Nickel-Doped Silver Sulfide: An Efficient Air-Stable Electrocatalyst for Hydrogen Evolution from Neutral Water.

A low-cost, platinum-free electrocatalyst for hydrogen (H2) generation via the water splitting reaction holds great promise to meet the demand of clean and sustainable energy sources. Recent studies are mainly concerned with semiconducting materials like sulfides, selenides, and phosphides of different transition metals as electrocatalysts. Doping of the transition metals within the host matrix is a good strategy to improve the electrocatalytic activity of the host material. However, this activity largely depends on the nature of the dopant metal and its host matrix as well. To exploit this idea, here, in the present work, we have synthesized semiconducting Ag2S nanoparticles and successfully doped them with different transition metals like Mn, Fe, Co, and Ni to study their electrocatalytic activity for the hydrogen evolution reaction from neutral water (pH = 7). Among the systems doped with these transition metals, the Ni-doped Ag2S (Ni-Ag2S) system shows a very low overpotential (50 mV) with high catalytic current in neutral water. The trend in electrocatalytic activity of different transition metals has also been explained. The Ni-Ag2S system also shows very good stability in ambient atmosphere over a long period of time and suffers no catalytic degradation in the presence of oxygen. Structural characterizations are carried out using X-ray diffraction, X-ray photoelectron spectroscopy, transmission electron microscopy, and energy-dispersive X-ray spectroscopy to establish the phase purity and morphology of the materials.

Introduction

The rapidly increasing environmental hazards created due to burning of fossil fuels and demand of nonconventional energy sources have motivated researchers to explore a new class of renewable carbon-free energy alternatives. Among the recently used traditional fuels, hydrogen is advocated as a promising clean and renewable energy carrier. A potential way to produce hydrogen is via water splitting either by light or electricity. The vital step involved in water splitting is the hydrogen evolution reaction (HER), which requires an efficient catalyst to lower its large overpotential value. To produce hydrogen fuel on a large scale, the electrocatalyst should have the capability to work in aqueous solutions at neutral or near-to-neutral pH with a small overpotential requirement and high catalytic rate.[1] The state-of-the-art electrocatalysts used for efficient hydrogen evolution reaction are Pt-group noble metals, but unfortunately there is a limitation in their use for large-scale hydrogen production because of their high price and low abundance. Therefore, in the present situation, exploration of Pt-free, efficient, earth-abundant electrocatalysts is highly desirable. There are several electrocatalysts that have already been explored including different semiconductor nanostructures of transition metal (TM) (Mo or W)-based sulfides,[2−15] selenides,[4,16−18] and phosphides,[19−24] which show very promising electrocatalytic activity.[25] However, most of the commonly used transition metal dichalcogenides are in the semiconducting 2H phase with lower conductivity. Because high conductivity promotes fast electron transfer in the electrocatalytic reaction, recent electrocatalysts need further improvement to enhance their HER performance. One of the finest strategies to improve the HER performance is to increase the active sites that act as molecular catalytic centers. However, the major problem of these catalysts is the poor activity and limited stability in neutral aqueous solution. Moreover, a little amount of oxygen can deactivate the electrocatalyst through the formation of its oxide and therefore during the long-term electrolysis process the cathode should be strictly separated from the oxygen-generating anode to avoid the oxygenation of the electrocatalyst.[26] To address these issues, further work is needed to design a new kind of electrocatalyst that will be useful in acid-free aqueous medium and work efficiently in aerobic conditions without losing its activity and stability. Thus, there is enormous scope to design electrocatalysts that have good conductivity and act in aerobic neutral medium where no extra source of proton is available.

