- Email: [email protected]
Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation
Nano Energy (]]]]) ], ]]]–]]]
1
Available online at www.sciencedirect.com
3 5 journal homepage: www.elsevier.com/locate/nanoenergy
7 9
COMMUNICATION
11
Q2
Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation
Q1
Prince Saurabh Bassia, Rajini P. Antonya, Pablo P. Boixb, Yanan Fanga, James Barbera,c,n, Lydia Helena Wonga,n,n
13 15 17 19 21 23 25 27
a
School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore b Energy Research Institute @ NTU, Nanyang Technological University, 50 Nanyang Drive, Research Techno Plaza, X-Frontier Block, Level 5, Singapore 637553, Singapore c Department of Life Sciences, Imperial College London, London SW7 2AZ, UK
29 31 33
Q6 Received 30 November 2015; received in revised form 13 January 2016; accepted 4 February 2016
35 37 KEYWORDS Hematite; Pseudobrookite; Band alignment; Impedance spectroscopy; Water splitting; Photoelectrochemical Cell
39 41 43 45 47 49
Abstract Q3 We have constructed a Fe2O3/Fe2TiO5 heterojunction based photoanode deposited on Fluorine
Doped Tin Oxide (FTO) substrate by initially fabricating hematite (Fe2O3) nanorods and subsequently pseudobrookite (Fe2TiO5) nanoporous thin film on top of them. Comparatively lower annealing temperature of 650 1C (usually 750 1C or above) was used to avoid degradation of FTO. The crystalline Fe2O3/Fe2TiO5 heterojunction shows a considerable enhancement in photocurrent density ca. 1.4 mA/cm2 and high surface charge separation efficiency of 85% at operating voltage of 1.23 V vs RHE as compared to its constituents. The crystalline heterojunction showed overall improvement in performance due to enhanced charge separation owing to the favorable band alignment with Fe2O3 nanorods and efficient injection of photogenerated holes through surface states into the electrolyte observed through Electrochemical Impedance Spectroscopy. & 2016 Published by Elsevier Ltd.
51 53 63 55 65 57
Q4
nn
Corresponding author at: School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore. E-mail addresses: [email protected] (J. Barber), [email protected] (L.H. Wong).
67
59 61
http://dx.doi.org/10.1016/j.nanoen.2016.02.013 2211-2855/& 2016 Published by Elsevier Ltd.
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
69 71
2
P.S. Bassi et al.
1
Introduction
63
3
The possibility of solar fuel production using Photoelectrochemical Cell (PEC) has developed considerably since Fujishima and Honda demonstrated in 1972, for the first time, solar water splitting using Titanium Dioxide (TiO2) as photoanode [1]. After that, efforts have been made to employ cheap, abundant and non-toxic semiconducting materials as photoanodes and to improve overall efficiency and performance of PEC devices. In the last decade, hematite (α-Fe2O3) based PEC devices have attracted many scientists in exploiting its advantages such as its good visible light absorption and stability but the reported Solar-toHydrogen (STH) efficiency (4% or less) is still far from the theoretical limit of 16.8% [2]. This is due to the high bulk/ surface recombination in hematite owing to its small minority carrier diffusion length, poor charge transport properties, slow water oxidation kinetics etc. Bulk recombination and surface recombination could be tackled by using nanostructuring, doping or surface treatments [2–4]. One way to improve the charge separation/transport in the bulk of photoanode is by forming a staggered gap (type II) band alignment based heterojunction with offsets in conduction band (ΔEC) and valence band (ΔEV) allowing facile transfer of electrons and holes and increased charge separation due to built-in potential [5]. To fabricate such heterojunctions, the key is to explore other stable and photo absorbing semiconductors to couple with hematite, which are crystalline in nature to avoid charge trapping on lattice disordering and share favorable band alignment and interface properties with hematite. Heterojunction of hematite with binary metal oxides have earlier been reported by various groups. WO3/Fe2O3 composite photoanodes [6], ZnO/Fe2O3 core-shell nanowires [7] and Fe2O3/TiO2 photoelectrodes [8] with n/n heterojunction structure yielded lower onset potential and improved performance of the PEC cell device as compared to their constituents. Heterojunction formed with ternary oxides like ferrites have also been reported. Branched array of Codoped Fe2O3 nanorods with MgFe2O4 [9] and Fe2O3:Ti/ ZnFe2O4 films [10] demonstrated good charge separation resulting in appreciable photocurrent density. These heterojunctions involved expensive element based oxides like V, Bi, W etc. which have to replace by abundant elements for a long sustainable solution. There is clearly a need to explore the heterojunction approach for large scale, cost-effective and efficient solar water splitting technology employing abundant materials with favorable band alignment and thermal/aqueous stability. We have previously reported about synthesis of an abundant mineral, Fe2TiO5, with a band gap of 2.1 eV which is stable in aqueous solutions [11]. In that work, we performed XPS/UPS measurements on pristine Fe2TiO5 and discovered that the work function (Ef) was around 4.77 eV whereas difference between work function and valence band (Ev Ef) was approximately 1.5 eV as extracted from XPS which helped us evaluate conduction band position since band gap of 2.1 eV is known from Tau plot analysis. Using this, band levels possessed by Fe2O3 and Fe2TiO5 integrate are joined as heterojunction in the schematic representation (in yellow) in Fig. 1. It shows easy electron
65
5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61
67 69 71 73 75 77 79 81 Fig. 1 Schematic representation of Fe2O3/Fe2TiO5 heterojunction showing photogenerated electron/hole transport.
injection to Fe2O3 owing to the conduction band alignment and hole transport to electrolyte due to valence band alignment between both materials. This band alignment has also been recently reported by Deng et al. [12]. There exists structural coherency between pseudobrookite and hematite which allows the overgrowth of pseudobrookite on hematite following the relation: 〈110〉 hematite || 〈101〉 pseudobrookite [13]. It also possesses the physical and chemical properties that are ideal for the construction of photoanodes for water oxidation using PEC. Liu et al. reported TiO2 based photoanodes with amorphous Fe2TiO5 overlayer acting as a visible light absorber which resulted in high performance and low onset potential [14]. Recently, Deng et al. demonstrated surface treatment of Fe2O3 with Fe2TiO5 for a surface passivating effect which enhanced photocurrent [12]. In both cases, the ultrathin Fe2TiO5 overlayer, fabricated by a solid state reaction between Fe and Ti based phases which limits the thickness of Fe2TiO5, was amorphous in nature. In present paper, we report Fe2O3/Fe2TiO5 heterojunction comprising of crystalline Fe2TiO5 overlayer capable of providing more distinct electrical junction as compared to amorphous layer. This heterojunction fabricated using low cost hydrothermal technique was characterized using impedance spectroscopy and bulk/surface charge separation efficiencies to investigate charge dynamics of the system. They were then coupled with CoOx cocatalyst to achieve enhanced performance for Fe2O3/Fe2TiO5 heterojunction based PEC device.
83 85 87 89 91 93 95 97 99 101 103 105 107 109 111 113
Results and discussion
115
The heterojunctions were synthesized on FTO substrates using the sequential hydrothermal synthesis of Fe2O3 nanorods followed by fabrication of nanoporous Fe2TiO5 films as mentioned in Section 2 (SI). A schematic representation of complete methodology is shown in Fig. 2. Reaction conditions to synthesize Fe2O3 nanorods were as earlier reported [15] whereas Fe2TiO5 films were fabricated under different reaction temperatures of 120 1C, 135 1C and 150 1C with a
117
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
119 121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21
fixed reaction time of 12 h. Pristine Fe2O3 nanorods and Fe2TiO5 films on bare FTO substrates were also prepared as control. To achieve significant photocurrents, hematite nanostructures on FTO substrates are usually thermally activated through sintering at 750 1C or higher to allow Sn diffusion into the semiconductor lattice [16]. But there is a trade-off between Sn diffusion into the semiconductor and degradation in conductivity of FTO substrate. Since we aim to study in-detail the heterojunction while avoiding degradation of FTO, the annealing temperature was kept at relatively lower sintering temperature of 650 1C. All the substrates were ramped up to 650 1C at 2 1C/min in air and maintained at this temperature for 5 h before naturally cooling down to room temperature. Fig. 3(a) shows the XRD pattern of heterojunction prepared with different reaction temperatures of 120 1C, 135 1C and 150 1C. Characteristic peaks from Fe2O3 (JCPDS no. 006-0502), Fe2TiO5 (JCPDS no. 009-0182) apart from FTO (JCPDS no. 0031114) can be observed which shows both crystalline hematite and pseudobrookite (Fe2TiO5) phase formation in heterojunction. This illustrates that hydrothermal synthesis route to
3
fabricate heterojunction provide a good control on forming a pure crystalline phase of Fe2TiO5 on crystalline Fe2O3 nanorods which is beneficial for reproducibility and in-depth study of Fe2O3/Fe2TiO5 heterojunction. To account for the photoactivity of photoanodes, it is imperative to observe the solar light absorption exhibited by semiconductor films. UV–vis absorption and reflectance spectra for Fe2O3/Fe2TiO5, Fe2O3 and Fe2TiO5 films are shown in Fig. 3(b). Pristine Fe2O3 nanorods showed much better absorption than pristine Fe2TiO5 thin film with a similar band edge due to larger thickness and higher absorption coefficient for the former. Pristine Fe2TiO5 films showed a relatively high reflectance in visible light region compared to Fe2O3. Therefore, due to this, Fe2O3/Fe2TiO5 heterojunction showed overall lower absorption in visible region (wavelengtho550 nm) than pristine Fe2O3 nanorods. The Scanning and Transmission electron microscopy were used to investigate the morphology and structure of Fe2O3, Fe2TiO5 and Fe2O3/Fe2TiO5 heterojunction. The top-view and the cross-section images are shown in Fig. 4. Crosssection image of Fe2O3-Fe2TiO5 heterojunctions, formed at
63 65 67 69 71 73 75 77 79 81 83
23
85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
103
43
Fig. 2 Process methodology of fabrication of optimized heterojunction and pristine films under 120 1C, 12 h reaction conditions.
