Ga(N,P) Growth on Si and Decomposition Studies of the N-P Precursor Di- tert-butylaminophosphane (DTBAP)

Article abstract of DOI:10.1021/acs.organomet.0c00078

III/V semiconductors containing small amounts of nitrogen (dilute nitrides) are promising for applications such as lasers and solar cells. Metal-organic vapor-phase epitaxy (MOVPE) is a widely used technique for growing III/V semiconductors on an industrial scale, and the growth of dilute nitrides with this method is promising for later successful market entry. The main issues of dilute nitrides are carbon incorporation and low nitrogen incorporation efficiency of the conventional N precursors. Due to the high N incorporation efficiency and the low decomposition temperature of the As and N precursor di-tert-butylaminoarsane (DTBAA), a similar P- and N-containing precursor, di-tert-butylaminophosphane (DTBAP), was synthesized and purified on a laboratory scale. Growth studies using this precursor were carried out in this work realizing Ga(N,P)/GaP multi quantum wells on Si and GaP substrates. The structures show evidence of N incorporation, and good layer structures were confirmed by high-resolution X-ray diffraction. Following the influence of different growth parameters on the N incorporation, the growth rate and surface morphology were characterized to set a foundation for possible growth applications in the future. DTBAP shows many advantages over the conventional N source 1,1-dimethylhydrazine (UDMHy) such as a much lower decomposition temperature of 310 °C and the realization of Ga(N,P) layers grown at temperatures as low as 475 °C with a high N incorporation of over 10percent. Furthermore, the gas-phase decomposition of DTBAP has been studied with a real-time fast Fourier transform quadrupole ion trap mass spectrometer attached inline to the MOVPE reactor. The decomposition of DTBAP behaves very similarly to the As analogue DTBAA. On the one hand, the tert-butyl groups attached to DTBAP decompose radically, leading to the formation of isobutane, and decompose, on the other hand, by β-H elimination, leading to the formation of isobutene. Furthermore, the decomposition products indicate a direct cleavage of the P-N bond of the molecule, resulting in the formation of aminyl radicals (NH2?). The formation of NH2? explains the high N incorporation efficiency of DTBAP at low temperatures as well as its limitations due to loss of NH3 at higher temperatures.

Full text of DOI:10.1021/acs.organomet.0c00078

pubs.acs.org/Organometallics
Article
Ga(N,P) Growth on Si and Decomposition Studies of the N−P
Precursor Di-tert-butylaminophosphane (DTBAP)
,§
,§
̈
Johannes Glowatzki,* Oliver Maßmeyer,* Marcel Koster, Thilo Hepp, Ebunoluwa Odofin,
̈
Carsten von Hanisch, Wolfgang Stolz, and Kerstin Volz
Cite This: Organometallics 2020, 39, 1772−1781
Read Online
Metrics & More
Article Recommendations
ACCESS
ABSTRACT: III/V semiconductors containing small amounts of nitrogen (dilute nitrides) are
promising for applications such as lasers and solar cells. Metal−organic vapor-phase epitaxy
(MOVPE) is a widely used technique for growing III/V semiconductors on an industrial scale,
and the growth of dilute nitrides with this method is promising for later successful market entry.
The main issues of dilute nitrides are carbon incorporation and low nitrogen incorporation
efficiency of the conventional N precursors. Due to the high N incorporation efficiency and the
low decomposition temperature of the As and N precursor di-tert-butylaminoarsane (DTBAA), a
similar P- and N-containing precursor, di-tert-butylaminophosphane (DTBAP), was synthesized
and purified on a laboratory scale. Growth studies using this precursor were carried out in this
work realizing Ga(N,P)/GaP multi quantum wells on Si and GaP substrates. The structures
show evidence of N incorporation, and good layer structures were confirmed by high-resolution
X-ray diffraction. Following the influence of different growth parameters on the N incorporation,
the growth rate and surface morphology were characterized to set a foundation for possible
growth applications in the future. DTBAP shows many advantages over the conventional N
source 1,1-dimethylhydrazine (UDMHy) such as a much lower decomposition temperature of 310 °C and the realization of
Ga(N,P) layers grown at temperatures as low as 475 °C with a high N incorporation of over 10%. Furthermore, the gas-phase
decomposition of DTBAP has been studied with a real-time fast Fourier transform quadrupole ion trap mass spectrometer attached
inline to the MOVPE reactor. The decomposition of DTBAP behaves very similarly to the As analogue DTBAA. On the one hand,
the tert-butyl groups attached to DTBAP decompose radically, leading to the formation of isobutane, and decompose, on the other
hand, by β-H elimination, leading to the formation of isobutene. Furthermore, the decomposition products indicate a direct cleavage
•
of the P−N bond of the molecule, resulting in the formation of aminyl radicals (NH2 ). The formation of NH2• explains the high N
incorporation efficiency of DTBAP at low temperatures as well as its limitations due to loss of NH3 at higher temperatures.
phase epitaxy (MOVPE). For later realization of devices in
mass production, MOVPE is a preferable method. Devices
such as lasers and LEDs are produced these days with this
method because of good scalability and high production
output.14 The disadvantages of MOVPE and the metal−
organic precursors are the unintentional incorporation of
impurities of alkyl groups and the restriction of growth
temperatures because of the necessary thermal decomposition
of the precursor. Large alkyl groups such as propyl and butyl
groups are preferable because carbon (C) incorporation is
probably more unlikely due to steric hindrance and the higher
stability of the forming radicals in comparison to methyl
1. INTRODUCTION
GaP-based III/V semiconductors containing small amounts of
nitrogen (“dilute nitrides”) have been investigated because of
their promising optoelectronic properties. The interaction of
the nitrogen (N) level with the conduction band (band
anticrossing model) leads to a strong reduction of the band gap
energy.1,2 The material system Ga(N,As,P) can be grown on
silicon (Si) without relaxation and also shows direct band gap
behavior.3−5 It is a candidate for a monolithically integrated
laser on Si. First optically and electrically pumped lasers were
demonstrated on GaP and Si substrates, respectively.6−8 This
material class with different compositions is also a candidate
for the top junction of a two-junction solar cell on Si.9 Si as a
substrate offers huge advantages in comparison to common
III/V substrates. It is abundant, nontoxic, and well established
in industry. In addition, solar cells, nanowires, and water-
splitting solar cells were demonstrated with Ga(N,P).10−13
The GaP-based material systems have been grown with
molecular beam epitaxy (MBE) and metal−organic vapor-
Received: February 5, 2020
Published: April 30, 2020
https://dx.doi.org/10.1021/acs.organomet.0c00078
© 2020 American Chemical Society
Organometallics 2020, 39, 1772−1781
1772

Organometallics
pubs.acs.org/Organometallics
Article
groups. However they also reduce the vapor pressure of the
precursor. For the growth of dilute nitrides the nitrogen
incorporation with the conventional precursor 1,1-dimethylhy-
drazine (UDMHy) is low and precursors with higher N
incorporation efficiency are desirable.15
starting materials tBuCl and PCl3 were purchased from Acros
Organics, and magnesium was obtained from Sigma-Aldrich.
