Effect of a Co-Based Oxygen-Evolving Catalyst on TiO2

Subscriber access provided by UNIV LAVAL

Article

Effect of a Co-Based Oxygen Evolving Catalyst on TiO2-Photocatalyzed Organic Oxidation Xiao Zhang, Xianqiang Xiong, Lianghui Wan, and Yiming Xu Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b01240 • Publication Date (Web): 30 Jul 2017 Downloaded from http://pubs.acs.org on August 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Effect of a Co-Based Oxygen Evolving Catalyst on TiO2-Photocatalyzed Organic Oxidation Xiao Zhang, Xianqiang Xiong, Lianghui Wan, and Yiming Xu* State Key Laboratory of Silicon Materials and Department of Chemistry, Zhejiang University, Hangzhou 310027, China.

ABSTRACT. Cobalt phosphate (CoPi) is a promising co-catalyst for the (photo)electrochemical oxidation of water over semiconductor electrodes in a phosphate solution, but CoPi effect on organic oxidation has been studied a little. Herein, we report a case-sensitive effect of CoPi on the TiO2 photocatalyzed oxidation of phenol, 4-chlorophenol (CP), and 2,4-dichlorophenol (DCP) in a phosphate-containing suspension at pH 7.0. A photochemical method was used to deposit Pt onto TiO2, followed by CoPi deposition onto Pt/TiO2 and TiO2. In all reactions, Pt/TiO2 and CoPi/TiO2 were always more and less active than TiO2, respectively. In comparison with Pt/TiO2, CoPi/Pt/TiO2 was less active for phenol oxidation, but more active for CP and DCP oxidation. The latter was also observed from the photocatalytic reduction of O2 into H2O2. For DCP oxidation in a phosphate-free suspension at pH 7, however, CoPi/Pt/TiO2 was much less active than either Pt/TiO2 or TiO2, ascribed to the dissolution of Co2+ ions that act as a recombination center. It is proposed that the CoIV species, formed from the hole oxidation of CoII/III in CoPi, are

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

surface-bound and short-lived. They can react with the adsorbed substrate (CP, DCP and H2O2) nearby, but would deactivate either in the absence of Pt (O2 reduction catalyst) or phosphate (CoPi repairer). Moreover, there is a synergism between the CoPi-mediated hole transfer and the Pt-mediated electron transfer, which would improve the efficiency of the charge separation, and consequently increase the rates of O2 reduction and organic oxidation.

1. INTRODUCTION Cobalt phosphate (CoPi) as a co-catalyst of water oxidation has received great attention in the recent years. 1−8 This CoPi is usually deposited onto a semiconductor film electrode through the (photo)electrochemical oxidation of Co2+ ions in a phosphate aqueous solution. The resulting precipitate contains both CoII and CoIII, and has a structure of cobalt oxide core likely terminated by phosphate anions.6 To date, CoPi has been deposited onto different semiconductors, including TiO2. All of the composite materials are claimed to be more active than blank semiconductor for the photoelectrochemical oxidation of water in a phosphate buffer solution. In this device, the electrons photogenerated from a semiconductor film electrode are externally transferred to a counter electrode, whereas the remaining photoholes undergo water oxidation to release O2. It is widely recognized that the enhanced water oxidation is due to the formation of a CoIV species from the hole oxidation of CoII/III in CoPi. If this holds, such CoIV species would be useful for organic oxidation. In neutral aqueous solution, many organic compounds have a one-electron reduction potential less positive than 1.3 V versus normal hydrogen electrode (NHE).9 The exact oxidation potential of CoIV is not known, but it would be more positive than 0.82 V vs NHE for


2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

water oxidation to O2. Then CoPi might be also capable of catalyzing organic oxidation over an irradiated semiconductor. However, such effect of CoPi has been little studied in the literature. Semiconductor photocatalysis has been widely studied as a new method of organic oxygenation and pollutant treatment.10,11 Different from the photoelectrochemical device, the photocatalytic system operates with an irradiated semiconductor, on which organic oxidation and O2 reduction occur simultaneously. As a result, the photoelectrons and holes easily recombine to heat, and the quantum yield of organic oxidation is usually very low. For example, the quantum yield of phenol oxidation, measured in an aerated aqueous suspension of P25 TiO2, is only 0.14 at 365 nm.12 In semiconductor photocatalysis, therefore, CoPi may play a different role from that in semiconductor photoelectrochemistry. To date, anatase TiO2 is the most studied photocatalyst for organic oxidation, mainly due to its low cost, high activity and stability, as well as good affinity and reactivity toward O2 in aqueous solution.13−16 Interestingly, Majima and coworkers have reported a positive effect of CoPi on the TiO2-photocatalyzed oxidation of 2 µM APF (3-paminophenyl fluorescein) in a phosphate aqueous solution containing 100 mM DMSO (dimethyl sulfoxide).17 Through a time-resolved diffuse reflectance spectroscopy, they have observed a decrease in the hole lifetime of TiO2, and ascribed it to the interfacial hole transfer from TiO2 to CoII/III in CoPi. In such suspension, however, DMSO is also organics and would compete with APF for the reactive species, whereas the colorful APF can absorb visible light, and its oxidation may occur through a dye-sensitized pathway. Therefore, the effect of CoPi on the TiO2photocatalyzed organic oxidation still remains unclear, and it is worthy being further investigated.


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

In this work, we report a case-sensitive effect of CoPi on the TiO2-photocatalyzed oxidation of phenol, 4-chlorophenol (CP), and 2,4-dichlorophenol (DCP) in a phosphate aqueous solution at pH 7. Reactions were carried out under UV light at wavelengths longer than 320 nm. Under such conditions, all of the model substrates did not absorb the incident light, and their oxidation would occur only through a TiO2 photocatalyzed pathway. In all reactions, however, a CoPi-deposited TiO2 was always less active than blank TiO2. To explore the effect of CoPi, a Pt-deposited TiO2 was made, followed by CoPi deposition. In comparison with Pt/TiO2, interestingly, CoPi/Pt/TiO2 was less active for phenol oxidation, but more active for CP and DCP oxidation, respectively. Several influencing factors were then examined, including organic adsorption, O2 reduction, CoPi loading, and catalyst stability in the absence and presence of phosphate. Furthermore, a possible mechanism responsible for the observed different effect of CoPi is discussed.

