Ternary synthesis of colloidal Zn3P2 quantum dots

Article abstract of DOI:10.1039/c4cc08068h

The synthesis and characterization of crystalline colloidal zinc phosphide quantum dots with observable excitonic transitions ranging between 424-535 nm (2.3-2.9 eV) are reported. A ternary combination of ZnEt₂, Zn(O₂CR)₂, and P(SiMe₃)₃, forms a pentanuclear zinc cluster on mixing followed by conversion to (Et₂Zn)P(ZnO₂CR)₂(SiMe₃) in a rate-determining step prior to quantum dot formation. This journal is

Full text of DOI:10.1039/c4cc08068h

ChemComm
View Article Online
View Journal
COMMUNICATION
Ternary synthesis of colloidal Zn3P2 quantum dots†
Benjamin A. Glassy and Brandi M. Cossairt*
Cite this:DOI: 10.1039/c4cc08068h
Received 13th October 2014,
Accepted 11th November 2014
DOI: 10.1039/c4cc08068h
www.rsc.org/chemcomm
The synthesis and characterization of crystalline colloidal zinc phos- large-scale synthesis, solution processable device fabrication, and
phide quantum dots with observable excitonic transitions ranging control over the optical and electronic band gap, a colloidal quantum
between 424–535 nm (2.3–2.9 eV) are reported. A ternary combination dot (QD) synthesis is desirable.12,13
of ZnEt2, Zn(O2CR)2, and P(SiMe3)3, forms a pentanuclear zinc cluster on
To date, there have been seven accounts of colloidal syntheses of
mixing followed by conversion to (Et2Zn)P(ZnO2CR)2(SiMe3) in a rate- Zn3P2.14–20 The earliest study showed formation of zinc phosphide
determining step prior to quantum dot formation.
clusters with a distinct lowest energy excitonic transition (LEET) at
300 nm, indicative of relatively uniform crystalline particles, with an
Rising global energy demands and diminishing fossil fuel reserves approximated 1 nm diameter. However, larger particles could not be
necessitate innovation in alternative energy sources. Solar radiation is obtained by this method.14 Green and O’Brien describe a synthesis
promising given it is environmentally benign and abundant, but the that gives a broad excitonic feature below 500 nm, but analysis
high cost of solar cell fabrication and materials, particularly high showed a high degree of polydispersity and a mixture of crystalline
quality crystalline silicon, prevents wide-spread deployment into the and amorphous material.16 Recent investigations have yielded zinc
power grid.1,2 A recent cost-benefit analysis was performed on 23 phosphide with no clear excitonic features in the absorption spectra
semiconducting materials with electronic structures suitable for and conflicting powder diffraction data, suggesting isolation of
photovoltaic applications to determine possible alternatives to crystal- different crystal phases of zinc phosphide with a broad size distribu-
line silicon. The study used the earth abundance of the semiconducting tion and lack of quantum confinement.18,19 These inconsistencies
materials, raw material extraction costs, and theoretical efficiency highlight the need for new, robust, and reproducible synthetic
of solar energy conversion to determine their energy production methods to obtain colloidal zinc phosphide quantum dots that are
potential. Of the 23 materials examined, nine were found to have the phase-pure, absorb in the visible region of the electromagnetic
capacity to meet or exceed annual worldwide electricity consumption spectrum, and have low polydispersity.21
with a significant cost-reduction over crystalline silicon.3 Among the
most promising of these materials was zinc phosphide.
