Quantum soldering of individual quantum dots

Article abstract of DOI:10.1002/anie.201206301

Making contact to a quantum dot: Single quantum-dot electronic circuits are fabricated by wiring atomically precise metal chalcogenide clusters with conjugated molecular connectors. These wired clusters can couple electronically to nanoscale electrodes and be tuned to control the charge-transfer characteristics (see picture). Copyright

Full text of DOI:10.1002/anie.201206301

Angewandte
Chemie
DOI: 10.1002/anie.201206301
Molecular Electronics
Quantum Soldering of Individual Quantum Dots**
Xavier Roy, Christine L. Schenck, Seokhoon Ahn, Roger A. Lalancette, Latha Venkataraman,*
Colin Nuckolls,* and Michael L. Steigerwald*
Here we describe a precise method to make electrical contact
to an individual quantum dot (QD). This supramolecular
construction connects the QD to its macroscopic environ-
ment, yet it does not disturb the nanoscopic quantum
mechanical confinement of the excitons in these small
solids. Quantum mechanical confinement has given rise to
the hallmark optical properties of QDs,[1–3] but it has been of
only limited use in electronic and opto-electronic applica-
tions[3] of QDs because of three interrelated problems: 1) the
lack of knowledge of how to make innocent electrical contact
to QDs; 2) the challenge of synthesizing atomically precise
QDs; and 3) not having the methods to efficiently wire
individual QDs in electrical devices. Robust electrical contact
to the core of QDs is essential in the development of QD-
based electronic devices[3–8] and for the extraction of hot
electrons[9] and the separation of charges from multiple
exciton states[10,11] in QD solar cells but yet it has only been
thoroughly explored in the context of thin films and bulk
samples of QDs[12–17] where function cannot be related to the
poorly characterized structure and quantum confinement is
compromised at best. Here we synthesize, for the first time,
a molecularly discrete, crystallographically defined, electron-
rich, metal chalcogenide cluster, Co6Se8,[18] that is capped with
conjugated, molecular connectors that can couple electroni-
cally to nanoscale electrodes. We show that these connectors
provide a well-defined electronic pathway for the transport of
charge carriers through a single QD. We measure the
conductance of individual QDs using a scanning tunneling
microscope based break-junction (STM-BJ) technique[19–21]
and compare our results with density functional theory.
Finally, we show that we can control the electronic
coupling between the core of the QD and the conducting
backbone of the connector by varying the connector structure
allowing us to differentiate between conductive molecular
connectors and insulating ones. These results establish
quantum mechanical design rules for controlling the elec-
tronic coupling to a QD for the creation of QD-based
electrical circuits.
The solid-state compound CoSe is an infrared bandgap
semiconductor.[22] We synthesized a series of atomically
precise cobalt selenide quantum dots[18,23,24] decorated with
different molecular connectors (L2–L5). Connectors L2–L4
have a phosphine end that coordinates to the cobalt atom in
the cluster and a thiomethyl end that is aurophilic. Connector
L5 lacks a thiomethyl group and serves as a control. We
selected this family of compounds based on the parent QD
Co6Se8(PEt3)6 (1) (Figure 1b) because its electron-rich core is
a reservoir of carriers, and its synthesis is amenable to a broad
range of phosphines. Single-crystal X-ray diffraction
(SCXRD) shows that the Co6Se8 core of the clusters, 1–5,
are isostructural (Figure 2), forming an octahedron of Co
atoms concentric with a cube of Se atoms. Cluster 4 packs
with its six molecular connectors grouped into two diametri-
cally opposed groups of three, resulting in an ideal conforma-
tion for bridging a linear gap between two electrodes, as
illustrated in Figure 1a.
[*] Dr. X. Roy, C. L. Schenck, Dr. S. Ahn, Prof. C. Nuckolls,
Dr. M. L. Steigerwald
Department of Chemistry, Columbia University
New York, NY 10027 (USA)
E-mail: cn37@columbia.edu
mls2064@columbia.edu
Homepage: http://nuckolls.chem.columbia.edu/
Prof. R. A. Lalancette
Department of Chemistry, Rutgers State University
Newark, NJ 07102 (USA)
We measured the conductance of both the individual QDs
2–5, and the free connectors, L2–L5 using a scanning tunnel-
ing microscope based break-junction (STM-BJ) technique.[19]
STM-BJ measurements use a gold tip and gold substrate to
repeatedly form and break gold point contacts in solutions of
the target compounds in 1,2,4-trichlorobenzene as solvent.
