Nanodiamond Charge Drift Reads Macrophage Inflammation: Chicago and Iowa in Advanced Materials, 4 February 2026
Quentir Medicine Monitor
Evidence-based insights for quantum medicine. Published by Quentir Systems LLC · August 28, 2026.
Nanodiamond quantum sensors have been read as thermometers inside living cells for more than a decade. One of them, a 70-nanometer crystal sitting inside a mouse macrophage, drifted downward by 0.27 megahertz over 200 seconds of measurement, which on the usual arithmetic means the cell warmed by 3.62 degrees Celsius. A team at the University of Chicago and the University of Iowa has now put a different quantity behind that same number.
Their paper, Probing cellular activity via charge-sensitive quantum nanoprobes, appeared in Advanced Materials on 4 February 2026, with Uri Zvi as first author and Denis R. Candido at Iowa, Aaron Esser-Kahn and Peter C. Maurer at Chicago as corresponding authors. It was received on 15 March 2025 and accepted on 15 January 2026. The University of Iowa publicized it on 26 August 2026, six months after the article went online, which is why it reaches this Monitor now.
Practical takeaway. The drift a nanodiamond shows inside a living cell can come from electrons moving between the diamond surface and the cell, not only from temperature. That is a caution for anyone quoting intracellular thermometry numbers, and it is also a new readout: the same charge signal separated resting macrophages from LPS-activated ones with a p-value of 0.00638.
What the zero-field splitting actually tracks, and the dipole term that was being dropped
The sensor is a nitrogen-vacancy center, an atomic defect in diamond whose spin can be initialized with green light, driven with microwaves and read out optically. Each 70-nanometer crystal used here carries roughly 100 of them. The measured quantity is the zero-field splitting, the frequency gap between the spin's ground sublevels, which sits near 2.87 gigahertz and moves by about 74 kilohertz for every kelvin. That temperature coefficient is what turned these particles into intracellular thermometers in the first place.
The complication is that the splitting also responds to electric fields, through dipole moments along and across the defect's axis. In a randomly oriented ensemble the standard axial and transverse terms average out to line broadening with no net shift, which is why an overall drift has usually been attributed to heat. Zvi and colleagues carried a second transverse dipole term, written d′, that is estimated to be about the same size as the conventional one and is routinely dropped. Keeping it changes the outcome: second-order perturbation theory gives a frequency shift proportional to the squared transverse field, so a randomly oriented ensemble does produce a systematic shift after all.
The physical driver is the diamond's own surface. Band bending at the surface ionizes substitutional nitrogen defects, the P1 centers, leaving a charge layer whose internal electric field the nitrogen-vacancy centers feel. Move electrons across that surface and the splitting moves with them.
Quantum pillar: sensing. Technology readiness: TRL 4 of 9. The theory, the coated particles and the fast readout were assembled and tested together under laboratory conditions, in a cultured mouse macrophage cell line rather than in tissue, an animal or a patient, and the cell counts behind the headline result are in single figures.
The saline experiment: minus 0.45 megahertz in buffer, and nothing in air or water
Bare 70-nanometer nanodiamonds were deposited on a coverslip and watched in phosphate-buffered saline, which approximates the ionic strength of biological fluid. After roughly one hour of laser excitation at about 0.1 milliwatts per square micrometer the splitting had fallen by an average of 0.45 megahertz, with the line shape broadening asymmetrically in the way the model predicts. The same particles measured in air and in pure water showed no systematic change, which points the effect at the chemistry of the surroundings rather than at the optics.
The team then coated the crystals with a 15-nanometer silica shell and repeated everything. The coated particles held their splitting steady through extended illumination. An Augmented Dickey-Fuller test returned 0.557 for the bare traces, consistent with a non-stationary drifting process, against 0.000306 for the core-shell traces, consistent with a stationary one. Allan deviation analysis put the short-time uncertainty about three times lower for coated particles, reaching a white-noise floor of 847 hertz.
Reading the splitting from a full spectrum is slow, so rapid tracking used a two-point scheme with a proportional-integral-derivative loop and simultaneous particle tracking, borrowed from established nitrogen-vacancy thermometry practice. That is what makes a 200-second time series inside a living cell possible at all.
What the shell changed in the cells: LDH, NF-κB and TNF-α at 50 micrograms per milliliter
Before any sensing, the group checked what the particles do to the cells. RAW-Blue macrophages, derived from the murine RAW 264.7 line, were incubated with bare or coated crystals, and each assay ran on its own schedule. Lactate dehydrogenase release, the standard membrane-damage readout, was measured at 6, 24 and 48 hours across concentrations from 10 to 200 micrograms per milliliter, and came out significantly lower with coated particles at every one of those points. NF-κB activation, read on a Quanti-Blue assay after overnight incubation, was reduced with coated particles above 50 micrograms per milliliter. Of thirteen cytokines and chemokines measured on a LEGENDplex panel, only TNF-α rose, and it rose less with the coated particles. Live-dead and EdU assays were run at the 24-hour point, where EdU incorporation ran 15 percent higher in coated-particle cells at the top concentration. On that basis the working concentration for live-cell measurement was set at 15 micrograms per milliliter.
This part deserves attention from anyone planning to use nanodiamonds in immunology. A sensor that provokes an inflammatory response is a poor instrument for measuring inflammation, and the paper is explicit that no mechanistic link between the shell's electrical effect and its biological effect has been established.
