Tailoring the performances of low operating voltage ... - OSA Publishing

Tailoring the performances of low operating voltage InAlAs/InGaAs avalanche photodetectors Yingjie Ma, Yonggang Zhang,∗ Yi Gu, Xingyou Chen, Suping Xi, Ben Du, and Hsby Li State Key Laboratory of Functional Materials for Informatics, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China ∗ [email protected]

Abstract: We present tailoring of the performances of thin multiplication layer InAlAs/InGaAs avalanche photodetectors (APDs) with operating voltages lower than 20 V. Their operating voltages, gain-voltage slopes and dark currents were successfully tailored by changing the electric field distributions in avalanche region. The thin multiplication layer APDs show small activation energies of the dark current ranging from 0.12 to 0.19 eV at temperatures above 220 K, suggesting a band-trap-band tunneling dominant dark current mechanism over this temperature range. The dark currents show very weak temperature dependences at temperatures lower than 175 K, which mainly originate from the band-to-band tunneling and the surface leakage currents. The spectral responsivity of those APDs show anomalous negative temperature coefficients at gain factors larger than 1, which is attributed to the enhanced phonon scattering effect of carriers in the avalanche region at higher temperatures. Good gain factor uniformity at a given bias is observed for those APDs, and the charge layer is found to help improve the gain uniformity. © 2015 Optical Society of America OCIS codes: (040.1345) Avalanche photodiodes (APDs); (040.0040) Detectors.

References and links 1. P. Martyniuk, J. Antoszewski, M. Martyniuk, L. Faraone, and A. Rogalski, “New concepts in infrared photodetector designs,” Appl. Phys. Rev. 1, 7671–7678 (2014). 2. W. R. Clark, K. Vaccaro, and W. D. Waters, “InAlAs-InGaAs based avalanche photodiodes for next generation eye-safe optical receivers,” Proc. of SPIE 6796, 67962H (2007). 3. J. C. Dries, B. Miles, and R. Stettner, “A 32 × 32 Pixel FLASH Laser Radar System Incorporating InGaAs PIN and APD Detectors,” Proc. of SPIE 5412, 250–256 (2004). 4. W. R. Clark, A. Davis, M. Roland, and K. Vaccaro, “A 1 cm × 1 cm In0.53 Ga0.47 As-In0.52 Al0.48 As Avalanche Photodiode Array,” IEEE Photon. Technol. Lett. 18, 19–21 (2006). 5. G. Q. Chen, J. L. Zhang, P. Wang, J. Zhou, L. Gao, and R. J. Ding, “A new ROIC with high-voltage protection circuit of HgCdTe e-APD FPA for passive active imaging,” Proc. SPIE 8562, 856226 (2012). 6. R. J. McIntyre, “A new look at impact ionization - Part I: A theory of gain, noise, breakdown probability, and frequency response,” IEEE T. Electron Dev. 46, 1623–1631 (1999). 7. S. Ong, K. F. Li, G. J. Rees, J. P. R. David, and P. N. Robson, “A simple model to determine multiplication and noise in avalanche photodiodes,” J. Appl. Phys. 83, 3426–3428 (1998). 8. T. Nakata, I. Watanabe, K. Makita, and T. Torikai, “InAlAs avalanche photodiodes with very thin multiplication layer of 0.1 μ m for high speed and low-voltage-operation optical receiver,” Electron Lett. 36, 1807–1809 (2000).

#240945 (C) 2015 OSA

Received 15 May 2015; revised 30 Jun 2015; accepted 13 Jul 2015; published 16 Jul 2015 27 Jul 2015 | Vol. 23, No. 15 | DOI:10.1364/OE.23.019278 | OPTICS EXPRESS 19278

