PROTON AND GAMMA-RAY IRRADIATION EFFECTS ON InGaP/GaAs HETEROJUNCTION BIPOLAR TRANSISTORS

 

B. Luo, J. W. Johnson and F. Ren

Department of Chemical Engineering

University of Florida, Gainesville, FL 32611

 

K. K. Allums, C. R. Abernathy and S. J. Pearton

Department of Materials Science and Engineering

University of Florida, Gainesville, FL 32611

 

R. Dwivedi, T. N. Fogarty and R. Wilkins

Center for Applied Radiation Research,

Prairie View A&M University, Prairie View, TX 77446

 

D. Schoenfeld

Department of Nuclear and Radiological Engineering

University of Florida, Gainesville, FL 32611

 

ABSTRACT

Large-area (75 mm emitter diameter) InGaP/GaAs Heterojunction Bipolar Transistors (HBTs) were irradiated either with 40MeV protons at fluences up to 5´109cm-2 or with Co60 g-rays to maximum doses of 500MRad.  Both types of radiation produced increases in generation-recombination leakage current in the emitter-base junction.  The dc current gain of the HBTs decreased monotonically with increasing g-ray dose, but was found to increase slightly for proton irradiation due to differential changes in the base and collector resistances.  The HBTs appear well-suited to space or nuclear-industry applications.

 

INTRODUCTION

InGaP/GaAs Heterojunction Bipolar Transistors (HBTs) are widely used in wireless communication applications such as hand-held phones and satellite communication links [1-10].  The radiation hardness of HBTs is one of the critical factors that need to be established for military, space and nuclear industry applications.  Previous results on AlGaAs/GaAs HBTs have indicated that these devices have superior g-radiation resistance to GaAs metal-semiconductor field effect transistors (MESFETs) due to the higher doping levels in the active regions [11-17].  There is little information available on the response of InGaP/GaAs HBTs to ionizing radiation.  The replacement of AlGaAs as the emitter material by InGaP has been prompted by improved resistance to oxidation,lower defect levels, higher etch selectivity and improved contact resistances [2, 5, 10].

In this paper, we report a study of the effects of both g-radiation and proton radiation on the dc performance of InGaP/GaAs HBTs.  The devices were exposed to either a 600Ci Cobalt60 source for accumulated doses of 175 or 500 MRad or to a 40MeV proton beam in the Texas A&M cyclotron for doses of 5´106 or 5´109 cm-2.  The proton irradiation conditions correspond to doses of 3.65 days to 10 years in low earth orbit.  We observed changes in both bulk and surface electrical properties of the HBTs.  The dc current gain decreased from 48 in control devices to 22 for HBTs exposed to g-doses of 500MRad.  By sharp contrast, the current gain increased slightly in proton-irradiated devices due to the base current  decreasing faster than the collector current .

EXPERIMENTAL

            The layer structure were grown by Metal Organic Molecular Beam Exitaxy on undoped, semi-insulating GaAs(100) substrates [18].  The n+ GaAs (n=3´1018 cm-3) sub-collector (0.6mm thick) was followed by an n+ GaAs (n=2´1016 cm-3) collector (0.4mm thick), p+ GaAs (p=7´1019 cm-3) base (0.07mm thick), n+ In 0.5Ga 0.5P (n=8´1018 cm-3) emitter (0.08mm thick) and a 0.02 mm grade to a n+ GaAs (n=1.5´1019 cm-3) emitter contact layer (0.2mm thick).  HBTs with emitter dimension 75mm were fabricated by wet chemical etching.  Lift-off of e-beam deposited Au/Ge/Ni and Ti/PT/Au was employed for ohmic contacts to the emitter/collector and base layer, respectively [19].

            The irradiations were performed in two separate facilities.  For the Co60 source, the calibration of radiation dose was performed with radiochromic films and ion chamber radiation meters [20].  The proton irradiations were performed at the Texas A&M cyclotron with 40MeV protons at fluences of 5´106 or 5´109 cm-2, with the higher dose equivalent to at least 10 years exposure in low earth orbit [21].  The projected range of the protons is > 50mm.  Measurements were taken ~50 hours after irradiation.  The device dc characteristics were measured at room temperature with an HP 4145B parameter analyzer.

RESULTS AND DISCUSSION

(a)    g-Irradiation

While devices irradiated with the 175MRad dose shown no visible changes in appearance, the contact metal morphology was found to be degraded after the 500MRad exposure (Figure 1).  This is not due to a temperature rise of the sample, but appears to be a direct result of the g-radiation on the metal or the oxide underneath the metal [17].

