-
Notifications
You must be signed in to change notification settings - Fork 10
/
ORNL-4831.txt
5523 lines (4188 loc) · 182 KB
/
ORNL-4831.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
CRTTTAAY cT T A e} T AR QRML-4837
,;- Rl T s s S, b . A
Oax ¥ AT LR L JBRA
i lflfffl M!flfflillifflllfl i
3 445k 00PLELE U
17
“q
DEVELOPMENT OF THE
VARIABLE.GAP TECUHNIQUE FOR
MEASURING THE THERMAL COMDUCTIVITY
OF FLUORIDE SALT MIXTURES
J.W. Coaks
Printed in the United States of America. Available from
National Technical Information Service
U.S. Department of Commerce
5235 Port Royal Road, Springfield, Virginia 22151
Price: Printed Copy $3.00; Microfiche $0.95
This report was prepared as an account of work sponsored by the United
States Government. Neither the United States nor the United States Atomic
Energy Commission, nor any of their amployees, nor any of their contractors,
subcontractors, or their employees, makes any warranty, express or implied, or
assumes any legal liability or responsibility for the accuracy, completeness or
usefulness of any information, apparatus, product or process disclosed, or
represents that its use would not infringe privately owned rights.
ORNI~L4831
UC-80 — Reactor Technology
Contract No. W-740S5-eng 26
Reactor Division
DEVELOPMENT OF THE VARIABLE-GA? TECHNIQUE FOR MEASURING
THE THERMAT, CONDUCTIVITY OF FLUORIDE SATLT MIXTURES
J. W. Cooke
FEBRUARY 1973
OAK RIDGE NATIONAL TABORATORY
Oak Ridge, Tennessee 37830
operated oy
UNTON CARBIDE CORPORATICN
for the
U.3. ATOMIC ENERGY COMMISSION
v
3 4456 002L5L2 O
iii
CONTENTS
NOMENCLATURE s 2 « & s & 8 4 4 s s a4 s s e s e s e s e e s aeoe s v
ABSTRACT + v ¢ v v v v v ¢ o & o o o o o »
1. INTRODUCTION . v & v 6 ¢ o a0 o v v o o o o o o o o o s o o o 3
2. METHODS FOR MEASURING MOLTEN-SATT THERMAL CONDUCTIVITIES
P
o
Selected Techniques for Fluids . . . . . . .
Transient hot wir€ . . .+ « ¢ ¢ v o v v o« v o o » s o o @
Transient hot foll ¢« ¢ v & 4 ¢ v v v o v v v o v« o o o o &
Necked~down sample technigue . . + . « « + + . .
Laminar heat f1ow .« + « 4 « o o v o« o o o o v v e e e e e
Parallel wall o o o & ¢ o s o o o o o o 5 o o o o o o o
N U1 U W W
The Variable-Gap Technique . . o o & + o o o o o & o « o
\
O,
W
Ge n@l‘&] deS c T' iptiorl L] L] » - - - . . . . . . . . . - . . 3
I(:j.ealized mOdel L] . . . ® L] . L] . -« L] . . . . . L L] . .
-3
Effect of radiaftion v v v v v v 6 v o o o o o o o o o o o 9
Effect of natural convectioln « + 2 o o o o o o o o o o o o & 12
Effect of heat shunting . . . . « « « v ¢ ¢« v « & « « . « . 16
Method of caleulalion .+ ¢ ¢ v v v v v 6 o ¢ o o o o o o o 19
FXPERIMENTAL APPARATUS . ¢ o v v v v v v v e s e v v o o o o o & 22
o
Thermal Conductivity Cell o + & o v v v ¢ v v ¢« v o o o o o o« & 22
TUYTIECE ¢ 6 v v 6 v o o e e o o o o o & o o o o o o o v v o o . 30
Hlectrical System o o o ¢ o o 4 o 4 o & o o + a o o o o« & o o 30
Instrumentation .+ . ¢ ¢ v ¢ o ¢ ¢ 4 6 i e e e e o o e s 8 s s e 32
T @ Ej}(PEB IIVEEINT_A.IJ PRO(JED ';S » s o ° . . » . . . . . . - . . . . . . 33
PN
Preliminary ProceduresS o o v o v 4 4 o o o s & s o o s o & o s 33
Operating Procedure — Fluld Specimen . . « « + « v « v o « « o « 34
Operating Procedure — Solid Specimen « « ¢« ¢ o ¢« ¢« ¢ o o &« o « o« 3
5. EXPERIMENTAL RESULTS ¢ o ¢ o o o ¢ o o o o o o o & o o o o o o « 3
Thermal Resistance CUIrvVES o &+ v ¢ o o o o o o o o o o 1 o o o 35
Thermal Conductivity . . « « v & v v v v o v v o v e « « « « « . ko
iv
6. DISCUSSION OF 'THE RESULTS
Comparison with Published Values . . .
