HAL Id: jpa-00216300
https://hal.archives-ouvertes.fr/jpa-00216300
Submitted on 1 Jan 1975
HAL
is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire
HAL, estdestinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
OPTICAL PROPERTIES OF CdIn2S4 SINGLE CRYSTALS
H. Nakanishi, S. Endo, T. Irie
To cite this version:
H. Nakanishi, S. Endo, T. Irie. OPTICAL PROPERTIES OF CdIn2S4 SINGLE CRYSTALS. Journal
de Physique Colloques, 1975, 36 (C3), pp.C3-163-C3-168. �10.1051/jphyscol:1975330�. �jpa-00216300�
JOURNAL DE PHYSIQUE Colloque C3, suppldment au no 9, Tome 36, Septembre 1975, page C3-163
OPTICAL PROPERTIES OF CdIn2S, SINGLE CRYSTALS
H. NAKANISHI, S. E N D 0 (*) a n d T. I R I E (*)
Department of Electrical Engineering, Faculty of Science and Technology Science University of Tokyo, Noda, Chiba-ken, Japan
R&umB. - L'absorption optique de monocristaux de CdIn2S4 a ete rnesurk a differentes tempe- ratures dans la region spectrale de 0,4 ii 0,8 p. On a trouve que le gap pour les transitions indirectes permises a un coefficient de temfirature de
-
9,5 X 10-4 eVldegrC entre 100 et 300 K et se situe h 2,7 eV, extrapole A 0 K.De la variation thermique du photocourant correspondant A I'inflexion de la courbe photq- courant en fonction de la temgrature, on trouve que le niveau d'energie des centres de recomb!- naison (centres sensibilisateurs) se situe a 0,7 eV au-dessus de la bande de valence. On trouve a partlr des courants stimules thermiquement qu'il y a trois types de niveaux de piegeages temporaires pour les Clectrons B environ 0,1,0,2 et 0,s eV en dessous de la bande de conduction.
Dans une etude des caractbistiques courant-tension, des oscillations du photocourant faible frequence (0,05
A
10 Hz) ont et6 observees. Deux types d'oscillations sont observees : soit sinusoi- dales, soit impulsionnelles. La frequence des oscillations croft avec I'augmentation de I'intensite d'illumination ainsi qu'avec celle de la tension appliquee. Ces oscillations sont vraisemblablement directement associees au vidage des pikges, la recombinaison et la trempe thermique.Abstract. -The optical absorption of CdIn2S4 single crystals has been measured at several temperatures in the spectral range from 0.4 to 0.8 p. The interband gap for the indirect allowed transition was found to have a temperature coefficient of - 9.5 X 10- J eV/deg between 100 K and 300 K and to have a value of 2.7 eV when extrapolated to 0 K.
From a temperature dependence of the photocurrent corresponding to the knee on the photo- current-temperature curve, the energy level of the recombination centers (sensitizing centers) was found to be about 0.7 eV above the valence band edge. It was found frommeasurement of thermally stimulated current that there were three kinds of temporary trap levels for the electrons at about 0.1, 0.2 and 0.5 eV below the conduction band.
In a study of the current-voltage characteristics, low frequency (0.05 Hz-l0 Hz) oscillations of the photocurrent were observed. Two types of the oscillations were observed i. e., sinusoidal type and pulse type. The frequencies of the oscillations increases with an increase of the light intensity and also they do with an increase of the applied voltage. The oscillations are probably directly asso- ciated with exhaustion of the traps, recombination and thermal quenching.
Introduction. - Recently the ternary compound CdIn2S, is attracting increasing attention. One of the reasons for this interest is the fact that CdIn,S, is a photosensitive semiconductor and has a possibility of use as pl~otoconductor in the visible region of the spectrum. The optical absorption, reflection, the photo- conductivity and the photoluminescence have been measured by several workers [l]-[9]. However, the results obtained by these authors were not always in agreement. We have succeeded in growing Iarge and homogeneous single crystals of Cdln2S4 and have measured several transport and optical properties [10]-[13]. The work presented here is the result of more detailed measurements of the optical properties of Cdln2S4 than have been made previously. The result of the study on the low frequency oscillations of the photocurrent in Cdln2S, will also be presented.
(*) Department of Electrical Engineering, Faculty of Engi- neering, Tokyo University of Science, Shinjuku-ku, Tokyo, Japan.
1 . Experimental procedure.
