-
Notifications
You must be signed in to change notification settings - Fork 10
/
ORNL-CF-59-12-64.txt
2065 lines (1235 loc) · 45.3 KB
/
ORNL-CF-59-12-64.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
ré 7&3&.
&
S
e ‘. ', __: ..i\
c 7t
DATE:
SUBJECT:
T0:
FROM:
OAK RIDGE NATIONAL LABORATORY
EXTERNAL TRANSMITTAL
W= CENTRAL FILES
~ Oak Ridge, Tenne.s see : 59_12_61;
‘ | (Revised)
Operated by . AUTHORIZED
UNION CARBIDE NUCLEAR COMPANY
Division of Union Carbide Corporation _ 0 R N l
NUMBER
Jenuary 12, 1960 | COPY NO. 7
,Mbltefi-Salt Breeder Reactors
Distribution
H. G. MacFherson
iflbstract
' The problems involved in building e molten-salt thermal-breeder
reactor are reviewed, and it is concluded that the most feasible
construction iz an externally-cooled reactor with the fuel salt
paesing through the reactor core in graphite tubes. A reactor
with 15% of the core volume occupied by fuel salt and 5% occupied
by fertile salt would have & net breeding ratio of about 1.06.
The specific power ie about 3.0 Mw(th) per kg of U-233, U-235,
and Pe in the entire reactor and chemical processing system. The
resulting doubling time is 13 full-power years. The cost of the
fuel eycle for & 1000-Mw(E) station with this breeding performance
ig estimated to be 1.2 mills/kwhr. The performance in terms of
‘material utilization is an output of 1.18 Mw(E) per kg of U-233,
' U-235 and Pa, and 3.2 Mw(E) per metric ton of thorium. The latter
| figure could be increased by & factor of two at & sacrifice of
0.01 in breeding ratio.
NOTICE .
This document contains information of ‘a preliminary nature
and was prepared primarily for internal use at the Oak Ridge
National Laboratory. It is subject to revision or correction
and therefore does not represent a final report. The information
is not to be abstracted, reprinted or otherwise given public
dissemination without the approval of the ORNL patent branch,
Legal and information Control Department.
?
—
¥
MOLTEN-SALT BREEDER REACTORS
The purpose of this memo 18 to examine and swmsarize the status of the
molten-selt reactor as meeting the requirements of a breeder with a
doubling time of not more than 25 years., . Included are a discussion of:
1, The practicebility of different types of breeder reactor
construction.
2. The power density attainable in the fuel selt.
3. The status and cost of the required chemicel processing schenme.
4. The breeding gain, specific pover, and doubling time coneistent
with reasonsble assumptions concerning items (1), (2), ana (3).
5. The feasibility end cost of molten-aalt‘reactors.
The.contents of this memo have not been subjected to analysis by the Thermal-
Breeder Eveluation group. Their work will be largely independent of this, and
their results, vhen available, will take precedence over the numbers used in
this memo. _
I. Reactor Construction
Three'types'bf'breeder regctor construction are discussed: the unit-fuel-
tube construction, the graphite-core-shell construction, and an internally-
cooled construction.
Unit Fuel Tube Construction - The type of construction that is believed to
be most practical at present for a molten-salt breeder reactor is one In
which the fuel salt passes through the reactor in graphite tubes. Graphite
moderator is massed outside of the fuel tubes in the core region of the
reactor, and blanket salt containing thorium surrounds the core. The blanket
galt also passes through small passages in the moderator graphite and cools
it. Fig. 1 (ORNL-IR-Dwg. 42242) gives a schematic representation of one edge
of such & reactor, showing a single fuel tube, one of many. Although Fig. 1
shows a re-entrant graphite tube with both inlet and outlet at the bottom end
to avoid problems of differential thermal expansion, it may also be possible
to use a construction in which the qul tubes go straight through the reactor.
The fuel tubes would be manufactured from fine-gxained extruded graphite
rendered impervious by one of a number of treatments‘available. This type.
of graphite has been shown to be the most impervious to molten salts; one
E
w9
o
&
+@ent of salt by volume when pressures of up to 150 psi are applied; in fact,
. blenket salt will be maintained under slight pressure with respect to the
such grade hae been used in contact with flowing salt streams for a year
with no evidence of attack or bulk penetration by the salt. Separate tests
have indicated that such a grade of graphite will soak up less than one per- |
one grade picked up less than 0.2% by volume of salt. Tubes 3-3/4 #n. ID x
5 in. OD are on order and will be tested within a few months.