NMR (nuclear magnetic resonance) spectra were recorded on
a Bruker AvanceHD 300 instrument with the following
multiplicities: s = singlet, d = doublet.
A novel arsenic (As)- and N-containing precursor (di-tert-
butylaminoarsane, DTBAA) for the growth of GaAs-based
dilute nitrides was investigated in the last few years and
showed a high N incorporation in Ga(N,As), (Ga,In)(N,As),
and Ga(N,As,Sb) and a low decomposition temperature.16−19
On the basis of the experience with this precursor a similar N-
and phosphorus (P)-containing precursor (di-tert-butylamino-
phosphane, DTBAP) was synthesized and tested in the
function of a precursor for MOVPE.
To get further insight into the N incorporation of the novel
precursor its thermal decomposition has been studied. Like the
As analogue DTBAA DTBAP exhibits a N−group V bond
instead of a direct C−N bond and the C is only present in
large alkyl groups. Due to prevention of the strong C−N bond
and steric restriction of the large alkyl groups it is expected that
C incorporation will be reduced in the growing layers using
these novel precursors.20 In this regard the precursors should
have an advantage in comparison to the commonly used N
precursor 1,1-dimethylhydrazine (UDMHy). UDMHy exhibits
strong N−N and C−N bonds as well as small alkyl groups. As
shown in former decomposition studies, this leads to formation
of CH4, N2, and larger stable fragments such as (CH3)2NH and
CH3NCH2 upon decomposition.21 These fragments probably
lead to comparably high C incorporation and low N
incorporation efficiencies due to the high bond strength,
especially for use at low growth temperatures.
2.2. tBu2PCl. A solution of tBuCl (55.1 mL, 0.50 mol, 2.5
equiv) and 150 mL of diethyl ether was added dropwise to a
suspension of magnesium (12.2 g, 0.50 mol, 2.5 equiv) and
200 mL of diethyl ether at room temperature. This Grignard
solution was then added dropwise to a solution of PCl3 (17.5
mL, 0.20 mol, 1.0 equiv) in 400 mL of diethyl ether at −40 °C.
After the mixture was warmed to rt, the formed precipitate was
filtered and washed twice with 40 mL portions of n-pentane.
Removal of the solvent in vacuo provided the raw product as a
colorless liquid which was purified via fractional distillation at
1
10 mmHg at 63 °C (18.0 g, 0.10 mol, 50% yield). H NMR
3
(300.13 MHz, C6D6): δ/ppm 1.09 [d, JHP = 12.1 Hz, 18H,
CH3]. 13C{1H}-NMR (75.47 MHz, C6D6): δ/ppm 27.9 [d,
2JCP = 17.1 Hz, CH3], 35.9 [d, 1JCP = 41.3 Hz, Cquart]. 31P{1H}-
NMR (121.54 MHz, C6D6): δ/ppm 148.1 [s, PCl].
2.3. tBu2PNH2. Gaseous ammonia was passed through a
solution of tBu2PCl (10 g, 0.55 mol, 1.0 equiv) and 10 mL of
diethyl ether for 1 h at −50 °C. The resulting precipitate was
filtered after it was warmed to rt and washed twice with 10 mL
portions of n-pentane. Removal of the solvent in vacuo
provided the raw product as a colorless liquid which was
purified via fractional distillation at 5 mmHg and 55 °C (8.06
1
g, 0.50 mol, 90% yield). H NMR (300.13 MHz, C6D6): δ/
3
ppm 0.92 [s, 2H, NH], 1.03 [d, JH,P = 11.2 Hz, 18H, CH3].
2
13C{1H}-NMR (75.47 MHz, C6D6): δ/ppm 28.1 [d, JC,P
=
15.4 Hz, CH3], 33.0 [d, 1JC,P = 21.5 Hz, Cquart]. 31P{1H}-NMR
(121.54 MHz, C6D6): δ/ppm 63.0 [s, PNH2].
In the following the synthesis, growth, and mass
spectrometer results of the novel precursor DTBAP are
presented. In the beginning N incorporation in GaP on Si
substrates will be demonstrated and systematic investigations
of the dependence on growth temperature and triethylgallium
(TEGa), tert-butylphosphane (TBP), and DTBAP partial
pressures will be shown. Subsequently the growth on GaP
will be demonstrated and compared to the growth on Si.
Finally, thermal decomposition studies of DTBAP via mass
spectrometry will be shown.
3. GROWTH AND CHARACTERIZATION OF
Ga(N,P)-CONTAINING SAMPLES
All samples were grown using MOVPE. An AIXTRON AIX
200 reactor system with gas foil rotation was used. The reactor
pressure was held at 50 mbar under a total gas flow of 6800
sccm for all experiments. Palladium cell purified hydrogen
(purity: 9.0) was used as the carrier gas. tert-Butylphosphane
(TBP), triethylgallium (TEGa), and DTBAP were used as
precursors. The vapor pressure of the novel precursor DTBAP
was unknown at the beginning of the experiments and was
estimated via consumption calculations afterward to be around
1 mbar at 20 °C. The samples were grown either on exact
semiinsulating Si (001) substrates with a 28 nm GaP
nucleation on top or on exact semiinsulating GaP (001)
substrates. For a high-quality surface both substrates were
processed with a pretreatment. The GaP/Si substrate was
heated (unstabilized and TBP stabilized) and a thin 12 nm
GaP buffer was grown under optimized conditions at 675 °C.
A thicker buffer was not possible because of relaxation at large
GaP thickness. The GaP substrate was processed with different
etching steps before pretreatment in the reactor. The samples
were heated and afterward overgrown with a 250 nm thick GaP
buffer. Test structures were grown to determine the N
incorporation as three times multi quantum wells (MQWs)
at temperatures between 475 and 575 °C. Cooling after buffer
growth was done under TBP stabilization to prevent P
desorption. Between 7 and 21 nm thick Ga(N,P) QWs and
thin 7−11 nm GaP barriers were grown. Thicker barriers were
avoided to prevent the samples from relaxation. The structure
2. SYNTHESIS OF tBu2PNH2 (DTBAP)
An early synthesis of tBu2PNH2 was performed by Scherer and
Schieder in 1968.22 In a first reaction of the Grignard reagent
tBuMgCl with PCl3 the compound tBu2PCl was prepared. This
reactant was aminated with gaseous ammonia to give the
desired product tBu2PNH2.22 The synthesis is sketched in
Scheme 1.
2.1. General Considerations. All working procedures
were conducted under rigorous exclusion of oxygen and
moisture using a Schlenk line and nitrogen/argon atmosphere.
Solvents were dried and freshly distilled before use. The
Scheme 1. Synthesis of tBu2PNH2 (DTBAP)
1773
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
was always finished with a QW and cooled to 350 °C under
TBP stabilization to determine the surface morphology
realistically. The TEGa, DTBAP, and TBP partial pressures
and also the growth temperature were varied systematically
and the influence on the N incorporation, growth rate, and
surface morphology were investigated.
