A Facile and Waste-Free Strategy to Fabricate Pt-C/TiO 2 Microspheres: Enhanced Photocatalytic Performance for Hydrogen Evolution

Afacileandwaste-freeflamethermalsynthesismethodwasdevelopedforpreparingPtmodifiedC/TiO 2 microspheres(Pt-C/TiO 2 ). The photocatalysts were characterized with X-ray diffraction, field emission scanning electron microscopy, transmission electron microscope, ultraviolet-visible (UV-vis) diffuse reflectance spectra, X-ray photoelectron spectroscopy, and thermogravimetry analysis. The photocatalytic activity was evaluated by hydrogen evolution from water splitting under UV-vis light illumination. Benefitting from the electron-hole separation behavior and reduced overpotential of H+/H 2 , remarkably enhanced hydrogen production was demonstrated and the photocatalytic hydrogen generation from 0.4 wt% Pt-C/TiO 2 increased by 22 times. This study also demonstrates that the novel and facile method is highly attractive, due to its easy operation, requiring no post treatment and energy-saving features.


Introduction
Photocatalytic water splitting for hydrogen evolution has attained great significance since hydrogen has been considered to be one of the ideal green fuels and can be used directly in fuel cells as well as in fuel cell-powered vehicles.In recent years, water splitting to produce hydrogen based on semiconductors has attracted tremendous attention [1][2][3][4][5][6].As a typical semiconductor material, TiO 2 has been widely studied in photocatalytic area due to its abundance, photostability, chemical inertness, and low toxicity [7][8][9][10].However, the high recombination rate of photogenerated electrons and holes in TiO 2 would limit its photocatalytic activity [11,12].The relatively low light harvesting efficiency behaves as another drawback of TiO 2 in its photocatalytic utilization [13,14].These factors strongly inhibit the extensive use of naked TiO 2 in solar-driven water splitting.
In the past decades, many attempts have been made to explore new avenues to prolong the life time of photogenerated carriers for higher quantum efficiency or to narrow the band gap of TiO 2 for a much larger visible fraction in its light absorption spectrum, such as modification of TiO 2 through implantation of metal ions [15,16], doping of nonmetal atoms [17,18], and combination with other components, including secondary semiconductors [19,20], noble metals [21,22], and dyes [23,24].
It was once proposed that carbon species incorporated in TiO 2 can act as sensitizers [25,26], thus enhancing the photocatalytic activity of TiO 2 by expanding light absorption range into visible region.There comes introduction of several nanostructured carbon materials such as carbon nanotubes [27] and graphene [28] for enhancement of photocatalytic H 2 production.Besides, Pt nanoparticles are excellent candidates to serve as cocatalysts since their Fermi-energy levels are normally more positive than the conduction band of TiO 2 .Thus it can attract the photon-excited electrons to suppress the possible electron-hole recombination [29,30].
Great efforts have been made to synthesize Pt modified TiO 2 composites with high electrochemical or photocatalytic activity [31][32][33][34][35][36].Ran et al. [2] recently addressed the fact that International Journal of Photoenergy cocatalysts especially Pt metals are a key factor in achieving higher photocatalytic activity in most semiconductor-based photocatalytic systems.Sreethawong et al. [34] prepared Ndoped nanocrystalline mesoporous-assembled TiO 2 ; various Pt contents were loaded onto the nanocomposites through the incipient wetness impregnation method.It was observed that the impregnation of Pt considerably improved its photocatalytic H 2 production activity and the optimum Pt loading content was 1.3 wt%.Ismail and coworkers [35] synthesized Pt/RuO 2 -TiO 2 photocatalysts by using photoreduction method and followed by calcination at 450 ∘ C for 4 h.The asprepared samples exhibited excellent photocatalytic activity for photooxidation of CH 3 OH under UV light irradiation.Wang's group [36] obtained Pt decorated TiO 2−x N x via a one-pot route.Enhanced photocatalytic activity for degradation of propylene under visible light was achieved.The improved performance was ascribed to the synergistic effect of oxygen vacancies and dopants of N and Pt.Although these methods have their own advantages, they are more or less limited by high temperatures or special facilities.Therefore, developing a more facile approach to fabricate Pt based TiO 2 photocatalysts is still very necessary.Recently, we have reported an easy route for constructing C/TiO 2 [37], Fe-C/TiO 2 [16], SiO x -C/TiO 2 [38], and Ta 2 O 5 [3] photocatalysts.Our strategy featured with short reaction time, easy operation, low cost, waste free and only one step to obtain the final TiO 2 .In the present work, we further introduced Pt to modify C/TiO 2 in order to achieve higher activity for the photocatalysts.The synthesis of Pt-C/TiO 2 nanomaterials via the simple flame thermal approach was achieved and their higher performance was revealed as desired.The influence of Pt content on the phase, microstructures, and optical and photocatalytic properties was investigated in detail.

