INDUSTRIAL AND ENGINEERING CHEMISTRY
December, 1928
sulated with sealing wax. Although there may be a slight effect of this distance of 8 cm. from center to center, as compared with 0.5 em. in the first case described, yet the effect is so small as to be fairly negligible. As the current density representing the extent of corrosion that occurs in a tin can must be exceedingly small, it is not likely that the conductivity of the electrolyte and hence the distance of two electrodes that might be set up within a can of fruit are pronounced factors in determining the rate or nature of corrosion. Application of Results t o Commercial Experience
The effect, from the standpoint of corrosion in canned fruits, in plain tin cans, as compared with that in enameled cans, is very striking. The purpose of enameled cans should be borne in mind. Enameled cans are used for colored fruits whose color is bleached by their action on plain cans. The bleaching action in plain cans is evidence that the coloring matter enters into the corrosion. The fruits which are generally packed in plain cans, as well as those which are ordinarily canned in enameled cans, do not tend to perforate unenameled cans, although the colored fruits generally have a tendency to perforate enameled cans. The action of the fruit on an unenameled can is generally distributed over the entire area of the can. In plain cans the area of tin exposed always enormously exceeds the area of iron exposed. The effect of the anodic tin in inhibiting iron corrosion is therefore pronounced, while the effect of the small area of cathodic iron exposed is almost negligible in increasing the amount of tin corrosion. In an enameled can a different condition exists. A vast proportion of the area of tin is eliminated from the corrosion picture by the enamel coating. It is true that portions of the iron which the tin does not coat may also be covered by the enamel coating. However, it is a t that portion of the can which is adjacent to the seam and other points where the plate undergoes a strain in forming the can that the bulk of the corrosion occurs. At such points both the enamel coat and the tin coat tend to be more or less fractured. At such points, therefore, the relative areas of tin and iron
1377
are of a vastly different proportion than in an unenameled can. At such points in an enameled can the area of the exposed iron apparently is sufficient to materially influence corrosion of the anodic tin. It is a common observation that such areas are, relatively early in the life of the can, completely detinned. The result is that, comparatively early in the life of many enameled cans, the area of iron exposed actually exceeds the area of tin exposed. The exposed iron, therefore, loses to a considerable extent the protective effect of the anodic tin. Local couples no doubt are set up on the iron, augmented by its non-homogeneous nature, whereby hydrogen is formed, and eventually pitting and perforation result. At any rate, it is common commercial experience that enameled cans give rise to hydrogen formation and perforations to a much greater extent than plain tin cans, even when the same fruit is considered. If, as pointed out by Walker and Lewis,I7 and to whom Lueck and Blair5 refer, the enamel film may act as a cathode to the iron, this would be an additional cause for perforation of the iron base plate in enameled cans. We have already stated that tin is anodic to iron, even in the less acid fruits in which hydrogen formation and perforations are the most common. It is, however, not so strongly anodic as in certain more acid fruits in which there is less tendency to hydrogen formation and perforation of the can. Our data on the effect of hydrogen-ion concentration on the single potential of tin offer an explanation for the relative nobility of the two metals under these varying conditions of acidity. The less pronounced tendency to hydrogen formation and perforations in the more acid fruits may be explained on the basis of the more effective protection of the iron by the tin, which is more markedly anodic. Tin is only mildly anodic to iron in certain varieties of black cherries, among the less acid fruits. These cannot be successfully canned because of the hazards of corrosion. It is believed that if tin were actually cathodic in any fruit, as has long been believed by many, the modern tin can would not be a practical container for such a fruit. 17
J. IND. END.CHEM.,1, 754 (1909).
Specific Gravity of Glycerol' L. W. Bosart and A. 0. Snoddy CHEMICAL DIVISION, THEPROCTER A N D GAMBLE COXPANY, IVORYDALE, Oruo
UR paper entitled ''Yew Glycerol Tables"2 was published to provide practical working tables based upon glycerol of the utmost possible purity, which, it was believed, could be relied upon for accuracy to the fourth decimal place for the percentages of glycerol. This work was done because none of the tables then in use seemed sufficiently accurate for either scientific or technical purposes. I n the case of some of the data published, no special precautions had been taken to insure a 100 per cent glycerol from which to start; in some, the determination of the specific gravity was only incidental t o other work; in others, no attempt was made a t accuracy beyond the third decimal place, although calculations showing the specific gravity in the fourth place were made; sometimes it was not clear whether apparent or true specific gravities were intended; and some of the tables were incomplete. I n short, none of the tables were entirely satisfactory. R e believed that
0
1 Received
* IND.