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 3 (a) XRD spectra for heterojunction for three different reaction temperatures (b) absorbance (solid lines) and reflectance spectra (dashed lines) for optimized heterojunction and pristine films deposited at 120 1C, 12 h. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
4
P.S. Bassi et al.
1
63
3
65
5
67
7
69
9
71
11
73
13
75
15
77
17
Fig. 4 (a) Cross-section FESEM Image, (b) top-view FESEM Image for Fe2O3/Fe2TiO5 Heterojunction film deposited under optimized conditions of 120 1C, 12 h.
19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
79 81
optimized condition of 120 1C for 12 h, with vertically aligned Fe2O3 nanorods coated with Fe2TiO5 nanoporous thin film (thickness 80 nm) on top is shown in Fig. 4(a). It confirmed the presence of Fe2TiO5 nanoporous thin film of thickness around 80 to 100 nm on top of Fe2O3 nanorods with a length of 500 600 nm and circular cross-section with a diameter of 60 nm. It is to be noted that the intermediate FeOOH phase 1D nanorods formed after hydrothermal synthesis of FeCl3 precursor solution results in square-cross section which after annealing turns into Fe2O3 nanorods with circular cross-section. Cross section images for both asdeposited FeOOH and annealed Fe2O3 phases are presented in Fig. S1 (SI). The top view image (Fig. 4(b)) shows coating of crystalline Fe2TiO5 particles on hematite nanorods. It could be observed that after Fe2TiO5 coating, there is no significant difference in morphology of Fe2O3 nanorods which understandably differ with as-deposited FeOOH nanorods. At higher temperature, 150 1C (Shown in Fig. S2, SI), the more vigorous reaction results in a thicker coating of pseudobrookite particles on the nanorods. The higher reaction rate also results in rupturing the nanorods and filling more particles between the gaps. As a reference, top-view FESEM image of pristine Fe2TiO5 film on FTO substrate is shown in Fig. S3 (SI). The d spacing calculated as shown in Fig. S4 (SI) were 0.27 and 0.35 and 0.49 nm corresponding to the (230), (101) and (200) planes respectively of orthorhombic Fe2TiO5 (Pseudobrookite) crystal indicating that the crystallinity of the heterojunctions were similar to that of the individual components. This was earlier confirmed by XRD analysis (in Fig. 3(a)) which showed existence of (230) and (101) planes of Fe2TiO5 in heterojunction. To observe the presence of Ti, elemental mapping and Energy Dispersive Xray analysis (EDAX) was also carried out as shown in Fig. S5. The elemental distribution in single Fe2O3 nanorod reveals uniformly distributed Fe and comparatively lower amount of Ti representative of contribution from Fe2TiO5 particles on top of nanorod. It is also noted that the EDAX did not detect Sn in heterojunction which is consistent with the report by Ling et al. stating that annealing at 650 1C induce negligible amount of Sn from FTO [17].
Photoelectrochemical characterizations of heterojunction and pristine films
83 85
To illustrate the function of Fe2TiO5 layer on Fe2O3 nanorods, linear sweep voltammetry was performed on Fe2O3/Fe2TiO5 heterojunction, fabricated under different reaction temperatures. As control, pristine Fe2O3 nanorods and Fe2TiO5 films were also characterized under similar conditions (all are shown in Fig. S6, SI). The highest photocurrent density was observed for heterojunction fabricated at 120 1C and the lowest for 150 1C films. Therefore, all the results hitherto presented were obtained with heterojunction and pristine electrodes synthesized at 120 1C, 12 h. Fig. 5(a) shows photocurrent density-applied bias (J–V) plot for optimized Fe2O3/Fe2TiO5 heterojunction, Fe2O3 nanorods and Fe2TiO5 thin films on FTO substrates. Fe2O3/Fe2TiO5 heterojunction shows high photocurrent density of 1.4 mA/cm2 at 0.23 V vs Ag/AgCl equivalent to 1.23 V vs RHE for pH=13.6 electrolyte (1 M NaOH). Pristine Fe2O3 shows a very low photocurrent of the order of 0.01 mA/cm2 at 1.23 V vs RHE which is comparable to that reported for pristine hematite nanorods annealed at 650 1C [17]. Pristine Fe2TiO5 thin films are observed to yield better photocurrents, of around 0.2 mA/ cm2, as compared to Fe2O3 nanorods at 1.23 V vs RHE which could be due to better charge transport properties of Fe2TiO5 as compared to Fe2O3. Very recently, Damian et al. reported about hematite-pseudobrookite heterojunctions exhibiting favorable charge transfer with improved photocurrent, in their paper on role of titanium oxide as a promoter for water oxidation [18]. To observe any improvement in charge recombination mechanisms and charge separation, photocurrent density vs voltage bias curve under chopped light were evaluated as shown in Fig. 5(b). Higher anodic/cathodic transient in chopped photocurrent–potential curves signifies the high recombination of photogenerated holes with electrons or other surface defects present in the semiconductor leading to accumulation of holes. Photocurrent transient in the pristine Fe2O3 electrodes was high due to its poor charge transport properties and slow water oxidation kinetics. On the other hand, Fe2TiO5
61
87 89 91 93 95 97 99 101 103 105 107 109 111 113 115 117 119 121 123
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21
electrodes showed photocurrent with lower transient than Fe2O3 due to better charge dynamics in the system. For heterojunction electrodes, the transient is appreciably lower, for lower potentials (o0.2 V vs Ag/AgCl), which diminish with higher potentials, presumably due to improved charge separation owing to the favorable band alignment between Fe2O3 and Fe2TiO5 films. It could also be due to improvement in hole transfer through surface states, the process which has been claimed by Klahr et al. as the primary step for water oxidation [19,20]. This would be further investigated in the upcoming sections. To evaluate the performance of photoelectrodes, Incident Photon to Current Efficiency (IPCE) for optimized Fe2O3/Fe2TiO5 heterojunction and pristine films was evaluated at 1.23 V vs RHE (formula stated in Section 4, SI) as shown in Fig. 6(a). It has a similar trend as the photocurrent measurements suggesting that for the same amount of photons, more carriers were generated for heterojunction as compared to pristine films. IPCE spectra for heterojunction and pristine Fe2TiO5 films for different reaction temperatures are as shown in Fig. S7 (SI), and shows their
5
correlation with photocurrent density. To determine the stability of the heterojunction, the amperometric curve at 1.23 V vs RHE was recorded for 2 hours continuously under standard illumination conditions. The stability of the photocurrent density of around 1.470.1 mA/cm2 throughout the measurement period of 2 h (as shown in Fig. 6(b)) demonstrates the potential of Fe2O3/Fe2TiO5 heterojunction as a photoanode for practical applications.
63 65 67 69 71
Investigation into charge dynamics using Electrochemical Impedance Spectroscopy
73 75
The heterojunction show a significant enhancement in the photocurrent density corroborated by higher IPCE as compared to pristine films indicating higher charge carrier collection at back contact. While carrier generation was almost similar for Fe2O3/Fe2TiO5 heterojunction and pristine Fe2O3 nanorods (as shown by absorption spectra in Fig. 3(b)), IPCE spectra clearly indicates that the charge separation and collection is more efficient for heterojunction. To investigate further the role of
77 79 81 83
23
85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
Fig. 5 (a) Linear sweep voltammetry and (b) chopped photocurrent–potential curves for Fe2O3, Fe2TiO5 and Fe2O3/Fe2TiO5 heterojunction for 120 1C, 12 h sample.
