MODULATION EFFECTS IN INDIUM FILMS

Electrocomponent Science and Technology1977, Vol. 4, pp. 19-22

(C) Gordon and Breach Science Publishers Ltd., 1977Printed in Great Britain

OPTICAL MODULATION EFFECTS IN INDIUM OXIDEFILMS

J. AVARITSIOTIS and R. P. HOWSON

Department ofElectrical and Electronic Engineering, University of Technology, Loughborough,Leicestershire, U.K.

(Received May 28, 19 76; in final form April 5, 19 77)

Tin doped indium oxide films produced by chemical vapour deposition and by d.c. sputtering have shown an opticalreflectance spectra which can be modulated in the region of the plasma edge in the near infra-red by an electric fieldapplied to the surface through an electrolyte. The effect produced could be quantitatively predicted by a theorydeveloped for surfaces whose optical properties were controlled by their free carrier properties and for the samplebeing in thin film form.

The plasma edge for films prepared by chemical vapour deposition on some occasions showed a modified plasmaedge effect which was attributed to surface plasmons allowed by roughness effects.

1. INTRODUCTION

Tin doped indium oxide films arouse interest in thatthey can be produced in an electrically conductingform whilst maintaining optical transparency to visibleradiation. They have found application in the radia-tion insulation of sodium lamps and domesticwindows, which leads to their exploitation in solarenergy converters, electro-optic devicesa and thinfilm optical wave-guides.4 They have provided veryconvenient electrodes for electro-chemical experi-ments where effects are being studied which occurclose to the surface of the electrode, The effect ofthe applied electric fields on the optical properties ofthe films has not been studied.

These films have optical properties which, for thespectral region where the transition from an opticallytransparent to an opaque highly reflecting stateoccurs, depend upon the properties of the freecarriers independent of electron transitions betweenallowed energy bands which are far removed fromthese wavelengths. This offers a system where thevalidity of recent quantitative theories applicable toconducting surfaces and to thin film samples can betested. Previous measurements have been made ongold and silver where there is considerable inter-ference from inter-band effects.6 The optical proper-ties of the film in the transition region are ofconsiderable interest in any optical application inthat the limit of the range of transparency and thelimit of the range of high reflectivity and lessemissivity are defined.

19

2. EXPERIMENTAL DETAILS

Films were used that had been prepared by chemicalvapour deposition (C.V.D.) and by d.c. sputtering;both on to glass substrates. The C.V.D. films wereprepared by the pyrolytic decomposition of indiumand stannic chloride in methyl alcohol on a substrateheated to about 500 C, at a rate of about 200 nm s-1

Sputtered films were made from prepared sources byD.C. sputtering of conducting mixed compactedsource oxides onto glass substrates at a pressure of10-4 torr at room temperature. The film thicknesseswere in the range 26 nm to 59 nm and were measuredfrom the thickness interference fringes shown as afunction of wavelength in the visible region of thespectrum where the films were transparent.

The free carrier concentration of Sn doped filmsof In20a films is shown to vary between approxi-mately 0.3 to 1.5 x 1021 cm-a dependent upon thenumber of oxygen vacancies introduced during thefabrication process,s These densities give plasmawavelengths in the near infra-red. Optical measure-ments in the region of most interest, the near infra-red, were made by point to point plotting of normalincidence reflectance in comparison to that of afreshly prepared aluminium coated mirror. A doublegrating monochromator was used with a Golaydetector and phase sensitive processing techniques.For the visible region of the spectrum a double beamrecording spectrophotometer (Unicam SP 5000) wasused. Inspection of the surfaces of films of similarthickness with a scanning electron microscope showed

20 J. AVARITSIOTIS AND R. P. HOWSON

that in general the films prepared by sputtering wereconsiderably smoother with graduated variations ofthe surface with a periodicity of about 250nm com-pared to much sharper variations at a period of about100nm for those prepared by C.V.D.An electric-field was applied to the surface of the

oxide film through the double layer of an electrolyticcell. The electrolyte, chosen after experimentalinvestigation, was aceto-nitrile which gave a largedouble layer capacitance (0.16 F/m2) such that alarge electric field could be applied to the surface buthad some absorption bands in the spectral region ofinterest which limited the measurements. The electro-lytic cell incorporated a sodium chloride windowthrough which the radiation passed onto the sample,and a large surface area platinum mesh counterelectrode. This system allowed that modulationvoltages of up to 20 V to be applied without anyevidence of electro-chemical activity and degradationof the reflectivity of the sample. Phase sensitivedetection techniques, with the modulating voltageproviding the reference, enabled sufficient sensitivityto be obtained for measurement of the relativelysmall electro-reflectance effect.

