Content uploaded by Oliver Bierwagen
Author content
All content in this area was uploaded by Oliver Bierwagen on Jan 24, 2023
Content may be subject to copyright.
J. Appl. Phys. 133, 045701 (2023); https://doi.org/10.1063/5.0131138 133, 045701
© 2023 Author(s).
Epitaxial synthesis of unintentionally doped
p-type SnO (001) via suboxide molecular
beam epitaxy
Cite as: J. Appl. Phys. 133, 045701 (2023); https://doi.org/10.1063/5.0131138
Submitted: 18 October 2022 • Accepted: 05 January 2023 • Published Online: 24 January 2023
Published open access through an agreement with Technische Informationsbibliothek
Kingsley Egbo, Esperanza Luna, Jonas Lähnemann, et al.
Epitaxial synthesis of unintentionally doped
p
-type
SnO (001) via
suboxide
molecular beam epitaxy
Cite as: J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138
View Online Export Citation CrossMar
k
Submitted: 18 October 2022 · Accepted: 5 January 2023 ·
Published Online: 24 January 2023
Kingsley Egbo,
1,a)
Esperanza Luna,
1
Jonas Lähnemann,
1
Georg Hoffmann,
1
Achim Trampert,
1
Jona Grümbel,
2
Elias Kluth,
2
Martin Feneberg,
2
Rüdiger Goldhahn,
2
and Oliver Bierwagen
1,b)
AFFILIATIONS
1
Paul-Drude-Institut für Festkörperelektronik, Leibniz-Institut im Forschungsverbund Berlin e.V., Hausvogteiplatz 5-7,
10117 Berlin, Germany
2
Institut für Physik, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2, 39106 Magdeburg, Germany
a)
Author to whom correspondence should be addressed: egbo@pdi-berlin.de
b)
bierwagen@pdi-berlin.de
ABSTRACT
By employing a mixed SnO
2
+ Sn source, we demonstrate suboxide molecular beam epitaxy (S-MBE) growth of phase-pure single-crystalline
metastable SnO (001) thin films on Y-stabilized ZrO
2
(001) substrates at a growth rate of ∼1.0 nm/min without the need for additional
oxygen. These films grow epitaxially across a wide substrate temperature range from 150 to 450 °C. Hence, we present an alternative
pathway to overcome the limitations of high Sn or SnO
2
cell temperatures and narrow growth windows encountered in previous MBE
growth of metastable SnO. In situ laser reflectometry and line-of-sight quadrupole mass spectrometry were used to investigate the rate of
SnO desorption as a function of substrate temperature. While SnO ad-molecule desorption at T
S
= 450 °C was growth-rate limiting, the SnO
films did not desorb at this temperature after growth in vacuum. The SnO (001) thin films are transparent and unintentionally p-type
doped, with hole concentrations and mobilities in the range of 0.9–6.0 × 10
18
cm
−3
and 2.0–5.5 cm
2
V
−1
s
−1
, respectively. These p-type SnO
films obtained at low substrate temperatures are promising for back-end-of-line (BEOL) compatible applications and for integration with n-
type oxides in pn heterojunctions and field-effect transistors.
© 2023 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://
creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/5.0131138
I. INTRODUCTION
Tin (II) oxide (SnO) is a valuable p-type oxide material useful
in several technological applications, such as in pn diodes, transis-
tors, solar cells, and solid-state gas sensing.
1,2
It is among the few
binary oxide materials that show unintentional p-type character
due to native defects.
3–7
Though SnO has a low indirect fundamen-
tal bandgap of 0.7 eV, an optical bandgap of ∼2.7–2.9 eV makes it
a suitable component for several transparent applications.
8,9
SnO
crystallizes in a layered tetragonal litharge structure with space
group P4/nmm, with four oxygen atoms and a Sn atom forming a
pyramid structure.
10,11
Due to a more dispersed valence band
maximum composed of hybridized Sn 5sand the O 2porbitals,
12,13
reasonably high mobilities compared to most p-type oxides have
been reported for polycrystalline and single-crystalline SnO thin
films.
14–16
As most metal oxides show a propensity for n-type con-
ductivity and due to the difficulty in their bipolar doping, efforts in
oxide electronics mostly depend on oxide heterojunctions. Hence,
SnO has become a major p-type oxide for oxide heterojunctions
due to its high mobility compared to other p-type oxides.
17–21
Unlike tin (IV) oxide (SnO
2
), SnO is metastable, and this pre-
sents a growth challenge with competing stable metallic Sn or SnO
2
as well as Sn
3
O
4
phases, which may coexist along with SnO phase
in the thin films. Though several growth techniques have been
employed for the growth of polycrystalline SnO,
20
phase-pure
single crystal SnO layers have been mostly obtained by electron
beam evaporation,
22
pulsed laser deposition,
8
and molecular beam
epitaxy. High-quality single-crystalline SnO has been grown by
plasma-assisted MBE (PA-MBE) using a metal Sn source
23
and
S-MBE using an oxide SnO
2
source.
24
However, both methods
present challenges; the plasma-assisted growth of SnO using a Sn
metal source requires the Sn effusion cell to be operated at very
high temperatures up to 1175 °C and the growth window is limited
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-1
©Author(s)2023
by the formation of SnO
2
and oxygen rich Sn compounds, such as
Sn
3
O
4
, requiring a rigorous fine tuning of a Sn/O flux ratio.
23
Also,
growth using an oxide SnO
2
charge involves high temperature
decomposition of the oxide into the SnO suboxide and a parasitic
oxygen background (2 SnO
2
→2 SnO + O
2
). Previously, it has been
proposed that the sublimation of a mixed oxide + metal charge in
an effusion cell can provide an effective suboxide flux for the
growth of oxides by MBE.
