Facet-dependent gas sensing properties of Cu2O crystals

Article abstract of DOI:10.1002/pssa.201600904

Well-defined polyhedral Cu₂O exposing different crystal facets are synthesized via a facile solution-phase method, and the facet-dependent gas sensing properties are investigated. The results show that cubic Cu₂O exposing (100) facets has higher sensitivity to NO₂, acetone and benzene than octahedral Cu₂O with (111) facets. More importantly, the cubic Cu₂O has high selectivity to NO₂ and acetone at varied working temperatures by different sensing mechanisms of adsorption, chemical reaction, and intrinsic carrier excitation, as verified by the catalytic and electric property measurements. The higher sensitivity and selectivity of cubic Cu₂O is explained by the relatively lower work function of (100) surfaces than that of (111), which induces a thicker surface hole accumulation layer and thus, a higher bulk hole activation energy (0.55 eV and 0.36 eV respectively). The synergistic effect of catalytic and electric properties gives rise to gas sensitivity decrease with the reduced percentage of (111) facets, which is beneficial to the gas sensing material design and also, is promising for the future highly sensitive and selective gas sensor fabrication.

Full text of DOI:10.1002/pssa.201600904

Phys. Status Solidi A, 1600904 (2017) / DOI 10.1002/pssa.201600904
a
www.pss-a.com
applications and materials science
Facet-dependent gas sensing
properties of Cu O crystals
2
1
,2
*,2,3
2
2
2
2
2
Zheng Xie , Ning Han , Wenhui Li , Yuzhou Deng , Shuyan Gong , Ying Wang , Xiaofeng Wu ,
,
1,2
2,3
*
*
Yingxia Li
, and Yunfa Chen
1
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, North Third Ring Road 15,
00029 Beijing, P.R. China
State Key Laboratory of Multiphase Complex Systems, Institute of Process Engineering, Chinese Academy of Sciences, North Second
Street 1, 100190 Beijing, P.R. China
Center for Excellence in Regional Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Jimei
Avenue 1799, 361021 Xiamen, P.R. China
1
2
3
Received 28 November 2016, revised 6 February 2017, accepted 6 February 2017
Published online 28 February 2017
Keywords catalysis, crystal facets, Cu_2O, gas sensing, surfaces, work function
*
Corresponding author: e-mail nhan@ipe.ac.cn, Phone:þ86-10-82544896, Fax: þ86-10-62525716
*
*
e-mail liyx@mail.buct.edu.cn, Phone/Fax:þ86-10-64419619
2
Well-defined polyhedral Cu O exposing different crystal effect of catalytic and electric properties gives rise to gas
facets are synthesized via a facile solution-phase method, sensitivity decrease with the reduced percentage of (111)
and the facet-dependent gas sensing properties are investi- facets, which is beneficial to the gas sensing material design
gated. The results show that cubic Cu
has higher sensitivity to NO , acetone and benzene than selective gas sensor fabrication.
octahedral Cu O with (111) facets. More importantly, the
cubic Cu O has high selectivity to NO and acetone at varied
2
O exposing (100) facets and also, is promising for the future highly sensitive and
2
2
2
2
working temperatures by different sensing mechanisms of
adsorption, chemical reaction, and intrinsic carrier excitation,
as verified by the catalytic and electric property measurements.
The higher sensitivity and selectivity of cubic Cu O is
2
explained by the relatively lower work function of (100)
surfaces than that of (111), which induces a thicker surface
hole accumulation layer and thus, a higher bulk hole activation
energy (0.55 eV and 0.36 eV respectively). The synergistic
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1
Introduction In recent years, indoor air pollution example, hierarchical structure design, doping by other
has aroused high attention due to its huge damage to atoms, and forming heterojunctions are all effective at
people’s health, such as headache [1, 2], fatigue [3], sick enhancing the sensitivity of the oxide sensing material (e.g.,
building syndrome [4, 5], and even cancer [6]. Metal oxide SnO and ZnO) by enhancing the surface interaction of the
2
semiconductor (MOX) gas sensor is believed to be one of gas and the sensing materials [8–10].
the most feasible detection methods because of its real-time
On the other hand, the surface properties are also highly
detection and low cost [7]. As there are stringent permissive dependent on the exposed surface planes due to the
levels of the air pollutant, it is necessary for the MOX difference in atom numbers and arrangement. Facet-
sensors to have high sensitivity. After years of development, dependent physical and chemical properties, such as
there are several strategies established to design and electronic [11–13] and catalytic [14–16] properties, of
synthesize highly sensitive MOX materials, with an aim polyhedral inorganic nanocrystals have caused widespread
to tailor the surface properties of the sensing materials. For concern in recent years. Taking polyhedral Cu O for
2
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
1
600904 (2 of 8)
2
Z. Xie et al.: Facet-dependent gas sensing properties of Cu O crystals
example, the (100) facets on cubes are reported to have Future experiments were also conducted at the same
lower electron work function than the (111) ones on conditions except for changing the mass of PVP (0.2, 0.6,
octahedra [17, 18], while photocatalytic oxidation activity 1.0, and 3.0 g for C, TC, TO, O-shaped Cu O).
