Role of acid in precursor conversion during InP quantum dot synthesis

Article abstract of DOI:10.1021/cm401289j

We have studied the speciation of P(SiMe₃)₃ during the synthesis of colloidal InP quantum dots in the presence of proton sources. Using ³¹P NMR spectroscopy, we show H_3-nP(SiMe ₃)_n formation on exposure of P(SiMe₃) ₃ to a variety of protic reagents including water, methanol, and carboxylic acid, corroborating observations of P(SiMe₃)₃ speciation during the hot injection synthesis of InP QDs. Quantitative UV-vis comparisons between InP growth from P(SiMe₃)₃ and HP(SiMe₃)₂ show unambiguously that when total H ⁺-content is accounted for, particle size, size dispersity, and concentration are indistinguishable for these two reagents. The dual role of myristic acid in P-Si bond cleavage and as a source of the myristate anion, an essential component of the quantum dot surface, is interrogated using tetrabutylammonium myristate, confirming that it is the protons that are responsible for increased quantum dot polydispersity. Together these data support the existence of a competing acid-catalyzed pathway in the conversion of P(SiMe₃)₃ to InP and demonstrate its impact. By preventing a constant solute supply and affecting the concentration of quantum dot surfactant over the course of the reaction, the existence of competing precursor conversion pathways is detrimental to formation of monodisperse colloids, explaining much of the irreproducibility in InP quantum dot syntheses to date.

Full text of DOI:10.1021/cm401289j

Article
pubs.acs.org/cm
Role of Acid in Precursor Conversion During InP Quantum Dot
Synthesis
Dylan C. Gary and Brandi M. Cossairt*
Department of Chemistry, University of Washington, Box 351700, Bagley Hall, Seattle, Washington 98195-1700, United States
S
* Supporting Information
ABSTRACT: We have studied the speciation of P(SiMe3)3
during the synthesis of colloidal InP quantum dots in the
presence of proton sources. Using 31P NMR spectroscopy, we
show H3‑nP(SiMe3)n formation on exposure of P(SiMe3)3 to a
variety of protic reagents including water, methanol, and
carboxylic acid, corroborating observations of P(SiMe3)3
speciation during the hot injection synthesis of InP QDs.
Quantitative UV−vis comparisons between InP growth from
P(SiMe3)3 and HP(SiMe3)2 show unambiguously that when
total H+-content is accounted for, particle size, size dispersity, and concentration are indistinguishable for these two reagents. The
dual role of myristic acid in P−Si bond cleavage and as a source of the myristate anion, an essential component of the quantum
dot surface, is interrogated using tetrabutylammonium myristate, confirming that it is the protons that are responsible for
increased quantum dot polydispersity. Together these data support the existence of a competing acid-catalyzed pathway in the
conversion of P(SiMe3)3 to InP and demonstrate its impact. By preventing a constant solute supply and affecting the
concentration of quantum dot surfactant over the course of the reaction, the existence of competing precursor conversion
pathways is detrimental to formation of monodisperse colloids, explaining much of the irreproducibility in InP quantum dot
syntheses to date.
KEYWORDS: precursor conversion, quantum dot, phosphide, III−V, mechanism
highly successful for other systems, such as CdSe, the
reactivities of currently employed phosphorus precursors for
InP, such as tris(trimethylsilyl)phoshphine (P(SiMe3)3),15−19
usually result in full consumption of the phosphorus precursor
during nucleation. As the crystallization of InP needs a source
of P3− or P0 (to react with In3+ or In0, respectively), there are a
limited number of feasible phosphorus precursors. Notable
alternatives to P(SiMe3)3 include phosphine gas (PH3),20−22
white phosphorus (P4),23,24 and more recently, tris-
(trimethylgermyl)phosphine, P(GeMe3)3.25
To improve upon current InP syntheses and rationally design
better systems, the roles of each of the precursors need to be
considered with respect to the formation of the soluble
semiconductor species. A previous investigation into the
formation of colloidal InP QDs from P(SiMe3)3 and indium
myristate (In(O2C(CH2)12CH3)3) has led to the proposal of an
amine-inhibited charge-dispersion SN2 transition state for the
initial formation of an In−P bond.2 Here, we report the
observation of rapid protonolysis of P(SiMe3)3 at 20 °C in the
presence of myristic acid, which is typically present in excess in
current hot-injection strategies using P(SiMe3)3.15−19 On the
basis of this observation, we propose a competing, acid
catalyzed, pathway to InP QDs from P(SiMe3)3 and In(O2C-
INTRODUCTION
■
The direct bandgap of III−V semiconductors allow these
materials to absorb and emit light with much greater efficiency
than indirect bandgap semiconductors such as silicon.1,2 III−V
materials also have higher carrier mobilities than other direct
bandgap semiconductors such as cadmium selenide and lead
sulfide.2−4 For these reasons, III−V semiconductors with
bandgaps in the visible region, such as gallium arsenide
(GaAs) and indium phosphide (InP), are being developed for
high-performance transistors, optoelectronics, and photovol-
taics.2,4−6
Quantum dot materials in the III−V family, in particular, are
seen as attractive for deployment in solid-state lighting and
biological imaging applications. This is primarily due to their
solution processability and superior emissive properties over
organic fluorophores, including broadband absorption with
narrow, symmetric emission profiles, as well as resistance to
photobleaching and high quantum yields.7−9 InP quantum dots,
specifically, offer a low-toxicity alternative to Cd-based materials
already in use in these light-emitting applications.10,11
Unfortunately, these materials are as yet unable to compete
with the CdSe because of the broad emission profiles that result
from the poor particle size dispersity afforded by current
synthetic methods. The current strategy to synthesize these
materials employs a hot-injection technique to temporally
separate the nucleation event from growth to achieve size
focusing.12−14 Although hot-injection methods have been
Received: March 5, 2013
Revised: May 22, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(CH2)12CH3)3, as depicted in Scheme 1. This pathway depicts
only a single protonolysis event, forming the secondary
Scheme 1. Precursor Conversion Reactions Leading to
Colloidal Indium Phosphide Nanocrystals, Including the
a
,
Suggested Acid-Catalyzed Pathway35
Figure 1. 31P{1H} NMR spectrum obtained upon mixing P(SiMe3)3
1
with HO2CCF3 at 20 °C. H-coupled spectra are shown in the insets.
