Nitrogen-vacancy center

The nitrogen-vacancy center is a point defect in the diamond lattice. It consists of a nearest-neighbor pair of a nitrogen atom, which substitutes for a carbon atom, and a lattice vacancy.

Bottom images are spatial photoluminescence (PL) maps before and after application of +20 V voltage to a planar Schottky diode. The top image outlines the experiment. The PL maps reveal the conversion of individual N-V⁰ centers into N-V⁻ centers that appear as bright dots.[1]

Two charge states of this defect, neutral N-V⁰ and negative N-V⁻, are known from spectroscopic studies using optical absorption,[2][3] photoluminescence (PL),[4] electron paramagnetic resonance (EPR)[5][6][7] and optically detected magnetic resonance (ODMR),[8] which can be viewed as a hybrid of PL and EPR; most details of the structure originate from EPR. The nitrogen atom on one hand has five valence electrons. Three of them are covalently bonded to the carbon atoms, while the other two remain non-bonded and are called a lone pair. The vacancy on the other hand has three unpaired electrons. Two of them form a quasi covalent bond and one remains unpaired. The overall symmetry, however, is axial (trigonal C_3V); one can visualize this by imagining the three unpaired vacancy electrons continuously exchanging their roles.

The N-V⁰ thus has one unpaired electron and is paramagnetic. However, despite extensive efforts, electron paramagnetic resonance signals from N-V⁰ avoided detection for decades until 2008. Optical excitation is required to bring the N-V⁰ defect into the EPR-detectable excited state; the signals from the ground state are presumably too broad for EPR detection.[9]

The N-V⁰ centers can be converted into N-V⁻ by changing the Fermi level position. This can be achieved by applying external voltage to a p-n junction made from doped diamond, e.g., in a Schottky diode.[1]

In the negative charge state N-V⁻, an extra electron is located at the vacancy site forming a spin S=1 pair with one of the vacancy electrons. As in N-V⁰, the vacancy electrons are "exchanging roles" preserving the overall trigonal symmetry. This N-V⁻ state is what is commonly, and somewhat incorrectly, called "the nitrogen-vacancy center". The neutral state has not yet been explored for spin manipulations.

The N-V centers are randomly oriented within a diamond crystal. Ion implantation techniques can enable their artificial creation in predetermined positions.[10]

Nitrogen-vacancy centers are typically produced from single substitutional nitrogen centers (called C or P1 centers in diamond literature) by irradiation followed by annealing at temperatures above 700 °C.[2] A wide range of high-energy particles is suitable for such irradiation, including electrons, protons, neutrons, ions, and gamma photons. Irradiation produces lattice vacancies, which are a part of N-V centers. Those vacancies are immobile at room temperature, and annealing is required to move them. Single substitutional nitrogen produces strain in the diamond lattice;[11] it therefore efficiently captures moving vacancies,[12] producing the N-V centers.

During chemical vapor deposition of diamond, a small fraction of single substitutional nitrogen impurity (typically <0.5%) traps vacancies generated as a result of the plasma synthesis. Such nitrogen-vacancy centers are preferentially aligned to the growth direction.[13]

Diamond is notorious for having a relatively large lattice strain. Strain splits and shifts optical transitions from individual centers resulting in broad lines in the ensembles of centers.[2] Special care is taken to produce extremely sharp N-V lines (line width ~10 MHz)[14] required for most experiments: high-quality, pure natural or better synthetic diamonds (type IIa) are selected. Many of them already have sufficient concentrations of grown-in N-V centers and are suitable for applications. If not, they are irradiated by high-energy particles and annealed. Selection of a certain irradiation dose allows tuning the concentration of produced N-V centers such that individual N-V centers are separated by micrometre-large distances. Then, individual N-V centers can be studied with standard optical microscopes or, better, near-field scanning optical microscopes having sub-micrometre resolution.[8][15]

Optical absorption and emission of the N-V⁻ center at room temperature.