Comparison with Theory . . . . .
Uncertainties in the Results .
Adequacy of the Experimental Apparatus
7. CONCLUSIONS
ACKNOWLEDGMENT'S
REFERENCES .
APPENDIX A. ADDITIONAT, DETATLS OF THE DESIGN OF THE APPARATUS .
APPENDIX B. EXPERMENTAT, DATA
APPENDIX C. PRECISION AND ERROR ANALYSTIS
X
NOMENCTLATURE
Specific heat, constant pressure, cal gt (ee)™
diameter of main heater assembly, cm
ratio of effective to total heater wire length
heat shunting factor
Guard heater factor
Acceleration due to gravity, cm sec””
Radiant heat-transfer coefficient, W em = (°C)™*
Current, amps
Thermal conductivity of fpecimen, k_ is that of the metal
cylindrical solid, W em~ (°C)7*
Mean index of refraction
Heat flux, W cm ?
Measured heat flux uncorrected for heat shunting, W cm™
Radius of cylindrical solid, cm
Temperature, °C; T, = °K
Temperature difference between upper and lower plates of
the conductivity cell, °C
. . . 1 2
Thermal resistance at zero specimen thickness, °C W~ cm
Heat~transfer coefficient, W em ™ (°¢)™*
voltage, volts
Integration variable
Specimen or gap thickness, cm
Radiative function defined in Eq. (8)
Coefficient of linear expansiom, (°C)7*
Coefficient of bulk expansion, (°C)7*
Emissivity of cylinder surface
mean absorption coefficient for radiant heat, cm™*
Viscosity, cP
Density, g em™°
Stefan-Boltzmann constant, W em™ (°K) ™%
Optical thickness
Electrical resistance, ohms
vi
Dimensionless Moduli
NGF Grashof number, gp®8 Al X3/gc u?
N, Prandtl number, Cpp/k
T 1A o .
NRa Rayleigh number, NGr NPr
N Reynolds number, oVD/u
Subscripts
f film
c critical
m metal
o at specimen thickness = Q
S specimen
X at specimen thickness = x
DEVELOPMENT OF THE VARIABLE~GAP TECHNIQUE FOR MEASURING
THE THERMAL CONDUCTIVITY OF FLUORIDE SALT MIXTURES
J. W. Cooke
ABSTRACT
The development and evaluation of the wvarigbhle-gap tech-
nique for measuring the thermal conductivity of molten fluoride
salts 1s described. A series of measurements were made of the
conductivities of several substances (Ar, He, H,0, Hg, and liquid
and solid heat-transfer salt) over & wide range of conductivities
[0.4 x 107 to 100 x 1072 W em™ (°c)™* ] and temperatures (40 to
950°C). The deviations of the results from published values
averaged iS%. The study demonstrates the accuracy and outstand-
ing versatility of the variable~gap technique.
Key words: thermwal conductivity, development, design,
measurement, fused salts, high temperature, MSER.
1. INTRODUCTION
High-temperature operations in the chemical processing and nuclear
industries have created a need for economical, efficient heat-transfer
media whose thermal propertieg are superior to those of organic and
gaseous coolants. Molten-salt mixtures have good heat-transfer prop-
erties compared with organic liquids, and their relative inertness and
low vapor pressure give them distinct advantages over 1liquid metals.
