CALIBRATION TABLES FOR DIFFERENT THERMOELEMENTS.
65
Storch, 2.physik. Chem., 19, 13 (1896). 23. Bray, T a n s . Am. Electrochem. Soc., 21, 143 (1912). 24. Sloan, THISJOURNAL, 32, 946 (1910). 2 5 . Kahlenberg, J. physic. Chem., 5, 339 (1901). 26. Washburn and MacInness, THISJOURNAL, 33, 1686 (1911). 27. Sprung, Wied. Ann., 159, I (1876). 28. Franklin and Kraus, Am. Chem. J., 23, 277 (1900); THISJOURNAL, 27, 191 i( 1901). 29. Franklin, 2. physik. Chem., 69, 272 (1909). 30. Fitzgerald, J. Phys. Chem., 16, 621 (1912). 31. Franklin and Gibbs, THISJOURNAL, 29, 1389 (1907). 32. Franklin, J . Phys. Chem., 15, 675 (19x1). 33 * Tammann, Wied. Ann., 69, 770 (1899). 34. Cohen, Wied. Ann., 45, 666 (1892). 35. Kohlrausch, Wied. Ann., 6, 196 (1879). 36. Sutherland, Phil. Mug.,50, 460 (1900). 3 7 . Young and Mitchell, THISJOURNAL, 26, 1389 (1904). 38. Euler, 2. physik. Chem., 25, 536 (1898). 39. Washburn, THISJOURNAL, 31, 322 (1909). 40. Denison, Z. physik. Chem., 43, 596 (1903). 41. Buchbok, 2. physik. Chem., 55,563 (1906). 42 * Washburn, THISJOURNAL, 33, 1461 (1911). 43. Lewis, THISJOURNAL, 34, 1631 (1912). 44. Kendall, J . Chem. Soc., 102, 1275 (1912). 45. Noyes and Kato, THISJOURNAL, 30,318 (1908) 46. Drucker, Z . physik. Chem., 62, 731 (1908). 47. Kohlrausch, Proc. Roy. SOL,71, 338 (1903). 48. Danneel, 2. electrochem., 11, 249 (1905). 49. Kraus and Bray, THISJOURNAL, 35, 1315 (1913). 22.
BOSTON, MASS.
CALIBRATION TABLES FOR COPPER-CONSTANTAN AND PLATINUM-PLATINRHODIUM THERMOELEMENTS. BY I,. H. ADAXE.
Received November 1, 1913.
In view of the fact that chemists are making more and more extensive use of thermoelements as a means of measuring high temperatures, i t seemed desirable t o present two tables which save muih trouble in converting microvolts into degrees. Nothing will be said here as to the precautions to be observed in the use of thermoelements,1 further .thanthat experience has shown the necessity of frequent recalibration in order t o insure trustworthy results. The experimental work required for such a calibration involves comparatively little time or labor, but interpolation between the fixed points can be accomplished accurately only by means of an empirical equation expressing the relation between electromotive force and temperature, and repeated recalculation of such an equation 1 These precautions have been discussed by W. P. White, Physic. Rev., 23, 449 (1906); Am. J. Sci., [4] 28, 479 (1909).
SF;
L. H. ADAMS.
may prove a somewhat tiresome task. This may be avoided by the use of a table representing a standard ciirvp in conjunction with an appropriate "deviation curve" : the advantages of this procedure have already been pointed out by Sosman.' Such 5tandard curves are arbitrary, but it is always advisable, for the sake OS accuracy, that they represent closely the actual form of curve characteristic of the thermoelement, so that the deviations may be as small as possible. Standard curves for use with platinum-platinrhodium2 and copperconstantan8 elements have been given in previous publications; but they now require amendment. Certain changes in the accepted temperature scale below 6jo" have necessitated a recalculation for the curve for the Pt-PtRh couple; and the superior qualities of the kind of constantan wire, which we hare lately succeeded in obtaining, made it desirable to construct for copper-constantan elements a new table which should conform more closely to the actual observations with that kind of wire. The C'oppei-constantan L'zrrve.-('l'able I.) The curve given in Table I represents very closely the observations for a sample of "Ideal" wire.4 This wire is of such uniformity-a characteristic which is very desirablethat three lengths cut from the same spool did not differ in electromotive force against copper by as much as I part in ~j,ooo. The table was calculated from the equation
E
= 74.672t -- 13892 (I
--
E being the electromotive force in microvolts, t the temperature (Centigrade) and e the base of the natural logarithms. The advantages of this form of equation, as compared with the ordinary power series, have been indicated in another paper;6 that it is capable of reproducing the observations. with high precision is evident from the following: Temp
E (ohs ) Microvoit~
E jcalc ) Sl~crovolts 960,9
Difference Microvolts *
24.30
961. I
49.59 78.75
2010.9
2010.8
0 .I
2391.7
3291.4
0.3
100.0
4276,
217.95 305.9
10248. Ij203,
1276) i10248) (1j203)
0.2
*
* *
R. B. Sosman, Am. J . Sci., [4] 30, 7 (19101. 30, 9 (1910). I,. H.Adams and J. Johnston, Am J . Sci., [4]32, 534 (1912). * Practically identical with constantan. Obtainable from the Electrical Alloys Co., Morristown, N. J ; size No. 30 B. & S., double silk-covered, is perhaps the most convenient. L. H. Adams, J . Wash. Acad., 3, 469 (1913). I microvolt corresponds in this case to about 0 . 0 2 ' * Icsed in computing the equation.
