Nanoscale Depth Control of Implanted Shallow Silicon Vacancies in Silicon Carbide

Nanoscale Depth Control of Implanted Shallow Silicon Vacancies in Silicon Carbide

We are an expert of semiconductor wafers in semiconductor industry, and we offer technology support and wafers selling for thousands of univerisities and industrial customers by our decades experience, including Cornell University, Stanford Univeristy, Peking University, Shandong Univerity, university of south carolina, Caltech Faraon lab (USA) ,University of Califomia, Ivine (USA),Singapore MIT Alliance for Research and Technology Centre (SMART), West Virginia University, Purdue Univerity, University of Califomnia, Los Angeles ,King Abdullah University of Science & Technology, Massachusetts Institute of Technology, University of Houston, University of Wisconsin, University of Science and Technology of China etc. Take the article: “Nanoscale Depth Control of Implanted Shallow Silicon Vacancies in Silicon Carbide” for example.

Published by:

Qiang Li,   *ab   Jun-Feng Wang, *ab   Fei-Fei Yan, *ab   Ze-Di Cheng, *ab   Zheng-Hao Liu, *ab   Kun Zhou, *ab   Li-Ping Guo, *c   Xiong Zhou, *c   Wei-Ping Zhang, *c   Xiu-Xia Wang, *d   Wei Huang, *e   Jin-Shi Xu,  *ab   Chuan-Feng Li  *ab  and  Guang-Can Guo *ab

Author affiliations
* Corresponding authors
a: CAS Key Laboratory of Quantum Information, University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China
b: CAS Center for Excellence in Quantum Information and Quantum Physics, University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China
c: Key Laboratory of Artificial Micro- and Nano-structures of Ministry of Education and School of Physics and Technology, Wuhan University, Wuhan, Hubei 430072, People’s Republic of China
d: Center for Micro- and Nanoscale Research and Fabrication, University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China
e: Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 201800, People’s Republic of China

Abstract

Color centers in silicon carbide have recently attracted broad interest as high bright single photon sources and defect spins with long coherence time at room temperature. There have been several methods to generate silicon vacancy defects with excellent spin properties in silicon carbide, such as electron irradiation and ion implantation. However, little is known about the depth distribution and nanoscale depth control of the shallow defects. Here, a method is presented to precisely control the depths of the ion implantation induced shallow silicon vacancy defects in silicon carbide by using reactive ion etching with little surface damage. After optimizing the major etching parameters, a slow and stable etching rate of about 5.5 ± 0.5 nm min−1 can be obtained. By successive nanoscale plasma etching, the shallow defects are brought close to the surface step by step. The photoluminescence spectrum and optically detected magnetic resonance spectra are measured, which confirm that there were no plasma-induced optical and spin property changes of the defects. By tracing the mean counts of the remaining defects after each etching process, the depth distribution of the defects can be obtained for various implantation conditions. Moreover, the spin coherence time T2* of the generated VSi defects is detected at different etch depths, which greatly decreases when the depth is less than 25 nm. The method of nanoscale depth control of silicon vacancies would pave the way for investigating the surface spin properties and the applications in nanoscale sensing and quantum photonics.

 

Received 12th July 2019, Accepted 30th July 2019

DOI: 10.1039/c9nr05938e

rsc.li/nanoscale

 

1 Introduction

In recent years, defects in silicon carbide (SiC) have emerged as a promising platform for quantum information processing from bright single photon sources1–7 to long coherence time spin qubits.8–13 Compared with the well-known diamond with nitrogen-vacancy (NV) centers, SiC exists in many different polytypes and has advantages such as commercial availability, wafer-scale growth, and mature micro and nanofabrication. Particularly, silicon vacancies (VSi) in 4H-SiC have stood out as favorable spin-qubit candidates that can be polarized and readout by laser and controlled by microwaves at room temperature.12–18 Moreover, they have near-infrared photo- stability, single photon emissions, and long spin coherence time (160 μs) at room temperature.12,14 It has been extended to above 20 ms using dynamical decoupling15 which is comparable to the results of NV centers in diamond.16 Furthermore, the  VSi   defects  have  also been used for high sensitivity quantum sensing, such as magnetic17–19 and temperature19 sensing, and microwave emitters.20

Except the natural VSi defects, there are some typical methods to generate VSi defects in 4H-SiC: high energy electron and neutron irradiation12–14 and ion implantation, 21–23 etc. However, little is known about the precise depth distribution of the generated VSi defects. Moreover, in order to investigate the surface spin properties of SiC24 and achieve high sensitivity to probe the external target samples, 25, 26 it is significant to generate shallow VSi defects.21 Since the detected signal relies on the dipolar coupling to the targeted external spins, inversely proportional to r3, where r is the distance between the VSi defects and target spins, the nanoscale depth control of shallow VSi defects becomes critical for the SiC- based nanoscale sensing.12, 24–27

In this work, we provide a method to nanoscale control the depth of the implantation generated shallow VSi in 4H-SiC using RIE etching. Generally, there are two methods to etch SiC, namely wet etching28 and dry etching including inductively coupled plasma (ICP) 29, 30 and reactive ion etching (RIE).31 Compared with the wet etching of SiC, which has some disadvantages including harsh etching conditions, fairly rough surface morphologies and poor uniformity of etching rate, the dry etching provides a smooth and uniform method for selective etching. Besides, dry etching carried out using RIE provides a slower etching than that of ICP, so the former is more suitable for nanoscale control of etching depths.

