^

TMl VltNl AND COHCLUtlON» CONTAINED IN tH(S OOCUMINT ARE THOK Of THt AUTHORI AND SHOULT

INO TMK omciAU POLICIII. DrHKR CXPRCtlKC; on IhlPLltD, Or T«| ADVANCiD nlSlARLH MOJICTI AOINCV OR THE U t. OOVCHNMRNr.

W OPTICAL CONVERSION PROCESSES

o •^ D. W. Trainor and S. A. Mani f~-^ Avco Everett Research Laboratory Inc.

FOREWORD

Contract No. : N00014-76-C-1162

DARPA Order No. : 1806 Amendment No. 36

Program Code No. : TE20

Short Title of Work: Optical Conversion Processes

Contractor; Avco Everett Research Laboratory, Inc. Everett, Massachusetts 02149

Principal Invistigator: Daniel W. Trainor, (617) 389-3000, Ext. 467

Scientific Officer: Director, Physics Program, Physical Sciences Divisioi Office of Naval Research 800 North Quincy Street Arlington, Virginia 22217

Effective Date of Contract: 15 September 1976

Contract Expiration Date: 15 November 1977

Amount of Contract: $233, 696

UNCLASSIFIED SECURITY CLASSIFICATION UF THIS PAGE (Whin DmtB Enlertd)

.IJ,THO^ r

REPORT DOCUMENTATION PAGE I REPORT NUMBER 2, GOVT ACCESSION NO. 3 RE CIPI ENT'S C AT ALOC NUMBE R

4. TITLE (mnd Sublltle)

Optical Conversion Processes • Semi-Annual Repwt' 15 MarÄ-^TT,- 15 Sep/ 1*77,

'. A1! D. W. /Trainor aad S. A. Mani i H

9. PERFORMING ORGANIZATION NAME AND ADDRESS

Avco Everett Research Laboratory, Inc. 2385 Revere Beach Parkway Everett, Massachusetts 02149

READ INSTRUCTIONS BEFORE COMPLETING FORM

I Wl UF HEPUR f » RfRVSD COVERED

RG. REPORT NUK Hl R

TgTbB'gwiNrnuMaeff

N00O14-76-C-1162

fit%%Etfmmmi eJr-JiZm

;CT TASK

I ' CONTROLLING OFFICE NAME AND ADDRESS

Defense Advanced Research Projects Agency DARPA Order No. 1806

14 MONITORING AGENCY NAME » ADDRESSfff imii iium ifuiimiiiaa riffifii $4* Office of Naval Research CflJ J ■*? CT Department of the Navy v---¥g¥&0 (•»■ren Data EnlntJ^t

r/P-

UNCLASSIFIED SECURITY CLASSIFICATION OP THIS PAOEfHTiwi Data Knl»t»d)

(20) issues involved. These include: the production of receptor candidates in the gas phase (typically refractory metals), the volumetric removal of the lower laser level in the stimulated Raman approach to prevent "bottle necking" and allow recycling of the atoms during the laser pulse, and thi consideration of overall system efficiency and scalability to Mgh power.

UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PAGEfWian Oa(a P.nltrtd)

;

TABLE OF CONTENTS

Section

I.

II.

III.

IV.

List of Illustrations

INTRODUCTION

THEORY A. Introduction

B. Stimulated Raman Cross Section

C. Phase Matching for the Four-Wave Process

D. The Third-Order Nonlinear Susceptibility

E. Conversion Efficiency

EXPERIMENTS

A. Introduction

B. Acceptor Candidate Production

C. Pump Laser Sources

D. Lasing Experiments

1. Atomic Iron

2. Iron Pentacarbonyl, Fe(Cü)5

SUMMARY

REFERENCES ,V-i'e Section

BJf Section n

•n D

V J; S

RV

IRlBOT/Ml'UBi'in TO — , ' ' ,- CIAL

/ri

Page

3

5

13

13

15

19

24

26

37

37

37

40

42

42

57

63

65

j

: s I

LIST OF ILLUSTRATIONS

Figure Page

1 Demonstrated Potentially Scalable Electronic ° Transition Lasers

2 Propagation of Converted KrF Photons 8

3 Schematic of the Parametric Down Conversion 14 Process

4 Energy Levels in Thallium 16

5 Schematic of Phase Matching 22

6 Phase-Matching Angle vs Output Wavenumber in 23 Thallium

7 Conversion Efficiency vs 3, Vi = v^ = 0 29 Conversion Efficiency vs (3, Yi = Y4 = ®

3J 8 Conversion Efficiency vs ß, yi = 0. 74" ^

9 Conversion Efficiency vs p, y. = 0. 2, y. = I 32

10 Max Conversion Efficiency vs Y4 ^

11 Map of A vs N for Certain Candidate Atoms 35 Considered for the Parametric Down Conversion

12 Anticipated Stimulated Raman Gain of Various 39 Accepted Candidates

13 Doubled Dye Laser System 41

14 Atomic Iron Laser 44

15 Optically Pumped Atomic Iron Laser 45

16 Atomic Iron Lasing Transitions 46

17 Spectral Distribution as a f (Atomic Iron Density) 47

18 Experimental Approach 49

19 Experimental Data 50

Figure Page

20 Discharge Cell Characteristics 51

21 Converted Laser Output vs Atomic Iron Density 52

22 Flash vs Discharge: Iron Laser Output 54

23 Production and Recycling of Metal Atoms 55

24 Quenching Gas Effects 56

25 Optically Pumped Iron Pentacarbonyl Lasing 58 Transitions

26 Experimental Data 60

27 Variation of Output with Fe(CO)5 Density 61

I. INTRODUCTION

The overall goal of this combined experimental and theoretical pro-

gram is to successfully and efficiently convert using scalable techniques

the output of a high power KrF laser into longer wavelengths so as to vastly

improve its propagation characteristics.

Since the first reported lasing of an inert gas halogen laser, a num-

ber of similar systems have demonstrated lasing characteristics. Operating

at various wavelengths, with different efficiencies, a major class of elec-

tronic transition lasers came into existence. Recently, analogous mercury

halide compounds showing similar formation kinetics have been shown to

läse in the visible, ' albeit: in high temperature (~ 275 C) cells (see

Figure 1). However, the most efficient laser reported to date in this group

is the KrF laser operating at 248 nm. It has also produced the highest

energy outputs reported utilizing e-beam pumping and e-beam controlled

discharge pumping and has a demonstrated capability for being scaled to

high average power. In certain applications, especially those requiring

transmission through the atmosphere, its short wavelength severely limits

its usefulness. This limitation in propagation at short wavelengths arises

due to absorption by atmospheric ozone and to Rayleigh scattering which in-

creases ad X as the wa\

for wavelengths < 3000 K.

-4 creases ad X as the wavelength gets shorter. Ozone absorption is severe

(1) Parks, J.H. , Appl. Phys. Letters 31_, 192(1977).

(2) Parks, J. H. , Appl. Phys. Letters jH. 297 (1977).

u. <

k «> c" ID CM « ^ x ao x H x

ffi u

;^

WAVELENGTH (nm) J u —I—

700 T X T

200 300

—,—

400 500 600 800 900 1000

62539-2

Figure 1 Demonstrated Potentially Scalable Electronic Transition Lasers

v

i

Figure 2 shows vertical transmission from a height of 3 km as a

function of wavelength. Also plotted are quantum e ficiency of conversion

from KrF wavelengths and the total percentage trarsmissioii of converted

KrF radiation. From the figure, it is apparent that to efficiently utilize

KrF laser radiation, its conversion wavelength should be be'ween 340 and

400 nm to maximize its atmospheric transmission with mini nal loss from

quantum yield considerations. Xenon fluoride lasers, while possessing a

more attractive wavelength for propagation, have not yet demonstraled the

combined efi'icier cy and energy density comparable to KrF. Any optical

conversion scheme for altering the wavelength of KrF laser radiation to

the 340 to 400 nm wavelength range could have higher overall efficiency

than the XeF laser if the photon conversion efficiency is > 40%. Such effi-

ciency for convei sion is a reasonable goal for the program we are discussing

here. For fupperting evidence, one can look over the past year at ;i number

of milestones that have been reported re'.evant to the optical conversion of

UV excimer lasers. With regard to ovei all conversion efficiency, an XeF

laser has been c< nverted, at near unit photon conversion efficiency, using

barium vapor. Also KrF conversion to a series of UV-visible linis due

to 6 Stokes and Z anti-Stokes transitions in high pressure molecular hydro-

(4) gen w-as reported showing good overall conversion efficiency. Also, a

number of accept Dr candidates identified as potential converters for KrF

(5) have been produced in the gas phase using scalable techniques. ] T view

of the above, it seem? reasonable and important to pursue scalable techniques

(3) Djet, N. and Burnham, R. , Appl. Phys. Letters 30, 473 (1977).

(4) Loree, T. R. . Sze, R, C. and Barker, D. L. , Apol. Phys. Letters M, 37 (1977).

