WorksPhysics and Astronomy
Solution-phase fluorination of nanodiamond: near-surface NV activation and spin relaxation
Abstract
Nanodiamonds hosting luminescent point defects, known as fluorescent nanodiamonds (FND), are a leading platform for quantum technology. The nitrogen-vacancy (NV) centre is the most intensively studied of these; its negatively charged state (NV) can serve as a qubit at room temperature, and its stability is governed by surface functional groups. We present two solution-phase fluorination routes for stabilising NV: direct C-F bond formation by decarboxylation with xenon difluoride via a radical mechanism, and the Balz-Schiemann reaction, which replaces surface amino groups with fluorine. The two routes were compared by infrared, X-ray photoelectron, energy-dispersive X-ray, Raman and photoluminescence spectroscopy. Both gave a high NV fraction, up to 90% on average and approaching 100% in fluorine-rich regions, which to our knowledge is among the highest reported for surface-terminated nanodiamonds of this size and, in particular, for fluorine termination. Frequency-domain relaxometry shows that the fluorinated particles retain a long spin-lattice relaxation time, and s for the XeF and Balz-Schiemann routes, several times the values reported for commercial HPHT nanodiamonds, although shorter than the s of the as-received material. Charge-state stability and spin lifetime therefore do not improve together: fluorination activates near-surface NV centres, which are the most exposed to surface noise but also the ones that dominate relaxometric sensing.
Introduction
The nitrogen–vacancy (NV) centre in diamond is among the most extensively studied point defects in the solid state. It consists of a substitutional nitrogen atom adjacent to a carbon vacancy; capture of an additional electron yields the negatively charged state, [1, 2]. The centre has attracted particular interest as a versatile platform for quantum sensing and other quantum technologies, largely because it possesses long-lived, optically addressable spin states and can therefore operate as a qubit even at room temperature [1, 3, 4]. Its electron spin can be initialised and read out optically by means of optically detected magnetic resonance (ODMR), providing a convenient interface between photons and single spins [3]. These capabilities underpin a broad range of quantum technologies—from nanoscale magnetometry, thermometry, and electrometry to single-molecule sensing and quantum-information processing [5, 6, 7].
All of these functions presuppose that the centre remains in its negative charge state, since lacks the spin-dependent fluorescence contrast on which optical readout relies. In practice, the charge state fluctuates due to the photoionisation of NV to NV under the very illumination used for spin initialisation and readout [1, 6]. A stable NV population is therefore a basic prerequisite for reliable operation. Equally important is the spin-coherence time, which sets the duration over which the spin retains phase information and thereby bounds the attainable sensitivity of any coherent measurement. Together, charge-state stability and long coherence times determine the performance of NV centres in many applications [5, 8, 9].
Magnetic fields, temperature, and electric fields are most often measured through changes in the ODMR spectrum [10]. Single centres can now resolve magnetic fields at the nT scale [11] and enable magnetic imaging with nm-scale spatial resolution [12]; temperature has been measured with sensitivities of mK Hz in bulk diamond (resolving variations as small as 1.8 mK, including inside living cells) and mK Hz in nanodiamonds [13]; and a.c. electric fields have been detected with a sensitivity of V cm Hz [14]. The charge state itself can also serve as the sensing signal because the NV/NV ratio can respond to the local electrostatic and chemical environment [15, 16, 17], Charge-state-based sensing has been used, for instance, to monitor the local electrochemical potential, with nanodiamond NV ensembles resolving potential swings as small as mV [15]. Finally, changes in report on nearby fluctuating spins, such as paramagnetic free radicals, and it can be detected all-optically with a simple initialise–wait–read fluorescence sequence, with no microwave driving required [8, 18, 19], well suited to biosensing. It has provided magnetic imaging of gadolinium ions with sub-cellular resolution at a sensitivity corresponding to statistically polarised spins [20], quantification of free radicals generated in chemical reactions and in single living cells and embryos [18, 21], and, when combined with responsive surface coatings, even pH measurements over the range 3–7 [9].
The sensitivity of all of these sensing schemes, however, depends critically on the proximity of the NV centre to the analyte. The dipolar field of a point-like magnetic target decays as with the separation , therefore, the relaxation rate it induces on the NV spin—which scales with the square of the coupling—falls off as [19, 5]. Accurate devices therefore require shallow centres residing within a few nanometres of the surface. Unfortunately, the very susceptibility to external perturbations that makes the NV centre—and any colour centre with comparable properties—such an excellent sensor also renders shallow centres unreliable. Dangling bonds and sp-like carbon defects at and near the surface constitute an electronic spin bath whose magnetic noise shortens both the coherence time and the relaxation time [22, 23, 24]; fluctuating surface charges generate electric-field noise that causes additional dephasing [25]; and electron-trapping surface states pin the Fermi level and induce upward band bending, which depopulates NV in favour of NV in the near-surface region [26, 27, 28]. Consequently, NV stability is strongly governed by surface chemistry and crystallographic orientation [29].
Considerable effort has therefore been devoted to engineering the diamond surface—of bulk crystals and nanodiamonds (NDs) alike—to improve the spin coherence and charge stability of shallow defects while retaining functionality for applications. Diamond surfaces are routinely cleaned in strongly oxidising acid mixtures, which graft oxygen-containing functional groups, most notably carboxyl (COOH) groups, that provide convenient moieties for further chemistry but simultaneously reshape the interfacial band structure [29, 30]. Indeed, the usual carboxyl, hydroxyl, and hydrogen terminations tend to destabilise NV through band bending and unfavourable electron-affinity effects [29, 26, 31]. Since the as-oxidised surface is thus far from optimal, numerous strategies for its further modification are being pursued.
Among these strategies, nitrogen-terminated surfaces have attracted growing interest owing to their ability to suppress electron spin noise and to promote NV charge-state stability [29, 32, 33, 34, 35]. Nitrogen termination has been achieved both by physical routes, such as plasma treatment [33, 36, 37, 38], and by wet-chemical routes [34, 39, 40, 41, 42, 43]. Plasma methods, however, can reduce the NV concentration and co-introduce oxygen [36, 37]. Intriguingly, such mixed N/O terminations can themselves further enhance NV stability [33, 42, 44, 45].
Beyond covalent termination, encapsulation in inorganic shells offers a complementary route: silica coatings and AlO layers passivate the surface and stabilise the NV charge state [46, 47]. Such surface engineering has recently extended the achievable of NDs substantially, reaching 4.7 ms in 100 nm NDs [48], ms after molten KNO oxidation [49], and 1 ms in 50 nm to 60 nm NDs upon silica coating [46].
Although many terminations have been examined experimentally and theoretically, fluorine termination is widely regarded as one of the most favourable, because its positive electron affinity (PEA) strongly stabilises the NV charge state [29, 44, 50]. Fluorine plasma treatment increased the stability of shallow ( nm) NV centres approximately four-fold, from 0.11% to % [51], and SF-plasma treatment proved essential for stabilisation nitrogen-delta-doped NV centres residing within 5 nm of the surface [52]. Consequently, the development of efficient yet mild fluorination procedures is of considerable interest.
The formation of C–F bonds, however, ordinarily demands highly reactive reagents, such as F or ClF, or otherwise harsh conditions, and controlled fluorination of diamond therefore remains a challenge. Owing to its extreme reactivity, fluorination with F requires specialised equipment and careful process control [53, 54, 55]. Even though, ClF provides even stronger fluorinating capability while enabling controlled C–F bond formation on diamond surfaces without significant structural degradation [53, 56], it is still far from being a safe reaction alternative. To circumvent the handling difficulties of these aggressive reagents, milder electrophilic fluorinating agents have been developed. The most prominent are Selectfluor (F-TEDA-BF; 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate)), that delivers electrophilic fluorine under mild conditions [57], and N-fluorobenzenesulfonimide (NFSI), a N–F reagent of attenuated reactivity that is valued for selective fluorinations [58]. In particular, Rodgers et al. introduced a gentle, visible-light-driven C–H activation route that fluorinates the diamond surface under ambient conditions while preserving the coherence of NV centres lying within 10 nm of the surface [59]. Moreover, a carboxyl-selective, silver-catalysed Hunsdiecker–Borodin-type reaction employing Selectfluor was shown to increase the NV fraction of the total NV population in NDs [60].
Despite these efforts, reports in which efficient fluorination is accompanied by a substantial improvement of NV properties remain scarce. For nanodiamonds, solution-phase fluorination has so far afforded only moderate gains in the NV fraction [60], while the photochemical route of Rodgers et al. preserved—but did not extend—the coherence of shallow centres in bulk diamond [59]. To the best of our knowledge, a fluorination procedure that simultaneously achieves high surface coverage and a marked improvement of both the charge-state stability and the spin relaxation of NV centres in nanodiamond has not yet been reported.
We previously reported an effective nitrogen termination route of nanodiamonds through the Hofmann degradation, which additionally reduces the sp carbon content and thereby affords a significant improvement in charge stability [43, 38, 45]. While the mixed N/O termination and the diminished graphitic content already improve the NV properties, the amino termination offers a further advantage: it constitutes a convenient functional group for surface fluorination under reaction conditions far milder than those required for oxygen-terminated surfaces. Nevertheless, the development of other fluorination reaction paths remains important.
