1. Introduction
Humans are facing an undisputed increase in energy consumption linked with doubtless environmental pollution due to global economic and social development [
1,
2,
3]. Thus, it is urgent to adopt sustainable strategies to mitigate global warming and improve the quality of life. Hydrogen is a well-known molecule that has been proposed as fuel due to its high energy density and because it is eco-friendly [
4,
5,
6]. However, the production of green hydrogen entails high electrolysis potential causing high-energy consumption, which hinders its up-scaling application and deployment. Therefore, great efforts are needed to pay for the transition to a hydrogen economy both in the development of cost-effective materials with high efficiency and simplifying the electrocatalytic system components for large-scale applications and reducing totals cost [
7,
8,
9]. It is known that green hydrogen can be produced by water electrolysis or water splitting using the excess of electrical energy from renewable sources such as solar panels or wind turbines, but this requires effective and durable electrode materials with high electrochemical activity [
10,
11]. In this context, the scientific community has devoted much effort to develo** new promising methods to fabricate electrodes with remarkable properties [
12,
13,
14,
15]. Electrocatalytic materials play a critical role in the transition to the hydrogen economy because they can reduce the required energy to achieve the anodic and cathodic reactions in the water splitting process [
16,
17,
18]. However, several proposed catalysts are far from commercial application due to their low electrocatalytic efficiency or because they are too costly [
19].
One pathway to solving the sluggish kinetics of anodic and cathodic water splitting is the implementation of noble metal nanoparticles, such as platinum-based electrocatalysts, that maximize active centers and fast electron transport [
20,
21,
22].
It is well known that acid electrolysis cells have some advantages, such as lower gas permeability and good proton conductivity [
23]. Nonetheless, the acid media limit the overall water splitting to noble metals, which are very expensive. The literature shows that even benchmark Pt- or Ru-based electrocatalysts have demonstrated poor performance in alkaline media, which is mainly attributed to the adsorption of intermediates that mitigate direct oxygen and hydrogen evolution [
24,
25]. One advantage of alkaline media is that water splitting can be achieved on noble-metal-free electrocatalysts. Still, the design of efficient electrocatalysts for alkaline media is challenging even now. Due to the importance of hydrogen generation, recent reviews focus on summarizing the pros and cons of using non-noble metals for HERs and OERs [
26,
27].
Another explored option for H
2 and O
2 generation is photoelectrochemical water splitting; i.e., H
2 evolves by a sunlight-assisted reaction [
28,
29,
30]. To this end, photoelectrochemical cells are fabricated to directly convert the harvested sunlight into chemical energy, consisting of an n-type photoanode and p-type photocathode immersed in a solution containing an adequate redox pair [
31]. In this way, great attention has been paid to using semiconductive materials as a promising tactic for improving water separation. These photoanodes need to have a suitable band gap, highly efficient light adsorption, fast electron transfer, and overall good stability [
31,
32]. A relatively limited number of reviews consider using noble metal and noble-metal-free electrocatalysts for water splitting and only focus on specific production pathways.
This review is intended to summarize the progress of material design with activity for hydrogen evolution reactions (HER) and oxygen evolution reactions (OER). First, we discuss the implementation of noble metals. Then, we systematically study research focused on noble-metal-free heterostructures. Also, direct solar water splitting is discussed, and some parameters are analyzed to gain insight into the versatility of this technology. Finally, we discuss future opportunities and challenges in a hydrogen economy.
3. Water Splitting on Noble Metal Electrocatalysts
The scientific community has envisaged that electrochemical water separation shall be the future for generating green hydrogen. Nonetheless, this process suffers from the sluggish and high overpotential for H
2O dissociation [
33,
34]. In this sense, much work has been performed to find the best electroactive catalysts for boosting the industrial production of green hydrogen resources at a large scale. Using noble-metal-based electrocatalysts seems to be the most viable tactic to improve the electrochemical electrode–electrolyte process in water splitting [
35]. Since these metals are expensive and scarce, it is mandatory to optimize their utilization. Researchers have proposed designing noble electrocatalysts by tuning their morphology, particle size, defects, and interface engineering to improve the electrocatalytic activity [
36,
37]. For example, Tang and coworkers have designed and developed Ru-Cr
2O
3 supported on N-doped graphene (NG), in which Ru stimulated hydrogen coupling, and Cr
2O
3 and NG facilitated the hydroxyl species [
38]. This electrocatalyst reaches a current density of 10 mA cm
−2 at a very low overpotential of 47 mV for HERs in alkaline media. Meanwhile, Zhang et al. investigated the performance of hollow Ru nanoparticles for HERs in acid electrolytes [
39]. This material was fabricated by a two-step process in which a galvanic replacement is present. First, Ni nanocrystals were synthesized using Ni(COOH)
2, terpilenol, dodecylamine, and 1-octodecene. Then, the presynthesized Ni nanocrystals were dispersed in dimethylformamide (DMF) and heated at 160 °C. After that, a solution of Ru was dropped into the aforementioned mixture. Surprisingly, hollow Ru nanoparticles exhibit a very low overpotential, 29 mV, when reaching a current density of −10 mA cm
−2. This value is very close to that obtained with Pt/C electrocatalysts, 25 mV, when the electrode delivers the same current density.
