, Pd(II) can bind to thiol-containing amino acids, proteins, DNA, and several biomoleculesand adversely affect the cellular processes [4]. In addition, fissionPd(II) is generated during the reprocessing of spent nuclear fuelsand high level radioactive liquid waste is increasing due to boomof nuclear power worldwide [5,6]. Furthermore, the release ofPd(II) and its accumulation in the environment can cause thehuman health problems such as eye irritation, skin problems, rhi-noconjunctivitis, asthma, etc. [7,8]. Therefore, the Pd(II) is strictlylimited to be 5–10 ppm level by the European Agency for the Eval-uation of Medicinal Products [9]. Thus, the research on the effectivemonitoring, separation and recovery of Pd(II) is of great impor-tance and in high demand.

Many analytical methods have been developed to determinethe presence of Pd(II) ions in clinical, environmental, industrial,and pharmaceutical samples such as spectrophotometry, atomicabsorption spectrometry, solid phase microextraction-high perfor-mance liquid chromatography, X-ray fluorescence, electrochemicalmethods, inductively coupled plasma-mass spectrometry (ICP-MS)and inductively coupled plasma-atomic emission spectrometry(ICP-AES) [10–14]. However, many of these are limited by instru-mentation cost, high training requirements, being cumbersome,time consuming and unsuitable, especially in developing or lessdeveloped countries [15,16]. From the stand point of analyticalchemistry, there is increasing demand to develop reliable, selec-tive, sensitive and robust methods to quantify and detect the Pd(II)ions in complex samples matrices. The reaction-based monitoringapproaches have demonstrated great advantages in-terms of selec-tivity and sensitivity for scientists [13,17,18]. They allow directanalysis of the specific elements by visual color optimization. How-ever, many of these sensing systems have been suffering from rig-orous test conditions and long response time. Therefore, the designof some novel and effective colorimetric methods for Pd(II) detec-tion that have a rapid response, are cost-effective, easy to use andcapable of being used in the field without reference solution is stillrequired.

The conventional method for Pd(II) separation is chemical pre-cipitation. This process has several disadvantages such as slowkinetics and solid–liquid-separation, huge chemical consumptionand high labor requirements. Also the process increases environ-mental pollution due to the use of toxic chemicals. Solvent extrac-tion (liquid–liquid extraction) is generally used for the separationand recovery of Pd(II) ions based on the soft donor atoms like ‘S’and ‘N’ containing ligands [19,20]. However, many of these ligandshave some drawback such as pH sensitivity, chemical stability andslower kinetics of extraction. The high cost of these compounds hasbeen a major hurdle in their use in the large-scale process forpotential applications in Pd(II) separation and recovery. There arealso several methods that have been reported to treat Pd(II) solu-tions such as electro-deposition, ion exchange, membrane separa-tion and adsorption [21–23]. Adsorption techniques have beenextensively applied in the fields of analytical chemistry and sepa-ration chemistry based on fast kinetics, with wide possibilities offunctionalization, high enrichment factors, less time consumptionand producing a low amount of secondary wastes [3,8,24]. Differ-ent materials, such as chelating resins, chemically-modified

activated carbon, functionalized silica, nanotubes, biomass orbio-adsorbent [20,25–31] have been used as adsorbents. However,most of the separation methods still suffer from low selectivity andsensitivity that limit their practical application. In connection tothis, we have used different ligand functionalized mesoporous sil-ica based materials for various metal ions monitoring and recoveryunder optimum conditions [11,16,26,32]. The ligand functionalizedmaterials have gained special attention due to their high surfacearea, long mechanical stability, high sorption efficiency, high selec-tivity and sensitivity.

In this study, we prepared functionalized ligand immobilizednano-conjugate adsorbent (NCA) based on organic–inorganic com-bination for efficient and sensitive Pd(II) detection and recoveryfrom waste samples. For preparation of the NCA, a N,N(octane-1,8-diylidene)di(2-hydroxy-3,5-dimethylaniline) (DHDM) wasincorporated into mesoporous silica by non-specific interactionvia hydrogen bonding, Van der Waals forces and reversible cova-lent bonds according to direct immobilization approach. Basedon Pearson Hard–Soft Acid–Base (HSAB) theory, the soft metal ionssuch as gold(III) and Pd(II) ions has high affinity to soft bases withdonor atoms as N > O [33]. The DHDM contained N and O donoratoms and exhibited a stable complex formation tendency withPd(II) at specific pH area. The detection data was also performedeven in the presence of diverse competing ions. The sorption andsensing experiments were carried out by both methods. Severalinfluencing parameters such as solution acidity, kinetics, initialmetal concentration, foreign ions, sorption capacity, color optimi-zation and elution condition were evaluated. The DHDM has excel-lent intramolecular charge transfer structure and desirablecolorimetrical properties and suitable for potential application inenvironmental and waste samples. It is also noted that the devel-oped materials based on direct organic–inorganic compositionare cost-effective and suitable to the large-scale treatment forPd(II) detection and recovery.

2. Materials and methods

2.1. Materials

All materials and chemicals were of analytical grade and used aspurchased without further purification. Tetramethylorthosilicate(TMOS), the triblock copolymers of poly(ethylene oxide-b-propyl-ene oxide-b-ethylene oxide) designated as F108 (EO141PO44EO141)and 1,8-Octanediol were obtained from Sigma–Aldrich CompanyLtd. USA. For pH adjustments in optical detection, buffer solutionsof 3-morpholinopropane sulfonic acid (MOPS), 2-(cyclohexyl-amino) ethane sulfonic acid (CHES) and N-cyclohexyl-3-aminopro-pane sulfonic acids (CAPS) were procured from Dojindo Chemicals,Japan, and KCl, HCl, NaOH from Wako Pure Chemicals, Osaka,Japan. The standard Pd(II) ions solutions, and other metal saltsfor the source of diverse metal ions were purchased from WakoPure Chemicals, Osaka, Japan. Ultra-pure water prepared with aMillipore Elix Advant 3 was used throughout in this work.