25
For instance, some of us have shown
that a reaction of mixed SnO
2
+ Sn metal charge to give SnO
(SnO
2
+Sn→2 SnO) as well as a non-negligible fraction of Sn
2
O
2
,
can be an efficient source of SnO flux, for suboxide-MBE growth of
SnO
2
.
25
This suboxide approach offers the advantages that the reac-
tion in the source takes place at lower effusion-cell temperatures,
free of parasitic oxygen formation, and growth of SnO can proceed
without the need to provide additional oxygen. It remains to be
clarified, however, if the fraction of Sn
2
O
2
in the suboxide flux, not
present when using pure SnO
2
source material, is detrimental for
the SnO growth.
Following this suboxide-MBE (S-MBE) approach using a
mixed source, we present in this study the growth of SnO thin films
without the need for plasma-activated oxygen or intentional back-
ground oxygen. SnO is grown heteroepitaxially on Y-stabilized ZrO
2
[YSZ (001)] and r-plane Al
2
O
3
in a substrate temperature, T
S
window from 50 to 650 °C. We investigate the impact of T
S
on the
phase purity, growth-rate-limiting SnO desorption as well as struc-
tural and transport properties of the SnO layers. A schematic com-
parison of the conventional method of PA-MBE using a metal
source and the S-MBE approach employed in this work is described
in Figs. 1(a) and 1(b), respectively. Detailed investigation of the
growth kinetics for our SnO growth using this S-MBE approach
indicates that the growth rate is limited by the desorption of the
adsorbed SnO molecules, which increase with the substrate tempera-
ture; however, grown single SnO crystalline layers are stable and
show negligible desorption at substrate temperatures ≤450 °C. We
show that while amorphous layers were obtained at T
S
= 50 °C, tex-
tured and single-crystalline phase-pure unintentionally (UID) doped
p-type SnO(001) films are grown between substrate temperatures of
150 and 450 °C. Above 550 °C, secondary Sn
3
O
4
and Sn phases are
present in the SnO layer. Obtained single-crystalline SnO(001)
layers at low T
S
between 150 and 250 °C show good UID transport
properties. It is well known that epitaxial and single-crystalline
channel layers result in superior qualities in devices such as thin
film transistors; however, epitaxial deposition of active layers usually
requires high substrate temperatures that exceed the BEOL limit.
Hence, these epitaxially grown SnO thin films obtained for T
S
between 150 and 250 °C can be promising for applications as active
layers for BEOL compatible device development.
II. EXPERIMENTAL
Approximately 100–190 nm-thick UID SnO(001) thin films
were deposited in an MBE chamber with solid-source effusion
cells. While most growth were performed on YSZ (001), r-plane
Al
2
O
3
substrates were also co-loaded for several growth runs. To
grow SnO, a SnO
2
+ Sn mixed source was sublimed from an effu-
sion cell with an Al
2
O
3
crucible at temperatures between 740
and 820 °C. The hot-lip of the used dual-filament cell was kept at
150 °C above the SnO
2
+ Sn-cell temperature. The resulting source
beam equivalent pressure (BEP), proportional to the particle flux
and measured using a nude filament ion gauge positioned at the
substrate location, is about 0.4–1.5 × 10
−7
mbar for all growth runs.
Before growth, the substrates, with 1 μm thick Ti sputter-deposited
on the backside to improve substrate radiative heating, were
oxygen-plasma treated in the growth chamber for 30 min at sub-
strate temperatures, T
S
between 400 and 700 °C using 1 standard
cubic centimeters per minute (SCCM) O
2
and 200 W plasma
power. The T
S
monitored in situ by a thermocouple between the
substrate and heating filament was varied between 50 and 650 °C
for different growth runs. The background pressure of the chamber
during SnO deposition was P
GC
∼5–8×10
−8
mbar without
plasma-activated oxygen and any intentional molecular oxygen.
The growth rate and the amount of desorbing flux were measured
in situ by laser reflectometry (LR) and line-of-sight quadrupole
mass spectrometry (QMS).
26
Different ex situ techniques were used to characterize the
grown SnO layers. A four-circle x-ray lab-diffractometer (X’pert Pro
MRD from Philips PANalytical) equipped with a Cu Kαradiation
source was used to investigate the crystallographic orientation of the
film and the epitaxial relationship with the substrate. The
out-of-plane orientation was analyzed by means of symmetric
on-axis 2Θ-ωscans with a 1 mm detector slit. The in-plane epitaxial
relationship between the SnO film and the YSZ substrates was mea-
sured by Φ-scans in a skew-symmetric geometry. On-axis rocking
curve ω-scans were used to investigate the crystalline quality of the
films and texture maps in a skew-symmetric geometry were used to
validate the phase purity. Bulk-sensitive room-temperature Raman
spectroscopy measurements in the backscattering geometry using a
solid-state laser at a wavelength of 473 nm were used to investigate
FIG. 1. (a) Schematic of the formation of SnO layers on a substrate using the
PA-MBE approach with Sn charge and (b) S-MBE approach using mixed
SnO
2
+ Sn charge. In the PA-MBE approach, a Sn metal cell operated at a very
high cell temperature supplies Sn metal fluxes, and a plasma source provides
activated oxygen leading to the formation of SnO on the film surface, while in
the S-MBE approach, SnO is formed in the crucible by the mixture and films
are deposited in a high vacuum without activated or molecular oxygen present.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-2
©Author(s)2023
the grown layers as described in Ref. 23. Surface morphology of the
grown layers was analyzed by atomic force microscopy (AFM) using
a Bruker Dimension Edge in the peak force tapping mode.