* Ib-id., j 4 ]
CALIBRATION TABLES FOR DIFFERENT THERMOELEMENTS.
67
The Pt-PtRh Cwve.-(Table 11.) For this element it proved impossible t o find a single equation, with a reasonable number of constants, which would adequately express the relation between temperature and electromotive force over the whole range between o o and 17js O, the melting point sf platinum; a circumstance which is hardly surprising, in view of the magnitude of the range. Accordingly, the curve given in Table I1 was calculated in three overlapping sections: 0-400°, 300-1200°, 1100-17jj~. The first section was computed from an exponential equation of the form,
+ B ( I -ect); the second from the equation,l E = -308 + 8.22942 + o.oo1649t2; E
=
At
and the third by inverting the table of Sosman2 so that the argument is in microvolts and not degrees. That this procedure introduced in this case no discontinuities in the slope of the curve a t the points of transition from one section to another, was established by the identity of the results calculated in both ways for the overlapping portions. Any slight inequalities were evened out by adjustment of the successive differences. Remarks.-It is understood, of course, that the cold junction is maintained at oo, which is most conveniently done by means of a mixture of ice and water contained in a vacuum-jacketed flask; if the cold junction is not a t oo, corrections similar to those given in the table by Foote must be a~plied.~ The standard temperatures employed are based mainly upon t h e series of gas thermometer determinations made by Day and S ~ s m a n ;this ~ scale does not differ materially from that advocated by the Bureau of Standardss or from that now in use a t the Reichsanstalt up to I I O O O , ~ and it is little likely to undergo any marked change for some time to come. The tables give temperatures (on the thermodynamic scale) and temperature differences for each IOO microvolts up to the limit of usefulness of each thermoelement; this form of table is more troublesome to calculate b u t is much more convenient to use. The last digit in each temperature value is given mainly for purposes of interpolation; otherwise it is meaningA. L. Day and R. B. Sosman, Am. J . Sci., [4]33, 528 (1912); Ann. Physik, 38, 863 (1912). R.B.Sosman, Am. J . Sci., [4]30, 9 (1910);Carnegie Inst. Wash., Pub. 157, 118. 3 P . D. Foote, Met. Chem. Eng., 1 1 329 (1913). Arthur L. Day and R. B. Sosman, Am. J . Sci., [4]29, 93-161 (1910);33, 517533 (1912) ; “High Temperature Gas Thermometry,” Carnegie Inst. Wash., Pub. 157; Jour. Physipue, [ 5 ] a, 727, 831, 899 (1912). See G. K.Burgess, Report 8th Intern. Congr. Appl. Chem., 22, 53-63; J . chim. PhYS., 11, 529 (1913). It should be remarked, however, that the official calibrations issued by the Reichsanstalt are, so far as we know, stiU based on their old temperature scale, which at 1000 O leads to temperature values about I .5 ’ higher than a o s e given here.
68
L. H. ADAMS.
TABLE E. 0
0.
2.60 100
2.60
200
5.17
800
7.73
400
10.28
27.72
2.56
30.15 32.57
2.55
12.81
2.52 600
15.33
700
17.83
so0
20.32
BOO
22.80
2.50
E.
