Ultrafine cobalt nanoparticles supported on reduced ...

Applied Surface Science 402 (2017) 294–300

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Full Length Article

Ultrafine cobalt nanoparticles supported on reduced graphene oxide: Efficient catalyst for fast reduction of hexavalent chromium at room temperature Tingting Xu, Jinjuan Xue, Xiaolei Zhang, Guangyu He ∗ , Haiqun Chen ∗ Key Laboratory of Advanced Catalytic Materials and Technology, Advanced Catalysis and Green Manufacturing Collaborative Innovation Center, Changzhou University, Changzhou, Jiangsu Province 213164, China

a r t i c l e

i n f o

Article history: Received 24 November 2016 Received in revised form 22 December 2016 Accepted 12 January 2017 Available online 16 January 2017 Keywords: Co-RGO Composite materials HCOOH Catalytic activity Cr(VI) reduction

a b s t r a c t A novel composite ultrafine cobalt nanoparticles-reduced graphene oxide (Co-RGO) was firstly synthesized through a modified one-step solvothermal method with Co(OH)2 as the precursor. The prepared low-cost Co-RGO composite exhibited excellent catalytic activity for the reduction of highly toxic Cr(VI) to nontoxic Cr(III) at room temperature when formic acid (HCOOH) was employed as the reductant, and its catalytic performance was even comparable with that of noble metal-based catalysts in the same reduction reaction. Moreover, Co-RGO composite could be readily recovered under an external magnetic field and efficiently participated in recycled reaction for Cr(VI) reduction. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The widespread use of chromium and its compounds in modern industry has caused extremely serious environmental pollution [1]. Hexavalent chromium (Cr(VI)) is a highly toxic pollutant with mutagenicity and carcinogenicity, which exists stably in the natural environment and is regarded as the third most common pollutant at hazardous waste sites [2]. Contrarily, trivalent chromium (Cr(III)) is relatively inert, nontoxic and accepted as an essential human nutrient [2,3]. Thus, researchers have been committed to search for a practical and economical technology for the conversion of Cr(VI) to Cr(III). Recent studies showed that the conversion of Cr(VI) to Cr(III) could be achieved by various reductants, such as HCOOH, H2 S, Fe(0), Fe(II)-bearing minerals and so on [4]. Among these reductants, HCOOH presented great potential for Cr(VI) reduction due to its superiorities in environmental protection, adequate sources and mild reaction conditions [4,5]. In the previous studies, noble metal-based catalysts were fabricated and played an important role in Cr(VI) reduction. For example, colloidal PdNPs [5], PtNPs@ProESM [6] and Pd-␥-Al2 O3 [7] were proved to be efficient catalysts for

∗ Corresponding authors. E-mail addresses: [email protected] (G. He), [email protected], [email protected] (H. Chen). http://dx.doi.org/10.1016/j.apsusc.2017.01.114 0169-4332/© 2017 Elsevier B.V. All rights reserved.

Cr(VI) reduction. However, these catalysts suffered from expensive price and high reaction temperature, which hindered their practical applications. Hence, developing advanced catalysts with low cost, high activity and good recyclability by a simple fabrication process for fast reduction of Cr(VI) at room temperature remains a challenge. In recent years, Co nanoparticles have drawn great concerns in the field of catalysis owing to their high catalytic activity, low price, nontoxicity as well as magnetic recyclability [8,9]. However, it is frustrating to find that pure Co nanoparticles with large size are prone to agglomerate together and unfavorable for catalytic reduction of Cr(VI). Therefore, preparing small, well-dispersed and non-agglomerated Co nanoparticles is very necessary. Our previous studies have clearly demonstrated that when metal nanoparticles such as Pt [10] and Cu [11] were decorated on RGO sheets, the unique synergistic effect between metal nanoparticles and RGO will endow the composite high catalytic activity and effective reusability. In addition, the aggregation and oxidation problems of metal nanoparticles could be solved because of the large specific surface area, high electron transport property and excellent adsorptivity of graphene. Herein, for the first time, ultrafine cobalt nanoparticles supported on reduced graphene oxide was successfully prepared via a modified one-step solvothermal method by using Co(OH)2 as the precursor, which achieved the growth of Co nanoparticles and the reduction of graphene oxide (GO) simultaneously. The