Clusters 2–4 can bind to the Au electrodes through thiomethyl
groups[25] whereas 5, which lacks thiomethyl groups, cannot.
The conductance across the Au gap is measured versus the tip/
substrate separation at an applied voltage of 500 and 750 mV
for L2–L5 and 2–5, respectively. In the inset of Figure 3a and
b, we show sample traces measured for 2, 4, L2, and L4. These
conductance traces show plateaus with lengths and conduc-
tance that characterize each compound, indicating that each
forms junctions.
Prof. L. Venkataraman
Department of Applied Physics and Applied Mathematics
Columbia University, New York, NY 10027 (USA)
E-mail: lv2117@columbia.edu
[**] This work was supported primarily through the Center for Re-
Defining Photovoltaic Efficiency Through Molecular-Scale Control,
an Energy Frontier Research Center (EFRC) funded by the U.S.
Department of Energy (DOE), Office of Science, Office of Basic
Energy Sciences under award number DE-SC0001085. L.V. thanks
the Packard Foundation for support. X.R. thanks the Natural
Sciences and Engineering Research Council of Canada for a post-
doctoral fellowship. C.L.S. is supported by the National Science
Foundation Graduate Research Fellowship under award number
DGE-1144155. We thank Brian Capozzi for help with the STM-BJ
measurements.
We created one-dimensional (1D) conductance and two-
dimensional (2D) conductance–displacement histograms
from the conductance traces.[21] Figure 3a and b show 1D
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206301.
Angew. Chem. Int. Ed. 2012, 51, 12473 –12476
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12473

.
Angewandte
Communications
Figure 1. Making electrical contact to an atomically precise QD. a) Schematic of a single-cluster junction formed between nanoscale electrodes
and molecular connectors. b) Molecular structure of the parent cluster Co6Se8(PEt3)6 (1). Carbon, black; cobalt, blue; phosphorus, orange;
selenium, green. c) Chemical structures of the different ligand investigated in this study.
previous measurements showing that longer mol-
ecules can bind further away from the apex of the
Au electrodes and change their binding site on the
electrode as the junctions are elongated.[21,26]
Cluster 5 lacks aurophilic thiomethyl function-
ality. Neither the 1D nor 2D conductance histo-
grams for 5 (Figure S2 and S3) shows peaks as that
of 4. This suggests that cluster 4 forms molecular
junctions by bonding its terminal thiomethyl
groups to the Au electrodes, while 5 does not.
Comparison of 2 and 3 demonstrates the effect
of the connector substitution pattern on the
conductance of the cluster. Although the meta-
substituted connector L3 shows a clear conduc-
tance peak that we ascribe to s-conduction,[27–29]
cluster 3 shows no peak (Figure S2). This indicates
that no end-to-end electronic pathway exists in 3.
There is also no stepwise pathway by which
a carrier can travel from one electrode to the
cluster core and then to the second electrode. Thus
when L3 binds to the cluster its 3-thiomethyl-
Figure 2. SCXRD characterization of the clusters. a–d) Molecular structures of
clusters 2–5, respectively. The hydrogen atoms and solvent molecules of crystalliza-
tion have been omitted. Carbon, black; cobalt, blue; phosphorus, orange; sulfur,
yellow; selenium, green. The Co-P, Co-Se, and Co-Co bond lengths for clusters 1–5
are in the range of 2.12–2.14 ꢀ, 2.32–2.36 ꢀ, and 2.88–2.97 ꢀ, respectively. These
distances change little in a given cluster and throughout the cluster series. The
methyl group on one of the sulfur atoms in cluster 4 is disordered between two
orientations (the site-occupancy-factor is 0.75:0.25).
À
phenyl substituent does not rotate around the P C
bond to enable conduction through a s-pathway
between the sulfur and the cluster core. These
results show that we can effectively modulate the
conductivity of a QD device by tuning the
conductance histograms generated using logarithm bins for 2,
4, L2, and L4. The histograms for the clusters do not overlap
those of the corresponding connectors confirming that stable
cluster junctions indeed form with 2 and 4. The lower
conductance of 2 and 4, when compared with that of L2 and
L4 is consistent with longer molecules spanning the junctions.
