Prasanta Kumar Sahooa,
Bharati Panigrahyb and
Dhirendra Bahadur*a
aIITB-Monash Research Academy, Indian Institute of Technology Bombay, Mumbai-400076, India. E-mail: dhirenb@iitb.ac.in
bSolid State and Structural Chemistry Unit, Indian Institute of Science, Bangalore, 560012, India
First published on 15th September 2014
The catalytic performance of metals can be enhanced by intimately alloying different metals with Reduced Graphene Oxide (RGO). In this work, we have demonstrated a simplistic in situ one-step reduction approach for the synthesis of RGO/Pt–Ni nanocatalysts with different atomic ratios of Pt and Ni, without using any capping agent. The physical properties of the as-synthesized nanocatalysts have been systematically investigated by XRD, FTIR, Raman spectroscopy, XPS, EDX, ICP-AES, and TEM. The composition dependent magnetic properties of the RGO/Pt–Ni nanocatalysts were investigated at 5 and 300 K, respectively. The results confirm that the RGO/Pt–Ni nanocatalysts show a super-paramagnetic nature at room temperature in all compositions. Furthermore, the catalytic activities of the RGO/Pt–Ni nanocatalysts were investigated by analyzing the reduction of p-nitrophenol, and the reduction rate was found to be susceptible to the composition of Pt and Ni. Moreover, it has been found that RGO/Pt–Ni nanocatalysts show superior catalytic activity compared with the bare Pt–Ni of the same composition. Interestingly, the nanocatalysts can be readily recycled by a strong magnet and reused for the next reactions.
In the last few decades, alloying two kinds of metal nanoparticles has received great interest because of the resulting exceptional electronic,23 optical,24 and catalytic properties,25 compared with those of the respective individual metal nanoparticles. For example, Y. Huang et al. have reported that Pt–Ni nanocrystals show a much better performance and durability than the commercial Pt black and commercial Pt/C catalysts for the oxygen reduction reaction.26 Recently, it has been demonstrated that graphene-supported metal alloys like Ni–Co,27 Zn–Ni,28 Fe–Pt,29 Fe–Ni30 and Pt–Ni31 exhibit an unusually high catalytic performance, which makes graphene an ideal substitute for other carbon materials as a catalyst support. The reduction of aromatic nitro compounds to amines is a very vital process in synthetic organic chemistry and in the industry for the fabrication of industrial products. Hence, the development of an effective, environmentally friendly and recyclable catalyst is anticipated for the reduction of aromatic nitro compounds to amines. Presently, the reduction of p-nitrophenol to p-aminophenol by NaBH4 is widely used as a model reaction to quantify the catalytic activity of various metals or alloy catalysts. For instance, T. Pal et al. have reported that Pt–Ni bi-metallic nanoparticles show a superior catalytic activity in the borohydrate reduction of p-nitrophenol compared with the monometallic Pt nanoparticles of comparable sizes.32
In the present work, the RGO/Pt–Ni nanocatalysts with different ratios of Pt and Ni have been synthesized by a simple, one-step reduction approach. The structural and magnetic properties of the as-synthesized nanocatalysts were studied. The catalytic studies of p-nitrophenol reduction by RGO/Pt–Ni nanocatalysts with varying ratios of Pt and Ni have been performed. The magnetic studies help in understanding their potential for the separation of these precious catalysts. The catalytic activity of the RGO/Pt–Ni nanocatalyst has been compared with that of bare Pt–Ni, RGO/Ni and RGO–Pt nanocatalysts as well as with some other reported bi-metallic and RGO/bi-metallic systems.27,28,33,34 The results obviously indicate an excellent catalytic activity of RGO/Pt–Ni nanocatalysts toward the reduction of p-nitrophenol as compared to bare Pt–Ni, RGO–Ni, RGO–Pt and other reported bi-metallic and RGO/bi-metallic systems.
RGO/Pt–Ni nanocatalysts with different Pt and Ni atomic ratios of 25:75, 33:67 and 50:50 were synthesized by adjusting the amount of the respective metal salts, keeping the GO amount constant (35 mg) in all cases. The total loading amount of Pt and Ni (Pt:Ni) in RGO/Pt–Ni (25:75), RGO/Pt–Ni (33:67) and RGO/Pt–Ni (50:50) was controlled such that it was approximately 40 wt%. For comparison, bare Pt–Ni (25:75), RGO–Ni (40 wt% of Ni) and RGO–Pt (40 wt% of Pt) were also synthesized in a similar way.
