Continuous Nucleation and Size Dependent Growth Kinetics of Indium Phosphide Nanocrystals

Article abstract of DOI:10.1021/acs.chemmater.0c01561

Aminophosphines derived from N,N′-disubstituted ethylenediamines (R-N(H)CH₂CH₂N(H)-R; R = ortho-tolyl, phenyl, benzyl, iso-propyl, and n-octyl) were used to adjust the kinetics of InP nanocrystal formation by more than 1 order of mag

Full text of DOI:10.1021/acs.chemmater.0c01561

pubs.acs.org/cm
Article
Continuous Nucleation and Size Dependent Growth Kinetics of
Indium Phosphide Nanocrystals
Brandon M. McMurtry, Kevin Qian, Joseph K. Teglasi, Anindya K. Swarnakar, Jonathan De Roo,
and Jonathan S. Owen*
Cite This: https://dx.doi.org/10.1021/acs.chemmater.0c01561
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Aminophosphines derived from N,N′-disubstituted
ethylenediamines (R−N(H)CH2CH2N(H)−R; R = ortho-tolyl,
phenyl, benzyl, iso-propyl, and n-octyl) were used to adjust the
kinetics of InP nanocrystal formation by more than 1 order of
magnitude. Ultraviolet−visible absorption and 31P nuclear magnetic
resonance measurements demonstrate that the rate of nanocrystal
formation is limited by the precursor reactivity. At low temperature
(180 °C), crystal nucleation is concurrent with growth throughout
the reaction, rather than occurring in a burst at early times. The low
temperature produces a narrow range of small sizes (d = 4.2−4.9
nm) regardless of the precursor used. Higher temperatures (up to
270 °C) promote growth to larger sizes (d ≤ 7.8 nm), shorten the nucleation period, and create conditions where the final size is
controlled by the precursor conversion reactivity. The temperature dependence is proposed to arise from growth kinetics that slow as
the nanocrystal size increases, a novel surface attachment limited size distribution-focusing mechanism. Such a mechanism supports a
narrow size distribution without separating the nucleation and growth phases.
Scheme 1. Synthesis of InP from Aminophosphines, Metal
Halides, and R′−NH2 (R′ = Oleyl)
INTRODUCTION
■
InP nanocrystals are leading replacements for CdSe nano-
crystals in solid-state lighting,1 luminescent displays,2,3 and
biological imaging.4,5 However, their broad emission line-
widths, chemical instability, moderate photoluminescence
quantum yields, and poor absorptivity at blue pump wave-
lengths present significant limitations that have slowed their
adoption.6,7 Although important advances in the synthesis of
InP nanocrystals have recently appeared,8−11 it remains
difficult to modify the microstructure of InP nanocrystals to
address the performance limitations described above. Well-
documented synthetic difficulties are partly the result of a
limited mechanistic understanding of InP nanocrystal for-
mation. Moreover, leading mechanistic hypotheses suggest that
InP formation does not follow a typical LaMer-like
sequence.12−17 A deeper understanding of InP nanocrystal
formation mechanisms is therefore of great interest
currently.18−23
The majority of synthetic approaches to InP use indium
carboxylates, In(O2CR)3 (R = n-alkyl), and tris-trimethylsilyl-
phosphine ((TMS)3P) to produce nanocrystals with narrow
optical absorption and photoluminescence (PL) features after
shelling with ZnS or ZnSe (PLfwhm = 36−50 nm).6,19,20 Unlike
II−VI and IV−VI nanocrystals, where the kinetics of precursor
conversion are rate limiting and the extent of nucleation can be
controlled by the precursor reactivity,14,15,17,19,24,25 the
nucleation and growth of III−V nanocrystals are, thus far,
less sensitive to the precursor design. For example, silyl,
germanyl, and stannyl pnictide derivatives adjust the kinetics of
InP formation from indium carboxylates, but have a minor
effect on the final size or dispersity.20,24−26 To explain these
results, a mechanism has been proposed where the formation
of stable cluster intermediates decouples the influence of the
precursor reactivity on the nucleation.20
Indium(III) halides (halide = Cl, Br, and I) and tris-
aminophosphines (e.g., P(NMe2)3) can also be used to
synthesize InP nanocrystals, albeit with broader optical features
(absorbance full-width at half-maxima (fwhm): 48−80
nm).22,27,28 Mechanistic studies have shown that the amino-
Received: April 11, 2020
Revised: April 21, 2020
https://dx.doi.org/10.1021/acs.chemmater.0c01561
© XXXX American Chemical Society
Chem. Mater. XXXX, XXX, XXX−XXX
A

Chemistry of Materials
pubs.acs.org/cm
Article
phosphine precursor rapidly transaminates with the oleylamine
solvent before undergoing redox disproportionation to InP,
according to Scheme 1.22,29,30 Although the nanocrystal
formation kinetics, polydispersity, and final size are sensitive
to the metal halide coreactant, it is unclear whether the
aminophosphine conversion reaction shown in Scheme 1 limits
the kinetics of nucleation and growth.
We hypothesized that aminophosphine derivatives would
provide a straightforward approach to adjust the kinetics of InP
formation and probe its influence on the nanocrystal
formation. By accessing a wide range of reactivity, the solute
production kinetics can be decoupled from the reaction
temperature, allowing the influence of temperature on the
crystal growth phase to be independently probed. In particular,
we sought aminophosphines that produce solute more slowly
than it is consumed by crystal growth (precursor conversion
limited growth), conditions that allow the extent of nucleation
and the final size to be controlled in other material systems.
These conditions can also evaluate whether the formation of
clusters decouples the solute supply from the growth.
Our effort led to the surprising conclusion that crystal
nucleation occurs slowly and steadily at low reaction
temperatures where nanocrystals are resistant to growth,
while higher temperatures support a more burst-like, albeit
slow, nucleation process. The slow continuous nucleation
kinetics starkly contrast with the LaMer model but nonetheless
provide narrow size distributions. These results can be
explained with a model where growth kinetics are limited by
attachment and depend on the nanocrystal sizea novel
mechanism that can enforce a narrow size distribution.
RESULTS
■
Cyclic ethylenediaminophosphines (1−5) are prepared in a
two-step sequence from phosphorus trichloride via the
Figure 1. (A) Reaction scheme for the conversion of 1−5 and
indium(III) halides to InP. (B) 31P{1H} NMR spectra (note that (*)
indicates a tri-n-octylphosphine oxide internal standard) and (C)
UV−vis absorption spectra of timed reaction aliquots for an InP
reaction with InI3 and 3a at 180 °C. (D) STEM image (left) for the
reaction of InCl3 and 3a at 180 °C and TEM image (right) of
nanocrystals from the reaction of InI3 and 5a at 270 °C.
corresponding diaminochlorophosphine (Scheme 2).31−36
A
Scheme 2. Aminophosphine Precursors Used in This Study
and form an unidentified black mixture, while 3−5 and their
parent diamines appear stable up to 290 °C. Thus, fewer
precursors were functional at the highest temperatures
evaluated here.
As described previously, the yield of InP can be determined
by measuring the optical absorption at λ = 413 nm.27 At 413
nm, a size independent extinction coefficient has been
calculated and the absorption from solutes such as clusters
with less than ∼20 InP units is negligible.20,37 The yield of InP
nanocrystals is compared with the conversion of the amino-
phosphine as measured by quantitative 31P NMR spectroscopy
(Figure 2A). When accounting for the formation of reactive
intermediates, these measurements confirm that the yield of
InP matches the conversion of molecular precursors. Mass
balance measurements show that >90 10% of InP produced
by the precursor conversion is accounted for by the absorbance
at λ = 413 nm (see the Supporting Information). The
precursor conversion kinetics can, therefore, be monitored
indirectly by following the absorption of nanocrystals at λ =
413 nm.
variety of N,N′ disubstituted ethylenediamines, including
sterically bulky N,N′-di(iso-propyl) and N,N′-di(ortho-tolyl)
derivatives 1 and 3, could be prepared in good yield and purity.
InP nanocrystals are prepared from 1 to 5, adapting
conditions previously optimized for P(NMe2)3 (Figure
1).22,27,28 Four equivalents of 1−5 are swiftly injected into a
colorless solution of indium(III) halide (InX3, X = Cl, Br, or I),
zinc chloride, and oleylamine at temperatures between 180 and
270 °C. Studies have shown that zinc halides improve the
spectral linewidths and participate in the precursor con-
version.27,30 Above 250 °C, ethylenediamines and cyclic
aminophosphines with aryl substituents (1 and 2) are unstable
Once the yield of InP reaches its maximum value, the
nanocrystals are isolated and purified. Powder X-ray diffraction
and transmission electron microscopy (Figures 1D and S1−
B
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
independent extinction coefficient (λ = 310 nm)41 where InP
clusters are known to absorb. These kinetics match those
measured at λ = 413 nm throughout the reaction. This strongly
suggests that any cluster intermediates are minor species. (2)
Clusters could not be separated from aliquots following
purification by precipitation; a minor (∼10%) reduction in the
ratio of the excitonic absorbance maximum and the absorbance
at 413 nm is observed that may be caused by scattering or the
absorbance of organic contaminants. (3) A mass balance was
performed by comparing quantitative 31P NMR measurements
of precursors, intermediates, and byproducts, with the InP
yield measured by ultraviolet−visible (UV−vis) absorption
spectroscopy. These measurements account for >90 10% of
all phosphorus in the reaction mixture. When considering mass
losses upon injection (∼5−10%), we conclude that the
majority of phosphorus is observed using NMR spectroscopy
and UV−vis absorption measurements. These experiments all
indicate that clusters are minor components of our reaction
mixture and do not meaningfully influence the absorption of
aliquots at 413 nm. Moreover, any small changes in the kinetics
caused by the presence of clusters cannot explain the
magnitude of changes in the absorption over time nor the
differences between precursors reported herein. We conclude
that the absorption measurements performed here selectively
monitor the yield of InP nanocrystals (within ∼10%) and are
not meaningfully influenced by cluster intermediates.
