A Chemically Soldered Polyoxometalate Single-Molecule Transistor

Article abstract of DOI:10.1002/anie.202002174

Polyoxometalates have been proposed in the literature as nanoelectronic components, where they could offer key advantages with their structural versatility and rich electrochemistry. Apart from a few studies on their ensemble behaviour (as monolayers or thin films), this potential remains largely unexplored. We synthesised a pyridyl-capped Anderson–Evans polyoxometalate and used it to fabricate single-molecule junctions, using the organic termini to chemically “solder” a single cluster to two nanoelectrodes. Operating the device in an electrochemical environment allowed us to probe charge transport through different oxidation states of the polyoxometalate, and we report here an efficient three-state transistor behaviour. Conductance data fits a quantum tunnelling mechanism with different charge-transport probabilities through different charge states. Our results show the promise of polyoxometalates in nanoelectronics and give an insight on their single-entity electrochemical behaviour.

Full text of DOI:10.1002/anie.202002174

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A chemically soldered polyoxometalate single-molecule transistor
Authors: Chuanli Wu, Xiaohang Qiao, Craig M. Robertson, Simon J.
Higgins, Chenxin Cai, Richard J. Nichols, and Andrea Vezzoli
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202002174
Link to VoR: https://doi.org/10.1002/anie.202002174

10.1002/anie.202002174
Angewandte Chemie International Edition
A chemically soldered polyoxometalate
single-molecule transistor†
Chuanli Wu, Xiaohang Qiao, Craig M. Robertson, Simon J. Higgins, Chenxin Cai, Richard J.
Nichols and Andrea Vezzoli*
This work is dedicated to the memory of Professor NJ Tao.
† Ms. Chuanli Wu, Ms. Xiaohang Qiao, Dr. Craig M. Robertson, Prof. Simon J. Higgins, Prof. Richard J. Nichols
and Dr. Andrea Vezzoli
Department of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, United Kingdom
e-mail for Dr. Andrea Vezzoli: andrea.vezzoli@liverpool.ac.uk
Ms. Chuanli Wu and Prof. Chenxin Cai
School of Chemistry and Materials Science, Nanjing Normal University, Nanjing 210023, People’s Republic of
China
Dr. Andrea Vezzoli
Stephenson Institute for Renewable Energy, University of Liverpool, Peach Street, Liverpool L69 7ZF, United
Kingdom
* corresponding author
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Abstract
Polyoxometalates have been proposed in the literature as nanoelectronic components, where they
could offer key advantages with their structural versatility and rich electrochemistry. Apart from a few
studies on their ensemble behaviour (as monolayers or thin films) this potential remains largely
unexplored. We synthesised a pyridyl-capped Anderson-Evans polyoxometalate and used it to fabricate
single-molecule junctions, using the organic termini to chemically “solder” a single cluster to two
nanoelectrodes. Operating the device in an electrochemical environment allowed us to probe charge
transport through different oxidation states of the polyoxometalate, and we report here an efficient three-
state transistor behaviour. Conductance data fits a quantum tunnelling mechanism with different charge
transport probabilities through different charge states. Our results show the promise of
polyoxometalates in nanoelectronics and give an insight on their single-entity electrochemical
behaviour.