(Scanning) transmission electron microscopy [(S)TEM] was used to
study the film’s microstructure. TEM observations were made with
a JEOL 2100F microscope operating at 200 kV. Cross-sectional TEM
specimens were prepared for observation in both〈011〉projections
of the YSZ substrate using standard mechanical polishing and dim-
pling, followed by argon ion-milling, starting at 3.0 keV and finish-
ing at 1.5 keV. The film domain structures and orientation were
further investigated by top-view electron backscatter diffraction
(EBSD) measurements in a scanning electron microscope operated
at 15 kV. Hall measurements in the van der Pauw geometry were
used to investigate the transport properties of the layers grown at
different substrate temperatures. Because of the low mobilities
obtained for some of the grown layers, a magnetic sweep method
was used to extract reliable Hall coefficients.
27
To compare the
optical properties of grown layers with existing literature,
24
spectro-
scopic ellipsometry measurement and modeling were performed on
a layer grown on r-plane Al
2
O
3
at a substrate temperature of 400 °C.
III. RESULTS AND DISCUSSIONS
A. Thermodynamics of the suboxide source and SnO
growth window
Previously from a quadrupole mass spectrometry study, we
have demonstrated that a reaction of a mixed oxide + metal charge
can serve as an efficient source of suboxides such as Ga
2
Oand
SnO.
25
By taking advantage of this suboxide source, we determine
the growth window and thermodynamic consideration for the
growth of metastable p-type SnO from a SnO
2
+ Sn mixture.
23,28
Figure 2(a) shows the calculated SnO
2
–Sn equilibrium phase
diagram as a function of stoichiometry, n
Sn
/(n
Sn
+n
SnO2
) and tem-
perature calculated at 10
−7
mbar typical for MBE growth without O
2
environment. This thermodynamic equilibrium diagram calculated
using FactSage
29
indicates that stable SnO in the solid phase can be
obtained at growth temperatures within 190–420 °C at a stoichiome-
try of n
Sn
/(n
Sn
+n
SnO2
) = 0.5. Here, ideal gas indicates gaseous
species of the constituent elements in the reaction.
25
This supports
our experimental data where phase-pure SnO was obtained between
150 and 450°C substrate temperatures during growth as discussed
below. Due to the absence of reactive oxygen during growth, we
can obtain phase-pure SnO for Sn stoichiometry of 0.5 in
n
Sn
/(n
Sn
+n
SnO2
), which was used throughout this growth experi-
ment. Figure 2(b) shows the corresponding phase diagrams for the
mixed charges at Sn stoichiometry, n
Sn
/(n
Sn
+n
SnO2
) = 0.5 indicating
the vapor pressure of suboxide SnO as a function of source tempera-
tures for the mixed charges. These mixed sources promote the avail-
ability of required suboxide vapor pressure at lower source
temperatures compared to the metal charge and solid oxide charge.
25
B. SnO flux, rate of desorption, and disproportionation
of the layer
In the identified growth window for SnO suboxide growth, the
growth rate is given by the difference in the amount of arriving
SnO species (proportional to the vapor pressure of the SnO
FIG. 2. (a) Equilibrium phase diagram of the SnO
2
–Sn system as a function of stoichiometry and temperature at a pressure of 10
−7
mbar typical for MBE growth without
intentional oxygen background. Stoichiometries of n
Sn
/(n
Sn
+n
SnO2
) =0, 0.5 and 1 correspond to SnO
2
; SnO and pure Sn, respectively. (b) Phase diagram of the SnO
2
–Sn
system for n
Sn
/(n
Sn
+n
SnO2
) = 0.5 as a function of temperature and pressure.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-3
©Author(s)2023
[cf. Figure 2(b)] at the cell temperature of the mixed source) and
the amount of desorbing species (which increases with increasing
substrate temperature). This promotes a simpler growth kinetics
compared to SnO growth by PA-MBE in which rigorous Sn/O flux
calibration is required to limit the formation of competing O-rich
and Sn-rich phases. The SnO flux impinging on a substrate from
the mixed SnO
2
+ Sn effusion cell is equivalent to the measured
BEP in the absence of an oxygen background. In order to charac-
terize the kinetics of the mixed SnO
2
+ Sn effusion cell, we mea-
sured the BEP as a function of source temperature from 740 to
800 °C in the absence of any active O
2
flow (see Fig. 3). The curve
shows the expected exponential dependence with an activation
energy for SnO of 2.4 eV, similar to the value obtained by our pre-
vious QMS results.
25
The incorporated cation flux on the substrate
(solid symbols in Fig. 3, obtained as the product of the measured
growth rate and the cation density) is proportional to the BEP as
expected for full cation incorporation due to the low growth tem-
perature of 50 °C, i.e., without desorption from the substrate.
Using line-of-sight quadrupole mass spectrometry, we identify
the rate of desorption of the SnO ad-molecules during growth as a
function of substrate temperature. Our measurement reveals that
the desorbing flux significantly decreases with decreasing substrate
temperature as shown in Fig. 4(a). The inset of Fig. 4(a) shows a
typical QMS spectrum recorded during SnO desorption at high
substrate temperatures. For a flux of 4 × 10
13
cm
−2
s
−1
reaching
the substrate with a substrate temperature of 450 °C, a growth
rate of 1.0 nm/min is expected; however, a SnO growth rate of
∼0.6 nm/min is obtained from the LR oscillation in Fig. S1 in the
supplementary material; hence, the desorption rate of the SnO
ad-molecules is ∼0.3 nm/min (1.4 × 10
13
cm
−2
s
−1
). From the plot
of desorption rate as a function of substrate temperature, an activa-
tion energy of desorption of ∼0.3 eV is obtained, and this low acti-
vation energy value indicates a high volatility of the suboxide
ad-molecules during growth. The desorption rate at higher sub-
strate temperatures is limited by the amount of flux reaching the
substrate as seen in Fig. 4(a). While these ad-molecules are volatile
during growth, to further understand the stability of the SnO mole-
cules within the film after growth, we perform a separate LR study
at a SnO film grown at a BEP of 1.3 × 10
−7
mbar. From Fig. 3, this
BEP corresponds to a flux at the substrate of 8.6 × 10
13
cm
−2
s
−1
(assuming full incorporation) and to a growth rate of 1.85 nm/min.