-
T h e samples were cut o u t of the single crystal grown from a melt with excess sulphur. Electrical resistivity of the samples in the dark is about 105 Q-cm. Samples for the optical measurements were thin slices with their surfaces polished by emery paper and diamond paste. A Nihon Bunko SS-50 grating monochrometer was employed. The absorption coefficient a and the sample thickness t are connected by the following equation neglecting multiple reflections,I = Io(l - R)' exp(
-
at)where I , and I are the incident and the transmitted light intensity, respectively and R is the reflectivity of the sample. A plot of the In ( I / I o ) vs t should give a straight line, the slope of which gives cr if R is constant throughout the measurement. The absorption coefficient was determined from such a plot of the relative transmittance of the sample having various thicknesses ranging from 0.03 m m to 0.2 mm. Measu-
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1975330
C3-164 H. NAKANISHI, S. E N D 0 A N D T. JRIE rements below room temperature were made using
a cryostat having quartz windows.
The photocurrent was recorded as potential drop across the series resistance which was selected to be low enough to insure constant voltage across the sample.
2. - Experimental results and analysis. - The absorption coefficient X was measured at several temperatures in the spectral range from 0.4 to 0.8 p.
For the shorter wave length than 0.4 p the apparatus was not sensitive enough to detect the weak transmitted light. In order to determine the energy gap for the direct transition, or2 is plotted as a function of the photon energy in figure 1. Extrapolating a linear por-
2.3 2.4 2.5 2.6 2.7
PHOTON E N E R G Y i e V )
FIG. 1. -- Square of the absorption coefficient of CdInzS4 single crystal at various temperatures as a function of photon
energy.
tjon to or2 = C), :he allowed direct energy gap AE at respective temperatures can be obtained as shown in figure 2. It can be seen that AE varies with tempe- rature T as
with AE(0) = 2.7 eV in the temperature range studied.
The value of AE(0) is in good agreement with that obtained by the reflection measurement by Fujita and Okada [9]. The optical absorption in a longer wave lenght may be due to indirect transitions. Theoreti- cally, square root of the absorption coefficient for the indirect transition should be proportional to the photon energy. Such a plot was made and it was found that a linear relationship was satisfied over the
0 100 2 00 300
TEMPERATURE ( K
FIG. 2. - T h e direct energy gap of CdlnnS4 as a function of temperature.
temperature range studied. This result yields the indi- rect energy gap of about 2.5 eV at 0 K .
Figure 3 shows the photocurrent as a function of the temperature for various light intensities when the sample is illuminated by the light from a mercury lamp. It is clear from the figure that thermal quenching is above a temperature near 200 K. From a tempera- ture dependence of the photocurrent corresponding to the knee on the photocurrent-temperature curve, the energy level of the recombination centers (sensitizing centers) was found to be about 0.7 eV above the valence band edge.
Thermally stimulated current (T. S. C.) was also measured in order to obtain information on the elec- tron traps. The T. S. C. was measured by heating the
- 3 3
10
3 4 5 6 7 8
IO~/TEMPERATURE
(K-')
FIG. 3. - Photocurrent in a single crystal of CdInzS4 as a function of temperature for various illumination intensities.
OPTICAL PROPERTIES O F CdInzS4 SINGLE CRYSTALS sample at constant rate from 80 K after illuminating
the sample by the light from a mercury lamp for 15 seconds. The current showed two broad maxima.
In order to determine the energy lcvels of the indivi- dual trap statcs, the decoyed T. S. C. [l41 was mea- sured. This method consists of heating in the dark the sample, previously illuminated, until1 the measured current i reaches a maximum, then the sample being cooled down immediately and, without illumination, reheated to obtain another current curvc. The pro- cess is repeated several times and for each heating the In i is plotted against 1/T. Welldelined straight lines are obtained with slopes corresponding to some acti- vation energies. The result is shown in figure 4. From
I I I I I
0 200 400 600 800 1000 1200
ELECTRIC FIELD ( V/cm)
FIG. 5.
-
Photocurrent vs electric field for a CdInzS4 single crystal. L , , L2, etc. express the relative light intensities withL, < L2 < L, < L',.
FIG. 4.
-
Decayed thermally stimulated current curves for a single crystal of CdIn2S4.this figure the threc kinds of trap levels were found at 0.1, 0.2 and 0.5 eV below the conduction band.
The trap depth of 0.2 and 0.5 eV are in good agree- ment with the activation energies of the donor levels determined from the Hall measurement. The density of the traps of 0.1 and 0.2 eV depth were estimated from the area of the T. S. C . curve both to be about 3 X 10'' cm-3.