The moderator graphite will be in the blanket salt environment, end the
fuel salt so that any leskage that develops will be from blanket salt to
fuel salt. Ieakage can be tolerated provided it is at a rate that 1s small
compared to the rate of chemicel processing of the fuel salt. The moderator
graphite could also be made from fine-grained extruded graphite to keep pick-
up of salt in it a¥'es low & level as possible. By confining the fuel to
tubes end pressurizing the blanket salt with respect to the fuel ealt, fis-
sloning within graphite will be kept to & minimum. As a result there is
little reason to expect buildup of fission-product poisons in the graphite.
In the re-entrant fuel-tube construction, two metal-to-graphite connections
are necessary. The connection to the central graphite tube need only be a
mechanicelly sound connection, sitwh as & slip fit, since a small leakage
here would only bypass s small amount of fuel from going through the reactor
core. The connection of the outer fuel tube to the metal wall of the reactor
should be reasonably tight, with leakage small relstive to chemical processing
rate of the fuel. The three possibilities for this joint are a flanged joint
- with a mechanical pressure seal, a frozen-salt plug seal, and a brazed metal-
to-graphite tube Jjunction. Babecock and Wilcox have experimented with pressure
type flanged Joints with some succees, and it is presumed that this will be a
feasible solution to the problem. The testing of the freeze plug technigue
is under way at Ozk Ridge, and early indications are that it will be possible
t0o braze graphite to INOR-8, probably by the use of pure molybdenum as en in-
termediate material to provide a match to the thermal expansion coefficient
to the graphite. _
Graphite Core Shell Construction - A simple construction for & small twa-
‘region reactor with a graphite core shell is shown in Fig. 2 (ORNL~-IR-Dwg.
37258). As ghown in the drawing, the core is made from three large blocks
of graphite, a top header, & bottom header, and a center section. The
diameter of the core is approximated 54 in., and the height of the center
section would be about 40 in. It is proposed that these graphite parts be
made from large-size molded-graphite blocks, that the blocks be rough machined
to shape, and that they then be impregnated and tmmted to make them nearly
impermeable to molten salts. It is possible that final machining on the in-
ternal parts would be done before treatment and the parts clamped together
during treatment to cement the headers on and yield a monolithie block con-
_ struction.
o
L
This monolithic construction is an altérnate to having the three graphite
blocks as separate pieces, clamped together and held in place by springs.
The monolithic structure is considered more desirable, but makes the
impervious treatment more difficult. The pressure contact should be sat-
isfactory, at least for the initiel reactor operation, on the basis of
"Babecock and Wilcox work. It is possible that distortions produced by
shrinkege accompanying radiation damage would reduce the effectiveness
of the seals. A greater worry is the effect of shrinkage on the internal
portion of the core. If trouble were encountered here, the interior of
the core could be made up of smaller graphite pieces, such as sticks of
extruded graphite. - -
B . .
Grephite has been made in larger sizes than called for in this reactor,
but not of the small grain size required to render it impervious. It is
believed that sultable material has been produced in diemeter of 39-1/2 in.
and in thicknessges of up to 20 in. Samples of this graphite are now being
procured, end teste of their penetrability by molten salts will be completed
" during FY 1960. If this graphite’appears suitable, it is believed that it
can be made in larger sizes, up to 5 £t in diameter, but at considerable cost
in production equipment and in development expense. The development of the
larger block graphite for molten-salt reactors is not now planned, but would
~ be a part of the cost of the first full-scale breeder reactor of this type.
(There is a possibility that the development of such graphite might be under-
teken by the Defense Department for other purposes before this time.)
Thies design of reactor requires a seal between the '¥raphite header and the
INOR-8 pipe passing into the blanket vessel. This joint need not be a her-
metic one, but should limit the leakage of blanket salt into the fuel to
some small fraction of the core processing rate. The problems of this joint
ere the same as those involved in the fuel tube construction.
The permissible pickup of molten salt by the graphite depends on the rates
of diffusion of uranium into the salt that is in the graphite and of the
fission products out of the graphite into the main fuel stream. Some infor-
mation on this subject will be determined in FY 1960 in the capsule experiments
&t the MTR. In the meantime, a reasonable assumption is that the diffusion of
fission products out balances the diffusion of them into the grephite. For a
poison effect of one percent, the pickup of fuel salt into the grephite should
‘be less than one or two volume percent, depending on the volume fraction of -
fuel passages in the core. If the large molded graphite turns out to have a
plckup of less than one percent, then it should be suitable for use as the bulk
- of the core graphite. If, on the other hand, it picks up more than 2% -galt
by volume, it would be preferable to use a hollow core shell with an interior
- construction of extruded graphiterstickSa. As previously indicated, such ex- .
‘truded grades have been shown to pick up a satisfactorily low level of salt.
