Chinese Journal of Chemical Physics  2019, Vol. 32 Issue (5): 521-524

The article information

Wen-long Zhang, Li-ren Lou, Wei Zhu, Guan-zhong Wang

Enhancing Fluorescence of Shallow Nitrogen-Vacancy Centers in Diamond by Surface Coating with Titanium Oxide Layers

Chinese Journal of Chemical Physics, 2019, 32(5): 521-524

http://dx.doi.org/10.1063/1674-0068/cjcp1904069

Article history

Accepted on: April 29, 2019
Enhancing Fluorescence of Shallow Nitrogen-Vacancy Centers in Diamond by Surface Coating with Titanium Oxide Layers
Wen-long Zhang , Li-ren Lou , Wei Zhu , Guan-zhong Wang
Dated: Received on April 9, 2019; Accepted on April 29, 2019
Hefei National Laboratory for Physical Sciences at the Microscale, and Department of Physics, University of Science and Technology of China, Hefei 230026, China
Abstract: We present an enhancement of the fluorescence of shallow ($<$10 nm) nitrogen-vacancy (NV$^-$) centers by using atomic layer deposition to deposit titanium oxide layers on the diamond surface. In this way, the shallow NV$^-$ center charge states were stabilized, leading to the increasing fluorescence intensity of about 2 times. This surface coating technique could produce a protective layer of controllable thickness without any damages to the solid-state quantum system surface, which might be an approach to the further passivation or packaging techniques for the solid-state quantum devices.
Key words: Shallow NV- center    Titanium oxide coating layers    Charge state stabilization    Fluorescence enhancement

The nitrogen-vacancy (NV) center in diamond has been considered to be prospective in both quantum information science and precision measurement owing to its outstanding spin properties at room temperature. For example, quantum registers based on the coupling of two NV centers have been experimentally demonstrated [1, 2]. Measuring techniques of electric and magnetic fields, temperature etc., by using NV$^-$ centers in diamond have driven to maturity stage. It is of interest to make the solid-state quantum system of diamond NV$^-$ center be a device for application. However, influenced by the surface energy band bending, the charge state of a shallow NV$^-$ center becomes unstable [4, 5], resulting in the reduction of its fluorescence intensity, which is not conducive to the applications of NV$^-$ centers. Up to now, a number of experimental schemes have been presented by many groups to enhance the shallow NV$^-$ center fluorescence intensity. One familiar method is to make the diamond surface O-terminated or N-terminated etc., called the surface termination [5-7]. However, thermal oxidation [8-10] or plasma etching [11-13] used to obtain the satisfactory surface termination damages the diamond surface irreversibly, etching away the very shallow NV$^-$ centers. Hence a method which can produce a protective layer without any surface damages to prevent from etching shallow centers away is being sought.

Surface coating with protective layers is a potential approach to solve the problems above. In previous work, the method of depositing silicon coating layers on the diamond surface was demonstrated, but no fluorescence enhancement of shallow NV$^-$ centers was reported [15]. On the other hand, some materials like aluminium oxide will be excited to produce fluorescence under the laser irradiation, which improves the background fluorescence noises. Therefore, a proper material to be used as the dielectric coating layers for diamond NV$^-$ centers should be carefully selected and tested.

In this letter, we demonstrate an enhancement of the fluorescence of shallow ($<$10 nm) NV$^-$ centers by depositing titanium oxide layers on the diamond surface without any surface damages. Titanium oxide is nonmagnetic and non-fluorescent at room temperature, therefore it is suitable for the surface coating layer. By performing atomic layer deposition (ALD), titanium oxide was deposited in nanoscale so that the protective layer thickness was controllable. Before and after ALD, we evaluated the diamond surface chemical bond precisely using X-ray photoelectron spectroscopy (XPS) and investigated the effect of the titanium oxide layer on the charge stability of a shallow NV$^-$ center in diamond using photoluminescence (PL) intensity mapping and Rabi oscillation measurements.