(5) Trainor, D.W. and Mani, S.A., Optical Conversion Procest es, Contract No. N00014-76-C-1162, Semi-AnnuaI Technical Report. 15 Sept, 1976 to 15 March 1977.

7

ATMOSPHERIC TRANSMISSION > AEROSOL

OOOI .6 .8 I 2

WAVELENGTH (/im)

HANDBOOK OF GEOPHYSICS .... AFCRL (1965) 08803

Figure 2 Propagation of Converted KrF Photons

that could efficiently convert KrF laser output to longer wavelengths. Two

non-linear optical conversion techniques that we have considered to achieve

this goal are stimulated Raman and parametric conversion processes.

Fox- the stimulated Raman process, pheJiomenologically, the acceptor

atom can be thought of as absorbing as incident KrF photon thereby making

a transition to an excited virtual state and then, with the emission of a

Raman photon at longer wavelengths, proceeding to a level near the ground

(initial) state. Through collisions with an efficient quenching gas, it can

return to the initial state for subsequent re-excitation by the KrF laser field,

i. e. , exhibit high efficiency by recycling the metal atoms. The Raman pro-

cess is enhanced when the virtual state is close to a real state.

Another method of "down conversion" to lower energy, longer wave-

length photons applicable to UV laser light is parametric down conversion.

In this process, conversion is achieved by the utilization of the non-linear

properties of the medium (the acceptor atom or molecules). Here an atom

in state 0 upon exposure to KrF laser light of frequency v 1 goes to a virtual

state 1 and re-emits three photons of frequencies v v , and v4 such th; t

v, =v,,+v„+v.. At the end of this process, the atom returns to its i litial 1 2 3 ^ state entirely by optical transitions. Once again, if the various atomic

transitions (v ,, v _, v .,, and v .) in the at 'jeptor are allowed and the dipc Le

moments are large, near resonant effects enhance the overall process ; uch

that efficient down-conversion hould be likely.

At AERL during the current reporting period, theoretical and experi-

mental research have been carried out on potentially efficient scalable

schemes for converting KrF photons to longer wavelengths. By theoretcal

calculations, we have identified a number of promising candidates to co-ivert

the KrF laser radiation to longer wavelengths using parametric processes

(see Table 1). This adds to the stimulated Raman candidates reported in

Ref. 5. Experimentally, we have modified a comn ercial KrF laser

(Tachisto Corp. , Needham, MA) to provide a focused outpul beam of nearly

a GW/cm and used this laser to convert to wavelengths near 300 nm using

atomic iron as the accepter candidate. The iron was produced using

scalable techniques in densities sufficiently high to provide single pasf-

amplified spontaneous emission when exposed to a ield of KrF photons

from an untuned laser source. In addition, we wer 2 able to demonstrate

laser action in the organo-metallic precursor (i.e., the FefCO)-). This

represents a likely laser pumped pL.;todissociation process followed by

inversion of a photofragmi nt produced in the initia1 itep. The results of

these combined theoretical and experimental efforts are summarized in

the following sections of this report.

(6) Trainor, D.W. and Mani, S.A., 30th Annual Gaseous Electronics Conference, Paper #LA-3, Oct. 20, 1977.

10

s

I if

TABLE 1. POTENTIAL CONVERSION CANDIDATES AND THEIR OUTPUT WAVELENGTHS

1 Candidate 1 Conversion Process Output Wavelength j X, nm

Iron Stimulated Raman 300, 304 Calcium 544 j

Palladium 332

Platinum 332 i

| Lead 309

| Hydrogen 277, 313, 360

Thallium Parametric Down Conversion

~380

Lead i it n -364

i Mercur/ ii it ~376

Lithiurr [ ii it ; -680 >

11

THIS

PAGE

IS MISSING

IN ·.

ORIGINAL

DOCUlviBN'l,

*

II. THEORY

A. INTRODUCTION

The stimulated Raman process and the calculations performed for

the ARPA program on optical conversion were discussed in detail in the last

(5) semi-annual report. x ' In this report, calculations performed on the four-

wave parametric down conversion will be discusset, In the parametric down

conversion process, an atom in state 0 interacting with the KrF laser light

of frequency v, goes to a virtual state 1 and remits three photons of fre-

quencies v?, v^ and v, suchthat v, - v? + v, + v. (Figure 3). At the

end of this process the atom returns to the initial state entirely by optical

transitions. None of the electromagnetic energy in the parametric process

is absorbed by an ideal medium in contradistinction to the stimulated Raman

Stokes process in which the atom is left in an excited state at the end of the

nonlinear process. Thus the problem of heat removal for a large scale con-

verter is less severe in the parametric approach.

An interesting candidate atom for the parametric down conversion of

KrF laser radiation is thallium which will be used as a concrete example in

the following discussion. In the down-conversion process, the photon at fre-

quency v7 is initially produced by spontaneous and stimulated Raman scat-

tering. Since there are usually several channels available for stimulated

Raman scattering, one should first calculate the Raman cross section for the

different possible final states of the atom and choose the one that is most

probable for determining v~. The frequencies v., and v4 are then chosen

13

ISBGEBUia FACBE g^ mjaD

i

PUMP

AE3

r^ \\vA

SIGNAL

^4

DIFFERENCE MIXING V^ 1/,- V^V^

X

G248C

(3) N ^01 ^"12 ^23 ^30

(AE,) (AE2) (AE3)

Figure 3 Schematic of the Parametric Down Conversion Process

14

(3) to maximize the nonlinear susceptibility, x • subject to phase matching

condition. In Section II,B, the stimulated Raman cross section is calculated

for the three possible final states of a thallium atom irradiated with a KrF

laser. J.n Section II. C, the dispersion of the four wives is calculated and (3)

phase matching is discussed. The nonlinear suscej tibility, x i ^ov t^16

fourwave process is calculated in Section II. D, while in Section II, E, the

coupled set of nonlinear equations for the field amplitudes is solved and the

resulting conversion efficiency is calculated.

B. STIMULATED RAMAN CROSS SECTION

The energy levels of thallium atom are shov n in Figure 4. Sponta-

neous or stimulated Raman scattering of KrF laser radiation of energy

-1 2 ~ 40225 cm by an atom in the ground state can only occur to the 6p PQ/T .

2 2 7p P1 /_ or 7p P-w? levels. Transitions to other states are forbidden by

parity selection rules. In this section, we shall calculate the stimulated

Raman cross section for these three possible final states. The calculation

of nonlinear susceptibilities is usually a tedious exercise. The susceptibility

formulae contain products of matrix elements of the dipole operator Q = e r ,

of the form < y J M | Q | y1 J' M1 > which connect the initial and final atomic

(7) states through a chain of intermediate (virtual) states. Yuratich and Hanna

have used Racah algebra to sum over the intermediate state M values and

write the nth order nonlinear susceptibility in a form that can be factorized

into two oarts, in one of which the relation between the fields is explicitly

displayed; the other contains the physics of the atom in the form of reduced

matrix elements (which can be related to oscillator strengths) and 6 j symbols

In what follows, we shall use the symbols of Ref. < as much as possible.

(7) Yuratich, M, A, and Hanna, D, C, , J. Phys, B: Atom. Molecule Phys. 9, 729 (1976),

15

50 r

40 ■8s

•.8p 7d

S o

w 30 i O

>

ft m 20

10

0 H276I

•7»

■6d

•7P P1/2, 3/2

6p 2P 3/2

6pZPl/2.

Figure 4 Energy Levels in Thallium

16

The Raman susceptibility x per atom is given by

( Paa Pec ) 12 J + 1 " 2 Jj + 1)

ca o

£ © (K)

EiJl K J

T i 1 J, 1

T 1^ T '

J < Yj Ji IIQII v2 J2 > < v2 J2 IIQII YJ >

V2 J2 W "'l ' *

1- jl I! (1) V2 J2 YJ P Y2 J2

In the above, n and n represent the initial and final state fractional ' Haa ^cc ^

populations, 2J + 1 and ZJj + 1 their corresponding degeneracies whereas

Y, J, represents the intermediate states; S2 T T=(E T-E T)/4i, id y2 J2 yj y2J2 yj

O) and U) are the pump and Raman frequencies respectively; < Y J II Ü || y' J' > p s

is the reduced matrix element and F is the damping term between levels

c and a. The factor 6 contains all the information about the angular

dependence between the pump and Raman fields. Even though the summation

in Eq. (1) extends over all the intermediate states, there are only three dom-

inant terms in the case of thallium: those in which the intermediate states

2 2 2 are 8s S. /_, 6d D, /_ and 7d D-s/?. The value of the reduced matrix ele-

ments is obtained through their relationship to the oscillator strength:

2 , j, 2 (2J + 1) f T , T, < YJ IIQII Y' J' > = ^ Ö YJ - ^ J (2) cm "y' J' yJ

The oscillator strengths for p to s and p to d transitions have been calcu-

lated by Anderson et al, and have also been experimentally determined by

(8) Anderson, E.M., Anderson, E.K., and Trusov, V.F., Opt. Spectr, (U.S.S.R. ) 20, 471 (1966).