2.2. Synthesis and characterization of DHDM ligand

The structure and preparation of the N,N0-(octane-1,8-diylid-ene)di(2-hydroxy-3,5-dimethylaniline) (DHDM) is shown inScheme 1. The DHDM was prepared by the reaction of 1,8-Octane-diol (one moles) and 2-hydroxy-3,5-dimethylaniline (two mole) inethanol and a small amount of acetic acid was added. The resultantmixture was then heated under reflux for 4 h and left to cool atroom temperature. The solid formed upon cooling was collectedby suction filtration. The separated product was recrystallized

OHCH3

H3C NH2

OHCH3

H3C N N

HOCH3

CH3

OO

H

H

HH

Re f

lux

4h

Et O

H/A

ctic

acid

2 +

Scheme 1. The structure and preparation steps of N,N0-(octane-1,8-diylidene)di(2-hydroxy-3,5-dimethylaniline) (DHDM) ligand.

M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619 613

using dichloromethane/methanol 1/1. Then the purpose materialswere dried at 50 �C for 24 h. The purity of the DHDM was analyzedby CHN elemental analyses. The observed values (C24H34N2O2)were C, 75.32%; H, 8.88%; N, 7.31% and the calculated values wereC, 75.39%; H, 8.90%; N, 7.33%. The product was characterized by 1HNMR spectroscopy. The 1H NMR (400 MHz, CDCl3): d 1.29 (H,methylene), 1.52 (H, methylene), 2.15 (H, methyl), 2.34 (H,methyl), 5.34 (H, aromatic C–OH), 6.92 (H, benzene), 7.12 (H, ben-zene), 7.50 (H, aldimine). 13C NMR (400 MHz, CDCl3): d 15.4 (CH,CH3), 21.6 (CH, CH3), 26.3 (CH, CH2), 28.6 (CH, CH2), 29.1 (CH,CH2), 122.2 (CH, benzene), 129.3 (CH, benzene), 130.8 (CH, ben-zene), 132.2 (CH, benzene), 136.8 (CH, benzene), 149.1 (CH, ben-zene), 163.7 (CH, C-imine).

2.3. Analyses

The NMR spectra was obtained on a Varian NMR System400 MHz Spectrometer. Transmission electron microscopy (TEM)was obtained by using a JEOL (JEM-2100F) and operated at theaccelerating voltage of the electron beam at 200 kV. The TEM sam-ples were prepared by dispersing the powder particles in ethanolsolution using an ultrasonic bath and then dropped on coppergrids. The N2 adsorption–desorption isotherms were measuredusing the 3Flex analyzer (Micromeritics, USA) at 77 K. The pore sizedistribution was measured from the BJH adsorption. Mesoporoussilica and NCA were pre-treated at 100 �C for 3 h under vacuumuntil the pressure was equilibrated to 10�5 Torr before the N2 iso-thermal analysis. The specific surface area (SBET) was measured byusing multi-point adsorption data from linear segment of the N2

adsorption isotherms using Brunauer–Emmett–Teller (BET) theory.The absorbance spectrum was measured by UV–Vis–NIR spectro-photometer (Shimadzu, 3700). The metal ions concentrations weremeasured by ICP-AES (SII NanoTechnology Inc.). The ICP-AESinstrument was calibrated using five standard solutions containing0, 0.5, 1.0, 1.5 and 2.0 mg/L (for each element), and the correlationcoefficient of the calibration curve was higher than 0.9999. In addi-tion, sample solutions having complicated matrices were not usedand no significant interference of matrices was observed.

2.4. Preparation of mesoporous silica and NCA

The mesoporous silica was prepared following the reportedmethods with a slight modification [34]. The preparation of meso-porous silica monoliths procedure was involved by adding of TMOSand triblock copolymers to obtain a homogenized sol–gel mixturebased on the F108/TMOS mass ratio. The liquid crystal phase was

achieved after quick addition of acidified solution and to promotehydrolysis of the TMOS around the liquid crystal phase assembly ofthe triblock copolymer surfactants. In typical conditions, the com-position mass ratio of F108:TMOS:HCl/H2O was 1.3:2:1, respec-tively. The homogeneous sol–gel synthesis was achieved bymixing F108/TMOS in a 200 mL beaker and then shaking at 60 �Cuntil homogeneous. The exothermic hydrolysis and condensationof TMOS occurred rapidly by addition of acidified aqueous solutionof HCl (at pH = 1.3) to this homogeneous solution. The methanolproduced from the TMOS hydrolysis was removed by a vacuumpump connected to a rotary evaporator at 45 �C. Then the materialswere dried at 45 �C temperature for 24 h to complete the dryingprocess. The organic moieties were removed by calcination at520 �C for 6 h under the normal atmosphere. After calcinations,the material was ground properly and ready to use for conjugationof organic ligand of DHDM.

The DHDM (50 mg) was dissolved in ethanol solution and 1.0 gmesoporous silica was added into the solution. Then the immobi-lization procedure was performed under vacuum at 30 �C untilDHDM saturation was achieved. The ethanol was removed by avacuum connected to a rotary evaporator at 45 �C and the resultingmaterial was washed with warm water to check the stability andelution of DHDM from mesoporous silica. The material was driedat 45 �C for 6 h and ground to fine powder for Pd(II) detectionand recovery operations. The DHDM immobilization amount(0.08 mmol/g) was determined by the following equation:

Q ¼ ðC0 � CÞ V=m ð1Þ

where Q is the adsorbed amount (mmol/g), V is the solution volume(L), m is the mass of inorganic silica materials (g), C0 and C are theinitial concentration and supernatant concentration of the DHDM,respectively.