The surface morphology of the samples was investigated
using a Nanoscope IIIa atomic force microscopy (AFM) setup
from Digital Instruments. The AFM images shown have a size
of 5 × 5 μm.
The structure of the samples was investigated using a
Panalytical X’Pert Pro high-resolution X-ray diffraction
(HRXRD) system with a Cu anode monochromated to the
Cu Kα line (wavelength of 0.15405 nm). By measurements
around the substrate (Si or GaP) (004) reflection and dynamic
simulations of the diffractograms with the X’Pert Epitaxy
smooth and fit software the N content and growth rate of the
Ga(N,P) layers could be determined. At different times during
the experiments, samples with nominally the same growth
conditions were prepared to ensure reproducibility and obtain
the error for different growth runs. The growth conditions
were as follows: Tgr = 525 °C, Pp(TBP) = 0 mbar, Pp(TEGa) =
7.1 × 10−3 mbar, and Pp(DTBAP) = 1.8 × 10−2 mbar. The
average N content of these samples is 2.12% with a standard
deviation of 0.13%. As this value is larger than the estimated
error from the simulations, this value is taken as the error for
the compositions. For the growth rate the estimated error of
the simulations leads to an error of 0.01 nm/s for the growth
rate, if not noted differently.
5. RESULTS AND DISCUSSION
The following part is organized as follows. In the beginning the
first evidence of nitrogen incorporation in GaP on the Si
substrate will be shown and experiments with varying growth
conditions will follow. Furthermore, Ga(N,P) growth on GaP
substrates will be shown and compared to growth on Si
substrates. Finally mass spectrometer measurements and
decomposition curves of the DTBAP will be shown.
5.1. MOVPE Growth Experiments on Si. The first
evidence of Ga(N,P) growth including sufficient nitrogen
incorporation using DTBAP was obtained with HRXRD for all
investigated samples.
As an example, the HRXRD diffractogram and dynamic
simulation of one sample grown with a partial pressure of
DTBAP of 1.8 × 10−2 mbar and a TEGa partial pressure of 7.1
× 10−3 mbar without TBP at a growth temperature of 525 °C
are shown in Figure 1.
4. MASS SPECTROMETER MEASUREMENTS OF
DTBAP
To study the thermal decomposition of DTBAP, a quadrupole
ion trap mass spectrometer from Carl Zeiss GmbH (iTrap)
was used in line with the MOVPE system. The setup is
designed to achieve measurement conditions as close to
realistic growth conditions as possible and has been
successfully used for gas-phase investigations of tert-butylar-
sane (TBAs) and DTBAA before.16,23 The ion trap uses
standard electron impact ionization with 70 eV for ion
generation. The ions were then captured by an oscillating
electric field applied to the ring electrode of the ion trap. The
frequency of the oscillation used is within the radio frequency
(RF) range. The mass range of interest can be selected by the
adjustment of the applied RF voltage, which changes the
strength of the electric field. Due to the oscillations of the ions
within the trap an induced mirror image current within the top
and bottom electrodes of the trap can be measured. This
current signal was then analyzed by fast Fourier trans-
formation, which enabled recording times for a single mass
spectrum below 2 s. Therefore, the setup is suitable for real
time analysis during the MOVPE process and the fast
measurement times allow high data point densities and good
statistics. Generally, quadrupole ion traps offer high sensitivity
leading to detection of impurities in the parts per trillion
concentration level with this setup.24
Figure 1. Example of an HRXRD diffractogram of a Ga(N,P)/GaP
multiquantum well on Si with dynamic simulation.
Simulation and measurement are in good agreement, and all
fringes are resolved well, which indicates high-quality layers. If
not noted otherwise, all samples show good HRXRD
measurements. Thus, in the following only the N content
and growth rate are plotted.
In Figure 2 the N incorporation and the growth rate are
plotted versus the TBP partial pressure for two different
DTBAP partial pressures. For all samples the TEGa partial
pressure as well as the growth temperature were kept constant
(7.1 × 10−3 mbar and 525 °C, respectively). An additional
supply of TBP during growth reduces the N incorporation.
The highest N incorporation was obtained using no TBP and a
higher DTBAP partial pressure, while the N incorporation was
reduced for lower DTBAP partial pressure and no TBP. This
also means that no additional P source is needed and DTBAP
delivers both elements N and P simultaneously. This behavior
was also seen for the As analogue DTBAA. In this case
Due to the setup, the precise measurement of the gas-phase
temperature at the point of gas collection was not possible and
only the surface temperature of the susceptor was measured.
By correlation of TBAs decomposition curves from ref 23 with
data taken from ref 25 the gas-phase temperature was
determined.
1774
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
Figure 2. N incorporation and growth rate in dependence of the TBP
partial pressure. DTBAP and TEGa partial pressures and growth
temperature were kept constant.
Ga(N,As) layers on GaAs were realized without an additional
As source.19
The N incorporation is low compared to the P
incorporation, which means that P is much more efficiently
incorporated. This can be related to the metastability of
Ga(N,P) grown on GaP. Under equilibrium conditions the
solubility of N in GaP is in the parts per million level. Only due
to the growth far from equilibrium can N incorporation in GaP
be realized, as is done in MOVPE.26,27 Also additional P from
the TBP is competing for the group V lattice sites and further
reduces the N incorporation. The N/P ratio seems to be
crucial for the N incorporation, as is also seen for growth
studies of Ga(N,P) with UDMHy and for the growth of
Ga(N,As) with the As analogue DTBAA.15,19 The TBP partial
pressure has no influence on the growth rate, since the group
III supply should be growth-limiting under these conditions.
This means also that the effective V/III ratio is larger than 1.
Since there is no need to add TBP and the highest N
incorporation was reached without TBP, further experiments
were carried out without TBP. The DTBAP partial pressure
was varied systematically at a growth temperature of 525 °C
while TEGa partial pressure was kept constant (7.1 × 10−3
mbar). The results of this experiment are shown in Figure 3a.
Increasing the DTBAP partial pressure increases the N
incorporation linearly. There are many reasons this depend-
ence could happen. As was mentioned before, the N/P ratio
seems to be crucial for N incorporation but the N/P ratio is
fixed to unity in the molecule itself. However, gas-phase
reactions in combination with surface and catalytic reactions
could lead to varying N/P ratios dependent on the DTBAP
partial pressure and the (N + P)/Ga ratio. One can also
consider the influence of the surface reconstruction, which may
change depending on the (N + P)/Ga ratio. A change of the
surface reconstruction on the V/III ratio has been widely
studied for GaAs (001) surfaces and is also seen for GaP (001)
surfaces.28−31 Even an increase in the N incorporation on more
Ga rich surfaces has been seen for Ga(N,As) grown in GaAs as
well as has been predicted for Ga(N,P).32,33 The complete
mechanism behind this incorporation behavior remains unclear
and has to be investigated in the future in more detail.
Figure 3. (a) N incorporation and growth rate dependence on the
DTBAP partial pressure. The TEGa partial pressure and growth
temperature were kept constant, and no TBP was added. (b) AFM
images (size 5 × 5 μm) of some samples from (a). DTBAP partial
pressures and RMS values are given as insets.
