Long Spin Relaxation Times in CVD-Grown Nanodiamonds

This paper reports a significant advancement in biosensing applications by utilizing an advanced heterogeneous nucleation technique to produce CVD-grown nanodiamonds with spin relaxation times nearly ten-fold longer than commercial equivalents, approaching bulk theoretical limits.

The Gist

Imagine you have a tiny, glowing diamond, no bigger than a speck of dust. Inside this diamond, there are tiny "defects" called Nitrogen-Vacancy (NV) centers. Think of these defects as tiny, glowing lightbulbs that can also act like super-sensitive compasses. They can detect magnetic fields and other invisible forces around them, making them perfect for sensing things inside living cells or chemical reactions.

However, there's a problem with the diamonds currently available in stores. They are like crumpled, jagged pieces of glass rather than smooth gems. Because they are so rough and damaged (created by crushing big diamonds into tiny ones), the "lightbulbs" inside them flicker out very quickly. In scientific terms, they lose their "spin" or memory too fast. This makes them poor sensors because they can't hold onto the information long enough to measure anything useful.

The Breakthrough: Building Diamonds from Scratch

The researchers in this paper decided to stop crushing big diamonds. Instead, they decided to grow their own from scratch, like baking a cake from batter rather than grinding up a finished cake.

They used a technique called Chemical Vapor Deposition (CVD). Imagine a giant, high-tech oven where they spray gas (like methane and hydrogen) onto a silicon surface. By carefully controlling the temperature and the gas, they coaxed the carbon atoms to stick together and grow into perfect, individual nanodiamonds.

To make sure these diamonds grew as separate, perfect little gems instead of a messy film, they first gave the silicon surface a tiny "scrub" with diamond dust. This created microscopic bumps that acted as starting points for the new diamonds to grow on.

The Results: A Super-Stable Glow

The results were impressive.

• The Old Way (Store-bought): The "lightbulbs" in commercial diamonds would flicker out in about 100 microseconds (a tiny fraction of a second).
• The New Way (Lab-grown): The diamonds grown in this lab kept their glow and "memory" for about 800 microseconds, with some lasting over 1.8 milliseconds.

This is like upgrading a flashlight that lasts for a split second to one that shines steadily for a long time. It's an 8-fold improvement. Because these diamonds are smoother and have fewer internal cracks, the "lightbulbs" inside are much more stable.

The "Shell" Experiment

The team also tried a clever trick to make the diamonds even better at sensing things right at their surface. They added a final "pulse" of nitrogen gas at the very end of the growth process to create a nitrogen-rich shell around the diamond.

Think of this like putting a thick, sticky coat of paint on a smooth ball. While the goal was to get more sensors near the surface, the thick nitrogen coat caused the diamond to grow in a messy, twinned way (like a crystal growing in two directions at once). This actually made the surface rougher and introduced more defects, which shortened the time the lightbulbs stayed on. So, while the idea was good, the execution showed that getting the "shell" just right is tricky and needs more work.

Why This Matters (According to the Paper)

The paper claims that by growing these diamonds carefully from the bottom up, they have created a batch of sensors that are:

1. Much more stable: They hold their quantum state much longer than anything currently sold.
2. More uniform: They are all roughly the same size (about 60 nanometers) and shape, unlike the jagged, irregular ones made by crushing.
3. Scalable: They showed a way to grow these on large surfaces and peel them off, meaning they could potentially make enough of these diamonds for real-world use, rather than just a few tiny samples.

In short, the researchers built a factory to grow perfect, smooth, glowing diamonds from gas, proving that "growing" is better than "grinding" when you need high-quality sensors.

Technical Summary

Technical Summary: Long Spin Relaxation Times in CVD-Grown Nanodiamonds

Problem Statement
Fluorescent nanodiamonds (FNDs) containing Nitrogen-Vacancy (NV) centers are promising tools for nanoscale sensing, including magnetic field detection and nano-NMR. However, their sensitivity is fundamentally limited by spin relaxation times, specifically the longitudinal relaxation time ($T_1$) and coherence time ($T_2$). Currently, commercially available FNDs, typically produced via High-Pressure High-Temperature (HPHT) diamond milling followed by irradiation and annealing, exhibit $T_1$ times in the range of 50–150 µs. This is significantly lower than the bulk diamond limit (~3 ms). The mechanical milling process introduces irregular shapes, surface defects, dangling bonds, and subsurface damage, while subsequent irradiation creates additional lattice strain and defect clusters. These factors degrade the quantum state characteristics necessary for high-sensitivity biosensing and room-temperature applications.