2
of (111) is higher than (100) due to the unique copper
atom alignments [19, 20]. Though Cu O is also widely
2.2 Characterization The crystalline phase was
2
used as gas sensors because of its intrinsic p-type property, determined by X-ray powder diffraction (XRD, Panalytical
such as Cu/Cu O hollow spheres [21], CuO/Cu O composite X’pert Pro, The Netherlands, Cu Ka radiation of
2
2
nanowires [22], porous Cu O/CuO cages [23], Cu O nano- l ¼ 0.15406 nm, 40 kV, 30 mA). The morphologies and
2
2
cages [24], the relationship of the facet-dependent physical microstructures were characterized by a field-emission
and chemical properties is not well investigated, though it is scanning electron microscope (FE-SEM, JEOL JSM-6700F,
essential because gas sensing property is related to electrical Japan 5 kV, 10 mA) and a transmission electron microscopy
resistance change caused by surface chemical catalytic (TEM, JEOL JEM-2010F, Japan, 200 kV, 100 mA). UV-Vis
reaction.
absorption spectra were recorded using a spectrophotometer
In the literature, ZnO sensing materials with different (Persee, TU-1901, China).
exposed planes have varied interaction properties with both
The gas sensing properties of Cu O with four
2
NO2 and ethanol and thus, have different sensing morphologies were examined by using a homemade tube-
performances [25, 26]. In this study, (100) and (111) furnace sensor test instrument, similar to our previous
exposed well-shaped cubes (C), truncated cubes (TC), report [28, 29]. Briefly, two Pt wires (0.5 mm ꢁ 1 m) were
2
truncated octahedra (TO), and octahedra (O) are synthe- fixed on an Al O substrate (1 ꢁ 2 cm ) by using a certain
2
3
sized, and the gas sensing properties are investigated in amount of Ag paste (Wuhan Youle Optoelectronics
detail taking typical air pollutants NO , acetone and benzene Technology Co., Ltd., China), which was annealed at
2
as probes. The results show that C-Cu O has the highest 550 8C for 30 min in the tube furnace to make good
2
sensitivity to NO , acetone and benzene compared with adhesion. After that, 20 mg as-synthesized Cu O was
2
2
other shapes, due to the different adsorption and catalytic ultrasonically dispersed into 2 mL ethanol and drop-coated
properties of the facets. And more importantly, the gas on the Al O substrate to form a sensor prototype. Then the
2
3
sensing maxima are obtained at distinguishingly different sensor was placed in the tube furnace and aged at 200 8C for
working temperatures showing the good selectivity to NO2 24 h to ensure good ohmic contact (Fig. S1). The targeted
and acetone. These results highlight the compromise of the gases with a series of concentrations from 1 to 10 ppm for
facet-dependent physical and chemical properties, and are acetone and benzene, 0.5–5 ppm for NO were mixed with
2
promising for gas sensor developments.
synthetic air (21.0 vol.% O , 79.0 vol.% N ) and standard
2 2
gas (49.8 ppm for benzene, 52.4 ppm for acetone, 48.4 ppm
2
2
Experimental
for NO in synthetic air, respectively, Beijing Hua Yuan Gas
2
.1 Synthesis of polyhedral Cu O Cu O with Chemical Industry Co., Ltd., China) controlled by two
2
2
four morphologies were synthesized according to the digital mass flow controllers at a total flow rate of
ꢂ1
literature [27] via a facile solution-phase method of 300 mL ꢀ min . The bias on the sensor was 5 V and the
analytical purity Cu SO ꢀ 5H O (Xilong Chemical current was recorded by using Keithley 2601 Sourcemeter
2
4
2
Reagent Co. Ltd., China), C H Na O ꢀ 2H O, Na CO , (Keithley Instrument Inc., USA) with accuracy of <ꢃ0.1%.
6
5
3
7
2
2
3
C H O ꢀ H O (Beijing Chemical Reagent Co. Ltd., China), The sensor responses to benzene and acetone are defined as
6
12
6
2
and PVP (Sigma–Aldrich Co., USA). In a typical procedure, the ratio of sensor resistance in target gas and in air (Rg/Ra),
1
1
mL 0.68 M Cu SO ꢀ 5H O solution was diluted into while that for NO is defined contrarily (Ra/Rg).
2
4
2
2
7 mL distilled water, and then 0.2 g PVP (K-30,
The catalytic activity evaluation of C-Cu O and
2
Mw ¼ 40000) was added into the solution. After being O-Cu O was carried out in a continuous-flow fixed-bed
2
stirred with a magnetic blender for about 15–20 min at quartz microreactor at a high space velocity (SV) of
ꢂ1 ꢂ1
room temperature, a clear light blue solution was 60,000 mL ꢀ g ꢀ h
.
