Photocorrosion of Cuprous Oxide in Hydrogen Production: Rationalising Self-Oxidation or Self-Reduction

Article abstract of DOI:10.1002/anie.201807647

Cuprous Oxide (Cu₂O) is a photocatalyst with severe photocorrosion issues. Theoretically, it can undergo both self-oxidation (to form copper oxide (CuO)) and self-reduction (to form metallic copper (Cu)) upon illumination with the aid of photoexcited charges. There is, however, limited experimental understanding of the “dominant” photocorrosion pathway. Both photocorrosion modes can be regulated by tailoring the conditions of the photocatalytic reactions. Photooxidation of Cu₂O (in the form of a suspension system), accompanied by corroded morphology, is kinetically favourable and is the prevailing deactivation pathway. With knowledge of the dominant deactivation mode of Cu₂O, suppression of self-photooxidation together with enhancement in its overall photocatalytic performance can be achieved after a careful selection of sacrificial hole (h⁺) scavenger. In this way, stable hydrogen (H₂) production can be attained without the need for deposition of secondary components.

Full text of DOI:10.1002/anie.201807647

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Photocorrosion of Cuprous Oxide in Hydrogen Production:
Rationalising Self-Oxidation or Self-Reduction?
Authors: Cui Ying Toe, Zhaoke Zheng, Hao Wu, Jason Scott, Rose
Amal, and Yun Hau Ng
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201807647
Angew. Chem. 10.1002/ange.201807647
Link to VoR: http://dx.doi.org/10.1002/anie.201807647
http://dx.doi.org/10.1002/ange.201807647

