Non-Equilibrium Quantum Materials for Electronics

Abstract

We review recent experimental advances in non-equilibrium quantum materials, focusing on current- and light-driven systems, transient and metastable phases, and non-equilibrium steady states. Emphasis is placed on current-driven phases and photoinduced control of quantum orders such as superconductivity, charge density waves, and ferroelectricity. We briefly outline the most relevant experimental results and discuss implications for future quantum and electronic technologies.

1. Introduction

Quantum materials [1]—systems in which electronic correlations, topology, and lattice dynamics intertwine—form the basis of many cutting-edge technologies. While the exploration of their equilibrium phases has revealed a vast landscape of emergent phenomena such as high-temperature superconductivity, multiferroicity, and topological order, it is becoming increasingly evident that many more striking functionalities emerge when they are driven far from equilibrium.

In this context, ultrafast control of quantum materials [2] has emerged as a vibrant frontier in condensed matter physics and materials science, and is rapidly becoming a central strategy in the search for next-generation (quantum) electronic devices, see Figure 1. Time-dependent perturbations such as ultrafast laser pulses, THz electric fields, or applied currents can access transient or metastable states, often revealing hidden orders or unconventional behaviors absent in the equilibrium phase diagram. These include possible light-induced superconductivity in cuprates [3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18] or fullerides [19,20,21,22,23,24], ultrafast switching of ferroelectric phenomena [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39], dynamical generation of charge density wave order [40,41,42,43,44,45] or excitonic order [46], and metal–insulator transitions induced by light pulses [47,48,49,50,51,52,53] or by steady currents [54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78].

The potential of out-of-equilibrium approaches is twofold. On one hand, they provide a powerful tool to probe fundamental mechanisms—such as electron-phonon coupling, symmetry breaking, or correlation effects—by perturbing the system away from its steady state and tracking its dynamical response with high temporal resolution. On the other hand, they have impressive technological potential, as dynamically generated states open ultrafast, low-power pathways to controlling material properties on demand.

For example, coherent light pulses in the terahertz to mid-infrared range have been shown to selectively excite lattice vibrations or electronic transitions, thereby engineering material properties in a nonthermal fashion. In systems such as ${\mathrm{K}}_3{\mathrm{C}}_{60}$ or cuprates, optical pulses have induced signatures of superconducting-like optical conductivity far above the equilibrium critical temperature [19,20]. Quantum devices based on photoinduced superconductivity could enable zero-resistance interconnects or switching elements that operate far above the equilibrium critical temperature, without the need for chemical doping or sustained cryogenics.

In some oxides, ferroelectric states have been induced on pico-seconds timescales [27]. Ultrafast control of polarization and domain walls in ferroelectric materials points to applications in non-volatile memory, logic devices, and neuromorphic computing, where switching speed and energy efficiency are critical.

From an applied perspective, these discoveries have the potential to shape new paradigms for electronic devices, whereby symmetry, topology, quantum correlations and electronic properties are modulated on demand via coherent driving. This can lead to a plethora of phenomena, including topologically protected currents, non-reciprocal transport, or Floquet-engineered quantum states, with potential applications in electronics, quantum information processing, and spintronics. Importantly, these functionalities are reversible and can be triggered selectively in time and space, opening the way to programmable material platforms.

The experimental realization of such states has been made possible by advances in ultrafast optical and electron spectroscopies [79], which allow sub-picosecond probing of structural, electronic, and magnetic dynamics. The ability to shape the temporal and spectral profile of the driving field—such as resonantly driving a phonon mode or exciting a specific charge transition—makes it possible to steer the system into tailored non-equilibrium regimes.

In this review, we present an overview of the most relevant experimental progress in this field, with an emphasis on light- and current-driven control of non-equilibrium phenomena in quantum materials. We review in particular light-induced superconductivity, ultrafast ferroelectric phenomena, and current-tuned metal–insulator transitions. We briefly review the experimental techniques used to investigate these phenomena, and focus on the most important breakthroughs and their impact on electronics. We highlight the implications of these phenomena for future electronic applications and identify the key challenges and opportunities in engineering non-equilibrium functionality in quantum materials.

2. Experimental Techniques for Ultrafast Phenomena

Investigating and controlling ultrafast dynamics in quantum materials demands advanced time-resolved probes capable of resolving changes in the electronic, structural, and magnetic degrees of freedom on femtosecond to picosecond timescales. Over the last two decades, a wide range of complementary experimental techniques has emerged, allowing impressive progress in our understanding of transient phenomena [2,79,80,81]. This section presents a brief overview of the most widely used approaches, organized by observable, see Figure 2.