In recent work, Ru clusters were developed on 2D-layered molybdenum metal carbide using an in situ reduction process [
40]. This material was used for electrocatalysts for HERs in neutral media and exhibited an overpotential of 73 mV when achieving a current density of −10 mA cm
−2. As can be observed from
Figure 1, the performance of Ru-based electrocatalysts is comparable to that obtained by using conventional Pt/C electrocatalysts. Furthermore,
Figure 1b) displays the obtained Tafel slopes for commercial Pt/C and Ru/Mo
2CT
x,, which are very similar. The authors mentioned that Ru/Mo
2CT
x offers good performance because this composite modulates the electronic structure resulting in facile H
2O dissociation and hydrogen desorption.
From the above-discussed research, it can be deduced that Pt-based electrocatalysts are currently the most active material for achieving water dissociation at low overpotential. In this regard, Wang and coworkers designed and fabricated low-cost Pt-based electrocatalysts to produce electrolyzed hydrogen water [
41]. This research group top-coated both sides of titanium plates with a Pt layer with three different thicknesses (100 nm, 200 nm, and 750 nm). The obtained coated electrode was used as a working electrode to study the electrolysis of acidic water and tap water. The electrochemical test showed the high efficiency of the 100 nm Pt layer sample in both acidic and tap water. The authors concluded that increased efficiency results from optimized H
2O, H
3O
+, and OH
− absorption on Pt (111) facets. Kori et al. reported Pt nanoparticles in a matrix of three-dimensional fibrous networks composed of peptide bolaamphiphile hydrogel [
42]. One relevant property of this material was the in situ reduction of the Pt species within the hydrogel, activating the synergistic effect. The electrochemical experiments show that this electrocatalyst can maintain a current density of −10 mA cm
−2 with an overpotential of 45 mV.
Another interesting work was conducted by Razavi et al., in which a metal–organic framework was used as support for Pt nanoparticles [
43]. To this end, a hydrothermal method was implemented to synthesize molybdenum sulfide nanosheets. Then, the desired amount of MoS was dispersed in deionized water, and an H
2PtCl
6 solution was added drop by drop under sonification. After that, the mixture was heated at 70 °C for 4 h. This electrocatalyst shows a slow overpotential of 41 mV dec
−1 in acid media with excellent stability for 20 h. In the same direction, Mei et al. studied a dedicated design based on 2D transition metal dichalcogenide (MoS
2, TiS
2, and TaS
2) nanosheets as support for Pt nanocatalysts. They adopted the electrochemical Li intercalation and exfoliation process to prepare the supports, while Pt nanoparticles were grown by the chemical reduction method [
44].
Smiljaníc et al. proposed a rational strategy to design a cost-effective Pt-based electrocatalyst [
45]. This research group uses conductive titanium oxynitride dispersed over reduced graphene as feasible support for Pt nanoparticles. The electrochemical measurements reveal that Pt/TiONx has a slightly lower electrochemical surface area than commercial Pt/C. Nevertheless, The CO strip** shows that Pt/TiON
x has the ability to oxidize the adsorbed CO at a lower potential, and this characteristic was attributed to a strong metal support interaction that activates the bifunctional effect and changes the adsorption energies of strongly bonded CO at the surface of Pt. The HER polarization curves in 0.1 mol L
−1 HClO
4 and recorded at 10 mV s
−1 demonstrate that Pt/TiON
x surpasses the commercial Pt/C activity. Due to the benefits of using Ti to develop advanced electrocatalysts for water electrolysis, much work has been reported. For instance, Amer et al. incorporate Pt nanoparticles into mesoporous titanium dioxide, and this composite was used as electrocatalysts for HERs in acid media [
46].
Iridium-based electrocatalysts are another group of promising nanomaterials regarding their catalytic activity, large pH window, and good stability [
47]. For example, the research group of Roy-conducted tests for the HERs and OERs on iridium grown on vertical graphene in both acid and alkaline media, observing overpotentials of 47 mV and 17 mV, respectively, to deliver a current density of 10 mA cm
−2 [
48]. A recent study proposed an engineering design to develop Ir nanoclusters embedded into a nitrogen-and-sulfur-doped graphene matrix, which is active at different pH levels for overall water splitting [
49]. This material was synthesized by mixing melamine, L-cysteine, and iridium chloride hydrate in ball-milling equipment and then thermally treated. The pH-universal overall water splitting experiments show that this electrocatalyst requires only 1.53, 1.42, and 1.45 V to deliver a current density of 10 mA cm
−2 in neutral, acid, and alkaline media. On the other hand, Kim et al. developed a self-supported water-splitting catalyst, VCoCOx@NF, using a one-step hydrothermal method to deposit a vanadium-doped cobalt carbonate bimetallic catalyst on NiVCoCOx@NF, which demonstrated excellent performance in overall water splitting in alkaline media, acting as a highly active bifunctional electrocatalyst for both OERs and HERs. It achieved a current density of 10 mA cm
−2 at overpotentials of 63 mV and 240 mV, respectively. The enhanced performance can be attributed to V species do** at the Co site, which reduces the overpotential by shifting the d-electron states of Co towards the Fermi level [
50].
As we know, the architecture of the materials plays a critical role in overall water splitting efficiency. In this sense, Bao et al. fabricated novel porous nano-hollows composed of Iridium with tunable wall thickness via the hydrothermal method [
51]. The OER experiments, in 0.5 mol L
−1 H
2SO
4 at 1 mV s
−1, show that the designed material performs comparably to the benchmark Ir/C, reducing its overpotential to 54 mV. Another approach was reported by Li et al. in which 1D iridium nanostructures were developed. The authors used a sacrificial template to obtain Ir-Te nanowires in a two-step synthesis process. The polarization curve for the OER displays that Ir-Te nanowires have almost the same overpotential to reach 10 mA cm
−2 in alkaline and acid media at 248 and 284 mV, respectively.