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
Photocorrosion of Cuprous Oxide in Hydrogen Production:
Rationalising Self-Oxidation or Self-Reduction
Cui Ying Toe,[a] Zhaoke Zheng,[b] Hao Wu,[a] Jason Scott,[a] Rose Amal,[a] and Yun Hau
Ng*[a,c]
Abstract: Cuprous Oxide (Cu2O) is a photocatalyst with severe
photocorrosion issues. Theoretically, it can undergo both self-
oxidation (to form copper oxide (CuO)) and self-reduction (to form
metallic copper (Cu)) by the photoexcited charges upon illumination.
There is, however, not much experimental understanding in the
“dominant” photocorrosion pathway. In this work, both
photocorrosion modes can be regulated by designing the conditions
of the photocatalytic reactions. We demonstrate that photooxidation
of Cu2O (in the form of suspension system) accompanied with
destructed morphology is kinetically favourable and is the prevailing
deactivation pathway. By knowing the dominant deactivation mode
of Cu2O, suppression of self-photooxidation together with
enhancement in its overall photocatalytic performance can be
achieved after a careful selection of sacrificial hole (h+) scavenger
thus, attaining stable hydrogen (H2) production without the need for
deposition of secondary components.
photocatalyst, understanding the photostability issues is hence
important as an anticipative solution.
Numerous studies have shown that Cu2O were easily
oxidized into CuO under illumination. Huang et al. demonstrated
the distinct tendencies of various sizes Cu2O transforming into
CuO during photodegradation of methyl orange.[6] Kakuta and
Abe also observed a gradual deactivation of Cu2O in water
reduction due to self-photooxidation.[7] In contrast, some other
works have indicated the reduction of Cu2O to Cu under photon
excitation in the presence of excess electrons (e-).[8] Efforts in
overcoming the photoreduction have been formulated by Tran
and co-workers using rGO to enhance e- transport efficiency of
Cu2O.[9] Knowing that the redox potentials of Cu2O fall between
its conduction and valance band energies, self-oxidation and
reduction reactions are kinetically and thermodynamically
possible. Thus, it is essential to understand Cu2O dominant
photocorrosion pathway before undertaking any route to unfold it.
This creates an important research question of Cu2O
photocorrosion: “Self-photoreduction or self-photooxidation?”.
Although extensive works have been performed to stabilize
Cu2O, no comprehensive experiment was done to identify this
vital aspect: the dominant photocorrosion pathway.
Over the past decades, much effort has been devoted to
contrive strategies in tackling photocorrosion issues of
photocatalysts, with focus leaning towards sulphide-based
materials.[1] Nevertheless, with the development of oxide-based
photocatalysts, cuprous oxide (Cu2O) has been explored to
suffer from detrimental photostability problems. Cu2O is a p-type
photocatalyst (band gap of 2.0 - 2.3eV)[2] that possesses high
carrier mobility (60 - 130 cm2V-1s-1).[3] Its energetic conduction
band brings extra merit for photocatalytic H2 generation (water
reduction reaction).[4] Although these superior properties of Cu2O
have been of great interest to researchers, the transformations
of Cu2O to CuO and Cu are still unavoidable due to the
thermodynamically feasible parasitic reactions of self-oxidation
and self-reduction.[5] Photocorrosion of Cu2O has significantly
hindered its practicability in long term applications. Of the
