Skip to main content
NCBI home page
Science Advances logoLink to Science Advances
. 2023 Feb 8;9(6):eadf0425. doi: 10.1126/sciadv.adf0425

A van der Waals heterojunction strategy to fabricate layer-by-layer single-molecule switch

Yu-Ling Zou 1, Qing-Man Liang 1, Taige Lu 1, Yao-Guang Li 1, Shiqiang Zhao 1, Jian Gao 1, Zi-Xian Yang 1, Anni Feng 1, Jia Shi 1, Wenjing Hong 1, Zhong-Qun Tian 1, Yang Yang 1,*
PMCID: PMC9908013  PMID: 36753541

Abstract

Single-molecule electronics offer a unique strategy for the miniaturization of electronic devices. However, the existing experiments are limited to the conventional molecular junctions, where a molecule anchors to the electrode pair with linkers. With such a rod-like configuration, the minimum size of the device is defined by the length of the molecule. Here, by incorporating a single molecule with two single-layer graphene electrodes, we fabricated layer-by-layer single-molecule heterojunctions called single-molecule two-dimensional van der Waals heterojunctions (M-2D-vdWHs), of which the sizes are defined by the thickness of the molecule. We controlled the conformation of the M-2D-vdWHs and the cross-plane charge transport through them with the applied electric field and established that they can serve as reversible switches. Our results demonstrate that the M-2D-vdWHs, as stacked from single-layer 2D materials and a single molecule, can respond to electric field stimulus, which promises a diverse class of single-molecule devices with unprecedented size.

Two single-layer graphene sheets and a single molecule were layered one by one to construct a new type molecular switch.

INTRODUCTION

Two-dimensional (2D) materials had attracted considerable interest in wide research fields, such as physics (1), chemistry (2), materials science (3), and electronics (4, 5). Their intriguing physical and chemical properties allow for the fabrication of versatile devices. In particular, in molecular electronics, graphene had been used to fabricate single-molecule junctions with high stability beyond the limits set by the surficial atom mobility of gold (6–9). On the basis of the nanogapped graphene electrodes, a range of single-molecule devices had been demonstrated, including diode (10), sensor (11, 12), field effect transistors (13, 14), and so on. However, for the as-fabricated graphene-based single-molecule devices, their minimum size is defined by the length of the molecule.

Alternative to these rod-like graphene molecular junctions, van der Waals integration of single-layer graphene electrodes and a single molecule offers a strategy for the further miniaturization of single-molecule electronic devices. The as-fabricated device, called single-molecule 2D van der Waals heterojunction (M-2D-vdWH) (15–17), is a layer-by-layer assembly of a bottom electrode of single-layer graphene, a sandwiched single molecule, and an upper electrode of single-layer graphene. The size of an M-2D-vdWH junction is determined by the thickness of the sandwiched molecule rather than its length. As the thickness is defined by the van der Waals radius of the largest atom in the molecule, it is naturally much smaller than the molecular length in most cases. The investigation of charge transport through molecular junctions could provide vital information for the design of molecular devices. For the conventional rod-like molecular junctions, the electric field had been proved to be a wonderful method in regulating charge transport (18–21) and catalyzing chemical reactions (22–25). Nevertheless, in the layer-by-layer M-2D-vdWHs, charge transports through the molecule with a cross-plane mode, which is distinct from the in-plane charge transport in the rod-like molecular junctions (26, 27). To date, whether and how the electric field will influence the charge transport through a single M-2D-vdWH are largely unexplored, which severely limits its wide applications.

In this work, we report the study on the influence of electric field on the cross-plane charge transport through the M-2D-vdWHs and demonstrate the fabrication of single-molecule switches with a layer-by-layer structure by the van der Waals heterojunction strategy. We fabricated a pair of single-layer graphene electrodes with nanogapped separation using a redesigned break junction setup. By using the van de Waals stacking interaction, we incorporated triphenyl derivatives into the nanogapped graphene layers and successfully fabricated a group of M-2D-vdWH devices. Combined experimental and theoretical investigations confirmed that they featured the layer-by-layer structure, which are distinctly different from the conventional rod-like single-molecule junctions. We further investigated the ability to modulate the charge transport through these molecular junctions by using applied bias and demonstrated their potential application in fabricating single-molecule switches with atomic thickness. This work shows the feasibility of single-molecule van der Waals heterojunction in the miniaturization of electronic devices beyond the conventional rod-like junction and opens an avenue for tuning the conformation and physicochemical properties of the layer-by-layer 2D van der Waals heterojunctions.

RESULTS

Fabrication and electrical characterization of M-2D-vdWHs

We redesigned and assembled a home-built cross-plane break junction (XPBJ) system based on a scanning tunneling microscope break junction setup to fabricate the M-2D-vdWHs of triphenylene (TP), triphenylphosphine (TPP), and triphenylamine (TPA), as shown in Fig. 1A. The rigid plane of TP (28, 29) and the stable triangular pyramidal structure of TPP (30) make them difficult to change the conformations as responded to the electric field. On the contrary, the chemical nature of TPA allows the fabrication of a responsive switch under the electric field. In particular, the thermodynamic barrier between the three-wing propeller (TWP) and triangular cone (TC) conformations of TPA is considerably lower (31). This feature allows us to regulate the conformation of TPA with the electric field and thus to fabricate the electrically controlled single-molecule switch.