3. THEORY

The position and sharpness of the plasma edge shownby the films is dependent upon the preparation of thefilms, as is the electrical conductivity, through thefree carder density, N, and the relaxation time, z.

The description of the optical properties aroundthe plasma edge is given by the theory of Drude, i.e.

( cop2 ) (1)e eL6o

2 "1"/(.D’l"-1

where e is the dielectric constant of the materialand eL that of the lattice alone. COp is the plasmafrequency given by

CO Ne2/m*eoer, (2)

m* is the effective mass of the free carriers. Thereflectance, R, is given by

(no n) + gR (3)

(no + n)2 + k2

where e (n ]k)2 n is the refractive index and kthe extinction coefficient. With some materials it ispossible to observe a kink in the plasma edge whichis attributed to an additional mode of resonance

due to surface plasmons allowed by surface rough-ness; the frequency is given from: 7

ee(COs) -e(COs)

where ee is the dielectric constant of the electrolyte,the other material at the interface. When CO" >>then the Drude model gives

pCO (4)

X/(ee/eL) +

The normalised changes in reflectance due to anelectric field at the surface of a thin film have beengiven as: 6

AR(I42,- W2)" RI"V (5)

RV ERS

where R and AR refer to the reflectance of the thinfilm system in terms of the surface effect R 1.

and

W

1 + (R[/2/R/) exp(-) cos ++ 2H(R/2/RI/2) exp(-/3) sin

+ (R2/R1) exp(-2) ++ 2(R/2/R/2) exp(-/3) cos

(6)

RIR2 exp(-2/) + R/2R/2exp(-) cos 0+ 2HR 2R 2 exp(-3) sin 0

1 + RxR2exp(-2) ++ 2R/2R/2exp(-{3) cos 0

(7)

COr(n2 k2 n) + 2nk

n2 k2 no + 2nkCOr(8)

{3 (2cod/c)k, and 0 are overall phase shifts,6 d isthe film thickness, R r , R2 r z’, r refersto the interface between electrolyte and filmmaterial and r2 to that of the film and the substrate.Finally,

4noqr

cm*eo-eA + e + eB CO’r ](CO2 7.2 - ) [(eA ee)2 +

(9)

where" eA n2 k2 eB 2nk, ee n no is therefractive index of the electrolyte and n that of thesubstrate.

OPTICAL MODULATION OF In Oa FILMS 21

C is the capacitance of the configuration used forthe application of the low frequency electric field onthe surface of the film (i.e., interface 1).

The important role of m*, r, eL in the spectraldistribution of the effect is evident; a large electro-reflectance effect is expected in materials where thefree carrier mobility, t (rim*q) is large and eL issmall. This in terms of reflectivity implies that a sharpreflectance plasma edge and a low value for thereflectivity minimum are the necessary features of amaterial to yield large reflectance modulation per-centages. The density of the free carriers determinesthe wavelength at which the plasma edge appears.

4. RESULTS

Figure shows the plasma edge of two films preparedby C.V.D. and illustrates the kink in the edge which isattributed to surface roughness and has previouslybeen observed by Kostlin. In general it is possible tosay that these films prepared by C.V.D. gave sharperplasma edges i.e. higher values of r and therefore freecarrier mobility than those sputtered and this isattributed to the greater structural order given by thehigher substrate temperatures that were used.

1.4

>, 1.2

1.1 p

80

60

40

2 3 4 5X (I.J m)

FIGURE The reflectivity spectra of two Sn doped InOafilms prepared by C.V.D. showing the effect of surfaceplasmons.

2 3 4 5 6

X (lm)

60

FIGURE 2 (a) Experimental and (b) calculated electro-reflectance spectra for a Sn doped Inz Oa film of 27nmthickness prepared by C.V.D. Also shown is the reflectancespectra and the absorption bands in the aceto-nitrile electro-lyte which prevented measurements being made.(Electrolytic capacitance, C, was 10/.tFm-2 ).

