Content uploaded by Oliver Bierwagen
Author content
All content in this area was uploaded by Oliver Bierwagen on Dec 16, 2022
Content may be subject to copyright.
1
Epitaxial synthesis of unintentionally doped p-type SnO (001) via suboxide
molecular beam epitaxy
Kingsley Egbo,1,* Esperanza Luna,1 Jonas Lähnemann,1 Georg Hoffmann,1 Achim Trampert,1 Jona
Grümbel,2 Elias Kluth, 2 Martin Feneberg, 2 Rüdiger Goldhahn, 2 Oliver Bierwagen1,†
1 Paul-Drude-Institut für Festkörperelektronik, Leibniz-Institut im Forschungsverbund Berlin e.V.,
Hausvogteiplatz 5–7, 10117 Berlin, Germany
2 Institut für Experimentelle Physik, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2,
39106 Magdeburg, Germany
*egbo@pdi-berlin.de
†bierwagen@pdi-berlin.de
ABSTRACT
By employing a mixed SnO2+Sn source, we demonstrate suboxide molecular beam epitaxy
growth of phase-pure single crystalline metastable SnO(001) thin films on YSZ (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 limitations of high Sn or SnO2 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-molecules desorption
at TS = 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 to 6∙1018 cm-3 and 2.0 to 5.5 cm2V-1s-
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 p-n
heterojunctions and field-effect transistors.
2
I. INTRODUCTION
Tin (II) oxide (SnO) is a valuable p-type oxide material useful in several technological
applications such as in p-n diodes, transistors, solar cells and solid-state gas sensing1,2. It is
among the few binary oxide materials that show unintentional p-type character due to native
defects3–7. Though, SnO has a low indirect fundamental bandgap of 0.7 eV, an optical bandgap
of ~2.7-2.9 eV makes it a suitable component for several transparent applications8,9. SnO
crystallizes in a layered tetragonal litharge structure with space group P4/nmm, with four oxygen
atoms and an Sn atom forming a pyramid structure10,11. Due to a more dispersed valence band
maximum composed of hybridized Sn 5s and the O 2p orbitals12,13 reasonably high mobilities
compared to most p-type oxides have been reported for polycrystalline and single crystalline
SnO thin films14–16. As most metal oxides show a propensity for n-type conductivity and due to
the difficulty in their bipolar doping, efforts in oxide electronics mostly depends 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 oxides17–21.
Unlike Tin (IV) oxide (SnO2), SnO is metastable, this presents a growth challenge with
competing stable metallic Sn or SnO2 as well as Sn3O4 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 evaporation22, pulsed laser deposition8 and molecular beam epitaxy. High
quality single crystalline SnO has been grown by plasma assisted MBE (PA-MBE) using a metal
Sn source23 and S-MBE using an oxide SnO2 source24. 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 by
the formation of SnO2 and oxygen rich Sn compounds such as Sn3O4, requiring a rigorous fine
tuning of Sn/O flux ratio.23 Also, growth using an oxide SnO2 charge involves high temperature
decomposition of the oxide into the SnO suboxide, leading to a parasitic high oxygen
background. Previously, it has been proposed that the sublimation of a mixed oxide + metal
charges in an effusion cell can provide an effective suboxide flux for the growth of oxides by
MBE 25. For instance, we showed that a reaction of SnO2 with Sn metal charge to give SnO
(SnO2 + Sn 2SnO) can be an efficient source of SnO flux, for suboxide MBE growth of SnO2
25. This suboxide approach also offers the advantage that the reactions; SnO2 + Sn 2 SnO
3
takes place at lower effusion-cell temperatures and growth of SnO can proceed without the need
for plasma activated oxygen or any intentional background oxygen.
Following this suboxide-MBE (S-MBE) approach, we present in this study the growth of SnO
thin films without the need for plasma activated oxygen or intentional background oxygen. SnO
is grown heteroepitaxially on Y-stabilized ZrO2; [YSZ (001)] and r-plane Al2O3 in a substrate
temperature, TS window from 50°C to 650°C. We investigate the impact of TS on the phase
purity, growth-rate-limiting SnO desorption as well as structural and transport properties of the
SnO layers. A schematic comparison of the conventional method of PA-MBE using a metal
source and the S-MBE approach employed in this work is described in Figure 1(a) and (b)
respectively. Detailed investigation of the growth kinetics for our SnO growth using this S-MBE
approach indicate that the growth rate is limited by the desorption of the adsorbed SnO molecules
which increases with the substrate temperature, 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 TS =50°C, textured and single-crystalline phase-pure
unintentionally (UID) doped p-type SnO(001) films are grown between substrate temperatures
of 150°C and 450°C. Above 550°C, secondary Sn3O4 and Sn phases are present in the SnO layer.