However, Fig. 4(b) shows only a growth rate of 1.2 nm/min
obtained from the oscillatory half period t
growth
of the laser reflec-
tometry signal
26
for growth at 450 °C substrate temperature,
which is the highest temperature where phase-pure SnO is
obtained. Hence, the SnO ad-molecules have a desorption rate of
∼0.65 nm/min during growth corresponding to a desorbing flux of
3.0 × 10
13
cm
−2
s
−1
. When the SnO shutter is closed, and the sub-
strate temperature maintained at 450 °C the reflected laser signal is
constant indicating negligible desorption of stabilized SnO species
on the substrate. This is in contrast to the high desorbing flux of
3.0 × 10
13
cm
−2
s
−1
for the SnO ad-molecules during growth at
450 °C, indicating that once the SnO molecule stabilizes in a layer,
its activation energy for desorption is higher than for directly
re-evaporated SnO ad-molecules. Finally, with an increase in the
substrate temperature to 550 °C, an onset of an oscillation is
observed due to SnO desorption or increasing roughness due to the
disproportionation of the grown layer. Note that this reflected
signal oscillation has a weak amplitude compared to the growth
oscillation pointing to the roughening of the layer and formation of
other Sn-compound phases. Therefore, while SnO ad-molecules
desorb during growth as indicated in Fig. 4(a), SnO layers formed
are very stable with negligible desorption rate at the growth
temperature.
C. Structural properties and epitaxial relation
Wide-angle symmetric 2Θ-ωXRD scans were used to investi-
gate the out-of-plane orientation of the layers. Figure 5(a) shows
representative results for layers grown at different YSZ (001) sub-
strate temperatures, wide-angle scans of selected samples between
10 and 120° are shown in Fig. S2(a) in the supplementary material,
and similar wide-angle scans for a sample grown on r-plane Al
2
O
3
are shown in Fig. S2(b) in the supplementary material. Typical
streaky RHEED patterns observed during the growth of SnO (001)
layers are also shown in Fig. S3 in the supplementary material. The
layer grown at a substrate temperature of 50 °C was amorphous;
hence, only substrate peaks are observed in the XRD scan. For
layers grown between 150 and 450 °C, only the SnO (001) and
higher order reflexes as well as the YSZ (100) substrate peaks are
present, indicating phase-pure (001) oriented single-crystalline SnO
films. This is in contrast with previous results on plasma-assisted
MBE growth of SnO in which phase-pure SnO was only possible at
a substrate temperature between 350 and 400 °C.
23
A very limited
substrate temperature window was observed in previous S-MBE
SnO growth on r-plane Al
2
O
3
(1–102) substrates using an SnO
2
FIG. 3. Arrhenius diagram of the incorporated Sn flux (solid symbols) during
MBE growth and the corresponding measured BEPs of the mixed SnO
2
+Sn
effusion cell (right axis, open symbols). An incorporated flux at the substrate of
4.8 × 10
13
cm
−2
s
−1
corresponds to a SnO growth rate of 1.0 nm/min and to a
BEP of ∼7×10
−8
mbar.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-4
©Author(s)2023
source: Mei et al. reported that their films grown below 370 °C
were amorphous while no deposition occurred above 400 °C.
24
Furthermore, we observe slightly sharper peaks in the 2Θ-ωscans
of the samples grown at 250–450 °C, indicating higher crystal
quality than that of layers grown at a lower temperature. However,
the increase is not linear with increasing substrate temperatures.
Once the substrate temperature is increased even further, to 550°C,
the presence of Sn
3
O
4
secondary phase and Sn peaks are observed
FIG. 4. (a) Arrhenius diagram of the desorbing flux as a function of the substrate temperature. The desorbing species were obtained from the line-of-sight QMS. Their flux
increases with increasing substrate temperatures and is limited at high temperatures by the provided SnO flux to the substrates. (b) Plot of the laser reflectometry signal
(with substrate rotation causing the high-frequency oscillation artifact) from the film surface during growth and thermal etching at higher substrate temperatures. The
opening and closing of the SnO
2
+ Sn-cell shutter as well as substrate temperatures are indicated.
FIG. 5. (a) XRD out-of-plane symmetric 2Θ-ωscan of SnO(001) layer on YSZ(100) grown at different substrate temperatures. The film grown at 50 °C was amorphous,
between 150 and 450 °C, phase-pure single-crystalline SnO(001) layers are obtained. At a substrate temperature of 550°C, secondary Sn
3
O
4
phases are observed and
growth at 650 °C showed a mixed phase with negligible SnO(001) contribution. (b) FWHM of the SnO 002 omega rocking curve as a function of the substrate temperature.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-5
©Author(s)2023
indicating a disproportionation of the phase-pure SnO to Sn-rich
and O-rich phases. This trend is continued with a further increase
in a substrate temperature up to 650 °C, where the intensity of a
mixed phase (metallic Sn, Sn
3
O
4
) dominates the XRD and the SnO
peak intensity becomes negligible. Figure 5(b) shows the FWHM of
the SnO (002) omega rocking curves as a function of a substrate
temperature for the grown layers. The FWHM maximum ranges
from a high value of 1.3° for the sample grown at a low substrate
temperature of 150 °C to values between 0.2 and 0.7° for samples
grown at higher substrate temperatures. Film roughness between 2
and 15 nm [root mean square (rms)] is obtained from AFM mea-
surements for the single-crystalline SnO (001) layers.
