вход по аккаунту



код для вставкиСкачать
Difficulty of carrier generation in orthorhombic PbO
Min Liao, Seiji Takemoto, Zewen Xiao, Yoshitake Toda, Tomofumi Tada, Shigenori Ueda, Toshio Kamiya, and
Hideo Hosono
Citation: Journal of Applied Physics 119, 165701 (2016);
View online:
View Table of Contents:
Published by the American Institute of Physics
Articles you may be interested in
Amorphous pnictide semiconductor BaZn2As2 exhibiting high hole mobility
Applied Physics Letters 109, 242105 (2016); 10.1063/1.4972039
Unusual defect physics in CH3NH3PbI3 perovskite solar cell absorber
Applied Physics Letters 104, 063903 (2014); 10.1063/1.4864778
Solid phase epitaxial growth of high mobility La:BaSnO3 thin films co-doped with interstitial hydrogen
Applied Physics Letters 108, 172101 (2016); 10.1063/1.4948355
SnS thin films prepared by H2S-free process and its p-type thin film transistor
AIP Advances 6, 015112 (2016); 10.1063/1.4940931
Antisolvent-assisted powder engineering for controlled growth of hybrid CH3NH3PbI3 perovskite thin films
APL Materials 5, 026101 (2017); 10.1063/1.4974942
Thermoelectric phase diagram of the SrTiO3–SrNbO3 solid solution system
Journal of Applied Physics 121, 185102 (2017); 10.1063/1.4983359
Difficulty of carrier generation in orthorhombic PbO
Min Liao,1 Seiji Takemoto,1 Zewen Xiao,1,2 Yoshitake Toda,1 Tomofumi Tada,1
Shigenori Ueda,3,4 Toshio Kamiya,1,2 and Hideo Hosono1,2,a)
Materials Research Center for Element Strategy, Tokyo Institute of Technology, Yokohama 226-8503, Japan
Materials and Structures Laboratory, Tokyo Institute of Technology, Yokohama 226-8503, Japan
Synchrotron X-ray Station at SPring-8, National Institute for Materials Science, Hyogo 679-5148, Japan
Quantum Beam Unit, National Institute for Materials Science, Tsukuba 305-0047, Japan
(Received 8 October 2015; accepted 12 April 2016; published online 26 April 2016)
Polycrystalline b-PbO films were grown by pulsed laser deposition in atmospheres ranging from
oxygen-poor (the oxygen pressure of 0.01 Pa) to oxygen-rich (13 Pa) conditions, and the oxygen
chemical potential was further enhanced by ozone annealing to examine hole doping. It was found
that each of the as-grown b-PbO films showed poor electrical conductivity, r < 1.4 107 S cm1,
regardless of the oxygen pressure. The density functional calculations revealed that native defects
including Pb and O vacancies have deep transition levels and extremely high formation enthalpies,
which indicates difficulty of carrier generation in b-PbO and explains the experimentally observed
poor electrical conductivity. The analysis of the electronic structures showed that the interaction
between Pb 6s and O 2p orbitals is weak due to the deep energy level of Pb 6s and does not raise
the valence band maximum (VBM) level unlike that observed in SnO, which is also supported by
ultraviolet photoemission spectroscopy measurements. The deep acceptor transition levels of the
native defects are attributed to the deep VBM of b-PbO. On the other hand, annealing b-PbO films
in reactive oxygen-containing atmospheres (i.e., O3) led to a significantly enhanced electrical conductivity (i.e., r > 7.1 102 S cm1) but it is the result of the formation of an n-type PbO2 phase
because oxygen chemical potential exceeded the phase boundary limit. The striking difference in
carrier generation between PbO and SnO is discussed based on the electronic structures calculated
by density functional theory. Published by AIP Publishing. []
Oxide semiconductors are attractive for a wide range of
device applications from thin-film transistors (TFTs) to solar
cells.1–3 Owing to the large electron affinity and the highly
dispersed conduction band minimum (CBM) composed
mainly of spatially spread isotropic ns0 orbitals of the metal
cations and existence of native shallow donors, oxide semiconductors usually exhibit unintentional n-type conductivity
and high electron mobilities.1,2 However, p-type doping is
more difficult than n-type one because fully occupied 2p
orbitals of oxygen ions tend to form deep (i.e., large ionization potential) and localized valence band maximum (VBM)
in most oxides of typical metals.4–6 In order to extend the
applications of transparent electronic devices, great efforts
have been made to design and fabricate high-performance
p-type oxide semiconductors.5–11 In 1997, some of the
authors proposed to employ hybridized orbitals of Cu 3d and
O 2p to raise the VBM level and delocalize the VBM states
(i.e., hole transport paths),5 and then a series of p-type transparent oxide semiconductors such as CuAO2 (A ¼ B, Al, Ga,
and In) and SrCu2O2 was found by following this chemical
design concept.7,8 However, TFTs with these channel layers
were not functional probably due to the low hole mobilities
and high density of band tail/defect states. To improve
the hole transport properties, layered oxychalcogenides
Author to whom correspondence should be addressed. Electronic mail:
LaCuOCh (Ch ¼ S and Se) with more extended orbitals for