They are applied in high-temperature fluxes, heat-treatment baths, and
electrolytic fuel cells, and as the fuel carrier and coolant for nuclear
reactors, such as the Molten-Salt Breeder Reactor (MSBR) experiments.l
This report describes an experimental technique for determining a key
thermal property of molten salts — thermal conductivity. The study
evaluates the method over a broad temperature range using a variety of
materials representing a wide range of conductivities.
Very few data on the thermal conductivities of molten salts, in
varticular, fluoride salts, have been published. Most of the existing
measurements for molten fluoride salt mixtures were made by members of
the MSBR group at Oak Ridge National Laboratory (ORNL) over 15 years
ago.” To extend the scope of the previous measurements, we have developed
an ahsolute, variable-gap technique to determine the thermal conductivity
of fluoride salt mixtures in ligquid and solid states at temperatures to
1000°C. This technique is particularly well suited to the measurement of
the conductivity of low-conducting, semitransparent fluids that must be
contained in inert surroundings at elevated temperatures. Most other
applicable methods suffer from one or more deficiencies 1f used under
these conditidns.
The variable-gap technique is examined in detail) and other techniques
are discussed briefly. The development of an experimental apparatus is
described, and experimental results are presented for the conductivities
of several calibrating fluids: Ar, He, heat-transfer salt (HTS), Ha0, and
Hg. These fluids represent a very wide range of conductivities [0.4 X 107°
to’ 100 X 107° W’cmfl(°C)"1] which were measured over a large temperature
range (40 to 950°C). The thermal conductivities of molten flucride salts
will be presented in a separate report to be published.
2. METHODS FOR MEASURING MOLTEN-SALT THERMAL, CONDUCTIVITIES
Thermal conductivity is one of the most difficult of all thermo-
physical properties to determine experimentally. The difficulties
primarily are due to the unreliability of temperature measurements, the
inadequacy of thermal insulation, and the simultaneous transfer of heat
by mechanisms other than conduction. Conductivity can be measured by
either steady-state, quasi-steady-state, or transient-state heat flow
systems. Experimental determinations using the steady-state methods
depend upon the attainmment of suitable boundary conditions that will allow
the Taplace equation to be solved for the temperature distribution. The
conducltivity is then calculated from the Fourier heat-transfer equation.
The transient method requires the solution of the diffusion equation with
suitable initial and boundary conditions for the thermal diffusivity co-
efficient; the density and heat capacity must be known to calculate the
conductivity. The quasi-steady-state methods are based on a solution of
*.
KNOz -NaNOz ~NaNOg (44-49-7 mole %).
the diffusion eguation for unique initial and boundary conditions such
that the thermal conductivity can ve directly determined.
In addition to conduction, radiation also way be present with trans-
parent substances, and the experimental technique must be capable of
identifying and separating these two mechanisms. In the case of fluilds,
convection may also be present. Thus, only a limited number of experl-
mental techniques are available for the determination of the conductivity
of fluids. Several techniques found in published investigations are
degcribed briefly in the following section, and the technigue used in the
present gtudies is described in a later section.
selected Technigues for Flulds
We will describe briefly each technique and discuss 1its advant
a
and disadvantages for the measurement of conductivity of molten salts.
Transient hot wire
In this technigue, the rate of change of ftemperature of a line heat
source zltuated in an Infinite medium is used to determine the conduc-
tivity of the medium. The line heat source consists of a wire (1 to 5
mils diam) placed axially in a cylinder (2 to 4 cm diam) filled with the
specimen. The wire ig heated by a steady current, and its temperature
is determined by the change in its electrical resistivity. After an
initial transient heating period, the log temperature becomes a
linear function of time until natural convection begins to occur. The
sleope of this linear function of tewperature wilth time can be related
directly to the conductivity of the specimen, Thus, this technique is
a gquasi~steady-state technique rather than a transient technique. It is
a common technique for determining conductivity of ligquids and has been
5
described in many publications.”™*
Tts simplicity, quickness, precision, and accuracy make this tech-
nigue useful for most liquids. With molten salts, however, a significant
amount of current can be shunted through the szalt itself due to tThe
relatively high electrical conductivity of the molten salts at elevated
temperatures. Since the degree of current shunting 1s very difficult to
predict, molten salt conductivity results obtained by this technique are
subject to questions which have not been sufficiently resolved to make
it suitable for this application.g
Transient hot foil
This technique’’® is siailar to the transient hot-wire technique
except in two respects: (1) a thin foil is substituted for the fine wire
to provide a plane heat source instead of a line source, and (2) the tem-
perature is measured with a front-wave-shearing laser interferometer.
