Polarization and temperature dependence of photoluminescence from zincblende and wurtzite InP nanowires

Article abstract of DOI:10.1063/1.2828034

We use polarization-resolved and temperature-dependent photoluminescence of single zincblende (ZB) (cubic) and wurtzite (WZ) (hexagonal) InP nanowires to probe differences in selection rules and bandgaps between these two semiconductor nanostructures. The WZ nanowires exhibit a bandgap 80 meV higher in energy than the ZB nanowires. The temperature dependence of the PL is similar but not identical for the WZ and ZB nanowires. We find that ZB nanowires exhibit strong polarization parallel to the nanowire axis, while the WZ nanowires exhibit polarized emission perpendicular to the nanowire axis. This behavior is interpreted in terms of the different selection rules for WZ and ZB crystal structures.

Full text of DOI:10.1063/1.2828034

Polarization and temperature dependence of photoluminescence from zincblende and
wurtzite InP nanowires
A. Mishra, L. V. Titova, T. B. Hoang, H. E. Jackson, L. M. Smith, J. M. Yarrison-Rice, Y. Kim, H. J. Joyce, Q. Gao
,
H. H. Tan, and C. Jagadish
Citation: Applied Physics Letters 91, 263104 (2007); doi: 10.1063/1.2828034
View online: http://dx.doi.org/10.1063/1.2828034
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/91/26?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Far field emission profile of pure wurtzite InP nanowires
Appl. Phys. Lett. 105, 191113 (2014); 10.1063/1.4901437
Photoluminescence study of Zn-doped wurtzite InP core-shell nanowiresa)
Appl. Phys. Lett. 102, 032105 (2013); 10.1063/1.4788925
Diameter dependent optical emission properties of InAs nanowires grown on Si
Appl. Phys. Lett. 101, 053103 (2012); 10.1063/1.4739001
Type II heterostructures formed by zinc-blende inclusions in InP and GaAs wurtzite nanowires
Appl. Phys. Lett. 97, 041910 (2010); 10.1063/1.3473779
Probing valence band structure in wurtzite InP nanowires using excitation spectroscopy
Appl. Phys. Lett. 97, 023106 (2010); 10.1063/1.3463036
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: 138.26.31.3
On: Mon, 22 Dec 2014 15:12:18