Furthermore, the growth rate does not seem to change, which
is expected because of the same reasons mentioned before in
the discussion of Figure 2. AFM images from the surfaces of
the samples grown with different DTBAP partial pressures are
shown in Figure 3b. At lower amounts of DTBAP the surfaces
are flat and show a small roughness but there are some larger
three-dimensional (3D) structures on it. Most likely these are
Ga droplets occurring because of a low effective (N + P)/Ga
ratio. The sample grown with Pp(DTBAP) = 2.5 × 10−2 mbar
is smooth without 3D features but is rougher than the surfaces
mentioned before. A higher Pp(DTBAP) leads to a very rough
surface with 3D islandlike structures. In addition, the HRXRD
measurement of this sample showed a single Ga(N,P) layer
instead of a three times multi quantum well structure. Probably
due to high strain the growth mode changed and only highly
defective layers were deposited after a specific thickness.
Further the TEGa partial pressure was varied systematically
at a growth temperature of 525 °C and a fixed DTBAP partial
pressure of 1.8 × 10−2 mbar. The N incorporation and growth
1775
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
rate are plotted in Figure 4. With increasing TEGa partial
pressure the N incorporation decreased and the last data point
Figure 4. N incorporation and growth rate dependent on the TEGa
partial pressure. The DTBAP partial pressure and growth temperature
were kept constant, and no TBP was added.
remained constant within the error or even increased again.
Possible explanations for the decreasing N content could be
most likely the lowering of the (N + P)/Ga ratio that resulted
in a lower N incorporation as seen in Figure 3a. In addition a
constant sticking coefficient at a certain temperature of the
active N species could explain this behavior, since by increasing
the growth rate less N can be incorporated due to the shorter
period of time per monolayer growth.
The increased N incorporation at higher TEGa partial
pressures perhaps originates from a change in the surface
reconstruction at low V/III ratios. At low V/III ratios the
surface can change to a more Ga rich surface reconstruc-
tion.30,31 For these more Ga rich surfaces an enhanced N
incorporation has been predicted in the literature and some
evidence for this has been seen for growth of Ga(N,As) on
GaAs (001) surfaces.32,33 The growth rate is increasing
proportionally to the TEGa partial pressure due to the
growth-limiting character of the group III supply in the group
V rich growth regime.
While the gas phase composition was kept constant, the
growth temperature was changed systematically. The DTBAP
and the TEGa partial pressures were kept constant at 2.5 ×
10−2 mbar and 7.1 × 10−3 mbar, respectively. In Figure 5a the
N content and the growth rate are plotted versus the growth
temperature. The N incorporation is increasing exponentially
(exponential behavior is discussed later in section 5.3) with a
decrease in the growth temperature, and a N content of about
10% was reached at a growth temperature of 475 °C. Growth
with this amount of N at this low temperature has not been
reported for MOVPE growth with the conventional N
precursor UDMHy.15 By further optimization of the growth
conditions even higher N contents should be realizable, which
could compete with N fractions of up to 16% realized in
Ga(N,P) layers grown by MBE.34 This exponential behavior of
the N incorporation implies a thermally activated loss process
with an activation energy, which will be shown in section 5.3.
One can think of loss processes such as desorption of the active
Figure 5. (a) N incorporation and growth rate dependent on the
growth temperature. TEGa and DTBAP partial pressures were kept
constant, and no TBP was added. (b) AFM images (size 5 × 5) of
some samples from (a). Temperature and RMS values are given as
insets.
nitrogen species from the surface or a loss due to gas-phase
reactions of the active species to more stable products, which
are not incorporated efficiently. This idea is supported by
•
formation of aminyl radicals (NH2 ) that form upon
•
decomposition. NH2 is likely to form NH3, which could
prevent N incorporation at higher temperatures and will be
discussed in section 5.3. The growth rate is increasing linearly
with the growth temperature. Possibly there is a larger supply
of the growth-limiting group III species due to a higher
diffusion rate from the boundary layer to the surface. Higher
decomposition of the Ga precursor should not be the reason,
since the decomposition temperature in our reactor is
measured as 250 °C, which is in agreement with the literature
data for TEGa.35,36 The desorption of surface-blocking alkyl
groups could also explain the increase in growth rate. The site
blocking seems to be the most likely explanation, since it has
been studied in different growth experiments with
MOVPE.37−39 In Figure 5b 5 × 5 μm AFM images of the
1776
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
surfaces from samples grown at different temperatures are
shown. For higher temperatures the surfaces do not show steps
or an otherwise distinct morphology but they are smooth with
wavylike modulation in the micrometer lateral range. The
wavylike modulation can be directly related to nucleation of
the GaP on the Si substrate, as it is not seen on GaP substrates.
This will be shown in section 5.2. For lower growth
temperatures, the surfaces are rougher and show 3D features
on top and the underlying surfaces are looking similarly
smooth as the samples grown at higher temperatures.
Roughening of the surface and 3D island formation could be
related to the higher N content and the lower diffusion rate on
the surface with decreased temperature. In addition samples of
the TEGa and DTBAP partial pressure series, which are
discussed in Figure 3a, showed rougher surfaces and similar
(but smaller) 3D features for higher N contents. It is
remarkable that the surface got smoother and showed fewer
3D features for the lowest growth temperature with the highest
amount of N. Probably the formation of these 3D features is
related to the strain induced by the high N content and the
high miscibility gap in this metastable material system.
Additionally, an activation energy should be necessary for
the change in growth mode from 2D to 3D growth which
occurs at higher temperatures. Thus, the formation seems to be
both temperature as well as strain induced.
liquid-phase epitaxy method.40 Lattice pulling effects were also
seen in the MOVPE growth of (Ga,In)N on GaN and also for
the dilute nitride material system (Ga,In)(N,As) on (Ga,In)-
As.41,42 An effect of lattice mismatch on solid composition was
also seen for MOVPE grown Ga(N,As) on GaP. In contrast to
the growth on GaAs the N incorporation was improved
drastically by a factor of about 10. This was explained by the
compressive strain of Ga(N,As) on GaP up to a fraction of
17.8% N which results in strain compensation. In contrast N
incorporation results in tensile strain over the whole
composition window for Ga(N,As)/GaAs.43 In this study the
compressive strain of Ga(N,P) on Si up to a N content of
around 2% also leads to the lattice pulling effect and allows a
higher incorporation due to the strain-compensating effect of
N at these compositions. This is in contrast to the growth on
GaP, where N incorporation introduces tensile strain for every
composition.
5.3. Decomposition Studies via Mass Spectrometry.
To provide a deeper insight into the N incorporation behavior
into Ga(N,P), the thermal decomposition of DTBAP (161 u)
has been investigated with an ion trap connected to the
MOVPE system. The detected mass spectrum at a temperature
of 50 °C for a supplied partial pressure of 3 × 10−3 mbar is
shown in Figure 7a. For comparison the detected intensities
5.2. MOVPE Growth Experiments on GaP. In addition
to the growth experiments on GaP/Si, a few samples were
grown under similar growth conditions on GaP substrates for
comparison. AFM images (size 5 x 5 μm) of samples grown
under same conditions on GaP/Si and GaP substrates are
shown in Figure 6.