a
R = (CH2)12CH3.
temperatures showed similar results.35 These results suggest
that even at room temperature, P(SiMe3)3 is readily converted
to H3‑nP(SiMe3)n, a fact with important mechanistic implica-
tions for the controlled synthesis of phosphide materials using
this P3− source.
phosphine, for simplicity. The full mechanistic picture including
all potential protonolysis products, H3‑nP(SiMe3)n, may be
found in the Supporting Information. This acid-dependent
pathway bears close analogy with the mechanism proposed by
Theopold and Douglas for the formation of small InP clusters
in solution using the molecular precursors RR′In−P(SiMe3)2 in
the presence of methanol (R = Cp*, R′ = Cl; R, R′ =
Me3SiCH2; R = Me3SiCH2, R′ = Cl; and R, R′ = Me3CCH2).26
It is known that P(SiMe3)3 can be used to prepare
HP(SiMe3)2 when treated with 1 equiv of methanol or 0.5
equiv of water.27,28 The corresponding primary phosphine and
PH3 can also be prepared analogously.22,27 Typical conditions
for InP nanocrystal preparation involve the in situ generation of
In(O2C(CH2)12CH3)3 from mixing 4 equiv of myristic acid and
In(O2CCH3)3 in octadecene at 100 °C under vacuum for 1
h.15,16,29−33 These conditions are assumed to fully convert to
indium acetate to indium myristate and acetic acid and to
remove any water or acetic acid from the reaction mixture. The
use of excess myristic acid to fully convert the indium acetate
means that a protic species remains in the reaction mixture
even if all of the acetic acid and water are removed. Combined
with the observations that carboxylic acid concentration greatly
affects quantum dot growth,15,34 this information strongly
suggests that under typical nanocrystal growth conditions,
P(SiMe3)3 may be converted to H3‑nP(SiMe3)n, leading to
competing nucleation pathways.
Second, we have analyzed P(SiMe3)3 speciation during InP
QD synthesis using optimized procedures adapted from Peng
and co-workers.15 Specifically, 0.4 mmol of In(O2CCH3)3 was
combined with 3.6 equiv of myristic acid in 4 g of dried and
distilled octadecene and was heated under vacuum (20 mtorr)
at 100 °C for 16 h. Freshly distilled P(SiMe3)3 (0.2 to 0.8
mmol) was dissolved in 2 g of dried and distilled octadecene in
an inert atmosphere glovebox and loaded into the injection
syringe. Spectra of aliquots taken within 5 s of phosphine
injection at 20 °C and at 315 °C in the presence of a 2-fold
excess of the phosphine precursor show phosphine speciation
comparable with that observed in the control experiments
discussed above.35 Prior to conversion of the phosphine precursor
into nanocrystals, P(SiMe3)3 protonolysis has occurred. Such
speciation was not observed in the earlier mechanistic work.16
The difference between the two experiments is the presence of
primary amine in the reactions. Primary amines were added to
slow the reaction kinetics by binding to the indium in
competition with phosphine. One possible alternative role of
primary amines is to deprotonate carboxylic acid (or water),
generating alkylammonium carboxylate (or hydroxide). Control
experiments run with P(SiMe3)3 in the presence of
octylammonium myristate indeed show significant variation in
the phosphine protonolysis profile, with a shift toward the
starting tertiary phosphine with analogous reaction times.35
This is likely a thermodynamic effect resulting from the higher
pKa of octyalmmonium myristate relative to myristic acid. Thus,
the inhibitory effect of amine may be due to modulation of the
acid-catalyzed pathway.
The above data suggest that the secondary phosphine should
be a suitable precursor for InP nanocrystal growth. To test this
hypothesis, we have prepared InP nanocrystals directly from
HP(SiMe3)2. Growth profiles of the syntheses using both the
tertiary and the secondary phosphines have been compared
using UV−vis absorption spectroscopy. The lowest energy
electronic transitions (LEET) were fit to Gaussian curves to
allow quantitative comparison of the peak widths (indicative of
polydispersity), amplitudes (indicative of particle concentra-
tion), and energies (indicative of particle size). Absorbance was
also monitored at 310 nm, where the extinction coefficient is
expected to be relatively independent of particle size.25,35,36 In
RESULTS AND DISCUSSION
■
We have taken a three-pronged approach to investigating the
role of acid on P(SiMe3)3 speciation and thus InP-QD
formation. First, control experiments have been performed to
assess the ability of various proton sources including water,
methanol, trifluoroacetic acid, myristic acid, and HCl, to
generate H3−nP(SiMe3)n. These experiments were carried out in
rigorously dried hydrocarbon solvent at 20 and 240 °C and
analyzed by proton-coupled 31P NMR spectroscopy. Repre-
sentative data revealing the speciation of P(SiMe3)3 in the
presence of 1 equiv of trifluoroacetic acid at 20 °C are shown in
Figure 1.