17

Gallagher and Lurio.' ' The signs of the matrix elements are determined

from the work of Vriens^ who used a combination of experimental and

theoretical results to deduce these for indium and thallium. The angular

dependence factor 9 for the Raman process has been calculated by

Yuratich and Hanna. ^ ' For our purposes, we shall assume that both the

pump and Raman fields are linearly polarized and for this case, we have

e(0) - j cos2 ß (3a)

G{1) = I sin2 p (3b)

e(2) = "£ (3 + cos2 p) (3c)

where ß is the angle between the pump and Raman fields.

The stimulated Raman emission cross section CTgojr i-6«. related

to the Raman susceptibility x by

-127rie 0 1r.7 o ;8 TT x 10 T . /AX aSRE = T ( c IP

) Xr (4)

where I is pump intensity in watts/cm and X is the wavelength of the

Raman radiation. The calculated CTgRE at the three Raman wavelengths

3083 Ä, 1. 649 fj. and 1. 975 |u, (when the atom goes respectively to 6p P3/2»

2 2 7p PWT and 7p Po/? states) are respectively

^SRE (6P S/Z) -1-72X 10*28^-' ^ (5a) ca

^SRE^P2^ =3-40x 10"27^-' ^ (5b) ca

aSRE (7P ^/z) = i-38 x 10"26 -^-> cm2 (5c)

ca

JT) Gallagher, A., and Lurio, A., Phys. Rev. 13b A. 87 (1964).

(10) Vriens, L., Opt. Commun. U, 396(1974).

18

where f is in cm" . The line width is taken to be the same in all three ca

cases since the dominant width is the pumping laser width (~ 100 cm" ),

natural and doppler widths being much less than 1cm . The largest stimu- 2 2

lated emission cross-section is for the 6p Pi/2 ■* ^P P3/2 transition and

thus v2 is expected to be ~ 5064 cm ,

C. PHASE MATCHING FOR THE FOUR-WAVE PROCESS

In any parametric process where the atomic state of the system re-

mains unchanged during the nonlinear optical convei-sion process, both mo-

mentum and energy of the photons have to be consevved. In the four wave

down conversion, this means that efficient down conversion will be achieved

only if the iT vectors of the three converted waves idd up to the iT vector of

the KrF pump wave. Since we are dealing with near resonant down conversion

it may be possible to phase match either intrinsically or by the addition of

buffer gas in the right proportion. The former method is to be preferred if

it is possible since it reduces the complexity of generating the right phase

matching condition.

In the case of intrinsic phase matching, again two possibilities exist.

In the first case, the frequency v4 is adjusted to produce the correct amount

of dispersion so that all waves propagate collinearly. This may be possible

because of the large dispersion of the eltctromagnetic wave at v. near an

atomic resonance. In the second case, the intrinsic phase matching can be

achieved by having the wave vectors S^, k^, and IT. corresponding to the

frequencies v?, v~ and v4 respectively propagate at slight angles to each

other such that their vectorial sum adds up to the pump wavevector k .

Obviously, this can be done only when the algebraic sum of k.,, k., and k.

is larger than k.. In a lossy medium the wave number k is larger than

19

the free space wave number at frequencies slightl r below a resonant fre-

quency. In the case of KrF laser down conversion in thallium angle phase

matching should therefore be possible for v A chosen to be less than the

2 2 -1 6p Pi/? to 7s S, /? transition energy (26477 cm ).

An important point that is to be considered in phase matching is the

real absorption of waves generated at v* by the medium. The closer ».

(3) is to a resonance, the larger is the absorption; x is also larger near a

resonance. However, the absorption coefficient in the wings of a Lorertzian

line decreases, as (AE)" , while X decreases as (AE) . Hence, v ,

should be chosen to minimize absorption while keeping x reasonably high.

For the case of thallium, collinear intrinsic phase matching seems pos dble

only where v, is less than 1 cm" from the 6p -* 7s resonance transition;

unfortunate y, absorption at this frequency is expected to be very large It

is therefore not a good operating point for the down converte -. Moncollinear

angle phase matching is therefore considered below for thall um.

The refractive index, n(v), at a frequency v is given by the Sellmeier

*. (11) equation

. r N. f.. KC , . e v i ii //v W 2 X.J [v.. - vj

where

r -- 2. 818 x 10"13 cm e

f.. = oscillator strength of the i-j transition

v.. - energy difference between i and j levels in cm

N. - number density of particles in the i level

(II) Miles, R. B., and Harris, E.E., I.E.E.E. - J.Q.E., QE-9. 470(1973).

20

Equation (6) is only valid far from any resonance; i.e., when |v - v..| is

at least 5 or 10 times the width of the line at v... Using f.. values from

(12) Ref. 8, N. from the vapor pressure curve of Nesmoyanov and v-- values

(13) from the tables of Moore, the wave number k at any frequency v( = Uj/Znc)

can be calculated as a function of medium temperature when thallium vapor

is in equilibrium with the liquid phase at that temperature.

For phase-matching the following two equations have to be satisfied:

(k. + k-) cos G + k2 cos (|) = k. (7)

and

(k4 + k ) sin 9 = k sin 4» , (8)

where 9 and (^ are the angles that k, ard k_ mak? with respect to pump

field k, (Figure 5). In the above, it is assumed that k, is parallel to k..

This is not a necessary condition for phase matching, but it gives the smallest

angle G. Also the roles of k-, k, and k, can be arbitrarily interchanged.

The initial generation of photons at v, takes place iue to the Raman process

and v? can come at any angle. The direction 6 can therefore be chosen

preferentially by making a cavity at frequency v? and making the KrF beam

come at an angle to the axis of the cavity. This was the reason for choosing

k1 and k? noncollinear. In Figure 6, (f) is plotted us functions of output wave

number v4 at different vapor temperatures. The precise operating point on

this curve is chosen by maximizing the conversion efficiency and minimizing

the absorption. These will be discussed more fully in the next two sections.

(12) Nesmeyanov, An. N, , Vapor Pressure of the Elements (Academic Press, N.Y. 1963).

(13) Moore, C.E., Atomic Energy Levels, Vol. Ill, (N.B.S. circular 467, Washington, D. C., 1958).

21

68135

Figure 5 Schematic of Phase Matching

22

lOOf i -p-—i—i—i i i -i i -i 1 1—I—r i i i i *i 1—t—i—n ' !

M2I«S

2 4 6 8 100

DETUNING, CM'1 6 8 I000

Figure 6 Phase-Matching Angle vs Output Wavenumber in Thallium

23

D. THE THIRD-ORDER NONLINEAR SUSCEPTIBILITY

Having identified the likely output wavelengths, we are now in a posi-

tion to calculate the third-order nonlinear susceptibility for the four wave

(3) down conversion. The third-order nonlinear susceptibility x (^^ UJ2, o^,

w/i if given byv'

/ ^ 3J + K + J9 1/7

6ft eo Jl.J2'J3 K (

< JllQllJi >< JillQIKz >< J2llQllJ3 *< J3llQllJ >

1 n\ *\ F(K) (9) 'l Jj 1 Ml J2 K '

where

m^ x £2)'K' * .3)'" x ^w

F(K) = ^JJ, + "2 + W3 + "4»

(3) It is now a simple matter to estimate x lor the parametric down

conversion process.

The dominant terms are when vJ = ^Pi/? ^1^1 = ^sl/2 0r ^^'

The absorption cross section in line center is given by

a = nr f/Av (14) o e ' K '

where Av is the width of the line in cm . For the 6p1 /_ -* 7s. /, transition

-3-1 in thallium, the natural width is ~ 2 x 10 cm , the doppler width at 1800K

-2-1 -1 is ~ 2 x 10 cm and collision broadened width at 1 atm is -0.2 cm

Taking the collision broadened width for Av, we find the absorption cross

-13 2 section in line center to be ~ 5. 5 x 10 cm . Since the natural line is

Lorentzian, the absorption cross section at a frequency tuned Av- cm away

from line center can be calculated from o (Av /^AiO . Thus for Av, ~ o c 3 3 -1 -19 2 200 cm , the absorption cross section is ~ 1.4 x 10 cm . The third-

order nonlinear susceptibility at this detuning frequency is about 2. 6 x

-30 10" e. s.u. /atom. It will be shown later that for thallium, this choice of

(3) A^,, produces a reasonable optimum of having large x and tolerable absorp-

tion at the output frequency. In the following section, we solve the coupled

nonlinear equations for the evolution of the field amplitudes taking into account

absorption. The density of parametric medium necessary and its length for

efficient conversion is also calculated.

E. CONVERSION EFFICIENCY

The coupled amplitude equations describing the parametric four-wave

conversion process are

dA

"dT = " PlA2A3A4 " Q1A1 (15)

dA

inr =+ ß2AiA3A4 (16)

26

!-*=i

i

dA.