The heights of the conductance peak for the clusters decrease
after measurement of several thousand traces, possibly
because of degradation under ambient conditions.
The 2D conductance–displacement histograms for 2 and
4, shown in Figure 3c and d, extend to about 0.7 and 1.8 nm,
respectively. This is significantly longer than for the corre-
sponding connectors (2D histograms for L2 and L4 are shown
in the Supporting Information in Figure S1) and agrees with
chemistry of the connectors by varying the substitution
pattern or removing of the aurophilic group.
Electronic absorption spectroscopy, cyclic voltammetry,
1H and 31P NMR spectroscopies, and electronic structure
calculations further characterize these molecular circuit
elements. The absorption spectra (Figures S4–S8) show that
modifying the connectors changes the Co6Se8 core little. The
three longer-wavelength absorptions that characterize
1 remain essentially unchanged in 2–5. The connectors L2–
L5 absorb in the near-UV and 2–5 show similar absorptions.
The spectra of the clusters are simply the sum of the spectra of
the isolated constituent parts. This conclusion is supported by
voltammetry of 2–5 (Figures S9–S12). Each cyclic voltammo-
gram shows one reversible reduction and two reversible
12474 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12473 –12476

Angewandte
Chemie
and they are much more sensitive to air than the
others. For example, we can record sharp, well-
1
defined H NMR spectra for all of the clusters, but
the spectra of 2 and 4 broaden rapidly after the
samples are exposed to air while those of the other
three clusters remain sharp. We suggest that these
two features, molecular conductance and chemical
reactivity, are two facets of the same fundamental
property: access to the Co6Se8 core that is granted
by L2 and L4 but forbidden by PEt3, L3, and L5.
We and others have shown that 1,4-disubstitu-
tion on phenyl rings can give conductive, conjugated
systems whereas compositionally similar 1,3-disub-
stitution gives insulating, cross-conjugated sys-
tems.[27–29] The present results extend this to include
substituted phenyl phosphine ligands in metal-con-
taining systems, and we use density functional
theory to study this. We modeled 2, 3 (Figure 4),
and
4 (Figure S13) with the simpler clusters,
(PMe3)5Co6Se8(L2) and (PMe3)5Co6Se8(L3), and
(PMe3)5Co6Se8(L4), respectively. The electronic
structures of the three model clusters are very
similar; the salient difference is in the orbitals that
are most nearly identified with the pp lone pairs on
the sulfur atoms. Comparison of these orbitals in the
model clusters indicates that the thiomethyl sub-
stituents are coupled more strongly to the cluster
core in 2 and 4 than in 3. We believe that the
corresponding orbitals in 2 and 4 provide conduits
through which electrons may move from the cluster
to its ambient surroundings—either to effect elec-
trical conduction in the break-junction or to medi-
Figure 3. Single-cluster junctions. Logarithm-binned conductance histograms con-
structed using over 5000 traces for a) connector L2 (pink) and cluster 2 (red) and
b) connector L4 (light blue) and cluster 4 (blue). The bin size is 100/decade. The
insets show individual conductance traces. c and d) 2D conductance histograms
for clusters 2 and 4, respectively. The conductance peaks extend over a distance of
0.7 nm for 2 and 1.8 nm for 4 relative to the break of the gold point contact.
oxidations—identical behavior (and at essentially identical
potentials) to 1. We see no redox processes for the connectors
on their own (L2–L5). Thus multiple charged states are
reversibly accessible in 2–5: the cluster core contains
a number of stored charge carriers that can be transferred
onto a macroscopic electrode. This voltammetry is comple-
mentary to the STM-BJ experi-
ate reaction with oxygen.
We supplemented these calculations with studies in which
the Co6Se8 core was protected with four spectator PMe3
ligands and two of the thiomethyl-containing ligands
((PMe3)4Co6Se8(L)2) to characterize the electronic communi-
cation between antipodal aurophilic sites. We observe that the
ments that show the transport of
charge carriers through conductive
molecular
connectors.