July 2, 1928.
ENG. CHEM., 19, 506 (1927).
our glycerol tables would be thoroughly satisfactory and would give results that could be relied upon for complete accuracy in the fourth decimal place a t least. This was the sole purpose of the work and therefore its only justification. Since the publication of our tables the third volume of the International Critical Tables has been published, and it contains, on page 121, tables for the absolute density of glycerol a t 15", 15.5", 20°, 25O, and 30' C. As the International Critical Tables are generally considered authoritative and as they show certain wide disagreements with ours, the whole question of the correct specific gravity of glycerol and its water dilutions is again opened. The International Critical Tables necessarily take into consideration the work of various investigators and probably reflect something of a composite of such work as is given the greatest credibility. Hence they are in fair agreement with ours for 15", 15.5', and 25" C., but in very poor agreement for 20" C. This is shown in Table I, where our values, originally given in terms of specific gravity a t 15'/15", 15.5O/15.5", 2O0/2Oo,
INDUSTRIAL AND ENGINEERING CHEMISTRY
1378
in International Critical Tables and in Tables of Bosart and Snoddy, Calculated t o Same Basis
Table I-Densities
I
DENSITY AT 15" C.
GLYCEROL
Table 11-Actual
I.C.T.(20" C.)
-
(1)
AT 15.5' C.
R. &
I.C.T.
GLYCEROL
s.
and Calculated Densities
1.02205 1.04695 1.07250 1.09920 1.12620 1.15375 1.18130 1.20840 1.23510 1.26105
DIFFERENCE
I,C.T.
Actual (2)
R.
(5)
1.0221 1.02198 1.02209 1.0470 1.04683 1.04690 1.0727 1.07264 1.07270 1.0995 1.09927 1.09931 1.1263 1.12622 1.12629 1.1533 1.15377 1.15380 1.1808 1.18127 1.18123 1.2079 1.20853 1.20850 1.2347 1.23505 1.23509 1.2609 1.26109 1.26108
$% ::
S.
(6)
$:,$ ;:$:: 2:
+O.OOO~O +0.00006
:!$:;
1 ::: ;
-0.00050 -0.00003 -0.00040 $0.00004 -0.00015 - O . O O O O ~
and 25"/25" C., have been recalculated to a bask of absolute density at each temperature, in other words, water of maximum density is taken as unity and weighings are reduced to vacuum.
AT 25O C.
AT 20" C.
B. 8: S.
I.C.T.
B. & S. (20' C.)
% 10 20 30 40 50 60 70 80 90 100
Vol. 20, No. 12
I.C.T.
B. & S.
I.C.T.
B. & S.
1,02209 1.04690 1.07270 1.09931 1.12629 1.15380 1.18123 1.20850 1,23509 1.26108
1.0207 1.0453 1.0706 1.0971 1.1239 1,1511 1.1784 1.2054 1.2320 1.2580
1.02071 1,04526 1.07071 1.09708 1.12375 1.15106 1.17841 1.20546 1.23198 1.25802
The coefficient of expansion of glycerol is fairly constant, a t least for glycerol of high concentration, as shown by Comey and Backma and confirmed by US.^ From our work it appears that the coefficient of expansion is practically uniform for a range of temperature of 15" to 25" C. and for concentrations from 80 to 100 per cent. For lower concentrations the coefficient of expansion varies, but for a given concentration it varies so slightly at different temperatures between 15" and 25" C. that we should expect. to find the specific gravity for glycerol of a given concentration a t 20" C. almost exactly halfway between that for 15" and that for 25" C. Indeed, if we were to prepare a table for 20" C. (showing values for intervals of 10 per cent glycerol), using the values we have obtained a t 15" and 25" c. and Placing the values for 20" c. midway between them, they would agree in every case a J. IND. 4
ENG.CIIEM., 2, 11 (1910).
