Optimization of the surface and structural quality of N-face InN grown by molecular beam epitaxy

Article abstract of DOI:10.1063/1.2335685

The authors demonstrate the impact of growth kinetics on the surface and structural properties of N-face InN grown by molecular beam epitaxy. Superior surface morphology with step-flow growth features is achieved consistently under In-rich conditions in a low-temperature region of 500-540 °C. Remarkably, off-axis x-ray rocking curve (ω scans) widths are found to be independent of the growth conditions. The band gap determined from optical absorption measurements of optimized InN is 0.651 eV, while photoluminescence peak emission occurs at even lower energies of ～0.626 eV. Hall measurements show room temperature peak electron mobilities as high as 2370 cm²/V s at a carrier concentration in the low 10¹⁷ cm^-3 region. Analysis of the thickness dependence of the carrier concentration demonstrates an-type surface accumulation layer with a sheet carrier concentration of ～3 × 10^-3 cm^-2.

Full text of DOI:10.1063/1.2335685

Optimization of the surface and structural quality of N-face InN grown by molecular
beam epitaxy
G. Koblmüller, C. S. Gallinat, S. Bernardis, J. S. Speck, G. D. Chern, E. D. Readinger, H. Shen, and M. Wraback
Citation: Applied Physics Letters 89, 071902 (2006); doi: 10.1063/1.2335685
View online: http://dx.doi.org/10.1063/1.2335685
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/89/7?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
N-type conductivity and properties of carbon-doped InN(0001) films grown by molecular beam epitaxy
J. Appl. Phys. 113, 033501 (2013); 10.1063/1.4775736
Characterization of high quality InN grown on production-style plasma assisted molecular beam epitaxy system
J. Vac. Sci. Technol. A 26, 399 (2008); 10.1116/1.2899412
Heteroepitaxial growth of In-face InN on GaN (0001) by plasma-assisted molecular-beam epitaxy
J. Appl. Phys. 97, 113520 (2005); 10.1063/1.1923166
Properties of InN layers grown on 6H–SiC(0001) by plasma-assisted molecular beam epitaxy
Appl. Phys. Lett. 84, 1671 (2004); 10.1063/1.1668318
Effects of film polarities on InN growth by molecular-beam epitaxy
Appl. Phys. Lett. 83, 251 (2003); 10.1063/1.1592309
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
89.132.144.249 On: Mon, 12 May 2014 14:18:23

APPLIED PHYSICS LETTERS 89, 071902 ͑2006͒
Optimization of the surface and structural quality of N-face InN
grown by molecular beam epitaxy
G. Koblmüller,a͒ C. S. Gallinat, S. Bernardis, and J. S. Speck
Materials Department, University of California, Santa Barbara, California 93106-5050
G. D. Chern, E. D. Readinger, H. Shen, and M. Wraback
U.S. Army Research Laboratory, Sensors and Electron Devices Directorate, 2800 Powder Mill Road,
Adelphi, Maryland 20783
͑Received 29 March 2006; accepted 26 June 2006; published online 14 August 2006͒
The authors demonstrate the impact of growth kinetics on the surface and structural properties of
N-face InN grown by molecular beam epitaxy. Superior surface morphology with step-flow growth
features is achieved consistently under In-rich conditions in a low-temperature region of
500–540 °C. Remarkably, off-axis x-ray rocking curve ͑␻ scans͒ widths are found to be
independent of the growth conditions. The band gap determined from optical absorption
measurements of optimized InN is 0.651 eV, while photoluminescence peak emission occurs at
even lower energies of ϳ0.626 eV. Hall measurements show room temperature peak electron
mobilities as high as 2370 cm2/V s at a carrier concentration in the low 1017 cm−3 region. Analysis
of the thickness dependence of the carrier concentration demonstrates a n-type surface accumulation
layer with a sheet carrier concentration of ϳ3ϫ1013 cm−2. © 2006 American Institute of Physics.