The Cu O catalyst (50 mg,
2
obtained naturally. In the meantime, 1 mL 0.74 M 40–60 mesh) was mixed with quartz sands (250 mg,
C H Na O ꢀ 2H O and 1 mL 1.2 M Na CO were mixed 40–60 mesh) at a ratio of 1:5 and loaded in the quartz
6
5
3
7
2
2
3
evenly, which was then dropped into the clear light blue microreactor with quartz wool packed at both ends of the
solution slowly under stirring constantly. The light blue catalyst bed. Typically, the target gas with a composition of
turbid emerged at first stage and then transferred into clear 50 ppm acetone or benzene in synthetic air was introduced
ꢂ1
dark blue solution, with no precipitate observed eventually. into the reactor at a flow rate of 50 mL ꢀ min . A gas
After about 10 min, 1 mL 1.4 M C H O ꢀ H O was slowly chromatograph (Shimadzu GC-2014) equipped with a flame
6
12
6
2
dropped into the solution, which was then kept in a water ionization detector (FID) was used to analyse the final
bath at a temperature of 80 8C for 2 h, and then cooled down products in the effluent gas. The concentration of CO in the
2
to room temperature naturally. The brick red precipitate was outlet gas was detected by another FID with a Ni conversion
washed several times with distilled water and absolute furnace for converting CO to CH . The conversion of
2
4
alcohol, and separated out by centrifugation at 4000 rpm for acetone or benzene was calculated as (C ꢂ C )/C , and
in
out
in
1
0 min, and finally dried in a vacuum oven at 80 8C for 6 h. CO yield was calculated as C /(C ꢁ n), where C and
2
CO2
in
in
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A (2017)
(3 of 8) 1600904
Cout are the concentrations in the inlet and outlet, and n is the hexagons in TEM as shown in Fig. 1b inferring that the
carbon numbers of acetone or benzene molecule.
products align along the (111) direction normal to the
substrate [27]. The corresponding SAED pattern indicates
that they are single-crystal structure (Fig. 1b, insert). Further
3
3
Results and discussion
.1 Morphology evolution and characterization increasing PVP amount to 1.0 g, TO-Cu O was obtained as
2
of Cu O Generally, the kinetics and thermodynamics have shown in Fig. S2c, the six corners of the particles look like
2
been known as the key factors in controlling the crystal being cut a bit from a octahedral and thus lead to form a
growth [30], the crystals with different morphologies can be truncated octahedral with six (100) facets and eight (111)
obtained by adjusting the experimental parameters, such as facets. It is noteworthy that the sizes of the particles increase
reaction temperature [31], reaction time [20], concentration from 1.03 mm of TC-Cu O to 1.31 mm, and meanwhile, the
2
of the reactants [32] or surfactant [33], and so on. In this particles still have a narrow size distribution. A dodecagon
paper, the amount of PVP was selected as unique with two unequal adjacent lengths has been exhibited in the
experimental variable to tune the morphologies of Cu O. TEM of TO-Cu O as shown in Fig. 1c, the corresponding
2
2
Firstly, C-Cu O was obtained at 80 8C for 2 h using 0.2 g SAED pattern reveals that they are single-crystal structure
2
PVP as FESEM images and particle size distribution (Fig. 1c, insert). Fully developed O-Cu O can be obtained
2
statistics shown in Fig. S2a. It is obvious that these particles when the amount of PVP increased to 3.0 g. From the SEM
are well-defined cubes with sharp edges and uniform shape. image in Fig. S2d, it can be seen that the Cu O particles have
2
Meanwhile, it can be clearly seen that the particle is mainly almost perfect octahedral morphology with 12 well-defined
composed of six (100) planes and the average edge length is edges and 6 sharp corners without truncating, and the
measured to be about 0.66 mm according to the statistics of length of the octahedra is 1.49 mm with a narrow size
1
00 C-Cu O particles. Further characterization by TEM distribution. Corresponding TEM pattern as shown in
2
(
Fig. 1a) found that C-Cu O particle exhibited a square Fig. 1d exhibits a hexagon with a high symmetry, and
2
pattern with high symmetry, and the corresponding selected meanwhile, SAED pattern indicates that the obtained
area electron diffraction (SAED) pattern reveals that it is a O-Cu O is a single-crystal structure (Fig. 1d, insert).
2
single-crystal structure with exposed (100) facets (Fig. 1a,
The crystal structures of products synthesized at
insert). TC-Cu O could be acquired as shown in Fig. S1b different conditions are examined by XRD as shown in
2
when the amount of PVP is increased to 0.6 g. It can be Fig. 2a, which reveal that all Cu O particles belong to cubic
2
clearly seen that eight corners of the particles are truncated phase (Cu O PDF 00-005-0667) without any impurity, such
2
to form eight triangles on these cross sections. The edge as copper and cupric oxide. Obviously, the strong and sharp
length of TC-Cu O with six (100) facets and eight (111) peaks indicate that obtained Cu O particles are with high
2
2
facets is 1.03 mm and has a narrow size distribution as crystallinity. To further explore the bandgaps of the
shown in the insert pattern of Fig. S2b. The particles exhibit products, UV-Vis absorption spectra of Cu O with different
2
morphologies are measured as shown in Fig. S3, where a
broad absorption bands is clearly seen ranging from 500 to
6
50 nm. According to Tauc equation, bandgaps are
2
calculated from the plots of (ahn) versus hn curves of
UV-Vis absorption spectra (Fig. 2b), indicating that the
band gaps decrease from 2.0 to 1.9 eV with the percentages
of (100) plane diminishing, that is the morphology of
Cu O transfer from cube to octahedral, in good accordance
2
with the literature [34, 35]. To confirm PVP on the
Cu O surface have been removed clearly, the TG curves of
2
C-Cu O and O-Cu O were carried out as shown in Fig. 2c.
