---
title: ND1 Centers in Diamond Optical Storage
url: https://www.emergentmind.com/topics/nd1-centers
type: topic
---

# ND1 Centers in Diamond Optical Storage

ND1 centers, in the context of recent diamond photonics and optical data-storage research, are treated as emissive defect centers in diamond whose characteristic optical signature lies in the violet/UV spectral region and whose dominant emission under \(1030\ \mathrm{nm}\) femtosecond excitation appears below \(400\ \mathrm{nm}\). A 2025 study positions ND1-rich defect sites as the information carrier in a high-density, three-dimensional optical data-storage platform for extreme environments, with binary data encoded as localized luminescent “pits” that can be deterministically written and optically read using the same near-infrared femtosecond source [2507.11363]. The same work presents ND1 centers not as an abstract color-center class but as a specific systems-level solution combining sub-micrometer write localization, three-photon-excited readout, and stability under thermal, magnetic, chemical, and long-duration storage stress. A central qualification is that the reported archival lifetime of “millions of years” is a projected claim grounded in defect robustness rather than a direct lifetime measurement [2507.11363].

## 1. Defect identity and spectroscopic definition

Within the reported storage architecture, ND1 centers are defined operationally by their optical behavior rather than by a fully developed microscopic defect model. The paper treats them as a specific emissive defect center in diamond with a dominant blue/violet-to-UV photoluminescence contribution under \(1030\ \mathrm{nm}\) femtosecond excitation, especially a peak below \(400\ \mathrm{nm}\). The authors cite Zaitsev for defect spectroscopy and state that the ground-to-excited-state transitions of ND1 lie in the \(3.2{-}3.6\ \mathrm{eV}\) region, which is central to their excitation argument [2507.11363].

Two roles are assigned to ND1 centers. First, they serve as the written information carrier: binary data are encoded as localized luminescent “pits,” that is, laser-modified sites containing a high density of ND1 defects. Second, they function as a spectrally advantageous readout center because their emission is in the violet/UV, which the paper argues improves diffraction-limited readout resolution relative to visible/NIR-emitting centers such as NV\(^-\) and GR1. The reported spectrum under \(1030\ \mathrm{nm}\) excitation spans from the violet to the NIR, but the dominant peak below \(400\ \mathrm{nm}\) is assigned to ND1, while NV\(^-\) emission at \(637\ \mathrm{nm}\) and GR1 emission at \(741\ \mathrm{nm}\) appear only as minor contributions under this excitation scheme [2507.11363].

The paper is explicit about what it does not establish. It does not provide a detailed defect-chemistry assignment beyond calling the emitters ND1 defects or ND1 centers, and it does not develop a full microscopic electronic-structure model. It also does not report polarization dependence of ND1 excitation or emission. A plausible implication is that the work is primarily a device and spectroscopy study rather than a definitive microscopic identification study. The authors do, however, infer that ND1 centers are electronically isolated from the bulk because their photoluminescence is excited by three-photon absorption at \(1030\ \mathrm{nm}\) rather than by exciting carriers across diamond’s full bandgap [2507.11363].

## 2. Laser writing and defect-generation mechanism

The reported platform creates ND1-rich pits by focusing near-infrared femtosecond pulses into single-crystal diamond at chosen three-dimensional coordinates. The experimental writing conditions are \(1030\ \mathrm{nm}\) wavelength, \(\sim 150\ \mathrm{fs}\) pulse duration, \(1\ \mathrm{MHz}\) repetition rate, a \(100\times\) objective with \(\mathrm{NA}=0.9\), and typical writing pulse energies of \(50{-}100\ \mathrm{nJ}\). The authors describe pit formation as a strong-field nonlinear process: multiphoton absorption of NIR photons occurs first, avalanche ionization follows, and these processes induce localized structural changes at the focal point. ND1 centers are formed together with many other emissive defects during this laser-driven structural modification [2507.11363].

The work distinguishes sharply between defect creation and defect excitation. For pit formation in pristine diamond, the authors invoke the diamond bandgap of about \(5.47\ \mathrm{eV}\) and the \(1030\ \mathrm{nm}\) photon energy of about \(1.2\ \mathrm{eV}\), concluding that the minimum number of photons needed to cross the bandgap is five. They state that “The band gap of diamond (about 5.47 eV) sets the minimum number photons to 5,” and therefore pit formation begins only in the intense central region of the focal spot. Using the simulated beam waist \(w_0=300\ \mathrm{nm}\), they estimate the seed radius for initiation as
\[
w_0/\sqrt{5} \approx 140\ \mathrm{nm}.
\]
After initiation, defect-assisted absorption permits growth beyond this seed region [2507.11363].

