High filling fraction gallium phosphide inverse opals by atomic layer deposition

Article abstract of DOI:10.1063/1.2387874

High filling fraction gallium phosphide (GaP) inverse opals were fabricated by atomic layer deposition within the void spaces of silica colloidal crystal templates. Depositions were performed from 400 to 500 °C using trimethylgallium and tris(dimethylamino)phosphine precursors. The resulting films were characterized by optical reflectance, which indicated infiltration as high as 100% of the conformal film growth maximum, corresponding to a volume filling fraction of 0.224. X-ray diffraction measurements confirmed the crystallinity of the film. These results indicate the fabrication of three-dimensional photonic crystals using a III-V optoelectronic material with sufficient dielectric contrast to form a full photonic band gap in the visible.

Full text of DOI:10.1063/1.2387874

High filling fraction gallium phosphide inverse opals by atomic layer deposition
E. Graugnard, V. Chawla, D. Lorang, and C. J. Summers
Citation: Applied Physics Letters 89, 211102 (2006); doi: 10.1063/1.2387874
View online: http://dx.doi.org/10.1063/1.2387874
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/89/21?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Optical characteristics of nanocrystalline AlxGa1−xN thin films deposited by hollow cathode plasma-assisted
atomic layer deposition
J. Vac. Sci. Technol. A 32, 031508 (2014); 10.1116/1.4870381
Atomic layer deposition of Al-doped ZnO thin films
J. Vac. Sci. Technol. A 31, 01A109 (2013); 10.1116/1.4757764
Photonic crystal thin films of GaAs prepared by atomic layer deposition
Appl. Phys. Lett. 89, 104103 (2006); 10.1063/1.2345359
Fabrication of inverted opal ZnO photonic crystals by atomic layer deposition
Appl. Phys. Lett. 86, 151113 (2005); 10.1063/1.1900957
High-filling-fraction inverted ZnS opals fabricated by atomic layer deposition
Appl. Phys. Lett. 83, 2566 (2003); 10.1063/1.1609240
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:
128.42.202.150 On: Sun, 23 Nov 2014 09:59:46

APPLIED PHYSICS LETTERS 89, 211102 ͑2006͒
High filling fraction gallium phosphide inverse opals by atomic layer
deposition
E. Graugnard,a͒ V. Chawla, D. Lorang, and C. J. Summersb͒
School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-
0245
͑Received 21 June 2006; accepted 25 September 2006; published online 20 November 2006͒
High filling fraction gallium phosphide ͑GaP͒ inverse opals were fabricated by atomic layer
deposition within the void spaces of silica colloidal crystal templates. Depositions were performed
from 400 to 500 °C using trimethylgallium and tris͑dimethylamino͒phosphine precursors. The
resulting films were characterized by optical reflectance, which indicated infiltration as high as
100% of the conformal film growth maximum, corresponding to a volume filling fraction of 0.224.
X-ray diffraction measurements confirmed the crystallinity of the film. These results indicate the
fabrication of three-dimensional photonic crystals using a III-V optoelectronic material with
sufficient dielectric contrast to form a full photonic band gap in the visible. © 2006 American
Institute of Physics. ͓DOI: 10.1063/1.2387874͔
Photonic band gap devices offer tremendous potential
for controlling the emission rates and propagation of light by
creating microcavity and waveguide structures with ex-
tremely high Q factors and low loss.1 For three-dimensional
͑3D͒ photonic crystal structures with a sufficiently high di-
electric contrast and appropriate crystal symmetry, an omni-
directional or full photonic band gap ͑FPBG͒ will exist
within the crystal, which forbids the propagation of light
within a certain range of frequencies.2,3 In order to form a
full photonic band gap in opal-based photonic crystals, the
face-centered-cubic lattice must be inverted and possess a
refractive index contrast greater than 2.8 for full infiltration
͑26% of volume͒ and 3.3 for complete conformal infiltration
͑22.4% of volume͒.4,5 To achieve these requirements, infil-
tration techniques have been developed for depositing high
refractive index material within the pores of an opal template
which is subsequently removed to create an inverted lattice
of air spheres in a high index backbone. Inverse opals with
full gaps in the infrared have been fabricated using Si, Ge,
GaAs, and InP,6–11 while other materials, including SnS2,
Sb2S3, GaP, and Ta3N5,12–15 have been investigated because
they offer the potential to form FPBGs in the visible.