N-V⁻ centers emit bright red light which can be conveniently excited by visible light sources, such as argon or krypton lasers, frequency doubled Nd:YAG lasers, dye lasers, or He-Ne lasers. Excitation can also be achieved at energies below that of zero phonon emission.[16] Laser illumination, however, also converts some N-V⁻ into N-V⁰ centers.[4] Emission is very quick (relaxation time ~10 ns).[17][18] At room temperature, no sharp peaks are observed because of the thermal broadening. However, cooling the N-V⁻ centers with liquid nitrogen or liquid helium dramatically narrows the lines down to a width of a few megahertz.

An important property of the luminescence from individual N-V⁻ centers is its high temporal stability. Whereas many single-molecular emitters bleach after emission of 10⁶–10⁸ photons, no bleaching is observed for the N-V centers at room temperature.[8][15]

Because of these properties, the ideal technique to address the N-V centers is confocal microscopy, both at room temperature and at low temperature. In particular, low temperature operation is required to specifically address only the zero-phonon line (ZPL).

Schematic energy level structure of the N-V⁻ center. Electron transitions between the ground ³A and excited ³E states, separated by 1.945 eV (637 nm), produce absorption and luminescence. The ³A state is split by 1027 gauss[5][6] (~12 µeV) and the ³E state by 508 gauss[19] (~5.9 µeV). Numbers 0, ±1 indicate spin value; splitting due to the orbital degeneracy is not shown.

In general the N-V⁻ center has a ground state triplet (³A), an excited state triplet (³E) and an intermediate state singlet (¹A).[20] ³A and ³E each contain the m_s=±1 spin states, where the two electron spins are aligned (either up, such that m_s=+1 or down, such that m_s=-1), and the m_s=0 spin state where the electron spins are antiparallel. Due to the magnetic interaction the energy of the m_s=±1 state is higher than that of the m_s=0 state. ¹A however only contains a spin state singlet with spin 0. If an external magnetic field is applied along the defect axis (the axis which crosses the nitrogen atom and the vacancy) of the N-V⁻ center, it does not affect the m_s=0 states, but it splits the m_s = ±1 levels (Zeeman effect).

In the absence of an external magnetic field, the ground and excited states are split by the magnetic interaction between the two unpaired electrons at the N-V⁻ center (see microscopic model): when two electrons have parallel spins (m_s=±1), their energy is higher than when spins are antiparallel (m_s=0). The farther apart the electrons are, the weaker their interaction energy D (roughly D ~1/r³).[6] Thus the smaller splitting in the excited state can be viewed in terms of larger electron-electron separation in the excited state. When an external magnetic field is applied to the N-V⁻ center, it does not affect the m_s=0 states nor the ¹A state (because it has S = 0), but it splits the m_s = ±1 levels. If a magnetic field is oriented along the defect axis and reaches about 1027 G (or 508 G) then the m_s = –1 and m_s = 0 states in the ground (or excited) state become equal in energy; they strongly interact resulting in so-called spin polarization, which strongly affects the intensity of optical absorption and luminescence transitions involving those states.[19]

This happens because transitions between electronic states are mediated by a photon which cannot change overall spin. Thus optical transitions must preserve the total spin and occur between levels of the same total spin. For this reason, transitions ³E↔¹A and ¹A ↔ ³A are non-radiative and quench the luminescence. Whereas m_s = −1 (excited state) ↔ m_s = 0 (ground state) transition was forbidden in the absence of an external magnetic field, it becomes allowed when a magnetic field mixes the m_s = −1 and m_s = 0 levels in the ground state. As a measurable outcome of this phenomenon, luminescence intensity can be strongly modulated by magnetic field.

