Surfactant-imposed interference in the optical characterization of GaP nanocrystals

Article abstract of DOI:10.1021/jp030294l

We report on the complications in optical characterization of GaP nanoparticles grown by colloidal chemistry. It is demonstrated that upon exposure to relatively mild heating conditions, the surfactants utilized to arrest growth and/or agglomerization lead to thermal decomposition products, which exhibit optical properties similar to those observed for nanocrystals with broad size distributions. The origin of this photoluminescence is still unknown; however, photoluminescence measurements of surfactants of different purities indicate that the impurities may be responsible for the emission. The results of this study demonstrate the potential complications associated with studying the optical properties of GaP nanocrystals prepared in the presence of commonly utilized surfactants. They suggest that more reliable assignment of the spectral features of nanocrystals can be afforded by a more detailed understanding of the optical properties associated with the surfactants and or their thermal decomposition products.

Full text of DOI:10.1021/jp030294l

11622
J. Phys. Chem. B 2003, 107, 11622-11625
Surfactant-Imposed Interference in the Optical Characterization of GaP Nanocrystals
Madalina Furis, Yudhisthira Sahoo, David J. MacRae, Felicia S. Manciu,
Alexander N. Cartwright,* and Paras N. Prasad*,†
Departments of Physics, Electrical Engineering, and Chemistry, Institute for Lasers, Photonics and
Biophtonics, UniVersity at Buffalo, State UniVersity of New York, New York 14260
ReceiVed: February 28, 2003; In Final Form: June 17, 2003
We report on the complications in optical characterization of GaP nanoparticles grown by colloidal chemistry.
It is demonstrated that upon exposure to relatively mild heating conditions, the surfactants utilized to arrest
growth and/or agglomerization lead to thermal decomposition products, which exhibit optical properties similar
to those observed for nanocrystals with broad size distributions. The origin of this photoluminescence is still
unknown; however, photoluminescence measurements of surfactants of different purities indicate that the
impurities may be responsible for the emission. The results of this study demonstrate the potential complications
associated with studying the optical properties of GaP nanocrystals prepared in the presence of commonly
utilized surfactants. They suggest that more reliable assignment of the spectral features of nanocrystals can
be afforded by a more detailed understanding of the optical properties associated with the surfactants and or
their thermal decomposition products.
Colloidal growth is an established method for the preparation
of semiconductor nanocrystals and typically involves heating
the precursors in the presence of a surfactant and/or coordinating
solvent.1-6 A variety of both II-VI and III-V semiconductor
nanocrystals have been successfully prepared by this method.
Although trioctylphosphine oxide (TOPO) and trioctylphosphine
(TOP) are most widely utilized, other surfactants such as
hexadecylamine (HDA), trioctylamine (TOA), etc., have found
use as well. In addition to controlling the growth of the
nanocrystals, the surfactants passivate their surfaces, thereby
limiting surface traps and the agglomerization of the nanocrys-
tals. However, to obtain highly crystalline nanocrystals, it is
necessary to conduct the reactions at relatively high tempera-
tures, 100-200 °C2,4 for the II-VI (such as CdSe) ionic
semiconductor nanocrystals, and in excess of 270 °C for several
days for III-V covalent semiconductors such as InP.3,5,6
At the temperature at which the semiconductor nanocrystals
are prepared, thermal decomposition of the surfactant results
in the formation of unidentified products in addition to the
nanocrystals. In most cases, isolation of the nanocrystals and
the removal of excess surfactant is achieved by washing with
methanol or acetone. To our knowledge, there has been no report
in the literature where the surfactant utilized in the preparation
of the nanocrystals has been completely removed from the
nanocrystal surface. In fact, the maximum surfactant exchange
has been reported to be between 80 and 90%.9,10 Therefore, it
is important to understand the nature of the optical properties
associated with the thermal decomposition products associated
with these surfactants in order to separate their contribution to
the absorption and emission spectra of the solutions containing
nanocrystals. We have successfully prepared GaP nanoparticles
using methods previously described,1-6 intended to be incor-
porated into hybrid inorganic/organic devices such as OLEDs.
