ACS Publications. Most Trusted. Most Cited. Most Read
My Activity
CONTENT TYPES

Charge Configuration Memory Devices: Energy Efficiency and Switching Speed

  • Anze Mraz*
    Anze Mraz
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    Faculty of Electrical Engineering, University of Ljubljana, Tržaška cesta 25, 1000 Ljubljana, Slovenia
    *Email: [email protected]
    More by Anze Mraz
  • Rok Venturini
    Rok Venturini
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    Faculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta 19, 1000 Ljubljana, Slovenia
  • Damjan Svetin
    Damjan Svetin
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    CENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Vitomir Sever
    Vitomir Sever
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Ian Aleksander Mihailovic
    Ian Aleksander Mihailovic
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Igor Vaskivskyi
    Igor Vaskivskyi
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Bojan Ambrozic
    Bojan Ambrozic
    CENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Goran Dražić
    Goran Dražić
    Jozef Stefan International Postgraduate School, Jamova cesta 39, 1000 Ljubljana, Slovenia
    Department of Materials Chemistry, National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
  • Maria D’Antuono
    Maria D’Antuono
    Dipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, Italy
    CNR-SPIN, Complesso Monte Sant’Angelo, Via Cinthia, 80126 Napoli, Italy
  • Daniela Stornaiuolo
    Daniela Stornaiuolo
    CNR-SPIN, Complesso Monte Sant’Angelo, Via Cinthia, 80126 Napoli, Italy
    Dipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, Italy
  • Francesco Tafuri
    Francesco Tafuri
    Dipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, Italy
    CNR-Istituto Nazionale di Ottica (CNR-INO), Largo Enrico Fermi 6, 50125 Florence, Italy
  • Dimitrios Kazazis
    Dimitrios Kazazis
    Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
  • Jan Ravnik
    Jan Ravnik
    Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
    More by Jan Ravnik
  • Yasin Ekinci
    Yasin Ekinci
    Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
    More by Yasin Ekinci
  • , and 
  • Dragan Mihailovic
    Dragan Mihailovic
    Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    CENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, Slovenia
    Faculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta 19, 1000 Ljubljana, Slovenia
    Jozef Stefan International Postgraduate School, Jamova cesta 39, 1000 Ljubljana, Slovenia
Cite this: Nano Lett. 2022, 22, 12, 4814–4821
Publication Date (Web):June 10, 2022
https://doi.org/10.1021/acs.nanolett.2c01116

Copyright © 2022 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY 4.0.
  • Open Access

Article Views

1809

Altmetric

-

Citations

LEARN ABOUT THESE METRICS
PDF (3 MB)
Supporting Info (1)»

Abstract

Current trends in data processing have given impetus for an intense search of new concepts of memory devices with emphasis on efficiency, speed, and scalability. A promising new approach to memory storage is based on resistance switching between charge-ordered domain states in the layered dichalcogenide 1T-TaS2. Here we investigate the energy efficiency scaling of such charge configuration memory (CCM) devices as a function of device size and data write time τW as well as other parameters that have bearing on efficient device operation. We find that switching energy efficiency scales approximately linearly with both quantities over multiple decades, departing from linearity only when τW approaches the ∼0.5 ps intrinsic switching limit. Compared to current state of the art memory devices, CCM devices are found to be much faster and significantly more energy efficient, demonstrated here with two-terminal switching using 2.2 fJ, 16 ps electrical pulses.

This publication is licensed under

CC-BY 4.0.
  • cc licence
  • by licence

Research in the area of novel memory devices has been intense in the recent decades, but there have been few breakthroughs that have led to implementation. (1−3) As a result, technologies such as cryocomputing that promise large improvements in energy consumption have been seriously hindered by the absence of a suitable fast and energy-efficient memory for more than two decades. (4,5) As an alternative to modern magnetic devices, manipulating charge, rather than spins, for data storage could be more efficient because it can be directly driven by electrical charge injection that promises to be extremely fast and efficient. Unfortunately, the direct coupling of an electronic two-level system to lattice degrees of freedom causes rapid state decoherence and dissipation, which fundamentally limits charge-storage based memory concepts. As an apparent solution to this problem, it was shown recently that topological protection that stabilizes different charge density wave (CDW) domain states in the quasi-2D layered transition metal chalcogenide 1T-TaS2 can be used for memory devices. (6,7) The basic mechanism for such a charge configuration memory (CCM) was shown to be unique to this material, (7−9) involving reversible charge reconfiguration from an insulating, spatially uniform CDW state to a metallic domain state (10) with accompanying restacking of the CDW along the direction perpendicular to the layers. (11) To develop the CCM concept into viable practical devices, a better understanding of the device characteristics is required, particularly the operational limitations, such as write speed, energy efficiency, and endurance, as well as size scaling limitations and contact material compatibility. For low temperature operation, minimization of dissipation is essential, so investigation of how the switching energy scales with device size is particularly important. The write (W) cycle of the CCM device was shown to be nonthermal, while the erase (E) cycle is at least partially thermal. (7) Intrinsic nonohmic behavior and dissipation at the contacts, (12) especially if energy barriers are formed as a result of interface chemistry, may limit the energy efficiency in small devices. (13) In this paper we investigate CCM scaling properties in the nonvolatile resistance switching region, which is particularly relevant for incorporation into cryocomputing environments where the device has potential applications. We investigate the electrical contact structure with high resolution electron microscopy, searching for possible interfacial layers that may limit device performance. We conclude by comparing the measured characteristics of CCM with other current and emerging memory devices.

The devices were nanofabricated by evaporating metal contacts on 1T-TaS2 single crystal flakes as can be seen on a scanning electron microscope (SEM) image in Figure 1a (see Methods in Suppporting Information). Switching characteristics were measured on many devices with very different gap sizes L (Figure 1b) between the Au electrodes, using a large range of “write” pulse lengths τW, from 16 ps to 600 ms. For short pulse switching, low-attenuation transmission line circuits were used in combination with picosecond digital electronics (see Methods in Supporting Information). We first present an analysis of the contact structure in Au/1T-TaS2 devices on Si/SiO2 substrates. A bright-field scanning transmission electron microscope (BF-STEM) image of a cross-section of two electrodes (one memory bit) and a zoom-in on the interface between the metal electrodes and 1T-TaS2 can be seen in Figure 1b and Figure 1c, respectively. The layered structure of the 1T-TaS2 crystal is clearly visible in Figure 1c. Part of the cross-section showing the entire structure of the device can be seen in Figure 1d. An energy-dispersive X-ray spectroscopy (EDS) analysis of the device’s interfaces is presented in Figure 1e, along the line scan marked with the red line in Figure 1d. Considering the overlap in the EDS spectrum for certain characteristic lines of elements (Ta M and Si K, Au M and S K), the individual layers of the fabricated CCM device can be identified (Figure 1e). Importantly, multiple EDS analyses (at ∼1 nm resolution) on different devices do not show any evidence of an oxide layer at the interface between the 1T-TaS2 crystal and the metal electrode, consistent with a previous report (13) and the voltage–current characteristics discussed below.