2. EXPERIMENTAL SECTION Materials. Anatase TiO2, chloroplatinic acid, horseradish peroxide (POD), and N,N−diethyl-pphenylenediamine (DPD) were purchased from Sigma−Aldrich. Other chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd., including Co(NO3)2, K3PO4, phenol, 4chloropehnol (CP), and 2,4-dichlorophenl (DCP). All solutions were prepared by a Milli-Q ultrapure water. The solution pH was adjusted with a dilute solution of HClO4 or NaOH. A Pt-loaded TiO2 (Pt/TiO2) was prepared by the photochemical reduction of Pt(VI).18,19 An aqueous suspension containing 0.9 g TiO2, 0.5 mM CH3OH and 120 µL H2PtCl6 was irradiated with a 300 W mercury lamp (Shanghai Yamin) for 3 h. Then the solid was collected, washed with water several times, and dried at 80 oC. A CoPi-loaded TiO2 (CoPi/TiO2), and a CoPi-


4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

loaded Pt/TiO2 (CoPi/Pt/TiO2) were also prepared by a photochemical method.17 A suspension containing 0.50 mM Co(NO3)2, and 6 g/L TiO2 or Pt/TiO2 in 0.1 M K3PO4 (KPi) at pH 7.0 was irradiated with a 300 W mercury lamp for 3 h. Then the solid was collected, washed with water several times, and dried at 80 oC. The amounts of H2PtCl6 and Co(NO3)2 remaining in the filtrates were measured by an inductively coupled plasma (ICP) mass spectroscopy. Then the weight percent of Pt in Pt/TiO2 was calculated to be 0.52 %, whereas the weight percents of Co in CoPi/TiO2 and CoPi/Pt/TiO2 were calculated to be 0.48 %, respectively. Characterization. X-ray diffraction (XRD) pattern was recorded on a D/max-2550/PC diffractometer (Rigaku), using a Cu Kα as the X-ray irradiation source. According to the fullwidth at half-maximum of the (101) anatase at two theta 25.3o, the average crystal diameter (dXRD) of TiO2 was calculated by using the Scherrer equation. Adsorption-desorption isotherm of N2 on solid was measured at 77 K on a Micromeritics ASAP2020 apparatus. From the adsorption isotherm, the Brunauer‒Emmett‒Teller (BET) specific surface area, a t-plot micropore volume, and total pore volume at a relative pressure of 0.995 were calculated, respectively. But the Barrett‒Joyner‒Halenda (BJH) average pore size was calculated from the desorption isotherm. Diffuse reflectance spectrum (DRS) was recorded on a Shimadzu UV-2550 with BaSO4 as a reference. X-ray photoelectron spectroscopy (XPS) was made on a Kratos AXIS UItra DLD spectrometer. The element binding energies were calibrated with C 1s at 284.8 eV. Scanning electron microscope (SEM) measurement was performed on a SU-8010, attached with energydispersive X-ray spectroscopy (EDS). High-resolution transmission electron microscope (TEM) image was obtained with a JEM-2100F. Photocatalysis and Analysis. Reactions were carried out at 25 oC in a Pyrex−glass reactor. Unless stated otherwise, all suspensions were prepared under fixed conditions (0.1 M KPi at pH


5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

7.0, 1.00 g/L catalyst, 0.43 mM phenol, 0.60 mM CP, and 0.60 mM DCP). A suspension containing necessary components was first stirred in the dark for 30 min, and then irradiated with a 300 W high pressure mercury lamp, equipped with a 320 nm cut-off filter. At given intervals, 2.0 mL of the suspension was withdrawn, and filtered through a 0.22 µm membrane, followed by analysis with HPLC (high performance liquid chromatography) on a Dionex P680 (Apollo C18 reverse column). The eluent was an aqueous solution of CH3OH at 50% for phenol, and 70% for CP and DCP, respectively. H2O2 was measured at 553 nm on an Agilent 8453 UV–visible spectrophotometer, through a POD-catalyzed oxidation of DPD.20 The amount of organic adsorption on solid in aqueous solution (q) was calculated from the difference between C0 and Ce, where C0 and Ce represent the concentration of organic substrate in aqueous solution before and after the suspension was stirred in the dark for 30 min. Table 1. Physical Parameters of the Catalysts Catalyst

ds (nm)

Asp (m2/g)

dp (nm)

Vm (cm3/g)

Vt (cm3/g)

qmax (µmol/g)b

K (mM−1)b

TiO2

13.6

141

83.8

0.0044

0.324

94

27

CoPi/TiO2

13.4

139

87.1

0.0021

0.345

81

17

Pt/TiO2

13.5

142

86.1

n.d

0.368

110

43

CoPi/Pt/TiO2

13.6

147

83.1

n.d

0.352

82

30

CoPi/Pt/TiO2c 13.7

127

81.8

n.d

0.310

−

−

a

dXRD, crystal diameter; Asp, surface area; dp, average pore size; Vm, micropore volume; Vt, total pore volume; n.d, not detectable. bqmax, maximum amount of adsorption; K, adsorption constant; calculated from the Langmuir adsorption isotherms of DCP in 0.1 M KPi (Figure S6). cSample collected after recycle experiment (Figure 4).

3. RESULTS AND DISCUSSION


6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Characterization. Solid was characterized with several techniques, and the results are shown in Figures S1−S4 of the Supporting Materials. Briefly, all samples showed a XRD pattern in good agreement with those of anatase TiO2 (PDF #65-5714). No diffractions of CoPi and Pt were observed, mostly due to their low contents or amorphous nature in the samples. By the Scherrer equation, the average size of anatase crystallites was calculated, which was about 13.5 nm for all samples. These observations indicate that the crystal phase of TiO2 remains unchanged after Pt and CoPi loading. In the adsorption−desorption isotherm of N2 on TiO2, there was a hysteresis loop, indicative of mesopores present in the sample. After TiO2 was loaded with CoPi or Pt, the solid micropore volume was decreased, but the solid total pore volume and average pore size were both increased (Table 1). These observations indicate that the micropores of TiO2 are occupied or blocked by the fine particles of CoPi or Pt, in accompany with construction of a new and large mesopore. After Pt/TiO2 was loaded with CoPi, the solid total pore and average pore size were both decreased, due to CoPi present in the mesopores. Among the samples, however, there was no substantial difference in the BET surface area. This is probably due to the Pt and CoPi loading that change the solid morphology in a different way. The SEM and TEM images of the samples showed that the TiO2 particles highly aggregated together, with an average crystallite diameter of approximately 20 nm. After TiO2 was loaded with Pt or CoPi, all of the relevant elements (Ti, O, Pt, Co, and P) were measurable by EDS. Through a high-resolution TEM image of CoPi/Pt/TiO2 (Figure S2), the lattice distances for anatase (101) and Pt (111) facets were measured to be 0.351 and 0.226 nm, respectively. But the crystal facets of CoPi in CoPi/TiO2 and CoPi/Pt/TiO2 were not observed, indicative of CoPi present in the amorphous form. Figure 1A shows the absorption spectra of the samples. There was a strong absorption band at wavelengths shorter than 400 nm, assigned to the intrinsic band gap transition (O2− → Ti4+).


7

Langmuir

After TiO2 was deposited by CoPi and Pt, the sample color changed from white to pale-yellow and gray, respectively. Meanwhile, CoPi/TiO2 also showed a weak absorption at wavelengths up to 450 nm, due to the forbidden d−d transition of cobalt cations in CoPi. The spectral base lines of Pt/TiO2 and CoPi/Pt/TiO2 also shifted upward in the whole region of visible light, due to the localized surface plasma resonance of Pt particles on TiO2.21 Figures 1B and C show the XPS spectra of Pt 4f and Co 2p, respectively, recorded with CoPi/Pt/TiO2. The binding energies of Pt 4f7/2 at 71.02 eV and Pt 4f5/2 at 74.27 eV are accordant with those for metallic Pt, whereas the peak at 76.2 eV is assigned to PtO2.21−23 The Co 2p peaks at 781.93 and 797.83 eV are attributed to 2p3/2 of Co2+ and 2p1/2 of Co3+, respectively.2,5,8,24 The binding energies of P 2p measured from CoPi/Pt/TiO2 and CoPi/TiO2 were 133.72 and 133.97 eV, respectively. This is indicative of P5+ that is present in CoPi, but not on the solid surface.2,8 Similar spectral data of Pt 4f and Co 2p were also obtained from Pt/TiO2 and CoPi/TiO2, respectively (Table S1). Moreover, all samples showed similar binding energies of Ti4+ (2p2/3, 459.2 ± 0.1 eV; 2p1/3, 464.9 ± 0.1), and O2− (lattice O 1s, 530.5 ± 0.2; surface O 1s, 532.6 ± 0.1 eV). These observations indicate that the surfaces of TiO2 are covered by a PtO2-containing Pt, and a CoII/III- containing CoPi, respectively. Moreover, there is no significantly specific interaction between the deposits and TiO2.