Introduction
Assemblies of molecules, or even single molecules, might in the future offer new approaches to
traditional nanoelectronics and nanofabrication, and contribute to scalability with a bottom-up self-
assembly approach and miniaturisation.[1] A key aspect of molecular electronics from a chemists’
perspective is however the possibility of exploring a large chemical space to (i) tune the device response
to the desired range and magnitude, obtaining in turn important insights on structure-properties
relationships and (ii) study phenomena unique to the molecular and quantum world. Since the first
pioneering studies in the late 1990s,[2] single-molecule devices with behaviour like semiconductor-
based diodes,[3,4] resistors,[5] switches[6–9] and transistors[10–13] have been demonstrated, and the
chemical complexity of the molecules used to fabricate junctions has rapidly increased. Stemming from
the original studies employing simple aliphatic and conjugated, rod-like, oligoaryl moieties,[14] it is now
common to read reports on molecular wires incorporating fused polyaromatic/heterocyclic systems,[15,16]
supramolecular complexes,[17,18] organometallic centres[19,20] and, more recently, polynuclear
clusters.[21–24] The latter are particularly interesting from a technology point of view for their electronic
behaviour, as the presence of multiple metallic centres imparts stability to several oxidation states, with
the cluster accommodating large charge variations. These electron-sink phenomena[25,26] (i.e. clusters
are able to accept and release electrons reversibly without significant changes in their structure)
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
resulted in great interest in their synthesis and applications, especially as electroactive materials to be
deployed in electronic, sensing and catalytic devices.[27,28]
Polyoxometalates (POMs) are a class of cluster compounds noteworthy for their stability and rich
electrochemistry.[29] The relatively high oxidation state of the metallic centres linked together by 휇2-oxido
ligands gives them less susceptibility to decomposition both in ambient atmosphere and in solution,
resulting in stability over multiple redox states. In recent years, much effort has been devoted to the
functionalisation of POMs with organic moieties,[30] to impart properties such as biocompatibility[31] and
luminescence,[32,33] and to provide additional metal-binding sites for the synthesis of coordination
polymers and three-dimensional frameworks.[34] Metallophilic termini can however be also exploited in
molecular electronics as electrode anchoring groups,[35,36] to ensure clear junction formation and provide
a strong mechanical stability and high electronic coupling between the molecule and the source/drain
electrodes. While perspectives on the promise of polyoxometalates as molecular electronic components
have indeed been featured in publications,[37,38] experimental reports are limited to a few niche studies,
mostly on their behaviour as a nanoscale ensemble (i.e. a monolayer or a thin film)[22,39,40] or immobilised
on a metal surface.[41] We therefore focussed on the synthesis of the Anderson-Evans cluster (Figure
1a and 1b) functionalised with pyridyl ligands 1, (NC5H4)-C-(CH2O)3≡[MnMo6O18]≡(OCH2)3-C-(C5H4N),
to then use it as molecular wire for the fabrication of chemically soldered[21] single-polyoxometalate
junctions. In this contribution, we explore its single-entity electrochemistry, and we find a remarkable
three-state transistor behaviour, with both ON/OFF ratios exceeding one order of magnitude.
Results and Discussion
We followed a published procedure[34] for the synthesis of compound 1 (more details in the SI) as a
tris(tetrabutylammonium) salt. To confirm the structure of the polyoxometalate, we grew crystals by
vapour diffusion of diethyl ether into a concentrated solution of 1 in dimethylformamide and obtained
samples suitable for XRD (Figure 1a-b).
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Figure 1: Details on the experiments presented in this study: a) top view of 1 (SCXRD structure).
b) Side view of 1 (SCXRD structure). See SI for more details on the structure of 1. c) Cyclic
voltammetry of 1 (10 mM) in in DMF with 0.1 M NBu4BF4 as support electrolyte, using a glassy
carbon working electrode, Pt wire counter electrode and Pt wire quasi reference electrode.
Ferrocene was added at the end of the experiment as a standard (see SI for calibration and
additional data). Scan speed 100 mV/s d) Structure of the ionic liquid BMIM-OTf used in this study.
e) Depiction of the single-POM junction in the 4-electrode configuration controlled by the
bipotentiostat. WE = working electrode; CE = counter electrode; RE = reference electrode. Legend
for panels (a) and (b): teal polyhedra: {MoO6}; orange polyhedra: {MnMo6}; grey spheres: C; blue
+
spheres: N; red ellipsoids: O. H atoms and counterions (3 x NBu4 ) not shown for clarity.
The compound itself showed an interesting electrochemical behaviour at a glassy carbon electrode.