9. Y. L. Goh, A. R. J. Marshall, D. J. Massey, J. S. Ng, C. H. Tan, M. Hopkinson, J. P. R. David, S. K. Jones, C. C. Button, and S. M. Pinches, “Excess Avalanche Noise in In0.52 Al0.48 As,” IEEE J. Quantum. Elect. 43, 503–507 (2007). 10. M. A. Saleh, M. M. Hayat, P. P. Sotirelis, A. L. Holmes, J. C. Campbell, B. E. A. Saleh, and M. C. Teich, “Impact-Ionization and Noise Characteristics of Thin III-V Avalanche Photodiodes,” IEEE T. Electron Dev. 48, 2722–2731 (2001). 11. D. J. Massey, J. P. R. David, and G. J. Rees, “Temperature Dependence of Impact Ionization in Submicrometer Silicon Devices,” IEEE T. Electron Dev. 53, 2328–2334 (2006). 12. X. G. Zheng, J. S. Hsu, J. B. Hurst, X. Li, S. Wang, X. Sun, A. L. H. Jr., J. C. Campbell, A. S. Huntington, and L. A. Coldre, “Long-Wavelength In0.53 Ga0.47 As-In0.52 Al0.48 As Large-Area Avalanche Photodiodes and Arrays,” IEEE J. Quantum Elect. 40, 1068–1073 (2004). 13. J. P. R. David and C. H. Tan, “Material Considerations for Avalanche Photodiodes,” IEEE J. Sel. Top. Quant. 14, 998–1009 (2008). 14. Y. J. Ma, Y. Zhang, Y. Gu, L. Zhou, X. Y. Chen, S. Xi, and H. Li, “Low Operating Voltage and Small Gain Slope of InGaAs APDs With p-Type Multiplication Layer,” IEEE Photon. Technol. Lett. 27, 661–664 (2015). 15. N. Duan, S. Wang, X. G. Zheng, X. Li, N. Li, J. C. Campbell, C. Wang, and L. A. Coldren, “Detrimental Effect of Impact Ionization in the Absorption Region on the Frequency Response and Excess Noise Performance of InGaAs-InAlAs SACM Avalanche Photodiodes,” IEEE J. Quantum Elect. 41, 568–572 (2005). 16. M. Nada, Y. Muramoto, H. Yokoyama, N. Shigekawa, T. Ishibashi, and S. Kodama, “Inverted InAlAs/InGaAs Avalanche Photodiode with Low-High-Low Electric Field Profile,” Jpn. J. Appl. Phys. 51, 02BG03 (2012). 17. S. Maimona and G. W. Wicks, “nBn detector, an infrared detector with reduced dark current and higher operating temperature,” Appl. Phys. Lett. 89, 151109 (2006). 18. Y. Gu, L. Zhou, Y. Zhang, X. Chen, Y. Ma, S. Xi, and H. Li, “Dark current suppression in metamorphic In0.83 Ga0.17 As photodetectors with In0.66 Ga0.34 As/InAs superlattice electron barrier,” Appl. Phys. Express 8, 022202 (2015). 19. J. F. Klem, J. K. Kim, M. J. Cich, G. A. Keeler, S. D. Hawkins, and T. R. Fortune, “Mesa-isolated InGaAs photodetectors with low dark current,” Appl. Phys. Lett. 95, 031112 (2009). 20. R. W. Hoogeveen, R. J. van der A, and A. P. Goede, “Extended wavelength InGaAs infrared (1.0-2.4 μ m) detector arrays on SCIAMACHY for space-based spectrometry of the Earth atmosphere,” Infrared Phys. Techn. 42, 1–16 (2001). 21. S. R. Forrest, M. DiDomenico, J. R. G. Sith, and H. J. Stocker, “Evidence for tunneling in reverse-biased III-V photodetector diodes,” Appl. Phys. Lett. 36, 580–582 (1980). 22. X. Ji, B. Liu, H. Tang, X. Yang, X. Li, H. Gong, B. Shen, P. Han, , and F. Yan, “2.6 μ m MBE grown InGaAs detectors with dark current of SRH and TAT,” AIP Adv. 4, 087135 (2014). 23. K.-W. Ang, J. W. Ng, G.-Q. Lo, and D.-L. Kwong, “Impact of field-enhanced band-traps-band tunneling on the dark current generation in germanium p-i-n photodetector,” Appl. Phys. Lett. 94, 223515 (2009). 24. A. Rogalski, K. Adamiec, and J. Rutkowski, Narrow-Gap Semiconductor Photodiodes (SPIE-The International Society for Optical Engineering, 2000), chap. 1-7. 25. W. W. Anderson, “Tunnel contribution to Hg1−x Cdx Te and Pb1−x Snx Te p-n junction diode characteristics,” Infrared Phys. 20, 353 (1980). 26. T. L. Tansley, “Temperature dependence of hole diffusion length from spectral response of Au-n+ GaAs photodiodes,” J. Phys. D: Appl. Phys. 5, 1146–1152 (1972). 27. L. Zhou, Y. Zhang, Y. Gu, Y. Ma, X. Chen, S. Xi, and H. Li, “Effects of material parameters on the temperature dependent spectral response of In0.83 Ga0.17 As photodetectors,” J. Alloy. Compd. 619, 52–57 (2015). 28. I. K. Sou, Z. H. Ma, Z. Q. Zhang, and G. K. L. Wong, “Temperature dependence of the responsivity of II-VI ultraviolet photodiodes,” Appl. Phys. Lett. 76, 1098–1100 (2000). 29. C. N. Harrison, J. P. R. David, M. Hopkinson, and G. J. Rees, “Temperature dependence of avalanche multiplication in submicron Al0.6 Ga0.4 As diodes,” J. Appl. Phys. 92, 7684–7686 (2002). 30. X. Jiang, M. Itzler, K. O’Donnell, M. Entwistle, M. Owens, K. Slomkowski, and S. Rangwala, “InP-Based Single-Photon Detectors and Geiger-Mode APD Arrays for Quantum Communications Applications,” IEEE J. Sel. Top. Quant. 21, 3800112 (2015). 31. M. Lahrichi, G. Glastre, E. Derouin, D. Carpentier, N. Lagay, J. Decobert, and M. Achouche, “240-GHz GainBandwidth Product Back-Side Illuminated AlInAs Avalanche Photodiodes,” IEEE Photon. Technol. Lett. 22, 1373–1375 (2010).