Catalytic water splitting with silver-based electrocatalysts has not yet been well explored. Being a very low band gap semiconductor (∼1.1 eV), Ag2S has good electrical conductivity. There are inadequate reports available on the electrocatalytic activity of Ag2S.[27−30] In addition, chemical doping of metal chalcogenides with transition metals to tune the physical properties is an elegant approach.[31,32] It has been proposed that transition metal (Fe, Ni, Co) doping adjacent to sulfur edge sites can enhance the HER activity by lowering the hydrogen adsorption free energy (ΔGH). Density functional theory calculations also support this fact. However, the lowering of Gibbs free energy as well as the enhancement of electrocatalytic activity depends on the nature of dopant metal in the particular host material, and it varies from one system to another system. In the case of MoS2, it has been reported that Ni doping can enhance its HER activity almost 3 times.[33] Actually, the transition-metal-doped MoS2 structure offers more active sites, which accelerate the hydrogen evolution reaction. It is reported that a Zn-doped MoS2 structure shows superior catalytic activity toward HER, whereas structures doped with other transition metals show less activity.[34] In a recent study, it has been reported that among different transition-metal-doped CoS2 structures, Mn-doped CoS2 shows the best electrocatalytic activity, which has been confirmed theoretically as well as experimentally. Theoretical analysis reveals that among different transition metals Mn is found to be the best dopant to have smallest Gibbs’ free energy (ΔGH).[35] With the concept of active sites and Gibbs free energy of hydrogen adsorption, our intention is to find the appropriate transition metal dopant that can accelerate the electrocatalytic activity of the Ag2S system. In this regard, we have synthesized different transition-metal (Mn, Fe, Co, Ni)-doped Ag2S nanoparticles through the reflux technique and the materials are characterized by different microstructural analyses like field emission gun transmission electron microscopy (FEG-TEM) and X-ray diffraction (XRD). Doping with every transition metal was confirmed by energy-dispersive X-ray spectroscopy (EDS) analysis. We have carried out electrochemical measurements with each of the four transition-metal-doped Ag2S nanoparticles along with pristine Ag2S. It is seen that among all of the doped systems Ni-doped Ag2S shows the best electrocatalytic hydrogen evolution reaction, whereas Mn-, Fe-, and Co-doped Ag2S materials show less activity than that of even pristine Ag2S. The Ni-doped Ag2S material shows a very low overpotential of 50 mV with quite higher catalytic current in neutral medium. The presence of Ni ions lowers the overpotential of pristine Ag2S (∼175 mV) in neutral medium. The material also has good stability toward hydrogen evolution even after 12 h in aerobic conditions. It can be used easily in ambient atmosphere without any activity loss. We have also measured the charge transfer resistance of the sample by electrochemical impedance spectroscopy during HER. It gives an Rct of 882 Ω at 50 mV potential, which gradually decreases to 120 Ω with an increase in forward bias, suggesting efficient charge transfer during HER.[36] Moreover, the trend in electrocatalytic activity of different transition-metal-doped Ag2S nanoparticles has been explained by difference in electronegativity and hydration enthalpy values of different transition metals with respect to silver. Finally, doping of transition metal (e.g., Ni) into a low-band-gap semiconductor (Ag2S) will result in a new pathway toward an efficient hydrogen evolution reaction for green energy production.

Results and Discussion

XRD

After successful synthesis of Ag2S and Ni-doped Ag2S (Ni–Ag2S), they are characterized primarily by XRD.

Figure a,b shows the XRD patterns of Ni–Ag2S and Ag2S, respectively. The XRD pattern represents sharp diffraction peaks, which confirm the formation of a crystalline Ag2S nanostructure (JCPDS no. 140072). It is seen that upon addition of very small amount of Ni there are no significant changes in the XRD pattern, and only XRD peaks get slightly intense. This indicates that substitution of silver by nickel does not remarkably change the crystal lattice parameters. The absence of any extra peak also discards the possibility of formation of a nickel sulfide phase during the reaction process. Ag2S doped with a higher concentration of Ni (Ni(L)–Ag2S) also shows no significant changes in the XRD pattern, as given in the Supporting Information (Figure S1). We have also performed XRD analyses of other transition-metal (Mn, Fe, Co)-doped Ag2S nanoparticles, which have been shown in the Supporting Information (Figure S1). It is worth mentioning that there are no peak shifts in the XRD patterns after incorporation of Ni/Mn/Fe/Co, which indicates that these transition metals are very compatible for substitution into the lattice structure of the Ag2S matrix and do not form any other compounds. In Mn–Ag2S, a small hump appears in the 10–30° region, which is due to the glass on which XRD of the sample is performed.[37]

Figure 1

(a) X-Ray diffraction patterns of Ni–Ag2S and (b) Ag2S; (c) X-ray photoelectron spectroscopy (XPS) spectrum of Ag 3d of Ni–Ag2S; (d) XPS spectrum of Ni 2p of Ni–Ag2S; and (e) deconvoluted spectra of S 2p of Ni–Ag2S.