In this work, we exploit two complementary solution-phase strategies to fluorinate nanodiamond surfaces and compare them directly: the Balz–Schiemann reaction via diazotisation of the surface amino groups and the fluorination with xenon difluoride (XeF). Xenon difluoride is a particularly attractive fluorinating agent, often showing higher reactivity than Selectfluor or NFSI depending on the reaction conditions [61, 62, 63]. Diamond fluorination with XeF is generally attributed to dissociation into highly reactive fluorine-containing species [64, 65], whereas reactions in solution frequently proceed via a single-electron transfer mechanism [61, 66, 67]. Both routes achieve a high fluorine surface content under comparatively mild conditions. We characterise the resulting fluorine-terminated nanodiamonds as a function of fluorine incorporation and examine the stability of the NV charge state in detail, showing that fluorination markedly increases the NV fraction and prolongs the spin–lattice relaxation time (), thereby improving the properties most relevant to quantum-sensing applications.
Methods
synthesis. The Balz–Schiemann (BS) route. HPHT diamond particles with a diameter of 140 nm and NV content of 3 ppm were purchased from Adámas Nanotechnologies. Amino-terminated diamond was synthesized as follows. From suspension (1 mg mL), 1 mL was transferred to a flask and evaporated to dry, and 5 mL of SOCl (Sigma Aldrich) was added. The mixture was sonicated in an ultrasonic bath at 37 kHz for 3 hours at 50 °C. The solvent was then evaporated under vacuum at 50 °C. Subsequently, 6 mL of an NH/dioxane mixture (2 mL of 0.4 M NH/dioxane diluted in 4 mL of anhydrous dioxane) was added under an argon flow and sonicated under the same conditions. The dioxane mixture was removed under vacuum at 50 °C and the nanoparticles were redispersed in 6 mL of water, which was followed by sonication at 37 kHz for 2 hours. The resulting amide intermediate was heated to 80 °C, and 0.5 g NaOH was added to mixture (VWR) and 0.4 mL Br (Sigma Aldrich) was added dropwise. The solution was stirred at 80 °C for 15 minutes and then sonicated at 80 °C for 10 minutes. Excess bromine was neutralised with ammonia solution, and the sample was purified by centrifugation until the pH reached approximately 8. For diazotisation, 200 μL of HF/pyridine (70/30%, Sigma-Aldrich) was added dropwise to the slightly alkaline suspension in a polypropylene tube and sonicated for 10 minutes at room temperature. Saturated NaNO (VWR) solution was added dropwise, keeping the temperature between 0 °C to 2 °C in an ice bath, with brief ultrasonic exposure for short 10-second periods. Excess NaNO₂ was detected using filter paper soaked in KI/starch. After the mixture reached room temperature, the tube was immersed in boiling water, resulting in a solution temperature of 87 °C to 88 °C for 20 minutes, then it was sonicated at the same temperature for 10 minutes. The cooled mixture was purified by centrifugation. Sample prepared by the described reaction route was labeled from now on as ND_NH₂-F.
The XeF₂ route. 1 mL of ND suspension was evaporated to dryness, placed in a Teflon crucible with 100 mg of XeF₂ (Thermo Fisher Scientific), and exposed to an argon stream. Subsequently, 1 mL of CDCl₃ (Sigma Aldrich) and 350 µL of HF/pyridine complex (Sigma Aldrich) were added. Deuterated chloroform was used to avoid the formation of C-H bonds. In the presence of non-deuterated solvents, the possible side-reactions decrease the yield of C-F bonds [67, 61, 62]. The crucible was sealed in a bomb tube and heated in a preheated jacket at 100 °C for 70 hours. The diamond was gradually transferred to the aqueous phase by adding isopropanol/water and re-evaporating in a warm ultrasonic bath. By-products were removed by repeated centrifugation.
FTIR. A Bruker Tensor 37 spectrometer equipped with an MCT detector was used for infrared spectroscopy between 400–4000 cm⁻¹ with 4 cm⁻¹ resolution. The sample concentration was 0.2 mg mL⁻¹ after purification, and 100 µL was drop-cast onto a Si plate to perform measurements at room temperature, under N₂ flow.
XPS. X-ray photoelectron spectroscopy was carried out with an Escalab Xi+ instrument (Thermo Fisher Scientific, Waltham, MA, USA). A microfocused monochromatic Al Kα X-ray gun was used; the diameter of the analysed area was 200 µm. Charge compensation was applied. The base pressure of the analysis chamber was mbar. Peak fitting was performed using CasaXPS version 2.3.19 rev1.0m, with Shirley background subtraction and Gaussian–Lorentzian GL(30) peak shapes.
EDS. Fluorine distribution was measured by energy-dispersive X-ray spectroscopy (EDS). The spectra were recorded with a TESCAN MIRA3 electron microscope equipped with an Element EDS system. The applied accelerating voltage was 8 kV. The relatively low voltage minimized fragmentation, ensured a small information depth below the particle surface, and it was yet sufficiently large to detect the K lines of all light elements.
PL and Raman. A Renishaw inVia Raman Microscope was used to record photoluminescence (PL) and Raman spectra with a 50× Leica objective. The excitation wavelength for PL measurements was 532 nm between 535–900 nm, while Raman spectra were recorded between 1053–2081 cm⁻¹ using 325 and 785 nm laser sources. The resolution was 1 cm⁻¹. These measurements were done on the same samples prepared for FTIR and EDS.
ODMR. For ODMR measurements, the samples were evaporated onto a 150 µm-thick glass plate. CW-ODMR measurements were performed using a home-built optical setup based on a Newport integrating sphere (IS) 819C-IS-5.3 (IS). The sample was placed on a coplanar waveguide antenna and connected to the bottom port of the IS. A 520 nm fibre-coupled laser from Roithner, delivering the excitation to an about 6 mm diameter spot through a 550 nm dichroic mirror (Thorlabs DMLP550T), served as an excitation source and was connected to a side port of the IS. The CW-ODMR spectra were recorded at a laser power of 220.5 mW, corresponding to a power density of approximately 0.78 W cm⁻² assuming uniform illumination. The radiation from the sample was collected through a biased silicon-based free-space photodetector (Thorlabs DET100A2) using a 700 nm longpass hard-coated filter (Thorlabs FELH0700) and a 645 nm longpass coloured-glass filter (Thorlabs FGL645M), and connected to the top port of the IS. The photocurrent from the detector output was converted into voltage using a 1 MHz bandwidth transimpedance amplifier (TIA) (Femto DHPCA-100) and connected to the input of a lock-in amplifier (Anfatec USBLockIn250). The microwave (MW) field near the ground-state spin resonance frequency (2870 ± 70 MHz) was generated by a MW generator (Vaunix LabBrick LSG-402) in conjunction with a high-power amplifier (Mini-Circuits ZHL-25W-63+) and applied to the sample through a coplanar waveguide terminated with a high-power (100 W) 50 Ω termination. The microwave excitation amplitude was modulated using a high-speed switch (Mini-Circuits ZASWA-2-50DRA+) and controlled by a TTL signal from the internal reference source of the lock-in amplifier. The microwave power applied for the CW-ODMR spectra was −36 dBm (251 nW). The spin-lattice relaxation time was determined with the same CW-ODMR setup by frequency-domain relaxometry, following the protocol of Verkholyuk et al. [68]. Instead of a pulsed initialise–wait–read sequence, the amplitude of the microwave field was modulated and the lock-in response was recorded while sweeping the modulation frequency from about 10 Hz to 20 kHz. The response passes through a maximum at a characteristic modulation frequency that is governed by the relaxation dynamics of the spin ensemble. The microwave generator was set to dBm, corresponding to 25 dBm at the sample after the amplifier, and was determined at four laser powers spanning 142 mW to 354 mW (approximately 0.50 W cm to 1.25 W cm). Identical microwave conditions were used for all three samples. Because continuous optical illumination itself contributes to the measured relaxation, was plotted as a function of laser power and fitted with a straight line; the intercept of this extrapolation to zero laser power yields the intrinsic relaxation time as . All values quoted below are such zero-power extrapolations. The approach retrieves over more than three orders of magnitude and remains applicable to inhomogeneous ensembles such as nanodiamond powders, while offering a substantial speed-up over pulsed relaxometry [68].
Results and discussion
Synthesis and chemical characterization
The processes of the two reactions are shown in Figure 1, which also schematically presents the surface.

Figure 1. (a) The fluorination reactions: the Balz–Schiemann (BS) reaction and the XeF-based route. (b) Schematic structure of the surface-modified diamond.
FTIR. The most noticeable change in the FTIR spectra after both fluorination reactions (Figure 2) is the appearance of new absorption bands at 1255 cm and 1265 cm for ND_NH-F and ND_XeF, respectively, together with a new feature around 800 cm in both samples. The positions of the former bands agree closely with the C–F stretching vibrations reported for fluorinated diamond materials [69, 70, 71, 55]. The deformation modes of CF groups, by contrast, are found at considerably lower wavenumbers in molecular fluorocarbons and fluoropolymers (typically –650 cm) [72]; the feature near 800 cm observed here is nevertheless consistent with C–F deformation modes stiffened by the rigid diamond lattice, in line with the systematic upward shift of the diamond-bound stretches relative to their fluoropolymer counterparts [70].