0.6

0.2 0.0

4

300

400

2

0 250

500

λ (nm)

(A)

300

350

400

450

Wavelength λ (nm)

500

(a) (b) (c) (d) 550

Pt 4f7/2 71.02

Intensity(a.u.)

FR

6

Co 2p3/2 781.93

Pt(IV) 76.08

0.4

Pt 4f5/2 74.27

(C) Co 2p1/2 797.83

Intensity (a.u.)

8

Kubelka-Munk unit FR

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

(B)

69

72

75

78

81

84

Binding energy (eV)

780

790

800

810

Binding energy (eV)


8

Page 9 of 31

Figure 1. (A) Absorption spectra of (a) TiO2, (b) CoPi/TiO2, (c) Pt/TiO2, and (d) CoPi/Pt/TiO2. XPS spectra of (B) Pt 4f, and (C) Co 2p, recorded with CoPi/Pt/TiO2. The dotted lines represent the curve fitting with a Lorentzian−Gaussian function.

0.45

0.6

0.6

0.5

(b)

0.35

0.5

CP (mM)

(a)

0.30 (d) 0.25

(b)

(b)

DCP (mM)

0.40

Phenol (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(a) 0.4 (c)

(a)

0.3 (c)

0.3

(A)

(c) 0.2

20

40

60

80

0

20

Irradiation time (min)

40

60

(C)

0.2

(d)

(B)

0.20 0

0.4

(d)

0.1

80

0

20

40

60

80



Figure 2. Photocatalytic degradation of (A) phenol, (B) CP, and (C) DCP in 0.1 M KPi, measured in the presence of (a) TiO2, (b) CoPi/TiO2, (c) Pt/TiO2, and (d) CoPi/Pt/TiO2. Table 2. Relevant Data of Organic Oxidation in Figure 2a Catalyst

R0 (phenol) R0 (CP) R0 (DCP) q (CP) q (DCP) R0/q (CP) R0/q (DCP)

TiO2

1.79

1.47

1.76

33

64

4.45

2.75

CoPi/TiO2

0.76

1.16

1.33

25

47

4.63

2.84

Pt/TiO2

3.48

3.35

5.05

39

77

8.58

6.56

CoPi/Pt/TiO2 3.37

5.90

8.43

37

58

16.0

14.5

a

R0 (µM/min), initial rate of organic oxidation; q (µmol/g), the amount of organic adsorption; R0/q (10−2 g L−1min−1), specific initial rate of organic oxidation.

Oxidation of Phenol. Figure 2 show the results of phenol, CP, and DCP oxidation over different catalysts, measured in 0.1 M KPi at pH 7.0 under UV light. In all cases, the concentration of organic substrate in aqueous phase decreased with the irradiation time. These curves satisfactorily fit the first-order rate equation (Figure S5), and the resulting apparent rate


9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

constants of organic oxidation (kobs) are given in Table S2. However, a linear fitting does not ensure that the reaction follows the first order kinetics.13,14 In fact, the value of kobs for organic oxidation is inversely proportional to Ce, the equilibrium concentration of organic substrate in solution, measured before light irradiation. To compare the catalyst activity, therefore, it is better to use the initial rate of organic oxidation (R0), determined at the initial irradiation time. In the present study, the value of R0 was derived from kobsCe, and the results are listed in Table 2. First, Pt/TiO2 was always more active than TiO2. The initial rates of phenol, CP and DCP oxidation, obtained from Pt/TiO2, were 1.94, 2.28, and 2.87 times larger than those, measured from TiO2, respectively. This is mostly due to metallic Pt in Pt/TiO2, that catalyzes the multielectron reduction of O2, and hence promotes the hole oxidation of organic substrate.18,19 Second, CoPi/TiO2 was always less active than TiO2. The initial rates of phenol, CP and DCP oxidation, obtained from CoPi/TiO2, were 2.36, 1.27, and 1.32 times smaller than those, measured from TiO2, respectively. The present result is opposite not only to that observed from APF oxidation,17 but also to the prediction described in the Introduction section. Third, in comparison with Pt/TiO2, CoPi/Pt/TiO2 was less active for phenol oxidation, but more active for CP and DCP oxidation, respectively. After Pt/TiO2 was deposited by CoPi, the initial rate of phenol oxidation was decreased by 1.03 times, whereas the initial rates of CP and DCP oxidation were increased by 1.76 and 1.67 times, respectively. These observations indicate that the effect of CoPi is determined by both organic substrate and photocatalyst used for assessment. Forth, the initial rate of CP oxidation over CoPi/Pt/TiO2 was 1.31 times larger than the sum of individual rates, measured from Pt/TiO2 and CoPi/TiO2, respectively. Such rate increase was also observed from DCP oxidation, which was 1.32 times. These observations imply that there is a cooperative effect of CoPi and Pt, which greatly promotes the TiO2 photocatalyzed oxidation of CP and DCP,


10

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

respectively. These results seem contradictory each other. However, understanding them is not straightforward. Several influencing factors need to be considered, including the adsorption of O2 and organic substrate on different solids, which will be discussed below. Effect of Organic Adsorption. In 0.1 M KPi at pH 7.0, phenol poorly adsorbed on TiO2, but CP and DCP highly adsorbed on TiO2 (Table 2). This is in line with the octanol-water partition coefficient of organic substrate (Log Kow) that increases in the order of phenol (1.46) < CP (2.39) < DCP (3.06). It means that the adsorbed substrates on solid are mostly in a molecular form (pKa = 9.98, 9.37, and 7.9 for phenol, CP and DCP, respectively). To better describe the sorption behaviors of different solids, the adsorption isotherms of DCP on those solids in 0.1 M KPi at pH 7.0 were separately measured in the dark, and the results are shown in Figure S6. All isotherms were Langmuir-type, and satisfactorily fit the Langmuir adsorption equation, q/qmax = KCe/(1 + KCe), where qmax is the maximum amount of adsorption, and K is the adsorption constant. The resulting adsorption parameters of DCP are listed in Table 1. The values of qmax increased in the order of CoPi/TiO2 < CoPi/Pt/TiO2 < TiO2 < Pt/TiO2. This trend in qmax among the solids does not match those changes in the solid BET surface area, and average pore size, but it is accordant with that change in the solid total pore volume, and/or in the solid micropore volume. For instance, Pt/TiO2 has the largest total pore volume, and hence the largest value of qmax, whereas TiO2 has the largest micropore volume and hence the second largest value of qmax. These observations indicate that DCP (0.69 nm in diameter) can enter the solid micropores and mesopores from aqueous solution. Since the solid pores are blocked by CoPi, the values of qmax and K, obtained from CoPi/TiO2 and CoPi/Pt/TiO2, are smaller than those measured from TiO2 and Pt/TiO2, respectively. However, similar micropore blocking of TiO2 and CoPi/TiO2 by Pt show an increased value of qmax and K. This is probably indicative of a strong interaction