X-ray Photoelectron Spectroscopy (XPS)

The chemical composition and oxidation states of the elements present in Ni–Ag2S have been further investigated by XPS analysis. Figure c shows high-resolution XPS spectra of the Ag 3d region. The two main peaks are located at 367.86 eV (Ag 3d5/2) and 373.87 eV (Ag 3d3/2), which confirm the presence of the Ag+ oxidation state in the Ni–Ag2S phase. No binding energy peak corresponding to metallic silver has been detected that excludes the possibility of formation of silver (Ag0) along with Ag2S.[27,38]

As shown in Figure d, the XPS spectrum of Ni 2p appears at binding energy values of 853.8 eV (2p3/2) and 871.3 eV (2p1/2). This indicates the presence of Ni2+ in the Ni–Ag2S structure.[39] The binding energy peaks of Ni 2p are less intense and broad in nature. This may be due to the presence of low nickel content in the nanostructure. Figure e shows the high-resolution deconvoluted spectra of S 2p. The deconvoluted spectra consist of three peaks. Of these three peaks, two components are centered at around 161.2 eV (2p3/2) and 162.4 eV (2p1/2). These two peaks are separated by spin–orbit splitting of 1.2 eV. According to the literature, the S 2p XPS spectrum of Ag2S appears at around 160.7 eV (2p3/2) and 161.9 eV (2p1/2).[38,40] This shifting in binding energy (0.5 eV) in our case may be due to the successful doping of nickel in Ag2S. The peak at around 163.8 eV indicates the presence of elemental sulfer, which justifies the slight excess of sulfer in the EDS analysis.

Microstructural Analysis

Figure represents field emission gun transmission electron microscopy (FEG-TEM) images of the Ni–Ag2S nanostructure. As shown in the transmission electron micrographs (Figure a–c), the Ni-doped Ag2S nanoparticles with an average diameter of 15–20 nm are grown.

Figure 2

(a–c) Low-resolution TEM images of Ni–Ag2S; (d) high-resolution TEM (HRTEM) image of Ni–Ag2S and the inset showing the fast Fourier transform (FFT) pattern of Ni–Ag2S; and (e) the corresponding EDS spectrum of Ni–Ag2S.

Figure d represents the high-resolution TEM image of Ni–Ag2S. It clearly shows the interplanar distances of 0.26 and 0.29 nm corresponding to the (1̅21) and (1̅12) planes, respectively, and the interplanar distance calculated from FFT (Figure d inset) is also found to be 0.265 nm, which is in good agreement with the XRD analysis. No lattice fringes associated with Ni or nickel sulfide phases have been detected. Therefore, to confirm the doping of Ni in the Ag2S nanostructure, energy-dispersive X-ray spectroscopy (EDS) analysis has been carried out, which has been shown in Figure e. The EDS spectrum of the material confirms the presence of Ni, Ag, and S elements. The atomic percentage of Ag/Ni/S is found to be 62.08:0.56:37.36. This indicates very low Ni content and slight excess of sulfur site in the Ni–Ag2S nanostructure. The FEG-TEM images of other transition-metal-doped Ag2S nanoparticles have been given in the Supporting Information (Figures S2–S4), which show very low atomic percentage of transition metal doping and also confirm the absence of other phases within the nanostructure. A higher percentage of Ni doping in Ni(L)–Ag2S has also been confirmed by EDS analysis given in the Supporting Information (Figure S5).