Figure 2. FTIR spectra of the two fluorinated and the as-received COOH-terminated samples. The different carbonyl species are indicated in a wide range between the vertical lines.
The applied fluorination reactions also decreased the intensity of the O–H stretching bands, the C=O stretching band, and the C–H stretching region was visibly modified, all of which indicate a substantial conversion of the pre-existing surface groups. A detailed band-by-band annotation of the 1000–1400 cm region is inherently difficult, because the C=O, C–O–C, and C–OH vibrations of residual oxygen-containing groups overlap with the CF stretching manifold in this window [69]. Notwithstanding this congestion, a distinct intensity increase around 1100 cm was apparent in both samples. This is consistent with the low-wavenumber component reported at 1092–1096 cm for fluorinated diamond surfaces [70, 56]. In aqueous media, the fluoride ion is strongly hydrated; therefore, HF behaves as a weak nucleophile only [73], so it is not suitable for effective replacement of surface OH groups. Previous studies have shown that anhydrous HF alone does not result in significant fluorination on the diamond surface [56]. Accordingly, the role of the pyridine–HF system employed in the present work is not to serve as a fluorine source but rather to act as a catalyst and a solvent for XeF [61, 63]. Finally, the absorption band at 1658 cm can be assigned to the pyridine/pyridinium hydrogen fluoride complex employed in the reaction; however, contributions from residual carboxyl groups and C=C skeletal vibrations cannot be excluded [69, 72, 74, 75].
XPS. The high-resolution C 1s and F 1s XPS spectra (Figure 3) also indicate that the applied reactions created C–F bonds on the nanodiamond surface. The C 1s spectrum of the as-received sample was dominated by the carbon peak at around 285 eV, accompanied by a small contribution at 284 eV. A separation of eV is consistent with previous reports on nanodiamond [76, 77]. The as-received sample also exhibited C–O and C=O components. After either surface modification, an additional component appeared at around 288.2 eV, which can be assigned to carbon in C–F bonds of appreciable ionic (semi-ionic) character [64, 78, 77]. A more pronounced change, however, was the growth of the component: the : peak-area ratio decreased from in the as-received material to in \mathrm{ND\mathchar`-XeF}_{2} and to in \mathrm{ND\mathchar`-NH}_{2}\mathchar`-F. We attribute the high content of \mathrm{ND\mathchar`-NH}_{2}\mathchar`-F to a by-product that could not be removed by the applied purification protocol. \mathrm{ND\mathchar`-NH}_{2}\mathchar`-F was washed repeatedly with water—each cycle consisting of centrifugation, removal of the supernatant and redispersion, which corresponds to an approximately tenfold dilution—and monitored by XPS after each washing cycle; no significant changes were observed in the monitored spectral windows. The presence and significance of this carbon will be discussed below. Here we merely note that the C 1s components appeared at identical positions in both samples, with varying relative weights but with unambiguous covalent character, corroborating the FTIR evidence of C–F bond formation at the surface. In the F 1s spectra, both samples exhibited the same two components, at 688.2 eV and 686 eV, binding energies characteristic of covalent and semi-ionic C–F bonding environments, respectively [64, 78, 79]. The quantity of the measured elements is indicated in Table 1.

Figure 3. XPS spectra of the two fluorinated and the as-received COOH-terminated samples.
| Sample | C (at.%) | F (at.%) | O (at.%) | Si (at.%) | N (at.%) | F:C ratio |
| COOH | 91.6 | 0 | 8.1 | 0.3 | 0 | 0 |
| XeF | 66.3 | 2.6 | 16 | 15 | 1 | 3.9 |
| NH F (BS) | 79.8 | 0.7 | 12 | 6.4 | 0.3 | 0.9 |
Table 1. F, C, O, Si and N content (at.%) determined by XPS for the as-received sample and for the F-terminated samples prepared by the XeF and BS methods.
It should also be mentioned that the absolute positions of the recorded photoelectron peaks cannot be determined free of doubt: highly insulating diamond particles are subject to charging and, in particular, to non-uniform (differential) charging, even when charge compensation with a low-energy electron flood gun is applied [80, 81]. Nevertheless, a single rigid shift of the binding-energy scale—chosen such that the diamond C 1s component falls at 285 eV—simultaneously brought the F 1s, O 1s, and N 1s lines into their expected binding-energy ranges. The fact that one common correction suffices for all elements indicates that the corresponding species reside on the same, electrically coupled surface.
According to XPS, \mathrm{ND\mathchar`-XeF}_{2} incorporated markedly more fluorine than \mathrm{ND\mathchar`-NH}_{2}\mathchar`-F, with measured F concentrations of 2.6% and 0.7%, respectively. To translate these values into surface coverages, we assume spherical particles and a C 1s information depth of –7 nm. The lower F content in the BS reaction can be explained by several factors. The main reason, which we have already seen in our earlier experiments, is that the Hofmann degradation is not
complete: unlike SiC, there are significantly more unreacted COOH groups according to the FTIR [38, 43]. Another important factor is the stability of the diazonium salt. Halliwell and Nyburg pointed out that the counterion influences its stability [82]. For example, diazonium salts containing smaller anions are more susceptible to hydrolysis, resulting in the formation of phenols rather than the corresponding haloarenes. The decrease in the carbon content can be attributed partly to surface reactions with carbon atoms and partly to the reduced coverage of the silicon substrate. The reduced coverage explains the higher Si content. This difference is clearly evident in the XPS C 1s spectra, whereas it can only be inferred from the FTIR data. Both fluorinated products exhibit a substantial increase in the proportion of -hybridized carbon. The lower carbon content can be attributed to the lower coverage of the Si plate. It should be noted, however, that XPS is highly surface-sensitive, and the measured composition may therefore overrepresent surface-enriched species. More than two-thirds of the signal originates from the top 3 nm depth layer, while nearly 80% comes from the top 5 nm. The observed increase in carbon is likely more dominant at the surface and may not be fully representative of the complete crystal [83, 84]. Removing a single-nm layer can result in a signal reduction of up to 50% [85]. Ducrozet and co-workers aimed to coat a diamond with an C layer like an onion skin; although the XPS signal was significant, the actual shell was only a single layer [86]. Based on the measured F concentration (2.6% and 0.7%), and assuming an XPS information depth of 5 nm to 10 nm and a spherical shape for the particles, the fluorine coverage is estimated to be approximately 20% to 40% of a monolayer after and 5% to 10% after the BS reaction.
Raman. The mild nature of the fluorination procedures and the preservation of the diamond structure were verified by Raman spectroscopy. This is a powerful method for investigating the allotropic forms of carbon due to its high sensitivity to the hybridization state of carbon atoms. Furthermore, the use of excitation lasers with different wavelengths allows discrimination between amorphous carbon that is structurally associated with the diamond lattice and amorphous carbon present only at the particle surface [87]. The Raman spectra are shown in Figure 4. The characteristic diamond peak is visible in all samples at 1335.5 cm, and the G band is at 1613 cm. We applied a UV laser because it clearly shows the diamond peak and it is possible to determine the integrity of the diamond lattice. Furthermore, the G band is visible in the UV Raman spectra as well. The latter can easily shift as a result of doping or strain [88].

Figure 4. Raman spectra of the COOH-terminated sample and the two products after fluorination. The spectra in the top row were measured with a 325 nm laser, the bottom row with a 785 nm laser.
Following the treatment, only a slight increase in the intensity of the G band is observed. In contrast, the sample prepared via the BS reaction exhibits a much more pronounced increase. However, the strong enhancement should not be interpreted as a direct measure of the amorphous carbon content, because the adsorbed aromatic compounds (e.g. pyridine) may contribute to it via resonance-excitation with 325 nm laser [89, 90]. To make sure whether the resonance effect caused the increase, the Raman spectra were recorded with 785 nm laser. The sample prepared by BS showed increase, but it is not as significant as measured with the UV laser. This result confirms the assumption that the large G-band can be explained in part by the resonant effect [87, 91].
Table 2 shows the half-width values of the diamond peaks determined with the UV laser. It confirms that the reactions did not cause any significant damage to the lattice structure.
| Sample | diamond band (cm) | half-width (cm) |
| COOH | 1335.5 | 10.4 |
| XeF | 1335.5 | 9.6 |
| NH F | 1335.3 | 10.8 |
Table 2. Measured positions and full widths at half maxima of the diamond band ( nm).
The sp carbon in ND_NH-F is thus best described as small aromatic islands rather than graphitic shells. The persistence of a prominent diamond line implies a thin, partial surface coverage on intact crystalline cores, in consistence with the sp dominated C 1s XPS spectrum and the absence of related vibration modes in the FTIR data. As a result of the Raman measurements, we could confirm the mild fluorination.