-I!
i‘-n--" ‘
oY
.0!
Internslly-Cooled Reactor - Various designs of internslly-cooled molten-
salt reactors have been suggested. One of the simplest is shown in concept
in Fig. 3. In this concept, the fuel is conteined in graphite tubes about
0.54n.ID x 0.7 in. OD that extend through the moderator, well into the
blenket region. The tubes are connected at each end by brazed joints to
e metal header system so that the fuel can be circulated slowly to keep it
uniform, to remove gaseous fission products, and to allow fuel concentration
adjustment as burnup proceeds. Presumably the tubes would have graphite in-
serte forcing the fuel to the periphery of the tubes in the core region and
occupying most of the internal volume of the tubes in the blanket region.
The heat generasted in the fuel would be transferred through the tube wall
to the blanket salt which is used a8 & coolant. This would probably limit
the heat generation to perhaps 50 kw per tube averaged over the reector,
and would therefore require 10,000 tubes for a reactor delivering 200 elec-
trical megawatts of power.
Although no brazed joint that is completely satisfactory in terms of com-
patibility with the salt has yet been demonstrated, there is little doubt
but that such & joint will be demonstrated during this fiscal year. The
use of molybdenum as en intermediate nipple connection has been demonstrated,
end brazing materials that wet graphite and ere compatible with the salt have
been found. Thus, in all probability, there will be no single technical
element of infeasibility remaining by the end of this fiscal year for an
internally-cooled reactor. Nevertheless, the concept of 10,000 tubes all
maintaining their integrity during a long reactor lifetime is not very attrac-
tive, at least to this writer. The adventage hoped for with internal cooling
ie a greater specific power, but it is doubtful if the internally-cooled
reactor cen achleve more than a factor of two in specific power over an
externally-cooled molten-salt reactor.
Summary - The unit-fuel-tube construction seems to be & feasible configuration
Tor & breeder reactor. By the end of this fiscal year it should be poseible
to specify suiteble types of graphite for both Prel tubes and moderator, and
to specify a satisfactory end connection for the .tubes. The construction
avolds most of the possible problems involved in sosking of fuel into the ‘
grephite since the fuel contacts only & small portion of the moderator graphite.
Furthermore, it will use the type of graphite that is now deemed least likely
to soak up fuel salt.
The graphite-core shell construction reqpires graphite of & size and quality
~ that is not immediately avallable, and will probably not be available without
the expenditure of a few million dollars of development money. If this expendi-
ture were made, the reactor construction would have & good chance of success.
However, the earlier availability of the fuel tube construction makes it the
first choice., ,
&
F
6
The internally-cooled reactor has some attraction in terms of higher specific
power and-is made up of elemente that individually seem quite feseible. The
great complexity of the core and the probable inacceseibility of it for minor
repairs make it unattractive at this stage of the technology. '
The reference reactor for the remainder of this memo i1s then teken to be
of the fuel tube construction, with 15 wol % of the core occupied by fuel
salt, 5% by blanket salt, and the remaining 80% by graphite. If the fuel
tubes are 3-3/4 in. ID, the fuel tubes will be spaced on 8-5/8 in. centers
on & square &rray. |
II. Power Densitz
The nuclear celculation will yield the breeding ratio and the uranium con-
centration required in the salt to make the reactor eriticel. In a circu-
lating fuel reactor, the latter figure must be combined with the power that
can be extracted per unit volume of fuel salt to yield a gross figure for
specific power. The power obtainable per unit volume of salt cen be errived
at in two ways: one is a general approach that looks at the fundamentsl
factors involved,; and the other 1s to lay out specific designs end eee what
thelr volumes are, and how much power they take care of. We will first look
at the problem generally and then examine specific layouts that have been
proposed. - o |
A ressonable value of the power density in the fuel can be estimated from
the total length of piping required in the system. The length of piping
considered 1s that required to carry the fuel salt into end out of the
reactor, through the blanket, through header connections, through the heat
exchanger and the equivalent length of piping represented by the pump volute
and expansion k. An average fluid velocity will be assumed through this
riping, and a saije._t_e_mperature range between reactor entrance and exit. This |
information combined with the volumetric specific heat of the palt determines
‘the amount of heat transferred per unit volume of salt.