The forward current-voltage (I-V) characteristics from the emitter-base (e-b) junction are shown in Figure 2 (top) as a function of g-dose.  Let us examine the effect of the radiation exposure on the three different regions of the I-V curves, namely the generation-recombination region (VF<0.7V), diffusion region (0.7<VF<1.1V) and the series-resistance section (VF>1.1V).  At very low bias (<0.3V), there is an obvious increase in leakage current whose magnitude is proportional to dose.  At slightly higher bias (0.3-0.5V), the higher dose HBT displays the effect of a reduction in free carrier density.  In the diffusion current region of the I-V curves, the junction ideality factor (n) was calculated and is shown in Figure 2 (bottom).  The n value increases only slightly with dose.  In the region dominated by series resistance, there is a clear increase in resistance for the highest gamma-dose.This is most likely due to a decrease in free carriers because of deep traps created by the absorbed g-rays.

The reverse I-V characteristics from the e-b junction are shown in Figure 3 (top) as a function of dose.  The breakdown voltage (VB) of this junction is determined by the competition between two mechanisms, namely creation of surface defects and bulk traps.  For doses less than 175MRad, the reverse breakdown is dominated by surface recombination defects which increase the leakage current and produce a reduction in VB (Figure 3, bottom).At higher doses the loss of free carriers by trapping into bulk defects dominates and leads to an increase of VB towards the control value.  These results are consistent with the forward I-V data.

A similar phenomenon was observed in the reverse breakdown characteristics from the base-collector (b-c) junction, as shown in Figure 4.

The common-emitter characteristics from the HBTs before and after g-irradiation are shown in Figure 5.  The radiation damage causes two effects.  The first effect is an increase in the turn-on voltage of the collector-emitter junction due to an increase in emitter resistance.  The second effect is a change in slope of the initial part of the curves, due to an increase in collector resistance.  These results are consistent with creation of deep electron trap states in both the emitter and collector layers.

The dc current gain in the HBTs was obtained from the Gummel plots for the different doses, as shown in Figure 6.  The gain was found to decrease monotonically with dose from a control value of 48 down to 22 after the 500MRad dose.  Both the base and collector currents were decreased by radiation exposure.

(b)Proton Irradiation

            The proton doses we employed did not degrade the contact metallization and did not have as large an effect on the forward e-b junction characteristics as the g-radiation.  Figure 7 (top) shows the IB-VBE characteristics as a function of proton dose, while the junction ideality factors are shown at the bottom of the figure.  There is a relatively minor increase in the generation-recombination region of the characteristics, as shown in more detail in Figure 8 (top).  In addition, the increase in series resistance of the junction is small (Figure 8, bottom).  These results show the remarkable resistance of InGaP/GaAs HBTs to proton damage and the highly promising nature of these devices for space-borne applications.

            Similar relatively small changes were observed in the reverse e-b junction I-V characteristics, as shown in Figure 9.  The decrease in breakdown voltage is only ~0.45V even after a dose of 5´109 proton·cm-2.  Changes of a similar magnitude were observed in the base-collector junction breakdown voltage,with an increase from –11.7V in the control devices to –12.3V in those exposed to 5´109 proton·cm-2.

            The common-emitter characteristics before and after the low dose exposure are shown in Figure 10 (top).  For this condition, there is no shift in the turn-on voltage or change in slope of the characteristics.  By sharp contrast to the case of g-irradiation, the collector current actually increases as a result of exposure to the proton beam.  The bottom of Figure 10 shows the ratio of the collector current after proton irradiation to its value in control HBTs.  These results suggest that at least some of the proton-induced damage has n-type nature and decreases the collector resistance.

            Figure 11 (top) shows the Gummel plots before and after proton irradiation, while the bottom of the Figure shows the changes in dc current gain relative to the control value.  Note that the gain remains above its initial value for both proton doses.  The reason for the increase in gain can be determined from the relative changes in base and collector sheet resistance (Figure 12), measured from transmission line data.  The base resistance increases with proton dose (i.e. the carrier density decreases due to the introduction of hole traps), whereas the collector sheet resistance decreases under the same conditions.  It is clear that the defects created by g-radiation and proton damage differ in their electrical properties and hence on the effect on HBT performance.

SUMMARY AND CONCLUSIONS

            The InGaP/GaAs HBTs show a remarkable resistance to both proton and g-ray induced degradation.  The proton-irradiated devices show small changes in junction ideality factor, generation-recombination leakage current and dc current gain at doses equivalent to approximately 10 years in low earth orbit.  The  g-irradiated HBTs show a decrease in current gain of roughly 50% after a dose of 500MRad.  The InGaP/GaAs HBTs appear to be very promising for terrestrial or space-borne applications where radiation-resistance is desirable.

ACKNOWLEDGEMENTS

            The work at UF is partially supported by ONR grant N00014-98-1-0204 and NSF grants CTS-991173 and DMR 0101438.  The work at Prairie View A&M is supported by NASA grant NCC9-114.