The reported synthetic approaches to colloidal zinc phosphide
utilize a zinc precursor (zinc perchlorate,14 zinc alkoxide,15 zinc
Zinc phosphide (Zn3P2) is ideally positioned as a next generation nanoparticles,17 dialkylzinc,16,18–20 zinc stearate,18 or zinc silylamide18),
photovoltaic material as it has a direct band gap of 1.50 eV, which a phosphorus precursor (phosphine gas,14,18 tristrimethylsilyl-
allows it to absorb a high percentage of the solar spectrum.4 phosphine,15,18,20 ditertbutylphosphine,16 white phosphorus,17 or
Zinc phosphide also has a high extinction coefficient,5 and a long trioctylphosphine19,20), and excess ligands such as hexameta-
minority-carrier diffusion length (5–10 mm); properties that are phosphate,14 pyridine,15,16 trioctylphosphine,16–20 or stearate.18 This
desirable in efficient photovoltaic materials.6 High-quality epitaxially communication discusses a novel synthesis that relies on the
grown Zn3P2 was first reported in 1992,7 and heterojunction opto- in situ generation of a highly reactive pentanuclear zinc cluster
electronic devices made with zinc phosphide have been fabricated, from ZnEt2 and Zn(O2CR)2 (R = (CH2)12CH3 [MA] or C(CH2)7CHQ
demonstrating the viability of industrial scale development of zinc CH(CH2)7CH3 [OA]) followed by its in situ reaction with P(SiMe3)3
phosphide semiconductor technologies.8–11 However, for cost-effective to generate (Et2Zn)P(ZnO2CR)2(SiMe3) prior to conversion to zinc
phosphide. The molecular precursor conversion is probed using
1H and 31P{1H} NMR spectroscopy. Through control of the
reaction temperature during nanocrystal growth, crystalline
University of Washington, Department of Chemistry, Box 351700, Bagley Hall,
Seattle, WA 98195-1700, USA. E-mail: cossairt@chem.washington.edu
† Electronic supplementary information (ESI) available: The additional figures
particles with clear excitonic transitions in the visible are
and information mentioned within this communication as well as details for all
obtained. Detailed characterization of the structure and compo-
sition of the resulting zinc-rich zinc phosphide nanocrystals by
the experiments discussed can be found in the supplementary information
document accompanying this communication. See DOI: 10.1039/c4cc08068h
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.

View Article Online
Communication
ChemComm
maxima suggests luminescence from mid-gap trap states.18 The
PL quantum yield of the synthesized Zn3P2 was 1%, typical of
related metal phosphide systems.18,25,26 The effect of concen-
tration on the final particle size was assessed (Fig. S4, ESI†) and
determined to be consistent with the predictions of classical
nucleation theory.27,28
NMR experiments were conducted to elucidate the mechanism of
precursor conversion and to understand the synergistic role of using
two different zinc precursors. When P(SiMe3)3 and ZnEt2 were
Scheme 1 Synthesis of zinc-rich zinc phosphide quantum dots.
HRTEM, TEM, ICP-OES, TGA, NMR, and powder XRD are
presented.
1
combined in C6D6 in a J-Young tube, the H and 31P{1H} NMR
spectra reveal no change over the course of 24 hours at room
temperature. There is also no change at 80 1C over the course of
one hour (Fig. S5, ESI†). In C6D6, P(SiMe3)3 and Zn(O2CR)2 react over
the course of hours (half-life of 5 hours for R = CH3) at room
temperature and do not form any resolvable phosphorus containing
intermediates (Fig. S6, ESI†). Zinc carboxylates and diethyl zinc,
however, are known to react to form [Zn5(O2CR)6(Et)4] clusters in
non-coordinating solvent, and we have identified this species in our
1H NMR data (Fig. S7, ESI†).29 When P(SiMe3)3 is combined with this
pre-formed zinc cluster, it reacts on the order of minutes (half-life of
o5 minutes) to form a phosphorus-containing intermediate with a
chemical shift of À276 ppm (Fig. S8, ESI†).
Our strategy to avoid the difficulties in forming this covalent
material was to pre-form a cluster containing Zn–P bonds, with initial
focus on silylphosphine and alkylzinc reagents inspired by an unusual
Zn6P4 cluster observed by Arno Pfitzner and co-workers.22 Our best
results were obtained by injecting a mixture of ZnEt2 and P(SiMe3)3
into Zn(O2CR)2 at 100 1C followed by heating to the growth tempera-
ture (Scheme 1). UV-vis spectra of timed aliquots for a typical synthesis
(Fig. 1A) show the progression of the LEET as a function of tempera-
ture and time. Particles with a clear excitonic feature only formed
when utilizing this heat-up method and all three precursors.23
The position of the final LEET is dependent on the final growth
temperature achieved and ranged from 424 to 535 nm, with absorp-
tion onsets extending beyond 600 nm (Fig. S2, ESI†).24 Photolumi-
nescence (PL) was observed for all samples grown to temperatures
Z255 1C and was red shifted from the absorbance peak by at least
0.37 eV (Fig. 1B and Fig. S3, ESI†). The red shift in fluorescence and
the lack of correlation of the emission profile from the absorption
To probe the molecular transformation of all three precursors,
Zn(MA)3, P(SiMe3)3, and ZnEt2 were mixed in a J-Young tube at room
1
temperature in C6D6 and was monitored by 31P{1H} and H NMR
spectroscopy. The time resolved 31P{1H} NMR spectra show the
disappearance of the P(SiMe3)3 singlet at À252 ppm along with the
concomitant appearance of a new singlet at À276 ppm, which suggests
that P(SiMe3)3 is converting to a new molecular species over the period
of hours at 22 1C (Fig. S9, ESI†). The 1H NMR spectra are difficult to
analyze due to the overlapping peaks from the myristate ligands,
however the complexity of the 1H NMR spectra can be significantly
reduced by replacing Zn(MA)2 with Zn(Ac)2 (Ac = O2CCH3).