" The composition of the fuel salt will be sbout é27mole % IiF, 37 mole % BeFp,
and 0.3 mole % UFy. The volumetric heat ecapacity of this mixture is about
1»2S‘ea1/cce°c at reactor temperature, or 7705_Btu/eu £t-OF. The melting 1
point of the fuel is about 850°F, and a figure about 100°F above this should ‘
be used as the minimum bulk fluid temperature. However6 the fertile salt has \
a melting point of 9759F, go that this, rather than 950°F, will be taken as
the minimum temperature of the fuel salt returning to the reactor. The J
maximm temperature of the salt leaving the reactor should be limited by the
corrosion tolerance of the metal alloy system, and with present knowledge
Gy
thie is set as 1300°F since very few loqps have been run ag yet at a higher
temperature. This 1e probably also & practical limit as set by the creep
resistance of the alloy INOR-8. The temperature range of 975 to 1300°F 1is
3250F, but & value of 300°F will be used as & reasonsble limit, presumably
from 975 to 12759F. With this AT and heat capacity, each cubic foot of
fuel transporte 23,200 Btu of heat each time it makes the heat transfer circuit.
It 1s difficult to set & natural 1limit on the maximum average flow velocity
that can be allowed for the salt in traversing its circuit. The only known
limiting factors are the pressure drop developed by the pumps and the pro-
portion of power one wishes to expend in pumping. A figure .of 17 ft per
- second has been picked somevhat arbitrarily for an average fluid velocity,
with velocities of up to 20 ft per second in the external system and &
lover velocity inside the reasctor. The major Justification for selection
of this figure is that it ylelds reasonable pressure drops (v100 psi) with
reasonably sigzed heat exchanger tubes and other plumbing fixtures, and with
‘it pumping powers are low. This velocity is well below the maximum velocity
assumed in the reactor portion of sodium-cooled reactors. It should be
recognized ;howvever, that this fluld velocity 1s not derived from any basic
constants of pature, but it does seem a reasonable essumption on the besis
- of present experience.
The effective length of the plumbing circuit will depend on the requirements
- for maintenance and on the necessary allowances for thermsl strains. These,
of course, depend on the type of reactor layout. Here, we will consider
pipes coming from the reactor to & pump, the pump feeding into & heat exchanger,
and the exit of the heat exchanger going back to the reactor. There will be '
header connections joining a number of fuel tubes together to feed into each
punp end heat exchanger. It would seem possible to have a total circuit length
of 80 ft for a reasonably sized power reactor, broken down approximately as followss
sl
In reactor core = - . 10.0
‘Through blanket ST 5.0
End connections to reaetor . 15.0
Heat exchanger : 0 15.0
“Allowance for pump and | L
expansion tank = - Te5
Miscellaneous SR 2.5
Connecting pipes 25.0
o
(2
ok
If this is the length of the circuit, the fuel salt will traverse the circuit
in k.7 seconds and the heat transfer rate will be 4940 Btu/sec-cu ft or
5.2 Mw/cu £t of fuel, or 184 kw/liter of fuel.
The remote maintenance demonstration facility in the 9201-3 building in Y-12
provides one check on the length of piping required for & proper layout of a.
~ reactor system. In the facility as it stands,; with its pump, dummy reactor,
riping layout with flange connections and dummy heat exchanger, the total
- effective length of the fuel galt circult is about k2 £t. If a true breeder
resctor were installed (with ite blanket dictating & larger path), and if a
full-scele real heat exchanger were installed, the piping length would in-
crease by ebout 30 £t to a total of 72 ft. Thie system would then be main-
tainable, but its power would be limited by the pump and piping presently
installed to about 67 Mw thermal., It would probably be desirable to have
ebout 100 Mw (thermal) in each pump-heat exchanger circuit for this type
of layout, so that with the increased capacity system, 80 £t is probably a
practicel length for the circuit piping in a reactor maintainable by the
canyon type equipment installed in this facility.
Another type of layout that is currently in favor is that celling for top
~ maintenance, and in which the entire fuel salt circuit is contained in a
single large vessel. This type of system allows greater compactness.
B. W. Kinyon has analyzed severel cases in a memo (reproduced ag Appendix I),
in which credit was taken only for a 200°F AT in the fuel. The results of
hie study indicate that with the 2009F AT and a fuel velocity of 15 ft per
second in the piping, power densities of sbout 5 Mw/cu £t can be obtained.
The total pressure drop in thie system is &bout 115 pei, of which 105 is
across the heat exchanger. This enalysislends further credence to the
belief that ebout 5.2 Mw/cu. £t or 184 kw/liter can be obtained for a molten-
salt breeder reactor.