Time-resolved optical pump–probe spectroscopies (TOS)—Optical pump–probe spectroscopy [81,82,83,84] has recently become a cornerstone of ultrafast science. In these experiments, a femtosecond laser pulse (pump) drives the system out of equilibrium, and a time-delayed pulse (probe) measures the transient changes in reflectivity or transmission. This approach measures changes in the dielectric function of the system, making it ideal for studying coherent phonons, transient states, and the dynamics and relaxation of hot carriers [85,86]—which has recently emerged as a way to probe directly non-equilibrium electron systems.

The frequency of the probe can be changed across a wide spectrum, ranging from optical probes—best suited for studying quasiparticle dynamics, electron-phonon coupling strength, and gap magnitudes—to THz probes, ideal for the investigation of low-energy phenomena such as collective modes, low-gap quasiparticles and superconducting phases.

More recently, the development of dual-delay schemes and rapid acquisition protocols has enabled correlation studies between different excitations. These techniques exploit sequences of pulses to generate frequency–frequency correlation maps [87,88], revealing anharmonicities and interactions between low-energy excitations that are otherwise not accessible, and disentangling the dynamics of electronic and vibrational modes.

Additionally, pump–probe spectroscopy that probe second harmonic generation (SHG) allow to investigate inversion symmetry breaking, enabling time-resolved tracking of ferroelectricity, charge ordering, and nematicity.

X-ray scattering—These scattering techniques take advantage of the small wavelength of X-rays to probe lattice structural dynamics, or changes in charge, spin, orbital electronic orders, with sub-picosecond resolution. Techniques include time-resolved X-ray diffraction (trXRD) [89] and time-resolved resonant inelastic X-ray scattering trRIXS [80,90].

Ultrafast Electron Probes—Ultrafast electron diffraction (UED) [91] provides direct access to atomic-scale dynamics, and is particularly well-suited to study lattice symmetry changes and electronic phases. In pump–UED experiments, a femtosecond laser excites the sample and a synchronized ultrafast electron pulse probes the transient diffraction pattern. The ability to track structural order parameters with femtometer-scale precision makes UED a powerful complement to optical probes.

A closely related technique, ultrafast scanning tunneling microscopy (USTM), combines ultrafast temporal resolution with spatially nanometer-resolved imaging. USTM enables visualization of photoinduced carrier diffusion and trapping at surfaces, and has been employed to study phase-separated systems or domain-structured materials [92,93,94,95,96,97].

Time- and angle-resolved photoemission spectroscopy (trARPES)—Time-resolved ARPES [98,99,100,101,102] allows to study the band structure and the single particle spectral function with momentum resolution and with picosecond time resolution. It has been employed to study a plethora of electronic orders.

Time-resolved scanning near-field optical microscopy (trSNOM)—This technique [103] employs near field microscopy to beat the diffraction limit and provide a time and frequency-resolved imaging of the system with nanometer scale spatial resolution. It has been used to track the evolution of phase domains [104], photoinduced metal–insulator transitions [51,52,105] or study plasmon-polariton propagation [106].

Time-resolved transport—Ultrafast transport can be probed via photoconductive switches [107] and has been used to investigate a variety of phenomena from light-induced superconducting states to band topology engineering in graphene [24,108].

Summary—The ensemble of ultrafast experimental techniques available today enables comprehensive probing of quantum materials far from equilibrium. Optical and THz probes excel at capturing carrier and low-energy dynamics; electron-based techniques offer direct insight into band structure; X-ray methods deliver atomic-scale insight into lattice dynanics; scanning techniques offer space-resolved imaging. Importantly, multi-messenger setups combining optical and structural probes allow to disentangle the effects of electronic and lattice dynamics. Together, they form an increasingly powerful platform for understanding and engineering emergent phases in driven quantum matter.

3. Photoinduced Transient Phases

Non-equilibrium drives—especially ultrafast optical pulses—have opened access to electronic and structural phases in quantum materials that are hidden or unstable at equilibrium.

These transient phases may be accessed via several mechanisms. A prototypical mechanism is the pump-driven switching of criticality, where an ordered phase is induced by the pump; this is illustrated in Figure 3a. At equilibrium, the free energy landscape does not favor any broken-symmetry phase. Upon pumping, the energy landscape is modified to favor a symmetry-breaking order parameter, driving the system into an ordered state. This mechanism has been proposed to explain the appearance of photoinduced superconducting-like states in potassium fulleride [109,110,111] and for the enhancement of excitonic order [112].

In another scenario, see Figure 3b, the equilibrium free energy already presents competing ordered phases which may be metastable or degenerate with the equilibrium state. The pump excites the system into these phases, which may then live for a short or long time, depending on decay processes. This mechanism has been proposed to explain the emergence of charge density wave or ferroelectric order in several systems [27,42,113].