Recently, Ma and coworkers reported an unusual Lu
1.25IrO
xOH
y catalyst that is amorphous in phase [
52]. The theoretical calculation demonstrates that the high performance of this composite is attributed to the greater d-hole-containing electronic state of Ir
V as a result of cationic vacancies.
4. Hydrogen Evolution Reaction on Noble-Metal-Free Electrocatalysts
At present, most electrocatalysts proposed for water splitting in acid and alkaline solutions are based on noble metals. Still, these chemical elements are not abundant on the earth’s crust, severely limiting their widespread application in daily life [
53]. The literature shows that different approaches have been proposed to develop efficient heterostructures for H
2 and O
2 generation via water electrolysis. For a better understanding, this section separately studies the anodic and cathodic reactions in water splitting on noble-metal-free electrocatalysts.
Hydrogen is a promising energy vector used in many electrochemical processes because of its advantages of high energy density and being environmentally benign. This gas is not naturally available, and technological strategies have been developed to generate it efficiently without using harmful substances. Non-precious metals have mainly been studied for their potential as electrocatalysts for the H
2 evolution reaction in the green electrochemical water splitting process [
54]. This hot topic is considered a sustainable alternative due to the low cost and abundance of these elements. Nevertheless, precious metal-free electrocatalysts suffer from slow kinetics and de-activation under the operational conditions of electrolyzers. In this context, many authors have proposed that the implementation of composites based on Ni, Cu, Co, and Fe are a short distance from commercial application. Conventional single-principal-element alloys suffer from insufficient water-splitting activity, so many efforts have been made to design multimetallic catalysts. For instance, the Shi research group proposes the implementation of Mo(NiFeCo)
4/Ni) as a hierarchical nickel network of electrocatalysts for water splitting [
55]. In recent work, Zhang et al. demonstrated the versatility of a high-valence Co
2FeAlMo
1.5 alloy for HERs in alkaline media [
56]. This compound was easily synthesized by the coprecipitation method in the presence of polyvinyl pyrrolidone and ammonia. The obtained powder was thermally treated at 750 °C for 10 h in a mixed gas atmosphere of H
2 and Ar at 5% and 95 %, respectively. The electrochemical results suggest that a synergistic effect of the alloy promotes HER due to the optimized crystal and electronic structure of this thermally treated alloy. Linear sweep voltammetry measurements in 1 mol L
−1 KOH show that this material has outstanding characteristics regarding its low overpotential, 71 mV at 10 mA cm
−2. The kinetic parameters and the stability were improved, showing a shift by 2 mV after 1000 cycles. Thus, it can be deduced that the intrinsic electrochemical activity of non-precious metals can be tuned by alloying them with high-valence and stable ones. The latter produces vacancies, dislocation, and interstitial sites that generate active sites and then improve the electrochemical activity of this material.
Figure 2 illustrates the structure of Co
2FeAlMo
x supported on carbon fiber paper and its interaction with H
2O species to successfully achieve the electrochemical water splitting.
Recently, rare earth elements have attracted enormous attention because of their chemical and physical behavior dictated by electronic configuration. The empty or half-filled 4f electron configuration in the Lanthanide group determines the optical and magnetic properties. Moreover, these outstanding elements can exhibit different valence states that change their ionic radii. Cai and coworkers studied the effect of lanthanide and phosphide incorporation into nickel–iron double-layer hydroxides [
57]. This electrocatalyst delivers a current density of 10 mA cm
−2 at an overpotential of 158 mV.
Sun’s group found that introducing Ln
2O
3 into the Ni matrix promotes hydrogen evolution reactions in alkaline media [
58]. The material composed of Ni/Yb
2O
3 was fabricated by selective high-temperature reduction. This compound shows good oxophilicity and stability, which reduces the energy of activation and corrosion under the operation condition. An interesting work was conducted by Zhang and coauthors in which La, Ce, and Nd were used to decorate ZIF-67 support, showing that these rare earth elements regulate the electronic coordination structure with the Co structure located in the support [
59].
Tan et al. reported a co-doped MoO
2/Mo
2N
3 heterostructure incorporating erbium [
60]. The latter leads to a change in electron density distribution that modifies the Gibbs free energy of intermediates, improving the HER.
Figure 3 presents the electrochemical response of these materials. From
Figure 3a, it can be observed that the sample Er-MoO
2/Mo
2N
3@P,NC surpasses the performance of the Pt/C material at high current densities and the EIS measurements.
Figure 3b reveals that the improvement of the proposed composite is due to its low charge transfer resistance, as can be inferred from the small semicircle diameter.
Transition metal nitrides have triggered interest in energy applications in recent years due to their chemical stability, excellent electrical conductivity, and tunable morphology [
61,
62]. Moreover, the nitrogen species in transition metals increase the electron density of the d-band, which is advantageous because the electronic structure is quite similar to the noble metals [
63]. In this context, Zhou et al. reported a Fe-N-WC nanoarray as an electrocatalyst for the HER [
64]. The compound was synthesized using ferrocene and a WO
3 nanoarray on carbon fiber paper and was thermally treated at 850 °C for 3 h under an Ar:NH
3 atmosphere.