Inverse opals fabricated from III-V materials are techno-
logically important due to the possibility of manipulating
their intrinsic optoelectronic properties, while also utilizing
those properties to create active photonic devices. However,
for GaP, which has both high index and transparency in the
visible spectral region, only low filling fractions or anoma-
lous behavior of the infiltrated material have been
reported.14,16 In the first study reported by Romanev et al.
using a cyclic chemical vapor deposition method with long
͑several hours͒ precursor hold times,16 the GaP filling frac-
tion was only ϳ3%, insufficient to form an inverse lattice.17
Using a similar cyclic chemical vapor deposition technique,
Palacios-Lidón et al. obtained higher filling fractions, but the
infiltrated GaP exhibited anomalous optical behavior result-
ing from a topology of spherical GaP grains within the opal
void spaces, deviating from surface limited growth.14 Al-
though the resulting inverse lattice deviated from a typical
inverse opal, certain structural parameters for the spherical
GaP topology were predicted to result in a FPBG.
In this letter, we report the fabrication of high filling
fraction GaP inverse shell opals by atomic layer deposition
͑ALD͒ and show the potential of GaP as a high index, trans-
parent material for photonic crystals. In previous work, ALD
of thin dielectric films has proven to be a powerful tool for
fabricating single and multicomponent inverse opals due to
its high precision and surface limited conformal growth
characteristics.15,18–21 Using ALD, GaP filling fractions as
high as 100% of the conformal film growth maximum were
achieved, and the inverse lattices were consistent with sur-
face limited film growth. The optical properties were in ex-
cellent agreement with photonic band calculations, indicating
that the formation of a full 3D photonic band gap should be
possible.
Opal films were formed on silicon substrates by forced
self-assembly of silica spheres within a 10 ␮m thick confine-
ment cell. The sphere diameters ranged from 297 to 450 nm,
resulting in face-centered-cubic films consisting of approxi-
mately 40–27 sphere layers, respectively, oriented with the
͗111͘ direction normal to the substrate. The samples were
infiltrated with a custom ALD reactor operated in the fill-
hold-purge mode to enhance precursor diffusion and efficient
precursor usage. In this mode, sequential precursor pulses
were delivered to the reaction chamber to a predetermined
partial pressure, held for a fixed time, and then purged from
the chamber. GaP ALD has been reported previously using
metal organic and hydride precursors.22–26 However, in this
study, optimal surface limited growth on SiO2 opal templates
was performed at 450 °C using trimethylgallium ͑TMGa͒
and tris͑dimethylamino͒phosphine ͑TDMAP͒, which is a
metal organic alternative to PH3 with a far lower toxicity
level.27 The TMGa pulses were held for 10 s at a partial
pressure of 300 mTorr and then purged with 300 SCCM
͑SCCM denotes cubic centimeter per minute at STP͒ of ni-
trogen for 20 s. The TDMAP was held for 10 s at 100 mTorr
and then purged for 60 s. A longer purge time for TDMAP
was used since its vapor pressure ͑3.5 Torr at 20 °C͒ is
a͒
Current address: Dept. of Engineering-Physics-Systems, Providence Col-
lege, Providence, RI 02918–0001.
b͒
Electronic mail: chris.summers@mse.gatech.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:
0003-6951/2006/89͑21͒/211102/3/$23.00 89, 211102-1 © 2006 American Institute of Physics
128.42.202.150 On: Sun, 23 Nov 2014 09:59:46

211102-2
Graugnard et al.
Appl. Phys. Lett. 89, 211102 ͑2006͒
FIG. 2. X-ray diffraction data for a GaP inverse opal. Labeled diffraction
peaks agree in both position and intensity with published data, confirming
the cubic GaP crystal structure. The inset shows an energy dispersion spec-
trum for an inverse opal, confirming the presence of only Ga and P within
the lattice.
structure exactly conforms to that expected for surface lim-
ited template-patterned polycrystalline film growth.
