.
LOFT
ORNL P.
2106
v
:
.
-
5
.
.
:
f
(
)
bent
u
.
PFEFEEEE
EEE
.
-
1
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS – 1963
ORNU P-
2106
CONF-666303-28
CFSTI PRICES
H.C. $ 2.00;MN. 50
MAY 10 1966
Neutron Cross Sections in the Energy Range 100 eV < E. < 100 key: Recent
Progress, Current Status, Future Outlook*
J. H. Gibbons
Oak Ridge National Laboratory
Oak Ridge, Tennessee
Invited paper to be given at Conference on Neutron Cross Section Technology,
Washington, D. C., March 22-24, 1966.

RELEASED FOR ANNOUNCEMENT
"IN FUCIEAR SCIENCE ABSTRACTS
Research sponsored by the U.S. Atomic Energy Commission under contract
with the Union Carbide Corporation.
LEGAL NOTICE
This report mo propared us in account of Government sponsored work. Neither the United
Sualas, aor the Commission, nor any person acting on behalf of the Commission:
A. Makes any warranty or representation, expressed or implied, with respect to the accu-
racy, completeness, or usefulness of the information contalped in the report, or that the use
of any information, apparatu, method, or process dlacloud in this report may not infringe
privately owned righto; or
B. Aorumor any liabiliuos with rospect to the use of, or for damages resulting from the
um of any information, apparatus, motod, or procesi disclosed in Wale report,
Aund in the above, "person acting on beball of the Commission" includes any om-
ploym or contractor of the Commission, or employee of such contractor, to the oxtent that
soch employee or contractor of the Commission, or employee of such contractor prepares,
disseminates, or provides account, any information pursuant to his employment or contract
with the Commission, or ble employment with such contractor.
Neutron Cross Sections in the Energy Range 100 eV < < 100 keV: Recent
Progress, Current status, Future Outlook*
J. H. Gibbons
Oak Ridge National Laboratory
Oak Ridge, Tennessee
ABSTRACT
The technology of measuring cross sections in the
energy range 100 eV < En < 200 keV has advanced signifi-
cantly over the past several years. The most notable de-
velopments have occurred in detectors for capture cross
section measurements, available neutron energy resolution
in genera, and absolute valzes for flux normalization.'
Reactor design needs for total and capture cross sections
can apparently be met with existing or feasible equipment.
The primary exceptions to this statement are highly radio-
active materials and some of the higher accuracy requests.
Reaction theory and measured reaction systematics are in-
sufficiently understood to enable more than an educated
guess at unmeasured capture cross sections.
The technology of cross section measurements in the energy range
from 100 eV to 100 keV has advanced on several fronts over the past few
years. Vast quantities of data have been obtained where literally no
data had existed before and considerable improvement in accuracy and reso-
lution has been achived in general. If one views the problem of nuclear,
data for reactor development, particularly fast breeders, from the vantage
point of the "outside looking in" it would seem that the needs that can
be classed as critical are not so much for improved resolution, but for
greater accuracy. Therefore we will concentrate on accuracy and consistency
much more strongly than on resolution in this presentation.
When one measures a neutron cross section (we will consider calcu-
lations later) there are in general three quantities that need to be de-
termined:
(1) number of neutrons, N(E)E, in a given energy interval,
(2) number of target nuclei,
(3) detection efficiency, (En, etc.).
Research sponsored by the U.S. Atomic Energy Commission under contract
with the Union Carbide Corporation.
A very important exception to the statement that these quantities are
needed is the transmission-type measurement in which (1) and (3) essentially
cancel out. This type of measurement is typified by a total cross section
measurement, described by Professor Rainwater, which provides the most basic
cross section information. The accuracy of a total cross section transmission
measurement is liited only by sample thickness and purity, multiple scattering
effects, accuracy of background subtraction, and the counting statistics.
Consequently it is not out of the question to obtain a total cross section
accuracy of about one percent in the non-resonant region or where one is
averaging over many resonances and accuracies of a few percent for resolved
resonance widths. In practice, however, accuracies of about 5% are normally
obtained. Most total cross section measurements these days in the 100 eV
to 100 keV range are obtained using either electron linacs or synchrotrons.
DC beam Van de Graaff techniques, once the major source of data for
En > 10 keV, have found better application at higher energies. This point
will be considered in detail later by H. W. Newson. The pulsed Van de Graaff,
however, continues to be a useful tool for total cross section measurements
for En 24 kev.2 The normally available energy resolution is about 19,
and this is frequently sufficient. Perhaps its best application is with
separated isotopes, the small (~ 1 mm2) neutron source size allows
measurements to be made with very small samples. For example, the cross
section for Sov is given in Fig. 1.3 The sample was about 330 mg of
vanadium, 44% enriched in 5 .
Unfortunately the transmission technique is not readily applicable
to the measurement of partial cross sections. In fact the only significant
transmission experiment of a partial cross section is that of "spherical
chell" transmission" which is related to the absorption cross section. In
a typical experiment a small, isotpic source is placed at the center of a
spherical shell whose wall thickness is of the order of one or more neutron
mean free paths thick. An energy insensitive neutron detector placed at :
least several shell diameters away is used to measure the ratio of counts
with shell on to shell off (Fig. 2). While a fiumber of measurements have
been attempted using this technique the accuracy of the results remain in
some question because of uncertainties in several rather difficult scattering
corrections that must be applied to the data.4-7 Bogart's recent work
(included in this conference) on multiple scattering effects in the resonance
region represents an important new level of understanding of this important
correction factor. In addition the shell transmission measurement can only
be used with discrete energy sources, a very restrictive condition for the
energy range under consideration. This discussion is not meant to minimize
the contribution of spherical shell measurements to the establishment of
absorption cross section standards; rather it is meant to indicate the
limited applicability of the technique. Certain characteristics of the
shell transmission technique are given in Table I. The lowest energy
neutron source used in such measurements is ~ 24 keV (Sb-Be, y-n). In
this case absolute accuracy for oa 100 mb is approximately 25%. For
