Highly distinguishable isomeric states of a tripodal arylazopyrazole derivative on graphite through electron/hole-induced switching at ambient conditions

Abstract

Manipulating the energy barrier and extending the half-life of nonequilibrium states in photochromic switches presents viable solutions for applying them in molecular electronics. Typically, the half-life of the Z isomer of azobenzene (AB) is a few days. Arylazopyrazole-based molecular switches are one of the profoundly explored systems in recent times due to their superior bidirectional photoswitching and long half-life (over a thousand days at room temperature) of Z isomers. Herein, we utilize an efficient solid-state photoswitchable fluorinated tripodal N-functionalized arylazo-3,5-dimethylpyrazole derivative (FNAAP) to envisage and access multiple metastable states on the surface. The tripodal molecule forms well-ordered, large crystalline domains on graphite through non-bonding interactions between the molecules. By injecting electron/hole pairs into the self-assembled molecules on a surface using a scanning tunneling microscope (STM) tip, they are switched between 8 states (EEE, EEZ, EZE, ZEE, EZZ, ZEZ, ZZE and ZZZ) in a tunneling junction at ambient conditions. Contrary to the degeneracy-controlled four states in solution phase, all the eight states are remarkably stable on the surface and are well distinguishable by the tunneling current passing through the molecule at the tunneling junction. The change in current upon switching between these states is approximately an order of magnitude. This is particularly notable at positive sample voltage compared to negative sample voltage. The exceptional stability of the states at ambient conditions provides an opportunity to use a single FNAAP molecule as an 8-bit operation unit, with a potential storage capacity of ≈200 Tbits per 1 cm² area using an atomically precise write and read tool like an STM tip.

Keywords: azo compounds; electron/hole-induced isomerization; isomerization; molecular switches; switching.

Figure 1

Top panel: Chemical structures of EEE, and ZZZ isomers of (FNAAP). Lower panel: Geometry-optimized ball and stick model of FNAAP. The EEE, EEZ, EZZ and ZZZ isomers are shown in sequence from left to right, respectively.

Figure 2

AFM phase images (a, b and c) of ultra-thin films of FNAAP deposited from ethanolic solution on HOPG (0001) and are obtained from independent areas. Domains in the 1D phase of FNAAP are marked by green and yellow dashed lines. The green and yellow three-fold arrows depict the orientation of the long edges of the islands with respect to each other. A few graphite step edges are marked by black dashed lines. Double-headed arrows in (c) indicate line-like features within the domains of the 1D phase, which is due to the moiré pattern originating from the super-lattice. (d) 2D-FFT taken on (c). The resonances corresponding to the super-lattice are indicated by green dashed lines. Two resonances are rotated by ≈120° and correspond to the major orientations of the domains in the 1D phase.

Figure 3

Constant current STM topographs (300 pA, 0.3 V) of the FNAAP adlayer on HOPG (a, b) deposited from the corresponding ethanolic solution. Imaging is performed at the HOPG–air interface. Green dashed lines depict two adjacent molecular lattices, and the green dashed oblique depicts the unit cell of the assembly. (c) Force field optimized geometry of the adlayer of FNAAP (EEE isomer) on bilayer graphite. The unit lattice vectors of the adlayer are a, b and γ is the angle between the vectors. Black arrows indicate the compact lattice directions of graphite. Magenta dashed lines depict possible non-bonding interactions between carbonyl groups and aromatic hydrogen atoms. Blue dashed lines depict the same between fluorine and aromatic/methyl hydrogen atoms.

Figure 4

(a) Current versus sample voltage (I–V) recorded on a single FNAAP within the assembly. The I–V measurements are recorded by stabilizing the current at ≈0.5 nA and at −2 V. Arrows indicate the switching of FNAAP molecules between E and Z forms. Insets show the zoomed section of the I–V curves indicated using dashed rectangles. Distinct levels of current can be discerned and are indicated by dashed magenta lines. (b) Statistical analysis of all the switching events (voltage value corresponding to the up and down arrows). The switching probability is obtained by taking the ratio of the number of times switching is observed in a given voltage window to the total number of I–V measurements performed. The binning of I–V measurements is 10 mV. Each data point depicts the average probability for switching in intervals of 50 mV. Blue and red arrowheads depict the threshold voltage for switching at negative and positive sample voltages, respectively.

Figure 5

(a) Current versus time (time trace) at selected voltage intervals acquired on the adlayer of FNAAP on HOPG at ambient conditions. The time traces are part of I–V curves measured by stabilizing the feedback at ≈0.5 nA and at −300 mV. The mean value of the sample voltage at which the time traces are recorded is marked by dashed arrows. Magenta dashed lines represent stable recurring levels of current, which correspond to different isomeric states. (b) Statistical analysis of the number of states as obtained from multiple time traces in intervals of ≈100 mV voltage versus sample voltage.