Skip to main content

Full text of "DTIC ADA174690: Expansion of Research on the Tuning and Stimulation of Nuclear Radiation"

See other formats








v ACCiAitMt't Catalog numoer 

4. TlTt« (m4 ******) 

Expansion of Research on the Tuning and 
Stimulation of Nuclear Radiation, 


Final Technical Report 
7/16/85 - 7/15/86 


7 authors 

1. ESSYKTSY AITMant NuMicko 

Carl B. Collins 



University of Texas at Dallas 

P.O.Box 830688 

Richardson, TX 75083-9688 



NR SDRE-102 412 

ii. controlling or rice nan* and aooru* 

Office of Naval Research 

U. reronT oat* 

October 1986 

Physics Division 

Arlinaton. VA 22217-5000 

i). Nuuacn of *»agcs 


U. pOMiyOftlMO AdENCY HAH* • AOORCSSOf ***•*■«# from Co*troUM4 Offlet) 

14. SCCURITV CLASS. (•! thlm r*fx>rt) 




H. CW3TAIAUTIOM STATEMENT (of thlc *opor<) 

Approved for public release; distribution unlimited 

IT. OISTRIEUTtON STATEMENT (•/ the obitroct on!oro4 In Block 7*. If #tfooml Bo^ori) 


K ■ Y WO AO* fCocttlnuo on r# rnrmm mtrfo it rw»<«H mn4 tOontlty by WocA rrnmtbor) 

Extreme Ultraviolet 

Tunable XUV 

Gamma ray laser 

2D ABSTRACT fConffnLt# nr\ rovorro mltfm It n#c###«n r Horttlfy by block tnmtbmr) 

This contract supported a critical phase of an ongoing project having 
the objective of determining the feasibility and means of realizing a gamma- 
ray laser. Previous work had served to establish that a gamma-ray laser is 
feasible if a certain combination of elementary properties can be found in a 
real nuclear material. Traditional technology could not support the evalu¬ 
ation of these laser-like properties of nuclei and new methodologies had to 

Continued on attachment. 




NAVAL postgraduate school 





h J 

20. Abstract 

be developed. Described here are highly significant successes in the 
development and demonstration of prototypes of two new devices needed 
for the screening of candidate laser materials in a realistic time 
frame, a flash x-ray device for pumping test materials and a Nuclear 
Raman Spectrometer to facilitate the search for certain arrangements 
of nuclear energy levels. 


for the period 

16 July 1985 

through 15 July 1906 


Office of Naval Research 
Contact N00014-85-K-0478 
Task No. NR SDRE-102 


Principal Investigator: Carl B. Collins 

The University of Texas at Dallas 
Center for Quantum Electronics 
P.0. Box 830688, Richardson TX 75083-0688 

Reproduction in whole, or in part, is permitted for any 
of the United States Government. 

♦This document has been approved for public release and 
distribution is unlimited. 


sale; its 







A Flash X-Ray Source of Intense 

Nanosecond-_fulses Produced at High 
Repetition Rates .9 

Nuclear Raman Spectrometer .15 




The universal concerns of those aspiring to develop a new 
type of high-power pulsed laser are to maximize the storage time 
of the pump energy and to minimize the bandwidth of the laser 
transition. Successes in either direction are generally rewarded 
by lessened demands upon both the peak level of input power and 
total fluence needed in the pump source to achieve threshold in 
the laser medium. By definition, a gamma-ray laser must draw its 
energy from nuclear excitation and so there exists the unique 
opportunity to exploit the Mossbauer effect to insure that the 
minimum bandwidth possible for the laser transition is actually 
achieved. Provided the macroscopic host in which the emitting 
nuclei are diluted is sufficiently rigid and provided tempera¬ 
tures are sufficiently reduced, nuclear transitions at energies 
in the range from 1-100 keV can occur with widths corresponding 
to the natural lifetimes of the levels. Over the past six years, 
our research group has described'*' ^ several viable means 
through which this excitation energy might be coupled at will to 
the radiation fields while maintaining the natural Mossbauer 
width. In such cases the cross section for stimulated emission 
could reach 10 cm , ten times more favorable than the value 
for the stimulation of 1.06 /jm from Nd^ + in YAG. 