The in-plane epitaxial relationship between the SnO (001)
layer and the substrate was investigated for the phase-pure SnO
(001) thin films by skew-symmetric Φ-scans of the SnO 112 and
YSZ 111 reflections with rotational angle Φaround the surface
normal (see Fig. S4 in the supplementary material). From Fig. 5(a),
we confirm the epitaxial relationship SnO(001)||YSZ(001) and SnO
(110)||YSZ(010) for the out-of-plane and in-plane directions,
respectively, indicating a 45° in-plane rotation of the SnO crystal
lattice with respect to the YSZ one. This is in agreement with our
previously grown PA-MBE thin films, where the same epitaxial
relations are observed.
23
To further verify the in-plane crystalline
texture of the grown layers, Fig. 6 shows the texture map of the
SnO 101 reflex. The texture map confirms the presence of a pre-
dominantly single domain as observed in the Φ-scans. The scan
shows four distinct peaks at a tilt angle of Ψ= 51.63° corresponding
to the tilt of the (101) plane with respect to the (001) planes. Other
Streaky peaks not labeled observed in the texture map are due to
the substrate holder as shown in Fig. S5 in the supplementary
material. Texture maps also show that distinct in-plane epitaxial
relationships are possible for these samples on YSZ(001) substrates
unlike randomly oriented samples observed for growths on Ga
2
O
3
substrates.
17,18
To clarify the phase composition of the grown epilayers beyond
the limited capabilities of our out-of-plane XRD scans, room-
temperature bulk-sensitive Raman spectroscopy measurements are
conducted on the grown layers. Following the peak assignment of
Eifert et al. (and references therein) and our previously reported
study on PA-MBE of SnO, the measured Raman spectra of all
S-MBE grown SnO films were compared to phonon modes due to
Sn, SnO, and Sn
3
O
423,30,31
as shown in Fig. 7. Raman spectra of the
single-crystalline samples grown between 150 and 450 °C show only
SnO B
1g
(113 cm
−1
)andA
1g
(211 cm
−1
) peaks strongly suggesting
that no secondary phases are present. In agreement with the x-ray
diffraction data, bulk-sensitive Raman scattering of the sample grown
at 550 °C indicates a weak contribution of the Sn
3
O
4
phase coexisting
with the SnO phase, while the sample grown at 650 °C showed pre-
dominantly Sn
3
O
4
peaks with weak contributions from Sn and SnO.
These samples at 550°C also show a slight shift in their SnO peak
positions likely due to the presence of these secondary phases.
Transmission electron microscopy (TEM) of the sample cross
section is used to investigate the microstructure of the SnO layer grown
at 400 °C, its interface to the YSZ substrate, and the epitaxial relation-
ship between the YSZ substrates and the SnO layers. The overview
bright field image of the SnO films as shown in Fig. 8(a) reveals a layer
composed of coalesced grains (their average diameter is about 150 nm),
a morphology that suggests a 3-D Volmer-Weber growth mode of SnO
on YSZ. The streak RHEED patterns obtained during growth shown
Fig. S3 in the supplementary material is,hence,likelyduetoreflections
FIG. 6. Texture scan along the (101) peak of SnO. Four peaks obtained for Φ:
0°–360° and Ψ:0°–90°. Peaks due to the (101) reflex are separated by 90° as
expected for the Wulff plot in Fig. S7 in the supplementary material.
FIG. 7. Bulk-sensitive Raman spectra measured with an excitation wavelength
of 473 nm. Vertical lines indicate the peak positions of dominant Raman active
phonon modes, in SnO B
1g
(113) and A
1g
(211) and in Sn
3
O
4
,A
g
and B
g
as
indicated by the color code.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-6
©Author(s)2023
FIG. 8. (a) Cross-sectional TEM image of the SnO(001) film grown at 450°C (b) High-angle annular dark-field (HAADF) STEM Z-contrast image of the SnO(001)/YSZ(001) hetero-
structure. (c) and (d) High-resolution transmission electron microscopy (HRTEM) images of the SnO(001) epilayer on YSZ(001) near the SnO/YSZ interface acquired along the [011]
zone axis of YSZ(100). Indexed reflections indicate an (001)
SnO
|| (001)
YSZ
and [010]
SnO
|| [1–10]
YSZ
epitaxial relationship as shown in (e) the FFT of the SnO/YSZ micrograph.
FIG. 9. (a) Hole concentration, p, and resistivity of phase-pure SnO (001) as a function of substrate temperature obtained from Hall measurements in the van der Pauw
geometry. (b) Hall mobilities of deposited thin films as a function of substrate temperature. The uncertainties for the hole concentration and mobility as shown by error bars
are approximately 2%–20%.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-7
©Author(s)2023
from the surface of these large grains and does not imply a
layer-by-layer growth mode. The grains have a single orientation and a
well-defined epitaxial relationship. Figure 8(b) displays a high-angle
annular dark-field (HAADF) STEM image of the SnO(001)/YSZ(001)
heterostructure with atomic number Z-contrast. In this image, SnO
shows a brighter contrast compared to YSZ due to its higher average Z.