With studies consistently revealing the inferior photoactivity
of cubic Cu2O, it has been demonstrated as the most unstable
morphology with unstable {100} facet.[10] Nonetheless, cubic
Cu2O can perform better in catalytic Cr(VI) reduction when
illumination was omitted.[11] Moreover, a high initial photocurrent
of 5.5 mAcm-2 exhibited by {100}-dominated Cu2O,[12] also
indicates the great photocatalytic potential of cubic Cu2O if
photocorrosion is suppressed. Secondly, spatial charge
separation has been observed in anisotropic faceted Cu2O
(multiple exposed facets) during co-catalyst deposition due to
the difference in surface potential.[13] To prevent the introduction
of additional complexity, isotropic faceted cubic Cu2O is used to
rule out the charge segregation effect promoted by the presence
of different facets. Although photoelectrochemical stability has
been previously investigated,[14] utilization of cubic Cu2O
suspension system could enable the examination of intrinsic
Cu2O photocorrosion without external influences. A systematic
study in several different conditions (i.e. the presence and/or
absence of e- and h+ scavengers) has been initiated to
investigate Cu2O dominant photocorrosion pathway. With a
comprehensive understanding, the performance was further
enhanced through the evaluation of photocatalytic H2 generation.
primary importance in developing Cu2O as
a promising
[a]
C. Y. Toe, H. Wu, Assoc. Prof. J. Scott, Prof. R. Amal, Dr. Y. H. Ng
Particles and Catalysis Research Group
School of Chemical Engineering
The University of New South Wales, NSW 2052, Australia
E-mail: yh.ng@unsw.edu.au
[b]
[c]
Prof. Z. Zheng
State Key Laboratory of Crystal Materials
Shandong University, Jinan 250100, China
Dr. Y. H. Ng
School of Energy and Environment, City University of Hong Kong,
Kowloon, Hong Kong SAR, P. R. China
First, cubic Cu2O was fabricated via a solution-phase route
as described in the Supporting Information (SI).[15] The cubic
morphology of the Cu2O sample was examined under scanning
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
electron microscopy (SEM, Figure 1a). Well-developed particles
with discernible edges were observed (inset of Figure 1a) and
explicitly uniform cube with an average edge length of 362 ± 56
nm were seen from Figure 1b. X-ray powder diffraction (XRD)
analysis was performed to investigate the crystal phase of the
sample. Development of distinct peaks readily indexed to cubic
crystalline Cu2O suggests the presence of pure Cu2O (Figure
1c). Selected area diffraction pattern (SAED, inset of Figure 1c)
was shown to have a crystal arrangement that possesses
properties of single crystal. Hereafter, the absorption band edge
was probed using UV-Vis diffuse reflectance spectroscopy
(Figure 1d). Its corresponding Tauc plot (inset Figure 1d)
displays a band gap of 2.10 eV, associating to a typical Cu2O
band gap.[2] The surface composition was also characterized
using X-ray photoelectron spectroscopy (XPS). Deconvolution of
Cu 2p3/2 spectrum (Figure S2a) demonstrated that the
dominated Cu species is Cu+ (932.1 eV), originated from
Cu2O.[16] The successful formation of Cu2O was also justified by
its Cu LMM spectrum (Figure S2b) exhibited only single Cu+
peak at 916.8 eV.[17]
Table 1. 5 h-illumination in different conditions for Cu2O photostability test.
Conditions
Purge with
Argon Air
Solution
Scavenging
System
Pure
10 v/v%
ethanol
water
1
2
3
4
-
-