2.5. Pd(II) sensing

In Pd(II) sensing, the NCA was immersed in a mixture of specificPd(II) ions concentrations (2.0 mg/L) and adjusted at appropriatepH of 1.0, 1.50, 3.50 (0.2 M of KCl with HCl), 5.00 (0.2 M CH3COO-HCH3COONa with HCl) and 7.01 (0.2 M MOPS with NaOH) in thespecific amount of the NCA (8 mg) at constant volume (10 mL)with shaking in a temperature-controlled water bath with amechanical shaker at 25 �C for 10 min at a constant agitation speedof 110 rpm to achieve good color separation. A blank solution wasalso prepared, following the same procedure for comparison withoptimum color difference. After color optimization, the solid mate-rials were filtered using Whatman filter paper (25 mm; Shibata fil-ter holder) and used color for assessment and signal intensity usingsolid-state UV–Vis–NIR spectrophotometer. The NCA was groundto fine powder to achieve homogeneity in the absorbance spectra.The limit of detection (LD) for Pd(II) ions was determined from thelinear part of the calibration plot according to the following equa-tion [26]:

LD ¼ KSb=m ð2Þ

where, K value is 3, Sb is the standard deviation for the blank and mthe slope of the calibration graph in the linear range, respectively.

2.6. Pd(II) sorption, elution/recovery and reuses studies

In sorption operations, the NCA was also immersed in Pd(II)ions concentrations and adjusted at specific pH values by addingof HCl or NaOH in 20 mL solutions. After stirring for 1 h at roomtemperature, the solid materials were separated by filtration sys-tem and Pd(II) concentrations in before and after sorption opera-tions were analyzed by ICP-AES. The batch single andmulti-component sorption experiments were performed at room

614 M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619

temperature (25 �C). During the sorption operation, the amount ofPd(II) adsorbed was determined according to the followingequations:

Mass balance qe ¼ ðC0 � Cf Þ V=M ðmg=gÞ ð3Þ

and metal ion uptake efficiency Re ¼ ðC0 � Cf ÞC0

� 100 ð%Þ ð4Þ

where V is the volume of the aqueous solution (L), and M is theweight of the NCA (g), C0 and Cf are the initial and final concentra-tions of Pd(II) ions in solutions, respectively.

To determine the kinetics performances, 10 mg of NCA wasadded to 20 mL solution containing 5.0 mg/L of Pd(II) ions. Themixture was then stirred, and solid materials were separated bythe filtration system at different time intervals, and the filtratesolution was analyzed by ICP-AES. In case of maximum removal,10 mg of NCA was also added in different concentration of Pd(II)ions and stirred for 3 h and filtrate solutions were analyzed byICP-AES.

To evaluate the most efficient eluting agent, first 30 mL of4.0 mM Pd(II) ion solution was adsorbed on the 30 mg NCA andthen elution experiments were carried out using various concen-trations of HCl acid solutions. Also the adsorbed Pd(II) ions ontoNCA was washed with deionized water several times and trans-ferred into 50 mL beaker. The washing solution was also checkedby ICP-AES. To perform the elution operation, 5 mL of the elutingagent was added, and then the mixture was stirred for 20 min.The concentration of Pd(II) ions released from the NCA into aque-ous phase was analyzed by ICP-AES. Then the NCA was reused sev-eral cycles to investigate the reusability from the stand point oflong-term use as cost-effective material.

All experiments in this study were duplicated to assure the con-sistency and reproducibility of the results.

3. Results and discussion

3.1. Mesoporous silica and nano-conjugate material (NCA)

The N2 isotherms exhibited typical IV adsorption behavior withH2 hysteresis loops of the mesoporous silica and functionalizedNCA. The N2 adsorption–desorption isotherms data are shown inFig. 1(a). The mesoporous silica exhibited appreciable surface mor-phology based on the surface area, pore size and pore volumes

Fig. 1. N2 adsorption–desorption isotherms of (a) mesoporous silica monolith and(b) nano-conjugate adsorbent (NCA) with different surface area, pore size and porevolumes.

measurement. However, the surface area of the NCA was reducedafter functionalization with DHDM ligand due to the anchoringof the DHDM molecules inside pore of the mesoporous silica(Fig. 1(b)). The data also clarified that the NCA had the higher sur-face area compared with the other materials which had compara-ble surface areas and pore volumes.

The mesoporous silica surface morphology and diameter wereobserved by using SEM, and the images are shown in Fig. 2(Aand B). The particles were uniform which indicated that the directtemplating method was believed to generate the bulk form of mes-oporous silica. It is also noted that the present rapid synthesis mes-oporous silica monoliths have an average particle size of 200 lm[32,34]. The TEM images are shown in Fig. 2(C and D), which showsthe high ordered mesoporous structure and well-arranged as mes-ocage structure of the mesoporous silica. The framework porosityconfirmed the interconnected regular pore distribution system ofthe silica material. Based on the observation, the TEM images werehexagonal arrays of mesopores and the silica has been typical hex-agonal high ordered mesoporous silica [26]. The NCA was fabri-cated by direct immobilization approach. The organic ligand ofDHDM has strong electrostatic interactions with abundant hydro-xyl containing mesoporous surface via hydrogen bonding, Van derWaals forces and reversible covalent bonds. The TEM images arealso shown in Fig. 2(E and F) and the structure of DHDM modifiedNCA exhibited well-ordered mesopores. Therefore, the DHDMligand has the ability to achieve flexibility in the specific activityof the electron acceptor or donor strength of the chemicallyresponsive visual color signal in detection and sorption operations.

