The formation mechanism of binary semiconductor nanomaterials: Shared by single-source and dual-source precursor approaches

Article abstract of DOI:10.1002/anie.201304958

One thing in common: The formation of binary colloidal semiconductor nanocrystals from single- (M(EEPPh₂)_n) and dual-source precursors (metal carboxylates M(OOCR)_n and phosphine chalcogenides such as E=PHPh₂) is found to proceed through a common mechanism. For CdSe as a model system ³¹Pa NMR spectroscopy and DFT calculations support a reaction mechanism which includes numerous metathesis equilibriums and Se exchange reactions. Copyright

Full text of DOI:10.1002/anie.201304958

.
Angewandte
Communications
DOI: 10.1002/anie.201304958
Nanocrystal Formation
The Formation Mechanism of Binary Semiconductor Nanomaterials:
Shared by Single-Source and Dual-Source Precursor Approaches**
Kui Yu,* Xiangyang Liu, Qun Zeng, Mingli Yang,* Jianying Ouyang, Xinqin Wang, and Ye Tao
Semiconductor materials are technologically important and
impact our daily lives in various aspects.[1–9] Both single-
source and dual-source precursor approaches (SSPA and
DSPA) to binary semiconductors have been documented.[9–27]
Such materials consist of elements from two groups such as
IIB and VIA. The single-source precursors (SSPs) consist of
the metallic and nonmetallic elements of the semiconductor
constituents in a single molecule.[9–14] DSPA uses separated
metallic-element and nonmetallic-element precursors, which
commonly involve metal carboxylates (M(OOCR)n such as
M = Zn, Cd, Pb, Cu, In) and phosphine chalcogenides (such as
E = PHR2 where E = S, Se, Te),[15–27] respectively. Despite the
large number of recipes developed for the various colloidal
semiconductor nanocrystals (NCs) in the past 20 years, there
is still little understanding of their formation mechanisms. To
realize the full potential of semiconductor materials, there is
an urgent need to advance our mechanistic understanding.
Filling this gap in our knowledge should have practical
implications such as lowering the high temperature currently
employed for syntheses and offering new avenues to optimize
the design of low-temperature approaches to novel semi-
conductor nanomaterials.
precursors.[15–27] Astonishingly, the lack of a common forma-
tion mechanism is actually accompanied by the same
31P NMR identification of RCOO-PPh2 (R = C17H33 99 ppm
(3 in Scheme 1) or C6H5 102 ppm) and Ph2P-PPh2 (ꢀ14 ppm,
4) for the various DSPAs to PbSe,[18] CdSe,[19–22] ZnSe,[23,24]
ZnS,[24] and ZnSeS,[24] together with C17H33COO-P(Se)Ph2
(77 ppm, 5) for the Se-based NCs.[18,21–24] Furthermore, the
=
conversion of Se PHPh2 to diphenyldiselenophosphinate
[18,19,22]
ꢀ
derivatives ( SeSePPh2) has been documented.
For
instance,[22] the formation of RCOOCdSeSePPh2 (c) was
=
proposed from a Cd(OA)2 + Se PHPh2 reaction after the
release of oleic acid (C17H33COOH or RCOOH, R = C17H33)
=
from (RCOO)2Cd(Se PHPh2)2 (b) followed by diphenyl-
ꢀ
phosphine (HPPh2 or DPP) from RCOOCd(Se PPh2)-
=
=
(Se PHPh2) (d) through cleavage of the Se P bond of the
=
Se PHPh2 coordination arm.