At the nuclear level the storage of excitation energies in 

1 2 

the Mossbauer range can approach tera-Joules (10 J) per liter 

for thousands of years. The stimulated release of this energy 

would occur at the rate at which resonant electromagnetic 

radiation passed through the laser medium and could lead to 


output powers as great as 3 x 10 Watts/liter. This is an 


astronomical level of intensity and has not been approached to 
within five orders of magnitude on earth by any means previously. 
The peak power from a one liter device would represent 0.03% of 
the total power output from the sun. At this level a relatively 
small laser rod of 5 mm in diameter and 1 cm long would emit a 
pulse of power comparable in amplitude to the total power 
developed by a 100 megaton nuclear device at the peak of its 
burn. Of course the output from the 100 megaton detonation would 
persist for a much longer duration, but with the laser the same 
total power would be delivered in a narrow output beam. 

Unfortunately, the quest for a gamma ray laser has been one 
of the longest unfruitful efforts in the field of laser science. 
Virtually all of the sustained pioneering work was done by 
Baldwin and Solem’s groups in the US and by Gol’danskii ’ s in 
the USSR and focused upon the single photon, brute force approach 
to pumping. Their work dealt extensively with concepts involving 
the use of a neutron flux for pumping the laser medium, either in 
real-time or as a preparatory step to be followed by a rapid 
separation of isotopes within their natural lifetimes. All 
proposals were concluded to require infeasibly high levels of 
particle fluxes to pump the inversions, exceeding even those 
available from nuclear explosions, and to require neutron 
moderators having virtually infinite thermal capacities. By 1980 
all conceivable variants of the single photon approach had been 
characterized as hopeless. In 1981 this "traditional" approach 
to a gamma ray laser was virtually abandoned with Baldwin’s 
publication of the monumental review 15 of all classical efforts. 

The involvement of our UTD Center for Quantum Electronics 



dates back to 1978 and can be conveniently divided into three 
periods of accomplishments. First was an NSF sponsored period of 
international theoretical effort from 1978 to 1981. It arose 
from a long program of previous activity focused upon fundamental 
interactions of coherent radiation with matter that had been 
conducted within NSF’s International Cooperative Science Program. 
Concerned at first with the problem of the correct gauge and 
basis sets to use in describing multiphoton process, we began to 
consider the impact of the work upon areas other than the usual 
atomic and molecular. As a result, the modernized concept of 
coherent pumping with optical radiation was introduced in a 


sequence of papers concerned with nonlinear processes mediated 
by virtual states of nuclear excitation and included the stimu¬ 
lated anti-Stokes scattering of intense but conventional laser 


radiation. The theoretical treatment served to estimate matrix 

elements for a new class of two-photon Mossbauer transitions 

making possible, in principle, the frequency upconversion of 

optical laser photons to gamma ray energies. 

In 1981 the implications of this theoretical renaissance to 

the prospects for a gamma rav laser based on several variants of 


upconversion were reviewed in an article appearing the following 

year. If strengthened by recent infusions of dressed state 
10 1 

theory, ’ that article still provides the most convenient 
review of the basic concepts and requirements for a viable gamma 
ray laser scheme. 

The second phase of our y-ray laser program at (JTD began in 
1981 with the initiation of a modest effort supported by ONR that 


was aimed at providing some experimental verification of the 
theoretical models. Successful beyond any reasonable expecta¬ 
tion, it produced a small but important group of experiments that 
have been recently reported. 9,11 " 13 These experiments to date 
have confirmed: 

13 That the matrix elements used in the previous research 
phase to obtain the favorable estimates of the threshold 
for laser output were correctly estimated; 

2) That extremely large ferromagnetic enhancements of the 
effective powers applied in the coherent pumping scheme 
can be obtained; 

3) That enhancements may also result from the modulation 
of the polarization of ferroelectric media in which 
coherently pumped nuclei could be diluted. 

The conclusion from these experiments is that the gamma-ray laser 
is definitely feasible if a sufficiently ideal isotope exists 
in reality. This is the single most critical issue to the 
development of a gamma-ray laser--the identity of the most nearly 
ideal candidate for upconversion. 

Despite the many applications of beautiful and involved 
techniques of nuclear spectroscopy, the current data base is 
inadequate in both coverage and resolution either to answer the 
question of whether an acceptable isotope exists or to guide in 
the selection of a possible candidate medium for a gamma-ray 
laser. Two new techniques for the measurement of nuclear 
properties with laser-grade precision resulted from the second 
phase of our research into the feasibility of a gamma-ray laser. 