Because this temperature measuring technique is extremenly sensitive, the
heat flux from the foll can be greatly reduced. However, the fluid speci-
men must be transparent as well as compatible with the material used in
the cell window.
The transient hot-foil method is more difficult to apply than the
transient hot-wire technique. Tis primary advantage is in the reduction
in the molten-salt ionization that results from a lower voltage along the
heat source. Consequently, the interface between the heated foill and the
molten salt remains polarized and the flow of current into the salt is
minimized. In practice, however, other voltage potentials may exist
within the cell, and some current will flow into the salt even though the
surfaces are polarized. Moreover, operation at elevated temperatures
presents formidable problems 1in the design and choice of materials for the
cell windows., Diamond is the only transparent material suitavle for use
with molten fluoride salts at high temperatures, but its high cost and
fabrication difficulties would restrict its use to very small apertures.
Neckec-down sample techunique
This method® is based on the measurement of the steady-state change
in resistance caused by electrical heating of a narrow bridge of the
sample mater al which joins two larger bodies of the same material. The
theory describling this phenomenon shows that the change in resistance
expressed by the voltage drop caused by the heating current does not
depend on the detalled shape of the narrow region. A necessary condition,
if this is to be true, is that no significant flow of heal occur outside
of the boundaries of the sample. For liquids, the narrow bridge 1is
N
maintained by contalining the liquid in a vessel separated into two parts
by o thin wall with a small aperture. The material of the wall must have
both a high thermal and a high electrical resistance. The technique is
classified as a quasi-steady~-state method.
Several uncertainties are associated with the methed, the first of
which is the losz of heat by conduction along the thin separating wall.
A second major problem arises from the possibility of convection. More-
over, with molten salts the possibility of polarization effects also
would need to be considered. The important advantages of the technigue
are the simplicity of the apparatus, the rapidity with which the measure-
ments can be made, and the reduction in the uncertainties in radiation
by the small size of the heated region.
Taminar heat flow
As)
The laminar flow method determines the conductivity of a fluid
flowing in a circular tube under carefully defined conditions. The wall
temperature of the tube 1s maintained uniform while the inlel and outlet
temperatures of the fluid are measured. In this method, radiation losses
can be neglected and the troublesome measurement of heat flux eliminated.
The main problems concern the prediction of velocity and temperature
profiles of the fluid at the entrance section of the tube and maintaining
a uniform wall temperature., Furthermore, the assumption of constant
physical propertiez of the fluid over the temperature range can intro-
duce significant error. Most of the published results of this technique
differ from the accepted values for thermal conductivity because the
hydrodynamic and thermal entry lengths were not properly assessed. ”
Parallel wall
with this method, the steady heat flow through the specimen and the
temperature drop 2eross it are measured. ' 1% The specimen is contained
between two parallel walls of plane, cylindrical, or spherical geometry.
This method is the most commonly used technique for measuring thermal
conductivity. Its seimplicity with regard to the analytic model and
experimental setup make it most attractive; however, the uncertainties
caused by convection, radiation, and stray heat flow at high temperatures
can be considerable. Reducing the heat flow uncertainties by decreasing
P
the specimen thickness and temperature drop can lead to large errors in
-+
these two measurements as well as in the heat flow. Thus, although the
method is simple, it requires care in application and way be unsuitable
for low-thermal conductivity fluids at high temperatures. ?