The resulting structures are explicitly described as structural defects rather than reversible optical-state modifications. This distinction is important for interpreting robustness claims. The paper contrasts ND1-rich structural defects with NV centers, noting that NV centers suffer from charge-state instability during optical reading, whereas ND1-based pits are presented as structurally and optically stable. At the same time, the paper stresses a trade-off: the written pits are not erasable under normal conditions, and once defects are present they can absorb at lower intensity than pristine diamond, which contributes to continued pit growth [2507.11363].

Surface-pit morphology was characterized by AFM. A typical pit is reported as about \(90\ \mathrm{nm}\) deep, with lateral widths of \(400\ \mathrm{nm}\) and \(500\ \mathrm{nm}\) in orthogonal directions. The lateral shape is described as somewhat random owing to highly nonlinear excitation and a large amount of heat deposition during writing. The authors further note that the pits are smaller than the simulated diffraction-limited focus, which they attribute to seeding only in the highest-intensity central region. This establishes the basic spatial scale of the written ND1-rich storage units while also indicating that their growth is governed by both thresholded initiation and subsequent defect-assisted expansion [2507.11363].

## 3. Three-photon excitation, luminescence, and intrinsic volumetric readout

A central result is that once ND1 centers have been created, their photoluminescence can be read at the same \(1030\ \mathrm{nm}\) wavelength by three-photon absorption rather than by band-to-band excitation. The paper supports this with two experimental observations. First, using two frequency-shifted beams overlapped to create intensity modulation at \(\phi=\phi_2-\phi_1=2\ \mathrm{kHz}\), the FFT of the UV photoluminescence below \(420\ \mathrm{nm}\) shows harmonic components at \(4\ \mathrm{kHz}\) and \(6\ \mathrm{kHz}\), which the authors interpret as signatures of three-photon absorption. Second, the photoluminescence shows a cubic dependence on excitation power, expressed in the text as \(I_{\mathrm{PL}} \propto P^3\), which the authors cite as further support for 3PA [2507.11363].

The energetic argument is similarly explicit. Three photons at \(1030\ \mathrm{nm}\) contribute approximately
\[
3 \times 1.2\ \mathrm{eV} = 3.6\ \mathrm{eV},
\]
which is resonant with the ND1 transition range of \(3.2{-}3.6\ \mathrm{eV}\). The paper states that this energy is nonresonant with the other common emissive centers considered, and uses that resonance mismatch to explain why ND1 emission dominates over NV and GR1 under the chosen excitation [2507.11363].

The reported optical consequences of this excitation mechanism are systems-relevant. Because multiphoton absorption is highly localized, excitation occurs only in the focal volume, and the readout is therefore intrinsically three-dimensional. The authors state that layers above and below the focal plane do not significantly contribute and report no interlayer crosstalk even in raw, unprocessed acquisitions from four vertically stacked planes. This is the basis for the paper’s claim of “intrinsic 3D reading,” and it is presented as an architectural simplification relative to earlier GR1-based approaches that required different wavelengths for writing and reading together with confocal microscopy [2507.11363].

Time-correlated single-photon counting on emission below \(420\ \mathrm{nm}\) yielded an average photoluminescence lifetime of less than \(2\ \mathrm{ns}\). The figure caption describes the decay as bi-exponential and states that lifetimes are shorter than \(10\ \mathrm{ns}\), but the main text uses the \(\sim 2\ \mathrm{ns}\) timescale as the practical read-speed limit. This short lifetime underlies the projected sequential read speed of about \(500\ \mathrm{Mbits/s}\), or \(\sim 60\ \mathrm{MBytes/s}\), which the authors compare with a cited Blu-ray reading speed of \(36\ \mathrm{Mbits/s}\) [2507.11363].

## 4. Experimental architecture and three-dimensional addressing

The storage medium in the reported implementation was single-crystal diamond from Element Six, with nitrogen impurity concentration below \(5\ \mathrm{ppb}\), NV concentration below \(0.03\ \mathrm{ppb}\), and described in the discussion as undoped pure diamond. These low impurity levels are stated to reduce background from pre-existing NV centers and thereby make the laser-written defect signal easier to isolate [2507.11363].

