HKU Researchers Uncover Piezoelectric Effect in Diamond Membranes, Upending a Century of Materials Science Consensus
University of Hong Kong team shows ultrathin polycrystalline diamond membranes generate voltage when bent, overturning the century-old classification of diamond as non-piezoelectric.
Overview
Diamond has long been prized for its hardness, thermal conductivity, and chemical inertness — but not for generating electricity. Since the early twentieth century, diamond has been categorized as a non-piezoelectric material, incapable of converting mechanical force into electric charge. A new study published in Science Advances by researchers at the University of Hong Kong challenges that classification. The team demonstrates that ultrathin polycrystalline diamond membranes do exhibit a measurable piezoelectric effect when bent, and that the response exceeds the performance of many conventional piezoelectric materials.
What We Know
The study, titled “Uncovering piezoelectric effect in polycrystalline diamond membranes,” was published in Science Advances (Volume 12, Issue 12) on March 18, 2026. It was led by Professor Zhiqin Chu, Associate Professor in the Department of Electrical and Computer Engineering, and Professor Yuan Lin, Professor in the Department of Mechanical Engineering, at HKU’s Faculty of Engineering. Equal contributors are Jixiang Jing, Bicong Wang, and Yumeng Luo, with Kwai Hei Li of the Southern University of Science and Technology’s School of Microelectronics among the corresponding authors. The work involved collaboration with Peking University and the Chinese Academy of Sciences, according to the paper’s author affiliations.
The core result: when ultrathin, flexible polycrystalline diamond membranes are bent, they generate stable voltage signals. As described in the Science Advances paper, the abstract states directly: “Diamonds have been regarded as nonpiezoelectric materials for more than one century. Here, we uncover a notable piezoelectric effect in ultrathin and ultraflexible polycrystalline diamond membranes.”
The peak performance came from membranes approximately 5 micrometers thick. At that thickness, the team recorded a piezoelectric voltage coefficient of approximately 82.2 millivolt meters per newton — a figure the authors say “surpasses many conventional piezoelectric materials,” according to the Science Advances paper. The maximum observed output voltage reached approximately 70 millivolts at a strain of 1.4 percent. The piezoelectric charge coefficient d₃₃ peaked at 4 picocoulombs per newton at 5-micrometer thickness, up from 2 picocoulombs per newton at 1 micrometer.
The effect also proved durable: the membranes maintained consistent output for more than 7,000 bending cycles at a 0.35 percent strain, the paper reports.
To fabricate the membranes, the team used an edge-exfoliation method that allows the exceptionally hard material to undergo large deformations, as Phys.org reported. First-principles calculations then identified the mechanism driving the effect: the piezoelectricity originates from asymmetry at grain boundaries within the polycrystalline structure. The paper states that “charge polarization accumulates at the grain boundaries as the imposed deformation increases, which in turn creates a potential difference between the upper and lower surfaces of the membrane,” as relayed by Phys.org. This grain-boundary asymmetry is why the effect arises in polycrystalline forms of diamond but not in single-crystal diamond — a distinction the research draws explicitly, per the PMC paper.
The team conducted extensive mechanical cycling tests under various controlled conditions to rule out measurement artifacts, including environmental noise and triboelectric interference, before confirming the voltage signals were genuine, according to Phys.org.
Potential Applications
Diamond’s combination of chemical inertness, biocompatibility, non-toxicity, and extreme hardness makes it unusual among piezoelectric candidates. The HKU press release identifies implantable medical devices as a priority application: the membranes could serve as self-generating power sources or deformation sensors inside the body, where conventional piezoelectric materials may corrode or cause adverse reactions. The release also points toward next-generation micro-energy systems and self-powered sensing technologies.
The Phys.org coverage of the study echoes that the findings open diamond to applications in implantable electronics and intelligent sensing — areas that require materials able to survive harsh biological environments over long periods.
What We Don’t Know
Several questions remain open. The study demonstrates piezoelectricity under laboratory bending conditions, but the path from ultrathin lab membranes to deployed medical or sensing devices involves additional engineering challenges — including integration with electronics, power extraction efficiency, and manufacturing scalability — that the paper does not address. The piezoelectric voltage coefficient of 82.2 mV·m/N surpasses many conventional materials in that specific metric, but a direct comparison to industry-standard materials such as lead zirconate titanate (PZT) across all performance axes is not provided by the published results.
The paper also does not disclose commercialization plans or industry partnerships. Whether the edge-exfoliation technique can be scaled to volumes required for consumer or medical applications, and at what cost, is not addressed in the current publication.
Analysis
The significance of the finding rests on the specificity of what has been overturned. The non-piezoelectric classification of diamond was not a casual assumption — it followed from diamond’s crystallographic symmetry in its pure single-crystal form, which forbids net polarization under stress. The HKU team’s contribution is to show that grain boundaries in polycrystalline membranes break that symmetry locally, producing a macroscopic effect strong enough to be useful. That distinction — single-crystal versus polycrystalline — is what makes the result credible while also scoping it appropriately: the century-old consensus about bulk single-crystal diamond stands, but it does not apply to the polycrystalline thin-film geometry the team studied.
For materials science, the result is a reminder that the properties of a substance at the nanoscale and in microstructured form can diverge substantially from its bulk properties. The work joins a broader wave of thin-film and two-dimensional materials research that has already upended assumptions about other well-studied materials.