Facet‐Control versus Co‐Catalyst‐Control in Photocatalytic H₂ Evolution from Anatase TiO₂ Nanocrystals

Abstract

Titanium dioxide (TiO₂) and, in particular, its anatase polymorph, is widely studied for photocatalytic H₂ production. In the present work, we examine the importance of reactive facets of anatase crystallites on the photocatalytic H₂ evolution from aqueous methanol solutions. For this, we synthesized anatase TiO₂ nanocrystals with a large amount of either {001} facets, that is, nanosheets, or {101} facets, that is, octahedral nanocubes, and examined their photocatalytic H₂ evolution and then repeated this procedure with samples where Pt co‐catalyst is present on all facets. Octahedral nanocubes with abundant {101} facets produce >4 times more H₂ than nanosheets enriched in {001} facets if the reaction is carried out under co‐catalyst‐free conditions. For samples that carry Pt co‐catalyst on both {001} and {101} facets, faceting loses entirely its significance. This demonstrates that the beneficial role of faceting, namely the introduction of {101} facets that act as electron transfer mediator is relevant only for co‐catalyst‐free TiO₂ surfaces.

The role of reactive facets of anatase TiO₂ crystallites on photocatalytic H₂ evolution is examined in this work. Pt co‐catalyst‐loaded and Pt‐free preferentially faceted nanocrystals are compared. As expected, both types of Pt‐decorated TiO₂ nanocrystals demonstrate significantly enhanced photocatalysis yet comparable H₂ evolution indicating that the beneficial effect of faceting is lost for the Pt‐decorated nanocrystals.

Introduction

With the depletion of fossil fuel and growing demand for green sustainable energy sources, photocatalytic H₂ evolution attracts increasing attention in recent decades. ^[1] Since the pioneering work of Fujishima and Honda on photocatalytic water splitting, ^[2] photocatalytic H₂ evolution using semiconductor materials has become a most promising clean energy technology with minimal impact on the environment. ^[3] Photocatalytic H₂ generation is based on the interaction of light with a suitable semiconductor and a reaction of the generated charge carriers with an aqueous environment (with or without sacrificial agents) to form the desired H₂. Essential reaction steps involve (i) absorption of photons and generation of electron (e ⁻) – hole (h ⁺) pairs, (ii) charge carrier separation, (iii) migration of e ⁻ or h ⁺ from the bulk to the surface, and, finally, (iv) charge carrier reaction (reduction of H⁺ to H₂ and a complementary photo‐oxidation reaction) at the surface of a photocatalyst. The overall performance of a photocatalyst is strongly determined by the kinetics of these four steps.

The semiconducting material absorbs photons with energies higher than its bandgap. ^[4] Numerous photocatalytic materials have been investigated but among available developed materials, titanium dioxide (TiO₂) is still considered one of the most promising photocatalysts due to its strong oxidizing and reducing capabilities, long‐term stability, non‐toxic nature, abundance, and low cost. ^[5] TiO₂ fulfills the thermodynamic conditions for full water splitting, that is: the conduction band (CB) edge is lower than the H⁺/H₂ reduction potential (0 V vs. NHE), while the valence band (VB) edge is higher than the H₂O/O₂ oxidation potential (+1.23 V vs. NHE). ^[6] Nevertheless, the overall performance of the photocatalyst is often determined by kinetic factors ^[7] and here namely the charge transfer reactions that depend on a variety of material parameters.

TiO₂ nanostructures exist mainly in three crystallographic polymorphs, i. e., anatase, rutile, and brookite. ^[8] Due to the longest lifetime of photocarriers, anatase is commonly the “best‐performing” semiconducting form of TiO₂, which makes it the most popular polymorph in the fields of catalysis, photocatalysis, and dye‐sensitized solar cells. It was shown that the photocatalytic water splitting performance of TiO₂ crystals varies with the surface physicochemical properties such as size, shape, crystal structure, and amount of exposed facets. ^[7b] Recently, great progress has been achieved in crystal facet engineering of anatase TiO₂ nanocrystals. ^[9] For instance, the reduction and oxidation reactivity of different anatase crystal facets were investigated by Majima et al. through single‐particle chemiluminescence imaging. ^[10] It was found that the H₂ evolution reaction of different anatase TiO₂ crystal facets can be rated as {101}>{001}>{100}, while that of O₂ evolution reaction is {101}≈{001}≈{100}.