FIG. 1. ͑a͒ SEM image of the top surface of a 297 nm GaP infiltrated opal
showing conformal, polycrystalline film growth. The inset shows a higher
magnification image at a small gap between spheres, selected to reveal the
conformal film growth. ͑b͒ Cross section through ͑111͒ plane of a 450 nm
inverse opal. The shell-like structure of the GaP film confirms the conformal
nature of the growth mode. The inset shows details of the air cavity structure
and the high quality of the conformal GaP backbone for a 350 nm inverse
opal.
The crystal structure of a GaP film deposited on a fused
silica substrate was assessed by x-ray diffraction measure-
ments, and the resulting data are plotted in Fig. 2. The x-ray
diffraction data show clear peaks matching the intensity and
position of the diffraction pattern reported for bulk GaP, con-
firming the cubic F͑−4͒3m͑216͒ phase of the film. The inset
of Fig. 2 shows an x-ray energy dispersion spectrum ac-
quired at 20 keV for the 450 nm inverse opal, confirming the
presence of only Ga and P in the inverse backbone.
The specular reflectance of the 450 nm GaP inverse opal
film, plotted in the right panel of Fig. 3, displayed a Bragg
peak at 1.0 eV ͑1245 nm or a normalized frequency of 0.51͒
with a peak width ͑full width at half maximum/EBragg͒ of
22%. The spectrum was acquired using a microreflectivity
system with a spot diameter of ϳ100 ␮m. Reflected light
was routed through a vis-IR spectrometer to an InGaAs pho-
todiode detector. The data are compared to a full photonic
band calculation, performed using the MIT PHOTONIC BAND
package,29 for a GaP inverse opal having a conformal infil-
tration of 100% and a refractive index of 3.34 ͑GaP index at
600 nm͒.30 The calculated gap between the second and third
much lower than TMGa ͑182 Torr at 20 °C͒. During the
nitrogen purges, the pressure within the reactor was
ϳ1 Torr.
Figure 1͑a͒ shows a scanning electron microscopy
͑SEM͒ image of the top surface of a 297 nm GaP infiltrated
opal after 100 ALD cycles. The GaP film is polycrystalline
with a fine scale ͑Ͻ10 nm͒ grain structure and is conformal
to the spherical surfaces, as can be seen from the growth in
the regions of missing spheres. The inset shows a high mag-
nification image at a small gap between neighboring spheres,
specifically selected to illustrate the fine conformal grain
structure. This film morphology is similar to opals infiltrated
with ZnS by ALD at 500 °C, which also grows as a poly-
crystalline film.28 The infiltrated opal was ion milled to ob-
tain access to the silica sphere network and was etched three
times in 2% HF for 15 min to remove completely the silica
spheres, as confirmed when the Bragg peak ceased to shift to
a shorter wavelength. Figure 1͑b͒ shows the ion-milled cross
section of a 450 nm inverse opal midway through the outer-
most ͑111͒ plane of spheres. The lattice consists of spherical
air cavities within a GaP backbone grown with 300 ALD
cycles. The conformal nature of the GaP film growth is con-
firmed by the shell-like structure observed in the image,
which leaves small air pockets at the interstitial points be-
tween the spheres and is predicted to enhance the photonic
properties.4,5 Nonconformal growth is not surface limited
and would have completely filled these void regions, as ob-
served in the metal organic chemical vapor desposition
growth reported previously.14 The dark spots observed within
the cavities result from the contact points of the silica
spheres and provide the connectivity pathway for the lattice.
These connection pores, ϳ30 nm in diameter, and air pock-
ets at the interstitials are shown in the high magnification
inset, which shows the high structural quality of the confor-
FIG. 3. Photonic band structure calculated for a conformal filling fraction of
0.224 ͑complete conformal infiltration͒ compared to the measured reflec-
tance spectrum from a 450 nm GaP inverse opal. As indicated by arrows, the
measured spectrum agrees very well with the calculated gap positions and
widths, shaded in gray along the ⌫-L direction. The solid gray bar indicates
the ϳ3% full photonic band gap of the structure. The excellent agreement
between the calculated bands and measured reflectance directly confirms the
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:
mal GaP shell for a 350 nm inverse opal. The backbone
high GaP filling fraction.
