ORIG. RUSSIAN: EXPERIMENTAL STUDY OF CRITICAL SYSTEMS WITH LI7H AND ZR H1.6 MODERATORS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP88-00904R000100110047-8
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 19
• Document Creation Date: December 22, 2016
• Document Release Date: August 26, 2009
• Sequence Number: 47
• Publication Date: May 1, 1964
• Content Type: STUDY

 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
Third United Nations
International Conference
on the Peaceful Uses
of Atomic Energy
A/CONF.28/P/878
US 3R
May 1964
Original: RUSSIAN
Confidential until official release during Conference
l:i:{PI f72NI1~P1'1'~+1, 1TU iJf O1'' CiiI'1'ICAi, I'.IY:J'1.'1;M:I 71I'1'II
Ili 711 and Zr 111.6 MO DINRATOIf;I .
V.A.I(uznelcov, Y.A'-.Prokhorov, I.I.Zalrharkin, I.lE'.:~omov, V.G.Maltsev
Recently special attention has been attracted to the use of
hydrides of metals as hydrogenous moderators. Hydrides of metals,
containing a great amounL of hydrogen per volume unit, have a
strong moderating power close to that of water.
In this paper the moderating power of lithium and zirconium
hydrides is considered.
Since the na-currtl mixture
ture cross section (6th = 71),
the natural isotope mixture is
neutron capture cross section
of lithium isotopes has a large cap-
the isotope Li7 comprising 92.5% of
of particular intere st p the thermal
of Li7 being 33 mb.
The separation of the lithium isotopes presents no difficulty,
due to the large relative difference of atomic weights (171,x).
When insignificant admixture of Li7 are present
des, or when small amounts of hafmium are found
in lithium hydri-
in zirconium hyd-
ride, both moderators have a sufficiently small capture cross sec-
tion. Besides the zirconium reduces the neutron moderation length
due to inelastic scattering (Table I).
T a b 1 e I.
Moderating Properties of Li7 Hydride, Zr Hydride and 'later.
Moderator
Formula
Density
(g/cm3)
Li7 hydride
Li7 H
0.8
Zr hydride
Zr H1 *6
5.0
Water
H2 0
1.0
Macroscopic Moderating
thermal neut- Power (epi-
ron
b
sorp
a
ti
-
h
l
erma
t
new
-
t
on cross sec- rons)
t ion
0.0231 1.16
0.0458 0.99
0.0222 1.28
} formula corresponds to the zirconium hydride composition,
used in the critical assemblies, described later on.
25 YEAR RE-REVIEW
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 T a b 1 e No
Core and Reflector Composition of the PP--I assemblies.
Element
Thickness
U0 2 (9096 0.48 g/om
enrichment)
Stainless
steel
Varied
Reflector
Reflector
CRITICAL ASSEMBLIES WITH HYDRIDE MODERATORS.
Lithium? hydride and zirconium hydride have been investigated
as moderators on the PP-4 physical assembly (zero-power reactor).
The detailed description of this assembly is to be found in
[1], therefore here only its main characteristics are mentioned
briefly .
Our physical assemblies are a set of aluminium tubes (50mm -
o.d., 1mm - wall thickness, 51mm - lattice spacing) placed into a
cylindrical containing vessel and surrounded by a biological shi-
elding of cast iron blocks filled with paraffin . The core and
reflector components are placed inside the aluminium tubes.
In the experiments described elements with a 47mm diameter
assemblies with various parts of volume occupied by air.
the main components of the core. These rings allow to make up
are used (see Table II). Thin aluminium rings are inserted between
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
Canning
Notes
Steel; In assemblies
0.2 mm with Zr1I1.6
CP In assemblies
7
0.1 mm with M.
11
Al 0.1nun
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 ? Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
of some effects occuring in thick elements (see e.g. (i J). In our
ease, however, these effects are small and the shielding factor
itself is close to unity.
On multiplying the experimentally obtained values of the
critical load and the shielding factor, we get values for (;Cz for
systems with homogenously arranged uranium.
T a b 1 e IV.
Values of Shielding factor (f) for Assemblies with
Lithium Hydride Moderators.
Assembly index
Uranium Layer
Thickness
Shielding,
factor
HL - 1
HL - 2 and
113 mg/cm2
0.844
HL - 3
HL- 4 and
58 mg/cm2
0.903
HL - 5
26 mg/cm2
0.948
3). The third correction eliminates aluminium present in the
form of aluminium -tubes of the core stacks (7 volume %) and spa-
cing rings, indispensable for the porosity of the system. By this
correction all the systems are brought to similar conditions as
to the quantity of structural mate: -_al (aluminium). The introduc-
tion of this correction is based on measurements of average core
reactivity coefficients of aluminium and uranium. It comprises
approximately-.8% of the critical load.