͓DOI: 10.1063/1.2335685͔
Over the past few years, InN has fueled tremendous in-
terest because the fundamental band gap energy has been
revised to approximately 0.7 eV.1–3 Consequently, InN is be-
coming increasingly attractive not only as a near-infrared
͑IR͒ material for optoelectronic applications, including its
possible performance as a terahertz emitter,4 but also for ex-
tending the emission of light emitting diodes from deep ul-
traviolet to near IR, if alloyed with GaN or AlN.5 Beyond
this, superior electronic transport properties, i.e., high elec-
tron peak velocities6 and mobilities, are expected due to its
small electron effective mass of Ϸ0.04m0.6,7 Indeed, a few
experimental studies of undoped InN point to room tempera-
ture electron mobilities exceeding 2000 cm2/V s.8,9
For the In-face orientation, numerous attempts, using
mainly plasma-assisted molecular beam epitaxy ͑MBE͒,
have been undertaken to generate good crystal and surface
quality for improved optical and transport properties. The
implementation of various buffer layers, such as AlN,10 low-
temperature InN,11 and GaN,9,12 has helped especially to re-
duce free electron carrier concentration and sharpen the
optical absorption edge, partly overcoming the lack of a
near-lattice-matched substrate. However, the low InN de-
composition temperature presents significant growth chal-
lenges, limiting the growth temperature to Ϸ500 °C.13,14
Most recently, the challenges of InN deposition have
been addressed by growing films along the thermally more
stable N-face orientation,15,16 where growth temperatures up
to 600 °C seem feasible. Higher structural quality InN films
with sharp step-flow surface features may be achieved, even
under N-rich growth conditions. This appears contradictory
to the established knowledge of high adatom diffusion under
purely metal-rich conditions, as in GaN ͑Ref. 17͒ or AlN
͑Ref. 18͒ growth. In this letter, we investigate these impor-
tant aspects of N-face InN growth, with special emphasis on
the effect of growth kinetics on the advancement of the
physical properties.
As substrates for N-face growth, we used high-quality
freestanding N-face GaN templates provided by Lumilog
with threading dislocation densities specified to be less than
1ϫ108 cm−2. To produce a surface free of contamination and
of good interface quality we have grown an Ϸ40 nm thick
GaN buffer layer under Ga-rich conditions at a temperature
of 700 °C. Both the GaN buffer and InN layer growths were
performed in an Epi 620 MBE system using standard effu-
sion cells for Ga and In, and an Epi Unibulb radio frequency
plasma source for supplying active nitrogen. A pyrometer
was used to measure the substrate temperature in situ. We
give all molecular fluxes in terms of potential growth rate in
units of nm/min, as calibrated from thickness measurements
of InN films grown under In-rich and N-rich conditions at
temperatures for which thermal decomposition was negli-
gible. High-resolution x-ray rocking curves ͑␻ scans͒ and
␻-2␪ scans were performed using a triple-axis diffractometer
with a resolution limit of ϳ10 arc sec. The structural quality
of the InN films was correlated with the surface morphology
by employing atomic force microscopy ͑AFM͒. In addition,
we have selected optimized InN films on N-face GaN layers
͑200 nm thick͒ grown on NOVASiC polished C-face 6H–SiC
substrates19,20 for optical spectroscopy, using a combination
of photoluminescence ͑PL͒ and optical absorption obtained
from reflectance and transmission measurements. To identify
transport properties, we performed Hall measurements in the
van der Pauw geometry on as-grown InN layers on the buff-
ered, semi-insulating C-face 6H–SiC substrates.
Reflection high energy electron diffraction ͑RHEED͒
was used to confirm the N-face growth direction prior to
each InN growth. By cooling the GaN buffer layer to tem-
peratures below ϳ620 °C we observed a distinct 3ϫ3
RHEED reconstruction, which is a well known characteristic
of a N-face GaN growth surface.21 The N-face InN layers
presented here were grown at different temperatures and
a͒
Electronic mail: gregor@engineering.ucsb.edu
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/2006/89͑7͒/071902/3/$23.00 89, 071902-1 © 2006 American Institute of Physics
89.132.144.249 On: Mon, 12 May 2014 14:18:23

071902-2
Koblmüller et al.
Appl. Phys. Lett. 89, 071902 ͑2006͒
rich conditions are similar to those commonly observed for
GaN growth under metal-rich conditions.17 Recent experi-
ments have demonstrated that the GaN growth front is stabi-
lized by a metallic Ga adlayer,23 which promotes quite high
N-adatom surface diffusion. Analogously, we have found an
In adlayer on the N-face InN surface under In-rich conditions
with a saturation coverage of 1 ML ͑as determined by In
adsorption experiments using RHEED Bragg spot intensity
variations͒.22 High-temperature InN growth under N-rich
conditions, however, is remarkably different from what is
known for GaN growth. As shown in Fig. 1͑c͒, the terrace-
like surface structure exhibits step-flow features, as also ob-
served by Xu and Yoshikawa.15 These smooth surfaces indi-
cate that high surface diffusion of In and N adatoms must be
prevalent even in N-rich growth. We assume that the diffu-
sion barrier for In adatoms on a bare InN surface is likely to
be smaller than the barrier for Ga migration on a GaN
surface.