To check the Ni doping effect on the electrical conductivity of Ag2S nanoparticles, we have also measured the electrical conductivity of Ag2S and Ni–Ag2S as a function of temperature, as shown in Figure S6. The standard four-probe method with pellet samples has been used for conductivity measurements. It is seen that the conductivity of both the samples is decreased exponentially with a decrease in temperature. This confirms the semiconductor-like behavior of both samples. The room temperature conductivities are found to be 8.4 × 10–3 and 5.5 × 10–3 S/m for Ag2S and Ni-doped Ag2S, respectively. Because of band structure modulation, Ni doping caused some lowering of the electrical conductivity of the virgin Ag2S nanoparticles. Doping of a divalent transition metal (Ni2+) in the matrix of monovalent Ag2S produces extra charge cloud on the Ni sites, which act as additional scattering centres, and scattering of conduction electrons with these decreases the conductivity of the doped material.

Performance of Electrocatalytic Activity in Hydrogen Evolution Reaction (HER)

We have used pristine Ag2S and different transition-metal-doped Ag2S nanoparticles as the active electrocatalysts for HER processes. We have measured their catalytic activity by linear sweep voltammetry experiments using the standard three-electrode setup. Here, the as-synthesized materials drop-casted on an indium tin oxide (ITO)-coated poly(ethylene terephthalate) (PET) sheet (1 × 1 cm2) (see electrode preparation) act as working electrodes and Ag/AgCl acts as a reference electrode. The platinum (Pt) and glassy carbon electrode is used as a counter electrode. For assessing the benefit of the electrocatalyst in practical applications, we have carried out all of the experiments in neutral medium (pH 7) under ambient conditions. We have used the KH2PO4–K2HPO4 buffer, pH 7, solution and 100 mM KPF6 solution as the working electrolyte and supporting electrolyte, respectively, in all of the cases.

In the first step, we have checked the HER activity of the as-synthesized Ag2S nanoparticles at pH 7. This shows ∼175 mV overpotential toward HER, as shown in Figure a. To check the effect of doping of transition metal (TM) in Ag2S nanoclusters, we have carried out HER measurements also on Ag2S nanoparticles doped with different TMs like Mn, Fe, Co, and Ni. Among them, Ni is found to be the best dopant candidate toward HER activity, whereas other dopants (Mn, Fe, and Co) show very low activity compared to that of even pristine Ag2S nanoparticles. Ni-doped Ag2S shows a very low onset potential of 50 mV with very high current density. However, Co-, Fe-, and Mn-doped systems show quite higher onset potentials of ∼210, 350, and 485 mV, respectively, with moderate current density, as shown in Figure a. From Figure a, we see that for pristine Ag2S the cathodic current starts to increase at around 70 mV overpotential and that after 170 mV it increases rapidly. However, for Ni-doped Ag2S nanoparticles, the overpotential drastically decreases. After only 5 mV overpotential, the cathodic current starts to increase, and then it increases rapidly after 50 mV.

Figure 3

(a) Polarization curves at a scan rate of 2 mV/s of Ag2S, Ni–Ag2S, Co–Ag2S, Fe–Ag2S, and Mn–Ag2S; (b) corresponding Tafel plots; (c) Nyquist plots at different overpotentials during the HER process with Ni–Ag2S in neutral medium; and (d) stability test of Ni–Ag2S in neutral medium and aerobic atmosphere.

The order of electrocatalytic activity is Ni–Ag2S ≫ Ag2S > Co–Ag2S > Fe–Ag2S > Mn–Ag2S. These results are well explained by considering electronegativity and hydration enthalpies of the TM dopants. Due to the diagonal relationship, silver (Ag) and nickel (Ni) have similarity in their chemical properties including their electronegativity values. As we go from Ni to Mn, the absolute electronegativity values of the metals decrease and therefore difference between the electronegativity of metal and sulfide increases, resulting in the enhancement of negative charge density on sulfur, which strongly adsorbs H+ ions from water. Because of this enhancement in negative charge density, desorption/release of hydrogen will be much more slower.[2] Apart from this trend in electrocatalytic activity, it is found that electrocatalytic activity increases sharply when Ag2S is doped with Ni atoms. Although the electronegativity values of Ni and Ag are almost the same, still Ag2S shows much better electrocatalytic activity when it is doped with nickel. Here, hydration enthalpy plays a key role. Ni has a much higher value of hydration enthalpy compared to that of Ag as well as other transition metals (Fe, Co, Mn). Therefore, it has the ability to attract more water molecules from the system to form a nickel aqua complex. Being a Lewis acid, Ni can attract the electron cloud from H2O molecules, which makes proton release easier during the hydrogen evolution reaction. In the present case, doping of bivalent Ni in the monovalent Ag site in Ag2S produces an extra positive charge cloud on the Ni sites, which will effectively attract the electron cloud of H2O molecules to release proton easily during the hydrogen evolution reaction.