An electronic grade (100)-oriented, 2 mm$\times$2 mm$\times$ 0.5 mm sized diamond substrate from "Element Six" ([13C]=1.1%, [N]$<$5 ppb) was used for the experiments. By using electron beam lithography, an array made of 60 nm diameter apertures as well as position marks (used to mark and track every NV center) was patterned on a 300 nm thick polymethyl methacrylate (PMMA) layer deposited beforehand on the diamond substrate surface [3, 13]. By ion implantation with the 14N$_{2}$$^{+} molecule energy of 5 keV and a fluence of 0.65\times10^{11} 14N_{2}$$^{+}$ cm$^{-2}$ through the apertures on the patterned PMMA layer template, the single NV center array was created in the diamond substrate with a probable depth of about 5 nm obtained by stopping and range of ions in matter (SRIM) in FIG. 1(a). The substrate was tilted at 7° to suppress the ion-channeling effect during implantation [16]. Afterwards, the implanted diamond substrate was annealed at 1050 ℃ in vacuum at 2$\times$10$^{-5}$ Pa for 2 h, and then boiled in a 1:1:1 mixture of sulfuric, nitric, and perchloric acids to remove any graphitic contamination at the surface.

 FIG. 1 (a) The SRIM simulations of the depth profiles of the implanted nitrogen atoms at an energy of 2.5 keV. (b) The boundary of masked and unmasked diamond surface areas characterized by AFM after ALD. (c) Ti 2p narrow scan spectra of the diamond sample surface before and after titanium oxide deposition. (d) Wide-range XPS spectra of the diamond sample surface before and after titanium oxide deposition.

A part of the diamond sample surface was masked with PMMA beforehand in order to characterize the thickness of the coating layer afterwards. The sample was then put into the ALD chamber (Atomic Layer Deposition machine, Picosun, Sunale R-200 Advanced) for depositing titanium oxide layer (Ti$_{x}$O$_{y}$ layer, abbreviated as TOL hereafter). The first atomic deposition cycle was H$_{2}$O cycle to make the hydroxyl ions absorb on the diamond surface. Then the second cycle of Ti deposition was applied so that the hydrogen ions of the hydroxy would be replaced by titanium ions. We repeated the above two cycles sequentially to deposit the oxide layer on the diamond surface. The average TOL thickness was confirmed to be about 4 nm by using AFM to measure the height of the boundary of masked and unmasked areas, as presented in FIG. 1(b). Before and after ALD, the diamond sample surface was characterized with XPS to verify if TOL had been deposited on the diamond sample surface.

FIG. 1(c) exhibits the high-resolution XPS titanium 2p spectra. The dotted line shows the result before ALD from which no titanium peaks can be obtained, while the solid line represents the result after ALD from which distinct peaks belonging to Ti 2p spectrum can be observed around 459 eV. Another proof to certify the success of depositing oxide protective layer on the diamond surface was the wide-range XPS spectra exhibited in FIG. 1(d), showing the relative intensity of oxygen and carbon peaks. The intensity of oxygen peak raised remarkably relative to that of carbon peak, revealing that after ALD, the main component of characterized sample surface changed from carbon to oxygen, i.e., the oxide layer had been indeed deposited on the diamond surface by ALD. Moreover, the appearance of Ti peaks in FIG. 1(d, bottom) demonstrated that the protective layer was made up of titanium oxide.

A custom-built confocal microscope system was used to observe and evaluate the properties of the single shallow NV$^-$ centers. With the position marks, we were able to track and measure the same single NV$^-$ centers (verified by the $g^{(2)}(0)$ values much less than 0.5) before and after the deposition. FIG. 2(a) shows the confocal PL-intensity mapping of the same center (NV-01) after different steps, obtained with a 532 nm excitation laser. This shallow center was created after ion implantation with the photon counts of about 6$\times$10$^3$ cps. After depositing TOL, the photon counts increased to 1.4$\times$10$^4$ cps, about twice the original. Similarly, the fluorescence intensities of all tracked single NV$^-$ centers showed consistent changes as presented in FIG. 2(b). The increase in fluorescence intensity indicated that the charge states of shallow NV$^-$ centers were stabilized [7].