Fig. 1. Electrical characterization of M-2D-vdWHs.

Fig. 1.

Open in a new tab

(A) Schematic diagram of the XPBJ setup and the conformational evolution of three M-2D-vdWHs under the applied electric field. Right: The chemical structures of TP, TPP, and TPA that were used to fabricate M-2D-vdWHs. (B to D) The 1D conductance histograms for TP, TPP, and TPA M-2D-vdWHs under the bias of 100, 200, and 300 mV, respectively. The inserts give the typical conductance-displacement traces.

The XPBJ system was equipped with a tailored electrolytic cell. A graphene-coated copper substrate served as the bottom electrode, while a bent graphene-coated copper wire served as the top electrode. During the break junction operation, a piezoelectric actuator was used to control the movement of the top electrode with a feedback circuit, thus controllably constructing the nanogaps in between the top and bottom graphene electrodes. When the size of the nanogap fit the molecular thickness in the electrolytic cell, the M-2D-vdWHs were eventually established between the top and bottom graphene electrodes by the van der Waals interactions. We used a home-built current-voltage (I-V) converter to monitor the formation of M-2D-vdWHs by recording the conductance evolution with a 20-kHz sampling frequency. To prevent the destruction of the M-2D-vdWHs, we set the upper limit of conductance to be 10–2.80 G0 (G0 = 2e2/h, quantum conductance). More than 1000 conductance traces were recorded for each single-molecule conductance measurement to conduct the subsequent statistical analysis. More details about the XPBJ setup, the preparation and characterization of graphene electrode, and the break junction operation were provided in figs. S1 to S3.

To determine the conductance of different states of the M-2D-vdWHs, we carried out the single-molecule conductance measurements with our customized XPBJ system. Before the electric characterization of M-2D-vdWHs, we first performed the control experiment in a pure solvent. Without the probe molecule, the recorded conductance-displacement traces showed no plateau at the conductance regime lower than 10–2.80 G0 (figs. S4 to S6). On the contrary, in the presence of probe molecules, legible plateaus appeared in the recorded conductance traces and peaks emerged in the constructed 1D conductance histograms (Fig. 1, B to D), which indicated the formation of M-2D-vdWHs. We then carried out the concentration gradient experiments (figs. S7 and S8) and found that the conductance value was independent of the concentration. These results confirmed that the measured conductance of the M-2D-vdWHs came from a single molecule.

To study the responsive behavior of M-2D-vdWHs under electric field, we performed the conductance measurements under different biases (fig. S9). For the TP and TPP M-2D-vdWHs (Fig. 1, B and C), there were no discernible differences in their 1D conductance histograms as the electric field changed. On the contrary, for the TPA M-2D-vdWHs (Fig. 1D), the conductance peak of TPA M-2D-vdWHs appeared in the 10–4.58 G0 under 100-mV bias. As the bias increased to 200 and 300 mV, the conductance peak emerged at 10–5.02 G0 and 10–5.36 G0, respectively. The single-molecule conductance value measured under 100 mV was about six times as large as that measured under 300 mV, showing the capability of electric field in modulating the conductance of TPA M-2D-vdWHs.

Thickness analysis of M-2D-vdWHs

To identify the molecular conformation of the M-2D-vdWHs, we adopted 10–3.0 G0 as the reference point to construct the 2D conductance-displacement histograms by overlaying all the conductance traces and then analyzed their distributions of plateau length (Figs. 2A and figs. S10 and S11) (32). For the pure solvent, there was only a peak that emerged at 0.27 nm (fig. S6), as caused by the through-space tunneling between the graphene electrode pair (26). For the length distributions of TP, TPP, and TPA M-2D-vdWHs under the 100 mV, there emerged another peak at 1.00, 1.21, and 1.20 nm, respectively.

Fig. 2. Conductance analysis of the M-2D-vdWHs devices.

Fig. 2.

Open in a new tab

(A) 2D conductance-displacement histograms of TPA M-2D-vdWHs under 200 mV bias. The inset shows the statistical distribution of plateau length. (B) Schematic of a TPA M-2D-vdWHs with the cross-plane charge transport mode. The theoretically simulated thickness is 0.96 nm. (C) Schematic of a TPA M-2D-vdWHs with the in-plane charge transport mode. The theoretically simulated thickness is 1.60 nm. (D) Raman spectra of TP, TPP, and TPA M-2D-vdWHs samples after the XPBJ operation. a.u., arbitrary units. (E) Evolution of formation probability of the M-2D-vdWHs with bias. (F) Distributions of plateau length for the M-2D-vdWHs under different biases.

We turned to simulate the thickness of M-2D-vdWHs with the density functional theory (DFT) calculation. We optimized the graphene molecular junctions of TP, TPP, and TPA for both of the cross-plane charge transport and the conventional in-plane charge transport modes and calculated their theoretical thickness accordingly (fig. S12 and tables S1 and S2). In particular, Fig. 2 (B and C) gives the two modes of TPA M-2D-vdWHs with graphene electrode pair. Their theoretical thicknesses were calculated to be 0.96 and 1.60 nm, which demonstrates the role of van der Waals heterojunction in reducing the size of a single-molecule device. For TPA M-2D-vdWH, the measured value is 1.20 nm, which is closer to the theoretical value of junction with cross-plane mode. Similar comparisons between the theoretical and experimental values are found in the TP and TPP M-2D-vdWHs, implying charge transport through all the as-fabricated M-2D-vdWHs with the cross-plane charge transport mode.