The Variable~Gap Technique
The variable-gap technique for measuring conductivity is a signifi-
cant improvement over the parallel wall method in that it takes advantage
of the fluidity of the specimen. By use of this technique the specimen
thickness can be varied continuously during the operation with a minimum
disturbance to the specimen composition or to the system temperature dis-
tribution. Also by varying the specimen thickness, the undesirable effectis
of several factors, including the errors caused by specimen voidg or
inhomogeneities, natural convection, radiative heat transfer, corrosion,
deposit formation, radial heat flow, thermocouple location, and thermo-
couple drift, can be greatly reduced. Since only the change in the speci-
men thickness and the change in the temperature across the specimen is
measured, the potential errors of these measurements are smaller and Cthe
influence of convection, radiation, and heat losses can be detected and
minimized. In addition, the apparatus can be used with little or no modi-
fication to measure the conductivities of solids and gases as well as
liquids. Considering the advantages of the method, it is surprising that
- . . 1
only limited use has been made of the variable-gap technique.*®r14” °
General description
The experimental apparatus is shown schematically in Fig. 1. Heat
from the main heater travels downward through the liquid sample region
(labeled '"variable gap™" in the figure) to a heat sink. Heat flow in the
upward and radial directions is minimized by appropriately located guard
heaters, and the heat flux into the sample is measured by the voltage and
current of the dec power to the main heater. The temperature drop across
the gap is determined by thermocouples located on the axial center line
in the metal surfaces defining the sample region. The sample thickness
is varied by moving the assembly containing the main heater and is measured
ORNL- DWG 72-10523R
,LIQUID LEVEL
SPACE
CONTAINER
WAL LS
/= SN NP e Ty -
B\ J : d GUARD HEATERS
oj_._ [ooooooooolanooourf%”oT'—@/’f i
X
QDDOOO)DOOOO()DOO o
\\\\\ RN R\ " ;( N \ \\
~
-\:‘~-MA|N HEATER
..
\\‘ “SVARIABLE GAP
~ .
~. .
~ HEAT SINK
x = THERMOCOUPLE LOCATIONS
Fig. 1. ©Schematic drawing of a variable-gap thermal conductivity
cell.
by a precision dial indicator. The system femperature level is main-
tained by a surrounding zone-controlled furnace.
Tdealized model
The measured temperature difference can be resolved into the temper-
ature drop across the sample gap; the temperature drops in the metal walls
defining the test region; the temperature drops in any solid or gaseous
films adhering to the metal surfaces; and errors assocciated with thermo-
couple calibration, lead-wire inhomogeneities in thermal gradient regions,
and instrument malfunctions. Neglecting the error term, we can write
Al = AT+ AT+ AT, (1)
S m T
where subscripts are sample, metal, and surface film, respectlvely.
For the sample region, the temperature difference is
AT = (a/A) sx /%, (2)
where Q/A is the heat flux, Ox 1s the gap width, and k, is the thermal
3
conductivity of the liquid sample. It is assumed that no natural convec-
tion exists in the sample region.
Similarly, the temperature drop in the confining horizontal metal
walls can be written
IKQn - (Q/A) AXm/km ’ (3)
where Axm<is the heat-flow path length in the metal walls, and km is the
wall conductivity. The heat flux Q/A is the same as in Eq. (2), assuming
that no radial heat flow and no bypass heat flow through the side (verti-
cal) walls of the sample cup. Since km is a function of temperature,
Eq. (3) can ve written separately for the upper and lower metal walls;
however, for the purposes of this analysis, the two regions are combined.
The f£ilm temperature difference is of the same form as the Al''s given
in Fgs. (2) and (3). If surface films are present but of constant and
known thickness, Axf, during the experiment, there is no effect on the
derived sample conductivity or on the associated error. IlHowever, a film
that grows or decays in an unknown way during the course of the measure-
ment introduces an error 1in Axs.
Combining the above expressions, we obtain:
k k kf
or
AT 1 éxm fi%f
g G ) 8
< m I
or, simplifying the notation,
AT 1 _
2 OB g
Q/A k Q/A o
where [AE/(Q/A)O] combines all the fixed resistances. This is of the
form,
vy =ax +b (6)
where a is the slope of tThis linear expression and is the reciprocal of
the sample conductivity. The intercept b combines all other resistances.