11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

between Pt and DCP in the solid mesopores. On the other hand, the amounts of CP and DCP adsorption, measured before light irradiation (Figure 2), also increased in the order of CoPi/TiO2 < CoPi/Pt/TiO2 < TiO2 < Pt/TiO2, respectively (Table 2). Furthermore, the amounts of DCP adsorption on TiO2 and Pt/TiO2 after CoPi deposition are decreased, but not increased. It means that the CoIII species in CoPi that are not reactive enough to oxidize DCP in aqueous solution. In general, the initial rate of organic degradation (R0) over TiO2 in aqueous solution is linearly proportional to the initial amount of organic adsorption (q), due to increase in the interfacial hole transfer from TiO2 to organics.25 To evaluate the relative activity for different photocatalysts, R0 needs to be normalized with q, and the results are shown in Table 2. According to this specific rate of R0/q, interestingly, CoPi has a positive effect on both the TiO2 and Pt/TiO2 photocatalyzed reactions. After CoPi deposition, the specific rates of CP and DCP oxidation over TiO2 are increased by 1.04 and 1.03 times, respectively, whereas the specific rates of CP and DCP oxidation over Pt/TiO2 are increased by 1.86 and 2.21 times, respectively. That is, Pt/TiO2 has an activity enhancement larger than does TiO2 upon CoPi deposition. Furthermore, the specific rate of organic oxidation, obtained from CoPi/Pt/TiO2, is also larger than the sum of individual specific rates, measured from Pt/TiO2 and CoPi/TiO2 (1.21 times for CP, and 1.54 times for DCP). These observations imply that CoPi is capable of catalyzing the hole oxidation of CP and DCP. Since Pt can catalyze the electron reduction of O2,18,19 such positive effect of CoPi is larger in the presence of Pt than that in the absence of Pt. As far as phenol is concerned, it weakly absorbs onto TiO2 in aqueous solution. Then it is difficult to examine the effect of phenol adsorption. However, there is a common feature that the rates of phenol oxidation over TiO2 and Pt/TiO2 are both decreased after CoPi loading (Table 2). Assuming that after CoPi deposition the amount of O2 adsorption on solid in aqueous solution is


12

Page 13 of 31

decreased. Then the electron reduction of O2 is slowed down, and the electron recombination with the photoholes is speeded up, and the hole oxidation of phenol is retarded. Since Pt can catalyze O2 reduction, such detrimental effect of CoPi loading would be less serious with Pt/TiO2 than with TiO2. After CoPi loading, in practice, the rate decrease of phenol oxidation over Pt/TiO2 (1.03 times) is indeed much less than that over TiO2 (2.36 times). Such effects of CoPi and Pt deposition on the rate of O2 reduction could be also inferred from CP and DCP oxidation. After CoPi deposition, the specific rate increases of CP and DCP oxidation over TiO2 are smaller than those over Pt/TiO2, respectively. After Pt deposition, however, the specific rate increases of CP and DCP oxidation over CoPi/TiO2 are larger than those over TiO2, respectively (3.46 vs.1.93 times for CP, and 5.11 vs. 2.39 for DCP). These observations illustrate again that there is indeed a cooperative effect of Pt and CoPi that greatly promotes the TiO2-photocatalyzed reaction. If the above proposal is operative, the photocatalytic reduction of O2 to H2O2 would also change from one catalyst from another, which will be shown blow.

0.5

12

(e)

(A)

0.4

(B)

H2O2 (mM)

(d)

9

H2O2 (uM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(c) 6 (a) 3

0.3 0.2 (a)

(c) 0.1

(b)

(d) 0.0

(b) In dark

under UV light

0 0

10

20

30

40

50

60

-60

-40


0

10

20

30

40

Reaction time (min)


13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

Figure 3. (A) Photocatalytic formation of H2O2 in presence of 0.60 mM DCP. (B) Adsorption and photocatalytic decomposition of H2O2 in absence of DCP. Experiment was performed in 0.1 M KPi at pH 7.0 in presence of (a) TiO2, (b) CoPi/TiO2, (c) Pt/TiO2, and (d) CoPi/Pt/TiO2.

Reduction of O2. It is proposed that the formation of H2O2 from the irradiated TiO2 can occur through the proton-coupled two electron reduction of O2, or through the hole oxidation of H2O, followed by •OH dimization.11,26 The latter pathway will be terminated in the presence of hole scavenger such as DCP. Figure 3A shows the result of H2O2 formation over different catalysts, measured under UV light in 0.1 M KPi at pH 7.0 containing 0.60 mM DCP. First, with TiO2, the concentration of H2O2 in aqueous phase increased with time toward a platform. Second, with CoPi/TiO2, the concentration of H2O2 also increased with time toward a platform. But the maximum concentration of H2O2, obtained from CoPi/TiO2, was about half of that, measured from TiO2. Third, with Pt/TiO2, the concentration of H2O2 quickly reached a maximum in 10 min, and then slowly decreased with time. Forth, with CoPi/Pt/TiO2, the concentration of H2O2 also increased, and then decreased with time. But the maximum concentration of H2O2, observed from CoPi/Pt/TiO2 at 40 min, was 1.47 times higher than that, observed from Pt/TiO2 at 10 min. Fifth, with all catalysts, the amount of H2O2 produced was about one order of magnitude smaller than the amount of DCP disappeared at given time (Figure 2C). These observations indicate that there is a competition between the formation and decomposition of H2O2. To compare the catalyst activity for the photocatalytic reduction of O2, therefore, the fate of H2O2 with different catalysts needs to be examined, which will be shown blow.


14

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

A separate experiment was then carried out with 0.50 mM H2O2 in 0.1 M KPi, and the result is shown in Figure 3B. In the dark, the amount of H2O2 adsorption on CoPi/TiO2 (37 µmol/g) was smaller than that on TiO2 (45 µmol/g), due to decrease in the solid micropore volume. Under UV light, the concentration of H2O2 in aqueous phase decreased with time. Control experiment without catalyst showed a negligible photolysis of H2O2. Then the observed decrease of H2O2 concentration with time is due to the TiO2-photocatalyzed decomposition of H2O2. After CoPi loading, interestingly, the initial rate of H2O2 decomposition, calculated at the first 5 min, was increased by 2.04 times, from 34.1 µM/min to 69.5 µM/min. Recently, Guillard and co-workers have reported that the initial rate of H2O2 decomposed is proportional to the initial amount of H2O2 adsorbed on TiO2 in aqueous solution.26 In the present case, a decrease of H2O2 adsorption still results into an increase of H2O2 decomposition. These observations indicate that CoPi has a positive effect on the TiO2 photocatalyzed decomposition of H2O2. In the presence of Pt/TiO2 or CoPi/Pt/TiO2, however, the concentration of H2O2 quickly decreased with time, even in the dark. Such fast disappearance of H2O2 is mainly due to the Pt-catalyzed decomposition of H2O2.16,19,27 Due to the adsorption and decomposition of H2O2 on solid, the total amount of H2O2 produced from O2 reduction should be higher than that measured in aqueous phase (Figure 3A). However, assessing the total amount of H2O2 is not straightforward. In comparison with TiO2, CoPi/TiO2 has a 1.2-fold decrease in the amount of H2O2 adsorption, but a 2.5-fold increase in the rate of H2O2 decomposition. If two photocatalysts have the same rates of O2 reduction, CoPi/TiO2 should have a 1.3-fold increase in the concentration of H2O2 in solution, as compared with TiO2. In practice, however, the maximum concentration of H2O2 in solution, obtained from CoPi/TiO2, is 1.91 times lower than that measured from TiO2 (Figure 4A). It implies that the real rate of O2 reduction over CoPi/TiO2 is lower than that of TiO2. In other words, the amount of O2 adsorption