Methodology
The authors propose a "bottom-up" synthesis approach using Chemical Vapor Deposition (CVD) to grow nanodiamonds directly on a substrate, thereby avoiding the surface damage associated with top-down milling. The core methodology involves:

1. Heterogeneous Nucleation: Instead of using seeds (like detonation nanodiamonds or diamondoids) which often lead to film formation, the authors employ a pre-engineered substrate treatment. Silicon substrates undergo ultrasonic vibration with diamond micro-particles to create a "nano-roughened" surface with carbon traces. This enhances the density of nucleation sites, allowing for the growth of well-separated, individual nanocrystals rather than continuous films.
2. CVD Growth Conditions: Nanodiamonds are grown in a microwave plasma-enhanced CVD reactor at temperatures ranging from 700 °C to 950 °C. The gas mixture consists of 1% methane in hydrogen. Two sample sets are prepared:
   • Low-doped samples: Grown with a low nitrogen background (10 ppm $N_2$).
   • Shell-doped samples: Grown with a final 30-second pulse of high nitrogen concentration (5 sccm $N_2$) to create a highly nitrogen-doped shell ($\delta$-doping) near the surface.
3. Characterization: Particle size and morphology are analyzed via Scanning Electron Microscopy (SEM). NV concentration and size distributions are correlated using Photoluminescence (PL) mapping. $T_1$ relaxation times are measured using a confocal setup with 532 nm laser pulses and microwave $\pi$-pulses. The authors employ a specific data analysis technique involving the subtraction of purely optical decay curves from microwave-driven curves to isolate spin relaxation from charge-state alteration artifacts and baseline drift.

Key Results

• Enhanced $T_1$ Times: The CVD-grown nanodiamonds, averaging 60 nm in size, demonstrate a mean $T_1$ time of approximately 800 µs, with maximum values exceeding 1.8 ms. This represents an 8-fold improvement over commercial HPHT nanodiamonds (typically ~100 µs) and approaches bulk theoretical values.
• Size and Temperature Dependence: The $T_1$ times remain relatively consistent across different particle sizes (up to 200 nm) and growth temperatures. The narrowest size distribution (mean ~58 nm) was achieved at 700 °C.
• NV Formation Efficiency: The study found that NV formation efficiency is significantly higher at lower growth temperatures (700 °C), showing a 6.25-fold increase compared to 850 °C. This is attributed to more efficient nitrogen incorporation in {111} facets, which are prevalent in the mixed-facet nanocrystals.
• Impact of Shell Doping: While the intention was to create NV-rich shells for surface sensing, the high nitrogen pulse induced significant re-nucleation and twinning on {111} facets. This resulted in a highly defective shell layer, which actually reduced the $T_1$ times (to a range of 0.5–0.9 ms) compared to the low-doped samples. The defects introduced by the high nitrogen concentration and multi-twinning act as a spin bath, limiting relaxation times.
• Measurement Artifacts: The authors identified that laser illumination can cause charge-state alterations and band bending changes on the nanodiamond surface, leading to baseline drift in PL measurements. They demonstrated that subtracting specific microwave-driven curves is necessary to accurately extract $T_1$ values, a step often overlooked in standard HPHT measurements where such drift is less pronounced.

Significance and Claims
The paper claims that direct CVD growth via heterogeneous nucleation is a viable route to producing high-quality FNDs with spin relaxation times approaching bulk diamond limits, overcoming the defect limitations inherent to milling-based production. The authors highlight that the regular crystal morphology and low density of defects in these CVD-grown particles are the primary drivers for the improved $T_1$ times.

However, the authors remain modest regarding immediate scalability and application readiness. They note that while $T_1$ times are significantly improved, $T_2$ coherence times have not seen a similar leap, likely due to the paramagnetic defect density (P1 centers) inherent to the CVD growth process. Furthermore, the attempt to engineer surface-sensitive NV shells via nitrogen pulsing currently results in defective, twinned surfaces that degrade performance. The paper concludes that while the method produces high-quality particles, further optimization is required to eliminate multi-twinning in doped shells and to improve the overall photoluminescence yield (which is currently lower than HPHT diamonds) and production volume for widespread biological and chemical sensing applications. The work establishes a foundational physics understanding of how growth conditions and surface engineering directly dictate the quantum sensing capabilities of nanodiamonds.