anions were developed.9,10 Although relatively high hole
mobility (up to 8 cm2 V1 s1) was attained in these p-type
oxide semiconductors, the high hole concentration
(>1018 cm3) makes them unsuitable for TFT applications.10
On the other hand, it was expected that an oxide, in which
the VBM is composed mainly of pseudo-closed ns2 orbitals
of p-block metal cations, would be a promising candidate for
p-type oxide semiconductors.11–13 As expected, effective ptype doping and relatively high hole mobility were attained
in SnO due to the hole contribution through the VBM composed of Sn2þ 5s2 orbitals,6,14,15 and thus p-type oxide-based
TFTs were successfully fabricated using SnO films as the
channel layers.16–22
In line with the ns2 character of Sn2þO2–, Pb2þO2– also
has an analogous ns2 (n ¼ 6) electronic configuration and is
expected to be another promising p-type oxide semiconductor accordingly.23 PbO is commonly found in two polymorphs, namely, a tetragonal (a-PbO, litharge) phase and an
orthorhombic (b-PbO, massicot) phase.24–26 a-PbO has a
small indirect band gap of 1.9 eV, while b-PbO has a large
indirect band gap (2.7 eV) comparable to the optical band
gap (direct-transition type band gap) of SnO.16,25 Because of
its important roles in industrial and technological applications, notably imaging devices, photovoltaic solar cells, and
lead-acid batteries, a great deal of attention has been paid to
PbO.23–33 Although the fabrication of PbO has widely been
reported,23,25–29 only a few experimental studies have
focused on the electrical properties of PbO.23,27,28 Some of
119, 165701-1
Published by AIP Publishing.
Liao et al.
these studies revealed that the oxygen concentration has a
great effect on the carrier generation (i.e., carrier density and
carrier polarity) in PbO.27,28 In particular, ozone exposure at
room temperature greatly enhances the electrical conductivity of b-PbO pellets;23 however, its mechanism is not clear
yet. Indeed, previous investigations on SnO have suggested
that an understanding of the electronic structure and defect
physics is vital to the optimization of carrier transport in
SnO films.13,15,17 However, very few works have been made
by focusing on the electronic structure and the defect chemistry of a-PbO,32,33 and the defect physics of b-PbO has not
yet been reported as far as we know.
In this work, b-PbO films were grown by pulsed laser
deposition (PLD) in atmospheres ranging from oxygen-poor
to oxygen-rich conditions, and the effect of ozone annealing
on the electrical properties of b-PbO films was systematically investigated. Their properties are discussed based on
crystal structure, electronic structure, and defect physics
with the help of first-principles density functional theory
(DFT) calculations.
b-PbO thin films were grown on silica glass substrates
by PLD with a KrF excimer laser at 300 C. Polycrystalline
b-PbO pellets prepared by spark plasma sintering were used
as targets. The base pressure in the deposition chamber was
3 105 Pa. The laser ablation of the PbO targets was carried out at a laser fluence of 2 J cm2 with a repetition rate
of 5 Hz for 8 min. The oxygen pressure (PO2) during deposition was varied from 0.01 to 13 Pa. To investigate the effect
of ozone annealing, b-PbO films grown at 300 C at
PO2 ¼ 13 Pa were subjected to post-annealing in an ozone
atmosphere at 200 C for various annealing times (tann) ranging from 3 to 120 min. The O3 gas was generated by silent
discharge of a pure O2 gas, and the O3 concentration was
measured to be approximately 3.74 g m3 using an ozone
meter.34 After the film fabrication, Au electrodes were
evaporated through a shadow mask on the fabricated PbO
films for electrical characterization.
The crystal structures of the fabricated films were characterized by X-ray diffraction (XRD) using Cu Ka1 radiation,
while the electronic structures of the samples were analyzed
by hard X-ray photoelectron spectroscopy (HAXPES) using
a high-resolution hemispherical electron analyzer at the
undulator beam line BL15XU35,36 of SPring-8 with the excitation energy of h ¼ 5953.4 eV. Ultraviolet photoemission
spectroscopy (UPS) measurements were performed by using
a hemispherical analyzer with a He I light source
(ht ¼ 21.2 eV). Optical properties of the films were investigated by measuring transmittance (Tr) and reflectance (R)
spectra. The absorption coefficient (a) was estimated
by a ¼ ln[(1–R)/Tr]/d, where d is the film thickness. Hall
mobility (lHall), electrical conductivity (r), and carrier concentration (N) of the films were evaluated by Hall effect
measurements using the van der Pauw configuration with an
AC modulation of magnetic field at room temperature.
The defect calculations were performed in the framework of DFT using the projector augmented wave method as
J. Appl. Phys. 119, 165701 (2016)
implemented in the VASP code.37 The plane wave cutoff
energy for all the calculations was set to 500 eV. The defect
models were based on the 3 3 3 supercell model, and a C
point-only k-mesh was used for defect enthalpy calculations.