Toluene solutions of GaP nanocrystals capped with TOPO
exhibit a broad photoluminescence (PL) spectrum upon UV
excitation (350 nm) with the maximum of the emission intensity
in the blue region as expected for GaP. However, our photo-
luminescence (PL) and photoluminescence excitation (PLE), as
well as time-resolved photoluminescence (TRPL) measurements
of toluene solutions containing just the heated TOPO, exhibited
a broad emission spectrum extended over the entire visible
region with a maximum located at a wavelength shorter than
500 nm. The PLE spectra and carrier recombination dynamics
are very similar to those expected from nanocrystals with a broad
size distribution. Under these circumstances, the identification
of the feature associated with carrier recombination in the
semiconductor nanocrystals becomes problematic, especially
where the emission of the nanocrystal is expected at wavelengths
shorter than 500 nm.
GaP nanocrystals were prepared by reacting GaCl3 with
P(SiMe3)3 in TOPO (90%) following a method similar to the
one presented in refs 1-6. Continuous wave (CW) PL and PLE
spectra were acquired in the frontal geometry using a Jobin
Yvon Fluorolog-3 spectrofluorometer equipped with a 400 W
Xe lamp as an excitation source and a Hamamatsu R928
photomultiplier (PMT) tube as a detector. All spectra were
corrected for the PMT response and/or the lamp emission
spectrum. The excitation source used for measuring the pho-
toluminescence intensity decay was the frequency-doubled 400
nm, 200 fs pulses output from a Coherent Rega 9000 regenera-
tive amplifier with a 250 kHz repetition rate. The solutions
containing the heated surfactants were diluted with toluene, and
their emission was collected from solutions in quartz cuvettes
using conventional optics in the frontal geometry. The photo-
luminescence signal was spectrally and temporally resolved by
a Chromex 250IS monochromator equipped with a Hamamatsu
C4334 streakcamera having a typical jitter of 50 ps.
The GaP nanocrystals were characterized by transmission
electron microscopy (TEM), energy dispersive spectroscopy
(EDS), and electron diffraction. The TEM image shown in
Figure 1a indicates that the nanocrystals are spherically shaped
with an average diameter of 14 nm and a large size distribution
* Corresponding authors.
† E-mail: pnprasad@acsu.buffalo.edu.
10.1021/jp030294l CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/27/2003

Optical Characterization of GaP Nanocrystals
J. Phys. Chem. B, Vol. 107, No. 42, 2003 11623
Figure 1. (a) Transmission electron microscopy image (b), energy
dispersive spectrum, and (c) electron diffraction pattern of GaP
nanocrystals grown by colloidal chemistry. The particles are character-
ized by an average diameter of 14 nm and a broad size distribution
((50%). (d) Comparison between the photoluminescence spectra of
heated 90% TOPO (dashed curve) and a solution containing GaP
nanocrystals (solid curve).
Figure 2. (a) Photoluminescence spectra from three different surfac-
tants commonly used in nanocrystals colloidal chemistry growth, and
(b) PLE spectra for the 90% TOPO heated at 150 °C taken at different
wavelengths across the PL feature. The peak of the PLE spectra shifts
as a function of wavelength, indicating the presence of several emitting
components in the heated TOPO solution.
((50%). The EDS spectrum (Figure 1b) exhibits an excess of
phosphorus, suggesting the presence of either excess surfactant
or other phosphorus-containing compounds. Electron diffraction
performed on the same grid exhibits a ring pattern as shown in
Figure 1c. Fourier transform far-infrared spectroscopic studies
on the same sample confirmed the crystallinity of the GaP
nanoparticles through the presence of transverse optical bulk-
like TO(Γ) phonon modes.8
Our preliminary results encouraged us to investigate the
optical properties of the nanocrystals. The photoluminescence
spectrum of a toluene solution containing the GaP nanocrystals
is presented in Figure 1d. The emission is broad and extends
over the entire visible spectrum with a maximum intensity in
the blue region as expected for GaP nanocrystals. A brief
statement made by Micic et al. alluded to the fact that the heated
surfactant (90% TOPO) might be emitting in the same wave-
length range as the GaP nanocrystals they fabricated.6 However,
no detailed studies of the optical properties of these heated
surfactants have been reported in the literature.