EDS analysis. The fluorine content was quantified from XPS survey spectra and from SEM-EDS measurements as well. The two techniques are complementary in their sampling: XPS averages over a spot of diameter and over only the outermost few nanometres, whereas EDS maps the lateral fluorine distribution across the sample with micrometre-scale resolution and a correspondingly larger, micrometre-deep information volume [92]. The average F content in the homogeneous distribution at the edge of the sample is 4.6%, while in the interior of the sample on the map—where a bright green spot indicates a high F content—the average F content is 4.7%. These values are high for an F monolayer in themselves, but the EDS information depth is much greater, and here this signal comes from aggregates of several micrometres, which thus shows a high coverage of the surface and confirms that particles with F-rich surfaces aggregate more easily with each other (due to their fluorophilic nature [93]). The element map confirms that fluorination affects oxygen-containing groups; where there is more F, there is less oxygen.
The average fluorine content is 4.6 % in the homogeneously covered edge region of the sample and 4.7 % in the interior but inhomogeneously in patches in the more hydrophobic environment.
Optical properties
PL. Despite the lower fluorine content and the sp carbon content of ND_NH-F, both fluorination routes produced a substantial increase in the measured NV fraction and its photosability. The NV/NV ratio was determined from photoluminescence (PL) spectra recorded at laser power densities spanning two orders of magnitude under 532 nm excitation.
The PL spectrum of an NV-containing sample comprises the emission of both charge states, and their contributions were quantified following the decomposition procedure of Ref. [94], in which the spectrum, normalised to its total intensity, is expressed as a linear combination of the two charge-state components [7, 38, 45]. Because the organic residue on sample (2) also contributes to the PL spectrum, the model function was extended by a background term and reads
where and are the reference emission spectra of the two charge states [95], and are the corresponding weighting coefficients, is a Gaussian with its position fixed below 600 nm, and accounts for the residual baseline and non-specific scattering. For the as-received material and ND_XeF, no background component was required; setting reduces (1) to the two-component decomposition of Refs. [94, 7, 38, 45]. The fits were performed in MATLAB. A representative fit is shown in fig:7a; as shown in fig:7b, the background component diminished with increasing laser power. From the fitted components, the NV fraction was calculated as
where and denote the integrated intensities of the respective weighted components.
The NV fractions obtained for the two fluorinated samples and the as-received material are shown in Figure 6 and listed in Table 3. Since fluorine resides on the ND surface and the EDS maps indicated some lateral inhomogeneity, the data are presented in two panels: Figure 6 shows the measurement points distributed across the sample, while Figure 6 shows measurements taken at the fluorine-rich locations identified by EDS (Figure 5). At all applied powers, the NV fractions of the fluorinated samples exceeded those of the as-received material. The two data sets also exhibited opposite power dependences: in the as-received sample the fraction decreased monotonically with increasing power (from 0.61 to 0.40), whereas in the fluorinated samples it increased monotonically and saturated near 0.90 at the highest powers. The two fluorination routes yield identical fractions within the measurement uncertainty at every power, although the fluorine coverage differs by a factor of about four. At the fluorine-rich locations (Figure 6), the fractions were higher still, reaching 0.9 at some points even at the lowest applied power.

Figure 5. EDS analysis of the fluorinated sample. Left: SEM image at the sample edge with the line-profile scan taken along the red line (30 points). Right: elemental maps of C, Si, F, and O recorded in the interior of the sample.

(a)

(b)
Figure 6. NV fraction as a function of laser power density for the fluorinated samples and the COOH-terminated starting material. (a) Average measurement points across the sample. (b) Measurements at the fluorine-rich locations identified by EDS.
| power density () | COOH | NH F | XeF |
| 171 | 0.61 (0.10) | 0.73 (0.08) | 0.76 (0.17) |
| 308 | 0.58 (0.11) | 0.80 (0.09) | 0.85 (0.08) |
| 1364 | 0.50 (0.15) | 0.86 (0.11) | 0.88 (0.07) |
| 2465 | 0.48 (0.15) | 0.89 (0.08) | 0.89 (0.08) |
| 14406 | 0.40 (0.17) | 0.90 (0.06) | 0.88 (0.09) |
Table 3. Applied laser power densities and the measured NV fractions.
The photostability of the NV state was further examined through optically driven charge cycling. The two charge states coexist in a dynamic, illumination-dependent equilibrium [96, 6, 7]. Interconversion proceeds by two-photon processes in both directions: ionisation of occurs when a first photon excites the centre and a second photon promotes the excited electron to the conduction band, whereas recombination to requires excitation of followed by electron capture from the valence band [6]. Under red illumination (633 nm), is still photoionised, but the reverse process is inefficient because absorbs only weakly at this wavelength; prolonged red illumination therefore shifts the population towards . Subsequent green illumination drives both processes and re-establishes the -rich steady state [6, 97]. The extent to which the population is depleted during red illumination, and recovered afterwards, thus provides a measure of the charge-state photostability.

Figure 7. (a) Representative spectral decomposition according to Eq. (1). (b) Decrease of the luminescent background with increasing laser power.
The PL spectra recorded before and after 10 min of 633 nm (3408 ) illumination are shown in Figure 8. During this period of time, we followed the intensity change at the ZPL. In both cases, it decreased somewhat over time, but it never became completely 0. In the as-received sample, the fraction decreased from 0.15 to 0.07 upon the discharge-recharge cycle at a power density of 3408 . Since oxygen-containing groups also possess a certain degree of stability effect, the centres may be more stable at those points or less stable, where the destabilising groups (e.g. COOH and H) are numerous. With this measurement, we wanted to demonstrate the high stability caused by the F termination. We compared this with a point in the as-received sample. In the fluorinated sample, by contrast, the fraction remained unchanged within the measurement uncertainty.

Figure 8. Discharge–recharge cycle of (a) the COOH-terminated and (b) the fluorine-terminated sample.
ODMR. The spin properties of the centres were characterised by ODMR. From CW-ODMR measurement the zero-field splitting parameters and were extracted. Besides the axial parameter , which originates from the electron spin-spin interaction, the transverse parameter is often nonzero in diamond NV centers and sensitive to transverse strain and to local electric fields, including those of nearby charges [1, 14, 98]. Together, the two parameters therefore probe whether fluorination alters the crystalline and electrostatic environment of the centres. The ODMR spectra are shown in Figure 9; a schematic of the measurement system is given in Figure 10. At zero applied magnetic field, the ensemble spectra exhibit two resonances at , and the spectra were fitted accordingly (coloured lines in Figure 9). The resulting and values are listed in Table 4.

Figure 9. ODMR spectra of (a) the COOH-terminated, (b) the XeF-treated, and (c) the BS-treated (NH F) NDs. The applied microwave power was dBm and the laser power 220.5 mW. The coloured lines are the fitted curves.
| COOH | XeF | NH F | |
| 2868.8 () | 2869.1 () | 2868.8 () | |
| 6.2 () | 6.2 () | 6.0 () |
Table 4. and parameters (MHz) determined for the COOH-terminated and fluorinated NDs.

Figure 10. Schematic of the ODMR measurement setup.
Both parameters were unchanged within the measurement uncertainty across the three samples. The invariance of indicates that neither treatment introduced measurable axial strain, and the invariance of that the ensemble-averaged transverse strain and quasi-static electric-field environment of the centres is likewise unaffected.
Spin–lattice relaxation ()
The ensemble spin–lattice relaxation time was measured on the same particles by CW frequency-domain relaxometry [68], using the ODMR setup described in the Methods; the results are summarised in Figure 11 and Table 5.
| Sample | (µs) |
| COOH (as-received) | 1173 ± 123 |
| XeF | 733 ± 56 |
| NH F (BS) | 712 ± 20 |
Table 5. Ensemble NV spin–lattice relaxation time for the as-received and F-terminated samples, obtained by extrapolating the measured relaxation rates to zero laser power. All three samples were measured under identical microwave conditions (25 dBm at the sample). The quoted uncertainties are the standard errors of the zero-power extrapolation.

Figure 11. (a) Position of the maximum of the lock-in response, , as a function of laser power for the three samples, with linear fits extrapolated to zero laser power; the filled symbols on the ordinate are the intercepts with their standard errors. The as-received point at 222 mW is the mean of two repeated measurements. (b) The resulting spin–lattice relaxation times . All data were recorded at 25 dBm microwave power at the sample.
Each value is an ensemble average over the very large number of nanoparticles contained in the illuminated spot, so that the reported characterises the particle population as a whole rather than individually selected crystals. The as-received, COOH-terminated particles show μs, of the same order as the value ( μs) reported for HPHT FNDs of comparable size [49]. Contrary to what the charge-state data alone might suggest, both fluorination routes shortened , to μs for ND_XeF2 and μs for ND_NH2-F, a reduction of about 40% in both cases. The two routes agree with each other within the measurement uncertainty, mirroring their behaviour in the fraction despite the four-fold difference in fluorine coverage (Table 1). The opposite signs of the two effects are not contradictory. Fluorination stabilises throughout the particle, including centres in the near-surface region that previously remained neutral; these are also the centres most strongly coupled to the residual surface spin bath, so that their recruitment raises the measured fraction while lowering the ensemble-averaged . The adsorbed organic layer seen by FTIR and Raman, and the increased content seen by XPS, provide further relaxation channels in the same near-surface region. The values reported here accordingly remain below those reached for nanodiamonds of comparable size by treatments that remove or bury the surface spin bath rather than repopulate —chemical-vapour-deposition growth combined with electron irradiation and high-temperature annealing (4.7 ms in 100 nm CVD NDs) [48], molten-KNO3 oxidation ( ms) [49], or silica over-coating ( ms) [46].
