Sulfides are frequently used as photocatalysts, since they absorb visible light better than many oxides. They have the disadvantage of being more easily photocorroded. This occurs mostly in oxidizing conditions; therefore, they are commonly used instead in reduction processes, such as CO2 reduction to fuels or H2 production.
1. Introduction
Photocatalysts are used for many purposes: energy-related applications, fine chemicals synthesis, environment protection, or detection of specific chemicals. Photocatalysis has been known for a long time. The first work on heterogeneous photocatalysis (to the researcher’s knowledge) was reported by Moore and Webster in 1913
[1]. The photoreduction to formaldehyde of CO
2 was described there, using iron or uranium oxide colloids and utilizing visible light. Since there is currently an urgent need to revert the increase of CO
2 in the atmosphere, this was certainly an important work. This work appeared just after the work by G. Ciamician
[2], which said in its last sentences, “So far, human civilization has made use almost exclusively of fossil solar energy. Would it not be advantageous to make better use of radiant energy?” Almost six decades later, an article was published by Fujishima and Honda
[3] who proposed using photoelectrochemistry, a practical way to photodissociate H
2O into O
2 and H
2.
Those authors used a single crystal of rutile TiO
2 in their work. Given that its bandgap is 3.0 eV, it absorbs light in the near-UV range, thus it is unsuitable for converting much of the solar spectrum. Other oxides like SrTiO
3, ZnO, or anatase TiO
2, having bandgaps above 3.0 eV, have a similar limitation; nonetheless, for some fine chemical syntheses and especially for environment protection, anatase-type TiO
2 remains unsurpassed as photocatalyst. Other materials have been tried in order to enlarge the amount of solar spectrum that can be used. Thus, anatase has been doped with cations or anions, and completely different oxides have been developed like BiVO
4 (with E
g = 2.4 eV); this material, which is mentioned frequently as able to photogenerate O
2, as well as the oxides mentioned before with bandgaps higher than 3.0 eV, are certainly resistant to photocorrosion. Additionally, (oxy)sulfides
[4,5,6[4][5][6][7],
7], (oxy)nitrides
[8,9[8][9][10][11],
10,11], selenides
[12,13[12][13][14][15],
14,15], several varieties of doped carbons
[16[16][17][18][19],
17,18,19], and a few more exotic materials were proposed in order to use a larger range of the solar spectrum. On the other hand, for energy applications, a co-catalyst is frequently required to facilitate the O
2 and/or H
2 evolution, the CO
2 reduction or the conversion of substances derived from biomass.
Many such materials can be also used in photoelectrochemical (PEC) systems. For example, Fe
2O
3 in the hematite phase (see structure in
[20]) has a small mobility of the photogenerated current carriers so that high recombination rates occur unless a very small thickness is used; it is, however, very actively studied for PEC uses thanks to its convenient bandgap (≈1.9 eV) and especially its abundance. Additionally, this material is resistant to photocorrosion. Examples of its use in photocatalysis can be found
[21,22,23][21][22][23].
Many reviews have appeared dealing with photocatalysis using sulfides for protection of the environment
[24] and transformation of organic molecules
[25], as well as others dealing with more general photocatalysts (including sulfides), devoted to H
2 generation
[26,27][26][27] and CO
2 reduction
[28]; sulfides containing several cations have been also studied for photocatalysis or energy harvesting purposes
[29,30][29][30]. Mixing sulfides with other phases has been utilized as well to better separate the photogenerated holes and electrons (leading, when both semiconductors absorb light, to the so-called Z-scheme)
[31].
It is well known that sulfides, particularly in oxidizing conditions, can be photocorroded; efforts are thus made in order to avoid or at least minimize this process
[32]. Therefore, sulfide photocatalysts are mainly used for photoreduction processes, as is the case of H
2 generation or CO
2 reduction. This might require using a sacrificial agent, e.g., sulphite or sulfide anions, which makes the process useful only for basic studies, except if the sacrificial agent is derived from biomass.
2. 2-Fold Coordinated Sulfides
One 2-fold coordinated sulfide that has been studied for photocatalysis is HgS (cinnabar). Its structure is given in
[33]; its bandgap is 2.05 eV, as shown in
[34]. Some other results in photocatalysis by HgS, combined with other phases, are found in
[35,36][35][36].
3. Tri- and Tetrahedrally Coordinated Sulfides
There are some sulfides which contain cations in trigonal planar coordination. For example, Bi
2S
3, with both 4- and 3-fold coordinated Bi (due to a lone pair present in Bi; its structure is reported in
[37]). Its bandgap is 1.3–1.7 eV and has been used for photocatalysis either alone
[38,39][38][39] or in combination with other phases
[40,41,42][40][41][42]. An additional example is CuS with covellite structure (structure reported in
[43]), which also has Cu in trigonal and tetrahedral coordination. Its bandgap is 1.75 eV and has also been used alone
[44,45][44][45] or combined with other materials, for photocatalysis applications
[46,47,48][46][47][48]. It must be noted that, in spite of its formula, it contains mainly Cu
+ [49], which means that disulfide ions exist in it.