Obtained single crystalline SnO(001) layers at low TS between 150-250°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 require high substrate temperatures that exceed the BEOL limit. Hence,
these epitaxially grown SnO thin films obtained for TS between 150-250°C can be promising for
applications as active layers for BEOL compatible device development.
4
Figure 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 SnO2+Sn charge. In the PA-MBE approach, a Sn metal
cell operated at 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 high vacuum without activated or
molecular oxygen present.
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 Al2O3
substrates were also co-loaded for several growth runs. To grow SnO, a SnO2+Sn mixed source
was sublimed from an effusion cell with an Al2O3 crucible at temperatures between 740-820°C.
The hot-lip of the used dual-filament cell was kept at 150°C above the SnO2+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 x10-7 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 plasma treated for 30 min
at substrate temperatures, TS between 400-700°C using 1 SCCM O2 and 200 W plasma power
in the growth chamber. The TS monitored in-situ by a thermocouple between the substrate and
heating filament was varied between 50-650 °C for different growth runs. The background
pressure of the chamber during SnO deposition was maintained at PGC ~5-8 x10-8 mbar without
plasma activated oxygen and any intentional molecular oxygen. The growth rate and the amount
5
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 4-circle x-ray
lab-diffractometer (X’pert Pro MRD from Philips PANalytical) equipped with a Cu Kα radiation
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-ω scans with a 1mm detector slit. The in-plane epitaxial relationship
between the SnO film and the YSZ substrates were measured 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 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 dimpling, followed by argon ion-
milling, starting at 3.0 keV and finishing at 1.5 keV. The film domain structures, and orientation
was further investigated by 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 coefficients27. To compare the optical properties
of grown layers with existing literature, spectroscopic ellipsometry measurement and modelling
was performed on a layer grown on r-plane Al2O3 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 Ga2O and
SnO25. 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 SnO2+Sn
6
mixture23,28. Figure 2(a) shows the calculated SnO2-Sn equilibrium phase diagram as function
of stoichiometry, nSn/(nSn + nSnO2) and temperature calculated at 10-7 mbar typical for MBE
growth without O2 environment. This thermodynamic equilibrium diagram calculated using
FactSage29 indicate that stable SnO in the solid phase can be obtained at growth temperatures
within 190-420°C at a stoichiometry of nSn/(nSn + nSnO2)=0.5. Here ideal gas indicate gaseous
species of the constituent elements in the reaction25. This supports our experimental data where
phase pure SnO was obtained between 150-450°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 nSn/(nSn + nSnO2 ) which was used throughout this growth
experiment. Figure 2(b) shows the corresponding phase diagrams for the mixed charges at Sn
stoichiometry, nSn/(nSn + nSnO2 ) = 0.5 indicating the vapour pressure of suboxide SnO as a
function of source temperatures for the mixed charges. These mixed sources promote the
availability of required suboxide vapour pressure at lower source temperature compared to the
metal charge and solid oxide charge25.
Figure 2: (a) Equilibrium phase diagram of the SnO2-Sn system as a function of stoichiometry and
temperature at a pressure of 10-7mbar typical for MBE growth without intentional oxygen background.
Stoichiometries of nSn/(nSn + nSnO2 ) = 0, 0.5 and 1 correspond to SnO2; SnO and pure Sn respectively. (b)
Phase diagram of the SnO2-Sn system for nSn/(nSn + nSnO2 )= 0.5 as a function of temperature and pressure.
B. SnO Flux, Rate of Desorption, and disproportionation of layer
In the identified growth window for SnO suboxide growth, the growth rate is given by the
difference of the amount of arriving SnO species (proportional to the vapor pressure of the SnO
7
[c.f. Fig. 2(b)] at the cell temperature of the mixed source) and the amount of desorbing species
(which scales with the 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 SnO2+Sn effusion cell is equivalent to the measured BEP in the absence of an oxygen
background. In order to characterize the kinetics of the mixed SnO2+Sn effusion cell, we
measured the BEP as a function of source temperature from 740-800°C in the absence of any
active O2 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 previous QMS results25.