This section surveys key experimental developments in this area, focusing on light-induced superconductivity, light-enhanced charge density wave (CDW) order, and metastable ferroic or excitonic states. Each case offers insights into how non-equilibrium phases hold great potential for future electronic devices.

3.1. Photoactive Superconductivity

One of the most striking demonstrations of phase control via light is the manipulation of superconducting states via light pulses. Several experiments have showed that it is possible to manipulate existing superconducting behavior by exciting the system with light pulses [114,115,116,117,118]. The most remarkable result is the transient induction of superconducting-like behavior above the equilibrium critical temperature in strongly correlated materials, particularly cuprates [3] and alkali-doped fullerides [19], see Figure 4.

3.1.1. Cuprate Superconductors

The work by Fausti et al. [3] first demonstrated that intense mid-infrared pulses resonant with a lattice vibration could induce a transient Josephson plasma edge in the reflectivity of stripe-ordered ${\mathrm{La}}_{1.675}{\mathrm{Eu}}_{0.2}{\mathrm{Sr}}_{0.125}{\mathrm{CuO}}_4$. This result suggested coherent superconducting coupling along the c-axis far above the equilibrium superconducting transition temperature. Follow-up experiments in ${\mathrm{La}}_{2-x}{\mathrm{Ba}}_x{\mathrm{CuO}}_4$ (LBCO) [4,5,6,8,9,13], and ${\mathrm{YBa}}_2{\mathrm{Cu}}_3{\mathrm{O}}_{6+x}$ (YBCO) cuprates [7,14,15,17,18], confirmed that the light-induced state exhibits features of superconductivity: a reflectivity edge, coherent interlayer coupling, and a suppression of charge order. On the other hand, an experiment in ${\mathrm{La}}_{2-x}{\mathrm{Sr}}_x{\mathrm{CuO}}_4$ (LSCO) [12] attributed the observed spectroscopic response to a photoinduced high-mobility metallic state rather than a superconducting state.

The mechanism proposed to explain the experiments involves non-linear phonon coupling [119]: the resonant excitation of anharmonically coupled phonons induces lattice distortions that transiently suppress competing stripe order. This idea, referred to as nonlinear phononics, was formalized in subsequent theoretical work and supported by trXRD experiments that directly observed melting of charge stripes [6].

3.1.2. Fulleride Superconductors: ${\mathrm{K}}_3{\mathrm{C}}_{60}$

In a different class of materials, namely potassium-doped fullerides, Mitrano et al. [19] showed that mid-infrared excitation of ${\mathrm{K}}_3{\mathrm{C}}_{60}$ induces a transient optical response characteristic of a superconducting state, even at temperatures far above the equilibrium $T_c$ (∼20 K). Reflectivity measurements in the THz regime indicated that the reflectivity saturates to one and the real part of the conductivity goes to zero for energies below ∼10 meV, indicating a superconducting gap. Moreover, the imaginary part of the conductivity diverges as ∼1/$\omega$, a behavior typical of superconductors. However, these observations are limited to the 1–10 meV range and are based on an extrapolation of raw data due to the differing penetration depths between pump and probe pulses.

Subsequent works revealed that the transient superconducting-like behavior could also be tuned via pressure [20], and that the photoinduced response was also visible in fluctuations [23] and ultrafast transport measurements [24].

Theoretical models linking these results to the emergence of a superconducting state have proposed several mechanism. Nonlinear phononics [21,109,110]—where the drive of an infrared active mode leads to a stationary displacement of a Raman phonon responsible for superconductivity—or a radiation-induced cooling of superconducting quasiparticles [120]. Other theories have disputed the superconducting nature of the non-equilibrium state, explaining it as a transient state with a negative conductivity [121,122]. Another explanation proposes that the experimental data are spurious results due to an extrapolation of raw data that ignores non-linear effects in the pump fluence, so that the superconducting state is actually a high mobility state induced by the drive [123,124].

3.1.3. Applications

Photoactive superconductivity holds great potential for future electronics, as it enables light-activated superconducting quantum devices. It suggests an architecture where electronic devices—which may be switching elements, transistors, or junctions—change state (e.g., from normal to superconducting, or vice versa) in response to light pulses rather than an electric gate or temperature, enabling reversible, ultrafast and energy efficient electrical processes. For example, a laser-triggered superconducting interconnect could conduct electrical signals with negligible resistance during the window of illumination, allowing high-speed, low-loss signal routing in integrated circuits. Similarly, photo-switchable superconducting junctions could act as logic or memory elements, toggling between normal resistive and superconducting states under light control, eliminating static power consumption and minimizing thermal dissipation.