We investigated the effects of four major parameters on the etching rate and obtained a slow and stable etching rate of about 5.5 ± 0.5 nm min−1. As a comparison, we found that for the as-received samples without implantation, there was no obvious change in the   photoluminescence (PL) intensity before and after plasma etching. For the implanted samples, through nanoscale plasma etching, the implantation generated shallow defects are brought close to the surface step by step. We measure the photoluminescence (PL) and optically detected magnetic resonance (ODMR) spectra to confirm that there are no significant plasma-induced optical and spin property changes of the detects.32, 33 Those results indicate that the RIE etching process caused little damage to the SiC surface. The effect is similar to that of soft plasma etching32 or removed plasma etching schemes33 for diamond but is different from that of the destructive surface etching.34 Then, we traced the mean counts of the implantation generated shallow defects at each step through which their depth distribution for various implantation conditions can be obtained. The photostability of shallow VSi defects remains unchanged during the etching process, even when those defects are very close to the surface (d < 10 nm). Furthermore, the spin coherence times are measured at different etched depths, which are used to investigate the influence of the surface noise. The nanoscale depth control of the shallow VSi defects with little surface damage is critical for its applications in investigating surface spin properties, 24 nanoscale sensing12, 24–27 and quantum photonics.21, 30, 35

2     Experiments

In the experiment, high-purity 4H-SiC epitaxy growth wafers with a thickness of about 7 μm are used (Xiamen Powerway Advanced Material Co., Ltd). The etching processes were performed by reactive ion etching (RIE, Oxford, Plasma Pro NGP 80) using sulfur hexafluoride (SF6) and oxygen (O2) as reactive gases. The process of etching and characterizing SiC samples is shown in Fig. 1. A 1.4 μm thick layer of positive photoresist S1813 is spin-coated onto the surface of the SiC sample which was spun at a speed of 4000 rpm as shown in Fig. 1(a). Through standard processes of ultraviolet lithography (SUSS MABA6) and development, a grating pattern photoresist is fabricated on the surface of the samples as a mark to protect SiC from etching, which is denoted in Fig. 1(b). Fig. 1(c) shows that the surface of the SiC sample is etched in the RIE and the grating pattern of photoresist is transferred to the SiC substrate. After removing the residual photoresist, we measured the surface morphology and the etched depth with an atomic force micro- scope (AFM, Bruker, Demension Icon), as shown in Fig. 1(d). The edges of the grating pattern are used as reference positions to determine the etching rate and depth. The PL intensity and ODMR signals at each etching step are measured by a home-built confocal microscopy.

 etching and characterizing of the SiC sample

Fig. 1 The etching and characterizing of the SiC sample. (a) A layer of 1.4 μm thick positive photoresist is spin-coated on the surface of the SiC sample. (b) Through standard processes of ultraviolet lithography and development, a grating structure pattern is transferred to the photoresist. (c) The surface of the SiC sample is etched using RIE and the photoresist plays the role of a mark to protect SiC from etching. (d) An AFM is used to measure the etching rate and depth. A home-built confocal microscopy (denoted as laser) is used to measure the PL intensity and ODMR signals.

There are four major parameters influencing the etching rate of SiC, including the power of the RIE, the pressure of the plasma chamber, the SF6 flow rate and the O2 flow rate. We comprehensively investigated their effects. The etching rate is found to increase almost linearly with the RF power, even when the power is relatively low, as shown in Fig. 2(a). It is because the increase in RF power promotes the dissociation of SF6 in the chamber and also enhances the kinetic energy and directionality of the colliding particles. However, Fig. 2(b) shows that the etching rate decreases as the pressure inside the plasma chamber rises. It is due to the fact that a higher pressure inside the chamber means a higher density of gas inside the reactive area, which increases the probability of collision of ionized particles with plasma sheath. These results are similar to that of etching diamonds with only oxygen gas.36 Moreover, we further studied the influence of the SF6 and O2 flow rates on etching rate, which are shown in Fig. 2(c) and (d), respectively. The effect of increasing SF6 flow rate on the SiC etching rate is small when the O2 flow rate is set to 2 sccm. Generally, increasing the SF6 flow rate can promote the etching rate before it saturates at a certain level. On the other hand, the etching rate increases with O2 concentration and reaches a maximum, which would then decrease again. The addition of O2 can maintain a high fluorine (F) radical concen- tration, which is responsible for the etching of SiC, by reacting with unsaturated CFn and SFn. The result is similar to that of inductively coupled plasma (ICP) etching.29

Etching rate as a function of various parameters

Fig. 2 Etching rate as a function of various parameters: (a) the RF power, (b) the pressure inside the chamber, (c) the SF6 flow rate, (d) the O2 flow rate. The other parameters are fixed and shown in the label box of each corresponding case.

In the experiment, a slow and stable etching rate of the SiC substrate is obtained by setting the SF6 flow rate to 9 sccm, the O2 flow rate to 2 sccm, the pressure inside chamber to 20 mTorr and the RF power to 30 W, respectively. Twelve etching experiments with different etching time under the same conditions mentioned above are performed and indicate an etching rate of 5.5 ± 0.5 nm min−1 through the surface morphology analysis with the AFM. A representative scanned image of an area 10 × 10 μm of the substrate etched for 60 seconds is shown in Fig. 3(a), which exhibits a clear boundary in the middle of the image and demonstrates that the height distribution of the etched area is uniform in nanoscale. Particularly, the height distribution along the blue dashed line in Fig. 3(a) is shown in Fig. 3(b). The average height is about 6 nm, which is consistent with the etching rate. Moreover, a magnified image of the blue 1 × 1 μm square area in Fig. 3(a) is shown in Fig. 3(c), in which the height distribution along a randomly chosen red dashed line is displayed in Fig. 3(d). These results demonstrate that we can realize a stable nanoscale selected depth control.