The generalized gradient approximation (GGA) with the
Perdew–Burke–Ernzerhof (PBE96)38 functional was used
for structural relaxations; however, PBE96 overestimated the
c-axis lattice parameter by 5% due to the absence of van der
Waals interactions between the neighboring layers in DFT
and also gave an underestimated band gap of 2.20 eV due to
the well-known band gap problem of DFT (the experimental
value is 2.81 eV). Since the formation enthalpy of an interstitial defect strongly depended on the interlayer separation, we
constrained the c-axis lattice parameter to the experimental
value.24,26 Then, total-energy calculations were performed for
the relaxed defect models using the Heyd–Scuseria–Ernzerhof
(HSE06) hybrid functional39 with a mixing parameter for the
exact-exchange term of 19%, which was optimized to reproduce the experimental band gap.
The formation enthalpy of a defect (D) in a charge state
q is given by40
DHD;q ðEF ; lÞ ¼ ED;q EH na la þ qðEF þ EV þ DVÞ;
where EH is the total energy of the supercell model of the
perfect crystal and ED,q is that of a supercell with a defect D
in the charge state q. na indicates the number of species a
added to (na > 0) or removed from (na < 0) the perfect supercell model when the defect is formed, and la is the chemical
potential of the species a with respect to that of an elemental
phase (lael) by la ¼ lael þ Dla (i.e., lPbel and lOel are taken
from elementary Pb (Fm3m) solid and gaseous O2, respectively). EF is the Fermi level with respect to the VBM level
(EV) in the perfect supercell. DV is added to correct the
energy shift due to the formation of the defect.
The chemical potentials lPb and lO vary between the
O-poor (Pb-rich) and O-rich (Pb-poor) limits under a constraint
by the equilibrium condition of b-PbO, i.e., DlPb þ DlO
¼ DH(b-PbO), where DH(b-PbO) is the formation energy of
b-PbO. The O-poor (Pb-rich) limit is given with respect to
elementary Pb (Fm–3m) (i.e., DlPb ¼ 0), while the O-rich
(Pb-poor) limit is given by the equilibrium condition
between b-PbO and Pb3O4 (Pbam) (i.e., DlO ¼ DH(Pb3O4)
– 3DH(b-PbO)). We also examined effect of impurity hydrogen. The chemical potential of hydrogen (lH) was taken
from a H2 molecule (lH ¼ 1=2 of the total energy of a H2
molecule model). The equilibrium Fermi level (EF,e) at
room temperature was determined by the condition of
charge neutrality.41
Figure 1(a) shows the crystal structure of b-PbO, which
is an orthorhombic structure with the space group Pbcm.23
Figures 1(b) and 1(c) show the out-of-plane and in-plane
XRD patterns of the as-grown PbO films at PO2 varied from
0.01 to 13 Pa. It is clear that all the PbO films crystallized to
the orthorhombic b-PbO structure. The out-of-plane XRD
Liao et al.
FIG. 1. (a) Crystal structure of b-PbO (space group Pbcm). The relaxed Hi
and VPb–2H models are also shown below (a). (b) Out-of-plane and (c) inplane XRD patterns of the films grown at PO2 varied from 0.01 to 13 Pa
(indicated in the figures).
patterns mainly exhibit a series of strong 0k0 diffraction
peaks, while the in-plane XRD patterns show strong h00 and
h0l diffraction peaks. It suggests that the as-grown b-PbO
films are polycrystalline with preferential b-axis orientation.
Only the film grown at 0.01 Pa includes a metallic Pb phase
as shown both by the out-of-plane and the in-plane XRD
patterns, but increasing the oxygen pressure removed the
metallic Pb phase.
Figure 2(a) plots electrical conductivity as a function of
PO2 for the as-grown b-PbO films. The electrical conductivity of the as-grown b-PbO films was almost unchanged with
PO2, and even the b-PbO film grown at 13 Pa still had an
extremely low electrical conductivity of 1.4 107 S cm1.