Since the same surfactant was used in the growth of the
nanoparticles discussed here, we measured, for completeness,
the PL spectrum of TOPO (90%) heated under the same
conditions as the growth process and diluted in toluene. The
spectrum is represented by the dashed line in Figure 1d. Clearly
the two spectra in this figure are essentially the same and suggest
the entire emission observed in the solution containing the GaP
nanoparticles is due to the surfactant as a result of being heated.
The observation of emission from heated surfactants points that
caution has to be exercised in assignment of the observed
emission to GaP nanocrystals.
Recently, a similar spectrum, measured for a mixture of
TOPO and HDA, has been reported in connection with the
optical studies of CdSe nanoparticles.7 However, unlike III-V
semiconductors, photoluminescence spectra from II-VI semi-
conductor nanocrystals typically exhibit sharp and strong
features that are distinguishable from the broad emission
observed for the surfactants.
Typical photoluminescence spectra obtained from three
different solutions containing heated TOPO (90%) and HDA
as well as unheated fresh TOA are presented in Figure 2a. The
excitation wavelength was 350 nm for all samples. The emission
is broad and extends over the entire visible region and is similar
in character to the emission from nanocrystals characterized by
a broad size distribution.11-14 Therefore, upon heating at high
temperatures, TOPO, HDA, and/or other potential decomposi-
tion products exhibit photoluminescence properties. In some
cases (for example TOA), the surfactant exhibits an emission
without any heating.
We have further investigated the optical properties of these
heated surfactants by recording the PLE spectra at several
wavelengths across the PL feature. The results of the measure-
ments are presented in Figure 2b. The excitation wavelength
corresponding to the maximum emission intensity changes at
different emission wavelengths, suggesting that there are several
emitting species present in the solution. This type of behavior
is also exhibited by semiconductor nanocrystals15-17 in the
quantum confinement size regime, where the maximum of the
PLE spectrum depends on the nanoparticle size. This behavior
is often utilized to identify the emission attributed to the
nanocrystals. Therefore, the heated surfactants not only exhibit
emission in the expected wavelength range for GaP nanocrystals,
but they also exhibit PLE spectral features similar to those
associated with nanocrystals.
The CW PL and PLE studies were found to be insufficient
to establish the nature of the photoluminescence or to isolate

11624 J. Phys. Chem. B, Vol. 107, No. 42, 2003
Furis et al.
Figure 4. PL spectra for 90% TOPO heated to 150 °C, taken after
different heating times. The emission intensity increases by a factor of
10 as the heating time increases from 0.5 h to 24 h. After 24 h, the
quantum efficiency is of the same order as the one reported for
nanocrystals.
to the thermal decomposition of the impurities present in the
surfactant rather then the surfactant itself. The impurities present
in commercial grade TOPO (90%) have been previously
reported to be dioctylmonophosphonic acid, mono-octyldiphos-
phonic acid, dioctylphosphonic acid, and dioctylphosphinate.21
However, none of these reported species incorporates conjugated
π-bonds, that could, upon heating, give rise to species expected
to exhibit luminescence properties. In an attempt to identify the
decomposition products that might be responsible for the
Figure 3. (a) Time-resolved PL spectra of 90% heated TOPO at
different delay times under 400 nm excitation, and (b) the PL intensity
decay as a function of time for the wavelength corresponding to the
peak PL intensity. A multicomponent fit to this decay shows that the
emission lifetimes are very similar to carrier radiative lifetimes for
semiconductor nanocrystals
1
the spectral components related to nanocrystals. Therefore, we
performed time-resolved photoluminescence measurements on
the same 90% TOPO solution using a 400 nm femtosecond
pulsed excitation. The measured time-resolved spectra at dif-
ferent delay times are presented in Figure 3a. The PL peak is
red-shifted when compared to the CW spectrum obtained with
350 nm excitation. Such a shift is expected considering the PLE
results, which show that different excitation wavelengths are
resonantly pumping different species, which contribute to the
broad PL feature. A plot of the PL intensity decay as a function
of time at the wavelength corresponding to the emission intensity
peak in Figure 2a is shown in Figure 3b. The decay has a
nonexponential character and can be attributed to the existence
of several different emitting species. The emission lifetimes
resulting from a multicomponent fit to this decay are in the
nanosecond range and are similar to the radiative recombination
lifetimes in direct band gap semiconductors. Again, this
spectroscopic technique has provided results that are similar to
those expected for nanoparticles. The nonexponential character
of the decay has been encountered in the time-resolved
studies18-20 of carrier recombination in nanocrystals and is a
result of the carrier lifetime dependence on the nanoparticle size.