An important property of the non-radiative transition between ³E and ¹A is that it is stronger for m_s = ±1 and weaker for m_s = 0. This property results in a very useful manipulation of N-V center, which is called optical spin-polarization. First, consider an off-resonance excitation which has a higher frequency (typically 2.32 eV (532 nm)) than the frequencies of all transitions and thus lays in the vibronic bands for all transitions. By using a pulse of this wavelength, people can excite all spin states and create phonons as well. For a spin state with m_s = 0, due to conservation of spin in transition, it will be excited to the corresponding m_s = 0 state in ³E and then go back to the original state. However, for a spin state with m_s = ±1 in ³A, after the excitation, it has a relatively high probability to jump to the intermediate state ¹A by non-radiative transition and go to the ground state with m_s = 0. After sufficient cycles, the state of the N-V center can be regarded as in the m_s = 0 state. Such a process can be used in the initialization of quantum state in quantum information processing.

There is an additional level splitting in the excited ³E state due to the orbital degeneracy and spin-orbit interaction. Importantly, this splitting can be modulated by applying a static electric field,[14][21] in a similar fashion to the magnetic field mechanism outlined above, though the physics of the splitting is somewhat more complex. Nevertheless, an important practical outcome is that the intensity and position of the luminescence lines can be modulated by applying electric or/and magnetic fields.

The energy difference between the m_s = 0 and m_s = ±1 states corresponds to the microwave region. Thus by irradiating the N-V centers with microwave radiation, one can change the relative population of those levels, thereby again modulating the luminescence intensity.

There is an additional splitting of the m_s = ±1 energy levels, which originates from the "hyperfine" interaction between the nuclear and electron spins. Thus finally, the optical absorption and luminescence from the N-V⁻ center consists of roughly a dozen sharp lines with a separation in the MHz-GHz range, and all those lines can be resolved, given proper sample preparation. The intensity and position of those lines can be modulated using the following tools:

1. Amplitude and orientation of magnetic field, which splits the m_s = ±1 levels in the ground and excited states.
2. Amplitude and orientation of elastic field (strain), which can be applied by, e.g., squeezing the diamond. Similar effects can be induced by applying an electric field,[14][21] and the electric field can be controlled with much higher precision.
3. Continuous-wave microwave radiation, which changes the population of the sublevels within the ground and excited state.[21]
4. Tunable laser, which can selectively excite certain sublevels of the ground and excited state.[21][22]
5. In addition to those static perturbations, numerous dynamic effects (spin echo, Rabi oscillations, etc.) can be exploited by applying a carefully designed sequence of microwave pulses.[23][24][25][26][27] The first pulse coherently excites the electron spins, and this coherence is then manipulated and probed by the subsequent pulses. Those dynamic effects are rather important for the practical realization of quantum computers, which ought to work at high frequency.

The above-described energy structure is by no means exceptional for a defect in diamond or other semiconductor.[28] It was not this structure alone, but a combination of several favorable factors (previous knowledge, easy production and excitation, etc.) which suggested the use of the N-V⁻ center.

The energy level structure of the N-V⁻ center was established by combining optically detected magnetic resonance (ODMR), electron paramagnetic resonance (EPR) and theoretical results, as shown in the figure. In particular, several theoretical works have been done, using the Linear Combination of Atomic Orbitals (LCAO) approach, to build the electronic orbitals to describe the possible quantum states, looking at the N-V center as a molecule. Moreover, group theory results are used, to take into account the symmetry of the diamond crystal, and so the symmetry of the N-V itself. The energy levels are labeled according to group theory, and in particular are labelled after the irreducible representations of the C_3V symmetry group of the defect center, A₁, A₂ and E. The numbers 3 in ³A and 1 in ¹A represent the number of allowable m_s spin states, or the spin multiplicity, which range from –S to S for a total of 2S+1 possible states. If S = 1, m_s can be −1, 0, or 1. The ¹A level is actually composed of two levels with a weak radiative decay in between in the near-infrared region.[29]

Spin dynamics in the N-V^– center on diamond. The primary transition between triplet ground and excited states is predominantly spin conserving. Decay via the intermediate singlets gives rise to spin polarization by preferentially switching spin from m_s = ±1 to m_s = 0. Both absorption and emission wavelengths are indicated,[30] since they differ due to Stokes shift.[31] (Correction: The wavelength of the 1E-1A transition should be 1042 nm. [32])