To further confirm the microscopic configuration of M-2D-vdWHs, we used Raman spectroscopy to characterize the coupling between the probe molecules and the graphene electrodes. Previous reports demonstrated that as a graphene layer was layered by a molecule, there would emerge a splitting in the G-band Raman peak of graphene. In comparison, there was no change in the G-band Raman peak if the molecule erected on the graphene because of the small overlapping area (33). In the collected Raman signals for the M-2D-vdWHs, as shown in Fig. 2D, we found that there was clear G-band splitting in all of them. In contrast, in the Raman signal that carried out in the pure solvent (fig. S13), there is no feature of G-band splitting. The combined analysis of the plateau length and Raman signal showed that, in the as-fabricated M-2D-vdWHs, the molecule lies flat on the graphene layer and that the M-2D-vdWHs feature atomic thickness.

Conformational analysis of M-2D-vdWHs under electric field

To explore the response of the M-2D-vdWHs to the electric field, we carried out a quantitative analysis of the through-space tunneling and through-molecule tunneling conductance traces. The change in the molecular conformation will lead to a different formation probability of the molecular junction. Vice versa, the formation probability can be used to monitor the evolution of molecular conformation. Here, the formation probability was calculated by the area ratio of the two peaks in the distribution histograms of plateau length (Fig. 2E and fig. S14), which were contributed by the through-space tunneling and the through-molecule tunneling, respectively (34). With the increase of electric field, the formation probability of TPA M-2D-vdWHs gradually decreased from 76.28% to 53.18% (Fig. 2E). In contrast, there is no discernible change in the formation probability of either TP or TPP M-2D-vdWHs (table S3). This finding indicated that the molecular conformation of TPA M-2D-vdWHs evolved with the electric field.

To identify the molecular conformation of the M-2D-vdWHs, we constructed their 2D conductance-displacement histograms from the data collected under different biases (Fig. 2F). When the bias increased from 100 to 300 mV, the plateau length of TP and TPP M-2D-vdWHs remained at approximately 1.00 and 1.21 nm, respectively. On the contrary, the plateau lengths for TPA M-2D-vdWHs at 100, 200 and 300 mV were 1.20, 1.31, and 1.49 nm, respectively, showing a notable increase. Considering the fact that the conductance of TPA M-2D-vdWHs decreased with the electric field (Fig. 1D) and the TC conformation of TPA M-2D-vdWHs has a thickness larger than the TWP one, these findings imply that, for the TPA M-2D-vdWHs, the high conductance at the low electric field comes from the TWP conformation, while the low conductance under the high electric field comes from the TC conformation.

To explore the role of electronic structure on the modulation efficiency of electric field, we fabricated more M-2D-vdWHs by using other three TPA derivatives, as sketched in Fig. 3A. We performed electrical measurements on the as-fabricated M-2D-vdWHs with 4-bromotriphenylamine (Br-TPA), 4-N,4-N-diphenylbenzene-1,4-diamine (NH2-TPA), and 4-methyltriphenylamine (CH3-TPA), which contain different substituents in the TPA backbone. The introduction of electron-withdrawing (electron-donating) substituents changed the density distribution of electron cloud and strengthened the interaction of aromatic π-π interactions. In contrast, the methyl substitution has a minor influence on the electronic structure of the molecule (35–37). Figure 3B shows the typical conductance-displacement traces and the 1D conductance histograms for Br-TPA M-2D-vdWHs under different biases. More detailed characterizations of the charge transport through Br-TPA M-2D-vdWHs, as well as that through NH2-TPA M-2D-vdWHs and CH3-TPA M-2D-vdWHs, were provided in figs. S15 to S17. The conductance of Br-TPA and NH2-TPA M-2D-vdWHs was modulated by a factor of 2.8 and 4.8 when the applied bias switched from 100 to 300 mV, respectively. Compared to that observed in the electrical characterizations of TPA M-2D-vdWHs, the modulation efficiency of electric field is lower in the cases of Br-TPA and NH2-TPA M-2D-vdWHs. In contrast, for CH3-TPA M-2D-vdWH, the modulation efficiency is comparable to that of TPA (Fig. 3C). To briefly conclude, TPA M-2D-vdWH promises to serve as the best candidate for fabricating single-molecule switch because it witnesses the largest modulation efficiency of electric field, yet with minimum thickness. In the following sections, we presented the characterizations on the switching behavior of TPA M-2D-vdWH.

Fig. 3. Conductance analysis of the TPA derivatives M-2D-vdWH devices.

Fig. 3.

Open in a new tab

(A) Chemical structure of TPA derivatives, including Br-TPA, NH2-TPA, and CH3-TPA. (B) 1D conductance histograms for Br-TPA M-2D-vdWHs under the bias of 100, 200, and 300 mV, respectively. The inset is the corresponding conductance-displacement traces. (C) Evolution of conductance of the TPA derivatives M-2D-vdWHs with bias. (D) Schematic evolution of TPA along with the electric field, from a TWP conformation with a high conductivity (“ON” state) to a TC conformation with a low conductivity (“OFF” state). (E) Reversible switching of TPA M-2D-vdWHs as the electric field was switched between 100 and 300 mV.