-

-
-

-
e-
Both
None
h+



-
-

XRD patterns in Figure 2a display the phases of post-
illuminated Cu2O samples in respective conditions. Apparent
formation of CuO was observed for samples illuminated in the
absence of h+ scavenger (condition 1 and 3); whereas the phase
of Cu2O remained unchanged for the other two samples
illuminated with ethanol as h+ scavenger (Figure 2a). To further
understand the photocorrosion mechanism, morphology
changes of post-illuminated samples were monitored. Noticeably,
Figure 2b.1 depicts obvious deformation of cubic morphologies
after illuminating in condition 1 (without h+ scavenger but with e-
scavenger). This sample shows a more severe transformation
when compared to sample illuminated in a non-scavenging
system (Figure 2b.3), due to the presence of O2 that boosted the
rate of e- withdrawal, leaving significant amount of h+ for self-
photooxidation. By periodically monitoring the morphology
changes in condition 1, Figure S3 manifested the evolution of
cubic Cu2O into irregular CuO clusters. The gradual growth of
CuO causes a total destruction of its cubic morphologies after 5
h. This is credited to the presence of e- scavenger, where
photoinduced e- was scavenged by O2 and the better charge
separation yields higher amount of free h+ for self-photooxidation.
40
(a)
(b)
35
30
25
20
15
10
5
0
150 200 250 300 350 400 450 500 550
1 μm
Size (nm)
(c)
(d)
(111)
(001)
(111)
2
1/nm
1.8 2.0 2.2 2.4 2.6
Careful examination of the SEM images for samples
hѵ (eV)
(200)
illuminated in condition
3 (Figure 2b.3) reveals obvious
(220)
(311)
(110)
dissolution of Cu2O from the centre of the particles, which is
associated to the instability of (100) facets in comparison with its
edges. In this instance, photoinduced e- are postulated to
undertake two different pathways: (a) recombination with the
photogenerated h+ and (b) consumed for the reduction of
dissolved Cu2+ ions (formed via self-oxidation) to Cu2O. When
the sample experienced loss of h+ through charge recombination,
rate of photocorrosion will be suppressed leading to a less
severe shape deformation when compared to condition 1. For
the second pathway, reduction of Cu2+ into Cu2O was validated
by HRTEM images (Figure S4), illustrating Cu2O (111)
characteristic lattice (2.46 Å) on the outer layer of the particle.[19]
20
30
40
50
2θ
60
70
80
400
500
600
700
800
Wavelength (nm)
Figure 1 (a) SEM image (inset: scale bar of 100nm), (b) size distribution, (c)
XRD pattern (inset: SAED pattern) and (d) UV-Vis absorption spectrum with
inset its corresponding Tauc plot of cubic Cu2O.
Since the performance of Cu2O is always restricted by its
photostability issues: self-oxidation into CuO [Eq. (S1)] or
reduction into Cu [Eq. (S2)], herein, Cu2O was irradiated in
several circumstances, utilizing e- scavenger (i.e. O2 in air) and
h+ scavenger (i.e. ethanol). It was known that the former
efficiently scavenges e- for O2 reduction reaction while the latter
is a well-established h+ scavenger that donates e- for self-
decomposition.[18] Specifically, illuminations of were performed in
four distinct conditions (i.e. non-scavenging system, system in
the presence of only e- or h+ scavenger and the presence of both
scavengers) as tabulated in Table 1.
This article is protected by copyright. All rights reserved.