Recently, a SSP, cadmium bis(diphenyldithiophosphinate)
(Cd(SSPPh2)2), together with cadmium oleate Cd(OA)2, was
used to synthesize CdS QDs at 2408C in 1-octadecene
(ODE).[14] Without Cd(OA)2, Cd(SSPPh2)2 did not produce
CdS QDs at 2408C. Various dithiophosphinato or diseleno-
phosphinato complexes have been synthesized, with the
general structures of M(EEPR2)n, where M = metal ions of
Group IIIA (trivalent), IIB (divalent), IVA (divalent), and IB
(monovalent), and R = alkyl, phenyl, and alkoxy groups.[28–32]
Some of these complexes have been used in SSPAs to make
E-based binary semiconductor NCs and thin films but only at
temperature higher than 3008C.[9–14]
To explore the common mechanism for the formation of
the various ME semiconductor nanomaterials from the SSPA
and DSPA, herein, CdSe was investigated as a model system
in detail. 31P NMR spectroscopy was used to monitor key
products involved in the formation of CdSe NCs from the
SSPA and DSPA which we have designed. As shown in
Scheme 1, the former addresses the reaction of cadium
bis(diphenyldiselenophosphinate) [Cd(SeSePPh2)2] (2) in
ODE, in the presence of Cd(OA)2 and DPP. The latter
deals with the Cd(OA)2 + SeDPP reaction in ODE. Also,
in situ absorption was used to monitor the formation of NCs
and density functional theory (DFT) calculations were
performed to study the reactants, intermediates (IM), tran-
sition states (TS), and products. On the basis of the NMR
spectroscopy, in situ absorption, and DFT calculations, we
propose an essentially identical mechanism for the two
approaches, as illustrated in Scheme 1 which seems to be
complicated but can be understood readily with numerous
metathesis equilibriums and Se exchange reactions.
Recent evidence suggests that the formation of various
binary semiconductor NCs by SSPAs and DSPAs may share
analogous mechanisms. A DSPA to CdS quantumm dots
(QDs) at 1608C in tetradecane (CH3(CH2)12CH3) was
reported from the reaction of cadmium stearate
=
(Cd(OOCC17H35)2) and diphenylphosphine sulfide (S
PHPh2).[25] DSAPs to E-based semiconductor QDs have
become popular with metal carboxylates as cation precursors
and diphenylphosphine chalcogenides E = PHPh2 as anion
[*] Dr. K. Yu, Dr. X. Liu, Dr. J. Ouyang, Dr. Y. Tao
National Research Council of Canada
Ottawa, ON, K1A 0R6 (Canada)
E-mail: kui.yu@nrc-cnrc.gc.ca
Q. Zeng, Prof. M. Yang, X. Wang
Institute of Atomic and Molecular Physics and State Key Laboratory
of Biotherapy, Sichuan University
Chengdu 610065 (P. R. China)
E-mail: myang@scu.edu.cn
[**] K.Y. thanks Dr. Keith Ingold for valuable discussions and Dr. Danial
Wayner for precious encouragement. The authors thanks Erik
Huisman for some of the in situ absorption measurements
presented and the Defence Research and Development Canada
Centre for Security Science Chemical, Biological, Radiological/
Nuclear, and Explosives Research and Technology Initiative (CRTI
09-0511RD “Next-generation stand-off radiation detection using
nanosensors”) for financial support.
Scheme 1 presents two manifolds of equilibriums, c, d,
e and 3 (top), and f, g, h, and 4 (bottom). Simply, the top
manifold leads to 3 and CdSe (clusters and NCs) (and 5); the
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304958.
11034
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039

Angewandte
Chemie
Scheme 1. The reaction mechanism proposed for the SSPA (red) and DSPA (blue) with Cd(SeSePPh2)2 (2) and Cd(OA)2 + SeDPP, respectively.
This mechanism include the metathesis reactions of x + DPP Q y + RCOOH, where x/y=c/f, d/g, e/h, and 3/4, and the Se exchange reaction
between 3/5. The formation of (CdSe)n clusters drives the equilibriums to the products (green). The compounds labeled by numbers were
detected by 31P NMR spectroscopy and summarized in Table S1, and those labeled by letters were studied by DFT (Schemes S2–S5).
bottom manifold results in 4 and CdSe (cluters and NCs). The
Se exchange reaction between 3 and 5 [Supporting Informa-
tion, Scheme S1, Eq. (6)], 3 + SeDPP Q 5 + DPP, is similar to
TOP + SeDPP Q TOPSe + DPP.[18,22] The top and bottom
manifolds are connected by the four metathesis reactions
which involve the reversible exchange of HPPh2 and
C17H33COOH between c and f, d and g, e and h, and 3 and 4
[Scheme S1, Eq. (1) to (4)]. The mechanism proposed is in
full agreement with our experimental observations: both the
SSAP and DSAP lead to the same products, which are the
three phosphorus-containing compounds (3, 4, and 5) and
CdSe NCs.