The write/read optical system used a femtosecond laser system (Impulse, Clark-MXR) as a common source for both writing and reading. The beam was directed by an \(800\ \mathrm{nm}\) short-pass dichroic mirror and focused by a Nikon \(100\times\), \(\mathrm{NA}=0.9\) objective. Intensity was controlled by a continuously variable neutral-density filter, and a high-speed shutter controlled laser exposure. Three-dimensional positioning was achieved by mounting the diamond on a motorized X–Y nanopositioning stage (HLD117NN, Prior Scientific) and controlling the \(z\) coordinate through independent motorized objective motion using a PS3H122R Z-axis drive (Prior Scientific). The \(x,y\) coordinates were therefore assigned by stage positioning and the \(z\) coordinate by objective displacement [2507.11363].

Readout used the same optical setup as writing, but at pulse energies below \(5\ \mathrm{nJ}\). The beam was raster scanned across each plane; written pits emitted blue photoluminescence, while unmodified “lands” remained dark. Photoluminescence was collected in the epi-direction through the same objective, residual excitation was blocked by an additional \(800\ \mathrm{nm}\) short-pass filter, and the signal was detected by an avalanche photodiode digitized by a 24-bit acquisition card. For lifetime measurements, the APD was replaced by a SPAD and timing was recorded with Time Tagger 20 (Swabian Instruments). The paper also states that software synchronized optical signals and stage motion [2507.11363].

The architecture is explicitly sequential. Pits are written one by one, and the pit size can be tuned by exposure time and laser pulse energy. The system supports both surface and bulk writing. The multilayer demonstration comprises four vertically stacked planes, and the paper states that layer-specific photoluminescence remains well resolved even in raw acquisitions. This demonstration supplies the direct experimental basis for three-dimensional addressing rather than leaving it as a purely theoretical consequence of nonlinear excitation [2507.11363].

| Parameter | Reported value |
|---|---|
| Writing wavelength | \(1030\ \mathrm{nm}\) |
| Pulse duration | \(\sim 150\ \mathrm{fs}\) |
| Repetition rate | \(1\ \mathrm{MHz}\) |
| Writing pulse energy | \(50{-}100\ \mathrm{nJ}\) |
| Stable readout pulse energy | \(<5\ \mathrm{nJ}\) |
| Objective | \(100\times\), \(\mathrm{NA}=0.9\) |
| Simulated lateral waist | \(w_0=300\ \mathrm{nm}\) |
| Simulated axial FWHM | \(620\ \mathrm{nm}\) |

## 5. Stability under extreme conditions and retention claims

The extreme-condition argument in the paper rests on a set of direct stress tests. During readout-like exposure at \(\sim 5\ \mathrm{nJ}\) pulse energy and \(\sim 150\ \mathrm{fs}\) pulse duration, continuous illumination for \(180\ \mathrm{min}\) produced no physical change, and the photoluminescence at the ND1 peak near \(400\ \mathrm{nm}\) remained stable. Over repeated photoluminescence measurements spanning one year under ambient conditions, no degradation was observed. After exposure to concentrated aqua regia and concentrated piranha solution for \(24\ \mathrm{h}\), the paper reports no spectral shift and no intensity loss; instead, a slight increase in photoluminescence is attributed to surface cleaning [2507.11363].

Thermal and magnetic tests were similarly broad in scope. The written defects were examined over temperatures from \(4\ \mathrm{K}\) to \(500\ \mathrm{K}\), with no degradation observed, and under magnetic fields up to \(5\ \mathrm{T}\), where the spectra did not change. The paper does not report detailed thermal shifts, quenching coefficients, temperature-dependent lifetime changes, Zeeman splitting, polarization analysis, or line-shape analysis; the result is specifically one of stability and absence of degradation under these conditions [2507.11363].

The work also frames readout as exposure to an extreme transient electric field. At a focus spot of area \(\sim 0.3\ \mu\mathrm{m}^2\), the estimated instantaneous electric field during readout is \(\sim 1.1\times10^{10}\ \mathrm{V\,m^{-1}}\), and the authors present this as evidence that the stored data withstand extreme electric fields during optical interrogation. The paper does not, however, report a controlled static electric-field dependence study. This distinction matters because the evidence concerns transient optical-field exposure rather than a calibrated DC or low-frequency field response [2507.11363].