At 180 °C, the least reactive precursors (2, 4, and 5) achieve
partial yield over the course of an hour, while more reactive
compounds reach full conversion in under an hour (Figure 2).
The rate of conversion and the yield from less reactive
precursors increase at higher temperatures, often reaching
100% yield as measured by UV−vis. However, the volatility of
some precursors and conversion byproducts lead to some
variability that is discussed in the Supporting Information.
At the reaction temperature, the aminophosphines (1−5)
transaminate with the oleylamine solvent, eventually leading to
the tetrakis oleylaminophosphonium coproduct (Figures 1B
and S9). These species are consistent with a conversion
mechanism where a tris oleylaminophosphine (6) intermediate
formed by transamination undergoes redox disproportionation
to form InP, as has been elucidated for the conversion of
P(NMe2)3 previously.22 A relatively large amount of 6 (∼50%)
is visible in the 31P NMR spectrum of aliquots when more
reactive precursors (1) are used, while trace amounts of 6 are
observed when more sluggish precursors (2−5) are used
(Figures 1B and S10). These observations suggest that
ethylenediamines reduce the precursor reactivity by inhibiting
the formation of 6, according to Scheme 3. Relative
transamination equilibrium constants (Keqrel) were measured
by heating ethylenediamines with 1 in oleylamine and
monitoring the ratio of 31P NMR signals for each (Figure
Figure 2. (A) Percentage of InP formed (black) and precursor species
remaining (white), and linear fit of early time points to determine Q0
(dashed line) for an InP reaction using 1c. (B) InP yield (points) and
Q0 fits (dashed lines) of P(NEt2)3 (red), 1b (orange), P(NMe2)3
(yellow), 3a (green), 2b (light blue), 5a (dark blue), and 4a (purple)
under InP-forming conditions shown in Figure 1A.
S3) verify that the nanocrystals have a zincblende crystal
structure and a quasi-tetrahedral shape at small sizes which
becomes spherical at large sizes. Nanocrystal sizes are
estimated from the energy of the absorption maximum using
a previously published sizing curve and could be tuned from d
= 4.2−7.8 nm by adjusting the synthesis temperature and
aminophosphine structure (see the Experimental Section and
Supporting Information).28
Reactions conducted with InI3 at 180 °C produce nano-
crystals with d = 4.2−4.9 nm regardless of the aminophosphine
used, including when a partial yield of InP is obtained. Larger
sizes (up to 7.8 nm) are obtained at higher temperature and
when InCl3 is used. We chose InI3 for more detailed kinetics
studies because of its tendency to produce smaller InP
nanocrystals, which provided more room to grow larger sizes
using less reactive precursors (i.e., 1−5).
InP nanocrystals with similarly narrow optical features are
obtained in all reactions described herein. Although no
meaningful PL is observed until a ZnSe shell is deposited
(Figure S4), the half-width at half maximum (hwhm) of the
lowest energy optical absorption feature can be compared with
previously reported values (Figure S6).27 The hwhm values
(26−50
2 nm) are equivalent to or narrower than the
original report when starting from InI3, InCl3, a variety of
aminophosphines, and various reaction temperatures (Figure
S6). Narrow size distributions (σ = 9−14%) were confirmed
using transmission electron microscopy (Figure S3). We
conclude that syntheses beginning from 1−5 and P(NMe2)3
produce narrow size distributions that are equivalent to the
original InX3/P(NR2)3 report. The insensitivity of the
absorption linewidth to the precursor used supports a
precursor conversion reaction that is orthogonal to the
crystallization steps.38,39 Finally, a recent study has shown
that trace water improves the size distribution obtained from
InX3/P(NR2)3, a variable not explored in our work.40
Scheme 3. Proposed Competitive Transamination
Preceding Conversion to [InP]i
The kinetics of InP formation is monitored by measuring the
optical absorption (λ = 413 nm) of timed aliquots.27 Although
large cluster intermediates (>[InP]20) could, in principle,
contribute to the absorption at this wavelength, no
spectroscopic signatures of clusters were observed herein nor
has any evidence for InP clusters been reported for syntheses
conducted with InX3/P(NR2)3. Several experiments were
performed to test whether persistent cluster intermediates
complicate our kinetics measurements (see the Supporting
Information): (1) kinetics were measured using another size
C
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
S12). These competition experiments verified that the affinity
of ethylenediamine for phosphorus (o-tol < i-Pr < Ph ≪ n-
octyl ≈ Bn) parallels the relative aminophosphine precursor
reactivity. We conclude that the ethylenediamines inhibit the
formation of 6 and thereby slow the conversion kinetics.
To further test this hypothesis, we measured the kinetics of
InP formation from P(NMe2)3 in the presence of added
ethylenediamines. At the reaction temperature, transamination
with the added ethylenediamines occurs prior to InP
formation, as could be observed with 31P NMR spectroscopy.
Increasing the amount of added N,N′-diphenylethylenedi-
amine and N,N′-di-iso-propylethylenediamine to more than 1
diamine/phosphorus slows the kinetics of InP formation;
additional equivalents lower Q0 below their respective
phosphines (2 and 3) (Figures 3 and S13). Added N,N′-di-
oleylamine for phosphorus, reducing the concentration of 6.
This explains why the aryl derivatives are more reactive
compared to the more basic aliphatic derivatives and why the
sterically bulky o-tolyl (1) derivative is more reactive than the
phenyl (2) analogue. Likewise, it explains why the bulkier iso-
propyl (3) derivative is more reactive than the benzyl (4) or n-
alkyl (5) analogues. Moreover, the similarity of the o-tolyl
derivative (1) and P(NMe2)3 implies that the sterically bulky
o-tolyl substituents destabilize the five-membered ring, making
it equally reactive toward transamination as the acyclic
P(NMe2)3. 4 and 5 combine high basicity and a low steric
profile, leading to the lowest concentration of 6 and the
slowest conversion reactivity. The trends also argue against an
alternative mechanism where the parent aminophosphines
(rather than 6) undergo redox disproportionation prior to
transamination. Such a reaction pathway might be expected to
be faster when more reducing alkyl derivatives are used, in
contrast with the trend in Figure 2B. We conclude that the
relative reactivity of the aminophosphines is governed by the
magnitude of the pre-equilibrium constant (Keq) depicted in
Scheme 3.
Adding N,N′-di-o-tolylethylenediamine to reactions with
P(NMe2)3 slightly speeds the reaction, as shown in Figure 3.
This suggests a parallel conversion of 1 directly to InP or a
spurious side reaction caused by the o-tolyl substituents.
Syntheses conducted with 1 also produce slightly broader
optical spectra and a small but steady decrease in absorbance at
413 nm at long reaction times; both support minor undesirable
side reactions when 1 is used. The weak benzylic C−H bonds
and the instability of aryl-substituted ethylenediamines above
250 °C also support this hypothesis. Without more detailed
analysis, it is unclear whether the diamines have other effects
on the reaction (e.g., chelation to indium) that affects the
kinetics or yield.
The 15× range of Q0 provided by 1−5 has a negligible
impact on the final nanocrystal size at 180 °C. Nanocrystals
with d = 4.2−4.9 nm are isolated from all aminophosphines
tested, including slowly reacting precursors such as 4 that only
achieve ∼25% yield over an hour at this low temperature. The
latter observation suggests that the extent of nucleation and the
InP yield are related, a finding that is inconsistent with a burst
of nucleation at early times.
At higher temperatures, larger nanocrystals are obtained
whose final size can be adjusted by the precursor reactivity.
Plotting the final nanocrystal volume versus Q0 at several
reaction temperatures demonstrates that the extent of
nucleation becomes more sensitive to the precursor reactivity
as the reaction temperature is increased (Figure 4). Note that
the nanocrystal sizes/volumes shown in Figure 4 are under-
estimated at large sizes where the shape becomes spherical
rather than tetrahedral. The underestimation is an artifact of
the sizing method described in the Experimental Section. Thus,
the influence of the reaction temperature on the extent of
nucleation is more pronounced than is already clearly visible in
Figure 4. This assumption only strengthens the mechanistic
findings described below.
Our results can be explained using the nucleation mass
balance model formalized by Sugimoto and described in
previous studies on PbS, PbSe, CdS, CdSe, AgCl, and
AgBr.13−15,17,42−46 This model explains how the nucleation
and the growth processes compete with one another for
supersaturated solutes (Scheme 4). The model defines
nucleation as the formation of “stable nuclei”, that is, crystals
Figure 3. Q0 for the reaction of P(NMe2)3 in the presence of varying
concentrations (i.e., equiv/indium) of ethylenediamines at 180 °C.
The Q0 value for the conversion of P(NMe2)3 without additional
diamine is shown by the empty circle and dashed line for reference.
o-tolylethylenediamine, on the other hand, slightly increases
the reactivity relative to P(NMe2)3 alone (see the Discussion
section below). Unsurprisingly the effect of the ethylenedi-
amines on the relative reactivity of these mixtures is consistent
with the relative reactivities of cyclicaminophosphines 1−5.
With the exception of the o-tolyl derivative, we conclude that
the ethylenediamines compete with oleylamine for phospho-
rus, inhibiting the formation of 6 and slowing the conversion
reaction.
DISCUSSION
■
The relative reactivity of the aminophosphines can be
explained by a transamination equilibrium between the parent
amino substituents and oleylamine, as shown in Scheme 3.
This pre-equilibrium controls the concentration of the reactive
intermediate, 6, which then undergoes rate determining
disproportionation. This explanation is consistent with our
observations that the most reactive aminophosphines produce
the highest amounts of 6 in NMR spectra of aliquots and that
added ethylenediamine inhibits the conversion reactivity. The
steady decrease in reaction rate on adding increasing
equivalents (0.25−2 diamine/phosphorus) of N,N′-di-iso-
propyl and N,N′-di-phenyl ethylenediamines to reactions
with P(NMe2)3 supports a rapid and reversible pre-equilibrium
preceding the rate determining conversion step.