2
2
From the figure, weight loss observed in O-Cu O-six and
2
not in C-Cu O-six for the major weight loss around
2
4
00 8C as revealed which can be attributed to normal
thermal decomposition of PVP [36], meanwhile, it can be
seen that the weight loss disappeared in O-Cu O-ten when
2
the numbers of ultrasonic treatment increased to ten. The
above results reveal that ten times ultrasonic washing could
remove the adsorbed PVP on the Cu O crystal surface and
2
exclude the impact on performance of Cu O facets.
2
3
.2 Growth mechanism of Cu O As for the
2
reaction process, the initial polyhedral Cu O particles
Figure 1 The TEM images of different morphologies Cu O, (a)
2
2
C-Cu O, (b) TC-Cu O, (c) TO-Cu O, (d) O-Cu O and the insets evolve into Cu O with different morphologies as a result of
2
2
2
2
2
are their corresponding SAED patterns.
anisotropic growth of the polyhedral particles, due to the
www.pss-a.com
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
1
600904 (4 of 8)
2
Z. Xie et al.: Facet-dependent gas sensing properties of Cu O crystals
obviously, and C-Cu O are produced eventually. When the
2
amount of PVP increases to 0.6 g, the interaction strength is
enhanced and the surface energy of (111) facets is decreased
effectively at the same time, which is beneficial to lower the
growth rate of (111) facets. As a result, (111) facets can be
able to retain a bit in the final morphology of Cu O and that
2
is the TC-Cu O. Further increasing the amount of PVP to
2
1
.0 g, the interaction strength is greatly enhanced and the
growth of the (111) facets is significantly reduced, while the
growth rate of (100) facets is maintained relatively
unchanged, such that TO-Cu O are formed with a high
2
percentage of (111) facets. When the amount of PVP is
3
.0 g, truncated corners belong to (100) facets disappear
gradually, perfect O-Cu O with well-defined edges and
2
sharp corners are generated eventually due to further
decreased growth on (111) facets compared to the (100)
facets. In a word, PVP as an effective capping agent can
change the surface energy of (111) facets or (100) facets,
Cu O crystals with different morphologies evolved from
2
cube to octahedral are synthesized successfully at last.
3
.3 Gas sensing properties of Cu O As MOX
2
sensors are working at high temperatures and Cu O is easily
2
to be oxidized into CuO at elevated temperatures, it is
therefore essential to determine the highest working
temperature of Cu O. From the TG curves and XRD
2
patterns before and after 250 8C annealing in Fig. S4,
Cu O will not turn into CuO when the temperature is below
2
2
50 8C, which can be chosen as a maximum operating
temperature. Figure 3a shows the electrical resistance
response of four kinds of Cu O morphologies versus
2
operating temperature when exposed to 10 ppm benzene.
Obviously, it can be seen that all responses of four
morphologies increase with the enhanced temperature and
get the highest response at 250 8C. It is noted that the
response of Cu O decrease with the reduced percentage of
2
(
100) facets from cube to octahedra at the same operating
Figure 2 The XRD patterns (a) and the tangent pattern of UV-vis
absorption spectra (b) for C-Cu O, TC-Cu O, TO-Cu O,
2
2
2
O-Cu O (PDF 00-005-0667), and TG curves (c) for
2
2 2
C-Cu O and O-Cu O with varied washing times.
different surface energies of varied planes [37]. In this
paper, PVP, is a surfactant that may have two roles. One is to
prevent the aggregation of the Cu O particles and the other
2
is to act as a shape-controller to facilitate the formation of
Cu O crystals with different morphologies [27].
2
As a capping or dispersant agent, PVP molecule with a
long chain can be adsorbed to the surface of Cu O particles
2
via physical and chemical adsorption. It is believed that PVP
tends to interact more strongly with the (111) facets than
with the (100) facets and thus lead to suppressed growth of
(
111) facets compared with (100) facets [33]. In the initial
stage, when 0.2 g PVP is added into the reaction system, the
capping effect of PVP to Cu O polyhedron is not enough or (c) 5 ppm NO , and (d) C-Cu O crystals to three gases at 5 ppm
Figure 3 The responses of Cu
morphologies to (a) 10 ppm benzene, (b) 10 ppm acetone,
2
O crystals with different
2
2
2
weaker, which cannot hinder the growth of (111) facets at a series of temperatures.