At 20 °C, 1 equiv of methanol was found to slowly and
selectively give rise to HP(SiMe3)2, whereas 0.5 equiv of water,
1 equiv of carboxylic acid, and 1 equiv of HCl all showed
nonselective formation of a mixture of HP(SiMe3)2, H2P-
(SiMe3), and PH3.35 Experiments performed at elevated
B
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 2. Data extracted from Gaussian fitting of the UV−vis spectra of aliquots taken at different times during InP nanocrystal growth using
P(SiMe3)3 and HP(SiMe3)2 precursors. Data points represent averages from three separate experiments with error bars at a 95% confidence interval.
(A) Energy of LEET versus time (indicative of particle size); (B) Full width at half-maximum of LEET versus time (indicative of particle size
dispersity); (C) Absorbance at the peak maximum of the LEET versus time (indicative of concentration); (D) Representative UV−vis traces for 8
min aliquots for P(SiMe3)3 and HP(SiMe3)2 reactions.
all experiments the precursors were prepared as described
above, the phosphine (0.2 mmol) injection temperature was
315 °C, and the growth temperature was maintained at 285 °C.
Two types of experiments were run for each phosphineone
set using 3 equiv of myristic acid (stoichiometric amount
needed to generate indium myristate and found empirically to
be optimal for HP(SiMe3)2) and one set using 3.6 equiv of
myristic acid (empirically determined to be optimal for
P(SiMe3)3). These data are summarized in Figure 2 and 3.
The data in Figure 2 reveal that the InP QD sizes, size
distributions,35 and concentrations are very similar in reactions
run with the two phosphines under optimal conditions. This
result suggests that in experiments with P(SiMe3)3, the added
0.6 equiv (1.2 equiv relative to P) of acid may serve to generate
HP(SiMe3)2. This result is corroborated by the data shown in
Figure 3, which compares the fwhm for particles grown with
P(SiMe3)3 and HP(SiMe3)2, both with 3 and 3.6 equiv of
myristic acid. These data show that, on average, QD
polydispersity is optimal when P(SiMe3)3 is used in
conjunction with an extra 0.6 equiv of acid, as observed
above, but when HP(SiMe3)2 is used, optimal QD polydisper-
sity is obtained with stoichiometric acid (3 equiv). The
observation that reactions using the tertiary phosphine require
an excess of acid to give the narrowest size distribution
demonstrates the significance of the acid-catalyzed pathway
shown in Scheme 1.
One convoluting effect of HP(SiMe3)2 being involved on the
path to InP QD formation is its ability to generate 1 equiv of
myristic acid on reaction with indium myristate. Myristic acid
not only leads to P−Si bond cleavage, as discussed extensively
above, but it also serves as a passivating agent for the QD
Figure 3. Data extracted from Gaussian fitting of UV−vis time aliquots
of InP growth from P(SiMe3)3 (A) and HP(SiMe3)2 (B) with 3 and
3.6 equiv of myristic acid. Data points represent averages from three
separate experiments.37.
C
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
surface in both its protonated form and as the myristate
anion.38−41 Studies have shown that the concentration of
surfactant has a dramatic effect on the growth profiles of
nanomaterials, influencing size,42,43 shape,44 and particle
dispersity.15 To isolate the roles of myristic acid as an agent
in P−Si bond rupture and as a QD ligand, we have conducted a
series of experiments in which In(O2C(CH2)12CH3)3 is
preformed and 0.6 equiv of tetrabutylammonium myristate,
[N((CH2)3CH3)4][O2C(CH2)12CH3], an aprotic myristate
anion source, is added.35 Notably, when this experiment is
performed using P(SiMe3)3 as the phosphorus source, QD
polydispersity shows a slight improvement over that obtained
with 3.6 equiv of myristic acid (0.01 eV lower average fwhm),35
suggesting that it is the myristate anion that is necessary for
maintaining particle dispersity. When the analogous experiment
is run with HP(SiMe3)2 as the phosphorus source, QD
dispersity suffers and is again analogous to experiments run
with 3.6 equiv of myristic acid. The reaction of HP(SiMe3)2
with indium myristate results in myristic acid formation during
the course of the experiment suggesting that the protons are
indeed responsible for the higher polydispersity as a result of
the acid-catalyzed pathway.
One caveat to these experiments is the increased thermal
sensitivity of tetrabutylammonium myristate-containing pre-
cursor solutions over myristic acid-containing precursor
solutions. Solutions containing tetrabutylammonium myristate
undergo a visible thermal decomposition near 315 °C,33 the
optimal injection temperature for our InP QDs. To prevent
decomposition, injections can be performed between 305 and
310 °C, which results in larger temperature drops upon
injection of the phosphine precursor and typically much poorer
QD size dispersity. If InP quantum dot syntheses using
P(SiMe3)3 are directly compared at 310 °C injection temper-
atures using 0.6 equiv of tetrabutylammonium myristate versus
0.6 equiv of myristic acid,35 significant improvements (0.05 eV
average lower fwhm, Figure 4) are observed using tetrabuty-
lammonium myristate, further reinforcing the detrimental
effects of protons in these syntheses.
((CH2)3CH3)4][O2C(CH2)12CH3], and P(SiMe3)3 under ideal
injection conditions (315 °C) show that in the absence of the
acid-catalyzed speciation pathway, precursor conversion,
nucleation, and crystal growth are still too rapid and occur
simultaneously (Figure 5). This observation underlines the
Figure 5. Representative UV−vis data obtained from timed aliquots
during InP QD synthesis from In(O2C(CH2)12CH3)3, 0.6 equiv of
[N((CH2)3CH3)4][O2C(CH2)12CH3], and P(SiMe3)3. Injection tem-
perature 315 °C, growth 285 °C.
need for further development of new molecular precursors to
allow for effective separation of nucleation and growth in the
synthesis of InP quantum dots. This is an ongoing area of
investigation in our laboratory.
