Cathodic Electrodeposition of Stoichiometric Cobalt Chalcogenide Thin Films

All chemicals were from Sigma-Aldrich and used without further purification. For voltammetry and EQCM, an EG&G Princeton Applied Research 263A instrument equipped with Power Suite electrochemistry software, a Seiko EG&G model QCA 922 instrument and an oscillator module (QCA 922-10), was used. A single compartment, three-electrode cell setup was used for electrochemical experiments at room temperature and comprised of an AT-cut, Pt-coated quartz crystal (geometric area, 0.2 cm²) working electrode, a Pt counterelectrode, and a Ag/AgCl/3 M NaCl reference electrode. All potentials below are quoted with respect to this reference electrode. For voltammetry, the potential scan rate was 20 mV s⁻¹. The cleanliness of the Pt working electrode surface was checked by cyclic voltammetry in 0.5 M H₂SO₄ by cycling potential between 0.8 V and −0.6 V until the voltammetric and EQCM frequency signals were stable. The electrolytes were degassed with high-purity nitrogen prior to the electrochemical measurements and nitrogen blanket was used during the measurements.

Film morphology and composition were obtained on a field emission scanning electron microscope (JEOL Model JSM-7601F) equipped with an energy-dispersive X-ray emission analysis (EDX) probe. X-ray photoelectron spectra were obtained using a Thermo Scientific X-ray photoelectron spectrometer with a monochromatic Al Kα X-ray source.

The two-step approach for the electrosynthesis can be successfully employed for the synthesis of CoS only when pre-deposited S film is reduced before any reduction of Co²⁺ ions occurs.²⁵ To examine this, a linear sweep voltammogram (LSV) combined with EQCM frequency changes, was first obtained for the Pt electrode in 0.1 M Na₂SO₄ containing 0.15 M CoCl₂. As shown in Fig. 1A, deposition of cobalt started at −0.75 V, evident from the EQCM frequency decrease (mass increase) accompanying the cathodic wave. The peak at −1.10 V was assigned to the reduction of Co²⁺ ions to Co.^17,26 A sulfur film, which was pre-electrodeposited at +0.6 V for 300 s in 0.1 M Na₂SO₄ containing 20 mM Na₂S, was reduced below ∼−0.70 V in the 0.1 M Na₂SO₄ blank electrolyte (Fig. 1B). The pre-deposited S film was almost completely stripped, as deduced from the comparison of frequency changes of Pt electrode before deposition and after stripping. The cathodic peak at −0.80 V was assigned to the reduction of S⁰ to S²⁻. The data in Figs. 1A and 1B thus demonstrated that a two-step approach could indeed be applied to the electrosynthesis of CoS thin films because Co²⁺ ions were reduced at more negative potentials than S film stripping.

Zoom In Zoom Out Reset image size

Figure 1C shows a linear sweep voltammogram simultaneously with EQCM frequency changes for a S-modified Pt electrode in 0.1 M Na₂SO₄ containing 0.15 M CoCl₂. The sulfur film was pre-electrodeposited on Pt at +0.6 V in 0.1 M Na₂SO₄ containing 20 mM Na₂S. Unlike in Fig. 1B, reduction of S to S²⁻ ions resulted in a frequency decrease and mass uptake. Thus, the reaction of electrogenerated S²⁻ with Co²⁺ in electrolyte resulted in CoS films being formed in situ on the Pt substrate. It is worth noting that the solubility product of CoS is very low, 5 × 10⁻²³,²⁷ attesting to the fact that the precipitation reaction is favored. The very small cathodic peak at ∼−0.46 V is assigned to the underpotential deposition of Co on the S-modified electrode surface.

Based on the data in Fig. 1, CoS films were potentiostatically synthesized using a S-modified Pt electrode at −0.7 V in 0.1 M Na₂SO₄ containing 0.15 M CoCl₂. A deposition potential of −0.7 V was selected based on the data in Fig. 1 since the potential was negative enough for S reduction, but insufficiently low for cobalt deposition to be triggered. However, rather surprisingly, EDX analyses of electrodeposited films at −0.7 V showed cobalt sulfide films with an atomic ratio of 2:1 with excess metallic Co content. This is attributed to the induced co-deposition of metal ion on S or CoS surface, which is well known for metal chalcogenides (e.g., CdSe, CdTe and CoSe).^18,28 This underpotential deposition of Co will result in CoS films with excess metallic Co.²⁶ This problem can be solved by employing complexing agents such nitrilotriacetic acid (NTA) to form a Co²⁺-NTA complex, which can shift the reduction potential of cobalt ions to more negative potentials.²⁶