Accordingly, the development of TiO₂ crystals with specific facets has attracted wide interest. The anatase (101) plane has the lowest surface energy being the most stable surface. ^[11] Yang et al. synthesized anatase TiO₂ crystals with 47 % {001} and 53 % {101} facets, ^[12] which was a breakthrough in the synthesis of high‐quality anatase TiO₂ crystals with a high percentage of reactive high energy {001} facets. In the meantime, different strategies to control the synthesized facets, such as wet‐chemistry (hydrothermal, solvothermal, and nonhydrolytic), gas oxidation, topotactic transformation, crystallization from amorphous TiO₂, and epitaxial growth have been developed. ^[11] Among them, the hydrothermal method is the most common technique for tailoring the exposed crystal facets due to the possibility of manipulating the nucleation and growth processes by controlling the experimental conditions. For instance, introducing capping agents (e. g., appropriate organic molecules, inorganic ions, or their mixtures) in the hydrothermal synthesis allows decreasing the surface energies of different surfaces, which has been extensively used to control the growth rates along different orientations.[ ¹² , ¹³ ] Consequently, nanocrystals with tunable percentages of particular facets can be formed.

In the present work, we synthesize anatase TiO₂ nanocrystals with a predominant amount of either {001} or {101} facets by hydrothermal methods leading to the formation of TiO₂ nanosheets (NS) with {001} facets or TiO₂ octahedral nanocubes (Oct) with {101} facets.[ ⁹ , ¹² ] The consensus exists in the literature that faceting is always beneficial and, in particular, an advantageous photocatalytic performance of {101}‐faceted anatase TiO₂ nanocrystals was provided against all other facets.[ ¹⁰ , ¹⁴ ] Here we show, however, that this is only appropriate for the co‐catalyst‐free state of TiO₂ nanocrystals. When Pt is deposited on the entire nanocrystal surface, unfavorable {001}‐faceted TiO₂ nanosheets demonstrated 400‐fold enhancement in their photocatalytic H₂ evolution performance as compared to their plain counterparts and even 2‐fold enhancement as compared to favorable {101}‐faceted Pt‐decorated anatase TiO₂ octahedral nanocrystals. This was achieved by the decoration of the entire TiO₂ nanocrystals surface, that is, both {001} and {101} facets, using immersion deposition process rather than the selective decoration of {101} facets as was demonstrated in the literature achieved by the commonly used photodeposition approach. ^[15]

Results and Discussion

Scheme 1a demonstrates the hydrothermal approaches used in this work to synthesize anatase TiO₂ nanosheets (NS) and octahedral nanocubes (Oct) enriched by either {001} or {101} facets, as described in the experimental section and as given in Refs. 9 and 11. Figure 1a–b shows representative FE‐SEM and HR‐TEM images of the TiO₂ NS revealing a well‐defined faceted morphology. The synthesized anatase TiO₂ NS are 43.5±19.3 nm wide in the basal plane and 5.9±1.3 nm thick (Figure 1a–b). TEM images of NS demonstrate lattice fringes with 0.35 nm spacing characteristic for anatase {101} orientation (d₁₀₁ =0.352 nm, JCPDS card #21‐1272). Notably, the determined {101} crystalline planes in NS show a 68° angle to the lateral view of the basal facets, in agreement with the theoretical angle between {101} and {001} facets, ^[16] therefore, the latter can be assigned to {001} surfaces.

Scheme 1.

Schematic representation of the hydrothermal processes forming anatase TiO₂ NS (route 1) or Oct (route 2), enriched by either {001} or {101} facets, respectively.

Figure 1.

Representative FE‐SEM and HR‐TEM images of hydrothermally synthesized anatase TiO₂ (a–b) NS and (c–d) Oct. Insets in (b, d) show a fast Fourier transform of the corresponding TEM images. The inset image in (b) is the top‐view of the TiO₂ NS.