103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 6 (a) IPCE spectra, for optimized heterojunction and pristine Fe2O3 and Fe2TiO5 photoelectrodes (b) stability test for Fe2O3/ Fe2TiO5 heterojunction prepared at 120 1C, 12 h. Signals recorded under applied bias of 1.23 V vs RHE. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
6 1 3 5 7 9 11 13 15 17 19 21 23
P.S. Bassi et al. of Rct,ss and Css. These were obtained from fitting, where the coincidence of the peak of Css and the valley of Rct,ss with the increase of the current supports the viability of the employed model. This model suggests the indirect hole transfer process through surface states into the electrolyte i.e. photogenerated holes trapped on the surface states have longer lifetime and hence take part in the water oxidation, as previously reported [22,23]. Fig. 8(b) shows a lower value of resistance for the trapping (Rtrap) for the heterojunction as compared to pristine Fe2O3 nanorods and Fe2TiO5 films which indicates the holes mediate via surface states for the former efficiently. This can contribute to a more efficient water oxidation in case of heterojunction, which is consistent with the correlation of dip in Rct,ss and peak in Css profile resulting in a more suitable charge transfer kinetics. In Fig. 8(c), the surface charge transfer resistance Rct,ss with potential bias for all the samples is presented. It is observed that the resistance reduced significantly for heterojunction suggesting easier hole injection into the electrolyte through surface states than in pristine electrodes. Fig. 8(d) shows Fig. 7 Model of equivalent circuit fitted with the Nyquist plots surface capacitance for pristine Fe2TiO5 films is low, with for heterojunction and pristine electrodes. the peak slightly shifted, suggesting the modification of the
Fe2TiO5 overlayer, Electrochemical Impedance Spectroscopy was performed and the measurement details are presented in Section 4, SI. The resulting Nyquist plots followed the behavior previously reported in these systems, and they were fitted with the corresponding equivalent circuit model (as shown in Fig. 7 below) [20,21]. In this model, Rtrap accounts for the resistance of the trapping of both photogenerated charge carriers at the surface states, whereas Rct,ss is the resistance to transfer of holes to the electrolyte through those surface states. The measured capacitance in the system is distributed in Cbulk (from the bulk material) and Css (from the surface states). Fig. 8(a) shows the photocurrent density–potential curve for optimized heterojunction (grey curve) along with the values
63 65 67 69 71 73 75 77 79 81 83 85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59 61
Fig. 8 (a) Current density–potential curve (grey), surface state capacitance (Css) and resistance through surface states (Rct,ss) with potential applied for optimized heterojunction (b) charge trapping resistance (Rtrap), (c) resistance through surface states (Rct,ss) and (d) surface state capacitance, for optimized heterojunction (blue square), pristine Fe2O3 nanorods (red circle) and pristine Fe2TiO5 nanoporous thin films (green triangle). Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
surface energetic by Fe2TiO5. Very low value of Css suggests a low surface defects density. For heterojunction, Css is similar to that for Fe2O3 since the overall surface area and hence defects density is dominated by the nanorods surface area given that the location of Fe2TiO5 is only on the top of nanorods. The Mott–Schottky plots (Shown in Fig. S8, SI) indicate that the heterojunction have the minimum slope which translates to the highest majority carrier density. This could be due to better charge separation and transport which inhibits bulk/surface recombination resulting in higher charge density.
Bulk/surface charge separation efficiency for heterojunction and pristine films Since both bulk and surface recombination play a part in enhancement of photocurrent, photoelectrochemical characterization was performed for all samples in the presence of a hole scavenger, H2O2 as shown in Fig. S9 (SI). Photooxidation of H2O2 allows the swift transport of photogenerated holes into the electrolyte and hence reduces or eliminates any surface recombination. This therefore provides a way to evaluate the extent of bulk recombination in the semiconductor [24]. Photocurrent potential curves for optimized heterojunction with (blue curve) and without hole scavenger (red curve) are shown in Fig. 9(a). There is an obvious enhancement in the photocurrent density and a cathodic shift in onset potential ( 150 mV) in the presence of H2O2 probably due to faster water oxidation kinetics. Fig. 9 (b) shows surface charge separation as η(catalysis) and bulk as η(separation) for optimized samples i.e. 120 1C, 12 h Bulk recombination is represented by η(separation) which accounts for the fraction of holes that do not recombine in bulk and reach semiconductor–electrolyte interface [24,25]. On the other hand, surface recombination is measured by η (catalysis) which determines the fraction of those holes on the semiconductor–electrolyte interface which oxidize water without recombining with defects/traps. As explained in-detail in Section 4 (SI), η(catalysis) and η (separation) values were extracted from photocurrent density obtained from water oxidation, JH2 O and from H2O2
7
oxidation JH2 O2 . Fig. 9(b) shows η(catalysis) curve (in solid line) obtained from the calculations. For Fe2O3/Fe2TiO5 heterojunction, η(catalysis) reaches almost 85% around 1.23 V vs RHE without any co-catalyst which indicates the overlayer of Fe2TiO5 on top of Fe2O3 nanorods is significantly inhibiting the surface recombination and hence enhancing the surface charge carrier efficiency. Pristine Fe2TiO5 films show a significant efficiency of 25% whereas pristine Fe2O3 nanorods attain a low efficiency of only 5% suggesting that hematite nanorods consists of high number of surface defects which limits its photocurrent density. Thus, it seems Fe2TiO5 overlayer enhances the catalytic activity on the surface of the hematite as earlier suggested by impedance spectroscopic analysis. On the other hand, bulk charge recombination was observed by η(separation) (in dash line) which was obtained as JH2 O2 =Jabs where Jabs is the total absorbed photocurrent density for films depending on their visible light absorption. The absorbance spectra along with the standard AM 1.5G solar spectrum is shown in Fig. S10 (SI) and Jabs was obtained for all the samples by their integration. Pristine Fe2O3 nanorod electrodes showed the lowest bulk charge separation efficiency (2% at 1.23 V vs RHE) validating the fact that hematite has poor charge transport properties in the bulk. The highest bulk charge separation efficiency of around 35% at 1.23 V vs RHE is shown by pristine Fe2TiO5 electrodes. Heterojunction had intermediate efficiency significantly better than hematite due to its coupling with a good charge carrier separation efficiency material i.e. Fe2TiO5. To further improve the performance of heterojunction, the system was coupled with CoOx catalyst using methodology previously reported by Liu et al. The oxidation activity of the holes was enhanced by CoOx catalyst which when coupled with Fe2TiO5 overcome the reaction barrier to effectively enhance hole transport across the semiconductor–electrolyte interface. Here, integrating this CoOx co-catalyst marginally enhances the photocurrent (10%) at 1.23 V vs RHE which shows that heterojunction without co-catalyst had low recombination of charge carriers. Improvement in onset potential of around 30 mV (Fig. 9(a)) indicates an improvement in surface kinetics for optimized heterojunction. This could be due to the presence of an energy band corresponding to CoOx catalyst which
63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 101 103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 9 (a) Current–potential curves for heterojunction coupled with co-catalyst or under the presence of hole scavenger (H2O2), (b) charge separation efficiency for surface [η(catalysis)] and bulk [η(separation)] in heterojunction. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
8
P.S. Bassi et al.
1 3 5 7 9 11 13 15 17
Fig. 10 Schematic representation of proposed charge transport model in Fe2O3/Fe2TiO5 nanoheterojunction.
19 21 23 25 27 29 31 33 35 37 39 41 43 45
regulates the hole transfer across to the electrolyte via the surface states. In the previous sections, the role of surface states was highlighted using impedance spectroscopy. As suggested by Klahr et al. surface states charging and discharging appears to be a process common to porous materials having large amount of surface defects. The hole transfer for water oxidation takes place significantly from surface trapped holes [19]. The role of surface defects as a recombination center limits the photocurrent at lower potential whereas at higher potential, the developed oxidized intermediates enhance the water oxidation kinetics. In the heterojunction devices, the higher photocurrent could be attributed to an improvement of the surface kinetics by these intermediates even at lower potentials, which is probably due to catalytic effect by Fe2TiO5 overlayer in enhancing indirect charge transfer from surface states. Very recently, Bisquert and group showed, through drift-diffusion simulations, that surface states could be beneficial for solar fuel production in providing good charge transfer kinetics and hence lower onset voltage and higher photocurrent [26]. Hence, we propose the enhancement in performance of heterojunction is attributed to the surface charge mediated water oxidation process (as shown in Fig. 10) i.e. the photogenerated holes being trapped in surface states in step 1 followed by step 2 where these oxidized species eventually enables the efficient water oxidation.
47 49
Conclusion
51
In summary, Fe2O3/Fe2TiO5 heterojunctions consisting of crystalline hematite and pseudobrookite showed superior PEC performance compared to pristine Fe2O3 nanorods. The photocurrent density at 1.23 V vs RHE for heterojunction was enhanced to 1.4 mA/cm2 without co-catalyst as compared to infinitesimally low photocurrent of 0.01 mA/cm2 exhibited by pristine Fe2O3 nanorods. High surface charge separation efficiency of 85% indicated the role of Fe2TiO5 in enhancing hole injection into the electrolyte. Heterojunction also showed an onset voltage of 0.9 V vs RHE which cathodically shifted further by around 30mV with the use of CoOx as a cocatalyst. Therefore, the fabrication of the
53 55 57 59 61
Fe2O3/Fe2TiO5 heterojunction was influential in enhancing charge separation through favorable band alignment and improvement in hole injection kinetics through surface states. These findings demonstrate a novel strategy to improve the performance of pristine Fe2O3 nanorods by fabricating crystalline heterojunction with an inexpensive and stable material (Fe2TiO5) using a simple and controllable hydrothermal technique. We believe that further research in such hybrid iron based semiconducting photocatalysts could be promising in artificial photosynthesis technology due to their cheapness, robustness, good light absorption and enhanced charge separation and transport properties.