III. Chemical Processing
~ The ohemical processing scheme proposed for the core circult is as follow5°
A small eide stream of the fuel salt will be fluorinated to remove Ufls
the fluoride volatility process. The UFg will be burned in Hp to UF, and
- will be placed in the reserve storage of UF) for reactor feed. The molten~
~ salt carrier, with most of the fission producte, will be stored for decay
of radicactivity to a suitable level, and then processed by the HF dissolu-
tion process, recovering the IiF and BeF D9 end eliminating most of the fission
products. The IdF-Ber will be adJusted in composition and added to the reactor
core stream again. The blanket will be processed by the fluoride volatility
process to remove the UFg on & frequent basis to keep the uranium inventory in
the blanket low. The UFg is burned to UF),, end the UF), produced is added to
n
»
the reserve supply pending reactor feed or sale. The Pa does not come out
of the blanket salt, but returns with it to the reactor system. By keeping
the ureanium et a low level in the blanket, the buildup of fission-product
poisons i very small and the blanket ie reprocessed completely only a few
times dnring reactor life. _
In practice, the freqnency of processing of. the blanket and fuel will be
determined by an economic belance. This balence is not struck here, the
approach being to see what processing rates are reqnired to achieve certain
nuclear aims end to examine the cost of these rates. A 1000-Mw(E) station
vill have a heat output of about 2500 Mw(th). At 5.2 Mw/cu ft, the circu-
leating fuel volume will be about 480 cu ft of fuel salt. When the fuel
comprises about 15% of the volume of the core, the uranium concentration’:
in the fuel salt ie sbout 1.2 kg/eu ft, ylelding a total uranium content
of the circulating fuel system of 575 kg of uranium (fissionable). About
3 kg will be burned per dzy, so that 191 days is the burnup time for the
fuel. TFor the variable fission-product poison to be kept at one percent,
& ten percent burnup is allowed before reproceseing, so that the entire
480 cu ft of core must be reprocessed every 19.l1 days of full-power operation.
This requires a chemical plant with e capacity of 9200 cu ft/yr, if the same
rercentage load fector is assumed for both reactor and chemical plant.
A rough calculation indicates that the required blanket or fertile stream
volume 1s between 2000 and 3000 cu ft for & system of resctors yielding
1000 Mw(E). . This volume is calculated on the basis of adequate coverage
of the reactor cores. However, if it is desired to keep the Pa absorbtion
down to 0.005, corresponding to a loss of breeding ratio of 0.01, then there
will have to be gbout 310,000 kg of thorium in the blanket system. Since
the blanket salt contains about 50 kg of thorium per cubic foot, this re-
quires a blanket volume of about 6200 cu ft. Thus the blanket volume can
be arbitrarily set to yield the desired Pa 1osses, and for this analysis,
6200 cu ft and 0.01 Pa loss is assumed.
The frequency of chemical processing of the blanket is set (aside from
eccnomics) by the deslre to keep the uranium 1nventory low and by the desire
to keep the fission-product buildup in the blanket small enough so that com-
" plete reprocessing of the blanket will not be required frequently. A desir-
able goal is.to keep the U and Pa in the blanket down to 30% of the fuel
dreuit inventory.. Reprocessing in ebout a 20-day cycle is required to accom-
plish this, and there is little benefit to faster processing because the Pa
holdup 1is limiting.
Processing at this rate will keep the U in the blanket circuit to about
one-tenth that in the core eircuit, or about 60 kg, Since uranium has =
about fifty times the neutron cross section of thorium, and since there
"
»
«\
10
are about 300,000 kg of Th in the blanket, there will be ebout one percent
as many U absorptions as thorium absorptions in the blanket. Thus after
ebout %en burnups of the core, the fission product level in the blanket will
glve sbout & one percent poison there. At an 80% load factor, this would be
after ebout 6.5 yr. Thus, in a 20-yr life, the fission-product poison level
in. the blanket might rise to 3% poison, and complete reprocesesing of the
blanket salt should be considered at that time.
The estimate of costs for chemical processing of the fuel and blenket at
the above rates (9200 cu ft per yr for the fuel and 113,000 cu ft per yr
‘for the blenket salt) is based on e report by Weinrich and Associates to
ORNL on "Process Design and Estimated Coste of Chemical Plants for Procese-
ing Molten Salt Fuels”. The larger plant estimated by them had & capacity
of 10,000 cu £t per yr of fuel salt, which is about the size required here
for the fuel salt circult. For the fuel processing plent, Weinrich and
Associates estimate & cost of $3,455,000, plus about §1,500,000 of shared
fecilities with the reactor plant. Crude adjustments to these figures made
by Osk Ridge personnel revised them upward to & total of about g9,830,000.