FIG. 2. (a) Electrical conductivity and (b) optical absorption spectra with
(aht)1/2–ht plots of the b-PbO films grown at different PO2. (c) UPS spectra
near the secondary electron cutoff region for the b-PbO film grown at
PO2 ¼ 13 Pa. The inset in (c) shows the UPS spectrum in the valence band
region. (d) Simplified energy level diagrams of b-PbO in comparison with
J. Appl. Phys. 119, 165701 (2016)
The poor conductivity is in accordance with the alreadyreported results and common both for a-PbO and b-PbO.23,27
However, this disappointing result is in stark contrast to
SnO, in which a far higher electrical conductivity (i.e.,
r > 2.0 102 S cm1) was attained easily by adjusting PO2
during film deposition or post-annealing.14,42
It is known that b-PbO single crystal has an indirect
transition type band gap with 2.7 eV.25 Here, we discuss
the change in the band gap with respect to PO2 based on the
optical band gap (Eg,ind) determined by (aht)1/2 – ht plot
(Fig. 2(b)). It is found that the b-PbO film grown at 0.01 Pa
exhibits a strong subgap absorption over 104 cm1 even in
the subgap region at photon energies ht < 1.0 eV. However,
the subgap absorption decreases with increasing oxygen
pressure, which would be attributed to the reduction of
metallic Pb in the b-PbO film. The (aht)1/2 – ht plots exhibit
linear regions, supporting the indirect transitions in the
as-grown b-PbO films. The Eg,ind of the b-PbO film grown at
0.01 Pa is estimated to be 2.7 eV, which is slightly smaller
than the reported value (Eg,ind ¼ 2.8 eV) of b-PbO films26 but
in good agreement with the value reported for b-PbO single
crystal.25 As the oxygen pressure increases from 0.01 to
13 Pa, the Eg,ind values estimated by extrapolating the linear
regions in the (aht)1/2 – ht plots appear to increase from 2.7
to 2.8 eV; however, all the (aht)1/2 values are the same at
ht 3.5 eV and the different Eg,ind values are obtained due
to the slightly different slopes of the (aht)1/2 – ht plots,
where the slope becomes sharper with increasing PO2. It
implies that the film grown at PO2 ¼ 13 Pa with the sharpest
slope and the smallest subgap absorption is the best-quality
b-PbO film and the Eg,ind value of 2.8 eV should be taken for
(0k0)-oriented b-PbO.
Since successful p-type and n-type doping of SnO can
be explained phenomenologically based on the band-edge
energies (i.e., a relatively small ionization potential makes
p-type doing feasible, while a large electron affinity makes
n-type doping feasible),6,15,43,44 here the band-edge energies
are measured by UPS in order to examine the feasibility of
carrier doping in b-PbO. Figure 2(c) shows the representative UPS spectra near the secondary electron cutoff region
for the b-PbO film grown at 300 C at PO2 ¼ 13 Pa. The EF
(adjusted to 0 eV in the figures) was calibrated using an Au
reference sample. The UPS measurements were carried out
with sample biases from 5 to 10 V, and the energy axes in
the figures were corrected with the applied bias. We can see
that the corrected cut-off energies fall in the same value
/ ¼ 4.8 eV (the work function defined as the difference
between the vacuum energy level (EVAC) and EF), guaranteeing the reliability of the measurement results. The valence
band region (the inset of Fig. 2(c)) shows that the EF is
located at 1.4 eV above the VBM (i.e., near the mid-gap),
which further supports the low electrical conductivity of the
b-PbO films. Figure 2(d) illustrates the energy level diagram
of b-PbO determined by the UPS results in comparison with
reported one for SnO.17 The VBM level (i.e., ionization
potential) of b-PbO is located at 6.2 eV below the vacuum
energy level, which is deeper than that of SnO (5.8 eV). The
origin of the deeper VBM of b-PbO will be discussed later.
On the other hand, because of the small indirect band gap of
Liao et al.
0.7 eV, SnO has a large electron affinity of 5.1 eV. However,
b-PbO has a significantly smaller electron affinity of 3.4 eV
due to its large indirect band gap (2.8 eV). From the viewpoint of its large ionization potential and small electron
affinity, both p-type and n-type doping would be difficult for
The relatively high electrical conductivity (i.e., easiness
of hole generation) in typical p-type Sn2þ-based semiconductors such as SnO and SnS would be understood theoretically based on DFT, i.e., Sn vacancies (VSn) have low
formation enthalpies (less than 2 eV) and act as shallow
acceptors with transition levels e(0/–n) within 0.1 eV above
the VBM.13,45,46 Here, we performed total-energy calculations for defect models to theoretically explain the difficulty
of carrier generation in b-PbO. The calculated DH of native
defects are plotted as a function of EF in Fig. 3 (solid lines).
Unlike VSn in SnO, Pb vacancy (VPb) can hardly be formed
in b-PbO in the p-type region (i.e., EF below the mid gap
1.4 eV) due to its prohibitively high DH under both O-poor
and O-rich limits. Moreover, the transition levels e(1/2)
and e(0/1) of VPb are located at 0.66 and 0.25 eV above the
VBM, respectively, which are much deeper than those of
VSn in SnO and SnS. Pb interstitial (Pbi) can be a donor, and
its (þ1/0) transition level is located above the CBM (not
seen in the figures) although the (þ2/þ1) transition level is
deep at 0.7 eV below the CBM; however, its DH in the
n-type region (EF > 1.4 eV) is extremely high and thus Pbi is
hardly generated. Although oxygen vacancy (VO) and interstitial (Oi) have relatively low DH, VO (donor) is stabilized
with the neutral charge in the n-type region and Oi (acceptor)
is stabilized with the neutral charge in the p-type region;
thereby, they contribute nothing to carrier generation. The
EF,e values at the O-poor and O-rich limits, which were
calculated by solving the charge neutrality equation selfconsistently, are 2.0 eV and 1.2 eV above the VBM, respectively, both of which are too deep to provide enough free
carriers for b-PbO. These results suggest that carriers can
hardly be generated through native defects in b-PbO.