Two quasi-reversible redox couples can be observed, and these have been related to the reduction and
oxidation of the central Mn atom (MnIII ⇄ MnII and MnIII ⇄ MnIV), a behaviour already found in amine-
and TTF-capped analogues.[42] For the pyridyl-capped POM 1, the reduction process is very clearly
electrochemically irreversible, with a peak-to-peak separation of ~600 mV at low scan rates. The
oxidation process is less irreversible, with a reduced peak-to-peak separation of ~180 mV (see SI for
more details).
We then fabricated junctions using the scanning tunnelling microscopy break junction (STM-BJ)
method,[43] by driving a Au tip towards a Au substrate under constant DC bias until it crashes and forms
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
a metallic contact having conductance > 5퐺ꢀ (퐺ꢀ = conductance quantum, 2푒ꢁ ℎ ≅ 77.48 휇푆). The tip
is then withdrawn at constant speed, and the metallic contact is thinned down to a single atom (having
conductance = 퐺ꢀ) and then broken. When the experiment is performed in a solution of a molecular
wire having metal-binding moieties at its termini, molecular junctions self-assemble in the freshly formed
nanogap. The process is repeated thousands of times while recording the current, and the conductance
⁄
⁄
of the junctions is then calculated as 퐺 = 퐼 푉ꢂꢃꢄꢅ. The traces obtained in the process (as current vs
electrode position) are then analysed statistically as histograms and density maps, to calculate the most
probable conductance value. The experiments were performed under electrochemical control, using a
bipotentiostat and a 4-electrode cell, with the Au substrate as working electrode, a platinum counter
electrode and a platinum reference electrode (Figure 1e). This setup was found to have an open circuit
potential of -0.23 V vs Fc/Fc+. The STM tip was coated with Apiezon wax[44] to reduce faradaic currents,
and constantly biased against the working electrode by the bipotentiostat, to ensure the two Au
electrodes act as source and drain. A tip-substrate bias of 200 mV was used throughout this study as
the smallest bias voltage to maximise the signal-to-noise ratio. The measurements were performed with
the four electrodes immersed in the ionic liquid BMIM-OTf (Figure 1c) with a 1 mM concentration of
compound 1. The use of ionic liquids ensures a high degree of molecule-gate coupling[10] and they are
an ideal medium for STM experiments and measurements, with low faradaic current attainable with
insulated tips and a wide electrochemical window.[45–47] More details about the STM-BJ experiments,
the electrochemistry of 1 in ionic liquid and the equipment used in this study can be found in the methods
section and in the SI.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Figure 2: Single-molecule electrochemical gating experiments on 1. a) heatmap of conductance
across the electrochemical window explored, with overlaid in white the gaussian fitting of the
conductance peaks. Double points are shown in the vicinity of the switching potential. b)
Conductance histograms for compound 1 at -0.2 V and -0.7 V electrochemical potential. c)
Conductance histogram of 1 at 0.6 V and 1.2 V potential. d) 2D conductance vs electrode
separation density map at 0.4 V potential. e) 2D conductance vs electrode separation density map
at 1.2 V potential. Heatmap in (a) compiled with 10 bins per conductance decade. Conductance
histograms in (b) and (c) compiled with 20 bins per conductance decade. 2D density maps in (d)
and (e) compiled with 100 bins per conductance decade and 50 bins per nanometre. More than
3500 individual breaking traces have been collected at each potential. In all heatmaps and density
plots green = low counts and red = high counts. The full dataset is presented in the SI.