This approach bypasses several limitations of conventional superconducting electronics, which require operation at cryogenic temperatures and often rely on fixed chemical doping or static strain to tune $T_c$ and coherence properties. In contrast, photoinduced superconducting properties that can be transiently accessed simply by shining light pulses—via reversible switching and without requiring modifications of the material—are highly desirable.

Superconducting interconnects—i.e., circuit elements that exhibit effectively zero electrical dissipation while in an active state—have already been proposed as integral components in superconducting qubits architectures [125]. In a photoinduced context, ultrafast light pulses may transiently turn a region of material into a superconducting (zero-resistance) state, allowing to route a signal with negligible losses between nodes in a circuit only during illumination.

Another recent work [115] proposed a light-controlled superconducting transistor where light intensity and phase directly modulate supercurrent magnitude and phase in a quantum dot Josephson junction, laying the groundwork for superconducting logic elements (diodes, transistors, etc.) [126,127].

Recent experiments on thin films of ${\mathrm{K}}_3{\mathrm{C}}_{60}$ exposed to mid-infrared pulses and integrated on a chip with picosecond electrical pulses [24]—which revealed a transient nonlinear current–voltage behavior characteristic of a critical current—may pave the way for ultrafast on-chip electrical switching devices [128].

Other applications include the results by Kazim et al. [117], who showed that Nb superconducting nanofilms switch to a high-resistivity state under optical illumination. In [116], Peña et al. demonstrated that visible light shone across a superconductor-insulator interface can dramatically increase $T_c$ and alter conductivity, providing a photoinduced superconductivity switching mechanism. In another experiment [114], UV light activated a molecular switch (spiropyran) to create an electric double layer; illumination converted a Mott-insulating organic crystal into a superconducting state at low temperature, reversibly controlled by light, functioning as a light-gated superconducting transistor.

Beyond interconnects, photo-activated superconductive devices could find other applications in ultrafast detectors (i.e., devices that switch to a superconducting state upon illumination, such as superconducting nanowire single-photon detectors [129]).

In summary, engineering superconductivity via light pulses may unlock a new paradigm for quantum electronic devices: ultrafast, low-loss electronic elements that function dynamically at temperatures above the conventional superconducting temperature $T_c$, circumventing the need for doping or continuous refrigeration, and paving the way toward novel superconducting electronics architectures.

3.2. Metastable Ferroelectricity and Multiferroic Ordering

The ultrafast control of ferroelectric phenomena has recently gained much interest because of their potential application to next-generation memory devices.

For example, Mankowsky et al. [26] demonstrated irreversible ultrafast reversal of the ferroelectric polarization in ${\mathrm{LiNbO}}_3$. The resonant excitation of an auxiliary phonon mode, switched the polarization transiently within ∼100 fs, as sensed via time-resolved SHG. Polarization reversal via pump pulse was also observed in Si-doped lead germanate [38]. These results were explained by Chen et al. [33], who showed how sequences of THz pulses along specific crystallographic directions can fully and deterministically reverse polarization in prototypical perovskite ferroelectrics, thus offering a clear pathway toward controlled ultrafast switching suitable for memory devices.

Polarization switching was also reported in ${\mathrm{CuInP}}_2{\mathrm{S}}_6$ [36] and in ferroelectric field effect transistors (FeFET) based on bilayer hexagonal boron nitride (hBN) [35]. Bilayer hBN is a sliding ferroelectric, and exhibits a polarization switching over nanosecond timescales, with endurance over ${10}^{11}$ cycles.

Several experiments on ${\mathrm{SrTiO}}_3$ [27,28], a quantum paraelectric at ambient conditions, demonstrated that intense mid-IR pulses induce a ferroelectric phase persisting for hours after the pump, as showed via time-resolved SHG. The mechanism is attributed to nonlinear phononics [119], where driving of a polar mode modifies the double-well potential to stabilize a ferroelectric minimum (see Figure 3a).

Light-induced ferroelectricity was also demonstrated in antiferromagnetic ${\mathrm{Cr}}_2{\mathrm{O}}_3$ [32]: a linearly polarized ultrafast light pulse transmitted through ${\mathrm{Cr}}_2{\mathrm{O}}_3$ (an antiferromagnet) induces a spontaneous electric polarization without moving lattice or spins. This enables control of the magnitude and direction of the polarization on ultrafast timescales. ${\mathrm{Cr}}_2{\mathrm{O}}_3$ is particularly interesting as it exhibits antiferromagnetic and ferroelectric orders, paving the way for ultrafast multiferroicity.

Ultrafast polarization reversal and deterministic THz switching enable ferroelectric memories–ferroelectric random access memory (FeRAM) [130,131] and FeFET [132,133,134]—with faster write-times (sub-ps to few-ps), demonstrating great potential for energy-efficient non-volatile memory in ferroelectrics. The bilayer hBN sliding-ferroelectric FeFET platform, combines low dimensionality, endurance, and fast switching, bringing atomically thin ferroelectric memories within reach, with great potential for integration with graphene channels.