Preparation of Pt-C/TiO 2 .
A series of Pt-C/TiO 2 nanocomposites were fabricated by using the aforementioned chemicals as starting materials, as illustrated in Scheme 1.Briefly, the specified weight of H 2 PtCl 6 ⋅6H 2 O was added into a standard beaker (150 mL) containing 35 mL of ethanol and 5 mL of TBOT by stirring.The obtained stable solutions were ignited by a match stick under ordinary conditions.Gray powders were obtained after gentle burning for about 30 min.The final samples with different amount of Pt: C/TiO 2 mass ratios were labeled as 0.1 wt%, 0.4 wt%, and 0.6 wt% Pt-C/TiO 2 , respectively.For comparison, C/TiO 2 powders were also prepared by a similar procedure except for the absence of Pt precursor [37].

Characterizations.
The structure and phase composition of the samples were examined by X-ray diffraction (XRD) on a Bruker D/8 advanced diffractometer using Cu K radiation.The morphologies were characterized by field emission scanning electron microscope (FE-SEM: Philips XL30) and transmission electron microscope (TEM: JEOL JEM-2011, Japan).A Cary-50 Scan UV-Vis spectrophotometer was used to measure the UV-vis diffuse reflectance spectra (DRS) with the scan range in the region of 200-800 nm.The instrument employed for X-ray photoelectron spectroscopy (XPS) studies was a RBD upgraded PHI-5000 C ESCA system (Perkin Elmer) with Al/Mg K radiation.The binding energy was corrected using the C1s level at 284.6 eV as an internal standard.Thermogravimetry analysis (TGA) was carried out on a SDT-Q600 (TA Company, USA) instrument with a heating rate of 10 ∘ C min −1 using oxygen as the purge gas.

Photocatalytic Activity Measurements.
The photocatalytic activity of samples was measured under UV-vis light illumination by using 500 W Xe lamp (CHF-XM35, Trusttech Co., Ltd., Beijing) as light source.The light intensity was kept at 180 mW cm −2 , which was measured by an optical power meter (1 L 1400 A, International Light).The experiments were carried out in a closed quartz reactor system.Typically, the photocatalysts (65 mg) were suspended in an aqueous methanol solution (80 mL of distilled water, 20 mL of methanol) by means of a magnetic stirrer within the reactor.Prior to the experiment, the mixture was dispersed by ultrasound treatment for 15 min, followed by purging N 2 gas for 30 min.The amount of evolved H 2 was determined by a GC7900 gas chromatograph (thermal conductivity detector, molecular sieve 5 A, 99.999% N 2 carrier).

Results and Discussion
3.1.XRD Patterns.Figure 1 shows the XRD patterns of the synthesized Pt-C/TiO 2 powders, and the patterns can be well indexed to the anatase TiO 2 , which were determined directly without any subsequent high temperature calcination.The peaks corresponding to Pt could not be detected in Figures 1(c)-1(e); this is mainly ascribed to the relatively low content of this element in the nanocomposites.Similar results can also be found in the previous report [31].The average crystallite sizes () calculated from the main diffraction peaks using Scherrer equation are present in Table 1.It can be seen that there is nearly no difference in the  value among all the samples (ca.20 nm), implying that the introduction of Pt did not alter the crystallite size of TiO 2 .