smaller cross sections the accuracy worsens due to a combination of poorer
statistics and larger correction factors.
Most shell transmission measurements have been performed with an
internal isotropic neutron source and external detectors. An alternative
geometry is an external "plane wave" source and an internal detector.8
Since there are no plane wave monoenergetic neutron sources this alternative
geometry is inherently less accurate because of the added correction
factors.
$
.
For the energy range under consideration the total cross section
consists solely of elastic scattering and capture. Thus one can consider
measuring scattering, rather than capture, to obtain the component cross
sections. This approach makes some sense at energies where In Poe 1.e.
for En si kev. A scattering cross section also has the advantage of being
closely akin to a transmission measurement. One recently reported experi-
bent was performed with a lithium loaded glass scintillator.9 So far,
however, it has proved to be much more practical for several reasons to
measure the capture cross section. The area of scattering cross sectior.
measurement is an example where thoughtful instrument development may well
reap benefits in more accurate partial cross sections for En I keV.
Before leaving the subject of neutron scattering we should note that
significant deviations from isotropic scattering (e.g. up to a factor of
two difference in do/an between oº and 180°) can occur for energies less
than 100 keV. It would be interesting to know what new user requirements
for ,elastic scattering angular distributions for En 100 keV may arise in
the next few years..
The most important partial cross section for Em < 100 keV is surely
capture. This cross section, frequently less than 10-309, has more often
than not eluded efforts to measure it accurately. The remainder of this
presentation is devoted to capture measurements. We have already touched
on the highly specialized and narrowly applicable transmission technique for
capture measurements. We now consider non-transmission experiments, which
means we must consider separately sources, samples, and detectors.
I. Neutron Sources
A convenient and natural division for neutron sources is whether or
not they are discrete in energy.
A. Discrete Energy Sources.
1. Photo-neutrons. Two reactions have provided virtually all
the isotopic keV range sources: Sb-Be and Th-D20. Some of their properties
are summarized in Table II. The Sb-Be source characteristics have bee
extensively investigated. It has become increasingly clear that the neutron
average energy, energy spread, and angular distribution are sensitive
functions of the physical source design. - A tremendous advantage of these
sources is that their emission rate varies only with the half life of the
y-ray emitter. A second tremendous advantage is that (because of their
compact size) they can be intercompared with world standard sources
(absolute calibration known to 2%) using a manganese bath or graphite
stack detector technique.
2. Neutrons from endothermic charged particle reactions.
a) ST(P,n) 'He. At threshold all neutrons appear near 0° due
to center-of-mass motion. The average energy is 65 kev. Monoenergetic neutrons
of less than 100 keV can be produced at back angles but it is better in this
case to use the (Li(pon) reaction.
b) 'Li(p,n)'Be. At threshold, as in a) above, all neutrons
appear near zero degrees with an energy of about 30 kev. "Monoenergetic"
neutrons, whose energy spread is determined by detector acceptance angle,
lithium target thickness, and proton energy spread are obtained for
emission angles > 90°.be
c) 31(pin)-cr. Although the neutron yield from this
reaction+3ad9 is vastly lower than a) or b) it has the advantage of slow
center-of-mass motion and consequently there is very little dependence of
neutron energy upon emission angle. This fact, coupled with the fact that
51cr, the residual nucleus from the (p,n) reaction is conveniently radio-
active, has enabled one group of investigators to use 450 geometry and
measure absolute activation cross sections versus energy. 14
B. Continuous Distribution Neutron Sources.
1. Fission Reactor, Neutrons from highly moderated fission
reactors have an energy distribution of 1/E in a portion of the region of
concern in this paper. Significant deviations from 1/E appear for
En 215 kev. Mechanical choppers are used to establish the time base for
time-of-flight measurements. Reactors have been largely supplanted by
electron linacs for studies in the energy range under consideration
2. Electron Linac. The neutron spectrum from linacs is, of
course, highly variable, mostly depending upon the exact geometry of the
source and the electron energy. Thus one cannot use source characteristics
to make an interpolation from one energy to another when determining
neutron flux. The principle advantages of the linac as a neutron source
are that (1) peak intensities are extremely high, (2) relatively short beam
burst widths are now feasible (< 5 nsec FWHM). A major disadvantage of the
linac is that since neutrons are produced in an evaporation process followed
by some moderation there are unwanted fast and slow components of the beam
pulse. There is little question but that the linac is a remarkably good
neutron source for energies from < 100 eV up to 30-50 keV for partial cross
sections and even higher for total cross sections. At the upper energy
range, and for reasonable flight paths, the y-flash and the several effects
due to fast neutrons provide a severe time-dependent background barrier for
capture measurements. Transmission measurements, on the other hand, are
no so adversely affected and can be carried out to considerably higher
energies. There is a reasonably good concensus anong linac users that there
is little early prospect for capture measurements above 30-50 keV. A
great advantage of linacs is their ability to produce a band of neutrons
that spans the energy range from essentially thermal to tens of keV and
consequently allow for fast cross section normalization to previously de-
termined absolute values in the eV and thermal range.
3. Synchrocyclotron. This sour.ce, like the electron linac,
produces fast pulses of evaporation neutrons. A valuable added feature of
the Columbia system_o is a synchronized mechanical chopper which removes the
unwanted low energy components of the beam. This device has been used almost
exclusively in transmission measurements and has produced a vast amount of
valuable, high resolution data on total cross sections.
ii .
4. Pulsed Van de Graaff. This source Todo has the advantage
of being able to produce neutrons only in the energy range of interest
since it uses endothermic (pen) reactions in thin targets. Neutrons can
be produced over a pre-selected band of energies in the range En z 10 kev.
This property results in allowing capture measurements to be made without
the large backgrounds inherent in linac experiments. The vastly lower peak