Cd(SeSePPh2)2. For the former approach, 0.005 mmol of 2
was used. For the latter approach, the amount of its limiting
reactants was 0.03 mmol. The collected 31P NMR spectra are
fairly similar. Compounds 3, 4, and 5 were obtained from the
both approaches and appear to share similar or identical
reaction pathways, as proposed in Scheme 1.
Moreover, an increase in the amount of Cd(OA)2
increases the 3-to-4 ratio for the both approaches. For the 2
+ 14Cd(OA)2 + 1DPP reaction (a), a small amount of 4 and
a large amount of 3 were observed, similar to the 2Cd(OA)2 +
1SeDPP reaction (d). For the 2 + 1Cd(OA)2 + 1DPP
reaction (c), a large amount of 4 and a tiny amount of 3 were
observed, similar to the 1Cd(OA)2 + 3SeDPP reaction (f).
Due to the equilibrium of 3 + DPP Q 4 + RCOOH weighted
toward the left (Figure S1a), the formation of the large
amount of 4 (particularly during the proceeding of Reaction
c) should be mainly from h instead of 3.
A significant amount of 5 was obtained in reactions (b)
and (e) (Figure 1), but almost no 5 was detected in reactions
(c) and (f) (with the relatively small Cd-to-Se feed molar
ratios) and in reactions (a) and (d) (with the relatively large
Cd-to-Se feed molar ratios). Such observations can be readily
understood by the fact that the reaction of Cd(OA)2 and
SeDPP is astonishingly fast (even at ꢀ558C),[22] while the
equilibrium of 3 + SeDPP Q 5 + DPP is weighted toward the
right (Figure S1a), similar to that of TOP + SeDPP Q SeTOP
+ DPP.[30,31] When Cd(OA)2 is in excess (reactions (a) and
(d)), SeDPP (such as released from d in the SSPA) should
react with Cd(OA)2 instead of 3 and little 5 could be obtained.
When Cd(OA)2 is not excess (reactions (b) and (e)), SeDPP
could react with 3 and an apparent amount of 5 could be
obtained. Thus, the fact, that the Cd-to-Se feed molar ratio
affects the relative amount of compounds 3, 4, and 5, supports
Furthermore, the feed molar ratio of the reactants affects
the relative amounts of the three phosphorus-containing
compounds obtained and the amount and size of the CdSe
NCs. RCOOH and DPP favor the top and bottom manifolds,
respectively. The formation of the CdSe NCs is thermody-
namically favored, which shifts the equilibriums to the
products. While Compounds 3, 4, and 5 have been previously
monitored by 31P NMR spectroscopy from the various
DSPAs,[18–24] the present study provides a common mechanism
for their formation from the DSPA with M(OOCR)n and E =
PHPh2 and from the SSPA with M(EEPPh2)n. With far-
reaching implications, our findings delineate the fundamental
understanding of the formation mechanism and endorse
subsequently the low-temperature design and syntheses of
various binary and alloyed NCs, bringing insight to the
empirical use of additives such as M(OOCR)n, DPP, and
RCOOH as a means to promote low-temperature syntheses
(particularly for the SSPA).
Figure 1 shows 31P NMR spectra of the SSPA and DSPA
at room temperature of 2 + Cd(OA)2 + DPP (a–c) and
Cd(OA)2 + SeDPP (d–f), respectively. Compound 2 is
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11035

.
