3.Physikalisches Institut - Universität Stuttgart

Spectroscopy of Single N-V Centers in Diamond

  Nitrogen-vacancy (N-V) center in diamond
  Single N-V centers at room temperature
  Spectroscopy of single N-V defects at low temperature
  Fluorescence spectroscopy on single N-V centers in diamond nanocrystals


Diamond and diamond-like materials are finding increasing use in various areas of modern technology because of their unique physical properties such as thermal conductivity, hardness, and high Debye temperature. Optical properties of diamond (which are mostly related to the presence of defects) are of great interest because of the proposed application of diamond as a laser source and high- efficiency photodetector for the UV [1]. More than 500 optical centers have been reported based on the absorption and luminescence spectra of diamond [2]. Half of them are believed to be impurity related. No other impurity exhibits in diamond such a variety of optical centers, as does nitrogen. Nitrogen-vacancy (N-V) defects show particularly strong electronic transition allowing to detect individual centers optically. Those defects show a remarkable photostability and they have been proposed as light sources for high-resolution scanning-probe microscopy and quantum cryptography [3][4][5]. The main goal of this paper is to give a short review on recent single-molecule studies on N-V centers in diamonds.

Nitrogen-vacancy (N-V) center in diamond

The nitrogen-vacancy center in diamond is created by irradiation damage and subsequent annealing of diamonds containing atomically dispersed nitrogen (type 1b crystals). Radiation damage creates vacancies in the diamond lattice. Annealing treatment leads to migration of vacancies towards nitrogen atoms creating the nitrogen-vacancy (N-V) defect. The figure on the left hand side shows the structure of N-V center. Mita and co-workers have reported based on neutron irradiation experiments that N-V center is negatively charged [6]. The excited state of the N-V center is an orbital doublet and the ground state is an orbital singlet (see figure on the right hand side). Optically detected magnetic resonance experiments demonstrated that the ground state is a spin triplet and the transition is 3A-3E [7][8]. The zero-phonon line associated with this transition is at 637 nm (1.945 eV) and exhibits a strong inhomogeneous broadening (30 cm-1) attributed to a large strain variation in the excited 3E state. The ground spin triplet is split into a singlet |0> and doublet |±1> separated by 2.88 GHz. Recent transient hole-burning studies [9] show that the excited state fine structure splitting is mostly dominated by spin-spin interactions. However, the detailed energetic structure of the excited 3E state is still under debate. Ensemble site-selective methods, i.e. spectral hole burning and ODMR, allow selection of centers by the optical frequency of 0-0 transition. This may still select centers with a range of strain and, hence, the various centers can be affected differently by magnetic and electric fields. The single molecule approach, which is free of any ensemble averaging, is thus a suitable method to probe the fine structure of N-V defects.

Single N-V centers at room temperature

"After a first report on the optical detection of single defect center in diamond [10], this system attached attention because of the unique photostability. Detection of single N-V centers at room temperature can be achieved using confocal microscopy (for details of single molecule spectroscopy technique see ref. [11] and references therein). A suitable defects concentration can be reached by the control of diamond irradiation dose. N-V defects can be efficiently excited via the strong phonon sideband using the 514 nm line of an Ar laser. Fig. 2a shows a confocal image of a N-V center containing diamond. Sample was irradiated with an electron dose 1012 e/cm2 and annealed for 2 hours at T= 800 ºC. Average distance between N-V centers is larger than resolution of a confocal microscope allowing to address single defects optically. Some of untreated synthetic 1b-type diamonds show defect concentrations suitable for single center detection [3]. The figure on the right hand side shows a 30x30 mm area of untreated diamond from Drukker International containing a few N-V defects. Contrary to single molecules isolated in a matrix or immobilized on surfaces, no photobleaching of N-V defects was observed, even under excitation intensity close to the saturation conditions. The reported saturation signal of about 5000 photocounts per second [3] is more than an order of magnitude lower that one can expect from two-level system with a lifetime corresponding to the N-V center radiative decay rate (11.6 ns [12]). Weak saturation fluorescence intensity can be explained in terms of presence of the metastable singlet state. Surprisingly, ensemble experiments give only a few indication of the presence of this state. Nearly degenerate four-wave mixing show a relaxation time of 0.6 s, which has been attributed to the decay of a metastable state [13]. In single center experiments, evidence for the population of the metastable state comes from the bunching observed in fluorescence autocorrelation function g(2)(t). At low excitation power the decay time of the g(2)(t) approach to 1/k31, whereas at excitation close to the saturation the decay rate is given by 2/k23.