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A (2017)
(5 of 8) 1600904
temperatures, so that the C-Cu O behaves with the best their optimum operating temperatures in Fig. 4d, which
2
sensing performance to benzene among the four kinds of shows a quasi-linear relationship for benzene and acetone.
morphologies, and the response reaches 1.24–10 ppm As for NO , it is a curve relationship at low concentration
2
benzene under 250 8C. On the other hand, the responses with a larger tangent slope and it will transfer into a quasi-
to 10 ppm acetone in Fig. 3b and to 5 ppm NO2 in linear relationship with a smaller slope nearly to saturation
Fig. 3c showed similar response relationships that the when the concentration exceeds 2.5 ppm.
response increases first and descends later as the
temperature increases, with the highest responses of
3
.4 Catalytic performance of Cu O To explore the
2
C-Cu O obtained at 200 and 100 8C for acetone and
2
origin of the different gas sensing properties of the
Cu O polyhedrons with varied shapes and catalytic
properties were assessed using a fixed-bed reactor as
showed in Fig. 5. The samples were mixed with quartz sands
and quartz wool with neglectable effect on acetone and
benzene degradation at operating temperatures below
50 8C [39]. Figure 5a and c shows the conversion of
benzene and acetone as a function of reaction temperature
over C-Cu O and O-Cu O catalysts, respectively. Appar-
ently, the catalytic activity to benzene and acetone increases
with the enhanced reaction temperature. It is noteworthy
that the catalytic activity of C-Cu O is higher than that of
O-Cu O at the same temperature, which would contribute
greatly to the higher gas response of C-Cu O than O-Cu O.
In the meanwhile, the conversion of acetone by both
C-Cu O and O-Cu O is higher than that of benzene, which is
nearly 100% for acetone at 250 8C but only 30–40% for
benzene. This might be the reason why both C-Cu O and
O-Cu O have higher response to acetone than to benzene as
shown in Figs. 3d and 4d. On the other hand, the acetone
conversion is only ꢄ40% at 200 8C where the highest
response is obtained, and there is also high response at
50 8C, where there is nearly no catalytic activity to both
acetone and benzene. Therefore, the adsorption of acetone
and benzene also contributes to the gas response at lower
working temperatures of Cu O. This is also applicable to the
gas sensing of NO with gas response maxima at 100 8C,
where there would only be the adsorption process because
NO , respectively. It can be clearly seen again that the
2
2
response decreases with the reduced percentage of (100)
facets and C-Cu O shows the highest response among the
2
four kinds of Cu O polyhedrons. All the response results are
2
summarized in Fig. 3d, which show the good selectivity
of C-Cu O to the NO and acetone. Specifically,
2
2
2
C-Cu O exhibits a far higher response to NO at 100 8C than
2
2
those of other two gases (nearly zero). So does the good
selectively to acetone at 200 8C. To exclude the particle size
effect on the performance of sensor response, 1.13 mm cubic
2
2
Cu O was prepared as shown in Fig. S5a according to the
2
2
literature [38]. Figure S5b shows the responses to 5 ppm
acetone where it is clearly seen that the responses are similar
2
2
2
for both cubic Cu O with the sizes of 1.13 and 0.66 mm. The
2
2
3
theoretical surface area can be calculated by 6L /L r, where
L is the size of C-Cu O and r is the density (6.0 g cm ). The
ꢂ3
2
2
2
specific surface areas of 1.13 and 0.66 mm C-Cu O are 0.88
2
2
ꢂ1
2
and 1.52 m g , which are very small as compared with
the nm scale crystals. Therefore, the main reason for the
performance difference is the crystal facets rather than the
particle size.
2
The dynamic sensing response curves of
C-Cu O toward benzene, acetone, and NO at a series of
1
2
2
concentrations at their optimized working temperatures are
shown in Fig. 4a–c. The response and recovery times are
shown to be less than 1 min, considering the time for the
incoming benzene to reach the stabilized concentration [10].
Further, the gas responses and concentrations are plotted at
2
2
NO cannot be converted at such low temperature. On the
2
2
Figure 4 The dynamic sensing characteristics of C-Cu O crystals
to different concentrations of (a) benzene, (b) acetone, (c) nitrogen Figure 5 A conversion of (a) benzene or (c) acetone as a function
dioxide, and (d) the responses of C-Cu O to three gases at their of reaction temperature and Arrhenius plots for the oxidation of
2
optimum operating temperatures.
2 2
(b) benzene or (d) acetone over the C-Cu O and O-Cu O.
www.pss-a.com
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
1
600904 (6 of 8)
2
Z. Xie et al.: Facet-dependent gas sensing properties of Cu O crystals
contrary, the gas response is relatively lower at the highest
conversion temperature of acetone (i.e., 250 8C) as
compared with that at optimum 200 8C, which infers the
resistance change is not only determined by the surface
adsorption/reaction process, but also by the intrinsic
electrical property changed, as discussed in the following
section.
To further demonstrate the facet-dependent catalytic
properties of C-Cu O and O-Cu O, the activation energy of
2
2
benzene and acetone are calculated on the kinetics of
catalytic oxidation of benzene [40, 41], toluene [40],
carbon monoxide [42], and so on, according to the
literature. Figure 5b and d shows the Arrhenius plots for the
oxidation of benzene and acetone over C-Cu O and
2
O-Cu O, respectively. Their apparent activation energies
2
ꢂ1 ꢂ1
are calculated from the conversion rate g (mol ꢀ g ꢀ s
)
of benzene and acetone below 20%. All the correlation
2
coefficients (R ) of the fitting lines for the test samples are
close to 1.0, which indicates the excellent linear relation-
ship of lng versus 1000/T. The activation energies for
C-Cu O and O-Cu O to benzene and acetone are 131.21,
2
2
ꢂ1
1
56.55, 52.18, 61.01 kJ ꢀ mol , respectively. Such a result
is in good consistency with the higher conversion for
C-Cu O than O-Cu O to both benzene and acetone.