APPLIED PHYSICS LETTERS 91, 263104 ͑2007͒
Polarization and temperature dependence of photoluminescence
from zincblende and wurtzite InP nanowires
a͒
A. Mishra, L. V. Titova, T. B. Hoang, H. E. Jackson, and L. M. Smith
Department of Physics, University of Cincinnati, Cincinnati, Ohio 45221-0011, USA
J. M. Yarrison-Rice
Department of Physics, Miami University, Oxford, Ohio 45056, USA
Y. Kim
Department of Physics, College of Natural Science, Dong-A University, Hadan 840, Sahagu,
Busan 604-714, Korea
H. J. Joyce, Q. Gao, H. H. Tan, and C. Jagadish
Department of Electronic Materials Engineering, Research School of Physical Sciences and Engineering,
Australian National University, Canberra, Australian Capital Territory 0200, Australia
͑
Received 4 October 2007; accepted 4 December 2007; published online 26 December 2007͒
We use polarization-resolved and temperature-dependent photoluminescence of single zincblende
ZB͒ ͑cubic͒ and wurtzite ͑WZ͒ ͑hexagonal͒ InP nanowires to probe differences in selection rules
͑
and bandgaps between these two semiconductor nanostructures. The WZ nanowires exhibit a
bandgap 80 meV higher in energy than the ZB nanowires. The temperature dependence of the PL
is similar but not identical for the WZ and ZB nanowires. We find that ZB nanowires exhibit strong
polarization parallel to the nanowire axis, while the WZ nanowires exhibit polarized emission
perpendicular to the nanowire axis. This behavior is interpreted in terms of the different selection
rules for WZ and ZB crystal structures. © 2007 American Institute of Physics.
͓
DOI: 10.1063/1.2828034͔
The large surface-to-volume ratio of semiconductor
nanowires has a profound effect on the energetics of nano-
wire growth and can result in new structures not observed in
the bulk. For instance, only recently has it been understood
that one-dimensional growth of InP nanowires ͑NWs͒ often
occurs preferentially in the hexagonal wurtzite phase, in con-
trast to the cubic zincblende structure ubiquitous in bulk
axis͒. For single NW measurements, NWs from the growth
substrate were sonicated into Methanol solution and dis-
persed onto a patterned silicon substrate. The NWs were sur-
veyed using a standard optical microscope to record the po-
sitions of ϳ20 NWs with respect to alignment marks. This
substrate was then mounted on the cold finger of a
continuous-flow liquid Helium cryostat. A 0.5 mW, 780 nm
CW Ti-sapphire laser was defocused to Ͼ10 ␮m to fully
illuminate a single NW. Photoluminescence ͑PL͒ from a
single NW was collected using a 50ϫ /0.5NA long working
distance microscope objective and a 250ϫ image of the NW
was projected onto the entrance slit of the spectrometer. A
matched set of Glan-Thompson linear polarizers and
Babinet-Soleil compensators were used both to control the
polarization of the laser excitation and to analyze the polar-
ization of the emitted PL.
1
–3
InP. Indeed, one can use the growth temperature to select
1
between one phase and another. Such a dramatic structural
change should also be reflected by changes in the energy
gap, the electron and hole masses, and recombination selec-
1
–3
tion rules. In this letter, we report polarized photolumines-
cence measurements as a function of temperature from
zincblende and wurtzite single InP nanowires.
InP NWs were grown using the gold catalyst-assisted
metal-organic vapor-liquid-solid growth mechanism. Un-
doped InP substrates oriented along the ͑111͒B direction
were functionalized by immersion in poly-L-lysine ͑PLL͒ so-
lution, rinsed in de-ionized water, and treated with a colloidal
solution of monodisperse 30 nm Au nanoparticles. NWs
were grown by horizontal flow metalorganic chemical vapor
deposition ͑MOCVD͒ at a pressure of 100 mbar, with a V/III
ratio of 110. The growth temperature was set at either 42 °C
In Fig. 1, we show PL spectra from three ZB ͑labeled A,
B, and C͒ and three WZ ͑labeled 1, 2, and 3͒ InP nanowires
͑
zincblende͒ or 48 °C ͑wurtzite͒. Structural analysis by
Transmission Electron Microscopy ͑Ref. 4͒ shows that the
4
2 °C grown NW resulted in a zincblende ͑ZB͒ crystal struc-
4
¯
¯¯
ture with the growth axis along the ͓111͔ direction. The ZB
NWs were 2–7 ␮m long with an average diameter of 80 nm.
InP NWs grown at 48 °C, in contrast, resulted in ϳ1 ␮m
long tapered NWs which exhibit hexagonal wurtzite ͑WZ͒
crystal structure with the growth axis along the ͓0001͔ ͑c
FIG. 1. ͑Color online͒ Typical PL spectra from several single ZB and WZ
InP nanowires at 15 K.
a͒Electronic mail: smithl@physics.uc.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: 138.26.31.3
003-6951/2007/91͑26͒/263104/3/$23.00 91, 263104-1 © 2007 American Institute of Physics
On: Mon, 22 Dec 2014 15:12:18
0