2.49
100
215.21
900
800 400
1.85 1.85
217.06 I
.85
I
.84
235.38
I
TOO
226.26
800
228.09
9 0
229,92
1.83
1.82
I
I .79
253.40 I
.81 .81
1.78
I
.80
262.29 264.06
1.80 265.83
248,03
1.79 249.82
96 23
2.22
83.17
2
91.91
2.12
106.91
2.19
2.18
z.Ir Iog.02
2.10 111.12
2.17
2.10
113.22
2.16
2.09
115 .SI
18000. 267.60
14000.
285.13
1.76 269.36
286.87 I
.76
271.12
288.61
1.76 1.76 I
1.77 281.65
1.77
1.73
1.75
1.73 295'54
1.75 1.74 I
285.13
1.73
293.81
283.39
77
1.74 1.74
.76
I ' 75
279.90
1.74
290.35 292.08
274.64
278.15
13
104.79 2.20
89,74
2.14
102.66 2.20
87.56
2.15
2.14
2.21
80.97
85.37
2.16
100.52
78.76
1.78
1.77
94.07
98.38
276.40
I
267 60
2.23
272.88
256.96 258.74
.23
76.54
.78
260.52
244'43
74.31
1.78
I.80
.83
1.79
255.18 I
1.82 231.74
1.82
246.23 I
1000
251.61
2
94.07
1'1000.
1.81
.87
2.25
249.82
.82
242.63
224.43
2.27
72.08
I
233.56
2.27
2.25 69.83
240.82
222.59
65.31
2.35
239.01
220.75
63.04
67.58
11000. 231 74
1.84 600
2.36
237.20
218.91
2.28
2.34
1.84 600
2.38
49.20
1oo00.
213.36
a .30
60.76
44.51 46.86
2.31 2.30
58.46
'
25.27
0
2.41
2.39 39.77
2.32
56.16
37.38
2.47 loo0
2.42
53.85
2.40
42 I5
2.48
2.33 51.53
34.98
2.53 500
2.45 2.43
2.57
72.08
49 I20
2j 2 7
0
4000.
3000.
2000.
1000.
297.26
1.72 1.72
298.98 I
300 70
74
.72
1.72
302.42
Standard calibration curve for copper-constantan thermo-element giving the temFor use in conjunction with a deviation curve determined by calibration of the Water.. . . . . . . . . Naphthalene. . . .
b. p. b. p. S n . . . . . . . . . . . . . . m. p. Benzophenone. . b. p. Cd . . . . . _ . . . _ . . . m.p. I
CALIBRATION TABLES FOR DIFFERENT THERMOELEMENTS.
69
I. 6000.
5000.
115.31
2.09 117.40
2.08
119.48
2.08 121 .56
135.91
139.96
2.06
133.88
2.03 135.91
15000.
192.73
1.72
1.70
321.19
r.71
1.69 322.88 I .69
I.7I
324.57
1.69
326.26
1.71
310.98
1.71 312.69
1.70 1.70
316.09
1.69 327.95
I .69
329.64 I
.68
I
.68
338.04 339.72 341.40 343 . 0 7 344.74 346.41
I .68
319.49
336.36
100'
4276 PV 10248 pV I1009 pV 1.5203 PV 16083 gV
900
1.86 213.36
1000
18000.
1. 0
353.09
....
1.68 ....
..
....
100
a00
I .68
....
1.67
....
..
300 400
1.67 1.67 ....
1.67
..
600 600
, .
1.67 ,349'75
353'09
perature and temperature differences for every 100 microvolts. particular element a t some of the follo ing fixed points: 217.95' 231.9 305.9' 320.9'
800
1.86 2 1 1 .SO
1.68 1.67 317.79 334.68 I ,68 351.42 1.70 I .67 1.70
700
1.86
1.89
348.08
331.32 333 . m
17000.
336.36
600
1.87 209.64
194.62
16000.
319.49
205.91
1.90
1.98
500
1.87 ro7.78
190.83
I .99
400
1.88
I .90
155.95
302.42
314.39
188.93
151.99
300
204.04
1.91
aoo
I .88
r.91 187.02
1.99 153.97
I .88
202.16
185.11
169.68
198,40
I.~I
1.95
100
1.89 200.28
183.20
167.73. 2.00
196.51
I .92
1.95
0
I .89
I ,92
181.28
1.96 165 .'78
150.00
2.04
309.27
163.82
148.00
131.84
179.36
I,97
2.00
2.05
1.93
I .96
2.0I
2.04
305 ' 85
161.86
143.99
2.06 129.80
177.43
159.89
E.
9000.