When Zn(Ac)2, P(SiMe3)3, and ZnEt2 are combined in C6D6 in a
J-Young NMR tube at room temperature, the reaction mixture is
initially heterogeneous due to the insolubility of Zn(Ac)2 in C6D6, but
becomes homogenous over time. The time resolved 31P{1H} NMR
spectra show the formation and disappearance of a singlet at
À276 ppm, as observed when using Zn(MA)2 (Fig. 2A). This reaction
also has a corresponding resolvable 1H NMR spectrum which shows
the formation and disappearance of the proposed phosphorus-
containing intermediate in addition to the pentanuclear [Zn5(Ac)6(Et)4]
cluster and the silylacetate co-product (Fig. 2B). Based upon the
1
integration of a H NMR spectrum taken two hours after all three
precursors were mixed together (Fig. S10, ESI†), we propose that the
long lived phosphorus containing intermediate is (Me3Si)P(ZnAc)2
with a ZnEt2 group datively bound (Fig. 2C). The assignment of
P-bound ZnEt2 is based on the observation of the downfield shift
and broadening of free ZnEt2 concomitant with the appearance of
the phosphorus-containing intermediate in addition to its subse-
quent shift back and sharpening on loss of this intermediate. There
is literature precedent for related tetrahedrally-coordinated phos-
phorus species displaying 31P{1H} NMR signals at À318.5 and
À270.3 ppm.30,31 We were unable to isolate and purify the proposed
(Et2Zn)P(ZnAc)2(SiMe3) species due to its limited lifetime and
instability under reduced pressure. This intermediate was also
Fig. 1 (A) UV-vis spectra showing the temporal evolution of zinc phosphide
quantum dots formed using Scheme 1. (B) Absorbance and room tempera-
ture photoluminescence spectra of isolated zinc phosphide quantum dots.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
formed via reaction of Zn(O2CR)2 with ZnEt2. The rate of reaction
between [Zn5(O2CR)6(Et)4] with P(SiMe3)3 is greatly enhanced relative
to ZnEt2 or Zn(O2CR)2 alone, and this reaction leads to formation of
the long-lived (Et2Zn)P(ZnO2CR)2(SiMe3) intermediate. This species
is the key molecular precursor preceding nucleation and growth of
the reported Zn3P2 QDs.
Transmission electron microscopy (TEM) images were collected
to determine the average particle size of the Zn3P2 QDs. Quantum
dots grown to 315 1C (LEET at 535 nm) had an average diameter of
2.9 Æ 0.6 nm based upon measurements made on 300 individual
particles using ImageJ32 (Fig. 3, histogram found in Fig. S15, ESI†),33
while quantum dots grown to 255 1C (LEET at 424 nm) had an average
diameter of 2.6 Æ 0.5 nm (Fig. S16, ESI†). QDs grown to the two
temperatures had average particle diameters within 0.3 nm, however
the energy of the LEET from the UV-vis spectra differed by 0.6 eV,
indicative of a high degree of quantum confinement. Calculation of
the expected average particle size based on the energy of the LEET
using the Brus equation with a range of reported reduced electron
and hole masses predict this behavior, with a maximum difference in
Fig. 2 (A) 31P{1H} and (B) 1H NMR spectra showing time evolution of the
zinc phosphide precursor conversion reactions at 22 1C. (C) Proposed
scheme for the zinc phosphide precursor conversion reactions.