The contract supporting the achievements being currently 


reported initiated a third phase of the project. The objective 
of this phase was to develop and to demonstrate prototypes of the 
hardware needed to instrument the two new measurement techniques 
that will be essential for the screening of candidate materials 
for a gamma-ray laser. Reported here will be the construction 
and evaluation of a flash x-ray device for pumping test materials 
and a Nuclear Raman Spectrometer to facilitate the search for 
certain necessary arrangements of nuclear levels. 


By involving two distinct steps, the schemes we have proposed 

for pumping a gamma-ray laser avoid the severe relationships 

between storage times and spontaneous powers wasted at threshold 


that were imposed on the single-step processes. Replacement 
power that is required falls within a technically accessible 
range avoiding damage to the laser medium. 

These two-step, upconversion processes for optically pumping 
nuclear reactions can be divided into two categories that cor¬ 
respond to the type of pumping employed: coherent and inco¬ 
herent, as shown in Fig. 1. The critical concept here is that 
either transfers the stored population to a state at the head of 
a cascade leading to the upper laser level. To be effective 
the pumping processes cannot transfer too many quanta of angular 
momenta from the fields and the cascade provides a mechanism for 
further changes that may be necessary to reach the laser levels. 
Then the ultimate viability of these pump schemes will depend 
upon: (1) spectroscopic studies locating a suitable configura¬ 

tion of nuclear energy levels, and C2] "kinetic" studies 



Figure 1 

Schematic diagram showing the energetically excited levels 
of a typical nucleus of interest to the development of a gamma- 
ray laser. Lifetimes of the stored energies in the isomeric level 
produced by the initial capture can range from days to hundreds 
of years. The first phase of the two-step process for the 
stimulated release of the stored energy is shown in the figure by 
the solid arrows. Both correspond to the use of longer 
wavelength radiation to lift a nucleus from the storage level to 
a higher level of excitation that has a much shorter lifetime. 

The arrow marked (1) illustrates the incoherent pumping of the 
storage level through the absorption of an x-ray that is resonant 
with the energy separation between the storage level and the next 
higher level of proper symmetry. The arrow marked Cl’) repre¬ 
sents the alternative process of coherent pumping through the 
non-resonant absorption of a photon from the radiation field in 
order to create a virtual or dressed state of excitation shown by 
the dashed level in the figure. In either case the gamma-ray 
output ultimately results from the upper laser level populated by 
a cascade occurring as a second step, as shown in the figure by 
either of the double arrows, (2) and C2’). 


providing an efficient path of cascading from the intermediate or 
dressed state to the upper laser level. 

Both the coherent and incoherent schemes for pumping a 
gamma-ray laser make stringent demands upon the arrangement of 
nuclear energy levels in the potential laser material. Both also 
depend upon the successful arrangement of an input source of 
radiation either to mix the properties of the storage level with 
those of some other state to release the metastability or to 
simply transfer the populations from the storage level to the 
other state. 

For the examination of the nuclear properties we introduced 

the techniques of Modulated Nuclear Radiation (MNR). This 

methodology is the nuclear analog of the optical double resonance 

studies which yielded much of the laser-grade database upon which 

rest the newer visible and UV lasers. A database of comparable 

quality for nuclear "kinetics" will be required for the screening 

of the candidate materials, and this will require a great amount 

of input radiation into implementations of the MNR process. 

Essential to the success of this technique is the accessibility 

of a source of pulses of x-rays of nanoseconds duration that can 


emit a total of 10 keV/keV of linewidth in a reasonably brief 
working period. Either laser plasmas or large e~beam machines 
can do this in a single shot, each of which requires about an 
hour of laboratory time to prepare; but costs are very high. As 
a result, none of these traditional light sources for the 
subAngstrom region could be used to complete an evaluation of the 
29 most attractive materials before the turn of the century. In 
the following section we report recent successes with a prototype 


flash x-ray device producing 35 mW of average power from 10 nsec 
pulses at energies near 8 keV, a level of performance approaching 
that of a large synchrotron light source. University of Texas 
System attorneys have considered this device to be patentable, and 
appropriate actions have been taken. 