"dT = +P3A1A2A4 (17»

dA

"dT = +P4A1A2A3 - a4A4 (18^

where A. is the amplitude of the electrit field of the jth wave, a. is the J J

linear absorption coefficient and ß. is the nonlinear coupling constant

(ß. = 3N k. x /4n.). Only waves 1 and 4 are assumed to have absorption

in the medium since they have resonance with the initial state of the atoms.

It is not difficult to include absorption of the other two waves as well. It

is convenient to rewrite the Eqs. (15) - (18) in a non-dimensional form with

the help of the following substitutions:

V: = T^VTT (19a:

; = •£ (i9b)

where A is the initial amplitude of the pump wave and L is the length of

the nonlinear medium. The reduced equations are,

dy

«H 1 — P y2 73 y4 - Yj Yi (20)

dy2 + Py, Y, YA (21) dt, f M '3 M

^3 + P Y, Y, YA (22) dt, ' r 71 72 M

dL = + PV! yzVs " ^4^4 (23;

^4

27

j

where . . /?

12 TT2 fe ^ cX4 VW1 W4/

I10 NL)«3),

Y1 = aj L and Y4 = Q4 L.

In the above Iin is the incident intensity of the purrp wave and N is the

2 density of the nonlinear medium, y. gives the normalized photon density

2 in the jth mode and the conversion efficiency is given by y, since v . is

the output frequency of interest. Equations (2ü) - (23) can be solved exactly

analytically in the absence of losses. The initial conditions are, y = 1,

y2 = Ö, y, = A and y . = 0 at £, = 0. The analytical solution involves

Jacobi's elliptic functions and the expression for the conversion efficiency

T] is given by

T! = A Sn2 (24) A + 1 - sn (u,p)

where

u = p To (1 + A)]1/2

r 6 - A I1/2 P ~15 (1 + A)J

It must be emphasized that the expression (24) is valid only for the

_3 case of no losses. Typical values of A and Ö might be 10 . When 6 = A,

we have p = 0 and the Jacobian elliptic function sn u reduces to sin u. Com-

plete conversion is obtained when sn u equals unity, which means u = ir/Z,

This implies that 3 should be equal to ~ n/Z ^7 fcr complete conversion.

Figure 7 shows a plot of conver.sion efficiency vs P for typical values of ö

and A. When we take into account the absorption of the waves, it is in general

28

50.0

H2760

Figure 7 Conversion Efficiency vs ß, y, - y^

29

not possible to obtain closed form analytical solutions. Equations (20) - (23)

ai'e, however, easily solved in a computer. Figures 8 to 10 show typical re-

sults of such computer runs. It is evident from the figures that absorption

at v. affects the generated output much less than absorption of the pump

wave. This is not surprising since the bulk of v . is generated ovt-r a small

length at the very end of the nonlinear medium while abosrption of v. takes

place over the entire length of the nonlinear medium.

For ö - 10 J, we find that we need P ^ 50 for efficient parametric

conversion. If one of the three generated waves is initially created by the

Raman process, the corresponding resonant denominator should be replaced

by the pump laser linewidth. Thus for a diffraction limited pump laser, the

conversion parameter 0 can be written as

ß = A N P/Ye (25)

where P is the KrF lasei power in watts, y« IS thi laser linewidth in cm

and A is ar effective cross section in cm /w. In the case when Av and

(3) Av, are very small, the expression for x , is dominated by a single term,

and A can be written as

9 ^2 7 [(gf)01 (gf)21 (gf)32 (g

f)033l/2 A - s r x 10 T T .2^2 Av, Av,,

4 TT ft c 13

1/2 |J K J2| v (2K -f l)

1^ 3

3(2J + 1) K U J, 1

J K J

1 J, 1 1 3

X{ÖK, 0 +26K, 2/^} ^

30

1 1 1 iii

Yx = 00 /

IO-' r* » io / j

lO"2

p* - / J

EF

FIC

IEN

CY

,

^ /

-^

CO

NV

ER

SIO

N

ö

5

/ _j

IO"6

/ -H

10-7 liii 1 1 1

H2762

10 20 30 40 50 60

Figure 8 Conversion Efficiency vs ß, y = Q, Y . = 1

31

o :/* U) 10 QC hi > Z o o -5

10

H2764

Figure 9 Conversion Efficiency vs (3, yi^Q-Z, yt-l

32

I

I

7\ =00

y, = 0.2

> o z UJ

Xl = 0.4

H2763

± ± 0 2 0.4 0.6 0.8 10 1.2

>4 1.4

Figure 10 Max Conversion Efficiency vs Y4

33

= 1.91 x lO"13 [(gf)01 (gf)21 (gf)32 (g£)02Y^/Av1 Av3 (2J + 1)

E (6K. 0 + 26K. 2) K > ' '1

J K J.

Jl 1

J K J.

1 J3 1

, cm /W (27)

The quantity A is an atomic parameter for a set of given frequencies. In

Figure 11 a map of constant P/y« is plotted in the coordinates of A and N.

Each atomic system is represented on this map by a horizontal line, whose

position depends upon the density of the medium that may be produced in the

gas phase at reasonable temperatures. The line of constant P/y« have been

plotted on the assumption that ß equals 50. Three elements are plotted on

this map. These are thallium, lithium and lead. From the map it is seen

that both lithium and thallium are good parametric down conversion candi-

dates from the point of view of ease of obtaining the operating density. How-

ever, both require angle phase matching for intrinsic phase matching. Collinei

phase matching conditions may be obtained by the addition of buffer vapor like

mercury.

17 -3 In the case of thallium an operating density of 10 cm over a path

length of 100' cm will give an integrated absorption depth of - 1.4 which is

acceptable for a down converter. This justifies our choice of Av^ ~ 200 cm

(3) to keep x reasonably large while the absorption cross section remains at a

tolerable level.

34

• 19 10

lo r

E o

10 -23 _

10 25

\ \ \ r I —T -

\ s \ \

\ ^■50

[- N. \ -\

1—

1126 C Pb \ 1525 > 1

k 925 Tl\ v 1 X. 1200

— NsBOO^fl-N

\6800 & 3660 Ä

\ \ \ \ \. A

r* v. x-, x \

"1

l \ \ \

\ ^

1 _l 1 i I

\

\ \ i I012

HI 140

10' 10 16 lO1» N, cm'3

ID2©

Figure 11 Map of A vs N for Certain Candidate Atoms Considered for the Parametric Down Conversion

35

J.

THIS

PAGE

IS

MISSING

IN .

ORIGINAL

Do~T T1\ tr •. ,',T'', · Li U lVl~l, -1-

III. EXPERIMENTS

A. INTRODUCTION

The successful demonstration of optical conversion ii any laser

system through stimulated Raman or Parametric conversion requires

optimization of a number of variables to achieve reasonable gain, namely,

overall system cross section, which is in turn influenced by the laser

intensity, bandwidth and wavelength, and also the ability to produce suffi-

cient densities of acceptor candidates to show reasonable coiversior. We

will describe the results and status of this program with regard to the

stimulated Raman candidates in the next two sections and thon d^scrbe how

these systems were coupled to producing lasing in a couple of candidates in

the section titled "Lasiag Experiments. "

B. ACCEPTOR CANDIDATE PRODUCTION

Under the early phases of this contract, the stimulated Ramsn and

direct optical puriping cross sections for various candidates were calcu-

lated. ' For tht stimulated Raman process, the total gain, g^, fc r a

single pass syste-n is equal to Ha L, where N is the density of atDms and

L is the active length of Raman medium. For a diffraction limited team,

the maximum valae of the intensity-length product is equal to P/X v-here P

is the pump laser power and X is the pump wavelength. This, goL can be

written as,

goL = ANP/y^ (28)

37

where

A = 9.55,.0-'4m^^y 7, V e

10 rl7

FLASH PHOTOLYSIS

m* HEAT PIPE

mm ROOM TEMPERATURE. GASES & LIQUIDS

HII29 N, cm'

Figure 12 Anticipated Stimulated Raman Gain of Various Accepted Candidates

39

therefore, to consider these effects using iron as ai example, since exneri-

ments with a glow discharge showed we could produce atomic iron densities

in excess of 8 x 1 0 atoms/cc using iron pentacarbonyi as the precursor.

'^his is equivalent to a thermal temperature near 1600 K. ) We were also

able to generate significant densities of atomic iron by the flash photo-

decomposition of Fe(CO)r in a buffer of argon.

These calculations, which were summarize 1 in our previous semi- (5)

annual report, showed the anticipated gain for direct optical pumping was

about 250 times greater than for the Raman process. We, therefore, had

every expectation that the atomic iron system should efficiently convert

the KrF pump radiation to longer wavelengths near 300 nm.