263104-2
Mishra et al.
Appl. Phys. Lett. 91, 263104 ͑2007͒
FIG. 2. ͑Color online͒ Integrated intensity of PL emission at 15 K from
single ZB NW A and WZ NW 1 as a function of linear polarization analyzer
angle.
FIG. 3. ͑Color online͒ PL spectra from single ZB NW A and WZ NW 1 as
a function of temperature. For comparison, PL spectra from a thick ͑100͒-
oriented epilayer is shown at 15 and 300 K.
measured at 15 K. The ZB InP ͑100͒ epilayer exhibits a
strong and narrow emission line centered at the free exciton
5
recombination energy of 1.418 eV. The three ZB NWs all
previous room temperature measurements of PL in WZ CdSe
NWs, which showed the polarization to be determined by
display a single broad ͓full width at half maximum ͑FWHM͒
ϳ20 meV͔ PL peak which is approximately centered at
1
3
dielectric contrast only. Quantum confinement plays no
role since the exciton Bohr radius ͑9 nm͒ in ZB InP is sig-
nificantly smaller then the diameter of the NWs ͑80 nm͒.
In contrast, a hexagonal wurtzite crystal ͑C6v point
group͒ exhibits a completely different symmetry. The s-like
1.418 eV. As described in Titova et al., while the NW PL
from the same growth substrate varies substantially from
4
wire to wire, the most intense emission is observed for NWs
whose PL is centered at 1.418 eV and these are the ones
displayed in Fig. 1. These ZB NWs also display lifetimes of
conduction band has ⌫ symmetry, while the p-like hole
5
00 ps–1 ns, lifetimes which are comparable to that ob-
7
4
bands split into three separate hole bands due to a combina-
tion of the spin-orbit interaction and crystal field
served in the ZB epilayer ͑1.7 ns͒. The emission width and
energy position is strongly affected by state filling and so it is
important to measure the NW at the lowest possible excita-
1
1,12
splitting.
The lowest energy hole states ͑highest lying
4
energy bands͒ having a symmetry of ⌫ ͑the so-called A
tion intensities.
9
band͒, while the two higher energy hole bands ͑B and C
In contrast, the WZ InP NWs ͑labeled 1, 2, and 3͒
display a somewhat narrower emission line ͑FWHM
ϳ17 meV͒ at 1.49 eV, 80 meV higher than for ZB InP
NWs. Existing published work on the bandgap in WZ InP
nanowires has been inconsistent. The 80 meV higher energy
11,12
bands͒ have ⌫ symmetry.
The lowest energy exciton
7
state should be between a ⌫ electron and a ⌫ hole, which
7
9
means that the ⌫ →⌫ recombination is dipole allowed only
7
9
for E perpendicular to the c axis of the nanowire, and for-
11,12
1
bidden for E parallel to the c axis.
Thus the lowest en-
WZ NW PL in Fig. 1 is consistent with Mattila et al. In
ergy exciton transition in WZ InP should be strongly polar-
ized perpendicular to the wurtzite c axis and, thus,
perpendicular to the nanowire axis.
contrast, Tragardh et al. used photocurrent spectroscopy of
In As1−xP ͑x ranging from 0% to 50%͒ segments inside an
x
InAs nanowire, to estimate that the WZ InP bandgap ranged
2
,3
Polarization analysis of the emission from wurtzite NW
from 90 to 230 meV above the ZB InP gap. More recently,
Reitzenstein et al. observed multiple emission lines from
single WZ InP nanowires in a lower energy range of
1
shows that the PL is polarized ͑49%͒ perpendicular to the
NW axis, consistent with growth of the NW along the ͓0001͔
axis. The degree of polarization in the WZ nanowire is
weaker than ZB nanowire because the dipole selection rules
of the wurtzite crystal are modulated by the dielectric con-
trast inherent for all semiconductor NWs.
6
ϳ1.44–1.47 eV. Such sharp lower energy emission lines
might result from exciton localization at defects as was seen
7
recently in CdS NWs.
The polarization of the PL emission from ZB and WZ
nanowires also display distinctive differences. To measure
this, circularly polarized 780 nm laser excitation was used to
excite all the linearly polarized electronic states with equal
probability, while the emitted PL was analyzed for linear
polarization. In Fig. 2, the circles show the variation of the
PL integrated intensity from ZB NW A as a function of the
In Fig. 3, we show the variation of the PL emission from
ZB NW A and WZ NW 1 as a function of temperature from
1
5 to 280 K and compare this with PL emission from the
epilayer at 15 and 300 K. Emission from both ZB and WZ
NWs shifts monotonically to lower energy with increasing
temperature, and the linewidths increase markedly. The in-
tensity of the PL becomes weaker, but never completely
quenches. In Fig. 4͑a͒, we plot emission energy versus tem-
perature for both the ZB and WZ NWs, along with a ZB
2 ␮m thick epilayer. The solid lines are fits to the modified
2
polarization angle, with the solid line a fit to cos ␪. The ZB
InP NW emission is strongly polarized PL ͑82%͒ parallel to
8
–10
the nanowire axis,
due to the strong dielectric contrast
between the approximately cylindrical NWs and the sur-
rounding air. In bulk InP, the recombination in ZB InP is
between the s-like conduction band electrons which has ⌫6
symmetry and the p-like doubly degenerate light and heavy
4
3
14
Varshni equation E=E0−␣T /͑T +␤͒. The emission en-
ergy is comparatively flat at low temperatures, but decreases
almost linearly at temperatures above 50 K. Interestingly, the
WZ NWs appear to follow most closely the temperature de-
pendence of the epilayer, though shifted up by 80 meV in
hole bands with ⌫ symmetry, and so the emitted light is
8
11,12
completely unpolarized.
ZB NW emission is polarized
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: 138.26.31.3
because of the dielectric contrast. This was also the case in
energy. In contrast, the ZB NW at high temperatures exhibits
On: Mon, 22 Dec 2014 15:12:18

