Skip to content

FRET Between NV Centers In Diamond And Chlorophyll Molecules - A Novel Resource For Multimodal Sensing And Imaging In Plant Cells

FieldValue
JournalAdvanced Optical Materials
Published2026-04-01
AuthorsSebastian Westrich, Nimba Oshnik, Nina Thiele, Nina Burmeister, Yanis Abdedou
DOI10.1002/adom.202600003

Shallow NV ensembles in single-crystal diamond transferred excitation energy to chlorophyll a and b. Their lifetimes fell from 14.4 ± 0.4 ns to 3.3 ± 0.5 ns at 7 ± 3 nm and 3.8 ± 0.6 ns at 9 ± 4 nm. Photobleaching the chlorophyll restored part of the lifetime, while ensembles at 40 and 72 nm showed no measurable change. Optically detected magnetic resonance remained visible under the chlorophyll layer, although its contrast fell to 4 ± 2%.

The NV⁻ fluorescence band spans 600 to 800 nm, with its zero-phonon line at 637 nm. The chlorophyll mixture has maximum absorption at 664 nm, so donor emission and acceptor absorption overlap. A 600 nm long-pass filter suppresses the NV⁰ zero-phonon line at 575 nm and much of its phonon sideband. Tri-acid treatment also stabilizes NV⁻, especially for shallow centers. These conditions indicate that the detected FRET signal is dominated by NV⁻.

Using random donor and acceptor orientations with κ² = 2/3 and a diamond refractive index of 2.4, the authors calculated a Förster radius of 3.6 ± 0.1 nm. They concluded that FRET was feasible for chlorophyll molecules placed on the surface.

Chlorophyll fluorescence covered 650 to 800 nm and peaked at 680 nm. This overlaps the NV emission, so the two signals cannot be separated spectrally in the combined sample. The extracted mixture had a lifetime of 0.63 ± 0.08 ns on silicon and 0.6 ± 0.1 ns on diamond. Prior in vivo measurements on comparable green plants gave 0.6 ns and 0.7 ns for chlorophyll a aggregates, consistent with the measured acceptor lifetime.

The study used three high-purity, (100)-oriented, electronic-grade single-crystal diamonds. Nitrogen and boron concentrations were below 5 ppb and 1 ppb, respectively. Every sample was polished to Ra below 1 nm. A stress-relief etch removed polishing damage before nitrogen implantation at a 7° angle.

Four implantation settings were used. They combined 4, 6, 30, and 60 keV nitrogen ions with fluences of 4 × 10¹¹, 3 × 10¹¹, 5 × 10¹⁰, and 3 × 10¹⁰ cm⁻². The reported depth distributions were 7 ± 3, 9 ± 4, 40 ± 11, and 72 ± 20 nm. SRIM supplied these estimates and tends to underestimate measured implantation depths.

The crystals were annealed at 800 °C for 2 hours in vacuum below 7.8 × 10⁻⁷ mbar. A 1:1:1 tri-acid cleaning at 500 °C for 1 hour removed residual nondiamond carbon. An assumed NV creation yield of about 1% placed fewer than 10 centers in each confocal focus.

Chlorophyll deposition and lifetime measurement

Section titled “Chlorophyll deposition and lifetime measurement”

Chlorophyll a and b were extracted from Arabidopsis thaliana with 95% ethanol. The authors spin-coated 4 µL onto each diamond at 2,000 rpm for 1 minute. Atomic force microscopy on test films showed aggregation and uneven coverage. Films spun at 1,500 to 3,500 rpm had an average roughness of Ra = 20 ± 8 nm. Areas without agglomerates were 14 ± 3 nm thick.

Measurements used positions without visible agglomerates to keep the donor-acceptor environment consistent. A custom confocal microscope supplied 520 nm pulses about 50 ps long at 5 to 10 MHz. A 600 nm long-pass filter selected the fluorescence, and the instrument response had a 440 ps FWHM. Two-component exponential fits separated chlorophyll and NV decays after convolution with that response.

Each sequence recorded the NV decay, bleached the chlorophyll for 20 minutes with 520 nm continuous-wave light at 280 µW, and repeated the measurement. Several surface positions were tested to account for coverage variation. Under continuous-wave excitation, chlorophyll emission fell to about 10% within 15 minutes. Under pulsed excitation at an average 6 µW, the count rate declined over about 45 minutes. Each lifetime integration lasted 300 s, so the authors treated the slower pulsed change as negligible.

Shallow ensembles show distance-selective quenching

Section titled “Shallow ensembles show distance-selective quenching”

The shallow NV ensembles were quenched below 4 ns. Their recovered lifetimes remained below the bare-diamond reference. The authors attribute this difference to residual chlorophyll that continues to provide a weak nonradiative decay channel.