Surface morphology of the etched SiC sample characterized by the AFM

Fig. 3 Surface morphology of the etched SiC sample characterized by the AFM. (a) Scanned image of a representative area 10 × 10 μm of the substrate etched for 60 s. (b) The height distribution along the blue line in (a). (c) The scanned image of a 1 × 1 μm square area described by the blue box in (a). (d) The height distribution along the red line in (c).

In order to verify the properties of the implanted shallow VSi defects at different depths, the as-received 4H-SiC samples are first implanted by 20 keV C+, He+ and H2+ ions with the same dose of 2 × 1013 cm−2, respectively. A home-built scanning confocal microscopy is then used (details can be found in Methods). Here, we show the results for the case with C+ ion implantation. The similar results with He+ and H2+ ion implantation can be found in the ESI. Fig. 4(a) shows the confocal fluorescent image of the VSi defects before etching. The mean counts is about 1 Mcps. The sample is then sequentially etched with RIE for 20 nm, 40 nm and 60 nm, respectively. The corresponding scanning images of fluorescence intensity in the same area are shown in Fig. 4(b), (c) and (d), respectively, in which the fluorescence intensity and the density of VSi defects are reduced as the etched depth increases. There are distinct boundaries in the right and upper areas of each figure, even when the layer of SiC containing VSi defects is almost etched (Fig. 4(d)), which are taken as positioning marks. The room temperature (RT) PL spectra of VSi defects at different etch depths are detected by a grating spectrometer (Horiba, iHR550), which are shown in Fig. 4(e). The black, red, blue and green solid lines correspond to the RT PL spectra of SiC samples etched 0 nm, 20 nm, 40 nm and 60 nm, respectively, all of which agree with the previous results of VSi defects.12,30 We further measured the low temperature (LT, 5 K) PL spectra of the VSi defects in SiC etched different depths, which are shown in Fig. 4(f). The two characterized zero phonon lines (ZPLs) of V1 (861 nm) and V2 (915 nm) according to the two types of VSi defects are obviously detected in each LT spectrum.14,37 Both the RT and LT spectra   prove that the etching process does not change the optical properties of VSi defects. We also measured the time traces of PL intensity of the VSi defects in the samples after etching for different depths at room temperature, in which the time bin is 10 ms. No photon blinking or photon bleaching is observed during the plasma etching process,32,33 as shown in Fig. 4(g). Moreover, we detected the ODMR spectra for different etched depths at room temperature. The number of scans (detection time) of the ODMR signals is fixed to be 31 times, which is the same for all the samples with different etched depths. The average results of a single measurement are exhibited in Fig. 4(h). Inferred from the fitting of the ODMR data, the resonant frequencies range from 70.6 MHz to 71.5 MHz with the full width at half maximum (FWHM) ranging from 12.5 MHz to 18.5 MHz, respectively, which are also consistent with previous results.12,14 Moreover, the contrasts of ODMR signals of VSi defects are almost same, which demonstrate that the etching processes do not change the spin properties.

Depth dependence of PL and ODMR spectra of VSi defects generated by 20 keV C+ ions with a dose of 2 × 1013 cm−2

Fig. 4 Depth dependence of PL and ODMR spectra of VSi defects generated by 20 keV C+ ions with a dose of 2 × 1013 cm−2. (a)–(d) Fluorescent images at the same position after different etched depths of 0 nm, 20 nm, 40 nm and 60 nm, respectively. (e) The room temperature PL spectra of the VSi defects at different etched depths. (f) The low temperature PL spectra of the VSi defects at different etched depths. Two characterized peaks labeled as V1 (861 nm) and V2 (915 nm), which are denoted by vertical black dashed lines, correspond to the zero photon lines of two types of VSi defects in SiC. (g) The time traces of PL intensity from the VSi defects in the samples after different  etched depths at room temperature, with sampling bin δt = 10 ms. (h) The ODMR signals of the VSi defects after different etched depths at room temperature. The red solid lines are the Lorentz fitting to the data.

We then studied the depth distribution of VSi defects gener-ated using the same implanted ions (C+) with different energies and a same dose of 2 × 1013 cm−2 by tracing the mean PL intensity during the etching process. In this experiment, the mean PL intensity is calculated in a scanning area of about  20 × 20 μm and is normalized by the mean PL intensity before etching. For each step, a 30s plasma etching process is performed on the surface of SiC sample, corresponding to an etched depth of about 2.75 nm. In this way, the defects can approach to the surface of the sample step by step until all the centers disappeared, from which the initial depth distribution of the shallow VSi defects can be obtained. The normalized mean PL intensity of the shallow VSi defects generated by 10 keV C+ ions implantation as a function of the etched depth is shown in Fig. 5(a). The blue dots represent the experimental results and the red solid line demonstrates the fitting result obtained from corresponding a Gaussian complementary error function32 (see ESI†). The normalized mean intensity decreases quickly when the etched depth is less than about 40 nm. After that, the mean PL intensity decreases slowly, and finally almost all the VSi defects disappeared at the etched depth of about 65 nm. The corresponding Gaussian depth distribution deduced from the fitting result is then denoted as the blue solid line in Fig. 5(b). The mean depth distribution is about 26.4 nm with a standard deviation of 17.2 nm. We further show the corresponding simulation results with the SRIM (stopping and range of ions in matter),32,38,39 which are shown as gray dashed lines in Fig. 5(a) and (b). Although the depth profile of VSi defects deduced from the experimental results are similar with the results of SRIM simulation, there are clear deviations between them. It may be due to the effect of ion channels in real experiments resulting in a deeper penetration of the ions propagating along the low-index crystallographic axes and planes, which are similar to the results of implanted NV centers in diamond.32,39