Reversible switching of M-2D-vdWHs

More than the feature of atomic thickness, systematic measurements over a wide bias range revealed that our graphene-based M-2D-vdWHs is applicable for a stable molecular switch under a strong electric field. Previous works demonstrated that, for the molecules trapped in Au electrode pair with a nanometer-sized gap, the electric field can accelerate the bond-forming process between them (38). Although this intriguing property had great potential in catalysis, it may cause the failure of a molecular switch because of the electrically induced chemical reaction. In the M-2D-vdWHs, the highest occupied molecular orbital (HOMO) of the molecule is close to the Fermi level of the graphene electrode (table S4); thus, the electron will be implanted into the molecule. Such electron transfer makes the molecule negatively charged and the adsorption site of graphene positively charged (8, 39). As a result, each molecule forms an electric dipole with the graphene underneath. The cohesive interaction between them leads to the enhanced binding of molecule and graphene. Moreover, the dipole-dipole interaction between adjacent molecules is repulsive. It moves the molecule away from each other (40, 41) and thus effectively hinders the dimerization of two adjacent TPA molecules.

To show the advantage of the M-2D-vdWH switch in stability, we compared the electric-induced switching behavior of TPA molecule in the macroscale electrochromic device and the TPA M-2D-vdWH switch. The fabrication of the electrochromic device is detailed in Methods. TPA solution was injected between two indium tin oxide (ITO) electrodes to establish a sandwich structure. Bias was applied between the two opposite ITO electrodes to generate an electric field. We performed ultraviolet (UV)–visible absorption measurement to monitor the evolution of a TPA molecule with the electric field (fig. S18). When the TPA electrochromic device underwent a voltage of 2.6 V for 60 s, its color gradually changed from transparent to black (fig. S19). As the bias increased, new UV-visible absorption peaks appeared at 480 and 690 nm. The 690-nm peak was attributed to the cationic radical generated by the oxidation of TPA, while the 480-nm peak was attributed to the formation of N,N,N′,N′-tetraphenylbenzidine (TPB) (42, 43). These results indicated that in the electrochromic device, where the electric field is about 103 V/m, the dimerization occurred and made the TPA molecule evolve to be TPB (fig. S20).

We then performed conductance measurement for TPB M-2D-vdWHs (fig. S21). It was found that the conductance of TPB M-2D-vdWHs under the bias of 100 mV was 10–5.39 G0, and the plateau length was 1.53 nm. Considering that these values were close to that of TPA M-2D-vdWHs under the bias of 300 mV, we designed and carried out the reversibility measurements to exclude the dimerization of TPA. The details are as follows. If the dimerization occurred with the higher bias (300 mV), the conductance of the M-2D-vdWHs would remain at 10–5.39 G0 when the bias was switched to the lower bias (100 mV). In contrast, if the higher bias only induced conformational change and no dimerization occurred, then the molecular conductance will be reversibly switched. We measured the conductance of M-2D-vdWHs by switching the applied bias between low bias and high bias (Figs. 3E and figs. S22 and S23). In both the positive and negative bias ranges, the electrical characterization results were consistent (figs. S24 to 27). This result indicated that it is the conformational change other than the chemical reaction that contributes to the switching in the conductance states under the electric field (Fig. 3D). It showed that, in the TPA M-2D-vdWHs, the electronic effect of graphene enabled the TPA molecule to retain its chemical structure and thus ensured the single-molecule switch work under the strong electric field as high as about 108 V/m (fig. S20).

Theoretically calculated transmissions of M-2D-vdWHs

We used the DFT-based simulations to further study the experimentally measured conductance. We first calculated the optimized geometries for these M-2D-vdWHs in the absence of electric field (Figs. 4A and fig. S28). For both the cross-plane and in-plane charge transport modes, the thicknesses of the M-2D-vdWHs were calculated and listed in table S1. The comparisons between the theoretical and experimental values confirmed that charge transport through the as-fabricated M-2D-vdWHs with the cross-plane charge transport mode. We have calculated the frontier molecular orbital energies for these M-2D-vdWHs in the absence of an electric field (table S4) and found that the HOMO–lowest unoccupied molecular orbital (LUMO) gaps for TP, TPP, and TPA are 4.89, 5.17, and 4.50 eV, respectively, which agrees with the measured UV-visible absorption spectra for these three molecules (fig. S29). With the optimized geometries and orbital energies, we calculated the transmission coefficients for these M-2D-vdWHs in the absence of an electric field (Fig. 4, B and H). The transmission coefficients near the Fermi energy (EF) of the graphene electrode (EF – EFDFT = 0 eV) was consistent with the measured conductance of these M-2D-vdWHs at the low bias (100 mV).

Fig. 4. Theoretical calculation and transmission pathways analysis.

Fig. 4.

Open in a new tab

(A) Optimized structures of TP, TPP, and TPA M-2D-vdWHs without the electrical field. For clarity, the screening regions were not shown. (B) Calculated transmission coefficients of TP, TPP, and TPA M-2D-vdWHs without the electrical field. (C) Evolution of the thickness of TP, TPP, and TPA M-2D-vdWHs with the electric field. (D) Evolution of the HOMO-LUMO gap of the sandwiched molecule with the electric field. (E) Calculated transmission coefficients of the TPA under different electric fields. (F to H) Theoretically simulated results of the TPA derivatives M-2D-vdWHs.