The amount of {001} facets can be evaluated according to the following equation [Eq. 1]:

$${\displaystyle P_{\left\{001\right\}}=\frac{2l^2}{2l^2+4ld}\times 100\%}$$ (1)

where P _{001} indicates the amount of {001} facets, l and d are the mean length and thickness of TiO₂ NS, respectively. ^[17] According to the calculations, the amount of {001} facets in the synthesized TiO₂ NS is 87.9 %. Brunauer‐Emmett‐Teller (BET) evaluation of NS yields a specific surface area of the obtained TiO₂ NS as 86.1 m² g⁻¹.

Figure 1c–d shows representative FE‐SEM and HR‐TEM images of the anatase TiO₂ octahedral nanocrystals (Oct). It was previously demonstrated that a low supersaturation benefits the growth of {101} facets, that is, reducing the amount of Ti(OH)₄ precursor favors the formation of octahedral TiO₂ nanocrystals. ^[11] Therefore, the hydrothermal process of the formation of anatase TiO₂ Oct was optimized along these lines (Table S1 and Figure S1, Supporting Information). Figure S1 demonstrates the influence of the amount of Ti(OH)₄ precursor and hydrothermal process temperature on the obtained TiO₂ Oct. As shown, fixing the amount of precursor to 1.0 g produces the most homogeneous TiO₂ crystals (Figure S1g). The rise in the process temperature contributes to their larger size. Furthermore, rapid quenching of the growth period is beneficial to a prominent crystallization and size distribution. ^[18] Therefore, the hydrothermal reaction of anatase TiO₂ Oct was quenched rapidly by cooling down the autoclave using cold tap water. Nevertheless, when the reaction duration reaches a critical point, a longer reaction time does not contribute significantly to the formation of larger TiO₂ Oct, which can be attributed to a fixed amount of reactant; still, a more uniform size distribution of TiO₂ Oct was obtained (Figure S1e–f). According to our experiments, the optimized amount of Ti(OH)₄ precursor (1.0 g), hydrothermal temperature (230 °C), and reaction duration of 40 h, as well as rapid reaction quenching, are crucial to obtain the most uniform octahedral TiO₂ Oct with a large percentage of {101} facets. The size distribution of anatase octahedral TiO₂ obtained under optimal conditions was estimated from FE‐SEM images and found as 12.1±4.2 nm. Considering the particle size and uniformity, the amount of {101} facets can be estimated according to Equations (2) and 3:

$${\displaystyle h=\sqrt{{\left(\frac{b-a}{2}\right)}^2+{\left(\frac{l}{2}\right)}^2}}$$ (2)

$${\displaystyle P_{\left\{101\right\}facets}=\frac{8\times \frac{(a+b)\times h}{2}}{8\times \frac{(a+b)\times h}{2}+2\times a^2}\times 100\%}$$ (3)

where a and b refer to the length of the two parallel sides of an isosceles trapezoid, l is the height of crystal, h is the height of the trapezoidal facet, and P _{101} facets stand for the percentage of {101} facets. ^[9] In this study, the a, b, l, and h parameters for the optimized sample were estimated as 2.5, 12.5, 15.0, 9.0 nm, respectively, resulting in the 97.7 % fraction of {101} facets. The corresponding BET isotherm demonstrates that the specific surface area of anatase TiO₂ Oct is 111.7 m² g⁻¹, that is, 1.3 times larger than that of NS.

The crystallinity of the as‐formed NS and Oct was analyzed by X‐ray diffraction (XRD) (Figure 2a). In both cases, the XRD patterns show sharp characteristic peaks of anatase TiO₂ at 2θ=25.3°, 37.8°, 38.6°, 48.1°, 54.0°, 55.1°, 62.7°, 70.5°, and 75.1° corresponding to (101), (001), (112), (200), (105), (211), (220), and (215) crystallographic plains, respectively (JCPDS card #21‐1272). ^[20] However, the TiO₂ NS photocatalyst shows an additional peak at 2θ=23.61° that can be attributed to TiOF₂ (JCPDS card #08‐0060) formed during the synthesis in HF, used as a capping reagent, while octahedral nanocrystals were synthesized under the F‐free conditions, that is, Ti(OH)₄ was utilized as the Ti precursor and isopropanol as capping agent and solvent. Therefore, prior to comparing the photocatalytic performance of anatase TiO₂ nanocrystals with various faceting, the TiOF₂ phase was removed.