Figure 1

Figure 1. CCM device, EDS, and work function analysis. (a) SEM image of a typical CCM device used for measurements with transmission line contacts. (b) BF-STEM image of a cross-section of a fabricated CCM device (one memory bit). (c) Zoom-in to the interface between the metal electrode and 1T-TaS2 crystal. (d) Zoom-in to a part of the cross-section with the EDS line scan marked. (e) EDS analysis of the zoomed-in section, where individual layers are identified: Au (yellow), Pd (green), Ta (dark blue), S (light blue), Si (red), O (black), Ti (pink). (f) Top panel shows an AFM image of a part of the CCM device with the line scan for the KPFM measurement marked. Bottom panel shows KPFM measurement of the work functions for 1T-TaS2, SiO2, and Au electrodes along the line scan. (g) Schematic band diagram of the device based on the KPFM measurements. On the left is a band diagram of an interface between the Au electrode and the nearly commensurate (NC) metallic state at room temperature, and on the right is an interface between the Au electrode and the commensurate (C) charge density wave state at cryogenic temperature.

To further ascertain the possible role of contacts in device functionality we measured the relative difference in work functions of the 1T-TaS2 and Au in a CCM device using a Kelvin probe force microscopy (KPFM). The upper panel in Figure 1f shows an atomic force microscope (AFM) image of a part of the CCM device, while the bottom panel in Figure 1f shows the KPFM potential difference as a function of the scanning position on the device. The scanning is performed across the 1T-TaS2 crystal, the Si/SiO2 substrate and the Au electrode, as indicated with the dashed line in Figure 1f, top panel. While the absolute value of was not calibrated, the relative difference of work functions is accurately determined,

in agreement with published values for = 5.5 ± 0.01 eV (single crystal) (14) and = 5.1–5.47 eV. (15) Judging from the KPFM measurement and the values from ref (15), it looks like our gold layer has facets with predominantly (100) or (110) orientation. The corresponding band diagrams for the room-temperature nearly commensurate (NC) metallic state and the low-temperature commensurate charge density wave (C-CDW) are shown in Figure 1g. Ideally electron injection from Au metal into the 1T-TaS2 CDW-gapped semiconductor with a gap eV  (16) takes place (Figure 1g), and the junction is expected to be ohmic. 1T-TaS2 reacts to charge injection by forming a conducting interface layer parallel to the interface. We thus do not expect a barrier to form at the interface with Au contacts due to carrier diffusion, which is consistent with the observed linear voltage–current curves (6) and the scaling behavior with device size and τW described in the following section.

While basic memristive switching of the CCM device at higher temperatures was discussed elsewhere, (6) a measurement that demonstrates millikelvin operation and switching from the high resistance (HI) state to low (LO) resistance state is illustrated by the RT curve in Figure 2a, for a device with intercontact distance nm. The entire measurement cycle is shown starting from 280 K, cooling and switching at 350 mK by a single write (W) pulse, and heating back to 280 K, where the system reverts to its original state. Upon heating we observe characteristic steps, previously attributed to relaxation of domains. (7,17) The insert to Figure 2a shows the RT curves on an expanded scale, showing a commonly observed hysteresis associated with the presence of additional phases (18,19) above 100 K.

Figure 2

Figure 2. Resistance switching and voltage–current characteristics of CCM devices. (a) Temperature dependence of the four-contact resistance R. Switching from RHI to RLO at 350 mK is caused by an electrical W pulse as indicated by the arrow. Heating above 90 K (red line) reverts the system to the RHI state. Inset to (a) shows the expanded scale of the RT curve. (b, c) Pulsed measurements of the VI curve for the W and E operations, respectively.

A typical voltage–current (VI) curve for a W cycle at 20 K is shown in Figure 2b for incrementally increasing current pulses through the CCM device with an intercontact distance nm (see insert to Figure 3b). In this measurement, the pulses were very long (), illustrating the versatility of the device (fast measurements are shown later). The VI measurement allows us to accurately determine the switching threshold for both the write and erase processes. We observe a nonlinear VI curve with an initial slope (dashed line) up to the mA. Above this current the voltage drops from 0.65 to 0.08 V and the device switches to a linear VI relation with resistance (dashed line), remaining in this state indefinitely at this temperature. The VI curve for W can be fit with

for (green line) and for (dashed line) in agreement with ref (6). To switch back to the pristine state, the erase (E) sequence is illustrated in Figure 2c where the current is ramped from to 0.4 mA. Initially we follow the RLO curve. Above a threshold mA the device switches to a high resistance state. As current is lowered to zero, the device remains in at low currents. The behavior is consistent with the previous VI measurements with 10 μs pulses, (6) implying that it is not dependent on the pulse length over the range τW = 10–5–1 s.

Figure 3

Figure 3. Speed and energy efficiency scaling at 20 K. (a) Switching energy density εW as a function of pulse length τW. The inset shows the actual pulse shapes. Red line shows linear scaling, and blue line shows departure from linearity at short τW. The data point at 1.9 ps was taken from ref (21). (b) Switching threshold voltage VW as a function of distance between the electrodes L. The inset shows a device with variable L used in the measurement. Different symbol colors are for different physical devices. (c) Endurance measurement showing cycling between RLO and RHI for 106 cycles. Each pair of points represents 2 × 104 W/E cycles.

Using 60 GHz Au transmission line contacts, fabricated on ∼50 nm thick 1T-TaS2 flakes on Si/SiO2 substrates (see Figure 1a, Methods in Supporting Information), we systematically study the W sequence of multiple CCM devices for different W pulse lengths τW ranging from 16 ps to 20 μs generated by electronic pulse generators in a single-pulse mode. For each pulse length, the pulse voltage was incrementally increased until the switching threshold was reached and recorded. Since the depends on sample geometry, in Figure 3a, we plot the switching energy density instead: ( is the crystal volume between electrodes, and is defined as the switching energy). The error bars include the systematic error of measurements and the estimated variation between different devices. The red line shows a linear relation , with . At shorter pulse lengths there is a departure from linearity denoted with the blue line, which most likely occurs due to an increase in transmission line losses in the GHz range. For the fastest device measured here (16 ps fwhm), the distance between contacts is and threshold W voltage , which gives a switching energy per bit of at 20 K. The resistance measured before and after switching is and , respectively. The resistance ratio is lower than the value observed with longer W pulses at low temperatures such as that shown in Figure 2a, which may be attributed to incomplete switching. However, both and resistance states have long-term stability at this temperature (20 K). To explore the switching capabilities of the CCM device in the ultrafast limit, an electro-optical setup that allows for ps pulse generation and detection was reported in ref (21), from which we include a data point = 1.9 ps in Figure 3a.

The scaling of the switching threshold voltage with intercontact distance is shown in Figure 3b. (A device with multiple L is shown in the inset.) Approximately linear scaling of is observed over nearly 2 orders of magnitude, 60 nm < L < 4 μm, that extrapolates to the origin, with a slight departure from linearity at the smallest . Considering the fact that the EDS analyses do not show any intermediate layer between Au and 1T-TaS2 and that the work function mismatch suggests ohmic contact behavior, we cannot directly attribute the observed small deviation from linearity to electrical contacts. One possible reason for the departure at small sizes may be related to the fact that the device size approaches the intrinsic 10–20 nm domain size measured in the low-resistance state of single crystals. (9,20) The efficiency of devices with feature sizes comparable to the domain size may be expected to decrease due to charge configuration pinning at the contacts.

In Figure 3c we show a typical endurance measurement at 20 K for a device with nm. Square pulses with and amplitude V are used for W, and asymmetric triangular E pulses with peak amplitude V and are used for erase (shown in the inset to Figure 3c). Each pair of corresponding and values shown by black and red dots, respectively, represents a measurement after 20 000 W/E cycles. We see that is remarkably stable over cycles, while initially increases slightly and later stabilizes after cycles when the measurement was terminated.