3.2. Pd(II) detection parameters

In optical chemical sensing, the development of receptors as‘‘indicator ligands’’ and surface-confinement materials as ‘‘carri-ers’’ is the key to broadening the applications of optical chemicalsensors/materials for wide-range detection of Pd(II) ions. Afteradding the NCA into the trace level of Pd(II) containing solutionunder optimum conditions, the NCA formed the color based oncharge transfer (intense p–p transition) transduction, which canbe observed by visual observation by the naked-eye. Among thevariable parameters, the solution pH is the most important factorfor detection of metals by the solid state mesoporous silica NCA.To evaluate the effect of solution acidity, the Pd(II) detection wasinvestigated at a wide range from 1.0 to 7.0. The signal intensityafter adding Pd(II) ions of the NCA was recorded at k = 398 nm todetermine the highest intensity difference between blank NCAand Pd(II) containing NCA. The signal intensity was high in theacidic region as judged from Fig. 3. Also the color change wasobserved by the indicator ligand (DHDM) in the liquid state asshown in Fig. 3 (inset). However, by increasing the pH value, thesignal intensity was decreased. Based on the results of the experi-ment, the maximum signal intensity was observed at pH 1.50. Inthis pH region, the specific color change indicated that the strongtendency to form stable complexation [Pd(II)–DHDM]n+. Then thesolution pH was adjusted to 1.50 to investigate the other subse-quent detection parameters in this study.

The signal intensity of NCA was increased upon addition ofPd(II) ions to investigate the various concentration Pd(II) detectionbased on color optimization by the prepared NCA. Increasing thesignal intensity with color optimization at different concentrationof Pd(II) ions is shown in Fig. 4(A). However, the observed signalintensity (at k = 398 nm) was not linearly proportional to the highconcentration level of Pd(II) ions. On the other hand, the remark-able signal intensity as a shoulder peak was observed accordingto the color change from the yellowish to blackish, which canclearly be perceived by the naked eye. The data also clarified thatthe color change upon Pd(II) concentration is a simple procedure

Fig. 2. SEM images of mesoporous silica (A and B); TEM images with ordered mesoporous silica (C and D) and TEM images of DHDM immobilized nano-conjugate adsorbent(E and F).

Fig. 3. The effect of solution pH in Pd(II) detection by the NCA under different pHarea. The equilibrated individually at different pH conditions with 2.0 mg/L of Pd(II)ions at 25 �C. The standard deviation was >3.0% for the analytical data of triplicateanalyses.

M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619 615

for the sensitive Pd(II) detection without using high-tech instru-ments. The color change tendency with different concentration ofPd(II) ions of NCA are presented in Fig. 4(A) (inset).

According to the color optimization of the NCA, the Pd(II) iondetection limit was evaluated from the linear part of the calibra-tion curve. The calibration curve is shown in Fig. 4(B). The dataexhibited the linearity when the Pd(II) concentration was from0.0282 to 0.939 lM (Fig. 4(B) inset). From the linear relationship,the correlation coefficient value was R2 = 0.9994. The regressionequation is Y = 0.09542 + 0.035669 X. According to the definitionof detection limit (Eq. (2)), the limit of detection for Pd(II) was0.14 lg/L. Increasing the signal intensity over 0.939 lM of Pd(II)concentration was nonlinear because of the saturation effects.

Therefore, the sensitive Pd(II) detection is possible in low level ofPd(II) concentrations by the NCA. The selective Pd(II) detectionwas also possible even in the presence of the several matrices (cal-ibration curve with the dotted line) as judged from Fig. 4(B).

The effects of competing ions, which often exists in the diverseions competition, on the detection of Pd(II) by the NCA was evalu-ated. However, the NCA formed yellowish to blackish color uponPd(II) addition, while other ions such as Ag+, Al3+, Cd2+, Co2+,Zn2+, Fe3+, Hg2+, Cu2+, Mg2+, Ni2+, Ca2+, Ru3+ and Pt2+ gave no visiblecolor changes as judged from Fig. 5(A) (inset). Large amounts ofcompeting metal ions had no interferences on Pd(II) detection,probably due to the lower tendency to form complex with DHDMligand at this pH region. Also the signal intensity of NCA with thedifferent metal ions was recorded, and data are shown inFig. 5(A). The data also clarified that the significant intensity wasnot observed upon the addition of these metal ions. Also the anionsdid not show any interfering effect based on signal intensity underthe optimum conditions. Therefore, the NCA was characteristic ofhigh selectivity toward Pd(II) ions over other foreign ions. Thestrong coordination between Pd(II) and DHDM is also illustratedin Scheme 2 based on the stable complex formation under theseoptimum conditions.

3.3. Pd(II) sorption parameters

In different pH area, Pd(II) can form stable chloride, hydroxy-chloride and hydroxide complexes, which are PdCl+, PdCl2, PdCl3

�,PdCl4

2�, PdCl3(OH)2�, PdCl2(OH)22�, PdCl(OH)3

2�, Pd(OH)2 andPd(OH)4

2� [20,35]. Therefore, the effect of solution pH on Pd(II)sorption by the ligand functionalized NCA was investigated. ThePd(II) sorption data under different pH solution are shown inFig. 5(B). The maximum Pd(II) sorption was observed at pH 1.50.The data also suggested that the Pd(II) sorption decreases overpH 1.5. It is also noted that the hydroxyl and hydroxy-chloride spe-cies are formed when the pH above 5.0. However, the ligand immo-bilized functional silica have a specific sorption ability at thespecific pH are for specific metal ions [32]. To evaluate the other

Fig. 4. (A) The color formation with increasing the Pd(II) ions concentrations corresponds to the increasing of signal intensity and (B) calibration plot with signal intensitymeasured at 398 nm with different concentrations of Pd(II) ions. The inlets in graphs (B) show the limit of detection with a linear fit in the linear concentration ranges. The Aand A0 are the signal intensity of the NCA after and before addition of Pd(II) ions. The dotted line represents the calibration plot of the Pd(II) ions in the presence of diversematrices. The error bars denote a relative standard deviation of P4.0% range for the analytical data of duplicate analyses.

