Signal generation in a hydrogenerated amorphous silicon detector

Lawrence Berkeley National Laboratory Recent Work Title SIGNAL GENERATION IN A HYDROGENATED AMORPHOUS SILICON DETECTOR Permalink https://escholarship.org/uc/item/1665g656 Authors Bureshi, S. Perez-Mendez, V. Kaplan, S.N. Publication Date 1988-10-01 eScholarship.org Powered by the California Digital Library University of California LBL-25848 Preprint Lawrence Berkeley Laboratory UNIVERSITY OF CALIFORNIA Physics Division Presented at the IEEE 1988 Nuclear Science Symposium, Orlando, FL, November 9-11, 1988 and to be published in the Proceedings ...:~i:IVE(... LAWRENCE BER!(El~Y LABORATORY .. JAN 3 Signal Generation in a Hydrogenated Amorphous Silicon Detector 1989 LIBRARY AND DOCUMENTS SECTto· S. Qureshi, V, Perez-Mendez, S.N. Kaplan, I. Fujieda, G. Cho, and R.A. Street October 1988 -]. Prepared for the U.S. Department of Energy under Contract Number DE-AC03-76SF00098. DISCLAIMER This document was prepared as an account of work sponsored by the United States Government. While this document is believed to contain correct information, neither the United States Government nor any agency thereof, nor the Regents of the University of California, nor any of their employees, makes any warranty, express or implied, or assumes any legal responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by its trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or the Regents of the University of Califomia. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Govemment or any agency thereof or the Regents of the University of California. ~.. " .. ~) Signal Generation in a Hydrogenated Amorphous Silicon Detector S. Qureshi, V. Perez-Mendez, S.N. Kaplan, I. Fujieda, and G. Cho Physics Division Lawrence Berkeley Laboratory 1 Cyclotron Road Berkeley, California 94720 R.A. Street Xerox Park Palo Alto, California 94304 October 1988 .. ,. SIGNAL GENERATION IN A HYDROGENATED AMORPHOUS SILICON DETECTOR S. Qureshi, V. Perez-Mendez, S. N. Kaplan, I. Fujieda, G. Cho Lawrence Berkeley Laboratory, Berkeley, CA 94720 and R. A. Street Xerox Park, Palo Alto, CA 94304 Abstract The signals produced in thick hydrogenated amorphous silicon p-i-n detectors were measured using incident light pulses with different mean free paths. The signal shapes as a function of bias potential were analyzed in terms of the relevant parameters: mobilities and mean free paths of the electrons and holes. These latter were measured by transient photoconductivity methods using a pulsed nitrogen-dye laser system. I. Introduction In this paper we analyze the characteristics of the signals produced in an amorphous silicon p-i-n detector subjected to various particle and photon radiations. Previously we1•2 and others 3 •4 have shown that alpha. particles and low energy protons produce signals in which there is some amplitude loss due to charge recombination in the detector. In later pa.pers 5 •6 we give the value for W, the average energy used in producing 1 e-h pair, measured using pulsed x-ra.ys. We also show that minimum ionizing particles (1.5 MeV electrons) produce ,...., 60 e-h pairs per micron of a-Si:H with no measurable recombination loss 7 . used model calculations that use measured values of mobility JL, carrier lifetime r, and the ionized dangling bond density Nd as fitting parameters. The model is used to explain the signal shape of Fig. 1 and is described in Section V. • • 510nm (£1ec1ron) o 760nm(£1ec. 4: Hole) .ri 0.8 - 5 E01 'Q; • 510nm • • 0.11 - 0 • • Cl) Ill 0 - 0 0.2 - • :I a. :; 0 0 • 0.4 0 • 0 0 The signals that we measure and analyze in this paper were produced by 1 microsecond pulses from diodes and lasers emitting light at ,\ = 760 nm and at 510 nm. The 760 nm light has a mean free path ~ 100 JLm in the a-Si:H and hence its signal simulates that of a minimum ionizing particle. The 510 nm light has an absorption mean free path ,...., 0.2 JLm and can be used to provide information on the individual contribution to the signal of