Angewandte
Communications
Figure 1. 31P NMR study of the SSPA of the three reactions of 2 in the presence of Cd(OA)2 and DPP (a–c) and the DSPA of the three reactions of
Cd(OA)2 + SeDPP (d–f). The spectra were collected at room temperature with each reaction proceeding at 5 min (1), 10 min (2), 20 min (3),
40 min (4), and 60 min (5). Compounds 3, 4, and 5 were obtained from the two approaches, with high Cd-to-Se feed molar ratios promoting the
3-to-4 ratio.
also the chemical mechanism illustrated by Scheme 1. Also,
the effect of such ratios on the nucleation and growth of CdSe
NCs appear analogous for the two approaches (Figure S1b).
The synthesis and characterization of 2 is illustrated in
Figure S1c. The dispersion of purified 2 in toluene was stable
with a chemical shift of 12.0 ppm and JP-Se of 480 Hz at 1008C
and no obvious decomposition was detected over 2 h. In the
presence of Cd(OA)2, the disappearance of 2 in toluene at
808C led to a clear solution, and a phosphorus resonance at
16.2 ppm and JP-Se of 496 Hz at 1008C was detected and
assigned to C17H33COO-Cd-SeSePPh2 (c). In the presence of
DPP and/or RCOOH, purified 2 in toluene/ODE remained
(as a white solid) even at 808C, with only a small amount of
SeDPP and 4 detected. In the presence of both Cd(OA)2 and
DPP, the reactivity of 2 in ODE increased profoundly at
temperature as low as 508C (Figure S1c-4); the presence of
both Cd(OA)2 and DPP is important for the SSPA to CdSe
NCs.
Figure 2 shows the increase of 4 from h by DPP and 3
from e by RCOOH. The 31P NMR spectra were collected
from the SSPA of 2 + 6Cd(OA)2 + xDPP (top) and 2 +
6Cd(OA)2 + xDPP + 1RCOOH (bottom) reactions at room
temperature, with x = 0.5 (a and b), 1 (c and d), and 4 (e and
f). Again, 0.005 mmol of Compound 2 was used. From
0.5DPP to 4DPP, the 3-to-4 ratio decreases together with
the disappearance of 5. Along the proceeding of reactions (e)
and (f) with 4DPP, the significant increase of 4 indicates
evidently its foremost formation from h instead of 3. Thus,
DPP seems to promote the formation of 4 from h together
with the suppression of 5 from 3 + SeDPP Q 5 + DPP. For
reactions (c) and (d) with 1DPP, reaction (d) produced more
3 and 5 at its early stage than reaction (c) did. Clearly, reaction
(d) (with 1RCOOH) suppressed 4 and promoted 3 (and thus
5). Along the proceeding of reaction (d), the remarkable
increase of 3 in the presence of a tiny amount of 4 specifies
outstandingly the probable formation of 3 from e (instead of
4). The effect of DPP on the promotion of 4 and the effect of
RCOOH on the promotion of 3 support the chemical
mechanism illustrated in Scheme 1.
Figure S2 demonstrates the enhanced nucleation and
growth of CdSe NCs from the SSPA by DPP and RCOOH.
Obviously, the increase of DPP (top, 2 + 6Cd(OA)2 + xDPP
+ 1.2OA with x = 0.1 and 0.2) and the increase of RCOOH
(bottom, 2 + 6Cd(OA)2 + 1DPP + (1.2 + y)OA with y = 0
and 0.8) increased the particle number. Note that RCOOH
and DPP were released from the DSPA even at ꢀ558C;[22]
thus, it is easy to understand that the effect of DPP and
RCOOH may not be obvious for the DSPA (Figure S2).