intensities are in part compensated by the higher (x 203) repetition rate,
Vineritets'..
' saios momenta bisita
-4-
131'
restricted energy band, and small source size.
5. Nuclear Explosives. The primary virtue of "bomb neutrons"
is that the instantaneous intensity can be so high that, even with flight
paths of about one kilometer, the flux is sufficient to enable measurements
on highly radioactive samples. This extremely important subject area is
covered in another section.
6. Lead Slowing-Down Time Spectrometer. This unique neutron
"source" utilizes the fact that neutrons suddenly introduced into a heavy
element medium slow down in time as a group. +9 Ne...trons from a pulsed (å,t:
source are introduced in the interior of a large lead cube. Then the
average neutron energy at the center of the cube is a function of the time!
lapsed. The principal difficulty with this "source" is that the energy
resolution is very poor (~ 15% for En < 1 keV and 35% for En ~ 10 keV).
Nevertheless there has been an impressive out-pouring of data, especially
for intermediate weight elements, over the past several years.
C. Flux Calibrations and Normalizations.
There are several ways to measure absolute numbers of neutrons
in the energy range under consideration.
1. One can produce neutrons in a nuclear reaction that has as
its product (in addition to the neutron) a convenient radionuclide whose
properties are well-known. This technique is of marginal benefit unless
the full (490) neutron flux is used in the cross section experiment, since
otherwise one must also know the angular distribution of the source. An
example of the use of this technique is the flux measurement of neutrons
from 51v(p,n)51cr. 15
2. One can immerse the source in a totally absorbing medium in
which essentially every capture results in a convenient radionuclide. This
classical technique is most frequently a manganese bath.
It has
been used in some of the most precise neutron flux measurements. Again,
3. As a variation of (2), one can place the source at the center
of a 41 counter whose absolute efficiency is well-known by intercomparison
by Macklin. 0 Its efficiency versus neutron energy is very nearly constant
and well-known. It is very insensitive to such perturbations as non-isotropy
of source, etc.
4. One can utilize a detector whose absolute efficiency may not
be known at the energy range in question but whose relative efficiency is
well-known and relatable to an absolute number at some other energy.
Examples are detectors using such sensors as 10B, SHe, 2350, Li, etc.,
whose absolute efficiencies can be determined either with a secondary standard
or independently using thermal or fast (by comparing to hydrogen recoil)
neutrons. Ultimately the most well-defined absolute cross sections which
are used as secondary values are selected thermal capture cross sections
(e.g. 1°B)2and the 1H(n,n) cross section at higher energies. These (boron
for thermal and H(n,n) for fast neutrons) cross sections are known to about
19. However by the time one arrives at any practical flux calibration in
the 100 EV < En < 100 keV range (by means of an intermediate flux shape
.5-
3tandard) the results are no more accurate than about 2% at best and more
usually 3-5%. Any attempts to attain flux accuracies of 2% or better in
th: energy range under discussion will have to be underwritten by a strong
need, especially for Em 10 keV. The reasons for this are fairly obvious.
a) Fast neutrons produce proton recoils in a variety
of detectors that are easily detected. The cross section H(n,n) has been
determined to very high accuracy for fast neutrons and provide an excellent
absolute standard. However it is very difficult to use recoil counters
at energies below a few hundred keV and exceedingly difficult below about
100 keV.
b) In a similar fashion the thermal cross sections for
several reactions, particularly tºB(ng) Li, YLi*, and 3He(n,p)T have been
very well determined by transmission. However, an extension of the
reaction cross section with high accuracy to higher energies becomes in-
creasingly difficult since other partial cross sections become increasingly
important and a simple transmission measurement no longer surfices. For
example, the total cross section for (He + n) is 99.36% absorption for
thermal neutrons but the two cross sections are of equal importance at about
70 keV. There are continuing efforts to increase the accuracy of these
cross sections, particularly that of 10B(nya). The concentration on boron
is for four principle reasons. First, the reaction has a very high cross
section. Secondly, the cross section is known to be nearly 1/v and without
resonance structure to about one hundred kilovolts. Thirdly, thin detectors
can be constructed using the 1°B(na) Li* component of the reaction that
have a very fast time response ~ 5 ns) and can therefore be used in fast
time-of-flight experiments. Lastly, in contrast to 3He(n,r) and Li(n,a)
this reaction produces a gamma ray rather than just a charged particle and.
thus provides radiation that can be used directly by the capture detector.
The boron reaction cross section has been investigated in the energy
range under concern by comparison with a counter with "constant" or at
least "known" efficiency, by a combination of total and scattering measure-
ments, by spherical shell transmission, and by measurements of the
10B(n,x) Li cross section by reciprocity from Lila,n)10B and combining
this with the branching ratio ''B(não)/10B(n,n) versus neutron energy.
The results have generally confirmed that the reaction cross sections are
in fact i/v in shape froin thermal to at least 50 keV, but it has become
increasingly clear that significant deviations from 1/v do occur at higher
energies. A current (if not perennial) topic of research is the exact
behavior of the partial cross sections for neutron energies between fifty
and several hundred key. This remains a very important research topic
since we badly need to more accurately bridge the gap between thermal values
and proton recoil techniques.
II. Samples and Sample Thickness Effects
..
.
.
It should no longer be a surprise to state that the cross section
one measures is normally a function of the sample thickness used. Professor
Rainwater has already talked to this point. There are in addition several
comments associated with samples that one should add, particularly for the
lower energies :
A. Chemical Purity.
Individual resonance effects are so important in the eV and low
V
6.
7
keV range that chemically negligible impurities (e.g. 1 ppm) can be
important. Perhaps a good example is 56Fe, which for years has hed a
capture resonance integral that is considerably greater than one can account
for with known individual resonances. Additional resonances have been
located;<3 but not nearly enough to account for the resonance integral.
One recent report, implied the presence of several resonances below about
a kilovolt that might account for the integral. However these have apparently
