Institute of Solid State Physics, University of Latvia, 8 Kengaraga Str. LV-1063 Riga, Latvia.
C. C. Yang
Department of Mechanical Engineering and Graduate Institute of Electro – Optical Engineering,
National Taiwan University, 1, Roosveld Road, Sec. 4, Taipei, Taiwan, R. O.C
Abstract
The deep level transient spectroscopy (DLTS) for investigation of trapping states in GaN muli quantum well structures has been installed and approved. In order to assess the usability of DLTS technique for study the deep trapping levels in multi-quantum well structures both the thermostimulated depolarization (TSD), thermostimulated capacitance relaxation (TSC) and DLTS are studied in p-n homojunction GaN blue diode. The TSD and TSC curves shows that cooling of the GaInN structure down to liquid helium temperatures results in complete trapping both the majority and the minority charge carriers at the shallow dopand levels. Thermal ionization of the dopand states occurs in the temperature region 40 – 60 K. An electron level at Ec-Et= 0.30 eV is observed. It is shown that DLTS technique extended wit DSD and TSC measurements is suitable for investigation of deep trapping states in MQW structures. Investigation of the photo-filling spectra of trapping states by DLTS it is prospective for selection of the localization of defect states responsible for deep trapping levels in the barrier, wall or interface region of MQW-s.
2.4.1.Introduction
Gallium nitride (GaN) processes some unique features such as large bandgap (3.39 eV), high thermal stability with a large breakdown field of 2MV/cm. These enable GaN to emerge as unique material for multiple applications particularly light emitting diodes, diode lasers, power electronic devices and solar blind ultraviolet detectors [1,2]. However, the performance of devices can be limited by the deep levels present within the GaN bandgap, which act as traps and/or recombination – generation centres. These deep levels cause non-radiative transitions and in some cases are responsible for long wavelength non-radiative recombination.
The deep levels in GaN, and GaInN, in general, are either related to structural defects like dislocations, native point defects and/or their complex such as nitrogen vacancies [3], or impurity-related defects.
Doping of GaN may introduce deep levels, in addition to shallow donor or acceptor levels. For example, it has been stated that Mg doping introduces a deep level at Ec –Et ~ 0.58-0.62 eV. [4]. For doped GaN trap levels at 0.28 and 0.45 eV are attributed to the Si-related defects [5].
41
At present experimental data concerning deep levels in Ga - related nitride materials condign on self in investigation of pure or doped GaN and GaN homostructures. Studies of trapping levels in GaN heterostructures including quantum well (QW) structures up to now have not been reported. Formation of deep levels in InGaN alloys may considerably affect the electrical characteristics of material as well as charge carrier generation – recombination kinetics.
It can been expected that by increase of the In – content in alloy a selective formation of localized states will occur. It is well known that the poor thermal stability of InN and low In vapor pressure over the metal phase result in the formation of In clusters, especially at temperatures T > 500 oC [7]. It is found that deviations in presently available experimental data on the bandgap of InN from 0.7 to 2 eV[8- 11] are linked to the precipitation of indium in the metallic phase that leads to additional optical losses due to light scattering and absorption by dispersed small particles [12].
Additional reason for formation of localized electron states is disorder in the interface region barrier-wall in the MQW-s of GaN/InGaN. Deep levels due to self – organized quantum dots (QD) may significantly affect the light generation processes. Particularly it is stated that self-organized InAs quantum dots (QD) embedded in GaAs act as both the electron and the hole trapping centres [6]. Thus, QD can act as radiative recombination centres.
However, investigations of hole and electron escape from a single layer QD in GaAs matrix by time-resolved capacitance spectroscopy show fundamentally different behavior. For electrons, besides conventional tunnel emission a thermally activated tunneling processes involving excited QD states dominates for high temperatures. For holes only thermal activation from the ground state directly to the GaAs valence band contributes. It is stated that the hole localization time at given temperature in the QD is significantly larger than the one of electrons. In GaInN/GaN both the carrier trapping and the strain-induced piezoelectric field, which generates the quantum confined Stark effect, has been involved for interpreting observed optical phenomena in samples with increasing silicon-doping concentration, such as the photoluminescence peak shift and decay time decrease followed by intensity increase [13-15]. The temperature dependence of photoluminescence has been interpreted assuming capture and thermoactivated delocalization of excitons from QD [13].