A much cruder estimate has been made of the additional plant cost to provide
for the rapid fluorination of the blanket salt. This was made by assuming
that multiplying the cost of the portion of the plent involved in fluorina-
tdon of the core salt and UF), recovery by five would give & plant of eleven
times the capacity. On thig assumption, the complete chemical plant for
treatment of both core and blanket salts would cost sbout $18,000,000 for
the 1000-Mw(g) plent. At & 29% annual charge and en 80% load factor, the
cost of the chemical plant, together with its operation, would be about
0.75 mill/kwhr. -
The total inventory of uranium and protectinfum in the reactor system is
estimated as follows:
In resctor fuel 575 kg
In blanket 180 kg
In chemical processing 30 kg
In storage | , 60 kg
- Total Bh5'kg
The uranium inventory at §15/g is $12,680,000, or $12.70/kw. At 4% this
is 0.07 mill/kvhr, or at 124, 1t is 0.22 mill/kwhr. The blanket and core
salts, including thorium inventory, will cost sbout $25,000,000, or £25/kw..
At 144 per year and 80% load factor, this amounts to 0.5 mill/kwhr. With a
net breeding ratio of 1.06, there would be 52.5 kg of fissionable uranfum
”»
;]
&)
11
produced per year, which would yleld ebout £790,000/yr or ebout 0.11 mill/kwhr.
Thus the total fuel cycle cost would be about.l.a.mdlls/kwhr on the basgis of
present uranium use charges.
It is obvious that considerable savings in fuel cycle cost can be made by
sacrificing doubling time. It is probable that if a breeding ratio of 1.00
wvere satisfactory, at least half of the salt and thorium inventory charge
could be avoided and the chemical plant chargee would be considerably reduced,
probably by at least one-third. Thue, & hold-own breeder might have power
costs as much es 0.5 mill/kvhr less than the doubling reactor.
In this enalysie of chemical processing, only processes on which there is
a faeir amount of laboratory data have been considered. With the fluid blanket,
an easy means of removing Pa 18 being sought. If it is found, then the blanket
holdup will be reduced, and the thorium inventory cen be reduced appreciebly.
IV. Performance as & Breeder
A number of grephite-moderated molten-salt resctor configurations have been
subjected to multigroup nuclear calculations with the Cornpone and Sorghum
codes.* The eriticality calculations cen be correlated quite well if one
plote the concentration of uranium in the core against the carbon absorptions
in the core. This is done in Fig. 4 (ORNL-IR-Dwg. 42240)., The plot shown
comes from reactors of equivalent spherical core diemeters of 3 ft, L ft,
5 £t and 14 ft. The fuels in the core have thorium concentration of 1 mole 4
ThF),. % mole % ThF), 7 mole % ThFy, and 13 mole % ThF,, and the volume frac-
tion’of fuel represented in the various calculations of 10%, 12.5%, 15%,
18.3% end 20%. It includes calculations of both the initial state of the
reactor with pure U-233 and the state achieved after 20 yr of operation with
e near equilibrium mixture of U-255 and U-235. Although the relationship is
not mathematical, there is a good empirical fit and the curve can be used
with fair confidence in predicting the uranium toncentration required in the
fuel salt. For the reference system with & 15 vol % fuel fraction in the
core and 0.04 neutron absorption in carbon, the concentration of fissionable
uranium required in the fuel salt ie 1.2 kg per cu ft. |
This figure, combined with the mumber of 5.2 Mv per- cu £t developed in
Section II sbove, ylelds a specific power in the fuel stream of L4.33 Mw/kg.
¥ 'MSR Quar. Prog. fipts._, om 268k, ORNL 2723, ORNL-2799-
3
»
@
12
The chemical processing and blanket holdupe of fuel lead to a total fission-
able uranium inventory of 845 kg, as described in Section III, so that the
over-all specific power is 2.96 Mw(th)/kg.
The effective value of eta for the fuel mixture will depend on the thermality
of the spectrum, which is related slso to the carbon gbsorptions per sbsorption
in fissionable urenium. At 0.0k ebsorption in carbon, the value of eta for the
isotope mixture is sbout 2.22, based on & thermal velue of 2.28 for U-233.
From the nucleer celeulations cited above, one can correlate the neutron absorp-
tions in the carrier salt in the core with the absorptions in carbon. This is
done in Fig. 5 (ORNL-IR-Dwg. 42239). The volume fraction of carrier salt is 20%,
compriced of 15% for the fuel and 5% for the fertile stream, so that the absorp-
tions in the carrier salt in the core are 0.04 for carbon sbsorption of 0.0k,
As described in the preceding section, the reference chemical processing plant
provides for keeping the variable fission-product polson fraction down to 0.01,
and the Pa losses (2 x absorptions) down to 0.0l. Uranium-236 will, of course,
build up from radietive captures in U-235. With a breeding ratio of 1.06, the
removal of U-236 by the salé of excess fuel will approximately equal the removal
by neutron absorptions, go that the U-236 poison will be approximately 0.0l. The
~ neutron losses to saturable non-volatile Pission products will be about 0.006, and
1f Xe-135 losses can be kept to 0.00k, the total saturable fission-product losses
will be limited to 0.0l. To keep the Xe-135 losses to 0.004 requires its removal
on a time eycle of about 6 minutes. The off-gas system can be designed to accom-
plish this by bypassing 2% of the pump flow through & degasser. This was the
degassing bypass flow in the ART pump.