Solution pH2 4 6 8 10

Eff

icie

ncy

(%)

0

20

40

60

80

100

(A) (B)

Fig. 5. (A) Effect of diverse ions to NCA under optimum conditions in color formation and signal intensity change. The competing ions and Pd(II) concentration was 20.0 and2.0 mg/L, respectively and (B) the Pd(II) sorption by the NCA in different solution acidity The RSD value was �4.0%.

N N

HO

H3C

CH3

HH

CH3

H3C

OH

N N

H3C

CH3

HH

CH3

H3C

OPd

Pd(II) at pH 1.50

0.20 M HCl &0.20 M Thiourea

O

Cl

Cl

Scheme 2. The possible stable complex formation of Pd(II) and DHDM at pH 1.50 and elution/regeneration with 0.20 M HCl–0.20 M thiourea.

616 M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619

sorption parameters, the solution pH was adjusted to 1.50 for sub-sequent Pd(II) sorption operations.

Sorption kinetic is strongly related with the physical features ofthe solid-state materials and especially the surface morphologysuch as surface area, pore size distribution of the material. There-fore, sorption rate based on contact time between adsorbate andmetal ions is an important parameter from the practical standpoint of practical use of materials. The influence of contact timeon Pd(II) sorption by the NCA is shown in Fig. 6(A). The Pd(II) sorp-tion efficiency was varied with the contact time. The data alsorevealed that the Pd(II) sorption was increased sharply with

increasing the contact time. However, more than 55% of the Pd(II)ions was absorbed in the first 10 min and equilibrium was reachedin only 30 min. The initial sorption was relatively fast, and such afast kinetic will benefit a highly efficient metal sorption. Hence,the contact time of 3 h ascertained the equilibrium Pd(II) sorptionfor high concentration of initial concentration to determine themaximum sorption capacity.

In order to understand the sorption behavior by the functional-ized NCA, the sorption ability was evaluated with increasing theinitial Pd(II) concentration. With increasing the initial Pd(II) con-centration, the sorption capacity was increased followed with a

Time in min.0 20 40 60 80

Eff

icie

ncy

(%)

0

20

40

60

80

100 (A) (B)

Fig. 6. (A) Effect of contact time for maximum Pd(II) sorption when Pd(II) concentration was 5.0 mg/in 20 mL and (B) effect of initial Pd(II) concentration and the Langmuirsorption isotherms with linear form (initial Pd(II) concentration range from 2.10 to 72.10 mg/L; solution pH 1.50; dose amount 10 mg; solution volume 30 mL and contacttime 3 h).

Table 1Comparison of Pd(II) sorption capacities using different forms of functionalizedmaterials.

Used functionalized materials Capacity (mg/g) Refs.

Modified silica gel 77.69 [3]Calixcrown grafted micro-sized silica 163.10 [6]Functionalized silica 83.00 [8]Amberlite XAD-2010 resin 12.80 [10]Conjugate adsorbent 164.20 [11a]Sensor ensemble adsorbent 163.13 [16a]MFT-resin 15.29 [20]Functionalized nanofibers 52.60 [23]NH2-MCM-41 285.00 [28]NH2-MCM-48 305.00 [28]Indian almond 47.17 [30]Protein-rich biomass 104.00 [31]Mesoporous adsorbent 172.53 [32b]Nano-conjugate material 213.67 This study

M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619 617

slow down to equilibrium capacity as judged from Fig. 6(B). Themaximum sorption capacity based on the experimental data was195.02 mg/g when the initial Pd(II) concentration was 72.10 mg/L. However, the NCA still had the potential to obtain higher sorp-tion capacity if the Pd(II) concentration continued to increasebeyond the initial amount. Therefore, it is important to study thesorption isotherms for understanding the Pd(II) distribution ontothe NCA surface. However, the simple Langmuir sorption isother-mal models are often successfully employed onto the ordered func-tionalized mesoporous materials. Moreover, the Langmuir modelrepresents the nonlinear sorption and suggests that metal sorptionoccurs on a homogeneous surface by monolayer sorption withoutinteraction between adsorbed species. The linear form of the Lang-muir model is given by following equation:

Ce=qe ¼ 1=ðKLqmÞ þ ð1=qmÞCe ðlinear formÞ ð5Þ

where Ce is the equilibrium Pd(II) concentration in the solution (mg/L) and qe (mg/g) is the sorption capacity of Pd(II) in the equilibriumstate. The fitted parameters qm (mg/g) and KL (L/mg) are the maxi-mum monolayer sorption capacity and the Langmuir sorption con-stant, respectively and can be calculated from the intercept andslope of the linear plot, with Ce/qe versus Ce as shown in Fig. 6(B)(inset). A plot of Ce/qe versus Ce over the entire concentration rangeproduces a straight line, which is an indication of the applicabilityof the Langmuir isotherm for the system under consideration. Themaximum sorption capacity was calculated to be 213.67 mg/g,which was close to the experimental data. The correlation coeffi-cients value (R2 = 0.9951) revealed that the Langmuir sorptionequation can be considered an accurate model and the sorptionmode of Pd(II) by the NCA may be mainly classified as monolayersorption coverage. Also the Langmuir sorption constant KL relatedto the affinity of the adsorbate, and the KL value was 1.16 L/mg.The high KL value implied the higher sorption capacity by theNCA. To predict the sorption efficiency of the sorption process, thedimensionless constant separation factor or equilibrium parameter,R was used; which is defined as:

R ¼ 1=1þ KLC0 ð6Þ

where, KL is the Langmuir constant and C0 is the initial concentra-tion of Pd(II). Values within the range 0 < R < 1 indicate favorablesorption. The NCA in the present study gave R values below the unit,which mean that the Pd(II) sorption on the NCA was favorable.