Taking the increase measured at the lowest laser power (0.61 → 0.76) and assuming a uniform NV distribution, the converted population corresponds to an equivalent surface shell of at most 4 nm, where is the volume-to-surface ratio of the particle. The spherical approximation used here gives an upper bound, since any faceted particle has a larger surface-to-volume ratio and hence a thinner implied shell. This is of the same order as the few-nanometre depth over which upward band bending is expected to deplete near the surface [26, 27], and is therefore consistent with the near-surface origin proposed above. We stress that this is an order-of-magnitude consistency check rather than a determination of the shell thickness: it assumes a uniform NV distribution and treats the photoluminescence-weighted fraction as a number fraction.
The relaxation times retained after fluorination nevertheless remain well suited to relaxometry. Values of ms are several times longer than the ms reported for commercial HPHT fluorescent nanodiamonds [48], and lie comfortably within the range exploited for the detection of free radicals and paramagnetic ions [8, 18, 20]. More importantly, because the relaxation rate induced by an external target falls off as [19, 5], the relaxometric signal is dominated by the NV centres lying closest to the surface. Fluorination converts precisely these centres from the sensing-inactive into , and thereby increases the number of usable sensors in the outermost shell of the particle, where the coupling to an analyte is strongest. On this view the moderate reduction of the ensemble-averaged is the expected signature of having recruited this near-surface population, and is traded against a larger effective sensing volume rather than against sensing capability as such.
Conclusions
This work set out to improve the quantum-optical properties of NV centres in nanodiamond by mild surface fluorination while preserving the integrity of the diamond lattice. Two solution-phase routes were developed and compared: a Balz–Schiemann-type reaction, adapted here to convert amino-terminated nanoparticles into C–F-terminated ones, in which HF serves as the fluoride source—a highly toxic reagent, yet one far easier to handle than elemental F or ClF—and the second involved XeF in the solution phase, where hydrofluoric acid (HF) acts both as a catalyst and as a solvent for XeF, as described in the literature. The formation of C–F bonds was established independently by FTIR and XPS, with fluorine contents of 2.6% (XeF) and 0.7% (BS). Both synthesis routes provided equal NV stabilisation, with the fraction saturating near 0.9 and photostability upon prolonged 633 nm illumination; the spin–lattice relaxation time, by contrast, was shortened by about 40 % by both treatments, while remaining near 0.7 ms and thus well within the range useful for relaxometric sensing.
Notably, the two routes yield the same charge-state behaviour within the measurement uncertainty despite a roughly four-fold difference in fluorine coverage, indicating again that mixed surface termination already suffices for stabilisation. We attribute the stabilisation to the combined effect of removing acceptor-like carboxyl groups, the positive electron affinity of the – termination, and the thinner adsorbed water layer on the more hydrophobic surface. The two routes also yield the same spin–lattice relaxation time within uncertainty, so the shortening of appears to be a generic consequence of the fluorinated surface rather than of the carbonaceous residue specific to the BS product. Because the relaxometric response to an external spin falls off as , the near-surface centres that fluorination converts into are precisely those that dominate sensing, so the shorter ensemble is better read as the cost of activating a larger and more strongly coupled sensor population than as a loss of performance; direct relaxometric measurements on an external target would be required to confirm that this trade is favourable in practice. Although not yet fully optimised, the BS method already matches the treatment in charge-state performance using bench-scale diazonium chemistry, and offers clear room for improvement through refined purification and reaction control. A remaining consideration is the increased hydrophobicity of fluorine-terminated surfaces, which lowers colloidal stability in aqueous media and is therefore relevant for biological applications; it can, however, be mitigated by secondary functionalisation or mixed terminations, and it is of little consequence for dry and solid-state sensing configurations. Taken together, these results establish mild, solution-phase fluorination as a simple and scalable route to charge-stable NV ensembles in nanodiamond, and motivate its extension to shallow NV centres in bulk diamond, where surface noise is most limiting.
Finally, we wish to emphasise the importance of the thorough characterisation that should accompany every diamond surface modification. Raman microscopy remains the fundamental tool for detecting, and even quantifying, graphitic carbon, and is therefore the natural benchmark of a not merely productive but clean surface reaction. Low-photon-energy Raman excitation, however, fails to detect condensed sp² phases whose optical gap exceeds the excitation energy: under visible and near-infrared excitation, the BS-modified nanodiamond showed no trace of sp² carbon whatsoever. Likewise, had we followed the routine XPS protocol of assigning the dominant C 1s component to sp³ diamond carbon, the binding-energy scale would have been miscalibrated, the spectra would still have indicated covalent C–F formation, and the extensive fluorine-passivated sp² layer would have gone unnoticed. Our own set of characterisation methods is certainly not exhaustive either; the present case nevertheless illustrates that conclusions drawn from any single, routine protocol can be qualitatively misleading, and that only mutually cross-validating measurements yield a self-consistent picture of the modified surface. We therefore advocate that studies of diamond surface chemistry report such orthogonal characterisation as a matter of course.
CRediT authorship contribution statement
Szabolcs Czene: Conceptualization, Methodology, Writing – original draft. Olga Krafcsik: Methodology, Writing – reviewing – review & editing. Nikoletta Jegenyés: Methodology, Writing – original draft. Dávid Beke: Writing – review & editing. László Péter: Methodology, Writing – review & editing. Vladimir Verkhovlyuk: Methodology, Writing – original draft. Ádám Gali: Conceptualization, Funding acquisition, Supervision, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
Data will be made available on request.
Acknowledgements
The authors thank Anton Pershin for valuable advice on the frequency-domain relaxometry measurements. This work was supported by the EUREKA project quNV2.0 (2021-17465/NP/BILAT_HU_DE_1, 2023–2026) via the National Research, Development and Innovation Office (NKFIH) under Grant No. 2020-1.2.3-EUREKA-2022-00022. The research reported in this paper and carried out at the Wigner Research Centre for Physics is supported by the infrastructure of the Hungarian Academy of Sciences.
Frequency-domain relaxometry spectra
Figure A.12 shows the CW-ODMR lock-in response as a function of the microwave amplitude-modulation frequency for the three samples at each of the laser powers used for the determination. In every sample the maximum shifts to higher modulation frequency as the laser power is increased, reflecting the additional, optically induced contribution to the measured relaxation rate; this power dependence is what the extrapolation to zero laser power removes. Between 14 and 35 scans were averaged per trace, and the sweeps start above 35 Hz so as to exclude the narrow interference spikes that dominate the response at lower modulation frequencies. The position of the maximum was obtained by fitting a parabola to the response in over decades about the maximum; the resulting is stable to within a few Hz against the choice of this window.

Figure A.12. Lock-in signal versus microwave amplitude-modulation frequency for (a) the as-received COOH-terminated, (b) the XeF-treated, and (c) the BS-treated (NH) nanodiamonds, at the laser powers used for the determination and a microwave power of 25 dBm at the sample. Each trace is normalised to its own maximum.