194.62
1.93
I .97
2.0I
146.00
125.69
I .97
157.92
1.92
8000.
175.50
2.02
141.98
123.63
304.14
2.03
137.94
2.07
127.75
7000.
I55,95
700 800 900
1000
Cold junction a t o o
L. H. ADAMS.
TABLE E. 0
0.
1000,
147. 1
0
17.8 100
17.8
200
34.5
800
50,s
400
6s ..1
16.7
Id 0
I2
15,I
I'?
89.0 94.1
700
107.8
800
I21
.z
11.8
331.5 II.j
13.7
312.3
231 2
II.j
242.7
3.53 . O II.4
900
1000
E. 0
254. I
14;.
10000.
1037.3
100
1045 Y
200
1054 4
500
1062.9
400
1071.4
11 .3
265.4
t
11000.
8.5
8.4 8.4
1222.6 1230.9
8.4
1096.9
800
1105.4
900
1113.8
1000
I122 2
8.5 8.5 8.4
1180.9 1189.2
8.3
1264.3 1272.6 1281.o
10.4
10.0
528.6 10.0
10.4
437. I
10.3
538.6 10.0
548.6
447.4
9.9
10.3
457.7
558.5
9.9
10.2
568.4
467.9
9.9
10.2
478.1
j78.3
1289.3
1372.4
14000.
8.4
1306.0
1380.7
8.3
9.3 8.3
1314.3 1322.6
8.3
1389 . o
8.3
8.3
1397.3
8.3
1405.6
8.3 8.2
8.3 1413.8
8.2
1422 .o 8.2
8.4
1430.2
1347.5
8.3 8.4
8.2
8.3 1355.8 1364. I
1438.4
8.3
8.2
1446.6 8.2
8.3
8.3 1280.3
10.1
518.6
8.3
8.3 1205.9
10.4
426.7
8.3
8.d
1197.6
8.4
416.3
508.5
1255.9
8.4
10.1
10.5
405.9
1247.6
1172.5
700
8.4
8.4
8.3 600
498.4
1297.7
1239.3
1164.2
EO0
10.6
IO.I
10.5
395.4
8.3
8.4
8.5
10.7
1214.2
1147.4 1155.8
10.7
10.2
488.3
13000.
8.4 1139.0
8.5
363.7
I O .8
1205.9
1130.6
8.j
10.9
374.3
1122.2
8.6
I1 .o
10.9
II.0 219.7
13.4
309.7 320.6
208.I
14.1 600
II .o
298.7 12.0
196.3 I4.0
600
II.I
3
478. I I0.6
384.9
287. 7
184.3
4000.
374.3 II.2
4
172,I
1j.8
265.4 276.6
159.7
5000.
2000.
1372.4
1454.8
Standard calibration curve for Pt-Pt.Rh (10% Rh) thermoelement, giving the temFor use in conjunction with a deviation curve determined by calibration of the Water. . . . . . . . Naphthalene. .
Sn ....... . . . . .
Benzophenone. Cd.. , ., . . . . . . Zn . . . . . . . . . . . . Sulfur.. . . . . . . . Sb . . . . . . . . . . . .
Al... . . . . . . . . .
b. p. b. p. m. p. b. p. m. p. m. p. b. p. m. p. m. p.
IOO' 2 I 7.950 O
231.9 305.9 O 320.9' 41p.4'~ e ~ 4 . 5 ~ ~ 630.0 658.7'
CALIBRATION TABLES FOR DIFFERENT THERMOELEMENTS.
5000.
578.3 588. I
6000.
675.3
9.8
684.8
597.9 607.7
9.5 713.3
9.4
656. I
741.5
665.7
760.2
9.6
16000.
8.2
1463.o
8.2 1471.2
8.2 8.3 1487.7
16000.
8.3 1545.8
8.3 1554.1
1570.8
8.3
1496.0
1512.6 1520.9 1529.2
1612.5
1537.5
1696.0
8.4
8.3 1620.9
m. p.
m. p. m. p. m. p.
m. p. m. p. m. p. m. p.
960.2' 1062.6' 1082.8'
....
100
aoo SO0
8.4 400
8.3
....
1746.0
8.3 1754.3
....
...
8.4
... .... ...
8.3
...
8.3
1704.3
9111 fiv 10296 pv 10534 pV 1201° 11941 pV 1391.5° I4230 pV 1452.6' 14973 PV I549;5 O 16144 PV 1755 18608 MV
0
...