The main results are shown in Figure 2. We explored the bias window between 1.2 V and -1.5 V vs Pt,
which is attainable in the ionic liquid we used and where 1 exhibited three distinct charge states. As can
be observed in Figure 2a, when the electrochemical potential is moved from zero towards more negative
values, there is a slight increase in conductance as the potential acts as an electrostatic gate and moves
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
the Fermi level towards a transport resonance. However, when the potential reaches -0.3 V vs Pt, the
cluster accepts an electron and its charge state switches[42,48] to -4, with a strong effect on the
conductance that drops by more than one order of magnitude, from 10-3.4 G0 to 10-4.6 G0. Example
conductance histograms pertinent to the -3/-4 switch are given in Figure 2b. Close to the equilibrium
potential, both charge states are present, and they almost equally contribute to the conductance
histogram (see all histograms in the SI and later in the manuscript for the analysis). As the potential is
made more negative, the electrostatic gate effect is reinstated, and conductance gradually climbs to 10-
4.3 G0. On the other hand, as the potential is made more positive from zero, the conductance is gradually
reduced due to the electrostatic gate effect, until the charge state of the POM switches to -2, and its
value falls to almost 10-5 G0. The effect of the -3/-4 switch on the conductance histograms can be seen
in Figure 2c. Statistical analysis of the conductance-electrode separation traces as 2D density plots
(see for example Figure 2d-e and the SI for the full dataset) shows that junction rupture length remains
constant, at approximately 0.4 – 0.5 nm, as 1 is oxidised or reduced. Significant structural or
configurational rearrangements as the oxidation state of 1 is switched can therefore be discounted.
Accounting for the electrode snapback (0.65 nm at room temperature)[7] we can calculate a junction
length at rupture of 1.05 – 1.15 nm, only slightly shorter than the molecular N–N distance obtained from
SCXRD (1.4 nm).
Two main things can be inferred from the above results. First, the observed electrostatic gate effect,
where conductance increases as the potential is made more negative, strongly suggests that charge
transport happens through a non-resonant, one-step tunnelling mechanism, assisted by the molecular
LUMO (pyridyls are known[36] to promote LUMO-based conductance). Second, that the oxidation and
reduction of the central Mn atom have a profound effect on the electronic structure of the POM, as the
conductance is electrochemically gated by approximately an order of magnitude in both cases. The
confirmation of a tunnelling mechanism, and the clear absence of a bell-shaped conductance
enhancement at the equilibrium potential, discount a possible interpretation based on a two-step
Kuznetsov-Ulstrup process (a hopping-type model associated with a conductance increase near the
redox potential).[49,50] A Nernstian model based on charge tunnelling through different redox states of a
molecule (in thermodynamic equilibrium depending on the applied electrochemical potential) has been
developed to aid the interpretation of data obtained from junctions based on anthraquinone-substituted
DNA[51] or electroactive ferrocenes.[52] In ferrocenes, which lie closer in size to our present system, the
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
mechanism of charge transport was unambiguously confirmed to be phase-coherent tunnelling by
analysis of the conductance decay with length, and the average conductance of the junction under
electrochemical control is given by the equation:
1
̅
(
)
퐺 = 퐺ꢆ +
퐺ꢆ − 퐺ꢁ
Eq. 1
푒
1 + 푒푥푝 ꢇ
(퐸 − 퐸ꢀ)ꢉ
ꢈ
푘ꢂ푇
Where 퐺ꢆ and 퐺ꢁ are, respectively, the conductance of the molecular junction in the higher and lower
oxidation states, 푒 is the electron charge, 푘ꢂ is the Boltzmann constant, 푇 is the temperature, 퐸 is the
̅
electrochemical potential and 퐸ꢀis the equilibrium potential. We calculated 퐺 as weighted arithmetic
mean of the conductance histograms Gaussian fitting:
퐺ꢆ퐴ꢆ + 퐺ꢁ퐴ꢁ
̅
퐺 =
Eq. 2
퐴ꢆ + 퐴ꢁ
using the peak centre as the value for 퐺 and the area under the Gaussian curve as 퐴, after removal of
the background tunnelling signal obtained from a control experiment performed in the pure, anhydrous
solvent. The weighted average of the experimental points and the Nernstian model are in good
agreement, as can be observed in in Figure 3. It is worth stressing that the Nernstian model is not a
fitting to the experimental conductance values, as no variable was parameterised. The agreement
between Eq. 1 and the empirical data thereby confirms non-resonant tunnelling as the dominant charge
transport mechanism, with the molecule either reduced or oxidised in the junction, depending on the
potential.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Figure 3: Gaussian fitting and Nernstian model. Examples of gaussian fitting of conductance
histograms of 1 at an electrochemical potential of (a) -0.3 V and (b) -0.4 V, with the background
tunnelling signal subtracted, used for the calculations of Gn and Sn. Further details on the
background subtraction process are available in the SI. c) Average conductance (orange dots) vs
equilibrium potential for the reduction process (E0 = -0.3 V vs Pt). d) Average conductance (orange
dots) vs equilibrium potential for the oxidation process (E0 = 0.9 V vs Pt). The grey line in (c) and
(d) is the Nernstian model obtained from Eq. 1. Histograms in (a) and (b) compiled with 100 bins
per decade.