Moreover, as ESI,† the XRD patterns of the RGO–Ni and Pt–Ni samples are given in Fig. S1.† RGO–Ni shows a face-centered cubic (fcc) structure of nickel (JCPDS 04-0802), whereas Pt–Ni (25:75) is also an fcc structure with a small shift of diffraction peaks similar to RGO/Pt–Ni (25:75). This shifting indicates the formation of the Pt–Ni alloy.
The FTIR spectra of the as-synthesized GO and the RGO/Pt–Ni nanocatalysts are compared in Fig. 2. Some characteristic peaks of the oxygenic functional groups of GO in Fig. 2(a) indicate that graphite successfully undergoes oxidation. Nearly, all the characteristic bands of oxygenic functional groups disappear in the FTIR spectra of RGO/Pt–Ni nanocatalysts (Fig. 2(b)–(d)). This suggests a successful transformation of GO into RGO in the reduction process. For RGO/Pt–Ni nanocatalysts, strong absorption bands at 1566, 1608 and 1618 cm−1 with Pt–Ni atomic ratios 25:75, 33:67 and 50:50, respectively, can be ascribed to the skeletal vibration of the graphene sheets.39
Fig. 2 FTIR patterns of (a) GO and RGO/Pt–Ni nanocatalysts at Pt–Ni atomic ratios of (b) 25:75, (c) 33:67 and (d) 50:50. |
The characteristic Raman spectra of GO and RGO/Pt–Ni nanocatalysts are shown in Fig. 3. Both GO and RGO/Pt–Ni nanocatalysts show the presence of the G and D bands. The intensity ratio (ID/IG) of the D to the G band is related to the average size of sp2 domains.40 The ID/IG ratio for GO is 0.94 and for RGO/Pt–Ni nanocatalysts with Pt–Ni atomic ratios of 25:75, 33:67 and 50:50 are 1.22, 1.24, and 1.21, respectively. The increase in ID/IG ratios in all the RGO/Pt–Ni nanocatalysts compared with the graphite oxide indicates that the graphite oxide has been successfully deoxygenated and reduced in the RGO/Pt–Ni nanocatalysts with different Pt and Ni concentration.
Fig. 3 Raman spectra of (a) GO and the RGO/Pt–Ni nanocatalysts at Pt–Ni atomic ratios of (b) 25:75, (c) 33:67 and (d) 50:50. |
The XPS measurement is used to evaluate the surface structures and chemical states of these RGO/Pt–Ni nanocatalysts. Fig. 4A and B show the XPS spectra of Pt 4f and Ni 2p in the RGO/Pt–Ni nanocatalyst with the atomic ratio 50:50. It has been observed that Pt exists predominantly in metallic form whereas Ni is oxidized at the catalytic surface during preparation. A slight negative shifting of the Pt 4f7/2 peak (Fig. 4C) in the RGO/Pt–Ni nanocatalysts with atomic ratios 50:50, 33:67 and 25:75 with respect to RGO–Pt may be caused by many factors. One of them is the transfer of electrons from Ni to Pt due to the electronegative difference between Ni (1.91) and Pt (2.28). This leads to a change in the electronic properties of Pt (lowering the density of state on the Fermi level) in RGO/Pt–Ni nanocatalysts of different atomic ratios.31 Such a change in the electronic properties of Pt due to alloying with Ni improves the catalytic performance. For example, it has been reported by several authors31,41,42 that the electron transfer from Ni to Pt may lower the density of states on the Fermi level and decrease the Pt–carbon monoxide (CO) bond energy (weaken the CO adsorption on Pt–Ni alloys), and thus improving the electrocatalytic activity of Pt–Ni alloys toward methanol oxidation. In addition, it has also been observed that with increasing Ni concentration, the Pt 4f7/2 peak gets more broadened. This may be due to the overlap of the Ni 3p peak with the Pt 4f7/2 peak by increasing the concentration of Ni in the Pt–Ni alloy. It has been reported that the binding energy at 68.9 eV can be assigned to the XPS peak position of Ni 3p.43,44 When we increase the concentration of Ni, the Ni 3p peak may appear and merge with the Pt 4f7/2 peak (as shown in the fitted Fig. 4D). Wakisaka et al.45 have reported that the decrease in electron density at the Fermi energy level results in a dull edged XPS peak for Pt–Co and Pt–Ru alloy systems as compared to pure Pt. In our Pt–Ni alloy system, by increasing the Ni concentration, the electron transfer from Ni to Pt gets enhanced which leads to a reduction in the density of states at the Fermi level, and this may result in the 4f7/2 XPS peak broadening.