The substituent effects on the conversion reactivity are also
consistent with the pre-equilibrium formation of 6 prior to the
rate determining step. Amino substituents with greater basicity
and lower steric profile compete most effectively with
D
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
reducing the conversion reactivity, we can access precursor
conversion limited kinetics at high temperature.
A similar reluctance to grow beyond a few nanometers has
also been reported in syntheses of InP from In(O2CR)3 and
P(SiMe3)3.1,8−10,47−49 Although the final size in these studies is
different from the present one, very different precursors,
kinetics, surfactants, and reaction temperatures can be
expected to cause significant differences in the growth
dynamics. Moreover, the formation of oxide byproducts is
thought to play an important role in these examples.50 Others
have shown that clusters persist in syntheses from In(O2CR)3
and P(SiMe3)3 at temperatures below 150 °C, much lower
temperatures than those used here.20 At higher temperatures,
these clusters ripen into larger nanocrystals. Although mass
spectrometry has identified clusters in synthesis mixtures at
nearly 300 °C, their chemical composition, abundance, and
lifetime in the reaction mixture are unknown and they may be
a minor component of the available solute.41 We conclude that
the ripening of clusters into nanocrystals is consistent with the
mechanism shown in Scheme 4, where clusters are a form of
the solute that is consumed by growth.
Figure 4. Plot of the final InP NC volume vs Q0 for five different InP-
forming conditions at 180, 200, 220, 250, and 270 °C. At higher
temperatures, the growth rate increases and addition of monomers to
the nanocrystal surface consumes a greater fraction of the solute.
A plot of the nanocrystal concentration versus time allows
the length of the nucleation phase to be estimated (Figure
5A,B). In all reactions studied here, nucleation continues over
Scheme 4. Nucleation Mass Balance Showing the
Production and Consumption of Solutes by Nucleation and
Growth
that do not redissolve but grow to their final size. Likewise,
growth is the sum of all processes that deposit monomers on
nanocrystals equal to or greater than the radius of stable nuclei
(i.e., r > r*). The rate can be expressed as the average rate of
change in the number of units per nanocrystal and includes the
consumption of unstable nuclei, clusters, oligomers, and
monomers. The model does not, however, consider mecha-
nisms where “stable nuclei” and growing nanocrystals
aggregate or Ostwald-ripen. These processes do not appear
to be important under our conditions as the concentration of
nanocrystals steadily increases during all kinetics measure-
ments in this study (see below). If such mechanisms were
occurring in parallel, they would only dampen the magnitude
of the trends observed herein, strengthening the mechanistic
conclusions reached below.
The mass balance framework illustrated in Scheme 4 leads to
an inverse relationship between the rate of crystal growth and
the number of nanocrystals, provided that the nucleation and
growth kinetics are limited by precursor conversion. Under
these conditions, the steady-state monomer concentration is
determined by the number of nanocrystals and the precursor
reactivity, and tuning the precursor conversion kinetics will
change the extent of nucleation and the final size.
Figure 5. Plot of nanocrystal concentration (pink) and InP yield
(gray) vs time for reactions of 3a at 180 °C (A) and 5a at 250 °C (B)
under InP-forming conditions. (C) Nucleation as a fraction of the
total reaction time vs temperature. The nucleation fraction is defined
as the fraction of [nanocrystal] and InP yield half-lives (t1/2,[NC]
1/2,InP). These values were extracted by fitting InP yield and
[nanocrystal] vs time plots to a single exponential function (y = A
−kt), the latter of which is displayed in Section 6 of the Supporting
/
t
e
Information.
a significant percentage of the total reaction time (Section 6 of
the Supporting Information). Interestingly, the length of the
nucleation phase depends on temperature. At low temperature,
where 4.2−4.9 nm nanocrystals are consistently obtained
(Figure 5A), the number of nanocrystals steadily increases
throughout the synthesis (nucleation half-life or t1/2,[NC] ≈ 500
s), while at high temperatures (Figure 5B), the nucleation
reaches completion at earlier times (t1/2,[NC] ≈ 30 s) and is
followed by growth. The steady eightfold increase in the
concentration of nanocrystals plotted in Figure 5A strongly
supports the slow continuous nucleation under these
However, the consistent final size obtained at low temper-
ature suggests that growth kinetics slow as the nanocrystal
reaches d = 4 nm. This is consistent with a mechanism that is
limited by growth kinetics, rather than precursor conversion.
Herein lies the value of the aminophosphine library: by
E
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
conditions. In all cases, nucleation persists for many minutes to
hours and often for more than 50% of the total reaction time.
The percentage of the reaction over which nucleation occurs
can be estimated from t1/2,[NC] and the time over which the InP
yield reaches 50% or the reaction half-life (t1/2,[InP]); we refer
to their ratio as the “nucleation fraction” and use this value to
estimate the relative duration of the nucleation phase at
different temperatures. These values are plotted at several
temperatures for all precursor combinations studied here in
Figure 5C. At low temperatures, nucleation continues over up
to half of the total reaction, while it reaches completion in the
first 10% of the reaction at higher temperatures. We conclude
that nucleation occurs in more of a “burst” at high
temperatures, while lower temperatures support slow con-
tinuous nucleation.
Although theoretical work on size dependent attachment
kinetics is uncommon, Finke and co-workers have recently
demonstrated a computational framework for this mechanism
using population balance modeling.57 In addition, a computa-
tional study of InP nanocrystal formation by Kulik and co-
workers showed that the distribution of indium atoms on
carboxylate-bound InP nanocrystal surfaces can dictate the
energetics of growth, which slows as the size increases.58,59
Likewise, it is well known that surfactants have a significant
influence on the final nanocrystal size and growth kinetics in
hot-injection syntheses and ripening experiments, including
recent work regarding the role of trace water on the size
distribution.17,37,40,60−64 It is therefore of great interest to
probe the intrinsic reactivity of nanocrystal surfaces toward
monomer attachment and how this reactivity is influenced by
size, shape, and surface ligands. New perspectives on this
reactivity can lead to improved mechanisms of size distribution
control.
Finally, it is interesting to consider whether the lack of a
nucleation burst and size dependent growth kinetics may be
relevant in other colloidal nanocrystal syntheses. In particular,
this mechanism could explain the recent report of narrower
size distributions obtained using a “heat-up” method.8 Low
temperatures at early reaction times could produce a narrow
distribution of small nanocrystals that then grow as the
temperature increases, effectively separating the nucleation and
growth stages. A related study by the same authors reports the
slow injection of InP solute into a solution of seed crystals at
low temperature.9 These conditions did not increase the seed
size which was hypothesized to be the result of continuous
nucleation. Both observations point to a mechanism where
continuous nucleation and size dependent growth kinetics are
operative in InP syntheses from indium carboxylates and
(TMS)3P.19,20,65 Interestingly, a size dependent growth rate
can explain the selective formation of atomically precise,
pyramidal CdSe clusters.60 A quantized growth mechanism,
where a small pyramid grows to the next larger size but does
not grow further, requires that monomers selectively attach to
small rather than large clusters.
Continuous nucleation has also been observed in syntheses
of CdSe nanocrystals monitored in situ using small angle X-ray
scattering.66 However, this study and other studies that analyze
the temporal evolution of the CdSe and PbS nanocrystal
concentration typically attribute the size distribution to a burst
of nucleation, as depicted by LaMer. It should be noted,
however, that measurements of the nanocrystal concentration
are challenging at length scales less than 4 nm, where precise
measurements of the nanocrystal shape, density, and
stoichiometry are required to determine the number of
monomer units per nanocrystal.67 This weakness is especially
true in syntheses that traverse a wide range of sizes. In the
present work, the average size evolves over a relatively small
range, reducing this source of uncertainty.
Finally, size dependent growth kinetics can help explain the
formation of a second population of nanocrystals during
deposition of a shell material onto a core nanocrystal (e.g., CdS
onto CdSe cores). If the rate of growth slows as the shell
thickens, it can lead to monomer supersaturation and
homogeneous nucleation. This observation helps explain the
importance of high-temperature conditions used in the
synthesis of core−shell nanocrystals with thick shells and
recent work on the synthesis of InP/ZnSe nanocrystals.3
Given the long nucleation times observed here, it is perhaps
surprising that relatively narrow size distributions are obtained.
The canonical picture of monodisperse nanoparticle formation
relies on a rapid “burst” of nucleation that is separate from
growth, as famously attributed to LaMer.16,51,52 In this picture,
the duration of the nucleation burst is thought to be related to
the size distribution. The distributions obtained here, as
estimated from the hwhm of the excitonic feature in the
absorption spectrum, are plotted versus the nanocrystal size in
Figure S6. Comparing these values versus temperature shows
that improved separation between nucleation and growth does
not correlate with narrower optical absorption features.
Although a detailed analysis of the size distribution is
beyond the scope of this study, it is clear that a form of size
focusing is required to explain our results. Previous work on
size distribution focusing describes how such a mechanism can
be a consequence of diffusion limited growth kinetics.53−55
However, neither the highly cited work of Peng and Alivisatos
nor subsequent papers consider the broad, size dependent
linewidths of single nanoparticles when estimating the size
distribution evolution with time. Recent studies on lead and
cadmium chalcogenide nanocrystals have shown that the
absorption and PL linewidth of a single nanocrystal are
intrinsically broadened (accounting for the majority of the
total width in monodisperse samples) and that the broadening
is strongly size dependent.13−15 Extracting a nanocrystal size
distribution from the PL linewidth is therefore incorrect if the
intrinsic width of a single size is not taken into account.