turned out to be due to small Mn and Co impurities.co We are about to con-
clude that the resonance integral, too, was measured with "impure" iron.
B. Corrections Associated with the Sample.
There are two coupled effects that must be considered when one
attempts to convert his data obtained with a "real world" sample to a cross
section for an infinitely dilute medium. These two effects are resonance
self-shielding and multiple scattering. 5-7,21
The degree of difficulty in making scattering corrections is roughly
proporticnal to the ratio o(peak)/(potential). The self-protection correc-
tion is worst for Is In As a consequence and largely thanks to the fact
of doppler broadening heavy element self-absorption and scattering correc-
tions are not large for energies down to about 10 keV, unless one uses
very thick sexaples such as spherical shells. Typical sizes of correction
factors are l-10% (known in turn to ~ 10%) on samples with an average
transmission of ~ 90% for En > 10 kev. A standard practice is to adjust
the sample thickness so that the correction factor remains in bounds, as
long as sufficient intensity is available.
The case of spherical shell transmission deserves special mention
since thicknesses used have been of the order of one to several mean free
paths. Until very recently thick sample multiple scattering corrections
assumed a smooth (non-resonant) cross section. This approximation is
reasonably valid in the higher energy range but was considered to be
questionable in the keV range. The recent Monte Carlo calculations of
Bogart and Semler now show conclusively that if one uses a more realistic
cross section (including doppler broadened resonances) the corrections can
differ significantly from the earlier values. It is interesting to note
that the corrections are dependent to first order upon statistical and
optical model parameters such as s- and p-wave strength functions and the
level width distribution function as well as the total cross section.
In summary, if the experimenter has sufficient intensity to keep
his sample thickness to o.1 neutron mean free path average then semi-
analytical corrections for scattering and absorption good to 1 or 2% are
feasible for energies down to about 10 kev. At lower energies, particularly
below a few keV the problem is much more complex since the peak to non-
resonance cross section is a 102. Detailed Monte Carlo calculations using
the real resonances involved are necessary if high accuracy is to be
achieved. Of course, the way to minimize this problem is to use sample
thicknesses of only ~ 0.2 mfp. at resonance, but this presents rather difficult
experimental problems.
III. Detectors
We divide this discussion into four general types of detectors.
A. Activation.
Although limited in use to those nuclides that activate with a
convenient half life and also only to monoenergetic beams, this method has
been very fruitful and is subject to relatively high accuracy (absolute
radioactivity counting) and sensitivity, since the counting is performed
separately from the neutron exposure. This technique has largely been ex-
hausted but it still finds valuable use in special cases such as recent
studies of capture in gold29 and sodium.30
B. "Total Absorption.
A major advance in capture measurement technology was effected
by the large liquid scintillator.3432! The detector geometry is virtually
49 and one observes the prompt y rays from capture, thus freeing ones self
from the restriction of activation. Major problems with large liquid
scintillators were the dependence of their efficiency upon the details of
the capture y-ray spectrum,27,32 and the large background counting rates
due to environmental radiation and particularly from radioactive samples.
In addition the time resolution of these devices (typically 10-20 nsec) has
been a limiting factor, especially in pulsed Van de Graaff work. However,
recently s. large improvement in the time resolution (a few nsec for
Ey > 0.5 MeV) has been reported. The largest liquid scintillator facility
for capture measurements, built by Haddad and colleagues 34 is shown in Fig. 3.
A typical y-ray pulse spectrum from neutron capture in a sample at the
center of the tank is shown in Fig. 4. The detector efficiency, proportional
to the integral under this curve when corrected for "total escape" proba-
bility, is rather accurately determined since there is only a small area
under the curve where one has to extrapolate to zero.
The large liquid scintillator has also been successfully applied
to the measurement of a = 0./07. The stated accuracy requirements for
a is such that it is not feasible to construct it from separate measure-
ments of fission and capture cross sections. One technique, 92,so applicable
for monoenergetic neutrons, involves the addition of a radiative neutron
absorber in the scintillator (gadolinium) that signals a fission event in
the sample by the combination of a prompt y-ray pulse followed several to
several tens of microseconds later by a pulse from the captured fission
severe restrictions are placed on the use of fast pulsed beam techniques.
With the exception or data using threshold neutrons at 30 and 65 keV the
data for 10 < En (keV) < 120 keV were obtained using a Van de Graaff pulsed
once every 32 usec (Fig. 5).31 More recently the delayed coincidence
technique to identify a fission event was removed by the development of a
technique that uses a fast fission counter as the neutron target. Jos
Thus the time delay required in the earlier system to classify the event
com
distribution neutron sources ard time-of-flight techniques. By this technique
a for 390 have been obtained from a few eV to more than a keV (Fig. 5).
The greatest difficulty in extending this technique to other nuclides such
as Su is the a-decay pile-up, which competes with fission counting
efficiency. The development of fast response fission counters that contain
sufficient material and have an acceptably small a pile-up remains an
important experimental challenge.
more recentesite.min.
C. Total Energy Detectors.
The advent of the large liquid scintillator left a residual desire
to find a detector which was insensitive to capture y-ray multiplicity
contestations are
nominate the
-8-
d
se
changes. This is most important in the study and comparison of neutron
capture in individual resonances. In addition it was hoped that a detector
with better time resolution might be found. This is particularly important
in certain pulsed Van de Graaff experiments where for intensity reasons
one wishes to use as short a flight path as possible. A final, pressing
motivation for a new detector was the need, particularly associated with
stellar nucleosynthesis problems, to measure capture cross sections of rare
isotopes (1.e. samples of a few grams weight). This meant that the detector
should not only be fast but also small. Moxon and Raeso developed a detector
that, when modified by Macklin, 39 provided most of the desired character-
istics. This "total energy" detector is the scintillation counter equivalent