C. PUMP LASER SOURCES

It was the original premise of this experimental approach to use a

double dye laser as the source to perform conversion experiments. This

system would have the advantage of being spectrally narrow (A X ~ 1 A

near Z48 nm) as well as tunable. Afte. much difficulty, the vendor (Phase-

R Corp. ) was able to provide a tunable, flashlamp pumped dye laser utiliz-

ing Coumerin 504 dye (Exciton Corp, ) and a cooled ADP crystal for fre-

quency doubling (see Figure 13). Output was measured to be near 5 mJ

in a 500 nsec pulse. This laser was found to be useful to perform absorp-

tion experiments of the organo-metallic/buffer gas photodecomposition/

discharge dissf "iation mixtures to determine loss processes at the pump

laser frequency. Referring to Figure 12, it can be seen that its output

(P/y ~10 Wem) is no' sufficient for many of these systams to be useful

as a pump laser. It should be po ible, however, to use it in a MOPA

configuration to be amplified in a discharge laser device to provide som »-

what tunable spectrally narrow, high power pulses.

40

H2 778

Figure 13 Doubled Dye Laser System

41

The conversion experiments were performed using a discharge

initiated KrF excimer laser (Tachisf-.o Corp. , Needham, MA). As delivered,

this laser provided approximately 50 mJ of output in a 20 nsec pulse with

a stated beam divergence near 4 mrad. The output is a rectangle ~15 mm

x 4 mm. To achieve values of P/y as high as possible, it was necessary

to improve the laser beam quality so it could be focussed tightly. This

(4) was accomplished using the technique described by Loree et al of LAKL

and consisted of replacing the supplied output coupling mirror with a 50 cm

plano-convex lens (suprasil). This provided an unstable cavity configur-

ation of quite a good beam quality which focussed near 50 cm from the

Tachisto exit port. Typical output was near 30-40 mJ in 20 nsec which,

therefore, provided approximately 1-2 MW of power. One and two shot

exposures of the output were analyzed to obtain the spectral width and it

was found to be near 10 A (~ 160 cm fwhm). This suggests that for this

pump laser a reasonable value for P/yis 1-2 x 10 Wem. Considerable

improvement can be made by narrowing the bandwi :1th with subsequent

increased ease of conversion demonstration (see F'gure 12),

D. LASING EXPERIMENTS

1, Atomic Iron

Asa check on our overall system, we optet to repeat the hydrogen

(4) conversion experiments of the LASL group usinf a 60 cm stainless steel

shock tube section filled to near 10 atm with hydrogen. We were able to

observe 4 or 5 Stokes shifted lines as well as one anti-Stokes line using a

simple prism spectrograph and polaroid film for observation. Satisfied

with our laser's performance and implied beam qu ility, we then

42

attempted our best candidate in terms of ease of demonstration, i, e, ,

atomic iron. We were able to show on our initial attempts, single pass

amplified spontaneous emission near 300 and 304 nm (see Figure 14) using

discharge dissociation of Fe(CO)r in 50 torr of neon to provide the

required atomic iron density. Subsequent spectra taken with a moderate

resolution spectrograph showed that for high iron densities we were seeing

output on four different lines (see Figure 15). The apparent doublet shape

of the KrF laser pulse is real and is likely due to the presence of a strong

absorber in the discharge cell. By superimposing some atonic mercury

spectra onto these type of data, we were able to correlate our observations

(15) of wavelength with the known transitions in the atomic iron system to

confirm the origin for these lasing lines. These conclusions are sum-

marized in Figure 16. We were able to delay the time the KrF pump laser

fired relative to the discharge and, thereby, probe the spectral output

variation with the decaying atomic iron density. The intensities of the

various lines were then probed through densitometer traces of the photo-

graphic plats. These data are plotted as intet aity peak heigit in arbitrary

units vs time on a semi-logarithmic scale (see Figure 17). From these

data, it is readily seen that the 300 nm transition shows tht greatest output

and persists with decreasing iron density due to its higher total gain in

agreement with theoretical calculations.

In addition to doing spectrally resolved measurements, we per-

formed experiments with calibrated photodiodes to establish temporal

characteristics and measure overall conversion efficiency {tee

15) Corliss, C.H, and Bozman, W. R. NBS Monograph 53 1962).

43

KrF 248nm

— IRON 300nm

\ IRON 304 nm

DISCHARGE

ALONE

KrF LASER

ALONE

LASER AND

DISCHARGE

HI226

Figure 14 Atomic Iron Laser

44

1

KrF PUMP LASER

1

ATOMIC IRON STIMULATED EMISSION

HI4e3 WAVELENGTH (nm xlO"1 )

Figure 15 Optically Pumped Atomic Iron Laser

45

50T

40-

i O

30--

2 o

>- o cc u z LU

20--

10--

0

HI422

Figure 16 Atomic Iron Lasing Transitions

46

9 8

7

CO 6

er * < er t 3 CD er <

I - 0.9

0

HZ 768

300nm'

340 nm 330nm

200 400 600 800

TIME(/iSEC) 1000 1200 1400

Figure 17 Spectral Distribution as a f (Atomic Iron Density)

47

Figure 18). As can be seen from the figure, we COT Id also measure

simultaneously, through atomic absorption techniques using an iron hollow

cathode, the production and decay of the various iron states produced in

the discharge cell. The shot to shot output of the KrF pump laser was also

monitored by an additional photodiode. Typical data are shown in Figure 19.

From the absorption information, it can be seen that the maximum

density of atomic iron in the lowest spin-orbit state ( D J is observed at

times near 300 /isec after the initiation of the discharge. Any direct dis-

charge production must occur on a faster time scale (see Figure 20). This

additional iron atom production beyond the off time of the discharge

probably arises from energy transfer reactions and radical reactions as

well as cascading of higher lying excited states of iron into the lowest

energy state. From the work of Callear and Oldman , it is likely that

the J = 0, 1, 2 and 3 levels of the D state under our experimental condi-

tions quickly (100 /isec) establish a Boltzmann distribution, then as a group

decay into equilibrium with ;he D. state. The characteristic time of this

latter process varies with the nature of the quenching gas present. For

50 torr of neon, its likely to be of the order of several msec, whereas

hydrogen was measured to relax these states with a rate constant of

7. 4 ± 0. 7 x 1 0 " cm /sec. By adding 3 torr of hydrogen to the 50 torr

of neon, we observed increased laser output at earlier times probably at

300 and 304 nm and at the expense of the 320 or 330 nm. Also, data show-

ing the time for peak signal (see Figure 21) coincide closely with the atomic

absorption data shown in Figure 19, supporting the induction time for

optimum iron production.

(16) Callear, A, B. and Oldman, R.I. Trans. Far. Soc. 63, 2888(1967).

48

TACHISTO LASER

50 cm LENS

HOLLOW CATHODE

PHOTODIODE

MONOCHROMATOR

HI937 V

MIRROR AT 248 nm

ETHYL ACETATE

NEUTRAL DENSITY AND NARROW BANDPASS FILTERS

PHOTODIODE

Figure 18 Experimental Approach

49

KrF

PUMP LASER

IRON LASER OUTPUT

PHOTODIODE SIGNALS

10 nsec

LASER FIRES

ATOMIC ABSORPTION

SIGNALS

|^ 200/1 sec

HI936

Figure 19 Experimental Data

50

RESONANT ABSORPTION METAL TO GLASS

SEAL

KrF PUMP LASER

- 30 CM H

50 TORR NEON + 0 1 TORR ORGANO-METALLIC

HI933

CURRENT PULSE

BROAD BAND LIGHT EMISSION

Figure 20 Discharge Cell Characteristics

51

X ~ 300nm

X

HZ 769

0.8 I 12

TIME (mSEC)

j 1.4 1.6 1.8

Figure 21 Converted Laser Output vs Atomic Iron Density

52

In addition to showing conversion of KrF with atomic iron produced

ia a discharge, we were able to demonstrate lasing with iron produced by

the photodecomposition of Fe(CO)(. in a buffer of Ar. In contiast to the

discharge, we saw prompt iron atom production (see Figure 22) and there-

fore chose to fire the laser at times near 40 /isec after the flash initiation.

The flashlamp temporal behavior is such that it exhibits a fwhm of nearly

8 jj.sec and a time to 98% extinction of approximately 40 jisec. Since we

routinely produced more iron atoms in the discharge than in tie flash

12 11 (e. g. , 4x 10 vs 3 x 10 atoms/cc), we observed higher coiversion

efficiency in the discharge metl od. These data suggest that ior the higher

iron densities, we demonstrated overall conversion efficiency of 1-2%. This

is in complete agreement with our expectations. As discussed in our propo-

sal, to produce higher conversion efficiency with these refractory candidates,

one must recycle the atoms during the KrF laser pulse time scale, i. e. ,

return the lower laser ^evel population to the initial state through collisions

with an efficient quenching gas for subsequent re-excitation by the laser

field (see Figure 23), In that the pump laser used in these experiments had

a 20 nsec pulse, it is difficult to demonstrate this eftect. By adding suffi-

cient quenching gas to relax on a time scale fast compared to 20 nsec, the

likelihood of competing unfavorably with the spontaneous emission lifetime

(~ 2 nsec) is assured whereas any scaleup KrF laser system is projected to

have pulse lengths of order 1 jisec and the recycling time requirement is

thereby significantly reduced. We did, however, investigate the effect of

some of the quenctüng candidates on the flash production of iron. These

results are shown in Figure 24. These data are interpreted as evidence thac

the initial production of iron atoms is not significantly affect« d by the

53

10 13

o

(A o

o I- <

in ■z.

