66 8
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
proportion of the time. On charges of all borings, especially in case of hand charging, briqueting would be desirable. Bundling of wire and similar scrap is almost essential. An electric furnace is like a motor truck in t h a t the less time t h a t is lost in loading and unloading, and the greater the number of hours per day it can be used. the lower the cost per ton melted in the one case and per ton mile in the other. Cutting down on waste time pays even better with the furnace than the truck, because the more power the furnace uses in 2 4 hrs., the lower the cost of the power per unit, while gasoline for the truck does not come much cheaper when bought in larger amounts. Another way t h a t power can be saved in electric melting is t o operate' on a definite schedule of power used. The Ajax-Wyatt takes power a t so steady a rate t h a t a time schedule works just as well on that type, but the granular resistor furnace and the arc furnaces may vary considerably in rate of power input. After a few test runs, one can, for any particular alloy, particular proportion of ingot, scrap, and borings, and for any particular weight of charge, make out a perfectlpdefinite schedule of kilowatt-hours needed on each heat. I n I O hrs. operation the kilowatt-hour per heat will be higher in the morning and approach or reach a lower constant figure a t the end of the day. By adhering t o such a schedule and running on kilowatt-hour input instead of on a time schedule a t a supposedly constant but actually variable kilowatt input, the heat can be brought out with astonishing regularity as t o temperature. Each furnace should have its individual kilowatt-hour meter, and one readable t o certainly not less than 5 kw.-hrs., and better I kw.-hr. With individual kilowatt-hour meters the performance of each furnace in a battery, and of each furnace tender can be watched. The kilowatt-hour meter is as important as the speedometer on a motor car or truck. Not all furnace makers supply them, but each furnace should have one. Inasmuch as the heat losses through the walls and the electrodes are approximately constant even though the rate of power input may change, it is obvious t h a t the higher the rate of power input, the more of it is usefully employed in melting. Suppose a furnace takes IOO kw. and loses 35 kw. through shell radiation and electrode losses, 6 5 kw. then do useful work. If the same furnace is given 1 2 5 kw. i t may lose 3 7% kw. in shell and electrode losses, but 87% kw. dQ useful work. The furnace will do one-third more work in a given time a t the higher rate. The upper limit of rate of input is t h a t a t which the local temperature is so high t h a t refractories fail or t h a t local overheating of the metal and consequent loss of volatile metals occur. It is quite probable t h a t on the larger furnaces automatic control of power input would be desirable. From the considerations noted above it is plain t h a t the advantages of electric furnaces are most marked where a number of large furnaces or of small efficient furnaces can be used and especially where they can be used 24 hrs. a day. Rolling mills, smelting and refining plants, brass foundries of large manufacturing
Vol.
11,
No. 7
plants and the larger jobbing foundries are the plants which can most obviously use electric brass furnaces a t a saving in melting costs. Such plants should think twice before deciding to go on with fuel-fired furnaces. The smaller plants and jobbing shops having a small output and a variable one, especially if they are so located t h a t power will cost much over 1 l / 4 cents per kw.-hr., should think twice before installing a n electric furnace. Even this class of furnace users may often find t h a t the electric furnace will save some money in the long run, although under normal peace conditions the amount saved will be small compared t o what i t was during the war or may be a t present. Whether electric brass melting has or has not as yet made much headway, it a t least has a foothold and apparently a firm one. It will be surprising indeed if inside the next decade the ratio of brass melted electrically t o t h a t melted by fuel does not surpass t h e ratio of steel melted electrically t o t h a t melted by fuel. All metallurgists will agree t h a t the electric steel furnace is here t o stay. They will soon agree t h a t the electric brass furnace is here t o stay, too. MORSEHALL,CORNELL UNIVERSITY, ITHACA, N E W YORK
CORROSION TESTS ON COMMERCIAL CALCIUM CHLORIDE USED IN AUTOMOBILE "ANTIFREEZE SOLUTIONS" 1 By PAULRUDNICK Received May 14, 1919
I n view of the serious shortage and great increase in cost of glycerin and alcohol caused by the European war, the sale of calcium chloride solutions t o prevent
.M 0
so
.m
I
I"'
FIG. I-CORROSIONTESTS, SOLUTION A
freezing of the cooling solution in motor vehicles was vigorously pushed. Three proprietary products of this type were submitted for an opinion as to whether 1 Read before the Division of Industrial Chemists and Chemical Engineers, 57th Meeting of the American Chemical Society, April 7 to 1 1 , 1919.