4). Measures have been taken that the ratio of the core
height to the equivalent core diameter should be close to unity.
The corrections for the form factor, evaluated according to data
quoted in 14] do not exceed 1%.
Table V and Fig. I present the critical parameters of the
assemblies after all four corrections have been introduced. The
introduction of corrections implies that the critical parameters
correspond to those of homogenous cylindrical assemblies (where
- 5 -
0
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8 0
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
the height and diameter are equal) with similar beryllium refle o-
tors whose effective thickness is 9cm. The strong dependence of the
critical parameters on the porosity of the system (or on the part
of volume occupied by the moderator EL;H ) should be noted. So
should be the fact that the critical load decreases (when EL;H _
const.) with the decrease of H to U concentration ratios right
to the ratio of 120, which is considerably lower than the optimal
nuclear concentration ratio in a water moderated reactor.
The last circumstance is probably due to the decreasing of the
thermal utilization factor in lithium hydride systems as compared
with water moderated systems, because of the large capture cross
section of lithium hydride (mainly, thanks to Li 6) in comparison
to that of water.
T a b 1 e V.
Crit:. cal Parameters of Assemblies with a Lithium hidride Moderator
1.Lattice distortion by control rods:
a change of load (k of U-235)
b change of volume (1)
2.Change of load due to shielding
(kg of U-235)
-
-
-0.78
-0.32
-7.4
-0.54
-0.22
-4.1
-0.40
-
-
-0.24
-0.08
-2.6
-0.21
3.Change of load due to eliminating
of aluminium (k~, of U-235)
0.12
0.46
0.16
0.04
0.01
Critical Parameters with Corrections:
Critical load (k of U-235)
4.30
5.55
3.86
4.32
3.78
Critical volume (1)
62.0
131.5
79.5
190
129
Parameters
HL-1
HL-2
I1L-3
HL-4
HL-5
Composition of Assembly:
.PN/ u215including shielding factor
122
245
220
445
464
Volume percentage of lithium hydride
Volume percentage of Uranium
30.5
30.3
40.3
1} 0.1
50.2
Critical Parameters without Correcti-
ons:
Critical load (k of U235)
4..
.95
4.32
4.52
4.06
Critical volume (1)
62.0
138.9
83.6
190
131.6
Corre etions:X)
The sign 10+" indicates load or volume increase after intro-
s duction of correction.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
1.2. Critical Assemblies with Zirconium Hydride Moderator .
Assemblies with zirconium hydride moderator have a relatively
hard neutron energy spectrum, being, characterized by the nuclear
concentration ratios of II to U235 12./I and 24.8. The assemblies
with a zirconium hydride moderator are surrounded by a beryllium ref-
lector 18cm thick (with the same volume percentage of beryllium
and aluminium as the lithium hydride assemblies).
There is a steel layer more than 200mm thick, i.e. of practi-
cally "infinite" thickness, around the beryllium reflector.
Uranium oxide with steel shells have been used in these asse-
mblies (see Table II). The composition of the two assemblies inves-
tigated is given in Table VI.
While determining the critical parameters, measures have been
taken that the ratio of the core height to its equivalent diameter
should be close to unity. This brings about a small critical num-
ber of core stacks (see Fig.2), which in turn requires a different
technique for determination of critical parameters of the assembli-
es, since each core stack changes considerably the effective mul-
tiplication factor.
To find the critical parameters, several systems with equal co-
re compositions but different in core height, that is with various
amounts of elementary cells (fuel-moderator complects) in a core
stack, have been assembled.
Some of the assemblies are subcritical, others insignificant-
ly supercritical. The multiplication factor (Keff) of the assemb-
lies is determined by means of control rods, whose efficiency has
been measured in these assemblies (beryllium blocks in an alumini-
um tube placed in the side beryllium reflector acting as control
rods).
Having found Keff as the function of the core height, the
latter is determined by linear interpolation for Keff=1 (see Fig-3).
In the same way the corresponding load of U-235 and the core
volume,, are obtained.
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
With the help of interpolation lines the value 11 /,&H (i.e.
the relative worth of the core volume) can be determined.
This term characterizes the neutron leakage from the core and
shows what changes the reactivity and fuel. load will undergo on
change of the height of the system.
Since the ratio of height to diameter of both systems is
close to unity (i.e. the form factor introduces no significant cor-
rections) the term kK! HH K / AV
V -
The relative worth of the core volume of assemblies HZ-1 and
HZr-2 turns out to be 30.2~eff and 31.0 fit , respectively (Pelf
being the effective part of delayed neutrons).