References
- [1]M. W. Doherty, N. B. Manson, P. Delaney, F. Jelezko, J. Wachtrup, L. C. L. Hollenberg, The nitrogen-vacancy colour centre in diamond, Physics Reports 528 (1) (2013) 1–45. doi:10.1016/j.physrep.2013.02.001.DOI
- [2]Á. Gali, Ab initio theory of the nitrogen-vacancy center in diamond, Nanophotonics 8 (11) (2019) 1907–1943. doi:10.1515/nanoph-2019-0154.DOI
- [3]A. Gruber, A. Dräbenstedt, C. Tietz, L. Fleury, J. Wrachtrup, C. von Borczyskowski, Scanning confocal optical microscopy and magnetic resonance on single defect centers, Science 276 (5321) (1997) 2012–2014. doi:10.1126/science.276.5321.2012.DOI
- [4]F. Jelezko, T. Gaebel, I. Popa, A. Gruber, J. Wrachtrup, Observation of coherent oscillations in a single electron spin, Phys. Rev. Lett. 92 (7) (2004) 076401. doi:10.1103/PhysRevLett.92.076401.DOI
- [5]R. Schirhagl, K. Chang, M. Loretz, C. L. Degen, Nitrogen-vacancy centers in diamond: Nanoscale sensors for physics and biology, Annual Review of Physical Chemistry 65 (2014) 83–105. doi:10.1146/annurev-physchem-040513-103659.DOI
- [6]N. Aslam, G. Waldherr, P. Neumann, F. Jelezko, J. Wrachtrup, Photo-induced ionization dynamics of the nitrogen vacancy defect in diamond investigated by single-shot charge state detection, New J. Phys. 15 (1) (2013) 013064. doi:10.1088/1367-2630/15/1/013064.DOI
- [7]G. Thalassinos, et al., Robust quantification of the diamond nitrogen-vacancy center charge state via photoluminescence spectroscopy, APL Photonics 10 (10) (2025). doi:10.1063/5.0284237.DOI
- [8]S. Fan, et al., Quantum sensing of free radicals in macrophages reveals early autophagy-lysosome regulation in an atherosclerosis cell model, Redox Biology 87 (2025) 103825. doi:10.1016/j.redox.2025.103825.DOI
- [9]T. Fujisaku, R. Tanabe, S. Onoda, R. Kubota, T. F. Segawa, F. T.-K. So, T. Ohshima, I. Hamachi, M. Shirakawa, R. Igarashi, ph nanosensor using electronic spins in diamond, ACS Nano 13 (10) (2019) 11726–11732, pMID: 31538479. arXiv:https://doi.org/10.1021/acsnano.9b05342, doi:10.1021/acsnano.9b05342. URL https://doi.org/10.1021/acsnano.9b05342
- [10]L. Rondin, J.-P. Tetienne, T. Hingant, J.-F. Roch, P. Maletinsky, V. Jacques, Magnetometry with nitrogen-vacancy defects in diamond, Reports on Progress in Physics 77 (5) (2014) 056503. doi:10.1088/0034-4885/77/5/056503.DOI
- [11]J. R. Maze, P. L. Stanwix, J. S. Hodges, S. Hong, J. M. Taylor, P. Cappellaro, L. Jiang, M. V. Gurudev Dutt, E. Togan, A. S. Zibrov, A. Yacoby, R. L. Walsworth, P. R. Hemmer, M. D. Lukin, Nanoscale magnetic sensing with an individual electronic spin in diamond, Nature 455 (7213) (2008) 644–647. doi:10.1038/nature07279.DOI
- [12]G. Balasubramanian, I. Y. Chan, R. Kolesov, M. Al-Hmoud, J. Tisler, C. Shin, et al., Nanoscale imaging magnetometry with diamond spins under ambient conditions, Nature 455 (7213) (2008) 648–651. doi:10.1038/nature07278.DOI
- [13]G. Kucsko, P. C. Maurer, N. Y. Yao, M. Kubo, H. J. Noh, P. K. Lo, H. Park, M. D. Lukin, Nanometre-scale thermometry in a living cell, Nature 500 (7460) (2013) 54–58. doi:10.1038/nature12373.DOI
- [14]F. Dolde, H. Fedder, M. W. Doherty, T. N"obauer, F. Rempp, G. Balasubramanian, T. Wolf, F. Reinhard, L. C. L. Hollenberg, F. Jelezko, J. Wrachtrup, Electric-field sensing using single diamond spins, Nature Physics 7 (6) (2011) 459–463. doi:10.1038/nphys1969.DOI
- [15]S. Karaveli, N. S. Zotev, A. Sarangan, H. Sari-Sarraf, C. M. Ajo-Franklin, D. E. Benson, R. Bleher, C. M. Ajo-Franklin, Modulation of nitrogen vacancy charge state and fluorescence in nanodiamonds using electrochemical potential, Proceedings of the National Academy of Sciences 113 (15) (2016) 3938–3943. doi:10.1073/pnas.1504451113.DOI
- [16]E. Bourgeois, A. Jarmola, P. Siyushev, M. Gulka, J. Hruby, F. Jelezko, D. Budker, M. Nesladek, Photoelectric detection of electron spin resonance of nitrogen-vacancy centres in diamond, Nature Communications 6 (2015) 8577. doi:10.1038/ncomms9577.DOI
- [17]B. J. Shields, Q. P. Unterreithmeier, N. P. de Leon, H. Park, M. D. Lukin, Efficient readout of a single spin state in diamond via spin-to-charge conversion, Phys. Rev. Lett. 114 (13) (2015) 136402. doi:10.1103/PhysRevLett.114.136402.DOI
- [18]L. Nie, Y. Zhang, L. Li, P. van Rijn, R. Schirhagl, ph sensitive dextran coated fluorescent nanodiamonds as a biomarker for hela cells endocytic pathway and increased cellular uptake, Nanomaterials 11 (7) (2021) 1837. doi:10.3390/nano11071837.DOI
- [19]J.-P. Tetienne, T. Hingant, L. Rondin, A. Cavaillès, L. Mayer, G. Dantelle, T. Gacoin, J. Wrachtrup, J.-F. Roch, V. Jacques, Spin relaxometry of single nitrogen-vacancy defects in diamond nanocrystals for magnetic noise sensing, Phys. Rev. B 87 (2013) 235436. doi:10.1103/PhysRevB.87.235436. URL https://link.aps.org/doi/10.1103/PhysRevB.87.235436DOI
- [20]S. Steinert, F. Ziem, L. T. Hall, A. Zappe, M. Schweikert, N. Götz, A. Aird, G. Balasubramanian, L. Hollenberg, J. Wrachtrup, Magnetic spin imaging under ambient conditions with sub-cellular resolution, Nature Communications 4 (2013) 1607. doi:10.1038/ncomms2588.DOI
- [21]F. Perona Martínez, A. C. Nusantara, M. Chipaux, S. K. Padamati, R. Schirhagl, Nanodiamond relaxometry-based detection of free-radical species when produced in chemical reactions in biologically relevant conditions, ACS Sensors 5 (12) (2020) 3862–3869. doi:10.1021/acssensors.0c01037.DOI
- [22]B. A. Myers, A. Das, M. C. Dartiailh, K. Ohno, D. D. Awschalom, A. C. Bleszynski Jayich, Probing surface noise with depth-calibrated spins in diamond, Phys. Rev. Lett. 113 (2014) 027602. doi:10.1103/PhysRevLett.113.027602. URL https://link.aps.org/doi/10.1103/PhysRevLett.113.027602DOI
- [23]Y. Romach, C. Müller, T. Unden, L. J. Rogers, T. Isoda, K. M. Itoh, M. Markham, A. Stacey, J. Meijer, S. Pezzagna, B. Naydenov, L. P. McGuinness, N. Bar-Gill, F. Jelezko, Spectroscopy of surface-induced noise using shallow spins in diamond, Phys. Rev. Lett. 114 (2015) 017601. doi:10.1103/PhysRevLett.114.017601. URL https://link.aps.org/doi/10.1103/PhysRevLett.114.017601DOI
- [24]S. Sangtawesin, B. L. Dwyer, S. Srinivasan, J. J. Allred, L. V. H. Rodgers, K. De Greve, A. Stacey, N. Dontschuk, K. M. O’Donnell, D. Hu, D. A. Evans, C. Jaye, D. A. Fischer, M. L. Markham, D. J. Twitchen, H. Park, M. D. Lukin, N. P. de Leon, Origins of diamond surface noise probed by correlating single-spin measurements with surface spectroscopy, Phys. Rev. X 9 (2019) 031052. doi:10.1103/PhysRevX.9.031052. URL https://link.aps.org/doi/10.1103/PhysRevX.9.031052DOI
- [25]M. Kim, H. J. Mamin, M. H. Sherwood, K. Ohno, D. D. Awschalom, D. Rugar, Decoherence of near-surface nitrogen-vacancy centers due to electric field noise, Phys. Rev. Lett. 115 (2015) 087602. doi:10.1103/PhysRevLett.115.087602. URL https://link.aps.org/doi/10.1103/PhysRevLett.115.087602DOI
- [26]M. V. Hauf, et al., Chemical control of the charge state of nitrogen-vacancy centers in diamond, Phys. Rev. B 83 (8) (2011) 081304. doi:10.1103/PhysRevB.83.081304.DOI
- [27]D. A. Broadway, N. Dontschuk, A. Tsai, S. E. Lillie, C. T. K. Lew, J. C. McCallum, B. C. Johnson, M. W. Doherty, A. Stacey, L. C. L. Hollenberg, J. P. Tetienne, Spatial mapping of band bending in semiconductor devices using in situ quantum sensors, Nature Electronics 1 (9) (2018) 502–507. doi:10.1038/s41928-018-0130-0.DOI
- [28]C. Shinei, H. Abe, T. Ohshima, T. Teraji, Change in charge state of nv center caused by monovacancy formation, Diamond and Related Materials 140 (2023) 110523. doi:10.1016/j.diamond.2023.110523.DOI
- [29]E. Janitz, K. Herb, L. A. Völker, W. S. Huxter, C. L. Degen, J. M. Abendroth, Diamond surface engineering for molecular sensing with nitrogen-vacancy centers, J. Mater. Chem. C 10 (37) (2022) 13533–13569. doi:10.1039/D2TC01258H.DOI
- [30]M. Barzegar Amiri Olia, P. S. Donnelly, L. C. L. Hollenberg, P. Mulvaney, D. A. Simpson, Advances in the surface functionalization of nanodiamonds for biological applications: A review, ACS Appl. Nano Mater. 4 (10) (2021) 9985–10005. doi:10.1021/acsanm.1c02698.DOI
- [31]F. A. Freire-Moschovitis, et al., The role of electrolytes in the relaxation of near-surface spin defects in diamond, ACS Nano 17 (11) (2023) 10474–10485. doi:10.1021/acsnano.3c01298.