63
Notes
77
Electronic supplementary information (ESI) available: Experimental Section, Morphological Characterization including FESEM/TEM images, Photoelectrochemical Characterization plots.
65 67 69 71 73 75
79 81 83
Acknowledgments
85
Financial support from the NTU Start Up Grant “Nanomaterials Q7 for Energy Harvesting” is kindly acknowledged. The authors are thankful to Mr. Mengyuan Zhang for TOC graphic.
87
Appendix A.
91
Supplementary material
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2016.02.013.
89
93 95 97
References [1] A. Fujishima, K. Honda, Nature 238 (1972) 37–38. [2] K. Sivula, F. Le Formal, M. Grätzel, ChemSusChem 4 (2011) 432–449. [3] D.K. Bora, A. Braun, E.C. Constable, Energy Environ. Sci. 6 (2013) 407–425. [4] P.S. Bassi, Gurudayal, L.H. Wong, J. Barber, Phys. Chem. Chem. Phys. 16 (2014) 11834–11842. [5] M.T. Mayer, Y.J. Lin, G.B. Yuan, D.W. Wang, Acc. Chem. Res. 46 (2013) 1558–1566. [6] P. Zhao, C.X. Kronawitter, X. Yang, J. Fu, B.E. Koel, Phys. Chem. Chem. Phys. 16 (2014) 1327–1332. [7] Y.K. Hsu, Y.C. Chen, Y.G. Lin, ACS Appl. Mater. Interfaces 7 (2015) 14157–14162. [8] M. Wang, M. Pyeon, Y. Gonullu, A. Kaouk, S. Shen, L. Guo, S. Mathur, Nanoscale 7 (2015) 10094–10100. [9] Y. Hou, F. Zuo, A. Dagg, P. Feng, Angew. Chem. 125 (2013) 1286–1290. [10] C. Miao, S. Ji, G. Xu, G. Liu, L. Zhang, C. Ye, ACS Appl. Mater. Interfaces 4 (2012) 4428–4433. [11] P.S. Bassi, S.Y. Chiam, Gurudayal, J. Barber, L.H. Wong, ACS Appl. Mater. Interfaces 6 (2014) 22490–22495. [12] J. Deng, X. Lv, J. Liu, H. Zhang, K. Nie, C. Hong, J. Wang, X. Sun, J. Zhong, S.-T. Lee, ACS Nano, , 2015. [13] J.F.W. Bowles, W.A. Deer, R.A. Howie, D.J. Vaughan, J. Zussman, Rock-Forming Minerals: Non-Silicates: Oxides, Q8 Hydroxides and Sulphides, Geological Society, 2011.
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
99 101 103 105 107 109 111 113 115 117 119 121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
[14] Q. Liu, J. He, T. Yao, Z. Sun, W. Cheng, S. He, Y. Xie, Y. Peng, H. Cheng, Y. Sun, Y. Jiang, F. Hu, Z. Xie, W. Yan, Z. Pan, Z. Wu, S. Wei, Nat. Commun. 5 (2014) 5122. [15] L. Xi, P.D. Tran, S.Y. Chiam, P.S. Bassi, W.F. Mak, H.K. Mulmudi, S.K. Batabyal, J. Barber, J.S.C. Loo, L.H. Wong, J. Phys. Chem. C 116 (2012) 13884–13889. [16] A. Annamalai, A. Subramanian, U. Kang, H. Park, S.H. Choi J.S. Jang, J. Phys. Chem. C 119 (2015) 3810–3817. [17] Y. Ling, G. Wang, D.A. Wheeler, J.Z. Zhang, Y. Li, Nano Lett. 11 (2011) 2119–2125. [18] D. Monllor-Satoca, M. Bartsch, C. Fabrega, A. Genc, S. Reinhard, T. Andreu, J. Arbiol, M. Niederberger, J.R. Morante, Energy Environ. Sci. 8 (2015) 3242–3254. [19] B. Klahr, S. Gimenez, F. Fabregat-Santiago, J. Bisquert T.W. Hamann, Energy Environ. Sci. 5 (2012) 7626–7636. [20] B. Klahr, S. Gimenez, F. Fabregat-Santiago, T. Hamann, J. Bisquert, J. Am. Chem. Soc. 134 (2012) 4294–4302. [21] Gurudayal, P.M. Chee, P.P. Boix, H. Ge, Y.N. Fang, J. Barber L.H. Wong, ACS Appl. Mater. Interfaces 7 (2015) 6852–6859. [22] S.R. Pendlebury, A.J. Cowan, M. Barroso, K. Sivula, J. Ye, M. Gratzel, D.R. Klug, J. Tang, J.R. Durrant, Energy Environ. Sci. 5 (2012) 6304–6312. [23] A.J. Cowan, C.J. Barnett, S.R. Pendlebury, M. Barroso, K. Sivula, M. Grätzel, J.R. Durrant, D.R. Klug, J. Am. Chem. Soc. 133 (2011) 10134–10140. [24] H. Dotan, K. Sivula, M. Gratzel, A. Rothschild, S.C. Warren, Energy Environ. Sci. 4 (2011) 958–964. [25] K. Itoh, J.O. Bockris, J Electrochem. Soc. 131 (1984) 1266–1271. [26] L. Bertoluzzi, P. Lopez-Varo, J.A. Jimenez Tejada, J. Bisquert, J. Mater. Chem. A (2015). Prince Saurabh Bassi is currently a PhD candidate, who is expected to be graduated soon, from School of Materials Science and Engineering at Nanyang Technological University (NTU), Singapore. He graduated with a B.Tech degree (Materials Science) in 2009 from Indian Institute of Technology (IIT), Kanpur, India. Prior to joining NTU for Ph.D., he worked as research associate in the field of organic electronics at Samtel Centre for Display Technologies, IIT Kanpur, India. His research interest focusses on synthesis of photocatalysts and investigation into charge carrier dynamics of semiconductor systems. Recently, he has been concentrating on exploration of novel systems like irontitanates, especially Fe2TiO5 nanoporous films, for application in solar water oxidation. Dr. Rajini P Antony is currently working as a K S Krishnan Research Fellow in Bhabha Atomic Research Center, Mumbai. She worked as a postdoctoral research fellow with Prof. Lydia Helena Wong, in School of Materials Science and Engineering, Nanyang Technological University Singapore during the period of 2013–2015. She obtained her Doctoral degree in Chemical sciences from Indira Gandhi Center for Atomic Research, Kalpakkam, India. Her research interests include Photoelectrochemical water splitting, Photocatalysis, Nano and porous material development by various synthesis routes for clean energy applications and electrocatalysis for water oxidation and reduction reactions.
9
Dr. Pablo P. Boix received his PhD. from the Universitat Jaume I (2012, Castelló, Spain). During this period, he analyzed the physical processes of optoelectrical devices including DSC, QDSC, organic photovoltaics and water splitting systems by impedance spectroscopy. In 2012 he joined ERI@N, where his research focuses on the electrochemical characterization and development of water splitting systems and perovskite solar cells, unveiling the working mechanisms which determine the performance of these optoelectrical devices.
61
Dr. Yanan Fang received her Ph.D. degree from Nanyang Technological University in Singapore, and a Bachelor of Science degree from Southeast University in Nanjing, China. She joined Energy Research Institute @NTU working on perovskite solar cell in 2012. She is currently holding a position of research fellow in School of Materials Science & Engineering, NTU. Her research interests include eco-materials, energyrelated materials and characterization of microstructures and defects in functional materials.
73
Dr. James Barber is Professor of Biochemistry, Senior Research Fellow at Imperial College London and the Cannon Visiting Professor to NTU in Singapore. He is a Fellow of the Royal Society, Member of European Academy and Foreign Member of the Swedish Royal Academy of Sciences. He has Honorary Doctorates of Stockholm University and the University of East Anglia and awarded several medals and prizes including Flintoff Medal of RSC, Novartis Medal (UK Biochem. Soc), Wheland Medal (Univ. of Chicago), Eni-Ital gas Prize and Interdiciplinary Prize Medal of the RSC. In 2009 he was the Lee Kuan Yew Distinguished Visitor to Singapore.
85
Dr. Lydia Helena Wong is currently an Assistant Professor at the School of Materials Science and Engineering, Nanyang Technological University (NTU) Singapore. She graduated with a Bachelor Degree of Applied Science (with Hons) in 2002 and Doctor of Philosophy in Materials Science and Engineering from NTU in 2007. She was also a Visiting Scholar in Department of Chemical Engineering in Stanford University. After her PhD, she was working as a Senior Engineer in Chartered Semiconductor (now: Global Foundries) before coming back to serve at NTU. Her research group currently focuses on the investigation of non-toxic and abundant metal oxides and chalcopyrite materials for clean energy application.