An immediate application of the device could be in cryocomputing, which has been heralded as an obvious solution to the overall challenge of reducing dissipation of computer systems. (5,22) In spite of huge research efforts and availability of superconducting circuits performing both single flux quantum (SFQ) and quantum information processing, (22) the absence of a fast low-energy cryogenic memory has prevented significant upscaling, (4,5) so CCM devices may offer a possible breakthrough. Considering the scaling limits of CCM devices, we find that while the measured value of is small compared with other current memory devices, the observed scaling laws suggest that can be reduced further by reducing and/or . Optimization of microwave electrical contacts seems to be essential in reducing the as well, since the losses in the transmission of GHz pulses are likely to be the cause for the departure from linear scaling in Figure 3a. Deviations from linearity might still be expected when approaches the intrinsic switching time . (23) These effects are likely to be important for device optimization, particulary for erase protocols. (7) Fundamentally, the measured is still significantly larger than the measured microscopic barrier EB = 15–21 meV ((2.4–3.4) × 10–6 fJ) obtained from thermal activation measurements. (17) This implies that in principle devices with much smaller can be built before reaching fundamental limits. It is also instructive to compare and with the lowest possible energy difference between two states that can be discerned thermodynamically on the basis of their entropy difference (the Brillouin–Landauer (BL) thermodynamic limit (24)). At K, fJ, so the presented CCM devices in Figure 3 are still far from this limit. For cryogenic memory applications, it is important that the CCM device could be driven by single flux quantum (SFQ) pulses. For a single SFQ pulse, the pulse energy may be estimated to = fJ, where we have assumed a typical critical current μA in the SFQ driver circuit. For a realistic CCM device with nm2 device area and a crystal thickness of 20 nm, the estimation from the linear fit in Figure 3a for switching with 2 ps (21) pulses gives fJ. Thus, on the basis of the presented scaling laws, CCM devices constructed using current fabrication techniques could be driven with single SFQ pulses, provided the coupling between the SFQ driver and the CCM microwave circuit is optimized.

In Figure 4 we compare operating parameters of the CCM with leading alternative technologies, including magnetic random-access memory (MRAM), phase change memory (PCM), and others. The smallest energy/bit values of 6 fJ/bit (25) and 8 fJ/bit (26) were reported for voltage-controlled magnetic anisotropy switching (MRAM) in Ta/CoFeB/MgO magnetic tunnel junctions (MTJ) and resistive switching in Ni/GeOx/HfON/TaN resistive random access memory (RRAM) devices, respectively (Figure 4b). The lowest theoretically predicted value for MTJs of a few (27) is comparable to the , which was already experimentally achieved in CCM. This value is still much higher than the theoretical limit of fJ predicted for CCM. PCMs have both significantly higher switching energies (>600 fJ  (28)) and switching voltages (>1 V  (29,30)) than CCM, and even the smallest memristors such as 6 nm crossbar memristor arrays have a write power of , (31) which is higher than the predicted ∼0.1 for a much bigger CCM device ( nm3), estimated from the linear fit in Figure 3a. We note that our demonstration of a man-made nonvolatile electronic memory device uses less energy per bit (2.2 fJ/bit) than a human brain (∼10 fJ/synapse) and potentially much less than artificial synapses. (32) The CCM’s advantage is that it is >9 orders of magnitude faster.

Figure 4

Figure 4. Measured switching energy EW and speed of leading memory devices: (a) switching energy in correlation with device area; (b) switching times τW plotted against switching energy. References: PCM, (28,29) RRAM, (26,33−35) STT-RAM, (36−38) MRAM, (25,39) nMem, (40,41) JJ-CMOS memory, (42) OST-RAM, (43) Mott memory, (44) SRAM, (45−47) and DRAM. (47)

Picosecond switching speeds of CCMs are similar to those reported by photomagnetic recording. (48) However, the switching energy density per bit is significantly smaller for CCM: (0.3 J/cm3 for CCM (21) vs 6 J/cm3 for photomagnetic recording), with the added advantage of two-contact electrical W/E and read (R) operations and the possibility of a very high packing density. For CCMs the current size limit is 10–20 nm, but for photomagnetic recording the size is limited by optical wavelengths used for W/E/R (hundreds of nm). Nanowire memory (nMem) (40,41) based on superconductive loops and nanowire cryotrons, Josephson magnetic random access memory (JMRAM), (49) superconducting quantum interference device (SQUID) memory, (50) and hybrid Josephson complementary metal oxide semiconductor (JJ-CMOS) memory (42,51,52) are all very promising solutions for a cryogenic memory. However, they are hard to scale down to the 10–20 nm regime and require additional driving periphery (4,53) (nMem, JMRAM and SQUID memory) or require voltage amplification for proper operation (4) (JJ-CMOS memory), which introduces additional dissipation into the circuits.

Thus, with presently demonstrated scalability (feature size, 60 nm to 4 μm; pulse length, 16 ps to 600 ms; voltage, 0.3–10 V) and with a wide operating temperature range (350 mK to 190 K), the CCM devices appear to be very versatile in comparison. The operating temperature range could be extended to higher T by appropriate choice of substrate. (17) So far, no other electronically ordered material has been found to exhibit significant metastability and thus 1T-TaS2 currently is the only material for CCM application. The device variability is small, as seen from the data on multiple devices shown in the scaling plots. Importantly for applications, thin films of 1T-TaS2 can be grown by various means (54−56) promising technological flexibility. For interfacing present CCM devices with single flux quantum devices, which is an obvious target application, nanocryotrons (nTrons) have been demonstrated to be a match in terms of output voltage, speed, and impedance. (57,58) However, nTrons could introduce additional design complexity and dissipation due to higher bias currents, (51) so direct driving by SFQ pulses might be preferable.

In comparison to complementary metal oxide semiconductor (CMOS) devices (42,51,52,59) and nanowire memories, (40,41) the CCM concept potentially offers advantages in terms of scaling, size, speed, energy efficiency, and operational simplicity. Compared with other fast memory, such as photomagnetic storage, (48) CCMs are more energy efficient and offer much higher data packing densities. The disadvantage is low-temperature operation, but considering its primary virtues are speed and energy efficiency, the target applications are in the cryocomputing environment. The presently measured energy efficiency (2.2 fJ/bit) is a consequence of very short write pulses needed, allowed by the inherent switching mechanism. Thus, the values shown here are limited by device size and transmission characteristics of the high-speed microwave circuit, not the intrinsic mechanism.

We conclude that while 1T-TaS2 appears to be unique in exhibiting combined set of properties useful for CCM operation, multilayer structures may introduce additional functionality that may result in improved performance and extended temperature range of operation. The additional degrees of freedom arising from interlayer interactions and proximity coupling (60) in 2D heterostructures may be expected to offer new possibilities for new CCM devices beyond the currently available materials.

Supporting Information

ARTICLE SECTIONS
Jump To

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.nanolett.2c01116.

