We will review salient features of two advanced nodes of an 8-Nb-layer fully planarized process developed recently at MIT Lincoln Laboratory for fabricating Single Flux Quantum (SFQ) digital circuits with very large scale integration (VLSI) on 200-mm wafers as well as SFQ benchmark circuits demonstrated to date. The emphasis will be on SFQ4ee and SFQ5ee nodes, where “ee” denotes the process is tuned for energy efficient SFQ circuits. The former has eight superconducting layers, a 2 Ω/sq resistor layer, and contact metallization. The latter has nine superconducting layers: eight Nb wiring layers with the minimum feature size of 350 nm and one high kinetic inductance (about 8 pH/sq) layer for forming compact bias inductors. It also features three resistive layers: a layer with high sheet resistance (~ 6.5 Ω/sq) for shunting and biasing of Josephson junctions (JJ), a layer of mΩ-range resistors for releasing unwanted flux quanta from superconducting loops of logic cells, and contact metallization for chip packaging. The technology utilizes one layer of Nb/Al-AlO_x/Nb Josephson junctions with critical current density of 100 μA/μm² and minimum diameter of 700 nm. Circuit patterns are defined by 248-nm photolithography and high density plasma etching. All circuit layers are fully planarized using chemical mechanical planarization (CMP) of SiO₂ interlayer dielectric. Results on characterization of JJs, inductors and other circuit components as well as benchmark circuits with JJ density up to 1.4∙10⁶ JJ/cm² will be presented. Our technology roadmap for SFQ circuits with 180 nm minimum linewidth and over 1M JJ count will also be discussed.

This work was sponsored by the IARPA under Air Force Contract FA872105C0002. Opinions, interpretations, conclusions, and recommendations are those of the authors and not necessarily endorsed by the United States Government.

The development of Josephson voltage standards for AC applications has made significant progress in recent years. Two different approaches have been in the focus of these developments. The first approach uses series arrays of overdamped Josephson junctions, in which the number of junctions per segment follows a binary sequence. The Josephson junctions are driven by a sinusoidal microwave. Each segment is operated by a fast bias source. Series arrays for output voltages of 10 V consist of about 70,000 Josephson junctions. The applications of these binary-divided series arrays are typically limited to a few kHz because of the operation principle as a multi-bit digital-to-analogue converter; they are often operated using sampling methods. The second approach is based on series arrays of overdamped Josephson junctions biased with a high-speed digital sequence of short current pulses. These pulse-driven arrays enable the synthesis of spectrally pure waveforms of frequencies up to the MHz range. The major breakthrough of rms voltages up to 1 V was recently reached by operating up to 8 arrays in series containing 63,000 Josephson junctions in total.

The fabrication of these large series arrays of overdamped Josephson junctions requires an adapted technology. Josephson junctions based on highly resistive, amorphous metal-silicide NbSi barriers are now used at PTB for both kinds of AC voltage standards. These junctions are robust and enable the fabrication of large series arrays with high reproducibility and yield. The characteristic voltage of the junctions can be tuned over a wide range for operating the arrays at different drive frequencies from 15 GHz for pulse-driven arrays to 70 GHz for binary-divided arrays. The NbSi layers are deposited by co-sputtering from two targets (Nb and Si). This paper discusses the development of binary-divided and pulse-driven arrays at PTB.

This work was partly supported by the EU within the EMRP JRP SIB59 Q-WAVE. The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union.

Superconducting-ferromagnetic hybrid devices have potential as a practical memory technology compatible with superconducting logic circuits and may help realize energy-efficient, high-performance superconducting computers. We have developed single Josephson junction devices with pseudo-spin-valve barriers. We observed changes in Josephson critical current depending on the magnetization state of the barrier (parallel or anti-parallel) through the superconductor-ferromagnet proximity effect [1]. This effect persists to nanoscale in contrast to the remanent (stray) field effect. In nanoscale devices, the magnetization states of the pseudo-spin-valve barriers could also be switched with applied bias currents which is consistent with the spin-transfer torque effect in room-temperature magnetic memory devices [2]. Our results demonstrate devices that combine scalable superconducting and spintronic effects promising for a nanoscale cryogenic memory technology.
[1] B. Baek, W. H. Rippard, S. P. Benz, S. E. Russek, and P. D. Dresselhaus, “Hybrid superconducting-magnetic memory device using competing order parameters.” Nat. Commun. 5, 3888, (2014).
[2] B. Baek, W. H. Rippard, M. R. Pufall, S. P. Benz, S. E. Russek, H. Rogalla, and P. D. Dresselhaus, “Spin-transfer torque switching in nanopillar superconducting-magnetic hybrid Josephson junctions.” Phys. Rev. Appl. 3, 011001 (2015).