The current-voltage characteristics were studied on the sample immersing in the liquid nitrogen and illu- minating with the light from a mercury lamp. Figurc 5 shows the photocurrent vs electric ficld intensity for several illumination levels. Abovc some critical field, negative differential conductance rcgion appeared.
In this region low frequency (0.05 Hz-l0 Hz) oscilla- tions of the current were obscrved. The region of the electric field in which oscillations are observed depends on the illumination level. Two types of the oscillations were observed corresponding to two different conditions. One of the types is pulse type
and another is sinusoidal one. In some cases the two types are observed alternately. Figures 6 and 7 show oscillograms of the pulse type and the sinusoidal type oscillations, respectively. For both types of the oscil- lations the frequencies depend on the applied field and the illumination level. Figures S and 9 show the fre- quency as a function of the applied field intensity and of the relative light intensity, respectively for the pulse type oscillations. Figures 10 and 1 1 are the similar plots as thosc in figures 8 and 9, respectively for the sinusoidal oscillations.
In order to study a dependence of the oscillation characteristic on the sample dimension, the oscilla- tions were observed for the various thicknesses along the direction perpendicular to thc current flow of a sample. The sample thicknesses examined were 1.7 mm, 1.5 mm, 1.0 mm and 0.6 mm. For the thick- ness of 1.7 mm the sinusoidal oscillations were observcd but the wave forms were irregular. For the thickness of 1.5 mm stationary sinusoidal oscillzt' c ions werc observed. When the thickness was reduced to 1.0 mm, the pulse type oscillations were observed. For the thickness of 0.6 mm breakdown occurred before oscillations were observed and hereafter the sample was no longer photosensitive. It is noted that as the sample thickness is reduced the oscillation wave form changes from sinusoidal to pulse type. For a critical sample size corresponding to the transition from inusoidal to pulse typc oscillations, the both typcs of the oscillations were often observed alternately.
C3-166 H. NAKANISHI, S. E N D 0 AND T. IRIE
FIG. 6. -Pulse type oscillations observed in a CdIn2S4 single crystal. Applied voltage : 368 V. Time scale : 2 sldiv.
0 10 20 30 40 LIGHT INTENSITY (rel. unit 1
FIG. 9. - Frequency vs light intensity for the pulse type oscilla- tions.
FIG. 7.
-
Sinusoidal type oscillations observed in a CdInzS4single crystal. Applied voltage : 483 V. Time scale : 1 sldiv.
'
0 1 I I I I I I
600 700 800 900
ELECTRIC FIELD (V/cm)
FIG. 8. - Frequency vs electric field intcnsity for pulse type oscillations. L4 and L5 express the relative light intensities with
L4 < Ls.
0 -
400 500 600 700
ELECTRIC FIELD (V/cm
FIG. 10. - Frequency vs electric field intensity for the sinusoidal type oscillations.
From the facts described above, it may be concluded that the mechanism of the oscillations is essentially the same as those of the oscillations observed in CdS and ~ d [15], ~ e1161 and that of the oscillations recently observed in CdIn2S, by Allakhverdiev et al. [17]. The sinusoidal oscillations are repetition of the current peak due to the exhaustion of the electron traps by avalanche-type process caused by the Joule heat and the successive recombination. The pulse type oscillations are, in addition, thought to be concerned with the thermal quenching because the current decreases to the value lower than the stationary one.
This interpretation is supported by the experiment on the size dependence of the oscillation characteristic above-stated, that is, as the sample size is reduced
OPTICAL PROPERTIES OF CdInzS4 SINGLE CRYSTALS C3-167
0 2 4 6 8 1 0
LIGHT INTENSITY (rel. unit)
FIG. 1 1 . - Frequency vs light intensity for the sinusoidal type oscillations.
the temperature rise in the sample becomes rapid and large, leading easily to the thermal quenching.
V. L. Vinetskii et al. [l 51 derived a condition for the instability as follows :
MLT' ePE2(cM/k T) > T Q , p c v , (3) where M is the trap concentration, 1 the generation rate of the conduction electrons, z the free electron life time, e the electronic charge, p the electron mobility E the electric field, the trap depth, k the Boltzmann constant, Q, = Q exp(- c,/kT) (Q is the density of states at the edge of the conduction band), p the density and C, is the specific heat of the sample.