8.4
....
....
perature and temperature differences for every 100 microvolts. particular element a t some of the following fixed points:
Ag . . . . . . . . . . . Au . . . . . . . . . . . c u. . . . . . . . . . . LizSiOa. . . . . . . . Diopside.. . . . . Ni . . . . . . . . . . . . Pd . . . . . . . . . . . Pt ............
E.
....
8.3
8.3 8.4
1687.6
1000
1037.3
1737.7
1679.3
8.3
1721 . o
8.3 1662.6
8.4
900
8.6
1729.3
1654.3
1595'8
1712.6
8.4
1670.9
1028.7
18000.
8.3
8.3 1604.2
8.4
1645.9
1587.5
8.3
1637.6
8.4
8.4
8.3 8.3
1629.2
700
8.6
8.8
8.3
600
800
1020.1
1704.3
8.3
8.7 8.6
8.8
950.4
17000.
8.3
1579.1
1504.3
941.6
9.1
8.3
8.3
932.8
1620.9
1011.5
8.9
9.1
861, I
1537.5
1562.4
1479.4
852.0
9.3 769.5
675.3
1454'8
9.3
400 600
8.7 1002.8
8.9
SO0
8.7 994.1
923.9
842.9
750.9
9.6
8.7
8.9
9.1
100
985.4
915.0
833.8
9.4
9.6
100
8.8
976.7
906. I
9.1
0
8.7
9.0
E.
8.8
968 .o
9.0 9.0
9.1
9.4
959.2
897.1
824.7
732 ' 1
9.7
888. I
9.2 815.6
722.7
9.7
646.5
806.4
9.0 9.0
9.2
9000.
950.4
879. I
9.2
9.4
9.7 627. I
788.0
870.1
9.2
797.2
703.8
9.7 617.4
9.3
9.5
8000.
861. I
778.8
694,3
9.8
636.8
9.5 9.5
9.8
7000.
769.5
7=
....
...
500 600
700
....
....
...
800 900 1000
Cold junction a t
0'.
less except at the lower end of the scale. Similarly, in the list of fixed points given a t the foot of Table 11, the magnitude of the uncertainty in the temperature wale is iuch that the temperatures aboze ~ooo' may be rounded ofT tn the nearest degree 'Mie reicrrnce curLt25 given in Tables I and I1 are intended for use in conjunction M ith the appropriate deviation curve. This correction curve is determined for each element by calibration a t several of the fixed points -preferably three or more--given at the foot of each table; whence it is simply constructed by plotting1 AE as ordinate (AE = E observed - E standard) against Eobs a\ abscissa, and joining up the various points. Then in order to obtain the temperatures corresponding to the electromotive force reading indicated by the element, the appropriate value of LE (as obtained from it> deviation curve by inspection) iq subtracted algebraicaliq- from the observed value of E before the latter is converted into degree5 by rneans of the table. There need be no apprehension of error in the use of this method even with deviations of as much as roo inicrovolts , especially if iufficient calibration points be taken within the specific temperature range, and if the deviation curve 50 obtained does not depart too far from a straight line. I t should he borne in mind that neither of these tables has an ahsolute significance, it represents mcrcly an arbitrary reference curve which is substantiallq the meau of the three elements (E, F, G) used by Day and Sosman as standards, for the curve does not differ from the average reading of these elements by more than I microvolt, except a t the somewhat less certain nickel point, where the divergence amounts to j microvolts. GEOPHYSICAL LAUOXATURY, C.\RXEGIB IN5TITUTIOIZ OP 1 y . 4 S € I I N G T U & ,
WASHINGTON. D C
NOTE. f i n Electrical Contact Vapor-Pressure Thermoregulator.-The change in the pressure of a saturated vapor with change in temperature has been made use of by various investigators as a means of automatic temperature control. Andreae' devised a thermoregulator in which a small quantity of some volatile liquid, such as ether, alcohol, or water, was utilized to increase the change of pressure per degree of a suitably enclosed volume of air or other gas. Lothar Meyer3 made use of the same principle. Benoit4 described a thermoregulator similar to the foregoing, with certain mechanical improvements. Kahlbaum5 It I S obvious that the required accuracy is secured by plotting on a small scale; a sheet of coordinate paper 2 0 X 2 0 cm is ample U-red Amalen, 4, 614 Ber , 16,io88 (1883) W e d Bezbl., 4, 296. Ber , 19, 2860 (1886)
