Silicon-vacancy (VSi) centers in silicon carbide (SiC) are expected to serve as solid qubits, which can be used in quantum computing and sensing. As a new controllable color center fabrication method, femtosecond (fs) laser writing has been gradually applied in the preparation of VSi in SiC. In this study, 4H-SiC was directly written by an fs laser and characterized at 293 K by atomic force microscopy, confocal photoluminescence (PL), and Raman spectroscopy. PL signals of VSi were found and analyzed using 785 nm laser excitation by means of depth profiling and two-dimensional mapping. The influence of machining parameters on the VSi formation was analyzed, and the three-dimensional distribution of VSi defects in the fs laser writing of 4H-SiC was established.

  • Silicon vacancies were fabricated by femtosecond laser writing of 4H-SiC without annealing.

  • A nondestructive, 3D, and rapid measurement method was used to characterize the distribution of silicon vacancy.

  • The defect formation model of silicon vacancy fabricated by femtosecond laser was established.

Silicon carbide (SiC) is a promising semiconductor with a wide bandgap, high critical electric field strength, high saturation drift velocity, and mature crystal growth technology.1,2 Because of these properties, SiC has become one of the most attractive materials for quantum sensing and metrology. Functional defects, such as color centers in SiC, have advantages of high brightness, emission in the red to near-IR spectral regions, long spin coherence time, and excellent magneto-optical properties, which make them very attractive for future applications in photonic quantum computing, network, and communication.3–7 In particular, silicon-vacancy (VSi) defects in SiC have been proven to serve as solid qubits, which can be used in quantum computing and sensing.8–12 

Many methods, such as particle irradiation13,14 and focused ion beams,15 have been used in the creation methods of color centers in SiC. To advance the application of quantum technology for color centers in SiC, it is necessary to ensure that the prepared color centers have reliable spin-related properties and to improve positioning accuracy. Kraus et al. presented a controlled generation of quantum centers in SiC via three-dimensional (3D) proton beam writing. The generation depth and resolution can be predicted by matching the proton energy to the material’s stopping power.16 Wang et al. provided a method to generate on-demand shallow single VSi defect arrays using ion implantation with a high conversion efficiency of approximately 80% and high concentration defect ensembles in 4H-SiC.17 Although the above described methods are mature, the complexity of the preparation of devices with certain requirements on the depth and position is relatively high, resulting in low preparation efficiency, unnecessary residual damage to the crystal lattice, and degraded properties of color centers.18,19 Recently, femtosecond (fs) laser writing has been proven to be a successful method in creating color centers in SiC. Castelletto et al. showed that VSi defects were successfully created in a hexagonal SiC using an fs laser, and the photoluminescence (PL) intensity of VSi in the irradiated area was related to the laser irradiation energy and annealing process.20 Chen et al. reported the controlled creation of single VSi centers in 4H-SiC using laser writing without any post-annealing process and discussed the mechanism of the laser writing process.21 

Although many studies showed the PL of defects in SiC, little attention has been paid to the 3D distribution of defects and color centers. Li et al. proposed a low surface disruption 3D characterization method using reactive ion etching to control the nanoscale depth of VSi defects in 4H-SiC generated via ion implantation.22 Compared with the destructive measurement method mentioned above, confocal spectroscopy is nondestructive, 3D, and rapid. Song et al. characterized the defects of PL signal in helium ion implanted with 4H-SiC via depth profiling and SWIFT mapping and revealed the 3D distribution of produced defects on this basis.23 In this work, the defect distribution of the fs laser-written 4H-SiC was characterized via confocal PL and Raman spectroscopy using two-dimensional (2D) mapping and depth profiling. The effects of the repetition rate, scanning speed, and pulse number of the fs laser on the VSi defect distribution were studied.

The host material used in this study was nitrogen-doped (0001) 4H-SiC substrate with 4° ± 0.5° off-axis from Xiamen Powerway Advanced Material Co., Ltd. A 1030-nm-wavelength, 350-fs-duration laser was used to create vacancies in 4H-SiC. The laser writing system irradiated the sample surface in a single-pulse irradiation mode through a two-axis laser scanner head (SCANcube III 10), with a fixed laser energy of 5 μJ per pulse. A six-line array was made with different repetition rates, pulse numbers, and scanning speeds. Irradiated regions with different parameters were labeled as R1–R6, as shown in Fig. 1 and Table 1. After surface cleaning using the piranha solution to remove surface organic contaminants, the laser-written sample was characterized.

Fig. 1.

Optical micrograph of a laser-written 4H-SiC surface. (a) Sample surface optical micrograph under 10× objective lens; (b)–(g) optical micrographs of R1–R6 under 100× objective lens.