This work was supported by NIST and by the US National Security Agency under agreement numbers EAO156513 and EAO176792.

Many challenges have been performed for applying superconductors to the practical devices and providing ultimate performances, and a cryogenic computer is one of most attractive applications of the superconductors among various examples of the device application. As a basic element of the supercomputing system, single-flux-quantum (SFQ) has been studied and expected as a next-generation computing device with high-speed processing and ultralow consumption power. Recently new types of the logic element have been demonstrated, e.g., low-voltage SFQ, energy-efficient SFQ, adiabatic quantum flux parametron, and reciprocal quantum logic. On the other hand, in order to realize large-scale cryogenic computing system, there are still two crucial challenges: (i) further reduction of the consumption power and (ii) realization of a high-density cryogenic memory. Regarding these problems, recently new trends to resolve them by introducing spintronics concept to superconducting devices appear. Basic unit of the superconducting spintronic devices is a magnetic Josephson junction (MJJ) consisting of a superconductor/ferromagnet/superconductor junction. In this work, we developed the MJJ based on niobium nitride (NbN) toward a realization of the key component of the ultra-efficient cryogenic computing system. The superconducting critical temperature of NbN is higher than that of Nb which is conventionally used for superconducting logic devices, thus the cooling cost for the cryogenic system is expected to be reduced using NbN-based MJJs, leading to ultralow consumption power of the total system [1]. We fabricated the NbN-based MJJs on a magnesium oxide (MgO) substrate. We evaluated the dependences of the Josephson current on the temperature and the thickness of the ferromagnetic barrier.
[1] K. Makise et al., IEEE Trans. Appl. Supercond. vol. 23, no. 3, pp. 1100804 (2013).

One of the crucial factors in achieving high performance of superconducting integrated circuits, such as analog-to-digital converters (ADCs), is sampling using a high frequency clock source with a low cycle-to-cycle jitter. As the superconductor ADC technology matures towards more complex designs for higher dynamic range performance, the need for synchronous clocking of multiple comparators continues to grow. Since high frequency external clock sources are expensive and make a significant contribution to the heat load of the system, a high-frequency and low jitter on-chip clock source using long Josephson junction (LJJ) is considered the preferred long term solution. Toward that end, we are working on improving an on chip 110 GHz clock source based on an unshunted long Josephson junction in annular geometry. Minimizing the additional jitter added by each fan-out of the clock signal is the effort’s goal. For synchronous clocking of up to 3 comparators, we are comparing clock distribution using superconducting passive transmission lines and a new approach using novel active transmission lines. We investigate with several lengths of the new transmission lines up to 980 µm and compare the performance to the signal distribution using PTLs.

This project was funded in part by Office of Naval Research. The authors thank Saad Sarwana and Anubhav Sahu for assistance in testing the fabricated chips.

Even though silicon is one of the most studied materials, superconductivity at ambient pressure in boron doped silicon has only been discovered in 2006 [1]. This is due to the extreme doping concentration required to trigger superconductivity in this system, more than three times the boron solubility limit in silicon. This concentration is impossible to reach using conventional micro-electronic techniques, but epitaxial superconducting Si:B thin films can be achieved using Gas Immersion Laser Doping.
We show that superconductivity is only observed for boron concentration values exceeding a threshold value, which is inversely proportional to the thin layer thickness. The critical temperature then rapidly increases and is fully determined by the boron dose [2] .
Superconducting silicon excites a great interest concerning its possible applications. The great asset of superconducting silicon is the possibility to take advantage from the silicon technology to elaborate built-in structures in a single Si crystal, avoiding the assembling step. Since Si:B doesn't transit directly from the semiconducting state to the superconducting one, but first becomes metallic, it allows the fabrication of a large range of nanodevices, in which superconductors, metals and semiconductors can be coupled through extremely clean, epitaxially grown interfaces.
We demonstrate the possibility of structuring the strongly boron-doped Si to realise superconducting micro and nano devices, and we show the first results on all silicon Superconductor/ Normal metal/ Superconductor Josephson junctions, SQUIDs, and microwave resonators.

References
[1] E. Bustarret et al., Nature 444, 465 (2006)
[2] A. Grockowiak et al., Phys. Rev. B 88, 064508 (2013)