the electron and hole motion, depending on whether the light is incident on the p or on the n side of the detector. These kinds of experiments provide detailed information about the internal field and charge collection, which allows the performance of a-Si:H particle detectors to be predicted. The material parameters: mobilities, and lifetimes for the electrons and holes, and the density of ionized dangling bond states were measured using the transient photoconductivity method and were done with a. pulsed dye laser as described below. • • • ••• 0 0 tOO 200 300 Figure 1 shows signals from 510 and 760 nm light in a 27 J.tm p-i-n detector as a function of bias. 510 nm light incident on the p side of the detector produces a signal where the holes are collected promptly and the electrons drift to the n side. The signal increases with bias as the depletion layer extends from the p side to then side of the detector. No appreciable hole signal is measured when light is incident from the n side of the detector at low bias. This is due to the fact that the depletion layer does not extend to the n side until a. sufficiently large bias is applied across the detector. The full depletion is indicated by the onset of hole collection and occurs a.t ,...., 375 V, as shown in Fig. 1. We have • • • • 400 • 600 Bias (V) Fig. 1. Signal vs bias curves for 760 - and 510 - nm diode light pulses. The sharp rise in the hole signal at 375V is described in text. III. Derivation of Signal Produced by 510 nm Light Pulses The induced charge dQ in a.n external circuit produced when a charge q is displaced by a. distance dx in a semiconductor detector bounded by two parallel plane electrodes with separation d is 8 - 10 dQ = q.dx d (1) II. Signals Produced by Light Pulses • 0 0 (Hole) :X: :; a. • • Equation (1) differs from the expression dQ = qdV VD frequently used in gas detectors which assumes a uniform electric field in the sensitive volume. The total distance moved by an excited electron or hole is given in terms of the electric field E and the lifetime T by Xmtu: = IJTE. Complete collection occurs when Xm 1u: is larger than the sample thickness. In general, however, the electric field is not uniform in the material. In an a-Si:H detector the i layer being the active detector region, it is of interest to know the electric field distri- "' 0.2pm) so that the total charge from the carriers is bution in it. Under reverse DC bias, the electric field distribution in the i layer is obtained by solving the Poisson equation. The model calculation assumes that 11 •12 , under sufficient DC bias, the fd d Qo = e Jo n(x)dx = enoXo(l- exp(- X <I2V Po dx2 =- coc - when~ Po = eNd; Nd Nd v(x} = pE(x} is constant, then the electric field falls off linearly. At low potential, the gap states are only partially ionized and the Poisson equation is of the form p(V) V<Vc dx2 =-~ density of carriers decays exponentially with time ~:; n(t) where p(V) = en'V and n 1 = "'7 X 10 14 cm- 3 , Vc V, being the critical voltage. where T / • (7) is the carrier lifetime. Q e fd = d lo IT' dx lo n(t)v(x)dt (8} where v( x} is the drift velocity of the carriers generated at x. The normalized charge collection efficiency, e:, is basically a function of the following parameters. eu :>,cJ ., e: G) = go = f(p, r, Nj, T,, ~. d, Xo) (9) Only the first three parameters are material related which we measured by standard transient photoconductivity experiments . l.i: 10. CD IV. Measurement of Mobility, Mobility-Lifetime Product, and Dangling Bond Density ~-'?~.'!'..... . jjj 300~- 1r/ We have used transient photoconductivity experiments to measure carrier mobility p, the mobility-lifetime product pr, and the ionized dangling bond density Nd in thicker (27 pm) detectors. A 100V 1o' t = n(O)exp(--] T . Finally, there is no loss of collected charge at the preamplifier. With these assumptions, the collected charge is expressed by = 1 volt). 