Treating the intrinsic single nanocrystal absorption/PL
linewidth as being much smaller than the broadening from
polydispersity will incorrectly lead to the conclusion that the
size distribution undergoes focusing during growth. Moreover,
this analysis assumes that all nanocrystals are produced in a
burst at early times, rather than a more complex distribution
that results from concurrent nucleation and growth over the
majority of the synthesis. These oversights invalidate previous
claims of size distribution focusing based on analysis of the
optical spectrum.
Likewise, a diffusion limited mechanism cannot explain our
results including (1) the sluggish growth of nanocrystals
beyond 4.9 nm at 180 °C, (2) a nucleation process that is more
rapid than growth beyond 4.9 nm over more than 1 order of
magnitude of solute supply kinetics, and (3) thermally
activated growth kinetics. Diffusion in liquids is not thermally
activated.56 This leaves the conclusion that the polydispersities
obtained here are the result of size dependent attachment
kinetics, a mechanism of size distribution focusing with an
unknown origin.
F
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
FEI Talos F200X. NMR spectroscopy was performed using Bruker
400 and 500 MHz spectrometers.
CONCLUSIONS
■
The mechanistic investigations reported here benefit from
relatively slow precursor conversion kinetics at both low and
high temperatures. Higher temperatures increase the rate of
growth inducing precursor conversion limited kinetics, much
like syntheses of PbS, PbSe, CdS, CdSe, AgCl, and AgBr
nanocrystals. Under these conditions, the cyclic amino-
phosphine precursor reactivity tunes the number of InP
nanocrystals and their final size. All reactions studied here
undergo slow, continuous nucleation concurrent with growth
yet achieve narrow size distributions. To explain these results,
the nanoparticle growth rate must be size dependent, a
mechanism that appears to be determined by the kinetics of
monomer attachment, rather than diffusion limitations. These
mechanistic proposals provide an experimentally validated
alternative to the LaMer model.
Synthesis of Cyclic Trisaminophosphines (1−5). N,N′-
Dioctylethylenediamine was synthesized according to a published
procedure.69 Purity was confirmed via H and 13C NMR spectros-
1
1
copy. H NMR (500 MHz, C6D6): δ 0.90 (t, 8H), 1.28 (m, 20H),
1.45 (quin, 4H), 2.56 (t, 4H), 2.66 (s, 4H). 13C{1H} NMR (125
MHz, C6D6): δ 14.00 (s), 22.74 (s), 27.50 (s), 29.45 (s), 29.70 (s),
30.54 (s), 31.93 (s), 49.91 (s), 50.10 (s).
2-Chloro-1,3-di(2-methylphenyl)-1,3,2-diazaphospholidine.
N,N′-Di(2-methylphenyl)ethylenediamine (6.004 g, 25.00 mmol),
triethylamine (17.675 g, 175.00 mmol), and dichloromethane (200
mL) are loaded into a Schlenk flask. The mixture is cooled to 0 °C on
a Schlenk line. Phosphorus trichloride (2.30 mL, 26.3 mmol) is added
dropwise and allowed to stir for 1 h. Precipitation of a white solid is
observed immediately upon addition. The reaction is allowed to
return to room temperature for 1 h, and volatiles are removed in
vacuo. The residue is dissolved in THF and filtered. Volatiles are
removed from the filtrate, and the resulting solid was recrystallized in
THF at −40 °C to yield a pale yellow solid. Yield 6.277 g (83%).
1
31P{1H} NMR (202 MHz, CDCl3): δ 148.27. H NMR (500 MHz,
EXPERIMENTAL SECTION
■
CDCl3): δ 2.46 (s, 6H), 3.72 (s, 2H), 4.12 (s, 2H), 7.20 (t, 2H), 7.28
(m, 4H), 7.46 (d, 2H). 13C{1H} NMR (125 MHz, CDCl3): δ 18.64
(d), 52.03 (d), 126.28 (d) 126.46 (d), 127.02 (d), 131.26 (s). Anal.
Calcd for C16H18ClN2P: C, 63.06; H, 5.95; N, 9.19. Found: C, 62.69;
H, 6.18; N, 8.99. MS (ASAP) m/z: calcd for [C16H18N2PCl + H+],
305.0974; found, 305.0958.
2-Chloro-1,3-dibenzyl-1,3,2-diazaphospholidine. The abovemen-
tioned procedure was followed using N,N′-dibenzylethylenediamine
(3.004 g, 12.5 mmol), triethylamine (7.575 g, 75.0 mmol),
phosphorus trichloride (1.15 mL, 13.1 mmol), and diethyl ether
(100 mL). The crude product is a dark orange liquid used without
further purification. Yield 3.076 g (81%). 31P{1H} NMR (202 MHz,
CDCl3): δ 167.83. 1H NMR (500 MHz, CDCl3): δ 3.29 (d, 4H), 4.29
(s, 4H), 7.36 (m, 2H), 7.44 (m, 8H). 13C{1H} NMR (125 MHz,
CDCl3): δ 50.05 (d), 51.47 (d), 127.81 (s), 128.65 (d). Anal. Calcd
for C16H18N2PCl: C, 63.06; H, 5.95; N, 9.19. Found: C, 62.78; H,
6.25; N, 9.22. MS (ASAP) m/z: calcd for [C16H18N2PCl + H+],
305.0974; found, 305.0978.
General Methods. All manipulations were conducted using
standard air-free techniques unless otherwise specified. Acetone
(≥99.5%), dichloromethane (≥99.5%, contains 40−150 ppm amylene
as a stabilizer), dimethylformamide (ACS reagent, ≥99.8%), lithium
aluminum hydride (powder, reagent grade, 95%), octanoic acid
(99%), sodium hydroxide (anhydrous, free-flowing, pellets ACS
reagent, ≥97%), tetrahydrofuran (≥99.0%, contains 250 ppm BHT as
inhibitor), toluene (≥99.5%), and triethylamine (≥99%) were
obtained from Sigma-Aldrich and used without further purification
for manipulations in air. Indium(III) chloride (99.999%, anhydrous),
indium(III) iodide (99.999%), tri-n-octylphosphine (97%), and
zinc(II) chloride (≥97, anhydrous) were obtained from Strem
Chemicals and used without further purification. N,N′-diphenylethy-
lenediamine (98%), octadecylamine (≥99.0%), selenium (powder,
−100 mesh, 99.99%, trace metal basis), and zinc stearate (technical
grade, 65%) were purchased from Sigma-Aldrich and used without
further purification. N,N′-di(o-tolyl)ethylenediamine (98%) was
purchased from Alfa Aesar and used without further purification.
Chloroform-d (99.8%) and benzene-d6 (99.5%) were obtained from
Cambridge Isotopes and stored in a glovebox over 3 Å molecular
sieves. Dichloromethane (≥99.5%, contains 40−150 ppm amylene as
a stabilizer), diethyl ether (≥99.9%, inhibitor-free), and tetrahy-
drofuran (≥99.0%, contains 250 ppm BHT as an inhibitor) were
obtained from Sigma-Aldrich, degassed, dried in a column packed
with activated alumina, and stored in a glovebox over 3 Å molecular
sieves for 24 h prior to use. Diethylamine (≥99.5%), triethylamine
(≥99%), N,N′-dibenzylethylenediamine (97%), N,N′-diisopropyle-
thylenediamine (99%), octadecene (technical grade, 90%), tetra-
ethylene glycol dimethyl ether (≥99%), trioctylamine (98%),
tris(diethylamino)phosphine (97%), and tris(dimethylamino)-
phosphine (97%) were obtained from Sigma-Aldrich, stirred over
calcium hydride overnight, distilled, and stored in a glovebox.
Oleylamine (technical grade, 70%) was obtained from Sigma-
Aldrich, stirred over calcium hydride overnight, and fractionally
distilled to remove lower boiling fractions.68 The desired fraction was
collected and stored in a glovebox. Oxalyl chloride (≥99%) and
phosphorus trichloride (99.999%, trace metal basis) were obtained
from Sigma-Aldrich, degassed by the freeze−pump−thaw method,
distilled, and stored in a Strauss flask under argon outside of the
glovebox. Tri-n-octylphosphine oxide (99%) was obtained from Strem
Chemicals and recrystallized.
2-Chloro-1,3-dioctyl-1,3,2-diazaphospholidine. The abovemen-
tioned procedure was followed using N,N′-dioctylethylenediamine
(1.778 g, 6.300 mmol), triethylamine (3.788 g, 37.50 mmol),
phosphorus trichloride (0.57 mL, 6.6 mmol), and THF (50 mL).
The crude product is a pale yellow liquid used without further
purification. Yield 1.974 g (91%). 31P{1H} NMR (202 MHz, C6D6): δ
1
163.29. H NMR (500 MHz, C6D6): δ 0.90 (t, 6H), 1.22 (m, 20H),
1.50 (t, 4H), 2.73 (m, 2H), 2.89 (m, 4H), 3.04 (m, 2H). 13C{1H}
NMR (125 MHz, C6D6): δ 13.99 (s), 22.70 (s), 27.00 (s), 28.71 (d),
29.28 (s), 31.83 (s), 47.32 (d), 49.79 (d). Anal. Calcd for
C16H18N2PCl: C, 61.96; H, 10.98; N, 8.03. Found: C, 61.73; H,
11.24; N, 8.04. MS (ASAP) m/z: calcd for [C18H39N2PCl + H+],
349.2539; found, 349.2518.
2-Chloro-1,3-diphenyl-1,3,2-diazaphospholidine was synthesized
according to a published procedure on a 26 mmol scale.70 Purity is
1
confirmed via 31P, H, and 13C NMR spectroscopy. 31P{1H} NMR
1
(162 MHz, CDCl3): δ 137.28. H NMR (400 MHz, CDCl3): δ 4.00
(s, 4H), 7.08 (t, 2H), 7.20 (m, 4H), 7.39 (t, 4H). 13C{1H} NMR
(125 MHz, CDCl3): δ 31.63 (t), 117.13 (d), 122.23 (d), 129.59 (s).