In operating the apparatus, Q/A is kept constant and AT is recorded as
/X 1ig varied. If other modes and paths of heat transfer exist within the
the specimen, the thermal resistance will not be a linear function of the
specimen thickness. However, the effect of these other forms of heat
transfer will be reduced as the specimen thickness is decreased. Taus,
the conductivity can be determined from the reciprocal slope evaluated at
zero specimen thickness.
Another approach to the determination of the sample conductivity can
be obtained by rearranging Fg. (6) as:
k = = (()
Agaln, 1f other modes and paths of heat transfer exist within the specimen,
the value of conductivity obtained from Fg. (7) will be the effective
value which will approach the true value only as the specimen thickness
approaches zerc.
Iiffect of radiation
Many investigators consider'only the radiation emitted by the wall
surfaces when evaluating the heat transfer through a medium separating the
walls. This assumption may be correct when the medium is a gas whose mean
absorption coefficient k is small and whose mean refraction index 1 is
near unity. Many media, however, absorb and emit significant amounts of
radiation. This internal radiation can contribute more to the heat trans-
fer from wall to wall than the radiation emitted by the wall surfaces.
Indeed, even at room.temperature; the heat transferred by radiation can
approacn 5% of that transferred by conduction in some organic fluids whose
specimen thickness is as small as 0.1 cm.
If some simplifying assumptions are made, an expression for the
radiant heat tranzfer can be derived. These assumptions are the existence
of a constant temperature gradient within the medium and the use of mean
values n and & independent of wavelength. The following equation was
617
- 1 ~ - 1 e 3
derived by Poltyz for the radiant heat flux:
I:
——— Ao
Q 16 1° 5 AT
— e O T (w) Y, (8)
A 30K X
10
where
1 1 — exp [~(t/v)]
(2 — ¢) j ve dv
T ol + (L —¢) exp [~(t/v)]
and
T = optical thickness of the medium = k/X,
T = average medium temperature, °K,
€ = emissivity of the wall surface,
T L -12 —2 o -4
o = Stefan-Boltzwann constant, 5.71 % 10 Woem © (°K)T%,
v = durmmy lntegration variable.
In Fig. 2, where Y is plotted as a function of r for various values
of ¢, the curve for e = O represents the hypothetical case in wnich the
radiation heat transfer between the walls is accomplished solely by the
inner radiation within the medium. The distance between the curve for
¢ = 0 and one of the upper curves (the appropriate plate emissivity
ORNL—-DWG 72-10524
08 — o
o
o
Y, FUNCTION
O
I
0.2 0.5 1 2 o) 10 20
T,0PTICAL THICKNESS
Fig. 2. Radiative function (Y) vs plate emittance and optical thick~
ness of the specimen.
11
curve) represents the contribution of the radiation emitted by the wall
surfaces to the total radiated heat flow from wall to wall.
Rguation (8) can be combined with the previously derived Eq. (9)
to obtain an expression for the total thermal resistance across a medium
separated by two parallel walls when the heat is being transferred simul-
taneously by conduction and radiation. That is,
Nk : 1 alk
—_— = L+ | — . (9)
™ ]A6 ?fd M3
/A \k + = o1/ 0/ 7,
In the limiting case where the optical thickness 71 approaches zero (i.e.,
very small infrared-absorbing medium),
3 ~
T e T Ref. 16
YT“&U N CI‘ ( ) ?
wnere
1 1 1 e
I U :
El 62 o - e
for
Thus, Ba. (9) simplifies to
AT _ , 1 Al
— (k + b n® €. 0 T Am) K+ -~:) . (lO)
Q/A Q/A
Figure 3 is a plot of the thermal resistance as a function of speci~
men thickness for varicus values of the absorptivity coefficient for a
specimen assumed to have a k = 0.003% W et (”C)*l, n=1.5, € a1l = 0.5,
and T = 10C0°C. TFor these values of conductivity and temperature,
the percentage of heat transferred by radiation is quite large. As the
sbszorption coefficient x decreases from gk = « (pure conduction) to g = O,
the percentage of radiated heal increases to a maximum at about ¢ = 2 and
~.
decreases until ¥ = 0. Within the interval o« > g > O these curves have
12
ORNL --DWG 72-140525
L T | —
7 = 1000 °C | Vi | | kmO =]
56 e =05 —t ! % /ir —
7 =0.0034 Wem™!oC™’ | 100 <
g (}8 .....