It can be expected that like other structure defects and impurities the QD will act as deep capture centers for charge carriers. By injection or photo-generation of free electrons and holes subsequent capture of charge carriers in QD can be important formation mechanism of excitons. As result, the composition and the concentration of In clusters significantly depends on the In concentration in WQ as well as the sample thermal annealing conditions.
The objective of the present work is application of the complex of methods of thermoactivation spectroscopy of defects for detailed characterization of localized s in InGaN/GaN quantum well structures. Different methods are utilized to investigate the relative concentration and the thermal activation energy of charge carrier release of various defects:
thermostimulated capacitance decay (TSCD), thermostimulated depolarization currents (TSDC) and deep level transient spectroscopy (TLDS).
2.4.2.Experimental
The TSCD, TSDC trap spectroscopy have been performed using closed cycle cryostat in the temperature region, starting from ~10 oK to about 350 oK with the linear heating rate being either 0.05 or 0.2 oKs-1. Polarization of the sample before TSDC measurements have been carried out in an external field (-30 V) either in the course of slow cooling from ~300 oK, or by the irradiation with photo-active light at 10oK.
42
The TLDS trap spectroscopy has been performed using conventional equipment. The closed cycle cryostat was equipped with two screened high impedance, non – vibrating wires attached to the sample holder. The sample mounted was on the sample holder, sample electrical contacts was connected with signal input –output wires using Cu or Ag wires Φ=
0.05 mm. For capacitance and conductance kinetics measurements the Model Boonton 7200 Capacitance meter was utilized. The type Agilent 33220A pulse generator was used as a sample polarization source. The measurement was carried out Five times with different rate windows from 1 ms to q16 ms. The bias used Ub = -1V; The pulse parameters: Up = 1V; ∆t=1 ms. The Commercial blue light diode was used for investigations.
2.4.3. Results
For the general characterization of the donor/acceptor doping governed diode conductivity parameters the temperature dependence of the conductivity and capacity was investigated. By cooling down up to about 80 K both the conductivity and the capacity decrease only slightly.
Figure 2.4.1. The depolarization current of the diode, polarized by cooling down of the sample to 10 K by applied bias -20 C, and heated with constant rate ).1 K/s
Further cooling results in decrease of conductivity from the value of about 80 µS to the very low value. The later is close to the sensitivity limit of the capacitance meter.
Simultaneously capacitance decreases from 64pF, to 0.2 pF. Results indicate that at low temperatures the donors are going occupied by electrons. The capacity at high temperatures being caused by p-n transition capacity at low temperatures reflects the sample capacity wit GaInN as dielectric media.
For study of the TSCD and TSDC the sample has been cooled down in external field (Ub=-25V) to T=10 K . Fig.2.4.1 represents the temperature dependence of depolarization current recorded by heating rate 0.1 K/s. Depolarization curve shows presence of two stages of depolarization kinetics. The observed kinetics indicates that two kinds of donor which differ in activation energy are present in MQW structure. It can be proposed the low temperature peak origins due to escape of electrons from donor states in the well region whereas the high temperature peak, characterized by more high activation energy correspond escape of electrons in GaN.
Results of the DLTS measurements provided at different temperatures are represented in Figure 2.4.2. By applying a voltage pulse the measured capacitance increases and remains
43
constant until the end of the pulse. At the end of the pulse the capacitance falls below the original level. After that time the capacitance increases exponentially.
Figure 2.4.2.
The capacitance transient kinetics obtained at different temperatures.
The Figure 2.4.3 shows the DLTS spectrum for the MQW structure obtained at different rate windows. At different rate windows the shape of the DLTS spectrum remains unchanged.
This result indicates that the single deep trapping level represents the main localized state in the MQW structure.
Figure 2.4.3. DLTS spectra of the GaN MQW structure obtained at different rate windows.
6 0 80 100 120 140 160 180 200 220 240 260 -2,5
-2,0 -1,5 -1,0 -0,5 0,0
∆C
Temperature t1/t
2 (msec) 0.25/2.5 0.35/3.5 0.45/4.5 0.5/5 1/10 2/20 4/40
44
Figure 2.4.4. Arrhenius plot of deep defect levet in the GaN MQW structure. The activation energy of the traps : 0.30 eV
2.4.4.Conclusion
In order to assess the usability of DLTS technique for study the deep trapping levels in multi-quantum well structures both the thermostimulated depolarization (TSD), thermostimulated capacitance relaxation (TSC) and DLTS are studied in p-n homo-junction GaN blue diode. The TSD and TSC curves shows that cooling of the GaInN structure down to liquid helium temperatures results in complete trapping both the majority and the minority charge carriers at the shallow dopand levels. Thermal ionization of the dopand states occurs in the temperature region 40 – 60 K.