10. u (6) density of carriers and the detrapping time is very long in comparison with the shaping time, T., of the electronics so that the 10' .g (5) Fourthly, the density of deep traps Nd is much larger than the The solution of this equation gives an exponentially dropping off field. Based on this calculation, Fig. 2 shows the electric field distribution in a 27 pm sample (ionized dangling bond density NJ enoXo Thirdly, mobility of the carriers is assumed constant in the region· of interest, independent of electric field, so that the velocity of carriers at any position, x, is simply proportional to the electric field at that position. V»Vc being the density of ionized defects. If )]::::: 0 dangling bonds are almost fully ionized and the Poisson equation has the form (deep depletion model] 0 6 10 16 20 26 pulsed nitrogen-dye laser system with 3-ns pulse width is used for this purpose at a wavelength of 510 nm, which has an absorption mean free path of "' 0.2pm in a-Si:H. 30 Distance (f.Lm) Fig. 2. Calculated electric field profiles in 27 pm p-i-n detector at various bias voltages. The calculation is based on measured ionized dangling bond density,...., 7 x 1014cm- 3 . In a conventional time-of-flight experiment 13 the drift mobility is determined by measuring the carrier transit time by applying a pulsed bias a few tens of microseconds before photo-injection of carriers. This ensures that at these short times very few dangling bonds are ionized. Hence, to a good approximation, the electric field is uniform through the bulk of the i layer. The photocurrent J resulting from the applied field is In order to calculate the charge collection due to one type of carrier injected near the appropriate contact (for example, holes at n contact), several assumptions are made. First, the exponentially varying electric field has negligible effect on the signal in comparison with the linear part so that the deep depletion model with linear electric field is assumed to be valid through the entire i layer. eN* E(x) = _ d (W- x) (2) E J = 1Jnepd (10) fOf where n is the number of photons absorbed, 1J is the carrier generation efficiency, J.l is the drift mobility of the carrier moving across the sample, and d is the thickness of i layer. where W, the depletion thickness is W= (3) The mobility is obtained from the transit time Tt, d <[2 p=-=- Secondly, the distribution of photo generated carriers is of the form X (4} n(x) = noexp(--) Xo For nondispersive transport T 1 is defined as the time at which J = 510nm, Xo drops by 50%, while for dispersive transport in which the mobility where Xo is mean absorption depth in a-Si:H (for>. ETt 2 VTt (11) .,/ is time-dependent, it is given by the change of slope in a log J log t plot. The mobility-lifetime product p.r is determined by measuring charge c'ollection (equal to J Jdt} as a function of electric field when a uniform electric field is applied. The charge collection is given by the Hecht equation, [ where Q ners. 1- exp( _ _ d p.rE = Q = 77ne corresponds 0 >] (12) to complete transit of all car- Fig. 3 shows time of flight data for electrons in the p-i-n 27 JLm sample for several different pulsed bias voltages. The electron transport is non-dispersive. The drift mobility 1-'e ,..., 1.3 ~ v 3ec is measured at 200V bias. Hole mobility is obtained from time of flight data shown in Fig. 4. Hole transport is dispersive with dispersion coefficient <X"' 0.6 and transit time is obtained from the change in slope of log J vs log t plot. A hole mobility value of 0.004 2 Viet is measured at 200V. The variation of electron and hole drift mobilities as function of electric field is shown in Fig. 5(a,b). It is seen that the mobility of holes increases by a larger percentage than for electrons which is a known effect of the dispersive drift mobility. In the range limited regime (JLrE~d}, the transient response is characterized by a deep trapping life time r. The charge collected is then given by (13} The field dependence of Q is used to find the JLT product by plotting Q vs E. JLT is obtained from the fit to the Hecht formula. The ionized dangling bond density is determined by measuring the transient photoconductivity current signal from the sample with equilibrium DC bias as follows. The depletion field depends on the distance x into the sample, so that the photocurrent is time dependent, corresponding to the drift of the charge packet down the field. p.E(t) J(t) = 7]ne-- (14} d For the carriers that start at x X= l 1000 -.:5 r--------------------., • 100 .ri 100 c~ • D a 0 0 • 10 . .....