of the thick wall geiger counter in that gammas are converted to electrons
for detection. The detector (Fig. 6 ) consists of a converter plate in
contact with a thin, fast plastic scintillator. By suitable adjustment
of converter plate composition (Fig. 7,8) and thickness one can produce a
detector whose efficiency is proportional to the gamma energy. Thus the
overall efficiency for defecting a radiative capture event is dependent only
upon binding energy and independent of cascade mode as long as the total
energy release is constant. The detector efficiency is ~ 3% vs, nearly 100%
for large liquid scintillators but the background level is at least two
orders of magnitude better. This type of detector has sufficient time
resolution ( 2 nsec), to allow meaningful measurements (~ 15% energy
resolution at 30 keV) of cross section vs. energy with a neutron flight
path of only a little over three, inches. An illustration of its signal
to noise for a 35 gm sample of 117Sn and ~ 30 keV neutrons is shown in
Fig. 9.
A later version of the same basic idea, due to a suggestion of
H. Maier-Leibnitz, 40 allows an impressive gain in detector efficiency.
The Moxon-Rae type detector capitalizes on physical properties of the
detector to produce a desired energy response. The new "total energy
detector" was developed utilizing the principle that if one knows the pulse
height vs. energy response matrix of any detector, a given overall response
can be generated by applying a suitable transformation or weighting to the
pulse height distribution (Fig. 10). This technique requires considerably
more elabcrate data acquisition apparatus and data processing procedures.
The net result, however, is a gain in detector efficiency of a 500%, i.e.
to about 15%. Preliminary data taken with this detector indicate that
the increased efficiency can enable us to measure small cross sections (a
few to tens of millibarns) using a 50 cm flight path and relatively smali
samples, giving an energy resolution of about 5 ns/m. This ability to use
longer flight paths also allows us to insert better collimation and
shielding, an important factor since we wish to extend measurements up
to several hundred keV.
Resuits
2
..
For energies less than a few keV there are two general kinds of
results emerging. One, typified by the niobium results in Fig. ll, is
classified by the medium-to-poor resolution available in the lead slowing-
down-time spectrometer and short flight path pulsed Van de Graaff experi-
ments. Individual resonance effects are soon washed out but one still
gets a good representation of the gross features of the cross section. A
recent scan of the cross section files at the Sigma Center at Brookhaven
shows that such data are now available on a wide variety of elements,
especially in the medium-weight range, The second kind of result is that
Startside the content to Arsitatea
-9-
.
of high resolution, obtained with liquid scintillators and total energy
detectors using linear accelerator sources. Figures 12 and 13 illustrate
these results. Individual resonance parameters such as I can be obtained
from a combination of these latter measurements (capture cross section)
with total cross sections, self-indication results, or scattering measure-
ments. Thus far the most productive combination has been the combination
of transmission and capture cross section. .
It should be reiterated that a capture measurement alone does not
determine a radiative width. The result of a single resonance capture measure-
ment is the capture area, Ay, the "single resonance integral," which is
approximately given by
4, -SO. (E)E - (VF,/)
(1)
There are many instances when the value of this resonance integral may be
known but the individual resonance parameters not known. Likewise in
situations where g and In have been determined by other experiments one
frequently knows Ane uf ultimate interest in reactor calculations, much more
accurately than rys the number normally reported in the literature. Finally,
one can obtain the capture resonance integral over partially resolved groups
of resonances (e.g. 250-500 eV range in Nb, Fig. 11 ) that may be of use to
reactor design but is of almost inconsequential use, per se, to the under-
standing of nuclear structure physics.
At energies greater than a few keV only lighter and magic nuclei have
observed (e.g. Fig. 14 ). Heavier elements in general have such close
resonance spacings that all one observes is a smooth average over many
resonances, such as shown in Fig. 15. The corrections necessary to transform
raw data to cross section for these two extremes are similar in that both
involve scattering and resonance self-shielding. However, in the case of
single resonances one uses explicit cross sections and resonance parameters
whereas in the case of heavy elements one uses statistical properties such
as average spacing and strength function as well as assumes the validity of
distribution functions such as the Porter-Thomas distribution of scattering
widths. These corrections have been developed to a reasonably satisfactory
state for thin samples if one is satisfied with an average cross section
accuracy of 2 5%. For better accuracies refinements such as thinner samples
and Monte Carlo techniques in analysis will have to be utilized.
The effects of l>o wave neutrons cannot be over-emphasized in the
0.1 < En (keV) < 100 range. For example, approximately 60% of the niobium
capture cross section at 10 keV is due to p-wave neutrons. In fact one
should include d-wave (4 = 2) effects for En > 100 keV but we don't know
enough about the systematics yet to be able to do so with confidence. Also,
such "constants" as (D) and (ry, can change significantly over an energy
interval as small as 100 keV. '
Another thing that should be mentioned about capture results in the
keV range is the discovery of a relatively large number of resonances not
observed in total cross section measurements despite the fact that experimental
resolution is considerably poorer than available for transmission measure-
ments. For example, in (pore + n) there are 2 resonances known from
-10-
transmission for 15 < En (keV) < 60, the second one being only recently
discovered.? However, it is observed that at least 4 resonances contribute
significantly to the capture cross section in this same energy region 19
(Fig. 16), two of which are to date unobservable even in a thick sample
transmission measurement (Fig. 17). Thus experimentalists can supply
capture resonance areas for which it is not yet possible to give a value
for the radiation widths, spin and parity, etc. In many other instances,
where the spin and neutron width are known, one can obtain a value of [y.
As a consequence our knowledge of the systematic behavior of radiative total
widths is rapidly growing. We are still a long way, however, from being
able to answer the question, for example, of the possible strong dependence
of ry on the parity of the capturing state.