2-Chloro-1,3-diisopropyl-1,3,2-diazaphospholidine was synthesized
according to a published procedure on a 26 mmol scale.71 Purity is
1
confirmed via 31P, H, and 13C NMR spectroscopy. 31P{1H} NMR
1
(202 MHz, CDCl3): δ 169.19. H NMR (500 MHz, CDCl3): δ 1.26
(d, 12H), 3.29 (d, 4H), 3.39 (m, 2H). 13C{1H} NMR (125 MHz,
CDCl3): δ 22.00 (s), 46.43 (d), 48.44 (d).
Instrumentation. UV−vis spectra were obtained using a
PerkinElmer Lambda 950 spectrophotometer equipped with
deuterium and halogen lamps. PL measurements were performed
using a Fluoromax 4 from Horiba Scientific. Powder X-ray diffraction
patterns were measured using a PANalytical X’Pert powder X-ray
diffractometer. Transmission electron microscopy and scanning
transmission electron microscopy images were collected using an
N,N-Diethyl-1,3-di(2-methylphenyl)-1,3,2-diazaphospholidin-2-
amine (1a). 2-Chloro-1,3-di(2-methylphenyl)-1,3,2-diazaphospholi-
dine (1.106 g, 3.600 mol), triethylamine (485 g, 4.80 mol), and
diethyl ether (20 mL) are loaded into a scintillation vial. Diethylamine
(0.415 mL, 4.80 mol) is added dropwise and allowed to stir for 20
min. Precipitation of a white solid is observed immediately upon
G
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
injection. The solid byproduct is filtered off. The resulting pale yellow
oil is used without further purification. Yield 1.087 g (88%). 31P{1H}
NMR (162 MHz, CDCl3): δ 101.10. 1H NMR (400 MHz, CDCl3): δ
0.73 (t, 6H), 2.48 (s, 6H), 3.02 (m, 4H), 3.39 (oct, 2H), 4.03 (sext,
2H), 7.04 (t, 2H), 7.20 (m, 4H), 7.30 (d, 2H). 13C{1H} NMR (125
MHz, CDCl3): δ 14.30 (s), 19.71 (d), 38.86 (d), 51.55 (d), 123.67
(d), 125.32 (d), 126.45 (s), 131.19 (s). Anal. Calcd for C20H28N3P:
C, 70.36; H, 8.27; N, 12.31. Found: C, 70.46; H, 8.39; N, 12.21. MS
(ASAP) m/z: calcd for [C20H28N3P + H+], 342.2099; found,
342.2104.
N-Octadecyl-1,3-di(2-methylphenyl)-1,3,2-diazaphospholidin-2-
amine (1b). 2-Chloro-1,3-di(2-methylphenyl)-1,3,2-diazaphospholi-
dine (1.106 g, 3.6 mmol), triethylamine (485 mg, 4.80 mmol), and
diethyl ether (15 mL) are loaded into a scintillation vial.
Octadecylamine (1.080 g, 4.000 mmol) is dissolved in diethyl ether
(5 mL) and added dropwise and allowed to stir for 20 min.
Precipitation of a white solid is observed immediately upon injection.
The solid byproduct is filtered off. Volatiles are removed from the
filtrate, and the resulting solid is recrystallized in THF at −40 °C to
yield a pale yellow solid. Yield 1.785 g (91%). 31P{1H} NMR (202
MHz, CDCl3): δ 90.60. 1H NMR (500 MHz, CDCl3): δ 0.91 (t, 3H),
1.28 (m, 34H), 2.35 (m, 1H), 2.47 (s, 6H), 2.90 (quin, 2H), 3.39 (m,
2H), 3.89 (m, 2H), 7.05 (t, 2H), 7.24 (m, 7H). 13C{1H} NMR (125
MHz, CDCl3): δ 14.17 (s), 19.61 (d), 22.73 (s), 26.87 (s), 29.39 (s)
29.61 (d) 29.72 (m), 31.96 (s), 33.06 (s), 42.00 (d), 51.00 (d),
123.85 (s), 125.21 (d), 126.52 (s), 131.15 (s). Anal. Calcd for
C34H56N3P: C, 75.93; H, 10.50; N, 7.81. Found: C, 75.89; H, 10.63;
N, 7.82. MS (ASAP) m/z: calcd for [C34H57N3P + H+], 538.4290;
found, 538.4298.
C, 75.70; H, 10.55; N, 8.22. MS (ASAP) m/z: calcd for [C32H52N3P +
H+], 510.3977; found, 510.3975.
N,N-Diethyl-1,3-diisopropyl-1,3,2-diazaphospholidin-2-amine
(3a). 2-Chloro-1,3-diisopropyl-1,3,2-diazaphospholidine (569 mg,
2.70 mmol), triethylamine (364 mg, 3.6 mmol), and diethyl ether
(30 mL) are loaded into a scintillation vial. Diethylamine (0.31 mL,
3.0 mol) is added dropwise and allowed to stir for 20 min.
Precipitation of a white solid is observed immediately upon injection.
The solid byproduct is filtered off. A second equivalent of
triethylamine (364 mg, 3.60 mmol) is added followed by the
dropwise addition of diethylamine (0.31 mL, 3.0 mmol) and allowed
to stir for 2 h. The white precipitate is filtered. The resulting deep
orange liquid is used without further purification. Yield 474 mg
1
(71%). 31P{1H} NMR (162 MHz, C6D6): δ 106.40. H NMR (400
MHz, C6D6): δ 1.02 (t, 6H), 1.16 (dd, 12H), 2.80 (m, 2H), 3.05 (m,
6H), 3.35 (m, 2H). 13C{1H} NMR (125 MHz, C6D6): δ 15.08 (d),
21.73 (dd), 39.09 (d), 44.33 (d), 46.94 (d). Anal. Calcd for
C12H28N3P: C, 58.75; H, 11.50; N, 17.31. Found: C, 59.13; H, 11.58;
N, 17.17. MS (ASAP) m/z: calcd for [C12H28N3P + H+], 246.2099;
found, 246.2099.
N,N-Diethyl-1,3-dibenzyl-1,3,2-diazaphospholidin-2-amine (4a).
2-Chloro-1,3-dibenzyl-1,3,2-diazaphospholidine (1.106 g, 3.600 mol),
triethylamine (485 mg, 4.80 mmol), and diethyl ether (20 mL) are
loaded into a scintillation vial. Diethylamine (0.41 mL, 4.0 mmol) is
added dropwise and allowed to stir for 20 min. Precipitation of a
white solid is observed immediately upon injection. The solid
byproduct is filtered off. The resulting orange liquid is used without
further purification. Yield 1.073 g (86%). 31P{1H} NMR (202 MHz,
CDCl3): δ 113.88. 1H NMR (500 MHz, CDCl3): δ 1.08 (t, 6H), 2.90
(m, 2H), 3.09 (quin, 4H), 3.17 (m, 2H), 3.94 (dd, 2H), 4.19 (dd,
2H), 7.26 (m, 2H), 7.35 (m, 8H). 13C{1H} NMR (125 MHz,
CDCl3): δ 15.18 (s), 43.93 (s), 44.58 (d), 48.78 (d), 127.57 (s),
128.12 (s), 128.63 (s). Anal. Calcd for C20H28N3P: C, 70.36; H, 8.27;
N, 12.31. Found: C, 70.27; H, 8.29; N, 12.07. MS (ASAP) m/z: calcd
for [C20H28N3P + H+], 342.2099; found, 342.2110.
N,N-Diethyl-1,3-dioctyl-1,3,2-diazaphospholidin-2-amine (5a).
2-Chloro-1,3-dioctyl-1,3,2-diazaphospholidine (1.106 g, 3.600 mol),
triethylamine (485 mg, 4.80 mmol), and diethyl ether (20 mL) are
loaded into a scintillation vial. Diethylamine (0.41 mL, 4.0 mmol) is
added dropwise and allowed to stir for 2 h. Precipitation of a white
solid is observed immediately upon injection. The solid byproduct is
filtered off. A second equivalent of triethylamine (485 mg, 4.80
mmol) is added followed by the dropwise addition of diethylamine
(0.41 mL, 4.0 mmol) and allowed to stir for 2 h. The white precipitate
is filtered. The resulting pale yellow liquid is used without further
purification. Yield 1.791 g (90%). 31P{1H} NMR (202 MHz, C6D6): δ
113.22. 1H NMR (500 MHz, C6D6): δ 0.90 (t, 6H), 1.03 (t, 6H), 1.29
(m, 20H), 1.59 (m, 4H), 2.92 (m, 6H), 3.07 (quin, 4H), 3.16 (m,
2H). 13C{1H} NMR (125 MHz, C6D6): δ 14.00 (s), 15.17 (s), 22.74
(s), 27.29 (s), 29.54 (d), 29.75 (d), 31.93 (s), 39.20 (d), 48.00 (d),
49.60 (d). Anal. Calcd for C20H28N3P: C, 68.52; H, 12.55; N, 10.90.
Found: C, 68.39; H, 12.51; N, 10.82. MS (ASAP) m/z: calcd for
[C22H49N3P + H+], 386.3664; found, 386.3666.
N-tert-Butyl-1,3-di(2-methylphenyl)-1,3,2-diazaphospholidin-2-
amine (1c). 2-Chloro-1,3-di(2-methylphenyl)-1,3,2-diazaphospholi-
dine (553 mg, 1.80 mmol), triethylamine (242 mg, 2.40 mmol), and
THF (10 mL) are loaded into a scintillation vial. tert-Butylamine
(0.207 mL, 2.00 mmol) is added dropwise and allowed to stir for 20
min. Precipitation of a white solid is observed immediately upon
injection. The solid byproduct is filtered off. Volatiles are removed
from the filtrate, and the resulting solid is recrystallized in THF at
−40 °C to yield a pale yellow solid. Yield: 620 mg (99%). 31P{1H}
1
NMR (202 MHz, CDCl3): δ 87.27. H NMR (500 MHz, CDCl3): δ
0.91 (t, 3H), 1.28 (m, 34H), 2.35 (m, 1H), 2.47 (s, 6H), 2.90 (quin,
2H), 3.39 (m, 2H), 3.89 (m, 2H), 7.05 (t, 2H), 7.24 (m, 7H).