Other neutron losses are estimated to be 0.03 in the blanket salt, 0.003 for
- delayed neutrons, and 0.002 for leakage. Considering the efficiency of the
fluoride volatility process, 0.005 may be adequate for chemicel processing
losses. Fission-product pickup by the graphite, assuming that the fuel tubes
soak up-<:0 5% by volume of fuel, givee a negligible loss unless there is pre-
ferential fission-product absorptionu
The total neutron losses now add up to about 0.16, which subtracted from an
ete of 2.22 yields & net breeding ratio of 1.06. It should be noted that
higher breeding ratios can be obtained by decreasing the volume fraction of
fuel in the core and by increasing the uranium-to-carbon ratio in the core.
However, these both lead to higher uranium inventories and consequently no
great improvement, if any, in doubling time. Furthermore, if thermal eta for
U-233 i 2.29, as is belleved in Osk Ridge, instead of the 2.28 assumed, the
breeding ratio is improved by nearly one point. With an over-all breeding
ratio of 1,06, ‘the doubling time is gbout 13 yr of full-power operation.
"V,_;FEasibiIity and Cost of Molten-Salt Reactors
The basic feasibility of molten-sslt reactors has been discussed in a section
of the book "Fluid Fuel Reactors®., This and later information have been
»
)
15
reviewed by the Fluid Fuel Reactors Task Force, and it was the concensus of the
group that, with minor exceptions, the feasibility of the molten-salt reactor
was established as far as materiels compatibility and handling ie concerned.
These exceptions concern the pickup of fuel salt by the graphite and the possible
.precipitation of UO> by gasee adsorbed on the graphite. Since the time the Fluld
Fuels Tesk Force met, the results of a one-year circulating sslt loop containing
grephite and of graphite impregnation studies bave shown that the graphite is
steble in contact with the salt, and that there are varieties of graphite that
will soak up less than 0.2% by volume of fuel salt, It has also been found possi-
ble to prevent UO2 precipitation by pretreatment of the graphite.
The Fluid Fuel Reactors Task Force further expressed doubt as to the economic
maintenance of fluid fuel reactors in general. For the molten-salt reactor,
this can be answered finaslly only in & reactor experiment, which has been pro-
posed. In the meantime, good progress has been made in devising suitable main-
tenance procedures for one type of reactor construction.
Since the time of the tasgk force, the design effort on molten-salt reactors has
been directed toward breeder resctors that take advantage of the compactness
that is possible as a result of the high temperature and good heat transfer pro-
perties of molten salts. Most of the designs developed have a compact primary
system, such as that described in Appendix I and illustrated in Fig. 6. In
these designs, the entire primary fuel circuit is contained inside a reactor
vessel, With this construction, a parallel comparison with solid-fuel-element
reectors is evident, in which the tubes of the primary heat exchangers of the
salt reactor are compared to the fuel tubes in the core of a solid-fuel-element
reactor; both contain fuel, both constitute the primary heat exchanger surface,
and in each case they are contained within the primary reactor vessel enclosure.
In a gimilar way, the pumps for circulating the fuel are compared to the control
rod mechanisms (and fuel shuffling mechaniem for the fast reactor). Both involve -
moving parts inside the reactor enclosure, and the pump, though bulky, is cer-
tainly simpler. Maintenance of pump and heat exchanger in the salt system is
by overhead withdrawl and replacement, and the operations required are comparable
to those required for the replacement of core sssemblies and repair of control
mechanisms in the reactors with solid fuel elements, particularly those cooled
with sodium: There is thus no reason to expect maintenance costs for the MSR
to be higher than, say, for the fast reactor.
As for capital costs, the higher temperature of the heat source and the very -
high heat capacity per unit volume of the salt (approximately k.4 times that
of sodium) lead to compactness of the entire system. The following table com-
pares pertinent factors of complexity and heat transfer with four reactors
using & sodium coolant. Comparing the MSR primary heat exchanger with the
reactor cores, it is simpler by virtue of having fewer tubes, and has about
the same efficiency as the fast reactore in terms of surface area.
The avoidance of an intermediate heat exchafiger for fhe'MBR, possible because
there is no violent water reaction and because the induced radiocactivity is
very short lived; 1s a further factor reducing capital cost. In the steam
»
"
5
14
- generator portion, the higher temperature of the salt coolant gives an -
edvantage by about a factor of two.