On the other hand, some reports revealed that hydrogenrelated impurities can significantly affect the electrical
FIG. 3. Formation enthalpies of native defects (solid lines) as a function of
the Fermi level under (a) O-poor and (b) O-rich limits. Those of hydrogenrelated impurities (dashed and dotted lines) are also shown.
J. Appl. Phys. 119, 165701 (2016)
properties of oxide semiconductors.47,48 Here, we theoretically examine the role of hydrogen-related impurities including H interstitials (Hi), VPb–H, and VPb–2H complexes, and
calculated DH for these defects are also plotted in Fig. 3. Hi
acts as a shallow donor and has negative DH values in a
wide range in the band gap, which is typical for oxide compounds.49,50 Such results indicate Hi should be formed in the
p-type region, because the interstitial H atoms (Hi) are
located near oxygen in these compounds and form -O–H
bonds (Figure 1(a)), which are more stable than the H–H
bonds for H2 molecules. On the other hand, it should be
noted that the DH of the VPb–H complex is 0.9 eV smaller
than the sum of DH values for its parent defects VPb and Hi,
and moreover the DH of the VPb–2H complex is 0.6 eV
smaller than the sum of DH values for VPb–H and Hi. It indicates that incorporated H atoms compensate VPb by forming
neutral VPb–2H complexes rather than forming Hi (in other
words, two H atoms passivate VPb). After passivating all the
VPb, the Hi acts as donors and raises the EF towards to the
CBM. However, the DH of Hi increases with raising the EF,
and it will become higher than the enthalpy of the reduction
reaction H2 þ PbO ! Pb þ H2O (–0.13 eV/H atom) when EF
is raised to 1.7 eV above the VBM (i.e., the H doping limit).
This means that further incorporation of H will reduce PbO
to Pb and H2O, rather than further producing Hi donors for
n-type conduction. Therefore, it is concluded that hydrogenrelated impurities are also unlikely to contribute enough
carriers to enhance the electrical conductivity of b-PbO.
The origin of the deep nature of defect states in b-PbO is
explained qualitatively based on molecular orbital theory,51
with a focus on the cation vacancies. For comparison, other
related oxide compounds with ns2 electronic configurations
(SnO and a-PbO) are also discussed. Figures 4(a) and 4(b)
show the calculated band structures and densities of states
(DOSs) for SnO, a-PbO, and b-PbO, in which the energy is
aligned by the O 2s core level. The simplified orbital energy
levels are schematically illustrated in Figure 4(c). Clearly, the
VBM of SnO is composited of the antibonding states of the
Sn 5s and O 2p states. However, the VBM would be formed
only by the O 2p–O 2p antibonding states if there was no contribution of Sn 5s orbitals. Consequently, once VSn is formed,
its energy level would be located far above the O 2p band
(i.e., deep in the band gap). Actually, the strong antibonding
interaction between the Sn 5s and O 2p orbitals raises the
VBM to a similarly higher level, causing the VSn states locate
slightly above the VBM, which is considered to be the main
reason for the strong p-type conduction in SnO.6,13,17,52
Unfortunately, the Pb 6s–O 2p antibonding interaction in
a-PbO is weaker than the Sn 5s–O 2p antibonding interaction
in SnO because the Pb 6s orbital is much deeper than the Sn
5s orbital,42 as qualitatively shown in Figs. 4(a) and 4(b). As
a result, the VBM level of a-PbO becomes deeper than that
of SnO even though a-PbO is isostructural to SnO, causing
the deeper VPb states. For b-PbO, the VBM is composed
mainly of the O 2p–O 2p antibonding states and the Pb 6s–O
2p antibonding interaction is even weaker than the a-PbO
case. Thus, as shown in the right panel of Figure 4(c), the VPb
states in b-PbO would be located much deeper in the band
gap. From these results, we conclude that the weak Pb 6s–O
Liao et al.
FIG. 4. Comparison in electronic structure between SnO and PbO. (a) Band
structures of SnO (left), a-PbO (middle), and b-PbO (right). (b) Total and
projected DOSs of SnO (top), a-PbO (middle), and b-PbO (bottom). The
energies are aligned by the O 2s core level for comparison. (c) Simplified
energy diagrams depicting the formation of VBM, CBM, and acceptor-like
cation vacancies in SnO (left), a-PbO (middle), and b-PbO (right).
2p antibonding interaction is responsible for the deep defect
transitions in b-PbO, similar to other poor electrically conductive metal oxides such as Bi2O3.53 The average effective
hole masses (mh*) for SnO, a-PbO, and b-PbO calculated
from the dispersions of the VBM bands (Fig. 4(a)) are 2.1m0,
3.0m0, and 5.8m0, respectively. It is known that the hole
mobility is primarily determined by lh ¼ es/mh* (rigid band
mobility. e is the elementary electric charge and s is the momentum relaxation time). In addition, a flat band and a flexible crystal structure like the layered one of b-PbO cause
small polaron hopping conduction, making the hole effective
mass much larger and the hole mobility much smaller.54
Thus, the exceptionally large mh* value of b-PbO further
J. Appl. Phys. 119, 165701 (2016)
explains the poor electrical conductivity of b-PbO even if
holes are formed.