Additionally, Ag
2S (structure reported in
[50]) and Cu
2S (structure reported in
[51]) contain, in their most stable phases, trigonally coordinated cations and have been used as well as photocatalysts. For example, Cu
2S, which has a bandgap (indirect) of 1.2 eV
[52], was utilized in degradation of herbicides combined with Bi
2WO
6 [53], in degradation of dyes in combination with TiO
2 of P25 type
[54] or with H
2O
2 [55]; combined with MoO
3, it has been used for H
2 generation, degradation of dyes, or reduction of Cr (VI)
[56]. Ag
2S, in turn, has an indirect bandgap of 1.0 eV
[57]; it has been utilized as a photocatalyst in a number of composites for generation of H
2 [58], degradation of dyes, or reduction of CO
2 [59], as well as in disinfection
[60]. Several Cu
2−xS structures were used as well as photocatalysts
[61].
A significant amount of sulfides contain only tetrahedrally coordinated cations. This is the case, for example, of ZnS (structure reported in
[62]). It has a rather large bandgap (3.4 eV
[63]), so that it can absorb only light in the UV range. It has been used, however, in photocatalysis for very different applications
[64,65,66][64][65][66]. Note that this material can adopt different shapes, influencing its photocatalytic and photophysical properties
[67].
The sulfide most studied for photocatalysis is probably CdS (structure reported in
[68]), also tetrahedrally coordinated. Due to the high mobility of its photo-excited electrons and holes, and its significant ability to absorb light in the visible range (λ < 500 nm; its bandgap is 2.48 eV
[63]), several reviews have studied its capabilities for generation of H
2 [69], organic chemistry transformations
[25] or degradation of dyes
[24]. One problem is the toxicity of Cd; this sulfide is also prone to photocorrosion, particularly in oxidizing conditions
[70]. Some attempts have been made to decrease this effect
[71,72][71][72]. On the other hand, CdS easily undergoes (as in the case of CdSe) quantum confinement effects. Therefore, its particle size can be tailored by irradiating it with monochromatic light in oxidizing conditions; any photocorrosion will finish when the size of its particles is so small that the single-wavelength light can no longer be absorbed
[73].
Another class of tetrahedrally coordinated sulfide photocatalysts that are studied include the chalcopyrite family. Thus, AgGaS
2, CuGaS
2, and CuInS
2, all of them with chalcopyrite structure (see structure for AgGaS
2 in
[74]), have gaps of 2.6, 2.3, and 1.5 eV, respectively, and even their alloys have been used as photocatalysts. CuGaS
2 [75[75][76],
76], (Ag,Cu)GaS
2 [77] or (Ag,Cu)(In,Ga)S
2 [78] as well as doped AgGaS
2 [79] are active for photogeneration of H
2; mixing CuGaS
2 with RGO-TiO
2 has photocatalytic activity in reducing CO
2 to CO
[80]. Some systems of this kind can be used for the photocatalytic elimination of dyes
[81], nitrate ions
[82] or NO
[83]; several years ago, a review dealt with the photocatalytic uses of CuInS
2 [84]. Additionally, kesterites, which have structures similar to chalcopyrites (see
[85]) and have a bandgap of 1.5 eV (like CuInS
2), were used as well as photocatalysts
[86].
4. Sulfides including 6-Fold Coordinated Cations
Materials that have been studied extensively as photocatalysts are WS
2 and (especially) MoS
2; there are several recent reviews on them
[87,88,89][87][88][89]. They have layered structures with cations in prismatic coordination, held together by van der Waals forces (see their 2H structures in
[90[90][91],
91], respectively). They are also polymorphs
[92,93][92][93]) and have indirect bandgaps of 1.35 eV and 1.23 eV, respectively
[94]. These bandgaps can be increased by decreasing their particle sizes; in fact, isolated trilayers of MoS
2 and WS
2 have, according to photoluminescence data, direct bandgaps of respectively 1.89 and 2.03 eV
[95]. This might position their conduction bands to levels more negative than the H
2|H
+ electrode potential
[96], so that H
2 photogeneration might be facilitated. MoS
2. with small-to-moderate particle size. is much more efficient photocatalytically if its particle size falls below 4–5 nm
[97,98][97][98]; this is certainly a quantum confinement effect. It must be noted, on the other hand, that there is another structure of MoS
2, termed 1T, which has octahedral, not prismatic, coordination (see structure in
[99]). It has metallic characteristics, so that it is very active in combining protons to achieve H
2 evolution
[100].
Other 6-fold coordinated sulfides have been studied for photocatalysis. This is the case of ZrS
2 (see structure in
[101]), which is also a layered structure held together by dispersion forces, but it has octahedral, not prismatic, coordination, at difference with 2H MoS
2 or WS
2 (i.e., it is similar to 1T MoS
2). Its bandgap is ca. 1.7 eV
[102]; however, as shown in
[103] if it is made in 2–3-layer shape, it may attain a 2.0 eV bandgap making it ideal for photo-generation of H
2 [103]. HfS
2 has a similar structure
[101] but has a smaller bandgap
[104]. Its isolation from ZrS
2 is difficult, however, and it has therefore rarely been used in photocatalysis
[105].
Additionally, FeS
2 with pyrite structure (including, in this case, only disulfide ions
[106]) also has an octahedral coordination to S atoms; its bandgap is 0.95 eV and has been used sometimes for photocatalysis
[107].
Finally, there is another octahedrally coordinated sulfide: PbS (its structure can be found in
[108]). However, its bandgap is rather small (ca. 0.5 eV or less, as shown in
[109]). Still, it has been considered for photocatalysis in combination with other phases
[110,111][110][111].