  • Description of materials and fabrication methods; description of DC and fast electrical measurement (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

ARTICLE SECTIONS
Jump To

  • Corresponding Author
    • Anze Mraz - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, SloveniaFaculty of Electrical Engineering, University of Ljubljana, Tržaška cesta 25, 1000 Ljubljana, SloveniaOrcidhttps://orcid.org/0000-0002-8030-2628 Email: [email protected]
  • Authors
    • Rok Venturini - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, SloveniaFaculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta 19, 1000 Ljubljana, Slovenia
    • Damjan Svetin - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, SloveniaCENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, Slovenia
    • Vitomir Sever - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    • Ian Aleksander Mihailovic - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    • Igor Vaskivskyi - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, Slovenia
    • Bojan Ambrozic - CENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, Slovenia
    • Goran Dražić - Jozef Stefan International Postgraduate School, Jamova cesta 39, 1000 Ljubljana, SloveniaDepartment of Materials Chemistry, National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, SloveniaOrcidhttps://orcid.org/0000-0001-7809-8050
    • Maria D’Antuono - Dipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, ItalyCNR-SPIN, Complesso Monte Sant’Angelo, Via Cinthia, 80126 Napoli, Italy
    • Daniela Stornaiuolo - CNR-SPIN, Complesso Monte Sant’Angelo, Via Cinthia, 80126 Napoli, ItalyDipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, Italy
    • Francesco Tafuri - Dipartimento di Fisica “Ettore Pancini”, Università di Napoli Federico II, Monte S. Angelo via Cinthia, 80126 Napoli, ItalyCNR-Istituto Nazionale di Ottica (CNR-INO), Largo Enrico Fermi 6, 50125 Florence, Italy
    • Dimitrios Kazazis - Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, SwitzerlandOrcidhttps://orcid.org/0000-0002-2124-2813
    • Jan Ravnik - Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
    • Yasin Ekinci - Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
    • Dragan Mihailovic - Complex Matter Department F7, Jozef Stefan Institute, Jamova cesta 39, 1000 Ljubljana, SloveniaCENN Nanocenter, Jamova cesta 39, 1000 Ljubljana, SloveniaFaculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta 19, 1000 Ljubljana, SloveniaJozef Stefan International Postgraduate School, Jamova cesta 39, 1000 Ljubljana, Slovenia
  • Author Contributions

    A.M., R.V., I.A.M., V.S., and I.V. performed electrical measurements experiments. A.M., D.Sv., D.K., J.R., and Y.E. fabricated the devices. M.D., D.St., and F.T. performed the 350 mK switching experiments. B.A. prepared the lamella for STEM analysis. G.D. performed STEM analysis. A.M., D.Sv., and D.M. devised the experiments. D.M. and A.M. wrote the paper.

  • Notes
    The authors declare no competing financial interest.

Acknowledgments

ARTICLE SECTIONS
Jump To

This project has received funding from the EU-H2020 research and innovation program under Grant Agreement 654360, NFFA-Europe, having benefited from the access provided by Paul Scherrer Institute in Villigen, Switzerland within the framework of the NFFA-Europe Transnational Access Activity. We acknowledge the help of L. Cindro from F9 at JSI on contact bonding. We thank the support from the Slovenian Research Agency (Grant P1-0040; Grant PR-08972 to A.M., Grant PR-10496 to R.V., Grant PR-06158 to A.K., Grant PR-07589 to J.R., Grant I0-0005 to D.S., Grant J2-3041 to G.D.), Slovene Ministry of Education, Science and Sport (Grant C3330-19-952005, Raziskovalci-2.1-IJS-952005), ERC AdG (Grant GA320602, TRAJECTORY), and ERC PoC (Grant GA767176, Umem4QC). We thank the CENN Nanocenter for the use of AFM, FIB, and DaLI. This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie Grant Agreement No 701647, PSI-FELLOW-II-3i. F.T. has been supported by the Project SQUAD-Programma STAR PLUS 2020.

References

ARTICLE SECTIONS
Jump To

This article references 60 other publications.