Fig. 1.

Optical micrograph of a laser-written 4H-SiC surface. (a) Sample surface optical micrograph under 10× objective lens; (b)–(g) optical micrographs of R1–R6 under 100× objective lens.

Close modal
Table 1.

Details of the laser-written regions.

Label Repetition rate (MHz) Pulse number Scanning speed (m/s)
R1  0.1 
R2 
R3 
R4  10 
R5 
R6  10 
Label Repetition rate (MHz) Pulse number Scanning speed (m/s)
R1  0.1 
R2 
R3 
R4  10 
R5 
R6  10 

The microscopic morphology of the machining areas with different parameters was characterized by the Dimension Icon atomic force microscope (AFM) from Bruker Corporation. The irradiated regions were scanned over an area of approximately 900 μm2 using a non-contact tapping mode. PL and Raman spectra were measured using Horiba XploRA PLUS and iHR550 confocal microscope in a backscattering geometric configuration at room temperature. Using 405 and 785 nm continuous-wave laser excitation with 1 μm spot size, a single spectrum focusing on the sample surface, depth profiles focusing on different depths of the sample (from 10 μm below the sample surface to surface), and 2D mapping were all measured in each irradiated region. A 405 nm laser has a small penetration depth, which is sensitive to near-surface signals. The damage area of the sample was mainly concentrated on the near surface, so the E2 mode intensity was measured under 405 nm laser excitation to determine the damage degree of SiC. The excitation efficiency of VSi changes for different excitation wavelengths. Hain et al. found that an optimal excitation wavelength for PL excitation measurements is approximately 770 nm.24 Compared with the 405 nm laser, a 785 nm laser is more suitable for the measurement of VSi signals. Objective lens (100× magnification) with a numerical aperture of 0.9 and 600 lines/mm grating were used. Single-spectrum and depth profiling measurements were performed in the standard working mode, and the acquisition time was 10 s for each spectrum for two accumulations. 2D mapping measurement was conducted in the SWIFT working mode (ultra-fast Raman imaging), and the acquisition time was 1 s for each spectrum for one accumulation. Some primary defects in SiC will produce fluorescent signals, so a baseline correction was performed in the single-spectrum and depth profiling measurements. The SWIFT working mode can only collect a small range of spectrum, so the baseline correction was not performed.

Fig. 1 shows the optical micrograph of the laser-written 4H-SiC. The permanent damage is visible on the SiC surface. For single-pulse fs laser machining, the damaged regions appear as adjacent single points, which is evident in R1 (100 kHz, 9 m/s, single pulse). For R2 (1 MHz, 9 m/s, single pulse), due to the increase in the repetition frequency, the distance between adjacent pulses decreases, and it is still possible to distinguish individual machining points at high magnification. For multi-pulse fs laser machining regions (R3–R6), the aperture of the ablative regions will increase with the increase in the number of irradiation pulses, resulting in the damage area appearing in the form of a continuous line. As the pulse number increases and the scanning speed slows down, the etching marks of SiC become more obvious, and the line width also increases.

The surface morphology of the laser-written regions was characterized by an AFM. Fig. 2(a)(f) shows the 2D/3D topography maps and cross-sectional profiles for the irradiated regions (R1–R6). The average width and maximum depth of the machining area in the optical micrograph and AFM results are shown in Table 2. The results of the AFM are consistent with those of the optical micrographs, single-pulse fs laser machining regions appear as adjacent single points, and multi-pulse fs laser machining regions appear as machining lines. For R1, from the height profile along the machining direction and perpendicular to the machining direction, there is 5–20 nm swelling around the machining point, and part of the central region is lower than the sample surface, which is consistent with the spatial distribution of the Gaussian beam intensity. For R2, two distinct uplift regions (one from 7 μm to 17 μm and another from 20 μm to 30 μm) can be distinguished from the height profile parallel to the machining direction, corresponding to two adjacent processing points in the optical micrograph. From the height profile parallel to the machining direction, the machining point is similar to the structure in R1. Instead of a material removal, the single-pulse fs laser machining regions produced tiny swells. This is because the sample is monocrystalline SiC, the amorphous layer is formed, and less sputtering effect occurred after the fs laser writing under a low irradiation energy. The enlargement of atomic spacing caused by amorphization was manifested as materials swelled. For multi-pulse machining areas (R3–R6), as the pulse number increased and the scanning speed slowed down, the etching marks of SiC became more obvious, the line width slightly increased, and the depth also increased. The AFM results from R3 to R6 also show that the maximum depth of the irradiation area gradually increased from 183 nm to 668 nm.

Fig. 2.