To monitor the optical properties of heated TOPO as a
function of heating time, a 0.5 mL aliquot was drawn every
hour and dissolved in toluene. The spectra in Figure 4 show
the resulting emission at various intervals of heating at 150 °C.
The results indicate a significant emission after only 0.5 h of
heating. Most importantly, the emission intensity increases as
a function of heating time. The quantum efficiency was
measured under 366 nm excitation using a solution of di-phenyl
anthracene in toluene as a reference. After 24 h of heating, the
quantum efficiency reaches 13.8%, which is similar to the
efficiency reported for many quantum dots. It is important to
note that most reported preparations of nanocrystals require
heating times in excess of 24 h.
observed properties, analytical tools such as H NMR, FTIR,
TLC, HPLC, and mass spectra were employed to study unheated
and heated (150 °C × 24 h) samples of 90% TOPO. In all these
measurements, the results on both types of samples showed no
apparent difference, indicating that any thermal decomposition
product may be present in too small of a concentration to be
1
easily identifiable by these techniques. For example, H NMR
spectra of both the heated and the unheated solutions in CDCl3
show no apparent difference in the spectra and exhibit no
resonances in the 3-8 ppm range. Thus, there is no evidence
of the presence of any compounds exhibiting aromaticity.
Admittedly, identifying the species responsible for such a
substantially efficient photoluminescence behavior is important
in order to minimize the interference due to surfactants on the
optical properties of the nanocrystals. On the basis of the
increasing intensity of PL with heating time, we believe that
the concentration of the responsible species increases with the
duration of heating.
In conclusion, surfactants widely used in the colloidal
synthesis of semiconductor nanocrystals exhibit photolumines-
cence properties related to thermal decomposition. The emission
spectra are very broad and extend over the entire visible region.
Both the PLE and the TRPL spectra and dynamics are very
similar to those expected for the GaP semiconductor nanocrys-
tals. Therefore, common optical techniques utilized in the
characterization of semiconductor nanocrystals cannot unam-
biguously distinguish between the PL resulting from the GaP
nanocrystals or the surfactant thermal decomposition products.
Furthermore, time-resolved photoluminescence studies could
unambiguously isolate the component due to nanocrystals only
if the emission is associated with surface or forbidden states,
characterized by very long (tens or hundreds of nanoseconds)
decay times. However, a short PL decay time does not
necessarily indicate the absence of PL from nanocrystals. Careful
and detailed PL, PLE, and TRPL studies of surfactants and/or
solvents heated through a process that reproduces the growth
conditions may help in isolating the surfactant component in
the PL spectra. Detailed studies of the photoluminescence
properties of surfactants are especially important for GaP
nanocrystals, which are expected to emit at wavelengths shorter
In an effort to determine the origins of the photoluminescence,
we compared the emission intensities from two TOPO solutions
of different purities: 90% and 99%. The emission intensity was
found to decrease by a factor of 100 in the 99% TOPO,
suggesting the photoluminescence properties might be related

Optical Characterization of GaP Nanocrystals
J. Phys. Chem. B, Vol. 107, No. 42, 2003 11625
(7) de Mello Donega, C.; Hickey, S. G.; Wuister, S. F.; Vannaekel-
bergh, D.; Meijerink, A. J. Phys. Chem. B 2003, 107, 489.
