Skip to content

Nanodiamond Sensing in Dynamic Environments With Fast‐Tracking Through Four‐Point Positioning

FieldValue
JournalAdvanced Functional Materials
Published2026-08-03
AuthorsGuoli Zhu, Ming-Zhong Ai, Zhi‐Yu Zhao, Weng-Hang Leong, Chen Shi-ning
DOI10.1002/adfm.77491

Four avalanche photodiodes raised the upper measurable diffusion coefficient of nanodiamond tracking to 1.5 µm²/s, compared with 0.16 µm²/s for a single-APD reference method. The four-point positioning system tracked a nanodiamond in 50 wt.% glycerol and water for 60 s, while the single-APD system lost the target within a few seconds. Parallel fluorescence collection supported faster tracking and ODMR sensing without a significant measured loss of sensitivity.

Parallel detection expands the tracking range

Section titled “Parallel detection expands the tracking range”

The 4PPT feedback cycle used a 20 ms fluorescence dwell and was about one order of magnitude shorter than six-point or eight-point sequential scanning with one APD. The gain came from four simultaneous detector channels, which determined the three-dimensional position within one feedback cycle.

Generated Brownian motion tests found similar error ranges for 4PPT at D = 0.5 µm²/s and for 1-APD at D = 0.05 µm²/s. The z-axis error was larger because the optical resolution was poorer in z. The authors also state that simultaneous collection quadrupled the photon-count-rate ceiling of one APD.

Four-point positioning couples tracking and ODMR

Section titled “Four-point positioning couples tracking and ODMR”

Real-time tracking used a home-built confocal microscope on an Olympus IX-73 frame. A 532 nm laser and a 60× oil immersion objective with NA 1.4 excited the NV centers. A three-axis piezoelectric stage moved the sample.

Three 50:50 cubic beamsplitters divided the fluorescence equally into four paths. Each path used a 10× collimating objective and an optical fiber coupled to a 4-pack single-photon-counting avalanche photodiode.

P1 and P2 were separated along y, while P3 and P4 were separated along x. The vertical arrangement placed the P1 to P2 separation toward positive z and the P3 to P4 separation toward negative z. The coordinate system was rotated 45° around z to include all four detector signals in the x and y estimates.

The four signals supplied three error inputs to proportional-integral-derivative controllers. Their output voltages repositioned the stage and recorded the three-dimensional trajectory. Tracking sampled fluorescence for 20 ms in each feedback cycle.

For ODMR, each microwave frequency had a 40 µs dwell. Counts from all four detectors were summed, and 80 frequency sweeps formed one full spectrum. The summed signal retained the same signal-to-noise ratio as the single-APD system under the same excitation conditions.

Temperature sensing used zero-field ODMR with 31 frequencies and a total time of about 0.1 s. A double-peak Lorentzian fit gave the zero-field splitting. Rotation sensing collected spectra under two different magnetic fields of about 50 G. Fitting both spectra to four NV orientations gave Euler angles α, β, and γ in a cycle of about 1 s. A single-field mode reduced acquisition to about 0.5 s but did not yield full three-dimensional rotation.

ODMR sensitivity showed no significant motion dependence

Section titled “ODMR sensitivity showed no significant motion dependence”

At D = 0.8 µm²/s, the zero-field ODMR spectrum matched the stationary reference. The temperature sensitivity was estimated from zero-field splitting scatter against inverse square root acquisition time. It showed no significant dependence on D across the tested range, which reached the order of 10⁻¹ µm²/s.

Rotation tests at D = 0.3 µm²/s used the dual-field protocol. The sensitivities of α, β, and γ likewise showed no significant dependence on D. The authors therefore report no significant degradation within the tested range, rather than exact invariance.

Thermorheology linked temperature and motion

Section titled “Thermorheology linked temperature and motion”

Thermorheology was tested in 70 wt.% glycerol and water because nanodiamond diffusion there exceeded the single-APD tracking limit. A nanodiamond was heated from ambient temperature to 60 °C and measured with a resistance temperature detector.

By 200 s, the zero-field splitting shift increased in agreement with the resistance temperature detector curve. The velocity derived from the trajectory also rose overall during heating. The authors listed Brownian motion and thermal convection as likely sources of the fluctuations.

At equilibrium temperatures of 30, 45, and 60 °C, multiple nanodiamonds were tracked. Their mean square displacement gave diffusion coefficients, and box plots showed an increase with temperature. The authors ascribed the spread to particle-size dispersion because diffusion depends on temperature, size, and viscosity.

Applying the Stokes-Einstein relation gave diameters of about 100 to 800 nm. Dynamic light scattering gave a consistent size estimate. Bulk viscosity was measured from 20 to 60 °C at a shear rate of 50 s⁻¹ and a 0.5 min sampling interval.