CONCLUSION
■
These results have important general implications for the
synthesis of colloidal III−V semiconductor nanocrystals. The
solution-phase synthesis of colloidal III−V semiconductors has
greatly lagged behind those of their II−VI and IV−VI
counterparts in terms of process reproducibility and particle
quality.14,16,25 For II−VI quantum dots, the ability to effectively
separate crystal nucleation and growth is enabled by precursor
conversion being single-mechanism and rate limiting. The
results presented here show that the presence of protic reagents
in the synthesis of phosphide materials, namely InP, opens up a
competing acid-catalyzed reaction pathway, preventing molec-
ular precursors from providing a constant solute supply and
hence inhibiting reproducibility and QD quality. Isolating the
dual roles of myristic acid as a proton source and a source of
myristate anion, a necessary component of the QD surface, has
allowed us to identify criteria for ideal InP size dispersity using
P(SiMe3)3, namely, elimination of protons using salt-based
precursors stable at elevated injection temperatures.
Analysis of representative growth profiles obtained from
experiments using In(O2C(CH2)12CH3)3, 0.6 equiv of [N-
While we have specifically focused on the fact that phosphine
protonolysis is detrimental to InP QD formation, studies
performed by Chaudret et al. have shown in a related study that
the high temperature injection of P(SiMe3)3 coincides with
ketonic decarboxylation of the carboxylic acids used in
traditional syntheses, producing water as a byproduct.33 The
water so generated can oxidize the quantum surface, generating
an InPOx shell around the core, inhibiting growth. Our work
demonstrates that water present in the reaction mixture has the
additional disadvantage of causing phosphine speciation. Taken
together, these studies demonstrate the need for improved
phosphorus precursor development and the exploration of
Figure 4. Representative UV−vis data obtained from timed aliquots
during InP QD synthesis from In(O2C(CH2)12CH3)3, 0.6 equiv of
[N((CH2)3CH3)4][O2C(CH2)12CH3], and P(SiMe3)3 (blue traces)
and In(O2C(CH2)12CH3)3, 0.6 equiv of HO2C(CH2)12CH3, and
P(SiMe3)3 (red traces). All injections were performed at 310 °C
(nonideal) to eliminate possible precursor decomposition effects. R =
(CH2)12CH3.
D
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
removed to allow the solution to come to room temperature while
maintaining stirring, returning the deep blue color.
surface-passivating ligands that do not contain acidic protons to
bypass such pathways.
Under a counter flow of nitrogen, the remaining septum was
exchanged for the solid addition funnel. The solid addition funnel was
rotated to deliver the red phosphorus to the sodium/potassium alloy
suspension. The apparatus was once again suspended in the silicone oil
bath and heated to a gentle reflux with vigorous stirring for a 24 h
period; during which a deep black suspension was produced.
After 24 h, the solution was allowed to come to room temperature.
The reflux condenser was replaced with a 100 mL dropping funnel
under a counter flow of nitrogen. Twenty milliliters of 1,2-
dimethoxyethane was cannula transferred from a Schlenk flask to the
dropping funnel. Two 9 mL aliquots of chlorotrimethylsilane (15.4 g,
141.8 mmol, 3.2 equiv) were drawn from an air-free container and
injected into the dropping funnel.
The chlorotrimethylsilane solution was added dropwise to the
sodium/potassium phosphide suspension at a rate of about 1 drop a
second. After the addition of the chlorotrimethylsilane solution, the
dropping funnel was exchanged for a Teflon stopper against a counter
flow of nitrogen. Electrical tape was wrapped several times around the
three-neck flask and each of the Teflon stoppers as well as the air-free
flow control gas adapter. The three-neck flask was carefully brought
into the glovebox with fifteen rapid cycles of partial vacuum.
The product solution was filtered through a pad of Celite to remove
the salts formed during the reaction. The filter pad was washed with
pentane, and the colorless to pale yellow solution was concentrated to
a total volume of 30 mL. The solution was transferred to a 100 mL 14/
20 neck Schlenk flask equipped with a small football-shaped magnetic
stir bar and was removed from the glovebox for distillation using a
vacuum-jacketed short-path distillation apparatus.
A first fraction containing solvent and any unreacted chlorotrime-
thylsilane was removed under full vacuum with heating at 30 °C. The
second distillation fraction, which contained the desired product in
pure form, was collected under full vacuum (20 mtorr) at
approximately 45 °C. The collected P(SiMe3)3 was stored in the
glovebox in the dark. Mass of distilled product: 8.94 g (35.7 mmol,
82% yield).
Synthesis of HP(SiMe3)2. HP(SiMe3)2 was prepared in pure form
following a procedure adapted from Buhro and co-workers.28
Tetrahydrofuran was substituted for diethyl ether.
HP(SiMe3)2 was prepared in situ according to the following
procedures: In a typical synthesis, 5.00 g of 1-octadecene was weighed
out into a 250 mL 24/40 neck Schlenk flask in a nitrogen filled
glovebox to which was added 145 μL (0.50 mmol) of P(SiMe3)3. The
flask was outfitted with a 24/40 septum.