Now, let us turn our attention to the reactions involving
HPPh2 shown in Scheme 1 such as c + HPPh2. Two pathways
shown in Scheme 2 are proposed, namely route I (c to d) and
route II (c to f). Route I, a reversed process of the release of
HPPh2 from d leading to c,[22] was studied by DFT. Again, the
CH3 group was used to mimic the phenyl and C17H33 groups in
our DFT calculations, such as dimetyl phosphine (DMP or
HPMe2) for DPP.[22] As computed in Scheme S2, the P atom of
HPMe2 attacks the Se atom of c leading to the formation of
IM1 (Gibbs free energy DG =+ 33.6 kJmolꢀ1). Then, a new
ꢀ
Se P bond forms between HPMe2 and c, while its neighboring
ꢀ
Se P bond is weakened, leading to TS1 (DG =
+ 76.0 kJmolꢀ1) and subsequently d (DG =+ 50.0 kJmolꢀ1).
Note that our DFT calculations were carried out at 298 K,
producing similar results with TPSSh, TPSS, and B3LYP
functionals. Although both total energy (DE) and DG
variations were computed, only the values of DG from the
11036 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039

Angewandte
Chemie
Figure 2. The effect of DPP and RCOOH studied by 31P NMR spectroscopy. The spectra were collected at room temperature with each of the six
reactions proceeding at 5 min (1), 10 min (2), 20 min (3), 40 min (4), and 60 min (5). Evidently, DPP promotes the formation of 4, while oleic
acid (RCOOH) assists the formation of 3.
HPPh2 (Figure S3a-1). With the PCy2 group instead of the
PPh2 group, 1’’ and 2’’ are analogous of 1 and 2, respectively.
The detection of SePPh2H from the 2 + HPPh2 reaction,
together with 2’’, SePPh2H, and SePCy2H from the 2 + HPCy2
reaction provided us extremely valuable information to
comprehend route I. The observation of SeDPP from these
ꢀ
two reactions suggests the probable breakage of Se P bond of
the -SeSePPh2 group of 2. Figure S3a-2 presents our proposed
chemical mechanism for the formation of 2’’ from the 2 +
HPCy2 reaction, which proceeds through a multi-step equi-
librium process. The P atom of HPCy2 attacks the Se atom of 2
ꢀ
resulting in the breakage of the original Se P bond. After-
ꢀ
ꢀ
Scheme 2. Two reaction pathways proposed for HPPh2 (where
R’=Ph2). Route I (top) illustrates the interaction of P and Se for c to d
and f to g. Route II (bottom) illustrates the interaction of P and Cd for
c to f, d to g, and e to h; this pathway is substitution/metathesis.
ward, the release of SeDPP leads to the formation of Cd
SePCy2 species. HPCy2 reacts with SeDPP resulting in
SeHPCy2 (Figure 3, bottom). SePCy2H sequentially reacts
ꢀ
ꢀ
with the Cd SePCy2 part to form a Cd(SeSePCy2) fragment.
Also, we designed two groups of reactions to support
Scheme 1 proposed. They are the 2 + Cd(OA)2 + HPCy2
TPSSh results are addressed, as the reactions took place at group (Figure S3b, left) and the Cd(OA)2 + SePCy2H group
a constant temperature and pressure. (Figure S3b, right). Compounds 3, 3’’ (C17H33COOPCy2,
To exam route I further, we designed three reactions. 134 ppm), 5, and 5’’ (C17H33COOP(Se)Cy2, 116 ppm) were
Figure 3 shows in situ 31P NMR spectra of the reactions of 2 + detected from the former, and compounds 3’’ and 5’’ from the
HPPh2 in DMSO (dimethyl sulfoxide) (a), 2 + HPCy2 latter.
(dicyclohexyl phospine) in ODE (b), and SeDPP + HPCy2
For route II, the P atom of HPPh2 interacts with the Cd
(c). The amounts of 2 were 0.0125 mmol (a) and 0.005 mmol atom.[22] As computed in Scheme S3a, the proton of HPMe2
(b), and that of SeDPP was 0.01 mmol (c). HPCy2 seems to be approaches one O atom of c, leading to the formation of IM3
more reactive than HPPh2 towards 2. This middle reaction (DG =+ 6.8 kJmolꢀ1). After going through a barrier of
completed in minutes at room temperature with the dissolu- 65.5 kJmolꢀ1, TS3 (DG =+ 72.3 kJmolꢀ1) is obtained, which
ꢀ
tion of 2 resulting in a clear solution. The chemical shift at has a six-membered ring with the breakage of one original O
54.8 ppm with
(SeSePCy2)2).