 

References

1.      See, for example, P. M. Asbeck, M. –C. Chang, K. C. Wang, and D. L. Miller, in Introduction to Semiconductor Technology: GaAs and Related Compounds, edited by C. T. Wang (Wiley-Interscience, New York, 1990), Chap. 4.

2.      K. Honjo. Solid-State Electron. 38, 1569(1995).

3.      P. M. Asbeck, Solid-State Electron. 38, 1691(1995).

4.      S. Tanaka, H. Hayama, A. Furukawa, T. Baba, M. Mizuta, and H. Honjo, Electron. Lett. 26, 1439(1990).

5.      H. F. Chau, W. Liu, and E. A. Beam III, Proceeding of the 7th International Conference on InP and Related Materials (IEEE, New York, 1995) pp. 640-644.

6.      E. A. Beam III, B. Brau, T. P. E. Broekaert, H. F. Chan, W. Liu, and A. C. Seabaugh, Mater. Res. Soc. Symp. Proc. 421, 3(1996).

7.      T. R. Fullowan, S. J. Pearton, R. F. Kopf, and R. P. Smith, J. Vac. Sci. Technol. B 9, 1445(1991).

8.      M. Hafizi and W. E. Stanchina, in InP HBTs: Growth, Processing, and Applications, edited by B. Jalali and S. J. Pearton (Artech Hourse, Boston, 1994), Chap. 5.

9.      S. Thomas III, K. K. Ko and S. W. Pang, J. Vac. Sci. Techno. A 13, 894(1995).

10.  C. Nquyen and M. Micovic, IEEE Trans. Electron. Dev. 48, 472(2001).

11.  M. A. Listvan. P. J. Vold and D. K .Arch, IEEE Trans. Nucl. Sci. NS-34 1664(1987).

12.  J. F. Sulzman, P. J. McNelty and A. R. Knudson, IEEE Trans. Nucl. Sci. NS-34 1667(1987).

13.  J. J. Liou, Phys. Stat. Solidi A 119, 337(1990).

14.  J. R. Srour and J. M. McGarrity, Proc. IEEE 76, 1443(1998).

15.  G. A. Schrantz, N. W. van Vonno, W. A. Krull, M. A. Rao, S. I . Cong and H. Kroemer, IEEE Trans. Nucl. Sci. NS-35, 1657(1988).

16.  Y. Song, M. E. Kion, A. K. Oki, M. E. Hafizi. J. B. Camou and K. W. Kobayashi, 11th GaAs IC Symp. Tech. Dig. 1989, pp.155-158.

17.  S. B. Witmer, S. D. Mittleman and S. J. Pearton, in InP HBTs: Growth, Processing and Applications, ed. B. Jalali and S. J. Pearton (Artech Hourse, Boston, 1995).

18.  C. R. Abernathy, Mat. Sci. Rep. R14, 203(1995).

19.  F. Ren, A. Y. Cho, Y. M. Kuo, S. J. Pearton, J. R. Lothian, D. L. Sino, R. G. Wilson and Y. K. Chen, Electron. Lett. 31, 46(1995).

20.  B. Luo, J. W. Johnson, D. Schoenfeld, S. J. Peartonaand F. Ren, Solid-State Electron. 45, 1149(2001).

21.  B. Luo, J. W. Johnson, F. Ren, K. K. Allums, C. R. Abernathy and S. J. Pearton, R. Dwivedi, T. N. Fogarty and R. Wilkins, A. M. Dabiran, A. M. Wowchack, C. J. Polley and P. P. Chow and A. G. Baca, Appl. Phys. Lett. (Oct. 8 issue, 2001).

 

Figure Captions

Figure 1.           Optical micrograph of HBT after 500MRad g-ray dose.

Figure 2.           E-B junction forward I-V characteristics (top) and ideality factor (bottom) as a function of g-ray dose.

Figure 3.           E-B junction reverse I-V characteristics (top) and reverse breakdown voltage (bottom) as a function of g-ray dose.

Figure 4.           B-V junction reverse I-V characteristics (top) and reverse breakdown voltage (bottom) as a function of g-ray dose.

Figure 5.           Common-emitter characteristics before and after g-irradiation at different doses.

Figure 6.           Gummel plots (top) and current gain (bottom) as a function of g-ray dose.

Figure 7.           E-B junction forward I-V characteristics (top) and ideality factor (bottom) as a function of proton dose.

Figure 8.           E-B junction forward I-V characteristics from series resistance dominated region (top) and generation-recombination region (bottom) as a function of proton dose.

Figure 9.           E-B junction reverse I-V characteristics (top) and reverse breakdown voltage (bottom) as a function of proton dose.

Figure 10.         Common-emitter characteristics (top) and normalized change in collector current (bottom) as a function of proton dose.

Figure 11.         Gummel plots (top) and normalized current gain (bottom) as a function of proton dose.

Figure 12.         Percentage change in base (top) or collector (bottom) sheet resistance as a function of proton dose.