T. Xu et al. / Applied Surface Science 402 (2017) 294–300

formed ultrafine Co nanoparticles were densely and compactly loaded on RGO sheets, which could provide quantities of active sites and be propitious to the improvement of the catalytic properties. The results manifested the as-prepared Co-RGO composite was an efficient catalyst for fast reduction of highly toxic Cr(VI) at room temperature. Moreover, the composite could be readily recovered under an external magnetic field. After eight cycles, it still exhibited considerably high catalytic activity for Cr(VI) reduction. 2. Experimental 2.1. Materials Cobalt nitrate (Co(NO3 )2 ·6H2 O), sodium hydroxide (NaOH), ethanol (CH3 CH2 OH), hydrazine hydrate (N2 H4 ·H2 O), and other materials purchased from Sinopharm Chemical Reagent Co., Ltd. (China). All chemicals were of analytical grade except industrial pure ethanol. 2.2. Synthetic procedures Graphite oxide (GO) was successfully prepared according to the modified Hummers method [12,13] by using the natural graphite powder as the raw materials. Co-RGO composites with different graphene contents were synthesized through a modified one-step solvothermal method and denoted as Co-RGOx (x = 5, 10, 15, 20, 25 wt%, respectively). A typical experimental procedure for the synthesis of Co-RGO10 composite with 10 wt% graphene was as follows: 0.306 g of the as-prepared colloidal GO (with GO content of 2.14 wt%) was uniformly dispersed in 40 mL of ethanol with sonication. 0.291 g of Co(NO3 )2 ·6H2 O (1 mmol) was dissolved in 30 mL of deionized water, which was adjusted to pH 11 by NaOH solution (6 M) for generating Co(OH)2 precipitation. After that, Co(OH)2 was separated from the mixture by vacuum filtration, washed with distilled water and ethanol respectively, and then dispersed in 30 mL of ethanol under sonication for 30 min again. The Co(OH)2 dispersion was added tardily into GO dispersion with continuous stirring for 1 h. Then, 6 M NaOH solution was added dropwisely in the above mixture until the pH of the mixture was 11. After vigorously stirring for 30 min, 3 mL of N2 H4 ·H2 O was mixed into the above the emulsion and stirred for another 30 min. The resulting stable homogeneous mixture was settled into a 100 mL Teflon-lined stainless autoclave and heated to maintain 180 ◦ C for 12 h under autogenous pressure (Fig. 1). The obtained sample was cooled to room temperature, and purified by filtration, washing and vacuum dried at 60 ◦ C for 10 h. Pure Co nanoparticles and RGO were also respectively prepared by the same method in the absence of GO or Co(OH)2 for comparison. 2.3. Characterization The phase structures of the as-prepared catalysts were analyzed by powder X-ray diffraction (XRD) employing a Bruker D8 Advanced diffractometer (Cu Ka, ␭ = 1.5418 Å) with the scanning angle ranging from 5◦ to 80◦ . Fourier transform infrared (FTIR) spectra were recorded on a Nicolet 370 FTIR spectrometer (Thermo Nicolet, USA) using pressed KBr pellets in the region of 500–4000 cm−1 . Raman spectra were acquired using a Renishaw inVia Reflex Raman microprobe. The Brunauer–Emmett–Teller (BET) surface areas of the catalysts were measured using an ASAP2010C surface aperture adsorption instrument (Micromeritics Instrument Corporation, USA) by N2 physisorption at 77 K. The content of Co of the catalyst was determined by an inductively coupled plasma-atomic emission spectrometer (ICP-AES, ICPS-7510, Shimadzu, Japan). Transmission electron microscopy (TEM) images were taken with a JEOL JEM-2100 microscope (JEOL, Japan) at an