The Schlenk flask was brought out of the glovebox and placed
under nitrogen after cycling the Schlenk line with nitrogen/vacuum
three times. A scintillation vial was filled with methanol and stoppered
with a rubber septum that was secured with electrical tape to ensure an
airtight seal. The methanol was sparged with nitrogen for 10 min, and
then 20 μL (0.50 mmol) of the sparged methanol was delivered to the
Schenk flask with vigorous stirring. The solution was allowed to stir for
approximately 10 h.
This discovery parallels important advances in our under-
standing of II−VI nanocrystal syntheses with respect to the
roles of impurities such as phosphonic acids,45−47 and the roles
of secondary phosphines in trioctylphosphine-chalcogenide-
based syntheses of II−VI and IV−VI QDs.48−50 All of these
studies underline the fact that understanding the molecular
reactions involved in precursor conversion to quantum dot
nuclei is critical to the development of synthetic methods that
adequately control colloidal QD synthesis. Although the
chemistries of III−V nanocrystal syntheses are not as mature
as those of their II−VI and IV−VI counterparts, the results
presented here illustrate that a concrete understanding of the
roles of protons in InP precursor conversion results in
improved control over InP nanocrystal synthesis.
EXPERIMENTAL SECTION
■
General Considerations. All glassware was dried in a 160 °C
oven overnight prior to use. All reactions, unless otherwise noted were
run under an inert atmosphere of dry nitrogen using a Vacuum
Atmospheres OMNI glovebox or using standard Schlenk technique
under an inert atmosphere of dry nitrogen or argon. Celite and 4 Å
molecular sieves (Sigma-Aldrich) were dried under reduced pressure at
200 °C for 16 h prior to use. 1-Octadecene (Sigma-Aldrich, 90%) and
1,2-dimethoxyethane (Sigma-Aldrich, ≥99%) were dried over calcium
hydride, distilled, and stored over 4 Å molecular sieves in a nitrogen-
filled glovebox. Indium acetate (Sigma-Aldrich, 99.99%), myristic acid
(Sigma-Aldrich, ≥99%), red phosphorus (Sigma-Aldrich, ≥99.99%),
and trimethylsilylchloride (Sigma-Aldrich, ≥99%) were used as
received. Sodium metal (Sigma-Aldrich) and potassium metal
(Sigma-Aldrich, 98%) were washed with hexane and had fresh surfaces
cut prior to use. Methanol (Sigma-Aldrich, ≥99.9%) was sparged with
nitrogen for 10 min prior to use. UV−vis spectra were acquired on a
Cary 5000 UV−vis-NIR spectrophotometer. NMR Spectroscopy:
NMR spectra were obtained using 300 or 499 MHz Bruker Avance
spectrometers. In 31P NMR experiments requiring integration, a single
pulse was used to allow quantitative comparison. 31P NMR spectra
were referenced externally with 85% H3PO4. All NMR data was
collected in a deuterated solvent (C6D6, Cambridge Isotope
Laboratories) that was dried with CaH2, distilled, and stored over 4
Å molecular sieves prior to use. TEM images were acquired on an FEI
Tecnai G2 F20 transmission electron microscope.
Synthesis of P(SiMe3)3.27 Caution! Sodium potassium alloy is
pyrophoric and will spontaneously ignite in air and will react violently with
water. P(SiMe3)3 is pyrophoric and will spontaneously ignite in air.
In a typical synthesis of P(SiMe3)3 targeting 11 g, 2.25 g (57.5
mmol, 1.3 equiv) of potassium and 1.80 g (78.3 mmol, 1.8 equiv) of
sodium were weighed out in scintillation vials in a nitrogen filled
glovebox. The sodium and potassium were combined in a 250 mL
three-neck round-bottom flask along with a football-shaped magnetic
stir bar. Approximately 100 mL of 1,2-dimethoxyethane was added to
the three-neck flask. One of the side necks of the flask was fitted with a
24/40 Schlenk line gas flow adapter that had been greased with H-
grease. Two 24/40 septa were then used to stopper the remaining two
necks of the flask. A 1.35 g portion (43.6 mmol of P, 1 equivalent) of
red phosphorus was placed into a solid addition funnel in the
glovebox. The solid addition funnel was greased with silicone grease
and stoppered with a 100 mL round-bottom flask in the glovebox.
The three-neck flask and solid addition funnel were taken out of the
glovebox. After cycling the Schlenk line with nitrogen/vacuum three
times, the three-neck flask was placed under nitrogen, and the flask was
outfitted with a reflux condenser under a counter flow of N2. The
three-neck flask was suspended in a silicone oil bath and heated to a
gentle reflux with vigorous stirring for 2 h. During this time period, the
solution changed colors from a vibrant blue to a dull green. The
apparatus was then raised from the silicone oil bath, and the bath was
An NMR tube was stoppered with a septum in the glovebox, and
the septum was secured with electrical tape. The NMR tube was
brought out of the glovebox, and a 0.4 mL aliquot of the octadecene
solution was injected into the NMR tube using a long metal needle
and a 1 mL disposable syringe. A small piece of parafilm was wrapped
around the top of the NMR tube to ensure the seal. A 31P{1H} NMR
spectrum was taken of the sample to confirm protonolysis of
P(SiMe3)3 to bis(trimethylsilyl)phosphine. If the spectrum confirmed
at least 85% conversion (typically near 100%) to the desired product,
then the Schlenk flask solution was used as a stock solution for the
indium phosphide nanoparticle synthesis.
Synthesis of Tetrabutylammonium Myristate. In a typical
synthesis targeting 5 g, 6.90 g (106 mmol) of a 40% by weight aqueous
solution of tetra-n-butylammonium hydroxide and 2.43 g (106 mmol)
of myristic acid were added to a 250 mL, 3-neck round-bottom flask.