The light-induced electronic dipole in antiferromagnetic ${\mathrm{Cr}}_2{\mathrm{O}}_3$ offers a potentially multiferroic operation mode: since only electronic ordering is affected, repeated switching at THz rates may be possible, with the advantage of reduced fatigue and reversible logic states relying purely on electronic symmetry rather than lattice displacement.

3.3. Photoinduced Insulator-to-Metal Transition

Vanadium dioxide (VO₂) is a prototypical strongly correlated oxide that undergoes a first-order insulator-to-metal transition near 340 K, accompanied by a structural distortion. The possibility of inducing this transition through ultrafast optical excitation, electric fields, or direct current injection has sparked intense research interest due to both potential applications in electronics [48].

A landmark study by Morrison et al. [47] used UED and mid-infrared TOS measurements to show that photoexcitation drives VO₂ into a metastable metal-like state without changing the lattice structure. As showed via trARPES, the metal state appears within ∼100 fs from the pulse [100], proving the decoupling of electronic and structural transitions, implying that a purely electronic route to metallicity is possible. This metastable state exhibits different properties from the equilibrium metal, highlighting a non-thermal pathway to phase control.

Nano imaging via trSNOM [51,52] and UED [49] revealed that strain also plays a role in the photoinduced switching of ${\mathrm{VO}}_2$ nanobeams.

These results demonstrate that ${\mathrm{VO}}_2$ has great potential for several electronic applications. ${\mathrm{VO}}_2$-based planar memristors [135] exploit nanoscale ${\mathrm{VO}}_2$ domains and focused electric fields to switch phase in ∼15 ps using femtojoule-scale voltage pulses, making them promising for ultrafast, energy efficient neuromorphic or memory circuits [136,137,138,139]. The application of ${\mathrm{VO}}_2$ to memory devices has also been studied in nanowires [140] and thin films [141,142] of ${\mathrm{VO}}_2$.

All these results, together with recent progress in device integration, especially in ultrafast memristive switching and hybrid photonic–electronic layouts, position ${\mathrm{VO}}_2$ as a promising material for ultrafast and low-energy applications to logic, memory, and signal processing.

3.4. More Light-Induced Phases

Many other experiments have demonstrated the photoinduced suppression or enhancement of different phases, including charge density wave order (CDW), spin waves, topological states and excitonic phases.

Charge density wave order—Several studies in transition-metal tellurides have demonstrated the ultrafast control of charge density wave and topological order. In ${\mathrm{LaTe}}_3$, Kogar et al. [42] used UED to observe a transient increase in the CDW order parameter, showing that low-fluence optical excitation enhances CDW amplitude instead of melting it. Follow-up work on ${\mathrm{LaTe}}_3$ [40] used TOS, UED and trARPES to reveal the nucleation and growth of a photoinduced CDW phase, forming domains via topological defects within ∼1 ps. In ${\mathrm{VTe}}_2$, which hosts CDW order intertwined with nontrivial band topology, Suzuki et al. [143] employed trXRD to track the ultrafast lattice transformation from a topological CDW phase to a non-topological phase. In ${\mathrm{CeTe}}_3$, Zhou et al. [144] were able to access a hidden CDW order via photoexcitation, as revealed by UED.

Spin waves—Advances were also recently made in the control of non-equilibrium spin waves in quantum magnets. It was demonstrated that photoexcitation pulses can reverse the magnetization [145], and that magnonic transport in insulators such as yttrium-iron-garnet (YIG) can be enhanced via static external fields [146,147]. Moreover, control of spin waves led to the creation of tunable nano-oscillators [148,149]. These spin-based phenomena do not exhibit Joule heating or thermal losses, establishing magnons as versatile carriers for low-power interconnects and electronic devices.

Topological and excitonic phases—Tuning of topological states via photoexcitation was studied in CDW systems, including ${\mathrm{1T-TaS}}_2$ [150]—where UED revealed reversible switching of topological order—and ${\mathrm{1T-TiSe}}_2$ [44,45]—where topological states were induced starting from a CDW phase. More experiments reported ultrafast light-induced anomalous Hall effect in graphene [108,151] and photoinduced chirality [39] and excitonic condensation [46].

Applications—The application of these non-equilibrium states to electronic devices is not as immediate as in the case of superconductivity and ferroelectricity. Nonetheless, the switching of charge density wave order shows potential for photo-switched electronic architectures as it is usually associated with metal–insulator transitions and anisotropic transport in the ordered phase. Photoinduced topological phases may also find application in ultrafast transistors and spintronic devices. Spin waves in quantum magnets can also find application in spintronic devices [152], enabling low-power logic gates [153,154], signal processing devices [148,149] and dissipationless magnonic transport [147].