implicated as the species directly responsible for the formation of expected radius of 0.4 nm between the two samples.34 Using the same
QDs in a typical heat-up synthesis. This intermediate was the only range of electron and hole masses, the exciton Bohr radius of Zn3P2
molecular species containing phosphorus observed by 31P{1H} NMR was calculated to range between 3 and 7 nm.35 This is consistent with
in an aliquot taken 30 seconds post injection from a typical synthesis our assignment of the Zn3P2 prepared here as quantum confined
(Fig. S11, ESI†) and no phosphorus containing molecular species semiconductor nanocrystals. Particles formed without ZnEt2 had an
were observed by 31P{1H} NMR in an aliquot taken after Zn3P2 QD average diameter of 2.6 Æ 0.5 nm (Fig. S17, ESI†), however these
formation. These observations indicate the loss of this intermediate particles exhibit no excitonic features in the visible and the sample is
by 31P{1H} NMR directly correlates with the formation of QDs and is likely largely amorphous (vide infra). High resolution TEM (HRTEM
not a by-product of the reaction. To further support this hypothesis, or phase-contrast TEM) was used to confirm the crystallinity and
a QD synthesis was performed in a J-Young tube to monitor the lattice spacing of QDs grown to 315 1C with all three precursors. FEI
formation and disappearance of the observed molecular intermedi- TrueImage software was used to reconstruct the exit wavefunction
ate during QD formation. In this experiment Zn(MA)2, ZnEt2, and using a series of images taken at different focus points. Analysis of the
P(SiMe3)3 were mixed in 1-ODE in a J-Young tube and heated up in a particle shown in Fig. 3 (left) showed lattice plane spacings of 2.6 and
sand bath and 31P{1H} NMR spectra were taken after the NMR tube 1.6 Å which are consistent with the (301) and (051) planes of the
had been heated to various temperatures. This series of 31P{1H} NMR tetragonal a-Zn3P2 phase, respectively.36
spectra show the growth of the intermediate species at À276 ppm
Air-free X-ray diffraction (XRD) was performed by sealing a solid
followed by its gradual disappearance, suggesting that this species is sample of synthesized particles under a Kaptons film. The collected
active in QD formation and is the only long-lived phosphorous diffraction pattern (Fig. S18, ESI†) agrees favourably with the pattern
containing molecular precursor observable prior to QD formation for Zn3P2 (tetragonal a-Zn3P2; COD 1010287). Upon air oxidization,
(Fig. S12, ESI†).
there is a change in diffraction pattern indicating these particles are
Further evidence in support of this hypothesis resulted from air sensitive (Fig. S19, ESI†). An air-free pattern of particles synthe-
pre-mixing all three molecular precursors (Zn(MA)2, ZnEt2, and sized without ZnEt2 shows a very intense peak at 191 2y which could
P(SiMe3)3) in 3 grams of 1-ODE at room temperature until complete be indicative of poor crystallinity (Fig. S20, ESI†). Scherrer analysis
conversion to the molecular species with a 31P{1H} NMR resonance
at –276 ppm was observed (8 hours). This solution, containing the
pre-formed precursor, was injected into 5 grams of 1-ODE at 100 1C
and heated to 315 1C. The formation of Zn3P2 QDs was observed,
similar to a typical synthesis in which this species was not purposely
pre-formed (Fig. S13, ESI†). To corroborate our hypothesis that the
pentanuclear zinc cluster (Fig. 2C) forms first in a typical synthesis, a
QD reaction was performed in which there was a 15 minute delay
between injecting ZnEt2 and P(SiMe3)3 into the Zn(OA)2 solution
(Fig. S14, ESI†). The size and spectroscopic characteristics of the
particles formed in this experiment were within error of particles
formed when ZnEt2 and P(SiMe3)3 are introduced simultaneously.
Fig. 3 (left) High resolution TEM image of a single zinc phosphide
The results of these experiments are consistent with the postulate
that in the ternary synthesis, a pentanuclear zinc cluster is initially
quantum dot with the FFT as an inset and (right) low resolution TEM image
of zinc phosphide quantum dots.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.

View Article Online
Communication
ChemComm
was performed on the particles grown using all three precursors Nanotechnology Infrastructure Network (NNIN). We thank Scott
using both the peaks at 321 and 451 2y37 and yielded an average Braswell for his help with TEM and HRTEM data collection. This
particle diameter of 1.1 and 2.8 nm, the second of which is consistent research was supported by start-up funds from the University of
with the TEM data (Fig. S21, ESI†) for a 0.4 mmol P(SiMe3)3 Washington and the University of Washington Innovation Award.
reaction.38,39 The peak at 321 2y was analyzed assuming one reflec-
Notes and references
1 R. R. Lunt, T. P. Osedach, P. R. Brown, J. A. Rowehl and V. Bulovic,
tion was the sole cause of the intensity, which may not be the case.