295

accelerating voltage of 200 kV. The reduction of Cr(VI) ions was performed by measuring the absorbance values at various time intervals using a Shimadzu UV-2700 UV–vis spectrophotometer at room temperature (25 ◦ C). 2.4. Catalytic activity measurement The reduction process of Cr(VI) to Cr(III) over Co-RGO composite was monitored by observing the absorbance changes of the reaction mixture between 300 and 500 nm with the main absorption peak at 350 nm. Typically, 40 mL of K2 Cr2 O7 aqueous solution (100 mg/L) and 1 mL of HCOOH (23 g/L) were added to a 100 mL beaker and uniformly stirred at room temperature. The pH of the reaction mixture was 2, which could be adjusted by the addition of NaOH solution to be 4 and 6, respectively. Besides, almost no changes of pH were measured during each reaction process. The absorption spectrum of the reaction solution was recorded and considered as 0 min data. Hereafter, 4 mg of Co-RGO composite powder was added to the reaction solution. The absorption spectra of the reduction process were first recorded at 1 min and then recorded at 2 min intervals. After the first reaction cycle, the Co-RGO composite was recovered by a permanent magnet, washed with distilled water and ethanol respectively, and vacuum dried before reused. 3. Results and discussion 3.1. Characterization of Co-RGO composites The typical XRD diffraction patterns of pure Co, Co-RGO10 , GO and RGO were presented in Fig. 2A. In the XRD pattern of Co-RGO10 composite, almost all the diffraction peaks of the characteristic were well consistent with standard hcp-type Co nanoparticles (JCPDS No. 05-0727), which indicated Co(OH)2 was converted into metallic Co after the modified solvothermal reaction. The diffraction peaks at 2 = 41.63◦ , 44.24◦ , 47.37◦ and 75.80◦ could be indexed to the (100), (002), (101) and (110) crystal planes of Co, respectively [14–16]. Compared with pure Co nanoparticles, sharper crystalline peak corresponded to (002) crystal plane was obvious observed in the pattern of Co-RGO10 composite, which indicated the introduction of graphene enhanced the crystallinity and orientation of Co nanoparticles [17]. However, neither the typical diffraction peak of GO (001) nor RGO (002) was discovered in the XRD pattern of CoRGO10 composite. This may be attributed to the fact that during the modified solvothermal reaction, GO was reduced to RGO and then exfoliated by the decorated Co nanoparticles, which destroyed the regular layer stacking of RGO [18,19]. Fig. 2B depicts the FTIR spectra of Co-RGO10 composite, RGO and GO. By comparison with the FTIR spectrum of GO, it could be obviously noted that the absorption intensities of characteristic peaks of oxygen-containing groups in the Co-RGO10 composite decreased considerably, which suggested GO in the Co-RGO10 composite was mostly reduced to RGO [20]. Besides, the positions of characteristic peaks in the Co-RGO10 composite were consistent with those of RGO and no new characteristic peaks were observed, further indicating the composite was synthesized by the hybridization of metallic Co and RGO. Ordinarily, the reduction of GO caused fragmentation generated along the reactive sites and produced new graphitic domains, leading to smaller size but more numerous of RGO sheets compared with GO before reduction [21]. Thus, it was reasonable to deem that much more active sites might contact with reactants for facilitating the occurrence of the reaction after the reduction of GO. The representative Raman spectra of Co-RGO10 composite, RGO and GO were displayed in Fig. 2C. As shown in the Raman spectrum of Co-RGO10 composite, the two characteristic peaks at 1347 cm−1

296


Fig. 1. Illustration of the synthesis procedure of Co-RGO composite.

Fig. 2. (A) XRD patterns of pure Co, Co-RGO10 , RGO and GO; (B) FTIR spectra of Co-RGO10 , RGO and GO; (C) Raman spectra of Co-RGO10 , RGO and GO; (D) Nitrogen adsorption/desorption isotherm of Co-RGO10 and pore size distribution of Co-RGO10 (inset).