July, 1919
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
669
day. The results are clearly shown in the curves obtained b y plotting the loss or gain in weight1 against the time of immersion. The most striking points developed are: I-Aluminum was attacked most severely, iron next, and copper least, as would be expected. 2-The rate of corrosion increased sharply on the eighteenth t o twentieth day of immersion. This is of special interest in view of the terms under which one of these products was offered for sale.2
$0
.IO
.o .IO
.20 .30 .IO
0 JO
.IO
20
0 .30
.IO
.40
.lo
IPON-COPPER-ALUMINUMTRIPLPT Al
.50.30
FIG.2-cORROSION
TESTS, SOLUTION J
such solutions might safely be used or whether harmful effects from corrosion might be expected. These products were accordingly tested for their effect on aluminum, copper, and cast iron by SUSpending polished plates of these metals in solutions of the concentration directed or furnished by the manufacturers. The plates were suspended in pairs of copper and aluminum, copper and cast iron, and aluminum and cast iron, and also in sets of all three, by means of copper wire attached t o the emergent ends of the respective plates. The solutions were continuously a6rated during the test period, and were kept at a temperature of 30' t o 40° C. The tests were continued for 2 8 days, the loss or gain tn weight of the plates being noted every other
FIQ.3-cORROSION
TESTS,SOLUTION T
The composition of the solutions varied from
28
to
30 per cent of calcium chloride and from 0.5 t o 3 . 0
per cent of magnesium chloride. Solution T contained formaldehyde. The reaction of the solution was determined by measuring their hydrogen-ion concentration according t o the well-known colorimetric method of S o r e n ~ o n . ~ 1 I n each curve gain in weight in grams is plotted above zero on the ordinate, loss of weight in grams below the zero. 2 The offer consisted of a 14-day trial. If a t the end of this time ne damage had been incurred, the product was to be paid for. I f any damage had occurred, the seller agreed t o have the necessary repairs made a t his expense and t o make no charge for the solution. a The formaldehyde was probably added to prevent corrosion in the hope t h a t it would act in the same way as in the case of hydrochloric acid and iron, a fact which was first brought t o my attention by Mr. A. V. H. Mory. The results show t h a t formaldehyde does not prevent corrosion b y calcium chloride. 4 Biochem. Z ,21'(1909), 131.
6 70
T H E JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY
Solution T was distinctly acid (Ph =4.5), solution J was practically neutral ( P h = 6 . 6 ) , while. solution A was distinctly alkaline (Ph = 8.0). The erratic behavior of solution A is probably due t o its alkaline reaction. Obviously calcium chloride is entirely unsuited for use in anti-freezing solutions in the cooling systems of motor vehicles in which aluminum or copper or both are in electrical contact with iron or steel. Rapid destruction, particularly of the aluminum parts, is t o be expected. Actual experience has abundantly confirmed the results of the tests described.l The writer desires t o acknowledge with thanks the assistance of Mary Hull, F. W. Waterman, and A. C. Haebich in carrying out the details of the work described in this paper. CHEMICAL LABORATORY AND COMPANY ARMOUR CHICAGO, ILLINOIS
Vol.
11,
No. 7
A NEW FORM OF DISTILLING BULB By J. S. MCHARGUE
Received April 23, 1919
Fig. I represents a new form of distilling bulb and Fig. 2 indicates t h e method of constructing the bulb from sections of glass tubing of different sizes. T h e T is made by sealing a section of tubing I O mm. inside diameter and 45 mm. long on t o a section 16 mm. inside diameter and 1 2 0 mm. long. A drain hole j mm. in diameter is blown out on opposite sides of t h e stem of the T, I j mm. from the arms and directly beneath them.
/OOmm.
-
I1 I
A SIMPLE WEIGHING BURETTE By DONALDW. MACARDLE Received April 8, 1919
I n many of the assays encountered in pharmaceutical work such as alkaloidal assay, a considerable degree of accuracy in titration is necessary. I n order t o avoid certain of the disadvantages of ordinary burettes, such as inaccuracy of bore, temperature variation, drainage, and difficulty of accurate reading, the writer has devised a simple weighing burette which obviates all corrections and allows nearly as rapid work as the ordinary style, with accuracy limited only by the sensitiveness of the end-point. T h e sketch is practically selfexplanatory. The glass stopcock was taken from a broken burette To use the burette, t u r n up with the cock closed, so t h a t the capillary is over the titrating vessel, and allow the heat of the hand t o force the standard solution into the capillary. Then when t h e cock is opened the liquid will flow freely, and there will be no danger of loss through the cock. When near the end-point, the cock may be closed and the solution forced i n drops by the heat of the hand. A predetermined quantity of solution cannot be delivered without repeated weighings, but a little practice will enable the user t o estimate the desired quantity within a few per cent. A I O O cc. flask with rubber stopper, cock, and capillary, weighs about 40 g. ANALYTICAL AND RESEARCH DEPARTMENT UNITEDDRUGCOMPANY BOSTON,MASSACHUSETTS 1 In addition to its corrosive nature, calcium chloride has anothei serious disadvantage, namely, the fairly good electrical conductivity of its solution. In the case of leaks or other accidents, short-circuiting of the ignition system may easily happen.
A section of glass tubing 60 mm. inside diameter and with walls 1.5 mm. thick is drawn off and blown out t o a uniform thickness and in the shape of a symmetrical dome a t one end. A hole of sufficient size t o admit the stem of the T-tube is blown out a t the apex of t h e dome. The stem of t h e T is inserted through this hole until the lower edges of t h e drain holes are opposite the rim of the hole in t h e large tube, and the two ?;.xes sealed together a t this point. The opposite end of the large tube is drawn off and blown out in such a w a y ac: t o complete a symmetrical bulb of uniform thickness and Go mm. in diameter. After blowing a hole of the proper size a t a point on the bulb opposite the one made for the stem of the T, a tube of 1 2 mm. inside diameter is sealed on t o make connection with a condenser tube. The form into which this tube is t o be bent may vary, depending upon the most convenient manner of connecting a distilling flask t o different types of condensers. I n making the first bend in this tube above the bplb i t is desirable t h a t i t be made in t h e direction t h a t will preserve a right and left symmetry with respect t o the arms of the T within the bulb.