Figure 6. AFM images (size 5 x 5 μm) of samples grown under the
same growth conditions but on different substrates (Si vs GaP). The
growth conditions are Pp(TEGa) = 7.1 × 10−3 mbar, Pp(DTBAP) =
2.5 × 10−2 mbar, and Tg = 525 °C. In the AFM images the N
incorporation and the RMS values are shown.
Figure 7. Detected mass spectra of DTBAP (161 u) at different
temperatures for a partial pressure of 3 × 10−3 mbar. (a) Electron-
induced fragmentation of DTBAP at 50 °C. Due to a comparably low
filament current of 1.3 A and a short time between the gas inlet and
the ionization with 70 eV, only a few fragments with low intensity are
detected. (b) Decomposition spectrum of DTBAP at 530 °C. Mainly
isobutane (58 u), tert-butyl radicals (57 u), isobutene (56 u), and
NH3 (17 u) are detected. The detected carbon chains at 40, 42, and
44 u can be attributed to electron-induced fragmentation of isobutane
and isobutene.
The surfaces look very similar, but the sample grown on
GaP/Si shows a wavylike modulation in the micrometer lateral
range as mentioned above and the sample grown on GaP does
not show these modulations. This implies that these
modulations occur due to the growth on Si. The N
incorporation of 2.1
0.1% in layers grown on GaP/Si is
significantly higher than that on GaP (1.4 0.1%). In general,
the effects of lattice pulling and lattice latching have been
described in the literature and can describe the observed
incorporation effect. The epitaxial growth of thin layers with a
lattice constant different from that of the substrate introduces a
strain energy due to lattice deformation, which increases the
total free energy of the system. One way to reduce the total
free energy is a change in the alloy composition and, therefore,
a change in lattice constant and mismatch.39 For the material
system (Ga,In)P the effect of lattice mismatch on the
composition of the solid was investigated in 1972 for the
are normalized to the signal of DTBAP at 161 u. At this
temperature no decomposition occurs and all fragmentation
products within this figure are induced by electron impact
ionization. Using optimized parameters within the ion trap, the
electron-induced fragmentation is reduced to a minimum.
Therefore, most DTBAP molecules are not fragmented by the
electrons and only a few fragments such as isobutane (58 u)
and probably C4H9P•• (88 u) are visible. This simplifies the
interpretation of the DTBAP decomposition.
1777
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
The thermally induced decomposition of DTBAP at a
temperature of 530 °C can be seen in Figure 1b. The intensity
of DTBAP completely vanishes at this temperature, indicating
complete decomposition of the DTBAP. As fragments mainly
isobutane (58 u), tert-butyl radicals (57 u), isobutene (56 u),
and NH3 (17 u) are detected. These can all be attributed to
thermal decomposition. Isobutane and tert-butyl radicals are
the products of radical decomposition of the C−P bond within
the molecule. This radical cleavage leads to the formation of
tert-butyl radicals that can react with available H radicals or H2
to form isobutane. Furthermore, there is the possibility of an
intramolecular coupling reaction of the tert-butyl group with an
H atom of the aminyl group, resulting in formation of
isobutane. These reaction pathways have been predicted by
density functional theory calculations for the As analogue
DTBAA and are on the one hand shown in decomposition
studies of this precursor. On the other hand these pathways are
also known for other metal−organic precursors with tert-butyl
groups such as TBAs.16,25,44 Another expected reaction
pathway of the tert-butyl groups is decomposition through β-
H elimination. In this case a H atom at one of the C atoms in a
β position within the tert-butyl group interacts with the p
orbitals of the phosphorus atom of the same molecule. This
leads to the formation of a CC double bond (formation of
isobutene) and a PH group. The corresponding reaction
scheme can be found in ref 16 for the DTBAA molecule. β-H
elimination seems to be a relevant decomposition pathway of
at least one of the tert-butyl groups of the DTBAP, since
isobutene was detected within the same magnitude as that for
the tert-butyl radicals and the isobutane.
pathway for P incorporation shown in Scheme 2. The aminyl
radicals are expected to be incorporated at low temperatures.
Scheme 2. Proposed Reaction Pathway
The temperature-dependent decomposition of DTBAP is
shown in Figure 8. All major decomposition products
In the decomposition spectrum also lighter alkyl groups are
visible. From former investigations of appropriate test gases,
the intensities of the smaller alkyl chains around 42 and 40 u
can be attributed to electron-induced fragmentation of
isobutane and isobutene. The test gases have been investigated
under the same ion trap settings and showed formation of
propene (42 u) and propyne (40 u) in the case of isobutane
and formation of propyne (40 u) for isobutene.
Figure 8. Decomposition curves of DTBAP (161 u) and DTBAA
(205 u) for partial pressures of 3 × 10−3 mbar and 2.5 × 10−3 mbar,
respectively. The most relevant decomposition fragments isobutane
(58 u), tert-butyl radicals (57 u), isobutene (56 u), and NH3 (17 u) of
the DTBAP decomposition are included. Data of DTBAA from ref 16
measured within the same reactor are included as a comparison.
Additionally, NH3 was detected with the same intensity as
that for DTBAP. The detected NH3 is attributed to the
•
formation of aminyl radicals (NH2 16 u) during the
decomposition of DTBAP and was also observed as a
decomposition product of DTBAA.16 This indicates a radical
decomposition of the N−P bond of the molecule, which leads
discussed above are included. As a comparison the
decomposition curve of the As analogue DTBAA is included
from ref 16. In comparison to DTBAA (Tdecomp = 160 10
°C) the DTBAP decomposes at a higher temperature of
to the formation of NH2 and DTBP• (145 u). No evidence
•
for DTBP• has been found within this measurement, but the
As analogue DTBA• has been detected in the case of DTBAA.
Therefore, it is expected that the DTBP• fragments are either
less stable or more reactive in comparison to DTBA• in the
decomposition of DTBAA. Either this is caused by dissociation
into isobutane, tert-butyl radicals, isobutene and P or the
DTBP• adsorbs first to an available surface before dissociation
and is therefore not detectable within the ion trap. No further
fragments containing P have been measured. Any elemental P
arising from the reaction would react to give P2 or P4, which is
seen as a coating of the liner in the reactor. The NH2• radicals
formed explain to some extent the N incorporation behavior
T
decomp = 310 10 °C, which is defined as the point where the
DTBAP signal drops to half of its intensity. This change in
decomposition temperature is known, for example, for
precursors such as TBAs (Tdecomp* = 350
10 °C) in
comparison to TBP (Tdecomp* = 400 10 °C). (Values marked
by * measured within the same reactor system. These
determined temperatures are 30−50 °C lower in comparison
to values from refs 25 and 45. This might be caused by the
different experimental setups, different surface areas within the
experiment, or different temperature measurement.) Immedi-
ately as the DTBAP decomposes, the intensities of isobutane,
tert-butyl radicals, isobutene and NH3 are increasing.