and 1583 cm−1 were well ascribed to the typical D and G bands of graphitic carbon materials, respectively [21–24]. Compared with those of GO, the D and G bands of the composite shifted to lower frequencies, indicating that GO has been reduced to RGO. Generally, the D band is attributed to defects, edges, as well as disordered carbon in the graphite structure, whereas the G band is in conformity to zone centre E2g vibration mode of the ordered sp2 hybridized carbon atoms. Besides, the intensity ratio of the D and G bands (ID /IG ) is mainly accepted to estimate the degree of graphitization and defect density of carbon materials [21,24]. From Fig. 2C, the values of ID /IG for Co-RGO10 composite and RGO were 1.24 and 1.07, respectively, which were significantly higher than that of GO (ID /IG = 0.83), further revealing the average size of the sp2 domains decreased upon reduction of the exfoliated GO. The BET surface area and porous structure of the Co-RGO10 composite were explored by the nitrogen adsorption/desorption isotherm. As illustrated in Fig. 2D, a typical type-IV isotherm with a distinct hysteresis loop was observed in the range from 0.45 to 1.0 of the relative pressure, which implied that the Co-RGO10 composite possessed abundant mesopores [25]. The Co-RGO10 composite had a large Brunauer–Emmett–Teller (BET) surface area of 93.33 m2 g−1 with a pore volume of 0.170 cm3 g−1 and a Barrett–Joyner–Halenda (BJH) desorption average pore diameter of 6.38 nm. Additionally,

the nitrogen adsorption/desorption isotherm of pure Co were also measured and presented in Fig. S1 (Supplementary material). The BET surface area obtained for the pure Co was just 14.43 m2 g−1 , which was noticeably lower than that of the Co-RGO10 composite. The result showed that the Co-RGO10 composite with large BET surface area was propitious to provide much more active sites, increase the collision probability between reactants, and thus enhance the catalytic properties. In addition, the content of Co in the Co-RGO10 composite was measured by ICP-AES. The test result showed that the content of Co out of 10.02 mg of Co-RGO10 composite was 8.98 mg. This meant the RGO content in the composite was about 1.04 mg (i.e. 10.38 wt%), which was basically in accordance with the calculated result. The microstructures of the Co-RGO10 composites and pure Co nanoparticles investigated through TEM were shown in Fig. 3 and S2 (Supplementary material). It could be clearly observed that RGO sheets showed a winkled and transparent veil-like morphology, which were fully exfoliated and uniformly decorated with ultrafine Co nanoparticles. The size distribution histogram manifested the average size of Co nanoparticles loaded on the RGO sheets was approximately 4.6 nm. In the modified solvothermal process, the graphene may play a significant role in advancing the formation of nanoparticles and limiting the size of nanoparticles [26]. These


297

Fig. 3. TEM images of Co-RGO10 composite and size distribution histogram of Co nanoparticles (inset).

small, well-dispersed Co nanoparticles contributed to the catalyst’s high performance. Furthermore, the introduction of RGO sheets were beneficial to increasing the BET surface area, diminishing the diffusion resistance as well as hastening the electron transfer in the reaction process, which further accounted for the improvement of catalytic activity. 3.2. Catalytic activity for Cr(VI) reduction The reductive conversion of Cr(VI) to Cr(III) using HCOOH as the reductant functions as a well-established model reaction for evaluating the catalytic activity of the Co-RGO composite. In the absence of the catalyst, the reduction of Cr(VI) proceeded extremely slowly, which indicated Cr(VI) could not be efficiently reduced by HCOOH alone (Fig. S3 in the Supplementary material). However, in the presence of the Co-RGO10 composite, the main absorption peak at 350 nm decreased swiftly and vanished within reaction time of 9 min (Fig. 4A). Meanwhile, the color of the reaction solution rapidly changed from yellow to colorless in this reaction process, visually confirming the thorough reduction of Cr(VI) to Cr(III) [27,28]. What’s more, the presence of Cr(III) in the colorless solution was verified by adding excess NaOH solution, which gave rise to the generation of a green solution (Fig. 4D inset). It was a characteristic phenomenon of the existence of Cr(III) owing to the formation of hexahydroxochromate(III) [29–31]. The inset of Fig. 4A illustrates that a pseudo-first-order kinetic Eq. (1) was applied to fit the reduction of Cr(VI) to Cr(III) in the condition of excess HCOOH. ln(Ct /C0 ) = −kt

(1)