Depth distribution of shallow VSi defects generated by C+ ion implantation

Fig. 5   Depth distribution of shallow VSi defects generated by C+ ion implantation. (a), (c) and (e) show the detected intensities as a function of the etched depth for the cases with C+ implantation energy of 10 keV, 30 keV and 40 keV, respectively. The blue dots, pink dots and purple dots in (a), (c) and (e) represent the mean PL intensities, respectively, with the red solid lines and gray dashed lines representing the Gaussian fittings and the SRIM simulation results, respectively. The blue, pink and purple solid lines in (b), (d) and (f) represent the depth distribution of VSi defects according the data shown in (a), (c) and (e), respectively, with the gray dashed lines representing the SRIM simulation results.

The depth distribution of the VSi defects in SiC generated by ion implantation depends on the initial energy. Fig. 5(c) and (e) show the mean PL intensity versus etched depth with the implanted energies of 30 keV and 40 keV, respectively. The pink dots in Fig. 5(c) and purple dots in Fig. 5(e) represent the experimental results with the red solid lines and the gray dashed lines representing the results of fitting and SRIM simulation, respectively. The duration time of each plasma etching process is set to 60 s, corresponding to an etched depth of about 5.5 nm in each step and the processes are repeated for 25 times in Fig. 5(c) and 30 times in Fig. 5(e). It is shown that the VSi defects implanted by 30 keV and 40 keV C+ ions have vanished at the etched depth of about 130 and 150 nm, respectively. The corresponding depth distributions of Fig. 5(c) and (e) are shown as the pink solid line in Fig. 5(d) and purple solid line in Fig. 5(f), respectively. The mean depths in Fig. 5(d) and (f) are 54.5 nm and 66.7 nm with the standard deviation of 33.3 nm and 41.4 nm, respectively. The gray dashed lines in Fig. 5(d) and (f) represent the simulation results of SRIM. It can be found that the depth of VSi defects increases as the implanted energy increases, which is coincident with the simulating results through the SRIM. However, the effect of the ion channel appears in all situations with different implantation energies.

We further investigate and control the depth of the VSi defects in SiC samples implanted by different types of ions. In the experiment, the C+, He+ and H2+ ions are used and the implantation energies are fixed to 20 keV with the same dose of 2 × 1013 cm−2. Fig. 6(a), (c) and (e) show the mean PL intensity versus etched depth of SiC samples which are pre-implanted with C+, He+ and H2+ ions, respectively. The brown dots in Fig. 6(a), green dots in Fig. 6(c) and orange dots in Fig. 6(e) represent the experimental results with the red solid lines exhibiting the fitting results and the gray dashed lines demonstrating the results of SRIM simulation. It is shown that VSi defects generated by 20 keV C+, He+ and H2+ ion implantation vanished at the etched depth of 110 nm, 190 nm and 170 nm, respectively. The corresponding depth distribution of the VSi defects generated by different implantation ions can be obtained, which are shown as the brown solid line in Fig. 6(b), green solid line in Fig. 6(d) and orange solid line in Fig. 6(f), respectively. The mean depth shown in Fig. 6(b), (d) and (f) are 38.9 nm, 99.6 nm and 97.1 nm with standard deviations of 25.3 nm, 30.5 nm and 23.1 nm, respectively. The distribution of VSi defects with C+ ion implantation is shallowest, which is due to the heaviest ion mass than the other two. There are also deviations between the experimental results and the SRIM simulation (gray dashed lines in Fig. 6(b), (d), (f), which implies that the effect of ion channel appears in all situations with different implantation ions.

Depth distribution of shallow VSi defects generated by different ions implantation with the same energy of 20 keV

Fig. 6 Depth distribution of shallow VSi defects generated by different ions implantation with the same energy of 20 keV. (a), (c) and (e) show the detected intensities as a function of the etched depth for the cases with C+, He+, H2 + implantations, respectively. The brown dots, green dots and orange dots in (a), (c) and (e) represent the mean PL intensities, respectively, with the red solid lines and gray dashed lines representing the Gaussian fittings and the SRIM simulation results, respectively. The brown, green and orange solid lines in (b), (d) and (f) represent the depth distribution of VSi defects according to the data shown in (a), (c) and (e), respectively, with the dashed lines representing the SRIM simulation results.