QO
o
o
540
L)
2
L 32 e e S O
2]
Lut
o 24
EIJ
P
WG 16
I i ‘
- . | |
8 <l ey : N
J | |
0 0.05 040 1 020 025 030 035 040
0.45
SPECIMEN THICKNESS {(cm)
Fig. 3. 'Thermal resistance of an infrared absorbing fluid having
assumed properties at various values of absorptivity, « (em™') vs specimen
thickness.
an inflection point producing what could be described as '"lazy S" curves.
Also shown in Fig. 3 is the recistance curve for a gas whose absorptivily
is near zero and whose index of refraction is near one. If the optical
properties of a specimen are known, Eq. (9) can be fitted to the experi-
mental data to obtain the slope (and thus the conductivity) at a specimen
thickness approaching zero; however, the mean optical properties must be
used and the temperature gradient within the specimen must be nearly
linear.
Fffect of natural convection
Under ideal conditions, no natural convectioan would be expected in a
fluid enclosed between two horizontal, parallel plates with one-dimensional
downward heat flow. In the real situation, however, small departures from
the ideal conditions can initiate and sustain convection currents within
the fluid. If the plates are not horizontal or parallel or if a tempera-
ture gradient exists in the horizontal direction, convection cells can
13
oceur. If, in addition, the vertical temperature distribution within the
specimen is not linear, but distorted by interfluld infrared apsorption,
the natural convection can be enhanced. Finally, vibrations, especially
those 1n resonance with the natural frequency of the enclosed fiuid, can
induce and enhance natural convection.
In order to initiate and sugtain buoyancy convection cells within
enclosed spaces, certain instability criteria must be satisfied. Raylelgh
was one of the first investigators to recognize that the instability
criterion could be related to certain limiting values of the dimensionless
modull NRa known as the Rayleigh number, which is
gp” 8 AT /x” Co b
P N — A
o, = Ny N = - ) (11)
& W s
o
where
.. .. " . 2
¢ = local acceleration due to gravity, cm/sec
. . , S S
¢ = dimensional constant = 1.0 g cm dyne ~ -sec
o = density of fluld, g/em”,
U = viscosity, cP,
x = gap distance, cum,
k¥ = thermal conductlvity, W oem (Pe)™,
C
= (specific) heat at constant pressure, cal g™+ (°C)™t,
8 = coefficient of bulk exzpansion, (°C)7*
The Rayleigh number, in essence, 1s the ratio of the product of the
buoyancy and inertial forces to the viscous forces. The limiting value
of the Rayleigh number to initiate and sustain convection cells has been
caleculated to be 1700 when the fluid layer is bound on both sides by
s0lid parallel and horizontal walls and is heated from velow. ® Recent
experimental studles by Norden ahd Usmanovl9 using an interferometer
technique show, however, that the departure from conductive to convective
mode of heat transfer can occur at NRa < 1700 for small specimen thick-
nesses.
Figure 4 shows a portion of the data taken from the above experi-
mental studies in which the critical temperature difference, AEC (the
temperature difference above which convection occurs), is plotted as a
ORNL-DWG 72-10526
RATURE DIFFERENCE
,_
=
o
AT,, CRITICAL TEMP
,,,,,,
o RN
1072 2 5 1071 2
SPECIMEN THICKNESS (cin)
5 10!
Fig, 4. Critical temperature above which convection occurs as a
function of specimen thickness for three liguids showing the departure
from theoretical criteria, NRa = 1700.
function of the specimen thickness for three liguids: eithylene glycol,
water, and ethyl alcchol. Also shown plotted are the curves for the
theoretical values of NRa = 1700. 'The area below each of the curves
plotted in Fig. 4 is stable (i.e., conduction only) and above the curves
Is unstable (i.e., convection occurs). ‘he experimental data can be
seen to have two linear slopes (n = 0.45 and 2.0) which merge with the
theoretical curve (n = 3) where
n
AN A = constant
15
The Nfiz = 1700 criteria could lead to a grogs overestimate of AEO for
=
smallhspecimen thicknesses.