Where C0 is the initial concentration of Cr2 O7 2− , Ct is the concentrations of Cr2 O7 2− at reaction time t and k is the apparent rate constant. The value of k was evaluated from the slope to be 0.474 min−1 for the Co-RGO10 composite, which demonstrated the composite is endowed with excellent catalytic activity for the reduction of Cr(VI) at room temperature. Furthermore, the turnover frequency (TOF, per mole of the catalyst can reduce the number of moles of Cr(VI) per unit time) was further used to assess the catalytic performance of the catalysts. A comparison of the catalytic reduction of Cr(VI) over Co-RGO10 and the previously reported catalysts was presented in Table 1 [6,29,30,32,33]. The calculated TOF of the Co-RGO10 composite was 4.9 × 10−2 mol mol−1 min−1 , which was much higher than that of other listed heterogeneous catalysts, suggesting the catalytic superiority of Co-RGO10 over other catalysts for the reduction of Cr(VI). It was demonstrated that pure RGO could not efficiently catalyze the activation of HCOOH for the reduction of Cr(VI) (Fig. S4 in the Supplementary material), indicating the catalytic active centers of the Co-RGO composites were Co nanoparticles during the process of Cr(VI) reduction. Hence, the effect of RGO content on Cr(VI) reduction over the Co-RGO composite was investigated and the results were summarized in Fig. 4B. Specifically, the cat-

alytic performance of the Co-RGO composite could be observed to increase first and then decrease thereafter with the increase of RGO content. And when the content of RGO was up to 10 wt%, the composite gave the highest the catalytic activity. Such phenomenon revealed the existence of the synergistic effect between Co nanoparticles and graphene. Graphene, a special two dimensional carbon-based material bearing a monolayer of carbon atoms sp2 -hybridized into a densely packed honeycomb structure, not only effectively prevented the aggregation of Co nanoparticles, conduced to their uniform distribution, increase the reaction surface area, but also decline the mass transfer resistance, which jointly promoted catalytic reaction of the composite [34]. In addition, graphene with high electrical conductivity was also conducive to electron transfer and then improved the catalytic activity of the composite. Last but not least, owing to the electrostatic interaction between RGO sheets and Cr(VI), Cr(VI) could adsorb on the surface of Co-RGO composite[30,33,35] (Fig. S5 in the Supplementary material), resulting in the collision probability between reactants increasing, which was also beneficial to the catalytic reaction. However, the increase of RGO content would inevitably bring about the decrease of the amounts of Co nanoparticles that played a key role in the catalytic reaction, which would gradually become a dominant factor affecting the reaction and resulted in a decrease in the reaction rate with the further increase of RGO content [26,36]. Therefore, an optimal ratio of Co nanoparticles and RGO existed for catalyzing Cr(VI) reduction. The reduction process of Cr(VI) was proved to be highly dependent on the pH of the solution. In this case, HCOOH was not only functioned as a reductant, but also played an important role in adjusting the pH of the reaction solution. The effect of different pH on the catalytic reduction of Cr(VI) was given in Fig. 4C. The reduction rate of Cr(VI) decreased drastically on increasing the pH, which is in accordance with the reaction mechanism of the Cr(VI) (Eq. (2)) [32], showing that highly acidic condition is favor for the Cr(VI) reduction. + 3+ Cr2 O2− + 7H2 O + 3CO2 7 + 8H + 3HCOOH → 2Cr

(2)

In order to research the influence of the reaction temperature on the reduction of Cr(VI) catalyzed by Co-RGO10 , the reactions were carried out at different temperatures and the results were provided in Fig. 4D. Apparently, the reduction rate of Cr(VI) increased significantly with the rise of temperature, indicating that the high temperature could accelerate the catalytic reduction of Cr(VI) [5,37]. This was because the appropriate elevated temperature could promote the effective decomposition of HCOOH into active H2 for Cr(VI) reduction, and meanwhile, the surface acidities of the catalysts would increase with the rise of temperature [37], which further facilitated the progress of the reaction. Although the asprepared Co-RGO10 composite displayed higher catalytic activity towards the complete reduction of Cr(VI) at high temperature, it is not appropriate for practical applications in terms of operation

298


Fig. 4. (A) UV–vis spectral evolution with time during the process of Cr(VI) reduction by HCOOH at 25 ◦ C in the presence of 4 mg of Co-RGO10 and the pseudo-first-order plot of −ln(Ct /C0 ) with time for the above reaction (inset); (B) Effect of RGO content on catalytic reduction of Cr(VI); (C) Effect of pH on catalytic reduction of Cr(VI); (D) Effect of temperature on catalytic reduction of Cr(VI) and the color change of Cr(VI) reaction solution (inset).