99
59
63 65 67 69 71
75 77 79 81 83
87 89 91 93 95 97
101 103 105 107 109 111 113 115 117 119
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
1
Available online at www.sciencedirect.com
3 5 journal homepage: www.elsevier.com/locate/nanoenergy
7 9
COMMUNICATION
11
Q2
Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation
Q1
Prince Saurabh Bassia, Rajini P. Antonya, Pablo P. Boixb, Yanan Fanga, James Barbera,c,n, Lydia Helena Wonga,n,n
13 15 17 19 21 23 25 27
a
School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore b Energy Research Institute @ NTU, Nanyang Technological University, 50 Nanyang Drive, Research Techno Plaza, X-Frontier Block, Level 5, Singapore 637553, Singapore c Department of Life Sciences, Imperial College London, London SW7 2AZ, UK
29 31 33
Q6 Received 30 November 2015; received in revised form 13 January 2016; accepted 4 February 2016
35 37 KEYWORDS Hematite; Pseudobrookite; Band alignment; Impedance spectroscopy; Water splitting; Photoelectrochemical Cell
39 41 43 45 47 49
Abstract Q3 We have constructed a Fe2O3/Fe2TiO5 heterojunction based photoanode deposited on Fluorine
Doped Tin Oxide (FTO) substrate by initially fabricating hematite (Fe2O3) nanorods and subsequently pseudobrookite (Fe2TiO5) nanoporous thin film on top of them. Comparatively lower annealing temperature of 650 1C (usually 750 1C or above) was used to avoid degradation of FTO. The crystalline Fe2O3/Fe2TiO5 heterojunction shows a considerable enhancement in photocurrent density ca. 1.4 mA/cm2 and high surface charge separation efficiency of 85% at operating voltage of 1.23 V vs RHE as compared to its constituents. The crystalline heterojunction showed overall improvement in performance due to enhanced charge separation owing to the favorable band alignment with Fe2O3 nanorods and efficient injection of photogenerated holes through surface states into the electrolyte observed through Electrochemical Impedance Spectroscopy. & 2016 Published by Elsevier Ltd.
51 53 63 55 65 57
Q4
nn
Corresponding author at: School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore. E-mail addresses: [email protected] (J. Barber), [email protected] (L.H. Wong).
67
59 61
http://dx.doi.org/10.1016/j.nanoen.2016.02.013 2211-2855/& 2016 Published by Elsevier Ltd.
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
69 71
2
P.S. Bassi et al.
1
Introduction
63
3
The possibility of solar fuel production using Photoelectrochemical Cell (PEC) has developed considerably since Fujishima and Honda demonstrated in 1972, for the first time, solar water splitting using Titanium Dioxide (TiO2) as photoanode [1]. After that, efforts have been made to employ cheap, abundant and non-toxic semiconducting materials as photoanodes and to improve overall efficiency and performance of PEC devices. In the last decade, hematite (α-Fe2O3) based PEC devices have attracted many scientists in exploiting its advantages such as its good visible light absorption and stability but the reported Solar-toHydrogen (STH) efficiency (4% or less) is still far from the theoretical limit of 16.8% [2]. This is due to the high bulk/ surface recombination in hematite owing to its small minority carrier diffusion length, poor charge transport properties, slow water oxidation kinetics etc. Bulk recombination and surface recombination could be tackled by using nanostructuring, doping or surface treatments [2–4]. One way to improve the charge separation/transport in the bulk of photoanode is by forming a staggered gap (type II) band alignment based heterojunction with offsets in conduction band (ΔEC) and valence band (ΔEV) allowing facile transfer of electrons and holes and increased charge separation due to built-in potential [5]. To fabricate such heterojunctions, the key is to explore other stable and photo absorbing semiconductors to couple with hematite, which are crystalline in nature to avoid charge trapping on lattice disordering and share favorable band alignment and interface properties with hematite. Heterojunction of hematite with binary metal oxides have earlier been reported by various groups. WO3/Fe2O3 composite photoanodes [6], ZnO/Fe2O3 core-shell nanowires [7] and Fe2O3/TiO2 photoelectrodes [8] with n/n heterojunction structure yielded lower onset potential and improved performance of the PEC cell device as compared to their constituents. Heterojunction formed with ternary oxides like ferrites have also been reported. Branched array of Codoped Fe2O3 nanorods with MgFe2O4 [9] and Fe2O3:Ti/ ZnFe2O4 films [10] demonstrated good charge separation resulting in appreciable photocurrent density. These heterojunctions involved expensive element based oxides like V, Bi, W etc. which have to replace by abundant elements for a long sustainable solution. There is clearly a need to explore the heterojunction approach for large scale, cost-effective and efficient solar water splitting technology employing abundant materials with favorable band alignment and thermal/aqueous stability. We have previously reported about synthesis of an abundant mineral, Fe2TiO5, with a band gap of 2.1 eV which is stable in aqueous solutions [11]. In that work, we performed XPS/UPS measurements on pristine Fe2TiO5 and discovered that the work function (Ef) was around 4.77 eV whereas difference between work function and valence band (Ev Ef) was approximately 1.5 eV as extracted from XPS which helped us evaluate conduction band position since band gap of 2.1 eV is known from Tau plot analysis. Using this, band levels possessed by Fe2O3 and Fe2TiO5 integrate are joined as heterojunction in the schematic representation (in yellow) in Fig. 1. It shows easy electron
65
5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61
67 69 71 73 75 77 79 81 Fig. 1 Schematic representation of Fe2O3/Fe2TiO5 heterojunction showing photogenerated electron/hole transport.
injection to Fe2O3 owing to the conduction band alignment and hole transport to electrolyte due to valence band alignment between both materials. This band alignment has also been recently reported by Deng et al. [12]. There exists structural coherency between pseudobrookite and hematite which allows the overgrowth of pseudobrookite on hematite following the relation: 〈110〉 hematite || 〈101〉 pseudobrookite [13]. It also possesses the physical and chemical properties that are ideal for the construction of photoanodes for water oxidation using PEC. Liu et al. reported TiO2 based photoanodes with amorphous Fe2TiO5 overlayer acting as a visible light absorber which resulted in high performance and low onset potential [14]. Recently, Deng et al. demonstrated surface treatment of Fe2O3 with Fe2TiO5 for a surface passivating effect which enhanced photocurrent [12]. In both cases, the ultrathin Fe2TiO5 overlayer, fabricated by a solid state reaction between Fe and Ti based phases which limits the thickness of Fe2TiO5, was amorphous in nature. In present paper, we report Fe2O3/Fe2TiO5 heterojunction comprising of crystalline Fe2TiO5 overlayer capable of providing more distinct electrical junction as compared to amorphous layer. This heterojunction fabricated using low cost hydrothermal technique was characterized using impedance spectroscopy and bulk/surface charge separation efficiencies to investigate charge dynamics of the system. They were then coupled with CoOx cocatalyst to achieve enhanced performance for Fe2O3/Fe2TiO5 heterojunction based PEC device.
83 85 87 89 91 93 95 97 99 101 103 105 107 109 111 113
Results and discussion
115
The heterojunctions were synthesized on FTO substrates using the sequential hydrothermal synthesis of Fe2O3 nanorods followed by fabrication of nanoporous Fe2TiO5 films as mentioned in Section 2 (SI). A schematic representation of complete methodology is shown in Fig. 2. Reaction conditions to synthesize Fe2O3 nanorods were as earlier reported [15] whereas Fe2TiO5 films were fabricated under different reaction temperatures of 120 1C, 135 1C and 150 1C with a
117
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
119 121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21
fixed reaction time of 12 h. Pristine Fe2O3 nanorods and Fe2TiO5 films on bare FTO substrates were also prepared as control. To achieve significant photocurrents, hematite nanostructures on FTO substrates are usually thermally activated through sintering at 750 1C or higher to allow Sn diffusion into the semiconductor lattice [16]. But there is a trade-off between Sn diffusion into the semiconductor and degradation in conductivity of FTO substrate. Since we aim to study in-detail the heterojunction while avoiding degradation of FTO, the annealing temperature was kept at relatively lower sintering temperature of 650 1C. All the substrates were ramped up to 650 1C at 2 1C/min in air and maintained at this temperature for 5 h before naturally cooling down to room temperature. Fig. 3(a) shows the XRD pattern of heterojunction prepared with different reaction temperatures of 120 1C, 135 1C and 150 1C. Characteristic peaks from Fe2O3 (JCPDS no. 006-0502), Fe2TiO5 (JCPDS no. 009-0182) apart from FTO (JCPDS no. 0031114) can be observed which shows both crystalline hematite and pseudobrookite (Fe2TiO5) phase formation in heterojunction. This illustrates that hydrothermal synthesis route to
3
fabricate heterojunction provide a good control on forming a pure crystalline phase of Fe2TiO5 on crystalline Fe2O3 nanorods which is beneficial for reproducibility and in-depth study of Fe2O3/Fe2TiO5 heterojunction. To account for the photoactivity of photoanodes, it is imperative to observe the solar light absorption exhibited by semiconductor films. UV–vis absorption and reflectance spectra for Fe2O3/Fe2TiO5, Fe2O3 and Fe2TiO5 films are shown in Fig. 3(b). Pristine Fe2O3 nanorods showed much better absorption than pristine Fe2TiO5 thin film with a similar band edge due to larger thickness and higher absorption coefficient for the former. Pristine Fe2TiO5 films showed a relatively high reflectance in visible light region compared to Fe2O3. Therefore, due to this, Fe2O3/Fe2TiO5 heterojunction showed overall lower absorption in visible region (wavelengtho550 nm) than pristine Fe2O3 nanorods. The Scanning and Transmission electron microscopy were used to investigate the morphology and structure of Fe2O3, Fe2TiO5 and Fe2O3/Fe2TiO5 heterojunction. The top-view and the cross-section images are shown in Fig. 4. Crosssection image of Fe2O3-Fe2TiO5 heterojunctions, formed at
63 65 67 69 71 73 75 77 79 81 83
23
85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
103
43
Fig. 2 Process methodology of fabrication of optimized heterojunction and pristine films under 120 1C, 12 h reaction conditions.