The Pd(II) sorption capacity is compared with various forms ofmaterials in Table 1 [3,6,8,11a,16a,20,23,28,30–32b]. The Pd(II)sorption capacity by the present NCA was relatively high due to

the spherical nanosized cavities with large surface area when com-pared to other materials. The differences of Pd(II) sorption onmaterials are due to the properties of function groups, surface mor-phology such as surface area, pore size and pore volumes of thefunctionalized materials.

The Pd(II) sorption was also evaluated in the presence of diverseions by the NCA. The Cu(II) and Zn(II) are base metals ions, whichare highly affected by Pd(II) sorption operation. Therefore, this willbe a great advantage to the NCA if Pd(II) is separated from thesetwo base metal ions. From the detection operation, these twometal ions were not affected on the Pd(II) detection system. There-fore, it is expected that the Pd(II) can be separated from other hardmetal ions by the NCA. Then the sorption studies were performedin the presence of diverse metal ions such as Cu2+, Zn2+, Ag+, Al3+,Cd2+, Co2+, Fe3+, Hg2+, Mg2+, Ni2+, Ca2+, Ru3+ and Pt2+ along withPd(II) ions. The sorption of competing metal ions is shown inFig. 7(B). The data clarified that the presence of competing metalions were not adversely affected the Pd(II) sorption even their con-centration was 5-fold higher than the Pd(II) concentration. In addi-tion, no masking reagents were used in this sorption operation, andthe Pd(II) sorption was higher than 95% indicating that the pre-sented NCA exhibited high selectivity to Pd(II) under optimumconditions.

The extraction/elution efficiency was evaluated on the ratio tothe mass desorbed from the NCA to the mass Pd(II) adsorbed ontothe NCA. Also the high sorption is not only the main performanceof the material but also elution is the significant character of the

Cu Zn Cd Fe Co Hg Al Mg Ni Ag Ca Ru PtPd+ions0

20

40

60

80

100Ef

ficie

ncy

(%)

Metal ions

(A) (B)

Fig. 7. (A) The effect of diverse metal ions sorption along with Pd(II) by the NCA. The competing metal ions were 10 mg/L and Pd(II) concentration was 2.0 mg/L and (B)extraction/elution operation with 0.20 M HCl–0.20 M thiourea after Pd(II) sorption by the NCA within ten reuses cycles. The RSD values were �5.0%.

618 M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619

potential material. Then the elution experiment was performed inbatch approach and suitable eluent of 0.20 M HCl–0.20 M thiourea(5.0 mL). The Pd(II) sorption and elution profiles are depicted inFig. 7(B). According to experimental observation, the similar sorp-tion and elution values were evaluated, and the NCA could usemany cycles as a cost-effective material in practical operation.However, the sorption efficiency was slightly decreased after the10th cycle of sorption–elution–regeneration operations.

4. Conclusions

This work demonstrated the Pd(II) detection and recovery usingfunctionalized ligand anchored nano-conjugate adsorbent (NCA).The ligand of N,N0-(octane-1,8-diylidene)di(2-hydroxy-3,5-dimeth-ylaniline) (DHDM) was synthesized and directly immobilized ontomesoporous silica for the preparation of NCA. The NCA formed sig-nificant color intensity based on the charge transfer transduction(p–p transition). The enhancement of color and signal intensityin the presence of Pd(II) was observed, and NCA could detect Pd(II)visible by naked-eye observation without using sophisticatedinstruments. The limit of detection to Pd(II) was 0.14 lg/L withoutsignificant interference of competing ions. Therefore, NCA can eas-ily identify the Pd(II) containing solution in the acidic regionregardless of the sample matrix and without pretreatment, indicat-ing potential on-site applications. The NCA followed the monolayersorption model described by Langmuir and the maximum sorptioncapacity was 213.67 mg/g. The presence of diverse competing ionsdid not affect the Pd(II) sorption capacity by the NCA. The NCAcould be regenerated and reused for many cycles without signifi-cant loss in its original performances using suitable the eluent of0.20 M HCl–0.20 M thiourea. Therefore, NCA is an efficient, eco-friendly material for Pd(II) capturing from waste samples. Thefunctionalized NCA possessed research value and applicationpotential for the simultaneous detection and recovery of Pd(II)from acidic media.

Acknowledgments

This research was partially supported by the Grant-in-Aid forResearch Activity Start-up (24860070) from the Japan Society forthe Promotion of Science. The authors also wish to thank the anon-ymous reviewers and editor for their helpful suggestions andenlightening comments. The authors are also greatly

acknowledged to Christopher Paul Taylor for his selfless Englishlanguage editing.

References

[1] T. Iwasawa, M. Tokunaga, Y. Obora, Y. Tsuji, Homogeneous palladium catalystsuppressing Pd black formation in air oxidation of alcohols, J. Am. Chem. Soc.126 (2004) 6554–6555.