Thinking of the N-V⁻ center as a multielectronic system, we can draw the diagram in the figure at the left, where the states are labeled according to their symmetry and with a left superscript that indicates with a 3 if it is a triplet (S=1) and with a 1 if it is a singlet (S=0). It is well accepted today that we have two triplet states and two intermediate singlet states.[33]

The optical excitations conserve the spin state, but there is a high probability of the states ${\textstyle \left|^{3}{\text{E}},\pm 1\right\rangle }$ decaying non-radiatively to the singlet state ${\textstyle \left|^{1}{\text{A}}_{1}\right\rangle }$, a phenomenon called intersystem crossing (ISC). This happens at an appreciable rate because the energy curve in function of the position of the atoms for the ${\textstyle \left|^{3}{\text{E}},\pm 1\right\rangle }$ state intersects the curve for the ${\textstyle \left|^{1}{\text{A}}_{1}\right\rangle }$ state. Therefore, for some instant during the vibrational relaxation that the ions undergo after the excitement, it is possible for the spin to flip with little or no energy required in the transition.[34] It is important to note that this mechanism also leads to a transition from ${\textstyle \left|^{3}{\text{E}},0\right\rangle }$ to ${\textstyle \left|^{1}{\text{A}}_{1}\right\rangle }$, but the rate of this ISC is much lower than the ${\textstyle \left|^{3}{\text{E}},\pm 1\right\rangle }$ states rate, therefore this transition is indicated with a thin line. The diagram also shows the non-radiative and infrared competing decay paths between the two singlet states, and the fine splitting in the triplet states, whose differences in energy correspond to microwave frequencies.

Some authors explain the dynamics of the N-V⁻ center by admitting that the transition from ${\textstyle \left|^{1}{\text{E}}\right\rangle }$ to ${\textstyle \left|^{3}{\text{A}}_{2},\pm 1\right\rangle }$ is small, but as Robledo et al. shows,[35] only the fact that the probability of decaying to ${\textstyle \left|^{1}{\text{A}}_{1}\right\rangle }$ is smaller for ${\textstyle \left|^{3}{\text{E}},0\right\rangle }$ than for ${\textstyle \left|^{3}{\text{E}},\pm 1\right\rangle }$ is enough to polarize the spin to m_s = 0.

Scanning thermal microscopy using the N-V center.
(a) Schematics of experimental setup. An electric current is applied to the arms of an AFM cantilever (phosphorus-doped Si, P:Si) and heats up the end section above the tip (intrinsic Si, i-Si). The bottom lens excites a diamond nanocrystal with a green laser light and collects photoluminescence (PL). The crystal hosts an N-V center and is attached to the AFM tip. A wire on the sample surface serves as the microwave source (mw). The temperature of the cantilever T_h is determined from the applied current and voltage.
(b) ODMR spectra of the N-V center at three temperatures. The line splitting originates from a ∼1 mT applied magnetic field.
(c) Thermal conductivity image of a gold letter E on sapphire. White circles indicate features that do not correlate with the AFM topography. (d) PL image of the AFM cantilever end and tip where the diamond nanocrystal appears as the bright spot. (e) Zoomed PL image of the N-V center in d.[36]

The spectral shape and intensity of the optical signals from the N-V⁻ centers are sensitive to external perturbation, such as temperature, strain, electric and magnetic field. However, the use of spectral shape for sensing those perturbation is impractical, as the diamond would have to be cooled to cryogenic temperatures to sharpen the N-V⁻ signals. A more realistic approach is to use luminescence intensity (rather than lineshape), which exhibits a sharp resonance when a microwave frequency is applied to diamond that matches the splitting of the ground state levels. The resulting optically detected magnetic resonance signals are sharp even at room temperature, and can be used in miniature sensors. Such sensors can detect magnetic fields of a few nanotesla[37] or electric fields of about 10 V/cm[38] at kilohertz frequencies after 100 seconds of averaging. This sensitivity allows detecting a magnetic or electric field produced by a single electron located tens of nanometers away from an N-V⁻ center.