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
measurement will also favour Cu2O self-reduction when e- is
flown to the p-type working photocathode.
(a)
Cu2O
CuO
Before
As verification to the above hypothesis, HRTEM and XPS
were performed to further investigate the composition of
samples illuminated in the presence of e- scavenger (i.e.
condition 1 and 2). TEM images (Figure 3a) again shows that
the sample illuminated in condition 1 was completely deformed.
Its surface lattice fringes are found to be the fingerprints of
crystalline CuO (002) with 2.52 Å lattice space (inset Figure
3a).[20] Evidently, the corresponding XPS spectrum (Figure 3b)
elucidates intensified peaks of Cu2+ at binding energies of 934.0
and 935.2 eV when compared to the fresh Cu2O. Conversely,
when cubic Cu2O was illuminated in the presence of both
scavengers (condition 2), its cubic shape was maintained
(Figure 3c) with HRTEM image (inset of Figure 3c) showing the
lattice space (3.01 Å) of Cu2O (110).[21] Likewise, its XPS
spectrum (Figure 3d) portrays the unchanged surface
composition, associating to the obtained XRD and SEM results.
These findings thus signify that regardless the presence of e-
scavenger, detrimental transformation of Cu2O into CuO will
occur in the absence of h+ scavenger. This indicates that Cu2O
self-photooxidation is more pronounced than the self-reduction
for a photocatalytic bias-free system. To eliminate the possible
contribution of highly reactive species (i.e. H2O2, superoxide and
hydroxyl radicals) towards Cu2O self-photooxidation, irradiations
were also performed in presence of AgNO3 and coumarin, as
elaborated in the SI (Figure S6).[22] Apart from the illuminated
systems, a preserved Cu2O phase after 5 h-dark reaction was
revealed (Figure S7), verifying that the corrosion of Cu2O into
CuO is attributed to photon illumination when h+ are
accumulated within the samples.
Air + water
1
2
Air + 10 v/v% ethanol
Argon + water
3
4
Argon + 10 v/v% ethanol
20
30
40
50
60
70
80
2θ
(b)
Absence of h+ scavenger Presence of h+ scavenger
1 2
100 nm
100 nm
3
4
100 nm
100 nm
Figure 2 (a) XRD patterns and (b) SEM images of samples illuminated in 1.
pure water purged with air (presence of e- scavenger), 2. 10 v/v% ethanol
purged with air (presence of both scavengers), 3. pure water purged with
argon (absence of both scavengers) and 4. 10 v/v% ethanol purged with argon
(presence of h+ scavenger).
Absence of hole scavenger (Condition 1)
Cu₂O
(a)
(b)
2.52Å
CuO
2nm
Cu(OH)₂
On the contrary, the severity of Cu2O self-photoreduction
was also monitored by investigating the Cu metal development,
specifically for samples illuminated in the absence of e-
scavenger. Surprisingly, Cu phase diffraction peaks was not
detected from all the XRD patterns (Figure 2a) despite the lack
of driving force for e- withdrawal from the e- sacrificial agent (i.e.
condition 3 and 4). This was further supported by XPS Cu LMM
auger spectrum where Cu0 peak (918.6 eV) was not detected
(Figure S5), manifesting the lower tendency for Cu2O to be
kinetically reduced into Cu metal in contrast to the self-oxidation
reaction. On the basis of all the characterizations above, it is
established that Cu2O self-oxidation into CuO is the dominant
photocorrosion pathway. Contradiction of this finding when
compared to some previous literatures that reported the self-
photoreduction of Cu2O can be explained by the existence of
external charge transfer influencing factors. For instance, Cu
formation was observed dominantly from the illumination of
anisotropic faceted Cu2O or heterojunction composite, whereby
internal bias is created due to the distinct band energies within
the particle.[8] Additionally, external applied bias for PEC
938
936
934
932
930
100nm
Binding Energy (eV)
Presence of hole scavenger (Condition 2)
(c)
(d)
Cu₂O
3.01Å
2nm
Cu(OH)₂
CuO
938
936
934
932
930
50nm
Binding Energy (eV)
Figure 3 (a) TEM image (inset: HRTEM), (b) XPS spectrum of sample
illuminated in condition 1, (c) TEM image (inset: HRTEM) and (d) XPS
spectrum of sample illuminated in condition 2.
This article is protected by copyright. All rights reserved.

