Role of Noble- and Base-Metal Speciation and Surface Segregation in Ni_2–xRh_xP Nanocrystals on Electrocatalytic Water Splitting Reactions in Alkaline Media

Abstract

Transition-metal phosphides have proven to be surprisingly active electrocatalysts for electrochemical water splitting, but the nature of the "active" catalyst depends strongly on the solution pH, the identity of the metals, and whether the reactions are anodic [oxygen evolution reaction (OER)] or cathodic [hydrogen evolution reaction (HER)]. In order to understand the origin of this activity, the synthesis of well-defined, compositionally controlled precatalysts is needed, as are detailed catalytic studies and physicochemical characterization/activity assessment of catalysts at different stages. While base-metal phosphides of Ni and Co have the advantage of being earth-abundant, in alkaline media, they are less active and less stable than noble-metal phosphides such as Rh 2 P. As a means to combine the abundant nature of base metals with the activity and stability of noble metals, the first synthesis of colloidal Ni 2−x Rh x P nanocrystals by arrested precipitation routes is reported along with their composition-dependent activity for electrocatalytic HER and OER. Phase-pure samples of Ni 2−x Rh x P were realized at the Ni-rich (hexagonal, Fe 2 P-type) end (x = 0.00, 0.25, 0.50) and Rh-rich (cubic, antifluorite-type) end (x = 1.75, 2.00). When assessed in terms of current density normalized to electrochemical surface area (ECSA) at a fixed potential, the most active precatalyst for OER is Ni 1.75 Rh 0.25 P, and for HER, it is Rh 1.75 Ni 0.25 P. Evaluation of X-ray photoelectron spectroscopy, transmission electron microscopy/ energy-dispersive spectroscopy and ECSA data before and after 10 h stability runs were performed. The data reveal surface compositions to be considerably richer in Ni and poorer in Rh and P relative to the bulk composition, particularly for Ni 0.25 Rh 1.75 P, where the surface ratio of Ni/Rh is nearly 2:1 and increases to 4:1 after HER catalysis. In all cases, surface phosphorus is completely depleted post catalysis, suggesting a sacrificial role for phosphide under alkaline conditions. Moreover, the activity of "Rh 1.75 Ni 0.25 P" for HER decreases over time, even as the ECSA continues to rise, attributed to a decrease in the more active and stable Rh sites relative to Ni on the surface. In contrast, the enhancement in OER activity of Ni 2 P with 12.5% Rh incorporation is attributed to restructuring upon phase segregation of Rh, suggesting that the noble metal may also play a sacrificial role and not directly participate in OER catalysis. The roles of minority noble metals (Rh) in base-metal phosphides for OER and of minority base metals in noble-metal (Rh) phosphides for HER are discussed in light of related data on Co 2−x Rh x P.

A Little Nickel Goes a Long Way: Ni Incorporation into Rh2P for Stable Bifunctional Electrocatalytic Water Splitting in Acidic Media

“…We are investigating phosphide solid solutions that combine noble- and base-metal components to explore possible synergies that may enable high activity and stability at lower noble-metal concentrations. Our studies on Co 2– x Rh x P ( x = 1.75 for HER and x = 0.25 for OER) and Ni 2– x Rh x P ( x = 0.25 for OER) in basic media showed higher electrocatalytic activities compared to their monometallic end members, with Ni 0.25 Rh 1.75 P ( x = 1.75) exhibiting a higher overpotential for HER relative to Rh 2 P when compared on geometrical surface area, but a lower overpotential when normalized against the electrochemical surface area (ECSA). ,, In acidic media, we found that adding 37.5% Co into the Rh 2 P cubic antifluorite structure to form Co 0.75 Rh 1.25 P retained the crystal structure and high (initial) HER activity of Rh 2 P . However, Co dissolution under HER conditions led to a loss in activity of ca 50% over 30 min before leveling off, whereas under positive potentials required for OER, the polarization curves were irreversible and broad, indicating the anode is being consumed. , …”

“…All reactions were carried out under an argon atmosphere using standard Schlenk line techniques. To synthesize Ni 2– x Rh x P our previously developed synthesis method was utilized . Briefly, (0.42– x ) mmol Ni to x mmol Rh was combined with 15.0 mL of oleylamine and 5.0 mL of 1-octadecene in a 200 mL Schlenk flask with an attached condenser.…”

“…Our studies on Co 2−x Rh x P (x = 1.75 for HER and x = 0.25 for OER) and Ni 2−x Rh x P (x = 0.25 for OER) in basic media showed higher electrocatalytic activities compared to their monometallic end members, with Ni 0.25 Rh 1.75 P (x = 1.75) exhibiting a higher overpotential for HER relative to Rh 2 P when compared on geometrical surface area, but a lower overpotential when normalized against the electrochemical surface area (ECSA). 7,8,13 In acidic media, we found that adding 37.5% Co into the Rh 2 P cubic antifluorite structure to form Co 0.75 Rh 1.25 P retained the crystal structure and high (initial) HER activity of Rh 2 P. 14 However, Co dissolution under HER conditions led to a loss in activity of ca 50% over 30 min before leveling off, whereas under positive potentials required for OER, the polarization curves were irreversible and broad, indicating the anode is being consumed. 14,34 In this study, we report the performance evaluation of our previously synthesized Ni 2−x Rh x P (x = 0.25, 0.5, and 1.75) nanomaterial catalyst 8 toward HER, OER, and overall water splitting (OWS) in acidic media, and in comparison to Rh 2 P and Ni 2 P end members.…”

“…7,8,13 In acidic media, we found that adding 37.5% Co into the Rh 2 P cubic antifluorite structure to form Co 0.75 Rh 1.25 P retained the crystal structure and high (initial) HER activity of Rh 2 P. 14 However, Co dissolution under HER conditions led to a loss in activity of ca 50% over 30 min before leveling off, whereas under positive potentials required for OER, the polarization curves were irreversible and broad, indicating the anode is being consumed. 14,34 In this study, we report the performance evaluation of our previously synthesized Ni 2−x Rh x P (x = 0.25, 0.5, and 1.75) nanomaterial catalyst 8 toward HER, OER, and overall water splitting (OWS) in acidic media, and in comparison to Rh 2 P and Ni 2 P end members. In contrast to Co 2−x Rh x P, we are able to identify (relatively) stable catalysts for both HER and OER, with the most stable bimetallic phosphide being Ni 0.25 Rh 1.75 P for both processes.…”

“…The development of inexpensive and efficient water electrolyzers for acidic media is a particular challenge because of the instability of the anodic catalysts, which perform the OER, generating electrons for hydrogen production at the cathode via the HER, and a lack of clear understanding of the OER mechanism and the nature of the catalytic sites under acidic conditions. − However, acidic water splitting is worth pursuing for several reasons: (1) due to the high mobility of H + ions relative to OH – ions, acidic media has a higher ionic conductivity than base, hence enabling a high current density; (2) acidic media dissolves CO 2 , thereby enabling long electrolyzer lifetimes relative to alkaline electrolyzers, where metal carbonates precipitate and poison active catalytic sites. ,,,− …”