Figures 2 and 3 show the effects observed with afilm prepared by C.V.D. and d.c. sputtering respec-tively; the reflectance and the experimental andtheoretical electro-reflectance are shown as a functionof wavelength. The theoretical effect is obtained byanalysing the experimental plasma reflectance edge,and adjusting the parameters in a computer programmegiving the reflectance of a thin film on a substrateaffected only by free carder and dielectric opticaleffects until a good match was obtained.1 This givesthe optical constants of the material in that region ofthe spectrum. This could then be used to substitute inthe theoretical formula, Eq. (5), to predict the electro-reflectance effect.

It can be seen that very good agreement isobtained both in the estimation of the magnitude ofthe effect and in its spectral distribution especiallyconsidering the approximation and assumptions

22 J. AVARITSIOTIS AND R. P. HOWSON

o 4.5

40

2 3 /. 5 6

FIGURE 3 (a) Experimental and (b) calculated electro-reflectance spectra and absolute reflectance spectra for a Sndoped In On film of 45 nm prepared by d.c. sputtering.(Electrolytic capacitance was 16btFm-2 ).

inherent in a macroscopic theory dealing with a thinlayer of enhanced carrier of atomic dimensions. Theelectro-reflectance of other samples with differentfree carrier effects, which are demonstrated by theposition and shape of the plasma edge, may beconfidently predicted.

2O

5. CONCLUSIONS

An electro-reflectance effect has been demonstratedin thin films of Sn doped In2 O3 which depends forits magnitude on the sharpness of the plasma reflect-ance edge and through that on the properties of thefree carriers, their relaxation time and effective massand hence their mobility. It has been shown that theexperimental electro-reflectance effect can be pre-dicted quantitatively by a theory based on enhanceddensity of free carriers produced in a very thinsurface layer by the applied electric field.

The specific values of electro-reflectance obtainedwill depend on the reflectance edge shown by thefilm and this in its turn will depend on preparationconditions. However, these conditions have not beenspecified in detail as it has been shown that theelectro-reflectance can be predicted over a wide rangeof reflectance edge values.

The conclusions drawn from the work reportedhere allow us to predict that for films of In2 O3which give higher relaxational mobilities, largerchanges of the reflectance induced by an electricfield will be obtained and can be used as a measure ofthe sharpness of the plasma edge as well as offering awavelength tuned optical modulator. Films of Zr-doped In2 O3 have been reported with mobilitiesconsiderably higher (170 cm2 V-1 s-1 )t than thoseused here (10 cm V- s-1 ) and would be expected tohave considerably greater electro-reflectance effects.

That films of In2 O3 have been shown to withstandup to 20 volts applied across a double layer in anelectrolytic cell with no evidence of surface activitysuggests a suitable system for electrochemical investi-gation of absorbed species and the kinetics of fasthomogeneous reactions involving electro-generatedspecies. Internal or external reflection spectroscopycould be used depending upon the transparency ofthe electrolyte solution.

ACKNOWLEDGEMENTS

We would like to thank Thorn Co. Ltd., and Plessey Co. Ltd.,who kindly provided samples of Sn doped in In20a and toSRC who supported this work.

REFERENCES

1. H. Kostlin, Philips Tech. Rev., 34, 242 (1974).2. H. J. J. van Boort and R. Groth, Philips Tech. Rev.. 29,

17 (1968).3. R. Clanger, Appl. Phys., 2, 247 (1973).4. D. B. Fraser and H. D. Cook, J. Electro. Chem. Soc.,

119, 1368 (1972).5. W. N. Hansen, Advances in Electro. Chem. and Electro.

Chem. Eng., 9, 50 (1973).6. R. P. Howson, J. N. Avaritsiotis and T. Fox, Thin Solid

Films, 30, 297 (1975).7. J. N. Avaritsiotis and R. P. Howson. To be published.8. H. Kostlin, R. Just and W. Lems, Phys. Star. Sol., 29,

No. 1, 89 (1975).9. E. A. Stern and R. A. Ferrell, Phys. Rev., 120, 130

(1960).10. J. N. Avaritsiotis and R. P. Howson. To be published.11. R. Groth, Phys. Star. Sol., 14, 69 (1966).

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2010

RoboticsJournal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation http://www.hindawi.com

Journal ofEngineeringVolume 2014

Submit your manuscripts athttp://www.hindawi.com

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation http://www.hindawi.com

Volume 2014

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

SensorsJournal of


Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks


MODULATION EFFECTS IN INDIUM FILMS

Documents

Transcript of MODULATION EFFECTS IN INDIUM FILMS