We studied the spin coherence properties of shallow VSi defects after different etched depths. Since a large amount of PL photons are needed in the lock-in methods, the SiC samples used in our experiments are implanted by 20 keV or 4 keV He+ ions with a dose of 2 × 1013 cm−2. Fig. 7(a) and (b) demonstrate the experimental results of Rabi oscillations at room temperature for the SiC samples implanted by 4 keV He+ ions without any etching process and after being etched about 20 nm, respectively. Comparing the decay of Rabi oscillations of VSi defect spins before and after etching, the decay of the latter is found to be faster than that of the former. The decay of Rabi oscillations is one of the characteristics of the environmental noise spectrum. The results imply that the coherence time of the defect spins after being etched about 20 nm is less than  that  without  any  etching  process.40,41 Furthermore, Fig. 7(c) demonstrates a representative measurement of Ramsey oscillations for the SiC samples implanted by 4 keV He+ ions without any etching process. Inferred from the fitting, the T2* of VSi defect spins is deduced to be 174 ± 13 ns. Fig. 7(d) summarizes the experimental results of T2* deduced from the fittings of Ramsey signals versus the depth of defects, in which the upper horizontal axis represents the etched depth and the lower horizontal axis represents the corresponding remaining depth defined as the difference between the etched depth and the depth  that  implanted defects just disappear completely. Purple dots represent the results of the 20 keV He+ ions implanted SiC samples. The T2* ranges from 177 ± 25 ns to 205 ± 22 ns and significant reduction of T2* is not observed with the increase of etched depth. The phenomenon may result from the reason that the remaining VSi defects are not so shallow that the surface noise cannot affect the coherence of most of the spins.42,43 In order to obtain a shallower depth distribution and relatively higher density of VSi defects,44 we also implant 4 keV He+ ions into the SiC samples with a dose of 2 × 1013  cm−2. The corresponding T2* of VSi defect spins without any etching process and after being etched about 20 nm are 174 ± 13 ns and 145 ± 16 ns, respectively. A significant decrease in T2* is observed after being etched about 20 nm, which is similar to the previous results of NV centers in diamonds.42,43

spin coherence properties and PL time traces of shallow VSi defects in the samples implanted by 4 keV or 20 keV He+ ions after different etched depths

Fig. 7   The spin coherence properties and PL time traces of shallow VSi defects in the samples implanted by 4 keV or 20 keV He+ ions after different etched depths. (a) and (b) The Rabi oscillation of shallow VSi defects in 4H-SiC sample implanted by 4 keV He+ ions before any etching process and after being etched about 20 nm, respectively (c) The Ramsey signal of shallow VSi defects produced by 4 keV He+  ions before any etching process. (d) T2* deduced from the fittings of Ramsey oscillation versus different etched depth and remaining depth. (e) The PL time traces with sampling bins δt = 10 ms at room temperature, obtained from VSi defects in the samples implanted by 4 keV He+ ions after different etched depth. (f) The enlargement of PL time traces in the randomly selected black window of (e). (g) The photon count statistics of time traces shown in (f)

Since the photostability is important for the defects in solid state,21 we also measured the time traces of PL intensity with a time bin of 10 ms for the samples with different etch depths and the results are shown in Fig. 7(e). It is worth noting that the PL intensity from the shallow VSi defects is stable without blinking or bleaching even after being etched about 40 nm, in which the average depth of VSi defects is estimated to be less than 5 nm. The enlargement of PL time traces in the randomly selected black window of Fig. 7(e) is shown in Fig. 7(f). The corresponding PL counting statistics are consistent with the Gaussian distribution and the statistical histogram and Gaussian fitting line are shown in the Fig. 7(g). The experimental results demonstrate that our nanoscale etching methods do not influence the photostability of the VSi defects in 4H-SiC, even when the depth of defects are very shallow (d < 10 nm).

These results indicate that shallow VSi defects near the surface of 4H-SiC experience larger environment noise, compared with those in bulk materials, which may be attributed to the influence of surface noise. In the future, systematic studies on the actual depth of individual VSi or VSiVC defects by measuring the spin coherence properties42,45 and the surface noise spectrum24,46 of the electron spin are extremely important for practical application of probing external sample spins.12,25–27 Nevertheless, these tasks are full of obstacles and challenges to date, also beyond the research scope of this paper.

3   Conclusions

In summary, we provide a little surface damage method to control the nanoscale depths of the VSi defects in 4H-SiC generated by ion implantation using reactive ion etching. We investigated the optimal etching parameters and obtained a slow and stable etching rate of about 5.5 ± 0.5 nm min−1. The measurement and control accuracy of the depths of the VSi defects can then be guaranteed. In our experiment, the shallow VSi defects are generated by implanting different ions (C+, He+ and H2+) with different energies. The optical and spin properties of these defects are detected during the etching process and no significant etching-induced changes are observed. For the as-received samples without implantation, the PL intensity is essentially not observed to increase at the same area after and before plasma etching. These results indicated that the RIE etching process caused little damage to the SiC surface, which played an important role in suppressing the increase of additional surface noise sources when make the defects approach SiC surface. The depth distribution of the VSi defects in the SiC samples generated by implantation can then be obtained by tracing the mean counts of the remaining defects step by step during the etching processes. The depth of VSi defects generated by He+ and H2+ were deeper than that generated by C+ ions with the same energy. With the increasing of implanted energy, the depth of defects will also increase. Finally, the influence of defect depth on the properties of spin coherence is investigated by the Ramsey and Rabi measurements. The results show that T2* greatly decreases when the depth is less than 25 nm.