Live-cell tracking separated membrane and cytoplasmic motion

Section titled “Live-cell tracking separated membrane and cytoplasmic motion”

On the plasma membrane of an MCF-7 cell, a fibronectin-coated nanodiamond stayed within a volume of about 1 µm³ over 900 s. Its typical speed was about 0.1 µm/s. The motion was consistent with binding to membrane receptors attached to the actin network.

Cell tracking was performed at 37 °C and 5% CO₂. In a HeLa cell, BSA-coated nanodiamonds were tracked in the cytoplasm. The trajectory split into two stages. From 0 to 480 s, high instantaneous speeds appeared often. After 480 s through 650 s, motion was slow and confined. Maximum speed exceeded 2 µm/s at about 90 s.

Analysis of 10 s segments gave an anomalous diffusion exponent of 1.6 for directed motion from 445 to 455 s. The later stage had exponents below 1, which indicates confinement within the measured motion.

Rotation tracking used one magnetic field because full three-dimensional tracking is slow when angular velocity is very large. During the first 300 s, the resonance frequency changed rapidly and was barely resolvable. From 300 to 480 s, the spectrum stabilized. The authors suggested that rotation was quenched or proceeded around the applied magnetic field.

They compared the confined rotation with directional translation to microtubule-associated trafficking. They also stated that organelle-specific staining or co-localization imaging is needed to identify the structures involved.

Feedback and geometry set the performance limits

Section titled “Feedback and geometry set the performance limits”

The feedback cycle remained the main speed constraint. The 20 ms interval included about 10 ms for piezo-stage motion and about 10 ms for data-acquisition communication. Tracking failed mainly when displacement exceeded the detection region before correction.

The tetrahedral geometry used lateral separations of about 200 nm and an axial separation of about 400 nm. Its size was about 0.6 µm in the x-y plane. Offsetting reduced the total collected fluorescence to about 60% of the maximum.

A single APD had a maximum translational tracking speed of about 1 µm/s. The 4PPT system reached instantaneous speeds of several µm/s. Full rotation tracking remained slow for rapidly rotating intracellular particles, so the single-field method traded full three-dimensional orientation for faster acquisition.

The demonstrated envelope covers diffusion coefficients up to 1.5 µm²/s and speeds of several µm/s. It overlaps diffusion coefficients of the order of 10⁻³ to 1 µm²/s and speeds above 1 µm/s reported for dynamic fluids and cells.

Device selection must therefore consider tracking speed, ODMR contrast during motion, particle dimensions, NV-center count, and surface chemistry. BSA coating was used to limit aggregation in culture medium, while fibronectin enabled membrane binding. Performance outside the tested particle class or diffusion range would require separate validation.

The paper uses fluorescent NV-center nanodiamond particles for fluid and live-cell tracking, with sizes in the hundreds of nanometers and more than 1000 NV centers each on average. It needs BSA coating to limit aggregation in culture medium or fibronectin coating for membrane binding. 6CCVD’s diamond materials catalog lists single-crystal and polycrystalline diamond grown by MPCVD, including material from 0.1 to 500 micrometers thick, but the listed facts do not specify this particle format, NV density, or coating chemistry, so a material adaptation should be checked through engineering support before sensor use.

View Original Abstract

ABSTRACT Nitrogen‐vacancy (NV) centers in nanodiamonds are excellent nanoscale sensors for measuring parameters such as temperature, magnetic field, and viscosity in complex fluidic environments, including living cells. However, the rapid motion imposes a significant challenge for continuous, real‐time tracking and sensing measurements. Here, we present a fast single particle tracking (SPT) method featuring a tetrahedral detection geometry for time‐efficient parallel fluorescence collection using four avalanche photodiodes (4‐APDs). This approach eliminates the temporal latency of traditional sequential scanning, achieving an order of magnitude improvement in the temporal resolution and the upper limit of the measurable diffusion coefficient compared to previously reported methods based on a single APD. The SPT is integrated with multi‐parameter quantum sensing based on optically detected magnetic resonance (ODMR) of NV centers. The sensitivities of ODMR‐based temperature and 3D rotation sensing are evaluated at different diffusion coefficients, which shows no significant degradation within our measurement range. We apply the system for thermorheology measurements in glycerol/water mixtures under thermal ramps. Additionally, we perform simultaneous translation and rotation tracking in live cells, revealing correlated translational and rotational dynamics. This approach advances multi‐parameter nanoscale sensing for soft matter and biological applications, paving the way for real‐time nanoscale sensing in highly dynamic fluidic environments.