  1. 1
    Holmes, D. S.; Kadin, A. M.; Johnson, M. W. Superconducting Computing in Large-Scale Hybrid Systems. Computer 2015, 48, 3442,  DOI: 10.1109/MC.2015.375
  2. 2
    Soloviev, I. I. Beyond Moore’s technologies: operation principles of a superconductor alternative. Beilstein J. Nanotechnol. 2017, 8, 26892710,  DOI: 10.3762/bjnano.8.269
  3. 3
    Janod, E. Resistive Switching in Mott Insulators and Correlated Systems. Adv. Funct. Mater. 2015, 25, 62876305,  DOI: 10.1002/adfm.201500823
  4. 4
    Tafuri, F., Ed. Fundamentals and Frontiers of the Josephson Effect; Springer Series in Materials Science, Vol. 286; Springer Nature, 2019; pp 632648.
  5. 5
    Holmes, D. S.; Ripple, A. L.; Manheimer, M. A. Energy-Efficient Superconducting Computing─Power Budgets and Requirements. IEEE Trans. Appl. Supercond. 2013, 23, 17016101701610,  DOI: 10.1109/TASC.2013.2244634
  6. 6
    Vaskivskyi, I. Fast electronic resistance switching involving hidden charge density wave states. Nat. Commun. 2016, 7, 11442,  DOI: 10.1038/ncomms11442
  7. 7
    Mihailovic, D. Ultrafast non-thermal and thermal switching in charge configuration memory devices based on 1T-TaS2. Appl. Phys. Lett. 2021, 119, 013106,  DOI: 10.1063/5.0052311
  8. 8
    Stojchevska, L. Ultrafast Switching to a Stable Hidden Quantum State in an Electronic Crystal. Science 2014, 344, 177180,  DOI: 10.1126/science.1241591
  9. 9
    Gerasimenko, Y. A.; Karpov, P.; Vaskivskyi, I.; Brazovskii, S.; Mihailovic, D. Intertwined chiral charge orders and topological stabilization of the light-induced state of a prototypical transition metal dichalcogenide. npj Quantum Mater. 2019, 4, 32,  DOI: 10.1038/s41535-019-0172-1
  10. 10
    Ritschel, T. Orbital textures and charge density waves in transition metal dichalcogenides. Nat. Phys. 2015, 11, 328331,  DOI: 10.1038/nphys3267
  11. 11
    Stahl, Q. Collapse of layer dimerization in the photo-induced hidden state of 1T-TaS2. Nat. Commun. 2020, 11, 1247,  DOI: 10.1038/s41467-020-15079-1
  12. 12
    Zhu, X.; Li, A. J.; Stewart, G. R.; Hebard, A. F. Detection of charge density wave phase transitions at 1T-TaS2/GaAs interfaces. Appl. Phys. Lett. 2017, 110, 181603,  DOI: 10.1063/1.4982964
  13. 13
    Tsen, A. W. Structure and control of charge density waves in two-dimensional 1T-TaS2. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 1505415059,  DOI: 10.1073/pnas.1512092112
  14. 14
    Perfetti, L. Femtosecond dynamics of electronic states in the Mott insulator 1T-TaS2 by time resolved photoelectron spectroscopy. New J. Phys. 2008, 10, 053019,  DOI: 10.1088/1367-2630/10/5/053019
  15. 15
    Michaelson, H. B. The work function of the elements and its periodicity. J. Appl. Phys. 1977, 48, 47294733,  DOI: 10.1063/1.323539
  16. 16
    Cho, D.; Cho, Y.-H.; Cheong, S.-W.; Kim, K.-S.; Yeom, H. W. Interplay of electron-electron and electron-phonon interactions in the low-temperature phase of 1 T – TaS 2. Phys. Rev. B 2015, 92, 085132,  DOI: 10.1103/PhysRevB.92.085132
  17. 17
    Vaskivskyi, I. Controlling the metal-to-insulator relaxation of the metastable hidden quantum state in 1T-TaS 2. Sci. Adv. 2015, 1, e1500168  DOI: 10.1126/sciadv.1500168
  18. 18
    Tanda, S.; Sambongi, T.; Tani, T.; Tanaka, S. X-Ray Study of Charge Density Wave Structure in 1T-TaS2. J. Phys. Soc. Jpn. 1984, 53, 476479,  DOI: 10.1143/JPSJ.53.476
  19. 19
    Wang, Y. D. Band insulator to Mott insulator transition in 1T-TaS2. Nat. Commun. 2020, 11, 4215,  DOI: 10.1038/s41467-020-18040-4
  20. 20
    Cho, D. Nanoscale manipulation of the Mott insulating state coupled to charge order in 1T-TaS2. Nat. Commun. 2016, 7, 10453,  DOI: 10.1038/ncomms10453
  21. 21
    Venturini, R. Ultra-Efficient Resistance Switching between Charge Ordered Phases in 1T-TaS2 with a Single Picosecond Electrical Pulse. arXiv 2022, 2202.13831
  22. 22
    Anders, S. European roadmap on superconductive electronics – status and perspectives. Phys. C Supercond. 2010, 470, 20792126,  DOI: 10.1016/j.physc.2010.07.005
  23. 23
    Ravnik, J.; Vaskivskyi, I.; Mertelj, T.; Mihailovic, D. Real-time observation of the coherent transition to a metastable emergent state in 1T–TaS2. Phys. Rev. B 2018, 97, 075304,  DOI: 10.1103/PhysRevB.97.075304
  24. 24
    Landauer, R. Irreversibility and Heat Generation in the Computing Process. IBM journal 1961, 5, 183,  DOI: 10.1147/rd.53.0183
  25. 25
    Grezes, C. Ultra-low switching energy and scaling in electric-field-controlled nanoscale magnetic tunnel junctions with high resistance-area product. Appl. Phys. Lett. 2016, 108, 012403,  DOI: 10.1063/1.4939446
  26. 26
    Cheng, C. H.; Chin, A.; Yeh, F. S. Ultralow Switching Energy Ni/GeOx/HfON/TaN RRAM. IEEE Electron Device Lett. 2011, 32, 366368,  DOI: 10.1109/LED.2010.2095820
  27. 27
    Wang, L. Voltage-Controlled Magnetic Tunnel Junctions for Processing-In-Memory Implementation. IEEE Electron Device Lett. 2018, 39, 440443,  DOI: 10.1109/LED.2018.2791510
  28. 28
    Liu, B. Multi-level phase-change memory with ultralow power consumption and resistance drift. Sci. Bull. 2021, 66, 22172224,  DOI: 10.1016/j.scib.2021.07.018
  29. 29
    Loke, D. Breaking the Speed Limits of Phase-Change Memory. Science 2012, 336, 15661569,  DOI: 10.1126/science.1221561
  30. 30
    Ding, K. Phase-change heterostructure enables ultralow noise and drift for memory operation. Science 2019, 366, 210215,  DOI: 10.1126/science.aay0291
  31. 31
    Pi, S. Memristor crossbar arrays with 6-nm half-pitch and 2-nm critical dimension. Nat. Nanotechnol. 2019, 14, 3539,  DOI: 10.1038/s41565-018-0302-0
  32. 32
    Xu, W.; Min, S.-Y.; Hwang, H.; Lee, T.-W. Organic core-sheath nanowire artificial synapses with femtojoule energy consumption. Sci. Adv. 2016, 2, e1501326  DOI: 10.1126/sciadv.1501326
  33. 33
    Havel, V.; Ultrafast switching in Ta2O5-based resistive memories. In 2016 IEEE Silicon Nanoelectronics Workshop (SNW) 82–83; IEEE, 2016;  DOI: 10.1109/SNW.2016.7577995 .
  34. 34
    Torrezan, A. C.; Strachan, J. P.; Medeiros-Ribeiro, G.; Williams, R. S. Sub-nanosecond switching of a tantalum oxide memristor. Nanotechnology 2011, 22, 485203,  DOI: 10.1088/0957-4484/22/48/485203
  35. 35
    Lee, S.; Sohn, J.; Jiang, Z.; Chen, H.-Y.; Philip Wong, H.-S. Metal oxide-resistive memory using graphene-edge electrodes. Nat. Commun. 2015, 6, 8407,  DOI: 10.1038/ncomms9407
  36. 36
    Ando, K. Spin-transfer torque magnetoresistive random-access memory technologies for normally off computing (invited). J. Appl. Phys. 2014, 115, 172607,  DOI: 10.1063/1.4869828
  37. 37
    Jan, G.; Demonstration of Ultra-Low Voltage and Ultra Low Power STT-MRAM designed for compatibility with 0x node embedded LLC applications. In 2018 IEEE Symposium on VLSI Technology 65–66; IEEE, 2018.
  38. 38
    Rehm, L. Sub-nanosecond switching in a cryogenic spin-torque spin-valve memory element with a dilute permalloy free layer. Appl. Phys. Lett. 2019, 114, 212402,  DOI: 10.1063/1.5094924
  39. 39
    Khalili Amiri, P. Electric-Field-Controlled Magnetoelectric RAM: Progress, Challenges, and Scaling. IEEE Trans. Magn. 2015, 51, 17,  DOI: 10.1109/TMAG.2015.2443124
  40. 40
    Zhao, Q.-Y. A compact superconducting nanowire memory element operated by nanowire cryotrons. Supercond. Sci. Technol. 2018, 31, 035009,  DOI: 10.1088/1361-6668/aaa820
  41. 41
    McCaughan, A. N.; Toomey, E.; Schneider, M.; Berggren, K. K.; Nam, S. W. A kinetic-inductance-based superconducting memory element with shunting and sub-nanosecond write times. Supercond. Sci. Technol. 2019, 32, 015005,  DOI: 10.1088/1361-6668/aae50d
  42. 42
    Konno, G.; Yamanashi, Y.; Yoshikawa, N. Fully Functional Operation of Low-Power 64-kb Josephson-CMOS Hybrid Memories. IEEE Trans. Appl. Supercond. 2017, 27, 17,  DOI: 10.1109/TASC.2016.2646911
  43. 43
    Krivorotov, I. N.; Ultrafast spin torque memory based on magnetic tunnel junctions with combined in-plane and perpendicular polarizers. In 70th Device Research Conference 211–212; IEEE, 2012.
  44. 44
    Pickett, M. D.; Stanley Williams, R. Sub-100 fJ and sub-nanosecond thermally driven threshold switching in niobium oxide crosspoint nanodevices. Nanotechnology 2012, 23, 215202,  DOI: 10.1088/0957-4484/23/21/215202
  45. 45
    Clerc, S.; A 0.32V, 55fJ per bit access energy, CMOS 65nm bit-interleaved SRAM with radiation Soft Error tolerance. In 2012 IEEE International Conference on IC Design & Technology 1–4; IEEE, 2012;  DOI: 10.1109/ICICDT.2012.6232860 .
  46. 46
    Wang, K. L.; Alzate, J. G.; Khalili Amiri, P. Low-power non-volatile spintronic memory: STT-RAM and beyond. J. Phys. Appl. Phys. 2013, 46, 074003,  DOI: 10.1088/0022-3727/46/7/074003
  47. 47
    Dieny, B. Opportunities and challenges for spintronics in the microelectronics industry. Nat. Electron. 2020, 3, 446459,  DOI: 10.1038/s41928-020-0461-5
  48. 48
    Stupakiewicz, A.; Szerenos, K.; Afanasiev, D.; Kirilyuk, A.; Kimel, A. V. Ultrafast nonthermal photo-magnetic recording in a transparent medium. Nature 2017, 542, 7174,  DOI: 10.1038/nature20807
  49. 49
    Dayton, I. M. Experimental Demonstration of a Josephson Magnetic Memory Cell With a Programmable π-Junction. IEEE Magn. Lett. 2018, 9, 15,  DOI: 10.1109/LMAG.2018.2801820
  50. 50
    Takeshita, Y. High-Speed Memory Driven by SFQ Pulses Based on 0-π SQUID. IEEE Trans. Appl. Supercond. 2021, 31, 16,  DOI: 10.1109/TASC.2021.3060351
  51. 51
    Tanaka, M. Josephson-CMOS Hybrid Memory With Nanocryotrons. IEEE Trans. Appl. Supercond. 2017, 27, 14,  DOI: 10.1109/TASC.2016.2646929
  52. 52
    Van Duzer, T. 64-kb Hybrid Josephson-CMOS 4 K RAM With 400 ps Access Time and 12 mW Read Power. IEEE Trans. Appl. Supercond. 2013, 23, 17005041700504,  DOI: 10.1109/TASC.2012.2230294
  53. 53
    Tolpygo, S. K. Superconductor digital electronics: Scalability and energy efficiency issues (Review Article). Low Temp. Phys. 2016, 42, 361379,  DOI: 10.1063/1.4948618
  54. 54
    Enomoto, H.; Kawano, T.; Kawaguchi, M.; Takano, Y.; Sekizawa, K. Van der Waals Growth of Thin TaS 2 on Layered Substrates by Chemical Vapor Transport Technique. Jpn. J. Appl. Phys. 2004, 43, L123L126,  DOI: 10.1143/JJAP.43.L123
  55. 55
    Sanders, C. E. Crystalline and electronic structure of single-layer TaS 2. Phys. Rev. B 2016, 94, 081404,  DOI: 10.1103/PhysRevB.94.081404
  56. 56
    Wang, X. Chemical Growth of 1 T -TaS 2 Monolayer and Thin Films: Robust Charge Density Wave Transitions and High Bolometric Responsivity. Adv. Mater. 2018, 30, 1800074,  DOI: 10.1002/adma.201800074
  57. 57
    Zhao, Q.-Y.; McCaughan, A. N.; Dane, A. E.; Berggren, K. K.; Ortlepp, T. A nanocryotron comparator can connect single-flux-quantum circuits to conventional electronics. Supercond. Sci. Technol. 2017, 30, 044002,  DOI: 10.1088/1361-6668/aa5f33
  58. 58
    Mraz, A.; Kabanov, V. V.; Mihailovic, D. Nanocryotron-driven Charge Configuration Memory. arXiv 2022, 2203.14586
  59. 59
    Mukhanov, O. A.; Kirichenko, A. F.; Filippov, T. V.; Sarwana, S. Hybrid Semiconductor-Superconductor Fast-Readout Memory for Digital RF Receivers. IEEE Trans. Appl. Supercond. 2011, 21, 797800,  DOI: 10.1109/TASC.2010.2089409
  60. 60
    Joshi, J. Charge density wave activated excitons in TiSe2–MoSe2 heterostructures. APL Mater. 2022, 10, 011103,  DOI: 10.1063/5.0067098