To further test the robustness of the junctions fabricated with the polyoxometalate 1, we performed
voltage ramps on stabilised junctions, in a 2-terminal device configuration. In this technique, a staircase
ramp is applied to the piezo controlling the tip position, so that a specific position is held for 100 ms.
The height of the staircase (1 nm) was chosen to ensure the formation of a gap of size commensurate
with the molecular length. During the hold portion of the ramp, the bias voltage is kept at a fixed value
(100 mV) for 25 ms, then ramped to obtain an 퐼/푉 characteristics or a high-voltage test, and then kept
again at the fixed value for the final 25 ms (Figure 4a). Data is then sliced and processed using the
automated algorithms described in the SI, to produce the 2d density maps presented in Figure 4. The
퐼/푉 characteristics show a good ohmic behaviour up to relatively high biases (~ 0.7 V, see SI for linear
scale plots) with no evident asymmetry or rectification arising from the presence of a molecular orbital
with energy sufficiently close[53,54] to the Fermi levels of the gold electrodes under the two-terminal
conditions employed. This further confirms off-resonant tunnelling as the charge transport mechanism,
and this would also account for the observed relatively low conductance. As a stress test, we drove the
junctions to biases higher than the one used to obtain the 퐼/푉 behaviour, and we found excellent
robustness under positive substrate bias, with the fabricated junctions surviving voltages up to 1.5 V.
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Figure 4: Two-terminal bias modulation experiments. Example (a) raw 퐼/푉 data (-1 to 1 V sweep)
and (b) raw high bias ramp curve (0.1 to 1.5 to 0.1 V sweep), with piezo and bias signal
superimposed. c) 2d density map of the 퐼/푉 characteristics for compound 1, compiled from 1049
traces. d) 2d density map of current vs time during the high voltage ramp (superimposed in light
grey), compiled from 922 traces. 2d maps in (c) and (d) compiled with 100 bins per current decade,
200 bins per volt and 2000 bins per second.
Conclusions
In conclusion, we have demonstrated here the use of a functionalised Anderson-Evans polyoxometalate
in a three-state electrochemical transistor single-molecule device. The MnMo6O24 oxide cluster core is
capped with two 4-pyridyl moieties that act as contact to the electrodes, allowing the fabrication of
single-POM junctions. The junctions showed a clear OFF-ON-OFF behaviour under electrochemical
control, with a difference in conductance of more than one order of magnitude between two adjacent
charge states. The transistor-like behaviour arises from the POM being in three distinct charge states
(-2, -3 and -4) depending on the applied electrochemical potential, and a clear electrostatic gating is
further visible in the conductance vs potential plot. The results can be modelled within the framework
of non-resonant tunnelling, with the charge states having a different tunnelling probability depending on
the charge state of the molecular wire. The fabricated junctions are robust even under high bias, further
showing the promise that POMs have in the field of molecular electronics, thanks to their multiple
charge/oxidation states and high stability, and our results pave the way to a further use of such
compounds in cluster electronics.
10
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Data Availability
The raw experimental STM-BJ data under electrochemical control, the processed data and the Python
code used for its analysis can be found on the University of Liverpool data repository at DOI:
10.17638/datacat.liverpool.ac.uk/962. The raw experimental STM-BJ data used to compile the 퐼/푉
characteristics and high bias behaviour, and the Labview code used for its analysis can be found on
the University of Liverpool data repository at DOI: 10.17638/datacat.liverpool.ac.uk/967.