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 temperature of 50°C, i.e., without desorption from
the substrate.
Figure 3: Arrhenius diagrams of the incorporated Sn flux (solid symbols) during MBE growth and the
corresponding measured BEPs of the mixed SnO2+Sn effusion cell (right axis, open symbols). An
incorporated flux at the substrate of 4.8×1013 cm-2s-1 corresponds to a SnO growth rate of 1.0 nm/min
and to a BEP of ~7× 10-8 mbar.
Using line-of-sight quadrupole mass spectrometry, we identify the rate of desorption of the SnO
ad-molecules during growth as function of substrate temperatures. Our measurement reveals that
the desorbing flux significantly decreases with substrate temperature as shown in Figure 4(a).
8
The inset of Figure 4(a) shows typical QMS spectrum recorded during SnO desorption at high
substrate temperature. For a flux of 4x1013 cm-2s-1 reaching the substrate with substrate
temperature of 450°C, a growth rate of 1.0 nm/min is expected, however an SnO growth rate of
~0.6nm/min is obtained from the LR oscillation in Figure S1, hence the desorption rate of the
SnO ad-molecules is ~0.3 nm/min (1.4x1013 cm-2s-1). From the plot of desorption rate as a
function of substrate temperature, an activation energy of desorption of ~0.3 eV is obtained, this
low activation energy value indicates a high volatility of the suboxide ad-molecules during
growth. The desorption rate at higher substrate temperatures is limited by the amount of flux
reaching the substrate as seen in Figure 4(a). While these ad-molecules are volatile during
growth, to further understand the stability of the SnO molecules within the film after growth, we
perform a separate LR study at a SnO film grown at a BEP of 1.3x10-7 mbar. From Figure 3, this
BEP corresponds to a flux at the substrate of 8.6x1013cm-2s-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 tgrowth of the laser reflectometry signal26 for growth at
450°C substrate temperature; which is the highest temperature where phase pure SnO is
obtained. Hence, the SnO ad-molecules has a desorption rate of ~0.65 nm/min during growth
corresponding to a desorbing flux of 3.0x1013 cm-2s-1. When the SnO shutter is closed and the
substrate 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 3.0x1013 cm-2s-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 increase in substrate temperature
to 550°C, an onset of an oscillation is observed due to SnO desorption or increasing roughness
due to disproportionation of grown layer. Note that this reflected signal oscillation has 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 Figure 4(a), SnO layers formed are very stable with negligible desorption
rate at growth temperature.
9
Figure 4. (a) Arrhenius diagrams of the desorbing flux as a function of the substrate temperatures, the
desorbing species obtained from the line-of-sight QMS increases with increasing substrate temperatures
and the desorbing flux is limited at high temperatures by the provided SnO flux to the substrates (b) Plot
of the laser reflectometry signal (with substrate rotation) from the film surface during growth and thermal
etching at higher substrate temperatures. The opening and closing of the SnO2+Sn cell shutter is indicated.
C. Structural Properties and Epitaxial Relation
Wide-angle symmetric 2Θ-ω XRD scans were used to investigate the out-of-plane orientation
of the layers. Figure 5 (a) shows representative results for layers grown at different YSZ(001)
substrate temperatures, wide-angle scans of selected samples between 10-120 deg are shown in
Fig. S2 (a)(Supplementary Material), similar wide-angle scan for a sample grown on r-plane
Al2O3 are shown in Figure S2(b). Typical streaky RHEED patterns observed during the growth
of SnO (001) layers are also shown in Fig. S3 (Supplementary Materials). 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 to 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-
400°C23. A very limited substrate temperature window was observed for previous S-MBE SnO
growth on r-plane Al2O3 (1-102) substrates using SnO2 source, Mei et.al., reported that their
films grown below 370°C were amorphous while no deposition occurred above 400°C24.