O

12

O

10'

H2770

LASER FIRES

h] ~4 x |°12 ^T POWER OBSERVED = I9KW

DISCHARGE

LASER FIRES

POWER OBSERVED = 0.73KW

200 400 600 800

TIML (MSEC) 1000

Figure 22 Flash vs Discharge: Iron Laser Output

54

MECHANISM

ORGANO-METALLIC

+

PHOTONS

\ QUENCHER KrF LASER

M + hv M

KINETICS

LOSS M + M + X— M2 + X T =0.5 msec

RECYCLE M+Q — M + Q T=3 nsec

H277I

Figure 23 Production and Recycling of Metal Atoms

55

S O

if)

O

<

CO z UJ Q

O

O

H2767

10

0

15 TORR CHA

• 50 TORR Na

L

50 TORR Ar -I

50 TORR Hz

200 400 600 800

TIME (fJLSEC)

1000 1200

Figure 24 Quenching Gas Effects

56

presence of the surrounding gases, however, different quenchers e:

>- z o CO

< o < I- z ÜJ a. z o

z g CO CO

s UJ

Q UJ I- < _l

s I- (O

E c

I I-

z

to ß O

d

d tn

o U < 5

T3

a £ d

0^

a. S a: Q. CO

., <

ü

a O

in

d

58

Fe(CO)4 + Fe(CO)5 (-) Fe2(CO)9

We did however, see emission on those lines even on the first fill, but no

special precautions were taken to isolate our storage containers from the

action of the room lights, etc, and therefore there is likely to be Fe2(CO)9

"impurities" in the iron pentacarbonyl.

Since the output showed prompt temporal behavior (see Figure 26),

it is likely to be lasing from an excited state of the FefCO),. or a photo-

fragment produced in the initial photon absorption step, i. e. ,

hv (KrF) + Fe(CO)5 -* FeCOg**

Fe(CO)5** -* Fe(CO)5* + hv

Fe(CO)5** - Fe(CO)4** + CO

Fe(CO)4** - Fe(CO)4* + h^

We did see a trend with increased iron pentacarbonyl density (see Figure

27) and this behavior can be explained by recognizing that thn extinction

-1 7 2(18) coefficient for Fe(COir absorption at 248 nm is 1, 7 x 10 ;m ,

so for 0. 2 torr, the distance for 50% extinction is 2. 7 cm. We observed

maximum output when the KrF pulse was focussed through a gas path

length of approximately 4 cm. Increased converted output would be

expected by reducing this to very short distances bvt severe damage to the

optical window occurred for these cases. Clearly, pumping the iron

carbonyl end-on (see Figure 20) in this matter is nc t optimal. We are,

therefore, planning on using a cylindrical lens to transversely pump the

Fe(CO)c in a dye cell configuration and look for output perpendicular to the

pump laser.

(18) Lundquist, R. T. and Cris, M. J, Org. Chem 2^7, 1167 1962).

59

OPTICALLY PUMPED IRON PENTACARBONYL LASER OUTPUT

KrF PUMP LASER

IRON PENTACARBONYL

LASER

TIME

H2772

Figure Z6 Experimental Data

60

I ■

0 H2?66

02 03

Fe {C0)5 DENSITY (TORR)

05

Figure 27 Variation of Output with Fe(CO)5 Density

61

i

These experiments on photodissociation of encapsulated metals

open a new field of candidates which do not require atom production techni-

ques. If further research indicates efficient conversion can be achieved,

the overall project would be simplified with regard to the engineering

aspects of a practical device. More work in this area is continuing with

emphasis being placed on transverse pumping, con/ersion efficiency,

and new candidates.

62

IV. SUMMARY

With the delivery and modification of an untuned high power KrF

laser, experiments began on optical conversion to longer wavelengths with

various candidates generated using scalable production techniques. For

ease of demonstration of principle, the candidates hydrogen, iron and

calcium were identified as compatible with the available power and band-

width of our pump laser and were therefore first considered. We lased

hydrogen in the manner described in Reference 4 to calibrate our experi-

mental setup and procedure. The next day we observed single pass

amplified spontar eous emission from atomic iron produced by the dis-

charge dissociation of Fe(CO)c-. In addition, we observed lasing in the

organo-metallic precursor (i.e. the iron pentacarbonyl) where the in-

version is from a molecular excited electronic state of the parent or a

fragment produced in the photodissociation process. Experiments to

evaluate this nev, class of conversion candidates and to demonstrate coi-

version in atomic calcium are continuing.

These results demonstrate that efficient conversion of KrF towards

longer wavelengths is achievable using an untuned laser as the pump. Also,

the technique of producing refractory metals in densities sufficiently high

for single pass amplified spontaneous emission by the decorr position o!

organometallic compounds is clearly established. ,_,_____

In addition to the stimulated Raman candidates, the t leoretical

effort has identif ed a number of parametric acceptor candicates for con-

verting KrF to mure propagating wavelengths. An experirm ntal evaluation

of some of these candidates will be undertaken shortly. 63

-~ .

THIS

PAGE

IS MISSING

IN ·.

ORIGINAL

DOCUlviEN'l,

REFERENCES

1. Parks, J.H., Appl. Phys. Letters 3_l_. 192(1977).

2. Parks, H.H., Appl. Phys. Letters 3j_, 297(1977).

3. Djeu, N. and Burnham, R. , Appl. Phys. Letters 30, 473 (1977).

4. Loree, T. R. , Sze, R. C. and Barker, D. L. , Appl. Phys. Letters 3_1_, 37 (1977)

5. Trainor, D. W. and Mani, S.A., Optical Conversion Proces.ies, Contract No. N00014-76-C-1162, Semi-Annual Technical Report, 15 Sept. 1976 to 15 March 1977.

6. Trainor, D.W. and Mani, S.A., 30th Annual Gaseous Electronics Conference, Paper #LA-3, 20 Oct. 1977.

7. Yuratich, M. A. and Hanna, D. C. , J. Phys. B. , Atom Molecular Phys. _9, 729 (1976)

8. Anderson, E. M. , Anderson, E. R. and Trusov, V. F. , Opt. Spectr. (USSR) 20, 471 (1966).

9. Gallagher, A. and Lurio, A., Phys. Rev. 136A, 87 (1964).

10. Vriens, L. , Opt. Commun. U., 396 (1974).

11. Miles, R.B. and Harris, E. E. , I. E. E. E. - J. Q. E. , QE~9, 470 (1973).

12. Nesmeyanov, An. N. , Vapor Pressure of the Elements. (Academic Press, N. Y. 1963).

13. Moore, C. E. , Atomic Energy Levels, Vol. III(N.B. S. circular 467, Washington, D. C. , 1958).

14. Fano. V. and Cooper, J. W. , Rev. Modern Phys. 40, 441 (1968).

15. Corliss, C.H. and Bozman , W. R. , NBS Monograph 53 (1962),

16. Callear, A.B. and Oldman, R.I,, Trans. Far. Soc. jj3f 2888(1967).

17. Eyber, G. , Z. Physik Chem. 144A. 1 (1929).

18. Lundquist, R. T. and Cris, M. J. , Org. Chem. 2J7, 1167(1962).

65

DISTRIBUTION LIST

Office of Naval He search. Department of the Navy, Ariington, VA 2221/ - Attn: PhytT« l'ro^rain (J (.cipiei I

Naval Research Laboratory, Department of the Navy, WaaMngton, 0. C. 2Ü370 - Attn: Technical Library 11 copy!

Oftice of the Director of Delenae, Research and Engineering, information Office Library Branch, The Pentagon, Ukahtngtun, D. C 2U)0I (I copy!

U.S. Army Research Office, Box CM, Duke Statiun, Durham, N. C. 27706(1 copy)

Defense Documentation Center, Cameron Station. Alexandria, VA Z2M4 (12 cnpicn)

Defender Information Analyais Center, Batlelle Memorial Institute, 505 KinR Avenue, Columbus. Oi! 4 5201 (1 copy)

Commanding Officer, Office of Naval Ri-search Branch Office, 536 South Clark Street, Chicago, IL 60615 (I copy!

New York Area Office, Offiv« of Naval Research, 715 Broadway i5tb Floor), New York, NY 10003 • Attm Dr. Irving Rowe |1 copy)

San Francisco Area Oi'fite. Office of Naval Research. 76Ü Msrket Street. Room 447, San Francisco, CA 94102 (1 copy)

Air Force Office of Scientific Research, Department of the Air Fürte, Wnahlngten, D-C. 22209 [1 copy)

Office of Naval Research Branch Office, I 030 East Green St reel, Pasadena, CA 91 I 06 - Attn: Dr, Robert Behiingcr (1 copy)

Code 102 IP (ONRLt. Office of Naval Research, 800 N. Quincy Street, Arlington. VA 22217 (6 copies!