J
To
P-Se = 484 Hz is assigned to 2’’ (Cd- Cd bond and the proton bridging the O and P atoms. The
ꢀ
confirm,
we
synthesized
1’’ breakage of the original H P bond results in the formation of
(Cy2PSeSeNH3R, R = C18H35) and 2’’, following the synthetic IM4 (DG =+ 61.9 kJmolꢀ1). Then HOAc is released, together
procedure to 1 and 2, respectively, but with HPCy2 instead of with the formation of f (DG =+ 53.1 kJmolꢀ1). Thus, our DFT
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11037

.
Angewandte
Communications
obtained from the top reaction, while a large amount of 3 and
a tiny amount of 3’ were obtained from the bottom reaction.
For the top reaction, thus, similar amounts of
C17H33COOCdSePPh2 (e) from C17H33COOCdSeSePPh2 (c)
and PhCOOCdSePPh2 (e’) from PhCOOCdSeSePPh2 (c’)
should have formed. For the bottom reaction, the amount of
e from c should be much more than that of e’ from the
reaction of Ph2PCdSePPh2 (h) + PhCOOH. Accordingly, the
formation of e from c (through d) seems to be more favored
than that of h from c (through f and g).
Finally, for the formation of (CdSe)n together with
compounds 3 and 4 from e and h, respectively, the energy
change is computed and summarized in Scheme S5. Intrigu-
ingly, DG is largely positive with n = 1, drops dramatically
with n = 2, becomes negative with n = 3, and further decreases
with n = 6 and 13. Thus, the formation of the (CdSe)n clusters
(n > 3) lowers significantly the DG values; taking the DG of c
as a reference, the total DG becomes negative with n = 13.
Consequently, the formation of the (CdSe)n clusters are
thermodynamically favored, driving the equilibriums to the
product direction. The experimental observation of (CdSe)13
(exhibiting the bandgap of ca. 350 nm) can be found else-
where[33] and in Figure S1b and S2.
In conclusion, the present study addresses a significant
gap in our understanding of the formation mechanism of
semiconductor materials from the DSPA with M(OOCR)n
and E = PHPh2 and from the SSPA with M(EEPPh2)n. Based
on our experimental data and DFT calculations, we have
proposed a common chemical mechanism shared by the SSPA
and DSPA to semiconductor nanomaterials. Also, the present
study provides the fundamental insight for rational design and
synthesis of semiconductor materials with enhanced particle
yield at low reaction temperature. Collectively, the proposed
mechanism suggests a set of design criteria to take advantage
of the equilibriums, providing new avenues for low-temper-
ature syntheses. Although our efforts focused on CdSe, the
formation mechanism (Scheme 1) can be readily extended to
other NC systems from the DSPA with M(OOCR)n and E =
PHPh2 and from the SSPA with M(EEPPh2)n.[9,18–24] We
envision that the proposed mechanism should be broadly
applicable to the design and synthesis of various binary and
alloyed semiconductor materials. With new perspectives, the
identification of the metathesis reactions should serve as
a platform to improve our capability to engineer semi-
conductor materials with smarter approaches to satisfy the
requirement on low-temperature syntheses with high quality.
Figure 3. Three reactions with HPR2 (where R=Ph (a) or Cy (b and c))
designed to investigate route I. a) The 31P NMR spectra were in situ
collected at room temperature before the addition of DPP (1), after the
addition of DPP of 5 min (2), 10 min (3), 45 min (4), 120 min (5), and
210 min (6). b) the 31P NMR spectra were obtained at 15 min (1) and
180 min (2) at room temperature, 30 min at 808C (3), and 30 min at
1208C (4).
calculations demonstrate that routes I and II experience
moderate energy barriers of 42.4 kJmolꢀ1 (TS1) and
65.5 kJmolꢀ1 (TS3), respectively, suggesting that the two
routes are kinetically feasible at 298 K (room temperature).