2
2
According to the comprehensive analysis of the above, the
C-Cu O has better catalytic performance than
a
2
O-Cu O, which may be an important explanation for
2
the corresponding sensing properties of benzene and
acetone.
Figure 6 The (a) response of four morphologies Cu O with
different exposed percentages of (100) facet to different gases at
2
their optimum operating temperatures at certain concentration and
(
b) their corresponding Arrhenius plots for electrical current.
3
.5 Mechanism analysis Generally, it is believed
that the sensing mechanism is controlled by the chemical
and electrical properties of the sensing material [43].
Obviously, the chemical [14–16] and electrical [11–13]
properties of polyhedron with different crystal facets, as
well as exposed percentage of certain crystal facets, are not
the same due to the difference of atom numbers and
arrangements on its surface. Figure 6a shows the response of
ꢂ
ꢂ
ꢂ
O2ꢂ þ e Ð 2O ðO dominate at 150 8C ꢂ 500 8CÞ
ð3Þ
Similar chemisorption occurs when NO is introduced
2
with higher electron affinity than oxygen, and the reactions
can be written as Eqs. (4) and (5). As for reducing gases,
benzene and acetone are electron donors which will donate
electrons by chemisorbed onto the surface or competitive
adsorptions with ion-sorbed oxygen, as shown in Eqs. (6)
and (7). These processes will contribute to the gas response
at low temperatures (i.e., <150 8C) where benzene and
acetone are not catalytically oxidized. However, when the
temperature is higher, they begin to be converted to CO as
shown in Fig. S6, which plots the CO yield as a function of
reaction temperature for the oxidation of benzene and
acetone over the C-Cu O and O-Cu O. Then, the reactions
of (8 and 9) start to occur [21, 29].
four morphologies Cu O with different exposed percentages
2
of (100) facet to different gases at their optimum operating
temperatures. It can be clearly seen that the responses to
benzene, acetone, and NO are decreased with the reduced
2
percentages of (100) facet, namely, with the transformation
of C-Cu O to O-Cu O.
2
2
Surface adsorption and catalytic reaction play important
roles in gas sensing processes. Firstly, the oxygen molecule
2
2
chemisorbs on the surface of Cu O particle and traps
2
electrons to form ionized oxygen species as shown in Eqs.
2
2
(1–3) [22]:
O ðgasÞ Ð O ðads:Þ
ð1Þ
ꢂ
2
2
ꢂ
NO2ðgasÞ þ e ! NO2 ðads:Þ
ð4Þ
ð5Þ
ꢂ
O ðads:Þ þ e Ð O
ꢂ
ðO2
ꢂ
dominate at 25 8C ꢂ 150 8CÞ
ð2Þ
ꢂ
2
2
ꢂ
NO2ðgasÞ þ O2 ðads:Þ þ 2e
ꢂ
! NO2
ꢂ
ðads:Þ þ 2O ðads:Þ
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A (2017)
(7 of 8) 1600904
All these results are schematically shown in Fig. 7.
ð6Þ The (100) facets have lower work function than (111)
facet, which is theoretically calculated by density
function theory and observed by Kelvin probe force
microscopy [17, 18, 45]. This lower work function of
(100) facets means more oxygen would be adsorbed on
C6H6ðgasÞ þ Cu2O Ð C6H6ðads:Þ ꢂ Cu2Oꢂ
þ
C6H6ðgasÞ þ O2ꢂ ðads:Þ Ð C6H6ðads:Þ þ O2ðgasÞ
ꢂ
þ e
(
100) facets trapping more electrons according to
ð7Þ
ð8Þ
ð9Þ
reactions (2) and (3), and thus, constructing a thicker
surface hole accumulation layer and a higher surface
potential barrier. This higher surface potential is in good
accordance with the higher thermal activation energy of
ꢂ
C H ðgasÞ þ 15O Ð 6CO ðgasÞ þ 3H OðgasÞ
6
6
2
2
ꢂ
þ 15e
holes in C-Cu O as schematically shown in Fig. 7b.
2
However, as a typical p-type semiconductor, the majority
of the concentration is dominant in the resistance because
there is neglectable barrier for holes to transport from
ꢂ
C H OðgasÞ þ 8O Ð 3CO ðgasÞ þ 3H OðgasÞ
3
6
2
2
ꢂ
þ 8e
particle to particle. In this case, C-Cu O changes its
2
resistance easily because of the relatively lower work
function and thus the higher concentration ion-sorbed
oxygen with higher catalytic activity. These results not
only show the facet dependent chemical and electrical
properties, but also promising for the highly selective
MOX sensor tailoring in p-type semiconductors for
the future high response and selectivity sensors and
noses.