Figure 8(c) displays high-resolution TEM (HRTEM) phase-contrast
micrographs of the SnO/YSZ(001) interface acquired along the [011]
zone axis of the substrate. Though HRTEM reveals the local presence
of steps at the SnO/YSZ interface, there is no noticeable misalignment
of the SnO(001) layer, which grows epitaxially on YSZ(001) following
the epitaxial relationship (001)
SnO
|| (001)
YSZ
and [010]
SnO
|| [1–10]
YSZ
.
The perfect epitaxial alignment is reflected in the Fast Fourier
Transform (FFT) pattern of the image in Fig. 8(d). The epitaxial rela-
tionship and crystallographic directions have been determined after a
comparison of the experimental data with simulated diffraction patterns
obtained using the JEMS simulation program.
32
Electron backscattering diffraction (EBSD) measured plan-
view in a scanning electron microscope was further used to investi-
gate the surface microstructure of our grown layers as the probing
depth of EBSD is limited to ∼20 nm. Figure S6(a) in the
supplementary material shows EBSD mapping of the SnO(001)
indicating uniformity of the crystal orientation without any crystal
twins on the layer. Figure S7(b) in the supplementary material
shows Kikuchi patterns from EBSD measurement and wireframe
representation of the indexed orientations.
D. Electrical properties of S-MBE grown UID SnO(001)
thin films
The charge carrier transport properties of the UID SnO (001)
thin films were obtained at room temperature by Hall
measurements in the van der Pauw geometry. For the van der
Pauw measurement, indium contacts are made on the corners of
the as grown UID SnO thin films prior to the measurement.
33
Due
to low hole mobilities observed in most p-type oxides, extracting
accurate Hall voltage using a single magnetic field value Bmay
become ambiguous. Here, to extract reliable Hall coefficients, a
Hall sweep between +0.8 and −0.8 T was carried out and the Hall
coefficient is obtained from the slope of the Hall voltage as shown
in Fig. S7 in the supplementary material. Varying room-
temperature UID hole densities, p
Hall
in the range of 0.9–
6.5 × 10
18
cm
−3
are obtained for the phase-pure SnO thin films
(S150–S450) as shown in Fig. 9(a). The sample grown at a low sub-
strate temperature of 150 °C shows a remarkable mobility of
2.2 cm
2
/V s, increasing the substrate temperatures resulted in the
growth of samples with average mobility of 4.5 cm
2
/V s as seen in
Fig. 9(b). Room-temperature electrical resistivities ρbetween 0.3
and 1.2 Ωcm are obtained for the deposited films. A slight spread
in transport properties for single-crystalline SnO (001) grown at
different temperatures may be related to the growth temperature
and crystallinity of the layers, but a consistent trend is not
observed. Obtained hole densities in our grown layers are about
two orders of magnitude higher than the value reported for UID
SnO grown by S-MBE from an SnO
2
source.
24
Table I summarizes
the room-temperature electrical properties for various reported
single-crystalline p-type SnO(001) thin films grown using different
techniques. The hole densities from these S-MBE grown layers are
also slightly lower than our previously reported values grown via
PA-MBE. This is likely due to the enhanced formation of Sn vacan-
cies due to the energetics of the different growth process. For
S-MBE growth, the SnO flux is reaching the substrate, this is
expected to decrease the formation of Sn vacancies and complexes
compared to the PA-MBE growth, where elemental Sn and
TABLE I. Comparison of electrical properties of epitaxial SnO(001) thin films grown using different techniques (S-MBE—suboxide MBE, PA-MBE—plasma-assisted MBE,
PLD—pulsed laser deposition, EBE—electron beam evaporation).
Material Method T
S
(°C)
Growth
P
O2
(Torr) Substrate
ω-FWHM
(deg) p
Hall
(cm
−3
)
Hole
mobility
(cm
2
/V s)
Resistivity
(Ωcm) Reference
SnO
(001)
S-MBE 380 5 × 10
−7
r-Al
2
O
3
0.007 2.5 × 10
16
2.4 101 24
SnO
(001)
PLD 575 1 × 10
−6
YSZ (001) 0.46 2.5 × 10
17
2.4 …8
SnO
(001)
PLD 200 6 × 10
−2
YSZ (001) 1.0 1.0 × 10
17
2.3 …35
SnO
(001)
EBE 600 …r-Al
2
O
3
2.9 5.6 × 10
17
0.1 110 9
SnO
(001)
EBE 600 ∼1×10
−6
r-Al
2
O
3
……195 22
SnO
(001)
PA-MBE 350–400 ∼5×10
−6
YSZ (001),
c-Al
2
O
3
0.4–1.9 1.8–9.7 × 10
18
1–6 0.25–2.0 23
SnO
(001)
S-MBE 150–450 ∼6×10
−8
YSZ (100) 0.2–1.3 0.9–6×10
18
2.5–5.5 0.3–1.2 This
Work
SnO
(001)
S-MBE 400 ∼6×10
−8
r-Al
2
O
3
1.3 7.0 × 10
17
1.4 7.2 This
Work
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-8
©Author(s)2023
activated oxygen is supplied. Nevertheless, we cannot rule out
unintentionally incorporated extrinsic dopants. To further control
the hole concentration of these SnO(001) thin films grown by sub-
oxide MBE, intentional extrinsic doping using Ga and La dopants
has been explored, transport measurements for these doped
samples show that Ga are efficient acceptor dopants while La acts
as compensating donors.
34
S550 films with dominant SnO also
showed p-type UID properties like the phase-pure SnO layers.
Amorphous layers grown at 50 °C were semi-insulating while the
mixed phase layer grown at 650 °C showed a n-type character.
