Rapid synthesis of high-quality InP nanocrystals

Article abstract of DOI:10.1021/ja057676k

A rapid new method for preparation of monodisperse InP-nanocrystals was developed. A highly reactive indium precursor and tris(trimethylsilyl)phosphine (TMS)₃P was reacted within a weakly coordinating solvent in the presence of a supporting protic agent. The yielded InP-nanocrystals had a very narrow size distribution without any size selection process. The precursor and ligand effects were considered as critical factors in control of nucleation and crystal growth process. Different ligands were introduced to study the reaction mechanism. The new method not only yielded the best InP-nanocrystals so far, but also includes the potential for preparation within a continuous flow reactor, because the utilized ester is liquid at room temperature. Copyright

Full text of DOI:10.1021/ja057676k

Published on Web 01/07/2006
Rapid Synthesis of High-Quality InP Nanocrystals
Shu Xu, Sandeep Kumar, and Thomas Nann*
Freiburg Materials Research Centre, UniVersity of Freiburg, Stefan-Meier-Strasse 21, 79104 Freiburg, Germany
Received November 11, 2005; E-mail: thomas.nann@fmf.uni-freiburg.de
Colloidal III-V semiconductor nanocrystals (NCs) have attracted
intense interest within the past 20 years, owing to their less ionic
lattice, larger exciton diameters, and reduced toxicity compared to
IIB-VI compounds. Nevertheless, the study and application of III-V
semiconductor nanocrystals are limited by the difficulty in their
synthesis. Because the molecular bonds in III-V semiconductors
are more covalent, it is very difficult to obtain a controllable
nucleation burst.1,2 Within the III-V group, the synthesis of InP
nanocrystals is the most extensively studied, but until now InP
nanocrystals synthesized by current chemical methods did not
achieve the same quality as that of most IIB-VI semiconductor
nanocrystals.3-10
Typical synthesis approaches for III-V semicondutor NCs in a
coordinating solvent are adaptations of the method for the IIB-VI
group. However, the common coordinating solvents, ligands for
IIB-VI system, and the similar precursors do not work well for the
synthesis of III-V semicondutor NCs. Both nucleation and crystal
growth processes in these approaches needed long reaction times,
all together 2-7 days, to yield crystalline particles. In addition, a
size-selective process was always necessary.3-6 Battaglia and Peng
developed an efficient method to synthesize InP nanocrystals in a
noncoordinating solvent. This method provided a fast, controllable
reaction and generated much higher-quality InP nanocrystals.7
Nevertheless, it is still a challenge to develop a rapid reaction
approach in coordinating solvents.
To the best of our knowledge, in noncoordinating solvent routes,
the indium precursor was always heated with fatty acids, and the
best-quality InP nanocrystals were obtained when the ratio of acid
to indium was 3:1.7,8 High-quality InP nanocrystals were also
produced when indium carboxylate was used as precursor without
addition of other ligands.9 Hence, we consider that the carboxylate
groups actually acted as an in situ high-selectivity coordination
ligand and led to the controllable nucleation and growth processes.
On the other hand, in strong coordinating solvents, the unselective
coordination of the solvent to indium and phosphorus results in
slower and more continuous nucleation. In such a case, it is very
difficult to separate the nucleation and crystal growth processes.
We supposed that a nucleation process similar to that in noncoor-
dinating solvents could also be obtained in coordinating solvents
when the coordinating effect of the solvent is much weaker than
that of the introduced strong ligands. We found that high-boiling
point esters could be effective weakly coordinating solvents for
control of the nucleation process. Further, we tried to accelerate
the nucleation burst by proper choice of reagents and reaction
conditions, which otherwise possibly limits the use of strong
coordinating ligands in producing high-quality InP nanocrystals.
Therefore, we devoted our efforts to synthesize InP nanocrystals
in weak coordinating solvents. High-quality InP nanocrystals were
synthesized via this approach. The as-prepared InP nanocrystals
had distinguishable absorption peaks and much narrower size
distribution than InP NCs prepared by currently available synthesis
methods.
Methyl myristate and dibutyl sebacate were chosen as weak
coordinating solvents, because they are nontoxic and relatively
inexpensive. Several long-carbon-chain fatty acids and amines were
also examined as ligands. Tris(trimethylsilyl)phosphine (TMS)3P
was used as the phosphorus precursor. All chemicals were purchased
from Aldrich and have a purity of at least 95%.