4. Current-Driven Phases

Electric currents and voltages constitute another type of non-equilibrium drive. Rather than being excited into a transient state, the system is driven into a non-equilibrium steady state (NESS), which may be a different phase from the equilibrium phase.

Experiments on several materials have demonstrated the possibility of resistive switching, i.e., driving the system from an insulating to a metallic state and vice versa, via an applied current or voltage. In this section, we briefly review the experimental results on the two most promising materials—namely ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ and ${\mathrm{VO}}_2$—and their applications to electronics.

4.1. Current-Driven Insulator-to-Metal Transition in ${\mathit{Ca}}_2{\mathit{RuO}}_4$

Calcium Ruthenate ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ is a layered 4d transition-metal oxide, see Figure 5a, that exhibits a Mott insulating state [155] at low temperature, stabilized by cooperative lattice distortions and orbital ordering. Recent experiments have demonstrated that a direct current can drive a sharp insulator-to-metal transition (IMT), accompanied by dramatic changes in resistivity, structural symmetry, and even magnetic ordering, so that ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ has emerged as a key platform for studying non-equilibrium phase transitions.

Evidence of a current-driven IMT in Ca₂RuO₄ was firstly reported by Nakamura et al. [54], who observed that applying DC current induces a reversible first-order transition from the insulating to a metallic state. The transition is accompanied by a structural transition, see Figure 5b, as later revealed by neutron scattering, XRD and RIXS [58,59,60,157], and is asymmetric, with the metallic phase always nucleating out of the negative electrode [105], see Figure 5e. The transition is observed even at temperatures well below the thermal IMT temperature, see Figure 5c,d, and at relatively low current densities (∼$1{\mathrm{A/cm}}^2$), suggesting a minor role of Joule heating.

The current induced transition was also studied in ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ thin films [61,62,158], and in bulk samples via ARPES and transport measurements [57,58], which found band dispersion changes and an anisotropic suppression of the insulating gap, and via neutron diffraction [60], which revealed that the current also induces a new orbital-ordered state in Mn-doped Ca₂RuO₄, characterized by suppressed anti-ferromagnetism. These observations suggest that the non-equilibrium metallic state is electronically distinct from the high-temperature equilibrium metal and arises from a complex interplay between electronic and lattice degrees of freedom.

On the other hand, thermal imaging by Mattoni et al. [159] showed that the IMT occurs at the same local temperature as the equilibrium transition, but this local heating is spatially confined and correlated with phase coexistence between metallic and insulating domains.

All these results have stimulated theoretical interest, on the origin of the current-driven transition. While DMFT simulations [160] suggest a purely electronic mechanism [58,161], other works [162,163] argue that Joule heating and heating due to Seebeck effects at the metal–insulator interface play an important role.

4.2. Current- and Voltage-Driven Switching in VO₂

Parallel to the photoinduced switching discussed in Section 3.3, current- and voltage-driven transitions have been widely demonstrated in nanoscale VO₂ devices [66,67,69,70,71,76,78,164,165,166,167].

Sufficiently strong electric fields (∼${10}^7$ V/m) can induce the collapse of the insulating phase and trigger metallic conduction even at room temperature. The transition is ultrafast [78], reversible, and characterized by random telegraph noise [67]. Zhou et al. [78] realized vertical metal–insulator–metal (MIM) structures exhibiting reversible switching on nanosecond timescales, while Ha et al. [69] reported high-frequency modulation of VO₂-based devices under RF driving fields. Crunteanu et al. [76] demonstrated robust electrical cycling over ${10}^8$ switching events, reinforcing the potential of VO₂ for applications in electronics.

While Joule heating contributes to the transition in many device configurations, several experiments and modeling studies point to a non-thermal origin. Shi et al. [66] developed a phase-field model incorporating electron correlation and local heating, showing that current can induce IMT by weakening electronic correlations without a global increase in temperature. An isostructural electronic IMT, consistent with the formation of an electronically driven metallic state, was observed in epitaxial films [71,165].

Experimental studies in electronic device architectures confirm practical relevance: low-threshold voltage switching with high ON/OFF contrast has been achieved in planar memristor and vertical MIM geometries, operating at room temperature with nanosecond switching speed and high endurance [66,164]. Other works using ${\mathrm{VO}}_2$ films confirm field-induced IMT reproducibly, with reversible switching over many cycles, demonstrating device stability and scalability [76,167].