To correlate the AFM surface morphologies with the
structural quality, we have systematically determined the
x-ray rocking curve ͑␻ scans͒ widths ͓full width at half
maximum ͑FWHM͔͒ of on-axis and off-axis reflections ͑the
latter measured in a skew-symmetric geometry͒ of several
InN layers grown in the temperature range between 500 and
595 °C. For the more important off-axis scans we observe no
dependence between FWHM values and growth conditions
͑both temperature and In/N flux ratio͒, with values scattered
FIG. 1. AFM micrographs ͓10ϫ10 ␮m2 and 1ϫ1 ␮m2 ͑insets͔͒ of InN
layers grown on freestanding GaN; growth temperatures and In/N flux ra-
tios are ͑a͒ 530 °C and 0.9, ͑b͒ 530 °C and 1.4, ͑c͒ 595 °C and 0.9, and ͑d͒
595 °C and 1.4. The height scale of the inset images is 2 nm for all samples,
but for the larger images it is 50 nm for ͑a͒ and ͑c͒ and 20 nm for ͑b͒ and
͑d͒.
¯
randomly between 1020 and 1200 arc sec for the ͑1022͒ re-
In/N flux ratios. To achieve different flux ratios, we kept the
N flux constant at 7.5 nm/min and varied only the In flux,
from N-rich ͑In/N=0.9͒ to In-rich droplet conditions
͑In/N=1.4͒. The designation of these growth regimes and
their intricate relationship with thermal decomposition have
been elaborated in detail and will be published elsewhere.22
In Fig. 1, typical AFM surface morphologies are shown for
nominally ϳ0.9 ␮m thick InN layers grown at two tempera-
tures ͑530 and 595 °C͒ and In/N ratios of 0.9 and 1.4, re-
spectively. The measured InN thicknesses from cross-
sectional scanning electron microscopy coincided with the
nominal thickness for layers grown at 530 °C ͑ϳ0.87 ␮m͒,
but were slightly reduced due to thermal decomposition for
layers grown at 595 °C ͑ϳ0.67 ␮m͒. InN layers grown un-
der N-rich growth conditions exhibit the roughest surfaces
with root-mean-square ͑rms͒ values of 54.1 nm ͓Fig. 1͑a͔͒
and 29.6 nm ͓Fig. 1͑c͔͒ over a 10ϫ10 ␮m2 area. In contrast,
when the growth was performed under In-rich droplet con-
ditions and low temperature ͑T=530 °C͒ significantly im-
proved surface quality was achieved ͑rms roughness
=6.6 nm͒ ͓Fig. 1͑b͔͒. Imaging the areas in between the few
and randomly distributed In droplets displays large atomi-
cally flat terraces with clear step-flow growth features that
are separated by occasional pits and coalescence boundaries.
These step-flow features are characterized by arrays of
curved terraces ͑as part of spiral growth hillocks͒ with step
heights of ϳ3 Å corresponding to the monolayer height of
c/2=2.88 Å for bulk InN. Increasing the growth temperature
to 595 °C ͓Fig. 1͑d͔͒ resulted in much rougher surface
͑rms=18.7 nm͒. In this case, a higher density of metallic In
droplets was observed on the surface ͑including small nuclei
of droplets indicated by the arrows͒, as a result of increased
In accumulation due to significant thermal dissociation of
InN and limited In desorption at these temperatures.22 Note
¯
flection and between 1320 and 1550 arc sec for the ͑2021͒
reflection. These results indicate that the twist about the c
axis is negligibly impacted by the growth conditions. In ac-
cordance with investigations of GaN by Heying et al.,24 this
means that the majority of pure edge-type threading disloca-
tions ͑TDs͒ are essentially invariant with growth conditions.