Furthermore, we observe slightly sharper peaks in the 2Θ-ω scans of the samples grown at 250-
450°C indicating higher crystal quality than that of layers grown at lower temperature. However,
the increase is not linear with increasing substrate temperature. Once the substrate temperature
10
is increased even further, to 550°C, the presence of Sn3O4 secondary phase and Sn peaks are
observed indicating a disproportionation of the phase-pure SnO to Sn-rich and O-rich phases.
This trend is continued with a further increase in substrate temperature up to 650°C where the
intensity of a mixed phases (metallic Sn, Sn3O4) dominate 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 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 0.2-0.7 for samples grown at higher substrate temperatures. Film roughness between 2-
15 nm (root mean squre, rms) are obtained from AFM measurements for the single crystalline
SnO (001) layers.
Figure 5(a) XRD out-of-plane symmetric 2Θ-ω scan of SnO(001) layer on YSZ(100) grown at different
substrate temperatures. The film grown at 50°C was amorphous, between 150°C-450°C, phase pure
single crystalline SnO(001) layers are obtained. At a substrate temperature of 550°C, secondary Sn3O4
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.
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 Φ around the surface normal (see supplementary
figure S4). From Figs. 5(a) and S4, we confirm the epitaxial relationship SnO(001)||YSZ(001)
11
and SnO(110)||YSZ(010) for the out-of-plane and in-plane directions, respectively, indicating
therefore 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 observed23. 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 predominantly
single domain as observed in the Φ-scans. The scan shows four distinct circular 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 labelled observed in the texture map are due to the substrate holder as
shown in Fig. S5 (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 Ga2O3 substrates17,18.
Figure 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 the supplementary
material (Figure S7).
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
12
Sn3O423,30,31 as shown in Fig. 7. Raman spectra of the single crystalline samples grown between
150-450°C show only SnO B1g (113 cm-1) and A1g (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 Sn3O4 phase
coexisting with the SnO phase, while the sample grown at 650°C showed predominantly Sn3O4
peaks with weak contributions from Sn and SnO. These samples at 550°C also show slight shift
in their SnO peak positions likely due to the presence of these secondary phases.
Figure 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 B1g (113) and A1g (211) and
in Sn3O4, Ag and Bg as indicated by the colour code.
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 relationship between the YSZ substrates and the SnO layers. The overview bright field
image of the SnO films as shown in Figure 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 Figure
S3 is hence likely due to reflections from the surface of these large grains and does not imply a
13
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) display 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 8(d).
Figure 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)
heterostructure. (c), (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
14
Electron backscattering diffraction (EBSD) measured plan-view in a scanning electron
microscope was further used to investigate the surface microstructure of our grown layers as the
probing depth of EBSD is limited to ~20 nm. Figure S6 (a) show EBSD mapping of the
SnO(001) indicating uniformity of the crystal orientation without any crystal twins on the layer.
Figure S7 (b) show 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 measurement32. Due to low hole mobilities observed in most p-type oxides, extracting
accurate Hall voltage using a single magnetic field value B may become ambiguous. Here, to
extract reliable Hall coefficients, a Hall sweep between +0.8T to -0.8T were carried out and the
Hall coefficient is obtained from the slope of the Hall voltage as shown in Figure S7
(Supplementary Materials). Varying room temperature UID hole densities, pHall in the range of
0.9-6.5x1018 cm-3 are obtained for the phase-pure SnO thin films (S150-S450) as shown in Figure
9(a). The sample grown at low substrate temperature of 150°C shows a remarkable mobility of
2.2 cm2/V.s, increasing the substrate temperatures resulted in the growth of samples with average
mobility of 4.5 cm2/V.s as seen in Figure 9(b). Room temperature electrical resistivities ρ
between 0.3-1.2 Ohm-cm are obtained for the deposited films. A slight spread in transport
properties for single crystalline SnO (001) grown at different temperatures maybe 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 2 orders of magnitude higher than the
value reported for UID SnO grown by S-MBE from an SnO2 source24. Table 1 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 vacancies 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 activated oxygen is supplied. Nevertheless, we cannot rule out unintentionally
incorporated extrinsic dopants. S550 films with dominant SnO also showed p-type UID
15
properties like the phase pure SnO layers. Amorphous layers grown at 50°C was semi-insulating
while the mixed phase layer grown at 650°C showed n-type character.
Figure 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.
