Skip to main content

Full text of "DTIC ADA079656: Infrared Nonlinear Optics in Semiconductors."

See other formats

^D-*079 656 


1979 P A FOLFF AF0SR-76>BS9R 





7. AUTHORr*; 

Peter A./Wolff 





j Massachusetts Institute of Technology —. 

Research Laboratory of Electronicsy (i/ 1 

77 Mass. Ave., Cambridge, MA 02139 v.11^ 


AFOSH/NE (/1 u 

Bldg. 410 V ^ i i Nii wn roF pages 

Bolling AFB, DC 20332 -12 

U. MONITORING AGENCY NAME & AOOR£SS('7f ditUtent horn Conttoiting OWe9) IS. SECURITY CLASS, (of thfa raporO 




Approved for public release; distribution unlimited. 

8. supplementary notes 

D D C 

KEY WORDS (ConHnuo on rovocMO »ido ft noeootory and idanllfy by block number) 

Infrared, nonlinear optics and semiconductors. 

JAN 21 1980 

IB^TRACT (Continue on rovarao aide It neceeamry and Identify by block number) 

A variety of nonlinear optical processes in semiconductors, have been 
investigated experimentally and theoretically. Among them are: nonlinear 
optical-impurity interactions, soin-induced four-photon mixing, FIR difference 
frequency generation, parametric excitation of plasma waves, nonequilibrium 
carrier dynamics under intense laser excitation, and enhanced four-ohoton 
mixing in the Gunn effect regime. The ultimate aoals of this work are the 
development of tunable IR sources; and the understandina of carrier dynamics ir 
semiconductors. ^ ^ ^ ^ 


iiNri A<;<;TFTFn 

- 2 - 

^OSR.tR. 79- 13 13 

Xn framed J/cn/Xetfr O^fXs m 

I. Research Objectives 

The purpose of this program has been to investigate a variety 
of infrared, nonlinear optical processes in semiconductors. The 
investigations have combined theory and experiment. Generally 
speaking, the processes studied have involved mobile or weakly bound 
electrons in these materials. Among them are: 

A. Nonlinear optical-impurity interactions in Ge and Si. 

B. Nonequilibrium carrier dynamics in n-Ge under intense laser 

C. Resonant, spin-induced,, four-photon mixing in (Cd,Hg)Te. 

D. Resonant, spin-induced FIR difference frequency generation 
in InSb and (Cd,Hg)Te. 

E. Optical properties of n-InSb samples with layered doping. 

F. Stimulated excitation of collective modes in n-InSb, n-(Cd,Hg)Te 
and other narrow gap materials. 

G. Studies via four-photon mixing of carrier dynamics in 
the Gunn effect regime In GaAs. 


AeoMalonTor ^ 




■ -J 

By . 


Availability Codes 






D D C 

rPTOrpm np[ 

JAN SI !980 



^0 1 XQ 

02 ( 



II. Status of Research Effort 


During the first year of this grant a laboratory containing 
equipment for performing infrared nonlinear optical experiments in 
semiconductors was established. This facility now includes: 

i. two simultaneously Q-switched CO 2 lasers with peak powers 
of 3 KW, 

ii. spectrometers, 

iii. IR detectors, 

iv. magnets, including a 70 Kg superconducting system, 

V. fast electronics, 

vi. optical tables, miscellaneous optical components, etc. 

The total value of the equipment is about $120K, of which $60K was 
provided by AFOSR, $21K by MIT. The rest was borrowed, scrounged, etc. 

The laboratory has been very successful. It is now used almost 
daily, by students, staff, and faculty, to perform IR nonlinear optical 
experiments in semiconductors. 

Research Projects 

A. Nonlinear Optical-Impurity Interactions in Ge and Si 
The first successful experiment in the IR laboratory was the 
observation of resonant, impurity-induced, four-photon mixing in n-Ge. 
The Ge crystal was pumped with two CO 2 laser beams at frequencies 
( 1 )^ and u) 2 . Power at frequency * 2wi - ug was monitored as a 

AIR r. , : v ■ ' , ■ . 

NOT •• 


a; . : • . 

D i 11' i ; . ^ , . 

Technical Irnornatioa ClTlcer 


function of the difference frequency, Au = - 0 ) 2 - Sharp resonances 

were observed when equalled the valley-orbit splitting of the donors. 
This effect has been observed in 6e:P, Ge:As, and Si:P. Until recently, 
tuning was achieved by using different pairs of CO 2 laser lines - a 
clumsy, inaccurate method. Lately, it has been demonstrated that the 
resonances can be observed with magnetic field tuning of the impurity 
energy levels. This technique will permit accurate measurements of 
energy level position vs. field. Theoretical calculations of the energy 
level structure are nearing completion; detailed experiments are in 
progress. This work will provide information concerning: 

i. diamagnetic compression of donor wave functions, 

ii. level anti-crossings due to off diagonal matrix elements 
of the Zeeman interaction, 

iii. donor-donor interactions, 

iv. excitation transport in donor systems. 

The theory of impurity-induced four-photon mixing has been worked 
out in detail, and is in good agreement with experiment. These studies 
determine impurity Raman cross-sections. 