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 3 (a) XRD spectra for heterojunction for three different reaction temperatures (b) absorbance (solid lines) and reflectance spectra (dashed lines) for optimized heterojunction and pristine films deposited at 120 1C, 12 h. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
4
P.S. Bassi et al.
1
63
3
65
5
67
7
69
9
71
11
73
13
75
15
77
17
Fig. 4 (a) Cross-section FESEM Image, (b) top-view FESEM Image for Fe2O3/Fe2TiO5 Heterojunction film deposited under optimized conditions of 120 1C, 12 h.
19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
79 81
optimized condition of 120 1C for 12 h, with vertically aligned Fe2O3 nanorods coated with Fe2TiO5 nanoporous thin film (thickness 80 nm) on top is shown in Fig. 4(a). It confirmed the presence of Fe2TiO5 nanoporous thin film of thickness around 80 to 100 nm on top of Fe2O3 nanorods with a length of 500 600 nm and circular cross-section with a diameter of 60 nm. It is to be noted that the intermediate FeOOH phase 1D nanorods formed after hydrothermal synthesis of FeCl3 precursor solution results in square-cross section which after annealing turns into Fe2O3 nanorods with circular cross-section. Cross section images for both asdeposited FeOOH and annealed Fe2O3 phases are presented in Fig. S1 (SI). The top view image (Fig. 4(b)) shows coating of crystalline Fe2TiO5 particles on hematite nanorods. It could be observed that after Fe2TiO5 coating, there is no significant difference in morphology of Fe2O3 nanorods which understandably differ with as-deposited FeOOH nanorods. At higher temperature, 150 1C (Shown in Fig. S2, SI), the more vigorous reaction results in a thicker coating of pseudobrookite particles on the nanorods. The higher reaction rate also results in rupturing the nanorods and filling more particles between the gaps. As a reference, top-view FESEM image of pristine Fe2TiO5 film on FTO substrate is shown in Fig. S3 (SI). The d spacing calculated as shown in Fig. S4 (SI) were 0.27 and 0.35 and 0.49 nm corresponding to the (230), (101) and (200) planes respectively of orthorhombic Fe2TiO5 (Pseudobrookite) crystal indicating that the crystallinity of the heterojunctions were similar to that of the individual components. This was earlier confirmed by XRD analysis (in Fig. 3(a)) which showed existence of (230) and (101) planes of Fe2TiO5 in heterojunction. To observe the presence of Ti, elemental mapping and Energy Dispersive Xray analysis (EDAX) was also carried out as shown in Fig. S5. The elemental distribution in single Fe2O3 nanorod reveals uniformly distributed Fe and comparatively lower amount of Ti representative of contribution from Fe2TiO5 particles on top of nanorod. It is also noted that the EDAX did not detect Sn in heterojunction which is consistent with the report by Ling et al. stating that annealing at 650 1C induce negligible amount of Sn from FTO [17].
Photoelectrochemical characterizations of heterojunction and pristine films
83 85
To illustrate the function of Fe2TiO5 layer on Fe2O3 nanorods, linear sweep voltammetry was performed on Fe2O3/Fe2TiO5 heterojunction, fabricated under different reaction temperatures. As control, pristine Fe2O3 nanorods and Fe2TiO5 films were also characterized under similar conditions (all are shown in Fig. S6, SI). The highest photocurrent density was observed for heterojunction fabricated at 120 1C and the lowest for 150 1C films. Therefore, all the results hitherto presented were obtained with heterojunction and pristine electrodes synthesized at 120 1C, 12 h. Fig. 5(a) shows photocurrent density-applied bias (J–V) plot for optimized Fe2O3/Fe2TiO5 heterojunction, Fe2O3 nanorods and Fe2TiO5 thin films on FTO substrates. Fe2O3/Fe2TiO5 heterojunction shows high photocurrent density of 1.4 mA/cm2 at 0.23 V vs Ag/AgCl equivalent to 1.23 V vs RHE for pH=13.6 electrolyte (1 M NaOH). Pristine Fe2O3 shows a very low photocurrent of the order of 0.01 mA/cm2 at 1.23 V vs RHE which is comparable to that reported for pristine hematite nanorods annealed at 650 1C [17]. Pristine Fe2TiO5 thin films are observed to yield better photocurrents, of around 0.2 mA/ cm2, as compared to Fe2O3 nanorods at 1.23 V vs RHE which could be due to better charge transport properties of Fe2TiO5 as compared to Fe2O3. Very recently, Damian et al. reported about hematite-pseudobrookite heterojunctions exhibiting favorable charge transfer with improved photocurrent, in their paper on role of titanium oxide as a promoter for water oxidation [18]. To observe any improvement in charge recombination mechanisms and charge separation, photocurrent density vs voltage bias curve under chopped light were evaluated as shown in Fig. 5(b). Higher anodic/cathodic transient in chopped photocurrent–potential curves signifies the high recombination of photogenerated holes with electrons or other surface defects present in the semiconductor leading to accumulation of holes. Photocurrent transient in the pristine Fe2O3 electrodes was high due to its poor charge transport properties and slow water oxidation kinetics. On the other hand, Fe2TiO5
61
87 89 91 93 95 97 99 101 103 105 107 109 111 113 115 117 119 121 123
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21
electrodes showed photocurrent with lower transient than Fe2O3 due to better charge dynamics in the system. For heterojunction electrodes, the transient is appreciably lower, for lower potentials (o0.2 V vs Ag/AgCl), which diminish with higher potentials, presumably due to improved charge separation owing to the favorable band alignment between Fe2O3 and Fe2TiO5 films. It could also be due to improvement in hole transfer through surface states, the process which has been claimed by Klahr et al. as the primary step for water oxidation [19,20]. This would be further investigated in the upcoming sections. To evaluate the performance of photoelectrodes, Incident Photon to Current Efficiency (IPCE) for optimized Fe2O3/Fe2TiO5 heterojunction and pristine films was evaluated at 1.23 V vs RHE (formula stated in Section 4, SI) as shown in Fig. 6(a). It has a similar trend as the photocurrent measurements suggesting that for the same amount of photons, more carriers were generated for heterojunction as compared to pristine films. IPCE spectra for heterojunction and pristine Fe2TiO5 films for different reaction temperatures are as shown in Fig. S7 (SI), and shows their
5
correlation with photocurrent density. To determine the stability of the heterojunction, the amperometric curve at 1.23 V vs RHE was recorded for 2 hours continuously under standard illumination conditions. The stability of the photocurrent density of around 1.470.1 mA/cm2 throughout the measurement period of 2 h (as shown in Fig. 6(b)) demonstrates the potential of Fe2O3/Fe2TiO5 heterojunction as a photoanode for practical applications.
63 65 67 69 71
Investigation into charge dynamics using Electrochemical Impedance Spectroscopy
73 75
The heterojunction show a significant enhancement in the photocurrent density corroborated by higher IPCE as compared to pristine films indicating higher charge carrier collection at back contact. While carrier generation was almost similar for Fe2O3/Fe2TiO5 heterojunction and pristine Fe2O3 nanorods (as shown by absorption spectra in Fig. 3(b)), IPCE spectra clearly indicates that the charge separation and collection is more efficient for heterojunction. To investigate further the role of
77 79 81 83
23
85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
Fig. 5 (a) Linear sweep voltammetry and (b) chopped photocurrent–potential curves for Fe2O3, Fe2TiO5 and Fe2O3/Fe2TiO5 heterojunction for 120 1C, 12 h sample.