The morphology and elemental composition of the RGO/Pt–Ni and bare Pt–Ni nanocatalysts were investigated by TEM, EDX, and ICP analysis. The TEM image of RGO–Ni nanocatalyst is shown in Fig. S2(a)† where Ni nanoparticles have a higher contrast, visible as dark dots and are spread out uniformly on the reduced graphene sheets. The average particle size is 65 nm. Fig. S2(b)† shows the TEM image of RGO–Pt nanocatalysts. Pt nanoparticles are in a nanocluster form with sizes varying from 20 to 80 nm. A higher magnification picture of the RGO–Pt nanocatalyst shows that Pt nanoclusters are aggregates of several individual Pt nanoparticles of 5 nm size. The TEM images of RGO/Pt–Ni nanocatalysts with different Pt–Ni atomic ratios are shown in Fig. 5a–c, each consisting of highly interconnected/aggregated crystalites with an average mean particle diameter of approximately 3–4 nm, which is in good agreement with the value estimated from the XRD data. The particle size was nearly equal for all RGO/Pt–Ni nanocatalysts synthesized with different Pt–Ni atomic ratios. Thus, the effects of the particle size on their magnetic and catalytic activities can be ignored. The HRTEM image of the RGO/Pt–Ni (25:75) nanocatalysts is shown as the inset of Fig. 2a, which reveals the polycrystalline characteristic of the alloy nanoparticles. The lattice spacing of 0.222 nm for (1 1 1) plane of the fcc structured Pt–Ni is larger than that (0.203 nm) of pure Ni.30 However, this value is slightly smaller than that of the (1 1 1) plane of Pt (0.23 nm),46 which may be due to the lattice contraction upon substitution of a Pt atom with a Ni atom. The TEM image of the bare Pt–Ni catalyst (Fig. 5d) shows that 3–4 nm-sized particles are aggregated to form an assemblage of around 50 nm. The size of the assemblage for bare Pt–Ni nanoparticles is higher compared with the RGO incorporated one, which results in a lower catalytic activity as discussed later. The loading amount of Pt and Ni (Pt:Ni) in bare Pt–Ni and RGO/Pt–Ni nanocatalysts with different atomic ratios is analyzed by EDX and ICP results and shown in Table 1. The atomic ratios of Ni and Pt determined by the EDX analysis are consistent with the results obtained by ICP. It is also found from the ICP result that the total amounts of Pt and Ni (Pt + Ni) on RGO sheets are in good agreement with the initial loading amount.
Nanocatalysts | Pt:Ni (EDX) | Pt:Ni (ICP-AES) | PtNi contenta |
---|---|---|---|
a The contents of Pt–Ni alloys in the samples were determined by ICP-AES. | |||
Pt–Ni (25:75) | 24.2:75.8 | 25.3:74.7 | 100 |
RGO/Pt–Ni (50:50) | 44.9:55.10 | 49.1:50.9 | 37.85 |
RGO/Pt–Ni (33:67) | 33.1:66.9 | 32.9:67.1 | 38.05 |
RGO/Pt–Ni (25:75) | 25.3:74.7 | 24.4:75.6 | 37.69 |
The field-dependent magnetic behavior of bare Pt–Ni and RGO/Pt–Ni nanocatalysts with different atomic ratios of Pt and Ni at room temperature (RT, 300 K) and low temperatures (LT, 5 K) are shown in Fig. 6a and b, respectively. The values of magnetization (M, at 5 Tesla), remanence (Mr), and coercivity (Hc) of RGO/Pt–Ni nanocatalysts at RT and LT are listed in Table S1.† The magnetization values of RGO/Pt–Ni nanocomposites decrease with the increase in the Pt content and are independent of the measured temperature, which is consistent with the result reported for bare Pt–Ni nanostructures.47 The higher the proportion of Ni, higher is the value of magnetization. From Fig. 6, it is evident that anisotropy plays an important role here, because the room temperature M–H curves of RGO/Pt–Ni nanocatalysts show definite magnetization but very small hysteresis and remenance values. Nevertheless, at LT, a large hysteresis and remenance are observed. The magnetization value of the RGO/Pt–Ni nanocatalyst at RT is lower than the reported magnetization values of the Pt–Ni alloy film.48,49 This may be attributed to the smaller particle size and the possible presence of the passivating surface layer of metal oxide (NiO) (formation of NiO at the surface of nanocatalysts is confirmed by XPS spectra).27