To facilitate the search for the scattering states 


energetically near isomeric levels, we had described a new 
technique of Nuclear Raman Spectroscopy that now promises to 
extend both the resolution and the tuning range available for 
high resolution nuclear spectroscopy of the type needed in the 
search for potentially resonant intermediate states needed for 
applications of the dressing process. 

The physical effect upon which this technique depends is 
that new transition frequencies are generated at the sums and 
differences between the frequency of the gamma transition and 
those of an integral number of photons dressing the nuclear 
states. By varying the frequency of the mixing radiation the 
energy of the sum frequency line can be swept, just as in the 
analogous processes implemented in the optical range and this 
provides the bases for the Nuclear Raman Spectroscopy (NRS) 
technique. It could be used to conduct Mossbauer spectroscopy 
with much higher resolution than previously achieved. This would 
be possible because no mechanical tuning needs to be used. In 
case microwave or infrared photons were used, the tuning range of 
Mossbauer spectroscopy would be dramatically increased. Viabili¬ 
ty of this NRS technique has now been fully demonstrated up to 
radiofrequencies. The most recent results are described in the 


following section. 


A Flash X-Rav Source of Intense Na nosecond Pulses Produced 

at-iiiflfr JIsils-S. inon Rates 

Central to the problem of using our MNR technique to eval¬ 
uate candidate materials for a rray laser is the accessibility 


to a source of pulsed X-rays that can emit at least 10 keV/keV 
in a reasonably brief working period. We mentioned above that 
laser plasmas can do this in a single shot, each of which requires 
about an hour of laboratory time to prepare. 

This section reports one of the achievements^ rea¬ 
lized in our Center for Quantum Electronics at UTD from this 
contract work. We have succeeded in demonstrating a compact 
flash X-ray device producing 35 mW of average power isotropically 
from 6 nsec pulses at energies near 8 keV. At this level we 
produce a photon flux in the K~lines that is only one order of 
magnitude lower than the design objective of the Advanced 
(Synchrotron) Light Source (ALS). In less than a minute our 
device delivers an integrated fluence comparable to that obtained 
in a similar line from a multi-kiloJoule, laser plasma shot. Of 
course, each of the larger devices has its unique advantages, 
more collimated beams from the synchrotrons and richer line 
spectra from the laser plasmas; but unless those capabilities are 
essential to a particular application, the laboratory scale 
system we report here may offer attractive support for many 
experiments in other applications, as well. Since it can be 
scaled to much higher energy K-lines and much larger sizes, it 
represents a proof-of-feasibi1ity of the large scale device, 

hyorogen thyratron 



























Figure 2 

Schematic drawing of the high repetition rate, 
device characterized in this work. 

flash x-ray 


needed for the irradiation of candidate laser materials that must 
be pumped at energies beyond those available from laser plasmas. 

As shown in Fig. 2, this first flash x-ray device consisted 

of three critical subassemblies: 1) a low impedance x-ray tube, 

2) a Blumlein power source and 3) a commutation system capable of 

operation at high repetition rates. The latter two components 

differed little from drivers we had developed for short pulse 


nitrogen ion lasers. For the present application the Blumlein 
was constructed from massive copper plates, potted with epoxy on 
outer surfaces to reduce corona and separated by layered Kapton 
(polyimide) dielectrics of 0.635 mm total thickness. The line 
impedance and transit times were calculated to be of the order of 
1.50 and 5.3 nsec, respectively. Capacitances of the switched 
and storage sides of the line were measured to be 3.5 and 3.2 nF, 
respectively. In operation the middle conductor was charged to a 
positive high voltage which could be varied to 25 KV and commuta¬ 
tion was effected by an EG&G 3202 hydrogen thyratron mounted in a 
grounded grid configuration. The average available input power 
was sufficient to support operation to a 100 Hz repetition rate. 

With emphasis on obtaining a singularly low inductance input 
to electrodes designed to give a filamentary source of radiation, 
the x-ray tube was constructed from cast materials, selected to 
minimize erosion and maximize heat transfer. Copper foil strips 
0.05 mm thick and 10 cm wide were fastened to the electrode mounts 
and then passed through the cast material forming the base of the 
x-ray head before it hardened. After emerging from the base, the 
foils were joined to the outermost copper plates of the Blumlein. 