Table 1 Cr(VI) reduction catalyzed by different catalytic systems. Catalyst

Temperature (◦ C)

Time (min)

TOF (mol mol−1 min−1 )

Reference

Co-RGO10 PtNPs@Pro-ESM [email protected] Ni-RGO10 [Cu4 (bpp)4 ][␤-As8 V14 O42 (H2 O)] AuNPs

25 45 25 25 25 25

9 15 15 4 70 100

4.9 × 10−2 8.6 × 10−4 2.4 × 10−2 1.4 × 10−2 1.9 × 10−6 1.1 × 10−2

This work [6] [29] [30] [32] [33]

simplicity and energy conservation. Consequently, we still selected 25 ◦ C as the appropriate reaction temperature for Cr(VI) reduction. Fig. 5A reflected that the catalytic reaction of Cr(VI) reduction was profoundly influenced by the initial Cr(VI) concentration. Almost no changes of pH were measured during each reaction process due to the excess HCOOH. Thus, the influence of the amount of the reductant HCOOH could be ignored. As could be seen from Fig. 5A, the catalytic reduction rate decreased rapidly with the increase of initial Cr(VI) concentration. Moreover, the reduction rate of Cr(VI) was just 83.6% when the catalytic reaction was carried out for 15 min in the reaction solution with 160 mg/L of the initial Cr(VI) concentration. This might arise from the following two main reasons: (1) The fixed dosage of catalyst only produce a certain amount of active metal centers, which catalyzed the effective decomposition of a certain amount of HCOOH to active H2 for Cr(VI) reduction. (2) The surface active metal centers could be blocked by the abundant Cr(VI), leading to the decrease of Cr(VI) reduction rate [37]. Hence, for the purpose of obtaining the complete reduction of Cr(VI) and convenience of experimental operations, we chose 100 mg/L as the optimum initial Cr(VI) concentration. The effect of the catalyst dosage on catalytic reduction of Cr(VI) was depicted in Fig. 5B. The catalytic reaction was performed using various amounts of Co-RGO10 composite (3, 4, 5, and 6 mg, respectively). As illustrated in Fig. 5B, the increasing amounts of Co-RGO10

composite resulted in the catalytic reduction rate of Cr(VI) increasing dramatically. It was inevitable that the amounts of surface active metal sites and adsorption of Cr(VI) would increase with the increase of the catalyst dosage, which could boost the effective decomposition of more amount of HCOOH to active H2 for the fast reduction Cr(VI) to Cr(III) and further indicated the amounts of reactive species was mainly dependent on the dosage of Co-RGO10 composite. It was found that when only 4 mg of Co-RGO10 composite was applied in the catalytic reaction, Cr(VI) could be completely reduced to Cr(III) within the reaction time of 9 min. Therefore, 4 mg was considered as the optimal catalyst dosage for Cr(VI) reduction in the economical point of view. Fig. 6A shows the catalytic stability of the Co-RGO10 composite for Cr(VI) reduction. Although the catalytic reaction rate of Cr(VI) reduction gradually declined with the increase of the frequency, the catalyst could be efficiently recycled and reused for at least eight consecutive cycles with a steady conversion rate of above 93%. Compared with two-step prepared [email protected] composite we reported before [29], where the conversion rate of Cr(VI) conspicuously decreased from 100% to 86% just after three cycles, the stability and recyclability of Co-RGO10 prepared by a one-step method are notably better. Moreover, the XRD pattern of Co-RGO10 composite after eight repeated catalytic reactions was also shown in Fig. 6B, which further demonstrated the good catalytic stability


299

Fig. 5. (A) Effect of initial Cr(VI) concentration on catalytic reduction of Cr(VI) by 4 mg of Co-RGO10 ; (B) Effect of the catalyst dosage on catalytic reduction of Cr(VI) with 100 mg/L of initial Cr(VI) concentration.