Figure 2.

(a) XRD patterns of the as‐formed, NaOH‐washed, and annealed at 450 °C TiO₂ NS, and as‐formed and annealed TiO₂ Oct. (b) 24 h photocatalytic H₂ evolution was obtained from bare and annealed anatase TiO₂ NS, and as‐formed and annealed Oct in 50 : 50 vol.% H₂O:MeOH electrolyte. (c–d) HR‐SEM images of TiO₂ NSs (c) and Oct (d) after annealing at 450 °C for 1 h.

We used two approaches to obtain pure anatase TiO₂ NSs: (i) thermal treatment of as‐formed NSs in air, and (ii) washing of the as‐formed NSs in an aqueous NaOH solution. After the thermal treatment at 450 °C for 1 h, the TiOF₂ peak in NSs disappears (Figure 2a, orange pattern). The corresponding annealed NS photocatalyst shows an increased peak intensity at 2θ=25.3° and 37.8°, indicating an improvement in the crystallinity. However, TiO₂ NS enriched in {001} facets deform (sinter) and agglomerate after the annealing as evident from Brunauer‐Emmett‐Teller (BET) measurements. The thermal treatment process leads to a drop in their surface area from 86.1 to 30.3 m² g⁻¹ (Figure 2c, and Table 1). The agglomeration and further sintering can be explained by the loss of the F‐terminated layer present mainly on {001} facets eventually leading to an increase in their surface energy.[ ¹¹ , ¹⁹ ] In contrast, the TiO₂ NSs after washing in an aqueous solution of 0.1 M NaOH show a total disappearance of the TiOF₂ peak while still preserving their NS morphology and anatase TiO₂ crystalline structure (Figure 2a, dark red pattern, and Figure S2a–b). The EDX analysis confirms the diminishing of fluorine in the NaOH‐washed TiO₂ NSs as compared to their as‐formed counterparts (Figure S2c–d). Therefore, we used the NaOH‐washed anatase TiO₂ NS nanocrystals enriched in {001} facets to analyze their photocatalytic performance and compare it to the anatase TiO₂ octahedral nanocrystals enriched in {101} facets. Noteworthy, in contrast to deformation and agglomeration of the NS, an intact octahedral morphology of the anatase TiO₂ nanocrystals was observed upon annealing in air at 450 °C for 1 h (Figure 2d). This is in line with the literature since the anatase (101) plane has the lowest surface energy making it the most stable surface. ^[11]

Table 1.

BET measurements of the as‐formed and annealed at 450 °C for 1 h in air NS and Oct nanocrystals.

Samples	BET [m2 g−1]
NS as‐formed	86.1
NS annealed at 450 °C	30.3
Oct as‐formed	111.7
Oct annealed at 450 °C	112.5

Since the surface composition can affect the H₂ production performance, X‐ray photoelectron spectroscopy (XPS) analysis of the as‐formed TiO₂ Oct and NaOH‐washed TiO₂ NSs was performed, and the results are summarized in Figure S3. In both cases, the high‐resolution Ti 2p XPS spectrum consists of two peaks at the binding energies of 458.3 eV and 464.1 eV (Figure S3a–b), corresponding to the Ti⁴⁺2p_3/2 and 2p_1/2 oxidation state of Ti, respectively (Figure S3a–b). ^[20] In the case of the O 1s spectra, the main peak of NaOH‐washed TiO₂ NSs is centered at 529.3 eV, and a shoulder at 530.5 eV is visible. The deconvolution of the O 1s peak leads to two main components: at 529.3 eV (84 %) corresponding to the bulk oxygen in TiO₂ (Ti−O−Ti linkages), and 530.5 eV (14 %), corresponding to surface hydroxyl groups (Ti−OH) (Figure S3c). ^[20] At the same time, the as‐formed TiO₂ Oct demonstrates a longer shoulder that was deconvoluted into three components: 529.3 eV (82 %), 530.6 eV (9 %), and 531.70 eV (6 %) corresponding to Ti−O−Ti, Ti−OH, and C−O/C=O bonds, respectively, as well as 533.0 eV (3 %), corresponding to H₂O (Figure S3d).[ ²⁰ , ²¹ ] Here the signal from C−O/C=O bonds and H₂O can be attributed to organic species and water adsorbed from the atmosphere. As shown, the XPS measurements demonstrate the comparable surface composition in both samples.