We then calculated the optimized geometries for these M-2D-vdWHs with a wide range of electric fields (figs. S30 to S35). Figure 4 (C and F) plotted the thickness evolution of the M-2D-vdWHs with the electric field. As the electric field increased, for the in-plane charge transport mode, the thickness of TPA molecular junction decreased, while for the cross-plane charge transport mode, the thickness increased (table S2). The latter is consistent with that observed in our experiments (Fig. 2F), which further confirmed that the as-fabricated TPA M-2D-vdWHs are M-2D-vdWHs. Moreover, the thickness of TPA M-2D-vdWHs increases with the increasing intensity of electric field, implying that the experimentally observed conductance switching of TPA M-2D-vdWH was attributed to the change of TPA conformation. In particular, there was a lesser thickness of the molecular junction if the TPA adopted the TWP conformation, while there was a greater thickness of the molecular junction if the TPA had switched to the TC conformation.

We further calculated the frontier molecular orbital energies for these molecules under different electric fields, and the values of the HOMO-LUMO gap were shown in Fig. 4 (D and G). As the electric field increased, for TP, there was almost no change in the HOMO-LUMO gap, while for TPP, the HOMO-LUMO gap decreased slightly. In contrast, for TPA derivatives, the HOMO-LUMO gap increased considerably as the electric field increased. Last, we placed the optimized molecular structure between two graphene electrodes to construct the single-molecule devices and calculated the transmission coefficients for TPA derivatives M-2D-vdWHs at different electric fields (Figs. 4E and figs. S36 to S39). As the intensity of electric field increased, it was found that the transmission coefficient of TPA derivatives M-2D-vdWHs reduced gradually, which were consistent with the experimental results (Fig. 3C).

The experimental and theoretical results allow us to reach the hypothesis for the mechanism of TPA M-2D-vdWH single-molecule switch. For TP, its rigid structure restricted the response under electric fields, which made it electrically inert. For TPP, with the increase of the electric field, the HOMO-LUMO gap decreased, but the length of junction increased. The competitive effect gave rise to the unchanged conductance in the experiment. For TPA derivatives, as the electric field increased, the HOMO-LUMO gap increased and the molecule changed from the short TWP conformation to the longer TC one. These two effects synchronously regulated the charge transport of TPA derivatives M-2D-vdWHs, resulting in the electric-induced switching behavior.

DISCUSSION

In conclusion, we developed a van der Waals heterojunction strategy to fabricate single-molecule switches in a layer-by-layer style. It was achieved by stacking a single molecule and two sheets of single-layer graphene layer by layer, followed by electrically modulating the cross-plane charge transport through them. Six triphenyl derivatives were used to fabricate the so-called M-2D-vdWHs, and the evolution of their single-molecule conductance along with the electric field was investigated both experimentally and theoretically. In particular, the thickness of TPA M-2D-vdWH was measured to be 1.20 nm, and its conductance switched between two states of 10–4.58 G0 and 10–5.36 G0 as the bias switched between 100 and 300 mV. Experiments and theoretical simulations demonstrated that the electric field altered the frontier molecular orbital energy and modulated the sandwiched TPA molecule between the TWP conformation and the TC one. The synchronous effects enable the TPA M-2D-vdWHs to act as a reversible single-molecule switch. The ability to modulate the cross-plane charge transport through M-2D-vdWHs not only offers opportunities for the design and further miniaturization of molecular electronic devices but also provides a strategy for a wide range of fields such as molecular assembly, biochemical reactions, catalysis, and materials sciences.

METHODS

Conductance measurement

In the break junction experiments, the size of the electrode gap was controlled with a module composed of a stepper motor and a piezoelectric stack. The stepping motor serves for the coarse and quick adjustment, while the piezoelectric stack serves for the more precise adjustment at the nanoscale. Under the control of the piezoelectric stack actuator, the upper electrode moves up and down accurately and controllably, which allows the repeated opening and closing of graphene electrodes to construct a large number of M-2D-vdWHs.

During the conductance measurement, A home-built I-V converter was used for data acquisition at a sampling frequency of 20 kHz, and a feedback circuit was used to modulate the bias to further construct different electric fields. To prevent the graphene layer from being damaged and forming Cu-Cu contacts during the break junction, we set the upper limit of conductance to 10–2.80 G0. Once the conductance exceeds this value during the measure, the actuator will immediately switch the direction of movement.

Electrochromic device fabrication

The manufacturing process of the liquid electrochromic device is as follows: After ultrasonic cleaning and nitrogen drying, a pair of ITO-coated conductive glass constituted the electrochromic cell. The resin spacer layer (2 mm in height) was fabricated by 3D printing technology and sandwiched between the electrodes, and then, epoxy resin was harnessed to seal up the spacer layer and the electrodes. The outside of the cell was covered with a blue film to avoid contamination of the ITO glass surface. TPA (0.01 mM) and tetrabutylammonium hexafluorophosphate (0.1 mM) were dissolved in propylene carbonate to prepare the electrolyte solution, bubbling the prepared electrolyte with nitrogen to remove oxygen using a syringe to inject it into an empty cell.

Theoretical calculation

The changing process of spatial conformation of TP, TPP, TPA, Br-TPA, NH2-TP, and CH3-TPA under different electric fields was studied by DFT. The molecular geometries were calculated and optimized with the Gaussian16 package. After structural optimization, we set an external electric field on the z axis and used the B3LYP/6-311(d, p) basis set for related calculations so that accurate energy, frontier molecular orbitals, and optimized molecular conformation can be obtained. Ensure that the final calculated structure is the most optimized structure by optimizing the molecular structure without imaginary frequency.