The application of large NaI crystals to the study of resonance
capture -ray spectraG943 has afforded us a helpful tool in capture
diagnostics. This technique can produce absolute, reduced partial radiative
widths for resonances whose parameters are knowns and the spectra from
resonances whose properties are not known can be helpful in determining the
spin and parity of the capturing state.43 One new interesting feature that
has already emerged from such studies is that for some elements in the weight
range 20 < A < 40 magnetic dipole transitions dominate over electric dipole
transitions. For example, capture in the 2º, 25 keV resonance in F(n,r)
decays via a number of relatively low energy magnetic dipole gammas instead
of the expected result of electric dipole gammas to and near the ground
state of 20F (Fig. 18).
A valid question is to what extent can one use nuclear reaction
theory to calculate capture cross sections. This approach was used some
years ago by people concerned with fission product poisoning and they got
their fingers burned because of lack of appreciation for p-wave neutron
capture contributions in the keV range. "Theory", in the context of neutron
capture is not at all akin to the elegant and accurate calculations of
scattering made possible, for example, by the non-local optical model.
Rather we use the term "theory" to describe the formal process which one
uses to manipulate empirically derived systematics to arrive at a desired
cross section. For example, one can express the capture cross section, in
terms of a statistical-optical model, as27
202
0=
Sess
;(n,d,j)><ry (7)>
— F(25)
(2)
<D_><ry>
Thus, if one knows (or can calculate) average values versus excitation
energy of radiative widths, level spacings, scattering widths, for various
resonance spins and parities, and neutron penetrabilities versus angular
momentum and energy he can calculate an absolute capture cross section. Un-
fortunately there are no easy shortcuts. In addition such a calculation
ignores possible direct capture processes that can be important, especially
in the MeV range. Also, this approach is obviously not valid when one is
interested in energy intervals of the order of the level spacing, for then
individual resonance characteristics become important.
-ll-
Reaction theory per se at its present state of development is of
very limited value in accurately predicting capture cross sections. On
the other hand empirically determined reaction systematics are proving to
be useful guides in cases where one is averaging over a statistically
significant number of levels. For example, a plot of stable isotope cross
sections for 65 keV neutrons, when empirically corrected for even-odd
effects, shows a remarkably small point scatter (Fig. 19). Unfortunately
this curve has very little to say about nuclides well off the valley of beta
stability or those heavier than bismith. Studies of capture as a function
of neutron excess may ultimately shed light on the former and Bell's
Interpretative studies44 of heavy element yields from bomb explosions are
already providing some of our best guesses about fast capture in such problem
nuclides as <33pa.
Conclusion
There has been a veritable explosion of data, particularly on capture
cross sections, over the past half decade. Where cross sections were non-
existent or were uncertain to a factor of two we now typically find results
with accuracy frequently approaching 10%. Even more significant is the
fact that results obtained using completely independent techniques are
more frequently agreeing with each other these days. This progress has not
come easily or by a few great breakthroughs; rather it is the collective
result of many individuals and small groups who are steadily chipping away
at the problems of neutron sources, flux standards, detectors, and correction
factors. A few of the more recent advances, however, deserve special
recognition: Moxon, Rae, and Macklin's work on total energy detectors,
Weston and deSaussure's fast, high efficiency fission counter, Haddad's
work on am improved massive liquid scintillator, Grench, et al.'s work
on gold activation using the v(p,:) reaction, and most recently Bogart's
Monte Carlo studies of shell transmission corrections. It is interesting:
to note that most of these advances have been made possible by small efforts
with a wide geographical representation rather than massive assaults.
The greatest challenge to the cross section measurer, from the point
of view of the user's interest, seems to me to lie in two main areas:
1) improvement in absolute accuracy and 2) ability to handle radioactive
samples. Advances in both areas are occurring but progress will probably
not come as easily as in the past. However, I am convinced that progress
will come, as it has in the past, from a variety of small groups with open
minds to innovation.
It is always well to end a presentation such as this one with a note
of realism. Thus, lest the reader concludes that all is now "well" in the
capture cross section business, I want to refer you to the results to date
of measurements on yttrium (Fig. 20). It should be clear that, despite all
I may have led you to believe, we are still in pretty bad shape. As Pogo
once said, "From here on down it's uphill all the way."
-12.
Table I
Spherical Shell Transmission Technique
Advantages
(1) Obviates need for neutron flux
calibration
Disadvantages
(1) Requires monoenergetic
source
(2)
Insensitive to detector efficiency
(to first order)
(2) Requires large amount of
sample
(3) Completely insensitive to mode of
decay of final nucleus.
(3) Requires extensive scattering
and self-shielding corre-
ctions, especially for
E s 10 keV.
Corrections
(1) Resonance self protection effects
(2) Multiple scattering in shell material
(3) Finite source/detector size effects
(4) Effects due to neutron energy degradation by scattering
Effects due to non-isotropic source
bermain.
-13-
Table II
Discrete Energy Neutron source, 100 eV < Enga < 100 keV with sufficient
intensity for practical use in partial cross section measurements.
Source
Energy (kev)
Typical Resolution, AE, (kev)
24
IV
no
(195)
V
no
1
30
Sb-Be (Y,n)
Th-D20 (y,n)
?Li(p,n)?Be at threshold
3r(p,n) He at threshold
7L1(p,n)?Be at 2 90°
52v(p,n)52cr at any
1
N
no
5 to > 100
5 to > 100
I
no
-14-
-
-
-
•
--
-
-
-V
-
-
--
-
K
-
References
1. P. F. Nichols, E. G. Bilpuch, and H. W. Newson, Ann. Phys. 8, 250 (1959).
2. See, for example, R. L. Macklin, P. J. Pasma, and J. H. Gibbons,
Phys. Rev. 136, B695 (1964).
3. W. M. Good and R. C. Block, private communication (1965).
4. H. W. Schmitt and c. W. Cook, Nucl. Phys. 20, 202 (1960).
5. H. W. Schmitt, ORNL-2883 (1960) unpublished.
6. L. Dresner, Nucl. Instr. and Methods 16, 176 (1962).
R. L. Macklin, Nucl. Instr. and Methods 26, 213 (1964).
8. R. L. Macklin, H. W. Schmitt, and J. H. Gibbons, Phys. Rev. 102, 797 (1956).
9. M. Ashgar, M. C. Moxon, and C. M. Chaffey, Proc. Int. Conf. on the
Study of Nuclear Structure with Neutrons, Antwerp, Belgium, July 1965.