The electrocatalytic activity order of different transition-metal (TM)-doped Ag2S nanoclusters has been further verified with higher doping concentration samples. Although the onset potentials are increased, the trend of electrocatalytic activity is very similar to that with lower doping concentration samples, as shown in Figure S7. This result also confirms the activeness of lower doping concentrations in Ag2S nanoparticles. Furthermore, to examine any effect of Pt dissolution, we have further checked the electrocatalytic activity of Ag2S and Ni–Ag2S using the glassy carbon electrode (non-Pt electrode) as the counter electrode under similar conditons.[41] In Figure S8, we do not see any difference from our previous results. Therefore, it is confirmed that platinum (Pt) has no role in improving the electrocatalytic activity of our synthesized materials.

To carry out a detailed study of the HER activity, we have investigated the kinetics of the process by plotting Tafel slopes and fitted with the equation η = a + b log(j) (where b is the Tafel slope). Three principal steps are used in the conversion of 2H+ to H2, commonly referred to as the Volmer (eq ), Heyrovsky (eq ), and Tafel (eq ) equations.Combinations of steps (eqs and 2) or (eqs and 3), i.e., Volmer–Heyrovsky or Tafel–Volmer, can lead to the production of molecular H2. A relatively smaller Tafel slope indicates the Tafel–Volmer mechanism, whereas a higher Tafel slope gives the Volmer–Heyrovsky mechanism during the HER process. In our case, all of the materials show quite higher Tafel slopes and the Volmer–Heyrovsky mechanism has been operated. Pristine Ag2S shows a Tafel slope of 154 mV/dec, and all of the TM-doped Ag2S also show quite higher Tafel slopes except for the Ni-doped system. It shows a quite lower Tafel slope of 99 mV/dec for the Ni–Ag2S system. These results also support our proposed explanation of higher hydration enthalpy of Ni for greater HER activity compared to other TM doped Ag2S nano systems.

To check the electron transfer process in the Ni–Ag2S system, we have measured the change of resistance of the sample during the HER process. From the Nyquist plot in Figure c, it is clearly seen that Ni–Ag2S has the resistance in a megaohm (MΩ) range before starting the HER process. Just after the commencement of HER, the resistance drops suddenly to 1.5 kΩ. As we have increased the applied potential, the resistance decreased sharply. Therefore, this strongly supports the charge transfer effect in the system during this HER process. The resistance (882 Ω) corresponding to charge transfer at 50 mV potential decreases to only 120 Ω with the increasing bias and suggests that better charge transfer makes the electrocatalyst more conducting during the HER process. Nyquist plots of bare Ag2S and Mn-, Fe-, and Co-doped Ag2S nanoparticles have been given in the Supporting Information (Figure S9).

In addition to low overpotential, the long-term stability of the electrocatalyst in aerobic conditions is another important parameter for its practical use. The durability of the material (Ni–Ag2S) has been assessed at neutral pH as well as in aerobic conditions. From Figure d, we can see that our synthesized material is quite stable after 14 h of electrocatalysis even at a higher overpotential (200 mV). Thus, our material shows quite good activity in aerobic conditions. We have also performed different characterizations like XRD and FEG-TEM of our Ni–Ag2S sample after the HER experiment to observe any structural changes in it. No significant structural changes have been observed for the nanostructure, as shown in Figure S10. Thus, this Ni-doped Ag2S material can be explored as an effective electrocatalyst in HER at pH 7 regarding its practical applications.