 FIG. 2 (a) PL-intensity mapping images of a representative shallow single NV$^-$ center before and after ALD with the 500-nm scale bar. (b) The increasing fluorescence intensities of 8 shallow single NV$^-$ centers after surface coating.

To further demonstrate the stability of the charge states after depositing TOL, the Rabi oscillations of shallow centers were measured. The contrast of Rabi oscillation $C$ is a significant parameter for evaluating the charge stability of the NV$^-$ center against the NV0 reported previously [17], which is defined as

 $\begin{eqnarray} C=\frac{F_{\text{top}}-F_{\text{bottom}}}{F_{\text{middle}}}=\frac{2(F_{\text{top}}-F_{\text{bottom}})} {F_{\text{top}}+F_{\text{bottom}}} \end{eqnarray}$ (1)

where $F_{\text{top}}, F_{\text{bottom}}$, and $F_{\text{middle}}$=$(F_{\text{top}}$+$F_{\text{bottom}})/2$ are the normalized fluorescence intensities of the top, bottom, and middle of the Rabi oscillations, respectively [7, 17]. The decrease in the $C$ value means that the charge state of a certain NV$^-$ center transfers to that of the NV0 center. The $C$ values of a typical single shallow NV$^-$ center before and after deposition TOL were derived from its Rabi oscillations in FIG. 3 (a) and (b), respectively. After deposition, the $C$ value ($C$=0.31) was higher than that ($C$=0.17) before deposition, which meant that the charge state of this NV$^-$ center with TOL on the diamond surface was sufficiently stable. The $C$ values of all measured shallow NV$^-$ centers exhibited obvious increase after depositon TOL and the statistical results of all 20 measured single NV$^-$ centers are shown in FIG. 3 (c) and (d). The spread of the Rabi contrast can be approximated by a Gaussian distribution. It can be obtained that the average $C$ value of all measured centers also increased after depositing TOL, corresponding to the results of single centers. Thus, the charge states of shallow NV$^-$ centers were indeed stabilized by surface coating with titanium oxide protective layers. This considerable stabilization in shallow center charge state could have two reasons: on one hand, surface effects are likely to be reduced under the oxide coating layer [14]; on the other hand, residual lattice defects from the implantation might be eliminated during the ALD process. In summary, the influences of the interface of diamond and external environment were reduced by surface coating with titanium oxide layers, hence the charge states of shallow centers could become stable.

 FIG. 3 Rabi oscillations of the same shallow single NV$^-$ center (a) before depositing titanium oxide and (b) after depositing titanium oxide. The statistical results of Rabi oscillation contrast of all the 20 measured shallow single NV$^-$ centers (c) before depositing titanium oxide and (d) after depositing titanium oxide. The solid lines in (c) and (d) are obtained by Gaussian fitting.

Consequently, we demonstrate that our surface coating with titanium oxide layers by ALD is effective for the charge stability of shallow NV$^-$ centers in diamond, revealed by the increasing Rabi contrast. The shallow center fluorescence intensity can be enhanced by 2 times. This work will enhance the future performance of applications using NV$^-$ centers and also provide a possible approach for passivation and packaging techniques of the solid-state quantum systems.

ACKNOWLEDGEMENTS

This work was supported by the National Natural Science Foundation of China (No.11374280 and No.50772110). The authors wish to thank Guo-ping Guo, Jie You and Yang Li from the Key Lab of Quantum Information for the support of electron beam lithography. We also thank Ming-ling Li at University of Science and Technology of China for the technical support of ALD.