Transmission calculation

The optimized molecular structures were inserted between the graphene electrodes to construct the graphene-molecule-graphene junction by the Atomistic Tool Kit software. The model devices consist of the periodic system of the left and right electrodes [one-layer (6 × 6) graphene surface, respectively] and a limited scattering region composed of molecular systems. The scattering region retains the screening area of the one-layer (6 × 6) graphene-electrode material layer. DFT was used to calculate the related electron transport characteristics for the model devices. The exchange-correlation potential adopts the generalized gradient approximation basis set; all elements adopt the double zeta polarization basis. For the transmission spectrum calculation, the energy range point was set as 501, 12 k-points in the z direction.

Acknowledgments

Funding: This research was supported by the National Natural Science Foundation of China (T2222002, 21973079, 22032004, 21991130, and 21905238), the Natural Science Foundation of Fujian Province (2021J06008), the State Key Laboratory of Physical Chemistry of Solid Surfaces (202002), and the “Nanqiang Outstanding Young Talents Program” of Xiamen University.

Author contributions: Y.Y. initiated and supervised the project. Y.Y., Y.-L.Z., and S.Z. designed the experiments. Y.-L.Z., Q.-M.L., Y.-G. L., and A.F. carried out the experiments. Y.-L.Z., J.G., and Z.-X.Y. analyzed the data. Y.-L.Z. and T.L. carried out the theoretical simulations. W.H. and J.S. wrote the control software for the break junction setup. Y.Y., Z.-Q.T., and Y.-L.Z. wrote the manuscript with input from all authors.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Sections S1 to S5

Figs. S1 to S39

Tables S1 to S4

Click here for additional data file. (5.3MB, pdf)