4.3. Applications in Electronics and Resistive Switching

The findings reviewed in Section 4.1 and Section 4.2 collectively suggest that current-driven IMTs in correlated oxides have great technological potential for next-generation electronics. In particular, the ability to reversibly and reliably toggle the system between insulating and metallic states using modest electrical currents opens exciting avenues for resistive switching applications. Other applications in electronics leverage field-driven transitions for low-energy processing of logic and memory signals.

Unlike traditional Mott insulators, where transitions require high fields or structural doping, ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ only require modest current densities or electric fields, demonstrating a low-power and directionally controllable switching behavior. These properties are highly desirable for developing Mott-based transistors [155] and neuromorphic computing elements [168,169]. Moreover, the sensitivity of the transition to lattice and orbital degrees of freedom implies that multi-functional responses—such as coupling to strain—could be integrated into complex device architectures, making ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$ a candidate for memristive switching.

The experimental results reviewed in Section 4.2 show the potential of VO₂ as a versatile platform for ultrafast, non-volatile switching electronics. Crunteneau et al. [76] demonstrated its ability to rapidly, consistently and reliably switch between insulating and metallic phases, with an operational lifetime of up to ${10}^8$ cycles. Devices based on ${\mathrm{VO}}_2$ devices could serve as two-state switches with minimal input energy (fJ) and fast operational times (ns), low-power Mott field-effect transistors, or active elements in THz modulators. Indeed, demonstrative experiments have already showed that such ${\mathrm{VO}}_2$ elements can be used to enable neuromorphic computing [136,138,139] and frequency-selective responses [137]. These combined advances position field-driven ${\mathrm{VO}}_2$ as a promising platform for beyond-CMOS computing architectures, and energy-efficient, ultra-fast switches [135] in future electronics.

5. Discussion

The experimental advances reviewed in the previous sections highlight the growing potential of non-equilibrium control in quantum materials as a foundational principle for future electronics.

Researchers have driven materials with ultrafast pulses of light or electrical current, managing to access an expanding landscape of transient and metastable phases—including superconductivity, ferroelectricity, charge density waves, and topological states—many of which are hidden in equilibrium and unreachable through conventional tuning parameters such as pressure or chemical doping.

One of the most promising features of these systems is that fundamental material properties—such as conductivity, polarization, and coherence—can be reversibly and dynamically tuned on ultrafast timescales. This opens the door to new classes of devices that operate through time-dependent functional states that can be changed within nanoseconds without the need to restructure the entire material architecture. In this framework, light or current are active control knobs—enabling electronic devices that are triggered by pulses, or programmed in time.

The implications for low-power, high-speed electronics are profound, as non-equilibrium phases are typically accessed within very short timescales (femtoseconds to picoseconds), with light pulses that require minimal energy, down to microJoules or even nanoJoules.

Light-induced superconductivity, for example, could provide ultrafast, dissipationless interconnects that operate above the equilibrium critical temperature, eliminating the need for continuous refrigeration. Ultrafast ferroelectric switching offers a new route to non-volatile memory elements with picosecond write speeds and minimal energy consumption. Current-induced metal–insulator transitions in correlated oxides point toward neuromorphic components and Mott transistors, operating beyond the limits of current technology.

In

Table 1. Overview of the most promising non-equilibrium phases (first column), the material on which they have been studied or implemented (second column), the corresponding applications to electronic devices (third column) and the relevant references (fourth column).

Phase	Material	Use Case	References
SC	on-chip ${\mathrm{K}}_3{\mathrm{C}}_{60}$, Nb nanowires	Superconducting switches and interconnects	[24,117,125,128]
	${\mathrm{K}}_3{\mathrm{C}}_{60}$	Superconducting transistors	[115,126,127]
IMT	doped ${\mathrm{VO}}_2$	Smart devices	[70,77,167]
	${\mathrm{VO}}_2$	Resistive Switch	[69,76,77,78,135,170]
	Nanoscale${\mathrm{VO}}_2$	Memory	[135,140,141]
	${\mathrm{VO}}_2$ memristor arrays	Neuromorphic computing	[136,137,138,139]
FE	${\mathrm{SrTiO}}_3$	Ultrafast memory	[27]
	hBN	Memory (FeFET)	[35]
	${\mathrm{Cr}}_2{\mathrm{O}}_3$	Electronic FeRAM	[32]

, we summarize the non-equilibrium phases that have the greatest potential for electronic applications. We include the most promising candidate materials, their potential for technological applications, along with the relevant literature references.

Beyond technological relevance, these non-equilibrium states are interesting for fundamental physics. They reveal how electron correlations, lattice dynamics, and topology conspire to form emergent orders—and how these orders can be manipulated, enhanced, or suppressed under dynamic conditions.

Looking forward, a key frontier will be the stabilization and control of these driven phases for extended durations and in scalable architectures. Integrating ultrafast drive mechanisms with existing electronic architectures will be essential. The convergence of non-equilibrium materials science, and electronics has the potential to spawn a new class of hybrid devices based on dynamical control of electronic properties.