Fig. 6 Magnetic hysteresis loops of the Pt–Ni(25:75) and RGO–Pt–Ni nanocatalysts with Pt–Ni atomic ratios of 25:75, 33:67 and 50:50 at (a) 300 K and (b) 5 K, respectively. |
These studies indicate that at a low temperature, the composites are ferromagnetic but at RT, they demonstrate a superparamagnetic behavior. It can be observed that the magnetization value is not saturated even after applying a strong magnetic field (50 kOe) due to the existence of exchange coupling between the ferromagnetic Ni and the adjacent anti-ferromagnetic NiO.50,51
The catalytic reduction took place due to the transfer of electrons from BH4− (donor) to p-nitrophenol (acceptor) through the nanocatalysts. The rate of electron transfer on the nanocatalysts surface was influenced by three processes: (a) adsorption of p-nitrophenol onto the nanocatalysts surface, (b) interfacial electron transfers and (c) desorption of p-nitrophenol left from the nanocatalysts surface. Since both the p-nitrophenol and p-aminophenol absorb in the UV-vis region, the progress of the reaction was monitored by UV-vis spectroscopy. It is well known that p-nitrophenol shows a strong absorption peak at 400 nm in an alkaline solution.55,56 As the reduction reaction proceeds, the intensity of the absorption peak at 400 nm gradually decreases, while a new peak appears at 300 nm, which is ascribed to the p-amino phenol. The progress of the reduction process is clearly visible to the naked eye, because the yellow color of the p-nitrophenol solution fades out slowly.
Fig. 7 shows the UV-vis spectra of the diluted reaction solution measured at intervals of 5 minutes using RGO–Ni, RGO–Pt, Pt–Ni (25:75) and RGO/Pt–Ni (25:75) as catalysts. It has been observed that the reduction reaction did not occur in the absence of catalysts or in the presence of pure RGO, even after two days of experimentation. However, in the presence of the catalyst, the absorption due to p-nitrophenol at 400 nm decreases, while there is an increase in the absorption at 300 nm as the reaction proceeds. The absorption at 300 nm corresponds to the formation of p-aminophenol. It is noticed that in comparison to RGO–Ni, RGO–Pt and Pt–Ni (25:75), the absorption intensity at 400 nm decreases much faster in the case of the RGO/Pt–Ni (25:75) nanocatalyst, as shown in Fig. 7(a–d), respectively. The conversion (%) of p-nitrophenol to p-aminophenol in 30 minutes are 12, 33.8, 40, 64.9, 78.4 and 86.5 for RGO–Ni, RGO–Pt, Pt–Ni (25:75), RGO/Pt–Ni (50:50), RGO/Pt–Ni (33:67) and RGO/Pt–Ni (25:75), respectively. The reduction rates of Pt–Ni, RGO–Ni, RGO–Pt and RGO/Pt–Ni nanocatalysts are compared in Fig. 5. It can be seen that the reduction rates in RGO/Pt–Ni nanocatalysts are in the following order: RGO/Pt–Ni (50:50) < RGO/Pt–Ni (33:67) < RGO/Pt–Ni (25:75), i.e., the catalytic activities increase with an increasing amount of Ni, whereas RGO–Ni, RGO–Pt and Pt–Ni (25:75) show a slower reduction rate than RGO/Pt–Ni (25:75) of the same composition.
Fig. 7 UV-vis absorption spectra of the reduction of p-nitrophenol by NaBH4 in the presence of (a) RGO–Ni, (b) RGO–Pt, (c) Pt–Ni (25:75) and (d) RGO/Pt–Ni (25:75) nanocatalysts. |
The kinetics of this reduction reaction is assumed to follow a pseudo-first-order to the concentration of p-nitrophenol when excess NaBH4 was used.57,58 Therefore, the kinetic equation of the reduction reaction may be given as follows:
kt = lnC0 − lnC = lnA0 − lnA |
Fig. 8 Plot of lnA400 vs. time for the kinetic studies of the reduction reaction of p-nitrophenol catalyzed by RGO–Ni, RGO–Pt, Pt–Ni and RGO/Pt–Ni nanocatalysts. |
It is clear from Fig. 8 that ln(Ct/C0) shows a good linear correlation (R2 > 0.99) with the reaction time for all catalysts confirming pseudo-first-order kinetics. The rate constant values are obtained from the pseudo-first-order reaction kinetics (using the slopes of the straight lines of ln(Ct/C0) versus time plot) for different catalysts and are given in Table 2.