1.

Introduction

Focal plane arrays (FPAs) based on In0.53 Ga0.47 As (referred as InGaAs hereafter) avalanche photodetector (APD) elements show an attractive application prospect in infrared imaging area which has been also regarded as one of the 4th generation infrared detection systems [1]. Benefiting from the internal gain of pixels, APD-FPAs provide a ultra-high imaging sensitivity #240945 (C) 2015 OSA


which is particularly suitable for detection in photon-starving environments, e.g. space remote sensing and laser radar (LADAR) imaging [2, 3]. In those applications, low operating voltage, small gain-voltage slope and high APD pixel gain uniformity are highly appreciated to improve the voltage compatibility for readout integrated circuits [4, 5]. Besides, low dark currents (Id ) are also crucial to maintain a high signal to noise ratio. Thin multiplication layer is generally exploited to lower the operating voltage of APDs taking advantage of the carrier’s ’dead-space effect’ [6–8]. The carrier transit time in the thin avalanche region is shortened which helps increase the APD’s bandwidth. On the other hand, as a result of the nonlocal nature of impact ionization, the excess avalanche noise of APDs decreases with decreasing thickness of multiplication region [9, 10]. However, as a trade-off, Id increases in thin multiplication layer APDs [8,11,12] since in order to achieve a same gain factor (M) as in thick multiplication layer APDs the electric (E) field in the thin avalanche region must be increased, which leads to a notable increase of band tunneling leakage in multiplication layer. Essentially, the gain characteristics of APDs are predetermined by the local E field intensity, the E field distribution and the carrier impact ionization threshold energy. This allows control of the operating voltage, the M and the gain-voltage slope via tailoring the E field profiles in the avalanche region. The threshold energy is an intrinsic parameter which directly depends on the material itself in multiplication layers. In0.52 Al0.48 As (referred as InAlAs hereafter) has been demonstrated to be a good electron multiplication material for InGaAs separate absorption and multiplication (SAM) APDs owing to its modest electron impact ionization threshold energy of 1.9-2.2 eV, a higher ionization coefficient ratio of electron to hole than that of hole to electron in InP, and a small excess noise factor. [9, 13] By adjusting the doping, the layer thickness and the layer architectures of the separate absorption, grading, charge and multiplication (SAGCM) InAlAs/InGaAs APDs, both wedge-shaped [14] and uniformed [15, 16] E fields have been demonstrated, showing improved gain-voltage slope or gain-bandwidth product (GBP). Small gain-voltage slope of APDs allows maintaining a more stable gain over a voltage fluctuation range. To achieve APDs with low operating voltage but small gain-voltage slope, the multiplication layer needs to be thin whereas the operating voltage range must be as large as possible. Our previous work has demonstrated the possibility of lowering the gainvoltage slope of thin multiplication layer InAlAs/InGaAs APDs by introducing an E gradient in the avalanche region [14]. In this work, we demonstrated the tailoring of the performances of thin multiplication layer InAlAs/InGaAs SAGCM APDs. The operating voltages, Id as well as gain-voltage slopes were tailored by controlling the E field distributions in avalanche region. Small gain-voltage slopes and low operating voltages less than 20 V with relatively low Id are realized. The Id is extensively studied and the dominating mechanisms in different temperature ranges are identified. Moreover, the temperature dependent behaviors of the spectral responsivity of those APDs at M larger or less than 1 are found to show opposite characteristics, which are explained by the enhanced phonon scattering effect of carriers in the avalanche region at high temperatures. The M uniformity at a given bias as well as its possible origins is also investigated. 2.