Heat-transfer measurements made during Norden and Usmanov?'s studies
showed the ratio of the effective conductivity to the true conductivity
for water to be 1.10 at a specimen thickness of 0.1% cm and an‘Nfia of
only 260. Thus, considerable care must be exercised to prevent natural
convection in fluids contained between parallel plates heated from below
as well as inclined, cylindrical, and spherical annuli heated from either
side.
e s N : ; A o)
A similar experimental study by Berkovsky and Fertmar®® was made
recently in which the specimen was heated from above and had a non-
uniform upper plate temperature distribution. This study showed that
when the nonuniformity of the temperature (T — T ) exceeded a certain
- max min
value with respect to the lower plate tewperature Tv’ no convection
{
occurred. It was found that if
T — T
max min
1,
T - T
max 0
no significant convection oceurs at a Rayleigh number of less than 10%.
No experimental results are reported for the amount of convection
taking place when the specimen is heated from above and the plates are
not exactly parallel or are slightly tilted. If we consider this case
analogous to The Berkovsky and Fertman study, convection would be avoided
at an NRa
the horizontal layer did not exceed that of the average AT across the
of less than 10% if the AT of the tilted layer above that of
plates. ©OGince Al &= Ax, the difference in the edge-to-edge separation
digtance between the plates with respect to each other, or with respect
to the horizontal, should not exceed their average separation distance.
To minimize convection due to vibrations, the conduction cell should
be well isolated from all sources of vibrations, particularly those with-
in the resonance frequency of the cell.
Effect of heat shunting
Some shunting of heat around the specimen 1s unavoidable even for
The most carefully designed thermal conductivity cells. The percent of
shunted heat as compared to heal flow through the specimen can be min-
imized by careful use of lnsulating materials, by guard heating, by using
large cell diameter to thnickness ratios, and by using zoned heal sourcesg
and sinks. The shunting problems becomes most acute for low-conductivity
cspecimens at elevated temperatures.
The apparatus described in the present report is designed to mini-
mize the shunting error with specimens having estimated thermal) conduc-
tivities in the range of 0.05 to 0.10 W em™ (°C)™*. The cell wall
thickness and the ratio of cell diameter to sample thickness are
optimized to reduce the heat shunted to less than 1% of the total heat
flow in the absence of heat guards. Unfortunately, the conductivities
of the salts of interest to the MSBR program were found to be an order
of magnitude lower than the range for which the apparatus was designed.
As a result, the radial guard heating was not adequate in a few cases to
prevent some heat shunting, and corrections were reqguired.
Figure 1 shows the complexity of the possible heat-transfer modes
and paths within the conductivity cell. Since neither the temperature
distribution along the cell wall nor the heat-transfer coefficients are
well known, the simplified model shown in Fig. 5 was selected as an
appropriate model for calculating the amount ©f heat shunting in the
system.
ORNL- DWG 72-10527
7 ;
;’/,:
7 NN 2
7 N} 77, REGION I
N Czzzz22%. "7 [ ecion i
’4/ : | 1
////'/ //////// 77 /f/,Ufiz///,/,; g //////?////{/// 7
=0
MODEL
Fig. 5. Model of the conductivity cell for heat shunting calcula-
tion.
17
By assuming =2 uniform heat flux and a uniform sink and wall temper-
ature equal to 0°C, an easy solution for the center-~line heat flux may
be obtained from the generalized heat conduction equation [Eg. (12)].
The radial heat-transfer coefficient Uz may then be decreased to account
for the guard heating.
The temperature distribultion within the cylindrical solid can be
18
described by the general heat conduction equation,
A%T 13T T
o — =0 (12)
3T r or D7z
where T is the temperature (°C) and r and 2z are the radial and axial
coordinates (cm) measured as shown in Fig. 5. Dividing the model into
O
two regilons, the boundary conditions for either region can be written:
) T (r,0)
k' ——— = - (13)
A7
3T {(r,I)
F} ’ : )
Fom——— Y ) Y €5 7 I B (1h)
A%