In this way any transverse constriction of the path of the dis¬ 
charge current between the Blumlem and the electrodes was avoided. 

The anode was simply a rod of copper partially buried in the 
cast material forming the base while the cathode was demountable. 

It consisted of a strip of 0.381 mm thick graphite further ground 
with a blade-like edge 10 cm wide and separated from the anode by 
a variable distance chosen to optimize performance. The electri¬ 
cal length of the strip was selected to give a resistance that 
was comparable but below the line impedance in order to assist in 
damping the ringing of the discharge current at times subsequent 
to the initial pulse which produced the x-rays. 

The discharge space was enclosed by a pressure shell, also 
fabricated from cast materials with an integral window of 0.076 
mm thick Kapton plastic film. The window aperture was covered 
with a graphite plate 0.127 mm thick to eliminate the emission of 
visible and UV light. Even with the cast construction and ready 
access to internal electrode spacings, operating pressures below 
3.0 mTorr were routinely maintained with a small mechanical pump. 

Precise measurements of time-resolved voltages and currents 

were rendered difficult by the extremely low impedance of the 

Blumlein and by the commutation of the thyratron in a grounded 

grid configuration on this particular decade of time scales, 1-10 

nsec. Operation generally produced the switching waveforms ex- 


pected of such Blumleins with the observation that the slower 
commutation of the thyratron prevented a full doubling of the 
applied d.c. voltage. The discharge seemed to develop at a value 
of voltage across the x-ray tube that was roughly 50% larger than 
the original charge voltage of the Blumlein. 


Outputs were detected with a block of iast scintillator 
plastic equivalent to NE114 with a nominal 7.0 nsec decay time. 
The resulting light output was measured with a faster photomulti¬ 
plier with 1.5 nsec resolution and recorded with a Tektronix 
7912AD transient digitizer. Numerical deconvolution was subse¬ 
quently employed to remove these instrumental time constants. 

The resulting data is shown in the inset to Fig. 3 to have a 
temporal width (FWHM) of about 10 nsec. 

Measurements of absolute intensities were made by comparing 
the time integrated fluorescence from the plastic detector when 
illuminated with geometrically attenuated x-ravs from the flash 
source directly with the level of excitation produced by a 
radioactive source of known characteristics. This technique was 
used to determine the dependence of the total energies in the x~ 
ray pulses as functions of the important experimental variables. 
Results are shown in Fig. 3 and suggest a rather surprisingly 
weak dependence of the output on anode-cathode separation over a 
substantial range of the smallest gaps. Moreover, an examination 
of the pulse durations at the different separations showed no 
variation that could be detected within the limits of resolution 
permitted by the deconvolution noise apparent in the inset to 
Fig. 3. 

Linder single pulse conditions the x~ray fluence delivered to 
an external target could be visually examined for uniformity by 
allowing it to fall upon a fluoroscopy screen of the type used 
in radiography. A uniform pattern, sharply delineated by the 
edges of the output aperture could be readily seen to result from 


Figure 3 

Total pulse energies emitted near 1.5 A under single shot 
conditions C2Hz) as functions of the critical experimental 
parameters: charge voltage and electrode spacing. While 

obtaining the variations shown, the other parameter was fixed at 
0.127 mm spacing and 24 kV, respectively. Shown in the inset is 
the time resolved intensity from a typical pulse at 24 kV and 
0.127 mm gap. 


the great majority of discharges. At higher repetition rates the 
x-ray pulse energies were found to increase markedly, in direct 
contrast to the behavior generally observed in other applications 
when such Blumleins were used for the deposition of energy into 
large volumes of gaseous laser media. Figure 4 shows that the 
resulting dependence of the average x-ray power emitted upon 
repetition rate seems to be greater than linear because of this 
enhancement of pulse energies at the higher values. 

Attenuations measured with a combination of K~edge filters 
and known thicknesses of aluminum foil indicated that at least 
25% of the total x-ray energy lay in the Cu K lines, in agree* 


20 21 

ment with previous observations and expectations. ’ If it is 
assumed that these lines have their customary width of around 

i n 

6eV, best values of output energy correspond to about 4.7 x 10 
keV/keV when expressed in terms customary for reporting laser 

plasma yields. Thus, in less than a minute of experimental time 


at 100 Hz repetition, over 2 x 10 keV/keV could be emitted. 