After obtaining critical parameters the following corrections
have been introduced:
1). Correction for reactor lattice distortions (namely, dis-
tortions of the reflector) by shut-down rod channels, introduced
in the same manner as the analogical correction for assemblies -
with a lithium hydride moderator.
2). Corrections connected with the presence of steel (uranium
elements shells) and aluminium in the reflector have been introdu-
ced with the aim of eliminating structural materials present in
different quantities in the systems.
These corrections have been deterin-1-ac l by means of experimen-
tally measured steel and aluminium reactivity coefficients.
Table VII brings the corrected critical parameters of the
assemblies.
The self-shielding effect of uranium elements due to their
considerable thickness is also an important correction, influencing
the critical load of the reactor.
To measure the shielding effect experiments have been made 2
where specific gamma-activities of "thick" uranium dioxide opera-
ting elements and "thin" ones composed of a uranyl uranate and
teflon mixture (see Table II) placed close to the uranium elements
have been compared. The shielding determined from the activity mea-
68 - 8 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
surements of the irradiated uranium elements by means of a scintil-
lation device with a sodium iodide crystal (activated tallium) tur-
ned out to equal f intern(HZ-1)=0.73? 0.02 and f intern(IIZ-2)=0.88?
0.02.
As mentioned above these magnitudes do not always give the de-
crease in critical load when we pass to a homogenous system. The
more so that the uranium thickness in this case is considerable
when compared to assemblies with lithium hydride moderators.
T a b 1eVI.
Core Composition of Assemblies with Zirconium Hydride Moderator.
HZ - I - 268 178--27.6 ~ 41 53
A
verage atomic weight of stainless steel in taken as 55.2.
Tab 1 e VII.
Critical Parameters of Assemblies with a Zirconium Hydride
Moderator (without shielding factor).
Assembly Nucleus density x-10 201/cmI
index II Zr U-235 Al Steelx)
HZ - 2 268 168 10.3 51 2?
Parameters
Assembl composition
H pPZ36 12.4 24.8
Volume percentage of Zr Hydride (%) 51.0 51.0
Volume percentage of U (?%) 4.9 2.4
Critical parameters without corrections
a critical load (k o -2 8.88 5.83
b critical volume (1) 11.1 14.4
CorrectionsX)
1. Reflector lattice distortion by emergency chan-
nels.
a AGO, k of U-235) -0.04 -0.05
b~ a v 1) -0.05 -0.13
2. Presence of Steel in Core and Reflector
a a GC, k of U-235) -0.71 -0.50
b A VC (15 -0.89 -1.22
3. Presence of Aluminium in Core and Reflector
a a G1, k of U-235) +0.34 +0.20
b 4 VC~, (1) +0.42 +0.119
Critical Parameters corrected
a cri ica1 load tk off -23 ) 8.47 5.48
b cr it i cal volume 1 10.6 13.5
x The sign "+" indicates the increase of critical load or vo-
lume ,fter introduction of correction.
r, c 9 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
NEUTRON SPECTRUM ENERGY MEASUREMENTS IN ASSEMBLIES
WITH A ZIRCONIUM HYDRIDE MODERATOR.
Experimental study of neutron spectra in multiplying media at
different moderator-fuel ratios is of considerable importance both
to the moderation and diffusion theory and to reactor physics.
Comparison of calculated and experimental results is of great
help in many respects: we may improve models and methods of calcu-
lation, correct systems of constants for multigroup calculation
and have a deeper insight into the physics of processes going on
in a nuclear reactor.
As a rule the most interesting energy region is the thermali-
zed neutron region, where various effects of chemical bonds of mo-
derator atoms can be observed in the neutron spectrum behaviour.
In nuclear reactors with hydrogenous moderators a comparative-
ly soft neutron spectrum is obtained, even with small nuclear con-
centration of hydrogen and fissile material. Therefore the great
interest for the formation of the neutron spectrum of various
hardness in reactors containing a zirconium hydride moderator is
quite Justified.
A characteristic feature of zirconium hydride as moderator is.
the fact that the bond between the proton in the lattice and the
zirconium atoms is rather rigid, the '?tion energy being 0.13ev.
That is why the neutron, when moderated, loses its energy by 0.13ev
portions and why the energy loss of a neutron with the original
energy less than O.'13ev is being impeded and is rather ineffective.
In consequence of this behaviour of the neutron in zirconium
hydride there is a considerable deviation in the neutron density
distribution from the Maxwell di-tribution at small absorption
amounts per hydrogen atom.