Sample | RGO–Ni | RGO–Pt | Pt–Ni (25:75) | RGO/Pt–Ni (50:50) | RGO/Pt–Ni (33:67) | RGO/Pt–Ni (25:75) |
---|---|---|---|---|---|---|
a Reaction conditions: p-nitrophenol, 0.01 mmoles; catalysts, 3 mg; reaction time, 30 min. | ||||||
k (× 10−3) min−1 | 4.4 | 13.7 | 17.3 | 35.5 | 51.7 | 67.2 |
R2 | 0.9968 | 0.9966 | 0.9904 | 0.9985 | 0.9964 | 0.9958 |
TOFa (× 1017) molecules per gram per second | 1.3 | 3.8 | 4.5 | 7.2 | 8.7 | 9.6 |
The rate constant for the RGO/Pt–Ni nanocatalysts is higher than those of RGO–Ni and RGO–Pt. This seems to be caused by a smaller particle size and the synergetic chemical coupling effects of Pt–Ni alloy, which shows higher catalytic activity compared with monometallic Pt and Ni.28,32,59 The RGO/Pt–Ni (25:75) nanocatalyst shows higher catalytic effect than Pt–Ni (25:75). This indicates that the catalytic activity of bare Pt–Ni (25:75) can be enhanced surprisingly by compositing it with RGO sheets. Such an enhancement in catalytic activity by compositing with RGO can be ascribed to (1) the adsorption of p-nitrophenol on the surface of RGO through π–π stacking interactions that provides an increase in the concentration of p-nitrophenol in the vicinity of Pt–Ni (25:75) on the RGO/Pt–Ni (25:75) nanocatalyst, leading to a strong contact between them; (2) the increase in local electron concentration by electron transmission from RGO to Pt–Ni (25:75), which causes an enhancement in the electron-uptake process by p-nitrophenol molecules; and (3) RGO prevents the aggregation of Pt–Ni (25:75) nanoparticles and hinders the facile loss of activity.22,60
The different rates of reduction of p-nitrophenol with NaBH4 using RGO/Pt–Ni nanocatalysts with variable compositions of Pt and Ni (atomic ratios 50:50, 33:67 and 25:75) may be attributed to the modification in the electronic structure and the effect of segregation of the materials on the alloy surface. It has been reported that in the case of the Pt–Ni alloy, the catalytic effect is due to the presence of active sites on Pt, and the surface is enriched with them, while Ni enhances the catalytic effect.61 The electronic structure of Pt in the Pt–Ni matrix appears to be affected when Ni is added as an alloying element, and the metal composition affects the electron density (ne) of the alloyed matrix (Ptx–Ni1-x).62
ne,Pt–Ni = xne,Pt + (1 − x)ne,Ni |
The reusability of the bare Pt–Ni (25:75) and RGO/Pt–Ni (25:75) nanocatalysts was tested for the reduction of p-nitrophenol by NaBH4. It has been seen in the RGO/Pt–Ni (25:75) nanocatalyst (Fig. 9) that after each cycle (three), the value of rate constant slightly decreases with an increase in the number of cycles. In contrast, the rate constant (k) for the bare Pt–Ni (25:75) drops drastically in the second cycle (Fig. S3†). These experiments confirm that the stability of the Pt–Ni (25:75) nanocatalyst was effectively improved by the incorporation of RGO sheets. Furthermore, RGO sheets as a supporting material could help in preventing the aggregation of the Pt–Ni (25:75) nanocatalyst and the damage of the RGO/Pt–Ni (25:75) nanocatalyst framework. Therefore, a high stability in the catalytic activity is due to the high stability of the RGO/Pt–Ni nanocatalyst.
Footnote |
† Electronic supplementary information (ESI) available: XRD patterns of RGO–Ni, RGO–Pt and Pt–Ni of atomic ratio 25:75. FTIR patterns of GO and RGO/Pt–Ni nanocatalysts. Raman spectra of GO and RGO/Pt–Ni nanocatalysts. XPS spectra of RGO/Pt–Ni nanocatalysts. HRTEM images of RGO–Ni and RGO–Pt nanocatalysts. Plots of ln(Ct/C0) of p-nitrophenol versus reaction time for 2 successive reaction cycles employing bare Pt–Ni (25:75) as the catalyst. The table contains room temperature (RT) and low temperature (LT) magnetic data of the RGO/Pt–Ni nanocatalysts of different atomic ratios. See DOI: 10.1039/c4ra07686a |
This journal is © The Royal Society of Chemistry 2014 |