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

on CoPi/TiO2 in aqueous solution is smaller than that on TiO2. This is ascribed to decrease in the solid micropore volume, as observed from CP, DCP and H2O2 adsorption. As for as Pt/TiO2 and CoPi/Pt/TiO2 are concerned, both of them have a fast decomposition of H2O2 in the dark. Then the observed maximum concentration of H2O2 in solution can be taken as evidence that the real rate of O2 reduction over CoPi/Pt/TiO2 is larger than that over Pt/TiO2. In comparison with TiO2, moreover, Pt/TiO2 has a faster decomposition of H2O2, and a larger maximum concentration of H2O2 in solution, and thus a higher real rate of O2 reduction. According to the above reasoning, the catalyst activity for O2 reduction increases in the order of CoPi/TiO2 < TiO2 < Pt/TiO2 < CoPi/Pt/TiO2. This trend in the apparent rate of O2 reduction is in good agreement with that in the initial rate of DCP oxidation (Table 2). On the other hand, the photocatalytic reduction of O2 to H2O2 was carried out in the presence of DCP. Then the different amounts of DCP adsorption on those solids in aqueous solution (Table 1) may have influence on the rate of O2 reduction. For this concern, the maximum concentration of H2O2 in aqueous phase, obtained from different solids at different time (Figure 3A), were tentatively divided by the relevant amounts of DCP adsorption (Table 2). But the resulting values (10−2 g/L) still follow an increasing order of CoPi/TiO2 (5.5) < TiO2 (7.6) < Pt/TiO2 (9.1) < CoPi/Pt/TiO2 (17.7). Therefore, we can conclude that after CoPi and Pt depositions, the rates of O2 reduction over the irradiated TiO2 are surely decreased and increased, respectively, which is accordant with the above proposal. Effect of CoPi Deposition. First, the deposition time (td) of CoPi onto Pt/TiO2 was examined, and the result is shown in Figure S7 and Table S3. In this experiment, a suspension containing 0.60 mM Co2+ in 0.1 M KPi at pH 7 was irradiated for different times, followed by collection, washing, and drying in an oven at 80 oC. In the dark, the amount of DCP adsorption on solid in 0.1 M KPi decreased with the increase of td. This is due to the amount of CoPi in Co/Pt/TiO2 that


16

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

increases with td, as measured by chemical analysis. Under UV light, the initial rate of DCP oxidation increased with td, and then decreased after reaching a maximum at 3 h. This trend in activity among the solids remained unchanged, when the initial rate of DCP oxidation was normalized with the amount of DCP adsorption. The rate increase of DCP oxidation with td confirms that CoPi has a positive effect on the Pt/TiO2 photocatalyzed oxidation of DCP. The rate decrease of DCP oxidation with td is probably due to excess CoPi that reduces the amount of O2 adsorption (Figure 3A), and/or the number of photons reaching Pt/TiO2 (Figure 1A). Otherwise, excess CoPi may act as a recombination center, as proposed for APF oxidation,17 and/or play as an inhibitor of the interfacial hole transfer, as proposed for water oxidation.8 Second, the deposition of CoPi onto Pt/TiO2 was made during the photocatalytic oxidation of DCP, and the results are shown in Figure S8 and Table S4. In this experiment, a suspension containing 1.0 g/L Pt/TiO2, 8.4 µM Co(NO3)2 and 0.54 mM DCP in 0.1 M KPi solution at pH 7.0 was irradiated, followed by DCP analysis. Before light irradiation, the amounts of DCP adsorption on solid in the absence and presence of Co2+ were 50.8, and 41.0 µmol/g, respectively. It means that all of Co2+ ions added in the suspension have been adsorbed onto Pt/TiO2. Under UV light, the concentration of DCP in aqueous phase decreased with time. But two curves obtained in the absence and presence of Co2+ overlapped each other, till an irradiation time of 60 min. After that, the rate of DCP oxidation in the presence of Co2+ was 1.19 times larger than that in the absence of Co2+. After the rate of DCP oxidation was normalized with the amount of DCP adsorption, the resulting specific rate of DCP oxidation in the presence of Co2+ was 1.48 times larger than that in the absence of Co2+. This is indicative of the CoPi deposition onto Pt/TiO2 that occurs at 60 min. Before 60 min, the hole oxidation of Co2+ is slower than the hole oxidation of DCP. This is due to the amount of Co2+ adsorption on Pt/TiO2 (8.4 µmol/g) that


17

Langmuir

is smaller than the amount of DCP adsorption (41 µmol/g) in the beginning. At 60 min, about 65% of DCP is oxidized, and the hole oxidation of Co2+ occurs rapidly. This CoPi/Pt/TiO2, prepared in the presence of DCP, is defined as Sample A. In comparison with Sample B (CoPi/Pt/TiO2 prepared without addition of DCP), Sample A was less active for DCP oxidation. The apparent rate and specific rates of DCP oxidation, obtained from Sample A, were 1.50 and 1.84 times smaller than those, measured from Sample B, respectively. This might be due to the fact that Co2+ loading in Sample A (8.4 µmol/g) is lower than that in Sample B (84 µmol/g). These observations indicate that the deposition of CoPi onto Pt/TiO2 can be achieved either off-site or on-site, both of which have a good performance of CoPi in participation of the Pt/TiO2 photocatalyzed reactions.

0.6 0.5

DCP (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

0.4 0.3 0.2 0.1 0.0 0

100

200

300

400

500

600


Figure 4. Recycling test for DCP oxidation over CoPi/Pt/TiO2 in 0.1 M KPi at pH 7. After each run, new solution of DCP was added into the irradiated suspension, followed by light irradiation.