Defense Advanced Research Projects Agency, 1400 Wilson Blvd , Arlington, VA 222UlJ - Attm Strategic TetlmoUigy Office it copy)

Office Director of Defense, Research k Engineering, The Pentagon, Waehingto^, 0. C. 20301 - Atln; Ansisiant Director (Space and Advant.ed Systems) it copy)

Office of the Assistant Secretary of Defense, System Analysis (Strategic Programs), Washington, D C, 203(11 - Attn: Mr (ierald H. McNlehOl« (1 < opj )

U.S. Arms Control and Disarmament Agency, Dept. of State Bldg. , Rm. 4931, Washington, D. C, 20451 -Atln: Dr. Charles Henkln (t copy!

Energy Research Deveiopmenl Agency, Division of Military Applications. Washington, D. C. 20545 (1 copy]

National Aeronautics and Space Administration, Lewis Research C.-nter, Cleveland. OH 441 35 - At»n: Dr. JohnW. Dunning, Jr, (Ico;yl (Aerospace Rpi. Engineer!

National Aeronautics fc Space AdminUtration, Code RR, FOB 10B, 600 Independence Ave, , SW, Washington. D.C. 2054(i 11 copy)

National Aeronautics and Space Administration, Ames Research Center, Moffett Field, CA 94035 - Altn: Dr Kenneth W. Billman (1 copy)

Department of the Army, Oftue of the Chief of RDs. A, Washington, DC 205 t (I - Altn; DASD-DD (1 copy! DA.MA-WSM-T I 1 copy!

Department of the Ar 1 the Deputy Chief of Staff fur Opei-atiuns K Plans, Washington, DC. 20310 -Attn: DAMÜ-RQD ■ (1 copy)

Ballistic Missile Defense Program Office (BMDPOj, 1 he A-ealth Building, 1 300 WlUo» Slvd., ArlingUm. VA 22209 - Attn.- Mr. Albert.I. Bast. Jr. il copy)

U.S. Army Missile Comtnami. Research *. Development Division. Redstone Arsenal. AL 38809 - Altn: Army High Energy Laser Programs (2 copies)

Commander. Rock Island Arsenal, Sock Islano. 11.61201. Attn; SARR1-1.R, Mr. J,\V, McOarvey {I copy!

Commanding Officer, U.S. Army MubiUty Equipment Ri. D Center, Ft Belvoir, VA 22060 - Attn: SMFFM-MW (i copy i

Commander. U.S. Army Armament Command. Roek Island. 11.6120) Altn: A MSA R - R DT ( 1 ^opy)

Director. Ballislic Missile Defense Advanced Technology Center, P O. BiJt 1300, HuntfvilleAL 15HÜ7 -Altn: ATC-OU copy) ACT-T f 1 copy)

Commander. U.S. Ar tny Material Command. Alrx.tndria, VA 22304 - Attm Mr. Paul Che rnoTi iAMCRD-T) (1 copy!

Commanding General, U.S. Army Mumtinns Command, Dover, NH l?80l - Altn; Mr, Gilbert F. Chesnuv IAMSMU-R) [I copy)

Dtrector, U S. Army Ballistics Res Lab, Aberdeen Proving Ground, MD 21005 - Attn; Dr Robert F.ichenhr rger ll copy!

Commandant, n S, Army. Air Defenue School. Ft- Bliss, TX 79916 • Attn: Air Defense Agency (1 copy) ATSA-CTD-MS (1 copy)

Commanding General, U S. Army Combat Dev Command, Ft. Belvoir, VA 22O'?0 - Atln: Director of Sjalenal, Missile Div, (1 copy!

Commander, U.S. Army Training b Doctrine Command, Ft. Monroe, V. . 23651 - Attn: ATCD-CF il copy)

Commander, U.S. Army Frankford Arsenal, Philadelphia. PA 19137 - Atln; Mr M. Eltiicl- SARFA-FC D Bldg. 201-3 (I copy)

Commander. U.S. Army Electronic* Command, Ft, Monmou'h, NJ 07703 - Altn: AMSEL-CT-L, Dr. R G Buaer(lcopy)

Comn^ander, US Army Combined Arms Combat Developments Activity, M. Leavenworth. KS 66027 (1 copy)

National Security Agency, Ft. GöO. G. Meade, MD 20755 - Altn; RC. Foi» A763 (1 copy)

Deputy Commandant for Combat fc 1 raining Developments. U.S. Army Ordnanre Center and School, Aberdeen Provin« Ground, MD 21005 Attn: ATSL-CTD-MS-R (1 copy)

Commanding Oflicer, USACDC CBR Agency, Ft. McClellan, AL 36201 -Attn; CDCCDR-MR (Mr. F. 0, Poerlil copy)

DiSTRIIlUTtON I.IST iCüMimit-dl

D.-paritiu-til of the Navy, Olficp ul tbo Chiff "f Nä\*1 Oppraliunn, The '^nUijon 5CT3'- V a »fi in «inn. DC. ^ü.^l ■ Attn; IÜ»» "MitI- 11 11 tupyi

Uttitt' ol Nasal Rpip»rirh Rrartth Oftuf, 4lt^ Summ-p Slrt>fl, Huiton. MA U22IÜ ■ Atlni Or. K't'.-i QlH'tl*1 11 vopyl

Di-partiupnt or the Navy, DepVily Chief "f Navy Matrfial (Drs . I. WaahirtRtofl, DC Z0i60 - Altit: Mr. H. OaylortHMAI UUMlMcopvl

Nasal Milltit Center, Puint MUJIU. CA ''J04i - Ann: i;«ry Cillifei (Cud- 5S5«! O . .yl

Naval Rfsearth Laboratory. \V »ahinulon, DC 203V' - AUn: iCudp 5»01-EOTK)> 11 iopy| Dr, K Lism^ituti - CoHv ^StU (1 copyl Dr. A. I. Schindler - Code f>000 (1 « \ y\ Dr, H Shenk,. . CuJc ^^04 tl mpyl Mr, D .1. McUughlln - Cudc 5560 (I copy) Dr, John L. Walih - Cueic 5503 (I opy)

Hijjh K ipr^y Laatr Project Oifite. D*-|3-rtmcnt of the Navy, Nasal Sc» Syslem» Ci. Pi«iUl. W»(ihint(ton, De:. JÜJ^O - Attti! Capl A S^ulnlck, USN (Ph. 22* U copyl

Supermlt-ndtnl, Naval PoBl^raduatc School, Mt.nU'rcy. CA 9J940 • AM'i: Librnry (Cotit* ? I 241 C jpy)

Navy Hadiaticn ] echrt-k ^i, Air pore WeApons t.ab (NI.Ui, Kirtland Ar'*, \M 87117 (I vnpy)

Naval Suri«c« VV.apona Center, White Oak. Silver Sprin«, MD 20910 - Attn; llr Leon H Schindel (Code 51ÜI (I copy) Dr. K LCTDV Harrta (Code 11 il (I copyl Mr. K, Knkeiihau» (Code 034) i! copy) Mr. I- \l .j-r iCude 04?) ! I copyl Tpchmcal Library 11 capyl

I .S, Naval Weapon« Cpnter. China Lake, CA ^SSS - At«! Technical Luwary (1 topy!

HQ I'SAF (AF/RDPS), Thp Penianun, Waihinntöii, D,C, 2U.U0 - Atln; Li, Col. A.J Chiotall ropyl

HQAFSC KiUW, Andrews AFh WlihlRgton, DC 20331 -AUn: Mai J M Walton (] copyl

HQ AFSC 10LCAW1. Andrt-ws APR. Washit.flton, D C, 20J3I - AUns Ma(. M. Axelrod il copy!