10. H. W. Schmitt, Nucl. Phys. 20, 220 (1960).
21. K. É. Larssc :, J. Nucl. Energy 6, 322 (1958).
d Meth
'
J. H. Gibbons and H. W. Newson, Fast Neutron Physics (Interscience
Publishers, Inc., New York, 1960), Part I.
13. J. H. Gibbons, R. L. Macklin, and H. W. Schmitt, Phys. Rev. 100, 167 (1955).
14. K. K. Harris, H. A. Grench, R. G. Johnson, and T. J. Vaughn, Nucl. Instr.
and Methods 33, 257 (1965).
15. J. H. Gibbons and R. L. Macklin, Nucl. Instr. and Methods 31, 330 (1965).
16. J. Rainwater, W. W. Havens, Jr., J. 8. Desjardius, and J. L. Rosen,
Rev. Sci. Instr. 31, 490 (1960).
-
-
17. W. M. Good, Neutron Time-of-Flight Methods (EURATOM, Brussels, 1960),
p. 309.
18.
C. D. Moak, W. M. Good, R. F. King, J. W. Johnson, H. E. Banta, J. Judish,
and W. H. dupreez, Rev. Sci. Instr. 35, 672 (1964).
19. Yu. P. Popov and F. L. Shapiro, Zh. Eksperim. 1. Teor. Fiz. 15, 683
(1962) (English transl.: Soviet Phys. - JETP 42, 988 (1962)7.
20. R. L. Macklin, Nucl. Instr. 1, 335 (1957).
21. H. W. Schmitt, R. C. Block, and R. L. Bailey, Nucl. Phys. 17, 109 (1960).
22. J. Als. Nielsen and 0. Dietrich, Phys. Rev. 133, B925 (1964).
23. J. A. Moore, H. Palevsky, and R. E. Chrien, Phys. Rev. 132, 801 (1963).
-15-
References (cont'd.)
24. M. C. Moxon, Proc. Int. Conf. on the study of Nuclear Structure with
Neutrons, Antwerp, Belgium, July 1965.
25. F. Mitzel and H. 8. Plendi, Nukleonik 6, 371 (1964).
R. C. Block, B. Steadman, and R. R. Fullwood, private communication (1966).
J. H. Gibbons, R. L. Macklin, P. D. Miller, and J. H. Neiler, Phys. Rev.
122, 182 (1961).
28. D. Bogart and T. T. Seculer, NASA TMX-52173 (1966). (See also contributed
paper, this conference).
29. K. K. Harris, H. A. Grench, R. G. Johnson, F. J. Vaughn, J. H. Ferziger,
and R. Sher, Nucl. Phys. 69, 37 (1965).
30. C. LeRigoleur, J. C. Bluet, H. Bell, and J. L. Leroy, Cadarache Centre
report No. DRP/SMNF 65/02 (unpublished).
31. B. C. Diven, J. Terrell, and A. Hemmendinger, Phys. Rev. 120, 556 (1960).
32. R. L. Macklin, J. H. Gibbons, and T. Inada, Phys. Rev. 129, 2695 (1963).
33. L. W. Weston, private communication (1966).
34. E. Haddad, 8. J. Freisenbahn, F. H. Frobner, and W. M. Lopez, Phys.
Rev. 140, B50 (1965).
35. J. C. Hopkins and B. C. Diven, Nucl. Sci. Eng. 12, 169 (1962).
36. L. W. Weston, G. deSaussure, and R. Gwin, Nucl. Sci. Eng. 20, 80 (1964).
37. G. deSaussure, L. W. Weston, R. Gwin, J. E. Russell, and R. W. Hockenbury,
Nucl. Sci. Eng. 23, 45 (1965).
38. M. C. Moxon and E. R. Rae, Nucl. Instr. and Methods 24, 445 (1.963).
39. R. L. Macklin, J. H. Gibbons, and T. Inada, Nucl. Phys. 43, 353 (1963).
40. H. Maier-Leibnitz, private communication (1963). See also F. Rau,
Nucleonik 5, 191 (1963).
See, for example, the paper by Frohner, Haddad, Lopez, and Friesenhahn
(this conference).
42. J. R. Bird, J. A. Biegerstaff, J. H. Gibbons, and W. M. Good, Phys.
Rev. 138, B20 (1965).
43. I. Bergqvist, J. A. Biggerstaff, J. H. Gibbons, and W. M. Good,
Phys. Rev. Letters 18, 323 (1965).
44. George I. Bell, Phys. Rev. 139, B1207 (1965).
-16-
Figure Captions
Fig. 1.
Total cross section of (pºv + n). The sample consisted of 330
gms of isotopically enriched Suv. These data were obtained with
a pulsed Van de Graaff accelerator.
Fig. 2. Apparatus for the measurement of neutron absorption cross sections
by spherical shell transmission.
Fig. 3• Experimental facility for measuring radiative capture cross
sections using a large (4000 liter) liquid scintillator.
Fig. 4 - Gold resonance neutron capture gamma ray pulse height distri.
butions in a large liquid scintillator. The upper curve was
obtained before steps were taken to eliminate capture of
scattered neutrons in the scintillator.
Fig. 5 - Alpha, the ratio of capture to fission, of SPU as a function
of energy
Fig. 6 - The 'Moxon-Rae-Macklin" detector consists of a plate that converts
gammas to electrons, which are in turn detected by a thin plastic
scintillator. This detector is very fast and compact, which
allows measurements at short flight paths. It is also very
insensitive to neutrons, which allows measurements where
oc «oscº
Fig. 7 - Efficiency per MeV of the Moxon-Rae detector versus gamma energy.
The ideal response would be a constant efficiency per MeV.
Fig. 8. Efficiency per MeV of the Moxon-Rae detector as modified by
Macklin. The efficiency per MeV is much more nearly constant
than the detector using pure graphite.
Fig. 9. Moxon-Rae detector counts versus time for a 35 gram sample
exposed to 30 keV neutrons from a pulsed Van de Graaff.
Fig. 10 - Pulse-height weighting function to create a plastic scintillation
detector response function that is proportional to gamma ray
energy
Fig. ll -
Capture cross section for NO + n). The data were obtained
using the "lead slowing down time spectrometer" which has relatively
poor energy resolution.
Fig. 12 -
C
C
Capture cross section for (In + n), obtained using a large liquid
scintillator and electron linac neutrons.
Fig. 13 - Capture cross sections of isotopes of tungsten. These data were
obtained with an ORNL scintillator tank on the RPI linac.
Fig. 14 - Capture cross section for ( Pb + n), obtained with a "total
energy detector" using neutrons from pulsed-bunched Van de Graaff.
Fig. 15 - Capture cross section for (NO + n). The various symbols stand
for different measurements, mostly by independent techniques.
The shape of the curve is determined by statistical nuclear properties.
-17-
Fig. 16 - Capture cross section for (°Fe + n). At least four resonances
contribute to the cross section for 15 < En (kev) < 60.
Fig. 17 - Neutron transmission through a thick sample of Fe. Only
two resonances are discernable for 15 < En (keV) < 60 keV.
Fig. 18 -
Gamma ray spectrum from 27 keV (2) resonance neutrori capture
in (19F + n). The spectrum due to thermal capture is indicated
by the vertical lines. Surprisingly the p-wave resonance
capture, which could lead to and near the ground state via
electric dipole emission, proceeds instead via magnetic dipole
to higher excited states.
Fig. 19 - Capture cross section for En = 65 keV versús atomic number.
The effects of neutron and proton shells are very clear. A
remarkably smooth curve is obtained if one weights even Z,
even N target cross sections by a factor of 2.2, independent
of A.
Fig. 20 -
Capture cross section for (°Sy + n). The resonance spacing
is a few key. The different symbols represent independent
experiments. This example was purposefully chosen to point
out that there are still some stable element cross sections
that are very poorly known.
..
..
..
...
...
.......
'! -.
*
-18-
اددا
سلسسلسلسلسسسلسسبسلسسلندا
3
4
5
6
7
8
9 10
En (kev)
۱ITIIIIII
30 40 50
60
UNCLASSIFIEO
ORNL-LA-OWG 241SIR
000
0
00
:
100
t0
i
.
110
000000
.
.
.
.
.
Pb-
PARAFFIN
BAC
.
WWW
mirage
06
PARAFFIN
COUNTERS
20
he Woman
INCHES
Sb-8e
SOURCE:
T
MATING
HEMISPHERES
CROSS SECTION
SAMPLE MATERIAL-
LONG COUNTER
ASSEMBLY
MOVABLE
CARRIAGE
TE
med
IT
II
.
DI
I
GUIDE FROM SOURCE
STORAGE CELL
bution
10
20 25
lo
1.060-in. DIA
SPHERE-
0.030-in. A
0.030-in. Be
0.940-in. DIA SO-
SIHONI