Florescence autocorrelation
Excitation - 514 nm
Room temperature
2 MHz-2 kHz
Florescence autocorrelation
Excitation - 532 nm
Room temperature
4 MHz 1 MHz-
Transient hole burning87 kHz--
Four wave mixing T=80 K--1.7 Hz

The table shows the population and depopulation rates of metastable state obtained in single-center and ensemble experiments. Note that the back rate is only 4.11 times slower than forward indicating the presence of thermally or/and optically activated deshelving occurring at room temperature. Intersystem crossing rates derived from different single center measurements are in a good agreement. However, disagreement between ensemble and hole- burning studies is remarkable. Single center and ensemble should provide the same results if they are carried out on the same type of defects. Difference between ensemble and single center studies can be attributed to selectivity of the single- molecule method. When detecting single centers, one selects them on certain photophysical parameters, i.e. brightness. Those centers can strongly differ from the average data obtained for ensemble with a variety of chemically identical but photophysically different defects. Examples of such inhomogeniety are experiments on spectral hole-burning. Narrow spectral holes can be only found on the red edge of the inhomogeneously broadened absorption band. Low temperature experiments on single centers also provide a possibility to select individual centers with different transition frequencies and can be helpful for finding correlations between different photophysical parameters of N-V centers.

Spectroscopy of single N-V defects at low temperature

Numerous ensemble experiments like non-degenerated four-wave mixing[13], photon echo [14], and spectral hole-burning [15][16][9] have been directed to unravel the energy level scheme of the N-V center. Under selective excitation of the pure electronic transition (637 nm) at low temperature N-V centers show a remarkable nonselective bleaching of the luminescence with two characteristic time constants of 1 ms and 40 s [17]. The fast (1 ms) component can be attributed to relaxation between spin sublevels. The slow (40 s) rate is possibly related to the decay of the metastable singlet state. This state becomes an efficient energy trap at low temperatures because of blocking of a thermally activated depopulation channel. Such trapping of energy is an important obstacle to detect single N-V centers at low temperature, because of strong decrease of saturation fluorescence intensity. The excited state population p for the system with a metastable state can be obtained by calculation of the excited-state population for a two-level system p2, and correct the result by including an additional factor: where k23 and t31 are the population rate and lifetime of a metastable singlet state respectively. Correction factor plays an important role at low temperature when depopulation rate of a singlet slows down to its spontaneous decay rate (seconds). This shelving explains the decrease of the single N-V center luminescence beyond a detectable level at temperatures below 80 K. Based on the temperature dependence of the single center fluorescence intensity, it was found that the metastable 1A state lies 300 cm-1 below the excited 3E state [18]. Ab initio cluster calculations reported by Goss et al. [19] predict 3E-1A splitting of 866 cm-1, which is in a reasonable agreement with experimentally determined value. Single centers, however, can be detected also at low temperature using so-called "deshelving technique." The idea of the method is to apply an additional laser, which repumps the population back towards short living excited triplet 3E state. With the aid of deshelving it was possible to detect fluorescence excitation spectra of the 3A-3E transition at T = 4 K [20]. Surprisingly, the linewidth of the excitation line was more than 4 orders of magnitude broader than expected lifetime-limited value (12 MHz). The origin of the strong line broadening can not be attributed only to dephasing because single center lines are 100 times broader than observed hole width (0.5 GHz). One of the possible interpretations could be an influence of a matrix. It must be taken into account that atomic nitrogen is always present in a high (500 ppm and more) concentration in 1b type diamond matrix. The single substitutional nitrogen is a donor with ionization energy of 1.7 eV (note, that acting of nitrogen as a donor can explain the fact, that the negative charge state is typical for defect centers in type 1b diamond). Single molecule experiments are always carried out using strong excitation intensities leading to effective nitrogen ionization in the local environment of the N-V defect. Resulting local field fluctuations may lead to a broadening of the spectral line. On the other hand, read-out process in hole-burning experiments requires moderate laser intensities allowing to observe narrow holes (the holewidth, however, still broader than a lifetime-limited value). It was found, that spectral diffusion rate is particularly sensitive to deshelving laser illumination (488 nm) [21]. Summarizing single center data available so far, one can conclude that low- temperature excitation spectra are strongly influenced by the effect of a metastable state population. This bottleneck can be overcome by using repumping laser, but deshelving procedure leads to the strong broadening of spectral lines. A metastable singlet state lies close to excited triplet state of N-V center, and a possible solution of the problem would be to find a system where a strain in the diamond lattice would lead to an inverse energetic order of the 3E and 1A states. Diamond nanocrystals showing strongly increased inhomogeneous broadening of the spectral lines can be a suitable candidate for high-resolution low-temperature spectroscopy on single N-V centers.