Cited By

This article is cited by 2 publications.

  1. Nabamita Chakraborty, Joy Ghosh, Saswata Goswami, Avinash Shaw, Biswajit Das, Abhijit Mallik, Kalyan K. Chattopadhyay. Perforated Turbostratic Graphene As Active Layer in a Nonvolatile Resistive Switching Memory Device. ACS Applied Electronic Materials 2023, 5 (4) , 2131-2144. https://doi.org/10.1021/acsaelm.3c00037
  2. Tiantian Huang, Rui Zhang, Lepeng Zhang, Peiran Xu, Yunkai Shao, Wanli Yang, Zhimin Chen, Xin Chen, Ning Dai. Energy-adaptive resistive switching with controllable thresholds in insulator–metal transition. RSC Advances 2022, 12 (55) , 35579-35586. https://doi.org/10.1039/D2RA06866D
  • Abstract

    Figure 1

    Figure 1. CCM device, EDS, and work function analysis. (a) SEM image of a typical CCM device used for measurements with transmission line contacts. (b) BF-STEM image of a cross-section of a fabricated CCM device (one memory bit). (c) Zoom-in to the interface between the metal electrode and 1T-TaS2 crystal. (d) Zoom-in to a part of the cross-section with the EDS line scan marked. (e) EDS analysis of the zoomed-in section, where individual layers are identified: Au (yellow), Pd (green), Ta (dark blue), S (light blue), Si (red), O (black), Ti (pink). (f) Top panel shows an AFM image of a part of the CCM device with the line scan for the KPFM measurement marked. Bottom panel shows KPFM measurement of the work functions for 1T-TaS2, SiO2, and Au electrodes along the line scan. (g) Schematic band diagram of the device based on the KPFM measurements. On the left is a band diagram of an interface between the Au electrode and the nearly commensurate (NC) metallic state at room temperature, and on the right is an interface between the Au electrode and the commensurate (C) charge density wave state at cryogenic temperature.

    Figure 2

    Figure 2. Resistance switching and voltage–current characteristics of CCM devices. (a) Temperature dependence of the four-contact resistance R. Switching from RHI to RLO at 350 mK is caused by an electrical W pulse as indicated by the arrow. Heating above 90 K (red line) reverts the system to the RHI state. Inset to (a) shows the expanded scale of the RT curve. (b, c) Pulsed measurements of the VI curve for the W and E operations, respectively.

    Figure 3

    Figure 3. Speed and energy efficiency scaling at 20 K. (a) Switching energy density εW as a function of pulse length τW. The inset shows the actual pulse shapes. Red line shows linear scaling, and blue line shows departure from linearity at short τW. The data point at 1.9 ps was taken from ref (21). (b) Switching threshold voltage VW as a function of distance between the electrodes L. The inset shows a device with variable L used in the measurement. Different symbol colors are for different physical devices. (c) Endurance measurement showing cycling between RLO and RHI for 106 cycles. Each pair of points represents 2 × 104 W/E cycles.

    Figure 4

    Figure 4. Measured switching energy EW and speed of leading memory devices: (a) switching energy in correlation with device area; (b) switching times τW plotted against switching energy. References: PCM, (28,29) RRAM, (26,33−35) STT-RAM, (36−38) MRAM, (25,39) nMem, (40,41) JJ-CMOS memory, (42) OST-RAM, (43) Mott memory, (44) SRAM, (45−47) and DRAM. (47)

  • References

    ARTICLE SECTIONS
    Jump To

    This article references 60 other publications.