Also, given the small particle size, the surface will have a larger
´
impact on lattice strain which would broaden the peak width.
Due to the similarity in the solubility of the zinc phosphide QDs and
zinc myristate, particles synthesized with this zinc reagent could not be
separated from excess zinc myristate. Zinc phosphide QDs were there-
fore prepared using more soluble zinc oleate for complete excess ligand
removal. 1H NMR data confirmed clean isolation of zinc-rich particles
capped only by oleate ligands (Fig. S22, ESI†). Particles grown to 255 1C
had a zinc to phosphorus ratio of 7.6 : 2 and particles grown to 315 1C
had a zinc to phosphorus ratio of 5 : 2 as determined by inductively
Adv. Mater., 2011, 23, 5712–5727.
2 A. E. Curtright, M. G. Morgan and D. W. Keith, Environ. Sci. Technol.,
2008, 42, 9031–9038.
3 C. Wadia, A. P. Alivisatos and D. M. Kammen, Environ. Sci. Technol.,
2009, 43, 2072–2077.
4 E. A. Fagen, J. Appl. Phys., 1979, 50, 6505–6515.
5 J. P. Bosco, S. B. Demers, G. M. Kimball, N. S. Lewis and H. A. Atwater,
J. Appl. Phys., 2012, 112, 093703.
6 N. C. Wyeth and A. Catalano, J. Appl. Phys., 1979, 50, 1403–1407.
7 T. Suda and K. Kakishita, J. Appl. Phys., 1992, 71, 3039–3041.
8 M. Bhushan and A. Catalano, Appl. Phys. Lett., 1981, 38, 39–41.
9 P. S. Nayar and A. Catalano, Appl. Phys. Lett., 1981, 39, 105–107.
coupled plasma atomic emission spectroscopy (ICP-OES). Thermo- 10 G. M. Kimball, N. S. Lewis and H. A. Atwater, Photovoltaic Specia-
lists Conference (PVSC), 2010 35th IEEE, 2010.
11 R. Yang, Y.-L. Chueh, J. R. Morber, R. Snyder, L.-J. Chou and
gravimetric analysis (TGA) was performed under nitrogen on particles
grown to 315 1C, showing a single weight loss transition centered near
Z. L. Wang, Nano Lett., 2006, 7, 269–275.
400 1C. Following TGA, 46% inorganic mass remained, consistent with 12 C. B. Murray, D. J. Norris and M. G. Bawendi, J. Am. Chem. Soc., 1993,
115, 8706–8715.
a Zn(O2CR)2-capped zinc phosphide formulation with less than a
`
13 L. Li, M. Protiere and P. Reiss, Chem. Mater., 2008, 20, 2621–2623.
mono-layer of zinc oxide included at the surface (Fig. S23, ESI†).33
14 H. Weller, A. Fojtik and A. Henglein, Chem. Phys. Lett., 1985, 117, 485–488.
To assess the potential of these particles for photovoltaic applica- 15 W. E. Buhro, Polyhedron, 1994, 13, 1131–1148.
16 M. Green and P. O’Brien, Chem. Mater., 2001, 13, 4500–4505.
17 S. Carenco, M. Demange, J. Shi, C. Boissiere, C. Sanchez, P. Le Floch
tions, a film of particles was electrodeposited onto a conductive
mesoporous nanoITO substrate using electrophoretic deposition
and N. Mezailles, Chem. Commun., 2010, 46, 5578–5580.
(Fig. S24, ESI†).40 Controlled potential electrolysis was performed 18 S. Miao, T. Yang, S. G. Hickey, V. Lesnyak, B. Rellinghaus, J. Xu and
A. Eychmu¨ller, Small, 2013, 9, 3415–3422.
19 E. J. Luber, M. H. Mobarok and J. M. Buriak, ACS Nano, 2013, 7,
under chopped white light (Fig. S25, ESI†) and showed a 100 nA
current increase with light irradiation indicating the possibility of
8136–8146.
using these particles as visible light absorbers.33
20 M. H. Mobarok, E. J. Luber, G. M. Bernard, L. Peng, R. E. Wasylishen
and J. M. Buriak, Chem. Mater., 2014, 26, 1925–1935.