263104-3
Mishra et al.
Appl. Phys. Lett. 91, 263104 ͑2007͒
substrate temperature. Striking changes in emission polariza-
tion and energy are observed which reflect the change in
symmetry and dipole selection rules. WZ NWs emit 80 meV
higher energy than do ZB NWs. The variation of the inten-
sity and emission energy with temperature for ZB and WZ
NW are similar except at high temperatures and apart from
the bandgap difference. Importantly, the emission for both
ZB and WZ nanowires is quite intense without passivation,
indicating notable low sensitivity to the surface of these
InP NWs.
A.M., L.V.T., T.B.H., H.E.J., L.M.S., and J.M.Y.-R. ac-
knowledge support from the Institute for Nanoscale Science
and Technology of the University of Cincinnati and the
National Science Foundation through Grant Nos. EEC/NUE
0
532495 and ECCS 0701703. The Australian authors ac-
knowledge support from the Australian Research Council.
Y.K. acknowledges support by the Korean Science and
Engineering Foundation ͑KOSEF͒ through Grant No. F01-
2
007-000-10087-0.
1
M. Mattila, T. Hakkarainen, M. Mulot, and H. Lipsanen, Nanotechnology
7, 1580 ͑2006͒.
FIG. 4. ͑Color online͒ ͑a͒ Peak position of the emitted PL as a function of
temperature for WZ NWs 1 ͑᭡͒ and 2 ͑ϩ͒ and ZB NWs A ͑b͒ and B ͑ϩ͒.
For comparison, the peak position of the emitted PL from a thick ͑100͒-
oriented InP epilayer is shown ͑᭺͒. ͑b͒ The integrated intensity I of WZ NW
1
2
J. Taagardh, A. I. Persson, J. B. Wagner, D. Hessman, and L. Samuelson,
J. Appl. Phys. 101, 123701 ͑2007͒.
3
H. Pettersson, J. Tragardh, A. I. Persson, L. Landin, D. Hessman, and L.
Samuelson, Nano Lett. 6, 4 ͑2006͒.
1
is plotted as I /I−1 as a function of 1/T, where I is related to the
0
0
integrated intensity at zero temperature. The exciton binding energy of
.1 meV is marked by arrow.
4
L. V. Titova, T. B. Hoang, J. M. Yarrison-Rice, H. E. Jackson, Y. Kim,
H. J. Joyce, Q. Gao, H. H. Tan, C. Jagadish, X. Zhang, J. Zou, and L. M.
Smith, Nano Lett. 7, 3383 ͑2007͒.
5
a significantly different slope compared to the bulk InP epil-
ayer. The reason for this difference is presently unclear.
In Fig. 4͑b͒, we plot I /I−1, where I reflects the low T
5
D. C. Reynolds, C. W. Litton, R. J. Almassy, S. B. Nam, P. J. Dean, and
R. C. Clarke, Phys. Rev. B 13, 2507 ͑1976͒.
6
S. Reitzenstein, S. Munch, C. Hofmann, A. Forchel, S. Crankshaw, L. C.
Chuang, M. Moewe, and C. Chang-Hasnain, Appl. Phys. Lett. 91, 091103
͑2007͒.
0
0
PL intensity, and I is the integrated intensity of the PL emis-
1
5
sion from the WZ InP nanowire. The fact that the PL can
be easily measured at room temperature indicates a low non-
radiative surface recombination rate. While the PL becomes
somewhat less intense in the temperature range between 15
and 50 K, the PL emission intensity decreases rapidly above
7
L. V. Titova, T. B. Hoang, H. E. Jackson, L. M. Smith, J. M.
Yarrison-Rice, J. L. Lensch, and L. J. Lauhon, Appl. Phys. Lett. 89,
0
53119 ͑2006͒.
J. F. Wang, M. S. Gudiksen, X. F. Duan, Y. Cui, and C. M. Lieber, Science
93, 1455 ͑2001͒.
8
2
9
0
H. E. Ruda and A. Shik, Phys. Rev. B 72, 115308 ͑2005͒.
T. B. Hoang, L. V. Titova, J. M. Yarrison-Rice, H. E. Jackson, A. O.
Govorov, Y. Kim, H. J. Joyce, H. H. Tan, C. Jagadish, and L. M. Smith,
Nano Lett. 7, 588 ͑2007͒.
50 K. The data display an abrupt change in slope at 56 K,
1
near the temperature which separates exciton dominated
emission at lower temperatures and electron-hole plasma
emission at higher temperatures. Temperature-dependent PL
data for single ZB InP nanowires are essentially equivalent,
which may indicate that the electron and hole masses are not
dramatically different between the ZB and WZ nanowires
To summarize, we have shown that the growth of ZB or
WZ InP nanowires can be controlled by changing the growth
11
J. L. Birman, Phys. Rev. 114, 1490 ͑1959͒.
J. L. Birman, Phys. Rev. Lett. 2, 157 ͑1959͒.
C. X. Shan, Z. Liu, and S. K. Hark, Phys. Rev. B 74, 153402 ͑2006͒.
M. Cardona, T. A. Meyer, and M. L. W. Thewalt, Phys. Rev. Lett. 92,
1
A. Chiari, M. Colocci, F. Fermi, Y. H. Li, R. Querzoli, A. Vinattieri, and
W. H. Zhuang, Phys. Status Solidi B 147, 421 ͑1988͒.
1
1
1
2
3
4
96403 ͑2004͒.
1
5
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: 138.26.31.3
On: Mon, 22 Dec 2014 15:12:18