- [32]S. Kawai, et al., Nitrogen-terminated diamond surface for nanoscale nmr by shallow nitrogen-vacancy centers, J. Phys. Chem. C 123 (6) (2019) 3594–3604. doi:10.1021/acs.jpcc.8b11274.DOI
- [33]L. Gaoxian, C. Wei, G. Nan, C. Shaoheng, L. Hongdong, Nitrogen-terminated diamond (111) surface for nitrogen-vacancy based quantum sensors, Diamond and Related Materials 142 (2024) 110813. doi:10.1016/j.diamond.2024.110813.DOI
- [34]C. Laube, et al., Defined functionality and increased luminescence of nanodiamonds for sensing and diagnostic applications by targeted high temperature reactions and electron beam irradiation, Mater. Chem. Front. 1 (12) (2017) 2527–2540. doi:10.1039/C7QM00241F.DOI
- [35]J. M. Abendroth, K. Herb, E. Janitz, T. Zhu, L. A. Völker, C. L. Degen, Single-nitrogen-vacancy nmr of amine-functionalized diamond surfaces, Nano Lett. 22 (18) (2022) 7294–7303. doi:10.1021/acs.nanolett.2c00533.DOI
- [36]H. Koch, W. Kulisch, C. Popov, R. Merz, B. Merz, J. P. Reithmaier, Plasma amination of ultrananocrystalline diamond/amorphous carbon composite films for the attachment of biomolecules, Diamond and Related Materials 20 (2) (2011) 254–258. doi:10.1016/j.diamond.2010.12.010.DOI
- [37]A. Artemenko, et al., Stability of the surface termination of nanocrystalline diamond and diamond-like carbon films exposed to open air conditions, Diamond and Related Materials 100 (2019) 107562. doi:10.1016/j.diamond.2019.107562.DOI
- [38]N. Jegenyes, et al., Materials and spin characteristics of amino-terminated nanodiamonds embedded with nitrogen-vacancy color centers, arXiv preprint arXiv:2502.07706 (2025). doi:10.48550/arXiv.2502.07706.DOI
- [39]K.-I. Sotowa, T. Amamoto, A. Sobana, K. Kusakabe, T. Imato, Effect of treatment temperature on the amination of chlorinated diamond, Diamond and Related Materials 13 (1) (2004) 145–150. doi:10.1016/j.diamond.2003.10.029.DOI
- [40]J. B. Miller, D. W. Brown, Properties of photochemically modified diamond films, Diamond and Related Materials 4 (4) (1995) 435–440. doi:10.1016/0925-9635(94)05209-3.DOI
- [41]M. Kaviani, P. Deák, B. Aradi, T. Frauenheim, J.-P. Chou, A. Gali, Proper surface termination for luminescent near-surface nv centers in diamond, Nano Lett. 14 (8) (2014) 4772–4777. doi:10.1021/nl501927y.DOI
- [42]R. Malkinson, M. K. Kuntumalla, A. Chemin, T. Petit, A. Hoffman, N. Bar-Gill, Enhanced quantum properties of shallow diamond atomic defects through nitrogen surface termination, J. Mater. Chem. C 12 (20) (2024) 7206–7213. doi:10.1039/D4TC00581C.DOI
- [43]S. Czene, et al., Amino-termination of silicon carbide nanoparticles, Nanomaterials 13 (13) (2023) 1953. doi:10.3390/nano13131953.DOI
- [44]K. J. Rietwyk, et al., Work function and electron affinity of the fluorine-terminated (100) diamond surface, Applied Physics Letters 102 (9) (2013) 091604. doi:10.1063/1.4793999.DOI
- [45]S. Czene, O. Krafcsik, N. Jegenyes, G. Bortel, Á. Gali, Direct amino-termination of nanodiamonds and investigation of their optical properties upon annealing, in: NANOCON 2025, 2025, pp. 373–379. doi:10.37904/nanocon.2025.5223.DOI
- [46]M. Barzegar Amiri Olia, et al., Functionalized fluorescent nanodiamonds with millisecond spin relaxation times, ACS Nano 19 (42) (2025) 36884–36895. doi:10.1021/acsnano.5c02407.
- [47]R. Kumar, S. Mahajan, F. Donaldson, S. Dhomkar, H. J. Lancaster, C. Kalha, A. A. Riaz, Y. Zhu, C. A. Howard, A. Regoutz, J. J. L. Morton, Stability of near-surface nitrogen vacancy centers using dielectric surface passivation, ACS Photonics 11 (3) (2024) 1244–1251. arXiv:https://doi.org/10.1021/acsphotonics.3c01773, doi:10.1021/acsphotonics.3c01773. URL https://doi.org/10.1021/acsphotonics.3c01773
- [48]A. Mameli, et al., The properties of the nitrogen-vacancy center in milled chemical vapor deposition nanodiamonds, Mater. Quantum Technol. 6 (1) (2026) 015202. doi:10.1088/2633-4356/ae4565.DOI
- [49]M. H. Alkhatni, et al., Advanced post-treatment strategy for quantum-grade fluorescent nanodiamonds, Front. Quantum Sci. Technol. 4 (2025). doi:10.3389/frqst.2025.1687810.
- [50]S. Cui, E. L. Hu, Increased negatively charged nitrogen-vacancy centers in fluorinated diamond, Applied Physics Letters 103 (5). doi:10.1063/1.4817651.
- [51]C. Osterkamp, J. Scharpf, S. Pezzagna, J. Meijer, T. Diamant, R. Jürgen Behm, B. Naydenov, F. Jelezko, Increasing the creation yield of shallow single defects in diamond by surface plasma treatment, Applied Physics Letters 103 (19). doi:10.1063/1.4829875.DOI
- [52]C. Osterkamp, J. Lang, J. Scharpf, C. Müller, L. P. McGuinness, T. Diamant, R. J. Behm, B. Naydenov, F. Jelezko, Stabilizing shallow color centers in diamond created by nitrogen delta-doping using SF6 plasma treatment, Applied Physics Letters 106 (11) (2015) 113109. doi:10.1063/1.4915305.DOI
- [53]N. N. Greenwood, A. Earnshaw, Chemistry of the Elements, Elsevier, 2012.
- [54]Y. Liu, Z. Gu, J. L. Margrave, V. N. Khabashesku, Functionalization of nanoscale diamond powder: Fluoro-, alkyl-, amino-, and amino acid-nanodiamond derivatives, Chem. Mater. 16 (20) (2004) 3924–3930. doi:10.1021/cm048875q.DOI
- [55]T. Ando, et al., Diffuse reflectance infrared fourier-transform study of the direct thermal fluorination of diamond powder surfaces, J. Chem. Soc., Faraday Trans. 91 (18) (1995) 3209–3212. doi:10.1039/FT9959103209.DOI
- [56]C. P. Kealey, T. M. Klapötke, D. W. McComb, M. I. Robertson, J. M. Winfield, Fluorination of polycrystalline diamond films and powders. an investigation using ftir spectroscopy, sem, energy-filtered tem, xps and fluorine-18 radiotracer methods, J. Mater. Chem. 11 (3) (2001) 879–886. doi:10.1039/B006800O.DOI
- [57]P. T. Nyffeler, S. G. Durón, M. D. Burkart, S. P. Vincent, C.-H. Wong, Selectfluor: Mechanistic insight and applications, Angewandte Chemie International Edition 44 (2) (2005) 192–212. doi:10.1002/anie.200400648.DOI
- [58]E. Differding, H. Ofner, A practical reagent for electrophilic fluorinations, Synlett 1991 (3) (1991) 187–189. doi:10.1055/s-1991-20673.DOI
- [59]L. V. H. Rodgers, et al., Diamond surface functionalization via visible light-driven c-h activation for nanoscale quantum sensing, Proceedings of the National Academy of Sciences 121 (11) (2024) e2316032121. doi:10.1073/pnas.2316032121.DOI
- [60]J. Havlik, et al., Benchtop fluorination of fluorescent nanodiamonds on a preparative scale: Toward unusually hydrophilic bright particles, Advanced Functional Materials 26 (23) (2016) 4134–4142. doi:10.1002/adfm.201504857.DOI
- [61]C. A. Ramsden, Xenon difluoride in the organic laboratory: a tale of substrates, solvents and vessels, Arkivoc 2014 (1) (2013) 109–126. doi:10.3998/ark.5550190.p008.436.DOI
- [62]C. Chatalova-Sazepin, et al., Xenon difluoride mediated fluorodecarboxylations for the syntheses of di- and trifluoromethoxyarenes, Org. Lett. 18 (18) (2016) 4570–4573. doi:10.1021/acs.orglett.6b02208.
- [63]M. Tramsek, B. Zemva, Synthesis, properties and chemistry of xenon(ii) fluoride, ChemInform 38 (21) (2007). doi:10.1002/chin.200721209.DOI
- [64]J. S. Foord, N. K. Singh, R. B. Jackman, A. Gutierrez-Sosa, S. Proffitt, K. B. Holt, Reactions of xenon difluoride and atomic hydrogen at chemical vapour deposited diamond surfaces, Surface Science 488 (3) (2001) 335–345. doi:10.1016/S0039-6028(01)01142-6.DOI
- [65]J. F. Morar, F. J. Himpsel, G. Hollinger, J. L. Jordan, G. Hughes, F. R. McFeely, C 1s excitation studies of diamond (111). i. surface core levels, Phys. Rev. B 33 (2) (1986) 1340–1345. doi:10.1103/PhysRevB.33.1340.DOI
- [66]M. A. Tius, Xenon difluoride in synthesis, Tetrahedron 51 (24) (1995) 6605–6634. doi:10.1016/0040-4020(95)00362-C.DOI
- [67]T. B. Patrick, S. Khazaeli, S. Nadji, K. Hering-Smith, D. Reif, Mechanistic studies of fluorodecarboxylation with xenon difluoride, J. Org. Chem. 58 (3) (1993) 705–708. doi:10.1021/jo00055a026.DOI
- [68]V. Verkholyuk, C. Bouchair, O. A. Anisimov, A. Pershin, A. Gali, Quantum relaxometry under continuous wave excitation, arXiv preprint arXiv:2608.07697 (2026). doi:10.48550/arXiv.2608.07697.