Experimental details

The InAlAs/InGaAs APDs are SAGCM-type devices and were grown on n+ -type (sulfurdoped) InP(100) substrates in a VG Semicon-V80H gas-source molecular beam epitaxy (GSMBE) system. N- and p-type dopants are silicon and beryllium, respectively. The epitaxial layer structure is sketched in Fig. 1(a), which in turn consists of heavily n-doped 200-nm-thick InP (2×1018 cm−3 ) and 800-nm-thick InAlAs (4×1018 cm−3 ) buffer layers, a thin p-doped InAlAs multiplication layer, a 70-nm-thick heavily p-doped (6×1017 cm−3 ) InAlAs charge layer, a 72-nm-thick p-doped (6×1016 cm−3 ) InAlGaAs grading layer, a 1.5-μ m-thick p-doped

#240945 (C) 2015 OSA


(6×1016 cm−3 ) InGaAs absorption layer, a 450-nm-thick heavily p-doped (3×1018 cm−3 ) InAlAs electron diffusion blocking layer and a 150-nm-thick heavily p-doped (3×1018 cm−3 ) InGaAs contact layer. To tailor the E field profiles in the avalanche region, three APD wafers denoted by APD-1, APD-2 and APD-3 were grown with multiplication layer thicknesses of 300, 150 and 150 nm, respectively. The doping densities of APDs 1, 2 and 3 are 3×1017 , 1×1017 and 6×1016 cm−3 , respectively. In order to realize a wedge-shaped E profile, the charge layer was removed for APD-1. The material parameters as well as the measured device performances of the three APDs are listed in Table 1. The wafers were processed into top-illuminated square mesa-type photodiodes without anti-reflection (AR) coating using standard photolithography, chemical wet-etching, Si3 N4 passivation and lift-off metallization techniques. The top InGaAs contact layer in the photo-sensitive area was removed using chemical wet-etching in a mixed solution of citric acid and H2 O2 after processing to eliminate photo absorption therein. A 1550 nm laser diode with an attenuated output power of 5 μ W from a single-mode fiber was used as light source in the photocurrent measurements. The temperature dependent dark currentvoltage (Id -V) and spectral responsivity characteristics were measured by bonding the device into a TO-56 package and then mounting in an opened cycle liquid nitrogen cryostat, which controls temperature in the range of 77-350 K.

Fig. 1. (a) Schematic epitaxial layer structure of APDs 1, 2 and 3. (b) Simulated E field distributions along the growth direction for the three APDs at a same M of 2.

Table 1. A list on the material parameters and the measured device performances of the three APDs. ’M-layer’ denotes ’multiplication layer’. Device APD-1

Charge

M-layer

M-layer doping

-layer

thickness (nm)

density (cm−3 )

VBR (V)

No

300

3×1017

20.5

APD-2

Yes

150

1×1017

APD-3

Yes

150

6×1016

#240945 (C) 2015 OSA

RT-Id at

Operating

0.9VBR (nA)

Voltage (V)

112

13.7-20.3

16.5

94

13.3-16.3

21.5

128

10.5-21.5


3.

Results and discussions

The actual room temperature (RT) carrier concentrations in the p-InAlAs multiplication layer of the three APDs are measured via capacitance-voltage (C-V). Results show neglectable differences with the nominal doping densities. Figure 1(b) shows the simulated E field profiles along the growth direction at RT for the three APDs at a same M of 2. The voltages at M=2 for APD1, APD-2 and APD-3 are determined to be -16.5, -14.2 and -16 V based on the photocurrent measurements. It is clearly seen that APD-1 has a wedge-shaped E profile in the 300-nm-thick avalanche region and a peak E field near the p-n junction interface. This peak E field exceeds 1100 kV/cm at M=2. APDs 2 and 3 with field control layers show more gentle gradients of E field in the thinner 150-nm-thick avalanche region. Besides, APD-3 has both lower E field gradient and peak E field intensity compared to APD-2 owing to the lower doping density in multiplication layer.