The measured photoluminescence exhibits fluctuations of about \(5\%\) around the average during prolonged monitoring, while laser pulse energy fluctuates by less than \(2\%\). The authors attribute the larger optical variation to amplification of fluctuations by the nonlinear process rather than to blinking or bleaching. In the same section, they delimit safe operating conditions: writing initiation requires more than \(50\ \mathrm{nJ}\), pits begin to grow once defect sites are present if the pulse energy exceeds \(20\ \mathrm{nJ}\), and stable readout is achieved below \(5\ \mathrm{nJ}\). The result is a defined separation between write, growth, and read regimes rather than a claim of unlimited readout robustness [2507.11363].

A common misconception would be to read the “millions of years” statement as an experimentally verified retention time. The paper does not provide such a direct test. The experimentally demonstrated retention evidence is no degradation over one year under ambient repeated measurements together with the thermal, magnetic, chemical, and readout-like stress tests. The longer archival claim is explicitly a projection based on defect robustness [2507.11363].

## 6. Relation to other defect platforms, analytical context, and limitations

The study positions ND1 centers against NV- and GR1-based optical storage schemes. NV centers are described as suffering from charge-state instability during optical readout, while structural defects are presented as more robust for extreme-environment archival storage. Compared with GR1-based systems, the ND1 architecture is said to simplify the optical setup because the same \(1030\ \mathrm{nm}\) source is used for both writing and reading and because the intrinsic volumetric localization of three-photon excitation removes the need for confocal sectioning [2507.11363].

At the same time, the paper is explicit about limitations. The storage medium is not rewritable: “the pits cannot be erased under normal conditions.” Amplified femtosecond lasers are currently required, making the system bulky and energy inefficient. Even if high-repetition-rate laser sources are available, actual write speed may be limited by mechanical movement during writing, similar to Blu-ray constraints. Pit-size optimization remains incomplete; the current pits are sub-micron but not yet pushed to the smallest possible dimensions. The authors suggest that doping studies may reduce write/read threshold energies and that spatio-temporal pulse shaping may shrink pits further and improve storage density. They also note a possible trade-off in doping: nitrogen-rich material might complicate the emission spectrum by adding stronger red-region contributions [2507.11363].

The work reports no experimentally demonstrated areal or volumetric storage density in bits/cm\(^3\) or bits/in\(^2\), and it does not provide a formal storage-capacity formula. Density arguments remain qualitative: violet/UV emission aids diffraction-limited resolution, three-dimensional layering increases density, and pulse shaping may reduce pit size toward Blu-ray-scale dimensions. Similarly, the projected write speed of \(>200\ \mathrm{MByte/s}\) is based on an external Yb-fiber laser system with \(1.83\ \mathrm{GHz}\) repetition rate and pulse energy above \(50\ \mathrm{nJ}\), not on the \(1\ \mathrm{MHz}\) laboratory system used for the core experiments [2507.11363].

Broader defect-spectroscopy literature supplies only indirect context for ND1 identification in the specific sense used here. A 2016 level anti-crossing spectroscopy study of NV\(^-\) centers does not explicitly mention ND1 or its common aliases, but it demonstrates how magnetic-field-dependent NV\(^-\) fluorescence can be used to infer parameters of other paramagnetic defect centers through coupled-spin Hamiltonian modeling [1609.07957]. This suggests a methodological route for future work if an ND1-related state were shown to be paramagnetic and magnetically coupled to NV\(^-\), but that paper itself provides no direct ND1-specific facts and therefore does not alter the 2025 paper’s operational definition of ND1 as an optically dominant UV/violet defect population in laser-written pits [1609.07957].

Taken together, the current literature presents ND1 centers less as a fully resolved microscopic defect species than as a laser-writeable, optically readable, structurally robust defect platform in diamond. The distinctive features established experimentally are deterministic creation of ND1-rich sub-micron pits at chosen three-dimensional positions, dominant UV/violet photoluminescence under \(1030\ \mathrm{nm}\) excitation, three-photon-driven intrinsic 3D readout, and no observed degradation under a defined set of extreme-condition tests. The principal unresolved issues concern microscopic defect assignment, rewritability, pit-size minimization, and quantitative storage-capacity benchmarking [2507.11363].

Source: https://www.emergentmind.com/topics/nd1-centers