Challenges

Although non-equilibrium phases in quantum materials exhibit rich phenomenology with great technological potential, several critical challenges still need to be overcome before these phenomena can be translated into scalable and robust technologies.

Theoretical interpretation—A central open question in the study of photoactive superconductivity concerns the true nature of the transient state observed in photoexcited cuprates and fullerides, as it is still unclear whether the light-induced state corresponds to genuine superconductivity, characterized by macroscopic phase coherence and Meissner effect, or rather to a high-mobility metallic state with superconducting-like optical features. Although the experimental signatures in ${\mathrm{K}}_3{\mathrm{C}}_{60}$ [19,20,21,22,23,24] and YBCO [7,14,15,16,17,18] have demonstrated transient responses that resemble those of a superconducting state, the lack of direct evidence for zero resistance or magnetic flux expulsion, as well as the mismatch in penetration depth between pump and probe pulses, leaves room for alternative interpretations.

Several mechanisms have been proposed, one of them invoking nonlinear phononics, where the driven phonon mode modifies the lattice potential landscape to stabilize a superconducting ground state [21,109,110]. Other theories propose radiation-induced quasiparticle cooling [120] or enhancement of superconducting fluctuations without full coherence [23]. On the other hand, critical analyses have argued that the observed conductivity spectra could also be fitted by non-superconducting models of transient metallic states with negative conductivity [121,122] or with non-linear effects in pump fluence [123]. Thus, further experimental efforts are required to conclusively establish whether these states are superconducting in the strict thermodynamic sense. Understanding this distinction bears direct implications for the design of superconducting electronics based on optical control.

A similar issue arises with the nature of the current-driven IMT in ${\mathrm{Ca}}_2{\mathrm{RuO}}_4$, for which a comprehensive explanation has still not been found. While some theoretical interpretations favor an electronic origin [58,161], other results point to some sort of heating mechanism [159,163]. A predominant role of thermal heating in the switching mechanism would limit applicability to resistive switching devices.

Reproducibility—A major obstacle, common to most results presented in this manuscript, is reproducibility across samples and platforms. Many of the reported photoinduced or current-driven effects are highly sensitive to sample quality, and device geometry, so that small variations in crystalline order, or contact resistance can lead to different outcomes. Standardizing the protocols for material preparation will be essential to move beyond proof-of-concept demonstrations.

Scaling and compatibilit—For practical use in new electronic devices, non-equilibrium quantum materials must be scaled up in size and need to be compatible with device architectures that support high-density packing, and low-power operation. This imposes stringent constraints on materials, thermal losses, and drive mechanisms.

Thermal losses—Thermal management is a critical challenge. Many of the transitions discussed — especially current-driven ones—involve localized heating [163,171] that may result in energy losses and hinder long-term device stability. Although some experiments have convincingly demonstrated non-thermal pathways (e.g., ultrafast transitions in ${\mathrm{VO}}_2$ or ferroelectric materials), distinguishing true electronic switching from thermally assisted mechanisms remains a challenge. In this sense, the investigation of hot-electrons dynamics [85,86] can help determining whether the non-equilibrium state arises simply from thermal heating or from true electronic mechanisms. Furthermore, repeated pulsed operation may lead to fatigue or excessive heating of the device.

Lifetime—Finally, the lifetime and coherence of the driven phases must be extended or stabilized for real-world operation. While ultrafast dynamics is beneficial for speed, many non-equilibrium states remain limited to picosecond–nanosecond timescales or require sustained external driving. Engineering ways to trap or sustain these states is an active area of research that could unlock persistent device functionality.

6. Outlook

In summary, the results presented in this review point to a paradigm shift in our approach to materials functionality. By embracing non-equilibrium control as a design principle, we gain access to a vastly richer set of electronic behaviors and device concepts.

Vanadium dioxide is the material with the highest level of technology readiness for novel electronic devices. It promises a large number of possible applications, which have been demonstrated in a laboratory setting, and some of them have even been tested for reproducibility and endurance. Ultrafast ferroelectricity is based on a large class of ferroelectric materials, and provides a versatile and promising platform for ultrafast memory applications, which have recently been demonstrated. Light-induced superconductivity in cuprates and fullerides has groundbreaking potential for electronics, but researchers in this field still need to reach a complete theoretical understanding and achieve reproducibility of the experimental results.

In summary, the path to the implementation of non-equilibrium phases on functional devices still requires coordinated advances in material growth, ultrafast driving techniques, device engineering, and theoretical understanding. Overcoming these challenges will be central to realizing the full potential of non-equilibrium quantum materials in next-generation electronics.