Fig. 6. (A) Catalytic stability of Co-RGO10 in Cr(VI) reduction; (B) XRD patterns of Co-RGO10 before the reaction and after recycled eight times for Cr(VI) reduction.

4. Conclusions

Fig. 7. Proposed mechanism of catalytic reduction of Cr(VI) over Co-RGO composite.

of the Co-RGO10 composite. The high stability of Co-RGO10 might be related to not only its relatively stable structure, but also the unique synergistic effect between Co nanoparticles and RGO sheets. Therefore, the Co-RGO composite possesses significant potential and far-reaching prospects for purifying the Cr(VI) contaminated waste water. On the basis of some references [28,30,31,33] and the results from the above control experiments, the plausible catalytic mechanism of Cr(VI) reduction over Co-RGO composite was proposed and described in Fig. 7. It could be considered that Co nanoparticles were the catalytic active centers and catalyzed the effective decomposition of HCOOH to active H2 and CO2 on the graphene surface. The adsorbed Cr(VI) was then efficiently reduced to Cr(III) by the active H2 . In addition, the high electron-transport property of RGO promoted the electron transfer from active H2 to Cr(VI), leading to the fast reduction of Cr(VI) at room temperature.

In summary, we have successfully developed Co-RGO composite through a facile one-step modified solvothermal route using Co(OH)2 as the precursor for the first time. The reduction of GO and the formation of Co nanoparticles were simultaneously accomplished in the synthesis process. The combination of salutary features of ultrafine Co nanoparticles and RGO sheets made the composite exhibit excellent catalytic activity for the fast reduction of Cr(VI) to Cr(III) at room temperature with the existence of HCOOH as well as high catalytic stability for Cr(VI) reduction. Under the optimum reaction conditions (40 mL of Cr(VI) solution with 100 mg/L as the initial Cr(VI) concentration, 1 mL of HCOOH as reductant and 4 mg of Co-RGO10 composite as catalyst at 25 ◦ C for 9 min), Co-RGO10 composite could efficiently catalyze the decomposition of HCOOH to active H2 for reducing Cr(VI) to Cr(III), which achieved the complete reduction of Cr(VI). In all, our new developed catalyst, which processed low cost, magnetic recyclability, high catalytic activity and simple fabrication process, would effectively solve the environment pollution problem of toxic Cr(VI).

Acknowledgements The financial supports from the National Natural Science Foundation of China (Nos. 51202020, 51472035, 51572036), the Science and Technology Department of Jiangsu Province (BY2016029-12, BE2014089, BY2015027-18), Changzhou Key Laboratory of Graphene-Based Materials for Environment & Safety (CM20153006, CE20160001-2) and the PAPD of Jiangsu Higher Education Institutions are gratefully acknowledged.

300


Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apsusc.2017.01. 114. References [1] K.K. Krishnani, S. Srinives, B.C. Mohapatra, V.M. Boddu, J. Hao, X. Meng, A. Mulchandani, J. Hazard. Mater. 252–253 (2013) 99–106. [2] A. Qian, P. Liao, S. Yuan, M. Luo, Water Res. 48 (2014) 326–334. [3] S.K. Prabhakaran, K. Vijayaraghavan, R. Balasubramanian, Ind. Eng. Chem. Res. 48 (2009) 2113–2117. [4] M. Yadav, Q. Xu, Chem. Commun. 49 (2013) 3327–3329. [5] M.A. Omole, I.O. K’Owino, O.A. Sadik, Appl. Catal. B: Environ. 76 (2007) 158–167. [6] M. Liang, R. Su, W. Qi, Y. Zhang, R. Huang, Y. Yu, L. Wang, Z. He, Ind. Eng. Chem. Res. 53 (2014) 13635–13643. [7] A. Dandapat, D. Jana, G. De, Appl. Catal. A: Gen. 396 (2011) 34–39. [8] L. Yang, N. Cao, C. Du, H. Dai, K. Hu, W. Luo, G. Cheng, Mater. Lett. 115 (2014) 113–116. [9] F. He, N. Niu, F. Qu, S. Wei, Y. Chen, S. Gai, P. Gao, Y. Wang, P. Yang, Nanoscale 5 (2013) 8507–8516. [10] H. Huang, H. Chen, D. Sun, X. Wang, J. Power Sources 204 (2012) 46–52. [11] H. Bi, G. He, R. Xu, Z. Song, J. Zhu, H. Chen, X. Wang, Curr. Org. Chem. 18 (2014) 3136–3140. [12] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [13] N.I. Kovtyukhova, P.J. Ollivier, B.R. Martin, T.E. Mallouk, S.A. Chizhik, E.V. Buzaneva, A.D. Gorchinskiy, Chem. Mater. 11 (1999) 771–778. [14] D. Zhao, Z. Nan, Funct. Mater. Lett. 9 (2016) 1650009. [15] Z. Ji, Y. Wang, J. Yang, X. Shen, Q. Yu, L. Kong, H. Zhou, RSC Adv. 6 (2016) 107709–107716. [16] Z. Ji, X. Shen, Y. Song, G. Zhu, Mater. Sci. Eng. B: Adv. Funct. Solid-State Mater. 176 (2011) 711–715.