• D v 200 V • 300 v .......-;.: ........-a-,0 0 D D DOot~a:b D e • D • e'b .s:: D.. e c D CD ·c:;; • ':. •'I c: <0 F = 0 and t = 0, x is related to t by 10 (15} JLE( t')dt' 100 1000 Time (nsec) E(x) can therefore be obtained directly from measurement of J(t) using equation (15}, provided E(t) and the mobility are known. Since the current is proportional to the field, E(t) can be found by comparing the current to that obtained for a known applied field and using equation (10). In this way, n77 does not need to be known, provided it is the same for pulsed and DC bias experiments. The ionized dangling bond density is deduced from the slope of linearly decreasing part of the DC bias photo-current transient by using equation (2}. Table 1 shows JLe, Jl.h, JI.Te, JI.Th and ionized dangling bond density N;i of various samples measured. Fig. 3. Transient photo current electron signal at different pulsed voltages using 510 nm laser light. Non-dispersive transit through 27 JLm p-i-n sample is seen clearly. 100 .ri .:5 10 c ~ (a-0.6) ~ u 0 0 .s:: D.. Table 1. Measured Parameters from 4 Samples • 200 v c G) 12 Thickness (J.Im) • J.le 27 12 ·~ 28 <0 Type p-i-n p-i-n n-i-p n-i-p Maker GSI GSI Xerox Xerox 1.2 1.2 1.1 1.4 2 (cm /Vs) Jl.eTe ( cm 2 /V) Jl.h (cm2 /Vs) Jl.h rh ( cm2 /V) Nd (cm- 3 ) 1.1 X 10- 7 1.2 0.004 1.0 X 10- 7 1.s x 10- 7 1.1 0.004 10- 8 1.2 7 X 1014 X X 0.004 10- 8 1.6 7 X 1014 X X 7.7 X 10 7 8.4 X 10 7 9.0 X 10 7 6.6 X 107 NdJl.hTh (1/cmV) 7.0 X 10 6 8.4 X 106 9.6 X 10 6 1.6 X 107 v 100 Fig. 4. Transient photo current hole signal from 27 JLm p-i-n detector at different pulsed bias voltages. Hole mobility is dispersive with dispersion parameter <X"' 0.6 at 300V. 6 X 1014 NdJl.eTe (1/cmV) 300 V Time (J.tsec) 10- 7 10- 8 D • 400 F 10 0.003 10- 8 2.7 6 X 1014 X • • 0.1 3 2.0,------------------. V. Fit to Hole Signal Using Measured Parameters and Nj '-'• -r Fig. 7 shows the fit to the measured hole signal of Fig. 1 using the model calculation of signal discussed in Section III. The dots in Fig. 7 are experimental data points for 510 nm light. The fit uses the measured values 1.6 .. ....... i.O cm 2 Vsec P.h = 0.004-0.5 8 & cm2 JlTh = 1.2 X 10- - V · sec Fig. 7 also shows calculated hole signal for three values of Nj. A value of N;i = 7.1 x 10 14 cm- 3 fits the measured signal FIG. 5-e 0.0 best. This value of Nj is in good agreement with the measured. 0.010 J dangling bond density obtained from the transient photoconductivity experiments. .. "' u 0.008 ...... > E 0.008 ~ :.s0 2 J.J.h- .004 cm /Vsec ..ci s ~ 0.8 2 J.J."T"- 1.2 x 1o·• cm /V > c 0 0.004 ::E ., CD ·o .g ;;;: 0 J: 0.8 Calculation w c 0.002 0 ~ FIG. 5·b 0.000 0.0 10.0 6.0 4 20.0 16.0 0.4 0 u Electric Field (10 V/cml 8.8xl0 cm" 3 Nd- 7.1x10 cm· 3 Nd- 7.4x10 cm' 3 ·CD ....> <tl Fig. 5. Electron and hole mobility vs electric field in 27 JLm p-i-n detector. a) Electron mobility; b) hole mobility. 0.2 a: 14 .• • 0 200 14 Experimental Q) Fig. 6 shows charge collection for sample of Fig. 3 at various DC biases from which the field profile and ionized dangling bond density is obtained. 