- [69]T. Petit, L. Puskar, First spectroscopy of nanodiamonds: Methods and interpretation, Diamond and Related Materials 89 (2018) 52–66. doi:10.1016/j.diamond.2018.08.005.DOI
- [70]V. Y. Osipov, N. M. Romanov, K. Kogane, H. Touhara, et al., Intrinsic infrared absorption for carbon–fluorine bonding in fluorinated nanodiamond, Mendeleev Communications 30 (1) (2020) 105–107. doi:10.1016/j.mencom.2020.01.028.DOI
- [71]E. M. Zagrebina, A. Generalov, A. Y. Klyushin, K. Simonov, N. Vinogradov, M. Dubois, L. Frezet, N. Martensson, A. Preobrajenski, A. S. Vinogradov, Comparative nexafs, nmr, and ftir study of various-sized nanodiamonds: As-prepared and fluorinated, J. Phys. Chem. C 119 (1) (2015) 835–844. doi:10.1021/jp510618s.DOI
- [72]J. Coates, Interpretation of infrared spectra, a practical approach, in: Encyclopedia of Analytical Chemistry, John Wiley & Sons, Ltd, 2006. doi:10.1002/9780470027318.a5606.
- [73]C. Nolte, J. Ammer, H. Mayr, Nucleofugality and nucleophilicity of fluoride in protic solvents, J. Org. Chem. 77 (7) (2012) 3325–3335. doi:10.1021/jo300141z.DOI
- [74]O. Shenderova, A. M. Panich, S. Moseenkov, S. C. Hens, V. Kuznetsov, H.-M. Vieth, Hydroxylated detonation nanodiamond: Ftir, xps, and nmr studies, J. Phys. Chem. C 115 (39) (2011) 19005–19011. doi:10.1021/jp205389m.DOI
- [75]N. S. Gill, R. H. Nuttall, D. E. Scaife, D. W. A. Sharp, The infra-red spectra of pyridine complexes and pyridinium salts, Journal of Inorganic and Nuclear Chemistry 18 (1961) 79–87. doi:10.1016/0022-1902(61)80372-2.DOI
- [76]O. Romanyuk, Š. Stehlík, J. Zemek, K. Aubrechtová Dragounová, A. Kromka, Utilizing constant energy difference between sp-peak and c 1s core level in photoelectron spectra for unambiguous identification and quantification of diamond phase in nanodiamonds, Nanomaterials 14 (7) (2024) 590. doi:10.3390/nano14070590.DOI
- [77]A. M. Panich, H.-M. Vieth, A. I. Shames, N. Froumin, E. Osawa, A. Yao, Structure and bonding in fluorinated nanodiamond, J. Phys. Chem. C 114 (2) (2010) 774–782. doi:10.1021/jp9078629.DOI
- [78]H. Hantsche, High resolution xps of organic polymers, the scienta escа300 database (review of book by g. beamson and d. briggs), Advanced Materials 5 (10) (1993) 778. doi:10.1002/adma.19930051035.
- [79]W. Kang, S. Li, Preparation of fluorinated graphene to study its gas sensitivity, RSC Advances 8 (41) (2018) 23459–23467. doi:10.1039/C8RA03451F.DOI
- [80]D. R. Baer, K. Artyushkova, H. Cohen, C. D. Easton, M. Engelhardt, T. R. Gengenbach, G. Greczynski, P. Mack, D. J. Morgan, A. Roberts, Xps guide: Charge neutralization and binding energy referencing for insulating samples, Journal of Vacuum Science & Technology A 38 (3) (2020) 031204. doi:10.1116/6.0000057.DOI
- [81]G. Greczynski, L. Hultman, X-ray photoelectron spectroscopy: Towards reliable binding energy referencing, Progress in Materials Science 107 (2020) 100591. doi:10.1016/j.pmatsci.2019.100591.DOI
- [82]H. F. Halliwell, S. C. Nyburg, The reaction of the benzenediazonium ion with certain anions in aqueous acid solution, J. Chem. Soc. (1960) 4603–4608doi:10.1039/JR9600004603.
- [83]T. A. Carlson, G. E. McGuire, Study of the x-ray photoelectron spectrum of tungsten-tungsten oxide as a function of thickness of the surface oxide layer, Journal of Electron Spectroscopy and Related Phenomena 1 (2) (1972) 161–168. doi:10.1016/0368-2048(72)80029-X.DOI
- [84]C. J. Powell, Practical guide for inelastic mean free paths, effective attenuation lengths, mean escape depths, and information depths in x-ray photoelectron spectroscopy, NIST, accessed: 2026-06-09 (2020). URL https://www.nist.gov/publications/practical-guide-inelastic-mean-free-paths-effective-attenuaDOI
- [85]D. R. Baer, et al., Challenges in applying surface analysis methods to nanoparticles and nanostructured materials, Journal of Surface Analysis (2005).
- [86]F. Ducrozet, et al., New insights into the reactivity of detonation nanodiamonds during the first stages of graphitization, Nanomaterials 11 (10) (2021) 2671. doi:10.3390/nano11102671.DOI
- [87]M. Veres, S. Tóth, M. Koós, Grain boundary fine structure of ultrananocrystalline diamond thin films measured by raman scattering, Applied Physics Letters 91 (3) (2007) 031913. doi:10.1063/1.2757122.DOI
- [88]A. C. Ferrari, J. Robertson, Resonant raman spectroscopy of disordered, amorphous, and diamondlike carbon, Phys. Rev. B 64 (7) (2001) 075414. doi:10.1103/PhysRevB.64.075414.DOI
- [89]E. B. Hughes, H. H. G. Jellinek, B. A. Ambrose, Pyridine, ultraviolet absorption spectrum and dissociation constant, J. Phys. Chem. 53 (3) (1949) 410–414. doi:10.1021/j150468a009.DOI
- [90]H. Wang, C. Zhu, J.-G. Yu, S. H. Lin, Anharmonic franck-condon simulation of the absorption and fluorescence spectra for the low-lying s1 and s2 excited states of pyridine, J. Phys. Chem. A 113 (52) (2009) 14407–14414. doi:10.1021/jp903585c.DOI
- [91]P. Rózsa, O. Krafcsik, S. Lenk, D. Beke, A. Gali, Thiourea modification of fluorescent nanodiamonds towards enhanced quantum sensing, Carbon Trends 23 (2026) 100618. doi:10.1016/j.cartre.2026.100618.DOI
- [92]J. J. Goldstein, D. E. Newbury, J. R. Michael, N. W. M. Ritchie, J. H. J. Scott, D. C. Joy, Scanning Electron Microscopy and X-Ray Microanalysis, 4th Edition, Springer, New York, NY, 2018. doi:10.1007/978-1-4939-6676-9.DOI
- [93]I. T. Horváth, J. Rábai, Facile catalyst separation without water: Fluorous biphase hydroformylation of olefins, Science 266 (5182) (1994) 72–75. doi:10.1126/science.266.5182.72.DOI
- [94]L. Rondin, et al., Surface-induced charge state conversion of nitrogen-vacancy defects in nanodiamonds, Phys. Rev. B 82 (11) (2010) 115449. doi:10.1103/PhysRevB.82.115449.DOI
- [95]S. T. Alsid, J. F. Barry, L. M. Pham, J. M. Schloss, M. F. O’Keeffe, P. Cappellaro, D. A. Braje, Photoluminescence decomposition analysis: A technique to characterize n-v creation in diamond, Phys. Rev. Applied 12 (2019) 044003. doi:10.1103/PhysRevApplied.12.044003.DOI
- [96]C. Shinei, et al., Equilibrium charge state of nv centers in diamond, Appl. Phys. Lett. 119 (25) (2021) 254001. doi:10.1063/5.0079687.DOI
- [97]K. Beha, A. Batalov, N. B. Manson, R. Bratschitsch, A. Leitensdorfer, Optimum photoluminescence excitation and recharging cycle of single nitrogen-vacancy centers in ultrapure diamond, Phys. Rev. Lett. 109 (9) (2012) 097404. doi:10.1103/PhysRevLett.109.097404.DOI
- [98]T. Mittiga, S. Hsieh, C. Zu, B. Kobrin, F. Machado, P. Bhattacharyya, N. Z. Rui, A. Jarmola, S. Choi, D. Budker, N. Y. Yao, Imaging the local charge environment of nitrogen-vacancy centers in diamond, Phys. Rev. Lett. 121 (24) (2018) 246402. doi:10.1103/PhysRevLett.121.246402.DOI