21 J. R. Heath and J. J. Shiang, Chem. Soc. Rev., 1998, 27, 65–71.
22 M. Westerhausen, G. Sapelza, M. Zabel and A. Pfitzner, Z. Naturforsch.,
B: J. Chem. Sci., 2004, 59, 1548–1550.
23 See Fig. S1 (ESI†) for absorbance of particles made without ZnEt2.
24 LEET is determined by fitting the spectra and taking the second-
derivative to locate the local minima.
In conclusion, this work describes a novel three precursor synth-
esis of colloidal crystalline Zn3P2 quantum dots that have clear
excitonic absorbance features ranging from 424 to 535 nm, depending
on the final growth temperature. NMR experiments revealed a critical
molecular transformation that only occurs when ZnEt2, Zn(O2CR)2,
and P(SiMe3)3 areallpresent.Theobserved(Et2Zn)P(ZnO2CR)2(SiMe3) 25 D. Battaglia and X. Peng, Nano Lett., 2002, 2, 1027–1030.
26 R. Wang, C. I. Ratcliffe, X. Wu, O. Voznyy, Y. Tao and K. Yu, J. Phys.
Chem. C, 2009, 113, 17979–17982.
27 V. K. LaMer and R. H. Dinegar, J. Am. Chem. Soc., 1950, 72,
intermediate was conclusively shown to be a direct precursor to the
crystalline zinc phosphide quantum dots obtained in this study. TEM,
ICP-OES, TGA, NMR, and XRD experiments showed that the resulting
particles were crystalline and confirmed their identity as zinc-rich
zinc phosphide QDs that ranged from 2.6 to 2.9 nm in diameter. In
4847–4854.
28 S. G. Kwon and T. Hyeon, Small, 2011, 7, 2685–2702.
29 K. L. Orchard, A. J. P. White, M. S. P. Shaffer and C. K. Williams,
Organometallics, 2009, 28, 5828–5832.
contrast to the synthesis of Zn3P2 QDs recently reported by Buriak and 30 O. Fuhr and D. Fenske, Z. Anorg. Allg. Chem., 2004, 630, 244–246.
31 G. Sapelza, P. Mayer and M. Westerhausen, Z. Anorg. Allg. Chem.,
2005, 631, 3087–3091.
32 C. A. Schneider, W. S. Rasband and K. W. Eliceiri, Nat. Methods,
co-workers which yields 15 Æ 2 nm particles from a combination of
ZnMe2 and P(SiMe3)3 with no strongly passivating ligands present, our
synthesis yields relatively monodisperse colloidal solutions of small
size nanocrystals as a result of carboxylate ligand passivation.20 The
stark contrast between the nanocrystal sizes reported here and by
2012, 9, 671–675.
33 See the ESI† for experimental details.
34 P. J. Reid, B. Fujimoto and D. R. Gamelin, J. Chem. Educ., 2013, 91,
280–282.
Buriak highlights the critical role of precursor reactivity and ligand 35 M. Kuno, Introductory Nanoscience: Physical and Chemical Concepts,
Garland Science, Taylor & Francis Group, New York, 2012, ch. 3,
pp. 29–60.
36 This larger particle gave the most detail during imaging.
binding strength on final particle size, and suggests a strategy for
achieving nanocrystals of intermediate size.41 Initial photoresponse
measurements demonstrated the possibility of using these particles 37 The discrepancy in expected versus observed peak intensity for the
45 2y reflection may be the result of a defective lattice and may be an
indication of the zinc-rich nature of this sample.
as visible light absorbers. Future experiments will focus on synthe-
sizing and characterizing larger particles that retain an excitonic
¨
38 P. Scherrer, Nachr. Ges. Wiss. Gottingen., 1918, 1990, 98–100.
feature as well as fabricating solid state devices to probe the effect 39 M. G. Bawendi, A. R. Kortan, M. L. Steigerwald and L. E. Brus,
J. Chem. Phys., 1989, 91, 7282–7290.
40 P. G. Hoertz, Z. Chen, C. A. Kent and T. J. Meyer, Inorg. Chem., 2010,
of QD composition and surface chemistry on device characteristics.
Part of this work was conducted at the University of
49, 8179–8181.
Washington NanoTech User Facility, a member of the NSF National 41 M. H. Mobarok and J. M. Buriak, Chem. Mater., 2014, 26, 4653–4661.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2014