I b i d . , 19, 506 (1927).
Table 111-Density and Percentage of Glycerol DENSITY AT:
GLYCEROL
15OC.
15.5'C.
2OoC.
25OC.
1.26108 1.25850 1.25590 1.25335 1.25080 1,24825 1.24560 1.24300 1.24035 1.23770 1.23510 1.23245 1.22975 1.22710 1.22445 1.22180 1.21915 1.21650 1.21380 1.21115 1.20850 1.20575 1.20305 1.20030 1.19760 1.19485 1,19215 1.18940 1.18670 1.18395 1.18125 1.17850 1.17575 1.17300 1.17025 1.16750 1.16475 1.16205 1.15930 1.15655 1.15380 1.15105 1.14830 1.14555 1.14280 1.14005 1.13730 1.13455 1.13180 1.12905
1.25802 1.25545 1.25290 1.25030 1.24770 1.24515 1,24250 1.23985 1.23725 1.23460 1.23200 1.22935 1.22665 1.22400 1.22135 1.21870 1.21605 1.21340 1.21075 1.20810 1.20545 1.20275 1,20005 1.19735 1.19465 1.19195 1.18925 1.18650 1.18380 1.18110 1.17840 1.17565 1.17295 1.17020 1.16745 1.16475 1.16200 1.15925 1.15655 1.15380 1.15105 1.14835 1.14560 1.14285 1.14015 1.13740 1.13465 1,13195 1.12920 1.12650
30'C.
82
81 80 79 78 77 76 75 74 73 72
71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51
EROL
15OC.
15.5O C.
20' C.
25' C.
30' C.
1.12870 1.12600 1.12325 1.12055 1.11780 1.11510 1.11235 1.10960 1.10690 1.10415 1.10145 1.09875 1.09605 1.09340 1.09070 1.08800 1.08530 1.08265 1.07995 1.07725 1,07455 1.07195 1.06935 1.06670 1.06410 1.06150 1.05885 1.05625 1.05365 1.05100
1.12845 1.12575 1.12305 1.12030 1.11760 1,11490 1.11215 1,10945 1.10670 1.10400 1.10130 1.09860 1.09590 1,09320 1.09050 1.08780 1.08515 1.08245 1.07975 1.07705 1.07435 1.07175 1.06915 1.06655 1.06390 1.06130 1.05870 1.05610 1.05350 1.05090 1.04825 1.04575 1.04325 1.04075 1.03825 1.03570 1.03320 1.03070 1.02820 1.02565 1.02315 1.02075 1.01835 1.01590 1.01350 1.01110 1.00870 1.00630 1.00385 1.00145 0.99905
1.12630 1.12360 1.12090 1.11820 1.11550 1,11280 1.11010 1.10740 1.10470 1.10200 1.09930 1.09665 1.09400 1.09135 1.08865 1.08600 1.08335 1.08070 1.07800 1.07535 1.07270 1.07010 1.06755 1.06495 1.06240 1,05980 1.05720 1.05465 1.05205 1.04950 1.04690 1.04440 1.04195 1.03945 1.03695 1.03450 1.03200 1.02956 1.02705 1.02455 1.02210 1,01970 1.01730 1.01495 1.01255 1.01015 1.00780 1.00540 1.00300 1.00060 0.99823
1.12375 1.12110 1.11840 1.11578 1.11310 1.11040 1.10775 1.10510 1.10240 1.09975 1.09710 1.09445 1.09180 1.08915 1.08655 1.08125 1.08390
1.12110 1.11845 1.11580 1.11320 1.11055 1.10795 1.10530 1.10265 1.10005 1.09740 1.09475 1.09215 1.08955 1.08690 1.08430 1.08165 1.07905 1.07645 1.07380 1.07120 1.06855 1.06605 1.06355 1.06105 1.05855 1.05605 1.05350 1.05100 1.04850 1.04600 1.04350 1.04105 1.03860 1.03615 1.03370 1.03130 1.02885 1.02640 1.02395 1.02150 1.01905 1.01670 1.01440 1.01205 1.00970 1.00735 1.00505 1.00270 1.00035 0.99800 0.99568
%
% 100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83
DENSITY AT:
GLYC-