B. Nonequilibrium Carrier Dynamics 

The four-photon mixing experiments (described above) imply large 
Impurity Raman cross-sections. This result led, in turn, to an 
attempt to demonstrate stimulated, impurity Raman scattering in n-6e. 
The experiment failed because the intense laser beams required 

stripped electrons from the donors. The carrier excitation process 


has since been studied via measurements of optical absorption vs. laser 

power. In cold n-6e crystals, the optical absorption decreases 

markedly (by more than a factor 10) in the power range 1-10 MM/cm . 

A kinetic model, which balances the photoionization rate from IS levels 

against the rate of carrier recombination with donors, gives an excellent 
fit to the data. These measurements determine the photoionization 
cross section, the donor recombination cross section, and the 
thermalization rate of hot carriers in n-Ge. 

C. Resonant, Spin-Induced, Four Photon Mixing 

Extensive studies of resonant, spin-induced, four photon mixing 
have been performed in (Cd,Hg)Te samples provided by Honeywell. In 
general, one observes several distinct spin resonances - a result 
indicating that the crystals are inhomogeneous. Such experiments are 
now being used by the Honeywell group as an analytic tool to determine 
g-valley and alloy composition as a function of position in (Cd,Hg)Te. 

The MIT work has focussed on homogeneous portions of the (Cd,Hg)Te 
crystals, which show exceedingly sharp (Q ^ 500) spin resonances. At 
higher laser powers (in the 500 W/cm range) the resonances broaden 
and saturate. The values of the spin relaxation times, T^ and T 2 , 
Inferred from these measurements are comparable to those found in 

A student supported by this grant, Asif Khan, spent the summer 
of 1978 at the Honeywell Laboratory in Minneapolis making further 
studies of IR nonlinearities in (Cd,Hg)Te<' Among the new processes 

he observed were spin-induced six-photon mixing and polarization-forbidden 
fbur-photon mixing. Khan has since become a staff member at Honeywell. 

He interacts regularly with the MIT IR group and also furnishes (Cd,Hg)Te 

D. Resonant, Spin-Induced FIR Generation 

The aim of this project is to generate tunable FIR via difference 
frequency generation. n-InSb samples, with spatially modulated doping, 
are the ideal mixing medium for such experiments. Until such crystals 
become available, it is planned to test the feasibility of the technique 
with difference frequency generation in bulk (Cd,Hg)Te. This material 
is preferable to InSb because its bandgap can be matched to COg lasers, 
whereas 5 p pumping is required In InSb. High field linewidth and 
absorption measurements are now being performed in (Cd,Hg)Te samples 
to determine the optimal configuration for the FIR experiment. 

E. Optical Properties of n-InSb Samples with Modulated Doping 
Optical reflectivity has been used to characterize n-InSb samples, 
with modulated doping, grown by Prof. Witt's group. These crystals 
are far too heavily doped (n = 10 -/cc) for FIR mixing experiments. 

The measurements indicate, however, that the modulation depth can 
be large (2 50%); hence that samples grown by Prof. Witt's technique 
may ultimately be suitable for phase-matched, FIR difference 
frequency generation. 

F. Excitation of Collective Modes in Narrow Gap Semiconductors 
Two mechanisms for exciting plasma waves in semiconductors have 
been studied theoretically. The first, involving nonlinear excitation 
by two laser beams, has the advantage of exciting plasmons of well 
defined‘wave vector. Their presence can be detected by radiation they 
emit at the plasma frequency. An experiment to test this idea in 
crystals of n-InSb is now being set up. 

The second plasmon excitation mechanism under consideration 
requires optical pumping of semiconductor crystals [such as (Pb,Sn)Te 
or (Cd,Hg)Te‘l whose bandgap can be made equal to their collective mode 
energy. When this condition is satisfied; minority carriers recombine 
via plasma wave emission. The process is exceedingly efficient, and 
reduces carrier lifetimes to the picosecond range. The reduction in 
lifetime is an interesting effect with several potential applications. 
However, the plasmon emission process could be even more important 
if means were found to channel the energy into a small subset o" tlie’ ' 
plasma modes. Under such conditions, stimulated plasma wave emission 
becomes a realistic possibility. Calculations are in progress to test 
several schemes, including finite geometry and the application of 
magnetic field, to control the plasmon spectrum, and thus permit 
selective excitation of long wave length plasmons. 

6. Studies of the Gunn Effect via Four-Photon Mixing 
Calculations imply that four-photon mixing by electrons in 
n-GaAs is greatly enhanced in the Gunn effect regime. A preliminary 

experiment by Prof. Muehlner has confirmed this suggestion. 

Work on this problem has been temporally discontinued due to 
Muehlner's departure. However, these studies will be resumed in the 
near future. 

III. Publications 

"Theory of Spin-Flip Line Shape in CdS," P.A. Wolff, O.G. Ramos and 

S- Yuen, Proceedings of 1st Soviet-American Symposium on the Theory 
of^Light Scattering in Solids (Publishing House "Nauka", Moscow, 

"Spin-Flip Nonlinearities in Semiconductors," P.A. Wolff, in Laser 
Photochemistry Tunable Lasers and Other Topics (Addison-Wesley, 

Reading, Mass. 1976). 