This would exceed the dose available from a shot of a large laser 
plasma system at these X-ray wavelengths corresponding to 8 keV. 
Nuclear Raman Spectrometer 

During this contract period a fully integrated Nuclear Raman 
Spectrometer was designed, constructed and has now begun to 
operate. Built from an Apple computer and a Wavetek frequency 
synthesizer it provides for the conduct of swept frequency 
spectroscopy of nuclear levels without the need to employ any 
mechanical effects for tuning that are required in conventional 
implementations of Mossbauer spectroscopy. The successes of the 



Figure 4 

Average powers emitted as X-rays near 1.5 A as a function of the 
repetition rate for a charge voltage of 20 kV. The linear 
approximation corresponds to the emission of a pulse energy 
independent of repetition rate. 


Figure 5 

Schematic representation of the experimental aggaratus for 
demonstrating the mixing of nuclear states in Fe. 


new NRS technique for nuclear spectroscopy indicate that a much 

higher resolution, by perhaps six orders of magnitude, can be 

achieved through a further upgrade of the apparatus. If the 

range of tunability does extend to the ferromagnetic spin 

resonance (FSR] frequency, then it will be possible to construct 

a swept frequency device capable of continuously tuning over a 

range of 10 linewidths, an enormous improvement in the state- 
of-the-art of nuclear spectroscopy. 

Construction details and operation of -a prototype NRS system 

1 2 

were reported in the literature earlier this year. That 
device is shown schematically in Fig. 5. It was first intended 
to be the nuclear analog of a spectrophotometer in which the 
dispersive element is replaced with a constant acceleration motor 
which moves a narrow line source of gamma radiation in order to 
sweep the wavelength by the resulting Doppler shift. The radio¬ 
frequency power was applied to a coil containing the absorption 
foil. It was pulsed for a duration equal to the time required 
for the full sweep of the range of velocities being examined. 

The output from a proportional counter monitoring the transmitted 
y intensity was gated into a multichannel scalar only during 
those times. 


The absorption spectrum of ground state Fe nuclei has long 
been considered a benchmark for techniques of Mossbauer 
spectroscopy. In pure polycrystal1 me iron foil the magnitude of 
the bulk magnetization is the same at all nuclei but varies 
widely in direction for nuclei in different magnetic domains. As 
a result, the standard 14.4 keV transition to the first excited 
state is split by the hyperfine energies resulting from different 


orientations of the spins of the states with respect to the local 
magnetization vector. Figure 6 shows typical data with the six 
allowed transitions being identified by the quantum numbers for 
spin projection with the order for the Mj being ground state 
followed by excited state. 

Also shown in Fig. 6 are the sidebands to these six basic 
transitions produced by dressing the nuclear states with 
radiofrequency (rfj photons from the applied field. For scale 
the energy of one rf photon is shown by the arrow enabling the 
sidebands in the figure to be readily identified. For example, 
if the line results from either the excited state being dressed 
upward in energy by one photon, or the ground state downward by 
one, the sideband line will appear displaced to lower transition 
energy (to the left) by the length of the arrow. 

The data of Fig. 6 was taken with the dispersion being 
obtained in the conventional manner with Doppler shifts. The 
true Nuclear Raman Spectroscopy (NRS) is based upon the use of 
one of these sum or difference frequency sidebands generated by 
mixing nuclear states as a tunable spectroscopic probe. By 
varying the frequency of the mixing radiation the energy of the 
sum frequency line would be swept, just as in the analogous 
processes implemented in the optical range. While the addition 
of a tunable sideband to a nuclear source is more instinctively 
attractive, the complementary experiment with an absorber proves 

the same principles with less expense. 


An extension of the previously described experiment was 
used to demonstrate the possibility for such Nuclear Raman 




a s i * t 









































0.96- jf t 




















61.7 MHz 



200 400 600 800 



Figure 6 

Typical data obtained for pure Fe at 10.1 W of radio - 1requency 
power at 61.7 MHz. The parent transitions are identified by (M-, 
M,), the projections of the nuclear spins in the initial and fiAal 
states, respectively, upon the axis of local magnetization M . 

The change in transition energy resulting from one additional 
radio-frequency photon is indicated. 