Figure 2b demonstrates the photocatalytic activity of both anatase TiO₂ NS and Oct, as well as the photocatalytic performance of their corresponding annealed counterparts, represented by the amount of H₂ generated in the 50 : 50 vol.% H₂O:MeOH mixture. It should be noted that TiO₂ NSs, used in all further experiments, are NaOH‐washed, for simplicity these samples will be named “bare” TiO₂ NSs. The TiO₂ NS or Oct photocatalysts were illuminated by UV light at λ=365 nm for 24 h and H₂ evolution was measured by gas chromatography. As shown, the H₂ evolution performance for both bare and annealed NS is significantly lower than that of as‐formed and annealed Oct being in good agreement with the literature. The H₂ evolution rate was calculated and found as 2.30×10² and 13.25×10² μL g⁻¹ h⁻¹ for bare NS and as‐formed Oct, respectively, demonstrating the >4‐fold enhanced photocatalytic activity (normalized to the available surface area according to the BET measurements) of Oct as compared to their NS analogs.

As was extensively discussed in literature, reaction sites for effective reduction reaction are located preferentially on the {101} facets – they act as electron exit sites of the anatase TiO₂ crystal, whereas {001} facets act as hole exit sites.[ ^10b , ²² ] Our results of H₂ evolution as shown in Figure 2b are well in line with the literature proofing again that the effective reduction, that is, H₂ evolution, is preferentially located on the {101} facets of the TiO₂ crystal rather than the {001} facets emphasizing once more a beneficial contribution of the exposed crystal facets on the photocatalyst efficiency.[ ^10b , ¹¹ , ²² ] In our case, TiO₂ octahedral nanocrystals with their large amount of {101} facets provide facilitated electron transfer towards H₂ evolution reaction as compared to TiO₂ NS, enriched mainly by {001} facets. Besides, a slight photocatalytic enhancement of TiO₂ Oct can be further ascribed to their 1.3 times larger surface area.

Although, the faceting of the crystals is successful in enhancing charge carrier transfer, in most practical applications, photocatalytic reactions targeting a maximum of H₂ evolution from anatase precursors widely employ co‐catalysts such as Pt, Au, Ru, or Pd. ^[23] To examine whether faceting is still beneficial after decorating the TiO₂ nanocrystals by a noble metal, Pt nanoclusters were deposited on the bare NS and as‐formed Oct using the immersion‐deposition technique as outlined in the experimental section. In contrast to photo‐deposition, the immersion‐deposition produces uniform decoration of Pt nanoclusters. The as‐formed Pt‐decorated TiO₂ NS and Oct samples were studied by HR‐SEM and TEM analysis. Figure 3a–e show the SEM, TEM images, and SEM‐EDX elemental analysis of TiO₂ NS and Oct after the Pt immersion‐deposition process. The presence of Pt nanoclusters can be observed by the appearance of bright dots on the SEM images (Figure 3a–b). As shown, Pt clusters cover the entire TiO₂ NS and Oct surface, that is, Pt nanoclusters are equally deposited on both {101} and {001} facets. The dark Pt dots on the TEM images indicate the formation of a cluster size of Pt in the range of 0.95±0.22 nm (Figure 3c–d). SEM‐EDX elemental analysis of TiO₂ NS and Oct after the Pt immersion‐deposition confirms a comparable amount of Pt loading for the two morphologies (Figure 3e).

Figure 3.

(a–b) SEM and (c–d) TEM images and (e) corresponding EDX spectra of Pt immersion‐deposited anatase TiO₂ (a, c) bare NS and (b, d) as‐formed Oct. (f) Photocatalytic H₂ evolution was obtained from anatase bare TiO₂ NS−Pt and as‐formed Oct−Pt in 50 : 50 vol.% H₂O:MeOH electrolyte.