Two kinds of donor levels governing the conductivity at high temperatures are observed in the MQW structures which have close activation energies. It is proposed that they characterize the dopand energy in walls and barriers respectively.
According to the DLTS spectra the single deep electron level at Ec-Et= 0.30 eV is observed. It is shown that DLTS technique extended with TSD and TSC measurements is suitable for investigation of deep trapping states in MQW structures. Investigation of the photo-filling spectra of trapping states by DLTS it is prospective for selection of the localization of defect states responsible for deep trapping levels in the barrier, wall or interface region of MQW-s.
References
1. Zhang X, Chua S J, Liu W & Li P, Phys. Stat. Solidi, 216, No.307, 307 (1999)
2. S. Nakamura, M. Senoch, S. Nagahama, N. Iwasa, T. Yamada, T. Matsushita, H. Kiyoka, and Y. Sugimoto, Jpn. J. appl. Phys., Part 2 35, L74, 1996.
3. Z-Q. Fang, J.W. Hemsky, D.C.Look, and M.P. Mack, Appl. Phys. Lett. 72, 448 (1998) 4. A. Hierro, S.A. Ringel, M. Hansen, J.S. Speck, U.K. Mishra and S.P. DenBaars, Appl. Phys
Lett. 77, 1499 (2000).
5. C.B. Soh, D.Z. Chi, H.F.Lim and S.J. Chua, Mat. Res. Soc. Symp. Proc. 719, F13.5 (2002).
3,9 4,0 4,1 4,2 4,3 4,4 4,5 4,6 4,7
- 7,0 - 6,5 - 6,0 - 5,5 - 5,0 - 4,5 - 4,0
∆E = 0.30 eV
ln(en/T2)
1000/T (K-1)
45
6. C. M. A. Kapetyn, M. Lion, R. Heitz, D. Bimberg, P.N. Brunkov, B. V. Volovik, S. G.
Konnikov, A. R. Kovsh, and V. M. Ustinov, Appl. Phys. Lett. 76, 1573 (2000).
7. S. V. Ivanov and V. N. Jmerik, in Vacuum Science and Technology: Nitrides as Seen by the Technology, ed. By T. Paskova and B. Monemar (Research Signpost, Kerala, 2002), p.
369.
8. L. Tansley and C.P. Foley, J. Appl. Phys. 59, 3941 (1986)
9. T. Inushima, V. V. Mamutin, V. a. Vekshin, S. V. Ivanov, T. Sakon, M. Motokawa, and S.
Okoya, J. Cryst. Growth 227/228, 481 (2001)
10. V. U. Davydov, A. A. Klochikhin, R. P. Seisyan, V. V. Emtsev, S. V. Ivanov, F. Bechstedt, J.
Furthmueler, H. Harima, A. V. Mudryi, J. Aderhold, O. Semchinova, and J. Graul, Phus.
Status Solidi B 229, R1 (2002).
11. J. Wu, W. Walukiewicz, K. M. Yu, J. W. Ager III, E. e. Haller, H. Lu, W. J. Schaff, Y. Saito, and Y. Nanishi, Appl. Phys. Lett. 80, 3967 (2002).
12. T.V. Shubina, S. V. Ivanov, V. N. Jmerik, D. D. Solnyshkov, V. A. Vekshin, P. S. Kop’ev, A.
Vasson, J. Leymarie, A. Kavokin, H. Amano, K. Shimono, A. Kasic, and B. Monemar., Phys. Rev. Lett. 92, 117407-1 (2004).
13. Yung-Chen Cheng, Cheng-Hua Tseng et al. Appl. Phys. Lett., June 2002
14. P. Ramvall, Y. Aoyagi, A. Kuramata, P. Hacke, K. Domen, and K. Horino, Appl. Phys. Lett.
76, 2994 (2000).
15. E. Oh, C. Sone, O. Nam, H. Park, and Y. Park, Appl. Phys. Lett. 76, 3242 (2000).
46
2.5. NANOSTRUCTURES AND CARRIER LOCALIZATION BEHAVIORS OF GREEN-LUMINESCENCE InGaN/GaN
QUANTUM-WELL STRUCTURES OF VARIOUS SILICON-DOPING