NV depth distributionBare diamond lifetimeLifetime with chlorophyllLifetime after bleaching
7 ± 3 nm14.4 ± 0.4 ns3.3 ± 0.5 ns9.3 ± 0.5 ns
9 ± 4 nm14.4 ± 0.4 ns3.8 ± 0.6 ns10.8 ± 0.6 ns
40 and 72 nm regions11.5 ± 0.5 ns at 40 nm; 12 ± 1 ns at 72 nm11.9 ± 1.1 ns12.1 ± 0.7 ns

The final row combines measurements from two implanted regions on one sample.

A representative measurement at 9 ± 4 nm gave 4.36 ± 0.03 ns before bleaching and 11.01 ± 0.05 ns after. The deeper regions gave 11.9 ± 1.1 ns before and 12.1 ± 0.7 ns after. Their values agree with bare lifetimes of 11.5 ± 0.5 ns at 40 nm and 12 ± 1 ns at 72 nm. The unchanged deeper response supports the short-range FRET mechanism.

A planar acceptor model describes ensemble coupling

Section titled “A planar acceptor model describes ensemble coupling”

The model replaces the single-pair z⁻⁶ distance law with z⁻⁴ for a donor above a planar acceptor ensemble. It treats the chlorophyll film as an effective 2D plane and averages decays over each Gaussian NV depth distribution. The calculation neglects the film’s finite thickness. Chlorophyll surface density could not be measured because the film was locally inhomogeneous and aggregated. Density therefore remains embedded in one effective coupling parameter.

Measured decays were fit with one exponential after the first nanosecond, which excluded chlorophyll emission and the instrument response. The fitted coupling lengths were 13 ± 2 nm for 7 ± 3 nm and 17 ± 2 nm for 9 ± 4 nm. These are collective donor-ensemble parameters, not single-pair Förster radii. Comparable donor-surface systems report interaction distances of about 1 to 40 nm, placing both values within the published range.

Optically detected magnetic resonance tests whether strong quenching removes the spin-dependent signal needed for magnetic readout. The bare ensemble produced 10 to 12% contrast without an external magnetic field. With chlorophyll, the average contrast fell to 4 ± 2%. A representative spectrum from the 9 ± 4 nm ensemble showed 6% contrast and a 10 MHz FWHM.

Chlorophyll fluorescence is not spin dependent and adds background to the microwave response. The authors retained only ODMR datasets without significant photoluminescence or lifetime changes before and after acquisition. Contrast recovered after bleaching. The authors interpret the retained ODMR signal as support for combining FRET proximity measurement with spin readout, although the contrast remains reduced under the acceptor layer.

The experiment is a surface model rather than an intracellular test. The NV ensembles remained in bulk single-crystal diamond, and chlorophyll was spin-coated on top. The authors propose delivering NV-containing nanodiamonds into plant cells with established gene-gun methods. That delivery step was not tested in this work.

The measured depth response constrains sensor design for this geometry. NV centers within roughly 10 nm can report surface-bound chlorophyll, while centers tens of nanometers deeper showed no measurable response. A practical sensor would therefore need controlled shallow implantation and low-strain preparation. The coating process must also control aggregation, which remained unresolved.

Other limits include the planar model’s neglected thickness and the unavailable acceptor-density measurement. Photobleaching restored only part of the shallow-NV lifetime, while chlorophyll lowered ODMR contrast. The current data therefore do not provide an independent acceptor-density calibration.

For plant-cell use, the authors suggest combining FRET proximity to chlorophyll with NV spin detection of radical species. That application remains a proposed direction because the present measurements do not place the sensor inside a plant cell.

For a chlorophyll-coupled NV sensor, the base material is a polished single-crystal diamond plate carrying shallow NV centers. The 6CCVD diamond materials catalog lists MPCVD-grown single-crystal plates and wafers up to 125 mm, with single-crystal polishing below Ra 1 nm. The paper still requires nitrogen implantation, 800 °C annealing, and surface preparation beyond that base material, which can be specified through the linked engineering support page.

View Original Abstract

ABSTRACT This work demonstrates efficient Förster resonance energy transfer (FRET) between ensembles of shallow nitrogen‐vacancy (NV) centers located and below a single‐crystal diamond surface and a naturally occurring fluorophore, here a mixture of chlorophyll a and b molecules extracted from Arabidopsis thaliana . The broad fluorescence band of NV centers spectrally overlaps with the chlorophyll molecule’s absorption, enabling FRET. Consequently, depositing a chlorophyll layer on the diamond surface reduces the NV fluorescence lifetime from approximately to below , indicating efficient FRET. Laser‐induced photobleaching of chlorophyll restores the unquenched NV lifetime. NV centers located deeper within the diamond ( and ) remain unaffected, confirming that the observed quenching originates from a short range FRET mechanism. The NV ensembles retain their optically detected magnetic resonance (ODMR) contrast while observing FRET, demonstrating preservation of their spin properties. Consequently, these proof‐of‐principle experiments demonstrate the feasibility of combining FRET‐based distance measurements with magnetic sensing using optically readable spins.