SEM and TEM Images.
The morphologies of the samples with various Pt content were characterized by SEM.As shown in Figure 2, the microsphere with a diameter of ∼0.5 to 2.0 m was observed for all of the prepared samples.It has been recognized that the microspheres can allow multiple reflections and scattering of incident light, which enhances light harvesting and increases the quantity of photogenerated electrons and holes [37].Therefore, it is anticipated that the photocatalytic performance will be improved for the samples.As compared with C/TiO 2 , the introduction of Pt does not cause noticeable change in morphology, but more microspheres are fused together with increasing Pt content, and this change is observed more clearly in Figure 2(d).These demonstrations reveal that limited fraction of Pt is required to obtain relatively dispersed microspheres.
Figure 3 depicts the TEM images of 0.4 wt% Pt-C/TiO 2 nanocomposite.As displayed in Figure 3(a), TiO 2 microsphere is composed of small primary nanoparticles of about 20 nm in diameter; this observation is in agreement with the XRD evaluation.The HRTEM image is presented in Figure 3(b); one can clearly observe that the lattice fringes are characteristics of the anatase TiO 2 crystal, in which the d-spacing of 0.35 nm corresponds to the distance between the (101) planes [39].The corresponding selectedarea electron diffraction (SAED) pattern of the nanocomposite (Figure 3 sample with Pt fraction up to 1.2 wt%.TEM image of the sample is shown in Figure 3(d).Interestingly, many spherical shaped particles with diameter of 2 to 5 nm can be easily found; the lattice fringes of 0.23 nm match the crystallographic planes of Pt (111) [31].This result further revealed the presence of Pt nanoparticles in the resulted samples.

Formation Mechanism of TiO 2
Spheres.Scheme 1 shows possible formation mechanism of TiO 2 microspheres.Selfgenerated water from ethanol combustion promoted the hydrolysis of TBOT molecules to form TiO 2 nanoparticles.Then nucleation and growth of the particles by coagulation and condensation occurred along the axial direction of the flame.After this period, the nanoparticles are surrounded by the solution, where the water content, gradually increasing in the system, leads to the continuous hydrolysis of TBOT.Nucleated clusters of TiO 2 particles quickly aggregated to form solid spheres and crystalized to anatase phase at the elevated temperature [40].Due to the rapid hydrolysis of TBOT and abrupt shrinkage of the droplets, during the firing in air, the final TiO 2 powders consist of a large amount of aggregated spheres [41].
When small amount of H 2 PtCl 6 ⋅6H 2 O is present in the solution, thermal decomposition of H 2 PtCl 6 ⋅6H 2 O would occur under the pyrolysis condition and Pt particles are simultaneously formed.Since the fast thermolysis and low fraction of H 2 PtCl 6 ⋅6H 2 O, the transformation to Pt reached so fast that the pyrolysis rate can continually exceed the hydrolysis and condensation rate of TBOT.Consequently, it is assumed that Pt nanoparticles may be embedded in the C/TiO 2 nanocomposite. is no absorption band from PtCl 6 2− complexes, which would exhibit a ligand-to-metal charge transfer band with a maximum at 262 nm and two bands at around 360 and 480 nm corresponding to d-d transitions [42,43].Furthermore, the absorption edges increased linearly with the increase of Pt concentration (see Table 1).The value of optical band gap energy of 3.15 eV was calculated for pure TiO 2 according to the equation of   (eV) = 1240/  [44], which is the same as theoretical value of the anatase phase.Narrowed band gap energies were obtained for the as-synthesized Pt-C/TiO 2 composites, and the data are presented in Table 1.The stronger absorption intensity and longer absorption edge can be attributed to the following factors: (i) carbon in samples can act as sensitizer, which will expand light absorption range into visible region [45]; (ii) Pt-C/TiO 2 photocatalysts probably exhibit surface plasmon resonance (SPR) due to the presence of Pt particles [46], which will extend the absorption edge to visible light.The SPR peak for the Pt nanoparticles is typically below 450 nm with broad peak shape; thus usually it can hardly be observed [47,48].