Conclusions

In summary, to explore a transition-metal-doped, low-band-gap, and air-stable electrocatalyst toward the hydrogen evolution reaction (HER), we have successfully synthesized different transition-metal (Mn, Fe, Co, Ni)-doped Ag2S nanoparticles via a simple wet chemical synthesis route. Among them, Ag2S doped with Ni shows the best electrocatalytic activity in neutral water medium, where nickel doping lowers Gibbs’ free energy of hydrogen adsorption, resulting in a very low overpotential of 50 mV. It also shows very good stability (>12 h) in ambient atmosphere. The charge transfer resistance (Rct) is also found to be as low as 120 Ω under the bias voltage indicating superior charge transport for better HER.

Experimental Section

Chemicals

Silver nitrate (AgNO3), nickel chloride hexahydrate (NiCl2·6H2O), manganese chloride (MnCl2), ferric chloride (FeCl3), cobalt chloride (CoCl2), oleylamine, 1-dodecanethiol, potassium hexafluorophosphate (KPF6), and pH 7 buffer are purchased from Sigma-Aldrich. Disodium hydrogen phosphate dihydrate (Na2HPO4·2H2O) and all of the spectroscopy-grade solvents like n-hexane, ethanol, and acetone are bought from Merck Chemicals. All of the chemicals are of analytical grade and used as received.

Synthesis of Catalysts

The Ni-doped Ag2S nanoparticle has been synthesized by a one-step temperature-controlled reflux method. Typically, 0.5 mmol of AgNO3 and 0.005 mmol of NiCl2·6H2O have been taken in 5 mL of oleylamine and allowed to stir under an inert atmosphere to form a homogeneous mixture. Then, 25 mL of 1-dodecanethiol is added to the mixture and the solution is allowed to keep at 120 °C for 1 h. Then, the resultant light green solution is slowly heated upto 240 °C and kept at this temperature for 1.5 h. The resultant blackish-gray solution is cooled down to room temperature, and a small amount of hexane is added to this solution. Precipitation is obtained by addition of excess ethanol and acetone to the solution. The final product is obtained by washing the precipitate several times by the re-precipitation method.

Mn-, Fe-, and Co-doped Ag2S nanoparticles were prepared by the same procedure with addition of MnCl2, FeCl3, and CoCl2 (i.e., 0.005 mmol), respectively.

Pure pan class="Gene">Ag2S is obtained by following the same procedure without addition of the transition pan class="Chemical">metal salt.

To check the effect of doping concentration, we have also synthesized another batch of doped Ag2S nanoparticles, namely, M(L)–Ag2S (M = Ni/Mn/Fe/Co) with a higher concentration of transition metal halides (0.01 mmol) following the same procedure.

Physical Characterization of Ag2S, Mn–Ag2S, Fe–Ag2S, Co–Ag2S, and Ni–Ag2S

Microscopic structures of these materials are determined by transmission electron microscopy (TEM), and energy-dispersive X-ray spectroscopy (EDS) has also been performed to analyze the elemental composition. A PPMS system with a closed-cycle cryostat (Janis) having a cold head (Sumitomo) has been used for the conductivity measurement of the samples. X-ray diffraction has been carried out by a X-ray diffractometer (RICH SEIFERT-XRD 3000P) using an X-ray generator (Cu, 10 kV, 10 mA, and wavelength 1.5418 Å). X-ray photoelectron spectroscopy (XPS) is performed using an OMICRON-0571 system. TEM is performed using JEOL 2011, which is equipped with an EDS tool. All of the electrochemical measurements are carried out using an electrochemical analyzer, model CHI760E.

Electrode Preparation

For the three-electrode system, we have used Pt (CHI760E) as the counter electrode, Ag/AgCl (CHI760E) as the reference electrode, and our as-synthesized different-transition-metal-doped Ag2S as well as bare Ag2S as the working electrodes. To prepare the working electrode, we have taken 5 mL of the as-synthesized Ag2S in hexane solvent and sonicated it for 30 min to prepare a homogeneous dispersion. Then, we have drop-cast the sample over a 1 × 1 cm2 ITO-coated poly(ethylene terephthalate) (PET) sheet to make the working electrode. Finally, we have dried the electrode in a vacuum oven before use.