For d!SeDPP + e (Scheme S2), the small increase in DG
of 4.0 kJmolꢀ1 is worthy of notice, together with the small
barrier of 7.8 kJmolꢀ1 (TS2). For f + DPP!h + SeDPP
(Scheme S3b), the DG increase is 50.8 kJmolꢀ1 with a barrier
of 54.2 kJmolꢀ1 (TS4). For
e
+
DPP!h
+ HOAc
Received: June 10, 2013
Published online: September 4, 2013
(Scheme S4), the DG increase is 51.1 kJmolꢀ1 with a barrier
of 82.5 kJmolꢀ1 (TS5). Thus, our DFT calculations suggest
that e is both more thermodynamically and kinetically
favored than h.
Keywords: metal–ligand interaction · molecular modeling ·
.
phosphines · nanocrystal formation mechanisms ·
semiconductor quantum dots
To test such a computational result, we designed two
reactions, 2 + 3Cd(OA)2 + 3Cd(OOCPh2)2 + 1DPP (Fig-
ure S4 top) and 2 + 6Cd(OA)2 + 3PhCOOH + 1DPP
(Figure S4 bottom). In addition to compounds 3 and 5,
compounds PhCOOPPh2 (3’, 102 ppm) and PhCOOP(Se)Ph2
(5’, 79 ppm) were also obtained, together with Ph2P-PPh2 (4,
ꢀ14 ppm). Interestingly, similar amounts of 3 and 3’ were
[1] S. Jun, E. Jang, Angew. Chem. 2013, 125, 707 – 710; Angew.
Chem. Int. Ed. 2013, 52, 679 – 682.
[2] Y. Shirasaki, G. J. Supran, M. G. Bawendi, V. Bulovic, Nat.
Photonics 2013, 7, 13 – 23.
11038 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039

Angewandte
Chemie
[3] R. Gill, M. Zayats, I. Willner, Angew. Chem. 2008, 120, 7714 –
7736; Angew. Chem. Int. Ed. 2008, 47, 7602 – 7625.
[4] F. Song, P. S. Tang, H. Durst, D. T. Cramb, W. C. W. Chan,
Angew. Chem. 2012, 124, 8903 – 8907; Angew. Chem. Int. Ed.
2012, 51, 8773 – 8777.
[5] E. H. Sargent, Nat. Photonics 2012, 6, 133 – 135.
[6] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S.
Strano, Nat. Nanotechnol. 2012, 7, 699 – 712.
[7] V. I. Klimov, A. A. Mikhailovsky, S. Xu, A. Malko, J. A.
Hollingsworth, C. A. Leatherdale, H. Eisler, M. G. Bawendi,
Science 2000, 290, 314 – 317.
[8] Z. Han, F. Qiu, R. Eisenberg, P. L. Holland, T. D. Krauss, Science
2012, 338, 1321 – 1324.
[19] B. M. Cossairt, J. S. Owen, Chem. Mater. 2011, 23, 3114 – 3119.
[20] K. Yu, Adv. Mater. 2012, 24, 1123 – 1132.
[21] R. Garcꢁa-Rodrꢁguez, H. Liu, J. Am. Chem. Soc. 2012, 134,
1400 – 1403.
[22] K. Yu, X. Liu, Q. Zeng, D. M. Leek, J. Ouyang, K. M. Whitmore,
J. A. Ripmeester, Y. Tao, M. Yang, Angew. Chem. 2013, 125,
4923 – 4928; Angew. Chem. Int. Ed. 2013, 52, 4823 – 4828.
[23] K. Yu, A. Hrdina, X. Zhang, J. Ouyang, D. M. Leek, X. Wu, M.