The nanoscale depth control of the shallow VSi defects is important for investigating the surface spin properties of SiC, near-surface VSi dynamics and nanoscale sensing.12,24–26 As compared with the NV centers of diamonds, highly sensitive nanoscale sensing based on the semiconductor material SiC is more fascinating for its wide application in the realm of   technologically mature devices. On the other hand, the etching method may also be used to prepare a single VSi defect in SiC with specific depth and to combine the nanostructures, such as nanowires30 and photonic crystals,35,47 to enhance the extraction efficiency of single photons, which are significant for quantum communication48 and quantum computation.49 Moreover, nanoscale depth control can also be used for other types of defects such as divacancy in SiC.10,11,50

4   Methods

Optical measurement

We used a home-built scanning confocal microscopy with an objective of 0.65 NA (Olympus, LCPLN50XIR) to determine the position and measure the mean PL intensity of remaining VSi defects in SiC samples. In all of the optical measurements, a 730 nm continuous wave laser with the pump power set to 1 mW, filtered by a band pass filter with a center wavelength of 720 nm and a bandwidth of 24 nm (Semrock, FF01-720-25), was employed to excite those color centers. A dichroic beams-plitter (Semrock, FF801-Di02-25×36) was used to separate the exciting laser and fluorescence. For the fluorescence measurement at room temperature, the SiC samples are mounted on a three-axis piezoelectric stage (PI, E-727.3SD) which is used to scan the samples. For the experiments at low temperature of 5 K, the SiC samples are mounted on a cooling stage in the Montana Instruments cryostat and a two-axis Galvo scanning system with silver-coated mirrors (Thorlabs, GVS012) is used to scan the exciting laser. The PL signal filtered by an 808  nm long pass filter (Semrock, LP02-808RU-25) is coupled to a single mode fiber or a multiple mode fiber and then guided to a single photon counting module (Excelitas, SPCM-AQRH-14) to detect the PL intensity, or to a grating spectrometer (Horiba, iHR 550) to measure the PL spectrum.

Spin coherence time measurement

For ODMR measurement, a microwave signal was generated by a signal generator (Agilent, N5181B) and then gated by a switch (Mini-Circuits, AZSWA-2-50DR+). After being amplified by an amplifier (Mini-Circuits, ZHL-20W-13SW+), it is fed to a 50 μm-width microwave copper wire above the surface of the SiC sample. The ODMR signal is obtained by modulating the microwave with a time period of about 2 ms and detecting the change in PL intensity by a single photon counting module (Excelitas, SPCM-AQRH-14).21,22 For the Rabi and Ramsey experiments, we used an infrared objective with N.A. of 0.85 (Olympus, LCPLN100XIR), the PL signals are collected through a multimode fiber and detected by a Femto (OE-200-Si). We measured the Rabi oscillations and performed the Ramsey experiments using the lock-in methods.44,51

Conflicts of interest

There are no conflicts to declare.

 

Acknowledgements

The authors thank Mao-Sheng Ma for his assistance in carrying out the XPS measurements at Instruments’ Center for Physical Science, University of Science and Technology of China. This work was supported by the National Key Research and Development Program of China (Grant No. 2016YFA0302700 and 2017YFA0304100), the National Natural Science Foundation of China (Grants No. 61725504, 11774335 and 11821404), the Key Research Program of Frontier Sciences, CAS (No. QYZDY-SSW-SLH003), Anhui Initiative in Quantum Information Technologies (AHY060300 and AHY020100), the Fundamental Research Funds for the Central Universities (Grants No. WK2030380017 and WK2470000026). This work was partially carried out at the USTC Center for Micro and Nanoscale Research and Fabrication.

 

References

1  S. Castelletto, B. C. Johnson, V. Ivády, N. Stavrias, T. Umeda, A. Gali and T. Ohshima, Nat. Mater., 2014, 13, 151–156.

2  S.   Castelletto,   B.   C.   Johnson,   C.   Zachreson,   D. Beke, I.  Balogh, T.  Ohshima I.  Aharonovich and  A.  Gali, ACS         Nano, 2014, 8, 7938–7947.

3  A.  Lohrmann,  N.  Iwamoto,  Z.  Bodrog,  S.    Castelletto,  T.  Ohshima, T.  J. Karle, A. Gali, S. Prawer, J. C. McCallum       and B. C. Johnson, Nat. Commun., 2015, 6, 7783.

4  B.   Lienhard,   T.    Schröder,   S.   Mouradian,   F.   Dolde,     T. T.  Tran, I. Aharonovich and D. Englund, Optica, 2016,   3,    768–774.

5  I.  A.  Khramtsov,  A.  A.  Vyshnevyy  and  D.  Y.  Fedyanin, npj   Quantum. Inf., 2018, 4, 15.

6  J.  F.  Wang,  Y.  Zhou,  Z.  Y.  Wang,  A.  Rasmita,  J.  Q.  Yang,  X. J. Li, H. J. von Bardeleben and W. B. Gao, Nat. Commun., 2018, 9, 4106.

7  Q. Li, J. Y. Zhou, Z. H. Liu, J. S. Xu, C. F. Li and G. C. Guo, J. Semicond., 2019, 40, 072902.

8  W. F. Koehl, B. B. Buckley, F. J. Heremans, G. Calusine   and D. D. Awschalom, Nature, 2011, 479, 84–87.

9   A.   L.   Falk,   B.   B.   Buckley,   G.   Calusine,   W.   F.    Koehl,      V. V. Dobrovitski, A. Politi, C. A. Zorman, P. X. L. Feng and         D. D. Awschalom, Nat. Commun., 2013, 4, 1819.