Though the catalytic activity becomes higher at
elevated temperatures, the gas response begin to decrease
after the maxima is obtained at medium temperature,
inferring that another dominant process is governing the
resistance of the material. Therefore, electrical properties
of Cu O with different crystal facets are then further
2
explored to demonstrate the intrinsic resistance changes.
The curves of resistance logarithmic versus temperatures
for C-Cu O, TC-Cu O, TO-Cu O, and O-Cu O as shown
2
2
2
2
4
Conclusions Cu O polyhedrons with four mor-
2
in Fig. S7 and Fig. 6b, show the corresponding Arrhenius
plots for electrical current of four morphologies of
phologies are successfully synthesized via a facile
solution-phase method, and the facet-dependent gas
sensing properties are investigated in detail.
Cu O, and it can be calculated that the hole activation
2
energy [44] decreases with the percentage of (100) facet
C-Cu O with six (100) facets shows higher responses to
reducing. C-Cu O with six (100) facets possess a
2
2
benzene, acetone, and NO than O-Cu O exposing eight
maximum value of 0.55 eV, while O-Cu O with eight
2
2
2
(
111) facets, which is in good consistence with their
(
111) facets gets a minimum value of 0.36 eV. These
catalytic activities to benzene and acetone. More
importantly, the gas sensing maxima are obtained at
different working temperatures illustrating good selectiv-
results show that intrinsic bulk carriers are easily excited at
elevated temperatures for O-Cu O. Consequently, if the
2
bulk hole concentration induced by thermal excitation is
much higher than that induced by surface adsorption/
reaction, the resistance will change a little, meaning a small
gas response.
ity to NO and acetone. Catalytic property measurements
2
reveal a lower apparent catalytic activation energy of (100)
facets than (111) ones, while electrical characterization
demonstrates a higher thermal excitation energy of bulk
holes on (100) facets than (111) ones, in good agreement
with the relatively lower work function of (100) compared
to (111). Therefore, the synergy of chemical and physical
properties gave rise to a better sensing property for
C-Cu O than O-Cu O eventually, and the C-Cu O has
2
2
2
good selectivity to NO and acetone at varied working
2
temperatures, promising for future highly selectively gas
sensor development.
Supporting Information Additional supporting infor-
mation may be found in the online version of this article at
the publisher’s web-site.
Acknowledgements This research was financially sup-
ported by the National Key Research and Development Program
Figure 7 The schematic diagrams (a) for the structure of (2016YFC0207101), the National Natural Science Foundation of
C-Cu O and O-Cu O after oxygen species adsorbed and (b) their China (51272253), and the State Key Laboratory of Multiphase
2
2
corresponding energy band diagrams.
Complex Systems (MPCS-2014-C-01).
www.pss-a.com
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
1
600904 (8 of 8)
2
Z. Xie et al.: Facet-dependent gas sensing properties of Cu O crystals
References
[23] Y. Wang, Y. L €u , W. Zhan, Z. Xie, Q. Kuang, and L. Zheng,
J. Mater. Chem. A 3, 12796 (2015).
[
1] Y. Chen, H. Qin, X. Wang, L. Li, and J. Hu, Sens. Actuators
B 235, 56 (2016).
2] F. Wantke, C. M. Demmer, P. Tappler, M. Gotz, and R.
Jarisch, Clin. Exp. All. 26, 276 (1996).
3] S. E. Anderson, J. R. Wells, A. Fedorowicz, L. F. Butter-
worth, B. J. Meade, and A. E. Munson, Toxicol. Sci. 97, 355
[
24] N. Tang, B. Chen, Y. Xia, D. Chen, and X. Jiao, RSC Adv. 5,
4433 (2015).
5
[
[
25] M. R. Alenezi, A. S. Alshammari, K. D. G. I. Jayawardena,
M. J. Beliatis, S. J. Henley, and S. R. P. Silva, J. Phys. Chem.
C 117, 17855 (2013).
[
[
26] S. L. Bai, L. Quan, P. G. Tang, Y. B. Zhao, R. X. Luo, A. F.
Chen, D. Q. Li, and N. Han, Ieee Sens. J. 16, 866 (2016).
27] Y. Sui, W. Fu, H. Yang, Y. Zeng, Y. Zhang, Q. Zhao, Y. Li,
X. Zhou, Y. Leng, M. Li, and G. Zou, Cryst. Growth Des. 10,
(
2007).
[
[
[
[
4] W. J. Kim, N. Terada, T. Nomura, R. Takahashi, S. D. Lee,
J. H. Park, and A. Konno, Clin. Exp. All. 32, 87 (2002).
5] P. Wolkoff, D. R. Crump, and P. T. C. Harrison, Environ. Int.
[
9
9 (2010).
8
7, 74 (2016).
[
[
[
[
[
28] N. Han, L. Chai, Q. Wang, Y. Tian, P. Deng, and Y. Chen,
Sens. Actuators B 147, 526 (2010).
29] W. Li, X. Wu, N. Han, J. Chen, X. Qian, Y. Deng, W. Tang,
and Y. Chen, Sens. Actuators B 225, 159 (2016).