To assess the optical properties of SnO layers grown via this
S-MBE route, spectroscopic ellipsometry measurement and modeling
is performed on ∼120 nm-thick phase-pure SnO(001) sample grown
on r-plane Al
2
O
3
at 400 °C. Figure S2(b) in the supplementary
material shows the wide-angle 2Θ-ωscan of the sample. Compared to
the sample grown on YSZ (001) at a similar substrate temperature,
Hall measurements for this sample show a slightly lower hole density
and mobility of ∼7.0 × 10
17
cm
−3
and ∼1.4 cm
2
/V s, respectively, and
higher resistivity of ∼7.2 Ωcm. This decrease in the transport proper-
ties is likely due to an increase in strain-induced dislocation density
caused by higher lattice mismatch of ∼12% in SnO/r-plane Al
2
O
3
hetero-interface compared to SnO/YSZ hetero-interface with ∼5%
lattice mismatch.
23,24
Ordinary and extraordinary complex dielectric
function (ε≡ε
1
+iε
2
) spectra of this thin film extracted from room-
temperature spectroscopic ellipsometry measurement are shown in
Fig. 10.Theε
2
spectrum in the ordinary direction shows an onset of
absorption at ∼2.7 eV similar to previously reported values.
24
IV. CONCLUSIONS
Using an intentional suboxide source comprising a mixed
SnO
2
+ Sn charge, we demonstrate the heteroepitaxial growth of
phase-pure, single-crystalline SnO (001) thin films on YSZ(001)
and r-Al
2
O
3
substrates. This S-MBE approach enabled the growth
of phase-pure SnO (001) films across a wider growth window and
substrate temperatures without plasma-activated oxygen or (un)
intentional molecular oxygen. Hence, overcoming the limitation of
the narrow growth window previously reported for PA-MBE
growth of SnO using a metal charge and S-MBE growth of SnO
using a SnO
2
source. We systematically characterized the S-MBE
growth kinetics, such as the growth rate and desorbing SnO fluxes
as a function of cell temperature, being an important step to
employing this suboxide approach to other material systems. Ex
situ XRD and Raman measurements showed that phase-pure
single-crystalline SnO is obtained for a wide substrate temperature
window between 150 and 450 °C. Transport and optical measure-
ments also confirm the p-type properties and optical transparency
of these layers. Hence, with the S-MBE approach, single-crystalline
phase-pure SnO(001) was achieved at the lowest substrate tempera-
ture of 150 °C so far. This possibility to achieve epitaxial single-
crystalline p-type SnO (001) thin films for our S-MBE grown
samples at low growth temperatures between 150 and 250 °C can
promote the integration of these p-type SnO layers for BEOL com-
patible device applications.
SUPPLEMENTARY MATERIAL
See the supplementary material for a typical RHEED image of
SnO layers acquired during growth. Wide-angle 2Θ-ωscans of SnO
(001) on YSZ(001) and r-Al
2
O
3
. Skew-symmetric Φ-scans of SnO
layer and YSZ substrate, EBSD mapping, and Hall magnetic field
sweep data.
ACKNOWLEDGMENTS
We would like to thank H.-P. Schönherr for MBE support,
D. Steffen for TEM sample preparation, and P. John for critically
reading the manuscript. This work was performed in the frame-
work of GraFOx, a Leibniz ScienceCampus partially funded by the
Leibniz Association.
AUTHOR DECLARATIONS
Conflict of Interests
The authors have no conflicts to declare.
Author Contributions
Kingsley Egbo: Conceptualization (equal); Data curation (lead);
Formal analysis (lead); Investigation (lead); Methodology (lead);
Writing –original draft (lead); Writing –review & editing (lead).
Esperanza Luna: Data curation (supporting); Formal analysis (sup-
porting); Investigation (supporting); Methodology (supporting);
Writing –original draft (supporting). Jonas Lähnemann: Formal
analysis (supporting); Investigation (supporting); Writing –review
& editing (supporting). Georg Hoffmann: Conceptualization (sup-
porting); Methodology (supporting). Achim Trampert: Formal
analysis (supporting); Investigation (supporting); Writing –review
& editing (supporting). Jona Grümbel: Data curation (supporting);
Formal analysis (supporting); Investigation (supporting); Writing –
original draft (supporting). Elias Kluth: Data curation
FIG. 10. Complex dielectric function of a SnO(001) thin film resolved into the
ordinary xy (solid lines) and extraordinary z(dashed lines) components obtained
from ellipsometry point-by-point fitting.
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-9
©Author(s)2023
(supporting); Investigation (supporting). Martin Feneberg: Formal
analysis (supporting); Investigation (supporting). Rüdiger
Goldhahn: Formal analysis (supporting); Investigation (support-
ing). Oliver Bierwagen: Conceptualization (equal); Data curation
(supporting); Formal analysis (supporting); Investigation (support-
ing); Project administration (lead); Resources (lead); Writing –
review & editing (supporting).
DATA AVAILABILITY
The data that support the findings of this study are available
within the article and its supplementary material.
REFERENCES
1
Z. Wang, P. K. Nayak, J. A. Caraveo-Frescas, and H. N. Alshareef, Adv. Mater.
28, 3831 (2016).
2
Y. Ogo, H. Hiramatsu, K. Nomura, H. Yanagi, T. Kamiya, M. Kimura,
M. Hirano, and H. Hosono, Phys. Status Solidi A 206, 2187 (2009).
3
A. Togo, F. Oba, I. Tanaka, and K. Tatsumi, Phys. Rev. B 74, 195128 (2006).
4
K. O. Egbo, C. P. Liu, C. E. Ekuma, and K. M. Yu, J. Appl. Phys. 128, 135705
(2020).
5
K. O. Egbo, C. E. Ekuma, C. P. Liu, and K. M. Yu, Phys. Rev. Mater. 4, 104603
(2020).