(8) Manciu, F. S.; Sahoo, Y.; MacRae, D. J.; Furis, M.; McCombe, B.
D.; Prasad, P. N. Appl. Phys. Lett. 2003, 82, 4059.
(9) Kuno, M.; Lee, J. K.; Dabbousi, B. O.; Mikulec, B. O.; Bawendi,
M. G. J. Chem. Phys. 1997, 106, 9869.
than 500 nm. A detailed understanding of the optical properties
associated with the solvents utilized in the nanocrystal prepara-
tion should permit more careful assignment of the spectral
features associated with nanocrystals.
(10) Bowen Katari, J. E.; Colvin, V. L.; Alivisatos, P. L. J. Phys. Chem.
1994, 98, 4109.
(11) Mic´ic´, O. I.; Ahrenkiel, S. P.; Bertram, D.; Nozik, A. J. Appl. Phys.
Lett. 1999, 75, 478.
(12) Mic´ic´, O. I.; Ahrenkiel, S. P.; Nozik, A. J. Appl. Phys. Lett. 2001,
78, 4022.
(13) Giessen, H.; Fluegel, B.; Mohs, G.; Peyghambarian, N.; Sprague,
J. R.; Mic´ic´, O. I.; Nozik, A. J. Appl. Phys. Lett. 1996, 68, 304.
(14) Wilcoxon, J. P.; Provencio, P. P.; Samara, G. A. Phys. ReV. B 2001,
64, 035417-1.
(15) Norris, J.; Efros, Al. L.; Rosen, M.; Bawendi, M. G. Phys. ReV. B
1996, 53, 16347.
(16) Bertram, D.; Mic´ic´, O. I.; Nozik, A. J. Phys. ReV. B 1998, 57,
R4265.
(17) Banin, U.; Lee, C. J.; Guzelian, A. A.; Kadavanich, A. V.;
Alivisatos, A. P.; Jaskolski, W.; Bryant, G. W.; Efros, Al. L.; Rosen, M. J.
Chem. Phys. 1998, 109, 2306.
Acknowledgment. This work was supported by a Defense
University Research Initiative on Nanotechnology Grant
F496200110358 through the Directorate of Chemistry and Life
Sciences of the Air Force Office of Scientific Research. The
authors thank Dr. B. D. McCombe for valuable discussions on
the FTIR spectroscopy of GaP nanoparticles.
References and Notes
(1) Alivisatos, P. A.; Xiaogang, P.; Liberato, M. U.S. Patent 6,306,-
736, 2001.
(2) Bawendi, M.; Stott, N. E. U.S. Patent Appl. 20020071952, 2002.
(3) Guzelian, A. A.; Katari, J. E. B.; Kadavanich, A. V.; Banin, U.;
Hamad, K.; Juban, E.; Alivisatos, A. P. J. Phys. Chem. 1996, 100, 7212.
(4) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc.
1993, 115, 5, 8706.
(5) Mic´ic´, O. I.; Sprague, J. R.; Curtis, C. J.; Jones, K. M.; Machol, J.
L.; Nozik, A. J.; Giessen, H.; Flueggel, B.; Mohs, G.; Peyghambarian, N.
J. Phys. Chem. 1995, 99, 7754.
(6) Mic´ic´, O. I.; Curtis, C. J.; Jones, K. M.; Sprague, J. R.; Nozik, A.
J. J. Phys. Chem. 1994, 98, 4966.
(18) Zhao, X.; Schoenfeld, O.; Komuro, S.; Aoyagi, Y.; Sugano, T. Phys.
ReV. B 1994, 50, 18654.
(19) Zacharias, M.; Fauchet, P. M. Appl. Phys. Lett. 1997, 71, 380.
(20) Mic´ic´, O. I.; Cheong, H. M.; Fu, H.; Zunger, A.; Sprague, J. R.;
Mascarenhas, A.; Nozik, A. J. J. Phys. Chem. B 1997, 101, 4904.
(21) Kolosky, M.; Vialle, J.; Cotel, T. J. Chromatogr. 1984, 299, 445.