14 Nd- 300 360 400 460 600 Bias (V) Fig. 7. Measured, •, and calculated hole signal threshold using 510 nm light pulses for different assumed values of ionized dangling bond density. Summary and Conclusions 100 • 100 ..ci D ~ ~ ;:; • • u 0 0 0 .s:::. The transient photoconductivity measurements on relatively thick samples of a-Si:H show ionized dangling bond densities in the range of 6-7 x 10 14 cm- 3 . The electron transport on the samples is found to be nondispersive with high field electron mobility 2 in the range of 1.1-1.3~; the hole mobility ranging from 0.003 200 V • JOO V c a. v 2 10 0.007 {/';. c is dispersive with <X"- 0.6. While the electron mobility increases by "' 20% at high field, hole mobility increases by "' 40% from the low field values. This increase in mobility is attributed to dispersion of the transport, but there may be the Poole-Frenkel effect with the non-dispersiveelectrons 14 • (JLT NJ)e • c D • CD ·u; c F"' • ' 0 values range from 6.6 x10 7 to 9 x10 7 and (JLrNd)h range from 10 20 30 40 7x10 6 1.6x107 • to These values are in good agreement with values reported on thinner samples 15 . We find that the fit to the hole signal based on model calculation discussed in Section III is very sensitive to the ionized dangling bond density value N;i. 60 Time (nsec) Fig. 6. Transient photo current electron signal at different DC bias voltages from 27 Jlm p-i-n detector using 510 nm light. This sensitivity and a good agreement between measured dangling bond density value from the transient photoconductivity experiment and the one obtained from the calculated fit suggests that this provides a useful method for the measurement of ionized dangling bond density in thick samples. 4 ./ Acknowledgements 6. I. Fujieda, G. Cho, S.N. Kaplan, V. Perez-Mendez, S. Qureshi, W. Ward and R.A. Street, "Detection of Charged Particles in Thick Hydrogenated Amorphous Silicon Layers," Symposium Proceedings, Materials Research Society, Vol. 118 (1988) 469. We thank Drs. C.C. Tsai (Xerox) and P. Bhat (GSI) for makiri.g these samples. This work was supported by the Director, • Office of Energy Research, Office of High Energy and Nuclear PhysiC!!, Division of High Energy Physics of the U. S. Department of E;;ergy under contract #DE-AC03-76SF00098. ... . 7. V. Perez-Mendez, S.N. Kaplan, G. Cho, I. Fujieda, S. Qureshi, W. Ward and R.A. Street, "Hydrogenated Amorphous Silicon Pixel Detectors for Minimum Ionizing Particles," Nucl. Instr. and Methods, A273, No. 1 (1988) 127. References 8. P.A. Tove and K. Falk, "Pulse Formation and Transit Time of Charge Carriers in Semiconductor Junction Detectors," Nucl. Instr. and Methods 29, (1964) 66. 1. V. Perez-Mendez, S.N. Kaplan, W. Ward, S. Qureshi and R.A. Street, "Signal, Recombination Effects and Noise in Amorphous Silicon Detectors," Nucl. Instr. and Methods, A260 (1987) 195. 9. S. Ramo, Proc. I.R.E. 27, (1939) 584. 10. G.K. Jen, Proc. I.R.E. 29, (1941) 345. 2. S.N. Kaplan, J. Morel, T.A. Mulera, V. Perez-Mendez, C. Schnurmacher and R.A. Street, "Detection of Charged Particles in Amorphous Silicon Layers," IEEE Trans. Nuclear Science NS-33 (1986) 351. 11. V. Perez-Mendez, J. Morel, S.N. Kaplan and R.A. Street, "Detection of Charged Particles in Amorphous Silicon Layers," Nucl. Instr. and Methods A252, (1986) 478. 3. B. Equer and A. Karar, "Effect of Primary Ionization in Amorphous Silicon Detectors," Nucl. lnstr. and Methods, A271 (1988) 574. 12. J.D. Cohen and D.V. Lang, "Calculation of the Dynamic Response of Schottky Barriers with a Continuous Distribution of Gap States," Phys. Rev. B, 25, No. 8, 5321 (1982). 4. J. Dube(!.u, T. Pochet, A. Karar, L.A. Hamel, B. Equer, J.P. Martin, S.C. Gujrathi, and A. Yelon, "Response of a-Si:H Detectors to Protons and Alphas," Symposium Proceedings, Materials Research Society, Vol. 118 (1988) 439. 13. R.A. Street, "Measurements of Depletion Layers in Hydrogenated Amorphous Silicon," Phys. Rev. B, 27, No. 8, (1983) 4924. 14. R.M. Hill, "Poole-Frenkel Conduction in Amorphous Solids," Phil. Mag. 23, 59 (1971). 5. S.N. Kaplan, I. Fujieda, V. Perez-Mendez, S. Qureshi, W. Ward and R.A. Street, "Detection of Minimum-Ionizing Particles in Hydrogenated An:orphous Silicon," Proceedings of the London Conference on i'osition-Sensitive Detectors, Sept. 1987, LBL-23961, Nucl. Inst. and Methods A273, Nos 2,3 (1988) 611. 15. R.A. Street, "Trapping Parameters of Dangling Bonds in Hydrogenated Amorphous Silicon," Appl. Phys. Lett. 41, No. 11, (1982) 1060. I " 5