The inflammation result, and how small the numbers behind it are
Macrophages were stimulated with 1.5 micrograms per milliliter of lipopolysaccharide, the bacterial cell-wall molecule that drives the TLR-4 inflammatory program. Tracking started between 20 and 100 minutes after stimulation and ran 200 seconds per particle. Every bare crystal in an activated cell drifted: five particles in four cells, mean total shift 0.27 megahertz downward, standard error 0.03. Against unstimulated cells the drift terms differed with a t-test p-value of 0.00638. Coated particles in activated cells, seven of them, showed no appreciable drift, which is the control that makes the temperature explanation hard to sustain, since a silica shell does not insulate against a warming cell. How many cells those seven particles sat in is stated two ways in the paper: the running text says five, the caption to Figure 4 says four. The discrepancy does not change the conclusion, and a reader quoting the sample size should quote both.
Three further checks narrow the mechanism. Changing pH between 3.5 and 7, and adding bovine serum albumin at 300 grams per liter, moved the splitting insignificantly. A lysate of LPS-stimulated macrophages moved it by 296 kilohertz downward, closely matching the live measurement, and left coated particles alone at p equal to 0.459. Hydrogen peroxide from 1 to 500 micromolar produced a smaller shift of 156 kilohertz, enough to implicate oxidative species without accounting for the full effect. The authors call this preliminary and say the exact metabolic process is an open question.
Five particles in four cells is a proof of principle, not a validated assay. The comparison group is a cell line, not primary human macrophages. The stimulus is a single canonical agonist at a single dose. A reader who wants to know whether this distinguishes clinically meaningful activation states will not find the answer here, and the paper does not claim it.
How Quentir Reads It
Two things happened in one paper, and they carry different weight. The first is corrective. A body of intracellular thermometry has reported temperature excursions inside single cells that thermodynamics struggles to accommodate, ranging in the Iowa summary from 2 to 18 degrees Fahrenheit, and this work supplies a thermodynamically consistent alternative source for those readings. It does not retract anyone's measurement. It changes what the measurement is allowed to mean without an added control, and that control now exists in a concrete form: run the same experiment with a passivated particle and see whether the drift survives.
The second is additive, and it is the part with clinical reach. If a diamond crystal reports on the redox chemistry around it, then the same hardware that was sold as a thermometer becomes a probe of metabolic state. The author list points at where that is heading. Kunle Odunsi directs the University of Chicago Medicine Comprehensive Cancer Center; Marina Garassino and Michele Ferro sit in hematology and oncology; Melody Swartz works on tumor immunology. The stated next targets are macrophage polarization states and antigen-specific T-cell killing of tumor cells. Both are named as future work rather than demonstrated here, and both sit alongside live-cell fluorescent reporters that immunology already uses; the paper's own case is that those reporters photobleach and drift, so a charge-based readout would complement them rather than replace them.
For a hospital or a clinical laboratory the honest position is that nothing here is buyable. The measurement requires a confocal microscope, a microwave waveguide printed under the coverslip, photon counting and a tracking loop, and it consumes 200 seconds per particle. What is worth acting on today is narrower and cheaper: any group running nitrogen-vacancy thermometry in cells should add the coated-particle control, and any claim of single-cell thermogenesis built on uncontrolled zero-field-splitting drift should be read as unresolved until that control is reported. Two earlier posts in this Monitor sit next to this one and answer different questions. Its post of 26 August 2026, "Cells That Grow Their Own Quantum Sensors," covered the effort to have the cell manufacture the sensor itself, which asks where the sensor sits. On 23 July 2026 it covered a Heriot-Watt project to listen for free-radical chemistry inside immune cells in a tumor, which asks about a different physical channel: that work reads radicals through their magnetic signature, while the paper here reads the electrostatic state of the sensor's own surface. Same diamond, same defect, two distinct quantities, and confusing them would make the readouts look interchangeable when they are not.
Sources
Primary source: Uri Zvi, Shivam Mundhra, David Ovetsky, Qing Chen, Aidan R. Jones, Stella Wang, Maria J. Román-Vazquez, Marie Kim, Udoka M. Ibeh, Michele Ferro, Kunle Odunsi, Marina C. Garassino, Michael E. Flatté, Melody A. Swartz, Denis R. Candido, Aaron Esser-Kahn and Peter C. Maurer, "Probing Cellular Activity Via Charge-Sensitive Quantum Nanoprobes," Advanced Materials 38(14):e05107, published 4 February 2026, open access under CC BY, from which the secondary transverse dipole model, the 0.45 megahertz shift in phosphate-buffered saline, the absence of any shift in air and water, the 15-nanometer silica shell, the Augmented Dickey-Fuller p-values of 0.557 and 0.000306, the 847 hertz Allan deviation floor, the lactate dehydrogenase, Quanti-Blue and LEGENDplex assays, the 15 micrograms per milliliter working concentration, the 1.5 micrograms per milliliter lipopolysaccharide stimulation, the 0.27 megahertz mean drift in five particles across four cells, the 0.00638 t-test p-value, the lysate and hydrogen peroxide controls, and the authors' own statement that the underlying metabolic process remains an open question are all taken. The coated-particle sample size for the stimulated condition is reported twice in that paper and not identically: the running text gives seven particles in five cells and the caption to Figure 4 gives seven particles in four cells, and both readings are carried above rather than silently reconciled. The University of Iowa's news release of 26 August 2026, written around co-corresponding author Denis Candido, is the item through which the paper entered this Monitor's daily pool and is the source for the 2 to 18 degrees Fahrenheit range attributed to earlier intracellular thermometry reports. Author affiliations, funding by the National Science Foundation, the National Institutes of Health and the United States Department of Energy, and the article's submission and acceptance dates are read from the PubMed Central record. The comparison with genetically encoded protein sensors draws on this Monitor's post of 26 August 2026, and the distinction between magnetic radical sensing and charge-sensitive zero-field-splitting drift draws on this Monitor's post of 23 July 2026 on the Heriot-Watt free-radical sensing project.