Figures 5(a)–5(c) show lHall, N, and r of the b-PbO films
annealed in an ozone atmosphere at 200 C for different tann.
The signs of all the Hall coefficients were negative, suggesting n-type conduction in the annealed PbO films. As the tann
increases from 3 to 120 min, the Hall mobility in the annealed
PbO films varies slightly, while the electron concentration in
the annealed PbO films increases obviously from 3.3 1020
to 1.0 1021 cm3. The electrical conductivity increases to
7.1 102 S cm1 after annealing the PbO film in the ozone
atmosphere at 200 C for 3 min, and the electrical conductivity increases further with further increase in tann. Since there
has been only PbO2 that is reported to exhibit a high electrical
conductivity at room temperature in the lead oxide family,23,55 we should consider the possibility that the significant
enhancement of electrical conductivity in the annealed PbO
films originates from the formation of the PbO2 phase.
To clarify the formation of PbO2 phase in the annealed
PbO films, the crystal structures of the annealed PbO films
were characterized by XRD. Figure 5(d) shows the out-ofplane XRD patterns of the annealed PbO films. One can see
in the out-of-plane XRD patterns that the full widths at half
maximum of the 010 and 030 peaks corresponding to the
b-PbO phase increase, while the relative intensities of
these two peaks decrease as tann is increased from 3 min.
Meanwhile, the 002 peak of PbO2 phase appears, and the
020 peak of b-PbO phase shifts toward the lower diffraction
angles corresponding to the 020 peak of the PbO2 phase. As
tann is further increased, the 010 and 030 peaks corresponding to the b-PbO phase disappear and the relative intensity of
the 002 peak corresponding to the PbO2 phase increases further. Figure 5(e) shows the in-plane XRD patterns of the
annealed PbO films. In accordance with the out-of-plane
XRD results, the relative intensities of the peaks corresponding to the b-PbO phase decrease, while the peaks corresponding to the PbO2 phase appear and become stronger as
tann increases. These results substantiate that PbO2 phase is
formed in the annealed PbO films and the content of the
PbO2 phase increases with increasing annealing time.
It is known that the Pb 4f core levels of PbO2 phase are
characterized by components with the well-screened and
poorly screened final states.56,57 To further verify the formation of PbO2 phase, core level spectra of the as-grown PbO
film and the annealed PbO film were measured by HAXPES,
as shown in Figs. 5(f) and 5(g). For the as-grown PbO film,
the Pb 4f core levels exhibit two well-defined and symmetric
peaks at 137.3 eV and 142.2 eV, which are consistent with
the reported values for Pb 4f7/2 and Pb 4f5/2 core levels of
Pb2þ, respectively.57,58 After annealing, the characteristic
peaks associated with the well-screened and poorly screened
final states of PbO2 appear, which further indicates the formation of the PbO2 phase in the annealed films. The O 1s
peak (Fig. 5(g)) for the as-grown PbO film can be decomposed into two peaks at 528.8 eV and 530.8 eV, which are
attributed to the lattice oxygen of PbO phase and the oxygen
in adsorbed water-hydroxyl complexes (OH (þH2O)),
respectively.57–59 The O 1s peak for the annealed PbO film
has three components, one at 528.8 eV is ascribed to the
Liao et al.
J. Appl. Phys. 119, 165701 (2016)
FIG. 5. (a) lHall, (b) N, (c) r, (d) out of
plane XRD patterns, and (e) in-plane
XRD patterns of the PbO films
annealed in an ozone atmosphere at
200 C for different annealing times.
The Pb 4f (f) and O 1s (g) core-level
HAXPES spectra of as-grown and 3min-annealed PbO films.
lattice oxygen of PbO2 phase and the other two at the higher
binding energies are assigned to adsorbed oxygen species
(OH and OH (þH2O)).56–59 Therefore, it is concluded that
the PbO2 phase is formed in the annealed PbO films and the
enhancement of the conductivity in the annealed PbO films
should be attributed to the formation of PbO2 phase.
The oxygen pressure dependence of structures, optical,
and electrical properties of b-PbO films grown at 300 C on
silica glass substrates was investigated. The results showed
that all the as-grown b-PbO films exhibited poor electrical
conductivities (r < 1.4 107 S cm1). Our DFT calculations revealed that the transition levels of Pb vacancy are
deep in b-PbO because of the weak Pb 6s–O 2p antibonding
interaction, which would be the main reason for the poor
charge conducting properties of b-PbO. The situation differs
totally from that in SnO. This striking difference primarily
comes from the difference in orbital interaction between Pb
6s–O 2p and Sn 5s–O 2p. The latter stronger interaction
pushes up the VBM primarily composed of the antibonding
Sn 5s–O 2p chemical bond, which results in hole generation
through the formation of Sn vacancy. Such situation is not
realized for the VBM of PbO due to weak interaction of Pb
6s–O 2p orbitals. Furthermore, the hole effective mass in
b-PbO is much larger than those of SnO and a-PbO, making
hole conduction much more difficult even if holes are doped.