Approximately 100 mL of acetonitrile was added to the flask. The flask
was outfitted with a reflux condenser and connected to the Schlenk
E
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
line. After stoppering the remaining two necks with Teflon stoppers,
the flask was placed under a flow of nitrogen. The flask was suspended
in a silicone oil bath and carefully heated to reflux. Reflux was
maintained for 2 h. Solvent was removed with a rotary evaporator, and
conversion to the desired product was confirmed by 1H NMR
tube rack for hexane and waste washes. Glass pipettes were placed
nearby along with pipet bulbs. The baseline and zero absorbance of the
UV−vis spectrophotometer were set using fluorescence cuvettes filled
with n-hexane in both sample beams. A timer was used to measure the
time from injection of the P(SiMe3)3 solution. 50 μL aliquots of the
reaction mixture were taken at specific times from injection using a
100 μL syringe and injected the test tubes with n-hexane. The
fluorescence cuvette from the UV−vis spectrophotometer used to
baseline the instrument from the front beam was washed once with n-
hexane and once with approximately one-third of the test tube solution
before being filled with the test tube solution with the reaction aliquot
and returned to the spectrophotometer for analysis.
NMR Control Experiments with P(SiMe3)3 and Various Acids.
For the control experiment with myristic acid, 0.007 g of myristic acid
was weighed out into a J-Young NMR tube. A Kim wipe was secured
over the top of the NMR tube with a rubber band, and the NMR tube
was transferred to the glovebox along with the J-Young NMR tube cap.
Approximately 0.4 mL of octadecene and 0.3 mL of toluene were
added to the NMR tube to dissolve the myristic acid. Nine microliters
of P(SiMe3)3 were then injected into the NMR tube, and the NMR
tube was sealed with the J-Young NMR tube cap and removed from
the glovebox for analysis. The octylammonium myristate control was
run analogously with 1 equiv of octylamine first premixed with the
myristic acid along with solvent prior to injection of P(SiMe3)3. The
1H NMR spectrum of the octylammonium myristate is provided in the
Supporting Information, showing complete conversion to octylammo-
nium myristate.
For all other acids, (including trifluoroacetic acid, HCl, sparged
methanol, and 18 MΩ Millipore water) Schlenk flasks were prepared
in the glovebox with approximately 3.2 g of octadecene for the
studying water, and 1.6 g of 1-octadecene for the study of each other
acid and were labeled according to which acid was to be studied. A 46
μL portion of P(SiMe3)3 was injected into each of the Schlenk flasks
with 1.6 g of 1-octadecene and 92 μL of P(SiMe3)3 was injected into
the Schlenk flask with 3.2 g of 1-octadecene. A small football-shaped
magnetic stir bar was added to each Schlenk flask.
Each Teflon valve was closed and each Schlenk flask was sealed with
a 24/40 Suba seal septum and brought out of the glovebox. Each
Schlenk flask was suspended from a claw clamp over a stir plate set to
stir. After cycling the Schlenk line three times with nitrogen, the flask
was placed under a flow of nitrogen. Methanol was sparged for 10 min,
and 6.3 μL of the sparged methanol was injected into the Schlenk flask
labeled methanol using a 25 μL syringe. An air-free container of
trifluoroacetic acid was suspended from a claw clamp over a stir plate
and connected to the Schlenk line. After cycling the line three times
with nitrogen, the flask was placed under a flow of nitrogen. The
Teflon valve of the air-free flask was exchanged for a rubber septum
under a counter-flow of nitrogen, and electrical tape was wrapped
around the septum to ensure an airtight seal. Twelve microliters of
trifluoroacetic acid was drawn from the air-free container using a 25 μL
syringe and injected into the Schlenk flask labeled trifluoroacetic acid.
Three microliters of 18 MΩ Millipore water was injected into the
Schlenk flask labeled water. Three NMR tubes sealed with septa and
electrical tape as described above were prepared in the glovebox and
taken out of the glovebox. A 0.4 mL portion of each Schlenk flask
solution was injected into an NMR using a 1 mL disposable syringe
and long needle, and each NMR tube was appropriately labeled for
analysis.
1
spectroscopy. The H NMR spectrum is provided in the Supporting
Information.
Synthesis of InP-QDs from P(SiMe3)3 and Indium Myristate.
In a typical synthesis, 0.117 g (0.40 mmol) of indium acetate, and
0.331 g (1.45 mmol), or 0.274 g (1.20 mmol) for the stoichiometric
studies, of myristic acid were weighed out into a 25 mL, 14/20 3-neck
round-bottom flask with a small football-shaped magnetic stir bar. The
three neck flask was secured with a claw clamp over a stir plate. A 14/
20 thermowell and a 14/20 reflux condenser were greased with H-
grease. The reflux condenser was inserted into the middle neck of the
round-bottom flask and the thermowell was intserted into one of the
side necks of the three-neck flask. A 14/20 hose adapter was greased
with silicone grease and inserted into the reflux condenser. A 4.00 g
portion of 1-octadecene was weighed out into a scintillation vial in the
glovebox, and the scintillation vial was capped. The scintillation vial
was removed from the glovebox, and the 1-octadecene was delivered
to the 3-neck flask with a glass pipet. The remaining neck of the 3-neck
flask was stoppered with a 14/20 Suba seal septum.
The hose adapter was connected to the Schlenk line, and the
apparatus was evacuated with vigorous stirring. The Schlenk flask was
then lowered into a heating mantle and wrapped with glass wool for
insulation. A temperature probe was inserted into the thermowell, and
both the heating mantle and the temperature probe were plugged into
a temperature controller. The temperature controller was set to 100
°C. The solution was allowed to off-gas acetic acid to generate the
indium myristate solution for approximately 12 h at 100 °C. The
apparatus was then placed under nitrogen and raised to 316 °C using
the temperature controller.