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
Arising from the rapid e- transfer for water reduction reaction
(sufficient over-potential of Cu2O conduction band), h+ was
easily accumulated, where the gathered h+ is the origin of severe
Cu2O photocorrosion. This is associated to the relatively lower
strength of Cu2O valance band that is insufficient for h+ to
participate in water oxidation (according to the experimentally
constructed energy diagram, Figure 4a). The inefficient
withdrawal of h+ by adsorbed water molecules will again prompt
Cu2O self-oxidation. Congruously, the presence of h+ scavenger
with suitable oxidizing potential would aid the h+ transfer, thus
improves the photostability of Cu2O. Although overall water
splitting using Cu2O has been reported,[23] the above findings
revealed that Cu2O is not an ideal candidate for water oxidation
but it emerges as an excellent photocatalyst for H2 evolution due
to its strong reducing potential.
generation for 15 h from Cu2O in Na2SO3 solution and its corresponding (d)
XRD pattern after reaction (inset: its SEM image).
Apart from the important role of h+ scavenger, its relative
redox strength (Figure 4a) is another essential aspect to be
considered since the scavenging ability determines the extent of
improvement that can be attained.[24] The favourable redox
2-
potential of SO3 leads to the enhanced h+ transport and thus
increases the amount of free e- for photoreduction. To the best
of our knowledge, this system is among the best when
compared to other reported Cu2O systems without the addition
of metals/co-catalysts (details are described in SI, Table S2).[10a,
25]
Conversely, ethanol that has
a lower redox potential
demonstrated
a
diminution in h+ scavenging ability
corresponding to its lower photoactivity. The phenomenon is
also proven by the utilization of other sacrificial h+ scavengers
(i.e. methanol and glycerol) as described in the SI (Figure S10).
With the withdrawal of photogenerated h+ on a time scale faster
than Cu2O self-oxidation, its photostability could be promoted.
As corroboration, H2 generation in Na2SO3 was performed for 15
h with two intermediate evacuations (Figure 4c). The constant H2
generation rate throughout 3 continuous cycles infers a stable
cubic Cu2O system in Na2SO3 solution. The Cu2O diffraction
peaks and apparent edges of cubic shape observed from the
XRD pattern (Figure 4d) and SEM image (inset Figure 4d)
conjointly demonstrate the excellent stability of Cu2O in Na2SO3.
From the view of practicability, utilizing Cu2O in water
splitting half reaction (H2 evolution) should be accompanied by
the oxidation of h+ scavenger, whereby, it would be beneficial if
these h+ scavengers are from the category of organic or
inorganic pollutants. Figure 4b displays the performance of
photocatalytic H2 evolution in pure water, presence of 10 v/v%
ethanol and 0.5 M Na2SO3 solution. Highest amount of H2 gas
was generated when Na2SO3 was used as the scavenger (23
µmol), which surpasses that of ethanol by 12 times (1.9 µmol).
By capturing photoexcited h+, ethanol was oxidized into CO2.
Alternatively, Na2SO3 enabled the scavenging of h+ to oxidize
2-
SO3 into SO42-, leaving free e- in the conduction band for H2
In short, there are two main findings in this work. The first
part demonstrated that accumulation of photoinduced h+ is the
dominant reason that causes Cu2O photocorrosion (i.e.
photooxidation). Irradiation in a few distinct conditions revealed
a noticeable transformation of Cu2O into CuO in the absence of
h+ scavenger, while Cu was not formed despite the absence of
e- scavenger. This is attributed to the higher e- consumption rate
for water reduction when compared to h+ that are incapable of
oxidizing water. As the photocorrosion issue was suppressed by
h+ scavenger, the second finding demonstrates Na2SO3 as a
evolution reaction. Details of the products from sulphite and
ethanol oxidation were shown in the SI (Figure S8). In contrast,
no H2 was detected when it was illuminated in water owing to the
depletion of e- via rapid electron-hole pair recombination, which
was contributed by the accumulation of h+ within the particle.
(a)
(b)
-1.5
-1.0
-0.5
0
Na2SO3
Ethanol
Water
CB
2–
SO42–/SO3
Cu2O/Cu
H+/H2
better h+ scavenger that enabled enhanced H2 evolution
CO2/CH3CH2OH
CuO/Cu2O
O2/H2O
26]
performance. Unlike other literatures,[10a,
this work can be
0.5
1.0
1.5
served as the highlight for stable H2 production from metal/co-
catalyst-free Cu2O system. Since accumulation of h+ within Cu2O
was shown to be detrimental, the efficient removal of h+ by
forming a z-scheme system with an excellent O2 evolving
photocatalyst (e.g. BiVO4, WO3 etc.) could potentially lead Cu2O
to a greater use in photocatalytic water splitting.[27]
VB
(c)
(d)
200 nm
Acknowledgements
This work was financially supported by the Australian Research
Council Discovery Project (DP180102540). The authors thank
the support provided by the UNSW Mark Wainwright Analytical
Centre.
20
30
40
50
2θ
60
70
80
Figure 4 (a) Energy diagram showing the conduction and valance band of
Cu2O (detailed experimental analyses to determine the band structure are
shown in the SI, Figure S9 and Table S1) with redox potentials of relevant
reactions, (b) 5 h of H2 generation with water, ethanol and Na2SO3, (c) H2
Keywords: Photocatalysis • Photocorrosion • Cuprous Oxide •
Photooxidation
This article is protected by copyright. All rights reserved.