Figure 3f demonstrates the photocatalytic H₂ evolution of both the Pt‐decorated TiO₂ bare NS and as‐formed Oct samples in the 50 : 50 vol.% H₂O:MeOH mixture. The samples were illuminated by UV light at λ=365 nm for 4 h and H₂ evolution was measured every 1 h. As expected, Pt decoration leads to the significant enhancement in the H₂ evolution rate for both facetted TiO₂ nanostructures, while in both these cases, the photocatalytic H₂ evolution rates demonstrate the same order of magnitude. The H₂ evolution rates of Pt‐decorated samples were estimated as 74.43×10³ and 52.71×10³  μL g⁻¹ h⁻¹ for the Pt‐decorated bare NS and as‐formed Oct, respectively. Furthermore, Pt‐decorated bare anatase TiO₂ NS enriched in {001} facets demonstrate ∼2‐fold enhanced photocatalytic performance over Pt‐decorated as‐formed TiO₂ Oct enriched in {101} facets (normalized to the entire surface area of NS and Oct). The latter emphasizes that the {101} preferred faceting matters strongly in the case of plain anatase TiO₂ nanostructures, that is, the preferential electron transfer is facet‐determined (rate‐determining step in heterogeneous reactions). Once decorated by Pt co‐catalyst, the Schottky junction is established leading to facile charge‐carrier transportation and separation reducing the recombination rate as well as providing efficient H⁰ recombination centers, that is, electric effects of Pt deposition becomes overall rate‐controlling; thus, the advantage of faceting that is a faster electron transfer kinetics on {101} facets vanishes entirely.

Conclusion

Anatase TiO₂ nanosheets enriched by {001} facets as well as anatase TiO₂ octahedral nanostructures enriched in {101} facets were synthesized and compared regarding their photocatalytic H₂ evolution performance with and without Pt photocatalysis. The plain TiO₂ nanocrystals of both types were decorated by immersion‐deposited Pt clusters. In contrast to photo‐deposition, immersion‐deposited Pt clusters of sub‐nm‐size were obtained on the entire TiO₂ nanocrystal surface, that is, on both {001} and {101} facets. Unlike octahedral nanostructures that improved their H₂ evolution rate by one order of magnitude (∼40‐fold enhancement), TiO₂ nanosheets demonstrated >400‐fold enhancement in their H₂ evolution rate once decorated by Pt clusters. This study indicates clearly that the preferential electron transfer is facet‐determined, that is, rate‐determining step in heterogeneous reactions, for co‐catalyst‐free anatase TiO₂ only. Once Pt‐decorated, both {001} and {101}‐enriched anatase TiO₂ nanocrystals demonstrate comparable H₂ evolution rates.

Experimental Section

Materials: Ti(OBu)₄ (98 %, Sigma‐Aldrich), HF (47 %, Carl Roth), TiCl₄ (99.9 %, Acros Organics), H₂PtCl₆ ⋅ 6H₂O (40.17 % metallic Pt weight concentration, Metakem), HCl (37 %, Sigma‐Aldrich), isopropanol (99.5 %, Sigma‐Aldrich), and NH₃ ⋅ H₂O (25 wt %, Merck) were used as received without further purification.

Synthesis of anatase TiO₂ nanosheets (NS) enriched by {001} facets: In a typical synthesis procedure, 1.2 mL HF was added dropwise to 10 mL of Ti(OBu)₄ placed in a 250 mL Teflon liner, under stirring at room temperature. The stirring was continued for the next 40 min before the Teflon containing the mixture was sealed in an autoclave, which was eventually transferred to a preheated oven at 200 °C. The reaction was completed after 24 h and the autoclave was allowed to cool in the oven naturally. After the hydrothermal reaction, the precipitates were collected and washed 3 times with ethanol and distilled water each, then dried in the oven at 75 °C overnight. To remove TiOF₂ crystalline phase, the as‐formed NSs were rather annealed in air at 450 °C for 1 h or washed in NaOH solution. The latter was performed by dispersing the as‐formed TiO₂ NSs in an aqueous solution of 0.1 M NaOH for 24 h under vigorous stirring followed by centrifugation at 4000 rpm and drying in the oven at 75 °C overnight.