XPS Analysis.
To investigate the chemical state of elements in the sample, XPS analysis is conducted and the results are presented in Figure 5. Figure 5(a) is the XPS spectra of a survey spectrum for a typical sample.Elements of Ti, O, C, and Pt can be detected in this spectrum.In the Ti 2p spectrum of TiO 2 (Figure 5(b)), there are two peaks centered at 458.5 eV and 464.2 eV, which are characteristic of the Ti 2p3/2 and Ti 2p1/2 for Ti (IV) [49,50], respectively.Figure 5(c) shows the XPS spectrum of C 1s of the sample.The peak at 284.6 eV is ascribed to elemental carbon, arising from the incomplete burning of organic compounds [38].The other small peak at higher binding energies of 288.5 eV is attributed to C-O bonds [51], which originates from the insufficient hydrolysis of TBOT.It is to be noted that the absence of the characteristic binding energy of 281.0 eV for C-Ti bonds indicates that the carbon did not exist as a dopant; that is, most of the carbon species in TiO 2 are present as elemental state [52].Figure 5(d) displays XPS spectrum for Pt 4f.Two peaks at around 71.4 eV and 74.7 eV are assigned to 4f7/2 and 4f5/2 of Pt (0), respectively [53].This result confirmed the presence of metallic Pt in the nanocomposite, which is in line with the observations from TEM and UV-vis DRS.presented in Figure 6.Three main weight losses can be seen at about 100 ∘ C, 300 ∘ C, and 600 ∘ C, respectively.The weight loss at 100 ∘ C originates from water and residual ethanol evaporation.The second step around 300 ∘ C is ascribed to the decomposition of residual organics on the surface of TiO 2 .Another peak is attributed to the oxidation of carbon species.The peak relevant to Pt oxidation has not been found within 800 ∘ C, suggesting the high stability of Pt in the nanocomposite.
3.7.Photocatalytic H 2 Production Activity.Photocatalytic activities of the samples were assessed and compared with that of commercial P25 by determining the hydrogen production from aqueous methanol solution.Such a sacrificial agent system is widely used in the studies of photocatalytic H 2 production [34,54,55].As depicted in Figure 7, the hydrogen evolution on P25 is negligible under the applied experimental conditions.It should be reasonable to detect hydrogen evolution from C/TiO 2 .Since carbon in TiO 2 is likely to carry out a charge transfer process and is responsible for the photosensitization of TiO 2 [45,56].The generated hydrogen amount is significantly increased when taking the samples of Pt-C/TiO 2 as photocatalysts.The improved photocatalytic performance is mainly because the modified Pt is able to capture electrons and decrease the overpotential of H + /H 2 , thus leading to the decrease of the electron-hole pairs recombination [57].The maximum H 2 evolution is achieved for about 5.1 mol on the sample of 0.4 wt%Pt-C/TiO 2 after 3 h irradiation.Significantly, the value is about 22 times that of P25.Nevertheless, the amount of H 2 decreased to 4.8 mol when Pt concentration gets to 0.6 wt%.This phenomenon agrees with that of previous reports [55,58].The reason probably is that excessive concentration of Pt nanoparticles will lead to abundant Pt-trapped electrons, which enable easier encounters between the diffusing holes and the Pt-trapped electrons.Consequently, Pt becomes more of electron-hole recombination centers instead of electron-hole separation enhancers [59].
It should be recognized that the activity for H 2 evolution is relatively low and more work is necessary to further increase the H 2 production.While in this work, the amount of Pt is relatively smaller to obtain the optimum photocatalytic activity compared with that reported in literatures [34,60].
The reduced Pt content on TiO 2 will be important both from a commercial and an ecological point of view.The photocatalytic efficiency of hydrogen evolution is expected to further improve through the better design of hydrogen evolution reactor, the modification of nanocomposites, and the optimization of energy levels of photocatalysts.Corresponding experiments are in progress in this laboratory and the results will be reported in due course.

Conclusions
In summary, we presented a novel and facile flame thermal method for the preparation of Pt modified C/TiO 2 microspheres in the first time and demonstrated their high photocatalytic performance for H 2 evolution.The as-prepared anatase TiO 2 powders show enhanced absorption behavior and obvious red shift in photoresponse towards visible light region.Compared to commercial P25, the resulted samples show significantly improved photocatalytic activity.Given the relatively simple, rapid, scalable, highly efficient, and energysaving properties of this method, it is reasonable to conclude that this technique will be transferable to the fabrication of other modified TiO 2 microspheres.

H 2
Scheme 1: (a) Schematic illustration of the fabrication procedure of Pt-C/TiO 2 composite.(b) Schematic representation of Pt-C/TiO 2 microsphere growth process.

Figure 4 2 −
displays the UV-vis diffuse reflectance spectra of the as-prepared samples.A strong absorption in the visible light region is observed for the Pt-C/TiO 2 photocatalysts.The results also indicate that PtCl 6 complexes transformed to Pt completely since there

Table 1 :
Crystal sizes, absorption edges, and band gap energies of the synthesized samples.
a  TiO 2 is calculated from the most intense diffraction peak (101) of XRD spectra of the samples.
3.6.TGA Curve.The thermal behavior of a typical sample is probed by measuring thermal analysis and TGA curve is