After annealing in an ozone atmosphere at 200 C for 3 min,
the PbO film exhibited strong n-type conduction with an
enhanced electrical conductivity of 7.1 102 S cm1 and a
high electron concentration of 3.3 1020 cm3. Moreover,
the electrical conductivity and the electron concentration
increase further with further increase in annealing time. This
phenomenon is primarily related to the formation of the
PbO2 phase in the annealed b-PbO films because the chemical potential of O exceeds the boundary limit.
This work was conducted under Tokodai Institute for
Element Strategy (TIES) funded by MEXT Elements
Strategy Initiative to Form Core Research Center. The
HAXPES measurements were performed with the approval
of NIMS Synchrotron X-ray Station (Proposal Nos.
2012B4612, 2013A4714, 2013A4715, 2013B4703, and
2013B4704). The authors would like to thank the staffs of
BL15XU, NIMS, and SPring-8 for their kind help. S.U. is
grateful to HiSOR, Hiroshima University, and JAEA/SPring8 for the development of HAXPES at BL15XU.
K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, and H. Hosono,
Nature 432, 488 (2004).
H. Hosono, Jpn. J. Appl. Phys. 52, 090001 (2013).
M. Quintana, T. Edvinsson, A. Hagfeldt, and G. Boschloo, J. Phys. Chem.
C 111, 1035 (2007).
J. Meng, P. Jena, and J. M. Vail, J. Phys.: Condens. Matter 2, 10371
H. Kawazoe, M. Yasukawa, H. Hyodo, M. Kurita, H. Yanagi, and H.
Hosono, Nature 389, 939 (1997).
H. Hosono, Y. Ogo, H. Yanagi, and T. Kamiya, Electrochem. Solid-State
Lett. 14, H13 (2011).
H. Kawazoe, H. Yanagi, K. Ueda, and H. Hosono, MRS Bull. 25, 28
M. Snure and A. Tiwari, Appl. Phys. Lett. 91, 092123 (2007).
K. Ueda and H. Hosono, J. Appl. Phys. 91, 4768 (2002).
H. Hiramatsu, K. Ueda, H. Ohta, M. Hirano, T. Kamiya, and H. Hosono,
Appl. Phys. Lett. 82, 1048 (2003).
U. V. Waghmare, N. A. Spaldin, H. C. Kandpal, and R. Seshadri, Phys.
Rev. B 67, 125111 (2003).
X. Q. Pan and L. Fu, J. Electroceram. 7, 35 (2001).
A. Togo, F. Oba, I. Tanaka, and K. Tatsumi, Phys. Rev. B 74, 195128
W. Guo, L. Fu, Y. Zhang, K. Zhang, L. Y. Liang, Z. M. Liu, H. T. Cao,
and X. Q. Pan, Appl. Phys. Lett. 96, 042113 (2010).
N. F. Quackenbush, J. P. Allen, D. O. Scanlon, S. Sallis, J. A. Hewlett, A.
S. Nandur, B. Chen, K. E. Smith, C. Weiland, D. A. Fischer, J. C. Woicik,
B. E. White, G. W. Watson, and L. F. J. Piper, Chem. Mater. 25, 3114
Y. Ogo, H. Hiramatsu, K. Nomura, H. Yanagi, T. Kamiya, M. Hirano, and
H. Hosono, Appl. Phys. Lett. 93, 032113 (2008).
Y. Ogo, H. Hiramatsu, K. Nomura, H. Yanagi, T. Kamiya, M. Kimura, M.
Hirano, and H. Hosono, Phys. Status Solidi A 206, 2187 (2009).
E. Fortunato, R. Barros, P. Barquinha, V. Figueiredo, S.-H. K. Park, C.-S.
Hwang, and R. Martins, Appl. Phys. Lett. 97, 052105 (2010).
P.-C. Hsu, W.-C. Chen, Y.-T. Tsai, Y.-C. Kung, C.-H. Chang, C.-J. Hsu,
C.-C. Wu, and H.-H. Hsieh, Jpn. J. Appl. Phys., Part 1 52, 05DC07 (2013).
L. Y. Liang, Z. M. Liu, H. T. Cao, Z. Yu, Y. Y. Shi, A. H. Chen, H. Z.
Zhang, Y. Q. Fang, and X. L. Sun, J. Electrochem. Soc. 157, H598 (2010).
Y.-J. Han, Y.-J. Choi, C.-Y. Jeong, D. Lee, S.-H. Song, and H.-I. Kwon,
IEEE Electron Device Lett. 36, 466 (2015).
Liao et al.
J. A. Caraveo-Frescas, P. K. Nayak, H. A. Al-Jawhari, D. B. Granato, U.
Schwingenschl€ogl, and H. N. Alshareef, ACS Nano 7, 5160 (2013).
H. Mizoguchi, H. Kawazoe, and H. Hosono, Chem. Mater. 8, 2769 (1996).
H. J. Terpstra, R. A. de Groot, and C. Haas, Phys. Rev. B 52, 11690 (1995).
K. Iinuma, T. Seki, and M. Wada, Mater. Res. Bull. 2, 527 (1967).
S. Venkataraj, O. Kappertz, R. Drese, C. Liesch, R. Jayavel, and M.
Wuttig, Phys. Status Solidi A 194, 192 (2002).