Figure 4 illustrates the preparation sequence obtained by following a sequential two-step approach [
64]. First, the WO
3 nanoarray was grown on a carbon fiber surface using hydrogen peroxide and tungstic acid. Then, ferrocene was thermally treated on the former using an in situ gas–solid reaction under NH
3 as a rich source of N.
In a separate work, Zhang et al. developed Ni
2P/@NC/NF using a two-step synthesis [
65]. Firstly, the nickel foam was prepared by a hydrothermal synthesis of Ni(NO
3)
2 6H
2O, NH
4F, and urea at 120 °C for 8 h. Then, a (Ni(OH)
2)@nitrogen-doped carbon nanosheet was created by using imidazole and Ni(OH)
2 in a tubular furnace at 280 °C for 2 h under an Ar atmosphere. Finally, the obtained sample was treated with NaH
2PO
2 in a furnace at 350 °C for 2 h.
Meanwhile, Gao et al. reported a MoP/NPC structure synthesized by mixing melamine phosphate, molybdenum chloride, and glucose [
66]. Then, the obtained gel was annealed at 900 °C in an Ar atmosphere. The LSV measurements in 1 mol L
−1 KOH show that MoP/NPC delivers a current of 10 mA cm
−2 with an overpotential of 163 mV. Similarly, Mukkavilli et al. fabricated Ta
3N
5-(O) nanofibers using electrospinning, and this 1D structure was implemented as electrocatalysis in a polymer electrolyte membrane electrolyzer [
67]. This material was developed by mixing tantalum ethoxide, polyvinyl pyrrolidine, ethanol, and acetic acid. Then, the sol was transferred to a syringe, and 16 kV was applied to form the nanofibers. The nanofibers were used to construct the electrolyzer electrode, and the test showed that the HER was achieved at an overpotential of 320 mV while maintaining a 10 mA cm
−2 current density.
As is well known, seawater is abundant worldwide and might be a natural resource for hydrogen production in alkaline media. Nevertheless, the anodic and cathodic reactions are hindered by chloride and chlorine, and only a few electrocatalysts can work in this harsh environment. In this sense, a novel report on nitride material was conducted by Ma et el. in which iron-rich phosphine/nickel nitride was proposed as an electrocatalyst for seawater splitting [
68]. This material only requires 113 and 212 mV to reach 100 mA cm
−2 for HER and OER, respectively.
Several different non-noble electrocatalysts have been reported. This includes Ni-based and Cu-based composites synthesized under controlled parameters to obtain materials with outstanding morphologies.
Table 1 lists some interesting electrocatalysts for the HER and their main electrochemical parameters. This includes tests conducted in acid, alkaline media, and benchmark electrocatalysts.
5. Oxygen Evolution Reaction on Noble-Metal-Free Electrocatalysts
A critical limitation in electrochemical water splitting is the large overpotential and low selectivity for the OER [
83]. To this end, non-noble metals have been studied as electrocatalysts in alkaline media. The OER might occur in two ways: a direct conversion or the formation of MOOH intermediates. It is commonly observed that the OER in alkaline media is achieved by the appearance of metal oxyhydroxides generating O
2 and H
2O molecules in consecutive steps [
84]. For instance, Cao et al. reported an ultrathin nickel hydroxide nanosheet with catalytic activity for the OER [
85]. The electrode was fabricated by the electrodeposition of α-Ni(OH)
2 nanosheets on nickel foam support. This research group studied the phase transformation of this compound by conducting X-ray diffraction measurements after OER tests. Interestingly, the diffractograms reveal that the compound suffers several phase transformations. The presence of Ni(OH)
2 and NiOOH phases suggests that the α-Ni(OH)
2 was converted to β-Ni(OH)
2 and γ-NiOOH during the OER.
The interconversion of metal oxyhydroxide species in alkaline media during oxygen evolution has already been studied in [
86] and is presented in
Figure 5. In the case of OERs, the reaction can be achieved via two pathways: peroxide generation and direct reaction.
Kang et al. prepared W@Ni(OH)
2 on conductive carbon cloth in order to study the interfacial electronic structure and its effect on the formation of oxyhydroxide species at the surface of the electrode [
87]. They used the magnetron sputtering technique to successfully deposit tungsten nanoparticles on a Ni(OH)
2 substrate. The electrochemical measurements reveal that this catalyst delivers 10 mA cm
−2 at an overpotential of 290 mV. The research group of Feng developed another interesting work in which FeOOH/NiOOH was grown on a nickel foam substrate by the impregnation method [
88]. They found that even after applying 435 mA cm
−2 for 30 h the catalysts do not have lattice distortion or volume expansion. Moreover, the material shows improved performance for water adsorption due to the high Lewis acidity with an overpotential of 163 mV.
Furthermore, NiFeMo alloys have been studied as prospective electrocatalysts for overall water splitting [
73]. Studies have shown that the electroactivity of this material is due to an induced tunable lattice strain of nickel that results in a d-band center change, improving the electronic interactions of the active sites. It is well known that multimetallic alloys can facilitate the adsorption and desorption of reactants during the OER. Cai et al. proposed the design of high-entropy alloys regarding the unusual physicochemical and mechanical properties of this type of compound [
89]. The material was readily prepared by mixing the Fe, Cu, Ni, Ir, and Mo precursors in a solution containing glucose and oleylamine. The sample was heated at 220 °C under a nitrogen atmosphere and was recovered by centrifugation using hexane and ethanol.