128.42.202.150 On: Sun, 23 Nov 2014 09:59:46

211102-3
Graugnard et al.
Appl. Phys. Lett. 89, 211102 ͑2006͒
1W. Bogaerts, R. Baets, P. Dumon, V. Wiaux, S. Beckx, D. Taillaert, B.
Luyssaert, J. Van Campenhout, P. Bienstman, and D. Van Thourhout, J.
Lightwave Technol. 23, 401 ͑2005͒.
bands was centered at a normalized frequency of 0.516 with
a gap width of 18.5%. The peak width of the measured spec-
trum is slightly larger than the calculated width and is typi-
cally broadened by disorder within the opal. The reflectance
of the Bragg peak in the inverse opal was 48%, which is
lower than the initial opal reflectance of 70%. We attribute
the low reflectance to scattering from the polycrystalline film
and are currently investigating annealing procedures to re-
duce surface roughness as well as techniques to produce opal
templates with a higher initial optical quality. The sharp de-
crease in spectral intensity below 0.75 eV corresponds to the
spectral limit of the detector. However, the measured spec-
trum agrees very well with the calculated band structure ͑and
the Bragg-Snell equation͒, confirming the conformal nature
of the GaP growth and giving an infiltration of 100% or
filling fraction of 0.224. Two higher energy peaks were also
observed at 1.55 and 1.76 eV and closely matched high-
order gaps in the calculated band structure. The calculation
also indicates a FPBG at a normalized frequency of 0.88, as
determined by the eighth and ninth bands at the W point.5
The excellent agreement between the measured and calcu-
lated band structures implies that a full photonic band gap of
ϳ3% may exist at 726 nm.
In summary, we have reported the fabrication of high
filling fraction GaP inverse opals using atomic layer deposi-
tion within silica opal templates, confirming fill-hold-purge
ALD as ideally suited to uniform film growth within highly
porous structures. X-ray diffraction data confirmed the high
quality of the GaP. Using this technique, inverse GaP opals
were fabricated with filling fractions as high as 0.224, corre-
sponding to 100% of the conformal film growth maximum
for an opal lattice. The optical properties of the GaP inverse
opal were well described by photonic band calculations, in-
dicating the potential to form a full photonic band gap in the
visible. The results demonstrate a reliable, controllable tech-
nique for fabrication of inverse opals with a high index,
transparent, semiconductor backbone, which avoids anoma-
lous film growth. As GaP is a leading optoelectronic III-V
semiconductor, it should be possible to exploit its properties
to fabricate active photonic crystal devices with even larger
full photonic band gaps using optimized structures.
2E. Yablonovitch, Phys. Rev. Lett. 58, 2059 ͑1987͒.
3S. John, Phys. Rev. Lett. 58, 2486 ͑1987͒.
4S. John and K. Busch, J. Lightwave Technol. 17, 1931 ͑1999͒.
5D. Gaillot, T. Yamashita, and C. J. Summers, Phys. Rev. B 72, 205109
͑2005͒.
6A. Blanco, E. Chomski, S. Grabtchak, M. Ibisate, S. John, S. W. Leonard,
C. López, F. Meseguer, H. Míguez, J. P. Mondla, G. A. Ozin, O. Toader,
and H. M. van Driel, Nature ͑London͒ 405, 437 ͑2000͒.
7E. Palacios-Lidon, A. Blanco, M. Ibisate, F. Meseguer, C. Lopez, and J.
Sanchez-Dehesa, Appl. Phys. Lett. 81, 4925 ͑2002͒.
8H. Míguez, E. Chomski, F. García-Santamaría, M. Ibisate, S. John, C.
López, F. Meseguer, J. P. Mondia, G. A. Ozin, O. Toader, and H. M. v.