For a typical experiment, 1 mL of ester and 0.4-0.45 mmol of
ligands were mixed in a three-neck flask and heated to 260 °C
under N2 flow; 0.1 mmol of (CH3)3In and 0.05 mmol of (TMS)3P
were dissolved in 0.5 mL of ester to form a clear solution which
was rapidly injected into the hot reaction flask. After injection, the
solution was cooled and maintained for growth at 200 °C. The
products were precipitated and washed with acetone and methanol
and redispersed in chloroform. The two esters generated similar
quality products.
We started our experiments from repeating the same reaction in
the ester as previously published in octadecene (ODE). Indium
carboxylate, or different ratios of acid to indium acetate, and
(TMS)3P were used as precursors. The ratio range of indium to
phosphorus for these experiments was varied between 1:2 and 4:1.
As our prospective result, the produced InP nanocrystals had less
distinguishable absorption peaks and broader emission peaks. The
full width at half-maximum (fwhm) of the photoluminescence (PL)
spectra of the best product was about 85 nm. The lower mono-
dispersity of produced nanocrystals indicates that esters still caused
a relatively slower nucleation process than noncoordinating solvents,
although they were better than strongly coordinating solvents such
as TOPO.
When trimethylindium was used instead of indium carboxylate
as the indium precursor, the quality of the InP nanocrystals was
dramatically improved with an enhanced PL emission and optimized
size distribution. In our experiments, trimethylindium easily dis-
solved in ester at room temperature and generated a soluble In-
ester complex. It is believed that this indium complex had relatively
higher activity than indium carboxylate. Therefore, it could react
with (TMS)3P more easily at high temperatures and lead to a higher
nucleation rate. We also found the best ligands are fatty acids in
our experiments. The temperature influence on our reactions was
investigated (Figure 1). The best temperature ranges for injection
and subsequent crystal growth processes were found to be 240-
280 and 180-210 °C, respectively. Hence, in our further studies,
we fixed the injection and growth temperatures to be 260 and 200
°C, respectively, as the standard conditions. The ratio of indium to
phosphorus was kept at 2:1 to maintain an indium-rich reaction.
When the ratio of fatty acid to indium was 3:1, the fwhm of our
produced InP nanoparticles PL spectra was reduced to about 60
nm.
Furthermore, some protic reagents were introduced into the
reactions to get a more rapid nucleation burst. It is believed that
protic reagents, such as MeOH and RNH2, hydrolyze (TMS)3P to
accelerate its reaction with indium compounds to produce InP
nanoparticles.4,10 In our study, we tried some amines as protic
9
1054
J. AM. CHEM. SOC. 2006, 128, 1054-1055
10.1021/ja057676k CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Suggested Reaction Scheme
noncoordinating solvents had very obvious differences. In our
experiments, we tried three fatty acids with different carbon chain
lengths, stearic acid (SA), myristic acid (MA), and oleic acid (OA),
as ligands to study the temporal evolution of the UV-vis spectra
of the InP nanocrystals. As shown in Figure 2, myristic acid had
the best hydrocarbon chain length for the balance of nucleation
and crystal growth. However, the three series of spectra did not
show as dramatic a difference in the size distribution as observed
for synthesis in the ODE system. The results indicate that the
influence of the ligand chain length on the reaction has been
weakened by some othersmore prominentsfactors, including the
reactive indium precursor and the effect of supporting protic
reagents.
Figure 1. PL and UV-vis absorption spectra (left) of the InP nanocrystal
obtained at different injection temperatures after growth for 2 min. TEM
image (right top) and XRD pattern (right bottom) of a typical sample. Stearic
acid (SA)/indium ratio was 4.2:1 for all the reactions. The mean diameter
of the particles in the TEM micrograph is 2.5 nm. Each spectrum was
recorded after mixing three batches of samples prepared under identical
conditions.
Since the hydrolyzation process in our reaction took place at
high temperature, resulting in a rapid nucleation burst followed by
a slow crystal growth process, it presents a very quick one-pot
approach to prepare high-quality InP nanocrystals. It is anticipated,
that the protons react with the phosphorus precursor to form in
situ highly reactive phosphine which acts as an active phosphorus
precursor for generation of high-quality nanocrystals. The reaction
pathway through In-ester complex intermediate as shown in
Scheme 1 is still under investigation and will be reported in a
forthcoming publication.