30] C. H. B. Ng and W. Y. Fan, J. Phys. Chem. B 110, 20801
6] Y. Zhang, B. Jiang, M. Yuan, P. Li, W. Li, and X. Zheng,
J. Sol-Gel Sci. Technol. 79, 167 (2016).
7] S.-J. Choi, B.-H. Jang, S.-J. Lee, B. K. Min, A. Rothschild,
and I.-D. Kim, Acs Appl. Mater. Interfaces 6, 2588
(
2014).
(
2006).
31] X. Wang, M. Chen, Y. He, and J. Zhu, J. Alloys Compd. 628,
1 (2015).
[
8] B.-Y. Kim, J. S. Cho, J.-W. Yoon, C. W. Na, C.-S. Lee, J. H.
Ahn, Y. C. Kang, and J.-H. Lee, Sens. Actuators B 234, 353
5
(
2016).
32] L. Tang, J. Lv, S. Sun, X. Zhang, C. Kong, X. Song, and Z.
[
9] J. Huang, L. Wang, C. Gu, Z. Wang, Y. Sun, and J.-J. Shim,
Sens. Actuators B 207, 782 (2015).
10] P. Hu, N. Han, D. Zhang, J. C. Ho, and Y. Chen, Sens.
Yang, New J. Chem. 38, 4657 (2014).
[
[
33] Y. Shang and L. Guo, Adv. Sci. 2, 1500140 (2015).
34] C.-H. Kuo, C.-H. Chen, and M. H. Huang, Adv. Funct. Mater.
[
Actuators B 169, 74 (2012).
1
7, 3778 (2007).
[
[
11] X. Xu and S. Peyman, Nano Lett. 9, 1999 (2009).
12] C.-S. Tan, H.-S. Chen, C.-Y. Chiu, S.-C. Wu, L.-J. Chen, and
M. H. Huang, Chem. Mater. 28, 1574 (2016).
[
[
[
[
[
[
[
[
[
35] Y. Zhang, B. Deng, T. Zhang, D. Gao, and A.-W. Xu, J. Phys.
Chem. C 114, 5076 (2010).
36] Z. J. Yu, M. R. Kumar, D. L. Sun, L. T. Wang, and R. Y.
Hong, Mater. Lett. 166, 287 (2016).
[
[
[
13] C.-S. Tan, S.-C. Hsu, W.-H. Ke, L.-J. Chen, and M. H.
Huang, Nano Lett. 15, 2155 (2015).
37] D.-F. Zhang, H. Zhang, L. Guo, K. Zheng, X.-D. Han, and Z.
14] J. Han, J. Meeprasert, P. Maitarad, S. Nammuangruk, L. Shi,
and D. Zhang, J. Phys. Chem. C 120, 1523 (2016).
15] C.-H. Kuo, I. M. Mosa, S. Thanneeru, V. Sharma, L. Zhang,
S. Biswas, M. Aindow, S. Pamir Alpay, J. F. Rusling, S. L.
Suib, and J. He, Chem. Commun. 51, 5951 (2015).
16] Q. Zhang and H. Wang, ACS Catal. 4, 4027 (2014).
17] L. Sungki, C.-W. Liang, and L. W. Martin, ACS NANO 5,
Zhang, J. Mater. Chem. 19, 5220 (2009).
38] H. Cao, X. Qian, C. Wang, X. Ma, J. Yin, and Z. Zhu, J. Am.
Chem. Soc. 127, 16024 (2005).
39] W. Tang, W. Li, D. Li, G. Liu, X. Wu, and Y. Chen, Catal.
Lett. 144, 1906 (2014).
40] S. Ord ꢀo ~n ez, L. Bello, H. Sastre, R. Rosal, and F. V. Dıez,
[
[
́
Appl. Catal. B 38, 144 (2002).
41] S. K. Gangwal, M. E. Mullins, J. J. Spivey, P. R. Caffrey, and
B. A. Tichenor, Appl. Catal. 36, 231 (1988).
42] V. A. Bondzie, S. C. Parker, and C. T. Campbell, Catal. Lett.
3
741 (2011).
[
18] Z. Zheng, B. Huang, Z. Wang, M. Guo, X. Qin, X. Zhang, P.
Wang, and Y. Dai, J. Phys. Chem. C 113, 14451 (2009).
19] Y. Xu, H. Wang, Y. Yu, L. Tian, W. Zhao, and B. Zhang,
J. Phys. Chem. C 115, 15288 (2011).
[
6
3, 147 (1999).
43] J. H. Kim, P. Wu, H. W. Kim, and S. S. Kim, Acs Appl.
[
[
20] C.-H. Kuo and M. H. Huang, Nano Today 5, 114 (2010).
Mater. Interfaces 8, 7175 (2016).
21] X. Zou, H. Fan, Y. Tian, M. Zhang, and X. Yan, Dalton [44] X. Liu, J. Hu, B. Cheng, H. Qin, and M. Jiang, Curr. Appl.
Trans. 44, 7811 (2015).
22] C.-L. Hsu, J.-Y. Tsai, and T.-J. Hsueh, Sens. Actuators B 224, [45] R. Li, X. Tao, R. Chen, F. Fan, and C. Li, Chem. Eur. J. 21,
5 (2016).
14340 (2015).
Phys. 9, 615 (2009).
[
9
ß 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com