"Spin Dynamics and Four-Photon Mixing in InSb," V.T. Nguyen, E.6. Burkhardt, 
and P.A. Wolff, Optics Comms. 2i, 145 (1976). 

"Coherent Effects in Spin-Flip Light Scattering," P.A. Wolff, R.N. Nucho, 
and R.L. Aggarwal, IEEE Jour, Quant. Elects. QE-12 , 500 (1976). 

"Dispersion of the Nonlinear Optical Susceptibility of n-Type Germanium," 
R.A, Wood, M.A. Khan, P.A. Wolff and R.L. Aggarwal, Optics Conms. 
il, 154 (1977). 

"Free-Electron Nonlinear Optical Processes in Semiconductors," P.A. Wolff, 
Proceedings of the 16th Scottish Universities Summer School , 

(Academic Press, New York, 1977j. 

"A Chemical Bond Approach to the Electric Susceptibility of Semiconductors, 

I, " R.N. Nucho, J.G. Ramos and P.A. Wolff, Phys. Rev. B IT^. 1843 

"A Chemical Bond Approach to the Electric Susceptibility of Semiconductors, 

II, " R.N. Nucho, J.G. Ramos and P.A. Wolff, Phys. Rev. B 21, 4835 

"Resonant Four Wave Mixing in n-Type Silicon," M.A. Khan, D.J. Muehlner, 
and P.A. Wolff, Optics Comms. 30, 206 (1979). 

- 10 - 




IV. Personnel 

P.A. Wolff, Prof. Physics 
O.J. Muehlner, Asst. Prof. Physics 
Andrew Wood, Sponsored Research Staff 
Jose Ramos, Visiting Scientist 
Graduate Students: 

Roger Nucho: Research Assistant 
Lynn Detwiler: Bell Fellow 
Roosevelt People: Bell Fellow 
Asif Khan: Research Assistant 
Kathy Kash: Research Assistant 

V. Degrees Granted 

"Thin Film Plasmons as a Nonlinear Mixing Mechanism for Difference 
Frequency Generation," L. Detwiler, M.S., E.E., February 1976. 

"Chemical Bond Approach to the Electric Susceptibility of Semiconductors, 
R.N. Nucho, Ph.D., Physics, June 1977. 

"Four-Photon Mixing in Semiconductors," M.A. Khan, Ph.D., Physics, 

May 1979. 




VI Interactions 

A. Bell Laboratories 

There is an ongoing interaction with Bell Laboratories 
via a consulting arrangement; Prof. Wolff visits Bell for a 
few days each month. He has regularly discussed nonlinear 
optics problems with Drs. Bridges, Geschwind, Gornik and 
Nguyen there. The interaction with T.J. Bridges has been 
especially fruitful. Bridges has performed four-photon 
mixing experiments, similar to those done at MIT, in {Cd,Hg)Te 
crystals grown by Cominco. 

B. National Magnet Laboratory 

We collaborate with Dr. Roshan Aggarwal of the National 
Magnet Laboratory. He participated, and was a co-author, in 
the four-photon mixing work in Ge, and in our attempts to 
demonstrate impurity-Raman scattering. We consider Dr. Aggarwal 
a member of the IR research group, and meet with him regularly 
(typically, twice a month). In addition. Dr. Aggarwal is 
supervising the experimental work of Roosevelt People. Prof. 
Margaret Weiler (formerly of NML) also meets with the group. 

She is an expert on the properties of (Cd,Hg)Te. 

C. Honeywell 

Dr. Paul Kruse has provided us with a large, excellent 
quality (Cd,Hg)Te crystal for use in nonlinear optic 
experiments. We expect, via Honeywell's AFOSR project, to 
receive a steady supply of such material. Dr. Kruse and 

- 12 - 




his associates have visited our laboratory several times in 
recent years. In addition, as indicated above, Asif Khan 
spent the summer of 1978 at Honeywell and is now a staff 
member there. He will facilitate future contacts with 

0. Prof. Witt, Materials Science, MIT 

Prof. Witt is preparing (with AFOSR support) n-InSb 
crystals with layered doping for our nolinsar optics work. 

We have already received some heavily doped material from 
him. There is continual interaction between his students 
and ours. 

E. Prof. Fielding Brown, Williams College 

Prof. Brown spent the past year on 'abbatical at MIT. 

He assisted us in setting up FIR lasers for use in stimulated 
impurity Raman scattering experiments, and generally advises 
us concerning the problems of FIR laser operation. 

F. Dr. Aram Mooradian, Lincoln Laboratories 

We have occasional contact with Drs. Brueck and Mooradian. 
Dr. Mooradian provided the {CdGe)As 2 crystal for doubling our 
COg lasers. 

G. Prof. Michael Salour, MIT 

Several members of Salour's group regularly attend our 
group meetings. They, too, are interested in nonlinear optical 
effects in semiconductors.