A 2.00 g portion of 1-octadecene was weighed out into a
scintillation vial in the glovebox. Fifty-eight microliters of P(SiMe3)3
was injected into this solution using a 100 μL syringe. This solution
was drawn into a 6 mL disposable syringe outfitted with 21G needle
and then the needle was inserted into the end of a 14/20 septum to
seal off the P(SiMe3)3 solution. The syringe was brought out of the
glovebox and injected into the indium myristate solution when the
temperature controller read 315 °C as the temperature of the solution
was falling, after adjusting the temperature to 285 °C, the quantum dot
growth temperature.
In experiments run using tetrabutylammonium myristate, a solution
of tetrabutylammonium myristate in octadecene was injected after the
overnight degassing step, and the total amount of octadecene was kept
the same as described above. To ensure quantitative transfer, the
tetrabutylammonium myristate solution was added by injection with a
glass syringe after heating the solution to 100 °C to melt the
ammonium salt.
Synthesis of InP-QDs from HP(SiMe3)2 and Indium
Myristate. In a typical synthesis, the indium myristate solution was
prepared in the same manner as that employed for the synthesis of
indium phosphide nanoparticles from P(SiMe3)3 and indium
myristate. The solution of bis(trimethylsilyl)phosphine was prepared
as previously described vide supra. The Teflon seal of the Schlenk flask
was closed off, and the Schlenk flask was inverted to draw 2.5 mL (2 g)
of the bis(trimethylsilyl)phosphine solution into a 6 mL disposable
syringe outfitted with a 21G needle. This solution was then
immediately injected into the solution of indium myristate as the
temperature fell through 315 °C in a manner analogous to the
synthesis performed with P(SiMe3)3.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional experimental details, NMR and UV−visible
absorption data, data analysis and TEM images. This material
is available free of charge via the Internet at http://pubs.acs.org.
In experiments run using tetrabutylammonium myristate, a solution
of tetrabutylammonium myristate in octadecene was injected after the
overnight degassing step, and the total amount of octadecene was kept
the same as described above.
UV−vis Monitoring Procedure. Prior to each injection of
P(SiMe3)3, a test tube rack was prepared with 13 test tubes. Each test
tube was filled with 6 mL of n-hexane using a 6 mL disposable syringe
outfitted with a 18G needle. Two beakers were placed beside the test
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: cossairt@chem.washington.edu.
F
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(25) Harris, D.; Bawendi, M. J. Am. Chem. Soc. 2012, 134, 20211−
20213.
Notes
The authors declare no competing financial interest.
(26) Douglas, T.; Theopold, K. H. Inorg. Chem. 1991, 30, 594−596.
(27) Becker, G.; Schmidt, H.; Uhl, G.; Uhl, W.; Regitz, M.; Rosch,
̈
ACKNOWLEDGMENTS
■
W.; Vogelbacher, U. J. Inorg. Synth. 1990, 27, 243−249.
(28) Goel, S.; Chiang, M.; Rauscher, D.; Buhro, W. J. Am. Chem. Soc.
1993, 115, 160−169.
We gratefully acknowledge the University of Washington for
funding this research. We acknowledge Prof. William Buhro
and Subhash Goel for helpful discussions regarding the
synthesis of HP(SiMe3)2. We acknowledge Prof. Daniel
Gamelin for feedback regarding the preparation and
presentation of this manuscript and Dr. Brian Walker for
helpful discussions. Part of this work was conducted at the
University of Washington NanoTech User Facility, a member
of the NSF National Nanotechnology Infrastructure Network
(NNIN).
(29) Xie, R.; Battaglia, D.; Peng, X. J. Am. Chem. Soc. 2007, 129,
15432−15433.
(30) Mutlugun, E.; Hernandez-Martinez, P.; Eroglu, C.; Coskun, Y.;
Erdem, T.; Sharma, V.; Unal, E.; Panda, S.; Hickey, S.; Gaponik, N.;
Eychmuller, A.; Demir, H. Nano Lett. 2012, 12, 3986−3993.
̈
(31) Li, L.; Reiss, P. J. Am. Chem. Soc. 2008, 130, 11588−11589.
(32) Thuy, U.; Huyen, T.; Liem, N.; Reiss, P. Mater. Chem. Phys.
2008, 112, 1120−1123.
(33) Cros-Gagneux, A.; Delpech, F.; Nayral, C. l.; Cornejo, A.;
Coppel, Y.; Chaudret, B. J. Am. Chem. Soc. 2010, 132, 18147−18157.
(34) Xu, S.; Kumar, S.; Nann, T.; Kumar, A. J. Am. Chem. Soc. 2006,
128, 1054−1055.
ABBREVIATIONS
LEET, lowest energy electronic transition; fwhm, full width at
half-maximum; Abs, absorbance; QD, quantum dot
■
(35) Please see the Supporting Information for more details.
(36) Micic, O.; Cheong, H.; Fu, H.; Sprague, J.; Mascarenhas, a.;
́ ́
Mascarenhas, A.; Nozik, A. J. Phys. Chem. B 1997, 101, 4904−4912.
(37) Average fwhm in wavelength range from 40 to 50 nm.
(38) Owen, J.; Park, J.; Trudeau, P.-E.; Alivisatos, A. J. Am. Chem. Soc.
2008, 130, 12279−12281.