J. C. Schottmiller, J. Appl. Phys. 37, 3505 (1966).
K. Wasa and S. Hayakawa, Jpn. J. Appl. Phys., Part 1 8, 276 (1969).
G. Benassi, N. Zambelli, M. Villani, D. Calestani, M. Pavesi, A.
Zappettini, L. Zanotti, and C. Paorici, Cryst. Res. Technol. 48, 245 (2013).
H. Zhang and J. Ouyang, Org. Electron. 12, 1864 (2011).
D. P. Boden, J. Power Sources 73, 56 (1998).
J. Berashevich, J. A. Rowlands, and A. Reznik, J. Phys.: Condens. Matter
25, 475801 (2013).
J. Berashevich, J. A. Rowlands, and A. Reznik, EPL 102, 47002 (2013).
K. Ide, Y. Kikuchi, K. Nomura, M. Kimura, T. Kamiya, and H. Hosono,
Appl. Phys. Lett. 99, 093507 (2011).
S. Ueda, J. Electron Spectrosc. Relat. Phenom. 190, 235 (2013).
S. Ueda, Y. Katsuya, M. Tanaka, H. Yoshikawa, Y. Yamashita, S.
Ishimaru, Y. Matsushita, K. Kobayashi, R. Garrett, I. Gentle, K. Nugent,
and S. Wilkins, AIP Conf. Proc. 1234, 403 (2010).
G. Kresse, Phys. Rev. B 54, 11169 (1996).
J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett. 77, 3865
J. Heyd, G. E. Scuseria, and M. Ernzerhof, J. Chem. Phys. 118, 8207
C. G. Van de Walle, J. Appl. Phys. 95, 3851 (2004).
D. B. Laks, C. G. Van de Walle, G. F. Neumark, P. E. Bl€
ochl, and S. T.
Pantelides, Phys. Rev. B 45, 10965 (1992).
M. Liao, Z. Xiao, F.-Y. Ran, H. Kumomi, T. Kamiya, and H. Hosono,
ECS J. Solid State Sci. Technol. 4, Q26 (2015).
J. Appl. Phys. 119, 165701 (2016)
A. Zunger, Appl. Phys. Lett. 83, 57 (2003).
J. Robertson and S. J. Clark, Phys. Rev. B 83, 075205 (2011).
Z. Xiao, F.-Y. Ran, H. Hosono, and T. Kamiya, Appl. Phys. Lett. 106,
152103 (2015).
F.-Y. Ran, Z. Xiao, Y. Toda, H. Hiramatsu, H. Hosono, and T. Kamiya,
Sci. Rep. 5, 10428 (2015).
Y. Hanyu, K. Domen, K. Nomura, H. Hiramatsu, H. Kumomi, H. Hosono,
and T. Kamiya, Appl. Phys. Lett. 103, 202114 (2013).
J. B. Varley, A. Schleife, A. Janotti, and C. G. Van de Walle, Appl. Phys.
Lett. 103, 082118 (2013).
F. Oba, A. Togo, I. Tanaka, J. Paier, and G. Kresse, Phys. Rev. B 77,
245202 (2008).
S. Limpijumnong, P. Reunchan, A. Janotti, and C. G. Van de Walle, Phys.
Rev. B 80, 193202 (2009).
W.-J. Yin, S.-H. Wei, M. M. Al-Jassim, and Y. Yan, Appl. Phys. Lett. 99,
142109 (2011).
J. P. Allen, D. O. Scanlon, L. F. J. Piper, and G. W. Watson, J. Mater.
Chem. C 1, 8194 (2013).
A. Walsh, D. J. Payne, R. G. Egdell, and G. W. Watson, Chem. Soc. Rev.
40, 4455 (2011).
H. Fr€
ohlich, Adv. Phys. 3, 325 (1954).
W. Mindt, J. Electrochem. Soc. 116, 1076 (1969).
D. J. Payne, G. Paolicelli, F. Offi, G. Panaccione, P. Lacovig, G. Beamson,
A. Fondacaro, G. Monaco, G. Vanko, and R. G. Egdell, J. Electron
Spectrosc. Relat. Phenom. 169, 26 (2009).
D. J. Payne, R. G. Egdell, D. S. L. Law, P.-A. Glans, T. Learmonth, K. E.
Smith, J. Guo, A. Walsh, and G. W. Watson, J. Mater. Chem. 17, 267
C. Malitesta, L. Sabbatini, P. G. Zambonin, L. P. Bicelli, and S. Maffi,
J. Chem. Soc., Faraday Trans. 1 85, 1685 (1989).
S. Yamamoto, H. Bluhm, K. Andersson, G. Ketteler, H. Ogasawara, M.
Salmeron, and A. Nilsson, J. Phys.: Condens. Matter 20, 184025 (2008).
Без категории
Размер файла
1 774 Кб
Пожаловаться на содержимое документа