Synthesis of octahedral anatase TiO₂ nanocubes (Oct) enriched by {101} facets: Initially, Ti(OH)₄ precursor was synthesized by adding 6.6 mL TiCl₄ to aqueous HCl dropwise under vigorous stirring in an ice bath. The TiCl₄ solution was then added to 5.5 wt.% NH₄OH dropwise under stirring. 4 wt.% NH₄OH was used to adjust the pH value to 7.0. After aging at room temperature for 2 h, the suspension was centrifuged and the precipitate was washed twice with DI water and ethanol. 1.0 g of Ti(OH)₄ precursor was dispersed in the mixture of 15 mL DI water and 15 mL isopropanol. After both 30 min stirring and ultrasonication, it was transferred to a 100 mL Teflon autoclave for hydrothermal synthesis in the oven (230 °C, 40 h). Upon cooling down to room temperature, the white powder could be separated by centrifugation at 4000 rpm for 10 min, followed by washing with deionized water and ethanol, each step was repeated 3 times. A series of Ti(OH)₄ precursors were synthesized by changing the amount of HCl and NH₄OH as listed in Table S1. A series of TiO₂ Oct were synthesized by changing the amount of precursor, reaction temperature and time, and the cooling method as listed in Table S2. The obtained as‐formed Oct was annealed in air at 450 °C for 1 h.

Pt immersion‐deposition process of as‐formed TiO₂ NS and Oct: 100 mg of hexachloroplatinic(IV) acid hydrate (H₂PtCl₆ ⋅ 6H₂O) was dissolved in 100 mL deionized water under vigorous stirring. After stirring for 10 min, 150 mg of either TiO₂ as‐formed NS or as‐formed Oct were added to this solution and the slurry was sonicated ultrasonically for 1 h. Upon completion, the nanoparticles were washed by centrifuging and rinsing with deionized water three times. Finally, the washed powders were collected and dried in air at 70 °C for 12 h.

Characterization methods: A field‐emission scanning electron microscope (FE‐SEM, Hitachi, S4800) equipped with an energy dispersive X‐ray detector (EDAX Genesis, fitted to SEM chamber) was used for characterization of the morphology and chemical composition of the samples. The crystallographic orientation of the surface exposed crystal facets for the prepared samples was verified using the high‐resolution transmission electron microscopy (HR‐TEM, Philips CM30). The crystal phase structure of the samples was obtained by X‐ray diffraction (XRD, X'pert Philips PMD diffractometer) operating with graphite monochromatized Cu_Kα radiation (wavelength: λ=1.54056 Å). The composition and the chemical state of the TiO₂ nanostructures were characterized using XPS (PHI 5600, US). The spectra were shifted according to the C 1s signal at 284.8 eV, and the peaks were fitted by MultiPak software. The specific surface area was performed through N₂ adsorption/desorption measurements at 77 K on a volumetric gas adsorption analyzer (Autosorb iQ XR, Anton‐Paar Quanta Tec, USA) up to 0.965. Before the analysis, the sample was degassed under high vacuum (10⁻⁷ Pa) at 200 °C for 12 h, while high purity (99.999 %) N₂ and He gases were used for the measurements. The Brunauer‐Emmett‐Teller surface area (BET) was determined concerning Rouquerol criteria for N₂ isotherm. A gas chromatograph (GC: GCMS‐QO2010SE, Shimadzu) with a thermal conductivity detector was used to measure the amount of H₂ generated. The sample was continuously stirred during irradiation and GC measurements were conducted to evaluate the amount of generated H₂.

Photocatalytic H₂ evolution measurements: In a typical experiment, 5 mg TiO₂ NS or Oct were dispersed in the 10 ml mixture of water and methanol (=50 : 50 vol.%) solution in a quartz tube. After Ar purging for 20 min and ultrasonic treatment for 15 min, this system was UV irradiated (λ=365 nm, 65 mW cm⁻²) under stirring conditions. The amount of generated H₂ was detected by the gas chromatography‐mass spectrometry system (GCMS‐QP2010 SE).

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

S. Qin, L. Shui, B. Osuagwu, N. Denisov, A. B. Tesler, P. Schmuki, ChemistryOpen 2022, 11, e202200010.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.