REFERENCES
■
(1) Vurgaftman, I.; Meyer, J. R.; Ram-Mohan, L. R. J. Appl. Phys.
2001, 89, 5815−5874.
(2) Adachi, S. Properties of Group-IV, III-V and II-VI Semiconductors;
John Wiley & Sons: West Sussex, U.K., 2005.
(39) Anderson, N. C.; Owen, J. S. Chem. Mater. 2013, 25, 69−76.
(40) Frederick, M.; Achtyl, J.; Knowles, K.; Weiss, E.; Geiger, F. J.
Am. Chem. Soc. 2011, 133, 7476−7481.
(3) Gupta, S.; Whitaker, J. F.; Mourou, G. A. IEEE J. Quantum
Electron. 1992, 28, 2464−2472.
(4) Schubert, E. F. Doping in III-V Semiconductors; Cambridge
University Press: Cambridge, U.K..
(41) Fritzinger, B.; Capek, R.; Lambert, K.; Martins, J.; Hens, Z. J.
Am. Chem. Soc. 2010, 132, 10195−10201.
(42) Yu, W. W.; Peng, X. Angew. Chem., Int. Ed. 2002, 41, 2368−
2371.
(43) Bullen, C.; Mulvaney, P. Nano Lett. 2004, 4, 2303−2307.
(44) Peng, X. G.; Manna, L.; Yang, W. D.; Wickham, J.; Scher, E.;
Kadavanich, A.; Alivisatos, A. P. Nature 2000, 404, 59−61.
(45) Wang, F.; Tang, R.; Buhro, W. E. Nano Lett. 2008, 8, 3521−
3524.
(46) Wang, F.; Tang, R.; Kao, J. L. F.; Dingman, S. D.; Buhro, W. E.
J. Am. Chem. Soc. 2009, 131, 4983−4994.
(47) Garcia-Rodriguez, R.; Hendricks, M. P.; Cossairt, B. M.; Liu, H.;
Owen, J. S. Chem. Mater. 2013, 25 (8), 1233−1249.
(48) Evans, C. M.; Evans, M. E.; Krauss, T. D. J. Am. Chem. Soc.
2010, 132, 10973−10975.
(49) Hendricks, M. P.; Cossairt, B. M.; Owen, J. S. ACS Nano 2012,
6, 10054−10062.
(50) Cossairt, B.; Owen, J. Chem. Mater. 2011, 23, 3114−3119.
(5) Casey, H. C. Devices for Integrated Circuits: Silicon and III-V
Compound Semiconductors; Wiley: New York, 1998.
(6) Li, T.; Mastro, M. III-V Compound Semiconductors: Integration
with Silicon-Based Microelectronics; CRC Press: Boca Raton, FL, 2011.
(7) Alivisatos, A. P.; Gu, W.; Larabell, C. Annu. Rev. Biomed. Eng.
2005, 7, 55−76.
(8) Chan, W. C. W.; Nie, S. Science 1998, 281, 2016−2018.
(9) Phillips, J. M.; Coltrin, M. E.; Crawford, M. H.; Fischer, A. J.;
Krames, M. R.; Mueller-Mach, R.; Mueller, G. O.; Ohno, Y.; Rohwer,
L. E. S.; Simmons, J. A.; Tsao, J. Y. Laser Photonics Rev. 2007, 1, 307−
333.
(10) Brunetti, V.; Chibli, H.; Fiammengo, R.; Galeone, A.; Malvindi,
M. A.; Vecchio, G.; Cingolani, R.; Nadeau, J. L.; Pompa, P. P.
Nanoscale 2013, 5, 307−317.
(11) Derfus, A. M.; Chan, W. C. W.; Bhatia, S. N. Nano Lett. 2004, 4,
11−18.
(12) Park, J.; Joo, J.; Kwon, S. G.; Jang, Y.; Hyeon, T. Angew. Chem.,
Int. Ed. 2007, 46, 4630−4660.
(13) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc.
1993, 115, 8706−8715.
(14) Peng, X. G.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc.
1998, 120, 5343−5344.
(15) Battaglia, D.; Peng, X. Nano Lett. 2002, 2, 1027−1030.
(16) Allen, P. M.; Walker, B. J.; Bawendi, M. Angew. Chem., Int. Ed.
2010, 49, 760−762.
(17) Green, M. Curr. Opin. Solid State Mater. Sci. 2002, 6, 355−363.
(18) Wells, R. L.; Gladfelter, W. L. J. Cluster Sci. 1997, 8, 217−238.
(19) Baek, J.; Allen, P. M.; Bawendi, M. G.; Jensen, K. F. Angew.
Chem., Int. Ed. 2011, 50, 627−630.
(20) Li, L.; Protier
2623.
̀
e, M.; Reiss, P. Chem. Mater. 2008, 20, 2621−
(21) Lim, T.; Ravi, S.; Bumby, C.; Etchegoin, P.; Tilley, R. J. Mater.
Chem. 2009, 19, 4852−4856.
(22) Xu, S.; Kumar, S.; Nann, T. J. Am. Chem. Soc. 2006, 128, 1054−
1055.
(23) Carenco, S.; Demange, M.; Shi, J.; Boissier
Floch, P.; Mezailles, N. Chem. Commun. 2010, 46, 5578−5580.
(24) Carenco, S.; Hu, Y.; Florea, I.; Ersen, O.; Boissiere, C.;
Mezailles, N.; Sanchez, C. Chem. Mater. 2012, 24, 4134−4145.
̀
e, C.; Sanchez, C.;
́
̀
́
G
dx.doi.org/10.1021/cm401289j | Chem. Mater. XXXX, XXX, XXX−XXX