Regarding the decomposition of the tert-butyl groups the
radical or intramolecular coupling reaction seems to be the
most relevant pathway over the whole temperature range, since
isobutane summed together with the tert-butyl radicals shows
the highest intensities. Nevertheless, the isobutane signal drops
again for higher temperatures of around 450 °C. This shows
•
seen in Figure 5a. At high temperatures NH2 is very likely to
form NH3 in the presence of H2 or H• and can therefore easily
desorb from the growth surface or be generated in the gas
•
phase. For low temperatures NH2 radicals are more likely to
stick to the surface, leading to the efficient incorporation of N.
These considerations lead to the predicted decomposition
1778
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
similarly to the DTBAA decomposition that one of the radical
pathways might be hindered and the β-H elimination reaction
becomes more relevant above this temperature. This is shown
by the isobutene intensity, which is still increasing, while the
isobutane intensity is already constant and remains stable
within the same magnitude at higher temperatures.
An Arrhenius plot of the DTBAP decomposition data from
Figure 8 and the nitrogen incorporation data from Figure 5a is
shown in Figure 9. From the linear regression one can extract
6. SUMMARY
In this work the novel N and P precursor di-tert-
butylaminophosphane (DTBAP) was introduced and the
very first MOVPE growth experiments and gas-phase
decomposition measurements were carried out. The first
evidence of significant N incorporation in GaP on Si and GaP
substrates, resulting in good structural quality samples, was
shown. The growth parameters were varied systematically, and
the influence on N incorporation, growth rate, and surface
morphology were determined to create a foundation for further
growth studies aiming toward device applications.
Offering the additional P precursor TBP leads to a lowering
of N incorporation due to competition for group V lattice sites
and a decreasing N/P ratio. Increasing the amount of TEGa
leads to a linear increase in growth rate and for higher V/III
ratios to a decrease in N incorporation.
As for the precursor DTBAA, an increasing partial pressure
of DTBAP leads to a linear increase in N incorporation. Since
the N/P ratio is fixed to unity within the molecule, this is
unexpected on first sight. Some mechanisms were discussed,
but this behavior has to be checked in future experiments to
give a reliable answer.
Increasing the growth temperature leads to an exponential
decrease of N incorporation in the grown Ga(N,P) samples. As
possible reasons thermally activated loss processes of the active
nitrogen species due to desorption or gas-phase reactions were
discussed. These correlate to the formation of aminyl radicals
found in the decomposition study of DTBAP. The observed
linear increase in growth rate is most likely related to a partial
surface blocking by adsorbed alkyl groups at low temperatures.
Additionally, a N content of over 10% was confirmed with
HRXRD for a growth temperature of 475 °C, which has not
been realized for Ga(N,P) samples grown by MOVPE before.
Nitrogen incorporation on the Si substrate was significantly
higher than that on GaP, which is related to lattice-pulling
effects caused by the strain compensation for growth on Si.
The decomposition temperature of DTBAP was determined
with Tdecomp = 310 °C from the decomposition curve, which
makes it suitable for low-temperature growth in MOVPE. The
decomposition studies suggest for DTBAP the same reaction
pathway as for DTBAA.16 The tert-butyl groups either undergo
radical and intramolecular coupling reactions leading to
formation of isobutane and tert-butyl radicals or dissociate
from the molecule by β-H elimination, which is detected by
the appearance of isobutene. These larger alkyl groups,
especially the stable isobutene, should lead to lower C
incorporation into the growing layer in comparison to
structures grown with UDMHy. Furthermore, the N−P bond
of DTBAP decomposes radically under the formation of
Figure 9. Arrhenius plot of the detected intensities of DTBAP and
NH3 together with the N concentration from Figure 5a. Activation
energies and attempt frequency factors A0 for the decomposition and
N incorporation can be determined from the slope according to the
Arrhenius equation shown in the inset.
the activation energies as well as the pre-exponential attempt
frequency factors A0. The activation energies of the DTBAP
decomposition and the NH3 production as well as the N
incorporation are all determined with 1.4
0.2 eV. The
attempt frequency factors are 23 2 1/s for DTBAP, 28
2
1/s for NH3, and 19 2 1/s for the N incorporation, which
are all within a similar frequency. Notably the agreement in
activation energies supports the assumption, on the one hand,
that the N−P bond breaks within the first step of the DTBAP
decomposition, leading to the formation of the aminyl radicals
•
(NH2 ) and, on the other hand, that the formation of the
•
aminyl radicals (NH2 ) is directly related to the N
•
aminyl radicals (NH2 ). The NH2• radicals seem to be directly
incorporation. As mentioned above, at low temperatures
•
related to the N incorporation. This is shown by the activation
energy of 1.35 eV for N incorporation from the growth studies,
NH2 is likely to stick to the growth surface, leading to N
•
incorporation. At higher temperatures NH2 can react with
•
available H2 or H• radicals to form NH3, which is stable under
these conditions. This could enhance the desorption rate of N
from the surface or prevent the adsorption to the surface by
formation within the gas phase. These findings could fully
explain the exponential dependence of N incorporation in
Ga(N,P) on the growth temperature discussed in Figure 5a.
which is in agreement to the one obtained from NH2•
formation during thermal decomposition. Therefore, NH2
should explain the high N incorporation at low temperatures
•
due to adsorption of NH2 . On the other hand, formation of
NH3 in the gas phase or at the surface leads to prevention of N
incorporation at higher temperatures.
1779
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
(2) Shan, W.; Yu, K. M.; Walukiewicz, W.; Wu, J.; Ager, J. W.;
Haller, E. E. Band Anticrossing in Dilute Nitrides. J. Phys.: Condens.
Matter 2004, 16 (31), S3355−S3372.
AUTHOR INFORMATION
■
Corresponding Authors
Johannes Glowatzki − Material Sciences Center and
Department of Physics, Philipps-Universitat Marburg, 35032
(3) Kunert, B.; Zinnkann, S.; Volz, K.; Stolz, W. Monolithic
Integration of Ga(NAsP)/(BGa)P Multi-Quantum Well Structures
on (0 0 1) Silicon Substrate by MOVPE. J. Cryst. Growth 2008, 310
(23), 4776−4779.
̈
Marburg, Germany; orcid.org/0000-0002-8440-2271;
Email: Johannes.Glowatzki@physik.uni-marburg.de
(4) Kunert, B.; Volz, K.; Koch, J.; Stolz, W. Direct-Band-Gap
Ga(NAsP)-Material System Pseudomorphically Grown on GaP
Substrate. Appl. Phys. Lett. 2006, 88 (18), 182108.
Oliver Maßmeyer − Material Sciences Center and Department
̈
of Physics, Philipps-Universitat Marburg, 35032 Marburg,
Germany; orcid.org/0000-0001-6101-5706;
(5) Kunert, B.; Volz, K.; Nemeth, I.; Stolz, W. Luminescence
Investigations of the GaP-Based Dilute Nitride Ga(NAsP) Material
System. J. Lumin. 2006, 121 (2), 361−364.