Figure 2A demonstrates the effect of NTA on the potential for Co deposition on the Pt substrate. Compared to Fig. 1A, where cobalt deposition started at ∼−0.75 V, cobalt reduction began at ∼−1.1 V instead in a 0.1 M Na₂SO₄ electrolyte containing 0.15 M CoCl₂ and 0.2 M NTA, the complexation shifting the deposition threshold by more than ∼350 mV. Thus, the frequency decrease (mass increase) below −1.1 V can be attributed to the cobalt deposition:²⁶

$$\begin{eqnarray}&&{\rm{Co}}{\left({\rm{NTA}}\right)}^{-}+2{{\rm{e}}}^{-}\,\to \,{{\rm{Co}}}^{0}+{{\rm{NTA}}}^{3-}\end{eqnarray} \tag{ 1 }$$

Zoom In Zoom Out Reset image size

Now, a two-step approach was employed again for the synthesis of CoS films using on the S-modified Pt electrode in 0.1 M Na₂SO₄ electrolyte containing both 0.15 M CoCl₂ and 0.2 M NTA. As shown in Fig. 2B, S stripping started at ∼−0.70 V (same as in Fig. 1B) and the electrogenerated S²⁻ ions reacted with cobalt species to produce CoS films on the Pt substrate. The frequency decrease in Fig. 2B with cathodic current due to S reduction starting at ∼−0.70 V was diagnostic of CoS film formation. The peak at ∼−1.2 V in Fig. 2B was due to bulk deposition of cobalt, in good agreement with Fig. 2A. Interestingly, a small peak at ∼−1.05 V was observed in Fig. 2B. Compared to Fig. 1A, it could be attributed to the reduction of uncomplexed, free Co²⁺ ions to Co.²⁶

For the synthesis of CoS films, pre-deposited S layers were potentiostatically electroreduced at −0.85 V in 0.1 M Na₂SO₄ electrolyte containing 0.15 M CoCl₂ and 0.2 M NTA until the EQCM frequency was stabilized, indicating complete stripping of the sulfur film. EDX analyses of the as-synthesized films afforded an elemental stoichiometric ratio of 1:1 (Co:S = 48.8%:51.2%). Figures 3A and 3B displayed the EDX elemental mapping images for Co and S, respectively, demonstrating uniform distribution of the elements on the film surface.

Zoom In Zoom Out Reset image size

Further evidence of the surface composition of the as synthesized CoS films was obtained from XPS as shown in Figs. 3C and 3D. For the Co 2p spectrum (Fig. 3C), binding energies at 796.6 and 781.1 eV could be assigned to 2p_1/2 and 2p_3/2, respectively, consistent with a Co²⁺ oxidation state.¹⁰ Another two peaks at 802.7 eV and 786.2 eV were attributed to the 2p⁵3d⁷ (ground state of Co²⁺) and 2p⁵3d⁸L, which exhibits a charge transfer from the ligand p-band to metal d-band.^10,11 For the S region (Fig. 3D), a peak at 162.6 eV could be assigned to 2p, indicating the valence state of S to be −2.^10,11 Another peak centered at 168.3 eV could be assigned to SO₄²⁻, probably originating from adsorbed electrolyte species. Note that bulk cobalt sulfate formation can be neglected owing to the high solubility of this compound.²⁹

The two-step approach could be extended to another Co-chalcogenide, CoSe. A linear sweep voltammogram was obtained using a Se-modified electrode in a 0.1 M Na₂SO₄ blank electrolyte to examine whether a two-step approach could be applied to the formation of CoSe films (Fig. 4A). The Se film was pre-deposited on the Pt substrate at −0.60 V for 300 s in 0.1 M Na₂SO₄ containing 20 mM SeO₂. Again, a cathodic wave centered at ∼−0.73 V was attributed to the reduction of Se to Se²⁻ and the EQCM frequency increase (mass decrease) up to that of a bare substrate after a cathodic scan clearly indicated complete stripping of the Se layer.

Zoom In Zoom Out Reset image size

Figure 4B shows a voltammogram along with EQCM frequency change for a Se-modified electrode in a 0.1 M Na₂SO₄ electrolyte containing 0.15 M CoCl₂ and 0.2 M NTA. Again, a cathodic wave with mass increase starting at ∼−0.70 V was assigned to the reduction of selenium to Se²⁻ and formation of CoSe by reaction of Se²⁻ with Co²⁺ in solution. Two cathodic peaks were seen due to cobalt deposition, as explained in Fig. 2B. The trends in frequency changes and currents due to Se reduction and subsequent reaction with cobalt species were very similar to those in Fig. 2B, although frequency changes were less pronounced here.