The SCXRD data is deposited on the Cambridge Crystallographic Dara Centre (deposition #1979403)
Acknowledgements
The authors thank D. Bethell for useful discussion on the Nernstian model and its interpretation. C.W.
acknowledges funding from the China Scholarship Council (grant no. 201806860023). A.V.
acknowledges funding from the Royal Society (University Research Fellowship URF\R1\191241) and
the School of Physical Sciences of the University of Liverpool (Early Career Researchers & Returners
Fund 2018). This work was further supported by UK EPSRC (grants EP/M005046/1, EP/M029522/1
and EP/M014169/1).
Conflict of Interest
The authors declare no conflict of interest.
References
[1]
[2]
[3]
M. A. Ratner, Nat. Nanotechnol. 2013, 8, 378–81.
M. A. Reed, C. Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science 1997, 278, 252–254.
M. L. Perrin, E. Galán, R. Eelkema, J. M. Thijssen, F. Grozema, H. S. J. van der Zant, Nanoscale
2016, 8, 8919–8923.
[4]
[5]
[6]
B. Capozzi, J. Xia, O. Adak, E. J. Dell, Z.-F. Liu, J. C. Taylor, J. B. Neaton, L. M. Campos, L.
Venkataraman, Nat. Nanotechnol. 2015, 10, 522–527.
M. H. Garner, H. Li, Y. Chen, T. A. Su, Z. Shangguan, D. W. Paley, T. Liu, F. Ng, H. Li, S. Xiao,
et al., Nature 2018, 558, 415–419.
A. K. Ismael, K. Wang, A. Vezzoli, M. K. Al-Khaykanee, H. E. Gallagher, I. M. Grace, C. J.
Lambert, B. Xu, R. J. Nichols, S. J. Higgins, Angew. Chemie Int. Ed. 2017, 56, 15378–15382.
11
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
[7]
S. Y. Quek, M. Kamenetska, M. L. Steigerwald, H. J. Choi, S. G. Louie, M. S. Hybertsen, J. B.
Neaton, L. Venkataraman, Nat. Nanotechnol. 2009, 4, 230–234.
[8]
Z. Liu, S. Ren, X. Guo, Top. Curr. Chem. 2017, 375, 1–33.
[9]
D. Taherinia, C. D. Frisbie, J. Phys. Chem. C 2016, 120, 6442–6449.
[10]
H. M. Osorio, S. Catarelli, P. Cea, J. B. G. Gluyas, F. Hartl, S. J. Higgins, E. Leary, P. J. Low,
S. Martín, R. J. Nichols, et al., J. Am. Chem. Soc. 2015, 137, 14319–14328.
[11]
[12]
[13]
R. J. Brooke, C. Jin, D. S. Szumski, R. J. Nichols, B. Mao, K. S. Thygesen, W. Schwarzacher,
Nano Lett. 2015, 15, 275–280.
S. Hosseini, C. Madden, J. Hihath, S. Guo, L. Zang, Z. Li, J. Phys. Chem. C 2016, 120, 22646–
22654.
N. Darwish, I. Díez-Pérez, P. Da Silva, N. Tao, J. J. Gooding, M. N. Paddon-Row, Angew.
Chemie Int. Ed. 2012, 51, 3203–3206.
[14]
[15]
D. J. Wold, R. Haag, M. A. Rampi, C. D. Frisbie, J. Phys. Chem. B 2002, 106, 2813–2816.
Z. Cai, W.-Y. Lo, T. Zheng, L. Li, N. Zhang, Y. Hu, L. Yu, J. Am. Chem. Soc. 2016, 138, 10630–
10635.
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
E. Leary, B. Limburg, A. Alanazy, S. Sangtarash, I. Grace, K. Swada, L. J. Esdaile, M. Noori, M.
T. González, G. Rubio-Bollinger, et al., J. Am. Chem. Soc. 2018, 140, 12877–12883.
K. Wang, A. Vezzoli, I. M. Grace, M. McLaughlin, R. J. Nichols, B. Xu, C. J. Lambert, S. J.
Higgins, Chem. Sci. 2019, 10, 2396–2403.