1.26415 1.26160 1.25900 1.25645 1.25385 1.25130 1.24865 1.24600 1.24340 1.24075 1.23810 1.23545 1.23280 1.23015 1.22750 1.22485 1.22220 1.21955 1.21690 1.21425 1.21160 1.20885 1.20610 1.20335 1.20060 1.19785 1.19510 1,19235 1.18965 1.18690 1.18415 1.18135 1.17860 1.17585 1.17305 1.17030 1.16755 1.16480 1.16200 1.15925 1.15650 1.15370 1.15095 1.14815 1.14535 1.14260 1.13980 1.13705 1.13425 1.13150
1.26381 1.26125 1.25865 1,25610 1.25350 1.25095 1,24830 1.24565 1.24305 1.24040 1.23775 1.23510 1.23245 1.22980 1.22710 1.22445 1.22180 1.21915 1.21650 1.21385 1.21120 1.20845 1.20570 1.20300 1.20025 1.19750 1.19480 1.19205 1,18930 1.18655 1.18385 1.18105 1.17830 1.17555 1.17275 1.17000 1.16725 1.16445 1.16170 1.15895 1.15615 1.15340 1,15065 1.14785 1.14510 1,14230 1.13955 1,13680 1.13400 1.13125
1.25495 1.25235 1.24975 1.24710 1.24450 1.24190 1,23930 1.23670 1.23410 1,23150 1.22890 1.22625 1.22360 1.22096 1.21830 1.21565 1.21300 1.21035 1.20770 1.20505 1.20240 1.19970 1.19705 1.19435 1.19170 1,18900 1.18635 1.18365 1.18100 1.17830 1.17565 1.17290 1.17020 1.16745 1.16470 1.16195 1.15926 1.15650 1.15375 1.15100 1,14830 1.14555 1.14285 1.14010 1.13740 1.13470 1.13195 1.12925 1.12650 1.12380
50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
1.07860 1.07600 1.07335 1.06815 1.07070 1.06560 1.06305 1.06055 1.05800 1.05545 1,05290 1.05035 1.04780 1.04525 1.04280 1.04035 1.03790 1.03545 1.03300 1.03055 1.02805 1,02560 1.02315 1.02070 1.01835 1.01600 1.01360 1.01125 1.00890 1.00655 1.00415 1.00180 0,99945 0.99708
December, 1928
INDUSTRIAL A N D ENGINEERING CHEMISTRY
within 0.0001 in specific gravity with the values we actually found, except for 10 per cent glycerol, in which case this figure is barely exceeded. Moreover, this is to be expected, as there is a gradual rise in the rate of expansion with the increased temperature as the glycerol solution becomes more dilute. This close agreement is shown in Table 11, which gives (1) the figures calculated for halfway between those for 15" and those for 25" C. as given in the International Critical Tables, (2) the figures given in the International Critical Tables for 20" C., (3) the figures calculated for halfway between those for 15" C. and those for 25" C. obtained from our table, (4) the figures according t o our table for 20" C., (5) the difference between the figures reported for 20" C. in the International Critical Tables and those calculated from them for 15" and 26" C., (6) the differences between the figures for 20" C. as actually found by us and as calculated from our results a t 15" and 25" C. Not only do our figures for 20" C. correspond almost exactly with our calculated figures, but in nearly every instance they correspond within 0.0001with the figures given in column (1) of Table 11, which shows the figures for 20" C.