Spectroscopy (NRS) with the first prototype system. Figure 7 
illustrates the concepts involved. The unsplit source energy 
corresponded to the wavelength shown in Fig. 7 by the heavy 
vertical line. At the lower mixing frequency shown in the upper 
data, the particular feature 5+ lies at longer wavelength than 
the source, while at higher frequency it lies at a lower value. 

It can be imagined that with a smooth change of frequencies 
between the two values, a point would be reached at which the 
transmission of y-photons from the source would be suddenly 
reduced. The realization of this supposition is seen in Fig. 8 . 
In addition to the absorption resonance resulting from the 
expected, 5+ line, another smaller feature appeared. It 
corresponded to 3 + +, the second order sideband to transition #3. 
It hints at the greater sensitivity of the NRS technique. 

In any case, the point of the experiment was not to 


characterize the energy levels of Fe but to demonstrate the 
effectiveness of this new NRS technique for nuclear spectroscopy 
in general, based on multiphoton effects. However, with this 
prototype system, frequencies could be changed only manually 
with rather elaborate retuning of the radiofrequency system being 
necessary for the acquisition of each point. Completed during 
the present contracting period was a full integration of 
component subassemblies that allows the frequencies to be 
stepped through a range in response to commands issued from the 
CApple) computer controller via the IEER-488 protocol. Dwell 
time at each frequency could be set by software at run time. 
Counts resulting from 7 -ray photons detected during the period of 

Two spectra t 
first-order s 
lies at w . 




Figure 8 

Gamma transmission intensity plotted as a function of rf photon 
frequency. The larger absorption line corresponds to the first 
order, sum-f r equency s^eband of the |l/2, -1/2 >*• | 3/2 , - 1/2 > 
nuclear transition of Fe, while the smaller line corresponds 
to the second-order, sum-f r equency sideband of the | 1 / 2 , 1 /2 >“* | 3/2 , 
-l/2> transition. 


mixing at a single frequency were stored at addresses identified 
with that frequency. 

Since no mechanical movements or Doppler shifts were 

involved in any way, instrumental resolution was set by the 

stability of the frequency synthesized, +100Hz. However, the 


transform of the lifetime of the 14.4 keV excited state of Fe 
is 2.3 MHz so no detail can be expected on a scale significantly 
smaller than MHz. Figure 9 presents the first NRS spectrum ever 
obtained that is densely packed with observations at contiguous 
frequency intervals. Instrumental resolution of both 60 and 30 
kHz are shown. 

An unexpected richness of structure is shown in Fig. 9. The 
feature to the right of center, magnified in the higher resolu¬ 
tion scan is the superposition of the (-1/2, -3/2) minus one 

photon sideband with the (+1/2, +3/2) plus one photon sideband, 
as expected. These two absorption peaks are not precisely coin¬ 
cident, rather being displaced by a frequency, v. such that hv^ 
is the chemical shift of energies between the radioactive source 
and absorber. The appearance of two peaks is clear in the high- 
resolution version and could be fit to two lineshape functions 
from which the chemical shift could be extracted. 

Some of the other features can be interpreted as expected 
sidebands but some are more difficult to identify. While those 
latter examples reproduce during remeasurement of the spectrum, 
their relative intensities seem quite sensitive to surface 
conditions of the absorber material and may reflect further 
shifts in energy caused by the proximity of a surface. Much more 
work is needed for an unequivocal interpretation. Nevertheless, 




> 7 

such data attest to the extraordinary power and resolution of 
the NRS system. 


The achievements of this contracted work are very 
significant to the critical path for development of a gamma~ray 
laser. A gamma-ray laser is feasible if the needed sequence of 
energy levels can be found in some real nucleus. Theory suffices 
to identify 29 prime candidates among the 1886 distinguishable 
nuclear materials, but further screening to eliminate inutile 
members of this prime set of materials cannot be done with 
conventional technology. 

Described in this report are the successful demonstrations 

of two new devices which can be scaled to a capacity able to 

evaluate the 29 candidates in a realistic time frame. One is 

the flash x~rav device needed to mount the nuclear analog of 

optical double resonance measurements to evaluate pumping 

efficiencies and requirements. The ultimate signal which can be 

made available to provide a basis for judgment of the merits of a 

particular candidate material will be directly proportional to 

the total keV/keV emitted by the x~ray source per hour of 

laboratory time. The table top device reported here has already 


achieved over 10 " keV/keV per hour in early tests. It was 
designed to represent 1/10 scale of a single module of the 
ultimate device which would be used in an array of 16 modules for 
the screening of candidates. Thus it is entirely reasonable to 
project an average spectral power of 2 x 10 21 keV/keV per hour in 
the range of photon energies from a few to 115 keV. 