W. Zhang, S. Gan, A. Vezzoli, R. J. Davidson, D. C. Milan, K. V. Luzyanin, S. J. Higgins, R. J.
Nichols, A. Beeby, P. J. Low, et al., ACS Nano 2016, 10, 5212–5220.
R. Davidson, J.-H. Liang, D. Costa Milan, B.-W. Mao, R. J. Nichols, S. J. Higgins, D. S. Yufit, A.
Beeby, P. J. Low, Inorg. Chem. 2015, 54, 5487–5494.
Y. Tanaka, Y. Kato, T. Tada, S. Fujii, M. Kiguchi, M. Akita, J. Am. Chem. Soc. 2018, 140, 10080–
10084.
X. Roy, C. L. Schenck, S. Ahn, R. A. Lalancette, L. Venkataraman, C. Nuckolls, M. L.
Steigerwald, Angew. Chemie Int. Ed. 2012, 51, 12473–12476.
J. de Bruijckere, P. Gehring, M. Palacios-Corella, M. Clemente-León, E. Coronado, J. Paaske,
P. Hedegård, H. S. J. van der Zant, Phys. Rev. Lett. 2019, 122, 197701.
C. Tang, L. Chen, L. Zhang, Z. Chen, G. Li, Z. Yan, L. Lin, J. Liu, L. Huang, Y. Ye, et al., Angew.
Chemie Int. Ed. 2019, 58, 10601–10605.
T. C. Ting, L. Y. Hsu, M. J. Huang, E. C. Horng, H. C. Lu, C. H. Hsu, C. H. Jiang, B. Y. Jin, S.
12
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
M. Peng, C. H. Chen, Angew. Chemie Int. Ed. 2015, 54, 15734–15738.
[25]
F. Fabrizi De Biani, C. Femoni, M. C. Iapalucci, G. Longoni, P. Zanello, A. Ceriotti, Inorg. Chem.
1999, 38, 3721–3724.
[26]
[27]
[28]
[29]
[30]
[31]
[32]
T. Albrecht, S. F. L. Mertens, J. Ulstrup, J. Am. Chem. Soc. 2007, 129, 9162–9167.
A. Mathew, T. Pradeep, Part. Part. Syst. Charact. 2014, 31, 1017–1053.
M. Stuckart, K. Y. Monakhov, Chem. Sci. 2019, 10, 4364–4376.
M. Sadakane, E. Steckhan, Chem. Rev. 1998, 98, 219–238.
H. N. Miras, J. Yan, D. L. Long, L. Cronin, Chem. Soc. Rev. 2012, 41, 7403–7430.
A. Blazevic, A. Rompel, Coord. Chem. Rev. 2016, 307, 42–64.
W. Salomon, A. Dolbecq, C. Roch-Marchal, G. Paille, R. Dessapt, P. Mialane, H. Serier-Brault,
Front. Chem. 2018, 6, 1–8.
[33]
P. Bolle, Y. Chéret, C. Roiland, L. Sanguinet, E. Faulques, H. Serier-Brault, P. A. Bouit, M.
Hissler, R. Dessapt, Chem. - An Asian J. 2019, 14, 1642–1646.
[34]
[35]
[36]
F. J. Yazigi, C. Wilson, D. L. Long, R. S. Forgan, Cryst. Growth Des. 2017, 17, 4739–4748.
J. Hihath, N. J. Tao, Semicond. Sci. Technol. 2014, 29, 054007.
E. Leary, A. La Rosa, M. T. González, G. Rubio-Bollinger, N. Agraït, N. Martín, Chem. Soc. Rev.
2015, 44, 920–942.
[37]
[38]
K. Y. Monakhov, M. Moors, P. Kögerler, Adv. Inorg. Chem. 2017, 69, 251–286.
L. Vilà-Nadal, S. G. Mitchell, S. Markov, C. Busche, V. Georgiev, A. Asenov, L. Cronin, Chem. -
A Eur. J. 2013, 19, 16502–16511.