Fig. 2. Solid and dash lines are the RT photo and dark I-V curves, respectively, for APD-1 (black), APD-2 (blue) and APD-3 (red). Mesa size: 20×20 μ m2 . The black open circle, the blue open square and the red open uptriangle indicate the gain-voltage curves for APD-1, APD-2 and APD-3, respectively. The RT photo and dark current-voltage (I-V) curves for the three APDs with a mesa size of 20×20 μ m2 are shown in Fig. 2. It is seen that APDs 1, 2 and 3 show multiplication layers punch through at around -13.7, -13.3 and -10.5 V, respectively, with relatively flat photocurrent responses. The corresponding unity-gain voltage (M=1) was defined to be -13.7, -13.3 and -10.5 V, respectively, by comparing the responsivities with a reference InGaAs p-i-n photodiode with the same absorption layer thickness [14]. M was extracted from photo current (I p ) by scaling the net photo-current (I p -Id ) with that at unity. The deduced corresponding gain-voltage curves are also shown in Fig. 2. The plotted M>1 voltage ranges for APDs 1, 2 and 3 are determined to be 13.7-20.3, 13.3-16.3 and 10.5-21.5 V, respectively, as listed in Table 1. The VBR was determined by defining that the breakdown takes place at a dark current of 10 μ A. It is found that APD-2 and APD-3 show the narrowest and the widest operating voltage ranges, respectively, leading to the largest and the least gain-voltage slopes. Such gain performance improvement of APD3 originates from the reduced doping concentration and thus the attenuated E field gradient in the avalanche region (Fig. 1). For APD-1 with a 300-nm-thick multiplication layer and a relatively high doping density, a moderate operating voltage range as well as gain-voltage slope is observed, which are likely owing to the wedge-shaped E field profile with a high E field #240945 (C) 2015 OSA


gradient. Those results tell that the operating voltage ranges and the gain-voltage slopes of thin multiplication InAlAs/InGaAs APDs can be readily tailored by control of the layer structures, the layer thicknesses and the doping densities in the avalanche regions.

Fig. 3. Arrhenius plot of Id for APDs 1, 2 and 3 biased at fixed voltages of -16.5, -14.2 and -15.8 V, respectively. Mesa size: 20×20 μ m2 . The fitted activation energies at temperatures above 250, 275 and 225 K for APD-1, APD-2 and APD-3, respectively, are indicated. Inset: temperature dependent Id -V curves for APD-1. The RT Id of APDs 1, 2 and 3 biased at 0.9 times breakdown voltage (0.9VBR ) were determined to be 112, 94, and 128 nA, respectively, as also listed in Table 1. Such relatively high Id of those APDs suggest that the high mean or peak E field intensity in thin multiplication layer APDs will affect their noise performance at high gains. APD-2 shows smaller Id at 0.9VBR than the two other APDs, possibly a combined result of the moderate doping density and the thinner multiplication layer thickness. To uncover the current leakage mechanisms in those thin multiplication layer APDs, temperature dependent Id measurements are carried out. Figure 3 shows the Arrhenius plot of Id for APDs 1, 2 and 3 biased at fixed voltages of -16.5, -14.2 and -15.8 V, respectively, which corresponds to a same gain factor of M=2 for each device at RT. It is seen that Id showed similar weak temperature dependences for APDs 1, 2 and 3 at temperatures below 150, 200 and 175 K, respectively. For higher temperatures, Id of all three APDs showed strong temperature dependencies which vary with 1/T in a linear relation. The activation energies of Id can be fitted by Id ∝exp(-Ea /kT), where k is the Boltzmann constant and T is the temperature. The fitted Ea at temperatures of T>250, T>275 and T>225 K for APD-1, APD-2 and APD-3 is 0.19, 0.16 and 0.12 eV, respectively, as indicated in Fig. 3. Generally, the dominated current leakage mechanism in photodiodes can be qualitatively determined via evaluating the ratio of Ea and the energy bandgap (Eg ) of the depletion region material. An Ea of ∼Eg indicates a diffusion component dominated Id , whereas an Ea of ∼Eg /2 suggests a generation-recombination component dominated Id [17–19]. Ea in an In0.83 Ga0.17 As p-i-n photodiode in temperature ranges of T>225 K and 175 K1. At M