A mechanical selector operating with the critical assembly
can be successfully used to investigate neutron spectra of highly
enriched systems with a broad energy range. The application of a
selector technique as part of a critical assembly has a number of ad-
C b
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
vantages when compared to a subcritical assembly in the column of
a powerful reactor. It is very convenient; that information on the
neutron spectrum is obtained together with other experimental mate-
riall that the experiments is conducted in conditions close to the
real ones, that there is no need of an external neutron source,
etc. The fast chopper of the PF-4 physical assembly enables us to
measure neutron spectra with a broad neutron energy range. Pig. It
gives the experimental set-up on neutron spectra measurements of
the critical assembly under investigation.
A beam of neutrons is passed from the investigated part of the
assembly by means of a rectangular channel through a slit collima-
tor to a mechanical chopper. The cross section of the channel clo-
se to the luminous surface is 50mm x 50mm. The rotor of the chop-
per is a hetinax disc 240mm in diameter in a casing of stainless
steel. The four slits of the chopper have a spindle-shaped form
(a 2mm slit width at the input and output and 4mm in the central
part of the rotor). There are two slit systems in the rotor that
are set under an angle of 900 to each other. This last condition
makes it possible to get four neutron impulses per one revolution
of the rotor, the recycling condition being fullfilled for the
flight base up to 11m. The rotor is designed for a rotation speed
up to 12000rpm, with a maximum flight base of the selector of 9m.
On the rotor surface at certain intervals from the slits there are
0.1 mm grooves containing magnetic material. When rotating the
magnetic reading device produce short electric pulses that are
used for starting the time analyser until a neutron impulse
enabling the background counting measurements appears. The backgro-
und counting is measured as a function of the angle of turning of
the chopper. This can prove important for a correct estimate of
the background when a sufficiently, intensive, hard neutron compo-
nent is present in the beam.
To reach a satisfactory statistical exactness in spectrum
measuring a comparatively high power level of the critical assembly
is required. This is connected with fuel activation. An increased
87u
- 11 -
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
Activity of the core stacks may sometimes hamper the setting up
of other experiments. The application of sufficiently effective
neutron detector and the increase of the transmission of
the selector lead to a decrease of the activation effect of the
core stacks.
He3 at a pressure of 18 atm, similar to the one described in [5]
serves as neutron detector.
The spherical casing of the chamber (100mm in diameter with
a well thickness of 1mm) is made of steel, the inner electrode ha-
ving a 5mm' diameter.
The working voltage -2.5kv. Amplified signals went after
discrimination into a 256 channel time analyser. Thanks to
the shielding and collimator system of the selector a rather good
ratio of the measured effect to the background was obtained. By
means of this mechanical selector neutron spectrum measurements
of the two assemblies with zirconium hydride moderators have been
accomplished.
In both assemblies the beam comes from the central region,
the luminous surface (50mm x 50mm) containing six elementary cells
of core stacks. This brings about the averaging of the neutron
spectra per cell. The axis of the beam channel coincides with the
centre of the core.
Fig-5 includes the experimental curves of the neutron flux
vs. energy, the curve reference being chosen arbitraril, When
the spectrum of the device was being treated a number of correcti-
on was made: corrections for the pervicusnessfunction, for the
detector effectiveness, the resolution,the distortion of the spec-
trum by absorption in air and in the detector wall, the neutron
flux scalar gradient and for the flux perturbation by the beam
output cavity.
The neutron scattering effect in air that took place in the
part of the tube close to the luminous surface was calculated se-
parat ely. ; 7 6
- 12 -
In the assembly investigated an ionizing chamber filled with
Approved For Release 2009/08/26: CIA-RDP88-00904R000100110047-8
 Approved For
The selector resolution amounted to M1 ~,mmae Additional
measurements of the neutron spectrum at decreased rotation speed
of the rotor was accomplished in assembly HZ-2 with the view of a
more detailed insight into the flux flow in the low-energy region.
The significant statistical error of experimental results is
to be explained by the insignificant effect and background differe-
nce in this neutron energy region.
From the given flux distributions it is to be concluded that
neutron density cannot be described in terms of the Maxwell dis-
tribut ion.
The comparatively narrow maximum flux close to the energy re-
gion of 0.13ev is significant. The high absorption per hydrogen
atom (-- 50b in HZ-1 and 25b in HZ-2x) gives rise to a strong
display of the effect of chemical bonds. The comparatively fast
decrease of the neutron flux with energies lower than 0.10ev is
another characteristic feature of the results quoted.
It should be noted that the experimentally measured values
of the subcadmium and supercadmium fissions of U235(CdRU_ 215-1 )
are relatively high - 0.8 and 1.8 for assemblies HZ-1 and HZ-2
respectively.
Nevertheless as mentioned before neutron fluxes decrease
fast with energy at