Table 1: Comparison of electrical properties of epitaxial SnO(001) thin films grown using
different techniques
Material
Methoda
Tdep [°C]
Growth
PO2(Torr)
Substrate
FWHM(°)
pHall [cm-3]
Mobility(cm2/
V.s)
Resistivity(Ω-
cm)
Ref
SnO(001)
S-MBE
380
5x10-7
r-Al2O3
0.007
2.5x1016
2.4
101
24
SnO(001)
PLD
575
1x10-6
YSZ(001)
0.46
2.5x1017
2.4
-
8
SnO(001)
PLD
200
6x10-2
YSZ(001)
1.0
1.0x1017
2.3
-
33
SnO(001)
EBE
600
-
r-Al2O3
2.9
5.6x1017
0.1
110
9
SnO(001)
EBE
600
~1x10-6
r-Al2O3
-
-
195
22
SnO(001)
PA-MBE
350-400
~5x10-6
YSZ(001)
,c-Al2O3
0.4-1.9
1.8-9.7x1018
1-6
0.25-2.0
23
SnO(001)
S-MBE
150-450
~6x10-8
YSZ(100)
0.2-1.3
0.9-6 x 1018
2.5-5.5
0.3-1.2
This
Work
SnO(001)
S-MBE
400
~6x10-8
r-Al2O3
1.3
7.0x1017
1.4
7.2
This
Work
a) Method of deposition: S-MBE-Suboxide Molecular Beam Epitaxy, PA-MBE- Plasma assisted MBE PLD
-Pulsed Laser Deposition, RFMS-RF Magnetron Sputtering, DCRS, DC Reactive Sputtering, EBE-
Electron Beam evaporation
16
To assess the optical properties of SnO layers grown via this suboxide MBE route, spectroscopic
ellipsometry measurement and modeling is performed on ~120 nm thick phase-pure SnO(001)
sample grown on r-plane Al2O3 at 400°C. Fig S2(b) shows the wide angle 2Θ-ω scan. Compared
to the sample grown on YSZ (001) at similar substrate temperature, Hall measurement for this
sample show a slightly lower hole density and mobility of ~7.0x1017cm-3 and ~1.4 cm2/V.s
respectively and higher resistivity of ~7.2 Ohm-cm. This decrease in the transport properties is
likely due to increase in strain-induced dislocation density caused by higher lattice mismatch of
~12% in SnO/r-plane Al2O3 hetero-interface compared to SnO/YSZ hetero-interface with ~5%
lattice mismatch23,24. Ordinary and extraordinary complex dielectric function (ε ≡ ε1 + iε2) spectra of
this thin film extracted from room-temperature spectroscopic ellipsometry measurement is shown in
Figure 10. The ε2 spectra in the ordinary direction shows an onset of absorption at ~2.7 eV similar to
previously reported values24.
Figure 10: Complex dielectric function of SnO(001) thin film resolved into the ordinary xy (solid
lines) and extraordinary z (dashed lines) components obtained from ellipsometry point-by-point
fitting.
17
IV. CONCLUSION
Using an intentional suboxide source comprising a mixed SnO2+Sn charge, we demonstrate the
heteroepitaxial growth of phase-pure, single crystalline SnO (001) thin films on YSZ(001)
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 narrow growth window
previously reported for PA-MBE growth of SnO using a metal charge and S-MBE growth of
SnO using a SnO2 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 employ 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°C-450°C. Transport and optical measurement also confirm
the p-type properties and optical transparency of these layers. Hence, with the S-MBE approach,
single crystalline phase-pure SnO was achieved at the lowest substrate temperature 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 150-250°C can promote the
integration of these p-type SnO layers for BEOL compatible device application.
SUPPLEMENTARY MATERIAL
See supplementary material for typical RHEED image of SnO layers acquired during growth.
Wide angle 2Θ-ω scans of SnO(001) on YSZ(001) and r-Al2O3. Skew-symmetric Φ-scans of
SnO layer and YSZ substrate, EBSD mapping and Hall magnetic field sweep data.
ACKNOWLEDGEMENTS
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
framework of GraFOx, a Leibniz ScienceCampus partially funded by the Leibniz Association.
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to declare
REFERENCES
18
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. Stat. Sol. (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, Journal of Applied Physics 128, 135705 (2020).