[2] J.A. Rodridguez, Physical and chemical properties of bimetallic surfaces, Surf.Sci. Rep. 24 (1996) 223–287.

[3] R.K. Sharma, A. Pandey, S. Gulati, A. Adholeya, An optimized procedure forpreconcentration, determination and on-line recovery of palladium usinghighly selective diphenyldiketone-monothiosemicarbazone modified silica gel,J. Hazard. Mater. 209–210 (2012) 285–292.

[4] International Programme on Chemical Safety, Palladium; EnvironmentalHealth Criteria Series 226, World Health Organization, Geneva, 2002.

[5] Z. Kolarik, E.V. Renard, Recovery of value fission platinoids from spent nuclearfuel Part I: general considerations and basic chemistry, Platinum Metal Rev. 47(2003) 74–87.

[6] Y. Leng, J. Xu, J. Wei, G. Ye, Amino-bearing calixcrown receptor grafted tomicro-sized silica particles for highly selective enrichment of palladium inHNO3 media, Chem. Eng. J. 232 (2013) 319–326.

[7] J. Kielhorn, C. Melber, D. Keller, I. Mangelsdorf, Palladium – a review ofexposure and effects to human health, Int. J. Hyg. Environ. Health 205 (2002)417–432.

[8] F. Bai, G. Ye, G. Chen, J. Wei, J. Wang, J. Chen, Highly selective recovery ofpalladium by a new silica-based adsorbent functionalized with macrocyclicligand, Sep. Purif. Technol. 106 (2013) 38–46.

[9] C.E. Garrett, K. Prasad, The art of meeting palladium specifications in activepharmaceutical ingredients produced by Pd-catalyzed reactions [review], Adv.Syn. Catal. 346 (2004) 889–900.

[10] H. Serencam, V.N. Bulut, M. Tufekci, A. Gundogdu, C. Duran, S. Hamza, M.Soylak, Separation and pre-concentration of palladium(II) from environmentaland industrial samples by formation of a derivative of 1,2,4-triazole complexon amberlite XAD-2010 resin, Int. J. Environ. Anal. Chem. 93 (2013) 1484–1499.

[11] (a) M.R. Awual, T. Yaita, S.A. El-Safty, H. Shiwaku, Y. Okamoto, S. Suzuki,Investigation of palladium(II) detection and recovery using ligand modifiedconjugate adsorbent, Chem. Eng. J. 222 (2013) 172–179;(b) M.R. Awual, M.A. Khaleque, Y. Ratna, H. Znad, Simultaneous ultra-tracepalladium(II) detection and recovery from wastewater using new class meso-adsorbent, J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.02.053.

[12] C. Locatelli, D. Melucci, G. Torsi, Determination of platinum-group metals andlead in vegetable environmental bio-monitors by voltammetric andspectroscopic techniques: critical comparison, Anal. Bioanal. Chem. 382(2005) 1567–1573.

[13] W.X. Ren, T. Pradhan, Z. Yang, Q.Y. Cao, J.S. Kim, Rapid responsive palladiumsensor under mild condition, Sens. Actuators B Chem. 171–172 (2012) 1277–1282.

[14] L.A. Simpson, R. Hearn, T. Catterick, The development of a high accuracymethod for the analysis of Pd, Pt and Rh in auto catalysts using a multi-collector ICP-MS, J. Anal. Atom. Spectrom. 19 (2004) 1244–1251.

[15] Z.X. Li, Y.H. Feng, Determination of the platinum group elements in geologicalsamples by ICP-MS after NiS fire assay and Te coprecipitation: ultrasound

M.R. Awual et al. / Chemical Engineering Journal 259 (2015) 611–619 619

assisted extraction of PGEs from Te precipitate, J. Anal. Atom. Spectrom. 21(2006) 90–93.

[16] (a) M.R. Awual, T. Yaita, Rapid sensing and recovery of palladium(II) usingN,N-bis(salicylidene)1,2-bis(2-aminophenylthio)ethane modified sensorensemble adsorbent, Sens. Actuat. B Chem. 183 (2013) 332–341;(b) M.R. Awual, M.M. Hasan, Fine-tuning mesoporous adsorbent forsimultaneous ultra-trace palladium(II) detection, separation and recovery, J.Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.03.013.

[17] A.L. Garner, F. Song, K. Koide, Enhancement of a catalysis-based fluorometricdetection method for palladium through rational fine-tuning of the palladiumspecies, J. Am. Chem. Soc. 131 (2009) 5163–5171.

[18] (a) M.R. Awual, T. Yaita, H. Shiwaku, Design a novel optical adsorbent forsimultaneous ultra-trace cerium(III) detection, sorption and recovery, Chem.Eng. J. 228 (2013) 327–335;(b) M.R. Awual, M. Ismael, T. Yaita, S.A. El-Safty, H. Shiwaku, Y. Okamoto, S.Suzuki, Trace copper(II) ions detection and removal from water using novelligand modified composite adsorbent, Chem. Eng. J. 222 (2013) 67–76;(c) M.R. Awual, M.M. Hasan, A novel fine-tuning mesoporous adsorbent forsimultaneous lead(II) detection and removal from wastewater, Sens. Actuat. BChem. 202 (2014) 395–403.

[19] R. Ruhela, J.N. Sharma, B.S. Tomar, M.S. Murali, R.C. Hubli, A.K. Suri,Dithiodiglycolamide: novel ligand with highest selectivity and extractabilityfor palladium, Tetrahedr. Lett. 52 (2011) 3929–3932.

[20] E. Birinci, M. Gulfen, A.O. Aydın, Separation and recovery of palladium(II) frombase metal ions by melamine–formaldehyde–thiourea (MFT) chelating resin,Hydrometallurgy 95 (2009) 15–21.