10  H. Seo,  A.  L.  Falk,  P.  V.  Klimov,  K.  C.  Miao,  G.  Galli and         D. D. Awschalom, Nat. Commun., 2016, 7, 12935.

11  D.  J.   Christle,   A.  L.   Falk,   P.   Andrich,   P.   V.   Klimov,      J.   U.   Hassan,   N.   T.   Son,   E.   Janzén,   T.   Ohshima and       D. D. Awschalom, Nat. Mater., 2015, 14, 160–163.

12  M. Widmann, S. Y.  Lee, T.  Rendler, N. T.  Son, H.     Fedder,      S.   Paik,   L.   P.   Yang,   N.   Zhao,   S.   Yang,   I.    Booker, A. Denisenko, M. Jamali, S. A. Momenzadeh, I. Gerhardt,    T.   Ohshima,  A.  Gali,  E.  Janzén  and  J.  Wrachtrup,    Nat.       Mater., 2015, 14, 164–168.

13  R.  Nagy,  M.  Widmann,  M.  Niethammer,  D.  B.  R. Dasari,      I.  Gerhardt,  Ö.  O.  Soykal,  M.  Radulaski,  T.   Ohshima, J.  Vučković,  N.  T.   Son,  I.  G.  Ivanov,  S.  E.    Economou, C.  Bonato,  S.  Y.  Lee  and  J.  Wrachtrup,  Phys.  Rev. Appl., 2018, 9, 034022.

14  F.  Fuchs,  B.  Stender,  M.  Trupke,  D.  Simin,  J.  Pflaum,  V.  Dyakonov and G. V.  Astakhov, Nat. Commun., 2015,    6,   7578.

15  D. Simin, H. Kraus, A. Sperlich, T. Ohshima, G. V.  Astakhov         and V. Dyakonov, Phys. Rev. B: Condens. Matter Mater. Phys.,       2017, 95, 161201(R).

16  N.   Bar-Gill,   L.  M.  Pham,   A.  Jarmola,   D.   Budker   and R. L. Walsworth, Nat. Commun., 2013, 4, 1743.

17  D. Simin, F. Fuchs, H. Kraus, A. Sperlich, P. G. Baranov,  G. V. Astakhov and V. Dyakonov, Phys. Rev. Appl., 2015,   4,       014009.

18  M.  Niethammer,  M.  Widmann,  S.  Y.   Lee,   P.     Stenberg,          O.  Kordina,  T.   Ohshima,  N.  T.   Son,  E.  Janzén      and  J. Wrachtrup, Phys. Rev. Appl., 2016, 6, 034001.

19  H.  Kraus,  V.  A.  Soltamov,  F.  Fuchs,  D.  Simin,  A.  Sperlich,          P.  G. Baranov, G. V.  Astakhov and V.  Dyakonov, Sci.  Rep.,   2014, 4, 5303.

20  H.  Kraus,  V.   A.  Soltamov,  D.  Riedel,  S.  Väth,  F.       Fuchs, A. Sperlich, P. G. Baranov, V. Dyakonov and G. V. Astakhov,       Nat. Phys., 2014, 10, 157–162.

21  J.  F.  Wang,  Y.  Zhou,  X.  M.  Zhang,  F.  C.  Liu,  Y.  Li,  K. Li,     Z. Liu, G. Z. Wang and W. B. Gao, Phys. Rev. Appl., 2017, 7,        064021.

22  J.  F.   Wang,  X.  M.  Zhang,  Y.  Zhou,  K.  Li,  Z.  Y.       Wang,        P. Peddibhotla, F. C. Liu, S. Bauerdick, A. Rudzinski, Z.   Liu        and W.  B. Gao, ACS Photonics, 2017, 4,  1054–1059.

23  H.  Kraus,  D.  Simin,  C.  Kasper,  Y.   Suda,  S.     Kawabata,  W.  Kada, T.  Honda,  Y.  Hijikata, T.  Ohshima,  V. Dyakonov   and G. V. Astakhov, Nano Lett., 2017, 17, 2865–2870.

24  T.     Rosskopf,    A.    Dussaux,    K.    Ohashi,    M.     Loretz,         R.  Schirhagl,  H.  Watanabe,  S.  Shikata,  K.  M.  Itoh and         C. L. Degen, Phys. Rev. Lett., 2014, 112, 147602.

25  F.  Z.  Shi, Q.  Zhang,  P.  F.  Wang,  H. B.  Sun,  J. R.   Wang,   X.  Rong,  M.  Chen,  C.  Y.   Ju,  F.  Reinhard,  H.  W.   Chen,        J. Wrachtrup, J. F.  Wang  and J. F.  Du, Science, 2015,    347,     1135–1138.

26  N.  Aslam,  M.  Pfender,  P.  Neumann,  R.  Reuter,  A. Zappe,      F.   F.   de  Oliveira,  A.  Denisenko,  H.  Sumiya,  S.    Onoda,        J. Isoya and J. Wrachtrup, Science, 2017, 357, 67–71.

27  J. R. Maze, P. L. Stanwix, J. S. Hodges, S. Hong, J. M. Taylor,          P.  Cappellaro, L. Jiang, M. V.  G. Dutt, E. Togan, A. S.   Zibrov,    A. Yacoby, R. L. Walsworth and M. D. Lukin, Nature,  2008,         455, 644–647.

28  R.  Cheung,  Silicon  carbide  micro  electromechanical  systems      for harsh environments, World Scientific., 2006.