Surface morphology of the laser-written areas. (a)–(f) Surface morphology of R1–R6; the insert graphs are the height profiles along the dashed lines indicated in the 3D morphology where the surface of the virgin area has a height of 0 μm.

Fig. 2.

Surface morphology of the laser-written areas. (a)–(f) Surface morphology of R1–R6; the insert graphs are the height profiles along the dashed lines indicated in the 3D morphology where the surface of the virgin area has a height of 0 μm.

Close modal
Table 2.

Depth and width of the machining regions according to the optical micrograph and AFM results.

Label Optical micrograph AFM
Average width (μm) Average width (μm) Max depth (nm)
R1  13  12.9  12 
R2  11  10.4  23 
R3  20  22.2  183 
R4  22  22.9  283 
R5  23  22.4  359 
R6  25  25.1  668 
Label Optical micrograph AFM
Average width (μm) Average width (μm) Max depth (nm)
R1  13  12.9  12 
R2  11  10.4  23 
R3  20  22.2  183 
R4  22  22.9  283 
R5  23  22.4  359 
R6  25  25.1  668 

fs laser writing can cause damage to the surface of SiC, so it is necessary to conduct Raman spectroscopy on the sample to determine the damage to the lattice quality. Fig. 3 shows the Raman spectra of different regions excited by the 405 nm laser. The Raman modes of 4H-SiC (E2, longitudinal optic phonon–plasmon coupled (LOPC), and second-order Raman mode) were observed in the pristine area and fabricated areas. A dramatic change was observed in the reduction of the E2 peak intensity, which can be attributed to the decrease in the crystal SiC content on the sample surface after the fs laser irradiation. Compared with R1, the E2 peak in the R2 region decreased more significantly. This phenomenon can be attributed to the fact that with the increase in the laser repetition frequency, the distance between adjacent pulses decreases, and the areas affected by a single pulse overlap, leading to a stronger structural damage in the SiC and further reduction of the E2 peak intensity. In addition, in the PL excitation region, the proportion of the modified area of R2 is greater than that of R1, which also leads to the weakening of the E2 mode intensity.

Fig. 3.

Raman spectra in the single-pulse machining regions (R1, R2) excited by the 405 nm laser.

Fig. 3.

Raman spectra in the single-pulse machining regions (R1, R2) excited by the 405 nm laser.

Close modal

To evaluate the uniformity of the damage caused by laser irradiation and repeatability of the single-point spectrum measurement in Fig. 3, 2D mapping focusing on the sample surface was performed by collecting spectra at each point of a 20 × 20-point array with a step size of 1.5 μm. An area of 30 μm × 30 μm was randomly selected from the irradiation regions (R1 and R2) as the scanning area. As shown in Fig. 4, the high energy density of the laser spot center produced relatively large damages in the irradiated central region, which has a lower E2 intensity than the surrounding part. The shape of the weak E2 intensity region is basically consistent with the results of the AFM and optical micrograph, which further proves that the fs laser has modified the SiC sample.

Fig. 4.

2D mapping of the E2 mode intensity excited by the 405 nm laser when focusing on the sample surface. (a) R1; (b) R2.

Fig. 4.

2D mapping of the E2 mode intensity excited by the 405 nm laser when focusing on the sample surface. (a) R1; (b) R2.

Close modal

The VSi in SiC refers to the color center formed by the vacancy defects of silicon atoms in its crystal lattice. There are two non-equivalent VSi defects in 4H-SiC, namely, V1 and V2 centers, whose zero phonon lines (ZPLs) are 862 and 917 nm at a temperature of 5.5 K, respectively.6 At room temperature, the fluorescence of VSi can be widened to a range from 850 nm to 950 nm.6,25Fig. 5(a) shows the PL spectral results of the 785 nm laser focusing on different areas of the sample surface at room temperature. In the pristine region and single-pulse machining region (R1–R2), only 4H-SiC Raman modes (E2, LOPC, and second-order Raman modes) appeared, whereas in the multi-pulse machining regions (R3–R6), an additional emission located in the range of 850–950 nm was observed and identified as VSi. The PL emission of the VSi under the same characterization setups, obtained from another SiC sample23 helium-ions-implanted with an energy of 30 keV and a dose of 5 × 1017 ions/cm2, was compared to support the origin of the PL from the laser-written area. In addition, the fs laser was used to machine an epitaxial-layer 4H-SiC sample (N-doping; concentration: approximately 1016 cm−3) to verify the universality of the VSi preparation method, where the fs laser processing parameters are the same as those presented in Table 1. Fig. 5(b) shows the spectral results focusing on the surfaces of the above sample machining areas at room temperature. The spectral shapes of the three samples are almost identical, with emission between 850 nm and 950 nm. Their different intensities can be attributed to the different densities of the prepared VSi. The above results indicate that fs laser writing, similar to ion implantation, is a feasible method to create VSi in SiC.

Fig. 5.

PL/Raman spectra excited by the 785 nm laser. (a) Irradiated regions (R1–R6) of the 4H-SiC substrate sample by fs laser writing; (b) Irradiated regions of different samples.

Fig. 5.

PL/Raman spectra excited by the 785 nm laser. (a) Irradiated regions (R1–R6) of the 4H-SiC substrate sample by fs laser writing; (b) Irradiated regions of different samples.

Close modal

3.3.1. 2D mapping of VSi

To further confirm the VSi distribution in the fs laser writing SiC sample, 2D mapping was performed in different regions. Because the VSi PL signal intensity is proportional to the VSi density, it is used as an indicator to characterize the VSi density. As shown in Fig. 5, the LOPC peak of SiC was located near 850 nm with the excitation of the 785 nm laser. To avoid the impact of LOPC on the 2D mapping results, the maximum value of the spectrum in the range of 875–950 nm was selected as the VSi signal intensity. An area of 50 μm × 50 μm was randomly selected from the irradiated regions (R1–R6) and pristine area on the sample surface. 2D mapping focusing on the surface of the above areas was obtained by collecting spectra at each point of a 20 × 20-point array with a step size of 2.5 μm. The VSi intensity of different points was presented according to the actual spatial distribution of the point array, as shown in Fig. 6. As shown in Fig. 6(a), the PL intensity of the pristine region is approximately 60. Therefore, in the results of the machining areas (Fig. 6(b)(g)), the part with a PL intensity of approximately 60 was identified as the part not affected by the fs laser. Compared with the pristine area, the single-pulse areas (R1–R2) produced evident PL signal attenuation, whereas the multi-pulse machining areas (R3–R6) produced significant PL signal enhancement. The results in R1 and R2 can be attributed to the fs laser damaging the SiC structure, causing the signal to weaken. In R3–R6, although the SiC structure was destroyed more seriously, the PL signal intensity was enhanced rather than weakened due to the formation of VSi. The shape of the region where the luminescence changed in Fig. 6 is consistent with the results of the AFM and optical micrograph, which proves that fs laser writing leads to the damage of SiC and the generation of VSi. The part with PL intensities below 60 in R1–R2 was defined as the SiC damage area, and their average widths were 33 and 30 μm (R1 and R2), respectively. The part with a PL intensity over 60 in R3–R6 was defined as the position with VSi, and their average widths are 23, 25, 26, and 28 μm (R3–R6), respectively. The above results combined with the data in Table 2 easily show that the width of the machining region in 2D mapping has increased compared to the AFM and optical micrograph results, which may be due to the lateral-resolution broadening effect of the PL method. For R3–R6, it is possible to produce VSi at the lateral position of the machining edge, which will promote luminescence area broadening. The distribution trend of VSi in the multi-pulse machining area was basically consistent. The PL intensity was the highest in the central region and gradually weakened to both sides of the machined grooves. The optical micrograph of the scanning area was superimposed with the 2D mapping results, as shown in Fig. 7. Consistency in the distribution of machining traces and fluorescence-enhanced regions further demonstrated that the VSi defects were caused by fs laser writing.

Fig. 6.

2D PL mapping of VSi excited by the 785 nm laser when focusing on the sample surface.

Fig. 6.

2D PL mapping of VSi excited by the 785 nm laser when focusing on the sample surface.

Close modal
Fig. 7.

Overlay of the optical micrograph and 2D PL mapping of fs laser machining SiC. (a) R5; (b) R6.

Fig. 7.

Overlay of the optical micrograph and 2D PL mapping of fs laser machining SiC. (a) R5; (b) R6.

Close modal

To study the effect of fs laser parameters on the VSi yield, a 5 μm × 5 μm area was selected from the central location of the surface irradiation area. The maximum PL intensity of each point in the above region within 875–950 nm was averaged, and the value was taken as the VSi signal intensity under different laser writing parameters. Similarly, the average E2 peak intensity of each point was taken to represent the SiC damage degree. The results are shown in Fig. 8. When the setups of the fs laser machining parameters were 5 m/s, 5 pulses, and 1 MHz, the maximum PL intensity was obtained. The comparisons of R3 with R5 and R4 with R6 show that when the pulse number is fixed, the slower the scanning speed, the smaller the processing point interval, and the higher the average PL intensity of VSi. As the scanning speed slowed down, the distance between adjacent pulses decreased. The modification of adjacent pulses can influence each other, enhance the accumulation effect, and promote the formation of VSi. Under a scanning speed of 9 m/s (R2, R3, and R4), the average PL intensity of VSi was positively correlated with the pulse number, which is similar to the laser writing defect generation in diamond.26 These dependencies could be explained by the photolytic mechanism of vacancy formation on the sample surface and their subsequent laser-stimulated diffusion in the bulk.27 However, at a scanning speed of 5 m/s (R5, R6), when the pulse number was 10, the average PL intensity of VSi was less than that of 5 pulses. Simultaneously, the damage of SiC is the most serious in R6, as shown in Fig. 8. The reason is temporarily attributed to too many pulses that caused serious damage to the SiC, led to the destruction of the previously generated VSi, and reduced the VSi signal intensity. The analysis of the above results shows that increasing the number of pulses and slowing down the scanning speed in a certain range will promote the formation of VSi due to the accumulation effect. If the scanning speed is too slow or there are too many pulses, the amorphous damage will increase, which inhibits the formation of VSi.

Fig. 8.

Average PL intensity of the E2 mode and VSi excited by the 785 nm laser. The error line represents the standard deviation.

Fig. 8.

Average PL intensity of the E2 mode and VSi excited by the 785 nm laser. The error line represents the standard deviation.

Close modal

3.3.2. VSi PL signal at different depths

The laser spot was focused on the sample surface, and the depth was set to 0. The depth value is negative when the Raman laser focal point moves below the surface by accurately controlling the Z axis. To ensure the accuracy of depth profiling, first, it was confirmed whether the surface of the sample was located at a depth of 0. The variation curve of the E2 peak intensity with depth in the pristine area was tested, as shown in Fig. 9(a). When the laser spot was focused inside the sample (−10 to 0 μm), the intensity amplitude of the E2 peak had little changes, and there was an inflection point at 0 μm. When the laser spot gradually left the sample (0 to 5 μm), the E2 peak intensity rapidly decreased, which proves that the depth of the surface sample is 0. To verify the repeatability of the single-point spectrum measurement, as shown in Fig. 9(a), six points on a line were selected in the pristine area with a spacing of 5 μm, and the E2 peak intensity changes with depth were explored, as shown in Fig. 9(b). There was a clear dividing line at 0 μm, the parts below the line (−10 to 0 μm) were basically the same, and the intensity above the line (0 to 5 μm) gradually decreased, which corresponds to the previous conclusion.

Fig. 9.

E2 peak intensity as a function of the depth excited by the 785 nm laser. (a) Pristine areas near the machining region; (b) Six points on a line in the pristine region with a step size of 5 μm.

Fig. 9.

E2 peak intensity as a function of the depth excited by the 785 nm laser. (a) Pristine areas near the machining region; (b) Six points on a line in the pristine region with a step size of 5 μm.

Close modal

Fig. 10 shows the variation of the PL intensity of the VSi signal as the focal point moves from the inside to the surface of the sample at room temperature. As the depth of focus moved inside the sample, the PL intensity of VSi first increased and then decreased and reached its maximum value at −1 μm. The fs laser intensity was proportional to the formation of VSi within a certain range. However, if the light intensity is too strong, then the material will be removed (the ablative region) or the amorphous layer will inhibit the formation of VSi. The AFM results show that the position at 0 μm is an ablative region, so some materials were removed, which resulted in a decrease in the PL intensity. The laser energy near the ablative region was still too strong, producing a thin layer of amorphous materials, similar to the analysis in Section 3.1. This condition inhibits VSi formation and reduces PL intensity at 0 μm. For areas below the amorphous layer, the absorption intensity was further reduced, the generated VSi defects existed in the optimal intensity at 1–2 μm below the surface, and then a highlighted fluorescence distribution appeared. Due to the continuous decrease in the laser intensity, the VSi yield decreased, and PL signals gradually weakened, as shown in Fig. 10. In addition, the damaged region on the sample top surface absorbed emitted radiation, resulting in a difference between the maximum PL intensity location and maximum VSi location. Modifying the effect requires further exploration, which is beyond the scope of this study.

Fig. 10.

VSi PL signal intensity of the irradiated regions as a function of the distance excited by the 785 nm laser.

Fig. 10.

VSi PL signal intensity of the irradiated regions as a function of the distance excited by the 785 nm laser.

Close modal

Then, a 2D mapping of the VSi PL signal intensity was measured at different depths in R5 under an excitation of 785 nm. The method is similar to that in Fig. 6 with a scanning area of 40 μm × 40 μm. As shown in Fig. 11, the intensity of VSi at different depths is always high in the center of the irradiated area and gradually decreases to both sides. This finding is consistent with the results of our analysis in Section 3.3.1. The strongest VSi strength was generated at approximately 1 μm below the surface, which is also consistent with the information in Fig. 10.

Fig. 11.

2D mapping of the VSi PL intensity in R5 at different depths. The intensity of each image is normalized with the strongest intensity at Z = −1 μm (Unify the upper and lower PL intensity ranges of the 2D mapping at different depths, and take the range of −1 μm as the standard).

Fig. 11.

2D mapping of the VSi PL intensity in R5 at different depths. The intensity of each image is normalized with the strongest intensity at Z = −1 μm (Unify the upper and lower PL intensity ranges of the 2D mapping at different depths, and take the range of −1 μm as the standard).

Close modal

Combined with the above 2D mapping results, the 3D distribution of the VSi PL signals in R5 is shown in Fig. 12. At the center of the machining area near the surface, there is a VSi dense region. The density of VSi decreases from the center to the periphery. According to the previous analysis, the depth of the dense area of VSi was approximately 1 μm below the surface. The obtained cross section (Fig. 12(b)) is not symmetric with respect to the vertical line through the center (∼20 μm). The laser was vertically incident, so VSi was expected to be distributed symmetrically along the machining center, which is not consistent with the results. As there may be drift errors in the translation stage during the actual testing process, the measurement positions of 2D mapping with different depths were inconsistent, which may be one of the reasons. Taking into account the possible drift of the stage during the test, the model needs to be revised later.

Fig. 12.

(a) 3D distribution of the VSi PL intensity in R5. (b) Profile along the dashed lines indicated in the 3D morphology.

Fig. 12.

(a) 3D distribution of the VSi PL intensity in R5. (b) Profile along the dashed lines indicated in the 3D morphology.

Close modal

3.3.3. Defect formation model

The 3D analysis of the PL and Raman signals is helpful to reveal the actual distribution of VSi produced by fs laser writing. From the presented results, we suggest the following model for the defect formation, as shown in Fig. 13. During the fs laser writing, the SiC samples absorbed light. The formation of VSi is positively correlated with the absorbed light energy within a certain range, but if the absorbed light energy is too strong, then the ablative region and amorphous layer may occur, inhibiting the formation of VSi. The incident laser is a Gaussian beam, and the light absorption of SiC sample presents a certain Gaussian distribution according to the laser light intensity and attenuates outward from the center. The horizontal distribution of VSi has the same trend, with the highest concentration in the middle and gradually decreasing to both sides. Longitudinally, the absorbed light intensity decreases with increasing depth. The surface absorption intensity of the SiC sample exceeded the material's ablative threshold, so an ablative region occurred, as shown in the pit structure in the AFM results. The absorbed light energy below the ablative region was still high, which caused amorphous damage in the material and inhibited the formation of VSi. With increasing depth, the absorbed energy further decreased, and there was an optimal absorbed intensity position with the highest VSi yield, and then the fluorescence distribution with high brightness appeared, as shown in the red region of Fig. 13. Next, due to the continuous decrease in light intensity, the VSi yield decreased. When the absorbed intensity dropped below the VSi formation domain value, almost no defects were produced, as shown in the green region of Fig. 13.

Fig. 13.

Schematic diagram of the VSi distribution of the laser-written SiC.

Fig. 13.

Schematic diagram of the VSi distribution of the laser-written SiC.

Close modal

In this study, the fs laser writing of the 4H-SiC sample was characterized by PL and Raman spectroscopy at room temperature. The PL peak in the range of 850–950 nm proves that this method can be used to fabricate VSi in SiC without annealing. Experimental results showed that increasing the number of pulses and slowing down the scanning speed in a certain range will increase the photon energy absorbed by the SiC sample and promote the formation of VSi. If the scanning speed is too slow or there are too many pulses, then the damage of SiC will be aggravated, which inhibits the formation of VSi. According to the surface morphology by AFM and PL spectral characterization results, a defect formation model was established. The PL intensity is the highest in the irradiated region and gradually weakens to both sides of the machined grooves. From the surface to the inside of the sample, the yield of VSi first increases and then decreases, and there is a VSi enrichment area 1 μm below the surface. Because the VSi signal is a broad peak at room temperature, it is necessary to measure the low-temperature PL spectrum to obtain the ZPL and thus determine the type of VSi produced. In addition, the subsequent annealing may increase the density of VSi or generate other defects on the color centers in SiC. Further studies on the post-annealing sample characterization and low-temperature PL spectra measurements are proposed for the future.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was supported by the National Natural Science Foundation of China (No. 51575389, 51761135106), the National Key Research and Development Program of China (2016YFB1102203), the State Key Laboratory of Precision Measuring Technology and Instruments (Pilt1705), and the ‘111’ Project by the State Administration of Foreign Experts Affairs and the Ministry of Education of China (Grant No. B07014). The authors would like to thank Prof. Dongqing Pang, Prof. Chen Xie from Ultrafast Laser Lab of Tianjin University for valuable discussions on this paper.

1.
Sugawara
Y.
Recent progress in SiC power device developments and application studies
. In:
ISPSD '03. 2003 IEEE 15th International Symposium on Power Semiconductor Devices and ICs
.
2003
. p.
10
18
.
2.
Wei
RS
,
Song
S
,
Yang
K
, et al 
Thermal conductivity of 4H-SiC single crystals
.
J Appl Phys
2013
;
113
(
5
).
3.
Koehl
WF
,
Buckley
BB
,
Heremans
FJ
, et al 
Room temperature coherent control of defect spin qubits in silicon carbide
.
Nature
2011
;
479
(
7371
):
84
87
.
4.
Castelletto
S
,
Johnson
BC
,
Ivady
V
, et al 
A silicon carbide room-temperature single-photon source
.
Nat Mater
2014
;
13
(
2
):
151
156
.
5.
Seo
H
,
Falk
AL
,
Klimov
PV
, et al 
Quantum decoherence dynamics of divacancy spins in silicon carbide
.
Nat Commun
2016
;
7
:
12935
.
6.
Nagy
R
,
Widmann
M
,
Niethammer
M
, et al 
Quantum properties of dichroic silicon vacancies in silicon carbide
.
Phys Rev Appl
2018
;
9
(
3
):
034022
.
7.
Zargaleh
SA
,
Hameau
S
,
Eble
B
.
Nitrogen vacancy center in cubic silicon carbide: A promising qubit in the 1.5 μm spectral range for photonic quantum networks
.
Phys Rev B
2018
;
98
(
16
):
165203
.
8.
Beke
D
,
Szekrényes
Z
,
Pálfi
D
, et al 
Silicon carbide quantum dots for bioimaging
.
J Mater Res
2013
;
28
(
2
):
205
209
.
9.
Anisimov
AN
,
Simin
D
,
Soltamov
VA
, et al 
Optical thermometry based on level anticrossing in silicon carbide
.
Sci Rep
2016
;
6
:
33301
.
10.
Niethammer
M
,
Widmann
M
,
Lee
SY
, et al 
Vector magnetometry using silicon vacancies in 4H-SiC under ambient conditions
.
Phys Rev Appl
2016
;
6
(
3
):
034001
.
11.
Soykal
Ö
,
Reinecke
TL
.
Quantum metrology with a single spin-3/2 defect in silicon carbide
.
Phys Rev B
2017
;
95
(
8
):.
12.
Tarasenko
SA
,
Poshakinskiy
AV
,
Simin
D
, et al 
Spin and optical properties of silicon vacancies in silicon carbide: A review
.
Phys Status Solidi B Basic Solid State Phys
2017
;
255
(
1
):
1700258
.
13.
Christle
DJ
,
Falk
AL
,
Andrich
P
, et al 
Isolated electron spins in silicon carbide with millisecond coherence times
.
Nat Mater
2015
;
14
(
2
):
160
163
.
14.
Widmann
M
,
Lee
SY
,
Rendler
T
, et al 
Coherent control of single spins in silicon carbide at room temperature
.
Nat Mater
2015
;
14
(
2
):
164
168
.
15.
Wang
JF
,
Zhang
XM
,
Zhou
Y
, et al 
Scalable fabrication of single silicon vacancy defect arrays in silicon carbide using focused ion beam
.
ACS Photon
2017
;
4
(
5
):
1054
1059
.
16.
Kraus
H
,
Simin
D
,
Kasper
C
, et al 
Three-dimensional proton beam writing of optically active coherent vacancy spins in silicon carbide
.
Nano Lett
2017
;
17
(
5
):
2865
2870
.
17.
Wang
JF
,
Li
Q
,
Yan
FF
, et al 
On-demand generation of single silicon vacancy defects in silicon carbide
.
ACS Photon
2019
;
6
(
7
):
1736
1743
.
18.
Acosta
VM
,
Bauch
E
,
Ledbetter
MP
, et al 
Diamonds with a high density of nitrogen-vacancy centers for magnetometry applications
.
Phys Rev B
2009
;
80
(
11
):
115202
.
19.
Yamamoto
T
,
Umeda
T
,
Watanabe
K
, et al 
Extending spin coherence times of diamond qubits by high-temperature annealing
.
Phys Rev B
2013
;
88
(
7
):.
20.
Castelletto
S
,
Almutairi
A
,
Kumagai
K
, et al 
Photoluminescence in hexagonal silicon carbide by direct femtosecond laser writing
.
Opt Lett
2018
;
43
(
24
):
6077
6080
.
21.
Chen
YC
,
Salter
PS
,
Niethammer
M
, et al 
Laser writing of scalable single color centers in silicon carbide
.
Nano Lett
2019
;
19
(
4
):
2377
2383
.
22.
Li
Q
,
Wang
JF
,
Yan
FF
, et al 
Nanoscale depth control of implanted shallow silicon vacancies in silicon carbide
.
Nanoscale
2019
;
11
(
43
):
20554
20561
.
23.
Song
Y
,
Xu
Z
,
Li
R
, et al 
Photoluminescence and Raman spectroscopy study on color centers of helium ion-implanted 4H-SiC
.
Nanomanuf Metrol
2020
;
3
(
3
):
205
217
.
24.
Hain
TC
,
Fuchs
F
,
Soltamov
VA
, et al 
Excitation and recombination dynamics of vacancy-related spin centers in silicon carbide
.
J Appl Phys
2014
;
115
(
13
):
133508
.
25.
Baranov
PG
,
Bundakova
AP
,
Soltamova
AA
, et al 
Silicon vacancy in SiC as a promising quantum system for single-defect and single-photon spectroscopy
.
Phys Rev B
2011
;
83
(
12
):
125203
.
26.
Kononenko
VV
,
Vlasov
 II
,
Gololobov
VM
, et al 
Nitrogen-vacancy defects in diamond produced by femtosecond laser nanoablation technique
.
Appl Phys Lett
2017
;
111
(
8
).

Jiayu Liu, State Key Laboratory of Precision Measuring Technology and Instruments, School of Precision Instruments and Opto-Electronics Engineering, Tianjin University, China. Mr. Liu is studying for Master degree. His research interests include: the preparation of silicon carbide color centers by femtosecond laser, Raman and fluorescence spectrum characterization.

Zongwei Xu, Associate Professor, State Key Laboratory of Precision Measuring Technology and Instruments, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University, China. His research interests include: defect engineering in wide band gap semiconductor, Micro/nanofabrication using focused ion beam, Raman and fluorescence spectrum, etc.

Ying Song, State Key Laboratory of Precision Measuring Technology and Instruments, School of Precision Instruments and Opto-Electronics Engineering, Tianjin University, China. Ms. Song is studying for PhD degree. Her research interests include: manufacture and spectral characterization of silicon carbide color centers, including the preparation of silicon carbide color centers by ion-implantation, three dimensional Raman and photoluminescence spectral characterization, and model of spectral depth profiling.

Hong Wang, Associate Professor, School of Materials Science and Engineering, Tianjin Polytechnic University. She mainly engaged in inorganic catalytic membrane materials, organic electrochemical synthesis, electrochemical research and development of electro-catalytic membrane reactor, sewagetreatment, and etc.

Bing Dong, State Key Laboratory of Precision Measuring Technology and Instruments, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University, China. Mr. Dong is studying for Master degree.

Shaobei Li, Tianjin BWT laser LTD, Tianjin, China, His research interests include: laser optics, femtosecond laser, and chirped pulse amplification(CPA).

Jia Ren, master of Physics, Changchun University of Science and Technology, research interests: laser-induced breakdown spectroscopy, ultra-fast laser precision machining.

Qiang Li, Postdoctor, CAS Key Laboratory of Quantum Information, School of Physical Sciences, University of Science and Technology of China, China. His research interests include: Controllable preparation and coherent manipulation of defect spins in Silicon Carbide, quantum network and quantum sensing based on solid-state color centers, etc.

Mathias Rommel, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) and Fraunhofer Institute for Integrated Systems and Device Technology IISB, Germany. His research interests include: focused ion beam (FIB), nanoimprint lithography, electrical scanning probe microscopy, and deep level transient spectroscopy (DLTS).

Xinhua Gu, Ph.D in Photochemical Science, his research interest is in the micromachining and the interaction of short pulse on different types of material.

Bowen Liu, Associate Professor, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University. His current research interests include photonic crystal fibers, PCF femtosecond laser amplification system.

Minglie Hu, Professor, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University. His current research interests includemode-locking laser oscillators and amplifiers, fiber lasers, linear and nonlinear propagation in photonic crystal fibers, and microstructure optical device.

Fengzhou Fang Professor, State Key Laboratory of Precision Measuring Technology and Instruments, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University. His research interests are in the areas of micro/nano manufacturing, optical freeform manufacturing, bio-medical manufacturing, ultra-precision machining and metrology.