Fluorescence spectroscopy on single N-V centers in diamond nanocrystals

So-called "explosion diamonds" – diamond nanocrystals obtained by detonation [22], – have been of great interest in a last decade. X-ray crystallography and Raman spectroscopy show that even nanometer-size crystals show a diamond- type crystal lattice [23]. Material properties of ultradisperse diamond differ from natural and synthetic diamond because considerable part of its atoms are located close to the grain surface and feel strongly its influence. As a consequence, the crystal and band structure is distorted within an important part of the volume and differs from those of a bulk diamond [24]. Photoluminescence properties of N3 defects (three nitrogen atoms replacing carbons in a lattice) are found to have distinguishing features compared with a spectrum in a bulk diamond: short-wavelength spectral shift and the change in the shape of the spectral line [25]. Recently detection of single N-V centers in diamond nanocrystals have been reported [26][27][28][29]]. Grangier and co-workers [28] observed a radiative decay rate modification related to modification of the surrounding refraction index. Low temperature emission spectra of N-V center in ultradisperse diamond differs from those in a bulk crystal by higher inhomogeneous broadening of their pure electronic transition lines. The figure on the left hand side shows the fluorescence emission spectrum of the single nanocrystallite at T = 1.6 K compared with the spectrum of the single defect in a bulk crystal. Note that well-resolved zero-phonon lines corresponding to emission of four N-V centers present in the nanocrystal are visible. A spread in the line position is about 5 nm indicating a large inhomogeneity of the local field in the vicinity of the center. The phonon sideband structure of the defect in a nanocrystal is different compared to bulk diamond. Spectral lines of N-V centers in ultra disperse diamond are narrower than in type 1b bulk samples. This effect can be related to the suppression of the dephasing processes because of the quantization of phonon states in nanocrystals. Quantization of acoustic phonon modes leading to unusual temperature dependence of homogeneous linewidth was recently observed in single quantum dots [30].


Optical access to single defect centers at low temperature opens the door to unraveling of their electronic structure. Two limitations of low-temperature technique are discussed. The first is related to the existence of a metastable singlet state irreversibly capturing excitation energy at cryogenic temperatures. A second bottleneck is the broadening of single central lines because of influence of the matrix. The paramagnetic nature of a triplet ground state makes the system suitable for single-spin nuclear magnetic resonance experiments. Efficient deshelving of a trapping state at room temperature make individual color centers a good candidate for production of single photons on demand.