    1. 1
      Holmes, D. S.; Kadin, A. M.; Johnson, M. W. Superconducting Computing in Large-Scale Hybrid Systems. Computer 2015, 48, 3442,  DOI: 10.1109/MC.2015.375
    2. 2
      Soloviev, I. I. Beyond Moore’s technologies: operation principles of a superconductor alternative. Beilstein J. Nanotechnol. 2017, 8, 26892710,  DOI: 10.3762/bjnano.8.269
    3. 3
      Janod, E. Resistive Switching in Mott Insulators and Correlated Systems. Adv. Funct. Mater. 2015, 25, 62876305,  DOI: 10.1002/adfm.201500823
    4. 4
      Tafuri, F., Ed. Fundamentals and Frontiers of the Josephson Effect; Springer Series in Materials Science, Vol. 286; Springer Nature, 2019; pp 632648.
    5. 5
      Holmes, D. S.; Ripple, A. L.; Manheimer, M. A. Energy-Efficient Superconducting Computing─Power Budgets and Requirements. IEEE Trans. Appl. Supercond. 2013, 23, 17016101701610,  DOI: 10.1109/TASC.2013.2244634
    6. 6
      Vaskivskyi, I. Fast electronic resistance switching involving hidden charge density wave states. Nat. Commun. 2016, 7, 11442,  DOI: 10.1038/ncomms11442
    7. 7
      Mihailovic, D. Ultrafast non-thermal and thermal switching in charge configuration memory devices based on 1T-TaS2. Appl. Phys. Lett. 2021, 119, 013106,  DOI: 10.1063/5.0052311
    8. 8
      Stojchevska, L. Ultrafast Switching to a Stable Hidden Quantum State in an Electronic Crystal. Science 2014, 344, 177180,  DOI: 10.1126/science.1241591
    9. 9
      Gerasimenko, Y. A.; Karpov, P.; Vaskivskyi, I.; Brazovskii, S.; Mihailovic, D. Intertwined chiral charge orders and topological stabilization of the light-induced state of a prototypical transition metal dichalcogenide. npj Quantum Mater. 2019, 4, 32,  DOI: 10.1038/s41535-019-0172-1
    10. 10
      Ritschel, T. Orbital textures and charge density waves in transition metal dichalcogenides. Nat. Phys. 2015, 11, 328331,  DOI: 10.1038/nphys3267
    11. 11
      Stahl, Q. Collapse of layer dimerization in the photo-induced hidden state of 1T-TaS2. Nat. Commun. 2020, 11, 1247,  DOI: 10.1038/s41467-020-15079-1
    12. 12
      Zhu, X.; Li, A. J.; Stewart, G. R.; Hebard, A. F. Detection of charge density wave phase transitions at 1T-TaS2/GaAs interfaces. Appl. Phys. Lett. 2017, 110, 181603,  DOI: 10.1063/1.4982964
    13. 13
      Tsen, A. W. Structure and control of charge density waves in two-dimensional 1T-TaS2. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 1505415059,  DOI: 10.1073/pnas.1512092112
    14. 14
      Perfetti, L. Femtosecond dynamics of electronic states in the Mott insulator 1T-TaS2 by time resolved photoelectron spectroscopy. New J. Phys. 2008, 10, 053019,  DOI: 10.1088/1367-2630/10/5/053019
    15. 15
      Michaelson, H. B. The work function of the elements and its periodicity. J. Appl. Phys. 1977, 48, 47294733,  DOI: 10.1063/1.323539
    16. 16
      Cho, D.; Cho, Y.-H.; Cheong, S.-W.; Kim, K.-S.; Yeom, H. W. Interplay of electron-electron and electron-phonon interactions in the low-temperature phase of 1 T – TaS 2. Phys. Rev. B 2015, 92, 085132,  DOI: 10.1103/PhysRevB.92.085132
    17. 17
      Vaskivskyi, I. Controlling the metal-to-insulator relaxation of the metastable hidden quantum state in 1T-TaS 2. Sci. Adv. 2015, 1, e1500168  DOI: 10.1126/sciadv.1500168
    18. 18
      Tanda, S.; Sambongi, T.; Tani, T.; Tanaka, S. X-Ray Study of Charge Density Wave Structure in 1T-TaS2. J. Phys. Soc. Jpn. 1984, 53, 476479,  DOI: 10.1143/JPSJ.53.476
    19. 19
      Wang, Y. D. Band insulator to Mott insulator transition in 1T-TaS2. Nat. Commun. 2020, 11, 4215,  DOI: 10.1038/s41467-020-18040-4
    20. 20
      Cho, D. Nanoscale manipulation of the Mott insulating state coupled to charge order in 1T-TaS2. Nat. Commun. 2016, 7, 10453,  DOI: 10.1038/ncomms10453
    21. 21
      Venturini, R. Ultra-Efficient Resistance Switching between Charge Ordered Phases in 1T-TaS2 with a Single Picosecond Electrical Pulse. arXiv 2022, 2202.13831
    22. 22
      Anders, S. European roadmap on superconductive electronics – status and perspectives. Phys. C Supercond. 2010, 470, 20792126,  DOI: 10.1016/j.physc.2010.07.005
    23. 23
      Ravnik, J.; Vaskivskyi, I.; Mertelj, T.; Mihailovic, D. Real-time observation of the coherent transition to a metastable emergent state in 1T–TaS2. Phys. Rev. B 2018, 97, 075304,  DOI: 10.1103/PhysRevB.97.075304
    24. 24
      Landauer, R. Irreversibility and Heat Generation in the Computing Process. IBM journal 1961, 5, 183,  DOI: 10.1147/rd.53.0183
    25. 25
      Grezes, C. Ultra-low switching energy and scaling in electric-field-controlled nanoscale magnetic tunnel junctions with high resistance-area product. Appl. Phys. Lett. 2016, 108, 012403,  DOI: 10.1063/1.4939446
    26. 26
      Cheng, C. H.; Chin, A.; Yeh, F. S. Ultralow Switching Energy Ni/GeOx/HfON/TaN RRAM. IEEE Electron Device Lett. 2011, 32, 366368,  DOI: 10.1109/LED.2010.2095820
    27. 27
      Wang, L. Voltage-Controlled Magnetic Tunnel Junctions for Processing-In-Memory Implementation. IEEE Electron Device Lett. 2018, 39, 440443,  DOI: 10.1109/LED.2018.2791510
    28. 28
      Liu, B. Multi-level phase-change memory with ultralow power consumption and resistance drift. Sci. Bull. 2021, 66, 22172224,  DOI: 10.1016/j.scib.2021.07.018
    29. 29
      Loke, D. Breaking the Speed Limits of Phase-Change Memory. Science 2012, 336, 15661569,  DOI: 10.1126/science.1221561
    30. 30
      Ding, K. Phase-change heterostructure enables ultralow noise and drift for memory operation. Science 2019, 366, 210215,  DOI: 10.1126/science.aay0291
    31. 31
      Pi, S. Memristor crossbar arrays with 6-nm half-pitch and 2-nm critical dimension. Nat. Nanotechnol. 2019, 14, 3539,  DOI: 10.1038/s41565-018-0302-0
    32. 32
      Xu, W.; Min, S.-Y.; Hwang, H.; Lee, T.-W. Organic core-sheath nanowire artificial synapses with femtojoule energy consumption. Sci. Adv. 2016, 2, e1501326  DOI: 10.1126/sciadv.1501326
    33. 33
      Havel, V.; Ultrafast switching in Ta2O5-based resistive memories. In 2016 IEEE Silicon Nanoelectronics Workshop (SNW) 82–83; IEEE, 2016;  DOI: 10.1109/SNW.2016.7577995 .