The other device reported here is the Nuclear Raman 


Spectrometer bringing the combination of extended resolution and 
extended tuning range needed for the identification of nuclear 
states useful in dressing isomeric levels. To dress an isomeric 
state requires a certain arrangement of nuclear levels that would 
make them indetectable to conventional techniques of nuclear 
spectroscopy. Our method of NRS is the only means found to date 
that can be used to search for this combination among the 29 best 
candidates. The successes of the new NRS apparatus for nuclear 
spectroscopy indicate that a much higher resolution, by perhaps 
six orders of magnitude, can be achieved through a reasonable 
upgrade of the apparatus. If the range of tunability does extend 
to the ferromagnetic spin resonance (FSR) frequency, then it will 

be possible to construct a swept frequency device capable of 


continuously tuning over a range of 10 linewidths, an enormous 
improvement in the state-of-the-art of nuclear spectroscopy. 



C. B. Collins, S. Olariu, M. Petrascu, and I. Popescu, 

Phys. Rev. Lett. 4_2, 1397 (1979). 

C. B. Collins, S. Olariu, M. Petrascu, and I. Popescu, 

Phys. Rev. C 20, 1942 (1979). 

S. Olariu, I. Popescu, and C. B. Collins, Phys. Rev. C 23.. 

50 (1981). 

S. Olariu, I. Popescu, and C. B. Collins, Phys. Rev. C 23. 
1007 (1981). 

C. B. Collins in Proreedinas of the Int,tj^Lnal_£ gn i er en ce 
on Lasers '80 . edited by C. B. Collins (STS Press, McLean, 

VA, 1981) p. 524. 

C. B. Collins in Laser Techniques t or _ Ext rdine ,.Ul tx avi ole t 
Rnectroscopy . edited by T. J. Mcllrath and R. R. Freeman 
(AIP Conference Proceedings No. 90, New York, 1982) p. 454. 

C B. Collins in Proceedings o f the International— Conteren ce 
on Lasers ’8 1 . edited by C. B. Collins (STS Press, McLean, 
VA, 1982) p. 291. 




Col 1ins, 



Lee, D. M. Shemwell, B. 

D. DePaola, S 

Olariu, and I . 

, I. 

Popescu, J. Appl. 

Phys. 53, 

, 4645 (1982). 







B. Coll ins, J. 

Op t. Soc. 

. Am. B 1, 812 


10. C. B. Collins and B. D. DePaola In Laser Techniques in the 
Extreme Ultraviolet , edited by S. E. Harris and T. B. 
Lucatorto (AIP Conference Proceedings No. 119, New York, 
1984) p. 45. 

11. C. B. Collins and B. D. DePaola, Optics Lett. JJ), 25 (1985). 


12. B. D. DePaola, S. S. Wagal, and C. B. Collins, J. Opt. Soc 
Am. B 2, 541 (1985). 

13. B. D. DePaola, S. S. Wagal, and C. B. Collins, J. Opt. Soc 
Am. B (pending). 

14. C. B. Collins, M. Barb, S. Oiariu, and I. I. Popescu 

15. G. C. Baldwin, J. C. Solem, and V. I. Goldanskii, Rev. Mod 
Phys. 53, 687 (1981). 

16. C. Cohen-Tannoud j i and S. Haroche, J. de Physique, 30., 125 

17. S. Haroche, Ann. Phys. f>, 189 (1971). 

18. C. B. Collins, F. Davanloo, and T. S. Bowen, Rev. Sci. 
Instrum. 57_, 863 (1986). 

19. C. B. Collins, IEEE J. Quantum Electron. QE~20 . 47 (1984). 

20. L. C. Bradley, A. C. Mitchell, Q. Johnson, and I. D. Smith 
Rev. Sci. Instrum. 55 . 25 (1984). 

21. Q. Johnson, A. C. Mitchell, and I. D. Smith, Rev. Sci. 
Instrum. 5.1.. 741 (1980).