Air Pore« Weapons Laboratory, Kirtland APH, NM«7ll? - Attn: Lti 11 copyl AL (1 copyl

HQ SAMSO (XRTDl, P.O. Box '127^0. Worldway Postal Center. Lo» Anypl.-». CA "UOU" - AUn; I.t. DorianOeMaio (XRTDM 1 copyl

AF Avionics ' nb (IFüi. Wright Pattenon APH, OH 45433 - Attn; Mr K, Hulthinaon (I copyl

Dppi, of the Air Forve, Air Force MateriaH Lab. (AFSC), Wright Patleracn AFB, OH 45433 - Altn: Maj, Haul Elder (LPSfl I copyl l.as.M- Window r.ruup

HQ Aeronautical Systems 0i\ - . Wrljlht Palter Bun API), OH 45433 - Altm XRF - Mr Clifford Fawcctt ll c py)

Rom« Air Pevelopment Conimarul. GriWli« APH, Rome, NY ' 5440 - Alln; Mr. H. UH« (OCSEl (t copy)

Il'4 Electrotnc» Systems Ptv. (KSt-l, L. C, Hanscom Field, Bedford. MA 01730 - AUn: Mr AUred F. Ahdrfaon XHTl i>pv i

Aerospace Research Lab».. (APi, Wri^ht Patterson AFH. OH 45433 -Attn; Ll. Cot. Max DUVIMIIK il .opy)

Dsfenae Intelligence Agency, Waihington, D. C. 20301 -Attn: Mr. St-ymour Berter iDIIh! (1 «opy;

Central Intelligence Agency. W'ashinnton. DC. 20505 -Attn: Mr Julian C Nail M copyl

Analytic Servtcei, Inc., 5613 Leesbur« Pike, Fall« Church, '-'A 22(Hl - Attn; Or. John Davla (1 copyi

Aerospace Cofp , P.O. Hnjt 92957, Los Angclva, CA 9#S09 - Attn; Dr. G. P. MUlburn fl copy)

Airesparch Manuf, Co,, 4851 - ^45 1 Sep^l veda His d, , LosAnneles, CA 9O00O • Attn: Mr. A. Colin Static lifle (1 copyl

Atlantic Resf-anh Corp,, Shirley Highway at FHsall Road, Alexandria, VA 22314 - Aim. Mr, Robert Naißmith (1 copy)

Avco Everett Research Lab.. 23«5 Revere Beach Parkway, Ever««, MA 02144 - Attn: Dr, Cvorge Sutton 11 copy) Dr. Jack Dau^herty (t copyl

Batielle Columbus Laboratories. 505 King Avenue. Columbus. OH 43201 - Attn: Mr. Fred TtetZel (STP1AC) 'I copy!

Bell Aerospace Co-, Buffalo, NY 14240 - Atln: Dr Wayne C. Solomon (1 copy)

Bm-.r^ Company, P.O. boxJ999. Seattle, W A 'iMI 24 - Attn; Mr. M.I. Gamble (2-. 460, MS KC-S8I (1 Copy)

Electro Optital Systems. 300 N HalsleaH. Pasadena, CA 91107 - Attn; Dr Andrew Jetie. " it copyl

ESL, inc, . 495 Java Drive, Sunnyvale, CA 'HQ»b - Attn: Arthur Einhorn ll copyl

DISTRIBUTION LIST (Conttnuedl

G^i.er»! Electric Co. , Space Dlviiion, P.O. BOK 8^55. PhiUdelphi». PA191UI - Attn; Dr. R. R. Sitllimontl (I copy)

General Electric Co. . 100 Plaitlci Avenue, Plttideld, MA01Z0I «Attn: Mr. D.Q. [UrrinKton (Rm. 1044) (1 copy)

General Rriearch Corp., P.O. Box 3587, Santa Barbar», CA 93105 - Attn: Dr. R. Holbrook (1 copy)

General Research Corp., IbOl Wllion Blvd.. Suit« 700, Arllnntton, VA 22209 - Atln; Dr. Gile» F. Criml (1 copy)

HePculeS, Inc., Industrial Syitem Dept. , Wilmington, DE 19899 ■ Attn; Dr, R. S. Vorlud copy)

Hercu'e«. Inc., P.O. Box 210, Cumberland, MD 21502 - Attn: Dr. R^lph 1BBFO, M/S 9 (t copy)

McDonnell DougUs Research Labs. . Dept. 220. Box 516. St. Louis, MO 63166 - Attn; Dr. D. *'- Arne» (1 copy)

MITRE Corp. . P.O. Box 208, Bedford, MA 01730 - Attn: Mr. A.C. Cron (1 copy)

North American Rockwell Crp. , Autonetics Div. , An»heim1 CA 92fl03 - Attn: Mr. TT. Kumagi. C/- Mail Code HA1B (I cop/)

Northrop Corp. , 3401 West Broadway, Hawthorne, CA 90Z50 - Attn: Dr. Gerard Masserjian, Laser Systems Dept. (1 copy)

Dr. Anthony N Pirri, Physical Sciences, Inc.. 18 Lakeside Office Park, Wakefield. MA 01HH0 (1 copy)

RAND Corp-, 1700 Main Street, Santa Monica, CA 90406 - A'.tn: Dr. C. R. Culp/Mr. CA. Carter (I copy)

Raytheon Co., 28 Seyon Street. Waltham. MA 0Z154 - Attn: Dr. F. A. Horrigan (Res. Div. ) (1 copy)

Raytheon Co., Boston post Road, Sudbury, MA 01776 - Atln: Dr. C. Sonnenschien (Equip, Div. | (1 copy)

Raytheon Co. , Bedford Labs, Missile Systems Div , Bedford. MA 01730 -Attn: Dr. H.A. Mehlhorn (1 copy}

Riverside Research Institute, 80 West End Street, New York. NV 1002i-Attn: Dr. L. H. O1 Neill (1 copy) Dr. John Boie (1 copy) (HPEGL Library) (I copy)

R&D Associates, Inc., P.O. Box 3580, Santa Monica, CA 90431 - Attn: Dr. R. E. LeLevier (1 copy!

Rockwell International Curpnration, Rucketdyne Division, Albuquerque District Office, 3636 Menaul Blvd. . NE, Suite 211, Albuquerque, NM 87)10- Alln: C.K. Kraus, Mgr. (1 copy)

SANDIACorp., P.O. Box 5800, Albuquerque. NM 87115 - Attn: Dr. Al N»rath (1 copy I

Stanford Retearch Institute, Menlo Park, CA 94025 - Attn; Dr. F. T. Smith (I copy)

Science Applications. Inc.. 1911 N. Ft. Meyer Drive, Arlington, VA 22209 - Atln: L. Peckam (I copy)

Science Applicaiions, Inc., P.O. Box 328, Ann Arbor. MI 48103 - Attn; R. E. Meredith (1 copy)

Science Applications, Inc., 6 Preston Court. Bedford, MA Ü1703 - Attn; R. Greenberg (1 copy)

Science Applications, Inc., P.O. Box 235J, La Jolla, CA 92037 - AUn: Dr. John Asmus (J copy)

System«, Science and Software, P.O. Box 1620. La Jolla. CA 9Z037 - Attn: Alan F. Ktetn(lcopy)

SyHtems Consultant«, Inc., 1050 3l8t Street, NW, Washington. D. C- 20007 - Attn; Dr. R.B. Keller (1 copy)

Thiokol Chemical Corp. , WASATCH Division, P.O. Box 524, Brigham City, UT 84302 - Atti; Mr. JE. Hansen (1 copy)

TRW Systems Group, One Space Park. Bldg. R-l, Rm. 1050. Redondo Beach, CA 90278 - Altn: Mr. Norman Campbell (1 copy)

United Techno' aie« Research Center. 400 Main Street, Ea»l Hartford, CT 06108 - Attn: Mr. G. 11. McLafferty (3 copie«)

DISTRIBUTION LIST (Continued)

Uniled Technologie! ae«eareh Center. PriH »nd Whitney Aircraft Div. , Florida Kt> D Center, Wept Palm Beath, FLIMUZAUn: Dr. R. A. Schmidthe (1 copyl Mr. td Pinaley (1 copy)

VARIAN Aiiociatet, Eh.lAC Division, 301 Induitrial Way, San Carloa, CA 94070 - Attn: Mr. Jat'* ■'■•inn O copy)

Vought Systema Divition. LTV Aetoapace Corp.. P.O. Box 5907, Dallaa. TX 75222 - Attn: Mr. F G. Simpaon, MS 254142 (I copy)

Weatinghouae Electric Corp.. Defenae and Space Center, Balt-Waih. International/■.>port - Box 746, Haltimon-, MD 2120J -Attn: Mr. W. F, Liat (I copy)

Weitinghouac Reaearch Laba-. Beulah Road, Churchill Boro, Pittaburgh, PA 15235 - ^ttn: Dr. E. P. Riedel 11 copy)

United Technoiogiei Reaaarch Center, Eaat Hartford, CT 06108 - Attn: A. J. DeMario (1 copy)

Airborne inatrumenta Laboratory. Walt Whitorlan Road, Melville, NY 11746 Attn: F. Paced copy)

General Electric RI(D Center, Schenectady, NY 12305 - Attn: Dr. Donald White {1 copy)

r eveland State Univeriity. Clevelcnd. OH 44115 - Attn; Dean Jack ^. alea (I copy)

EXXON Reaearch and Engineering Co., P.O. Box 8, Linden, NJ 0/036 -Attn: D. CrafsUin i i copy)

University of Maryland. Department of Phyaic« and Aatronomy, College Park, MD 20742 - Attn; D. Currit (1 copy)

Sylvanla El ctric Producta, Inc., 100 Fergeaon Drive. Mountain View. CA 94040 - Attn: L. M. Oaterink (I copy)

North American Rockwell Corp. . Autonettca Diviaton, 3370 Miratoma Avenue, Anaheim. CA 92803 - Attn: t