[21] T. Kakoi, M. Goto, F. Nakashio, Separation of platinum and palladium by liquidsurfactant membranes utilizing a novel bi-functional surfactant, J. Membr. Sci.120 (1996) 77–88.

[22] V.N. Bulut, M. Tufekci, C. Duran, M. Soylak, H. Kantekin, Selective solid phaseextraction for separation and preconcentration of palladium from gold ore andanode slime after complexation with a N4O2 mixed donor ligand derivative,Clean – Soil Air Water 38 (2010) 678–683.

[23] O.E. Fayemi, A.S. Ogunlaja, P.F.M. Kempgens, E. Antunes, N. Torto, T. Nyokong,Z.R. Tshentu, Adsorption and separation of platinum and palladium bypolyamine functionalized polystyrene-based beads and nanofibers, Miner.Eng. 53 (2013) 256–265.

[24] D.P. Chen, C. Yu, C.Y. Chang, Y. Wan, J.M.J. Frechet, W.A. Goddard, M.S. Diallo,Branched polymeric media: perchlorate-selective resins from hyperbranchedpolyethyleneimine, Environ. Sci. Technol. 46 (2012) 10718–10726.

[25] Z. Tu, S. Lu, X. Chang, Z. Li, Z. Hu, L. Zhang, H. Tian, Selective solid-phaseextraction and separation of trace gold, palladium and platinum usingactivated carbon modified with ethyl-3-(2-aminoethylamino)-2-chlorobut-2enoate, Microchim. Acta 173 (2011) 231–239.

[26] (a) M.R. Awual, M. Ismael, Efficient gold(III) detection, separation andrecovery from urban mining waste using a facial conjugate adsorbent, Sens.Actuat. B Chem. 196 (2014) 457–466;

(b) M.R. Awual, M.A. Khaleque, M. Ferdows, A.M.S. Chowdhury, T. Yaita, Rapidrecognition and recovery of gold(III) with functional ligand immobilized novelmesoporous adsorbent, Microchem. J. 110 (2013) 591–598;(c) M.R. Awual, M.M. Hasan, Novel conjugate adsorbent for visual detectionand removal of toxic lead(II) ions from water, Micropor. Mesopor. Mater. 196(2014) 261–269.

[27] S.Y. Kim, Y. Xu, T. Ito, Y. Wu, T. Tada, K. Hitomi, E. Kuraoka, K. Ishii, A novelpartitioning process for treatment of high level liquid waste usingmacroporous silica-based adsorbents, J. Radioanal. Nucl. Chem. 295 (2013)1043–1050.

[28] H. Ebrahimzadeh, N. Tavassoli, M.M. Amini, Y. Fazaeli, H. Abedi, Determinationof very low levels of gold and palladium in wastewater and soil samples byatomic absorption after preconcentration on modified MCM-48 and MCM-41silica, Talanta 81 (2010) 1183–1188.

[29] R. Li, Q. He, Z. Hu, S. Zhang, L. Zhang, X. Chang, Highly selective solid-phaseextraction of trace Pd(II) by murexide functionalized halloysite nanotubes,Anal. Chim. Acta 713 (2012) 136–144.

[30] P. Ramakul, Y. Yanachawakul, N. Leepipatpiboon, N. Sunsandee, Biosorption ofpalladium(II) and platinum(IV) from aqueous solution using tannin fromIndian almond (Terminalia catappa L.) leaf biomass: kinetic and equilibriumstudies, Chem. Eng. J. 193–194 (2012) 102–111.

[31] T. Maruyama, H. Matsushita, Y. Shimada, I. Kamata, M. Hanaki, S. Sonokawa, N.Kamiya, M. Goto, Proteins and protein-rich biomass as environmentallyfriendly adsorbents selective for precious metal ions, Environ. Sci. Technol.41 (2007) 1359–1364.

[32] (a) M.R. Awual, T. Yaita, S.A. El-Safty, H. Shiwaku, S. Suzuki, Y. Okamoto,Copper(II) ions capturing from water using ligand modified a new typemesoporous adsorbent, Chem. Eng. J. 221 (2013) 322–330;(b) M.R. Awual, I.M.M. Rahman, T. Yaita, M.A. Khaleque, M. Ferdows, PHdependent Cu(II) and Pd(II) ions detection and removal from aqueous mediaby an efficient mesoporous adsorbent, Chem. Eng. J. 236 (2014) 100–109;(c) M.R. Awual, Investigation of potential conjugate adsorbent for efficientultra-trace gold(III) detection and recovery, J. Ind. Eng. Chem. 20 (2014) 3493–3501.

[33] G.R. Pearson, Hard and soft acids and bases, J. Am. Chem. Soc. 85 (1963) 3533–3539.

[34] (a) M.R. Awual, T. Yaita, Y. Okamoto, A novel ligand based dual conjugateadsorbent for cobalt(II) and copper(II) ions capturing from water, Sens. Actuat.B Chem. 203 (2014) 71–80;(b) M.R. Awual, M.M. Hasan, A. Shahat, Functionalized novel mesoporousadsorbent for selective lead(II) ions monitoring and removal from wastewater,Sens. Actuat. B Chem. 203 (2014) 854–863;(c) M.R. Awual, M.M. Hasan, T. Ihara, T. Yaita, Mesoporous silica based novelconjugate adsorbent for efficient selenium(IV) detection and removal fromwater, Micropor. Mesopor. Mater. 197 (2014) 331–338.

[35] M.A. Barakat, M.H.H. Mahmoud, Y.S. Mahrous, Recovery and separation ofpalladium from spent catalyst, Appl. Catal. A Gen. 301 (2006) 182–186.