13C{1H} NMR (125 MHz, CDCl3): δ 19.57 (d), 32.78 (d), 49.79 (d),
123.48 (d), 124.75 (d), 126.39 (s), 130.90 (s). Anal. Calcd for
C20H28N3P: C, 70.36; H, 8.27; N, 12.31. Found: C, 70.11; H, 8.38; N,
12.12. MS (ASAP) m/z: calcd for [C20H28N3P + H+], 342.2099;
found, 342.2094.
N,N-Diethyl-1,3-diphenyl-1,3,2-diazaphospholidin-2-amine (2a) is
prepared according to a published procedure on a 3.55 mmol scale.34
Purity is confirmed via 31P, 1H, and 13C NMR spectroscopy. 31P{1H}
1
NMR (202 MHz, CDCl3): δ 95.24. H NMR (500 MHz, CDCl3): δ
0.91 (t, 6H), 3.05 (m, 4H), 3.68 (m, 2H), 3.91 (m, 2H), 6.86 (t, 2H),
7.07 (d, 4H), 7.29 (m, 4H). 13C{1H} NMR (125 MHz, CDCl3): δ
14.25 (s), 39.72 (d), 46.28 (d), 115.05 (d), 118.55 (s), 129.00 (s).
N-Octadecyl-1,3-diphenyl-1,3,2-diazaphospholidin-2-amine
(2b). 2-Chloro-1,3-diphenyl-1,3,2-diazaphospholidine (1.006 g, 3.600
mol), triethylamine (485 mg, 4.80 mmol), and diethyl ether (15 mL)
are loaded into a scintillation vial. Octadecylamine (1.080 g, 4.000
mmol) is dissolved in diethyl ether (5 mL) and added dropwise. The
mixture is allowed to stir for 20 min. Precipitation of a white solid is
observed immediately upon injection. The solid byproduct is filtered
off. Volatiles are removed from the filtrate, and the resulting solid is
recrystallized in THF at −40 °C to yield a pale yellow solid. Yield
1.545 g (83%). 31P{1H} NMR (500 MHz, CDCl3): δ 84.77. 1H NMR
(202 MHz, CDCl3): δ 0.91 (t, 3H), 1.29 (m, 33H), 2.58 (m, 1H),
2.75 (quin, 2H), 3.69 (m, 2H), 3.83 (m, 2H), 6.89 (t, 2H), 7.17 (m,
4H), 7.30 (t, 4H). 13C{1H} NMR (125 MHz, CDCl3): δ 14.16 (s),
22.72 (s), 26.78 (s), 29.23 (s), 29.40 (s), 29.56 (d), 29.72 (m), 31.95
(s), 32.11 (s), 41.48 (s), 46.66 (d), 115.20 (d), 118.81 (s), 129.04
(s). Anal. Calcd for C32H52N3P: C, 75.40; H, 10.28; N, 8.24. Found:
Precursor Conversion Kinetics. A three-neck round-bottom
flask is loaded with indium(III) iodide (112 mg, 0.230 mmol),
zinc(II) chloride (153 mg, 1.13 mmol), and oleylamine (3.659 g,
4.500 mL, 13.70 mmol) and outfitted with a glass adapter for a
temperature probe. The desired phosphorus precursor (0.9 mol of
phosphorus) and trioctylamine (405 mg, 0.500 mL, 1.10 mmol) or
tetraethylene glycol dimethyl ether (in the case of 2a, 504.5 mg,
0.5000 mL, 1.100 mmol) are loaded into a 4 mL vial. Both vessels are
transferred to a Schlenk line and placed under argon. The three-neck
round-bottom flask was brought to the desired temperature, and the 4
mL vial was brought to 100 °C (in some cases to dissolve the
phosphorus precursor). The phosphorus precursor solution was
quickly injected into the mixture of metal salts and allowed to react
for the appropriate amount of time. Quantitative aliquots are taken
sequentially throughout the course of the reaction for analysis by
1
UV−vis and NMR spectroscopy (31P and H).
H
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Aliquots (0.1 mL) for UV−vis absorption spectroscopy measure-
ments are injected into a preweighed vial in air and subsequently
diluted with a known amount of toluene (5 mL). Aliquots (0.1 mL),
for 1H and 31P NMR analyses, are injected into a septum-sealed NMR
tube containing a stock solution (30 mM) of tri-n-octylphosphine
oxide in C6D6. Upon injection, the NMR tube is placed into a
CO2(s)−acetone bath to prevent further transamination. 31P NMR
Authors
Brandon M. McMurtry − Department of Chemistry, Columbia
University, New York, New York 10027, United States;
orcid.org/0000-0002-3624-941X
Kevin Qian − Department of Chemistry, Columbia University,
New York, New York 10027, United States
Joseph K. Teglasi − Department of Chemistry, Columbia
University, New York, New York 10027, United States
Anindya K. Swarnakar − Department of Chemistry, Columbia
University, New York, New York 10027, United States
Jonathan De Roo − Department of Chemistry, Columbia
University, New York, New York 10027, United States;
Department of Chemistry, University of Basel, Basel 4058,
Switzerland; orcid.org/0000-0002-1264-9312
1
spectra are collected without the decoupling of H nuclei and with
delay times (7 s) that are 5× longer than the longest T1 relaxation
time (1.4 s) to assure that the measurements are quantitative. Exact
concentrations of reaction species are determined by comparing
relative integrations of their 31P NMR spectroscopy signals to that of
1
the tri-n-octylphosphine oxide-C6D6 stock solution. H NMR spectra
are also collected.
Determination of Nanocrystal Size and Yield. The average
nanocrystal size in each aliquot is calculated from the peak wavelength
of the first excitonic transition using a previously reported InP
nanocrystal sizing curve.28 The tetrahedral shape model was chosen
for convenience (a new curve relating both size/shape to the excitonic
transition energy was not attempted). This approach intentionally
underestimates the volume of large nanocrystals; using the sizing
curve for nanocrystals with a spherical shape published by the same
authors, instead, would magnify the conclusion that nucleation is
continuous and growth is size dependent. Thus, our size estimates are
conservative and intended to increase the confidence in our
mechanistic model (see the Supporting Information for further
discussion of error analysis).
The concentration of [InP]i is determined from the absorbance at λ
= 413 nm using a previously reported extinction coefficient (see the
Supporting Information for details).27 The concentration of nano-
particles formed is extracted according to previously described
methods.27
Dependence of the Reaction Rate on Diamine Concen-
tration. A three-neck round-bottom flask is loaded with indium
iodide (112 mg, 0.230 mmol), zinc chloride (153 mg, 1.13 mmol), the
desired diamine (0.45 mmol), and oleylamine (3.659 g, 4.500 mL,
13.70 mmol) and outfitted with a glass adapter for a temperature
probe. Tris(dimethylamino)phosphine (147 mg, 0.900 mmol) and
trioctylamine (405 mg, 0.500 mL, 1.14 mmol) are loaded into a
separate 4 mL vial. The remaining procedures are identical to those
discussed above.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.chemmater.0c01561
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
under award number 1710352. We would like to acknowledge
Columbia University’s Nanoscience Initiative for use of its
shared facilities and Amirali Zangiabadi for assistance in
collecting transmission electron microscopy and scanning
transmission electron microcopy images. Powder X-ray
diffraction patterns were collected at Columbia University’s
Shared Materials Characterization Laboratory. We would like
to acknowledge Leslie S. Hamachi and Michael L. Steigerwald
for helpful discussion.
REFERENCES
■
(1) Cao, F.; Wang, S.; Wang, F.; Wu, Q.; Zhao, D.; Yang, X. A
Layer-by-Layer Growth Strategy for Large-Size InP/ZnSe/ZnS Core−
Shell Quantum Dots Enabling High-Efficiency Light-Emitting Diodes.
Chem. Mater. 2018, 30, 8002−8007.
Purification of InP Nanocrystals. All purification steps were
performed in air. After returning to room temperature, the reaction
mixture is precipitated using an excess of acetone (∼35 mL) and
centrifuged. The supernatant is disposed of, and the colored residue is
redispersed in toluene (∼3 mL) and subsequently precipitated with
acetone (∼30 mL). The nanocrystals are washed an additional three
times with toluene and acetone.
(2) Dupont, D.; Tessier, M. D.; Smet, P. F.; Hens, Z. Indium
Phosphide-Based Quantum Dots with Shell-Enhanced Absorption for
Luminescent Down-Conversion. Adv. Mater. 2017, 29, 1700686.
(3) Won, Y.-H.; Cho, O.; Kim, T.; Chung, D.-Y.; Kim, T.; Chung,
H.; Jang, H.; Lee, J.; Kim, D.; Jang, E. Highly Efficient and Stable
InP/ZnSe/ZnS Quantum Dot Light-Emitting Diodes. Nature 2019,
575, 634−638.
(4) Choi, H. S.; Kim, Y.; Park, J. C.; Oh, M. H.; Jeon, D. Y.; Nam, Y.
S. Highly Luminescent, Off-Stoichiometric CuX in YS2/ZnS
Quantum Dots for near-Infrared Fluorescence Bio-Imaging. RSC
Adv. 2015, 5, 43449−43455.
(5) Gao, X.; Cui, Y.; Levenson, R. M.; Chung, L. W. K.; Nie, S. In
Vivo Cancer Targeting and Imaging with Semiconductor Quantum
Dots. Nat. Biotechnol. 2004, 22, 969−976.
(6) Tamang, S.; Lincheneau, C.; Hermans, Y.; Jeong, S.; Reiss, P.
Chemistry of InP Nanocrystal Syntheses. Chem. Mater. 2016, 28,
2491−2506.