REFERENCES AND NOTES

  • 1.Burch K. S., Mandrus D., Park J.-G., Magnetism in two-dimensional van der Waals materials. Nature 563, 47–52 (2018). [DOI] [PubMed] [Google Scholar]
  • 2.Chia X., Pumera M., Characteristics and performance of two-dimensional materials for electrocatalysis. Nat. Catal. 1, 909–921 (2018). [Google Scholar]
  • 3.Low T., Chaves A., Caldwell J. D., Kumar A., Fang N. X., Avouris P., Heinz T. F., Guinea F., Martin-Moreno L., Koppens F., Polaritons in layered two-dimensional materials. Nat. Mater. 16, 182–194 (2017). [DOI] [PubMed] [Google Scholar]
  • 4.Zhu K., Wen C., Aljarb A. A., Xue F., Xu X., Tung V., Zhang X., Alshareef H. N., Lanza M., The development of integrated circuits based on two-dimensional materials. Nat. Electron. 4, 775–785 (2021). [Google Scholar]
  • 5.Dai C., Liu Y., Wei D., Two-dimensional field-effect transistor sensors: The road toward commercialization. Chem. Rev. 122, 10319–10392 (2022). [DOI] [PubMed] [Google Scholar]
  • 6.Jia C., Migliore A., Xin N., Huang S., Wang J., Yang Q., Wang S., Chen H., Wang D., Feng B., Liu Z., Zhang G., Qu D.-H., Tian H., Ratner M. A., Xu H. Q., Nitzan A., Guo X., Covalently bonded single-molecule junctions with stable and reversible photoswitched conductivity. Science 352, 1443–1445 (2016). [DOI] [PubMed] [Google Scholar]
  • 7.Zeng B.-F., Wei J.-Y., Zhang X.-G., Liang Q.-M., Hu S., Wang G., Lei Z.-C., Zhao S.-Q., Zhang H.-W., Shi J., Hong W., Tian Z.-Q., Yang Y., In situ lattice tuning of quasi-single-crystal surfaces for continuous electrochemical modulation. Chem. Sci. 13, 7765–7772 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wen H., Li W., Chen J., He G., Li L., Olson Mark A., Sue Andrew C. H., Stoddart J. F., Guo X., Complex formation dynamics in a single-molecule electronic device. Sci. Adv. 2, e1601113 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Zeng B.-F., Wang G., Qian Q.-Z., Chen Z.-X., Zhang X.-G., Lu Z.-X., Zhao S.-Q., Feng A.-N., Shi J., Yang Y., Hong W., Selective fabrication of single-molecule junctions by interface engineering. Small 16, 2004720 (2020). [DOI] [PubMed] [Google Scholar]
  • 10.Song P., Guerin S., Tan S. J. R., Annadata H. V., Yu X., Scully M., Han Y. M., Roemer M., Loh K. P., Thompson D., Nijhuis C. A., Stable molecular diodes based on π–π interactions of the molecular frontier orbitals with graphene electrodes. Adv. Mater. 30, 1706322 (2018). [DOI] [PubMed] [Google Scholar]
  • 11.Traversi F., Raillon C., Benameur S. M., Liu K., Khlybov S., Tosun M., Krasnozhon D., Kis A., Radenovic A., Detecting the translocation of DNA through a nanopore using graphene nanoribbons. Nat. Nanotechnol. 8, 939–945 (2013). [DOI] [PubMed] [Google Scholar]
  • 12.Rossini A. E., Gala F., Chinappi M., Zollo G., Peptide bond detection via graphene nanogaps: A proof of principle study. Nanoscale 10, 5928–5937 (2018). [DOI] [PubMed] [Google Scholar]
  • 13.Xu Q., Scuri G., Mathewson C., Kim P., Nuckolls C., Bouilly D., Single electron transistor with single aromatic ring molecule covalently connected to graphene nanogaps. Nano Lett. 17, 5335–5341 (2017). [DOI] [PubMed] [Google Scholar]
  • 14.Xie L., Liao M., Wang S., Yu H., Du L., Tang J., Zhao J., Zhang J., Chen P., Lu X., Wang G., Xie G., Yang R., Shi D., Zhang G., Graphene-contacted ultrashort channel monolayer MoS2 transistors. Adv. Mater. 29, 1702522 (2017). [DOI] [PubMed] [Google Scholar]
  • 15.Alexeev E. M., Ruiz-Tijerina D. A., Danovich M., Hamer M. J., Terry D. J., Nayak P. K., Ahn S., Pak S., Lee J., Sohn J. I., Molas M. R., Koperski M., Watanabe K., Taniguchi T., Novoselov K. S., Gorbachev R. V., Shin H. S., Fal’ko V. I., Tartakovskii A. I., Resonantly hybridized excitons in moiré superlattices in van der Waals heterostructures. Nature 567, 81–86 (2019). [DOI] [PubMed] [Google Scholar]
  • 16.Xu S., Al Ezzi M. M., Balakrishnan N., Garcia-Ruiz A., Tsim B., Mullan C., Barrier J., Xin N., Piot B. A., Taniguchi T., Watanabe K., Carvalho A., Mishchenko A., Geim A. K., Fal’ko V. I., Adam S., Neto A. H. C., Novoselov K. S., Shi Y., Tunable van Hove singularities and correlated states in twisted monolayer-bilayer graphene. Nat. Phys. 17, 619–626 (2021). [Google Scholar]
  • 17.Pham P. V., Bodepudi S. C., Shehzad K., Liu Y., Xu Y., Yu B., Duan X., 2D heterostructures for ubiquitous electronics and optoelectronics: Principles, opportunities, and challenges. Chem. Rev. 122, 6514–6613 (2022). [DOI] [PubMed] [Google Scholar]
  • 18.Xiang L., Zhang P., Liu C., He X., Li H. B., Li Y., Wang Z., Hihath J., Kim S. H., Beratan D. N., Tao N., Conductance and configuration of molecular gold-water-gold junctions under electric fields. Matter 3, 166–179 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Sarkar A., Rajaraman G., Modulating magnetic anisotropy in Ln(III) single-ion magnets using an external electric field. Chem. Sci. 11, 10324–10330 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Mangel S., Skripnik M., Polyudov K., Dette C., Wollandt T., Punke P., Li D., Urcuyo R., Pauly F., Jung S. J., Kern K., Electric-field control of single-molecule tautomerization. Phys. Chem. Chem. Phys. 22, 6370–6375 (2020). [DOI] [PubMed] [Google Scholar]
  • 21.Foroutan-Nejad C., Andrushchenko V., Straka M., Dipolar molecules inside C70: An electric field-driven room-temperature single-molecule switch. Phys. Chem. Chem. Phys. 18, 32673–32677 (2016). [DOI] [PubMed] [Google Scholar]
  • 22.Yang C., Liu Z., Li Y., Zhou S., Lu C., Guo Y., Ramirez M., Zhang Q., Li Y., Liu Z., Houk K. N., Zhang D., Guo X., Electric field–catalyzed single-molecule Diels-Alder reaction dynamics. Sci. Adv. 7, eabf0689 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Meir R., Chen H., Lai W., Shaik S., Oriented electric fields accelerate Diels-Alder reactions and control the endo/exo selectivity. ChemPhysChem 11, 301–310 (2010). [DOI] [PubMed] [Google Scholar]