[1] Koizumi, S., Watanabi, K., Hasegawa, M., and Kanda, H Science 292 (2001) 1899
[2] The Properties of Diamond, Academic Press, London, New York (1979)
[3] Kurtsiefer, C., Mayer, S., Zarda, P., and Weinfurter, P. Physical Review Letters 85 (2000) 290
[4] Martin, J., Wannemacher, R., Teichert, J., Bischoff, L., and Köhler, B. Applied Physics Letters 75 (1999) 3096
[5] Brouri, R, Beveratos, A., Poizat, J.-P., and Grangier, P. Physical Review A 62 (2000) 0683171-1
[6] Mita, Y. Physical Review B 53 (1996) 11360
[7] Reddy, N. R. S., Manson, N. B., and Krausz, E. R. Journal of Luminescence 38 (1987) 46
[8] van Oort, E. Photodynamics of N-V defect in diamond, PhD thesis, University of Amsterdam, 1990
[9] Martin J.P.D. Journal of Luminescence 81 (1999) 237
[10] Gruber, A., Dräbenstedt, A., Tietz, C., Fleury, L., Wrachtrup, J., and von Borczyskowski, C. Science 276 (1997) 2012
[11] Tamarat Ph., Maali A., Lounis B., and Orrit M. The Journal of Physical Chemistry A 104 (2000) 1
[12] Collins, A. T., Thomaz, M. F., and Jorge, M. I. B. Journal of Physics C: Solid State Physics 16 (1983) 2177
[13] Redman, D. A., Brown, S., Sands, R. H., and Rand, S. C. Physical Review Letters 67 (1991) 3420
[14] Lenef, A., Brown, S. W., Redman, D. A., Rand, S. C., Shigley, J., and Fritsch, E. Physical Review B 53 (1996) 13427
[15] Harley, R. T., Henderson, M. J., and Macfarlane, R. M. Journal of Physics C: Solid State Physics 17 (1984) L233
[16] Redman, D., Brown, S., and Rand, S. C. Journal of the optical Society of America B 9 (1992) 768
[17] Kiisk, V., Palm, V., Suisalu, A., and Sildos, I. Journal of Luminescence 86 (2000) 349
[18] Dräbenstedt, A., Fleury, L., Tietz, C., Jelezko, F., Kilin, S., Nizovtzev, A., and Wrachtrup, J. Physical Review B 60 (1999) 11503
[19] Goss, J., Jones, R., Breuer, S., Briddon, P., and Oberg, S. Physical Review Letters 77 (1996) 3041
[20] Dräbenstedt, A., Tietz, C, Jelezko, F, Kilin, S., Nizovtsev, A. P., and Wrachtrup, J. Acta Physica Polonica A 96 (1999) 665
[21] Jelezko, F, Tietz, C., and Wrachtrup, J. to be published (2001)
[22] Lyamkin, A., Petrov, E., Ershov, A., Sakovitch, G., Staver, A., and Titov, V Sov.Phys.Dokl. 33 (1988) 705
[23] Yoshikawa, M., Mori, Y., Obata, H., Maegawa, N., Katagiri, G., Ishida, H., and Ishitani, A. Applied Physics Letters 67 (1995) 694
[24] Kopman, M., Terukov, E, Gordeev, S., Zhukov, S., and Nikolaev, Yu. Phys.Solid State 39 (1997) 1928
[25] Mikov, S., Igo, A., and Gorelik, V. Phys.Solid State 41 (1999) 1012
[26] Wrachtrup, J. Physikalische Blätter 56 (2000) 14
[27] Dräbenstedt, A. Hochauflösende Spektroskopie und Mikroskopie einzelner Moleküle und Farbzentren bei tiefen Temperaturen, PhD Thesis, Chemnitz University of Technology, 1999.
[28] Beveratos, A., Brouri, R, Gacoin, T., Poizat, J.-P., and Grangier, P. arXiv:quant-phys (2001) 0104028
[29] Kühn, S., Hettigh, C., Schmitt, C., Poizat, J.-P., and Sandoghdar, V. J.Microscopy (to be published) (2001)
[30] Palinginis, P. and Wang, H. Applied Physics Letters 78 (2001) 1541
[31] Davies, G. Properties and growth of Diamond, INSPEC, London (1994)