(7) Chen, O.; Zhao, J.; Chauhan, V. P.; Cui, J.; Wong, C.; Harris, D.
K.; Wei, H.; Han, H.-S.; Fukumura, D.; Jain, R. K.; Bawendi, M. G.
Compact High-Quality CdSe−CdS Core−Shell Nanocrystals with
Narrow Emission Linewidths and Suppressed Blinking. Nat. Mater.
2013, 12, 445−451.
(8) Ramasamy, P.; Kim, N.; Kang, Y.-S.; Ramirez, O.; Lee, J.-S.
Tunable, Bright, and Narrow-Band Luminescence from Colloidal
Indium Phosphide Quantum Dots. Chem. Mater. 2017, 29, 6893−
6899.
ZnSe Shelling of InP Nanocrystals. InP nanocrystals are shelled
with ZnSe according to a previously reported procedure.27
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.chemmater.0c01561.
■
sı
Supplementary figures (S1−S14), discussions of yield
variability, calculation of nanocrystal concentration and
nucleation fraction, Q0 fits, traces associated with
1
reaction aliquots, and 31P, H, and 13C NMR spectra
for all newly synthesized compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Jonathan S. Owen − Department of Chemistry, Columbia
University, New York, New York 10027, United States;
orcid.org/0000-0001-5502-3267; Email: jso2115@
columbia.edu
(9) Ramasamy, P.; Ko, K.-J.; Kang, J.-W.; Lee, J.-S. Two-Step “Seed-
Mediated” Synthetic Approach to Colloidal Indium Phosphide
I
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Quantum Dots with High-Purity Photo- and Electroluminescence.
Chem. Mater. 2018, 30, 3643−3647.
(10) Li, Y.; Hou, X.; Dai, X.; Yao, Z.; Lv, L.; Jin, Y.; Peng, X.
Stoichiometry-Controlled InP-Based Quantum Dots: Synthesis,
Photoluminescence, and Electroluminescence. J. Am. Chem. Soc.
2019, 141, 6448−6452.
Colloidal Quantum Dots in Tetrahedral Shape. Angew. Chem., Int. Ed.
2016, 55, 3714−3718.
(29) Buffard, A.; Dreyfuss, S.; Nadal, B.; Heuclin, H.; Xu, X.;
́
Patriarche, G.; Mezailles, N.; Dubertret, B. Mechanistic Insight and
Optimization of InP Nanocrystals Synthesized with Aminophos-
phines. Chem. Mater. 2016, 28, 5925−5934.
(11) Xu, Z.; Li, Y.; Li, J.; Pu, C.; Zhou, J.; Lv, L.; Peng, X. Formation
of Size-Tunable and Nearly Monodisperse InP Nanocrystals:
Chemical Reactions and Controlled Synthesis. Chem. Mater. 2019,
31, 5331−5341.
́
(30) Laufersky, G.; Bradley, S.; Frecaut, E.; Lein, M.; Nann, T.
Unraveling Aminophosphine Redox Mechanisms for Glovebox-Free
InP Quantum Dot Syntheses. Nanoscale 2018, 10, 8752−8762.
(31) Bent, E. G.; Haltiwanger, R. C.; Norman, A. D. Intermediates in
the 1,2-Diaminobenzene/Tris(diethylamino)phosphine Transamina-
tion Reaction. Inorg. Chem. 1990, 29, 4310−4318.
̌
(12) Abe, S.; Capek, R. K.; De Geyter, B.; Hens, Z. Tuning the
Postfocused Size of Colloidal Nanocrystals by the Reaction Rate:
From Theory to Application. ACS Nano 2012, 6, 42−53.
(13) Campos, M. P.; Hendricks, M. P.; Beecher, A. N.; Walravens,
W.; Swain, R. A.; Cleveland, G. T.; Hens, Z.; Sfeir, M. Y.; Owen, J. S.
A Library of Selenourea Precursors to PbSe Nanocrystals with Size
Distributions near the Homogeneous Limit. J. Am. Chem. Soc. 2017,
139, 2296−2305.
(14) Hamachi, L. S.; Jen-La Plante, I.; Coryell, A. C.; De Roo, J.;
Owen, J. S. Kinetic Control over CdS Nanocrystal Nucleation Using a
Library of Thiocarbonates, Thiocarbamates, and Thioureas. Chem.
Mater. 2017, 29, 8711−8719.
(15) Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La
Plante, I.; Owen, J. S. A Tunable Library of Substituted Thiourea
Precursors to Metal Sulfide Nanocrystals. Science 2015, 348, 1226−
1230.
(16) LaMer, V. K.; Dinegar, R. H. Theory, Production and
Mechanism of Formation of Monodispersed Hydrosols. J. Am.
Chem. Soc. 1950, 72, 4847−4854.
(17) Owen, J. S.; Chan, E. M.; Liu, H.; Alivisatos, A. P. Precursor
Conversion Kinetics and the Nucleation of Cadmium Selenide
Nanocrystals. J. Am. Chem. Soc. 2010, 132, 18206−18213.
(18) Gary, D. C.; Cossairt, B. M. Role of Acid in Precursor
Conversion During InP Quantum Dot Synthesis. Chem. Mater. 2013,
25, 2463−2469.
(19) Gary, D. C.; Glassy, B. A.; Cossairt, B. M. Investigation of
Indium Phosphide Quantum Dot Nucleation and Growth Utilizing
Triarylsilylphosphine Precursors. Chem. Mater. 2014, 26, 1734−1744.
(20) Gary, D. C.; Terban, M. W.; Billinge, S. J. L.; Cossairt, B. M.
Two-Step Nucleation and Growth of InP Quantum Dots via Magic-
Sized Cluster Intermediates. Chem. Mater. 2015, 27, 1432−1441.
(21) Koh, S.; Eom, T.; Kim, W. D.; Lee, K.; Lee, D.; Lee, Y. K.; Kim,
H.; Bae, W. K.; Lee, D. C. Zinc−Phosphorus Complex Working as an
Atomic Valve for Colloidal Growth of Monodisperse Indium
Phosphide Quantum Dots. Chem. Mater. 2017, 29, 6346−6355.
(22) Tessier, M. D.; De Nolf, K.; Dupont, D.; Sinnaeve, D.; De Roo,
J.; Hens, Z. Aminophosphines: A Double Role in the Synthesis of
Colloidal Indium Phosphide Quantum Dots. J. Am. Chem. Soc. 2016,
138, 5923−5929.
(23) Xie, L.; Harris, D. K.; Bawendi, M. G.; Jensen, K. F. Effect of
Trace Water on the Growth of Indium Phosphide Quantum Dots.
Chem. Mater. 2015, 27, 5058−5063.
(24) Franke, D.; Harris, D. K.; Xie, L.; Jensen, K. F.; Bawendi, M. G.
The Unexpected Influence of Precursor Conversion Rate in the
Synthesis of III-V Quantum Dots. Angew. Chem., Int. Ed. 2015, 54,
14299−14303.
(25) Harris, D. K.; Bawendi, M. G. Improved Precursor Chemistry
for the Synthesis of III-V Quantum Dots. J. Am. Chem. Soc. 2012, 134,
20211−20213.
(26) Joung, S.; Yoon, S.; Han, C.-S.; Kim, Y.; Jeong, S. Facile
Synthesis of Uniform Large-Sized InP Nanocrystal Quantum Dots
Using Tris(tert-butyldimethylsilyl) Phosphine. Nanoscale Res. Lett.
2012, 7, 93−100.
(27) Tessier, M. D.; Dupont, D.; De Nolf, K.; De Roo, J.; Hens, Z.
Economic and Size-Tunable Synthesis of InP/ZnE (E= S, Se)
Colloidal Quantum Dots. Chem. Mater. 2015, 27, 4893−4898.
(28) Kim, K.; Yoo, D.; Choi, H.; Tamang, S.; Ko, J.-H.; Kim, S.;
Kim, Y.-H.; Jeong, S. Halide-Amine Co-Passivated Indium Phosphide
(32) Hill, T. G.; Haltiwanger, R. C.; Thompson, M. L.; Katz, S. A.;
Norman, A. D. Cis/Trans Isomerization and Conformational
Properties of 2, 4-Bis (primary amino)-1, 3, 2, 4-diazadiphospheti-
dines. Inorg. Chem. 1994, 33, 1770−1777.
(33) Abrams, M. B.; Scott, B. L.; Baker, R. T. Sterically Tunable
Phosphenium Cations: Synthesis and Characterization of Bis-
(arylamino)phosphenium Ions, Phosphinophosphenium Adducts,
and the First Well-Defined Rhodium Phosphenium Complexes.
Organometallics 2000, 19, 4944−4956.
(34) Huang, H.; Kang, J. Y. Mitsunobu Reaction Using Basic Amines
as Pronucleophiles. J. Org. Chem. 2017, 82, 6604−6614.
(35) Wucher, P.; Roesle, P.; Falivene, L.; Cavallo, L.; Caporaso, L.;
̈
Gottker-Schnetmann, I.; Mecking, S. Controlled Acrylate Insertion
Regioselectivity in Diazaphospholidine-Sulfonato Palladium(II) Com-
plexes. Organometallics 2012, 31, 8505−8515.
́
(36) Nifantev, E. E.; Sorokina, S. F.; Vorobeva, L. A.; Bekker, A. R.,
1,3-Dibenzyl-1,3,2-diazaphospholidines and -Hexahydro-1,3,2-di- aza-
phosphorines. Russ. J. Gen. Chem. 1985, 55, 269−275. DOI: 10.1002/
chin.198540252
(37) Friedfeld, M. R.; Johnson, D. A.; Cossairt, B. M. Conversion of
InP Clusters to Quantum Dots. Inorg. Chem. 2018, 58, 803−810.