Two-dimensional (2D) materials such as graphene, reduced graphene oxide (rGO), layered double hydroxides, transition metal oxides, and others have gained interest due to their physical, chemical, and optical properties for energy systems [
90,
91]. These materials have a large specific area that exposes the active sites, and for this reason, they have been proposed for constructing electrochemical devices. For example, Yu et al. prepared MoS
2-Ni
3S
2 on graphene as a growth template by one hydrothermal method [
92]. An interesting point of this research is that the designed heterostructure has the ability to achieve both the HER and the OER simultaneously. This is very important considering that most of the fabricated electrocatalysts are incompatible with performing both reactions, thus limiting the practical application of electrolytic water [
93]. Another group of researchers synthesized 2D VSe
2/rGO composite using a facile one-pot hydrothermal process [
94]. This dichalcogenide material was used to construct electrodes for the OER in 0.5 mol L
−1 KOH. From the experimental results, this group concludes that the high electroactivity of this composite is mainly due to the presence of rGO, which increases the electron transfer kinetic and synergistic effect.
An impressive and prospective electrocatalyst was developed by Ji et al. who used vertical graphene nanosheets and carbon fibers to obtain a multilayer stacked MoS
2/FeCoNi(OH)
x. They used chemical vapor deposition and subsequent electrodeposition [
95]. The high performance of this composite was attributed to the easy access to electrolytes, elevated area, and optimization of the electronic structure.
Another pursued two-dimensional material is transition metal carbide (MXene), which has good conductivity and stability. This extraordinary composite has been proposed as the material of choice in the field of energy conversion [
96,
97]. Using consecutive synthesis, Zeng and coworkers reported a honeycomb MXene/NiFePx-NC heterostructure [
98]. Firstly, the Ti
3AlC
2 MAX phase was treated with HF to obtain a multi-layered Ti
3C2T
x MXene. Secondly, the obtained material was mixed with a NiCl
2 6H
2O water solution and sodium citrate. Then, K
3[Fe(CN)]
6 was added to the slurry and kept at room temperature for 10 h. The resulting sample was recovered by centrifugation, washed several times, and dried at 60 °C for 12 h under a vacuum.
Metal–organic frameworks (MOF) are compounds consisting of metal species linked to organic molecules via coordination bonds. They are easily tunable to improve high porosity and chemical reactivity [
99,
100]. Previous studies show that the catalytic performance in an MOF can be enhanced by adding different elements [
101]. For example, Yan et al. fabricated two new Ni/Co-MOF as electrocatalysts for the OER, showing promising activity by delivering a current density of 10 mA cm
−2 at an overpotential close to 350 mV [
102]. They stated that the mass activity of these novel materials surpasses the activity of RuO
2 by more than 3 times. Nevertheless, the stability of this compound shows short-term electrochemical durability.
Figure 6 presents the electrochemical performance of Co-MOF and Ni-MOF for the OER in alkaline media. It should be highlighted that Ni-MOF electrocatalysts deliver high current densities at a lower potential in comparison with other metals (
Figure 6a). In addition, the EIS measurements indicate that the OER is mainly favored due to an increase in electron transfer rate,
Figure 6b).
Similarly, Chen et al. found that the hybrid nanostructure of a zeolitic imidazole framework, ZIF-67/MIL-88 (Fe and Ni), has an overpotential of 269 mV at a current density of 10 mA cm
−2 (close to 1.51 V vs. RHE) [
54]. To assess the potential of this compound, this research group fabricated a complete cell to study the overall water splitting process (OWS).
Figure 7 displays the performance of this synthesized material for the OWS [
54]. At first glance, it can be observed (
Figure 7a) that the proposed noble-metal-free electrode has almost the same onset potential for the OER in comparison with the Pt/C electrode. Nevertheless, at high current densities, the Co-M-Fe/Ni(150) experiences detrimental effects that result in high overpotentials. On the other hand,
Figure 7b shows the chrono-potentiometric response of this material when 1.524 V vs. RHE is applied. From that, it can be concluded that the composite has good stability to achieve the OWS.
NiFe-LDH self-supporting electrocatalysts were synthesized by hydrothermal reaction at 120 °C for 12 h [
53]. The increased activity of this composite was due to the hierarchical structure and high conductivity of iron foam.
Figure 8 illustrates the consecutive hydrothermal reactions. First, terephthalic acid molecules were coordinated with Fe species released from the porous iron foam. Then, the obtained sample was further treated by adding a NiCl
2 6H
2O solution. The presence of ethanol directs the nickel precursor hydrolysis, and H
+ species are formed that etch the Fe-MOF. The released Fe ions interact with the Ni ones and form NiFe-LDH on the surface of Fe-MOF.
Another relevant work was developed by Shen et al. in which controlled tridimensional ZIF-67/NiCo-S/NF was prepared by a feasible strategy consisting of a simple hydrothermal process, stable stirring, and high-temperature treatment [
74]. As the other reported electroactive MOF, this material shows an excellent synergetic effect, high surface area, good electron transfer, and excellent stability. The previous works show that MOF is an appropriate way to prepare 3D structures with the potential for water splitting.
Despite the improvement in the design of novel and promising electrocatalysts for water splitting, the OER is still the bottleneck for large-scale water electrolysis. An approach to face the low efficiencies is the oxidation of organic molecules such as methanol, ethanol, urea, etc., which are more thermodynamically favorable [
76]. In this sense, ethanol has been considered the fuel of choice for H
2 and O
2 generation because it is easy to oxidize and can be produced from the fermentation of biomass [
103]. Sheng et al. designed a Ni-Fe-P/NF electrode for a hybrid ethanol–water electrolyzer that outperforms the sole water splitting. Furthermore, this research group found that this system can produce acetic acid as a highly valuable chemical byproduct.