In contrast, we observe a slight dependence of the on-axis
͑0002͒ FWHM values ͓i.e., representing minority pure screw
and mixed TDs ͑Ref. 24͔͒ on the growth conditions, giving
the lowest FWHM ͑ϳ320 arc sec͒ for layers grown under
In-rich conditions in a moderately low temperature range
͑500–540 °C͒. These values are comparable with the best
values reported for N-face InN ͑Ref. 15͒ and are in general
slightly better than those realized for In-face InN.14,25
The significance of choosing optimum conditions of low
growth temperature and In-rich conditions for the optical
properties of N-face InN is illustrated in Fig. 2. Room-
temperature PL, reflectance, and transmission data were ob-
tained on optimized InN layers grown on N-face GaN buffer
layers on C-face SiC substrates. The absorption coefficient
was determined from both the reflectance and transmission
data, and its square is plotted versus photon energy to deter-
mine the direct band gap of the InN. The intercept of the
linear fit with the energy axis provides a band gap of
0.651 eV. The PL peak emission from a thick sample was
determined using a Lorentzian fit to the data to occur at
ϳ0.626 eV, with a FWHM of 54 meV. The slight redshift
with respect to the band gap implies that the emission ema-
nates primarily from the bandtail states. The band gap and
PL peak emission energy are among the lowest reported val-
ues for wurtzite InN.1–3,14,15,26
Finally, the transport properties were evaluated using an
InN series with different thicknesses under identical and op-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
that the step-flow surface morphologies generated under In-
timized conditions ͑T=535 °C, In/N=1.4͒ on a 200 nm
89.132.144.249 On: Mon, 12 May 2014 14:18:23

071902-3
Koblmüller et al.
Appl. Phys. Lett. 89, 071902 ͑2006͒
structural quality and smooth step-flow-like surface mor-
phologies were achieved in a rather low-temperature region
of 500–540 °C under In-rich conditions. The structural qual-
ity of InN was facilitated by the multiple effect of negligible
thermal decomposition at these temperatures and the autosur-
factant action of a 1 ML thick In adlayer on the surface. The
band gap of 0.651 eV and room-temperature PL peak emis-
sion at 0.626 eV of our optimized InN are among the lowest
reported values. In addition, Hall mobilities reached
2370 cm2/V s at bulk carrier concentrations in the low
1017 cm−3, and a surface accumulation layer with an electron
density of ϳ3ϫ1013 cm−2 was determined.
The authors gratefully acknowledge C. G. Van de Walle
and D. Segev for many helpful discussions. This work was
supported by DARPA ͑CNID program͒ and AFOSR ͑D. J.
Silversmith, program manager͒.
FIG. 2. Band gap obtained from absorption data and PL spectrum from
optimized N-face InN samples.
1V. Yu. Davydov, A. A. Klochikhin, R. P. Seisyan, V. V. Emtsev, S. V.
Ivanov, F. Bechstedt, J. Furthmüller, H. Harima, A. V. Mudryi, J.
Aderhold, O. Semchinova, and J. Graul, Phys. Status Solidi B 229, R1
͑2002͒.
thick N-face GaN buffer layer grown on semi-insulating
C-face 6H–SiC substrates. To facilitate this we performed
Hall measurements at 300 K on the as-grown InN layers
͑containing a small density of In droplets͒, with their results
shown in Fig. 3. The Hall electron mobility increases con-
tinually with increasing layer thickness, reaching a maxi-
mum mobility of 2370 cm2/V s for a layer thickness of
4.4 ␮m. These electron mobilities are almost a factor of 2
higher, and the corresponding bulk carrier concentration of
2.8ϫ1017 cm−3 is nearly one order of magnitude lower than
those reported for the best MBE grown N-face InN.15 By
linear extrapolation of the sheet carrier concentration of the
InN layers to zero thickness ͑for a series of samples grown
to different InN thickness͒, a surface electron accumula-
tion layer density of ϳ3ϫ1013 cm−2 was determined. The
surface carrier concentration is in close agreement with
the ϳ2.5ϫ1013 cm−2 density determined independently by
energy electron loss spectroscopy and theoretical
calculations.27 Despite this consistency, we suggest that a
more detailed analysis is needed for differentiating between
bulk conduction and surface or interface conduction.
2J. Wu, W. Walukiewicz, K. M. Yu, J. W. Ager, E. E. Haller, H. Lu, W. J.
Schaff, Y. Saito, and Y. Nanishi, Appl. Phys. Lett. 80, 3967 ͑2002͒.
3T. Matsuoka, H. Okamoto, M. Nakao, H. Harima, and E. Kurimoto, Appl.
Phys. Lett. 81, 1246 ͑2002͒.
4R. Ascazubi, I. Wilke, K. Denniston, H. Lu, and W. J. Schaff, Appl. Phys.
Lett. 84, 4810 ͑2004͒.
5M. Higashiwaki and T. Matsui, Jpn. J. Appl. Phys., Part 2 41, L540
͑2002͒.
6V. M. Polyakov and F. Schwierz, Appl. Phys. Lett. 88, 032101 ͑2006͒.