...
Apparatus for the Measurement of Neutron Absorption Cross Sections.
SAMPLE
LIQUID SCINTILLATOR SOLUTION (ACTIVE)
LIQUID SCINTILLATOR SOLUTION (INACTIVE;
2 LEAD
-6° BORON-PARAFFIN
.
-
**
*
+
UGHT TIGHT
PLYWOOD BOX
- 5" PHOTOMUL-
TIPLIER TUBES
.
- SELF-INDICATION
WELL
+
.
1
i
AN
BFG MONITORS
... we
NEUTRON
BEAM
DIRECTION
O
TANK ACCESS
DOOR FOR NORTH
TUBE BANK
ON
fed
Til
COLLIMATOR
SHIELDING
SHIELDING
TANK ACCESS DOOR FOR
SOUTH TUBE BANK
7. 39,2..11

i
2017*

3000
GAMMA PULSE HEIGHT
DISTRIBUTION FROM
5 V GOLD RESONANCE
NO LINER OR METHYL
BORATE.
00
COUNTS PER CHANNEL
Bas googoo 00
1000
GAMMA RAY ENERGY (MOV).

O AMBIENT
I BACKGROUND
3000
GAMMA PULSE HEIGHT
DISTRIBUTION FROM
5 V GOLD RESONANCE
B'O LINER 10% METHYL
BORATE IN CENTER
SECTION.
PER CHANNEL
2000
ecco o que solo
1000
ooooooooooo
-
-
.. ...
.
.
w
MACHINE
BACKGROUND
- . r
..
6 7 8
GAMMA RAY ENERGY (MOV)
ORNL-DWG 64-5817R2
TTTTIIIIIIIIIIIIII
+ DIRECT MEASUREMENTS BY ORNL AND RPI
DIRECT MEASUREMENTS BY ORNL
O DIRECT MEASUREMENTS BY LOS ALAMOS (NUCL. SCL ENG.
12, 169 (1962)]
- A INTEGRAL MEASUREMENTS BY KAPL (PROC. INTERN. CONF.
PEACEFUL USES AT. ENERGY, GENEVA, 1955, 4, 315)
O CALCULATIONS BASED ON MEASUREMENTS OF OF AT SACLAY
AND OF O, AT HARWELL (AERE-M-4272)
- O FROM MEASUREMENTS OF 9, 0, AND G, AT HARWELL (PRIVATE
COMMUNICATION)
fillol
o lºlol loll 197411
Mutta
1,000,000

10
100
1000
10,000
400,000
Ę (OV)
!
.
-** **
..--
?!
PHOTOMULTIPLIER &
PLASTIC SCINTILLATOS
LICH COLLIMATOR
NEUTRON TARGET
PROTON TARGET
CONVERTER PLATE
".
..
MON
ANOMAT
60723

.
UNCLASSIFIED
ORNL-LR-DWG 75650
10-2
EFFICIENCY / Ey
GRAPHITE SLAB
THICKNESS: 2.54 cm
SOURCE DISTANCE: 5.08 cm
ELECTRON DENSITY: 5.4 x 1023 cm 3
I CALCULATED CURVE FOR ZERO BIAS
COMPTON AND PAIR FIRST
INTERACTIONS
II EMPIRICAL CALIBRATION - FINITE BIAS
10-3
0
2
4
6 - 8
Ey (Mev)
10
12
14

UNCLASSIFIED
ORNL-LR-DWG 75651
II
EFFICIENCY/EY
Biz O3 +C MIXTURE
THICKNESS : 2.38 cm
SOURCE DISTANCE : 5.08 cm
ELECTRON DENSITY: 8x1023 cm 3
COMPOSITION: C-78.5 at. %
Bi- 8.6 at. %
0- 12.9 at. %
CALCULATED CURVE FOR ZERO BIAS
COMPTON AND PAIR FIRST
INTERACTIONS
II EMPIRICAL CALIBRATION – FINITE BIAS
10-3
0
2
4
10
12
14

6 8
Ey (Mev)
jantar
minnetta
nt
en
UNCLASSIFIED
ORNL-LR-DWG 68051
300
_
GRAPHITE MOXON-RAE DETECTOR
30-kev NEUTRON CAPTURE IN
- Sn"7 (35 -GRAM SAMPLE)
06 = 390 1 80 mb
PROMPT
GAMMA-RAY
PEAK
NUMBER OF COUNTS PER CHANNEL/14 min
1
140
230

150 160 170 180 190 200 210 220
TIME DELAY CHANNEL NUMBER (0.85 nsec/channel)
UNCLASSIFIED
ORNL-DWG 64-8387
1400
2.64 cm
1200
100
10.16 cm
00
7.62
800
G (1), WEIGHTING FUNCTION
TWO PLASTIC SCINTILLATORS
10.16-cm diam X 7.62-cm THICK
SOURCE TO DETECTOR FACE
DISTANCE 2.64 cm
G (I), WEIGHTING FUNCTION
400
200
2
4
6
14
16
18
20

8 10 12
I, PULSE HEIGHT (mec)
Weighting Function for Total Gamma-Ray Energy.
ORNL-DWG 66-2212
.
N693
Oc (barns)
0.01
0.02
0.05
0.1
0.2 0.5
En (kėV)
1
2
5
10

MELMARATY W T
TARGET DATA
100000
10
NOTION EGY (en)
Le-40546

-
-
-
..
.

..:
.
CARLOWG 65-2424
NEUTRON ENERGY (OV)
1400 1200 000
8000 6000
4000
2000
1600
900
&
3000 2400
TTTT
estimui
182 w
:::::
::
0
TTTTTTTTTTTT
TTTP
184W
so
** :
CAPTURE CROSS SECTION (borno)
тттттттттттт —
TT
LULUILL
486W
sot
...
.::
183w
"roo
TULOS
Lanciranimiz var
.
125
200
300
400
800
300
100
500
600
CHANNEL: NUMBER (008 usec/channel)
-
-
-
-
-

ORNL-DWG 63-7850
Pb 207
CAPTURE CROSS SECTION (mb)
10
15
20
50
55
60
25 30 35 40 45
NEUTRON ENERGY (keV)
ORNL-DWG 66-2211

1000
500
11
4, No 93
(qw)
200 op de fotocaba
20
0.01
0.02
0.05
0.2
0.
5
1
0.1
En (MeV)
ORNL-DWG 63-7847
CAPTURE CROSS SECTION (mb)
10
15
20
25 30 35 40 45
NEUTRON ENERGY (keV)
50
55
60


arl-OWG 64-754
1.
09
08
hogy nem
a6
TRANSMISSION
Foss
18.4 barns/atom
0.2
TTTT
TTTTTT
30
NEUTRON ENERGY (kev)
So

ORNL-DWG 63-5741R
т
F19(n, y) F20
ERES = 27 keV; 1" = 2-
т
т
т
COUNTS/AE (arbitrary units)
Т
Т
0
1
2
5
6
3 4
ENERGY (MeV)
'
3+2+
(2,3,410 atd
(1,2,3)
PREDOMINANTLY
(1.21+
? \(2,3)+
+
-
-
-
.
-
-
-
-
-
-
-
ORNL-LR-DV'G 70890R 2
-
• ODD Z TARGET
O EVEN Z TARGET WITH E-E COMPONENT
WEIGHTED BY MULTIPLYING BY A FACTOR
OF 2.2
..
.
_____
_
_
2 x 103

.....
....
...
.
Coff (mb) at 65 kev
_
__
_
_
_
0
20
40
60
80
100
120
940
Average Capture Cross Sections for 65-kev Neutrons.

*---*
-
-
-
-
-
ORNL-DWG 66-2213
8 8 8
439789
☺
oc (mb)
.
ow so voo
0.02
0.05
0.1
0.2
0.5
En (MeV)
*
*
-
*-
-
.
2.
V
-
--
47
END
22
-
-
DATE FILMED
6 / 7 / 66