[Paper Review] Bipolar electrical switching in metal-metal contacts

Gaurav Gandhi, Varun Aggarwal|arXiv (Cornell University)|Jun 4, 2013

Advanced Memory and Neural Computing9 references3 citations

TL;DR

This paper demonstrates that simple metal-metal point contacts or granular metallic arrangements—historically known as coherers—exhibit bipolar electrical switching and memristive behavior, functioning as the first canonical physical implementation of the memristor. The key finding is that resistance states are programmable via maximum current magnitude and polarity, enabling reversible, multi-state memory operation without specialized materials or nano-scale fabrication.

ABSTRACT

Electrical switching has been observed in carefully designed metal-insulator-metal devices built at small geometries. These devices are also commonly known as memristors and consist of specific materials such as transition metal oxides, chalcogenides, perovskites, oxides with valence defects, or a combination of an inert and an electrochemically active electrode. No simple physical device has been reported to exhibit electrical switching. We have discovered that a simple point-contact or a granular arrangement formed of metal pieces exhibits bipolar switching. These devices, referred to as coherers, were considered as one-way electrical fuses. We have identified the state variable governing the resistance state and can program the device to switch between multiple stable resistance states. Our observations render previously postulated thermal mechanisms for their resistance-change as inadequate. These devices constitute the missing canonical physical implementations for memristor, often referred as the fourth passive element. Apart from the theoretical advance in understanding metallic contacts, the current discovery provides a simple memristor to physicists and engineers for widespread experimentation, hitherto impossible.

Motivation & Objective

• To identify a physically realizable, simple device that exhibits memristive behavior as defined by Leon Chua.
• To resolve the long-standing absence of a canonical physical implementation of the memristor in electronic circuits.
• To demonstrate that bipolar resistance switching—essential for non-volatile memory—can occur in basic metallic contacts without specialized materials.
• To challenge the long-held thermal mechanism explanation for coherer behavior, showing it insufficient to explain observed multi-state, reversible switching.
• To establish that the state variable governing resistance is the maximum current passed through the device, not temperature or oxide formation.

Proposed method

• Conducted electrical measurements on point-contact and granular metallic coherers (e.g., iron filings, ball bearings) under controlled voltage and current excitation.
• Applied bipolar voltage pulses with varying amplitudes and polarities to induce resistance switching and observe hysteresis in I-V characteristics.
• Used a pinched hysteresis loop in the I-V plane as the defining fingerprint for memristive behavior, consistent with Chua’s theoretical framework.
• Mapped resistance states as a function of maximum current magnitude and direction, revealing distinct forward and reverse switching behavior.
• Performed comparative analysis with thermal models (e.g., Eccles’ thermistor equation) and found them inadequate to explain the observed non-thermal, reversible switching.
• Conducted preliminary experiments with polished gold balls to test whether oxide layers were necessary, suggesting the effect may arise from intrinsic contact physics.

Experimental results

Research questions

• RQ1 Can a simple metal-metal contact exhibit memristive behavior without specialized materials or nano-scale fabrication?
• RQ2 Is the resistance switching in coherers truly reversible and programmable via electrical signals, or is it unidirectional and fatigue-prone as previously believed?
• RQ3 Does the observed switching depend on current magnitude and polarity, indicating a state-dependent resistance governed by a non-thermal mechanism?
• RQ4 Can the coherer be considered the canonical physical implementation of the memristor, given its simplicity and adherence to Chua’s theoretical definition?
• RQ5 What is the physical origin of the resistance switching—thermal effects, oxide formation, or contact polarization—given that the phenomenon persists even in polished gold contacts?

Key findings

• The coherer and auto-coherer exhibit a pinched hysteresis loop in the I-V plane, confirming their memristive nature as per Chua’s definition.
• The device can be electrically programmed into multiple stable resistance states by applying current pulses of varying magnitude and polarity.
• Resistance switching is bipolar: the device can be switched from a high-resistance state to a low-resistance state and back to the original high-resistance state via appropriate current signals.
• The state variable is the maximum current passed through the device, not temperature or oxide thickness, and switching behavior differs for forward and reverse current directions.
• The observed behavior cannot be explained by thermal mechanisms such as Joule heating or oxide melting, as these would not account for the direction-dependent, reversible switching.
• The phenomenon persists even in polished gold contacts, suggesting that the effect is intrinsic to metal-metal point contacts and not dependent on surface oxides or defects.