Email: Oliver.Maßmeyer@physik.uni-marburg.de
Authors
(6) Borck, S.; Chatterjee, S.; Kunert, B.; Volz, K.; Stolz, W.; Heber,
̈
Marcel Koster − Material Sciences Center and Department of
J.; Ruhle, W. W.; Gerhardt, N. C.; Hofmann, M. R. Lasing in Optically
̈
Pumped Ga(NAsP)/GaP Heterostructures. Appl. Phys. Lett. 2006, 89
(3), 031102−3.
(7) Kunert, B.; Reinhard, S.; Koch, J.; Lampalzer, M.; Volz, K.; Stolz,
W. First Demonstration of Electrical Injection Lasing in the Novel
Dilute Nitride Ga(NAsP)/GaP-Material System. Phys. Status Solidi C
2006, 3 (3), 614−618.
(8) Liebich, S.; Zimprich, M.; Beyer, A.; Lange, C.; Franzbach, D. J.;
Chatterjee, S.; Hossain, N.; Sweeney, S. J.; Volz, K.; Kunert, B.; Stolz,
W. Laser Operation of Ga(NAsP) Lattice-Matched to (001) Silicon
Substrate. Appl. Phys. Lett. 2011, 99 (7), 071109.
(9) Geisz, J. F.; Olson, J. M.; Friedman, D. J.; Jones, K. M.; Reedy, R.
C.; Romero, M. J. Lattice-Matched GaNPAs-on-Silicon Tandem Solar
Cells. Conf. Rec. Thirty-first IEEE Photovolt. Spec. Conf. 2005, 695−
698.
(10) Sukrittanon, S.; Liu, R.; Breeden, M. C.; Pan, J. L.; Jungjohann,
K. L.; Tu, C. W.; Dayeh, S. A. Radial Direct Bandgap P-i-n GaNP
Microwire Solar Cells with Enhanced Short Circuit Current. J. Appl.
Phys. 2016, 120 (5), 055702.
̈
Chemistry, Philipps-Universitat Marburg, 35032 Marburg,
Germany
Thilo Hepp − Material Sciences Center and Department of
̈
Physics, Philipps-Universitat Marburg, 35032 Marburg,
Germany
Ebunoluwa Odofin − Material Sciences Center and Department
̈
of Physics, Philipps-Universitat Marburg, 35032 Marburg,
Germany
̈
Carsten von Hanisch − Material Sciences Center and
̈
Department of Chemistry, Philipps-Universitat Marburg, 35032
Marburg, Germany
Wolfgang Stolz − Material Sciences Center and Department of
̈
Physics, Philipps-Universitat Marburg, 35032 Marburg,
Germany
Kerstin Volz − Material Sciences Center and Department of
̈
Physics, Philipps-Universitat Marburg, 35032 Marburg,
Germany
(11) Sukrittanon, S.; Kuang, Y. J.; Dobrovolsky, A.; Kang, W. M.;
Jang, J. S.; Kim, B. J.; Chen, W. M.; Buyanova, I. A.; Tu, C. W.
Growth and Characterization of Dilute Nitride GaNxP1−x Nanowires
and GaNxP1−x/GaNyP1−y Core/Shell Nanowires on Si (111) by
Gas Source Molecular Beam Epitaxy. Appl. Phys. Lett. 2014, 105 (7),
072107.
(12) Sukrittanon, S.; Liu, R.; Ro, Y. G.; Pan, J. L.; Jungjohann, K. L.;
Tu, C. W.; Dayeh, S. A. Enhanced Conversion Efficiency in Wide-
Bandgap GaNP Solar Cells. Appl. Phys. Lett. 2015, 107 (15), 153901.
(13) Kargar, A.; Sukrittanon, S.; Zhou, C.; Ro, Y. G.; Pan, X.; Dayeh,
S. A.; Tu, C. W.; Jin, S. GaP/GaNP Heterojunctions for Efficient
Solar-Driven Water Oxidation. Small 2017, 13 (21), 1603574.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.0c00078
Author Contributions
§J.G. and O.M. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the German Research Foundation
(GRK 1782: “Functionalization of Semiconductors”), and the
support for the mass spectrometer (iTrap) was provided by
Carl Zeiss SMT GmbH.
(14) Jurgensen, H. Large-Scale MOVPE Production Systems.
̈
Microelectron. Eng. 1992, 18 (1−2), 119−148.
(15) Kunert, B.; Koch, J.; Torunski, T.; Volz, K.; Stolz, W. MOVPE
Growth Experiments of the Novel (GaIn)(NP)/GaP Material System.
J. Cryst. Growth 2004, 272 (1−4), 753−759.
ABBREVIATIONS
■
(16) Maßmeyer, O.; Inacker, S.; Hepp, T.; Glowatzki, J.;
AFM, atomic force microscopy; dilute nitrides, III/V semi-
conductors containing small amounts of nitrogen; DTBA•, di-
tert-butylarsane radical; DTBAA, di-tert-butylaminoarsane;
DTBAP, di-tert-butylaminophosphane; DTBP•, di-tert-butyl-
phosphane radical; HRXRD, high-resolution X-ray diffraction;
LEDs, light-emitting diodes; MBE, molecular beam epitaxy;
MOVPE, metal−organic vapor-phase epitaxy; MQWs, multi
quantum wells; NH•, aminyl radical; NMR, nuclear magnetic
resonance; QW, quantum well; RMS, root mean square; TBAs,
tert-butylarsane; TBP, tert-butylphosphane; TEGa, triethylgal-
lium; UDMHy, 1,1-dimethylhydrazine
̈
Nattermann, L.; Sterzer, E.; Ritter, C.; Von Hanisch, C.; Stolz, W.;
Volz, K. Decomposition Mechanisms of Di-Tert-Butylaminoarsane
(DTBAA). Organometallics 2019, 38 (16), 3181−3186.
̈
(17) Sterzer, E.; Ringler, B.; Nattermann, L.; Beyer, A.; von Hanisch,
C.; Stolz, W.; Volz, K. (GaIn)(NAs) Growth Using Di-Tertiary-Butyl-
Arsano-Amine (DTBAA). J. Cryst. Growth 2017, 467, 132−136.
(18) Sterzer, E.; Maßmeyer, O.; Nattermann, L.; Jandieri, K.; Gupta,
̈
S.; Beyer, A.; Ringler, B.; Von Hanisch, C.; Stolz, W.; Volz, K. 1 EV
Ga(NAsSb) Grown by MOVPE Using Di- Tertiary -Butyl-Arsano-
Amine (DTBAA). AIP Adv. 2018, 8 (5), 055329.
(19) Sterzer, E.; Beyer, A.; Duschek, L.; Nattermann, L.; Ringler, B.;
̈
̈
Leube, B.; Stegmuller, A.; Tonner, R.; Von Hanisch, C.; Stolz, W.;
Volz, K. Efficient Nitrogen Incorporation in GaAs Using Novel Metal
Organic As-N Precursor Di-Tertiary-Butyl-Arsano-Amine (DTBAA).