18

Page 19 of 31

Catalyst Stability. A recycling test was carried out by using CoPi/Pt/TiO2 as photocatalyst. Figure 5 show the result of DCP oxidation in 0.1 M KPi under UV light. From the first run to the last, the apparent rate constant of DCP oxidation decreased from 0.0243 min−1 to 0.0201 min−1. Such decrease of the reaction rate with time is mostly due to the intermediates of DCP oxidation that compete with DCP for the reactive species, and due to the net amount of catalyst in the suspensions that decreases from one run to another. After the recycling test, moreover, the solid was collected, and dried at 60 oC, Then the sample was characterized with XRD, SEM, TEM, N2 adsorption, and absorption spectroscopy, and the results are shown in Figures S9, and S1−S4. After 10 h, the sample showed no changes in the crystallite diameter of anatase (13.7 nm), and in the lattice distances of (101) anatase (0.350 nm) and (111) Pt (0.226 nm). But the BET surface area, and total pore volume of the irradiated sample were decreased by 1.15 times, respectively, as compared with those of the un-irradiated sample (Table 1). This is probably due to the reaction intermediates that adsorb on the solid, and/or due to the solid fine particles that are lost during DCP analysis. These observations indicate that CoPi/Pt/TiO2 is very stable, and can be repeatedly used for DCP oxidation in 0.1 M KPi.

0.56

0.5

(b)

(c)

DCP (mM)

DCP (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

0.4

0.3

(a)

0.2

0.48

(b)

(c)

0.40

(b) (a)

(B)

(A)

(a)

0.32 0.1 0

20

40

60

80

100

120

0


20

40

60

80



19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

Figure 5. (A) DCP oxidation over CoPi/Pt/TiO2 at initial pH 7, (a) in 0.1 M KPi, and (b) in aqueous solution. (B) DCP oxidation over TiO2 (solid symbols), and Pt/TiO2 (open symbols) in aqueous solution at initial pH 7, containing (a) no Co2+, (b) 4.2 µM Co2+, and (c) 8.4 µM Co2+.

Effect of Phosphate Buffer. Figure 5A shows the result of DCP oxidation over CoPi/Pt/TiO2 in aqueous solution at initial pH 7.0, measured either in the absence or presence of 0.1 M KPi. In both cases, the amounts of DCP adsorption on solid were similar. Under UV light, however, the initial rate of DCP oxidation in the absence of phosphate (0.98 µM/min) was 8.56 times smaller than that in the presence of phosphate (8.39 µM/min). This is indicative of CoPi/Pt/TiO2 that suffers from a serious deactivation in a phosphate-free solution. Such problem has been found from water oxidation in the literature. For example, the electrochemical oxidation of water over a CoPi/TiO2 electrode in 0.1 M NaClO4 shows a dark current much lower than that in 0.1 M NaPi. This is ascribed to the requirement of phosphate anions as proton acceptor for O2 evolution.6 For the photoelectrochemical oxidation of water over a CoPi/Fe2O3 electrode, the photocurrent in 1 M NaOH is comparable to that in 1 M NaOH and 0.1 M KH2PO4.24 For the photoelectrochemical oxidation of water, a CoPi/TiO2 electrode has a good performance in 0.1 M KPi at pH 1−14.8 These literature studies indicate that CoPi can maintain its activity for water oxidation, either in a phosphate-containing solution or in a highly alkaline solution. In the present case, CoPi may dissolve from CoPi/Pt/TiO2 into a phosphate-free aqueous solution at pH 7. If the dissolved CoPi species in solution are inert, similar activities of CoPi/Pt/TiO2 and Pt/TiO2 should be observed in a phosphate-free solution. However, this is not the case, as will be shown below.


20

Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5B shows the result of DCP oxidation over Pt/TiO2 and TiO2 in a phosphate-free solution at initial pH 7.0. In the absence of Co2+, the initial rates of DCP oxidation over TiO2 and Pt/TiO2 were 2.42 and 3.77 µM/min, respectively. Surprisingly, these rates were 2.47 and 3.85 times larger than that measured from CoPi/Pt/TiO2 (0.98 µM/min), respectively. It implies that some species, dissolved from CoPi/Pt/TiO2 in a phosphate-free aqueous solution, are detrimental to the photocatalytic reaction. To practice this hypothesis, the effect of Co2+ ions was examined. On the addition of 4.2 and 8.4 µM Co2+, the initial rates of DCP oxidation over Pt/TiO2 were decreased by 1.30 and 1.34 times, respectively (Table S5). On the addition of 4.2 and 8.4 µM Co2+, the initial rates of DCP oxidation over TiO2 were also decreased by 1.60 and 2.12 times, respectively. Moreover, the amount of DCP adsorption on solid was decreased on the addition of Co2+. After normalization with the amount of DCP adsorption, however, the resulting specific rates of DCP oxidation over TiO2 and Pt/TiO2 were still decreased on the addition of Co2+. It is worth noting that the amount of Co2+ adsorption on Pt/TiO2 in a phosphate-free solution (1 µmol/g) was much lower than that on Pt/TiO2 in 0.1 M KPi (8.2 µmol/g). These observations indicate that Co2+ ions in a phosphate-free solution are detrimental to all of the TiO2 and Pt/TiO2 photocatalyzed oxidation of DCP. The exact reason for that is not known, but Co2+ ions may act as a recombination center of electrons and holes. Palmisano and coworkers have reported that after TiO2 is loaded with 0.22, 0.74, 1.48 wt% Co2+, the initial rates of 4-nitrophenol degradation in aqueous solution are decreased by 1.90, 2.11, and 6.33 times, respectively. Meanwhile, the second-order rate constants of electron−hole recombination in the picosecond time scale are increased by 1.64, 1.97, and 2.14 times, respectively.27 In aqueous solution at pH 7.0, the potential for the Co2O3/Co2+ couple is 0.50 V vs NHE,2 which is less and more positive than the valence and conduction band edge potentials of anatase TiO2, respectively (Scheme 1). This


21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

would allow the hole oxidation of Co2+ to Co3+, followed by the electron reduction of Co3+ to Co2+. Such null cycle of Co2+ would consume a pair of electrons and holes, consequently decreasing the rates of O2 reduction and DCP oxidation. In 0.1 M KPi, however, Co2+ highly adsorbs on solid, and CoPi sticks to the solid, due to its low solubility. As a result, the rate of DCP oxidation over Pt/TiO2 in 0.1 M KPi in the presence of Co2+ is increased after 60 min of irradiation time. Accordingly, one can conclude that the deactivation of CoPi/Pt/TiO2 in a phosphate-free solution is due to the detrimental Co2+ ions that are dissolved from the solid in aqueous solution. Therefore, phosphate anions should be present in the photocatalytic system, as so to keep CoPi immobilized on solid, and to maintain the catalytic activity of CoPi as well. Possible Mechanism. In TiO2 photocatalysis, the electrons and holes are photogenerated, and consumed in a pair. Since Pt can catalyze the electron reduction of O2, its deposition on TiO2 greatly promotes the hole oxidation of phenol, CP and DCP, respectively. This is equivalent to an improvement in the efficiency of the charge separation, and/or in the apparent photocatalytic activity of TiO2. After CoPi is deposited on TiO2, the solid micropore volume is reduced, and the amount of O2 adsorption on solid in aqueous solution is decreased. As a result, the electron reduction of O2 is slowed down, and the hole oxidation of phenol is retarded. When organic adsorption is considered, however, the specific rates of CP and DCP oxidation over CoPi/TiO2 are slightly larger than those over TiO2, respectively. Since CoIII species in CoPi are not reactive enough to oxidize CP and DCP in the dark, it follows that CoPi has participated into the TiO2photocatalyzed oxidation of CP and DCP, respectively.