Wang et al. prepared a multifunctional material composed of copper–nickel sulfide anchored on graphene with large vacancies [
104]. This group found that by implementing this type of material for water splitting and alcohol oxidation, it is possible to obtain H
2 and value-added organic products with a high-efficiency rate. The catalyst was prepared using a two-step process hydrothermal method. First, GO was mixed with the Ni and Cu chemical precursors in a urea solution. Then, the sample was hydrothermally treated at 160 °C for 6 h. Finally, the recovered sample was calcinated at 350 °C for 2 h. In a separate work, Lopez-Fernández et al. studied hybrid water–ethanol electrolysis on Ni-based electrocatalysts [
105]. They used the magnetron sputtering in an oblique angle deposition to prepare active Ni on a carbon paper substrate.
Wu and coworkers successfully modified SrTi
0.1Fe
0.85Ni
0.05O
3−δ perovskite with carbon nanotubes using the facile chemical vapor deposition method [
106]. This group introduced carbon nanotubes with the idea of enhancing the electrical conductivity and specific surface area of this perovskite-oxidized compound. Surprisingly, this modified compound displayed better performance for the overall water splitting in comparison with the untreated one, and its performance was similar to that reported for cells fabricated with IrO
2 and Pt/C in alkaline electrolyte.
But not only ethanol can be used as sacrificial fuel in water splitting. Huang et al. proposed the implementation of cobalt/iron heterostructures as electrocatalysts for OERs and benzyl alcohol oxidation [
107]. They observed that defective oxygen is responsible for the facile adsorption/dissociation of intermediates during electrolysis. In this context, Woo et al. reported the collaborative electrochemical oxidation of alcohols and aldehyde groups [
108]. They claimed that this strategy is promising to produce high-value products considering that alcohol oxidations are thermodynamically more favorable.
Table 2 presents a summary of different electrocatalysts for the OER.
Electrospinning is a feasible technique for designing and develo** outstanding materials in energy conversion technology. A detailed study was conducted by Zhang et al. in which electrospinning perovskite SmBa
0.5Sr
0.5Co
2O
6−δ was proposed to control the physical and morphological characteristics of this material, providing new insight into the design of advanced materials [
125]. The XRD patterns demonstrate the formation of a single perovskite phase with a cubic structure after heat treatment at 800 °C. This material was used as an electrocatalyst for the OER in alkaline media, which could deliver 10 mA cm
−2 at 0.37 V overpotential and a Tafel slope close to 46 mV dec
−1. The authors state that this cubic perovskite material is improved due to its high electrical conductivity and electrochemically effective surface area (estimated from the double later region).
Since electrospinning is a cost-effective method, several nanostructures have been developed by adopting this approach. For example, hybrid crystalline–amorphous La
0.33SrCo
0.5Fe
0.5O
x was developed by electrospinning and used as electrocatalysts for OER in water splitting [
126]. Electrochemical measurements in 0.1 mol L
−1 KOH media confirm that hybrid perovskite can achieve an OER at a 250 mV overpotential at a current density of 10 mA cm
−2, surpassing the performance of the pure phase.
6. Photocatalysts for Water Splitting
Historically, humankind has been interested in using natural energy sources like wind, solar, tidal, and others to produce electrical energy. Nonetheless, the major issue of these renewable energies is their intermittency [
127,
128]. In this regard, hydrogen is considered a viable potential energy conversion vector because it can be produced from the aforementioned renewable sources and stored when required [
129,
130]. Photoelectrochemical water splitting has been gaining interest because it can produce clean energy, and most of the designed photocatalysts have low-cost chemical elements and are environmental-friendly [
131,
132,
133,
134].
The shortcomings of the current photocatalysts are poor stability, low sunlight adsorption, and rapid recombination [
135,
136]. Thus, the design of robust photocatalysts that permit water splitting with high efficiency is essential, and this boosts the practical applications [
137,
138,
139]. Some proposed strategies for improving efficiency are using multiple elements, designing heterostructures, and tuning the material properties for defects using a controlled morphology. The photocatalytic reaction is achieved by the adsorption of sunlight, which changes the semiconductor’s electronic state, producing electrons and holes [
140,
141,
142]. If the band gap is larger than the potential required for the water splitting, 1.23 V, and the semiconductor satisfies the thermodynamic requirements, the generated electrons can migrate and assist the reaction. In the case of water splitting, the generated electrons reduce hydrogen ions, and holes interact with oxygen anions, according to the reaction
[
143,
144,
145]. The operation and schematic of a photocatalytic cell is illustrated in
Figure 9.
Monocompounds have been widely used since the discovery of H
2 generation (1972) via water splitting using TiO
2 as a photoanode [
146,
147]. However, the efficiency of these materials is limited and cannot be used for large-scale applications. In this sense, hetero-compounds have been proposed as the catalysts of choice because they have a wide range band gap and slow electron recombination [
148,
149].
For instance, Tai et al. developed for the first time a type-II heterostructure composed of PdS@MIL-125-NH
2@ZnS [
151]. This material was fabricated in three stages. First, NH2-MIL-125 (Ti) was synthesized by mixing titanium tetra butoxide, NH2-BDC, and dimethylformamide in the solvothermal process. Secondly, Pd@M125 was prepared using the double solvent method and hexane. Lastly, PdS@M125@ZnS was also obtained by the double solvent method, in which zinc acetate dehydrates and sodium sulfide nonahydrate were used as Zn and S precursors. The characterization results reveal that this heterostructure significantly improves the evolution of H
2 attributed to abundant active sites and high open porosity that help the exchange of reactants.