103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 6 (a) IPCE spectra, for optimized heterojunction and pristine Fe2O3 and Fe2TiO5 photoelectrodes (b) stability test for Fe2O3/ Fe2TiO5 heterojunction prepared at 120 1C, 12 h. Signals recorded under applied bias of 1.23 V vs RHE. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
6 1 3 5 7 9 11 13 15 17 19 21 23
P.S. Bassi et al. of Rct,ss and Css. These were obtained from fitting, where the coincidence of the peak of Css and the valley of Rct,ss with the increase of the current supports the viability of the employed model. This model suggests the indirect hole transfer process through surface states into the electrolyte i.e. photogenerated holes trapped on the surface states have longer lifetime and hence take part in the water oxidation, as previously reported [22,23]. Fig. 8(b) shows a lower value of resistance for the trapping (Rtrap) for the heterojunction as compared to pristine Fe2O3 nanorods and Fe2TiO5 films which indicates the holes mediate via surface states for the former efficiently. This can contribute to a more efficient water oxidation in case of heterojunction, which is consistent with the correlation of dip in Rct,ss and peak in Css profile resulting in a more suitable charge transfer kinetics. In Fig. 8(c), the surface charge transfer resistance Rct,ss with potential bias for all the samples is presented. It is observed that the resistance reduced significantly for heterojunction suggesting easier hole injection into the electrolyte through surface states than in pristine electrodes. Fig. 8(d) shows Fig. 7 Model of equivalent circuit fitted with the Nyquist plots surface capacitance for pristine Fe2TiO5 films is low, with for heterojunction and pristine electrodes. the peak slightly shifted, suggesting the modification of the
Fe2TiO5 overlayer, Electrochemical Impedance Spectroscopy was performed and the measurement details are presented in Section 4, SI. The resulting Nyquist plots followed the behavior previously reported in these systems, and they were fitted with the corresponding equivalent circuit model (as shown in Fig. 7 below) [20,21]. In this model, Rtrap accounts for the resistance of the trapping of both photogenerated charge carriers at the surface states, whereas Rct,ss is the resistance to transfer of holes to the electrolyte through those surface states. The measured capacitance in the system is distributed in Cbulk (from the bulk material) and Css (from the surface states). Fig. 8(a) shows the photocurrent density–potential curve for optimized heterojunction (grey curve) along with the values
63 65 67 69 71 73 75 77 79 81 83 85
25
87
27
89
29
91
31
93
33
95
35
97
37
99
39
101
41
103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59 61
Fig. 8 (a) Current density–potential curve (grey), surface state capacitance (Css) and resistance through surface states (Rct,ss) with potential applied for optimized heterojunction (b) charge trapping resistance (Rtrap), (c) resistance through surface states (Rct,ss) and (d) surface state capacitance, for optimized heterojunction (blue square), pristine Fe2O3 nanorods (red circle) and pristine Fe2TiO5 nanoporous thin films (green triangle). Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41
surface energetic by Fe2TiO5. Very low value of Css suggests a low surface defects density. For heterojunction, Css is similar to that for Fe2O3 since the overall surface area and hence defects density is dominated by the nanorods surface area given that the location of Fe2TiO5 is only on the top of nanorods. The Mott–Schottky plots (Shown in Fig. S8, SI) indicate that the heterojunction have the minimum slope which translates to the highest majority carrier density. This could be due to better charge separation and transport which inhibits bulk/surface recombination resulting in higher charge density.
Bulk/surface charge separation efficiency for heterojunction and pristine films Since both bulk and surface recombination play a part in enhancement of photocurrent, photoelectrochemical characterization was performed for all samples in the presence of a hole scavenger, H2O2 as shown in Fig. S9 (SI). Photooxidation of H2O2 allows the swift transport of photogenerated holes into the electrolyte and hence reduces or eliminates any surface recombination. This therefore provides a way to evaluate the extent of bulk recombination in the semiconductor [24]. Photocurrent potential curves for optimized heterojunction with (blue curve) and without hole scavenger (red curve) are shown in Fig. 9(a). There is an obvious enhancement in the photocurrent density and a cathodic shift in onset potential ( 150 mV) in the presence of H2O2 probably due to faster water oxidation kinetics. Fig. 9 (b) shows surface charge separation as η(catalysis) and bulk as η(separation) for optimized samples i.e. 120 1C, 12 h Bulk recombination is represented by η(separation) which accounts for the fraction of holes that do not recombine in bulk and reach semiconductor–electrolyte interface [24,25]. On the other hand, surface recombination is measured by η (catalysis) which determines the fraction of those holes on the semiconductor–electrolyte interface which oxidize water without recombining with defects/traps. As explained in-detail in Section 4 (SI), η(catalysis) and η (separation) values were extracted from photocurrent density obtained from water oxidation, JH2 O and from H2O2
7
oxidation JH2 O2 . Fig. 9(b) shows η(catalysis) curve (in solid line) obtained from the calculations. For Fe2O3/Fe2TiO5 heterojunction, η(catalysis) reaches almost 85% around 1.23 V vs RHE without any co-catalyst which indicates the overlayer of Fe2TiO5 on top of Fe2O3 nanorods is significantly inhibiting the surface recombination and hence enhancing the surface charge carrier efficiency. Pristine Fe2TiO5 films show a significant efficiency of 25% whereas pristine Fe2O3 nanorods attain a low efficiency of only 5% suggesting that hematite nanorods consists of high number of surface defects which limits its photocurrent density. Thus, it seems Fe2TiO5 overlayer enhances the catalytic activity on the surface of the hematite as earlier suggested by impedance spectroscopic analysis. On the other hand, bulk charge recombination was observed by η(separation) (in dash line) which was obtained as JH2 O2 =Jabs where Jabs is the total absorbed photocurrent density for films depending on their visible light absorption. The absorbance spectra along with the standard AM 1.5G solar spectrum is shown in Fig. S10 (SI) and Jabs was obtained for all the samples by their integration. Pristine Fe2O3 nanorod electrodes showed the lowest bulk charge separation efficiency (2% at 1.23 V vs RHE) validating the fact that hematite has poor charge transport properties in the bulk. The highest bulk charge separation efficiency of around 35% at 1.23 V vs RHE is shown by pristine Fe2TiO5 electrodes. Heterojunction had intermediate efficiency significantly better than hematite due to its coupling with a good charge carrier separation efficiency material i.e. Fe2TiO5. To further improve the performance of heterojunction, the system was coupled with CoOx catalyst using methodology previously reported by Liu et al. The oxidation activity of the holes was enhanced by CoOx catalyst which when coupled with Fe2TiO5 overcome the reaction barrier to effectively enhance hole transport across the semiconductor–electrolyte interface. Here, integrating this CoOx co-catalyst marginally enhances the photocurrent (10%) at 1.23 V vs RHE which shows that heterojunction without co-catalyst had low recombination of charge carriers. Improvement in onset potential of around 30 mV (Fig. 9(a)) indicates an improvement in surface kinetics for optimized heterojunction. This could be due to the presence of an energy band corresponding to CoOx catalyst which
63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 101 103
43
105
45
107
47
109
49
111
51
113
53
115
55
117
57
119
59
121
61
Fig. 9 (a) Current–potential curves for heterojunction coupled with co-catalyst or under the presence of hole scavenger (H2O2), (b) charge separation efficiency for surface [η(catalysis)] and bulk [η(separation)] in heterojunction. Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
123
8
P.S. Bassi et al.
1 3 5 7 9 11 13 15 17
Fig. 10 Schematic representation of proposed charge transport model in Fe2O3/Fe2TiO5 nanoheterojunction.
19 21 23 25 27 29 31 33 35 37 39 41 43 45
regulates the hole transfer across to the electrolyte via the surface states. In the previous sections, the role of surface states was highlighted using impedance spectroscopy. As suggested by Klahr et al. surface states charging and discharging appears to be a process common to porous materials having large amount of surface defects. The hole transfer for water oxidation takes place significantly from surface trapped holes [19]. The role of surface defects as a recombination center limits the photocurrent at lower potential whereas at higher potential, the developed oxidized intermediates enhance the water oxidation kinetics. In the heterojunction devices, the higher photocurrent could be attributed to an improvement of the surface kinetics by these intermediates even at lower potentials, which is probably due to catalytic effect by Fe2TiO5 overlayer in enhancing indirect charge transfer from surface states. Very recently, Bisquert and group showed, through drift-diffusion simulations, that surface states could be beneficial for solar fuel production in providing good charge transfer kinetics and hence lower onset voltage and higher photocurrent [26]. Hence, we propose the enhancement in performance of heterojunction is attributed to the surface charge mediated water oxidation process (as shown in Fig. 10) i.e. the photogenerated holes being trapped in surface states in step 1 followed by step 2 where these oxidized species eventually enables the efficient water oxidation.
47 49
Conclusion
51
In summary, Fe2O3/Fe2TiO5 heterojunctions consisting of crystalline hematite and pseudobrookite showed superior PEC performance compared to pristine Fe2O3 nanorods. The photocurrent density at 1.23 V vs RHE for heterojunction was enhanced to 1.4 mA/cm2 without co-catalyst as compared to infinitesimally low photocurrent of 0.01 mA/cm2 exhibited by pristine Fe2O3 nanorods. High surface charge separation efficiency of 85% indicated the role of Fe2TiO5 in enhancing hole injection into the electrolyte. Heterojunction also showed an onset voltage of 0.9 V vs RHE which cathodically shifted further by around 30mV with the use of CoOx as a cocatalyst. Therefore, the fabrication of the
53 55 57 59 61
Fe2O3/Fe2TiO5 heterojunction was influential in enhancing charge separation through favorable band alignment and improvement in hole injection kinetics through surface states. These findings demonstrate a novel strategy to improve the performance of pristine Fe2O3 nanorods by fabricating crystalline heterojunction with an inexpensive and stable material (Fe2TiO5) using a simple and controllable hydrothermal technique. We believe that further research in such hybrid iron based semiconducting photocatalysts could be promising in artificial photosynthesis technology due to their cheapness, robustness, good light absorption and enhanced charge separation and transport properties.
63
Notes
77
Electronic supplementary information (ESI) available: Experimental Section, Morphological Characterization including FESEM/TEM images, Photoelectrochemical Characterization plots.
