Non-equilibrium lifetimes of DNA under electronic current in a molecular junction

TL;DR

The paper addresses how tunnelling electronic currents influence the mechanical stability and lifetime of DNA in a molecular junction. It employs non-equilibrium Keldysh-Langevin molecular dynamics to couple a single classical collective coordinate to a quantum DNA Hamiltonian, deriving an effective potential $U(x)$, mean first-passage times $\langle \tau \rangle$, and a coordinate-dependent temperature $T(x)$. Key findings include a non-monotonic voltage dependence of stability: low biases destabilise the junction by lowering barriers and shifting the potential minimum, while higher biases can restore partial stability as barriers re-emerge, alongside a Landauer blowtorch where spatial fluctuations govern barrier crossing. The results reveal that current-induced forces, dissipation, and fluctuations collaboratively govern non-equilibrium lifetimes, with implications for DNA-based nanodevices and broader biomolecular junctions.

Abstract

We investigate the non-equilibrium mechanical motion of double-stranded DNA in a molecular junction under electronic current using Keldysh-Langevin molecular dynamics. Non-equilibrium electronic force reshapes the effective potential energy surface, and along with electronic viscosity force and stochastic force, governs voltage-dependent dynamics of DNA's collective mechanical coordinate. We compute mean first-passage times to quantify the non-equilibrium lifetime of the DNA junction. At low voltage biases, electron-mechanical motion coupling destabilises DNA by shifting the potential minimum towards critical displacement and suppressing barriers, shortening lifetimes by several orders of magnitude. Unexpectedly, at higher voltages the trend reverses: the potential minimum shifts away from instability and the barrier re-emerges, producing re-stabilisation of the junction. In addition, we demonstrate the Landauer blowtorch effect in this system: coordinate-dependent fluctuations generate a spatially varying effective temperature, changing current-induced dynamics of mechanical degrees of freedom. Apparent temperatures of DNA mechanical motion increase far above ambient due to current-induced heating, correlating with suppressed electronic current at stronger couplings. Our results reveal a non-equilibrium interplay between current-driven forces, dissipation, and fluctuations in DNA junctions, establishing mechanisms for both destabilisation and recovery of DNA stability under electronic current.

Figures (4)

• Figure 1: (a) Effective potential energy surface and spin-resolved currents as a function of displacement, with the critical instability threshold at $x =$ 1 a.u. (b) Voltage-dependent shift of the potential minimum, which moves toward destabilising displacements at stronger electron-mechanical motion coupling. (c) Corresponding barrier height at the critical displacement.
• Figure 2: Voltage dependence of the ratio of non-equilibrium MFPT to equilibrium MFPT for different electron - mechanical motion coupling strength: $\chi = 2$ meV (black), $\chi = 5$ meV (blue), $\chi = 10$ meV (red).
• Figure 3: (a) Voltage dependence of MFPTs ratio computed with coordinate-dependent effective temperature $T(x)$ (solid) that means including electronic friction and fluctuating forces, compared to constant-temperature dynamics at $T=300$ K (dashed). (b) Probability densities from trajectories at $\chi = 10$ meV and $V = 0.3$ V. Including electronic viscosity and noise (blue dashed) reduces the probability near the threshold critical displacement compared to the distribution from non-equilibrium electronic forces and environmental dissipation alone (orange dashed). The locally elevated effective temperature $T(x)$ (red) illustrates the stabilising blowtorch effect. (c) Same comparison at $V = 1$ V, showing even stronger broadening and access to high-displacement states as $T(x)$ rises sharply near the critical threshold displacement -- destabilising blowtorch effect.
• Figure 4: (a) Trajectory-averaged apparent temperature of the DNA mechanical coordinate (computed as average kinetic energy) as a function of applied voltage for different electron-mechanical motion couplings. (b) Corresponding trajectory-averaged current through the DNA junction. Spin polarisation $<14\%$ is evident in the difference between spin up (solid) and spin down (dashed).