Carbon-nitride-based photocatalysts are another group of materials that have been applied in the design of advanced active photoanodes [
152,
153,
154,
155,
156]. This compound has promising properties such as acceptable stability and a unique 2D structure because its electronic structure can be easily tuned [
157,
158]. In this direction, the research group of Ajmal designed a novel photocatalyst in which an Fe
2TiO
5 donor-π-acceptor was incorporated into the carbon nitride with a copolymer molecule to endorse the solar water splitting process. The fabricated compound decreases the band gap from 3.25 to 2.55 eV, displaying promising photocatalytic activity.
Here, it should be mentioned that some researchers proposed using electron donors like methanol, ethanol, ethylene glycol, and others. Still, there is controversy because the generated electrons come from the oxidation of the organic molecule and not from water [
159,
160]. Since organic molecule reformation can play a significant role in green H
2 generation, we consider some works related to this process.
Meanwhile, Dai et al. developed a phenyl-grafted nitride compound with a broad spectrum for light adsorption compared to melamine nitride material [
161]. The photoelectrochemical measurements in triethanolamine aqueous solution show that this composite can produce 2846 µmol h
−1g
−1 of hydrogen, which was five times that of melamine material under visible light. The author stated that this outstanding response is due to the inhibition of the recombination of electron–hole pairs due to the presence of phenyl species.
On the other side, using perovskite-type materials have been explored as a sustainable tactic to produce green hydrogen via water splitting. This compound is characterized by its simplicity, low cost, and non-toxicity and because its chemical and physical properties can be easily tuned [
162,
163,
164,
165]. In this direction, Salem et al. studied hydrogen-doped SrSnO
3 perovskite as having promising potential as photocatalysts for water splitting [
166]. The theoretical calculations reveal that the formation of interstitial hydrogen defects into SrSnO
3 perovskite structure favors substitutional hydrogen defects. Furthermore, a high concentration of oxygen vacancies in the perovskite matrix results in defect states that ultimately change the band gap below 1.3 eV of the conduction band minimum.
Tam et al. synthesized a 2D porous SrNbO
2N doped with Zr as photocatalysts for water splitting [
167]. The electrochemical results show that the derived material can produce 2 mA cm
−2 of current density at 1.23 V vs RHE for water splitting under 1.5 G simulated sunlight. Due to its suitable band gap and available chemical elements, the ternary chalcogenide copper indium sulfide (CuInS2) semiconductor has also been used to pursue water splitting [
168]. Yu et al. recently designed CuInS
2 supported on porous SiO
2 nanotubes using electrospinning methodology [
169]. The experimental measurements were achieved in a 100 mL quartz reactor with 0.35 mol L
−1 Na
2S and 0.25 mol L
−1 Na
2SO
3 as a sacrificial agent. The results indicated that this structure reached a hydrogen production rate of 367 µmol g
−1h
−1, which is almost 3.1 times that of the unsupported one.
Chen et al. developed a 2D/2D Cd
0.5Zn
0.5S/CuInS
2 nanostructure with potential applications for hydrogen generation under visible light irradiation [
170]. This research group used 0.35 to 0.25 mol L
−1 Na
2S/Na
2SO
3 as a sacrificial reagent mixed with 25 mg of the photocatalysts into a photoreactor illuminated with simulated sunlight from a 300 W Xe lamp. The test shows that the proposed structure has 0.79 mmol g
−1h
−1. MoS
2/rGO was used as a co-catalyst to inhibit the photocorrosion of a Cu
2O compound and, at the same time, reduce the recombination rate of the generated charge carriers [
171]. The aforementioned material was synthesized by the hydrothermal process conducted in three stages: the synthesis of Cu
2O, rGO-MoS
2, and finally, Cu
2O-MoS
2/rGO. The photoelectrochemical tests were conducted in a standard three photoelectrochemical cell setup in a 0.5 mol L
−1 Na
2SO
4 electrolyte solution. The working electrode was constructed by using the obtained compound and transferred to an indium tin oxide conductive glass with an active area of 0.25 cm
2, while the counter electrode was a Pt wire and Ag/AgCl as the reference electrode. The modified material delivers a higher current density: 8.46 mA cm
−2 at 0.95 V.