    34. 34
      Torrezan, A. C.; Strachan, J. P.; Medeiros-Ribeiro, G.; Williams, R. S. Sub-nanosecond switching of a tantalum oxide memristor. Nanotechnology 2011, 22, 485203,  DOI: 10.1088/0957-4484/22/48/485203
    35. 35
      Lee, S.; Sohn, J.; Jiang, Z.; Chen, H.-Y.; Philip Wong, H.-S. Metal oxide-resistive memory using graphene-edge electrodes. Nat. Commun. 2015, 6, 8407,  DOI: 10.1038/ncomms9407
    36. 36
      Ando, K. Spin-transfer torque magnetoresistive random-access memory technologies for normally off computing (invited). J. Appl. Phys. 2014, 115, 172607,  DOI: 10.1063/1.4869828
    37. 37
      Jan, G.; Demonstration of Ultra-Low Voltage and Ultra Low Power STT-MRAM designed for compatibility with 0x node embedded LLC applications. In 2018 IEEE Symposium on VLSI Technology 65–66; IEEE, 2018.
    38. 38
      Rehm, L. Sub-nanosecond switching in a cryogenic spin-torque spin-valve memory element with a dilute permalloy free layer. Appl. Phys. Lett. 2019, 114, 212402,  DOI: 10.1063/1.5094924
    39. 39
      Khalili Amiri, P. Electric-Field-Controlled Magnetoelectric RAM: Progress, Challenges, and Scaling. IEEE Trans. Magn. 2015, 51, 17,  DOI: 10.1109/TMAG.2015.2443124
    40. 40
      Zhao, Q.-Y. A compact superconducting nanowire memory element operated by nanowire cryotrons. Supercond. Sci. Technol. 2018, 31, 035009,  DOI: 10.1088/1361-6668/aaa820
    41. 41
      McCaughan, A. N.; Toomey, E.; Schneider, M.; Berggren, K. K.; Nam, S. W. A kinetic-inductance-based superconducting memory element with shunting and sub-nanosecond write times. Supercond. Sci. Technol. 2019, 32, 015005,  DOI: 10.1088/1361-6668/aae50d
    42. 42
      Konno, G.; Yamanashi, Y.; Yoshikawa, N. Fully Functional Operation of Low-Power 64-kb Josephson-CMOS Hybrid Memories. IEEE Trans. Appl. Supercond. 2017, 27, 17,  DOI: 10.1109/TASC.2016.2646911
    43. 43
      Krivorotov, I. N.; Ultrafast spin torque memory based on magnetic tunnel junctions with combined in-plane and perpendicular polarizers. In 70th Device Research Conference 211–212; IEEE, 2012.
    44. 44
      Pickett, M. D.; Stanley Williams, R. Sub-100 fJ and sub-nanosecond thermally driven threshold switching in niobium oxide crosspoint nanodevices. Nanotechnology 2012, 23, 215202,  DOI: 10.1088/0957-4484/23/21/215202
    45. 45
      Clerc, S.; A 0.32V, 55fJ per bit access energy, CMOS 65nm bit-interleaved SRAM with radiation Soft Error tolerance. In 2012 IEEE International Conference on IC Design & Technology 1–4; IEEE, 2012;  DOI: 10.1109/ICICDT.2012.6232860 .
    46. 46
      Wang, K. L.; Alzate, J. G.; Khalili Amiri, P. Low-power non-volatile spintronic memory: STT-RAM and beyond. J. Phys. Appl. Phys. 2013, 46, 074003,  DOI: 10.1088/0022-3727/46/7/074003
    47. 47
      Dieny, B. Opportunities and challenges for spintronics in the microelectronics industry. Nat. Electron. 2020, 3, 446459,  DOI: 10.1038/s41928-020-0461-5
    48. 48
      Stupakiewicz, A.; Szerenos, K.; Afanasiev, D.; Kirilyuk, A.; Kimel, A. V. Ultrafast nonthermal photo-magnetic recording in a transparent medium. Nature 2017, 542, 7174,  DOI: 10.1038/nature20807
    49. 49
      Dayton, I. M. Experimental Demonstration of a Josephson Magnetic Memory Cell With a Programmable π-Junction. IEEE Magn. Lett. 2018, 9, 15,  DOI: 10.1109/LMAG.2018.2801820
    50. 50
      Takeshita, Y. High-Speed Memory Driven by SFQ Pulses Based on 0-π SQUID. IEEE Trans. Appl. Supercond. 2021, 31, 16,  DOI: 10.1109/TASC.2021.3060351
    51. 51
      Tanaka, M. Josephson-CMOS Hybrid Memory With Nanocryotrons. IEEE Trans. Appl. Supercond. 2017, 27, 14,  DOI: 10.1109/TASC.2016.2646929
    52. 52
      Van Duzer, T. 64-kb Hybrid Josephson-CMOS 4 K RAM With 400 ps Access Time and 12 mW Read Power. IEEE Trans. Appl. Supercond. 2013, 23, 17005041700504,  DOI: 10.1109/TASC.2012.2230294
    53. 53
      Tolpygo, S. K. Superconductor digital electronics: Scalability and energy efficiency issues (Review Article). Low Temp. Phys. 2016, 42, 361379,  DOI: 10.1063/1.4948618
    54. 54
      Enomoto, H.; Kawano, T.; Kawaguchi, M.; Takano, Y.; Sekizawa, K. Van der Waals Growth of Thin TaS 2 on Layered Substrates by Chemical Vapor Transport Technique. Jpn. J. Appl. Phys. 2004, 43, L123L126,  DOI: 10.1143/JJAP.43.L123
    55. 55
      Sanders, C. E. Crystalline and electronic structure of single-layer TaS 2. Phys. Rev. B 2016, 94, 081404,  DOI: 10.1103/PhysRevB.94.081404
    56. 56
      Wang, X. Chemical Growth of 1 T -TaS 2 Monolayer and Thin Films: Robust Charge Density Wave Transitions and High Bolometric Responsivity. Adv. Mater. 2018, 30, 1800074,  DOI: 10.1002/adma.201800074
    57. 57
      Zhao, Q.-Y.; McCaughan, A. N.; Dane, A. E.; Berggren, K. K.; Ortlepp, T. A nanocryotron comparator can connect single-flux-quantum circuits to conventional electronics. Supercond. Sci. Technol. 2017, 30, 044002,  DOI: 10.1088/1361-6668/aa5f33
    58. 58
      Mraz, A.; Kabanov, V. V.; Mihailovic, D. Nanocryotron-driven Charge Configuration Memory. arXiv 2022, 2203.14586
    59. 59
      Mukhanov, O. A.; Kirichenko, A. F.; Filippov, T. V.; Sarwana, S. Hybrid Semiconductor-Superconductor Fast-Readout Memory for Digital RF Receivers. IEEE Trans. Appl. Supercond. 2011, 21, 797800,  DOI: 10.1109/TASC.2010.2089409
    60. 60
      Joshi, J. Charge density wave activated excitons in TiSe2–MoSe2 heterostructures. APL Mater. 2022, 10, 011103,  DOI: 10.1063/5.0067098
  • Supporting Information

    Supporting Information

    ARTICLE SECTIONS
    Jump To

    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.nanolett.2c01116.

    • Description of materials and fabrication methods; description of DC and fast electrical measurement (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

You’ve supercharged your research process with ACS and Mendeley!

STEP 1:
Click to create an ACS ID

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

MENDELEY PAIRING EXPIRED
Your Mendeley pairing has expired. Please reconnect