Reference
 [1] P. Neumann, R. Kolesov, B. Naydenov, J. Beck, F. Rempp, M. Steiner, V. Jacques, G. Balasubramanian, M. Markham, D. Twitchen, S. Pezzagna, J. Meijer, J. Twamley, F. Jelezko, and J. Wrachtrup, Nat. Phys. 6, 249(2010). DOI:10.1038/nphys1536 [2] F. Dolde, I. Jakobi, B. Naydenov, N. Zhao, S. Pezzagna, C. Trautmann, J. Meijer, P. Neumann, F. Jelezko, and J. Wrachtrup, Nat. Phys. 9, 139(2013). DOI:10.1038/nphys2545 [3] J. Wang, F. Feng, J. Zhang, J. Chen, Z. Zheng, L. Guo, W. Zhang, X. Song, G. Guo, L. Fan, C. Zou, L. Lou, W. Zhu, and G. Wang, Phys. Rev. B 91, 155404(2015). DOI:10.1103/PhysRevB.91.155404 [4] A. N. Newell, D. A. Dowdell, and D. H. Santamore, J. Appl. Phys. 120, 383(2016). [5] M. Hauf, B. Grotz, B. Naydenov, M. Dankerl, S. Pezzagna, J. Meijer, F. Jelezko, J. Wrachtrup, M. Stutzmann, F. Reinhard, and J. A. Garrido, Phys. Rev. B 83, 081304(2011). DOI:10.1103/PhysRevB.83.081304 [6] M. Kaviani, P. DeałĄk, B. Aradi, T. Frauenheim, J. P. Chou, and A. Gali, Nano lett. 14, 4772(2014). DOI:10.1021/nl501927y [7] T. Kageura, K. Kato, H. Yamano, E. Suaebah, M. Kajiya, S. Kawai, M. Inaba, T. Tanii, M. Haruyama, K. Yamada, S. Onoda, W. Kada, O. Hanaizumi, T. Teraji, J. Isoya, S. Kono, and H. Kawarada, Appl. Phys. Express 10, 055503(2017). DOI:10.7567/APEX.10.055503 [8] M. Loretz, S. Pezzagna, J. Meijer, and C. Degen, Appl. Phys. Lett. 104, 033102(2014). DOI:10.1063/1.4862749 [9] M. Kim, H. J. Mamin, M. H. Sherwood, and C. T. Rettner, Appl. Phys. Lett. 105, 042406(2014). DOI:10.1063/1.4891839 [10] J. Wang, W. Zhang, J. Zhang, J. You, Y. Li, G. Guo, F. Feng, X. Song, L. Lou, W. Zhu, and G. Wang, Nanoscale 8, 5780(2016). DOI:10.1039/C5NR08690F [11] S. Cui, A. S. Greenspon, K. Ohno, B. A. Myers, A. C. B. Jayich, D. D. Awschalom, and E. L. Hu, Nano Lett. 15, 2887(2015). DOI:10.1021/acs.nanolett.5b00457 [12] F. Fávaro de Oliveira, S. A. Momenzadeh, Y. Wang, M. Konuma, M. Markham, A. M. Edmonds, A. Denisenko, and J. Wrachtrup, Appl. Phys. Lett. 107, 073107(2015). DOI:10.1063/1.4929356 [13] W. Zhang, J. Zhang, J. Wang, F. Feng, S. Lin, L. Lou, W. Zhu, and G. Wang, Phys. Rev. B 96, 235443(2017). DOI:10.1103/PhysRevB.96.235443 [14] T. Staudacher, F. Ziem, L. Häussler, R. Stöhr, S. Steinert, F. Reinhard, J. Scharpf, A. Denisenko, and J. Wrachtrup, Appl. Phys. Lett. 101, 212401(2012). DOI:10.1063/1.4767144 [15] Y. Romach, C. Müller, T. Unden, L. Rogers, T. Isoda, K. Itoh, M. Markham, A. Stacey, J. Meijer, S. Pezzagna, B. Naydenov, L. P. McGuinness, N. Bar-Gill, and F. Jelezko, Phys. Rev. Lett. 114, 017601(2015). DOI:10.1103/PhysRevLett.114.017601 [16] J. Nakata, Phys. Rev. B 60, 2747(1999). DOI:10.1103/PhysRevB.60.2747 [17] H. Yamano, S. Kawai, K. Kato, T. Kageura, M. Inaba, T. Okada, I. Higashimata, M. Haruyama, T. Tanii, K. Yamada, S. Onoda, W. Kada, O. Hanaizumi, T. Teraji, J. Isoya, and H. Kawarada, Jpn. J. Appl. Phys. 56, 04C(2017).