In summary, a novel and rapid method was developed for the
synthesis of high-quality InP nanocrystals via an effective and
simple reaction. Fatty acid esters provide a weak coordination
condition similar to that by noncoordinating solvents. A highly
reactive indium precursor and protic reagents were used to
accelerate the nucleation process. The as-prepared nanocrystals had
the most narrow size distribution compared to other NCs generated
by current methods. Further investigation would extend this method
for the synthesis of other III-V and IIB-VI semiconductor nano-
crystals.
Figure 2. (A) UV-vis spectra of InP nanocrystals obtained with different
ratios of ligands, dioctylamine (DOA), trioctylamine (TOA), and stearic
acid (SA), after growth for 5 min. (B, C, D) Temporal evolution of the
UV-vis spectra of InP nanocrystals grown with different ligands. Ligand/
In ratio was 4.2:1 for all reactions.
reagents, such as dioctylamine and hexadecylamine. The excessive
fatty acid and oleic acid can also serve as protic reagents themselves.
We compared the UV-vis absorption spectra of the nanocrystals
obtained in the presence of some different protic and nonprotic
reagents. The results are shown in Figure 2 A. When the ratio of
stearic acid to indium was 3:1, the absorption spectrum of the as-
prepared NCs had an indistinguishable peak. The absorption peak
became clear gradually with an increasing content of stearic acid.
The best distinguishable absorption peaks were obtained when the
ratio was between 4:1 and 4.5:1. Alternatively, when the excessive
acid was replaced by the same molar amount of protic reagents,
such as dioctylamine (DOA), the as-prepared nanocrystals had a
similar quality. However, the replacement of the excessive stearic
acid by nonprotic reagents, such as trioctylamine (TOA), resulted
in lower-quality nanocrystals.
In the presence of the respective supporting protic reagents, we
successfully reduced the fwhm of the PL spectra of resulting InP
nanocrystals to about 48 nm. Transmission electron microscope
(TEM) images of our samples exhibited that the InP nanocrystals
were dot-shaped and very well dispersed. The powder X-ray
diffraction (XRD) patterns showed the clear InP nanocrystal peaks
of zinc blende structure. Figure 1 depicts one TEM micrograph
and XRD pattern of a typical InP sample.
Acknowledgment. We acknowledge the help of Dr. Ralf
Thomann for measuring the TEM micrographs and of Dr. Martin
Ade for the XRD-measurements. S.X. thanks the state of Baden-
Wu¨rttemberg for financial support within the project 24-7532.22-
1-14/1.
References
(1) Masala, O.; Seshadri, R, Annu. ReV. Mater. Res. 2004, 34, 41-81.
(2) Heath, J. R.; Shiang, J. J. Chem. Soc. ReV. 1998, 27, 65-71.
(3) Micic, O. 1.; Curtis, C. J.; Jones, K. M.; Sprague, J. R.; Nozik, A. J. J.
Phys. Chem. 1994, 98, 4966-4969
(4) Micic, O. I.; Ahrenkiel, S. P.; Nozik, A. J. Appl. Phys. Lett. 2001, 78,
4022-4024.
(5) Guzelian, A. A.; Katari, J. E. B.; Kadavanich, A. V.; Banin, U.; Hamad,
K.; Juban, E.; Alivisatos, A. P.; Wolters, R. H.; Arnold, C. C.; Heath, J.
R. J. Phys. Chem. 1996, 100, 7212-7219.
(6) Nedeljkovic, J. M.; Micic, O. I.; Ahrenkiel, S. P.; Miedaner, A.; Nozik,
A. J. J. Am. Chem. Soc. 2004, 126, 2632-2639.
(7) Battaglia, D.; Peng, X. G. Nano Lett. 2002, 2, 1027-1030.
(8) Gerbec, J. A.; Magana, D.; Washington, Aaron.; Strouse, G. F. J. Am.
Chem. Soc. 2005, 127, 15791-15800
The influence of the hydrocarbon chain length of the fatty acid
was studied as well. In noncoordinating solvents, the chain length
of the fatty acid had the most distinct effect for the balance of
nucleation and growth. The UV-vis spectra of nanocrystals
obtained with fatty acids of different chain lengths in these
(9) Lucey, D. W.; MacRae, D. J.; Furis, M.; Sahoo, Y.; Cartwright, A. N.;
Prasad, P. N. Chem. Mater. 2005, 17, 3754-3762.
(10) Douglas, T.; Theopold, K. H. Inorg. Chem. 1991, 30, 594-596.
JA057676K
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1055