[39]
[40]
[41]
[42]
S. Sherif, G. Rubio-Bollinger, E. Pinilla-Cienfuegos, E. Coronado, J. C. Cuevas, N. Agraït,
Nanotechnology 2015, 26, 291001.
M. Laurans, K. Dalla Francesca, F. Volatron, G. Izzet, D. Guerin, D. Vuillaume, S. Lenfant, A.
Proust, Nanoscale 2018, 10, 17156–17165.
O. Linnenberg, M. Moors, A. Notario-Estévez, X. López, C. De Graaf, S. Peter, C. Baeumer, R.
Waser, K. Y. Monakhov, J. Am. Chem. Soc. 2018, 140, 16635–16640.
A. Boulmier, A. Vacher, D. Zang, S. Yang, A. Saad, J. Marrot, O. Oms, P. Mialane, I. Ledoux, L.
Ruhlmann, et al., Inorg. Chem. 2018, 57, 3742–3752.
[43]
[44]
B. Xu, N. Tao, Science 2003, 301, 1221–1223.
R. Kazinczi, E. Szõcs, E. Kálmán, P. Nagy, Appl. Phys. A Mater. Sci. Process. 1998, 66, S535–
S538.
13
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
[45]
[46]
[47]
T. Albrecht, K. Moth-Poulsen, J. B. Christensen, J. Hjelm, T. Bjørnholm, J. Ulstrup, J. Am. Chem.
Soc. 2006, 128, 6574–6575.
N. J. Kay, R. J. Nichols, S. J. Higgins, W. Haiss, G. Sedghi, W. Schwarzacher, B.-W. Mao, J.
Phys. Chem. C 2011, 115, 21402–21408.
Y. Li, M. Baghernejad, A.-G. G. Qusiy, D. Zsolt Manrique, G. Zhang, J. Hamill, Y. Fu, P.
Broekmann, W. Hong, T. Wandlowski, et al., Angew. Chemie Int. Ed. 2015, 54, 13586–13589.
[48]
[49]
[50]
Y. Yan, B. Li, Q. He, Z. He, H. Ai, H. Wang, Z. Yin, L. Wu, Soft Matter 2012, 8, 1593–1600.
A. M. Kuznetsov, J. Ulstrup, J. Electroanal. Chem. 2004, 564, 209–222.
J. Zhang, A. M. Kuznetsov, I. G. Medvedev, Q. Chi, T. Albrecht, P. S. Jensen, J. Ulstrup, Chem.
Rev. 2008, 108, 2737–2791.
[51]
[52]
L. Xiang, J. L. Palma, Y. Li, V. Mujica, M. A. Ratner, N. Tao, Nat. Commun. 2017, 8, 14471.
Y. Li, H. Wang, Z. Wang, Y. Qiao, J. Ulstrup, H. Y. Chen, G. Zhou, N. Tao, Proc. Natl. Acad.
Sci. U. S. A. 2019, 116, 3407–3412.
[53]
[54]
M. L. Perrin, E. Galan, R. Eelkema, F. Grozema, J. M. Thijssen, H. S. J. Van Der Zant, J. Phys.
Chem. C 2015, 119, 5697–5702.
A. Z. Thong, M. S. P. Shaffer, A. P. Horsfield, Sci. Rep. 2018, 8, 9120.
14
This article is protected by copyright. All rights reserved.

10.1002/anie.202002174
Angewandte Chemie International Edition
Table of Contents
An Anderson-Evans polyoxometalate functionalised with 4-pyridyl
termini has been used to fabricate single-entity junctions in an
electrochemical environment. As the charge state of the
polyoxomolybdate is changed by the electrochemical potential,
fabricated devices showed a tunnelling transistor behaviour, with
OFF-ON-OFF conductance gating by more than one order of
magnitude.
Keywords





Polyoxometalates
Molecular Electronics
Charge Transfer
Molecular Devices
Electrochemical Transistor
15
This article is protected by copyright. All rights reserved.