7S. K. O’Leary, B. E. Foutz, M. S. Shur, and L. F. Eastman, Appl. Phys.
Lett. 87, 222103 ͑2005͒.
8J. Wu, W. Walukiewicz, W. Shan, K. M. Yu, J. W. Ager, S. X. Li, E. E.
Haller, H. Lu, and W. J. Schaff, J. Appl. Phys. 94, 4457 ͑2003͒.
9C. S. Gallinat, G. Koblmüller, S. Bernardis, J. S. Brown, J. S. Speck, G. D.
Chern, E. D. Readinger, H. Shen, and M. Wraback, Appl. Phys. Lett. 89,
032109 ͑2006͒.
10H. Lu, W. J. Schaff, J. Hwang, H. Wu, G. Koley, and L. F. Eastman, Appl.
Phys. Lett. 79, 1489 ͑2001͒.
11Y. Saito, N. Teraguchi, A. Suzuki, T. Araki, and Y. Nanishi, Jpn. J. Appl.
Phys., Part 2 40, L91 ͑2001͒.
12C. J. Lu, L. A. Bendersky, H. Lu, and W. J. Schaff, Appl. Phys. Lett. 83,
2817 ͑2003͒.
13E. Dimakis, E. Iliopoulos, K. Tsagaraki, and A. Georgakilas, Appl. Phys.
Lett. 86, 133104 ͑2005͒.
In summary, systematic growth kinetics studies were
conducted to optimize the surface and structural properties of
N-face InN grown on freestanding GaN. InN films with high
14T. Ive, O. Brandt, M. Ramsteiner, M. Giehler, H. Kostial, and K. H. Ploog,
Appl. Phys. Lett. 84, 1671 ͑2004͒.
15K. Xu and A. Yoshikawa, Appl. Phys. Lett. 83, 251 ͑2003͒.
16H. Naoi, F. Matsuda, T. Araki, A. Suzuki, and Y. Nanishi, J. Cryst. Growth
269, 155 ͑2004͒.
17B. Heying, R. Averbeck, L. F. Chen, E. Haus, H. Riechert, and J. S. Speck,
J. Appl. Phys. 88, 1855 ͑2000͒.
18G. Koblmüller, R. Averbeck, L. Geelhaar, H. Riechert, W. Hoesler, and P.
Pongratz, J. Appl. Phys. 93, 9591 ͑2003͒.
19S. Rajan, M. Wong, Y. Fu, F. Wu, J. S. Speck, and U. K. Mishra, Jpn. J.
Appl. Phys., Part 2 44, L1478 ͑2005͒.
20E. Monroy, E. Sarigiannidou, F. Fossard, N. Gogneau, E. Bellet-Amalric,
J.-L. Rouviere, S. Monnoye, H. Mank, and B. Daudin, Appl. Phys. Lett.
84, 3684 ͑2004͒.
21A. R. Smith, R. M. Feenstra, D. W. Greve, M. S. Shin, M. Skowronski, J.
Neugebauer, and J. E. Northrup, Appl. Phys. Lett. 72, 2114 ͑1998͒.
22G. Koblmüller, C. S. Gallinat, and J. S. Speck ͑unpublished͒.
23G. Koblmüller, R. Averbeck, H. Riechert, and P. Pongratz, Phys. Rev. B
69, 035325 ͑2004͒.
24B. Heying, X. H. Wu, S. Keller, Y. Li, D. Kapolnek, B. P. Keller, S. P.
DenBaars, and J. S. Speck, Appl. Phys. Lett. 68, 643 ͑1996͒.
25V. V. Mamutin, V. A. Vekshin, V. Y. Davydov, V. V. Ratnikov, T. V.
Shubina, S. V. Ivanov, P. S. Kopev, M. Karlsteen, U. Södervall, and M.
Willander, Phys. Status Solidi A 176, 247 ͑1999͒.
FIG. 3. Dependence of room-temperature Hall mobility and sheet carrier
concentration on layer thickness for InN grown on N-face GaN on C-face
6H–SiC templates. A linear extrapolation of the carrier concentration to zero
thickness yields an estimate of ϳ3ϫ1013 cm−2 for the surface electron
26M. Higashiwaki and T. Matsui, J. Cryst. Growth 269, 162 ͑2004͒.
27I. Mahboob, T. D. Veal, C. F. McConville, H. Lu, and W. J. Schaff, Phys.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
density.
Rev. Lett. 92, 036804 ͑2004͒.
89.132.144.249 On: Mon, 12 May 2014 14:18:23