The
1H NMR spectra of 2–5 are essen-
tially the same as for L2–L5. The
single 31P NMR resonance for 2–5 is
significantly broader than and is
shifted downfield from that of L2–
L5, respectively, by about 75 ppm.
These data show that in most
ways the clusters 1–5 are essentially
identical; the differences in elec-
tronic absorption, chemical struc-
ture (determined both in the solid
by SCXRD and in solution by
NMR spectroscopy), and electronic
structure (determined by cyclic vol-
tammetry) are minor. In only two
aspects do these clusters differ: 2
Figure 4. Model computational studies of clusters 2 and 3 using density functional theory. The
orbitals associated with the sulfur pp lone pairs for the models (PMe3)5Co6Se8(L2),
and 4 are electrically conductive, (PMe3)5Co6Se8(L3), (PMe3)4Co6Se8(L2)2, and (PMe3)4Co6Se8(L3)2 are shown.
Angew. Chem. Int. Ed. 2012, 51, 12473 –12476
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12475

.
Angewandte
Communications
[8] B. M. Boardman, J. R. Widawsky, Y. S. Park, C. L. Schenck, L.
Venkataraman, M. L. Steigerwald, C. Nuckolls, J. Am. Chem.
Soc. 2011, 133, 8455 – 8457.
[9] W. A. Tisdale, Science 2010, 328, 1543 – 1547.
[10] J. B. Sambur, T. Novet, B. A. Parkinson, Science 2010, 330, 63 –
66.
[11] A. J. Nozik, Inorg. Chem. 2005, 44, 6893 – 6899.
[12] D. V. Talapin, C. B. Murray, Science 2005, 310, 86 – 89.
[13] I. Robel, V. Subramanian, M. Kuno, P. V. Kamat, J. Am. Chem.
Soc. 2006, 128, 2385 – 2393.
[14] S. Geyer, V. J. Porter, J. E. Halpert, T. S. Mentzel, M. A. Kastner,
M. G. Bawendi, Phys. Rev. B 2010, 82, 155201.
[15] Y. Liu, M. Gibbs, J. Puthussery, S. Gaik, R. Ihly, H. W. Hillhouse,
M. Law, Nano Lett. 2010, 10, 1960 – 1969.
[16] A. T. Fafarman, W.-k. Koh, B. T. Diroll, D. K. Kim, D.-K. Ko,
S. J. Oh, X. Ye, V. Doan-Nguyen, M. R. Crump, D. C. Reifs-
nyder, C. B. Murray, C. R. Kagan, J. Am. Chem. Soc. 2011, 133,
15753 – 15761.
[17] T. P. Osedach, N. Zhao, T. L. Andrew, P. R. Brown, D. D.
Wanger, D. B. Strasfeld, L.-Y. Change, M. G. Bawendi, V.
Bulovic, ACS Nano 2012, 6, 3121 – 3127.
essential electronic structure of the cluster is unchanged; the
significant differences again appear in the orbitals most
readily associated with the sulfur pp lone pairs (and also the
=
C C bond in the case of L4). We find that there is a clear,
single-orbital pathway between the two antipodal points in
(PMe3)4Co6Se8(L2)2 and (PMe3)4Co6Se8(L4)2 that is absent in
(PMe3)4Co6Se8(L3)2. These results are consistent with our
experimental observations that 2 and 4 are electrically
conductive and air-sensitive.
The compound Co6Se8(PEt3)6 is part of a large and varied
family of metal chalcogenide molecular clusters that includes
Chevrel phase analogues such as Cr6Te8(PEt3)6,[30] Mo6S8-
(PEt3)6,[31,32] and W6S8(PCy3)6,[4,33] and larger compounds that
contain tens to hundreds of metal atoms.[34] We anticipate that
our approach will provide a route to integrating clusters with
tunable optical, electronic, and magnetic properties into
electrical circuits and devices.
Quantum dots can act as reservoirs of electrical carriers or
electronic excitations but this is valuable only to the extent
that the charge or excitation can be removed from the cluster.
In this study, we demonstrate a method to make contact and
extract charge from one simple prototype. These results thus
provide design rules for the preparation of structurally and
electronically discrete quantum dots at molecular scale that
reliably connect to nanoscale electrodes in a selective, well
understood, and controllable fashion. Our work paves the
way to incorporating these molecular electronic elements into
circuits. We anticipate that this approach will be widely
applicable to other quantum dot systems, thereby enabling
a multitude of studies including the extraction of charges from
multiple exciton states and extraction of hot carriers.
[18] S. M. Stuczynski, Y. U. Kwon, M. L. Steigerwald, J. Organomet.
Chem. 1993, 449, 167 – 172.
[19] B. Q. Xu, N. J. J. Tao, Science 2003, 301, 1221 – 1223.
[20] A. Nitzan, M. A. Ratner, Science 2003, 300, 1384 – 1389.
[21] M. Kamenetska, M. Koentopp, A. C. Whalley, Y. S. Pakr, M. L.
Steigerwald, C. Nuckolls, M. S. Hybertsen, L. Venkataraman,
Phys. Rev. Lett. 2009, 102, 126803.
[22] W. Maneeprakorn, M. A. Malik, P. OꢂBrien, J. Mater. Chem.
2010, 20, 2329 – 2335.
[23] M. L. Steigerwald, Polyhedron 1994, 13, 1245 – 1252.
[24] M. L. Steigerwald, T. Siegrist, S. M. Stuczynski, Inorg. Chem.
1991, 30, 4940 – 4945.
[25] Y. S. Park, A. C. Whalley, M. Kamenetska, M. L. Steigerwald,
M. S. Hybertsen, C. Nuckolls, L. Venkataraman, J. Am. Chem.
Soc. 2007, 129, 15768 – 15769.
[26] J. S. Meisner, M. Kamenetska, M. Krikorian, M. L. Steigerwald,
L. Venkataraman, C. Nuckolls, Nano Lett. 2011, 11, 1575 – 1579.
[27] S.-H. Ke, W. Yang, H. U. Baranger, Nano Lett. 2008, 8, 3257 –
3261.
Received: August 6, 2012
Published online: November 7, 2012
[28] D. M. Cardamone, C. A. Stafford, S. Mazumdar, Nano Lett.
2006, 6, 2422 – 2426.
[29] S. V. Aradhya, J. S. Meisner, M. Krikorian, S. Ahn, R. Para-
meswaran, M. L. Steigerwald, C. Nuckolls, L. Venkataraman,
Nano Lett. 2012, 12, 1643 – 1647.
[30] B. Hessen, T. Siegrist, T. Palstra, S. M. Tanzler, M. L. Steiger-
wald, Inorg. Chem. 1993, 32, 5165 – 5169.
[31] T. Saito, N. Yamamoto, T. Yamagata, H. Imoto, J. Am. Chem.
Soc. 1988, 110, 1646 – 1647.
[32] S. Jin, F. Popp, S. W. Boettcher, M. Yuan, C. M. Oertel, F. J.
DiSalvo, J. Chem. Soc. Dalton Trans. 2002, 3096 – 3100.
[33] S. Jin, D. Venkataraman, F. J. DiSalvo, Inorg. Chem. 2000, 39,
2747 – 2757.
[34] J. F. Corrigan, O. Fuhr, D. Fenske, Adv. Mater. 2009, 21, 1867 –
1871.
Keywords: charge extraction · molecular wiring · quantum dots ·
scanning probe microscopy · single-molecule electronics
.
[1] M. G. Bawendi, M. L. Steigerwald, L. E. Brus, Annu. Rev. Phys.
Chem. 1990, 41, 477 – 496.
[2] P. A. Alivisatos, Science 1996, 271, 933 – 937.
[3] D. V. Talapin, J.-S. Lee, M. V. Kovalenko, E. V. Shevchenko,
Chem. Rev. 2010, 110, 389 – 458.
[4] M. Yuan, B. ꢀlgꢁt, M. McGuire, K. Takada, F. J. DiSalvo, S. Lee,
H. AbruÇa, Chem. Mater. 2006, 18, 4296 – 4306.
[5] D. L. Klein, R. Roth, A. K. L. Lim, P. A. Alivisatos, P. L.
McEuen, Nature 1997, 389, 699 – 701.
[6] M.-H. Jo, Nano Lett. 2006, 6, 2014 – 2020.
[7] D. I. Gittins, D. Bethell, D. J. Schiffrin, R. J. Nichols, Nature
2000, 408, 67 – 69.
12476 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12473 –12476