  • 24.Huang X., Tang C., Li J., Chen L.-C., Zheng J., Zhang P., Le J., Li R., Li X., Liu J., Yang Y., Shi J., Chen Z., Bai M., Zhang H.-L., Xia H., Cheng J., Tian Z.-Q., Hong W., Electric field–induced selective catalysis of single-molecule reaction. Sci. Adv. 5, eaaw3072 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yang C., Shen P., Ou Q., Peng Q., Zhou S., Li J., Liu Z., Zhao Z., Qin A., Shuai Z., Tang B. Z., Guo X., Complete deciphering of the dynamic stereostructures of a single aggregation-induced emission molecule. Matter 5, 1224–1234 (2022). [Google Scholar]
  • 26.Zhao S., Wu Q., Pi J., Liu J., Zheng J., Hou S., Wei J., Li R., Sadeghi H., Yang Y., Shi J., Chen Z., Xiao Z., Lambert C., Hong W., Cross-plane transport in a single-molecule two-dimensional van der Waals heterojunction. Sci. Adv. 6, eaba6714 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Zhao S., Deng Z.-Y., Albalawi S., Wu Q., Chen L., Zhang H., Zhao X.-J., Hou H., Hou S., Dong G., Yang Y., Shi J., Lambert C. J., Tan Y.-Z., Hong W., Charge transport through single-molecule bilayer-graphene junctions with atomic thickness. Chem. Sci. 13, 5854–5859 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Dasari R. R., Wang X., Wiscons R. A., Haneef H. F., Ashokan A., Zhang Y., Fonari M. S., Barlow S., Coropceanu V., Timofeeva T. V., Jurchescu O. D., Brédas J.-L., Matzger A. J., Marder S. R., Charge-transport properties of F6TNAP-based charge-transfer cocrystals. Adv. Funct. Mater. 29, 1904858 (2019). [Google Scholar]
  • 29.Iwane M., Tada T., Osuga T., Murase T., Fujita M., Nishino T., Kiguchi M., Fujii S., Controlling stacking order and charge transport in π-stacks of aromatic molecules based on surface assembly. Chem. Commun. 54, 12443–12446 (2018). [DOI] [PubMed] [Google Scholar]
  • 30.Geva R., Levy N. R., Tzadikov J., Cohen R., Weitman M., Xing L., Abisdris L., Barrio J., Xia J., Volokh M., Ein-Eli Y., Shalom M., Molten state synthesis of nickel phosphides: Mechanism and composition-activity correlation for electrochemical applications. J. Mater. Chem. A 9, 27629–27638 (2021). [Google Scholar]
  • 31.Zhang S., Dai Y., Luo S., Gao Y., Gao N., Wang K., Zou B., Yang B., Ma Y., Rehybridization of nitrogen atom induced photoluminescence enhancement under pressure stimulation. Adv. Funct. Mater. 27, 1602276 (2017). [Google Scholar]
  • 32.Caneva S., Gehring P., García-Suárez V. M., García-Fuente A., Stefani D., Olavarria-Contreras I. J., Ferrer J., Dekker C., van der Zant H. S. J., Mechanically controlled quantum interference in graphene break junctions. Nat. Nanotechnol. 13, 1126–1131 (2018). [DOI] [PubMed] [Google Scholar]
  • 33.Dong X., Shi Y., Zhao Y., Chen D., Ye J., Yao Y., Gao F., Ni Z., Yu T., Shen Z., Huang Y., Chen P., Li L.-J., Symmetry breaking of graphene monolayers by molecular decoration. Phys. Rev. Lett. 102, 135501 (2009). [DOI] [PubMed] [Google Scholar]
  • 34.Zhan C., Wang G., Yi J., Wei J.-Y., Li Z.-H., Chen Z.-B., Shi J., Yang Y., Hong W., Tian Z.-Q., Single-molecule plasmonic optical trapping. Matter 3, 1350–1360 (2020). [Google Scholar]
  • 35.Sinnokrot M. O., Sherrill C. D., Substituent effects in π−π interactions: Sandwich and T-shaped configurations. J. Am. Chem. Soc. 126, 7690–7697 (2004). [DOI] [PubMed] [Google Scholar]
  • 36.Neel A. J., Hilton M. J., Sigman M. S., Toste F. D., Exploiting non-covalent π interactions for catalyst design. Nature 543, 637–646 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kumar A., Tyagi P., Reddy M. A., Mallesham G., Bhanuprakash K., Rao V. J., Kamalasanan M. N., Srivastava R., Chemical structure dependent electron transport in 9,10-bis(2-phenyl-1,3,4-oxadiazole) derivatives of anthracene. RSC Adv. 4, 12206–12215 (2014). [Google Scholar]
  • 38.Zheng J., Liu J., Zhuo Y., Li R., Jin X., Yang Y., Chen Z.-B., Shi J., Xiao Z., Hong W., Tian Z.-q., Electrical and SERS detection of disulfide-mediated dimerization in single-molecule benzene-1,4-dithiol junctions. Chem. Sci. 9, 5033–5038 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Oh S., Crommie M. F., Cohen M. L., Simulating the nanomechanical response of cyclooctatetraene molecules on a graphene device. ACS Nano 13, 1713–1718 (2019). [DOI] [PubMed] [Google Scholar]
  • 40.Tsai H.-Z., Omrani A. A., Coh S., Oh H., Wickenburg S., Son Y.-W., Wong D., Riss A., Jung H. S., Nguyen G. D., Rodgers G. F., Aikawa A. S., Taniguchi T., Watanabe K., Zettl A., Louie S. G., Lu J., Cohen M. L., Crommie M. F., Molecular self-assembly in a poorly screened environment: F4TCNQ on graphene/BN. ACS Nano 9, 12168–12173 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Stradi D., Garnica M., Díaz C., Calleja F., Barja S., Martín N., Alcamí M., Vazquez de Parga A. L., Miranda R., Martín F., Controlling the spatial arrangement of organic magnetic anions adsorbed on epitaxial graphene on Ru(0001). Nanoscale 6, 15271–15279 (2014). [DOI] [PubMed] [Google Scholar]
  • 42.Blanchard P., Malacrida C., Cabanetos C., Roncali J., Ludwigs S., Triphenylamine and some of its derivatives as versatile building blocks for organic electronic applications. Polym. Int. 68, 589–606 (2019). [Google Scholar]
  • 43.Yurchenko O., Freytag D., zur Borg L., Zentel R., Heinze J., Ludwigs S., Electrochemically induced reversible and irreversible coupling of triarylamines. J. Phys. Chem. B 116, 30–39 (2012). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Sections S1 to S5

Figs. S1 to S39

Tables S1 to S4

Click here for additional data file. (5.3MB, pdf)

Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES