Open Access
Add to BibSonomy
Abstract
A monolithic diamond photodetector with microlenses is fabricated by etching microlens arrays (MLAs) on single crystal diamond surface and patterning tungsten electrode strips on the edge of these arrays. Firstly, compact MLAs are etched on half of diamond sample surface by thermal reflow method. Secondly, via magnetron sputtering technique, two sets of interdigitated tungsten electrodes are patterned on the sample surface, one set is on the edge of MLAs, the other set is on the planar area. The optoelectronic performances of photodetectors have been investigated and indicated that the photocurrent of microlens photodetector increases by 74.8 percent at 10 V under 220 nm UV light illumination by comparing with that in planar case. Simulations of photodetectors' electrical and optical properties have been carried out, illustrating an improvement of charge collection ability and light absorption efficiency in microlens case. Furthermore, the present device structure can be extended to other semiconductor photodetectors.
© 2017 Optical Society of America
1. Introduction
Solar-blind ultraviolet photodetector based on wide band gap semiconductor has drawn much attention to various applications such as environment security, information technology, medical treatment, astronomical observation, and so on [1–3]. More recently, many new materials or structures have been utilized to improve the performance and extend the functions of conventional photodetectors [4–7]. Based on planar metal-semiconductor-metal (MSM) photodetector, researchers have developed more new device configurations to achieve higher photo sensitivity. For instance, three dimensional electrode structures have been developed to achieve high charge collection efficiency [8]. And three dimensional photodetector surface structures, such as nanorods and nanocorns, have been used to enhance the light absorption efficiency [9,10]. Diamond has band gap of 5.5 eV which makes it a naturally solar blind. Due to its unique properties such as hardness, chemical inertness, high carrier mobility and thermal conductivity, diamond is an ideal material for ultraviolet photodetector [11]. Along with its advances in growth and surface patterning techniques [12–15], diamond is drawing much attention for new concept photodetection applications. Recently, several kinds of three dimensional electrode structures have been realized in diamond detector and achieved higher charge collection efficiency. Three dimensional buried and groove-shaped electrode structure photodetectors are fabricated and exhibit higher photoresponse in comparison with that in planar cases [16, 17]. And, a down-top process is utilized to fabricate three dimensional diamond photodetector whose epitaxial layer is grown between metal electrodes [18]. In addition, three dimensional buried graphitic electrode structure is also achieved by femto-laser machining, showing a positive effect on charge collection efficiency in diamond radiation detectors [19, 20]. However, most of these new configuration diamond detectors are focused on charge collection only, there are few reports on three dimensional surface photodetectors utilizing bulge configurations to improve both charge collection and light incident efficiencies.
In this paper, a monolithic diamond photodetector with microlenses is fabricated by etching microlens arrays (MLAs) on single crystal diamond surface and patterning tungsten electrode strips on the edge of these arrays. A planar photodetector has also been prepared on the same sample surface as a comparison. By investigating their electrical properties, microlens photodetector shows an improved photoresponse.
2. Experimental details
2.1. Fabrication of microlens diamond photodetector
MLAs structures were patterned on diamond substrate using conventional thermal reflow and inductively coupled plasma (ICP) etching methods [21]. The substrate used in this experiment was chemical vapor deposited (CVD) diamond with dimension of 3 × 3 × 0.5 mm3. Diamond MLAs fabrication process are schematically illustrated in Fig. 1. The SPR 220 photoresist (PR) was coated on diamond substrate by spinning coater with a speed of 6000 rpm, resulting in a PR thickness of about 2.5 μm. The standard photolithography process was used to form hexagonal PR pillars. Space between two neighbor pillars was about 3 μm. By holding the sample on a hot plate for 25 s at 160 °C, the pillars melted and formed spherical segments. Then, the PR patterns were transferred onto diamond surface using an ICP etch process with O2 and Ar as the etch gas. Arrays of closed-packed hexagonal microlenses with diameter of approximately 15 μm were fabricated on half surface of substrate, and the other half was planar surface. About 100 nm thick interdigitated tungsten (W) electrodes were patterned on both of microlens and planar surface through standard photolithography technique and radio frequency magnetron sputtering method. The width of electrode was 7.5 μm and the distance between two electrodes was 6.5 μm, leading to an optical receiving area of about 0.231 mm2. After above process, two photodetectors were fabricated in both microlens and planar areas on diamond substrate as shown in Fig. 1.
2.2. Characterizations
The quality of diamond substrate was evaluated by Raman spectrometer with a 50 × objective lens and a 532 nm wavelength laser. The morphologies of diamond photodetectors were characterized by scanning electron microcopy (SEM), atomic force microscopy (AFM) and laser confocal scanning microscopy (LCSM). The current-voltage (I-V) relations of as-fabricated photodetectors were investigated with Agilent B1505A power device analyzer. The external voltage is applied between the metal contacts generating an electric field perpendicular to the incident light across the device by a two-point probe method [22], which is also illustrated in inset of Fig. 1. The incident direction of light is perpendicular to the sample surface in our experiment. The optoelectronic properties were evaluated with a Keithley 6487 picoammeter/voltage source, a 500 W Xe lamp source and a monochromator. The light power at the sample surface was measured by a commercial UV-enhanced Si detector. To investigate the influence of microlens geometry on photodetector properties, COMSOL Multiphysics and FDTD Solutions software were used to simulate the electric field and light distributions, respectively [19, 23].
3. Results and discussion
3.1. Morphology
Both monolithic microlens and conventional planar diamond photodetectors are fabricated on halves of one diamond respectively, and the microlens photodetector is shown in Fig. 2(a). Electrodes in these two type photodetectors have the same horizontal width and inter-distance. Figure 2(b) displays magnified microlens area on diamond substrate before electrode deposition. Figures 2(c) and 2(d) show detail configurations of microlens diamond photodetectors. Compared with conventional planar photodetector, microlens photodetector has quite different geometry. In microlens case, electrodes are symmetrically fabricated on two edge sides of each microlens, which are consistent with original design as shown in Fig. 1. Each microlens has diameter of about 15 μm and height of about 300 nm. Based on the geometry and optical theory, the radius of curvature, focal length (f) can be estimated using the following formulas [15].
Fig. 2 (a) SEM image of fabricated microlens photodetector. (b, c) SEM images of fabricated microlens before and after electrode deposition with high magnification, dashed line is guide to emphasize the geometry of microlens. (d) Three dimensional LCSM images of fabricated microlens photodetector, black dash lines are guide to emphasize electrodes.
where ROC is radius of curvature, r is the radius of the microlens, and n is the material refractive index. Here, refractive index of diamond is 2.42. By Eqs. (1) and (2), ROC and f are calculated to be 93.9 and 66.13 μm, respectively. Since the small diameter of microlens results in small value for radius of curvature, the value of f is low. Beside, AFM measurement results indicate roughness of 0.845 and 0.581 nm for microlens and planar areas, respectively. These values are achieved with average of roughness measured at three different 1 × 1 μm2 areas.
The cross section profiles of planar and microlens diamond photodetectors are sampled and shown in Fig. 3. Thickness, width and inter-distance of electrodes are about 100 nm, 7.5 μm and 6.5 μm, respectively. As shown in Fig. 3, photo active area between two electrodes in microlens photodetector is larger than that in planar one. Meanwhile, electrodes are deposited to follow the spherical shape of the diamond surface, causing a different electric field distribution.
3.2. Optoelectronic performances
Figure 4(a) exhibits the dark currents of microlens and planar photodetectors, whose values are extremely low and close to each other due to the same intrinsic electrical property from one diamond substrate. Photocurrents of these two photodetectors are investigated by using UV light illumination. Figure 4(b) shows the I-V characteristics of microlens and planar photodetectors illuminated with 220 nm wavelength light. The significant increase of photocurrents clearly indicates that two diamond based photodetectors are UV light sensitive, which attributed to the photo generation of free carriers and the electrical transport through the diamond surface. Photocurrents of microlens and planar photodetectors indicate the values of 2.501 and 1.434 pA at 10 V, respectively, illustrating that the microlens one has larger photocurrent. The signal to dark ratios are 24 and 19.2 dB at 10 V for microlens and planar photodetectors, respectively. These pA level photocurrents would be ascribed to the Schottky contact between tungsten and diamond from the I-V curve shown in Fig. 4, and also to the smaller photo active area compared with that in previous work [11], which could lead to a relatively poor charge collection and low amount of photogenerated carriers. In addition, it may also be due to low quality of diamond whose values of full width at half maximum (FWHM) are 4.409 and 4.472 cm−1 for microlens and planar areas respectively evaluated using Raman spectrometer. The comparison results indicate that microlens have a positive effect on photo induced current.
Fig. 4 I-V characteristics of diamond photodetectors. (a) Dark currents and (b) photocurrents under illumination of 220 nm light of microlens and planar diamond photodetectors. The incident light power density is 742 nW/mm2.
Figure 5 shows the responsivity of both microlens and planar photodetectors investigated at a bias voltage of 10 V. Here, photo active area of microlens photodetector is calculated to be 0.2315 mm2, which is larger than that of planar one. The responsivity of two photodetectors show a similarity in spectrum shape. It is mainly stemmed from the intrinsic photo response performance of the diamond substrate. There are peaks at around 300 nm for both photodetectors, which is ascribed to sub-bandgap defects in the substrate [1, 22]. Responsivity of microlens photodetector is higher than that of planar case. The UV to visible rejection ratio of microlens and planar photodetectors for 220 versus 400 nm are 20.4 and 13.9, respectively. All these results indicate that microlens structure can enhance the response ability and rejection ratio. In addition, the improvement efficiency of responsivity in our work is comparable to the efficiency of 50% reported in three dimensional electrode diamond photodetector [17]. However, the rejection ratio is lower than that reported in literature by several orders [11, 24, 25], which is mainly ascribed to the differences of diamond qualities and experiment conditions. With further optimization of experiment conditions, higher photocurrent and rejection ratio should be attainable with microlens structures.
3.3. Mechanism
The larger photocurrent response in microlens diamond photodetector stems from two aspects. In the first, the electrodes deposited on bulged diamond microlenses result in a non-planar electrode configuration as shown in Fig. 3(b), which would enhance the charge collection efficiency of the photodetector. To investigate the geometry impact on charge collection efficiency, simulations of electric field distribution in photodetectors with microlens and planar configurations were carried out with COMSOL software, whose results are shown in Fig. 6. These illustrations of simulated models are shown in insets of Fig. 6(a) and 6(b). The presence of charges trapped at the metal-diamond interface is assumed to be negligible in simulations. Figures 6(a) and 6(b) show the electric field distributions of microlens and planar photodetectors at 10 V in the cross section along the black dash lines indicated in the insets of Fig. 6. Their electric field distribution profiles are slightly different, which is due to the geometry difference between two photodetectors caused by microlens structure with lens height of 300 nm. In both cases, the electric field intensities along the red dash lines are calculated and as shown in Fig. 6(c). It is shown the difference of electric field intensity between microlens and planar photodetectors. Within 10 μm thickness, electric field intensity in microlens photodetector is slightly higher than that of the planar one, which suggests stronger electron capturing ability.
Fig. 6 Electric field simulations in (a) microlens and (b) planar geometry. Insets: top view of simulated configurations. (c) Electric field intensity versus depth along red dash lines as shown in 6(a) and 6(b).
The larger electric field in microlens case could be explained by oblique electrodes which work as a capacitor. It is because electrostatic force of capacitor has a relation with electrode plate angle [26].
where F is electrostatic force, E is electric field intensity, q is value of electric charge, and β is angle between two electrodes plate. Here, β values are 178.3° and 180° for electrodes on microlens and planar surfaces respectively. From Eqs. (3) and (4), it is clear that non-planar electrodes have positive effect on electric field intensity.
In the second, microlens structure on diamond could enlarge photo incident area and improve the light distribution in diamond, which may enhance light absorption efficiency and photo induced carrier density. In our case, photo active area increase by about 480 μm2 in microlens photodetector. Additionally, due to converging ability of microlens, light intensity per unit volume in illuminated area is higher than that in planar part, which may increase the photo induced carrier number. To investigate the light intensity distributions in microlens photodetector, three dimensional optical simulations are carried out by FDTD Solutions software. The wavelength of incident light and refractive index of diamond are 220 nm and 2.729, respectively [27]. According to simulation results of light distribution in microlens and planar photodetectors, as shown in Fig. 7(a), light is converged by microlens and indicated by red dash lines. The convergence suggests that more photons are gathered in shallow diamond layer, from where photo induced carriers are captured by electrodes with short drift length. Figure 7(b) displays the light intensity calculated along with the red solid lines indicated in Fig. 7(a). It clearly indicates that light intensity in microlens photodetector is higher than that in planar one. With these simulation results, the incoming light intensity in microlens photodetector is calculated to be about 1.2 times larger than that in planar case. The relationship between photocurrent (I) and light intensity can be expressed with the formula as followed [28].
Fig. 7 (a) Light distribution simulation of microlens and planar diamond photodetectors. (b) Light intensity versus depth along red solid lines as shown in 7(a).
where η and q are quantum efficiency and electronic charge, respectively, which are unity for microlens and planar photodetectors due to the same experiment conditions of both cases. G and N are photocurrent gain and number of photons, respectively. From Eq. (5), larger light intensity would conduce to enlargement of photocurrent. Meanwhile, since photocurrent gain can increase with increasing light power [29], the photocurrent may be further improved.
Besides, based on our results, UV-light response in microlens photodetector could be further enhanced by improving electrode contact and the diamond quality. The detail results will be reported in future work.
4. Conclusions
A monolithic microlens diamond photodetector has been fabricated with photolithography, thermal reflow process, inductively coupled plasma etching and magnetron sputtering. Compared with conventional planar configuration, the photocurrent and rejection ratio of microlens photodetector have been increased about 74.8 and 46.8 percent at 10 V under 220 nm light illumination, respectively. Electrical and optical simulation results reveal that photocurrent increase is ascribed to electric field intensity and light intensity enhancement in microlens photodetector. Furthermore, the present device configuration can be extended to other semiconductor photodetectors.
Funding
National Natural Science Foundation of China (NSFC) (61627812, 61605155), Technology Coordinate and Innovative Engineering Program of Shaanxi (2016KTZDGY02-03) and Postdoctoral Science Foundation of China (PSFC) (2015M580850).
Acknowledgments
The authors are thankful to Ms. Dai from International Center for Dielectric Research (ICDR), Xi’an Jiaotong University for her help in SEM measurement.
References and links
1. M. Liao, L. Sang, T. Teraji, M. Imura, J. Alvarez, and Y. Koide, “Comprehensive Investigation of Single Crystal Diamond Deep-Ultraviolet Detectors,” Jpn. J. Appl. Phys. 51, 090115 (2012).
2. L. Sang, M. Liao, and M. Sumiya, “A comprehensive review of semiconductor ultraviolet photodetectors: from thin film to one-dimensional nanostructures,” Sensors (Basel) 13(8), 10482–10518 (2013). [PubMed]
3. E. Monroy, F. Omnès, and F. Calle, “Wide-bandgap semiconductor ultraviolet photodetectors,” Semicond. Sci. Technol. 18, R33 (2003).
4. A. S. Pratiyush, S. Krishnamoorthy, S. V. Solanke, Z. Xia, R. Muralidharan, S. Rajan, and D. N. Nath, “High responsivity in molecular beam epitaxy grown β-Ga2O3 metal semiconductor metal solar blind deep-UV photodetector,” Appl. Phys. Lett. 110, 221107 (2017).
5. A. Szyszka, L. Lupina, G. Lupina, M. Mazur, M. Schubert, P. Storck, S. Thapa, and T. Schroeder, “Enhanced ultraviolet GaN photo-detector response on Si (111) via engineered oxide buffers with embedded Y2O3/Si distributed Bragg reflectors,” Appl. Phys. Lett. 104, 011106 (2014).
6. J. Du, J. Xing, C. Ge, H. Liu, P. Liu, H. Hao, J. Dong, Z. Zheng, and H. Gao, “Highly sensitive and ultrafast deep UV photodetector based on a β-Ga2O3 nanowire network grown by CVD,” J. Phys. D Appl. Phys. 49, 425105 (2016).
7. M. Kim, S.-M. Choi, H. A. Yoon, S. K. Choi, J.-U. Lee, J. Kim, S. W. Lee, Y.-W. Son, and H. Cheong, “Photocurrent generation at ABA/ABC lateral junction in tri-layer graphene photodetector,” Carbon 96, 454–458 (2016).
8. S. I. Parker, C. J. Kenney, and J. Segal, “3D — A proposed new architecture for solid-state radiation detectors,” Nucl. Instrum. Methods in Phys. Rese. A 395, 328–343 (1997).
9. D. Sang, H. Li, S. Cheng, Q. Wang, J. Liu, Q. Wang, S. Wang, C. Han, K. Chen, and Y. Pan, “Ultraviolet photoelectrical properties of a n-ZnO nanorods/p-diamond heterojunction,” RSC Advances 5, 49211–49215 (2015).
10. H. Fan, M. Sun, P. Ma, M. Yin, L. Lu, X. Xue, X. Zhu, D. Li, and J. Ma, “UV photodetectors based on 3D periodic Au-decorated nanocone ZnO films,” Nanotechnology 27(36), 365303 (2016). [PubMed]
11. M. Liao, Y. Koide, and J. Alvarez, “Schottky-barrier photodiode with interdigitated-finger geometry: Application to diamond,” Appl. Phys. Lett. 90, 123507 (2007).
12. D. S. Hwang, T. Saito, and N. Fujimori, “New etching process for device fabrication using diamond,” Diamond Related Materials 13, 2207–2210 (2004).
13. M. Karlsson and F. Nikolajeff, “Diamond micro-optics: microlenses and antireflection structured surfaces for the infrared spectral region,” Opt. Express 11(5), 502–507 (2003). [PubMed]
14. C. Lee, H. Choi, E. Gu, M. Dawson, and H. Murphy, “Fabrication and characterization of diamond micro-optics,” Diamond Related Materials 15, 725–728 (2006).
15. T.-F. Zhu, J. Fu, W. Wang, F. Wen, J. Zhang, R. Bu, M. Ma, and H.-X. Wang, “Fabrication of diamond microlenses by chemical reflow method,” Opt. Express 25(2), 1185–1192 (2017). [PubMed]
16. Z. Liu, F. Li, W. Wang, J. Zhang, F. Lin, and H.-X. Wang, “Effect of depth of Buried-In Tungsten Electrodes on Single Crystal Diamond Photodetector,” MRS Adv. 1, 1099–1104 (2016).
17. K. Liu, B. Dai, V. Ralchenko, Y. Xia, B. Quan, J. Zhao, G. Shu, M. Sun, G. Gao, and L. Yang, “Single crystal diamond UV detector with a groove-shaped electrode structure and enhanced sensitivity,” Sens. Actuators A Phys. 259, 121–126 (2017).
18. Z. Liu, J.-P. Ao, F. Li, W. Wang, J. Wang, J. Zhang, and H.-X. Wang, “Fabrication of three dimensional diamond ultraviolet photodetector through down-top method,” Appl. Phys. Lett. 109, 153507 (2016).
19. B. Caylar, M. Pomorski, and P. Bergonzo, “Laser-processed three dimensional graphitic electrodes for diamond radiation detectors,” Appl. Phys. Lett. 103, 043504 (2013).
20. J. Forneris, A. L. Giudice, P. Olivero, F. Picollo, A. Re, M. Marco, F. Pompili, C. Verona, G. V. Rinati, M. Benetti, D. Cannata, and F. D. Pietrantonio, “A 3-dimensional interdigitated electrode geometry for the enhancement of charge collection efficiency in diamond detectors,” Europhys. Lett. 108, 18001 (2014).
21. T.-F. Zhu, J. Fu, F. Lin, M. Zhang, W. Wang, F. Wen, X. Zhang, R. Bu, J. Zhang, J. Zhu, J. Wang, H.-X. Wang, and X. Hou, “Fabrication of diamond microlens arrays for monolithic imaging homogenizer,” Diamond Related Materials 80, 54–58 (2017).
22. M. Liao, Y. Koide, and J. Alvarez, “Thermally stable visible-blind diamond photodiode using tungsten carbide Schottky contact,” Appl. Phys. Lett. 87, 022105 (2005).
23. B. J. M. Hausmann, M. Khan, Y. Zhang, T. M. Babinec, K. Martinick, M. Mccutcheon, P. R. Hemmer, and M. Lončar, “Fabrication of diamond nanowires for quantum information processing applications,” Diamond Related Materials 19, 621–629 (2010).
24. A. Balducci, M. Marinellia, E. Milani, M. E. Morgada, A. Tucciarone, and G. Verona-Rinati, “Extreme ultraviolet single-crystal diamond detectors by chemical vapor deposition,” Appl. Phys. Lett. 86, 193509 (2005).
25. A. BenMoussa, A. Soltani, K. Haenen, U. Kroth, V. Mortet, H. A. Barkad, D. Bolsee, C. Hermans, M. Richter, and J. C. Dejaeger, “New developments on diamond photodetector for VUV solar observations,” Semicond. Sci. Technol. 23, 035026 (2008).
26. F. E. H. Tay, J. Xu, Y. Liang, V. J. Logeeswaran, and Y. Yao, “The effects of non-parallel plates in a differential capacitive microaccelerometer,” J. Micromech. Microeng. 9, 283 (1999).
27. A. M. Zaitsev, Optical properties of diamond: A data handbook (Springer Science & Business Media, 2013).
28. R. S. Sussmann, CVD Diamond for Electronic Devices and Sensors (Wiley, 2009).
29. M. Liao, X. Wang, T. Teraji, S. Koizumi, and Y. Koide, “Light intensity dependence of photocurrent gain in single-crystal diamond detectors,” Phys. Rev. B 81, 033304 (2010).