22

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Scheme 1. Possible mechanism for the cooperative effect of Pt and CoPi in KPi solution.

We propose that the CoIV species produced from the hole oxidation of CoII/III in CoPi are surface-bound and short-lived (Scheme 1). They can quickly react with the surface adsorbed substrate nearby, together with regeneration of CoPi on TiO2 in the presence of phosphate. In neutral aqueous solution, the one-electron reduction potentials of phenol, CP, and DCP are 1.03, 0.99, and 0.93 V vs. NHE, respectively. 9 The exact oxidation potential of CoIV is not known, but it would be more positive than that potential for water oxidation to O2, which is 0.82 V vs. NHE plus over-potential. Therefore, CoIV would be reactive toward these organics. Note that in the cyclic voltammograms of CoPi, deposited on a conducting electrode, there is only one reductive wave, attributed to the reduction of CoIII to CoII (Figure S10).30,31 Since CoPi is firmly immobilized on TiO2, it would capture more photoholes than does organic substrate (CP and DCP). As a result, the efficiency of the charge separation is improved, and the specific rates of CP and DCP oxidation are increased. In the presence of Pt, the electron reduction of O2 is accelerated. Then, the Pt-mediated electron transfer would promote the CoPi-mediated hole transfer, and vice visa. This would further improve the efficiency of the charge separation. As a


23

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 31

result, the specific rates of CP and DCP oxidation, obtained from CoPi/Pt/TiO2, are larger than the sum of individual rates, measured from Pt/TiO2 and CoPi/TiO2, respectively.. Different from CP and DCP, phenol poorly adsorbs on TiO2 in aqueous solution. Then phenol oxidation by the trapped holes of TiO2 (≡Ti−O−•) or by CoIV on the solid surface would be slow. Since the deposited CoPi on TiO2 inhibits the adsorption of O2 from aqueous solution, the electron recombination with ≡Ti−O−• and CoIV would be increased. As a result, the apparent rate of phenol oxidation on CoPi/TiO2 is much lower than that on TiO2. Since Pt can catalyze the multi-electron reduction of O2, the apparent rate of phenol oxidation on CoPi/Pt/TiO2 is not much lower than that on Pt/TiO2. On the other hand, after CoPi deposition, the specific rates of H2O2 decomposition over TiO2 without addition of organics are increased by 2.48 times. Recall that after CoPi loading the specific rates of CP and DCP oxidation over TiO2 are increased only by 1.03 and 1.04 times, respectively. This discrepancy between H2O2, CP and DCP is due to the fact that the decomposition of H2O2 over the irradiated TiO2 can occur through a reductive and oxidative pathway.26 That is, the electron reduction of H2O2 would be promoted by the CoPimediated hole oxidation of H2O2, and vice visa. But the CoPi-mediated hole oxidation of CP and DCP is somewhat balanced by decrease in the adsorption and reduction of O2. In comparison with O2, moreover, H2O2 highly adsorbs onto TiO2 in aqueous solution. This would favor the electron reduction of H2O2. To exploit the positive effect of CoPi, therefore, the reduction of O2 reduction should be improved, through a co-catalyst (Pt), or through an enrichment of O2 onto TiO2 from aqueous solution. Otherwise, CoIV species would recombine with the photoelectrons of TiO2. This would consume a pair of electrons and holes, and decrease the quantum efficiency of the charge carriers for O2 reduction and organic oxidation.


24

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

4. CONCLUSIONS It has been widely reported that the deposited CoPi on a semiconductor film photoanode can promote the hole oxidation of water to O2 in a phosphate solution, when the photoelectrons are externally transferred onto a cathode at an applied potential. In this work, we have demonstrated that the deposited CoPi on a powdered TiO2 can also facilitate the hole oxidation of organics in a phosphate-containing suspension, when the photoelectrons are internally transferred onto Pt sites for O2 reduction. However, this positive effect of CoPi operates only with a highly adsorptive substrate on TiO2 in aqueous solution (CP, DCP, and H2O2). Accordingly, we propose that the CoIV species formed from the hole oxidation of CoII/III in CoPi are surface-bound and short-lived. They quickly oxidize the surface adsorbed substrate nearby, but would recombine with the photoelectrons of TiO2 in the absence of efficient electron acceptor (Pt or H2O2). Such CoPimediated hole transfer would cooperate with the Pt-mediated electron transfer, further improving the efficiency of the charge separation. Moreover, we also propose that phosphate anions act as a repairer of CoPi on the surface of TiO2. Otherwise, the dissolved Co2+ ions in aqueous solution are detrimental to the photocatalytic reactions of TiO2 and Pt/TiO2. To maintain the activity of CoPi, it is recommended that the photocatalytic reaction may be performed in a non-aqueous solvent, and/or using precipitators such as borate.29 This information would be useful for organic oxygenation, environmental remediation, and (photo)electrochemical oxidation of water.

ASSOCIATED CONTENT Supporting Information


25

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 31

XRD patterns, N2 adsorption–desorption isotherms, absorption spectra, XPS data, kinetic fitting of organic oxidation, DCP adsorption isotherm, CoPi deposition method, Co2+ effect, and CV curves for a CoPi electrode. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Fax: +86-571-87951895. Tel: +86-571-87952410. Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by the Funds for Creative Research Group of NSFC (No. 21621005), and by the General Project of NSFC (No. 21377110).

REFERENCES (1)

Kanan, M. W.; Nocera, D. G. In situ Formation of an Oxygen-Evolving Catalyst in Neutral Water Containing Phosphate and Co2+. Science 2008, 321, 1072−1075.


26

Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(2)

Steinmiller E. M. P.; Choi, K. S. Photochemical Deposition of Cobalt-Based Oxygen Evolving Catalyst on a Semiconductor Photoanode for Solar Oxygen Production. Proc. Natl. Acad. Sci. USA. 2009, 106, 20633−20636.

(3)

Zhong D. K.; Cornuz, M.; Sivula, K.; Graetzel, M.; Gamelin, D. R. Photo-assisted Electrodepostion of Cobalt-Phosphate (Co−Pi) Catalyst on Hematite Photoanodes for Solar Water Oxidation. Energy Environ. Sci. 2011, 4, 1759−1764.

(4)

Seabold J. A.; Choi, K. S. Effect of a Cobalt-Based Oxygen Evolution Catalyst on the Stability and The Selectivity of Photo-Oxidation Reactions of a WO3 Photoanode. Chem. Mater. 2011, 23, 1105−1112.

(5)

Pilli S. K.; Furtak, T. E.; Brown, L. D.; Deutsch, T. G.; Turner, J. A.; Herring, A. M. Cobalt-Phosphate (Co-Pi) Catalysts Modified Mo-Doped BiVO4 Photoelectrodes for Solar Water Oxidation. Energy Environ. Sci 2011, 4, 5028−5034.

(6)

Khnayzer R. S.; Mara, M. W.; Huang, J.; Shelby, M. L.; Chen, L. X.; Castellano, F. N. Structure and Activity of Photochemically Deposited “CoPi” Oxygen Evolving Catalyst on Titania. ACS Catal. 2012, 2, 2150−2160.