6
K. O. Egbo, M. Kong, C. P. Liu, and K. M. Yu, J. Alloys Compd. 835, 155269
(2020).
7
M. Nolan and S. D. Elliott, Phys. Chem. Chem. Phys. 8, 5350 (2006).
8
Y. Ogo, H. Hiramatsu, K. Nomura, H. Yanagi, T. Kamiya, M. Hirano, and
H. Hosono, Appl. Phys. Lett. 93, 032113 (2008).
9
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).
10
A. Seko, A. Togo, F. Oba, and I. Tanaka, Phys. Rev. Lett. 100, 045702 (2008).
11
J. P. Allen, D. O. Scanlon, S. C. Parker, and G. W. Watson, J. Phys. Chem. C
115, 19916 (2011).
12
A. Walsh and G. W. Watson, Phys. Rev. B 70, 235114 (2004).
13
D. B. Granato, J. A. Caraveo-Frescas, H. N. Alshareef, and
U. Schwingenschlögl, Appl. Phys. Lett. 102, 212105 (2013).
14
J. A. Caraveo-Frescas, P. K. Nayak, H. A. Al-Jawhari, D. B. Granato,
U. Schwingenschlögl, and H. N. Alshareef, ACS Nano 7, 5160 (2013).
15
M. Minohara, A. Samizo, N. Kikuchi, K. K. Bando, Y. Yoshida, and Y. Aiura,
J. Phys. Chem. C 124, 1755 (2020).
16
S. A. Miller, P. Gorai, U. Aydemir, T. O. Mason, V. Stevanović, E. S. Toberer,
and G. J. Snyder, J. Mater. Chem. C 5, 8854 (2017).
17
M. Budde, D. Splith, P. Mazzolini, A. Tahraoui, J. Feldl, M. Ramsteiner,
H. von Wenckstern, M. Grundmann, and O. Bierwagen, Appl. Phys. Lett. 117,
252106 (2020).
18
K. Tetzner, K. Egbo, M. Klupsch, R.-S. Unger, A. Popp, T.-S. Chou,
S. B. Anooz, Z. Galazka, A. Trampert, O. Bierwagen, and J. Würfl, Appl. Phys.
Lett. 120, 112110 (2022).
19
Z. Wang, P. K. Nayak, A. Albar, N. Wei, U. Schwingenschlögl, and
H. N. Alshareef, Adv. Mater. Interfaces 2, 1500374 (2015).
20
K. J. Saji, Y. P. Venkata Subbaiah, K. Tian, and A. Tiwari, Thin Solid Films
605, 193 (2016).
21
A. Parisini, P. Mazzolini, O. Bierwagen, C. Borelli, K. Egbo, A. Sacchi, M. Bosi,
L. Seravalli, A. Tahraoui, and R. Fornari, J. Vac. Sci. Technol. A 40, 042701
(2022).
22
X. Q. Pan and L. Fu, J. Electroceram. 7, 35 (2001).
23
M. Budde, P. Mazzolini, J. Feldl, C. Golz, T. Nagata, S. Ueda, G. Hoffmann,
F. Hatami, W. T. Masselink, M. Ramsteiner, and O. Bierwagen, Phys. Rev. Mater.
4, 124602 (2020).
24
A.B.Mei,L.Miao,M.J.Wahila,G.Khalsa,Z.Wang,M.Barone,N.J.Schreiber,
L.E.Noskin,H.Paik,T.E.Tiwald,Q.Zheng,R.T.Haasch,D.G.Sangiovanni,
L. F. J. Piper, and D. G. Schlom, Phys.Rev.Mater.3, 105202 (2019).
25
G. Hoffmann, M. Budde, P. Mazzolini, and O. Bierwagen, APL Mater. 8,
031110 (2020).
26
P. Vogt and O. Bierwagen, Appl. Phys. Lett. 106, 081910 (2015).
27
F. Werner, J. Appl. Phys. 122, 135306 (2017).
28
K. M. Adkison, S.-L. Shang, B. J. Bocklund, D. Klimm, D. G. Schlom, and
Z.-K. Liu, APL Mater. 8, 081110 (2020).
29
C. W. Bale, E. Bélisle, P. Chartrand, S. A. Decterov, G. Eriksson, A. E. Gheribi,
K. Hack, I.-H. Jung, Y.-B. Kang, J. Melançon, A. D. Pelton, S. Petersen,
C. Robelin, J. Sangster, P. Spencer, and M.-A. Van Ende, Calphad 55, 1 (2016).
30
B. Eifert, M. Becker, C. T. Reindl, M. Giar, L. Zheng, A. Polity, Y. He,
C. Heiliger, and P. J. Klar, Phys. Rev. Mater. 1, 014602 (2017).
31
F. Wang, X. Zhou, J. Zhou, T.-K. Sham, and Z. Ding, J. Phys. Chem. C 111,
18839 (2007).
32
P. A. Stadelm ann , JEMS (JEMS-SWISS, Chemin Rouge 15 CH-1805 Jongny, n.d.).
33
O. Bierwagen, T. Ive, C. G. Van de Walle, and J. S. Speck, Appl. Phys. Lett. 93,
242108 (2008).
34
K. Egbo, J. Lähnemann, A. Falkenstein, J. Varley, and O. Bierwagen,
arXiv:2212.06350 (2022).
35
H. Hayashi, S. Katayama, R. Huang, K. Kurushima, and I. Tanaka, Phys.
Status Solidi RRL 9, 192 (2015).
Journal of
Applied Physics ARTICLE scitation.org/journal/jap
J. Appl. Phys. 133, 045701 (2023); doi: 10.1063/5.0131138 133, 045701-10
©Author(s)2023