29  J. Choi, Silicon carbide nanowires: from fabrication to related       devices, PHD Thesis, 2013.

30  M. Radulaski, M. Widmann, M. Niethammer, J. L. Zhang,    S.-Y.   Lee,   T.   Rendler,   K.   G.   Lagoudakis,   N.   T.  Son,        E. Janzén, T.  Ohshima, J. Wrachtrup and J. Vučković,   Nano  Lett., 2017, 17,  1782–1786.

31  P. H. Yih, V.  Saxena and A. J. Steckl, Phys. Status Solidi    B,    1997, 202, 605–642.

32  F. F. de Oliveira, S. A. Momenzadeh, Y. Wang, M.   Konuma,       M.    Markham,    A.    M.    Edmonds,    A.    Denisenko   and J. Wrachtrup, Appl. Phys. Lett., 2015, 107,   073107.

33  S.    Cui,    A.    S.    Greenspon,    K.    Ohno,    B.    A.    Myers,        A. C. B. Jayich, D. D. Awschalom and E. L. Hu, Nano   Lett.,          2015, 15, 2887–2891.

34  M.  Kim,  H.  J.  Mamin,  M.  H.  Sherwood,  C.  T.   Rettner,   J. Frommer and D. Rugar, Appl. Phys. Lett., 2014, 105, 042406.

35  M.  Radulaski,  T.  M.  Babinec,  S.  Buckley,  A.   Rundquist,        J.  Provine,  K.  Alassaad,  G.  Ferro  and  J.  Vučković,   Opt.  Express, 2013, 21, 32623–32629.

36  J. Evtimova, W.  Kulisch, C. Petkov, E. Petkov, F.    Schnabel,   J. P. Reithmaier and C. Popov, Diamond Relat. Mater.,  2013,       36, 58–63.

37  T.   C.  Hain,  F.   Fuchs,  V.   A.  Soltamov,  P.   G.    Baranov,      G. V.  Astakhov, T.  Hertel and V.  Dyakonov, J. Appl.   Phys.,   2014, 115, 133508.

38  J. F. Ziegler, M. D. Ziegler and J. P. Biersack, Nucl.   Instrum.     Methods Phys. Res., Sect. B, 2010, 268,  1818–1823.

39  S.   Pezzagna,   D.   Rogalla,   D.   Wildanger,   J.   Meijer and A. Zaitsev, New J. Phys., 2011, 13, 035024.

40  V.    V.    Dobrovitski,    A.    E.    Feiguin,    R.    Hanson   and         D. D. Awschalom, Phys. Rev. Lett., 2009, 102, 237601.

41  L.  P.  McGuinness,  L.  T.  Hall,  A.  Stacey,  D.  A. Simpson,        C. D. Hill, J. H. Cole, K. Ganesan, B. C. Gibson, S. Prawer,  P. Mulvaney, F.  Jelezko, J. Wrachtrup, R. E. Scholten and    L. C. L. Hollenberg, New J. Phys., 2013, 15,   073042.

42  J. F. Wang, W. L. Zhang, J. Zhang, J. You, Y. Li, G. P. Guo,        F.  P. Feng, X. R. Song, L. R. Lou, W.  Zhu and G. Z.    Wang,       Nanoscale, 2016, 8, 5780–5785.

43  W.  L.  Zhang, J.  Zhang, J.  F.  Wang, F.  P.  Feng, S. R.   Lin,         L. Lou, W.  Zhu and G. Z. Wang, Phys. Rev. B: Condens.        Matter Mater. Phys., 2017, 96, 235443.

44  J. F. Wang, Q. Li, F. F. Yan, H. Liu, G. P. Guo, W. P. Zhang,        X. Z, L. P.  Guo, Z. H. Lin, J. M. Cui, X. Y.  Xu, J. S. Xu, C. F. Li and G. C. Guo, ACS Photonics, 2019, 6,    1736–1743.

45  B.   K.   Ofori-Okai,   S.   Pezzagna,   K.   Chang,   M. Loretz,    R.  Schirhagl,  Y.   Tao,  B.  A.  Moores,  K.     Groot-Berning,       J. Meijer and C. L. Degen, Phys. Rev. B: Condens. Matter     Mater. Phys., 2012, 86, 081406(R).

46  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 and F.    Jelezko,      Phys. Rev. Lett., 2015, 114,  017601.

47  A.   Lohrmann,   B.   C.   Johnson,   J.   C.   McCallum      and    S. Castelletto, Rep. Prog. Phys., 2017, 80,   034502.48      G.  Brassard,  N.  Lütkenhaus,  T.  Mor and  B.  C. Sanders,   Phys. Rev. Lett., 2000, 85, 1330–1333.

49  E. Knill, R. Laflamme and G. J. Milburn, Nature, 2001,  409,        46–52.

50  D. J.  Christle, P.  V.  Klimov,  C. F.  de  las Casas, K.    Szász,         V.   Ivády,  V.   Jokubavicius,  J.  U.  Hassan,  M.     Syväjärvi,          W.  F.  Koehl, T.  Ohshima, N. T.  Son, E. Janzén, Á. Gali  and       D. D. Awschalom, Phys. Rev. X, 2017, 7, 021046.

51  Y.  Zhou, J. F.  Wang, X. M. Zhang, K. Li, J. M. Cai and        W. B. Gao, Phys. Rev. Appl., 2017, 8, 044015.

Share this post