7. Mechanism and Activity Descriptors for HER and OER
Searching for a feasible electrocatalyst for water separation has been the holy grail in material science and electrochemistry. But to face this challenge, it is essential to fully understand each phenomenon that occurs during water splitting so that it can be controlled and assisted. In this regard, many attempts have been made to understand the reaction mechanism during the OERs and HERs at different electrode surfaces. Guan and coworkers proposed a design strategy for which the ultimate goals were to create a material with multiple active sites capable of shortening the reaction pathway [
172]. DFT was used to study H
2O adsorption on the surface, H
2O dissociation, OH
− desorption, and the concomitant of adsorbed H
*. The authors state that material oxygen vacancies promote the adsorption of H
2O, while transition metal cations serve as active sites for H
* species. Thus, if a material is designed that way, an acceptable HER is expected for water splitting because there is a shortened reaction pathway for intermediates. In this regard, Fe
2O
3@NiO heterostructures were prepared by a facile one-step hydrothermal route and further thermally treated [
173]. This composite was used as an electrocatalyst for the overall water splitting, showing an overpotential of 224 mV and 187 mV for the OER and HER when driving a current density of 10 mA cm
−2. The enhanced performance of this electrocatalyst was attributed to the rich oxygen vacancies of Fe
2O
3 and the synergistic effect between both metals. Li at al. conducted interesting research focused on determining the effect of the surface and the bulk oxygen vacancies on La
0.7Sr
1.3Co
2O
6−δ by using X-ray absorption spectroscopy and magnetization measurements [
174]. They claimed that excessive bulk oxygen vacancies are not desired because they inhibit the presence of the O 2p bridge (concluded from the paramagnetic state of the samples), which mitigates the OER. Meanwhile, materials with surface oxygen vacancies promote the adsorption of electroactive species such as OH
− due to available lattice sites. Heterogeneous electrocatalyst strain and reconstruction are other structural manipulation engineering strategies usually proposed to enhance the overall water separation due to their ability to tailor the material electronic surface [
175]. In this direction, Li
2Co
2O
4 was prepared via a sol–gel process by mixing the chemical precursors in the desired molar ratio and evaporating to form the raw gel that was then thermally treated at different temperatures [
176]. The differences in electrochemical activity for the OER of each sample were studied by X-ray absorption near-edge structure (XANES). The results verify that the lattice oxygen is volatilized in thermal treatment at high temperatures, lowering the Co valence state. The latter explains why the sample treated at 1000 °C experienced a large structural reconstruction in comparison with the one treated at 550 °C. Operando Raman spectra reveal enhanced tensile strain on the electrocatalyst treated at 550 °C during and after the OER that triggered the OH
− adsorption. From a mechanistic point of view and based on density functional theory (DFT), OH
− adsorption improves the OER because it stimulates the adsorbate evolution mechanism and the lattice oxygen mechanism. In the same direction, Wang et al. developed a tantalum-modified NiFe layered double hydroxide via hydrothermal synthesis [
177]. They observe a lattice expansion of the layered double hydroxide in conjunction with electronic structure modifications resulting from a strain effect. The incorporated deformation creates active sites that decrease the overpotential at 260 mV for the OER when driving a current density of 50 mA cm
−2.
To deal with the low efficiency of current electrocatalysts, it is mandatory to visualize and understand the main activity descriptors in the water splitting process that initiate their activity. In this direction, Kumar et al. studied all the existing activity descriptors in metal phthalocyanines as electrocatalysts for the OER and HER [
178]. Because oxygen reactions include O* and OH* reaction intermediates, these authors proposed the implementation of Volcano plots, based on DFT, that were constructed considering the intermediate free energy. In fact, Gibbs free energy change has been proposed as a robust descriptor to predict the OER and HER. This is based on the Sabatier statement considering that catalysts should have a variation in Gibbs free energy close to zero.
Guan et al. designed ABO
3−δtype perovskite electrocatalysts for HERs based on coordination rationales as a unifying descriptor [
179]. The introduced A-site ionic electronegativity in perovskite structures provides a better volcano trend for the HER. The modified material shows excellent activity for the HER with a very small Tafel slope, 27.6 mV dec
−1, making this material the state of the art. Because some mechanisms related to activity descriptors are not easy to understand, the scientific community has proposed implementing in situ measurements [
180]. For example, the mechanism of the metal cation size trend into the OER on Ni-Fe oxyhydroxide electrodes was studied by conducting in situ X-ray absorption spectroscopy [
181]. Those measurements reveal that the Ni
2+/(3+δ)+ redox peak is influenced by the cation size. Big cations introduce a displacement at lower potentials, decreasing the OER onset potential. Furthermore, it was concluded that the phenomenon of the cation size effect was an indirect pH effect. In other words, the electrolyte pH dictated the OER activity as the parameter that modifies the Lewis acidity/basicity of the Ni-Fe oxyhydroxide electrode interconversion. Another interesting operando spectroscopy study was applied to Ru-Ir nanomaterials with coral-like structures during water splitting in acid media [
182]. This electrocatalyst is promising because it only requires 165 mV to reach a current density of 10 mA cm
−2 for the OER. X-ray operando absorption near-edge spectroscopy and atomic-resolution electron microscopy were used to elucidate the outstanding performance of these electrocatalysts. The results show that changing the applied potential from the open circuit potential to 1.25 V shifts the adsorption edge from metal Ru to oxidized RuO
2. The analysis concluded that the hexagonal (0001) facet protects the Ru from corrosion and is responsible for the enhanced OER in the coral-like structure. Similar results were observed in NiMoO
4 when the electrode was summited to an applied potential bias [
183]. In this case, the catalysts preferentially form γ-NiOOH structures, known as active sites, to achieve the OER. In another work, high-valence iron was used to modify NiMoO
4 heterostructures using oxygen plasma, and in situ Raman measurements were conducted to analyze the irreversible reconstruction [
184]. Phase changes during the OER were analyzed in the potential windows of 1.18 to 1.43 V, in which NiOOH was the main remaining phase during water electrolysis at a potential close to 1.43 V. Operando UV-vis spectroscopy coupled with electrochemical measurements is another widely used technique to determine electrocatalyst reaction mechanisms. A series of modified nickel heterostructures has been studied in detail by this in situ technique, demonstrating a shift in the Ni
2+/Ni
3+ redox peak in the order of Mn < Co < Fe < Zn, which is attributed to a decrease in oxygen binding energy [
185].