1379
calculated from the International Critical Tables for 15" and 25" C. As the densities for glycerol published in the International Critical Tables (especially those a t 20" C.) did not seem entirely satisfactory and as tables for absolute densities are desirable for certain purposes, we have recalculated our tables to this standard, as shown in Table 111. The density table for 30" C. is also given. This is not based on actual determinations a t 30" C. but on our determinations a t the lower temperatures, taking into consideration the estimated rate of expansion between 25" and 30" C. This table is given because the International Critical Table for 30" C. appears to us to be out of line with the results for 15" and 25" C. I n conclusion, we wish to disclaim any desire to criticize the work of the compilers of the glycerol tables published in the International Critical Tables. This work has been very admirably done and the best use has been made of the available material, all of which had to be given due consideration. It is our belief, however, that the tables are unsatisfactory as a working basis where accuracy in the fourth decimal place is necessary.
Improved Laboratory Rectifying Column' Melville J. Marshall DEPARTMENT OF C H E M I S T R Y ,
U N I V E R S I T Y OF
AGGED laboratory columns with reflux have been de-
L
scribed by Leslie,2 peter^,^ and others. Recently Marshall and 9utherland4have obtained figures demonstrating the advantages to be gained by using columns of this type instead of the plain Hempel variety. If the vapor coming from the column is totally condensed, one part returned as reflux and another part drawn off as product, as in commercial apparatus, a certain hold-up of reflux occurs, which decreases the sharpness of the cuts. For this reason, in columns previously described reflux has been obtained by partial condensation of the vapor from the column, the residual vapor being condensed in a separate condenser as product. This involves practically no hold-up of reflux, but it is difficult to control the reflux ratio to the best advantage. To avoid this hold-up, and a t the same time to permit easy regulation of reflux, a shallow, circular glass trough was first constructed above the column proper. The reflux overflowed from this trough into the column, a certain fraction being drawn off as product through a stopcock from the bottom of the trough. Later it was found that this arrangement could be considerably simplified, the hold-up decreased still further, and the insertion of a thermometer facilitated, which was rather difficult in the first type. This last arrangement is shown in the sketch. The column E is filled with beads and lagged in any suitable manner. The vapor passes through the side tube A , and is condensed in 1 Received
June 14, 1928. "Motor Fuels," p. 55.5, Chemical Catalog Co., 1923. IND. END.CHSM.,18,69 (1926). 1 Zbid., 19, 735 (1927).
2
*
BRITISHC O L U M B I A ,
VANCOUVER,
B. c .
the reflux condenser C. The reflux enters the column a t A and is quickly distributed by the beads uniformly through the column. The reflux ratio is controlled by adjusting the stopcock, which requires no grease, as it is sufficiently sealed by the condensate. I n order that the product may drain properly only a very slight depression is necessary in the side tube a t A . To decrease hold-up to a minimum the stopcock is placed as close to the side tube as possible. For general purposes a ground glass joint a t B is preferable, although in the experimental column a fused glass joint was used. This column has already been used in two investigations with marked success. It is possible to adjust the reflux ratio accurately and rapidly a t any value up to total reflux. Once adjusted, the column can be operated with very little attention. Under some circumstances, where the material is not decomposed by long boiling, the column can be left unattended for a time by simply closing the stopcock and operating a t total reflux without removing the heat supply. The method used commercially of operating a t total reflux to concentrate the lighter component a t the top of the column, when removing the last amounts of a fraction, can be employed with particular facility, because of the rapidity with which the reflux ratio can be altered.
Indian Forest Research Institute The Government of India has decided to appoint a committee of inquiry into the future activities of the forest research institute at Dehra Dun. The approval of the standing finance committee for an expenditure of $14,500 has been obtained t o cover the costs of the committee, which started work in November. The committee will consist of a chairman of administrative experience in India, a business man having knowledge of Indian conditions, a scientist, and a secretary chosen from the Indian forest service, Its work will probably not extend over a period of more than two months. It will report on the organization, working, and control of the institute. It will suggest methods of coordination to achieve scientific accuracy and also a system of training Indians for scientific research. The institute when finished will cost about $4,000,000and is expected to be one of the best organizations of its kind in the world.