Typical values of the parameters in the case of CdIn2S,
results in the value of the threshold field for instability, E,, E 103 V/cm. This value agres in order of magni- tude with the observed values.
When a sample was subjected to additional illumi- nation at 6 000
A,
the oscillations, both of the sinu- soidal type and the pulse type, were stopped. The additional illumination was removed, then the oscil- lations again started. The photon energy correspondingto 6 000
A
is about 2.06 eV and is approxiniately equal to the energy difference between the conduction band and the sensitizing level or that between the trap level and the valence band.3. Conclusions. - The interband gap for the direct allowed transition of CdTn2S4 single crystal was found from the optical absorption measurement to have a temperature coefficient of - 9.5 X 10-4 eV/deg between 100 K and 300 K and to have a value of 2.7 eV when extrapolated to OK. The absorption edge was found to have low energy tail. If we attribute this tail to the indirect transition, we obtain the indirect gap as about 2.5 eV at 0 K. This value is nearly the same as that obtained by the electrical measurement.
FIG. 12.
-
Energy level diagram for the electron traps and the sensitizing centers in our samples of CdInzS4.////////////////
Conduction
Band
/ /
The low frequency oscillations of photocurrent in Cdln2S, are closely connected with the electron traps the energy levels of which are 0. l , 0.2 and 0.5 eV below conduction band and the sensitizing centers the level of which is 0.7 eV above the valence band. Figure 12 is the energy level diagram of these electron traps and these sensitizing centers. The mechanism of the oscillations is essentially the same as those of the oscillations observed in CdS and CdSe.
A - o - l e v 0 . 2 ~
D I
ID2
Acknowledgment. - The authors are much indebted to Mr. T. Kolniya for help with the experiments.
A
0.5eV
References
A I 3 3
[ l ] K O E L M A N S , H . ~ ~ ~ G K I M M E I S S , H . G . , Physica25(1959)1287. [4] ARDULLAEV. G . B., .~NTONOV, V. B., GUSEINOV, D. T., [2] BEUN, J. A., NITSCHE, R. and LICHTENSTEIGEK, M,, Physica NANU, K. Kh. and SALAEV, E. Yu., SOV. Phys.-Semi-
27 (1961) 448. cor~d. 2 (1969) 878.
[3] BOLTIVETS, N . S., DKOBYAZKO, V. P. and MITYUREV, V. K., [ S ] BEUN, J . A., NITSCHE, R. and LICHTENSTRIGER, M., Physica
Sov. Phys. Sen~icond. 2 (1969) 867. 26 (1960) 647.
C3-168 H. NAKANISHI, S. E N D 0 AND T. IRIE [6] RADAUTSAN, S. I., Smnu, N. N., TEZI-EVAN, V. E., SI-IER-
BAN, K. F. and BARAN, N. P., Phys. Stut. Sol. (b) 57 (1973) K 93.
[7] CZAJA, W. and KRAUSHAUER, L., Plzys. Slat. Sol. 33 (1969) 191.
[8] RADAUTSAN, S. I., MOLODYAK, I. P., SYRBU, N. N., TEZLE- VAN, V. E. and SRIPITKA, M. A., Phys. Staf. Sol. 49 (1972) K 175.
[9] FUJITA, H. and OKADA, Y., Japan J. Appl. Phys. 13 (1974) 1823.
[l01 SUDO, I.,ENvo, S., IRIE, T. and NAKANISHI, H., J. Phys. SOC.
Jupan 31 (1971) 949.
[l11 ENDO, S., SIJDO, I. and IKIE, T., J . Phys Soc. Japan 29 (1970) 519.
[l21 NAKANISHI, H., ENDO, S. and IRIE, T., Jupan. J. ApN. Phys.
12 (1973) 1646.
[l31 ENDO, S. and IRIE, T., J. P h y ~ . SOC. Japan 36 (1974) 1210.
[l41 BUBE, R. H., DUSSEL, A., Ho, C. T. and MILLER, L. D., J.
Appl. Phys. 37 (1966) 21.
[l51 VINETSKII, V. L., SOY. Phys.-Solid State 11 (1969) 1140.
[l61 VINETSKII, V. L., SHAKHOVTSOVA, S. l . and KOKOZEKKO, I D., Phys. Stal. Sol. 42 (1970) 95.
[l71 ALLAKHVERDIEV, K. R., ANTONOV, V. B., NANI, R. Kh. and SAI-AEV, g . Yu., SOV. Phys.-Sertricotrd. 6 ( 1 972) 507.