Compositional analyses of the as-deposited films resulted in CoSe with an atomic ratio of 1:1 (50.2%:49.8%) and EDX elemental mapping images demonstrated again uniform distribution of Co and Se on the Pt surface (Figs. 5A and 5B). The surface composition of the films was obtained from XPS analyses of as-deposited CoSe films. For the Se region (Fig. 5D), a peak centered at 54.5 eV was assigned to Se 3d_5/2, consistent with the existence of Se²⁻.^2,3,30 Another peak at 59.7 eV could be attributed to Se 3d_3/2 in SeO₂ resulting from surface oxidation.^30,31 The high-resolution XPS peaks observed in Fig. 5C from cobalt were in agreement with corresponding results in Fig. 3C.

Zoom In Zoom Out Reset image size

The data above demonstrate that the two-step electrodeposition sequence can be combined with metal complexation to secure the electrosynthesis of stoichiometric CoQ (Q = S or Se) without the need for any post-deposition steps. Next, the use of these films in a practical application context, is presented. For this purpose, a redox probe, namely iodide/triiodide, was selected. This redox couple is commonly used as a shuttle in dye-sensitized solar cells where Co-based chalcogenides have been shown to serve as a counterelectrode.²⁵

Electrocatalytic behavior of electrosynthesized CoQ (Q = S or Se) films

Figure 6A shows cyclic voltammograms for CoS and CoSe electrodes prepared by the approach described above in 0.1 M LiClO₄ containing 10 mM LiI and 1 mM I₂. For comparison, a voltammogram obtained using a Pt foil electrode is also shown in Fig. 6A. As shown in the figure, two pairs of oxidation and reduction peaks were observed in all the voltammograms. The pair of peaks Ox₁ and Red₁ at relatively negative potentials could be attributed to redox reaction 2 and the other peaks Ox₂ and Red₂ at more positive potentials were assignable to reaction 3:^3,19,23

$$\begin{eqnarray}&&{{{\rm{I}}}_{3}}^{-}+2{{\rm{e}}}^{-}=3{{\rm{I}}}^{-}\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&3{{\rm{I}}}_{2}+2{{\rm{e}}}^{-}={{2{\rm{I}}}_{3}}^{-}\end{eqnarray} \tag{ 3 }$$

Since the main role of the counterelectrode in DSSCs is to effectively catalyze the reduction of I₃^– to I^–, the redox peaks Ox₁ and Red₁ were analyzed based on the reduction current density (J_Red) and the peak-to-peak potential separation (E_pp). Generally, larger J_Red and smaller E_pp are diagnostic of better electrocatalytic activity.²¹ Values for J_Red and E_pp (Table I) increased in the order: Pt < CoSe < CoS and Pt < CoS < CoSe, respectively, demonstrating the comparable electrocatalytic performance for the electrodeposited CoS electrode to the Pt "standard." The improved electrochemical activity of electrochemically deposited CoS and CoSe electrodes may largely be due to a higher active surface area, in agreement with previous results.¹⁹

Zoom In Zoom Out Reset image size

Table I.  Comparison of electrocatalytic activities of counterelectrodes for the reduction of I₃⁻ to I⁻.^a)

	Pt	CoS	CoSe
JRed (mA cm−2)	1.235	1.875	1.825
Epp (mV)	285	355	407

^a)Data from Fig. 6A.

To assess electrochemical stability, twenty consecutive cyclic voltammograms were obtained for the CoS electrode in an acetonitrile solution containing 0.1 M LiClO₄, 10 mM LiI and 1 mM I₂ as shown in Fig. 6B. No substantial changes in J_Red and E_pp were observed for the CoS electrode after 20 cycles, indicating good electrochemical stability. A linear relationship between current density of Ox₁ (or Red₁) and square root of the scan rate was observed using the CoS electrode (not shown). This implies that the redox reaction is diffusion limited for the transport of the I⁻/I₃⁻ redox couple and there is no specific interaction between the redox couple and CoS electrode surface.

In summary, CoS and CoSe films were successfully electrosynthesized by the two-step approach using S- or Se-modified electrode in a Co²⁺-NTA complexing electrolyte. NTA was employed to shift the Co deposition potential to more negative potentials, and consequently, excess Co co-deposition could be avoided. Thus, stoichiometric CoQ (Q = S or Se) films with the component elements in a 1:1 ratio could be electrosynthesized without any post-deposition step unlike in previous studies. The as-synthesized CoS and CoSe films were successfully applied for the reduction of triiodide ions and found to be promising counterelectrode materials for possible use in DSSCs. Finally, the approach described here can be utilized for the facile synthesis of other metal chalcogenides in general.

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), Ministry of Education (Grants NRF-2016R1D1A1B02010133 and NRF-2018R1D1A1B07040855). We thank the two anonymous reviewers for constructive criticisms of an earlier manuscript version.