Driel, Adv. Mater. ͑Weinheim, Ger.͒ 13, 1634 ͑2001͒.
9Y.-C. Lee, T.-J. Kuo, C.-J. Hsu, Y.-W. Su, and C.-C. Chen, Langmuir 18,
9942 ͑2002͒.
10H. M. Yates, M. E. Pemble, E. Palacios-Lidón, F. García-Santamaría, I.
Rodriguez, F. Meseguer, and C. López, Adv. Funct. Mater. 15, 411
͑2005͒.
11I. M. Povey, D. Whitehead, K. Thomas, M. E.Pemble, M. Bardosova, and
J. Renard, Appl. Phys. Lett. 89, 104103 ͑2006͒.
12M. Müller, R. Zentel, T. Maka, S. G. Romanov, and C. M. Sotomayor
Torres, Adv. Mater. ͑Weinheim, Ger.͒ 12, 1499 ͑2000͒.
13B. H. Juárez, M. Ibisate, J. M. Palacios, and C. López, Adv. Mater. ͑Wein-
heim, Ger.͒ 15, 319 ͑2003͒.
14E. Palacios-Lidón, H. M. Yates, M. E. Pemble, and C. López, Appl. Phys.
B: Lasers Opt. 81, 205 ͑2005͒.
15A. Rugge, J. S. Park, R. G. Gordon, and S. H. Tolbert, J. Phys. Chem. B
109, 3764 ͑2005͒.
16S. G. Romanov, R. M. D. L. Rue, H. M. Yates, and M. E. Pemble, J. Phys.:
Condens. Matter 12, 339 ͑2000͒.
17H. M. Yates, W. R. Flavell, M. E. Pemble, N. P. Johnson, S. G. Romanov,
and C. M. Sotomayor-Torres, J. Cryst. Growth 170, 611 ͑1997͒.
18A. Rugge, J. S. Becker, R. G. Gordon, and S. H. Tolbert, Nano Lett. 3,
1293 ͑2003͒.
19J. S. King, C. W. Neff, C. J. Summers, W. Park, S. Blomquist, E. Forsythe,
and D. Morton, Appl. Phys. Lett. 83, 2566 ͑2003͒.
20J. S. King, E. Graugnard, and C. J. Summers, Adv. Mater. ͑Weinheim,
Ger.͒ 17, 1010 ͑2005͒.
21J. S. King, E. Graugnard, and C. J. Summers, Appl. Phys. Lett. 88,
081109 ͑2006͒.
22Y. Sakuma, K. Kodama, and M. Ozeki, Appl. Phys. Lett. 56, 827 ͑1990͒.
23M. Ozeki, K. Kodama, Y. Sakuma, N. Ohtsuka, and T. Takanohashi, J.
Vac. Sci. Technol. B 8, 741 ͑1990͒.
24A. Usui, Proc. IEEE 80, 1641 ͑1992͒.
25T. Soga, T. Jimbo, and M. Umeno, J. Cryst. Growth 146, 554 ͑1995͒.
26H. Isshiki, Y. Aoyagi, and T. Sugano, Appl. Surf. Sci. 112, 122 ͑1997͒.
27N. Y. Li, H. K. Dong, W. S. Wong, and C. W. Tu, J. Cryst. Growth 164,
112 ͑1996͒.
28J. S. King, C. W. Neff, S. Blomquist, E. Forsythe, D. Morton, and C. J.
Summers, Phys. Status Solidi B 241, 763 ͑2004͒.
The authors thank D. P. Gaillot and J. S. King for valu-
able assistance with this work. This work was supported by
the U.S. Army Research Office under MURI Contract No.
DAAD19-01-1-0603.
29S. G. Johnson and J. D. Joannopoulos, Opt. Express 8, 173 ͑2001͒.
30D. E. Aspnes and A. A. Studna, Phys. Rev. B 27, 985 ͑1983͒. Thin film
reflectance measurements were used to independently confirm the index of
the ALD grown GaP film.
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:
128.42.202.150 On: Sun, 23 Nov 2014 09:59:46