[17] L. Liu, Z. Shen, S. Liang, M. Yi, X. Zhang, S. Ma, J. Mater. Sci. 49 (2014) 321–328. [18] C. Xu, X. Wang, J. Zhu, J. Phys. Chem. C 112 (2008) 19841–19845. [19] W. Zhang, Q. Yao, X. Wu, Y. Fu, K. Deng, X. Wang, Electrochim. Acta 200 (2016) 131–141. [20] Y. Fu, X. Wang, Ind. Eng. Chem. Res. 50 (2011) 7210–7218. [21] J. Shen, B. Yan, M. Shi, H. Ma, N. Li, M. Ye, J. Mater. Chem. 21 (2011) 3415–3421. [22] X. Meng, H. Sun, J. Zhu, H. Bi, Q. Han, X. Liu, X. Wang, New J. Chem. 40 (2016) 2843–2849. [23] Y. Fu, J. Zhu, C. Hu, X. Wu, X. Wang, Nanoscale 6 (2014) 12555–12564. [24] X. Meng, J. Zhu, H. Bi, Y. Fu, Q. Han, X. Wang, J. Mater. Chem. A 3 (2015) 21682–21689. [25] X. Xu, J. Shen, N. Li, M. Ye, Int. J. Hydrogen Energy 40 (2015) 13003–13013. [26] T. Xu, G. He, Y. Zhao, H. Gu, Z. Jiang, Q. Chen, X. Sun, H. Chen, Appl. Surf. Sci. 389 (2016) 840–848. [27] H. Saikia, B.J. Borah, Y. Yamada, P. Bharali, J. Colloid Interface Sci. 486 (2017) 46–57. [28] Y. Guo, D. Wang, X. Liu, X. Wang, W. Liu, W. Qin, New J. Chem. 38 (2014) 5861–5867. [29] L. Liu, J. Xue, X. Shan, G. He, X. Wang, H. Chen, Catal. Commun. 75 (2016) 13–17. [30] K. Bhowmik, A. Mukherjee, M.K. Mishra, G. De, Langmuir 30 (2014) 3209–3216. [31] B.J. Borah, H. Saikia, P. Bharali, New J. Chem. 38 (2014) 2748–2751. [32] X. He, Y. Liu, K. Gong, Z. Han, X. Zhai, Inorg. Chem. 54 (2015) 1215–1217. [33] R.K. Gupta, V. Kumar, A. Srivastava, D.S. Pandey, RSC Adv. 6 (2016) 40911–40915. [34] W. Zhang, Z. Zhou, X. Shan, R. Xu, Q. Chen, G. He, X. Sun, H. Chen, New J. Chem. 40 (2016) 4769–4774. [35] Z. Chen, Y. Li, M. Guo, F. Xu, P. Wang, Y. Du, P. Na, J. Hazard. Mater. 310 (2016) 188–198. [36] Y. Zhao, G. He, W. Dai, H. Chen, Ind. Eng. Chem. Res. 53 (2014) 12566–12574. [37] S. Li, L. Tang, G. Zeng, J. Wang, Y. Deng, J. Wang, Z. Xie, Y. Zhou, Environ. Sci. Pollut. Res. 23 (2016) 22027–22036.