5 K.O. Egbo, C.E. Ekuma, C.P. Liu, and K.M. Yu, Phys. Rev. Materials 4, 104603 (2020).
6 K.O. Egbo, M. Kong, C.P. Liu, and K.M. Yu, Journal of Alloys and Compounds 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, Applied Physics
Letters 96, 96 (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, Journal of Vacuum Science & Technology A 40, 042701 (2022).
22 X.Q. Pan and L. Fu, Journal of Electroceramics 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. Materials 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. Materials 3, 105202 (2019).
25 G. Hoffmann, M. Budde, P. Mazzolini, and O. Bierwagen, APL Materials 8, 031110 (2020).
26 P. Vogt and O. Bierwagen, Appl. Phys. Lett. 106, 081910 (2015).
27 F. Werner, Journal of Applied Physics 122, 135306 (2017).
28 K.M. Adkison, S.-L. Shang, B.J. Bocklund, D. Klimm, D.G. Schlom, and Z.-K. Liu, APL Materials
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. Materials 1, 014602 (2017).
31 F. Wang, X. Zhou, J. Zhou, T.-K. Sham, and Z. Ding, J. Phys. Chem. C 111, 18839 (2007).
32 O. Bierwagen, T. Ive, C.G. Van de Walle, and J.S. Speck, Appl. Phys. Lett. 93, 242108 (2008).
33 H. Hayashi, S. Katayama, R. Huang, K. Kurushima, and I. Tanaka, Phys. Status Solidi RRL 9, 192
(2015).
19
Supplementary Material for:
Epitaxial synthesis of unintentionally doped p-type SnO (001) via suboxide
molecular beam epitaxy
Kingsley Egbo,1,* Esperanza Luna,1 Jonas Lähnemann,1 Georg Hoffmann,1 Achim Trampert,1 Jona
Grümbel,2 Elias Kluth, 2 Martin Feneberg, 2 Rüdiger Goldhahn, 2 Oliver Bierwagen1,†
1 Paul-Drude-Institut für Festkörperelektronik, Leibniz-Institut im Forschungsverbund Berlin e.V.,
Hausvogteiplatz 5–7, 10117 Berlin, Germany
2 Institut für Experimentelle Physik, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2,
39106 Magdeburg, Germany
*egbo@pdi-berlin.de
†bierwagen@pdi-berlin.de
Figure S1: Laser reflectometry period signal for SnO thin film grown at a substrate temperature of
450°C and flux of 4x1013 cm-2s-1. This corresponds to a growth rate of ~0.6nm/min. A 650 nm
laser reflected at an angle of 60° with respect to the substrate normal was used for the laser
reflectometry measurement.
20
Figure S2(a): Wide angle 2Θ-ω scan of S150 and S450 SnO(001) thin films on YSZ(001).
Figure S2(b): Wide angle 2Θ-ω scan SnO(001) layer grown on r-plane Al2O3 at 400°C
21
Figure S3: Typical streaky RHEED images suggest smooth surface with some islands during
growth
The cubic four-fold rotational symmetry of the YSZ substrate is reflected by the four black
peaks seen in Figure S4. The red peaks are the peaks due to the SnO(112) skew symmetric
planes for S450 also showing 4 peaks due to the rotational symmetry hence indicating
absence of multiple rotational domains in the epilayer1. The projection of the (111) peak of
the YSZ onto the 100 plane is rotated by an angle of Φ = ±45° to the projection of the 112
peak unto the (001) plane of SnO.
Figure S4: Phi-scans of the SnO and YSZ layers
120nm SnO
YSZ(100)
SnO<110>
22
SnO (001) Wulff plot and Texturemap of the substrate holder
Figure S5: Calculated WULFF plot of the SnO(001) out-of-plane layer and Texturemap of
substrate holder used in measurements.
EBSD Analysis of the crystal orientation in the surface of grown SnO layer and YSZ substrate
23
Figure S6. (a) EBSD Mapping of the SnO(001) layer showing uniformity across the surface of
the sample (b) Kikuchi patterns of the SnO(001) epilayer and wireframe representation of the
indexed orientations.
Hall Magnetic Field sweep
Figure S7. Typical magnetic field sweep used to extract the Hall coefficient of grown layers.
References
1 M. Grundmann, T. Böntgen, and M. Lorenz, Phys. Rev. Lett. 105, 146102 (2010).