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
[1]
a) L. Jia, D.-H. Wang, Y.-X. Huang, A.-W. Xu, H.-Q. Yu, J. Phys. Chem.
C 2011, 115, 11466-11473; b) H. Matsumoto, T. Sakata, H. Mori, H.
Yoneyama, J. Phys. Chem. 1996, 100, 13781-13785.
[2]
[3]
W. Ching, Y.-N. Xu, K. Wong, Phys. Rev. B 1989, 40, 7684-7695.
a) S. Ishizuka, T. Maruyama, K. Akimoto, Jpn. J. Appl. Phys. 2000, 39,
L786; b) F. F. Abdi, S. P. Berglund, J. Phys. D 2017, 50, 193002.
H. Gao, J. Zhang, R. Wang, M. Wang, Appl. Catal., B 2015, 172, 1-6.
J. A. Altermatt, S. E. Manahan, Inorg. Nucl. Chem. Lett. 1968, 4, 1-4.
L. Huang, F. Peng, H. Yu, H. Wang, Solid State Sci. 2009, 11, 129-138.
S. Kakuta, T. Abe, Electrochem. Solid-State Lett. 2009, 12, P1-P3.
a) C.-M. Wang, C.-Y. Wang, J. Nanophotonics 2014, 8, 084095-
084095; b) Z. Xi, C. Li, L. Zhang, M. Xing, J. Zhang, Int. J. Hydrogen
Energy 2014, 39, 6345-6353.
[4]
[5]
[6]
[7]
[8]
[9]
P. D. Tran, S. K. Batabyal, S. S. Pramana, J. Barber, L. H. Wong, S. C.
J. Loo, Nanoscale 2012, 4, 3875-3878.
[10] a) Y. Kwon, A. Soon, H. Han, H. Lee, J. Mater. Chem. A 2015, 3, 156-
162; b) Z. Zheng, B. Huang, Z. Wang, M. Guo, X. Qin, X. Zhang, P.
Wang, Y. Dai, J. Phys. Chem. C 2009, 113, 14448-14453.
[11] A. K. Mishra, D. Pradhan, Cryst. Growth Des. 2016, 16, 3688-3698.
[12] Y. Li, X. Yun, H. Chen, W. Zhang, Y. Li, Phys. Chem. Chem. Phys.
2016, 18, 7023-7026.
[13] R. Li, X. Tao, R. Chen, F. Fan, C. Li, Chem. - Eur. J. 2015, 21, 14337-
14341.
[14] X. Chang, T. Wang, P. Zhang, Y. Wei, J. Zhao, J. Gong, Angew. Chem.,
Int. Ed. 2016, 55, 8840-8845.
[15] X. Liang, L. Gao, S. Yang, J. Sun, Adv. Mater. 2009, 21, 2068-2071.
[16] Z. Zhang, S.-S. Wang, R. Song, T. Cao, L. Luo, X. Chen, Y. Gao, J. Lu,
W.-X. Li, W. Huang, Nat. Commun. 2017, 8, 488.
[17] a) S. Poulston, P. Parlett, P. Stone, M. Bowker, Surf. Interface Anal.
1996, 24, 811-820; b) D. Tahir, S. Tougaard, J. Phys.: Condens. Matter
2012, 24, 175002.
[18] a) T. Tan, D. Beydoun, R. Amal, J. Photochem. Photobiol., A 2003, 159,
273-280; b) J. Schwitzgebel, J. G. Ekerdt, H. Gerischer, A. Heller, J.
Phys. Chem. 1995, 99, 5633-5638.
[19] W. Z. Wang, G. Wang, X. S. Wang, Y. Zhan, Y. Liu, C. L. Zheng, Adv.
Mater. 2002, 14, 67-69.
[20] Y. Guo, Y.-T. Xu, B. Zhao, T. Wang, K. Zhang, M. M. Yuen, X.-Z. Fu, R.
Sun, C.-P. Wong, J. Mater. Chem. A 2015, 3, 13653-13661.
[21] C. Lu, L. Qi, J. Yang, X. Wang, D. Zhang, J. Xie, J. Ma, Adv. Mater.
2005, 17, 2562-2567.
[22] a) V. Cristino, S. Marinello, A. Molinari, S. Caramori, S. Carli, R.
Boaretto, R. Argazzi, L. Meda, C. A. Bignozzi, J. Mater. Chem. A 2016,
4, 2995-3006; b) C. Y. Toe, H. L. Tan, C. Boyer, A. Rawal, S. C.
Thickett, J. Scott, R. Amal, Y. H. Ng, J. Mater. Chem. A 2017, 5, 4568-
4575.
[23] J. Kondo, Chem. Commun. 1998, 357-358.
[24] a) T. J. Bruno, P. D. Svoronos, CRC handbook of basic tables for
chemical analysis, CRC press, 2003; b) C. Lamy, E. Belgsir, J. Leger, J.
Appl. Electrochem. 2001, 31, 799-809.
[25] G. Wang, R. van den Berg, C. de Mello Donega, K. P. de Jong, P. E.
de Jongh, Appl. Catal., B 2016, 192, 199-207.
[26] C. G. Morales-Guio, L. Liardet, M. T. Mayer, S. D. Tilley, M. Grätzel, X.
Hu, Angew. Chem., Int. Ed. 2015, 54, 664-667.
[27] a) K. Iwashina, A. Iwase, Y. H. Ng, R. Amal, A. Kudo, J. Am. Chem.
Soc. 2015, 137, 604-607; b) Y. Sasaki, A. Iwase, H. Kato, A. Kudo, J.
Catal. 2008, 259, 133-137.
This article is protected by copyright. All rights reserved.

10.1002/anie.201807647
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
The two possible photocorrosion
pathways of photoexcited Cu2O are
self-photoreduction into Cu metal or
self-photooxidation into CuO. A
systematic approach used in this
study has demonstrated that
photocorrosion of Cu2O photocatalyst
is dominated by self-photooxidation
via h+ consumption.
Cui Ying Toe, Zhaoke Zheng, Wu Hao,
Jason Scott, Rose Amal and Yun Hau
Ng*
Page No. – Page No.
Self-photooxidation
Photocorrosion of Cuprous Oxide in
Hydrogen Production: Rationalising
Self-Oxidation or Self-Reduction.
100 nm
Self-photoreduction
This article is protected by copyright. All rights reserved.