The lest two rowe of the table indicate how the high heat capacity of the
salt, even using a conegervative bulk AT, can reduce the pump capacity end
eystem piping requirements by at least a factor of two.
On the basis of this analysis, even after allowing for the high cost of INOR-8
and of the salt coolant, the ecapital costs of a molten-salt reactor should be
less than for the sodium-cooled reactors.
Sodiuwmn-Cooled Reactors
' Advanced
MSR Hellam P/604* Fermi Fast Reactor
Net electrical Mv assumed
for reactor 333 80 205 ok 283
Fuel tubes per MwE 58 28 1%0 264
Primary heat exchsnger -
tubes per MwE 19
Primary heat transfer 22 111 38 14.6 25
surface per MVE (fuel
tubes in case of sodium-
cooled reactors) sq ft
per MwE
Intermediate heat exchanger
sq £t per MwE - | 935 170 160 g2
Steam generator, super- 120 237 -:214 345 200
" heater and reheater
.surface gq ft per MWE
- Coolant flow data (avg) L | o
Bulk AT assumed . -~ 150°F 338°F 275°F 250°F 350°F
“gal/min flow per MwE 106 2k 263 318 214
* 8. Ievy et al, "Advanced Design of a Sodium-Cooled Thermal Reactor for
Power Generation", 1958 Geneva Conference Paper P/60L.
o
b
»
15
The moet careful cost estimates of molten-salt reactor construction have
been made by G. D. Whitman and ere included in: (1) ORKRL 2634; (2) ORNL-
CF-59-1-26; and (3) ORNL 2796. The three cases include two power reactors
of 640- and 860-Mw (thermal) caepacity, and an experimental reactor of 30-Mw
capacity. A reasonable extrapolation of these costs to the 2500-Mv (thermal)
etation required for 1000 Mw(E) ylelds capital costs of from $170 to $200
per kw. This estimate 1s for & first plant, but does not include development
coets. When these capitel coste are combined with the fuel eycle costes esti-
mated in Section III of 1.2 Mw/kvhr and & reasonable operation and mainten-
ance estimate of 1 mill/kwhr, one gets & power cost in the neighborhood of
6 mills/kxwhr for the first such large breeder reasctor plant. Presumably one
could expect lower coste than thies as a result of prior prototype reactor
construction and operation. It is difficult to attempt to prediect ultimate
costs, however, until experience ‘has been had with at least an experimental
reactor. .
»
16
APPENDIX I.
INTRA-LABORATORY CORRESPONDENCE
Oak Ridge National Iaboratory
October 22, 1959
To: H. G. MacPherson
ce: L. G. Alexander
J. W, Miller
File (BWK)
Subject: Volumes and Pressure Drops for Molten Salt Breeder Reactors
The following table is a comparison of two reactor sizes, each with two flow
velocities in the external piping.
The heet exchanger has been designed on the basis of 0.300 in. inside diasmeter
x 0.035 in. wall tubing in a 45° helix, with edjacent coils wound opposite hand.
Fuel temperatures are taken as 1275 and 1075°F; coolant temperatures as 1150
and 10000F. The use of 1/4 in. ID x 5/16 in. OD tubing would decrease the heat
exchanger length by 21%, increase the number of tubes by 50%, and inereasge the.
diameter by about 15%. The fuel volume external to the core would be reduced
by about 10%, which might overweigh the problems introduced by the other changes.
The fuel volume might be reduced by considering the entire flow in the center of
the heat exchanger as "pump suction" and using & higher flow velocity. This would
be about 10% of the fuel outside the core for the higher flow rate cases. :
The attached sketch (Fig. 6) is approxihately to scale for the smaller reactor
with 20 ft per second fuel -veloéity in thé piping.
/e/ B. W. Kinyon
:nh S
Enclosure
Qi
17
VOLUMES AND PRESSURE DROPS FOR MOLTEN SALT BREEDER REACTORS
Reactor Power, MWwE (net)
Station Efficiency,%
Reactor Power, MwT
Blanket Power, %
Core Power, MwT
Fuel Tempersture Exit, °F o
Fuel Tempersature Entrance, F
AT in Fuel, °F 5
Volume Flow of Fuel, ft”/sec
Flow Velocity,in Core, ft/sec
Flow Area, ft
Volume Fraction in Core
Core Cross Section, ££2
Core Diameter, ft
Diameter of Equ%valent-Sphere, £t
Core Volume, ft
Fuel Volume in Core, £t
Blanket Thickness, ft
Blanket Volume, £t2