J. Cryst. Growth 2016, 439, 19−27.
(20) Zanella, P.; Porchia, M.; Rossetto, G.; Brianese, N.; Ossola, F.;
Williams, J. O. Organometallic Precursors in the Growth of Epitaxial
REFERENCES
■
(1) Wu, J.; Shan, W.; Walukiewicz, W. Band Anticrossing in Highly
Mismatched III−V Semiconductor Alloys. Semicond. Sci. Technol.
2002, 17, 860−869.
1780
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781

Organometallics
pubs.acs.org/Organometallics
Article
Thin Films of III-V Semiconductors by Metal-Organic Chemical
Vapor Deposition (MOCVD). Chem. Mater. 1991, 3 (2), 225−242.
(21) Lee, R. T.; Stringfellow, G. B. Pyrolysis of 1,1 Dimethylhy-
drazine for OMVPE Growth. J. Electron. Mater. 1999, 28 (8), 963−
969.
(22) Scherer, O. J.; Schieder, G. Metallorganische Aminophosphine
Und Phosphinimine. Chem. Ber. 1968, 101 (12), 4184−4198.
(23) Nattermann, L.; Maßmeyer, O.; Sterzer, E.; Derpmann, V.;
Chung, H. Y.; Stolz, W.; Volz, K. An Experimental Approach for Real
Time Mass Spectrometric CVD Gas Phase Investigations. Sci. Rep.
2018, 8 (319), 1−7.
(42) Sanorpim, S.; Katayama, R.; Onabe, K.; Usami, N.; Nakajima,
K. Lattice-Latching Effect in Metalorganic Vapor Phase Epitaxy
Growth of InGaAsN Film Lattice-Matched to Bulk InGaAs Substrate.
Jpn. J. Appl. Phys. 2010, 49 (4), 040202.
(43) Kunert, B.; Volz, K.; Koch, J.; Stolz, W. MOVPE Growth
Conditions of the Novel Direct Band Gap, Diluted Nitride Ga(NAsP)
Material System Pseudomorphically Strained on GaP-Substrate. J.
Cryst. Growth 2007, 298, 121−125.
(44) Stringfellow, G. B. Development and Current Status of
Organometallic Vapor Phase Epitaxy. J. Cryst. Growth 2004, 264
(4), 620−630.
(45) Li, S. H.; Larsen, C. A.; Buchan, N. I.; Stringfellow, G. B.
Pyrolysis of Tertiarybutylphosphine. J. Electron. Mater. 1989, 18 (3),
457−464.
(24) Chung, H. Y.; Aliman, M.; Fedosenko, G.; Laue, A.; Reuter, R.;
Derpmann, V.; Gorkhover, L.; Antoni, M. Very Sensitive Real-Time
Inline Process Mass Spectrometer Based on FFT Ion Trap
Technique. Annu. SEMI Adv. Semicond. Manuf. Conf. ASMC 2016
27th 2016, 263−266.
(25) Larsen, C. A.; Buchan, N. I.; Li, S. H.; Stringfellow, G. B.
Decomposition Mechanisms of Tertiarybutylarsine. J. Cryst. Growth
1989, 94 (3), 663−672.
(26) Stringfellow, G. B. Calculation of the Solubility and Solid-Gas
Distribution Coefficient of N in GaP. J. Electrochem. Soc. 1972, 119
(12), 1780−1782.
(27) Kaprinski, J.; Jun, J.; Grzegory, I.; Bugajski, M. Crystal Growth
of GaP Doped with Nitrogen under High Nitrogen Pressure. J. Cryst.
Growth 1985, 72 (3), 711−716.
(28) Ohtake, A. Surface Reconstructions on GaAs(001). Surf. Sci.
Rep. 2008, 63 (7), 295−327.
(29) Richter, W. In-Situ Observation of MOVPE Epitaxial Growth.
Appl. Phys. A: Mater. Sci. Process. 2002, 75 (1), 129−140.
̈
(30) Doscher, H.; Hannappel, T. In Situ Reflection Anisotropy
Spectroscopy Analysis of Heteroepitaxial GaP Films Grown on
Si(100). J. Appl. Phys. 2010, 107 (12), 123523.
(31) Law, D. C.; Sun, Y.; Hicks, R. F. Reflectance Difference
Spectroscopy of Gallium Phosphide(001) Surfaces. J. Appl. Phys.
2003, 94 (9), 6175−6180.
(32) Maßmeyer, O.; Sterzer, E.; Nattermann, L.; Stolz, W.; Volz, K.
Influence of UDMHy on GaAs (0 0 1) Surface Reconstruction before
and during Growth of Ga(NAs) by MOVPE. Appl. Surf. Sci. 2018,
458 (15), 512−516.
(33) Zhang, S. B.; Zunger, A. Surface-Reconstruction-Enhanced
Solubility of N, P, As, and Sb in III-V Semiconductors. Appl. Phys.
Lett. 1997, 71 (5), 677−679.
(34) Bi, W. G.; Tu, C. W. N Incorporation in GaP and Band Gap
Bowing of GaNxP1−x. Appl. Phys. Lett. 1996, 69 (24), 3710−3712.
(35) Lee, P. W.; Omstead, T. R.; McKenna, D. R.; Jensen, K. F. In
Situ Mass Spectroscopy and Thermogravimetric Studies of GaAs
MOCVD Gas Phase and Surface Reactions. J. Cryst. Growth 1987, 85
(1−2), 165−174.
(36) Yoshida, M.; Watanabe, H.; Uesugi, F. Mass Spectrometric
Study of Ga(CH3)3 and Ga(C2H5)3 Decomposition Reaction in H2
and N2. J. Electrochem. Soc. 1985, 132 (3), 677−679.
(37) French, C. L.; Foord, J. S. Reaction Kinetics for the CBE
Growth of GaAs from Triethylgallium; Computer Modelling Studies
Incorporating Recent Surface Spectroscopic Data. J. Cryst. Growth
1992, 120 (1−4), 63−70.
(38) Murrell, A. J.; Wee, A. T. S.; Fairbrother, D. H.; Singh, N. K.;
Foord, J. S.; Davies, G. J.; Andrews, D. A. Surface Chemical Processes
in Metal Organic Molecularbeam Epitaxy; Ga Deposition from
Triethylgallium on GaAs(100). J. Appl. Phys. 1990, 68 (8), 4053−
4063.
(39) Stringfellow, G. B. Organometallic Vapour Phase Epitaxy: Theory
and Practice, 2nd ed.; Academic Press: 1999.
(40) Stringfellow, G. B. The Importance of Lattice Mismatch in the
Growth of GaxIn 1-XP Epitaxial Crystals. J. Appl. Phys. 1972, 43 (8),
3455−3460.
(41) Stringfellow, G. B. Microstructures Produced during the
Epitaxial Growth of InGaN Alloys. J. Cryst. Growth 2010, 312 (6),
735−749.
1781
https://dx.doi.org/10.1021/acs.organomet.0c00078
Organometallics 2020, 39, 1772−1781