(7)

Liu D. N.; Jing, L. Q.; Luan, P.; Tang, J. W.; Fu, H. G. Enhancement Effects of Cobalt Phosphate Modification on Activity for Photochemical Water Oxidation of TiO2 and Mechanism Insights. ACS Appl. Mater. Interfaces 2013, 5, 4046−4052.

(8)

Ai G.; Mo, R.; Li, H.; Zhong, J. Cobalt Phosphate Modified TiO2 Nanowire Arrays as Cocatalyst for Solar Water Splitting. Nanoscale 2015, 7, 6722−6728.


27

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(9)

Page 28 of 31

Li, C.; Hoffman, M. Z. One–Electron Redox Potentials of Phenols in Aqueous Solution. J. Phys. Chem. B 1999, 103, 6653–6656.

(10) Maldotti, A.; Molinari, A.; Amadelli, R. Photocatalysis with Organized Systems for the Oxofunctionlization of Hydrocarbons by O2. Chem. Rev. 2002, 101, 3811–3836. (11) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Environmental Application of Semiconductor Photocatalysis. Chem. Rev. 1995, 95, 69–96. (12) Emeline, A. V.; Zhang, X.; Jin, M.; Murakami, T.; Fujishima, A. Application of a “Black Body” Like Reactor for Measurements of Quantum Yields of Photochemical Reactions in Heterogeneous Systems. J. Phys. Chem. B 2006, 110, 7409–7413. (13) Ollis, D. F. Kinetics of Liquid Phase Photocatalyzed Reactions: An Illumination Approach. J. Phys. Chem. B. 2005, 109, 2439–2444. (14) Emeline, A. V.; Ryabchuk, V. K.; Serpone, N. “Dogmas and Misconceptions in Heterogeneous Photocatalysis. Some Enlightened Reflections.” J. Phys. Chem. B. 2005, 109, 18515–18521. (15) Ryu, J.; Choi, W. Substrate-Specific Photocatalytic Activities of TiO2 and Multiactivity Test For Water Treatment Application. Environ. Sci. Technol. 2008, 42, 294−300. (16) Xiong, X.; Zhang, X,; Xu, Y. Incorporative Effect of Pt and Na2CO3 on TiO2Photocatalyzed Degradation of Phenol in Water. J. Phys. Chem. C 2016, 120, 25689– 25696.


28

Page 29 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(17) Tachikawa T.; Zhang, P.; Bian, Z. F.; Majima, T. Efficient Charge Separation and photooxidation on Cobalt Phosphate-Loaded TiO2 Mesocrystal Superstructures. J. Mater. Chem. A 2014, 2, 3381−3388. (18) Hufschmidt, D.; Bahnemann, D.; Testa, J. J.; Emilio, C. A.; Litter, M. I. Enhancement of the Photocatalytic Activity of Various TiO2 Materials by Platinisation. J. Photochem. Photobiol. A 2002, 148, 223–231. (19) Lee, J.; Choi, W.; Photocatalytic Reactivity of Surface Platinized TiO2: Substrate Specificity and the Effect of Pt Oxidation State. J. Phys. Chem. C 2005, 109, 7399−7406. (20) Bader, H.; Sturzenegger, V.; Hoigné, J. Photometric Method for the Determination of Low Concentrations of Hydrogen Peroxide by the Peroxidase Catalyzed Oxidation of N,NDiethyl-p-phenylenediamine (DPD). Water Res. 1988, 22, 1109–1115. (21) Yu, J.; Li. Q.; Jaroniec, M. Hydrogen Production by Photocatalytic Water Splitting over Pt/TiO2 Nanosheets with Exposed (001) Facets. J. Phys. Chem. C 2010, 114, 13118– 13125. (22) Saveleva, V. A.; Papaefthimiou,V.; Daletou, M. K.; Doh, W. H.; Ulhaq-Bouillet, C.; Diebold, M.; Zafeiratos, S.; Savinova, E. R. Operando Near Ambient Pressure XPS (NAP-XPS) Study of the Pt Electrochemical Oxidation in H2O and H2O/O2 Ambients. J. Phys. Chem. C 2016, 120, 15930−15940. (23) Alexander, V.; Vorontsov, A. V.; Savinov, E. N.; Jin, Z. Influence of the Form of Photodeposited Platinum on Titania upon Its Photocatalytic Activity in CO and Acetone oxidation. J. Photochem. Photobiol. A 1999, 125, 113−117.


29

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

(24) McDonald, K. J.; Choi, K-S. Photodeposition of Co-Based Oxygen Evolution Catalysts on α-Fe2O3Photoanodes. Chem. Mater. 2011, 23, 1686–1693. (25) Lv, K.; Xu, Y. Effects of Polyoxometalate and Fluoride on Adsorption and Photocatalytic Degradation of Organic Dye X3B on TiO2: The Difference in the Production of Reactive Species. J. Phys. Chem. B 2006, 110, 6204–6212. (26) Sahel, K.; Elsellami, L.; Mirali, I.; Dappozze, F.; Bouhent, M.; Guillard, C. Hydrogen Peroxide and Photocatalysis. Appl. Catal. B 2016, 188, 106–112. (27) Di Paola, A.; Marcì, G.; Palmisano, L.; Schiavello, M.; Uosaki, K.; Ikeda, S.; Ohtani, B. Preparation of polycrystalline TiO2 photocatalysts impregnated with various transition metal ions: characterziation and photocatalytic activity for the degradation of 4nitrophenol, J. Phys. Chem. B 2002,106, 637−645. (28) Kim, J.; Lee, C. W.; Choi, W. Platinized WO3 as an Environmental Photocatalyst that Generates OH Radicals under Visible Light. Environ. Sci. Technol. 2010, 44, 6849–6857. (29) Kurosu, H.; Yoshida, M.; Mitsutomi, Y.; Onishi, S.; Abe, H.; Kondoh, H. In Situ Observations of Oxygen Evolution Cocatalysts on Photoelectrodes by X-ray Absorption Spectroscopy: Comparison between Cobalt-Phosphate and Cobalt-Borate. Electrochem. 2016, 84, 779–783. (30) Klingan, K.; Ringleb, F.; Zaharieva, I.; Heidkamp, Chernev, P.; Gonzalez−Flores, D.; Risch, M.; Fischer, A.; Dau, H. Water Oxidation by Amorphous Cobalt-Based Oxides: Volume Activity and Proton Transfer to Electrolyte Bases. ChemSusChem. 2014, 7, 1301– 1310.


30

Page 31 of 31

(31) Ma, Y.; Kafizas, A.; Pendlebury, S. R.; Le Formal, F.; Durrant, J. R. Photoinduced Absorption Spctroscopy of CoPi on BiVO4: The Function of CoPi during Water Oxidation. Adv. Funct. Mater. 2016, 26, 4951–4960.

Table of Contents Graphic Number of Manuscript: Authors: Xiao Zhang, Xianqiang Xiong, Lianghui Wan, and Yiming Xu* Title: Effect of a Co-based oxygen evolving catalyst on TiO2-photocatalyzed organic oxidation

Relative Oxidation Rate

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

TiO2 CoPi/TiO2 Pt/TiO2 CoPi/Pt/TiO2

phenol

4-CP

2,4-DCP


31

Effect of a Co-Based Oxygen-Evolving Catalyst on TiO2

Recommend Documents