Gong, D. Wilkinson, C. Li, Chem. Commun. 2011, 47, 8811 –
8813.
[24] K. Yu, A. Hrdina, J. Ouyang, D. Kingston, X. Wu, D. M. Leek, X.
Liu, C. Li, ACS Appl. Mater. Interfaces 2012, 4, 4302 – 4311.
[25] H. H.-Y. Wei, C. M. Evans, B. D. Swartz, A. J. Neukirch, J.
Young, O. V. Prezhdo, T. D. Krauss, Nano Lett. 2012, 12, 4465 –
4471.
[9] M. A. Malik, M. Afzaal, P. OꢀBrien, Chem. Rev. 2010, 110, 4417 –
4446.
[10] J. Sun, W. E. Buhro, Angew. Chem. 2008, 120, 3259 – 3262;
Angew. Chem. Int. Ed. 2008, 47, 3215 – 3218.
[11] C. Byrom, M. A. Malik, P. OꢀBrien, A. J. P. White, D. J. Williams,
Polyhedron 2000, 19, 211 – 215.
[12] C. Q. Nguyen, M. Afzaal, M. A. Malik, M. Helliwell, J. Raftery,
P. OꢀBrien, J. Organomet. Chem. 2007, 692, 2669 – 2677.
[13] N. O. Boadi, M. A. Malik, P. OꢀBrien, J. A. M. Awudza, Dalton
Trans. 2012, 41, 10497 – 10506.
[14] M. P. Hendricks, B. M. Cossairt, J. S. Owen, ACS Nano 2012, 6,
10054 – 10062.
[15] M. R. Buck, R. E. Schaak, Angew. Chem. 2013, 125, 6270 – 6297;
Angew. Chem. Int. Ed. 2013, 52, 6154 – 6178.
[16] L. De Trizio, M. Prato, A. Genovese, A. Casu, M. Povia, R.
Simonutti, M. J. P. Alcocer, C. DꢀAndrea, F. Tassone, L. Manna,
Chem. Mater. 2012, 24, 2400 – 2406.
[17] F. Wang, W. R. Buhro, J. Am. Chem. Soc. 2012, 134, 5369 – 5380.
[18] C. M. Evans, M. E. Evans, T. D. Krauss, J. Am. Chem. Soc. 2010,
132, 10973 – 10975.
[26] K. Yu, J. Ouyang, D. M. Leek, Small 2011, 7, 2250 – 2262.
[27] See Supporting Information: K. Yu, et al., ACS Appl. Mater.
Interfaces 2013, 5, 2870 – 2880.
[28] W. Kuchen, H. Hertel, Angew. Chem. 1969, 81, 127 – 135; Angew.
Chem. Int. Ed. Engl. 1969, 8, 89 – 97.
[29] R. G. Cavell, E. D. Day, W. Byers, P. M. Watkins, Inorg. Chem.
1972, 11, 1759 – 1772.
[30] R. P. Davies, C. V. Francis, A. P. S. Jurd, M. G. Martinelli, A. J. P.
White, D. J. Williams, Inorg. Chem. 2004, 43, 4802 – 4804.
[31] C. Q. Nguyen, A. Adeogun, M. Afzaal, M. A. Malik, P. OꢀBrien,
Chem. Commun. 2006, 2182 – 2184.
[32] P. K. Liao, D. R. Shi, J. H. Liao, C. W. Liu, A. V. Artemꢀev, V. A.
Kuimov, N. K. Gusarova, B. A. Trofimov, Eur. J. Inorg. Chem.
2012, 4921 – 4929.
[33] Y. Wang, Y. Liu, Y. Zhang, F. Wang, P. J. Kowalski, H. W. Rohrs,
R. A. Loomis, M. L. Gross, W. E. Buhro, Angew. Chem. 2012,
124, 6258 – 6261; Angew. Chem. Int. Ed. 2012, 51, 6154 – 6157.
Angew. Chem. Int. Ed. 2013, 52, 11034 –11039
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11039