(38) The hwhm values reported here are ∼30% broader than the
original report, which appears to be a consequence of the different
methods to analyze the spectra. The previous study measured the
fwhm by calculating the first derivative of the absorbance spectrum
and determining a maximum/minimum to extract the inflection point
in the lowest energy optical transition. This analysis does not
accurately capture the behavior of more complex peak shapes nor the
contribution of overlapping transitions on the higher energy side of
the excitonic features. Using this method to analyze the spectra
reported here gave values equal to or smaller than the original report.
(39) Owen, J.; Brus, L. Chemical Synthesis and Luminescence
Applications of Colloidal Semiconductor Quantum Dots. J. Am.
Chem. Soc. 2017, 139, 10939−10943.
(40) Vikram, A.; Zahid, A.; Bhargava, S. S.; Keating, L. P.; Sutrisno,
A.; Khare, A.; Trefonas, P.; Shim, M.; Kenis, P. J.A. Mechanistic
Insight into Size-focused Growth of Indium Phosphide Nanocrystals
in the Presence of Trace Water. Chem. Mater. 2020, 32, 3577−3584.
́
(41) Xie, L.; Shen, Y.; Franke, D.; Sebastian, V.; Bawendi, M. G.;
Jensen, K. F. Characterization of Indium Phosphide Quantum Dot
Growth Intermediates Using MALDI-TOF Mass Spectrometry. J. Am.
Chem. Soc. 2016, 138, 13469−13472.
(42) Sugimoto, T.; Shiba, F. Spontaneous Nucleation of
Monodisperse Silver Halide Particles from Homogeneous Gelatin
Solution II: Silver Bromide. Colloids Surf., A 2000, 164, 205−215.
(43) Sugimoto, T.; Shiba, F.; Sekiguchi, T.; Itoh, H. Spontaneous
Nucleation of Monodisperse Silver Halide Particles from Homoge-
neous Gelatin Solution I: Silver Chloride. Colloids Surf., A 2000, 164,
183−203.
(44) Hamachi, L. S.; Yang, H.; Jen-La Plante, I.; Saenz, N.; Qian, K.;
Campos, M. P.; Cleveland, G. T.; Rreza, I.; Oza, A.; Walravens, W.;
Chan, E. M.; Hens, Z.; Crowther, A. C.; Owen, J. S. Precursor
Reaction Kinetics Control Compositional Grading and Size of
CdSe1‑XSX Nanocrystal Heterostructures. Chem. Sci. 2019, 10,
6539−6552.
J
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
(45) Chu, D. B. K.; Owen, J. S.; Peters, B. Nucleation and Growth
Kinetics from Lamer Burst Data. J. Phys. Chem. A 2017, 121, 7511−
7517.
(46) Rempel, J. Y.; Bawendi, M. G.; Jensen, K. F. Insights into the
Kinetics of Semiconductor Nanocrystal Nucleation and Growth. J.
Am. Chem. Soc. 2009, 131, 4479−4489.
Indium Phosphide Nanocluster. J. Am. Chem. Soc. 2016, 138, 1510−
1513.
́
(66) Abecassis, B.; Bouet, C.; Garnero, C.; Constantin, D.; Lequeux,
N.; Ithurria, S.; Dubertret, B.; Pauw, B. R.; Pontoni, D. Real-Time In
Situ Probing of High-Temperature Quantum Dots Solution Synthesis.
Nano Lett. 2015, 15, 2620−2626.
(67) Jasieniak, J.; Smith, L.; van Embden, J.; Mulvaney, P.; Califano,
M. Re-Examination of the Size-Dependent Absorption Properties of
CdSe Quantum Dots. J. Phys. Chem. C 2009, 113, 19468−19474.
(68) Baranov, D.; Lynch, M. J.; Curtis, A. C.; Carollo, A. R.;
Douglass, C. R.; Mateo-Tejada, A. M.; Jonas, D. M. Purification of
Oleylamine for Materials Synthesis and Spectroscopic Diagnostics for
Trans Isomers. Chem. Mater. 2019, 31, 1223−1230.
(47) Kim, Y.; Ham, S.; Jang, H.; Min, J. H.; Chung, H.; Lee, J.; Kim,
D.; Jang, E. Bright and Uniform Green Light Emitting InP/ZnSe/ZnS
Quantum Dots for Wide Color Gamut Displays. ACS Appl. Nano
Mater. 2019, 2, 1496−1504.
(48) Liu, Z.; Kumbhar, A.; Xu, D.; Zhang, J.; Sun, Z.; Fang, J.
Coreduction Colloidal Synthesis of III-V Nanocrystals: The Case of
InP. Angew. Chem., Int. Ed. 2008, 47, 3540−3542.
(69) Block, M. A. B.; Hecht, S. Alternating (Squaraine−Receptor)
Sensory Polymers: Modular One-Pot Synthesis and Signal Trans-
duction Via Conformationally Controlled Exciton Interaction.
Macromolecules 2004, 37, 4761−4769.
(49) Shen, W.; Tang, H.; Yang, X.; Cao, Z.; Cheng, T.; Wang, X.;
Tan, Z.; You, J.; Deng, Z. Synthesis of Highly Fluorescent InP/ZnS
Small-Core/Thick-Shell Tetrahedral-Shaped Quantum Dots for Blue
Light-Emitting Diodes. J. Mater. Chem. C 2017, 5, 8243−8249.
(50) Baquero, E. A.; Virieux, H.; Swain, R. A.; Gillet, A.; Cros-
Gagneux, A.; Coppel, Y.; Chaudret, B.; Nayral, C.; Delpech, F.
Synthesis of Oxide-Free InP Quantum Dots: Surface Control and H2-
Assisted Growth. Chem. Mater. 2017, 29, 9623−9627.
(70) Marre, M. R.; Sanchez, M.; Brazier, J. F.; Wolf, R.; Bellan, J. La
̀
Liaison Phosphazene Dans Quelques Nouveaux Iminophospholanes.
́
́
Diversite De Comportement Reactionnel. Can. J. Chem. 1982, 60,
456−468.
(71) Krysiak, J.; Lyon, C.; Baceiredo, A.; Gornitzka, H.; Mikolajczyk,
M.; Bertrand, G. Stable Optically Pure Phosphino(silyl)carbenes:
Reagents for Highly Enantioselective Cyclopropanation Reactions.
Chem.Eur. J. 2004, 10, 1982−1986.
(51) Yin, Y.; Alivisatos, A. P. Colloidal Nanocrystal Synthesis and
the Organic-Inorganic Interface. Nature 2005, 437, 664−670.
̈
(52) Whitehead, C. B.; Ozkar, S.; Finke, R. G. LaMer’s 1950 Model
for Particle Formation of Instantaneous Nucleation and Diffusion-
Controlled Growth: A Historical Look at the Model’s Origins,
Assumptions, Equations, and Underlying Sulfur Sol Formation
Kinetics Data. Chem. Mater. 2019, 31, 7116−7132.
(53) Sugimoto, T. Underlying Mechanisms in Size Control of
Uniform Nanoparticles. J. Colloid Interface Sci. 2007, 309, 106−118.
(54) Peng, X.; Wickham, J.; Alivisatos, A. P. Kinetics of II-VI and III-
V Colloidal Semiconductor Nanocrystal Growth: “Focusing” of Size
Distributions. J. Am. Chem. Soc. 1998, 120, 5343−5344.
(55) Reiss, H. The Growth of Uniform Colloidal Dispersions. J.
Chem. Phys. 1951, 19, 482−487.
(56) Cussler, E. L. Diffusion: Mass Transfer in Fluid Systems;
Cambridge University Press: Cambridge, 2009.
̈
(57) Handwerk, D. R.; Shipman, P. D.; Whitehead, C. B.; Ozkar, S.;
Finke, R. G. Mechanism-Enabled Population Balance Modeling of
Particle Formation en Route to Particle Average Size and Size
Distribution Understanding and Control. J. Am. Chem. Soc. 2019, 141,
15827−15839.
(58) Kim, J. Y.; Steeves, A. H.; Kulik, H. J. Harnessing Organic
Ligand Libraries for First-Principles Inorganic Discovery: Indium
Phosphide Quantum Dot Precursor Design Strategies. Chem. Mater.
2017, 29, 3632−3643.
(59) Zhao, Q.; Kulik, H. J. Electronic Structure Origins of Surface-
Dependent Growth in III−V Quantum Dots. Chem. Mater. 2018, 30,
7154−7165.
(60) Beecher, A. N.; Yang, X.; Palmer, J. H.; LaGrassa, A. L.; Juhas,
P.; Billinge, S. J. L.; Owen, J. S. Atomic Structures and Gram Scale
Synthesis of Three Tetrahedral Quantum Dots. J. Am. Chem. Soc.
2014, 136, 10645−10653.
(61) Bullen, C. R.; Mulvaney, P. Nucleation and Growth Kinetics of
CdSe Nanocrystals in Octadecene. Nano Lett. 2004, 4, 2303−2307.
(62) van Embden, J.; Mulvaney, P. Nucleation and Growth of CdSe
Nanocrystals in a Binary Ligand System. Langmuir 2005, 21, 10226−
10233.
(63) Yordanov, G. G.; Yoshimura, H.; Dushkin, C. D. Fine Control
of the Growth and Optical Properties of CdSe Quantum Dots by
Varying the Amount of Stearic Acid in a Liquid Paraffin Matrix.
Colloids Surf., A 2008, 322, 177−182.
(64) Yu, W. W.; Peng, X. Formation of High-Quality CdS and Other
II-VI Semiconductor Nanocrystals in Noncoordinating Solvents:
Tunable Reactivity of Monomers. Angew. Chem., Int. Ed. 2002, 41,
2368−2371.
(65) Gary, D. C.; Flowers, S. E.; Kaminsky, W.; Petrone, A.; Li, X.;
Cossairt, B. M. Single-Crystal and Electronic Structure of a 1.3 nm
K
https://dx.doi.org/10.1021/acs.chemmater.0c01561
Chem. Mater. XXXX, XXX, XXX−XXX