65 67 69 71 73 75
79 81 83
Acknowledgments
85
Financial support from the NTU Start Up Grant “Nanomaterials Q7 for Energy Harvesting” is kindly acknowledged. The authors are thankful to Mr. Mengyuan Zhang for TOC graphic.
87
Appendix A.
91
Supplementary material
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/ j.nanoen.2016.02.013.
89
93 95 97
References [1] A. Fujishima, K. Honda, Nature 238 (1972) 37–38. [2] K. Sivula, F. Le Formal, M. Grätzel, ChemSusChem 4 (2011) 432–449. [3] D.K. Bora, A. Braun, E.C. Constable, Energy Environ. Sci. 6 (2013) 407–425. [4] P.S. Bassi, Gurudayal, L.H. Wong, J. Barber, Phys. Chem. Chem. Phys. 16 (2014) 11834–11842. [5] M.T. Mayer, Y.J. Lin, G.B. Yuan, D.W. Wang, Acc. Chem. Res. 46 (2013) 1558–1566. [6] P. Zhao, C.X. Kronawitter, X. Yang, J. Fu, B.E. Koel, Phys. Chem. Chem. Phys. 16 (2014) 1327–1332. [7] Y.K. Hsu, Y.C. Chen, Y.G. Lin, ACS Appl. Mater. Interfaces 7 (2015) 14157–14162. [8] M. Wang, M. Pyeon, Y. Gonullu, A. Kaouk, S. Shen, L. Guo, S. Mathur, Nanoscale 7 (2015) 10094–10100. [9] Y. Hou, F. Zuo, A. Dagg, P. Feng, Angew. Chem. 125 (2013) 1286–1290. [10] C. Miao, S. Ji, G. Xu, G. Liu, L. Zhang, C. Ye, ACS Appl. Mater. Interfaces 4 (2012) 4428–4433. [11] P.S. Bassi, S.Y. Chiam, Gurudayal, J. Barber, L.H. Wong, ACS Appl. Mater. Interfaces 6 (2014) 22490–22495. [12] J. Deng, X. Lv, J. Liu, H. Zhang, K. Nie, C. Hong, J. Wang, X. Sun, J. Zhong, S.-T. Lee, ACS Nano, , 2015. [13] J.F.W. Bowles, W.A. Deer, R.A. Howie, D.J. Vaughan, J. Zussman, Rock-Forming Minerals: Non-Silicates: Oxides, Q8 Hydroxides and Sulphides, Geological Society, 2011.
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013
99 101 103 105 107 109 111 113 115 117 119 121 123
Crystalline Fe2O3/Fe2TiO5 Heterojunction Nanorods with efficient charge separation 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57
[14] Q. Liu, J. He, T. Yao, Z. Sun, W. Cheng, S. He, Y. Xie, Y. Peng, H. Cheng, Y. Sun, Y. Jiang, F. Hu, Z. Xie, W. Yan, Z. Pan, Z. Wu, S. Wei, Nat. Commun. 5 (2014) 5122. [15] L. Xi, P.D. Tran, S.Y. Chiam, P.S. Bassi, W.F. Mak, H.K. Mulmudi, S.K. Batabyal, J. Barber, J.S.C. Loo, L.H. Wong, J. Phys. Chem. C 116 (2012) 13884–13889. [16] A. Annamalai, A. Subramanian, U. Kang, H. Park, S.H. Choi J.S. Jang, J. Phys. Chem. C 119 (2015) 3810–3817. [17] Y. Ling, G. Wang, D.A. Wheeler, J.Z. Zhang, Y. Li, Nano Lett. 11 (2011) 2119–2125. [18] D. Monllor-Satoca, M. Bartsch, C. Fabrega, A. Genc, S. Reinhard, T. Andreu, J. Arbiol, M. Niederberger, J.R. Morante, Energy Environ. Sci. 8 (2015) 3242–3254. [19] B. Klahr, S. Gimenez, F. Fabregat-Santiago, J. Bisquert T.W. Hamann, Energy Environ. Sci. 5 (2012) 7626–7636. [20] B. Klahr, S. Gimenez, F. Fabregat-Santiago, T. Hamann, J. Bisquert, J. Am. Chem. Soc. 134 (2012) 4294–4302. [21] Gurudayal, P.M. Chee, P.P. Boix, H. Ge, Y.N. Fang, J. Barber L.H. Wong, ACS Appl. Mater. Interfaces 7 (2015) 6852–6859. [22] S.R. Pendlebury, A.J. Cowan, M. Barroso, K. Sivula, J. Ye, M. Gratzel, D.R. Klug, J. Tang, J.R. Durrant, Energy Environ. Sci. 5 (2012) 6304–6312. [23] A.J. Cowan, C.J. Barnett, S.R. Pendlebury, M. Barroso, K. Sivula, M. Grätzel, J.R. Durrant, D.R. Klug, J. Am. Chem. Soc. 133 (2011) 10134–10140. [24] H. Dotan, K. Sivula, M. Gratzel, A. Rothschild, S.C. Warren, Energy Environ. Sci. 4 (2011) 958–964. [25] K. Itoh, J.O. Bockris, J Electrochem. Soc. 131 (1984) 1266–1271. [26] L. Bertoluzzi, P. Lopez-Varo, J.A. Jimenez Tejada, J. Bisquert, J. Mater. Chem. A (2015). Prince Saurabh Bassi is currently a PhD candidate, who is expected to be graduated soon, from School of Materials Science and Engineering at Nanyang Technological University (NTU), Singapore. He graduated with a B.Tech degree (Materials Science) in 2009 from Indian Institute of Technology (IIT), Kanpur, India. Prior to joining NTU for Ph.D., he worked as research associate in the field of organic electronics at Samtel Centre for Display Technologies, IIT Kanpur, India. His research interest focusses on synthesis of photocatalysts and investigation into charge carrier dynamics of semiconductor systems. Recently, he has been concentrating on exploration of novel systems like irontitanates, especially Fe2TiO5 nanoporous films, for application in solar water oxidation. Dr. Rajini P Antony is currently working as a K S Krishnan Research Fellow in Bhabha Atomic Research Center, Mumbai. She worked as a postdoctoral research fellow with Prof. Lydia Helena Wong, in School of Materials Science and Engineering, Nanyang Technological University Singapore during the period of 2013–2015. She obtained her Doctoral degree in Chemical sciences from Indira Gandhi Center for Atomic Research, Kalpakkam, India. Her research interests include Photoelectrochemical water splitting, Photocatalysis, Nano and porous material development by various synthesis routes for clean energy applications and electrocatalysis for water oxidation and reduction reactions.
9
Dr. Pablo P. Boix received his PhD. from the Universitat Jaume I (2012, Castelló, Spain). During this period, he analyzed the physical processes of optoelectrical devices including DSC, QDSC, organic photovoltaics and water splitting systems by impedance spectroscopy. In 2012 he joined ERI@N, where his research focuses on the electrochemical characterization and development of water splitting systems and perovskite solar cells, unveiling the working mechanisms which determine the performance of these optoelectrical devices.
61
Dr. Yanan Fang received her Ph.D. degree from Nanyang Technological University in Singapore, and a Bachelor of Science degree from Southeast University in Nanjing, China. She joined Energy Research Institute @NTU working on perovskite solar cell in 2012. She is currently holding a position of research fellow in School of Materials Science & Engineering, NTU. Her research interests include eco-materials, energyrelated materials and characterization of microstructures and defects in functional materials.
73
Dr. James Barber is Professor of Biochemistry, Senior Research Fellow at Imperial College London and the Cannon Visiting Professor to NTU in Singapore. He is a Fellow of the Royal Society, Member of European Academy and Foreign Member of the Swedish Royal Academy of Sciences. He has Honorary Doctorates of Stockholm University and the University of East Anglia and awarded several medals and prizes including Flintoff Medal of RSC, Novartis Medal (UK Biochem. Soc), Wheland Medal (Univ. of Chicago), Eni-Ital gas Prize and Interdiciplinary Prize Medal of the RSC. In 2009 he was the Lee Kuan Yew Distinguished Visitor to Singapore.
85
Dr. Lydia Helena Wong is currently an Assistant Professor at the School of Materials Science and Engineering, Nanyang Technological University (NTU) Singapore. She graduated with a Bachelor Degree of Applied Science (with Hons) in 2002 and Doctor of Philosophy in Materials Science and Engineering from NTU in 2007. She was also a Visiting Scholar in Department of Chemical Engineering in Stanford University. After her PhD, she was working as a Senior Engineer in Chartered Semiconductor (now: Global Foundries) before coming back to serve at NTU. Her research group currently focuses on the investigation of non-toxic and abundant metal oxides and chalcopyrite materials for clean energy application.
99
59
63 65 67 69 71
75 77 79 81 83
87 89 91 93 95 97
101 103 105 107 109 111 113 115 117 119
Please cite this article as: P.S. Bassi, et al., Crystalline Fe2O3/Fe2TiO5 heterojunction nanorods with efficient charge separation and hole injection as photoanode for solar water oxidation, Nano Energy (2016), http://dx.doi.org/10.1016/j.nanoen.2016.02.013