Using the same mechanism, the N-V⁻ centers were employed in scanning thermal microscopy to measure high-resolution spatial maps of temperature and thermal conductivity (see image).[36]

Another possible use of the N-V⁻ centers is as a detector to measure the full mechanical stress tensor in the bulk of the crystal. For this application, the stress-induced splitting of the zero-phonon-line is exploited, and its polarization properties.[39] A robust frequency-modulated radio receiver using the electron-spin-dependent photoluminescence that operated up to 350 °C demonstrates the possibility for use in extreme conditions.[40]

In addition to the quantum optical applications, luminescence from the N-V⁻ centers can be applied for imaging biological processes, such as fluid flow in living cells.[41] This application relies on good compatibility of diamond nano-particles with the living cells and on favorable properties of photoluminescence from the N-V⁻ centers (strong intensity, easy excitation and detection, temporal stability, etc.). Compared with large single-crystal diamonds, nanodiamonds are cheap (about 1 USD per gram) and available from various suppliers. N-V⁻ centers are produced in diamond powders with sub-micrometre particle size using the standard process of irradiation and annealing described above. Due to the relatively small size of nanodiamond, NV centers can be produced by irradiating nanodiamond of 100nm or less with medium energy H+ beam. This method reduces the required ion dose and reaction, making it possible to mass produce fluorescent nanodiamonds in ordinary laboratory.[42] Fluorescent nanodiamond produced with such method is bright and photostable, making it excellent for long-term, three dimensional tracking of single particle in living cell.[43] Those nanodiamonds are introduced in a cell, and their luminescence is monitored using a standard fluorescence microscope.[44]

Further N-V⁻ center has been hypothesized to be a potential bio-mimetic system for emulating radical pair spin dynamics of the avian compass.[45][46]

Stimulated emission from the N-V⁻ center has been demonstrated, though it could be achieved only from the phonon side-band (i.e. broadband light) and not from the ZPL. For this purpose, the center has to be excited at a wavelength longer than ~650 nm, as higher-energy excitation ionizes the center.[47]

The first continuous-wave room-temperature maser has been demonstrated.[48][49] It used 532-nm pumped N-V⁻ centers held within a high Purcell factor microwave cavity and an external magnetic field of 4300 G. Continuous maser oscillation generated a coherent signal at ~9.2 GHz.

The N-V center can have a very long spin coherence time approaching the second regime.[50] This is advantageous for applications in quantum sensing[51] and quantum communication.[52] Disadvantageous for these applications is the long radiative lifetime (~12 ns[53][54] ) of the N-V center and the strong phonon sideband in its emission spectrum. Both issues can be addressed by putting the N-V center in an optical cavity.[55]

The microscopic model and most optical properties of ensembles of the N-V⁻ centers have been firmly established in the 1970s based on the optical measurements combined with uniaxial stress[2] and on the electron paramagnetic resonance.[5][6] However, a minor error in EPR results (it was assumed that illumination is required to observe N-V⁻ EPR signals) resulted in the incorrect multiplicity assignments in the energy level structure. In 1991 it was shown that EPR can be observed without illumination,[7] which established the energy level scheme shown above. The magnetic splitting in the excited state has been measured only recently.[19]

The characterization of single N-V⁻ centers has become a very competitive field nowadays, with many dozens of papers published in the most prestigious scientific journals. One of the first results was reported back in 1997.[8] In that paper, it was demonstrated that the fluorescence of single N-V⁻ centers can be detected by room-temperature fluorescence microscopy and that the defect shows perfect photostability. Also one of the outstanding properties of the N-V center was demonstrated, namely room-temperature optically detected magnetic resonance.

Wikimedia Commons has media related to Nitrogen-vacancy center in diamond.

• Crystallographic defects in diamond
• Crystallographic defect
• Material properties of diamond