T H E JOURNAL O F INDUSTRIAL A N D ENGINEERING CHEMISTRY
Aug., 1918 IOO
a
0.02
(e--)]
[
]
1,000,000
(6)
2,240 g a = Price per long ton (in dollars) of the coal as received. b = Per cent water in coal as received. c = Cost per long t o n (in dollars) of the dry coal. d = Per cent of ash on the dry basis. e = Per cent ash selected as t h e standar.d. f = Cost per l F g t o n (in dollars) of the dry coal corrected for ash. g = B. t. u. per lb. dry coal. x = Cost per million B. t. u. (in dollars). The theoretical considerations upon which these equations are based are all discussed in the previous paper. Equations z and 5 apply when the ash is greater t h a n standard and 3 and 6 when the ash is less than standard.
__( d - e ) 2 is an algebraic expression
629
This chart shows how even a very complicated equation, such a s 5 , involving several multiplications and divisions, as well as additions, subtractions, and a square can be readily and easily solved by a properly constructed alinement chart. I t is impossible in this paper t o go into the mathematical details governing the construction of the chart, but the reader is referred t o “A Manual of Chemical Nomography” b y Dr. Horace G. Deming’ for information which should make the matter clear. The chart is, in fact, a sort of adaptation of the calculating device known as the nomon.2 ATLANTICSUGAR REFINERIES,LIMITED ST. JOHN,N. B., CANADA
NOTE ON THE USE OF THE DIPPING REFRACTOMETER B y WYATTW. RANDALL Received June 3, 19 18
Experiments recently made in this laboratory seem t o the writer t o justify the publication of a note of which happens t o represent almost exactly the price warning t o chemists who may have occasion t o use the deduction t o be made for excess ash as given in the dipping refractometer for exact determination of the tables of the Bureau of Mines Bulletins, and accounts refractive properties of liquids, especially where the for the increased labor charges, diminished efficiency latter are rather volatile. of combustion, etc., resulting from high ash coal. The Two samples of whiskey containing a n unusually values which the Government deducts from the price per ton t o be paid we add t o t h e cost per ton. When low percentage of alcohol were under examination by the ash is below standard a premium of 2 cents per Mr. C. 0. Miller. The density of each of the alcoholic ton for each whole ‘per cent less is paid. This ex- distillates having been detkrmined with the aid of the plains Equation 3. Equation 5 is a combination of I , pycnometer, the refractometer reading a t 2 0 ° C. was made as a means of estimating the amount of methyl 2 and 4, and Equation 6 of I , 3 and 4. alcohol, should any be present. I n order t o prevent The tedious arithmetical calculations which would any inaccuracy of reading through evaporation of be required t o solve these equations are all eliminated alcohol, in each case the distillate was placed by Mr. by the use of the chart shown herewith. The directions Miller in the metal cup secured by a bayonet joint for use are given in the cut. T o illustrate its use supto the instrumznt. The readings gave a percentagepose i t i s desired t o know which of two coals, A or B, of-alcohol-by-weight which differed notably from t h a t is the more economical, t h e prices and analyses being found by the use of the pycnometer, and which indias follows: cated the presence in each distillate of about 1.25 A B Price per long ton . . . . . . . . . . . . . . . . . . . 4.54 $5.38 per cent of methyl alcohol t o 98.75 per cent of ethyl. 3.50 Per cent water.. ................... 4.30 Similar results were obtained by two other chemists, P e r c e n t dry a s h . ................... 10.40 6.00 working independently. As the presence of methyl B. t. u. per Ib. d r y coal.. ............ 13,550 14,350 alcohol in any noticeable amount in these whiskeys By means of a ruler, a drawing triangle, a fine silk thread, or was a matter of importance, all the distillations and best of all a strip of celluloid with a straight line ruled on its dkterminations were carefully repeated by Mr. Miller; under side, connect 4.54 on the left-hand axis with 4 . 3 on while the general results were the same, the figures the inclined axis. The line intersects the “Price per dry ton” obtained were not as close as was considered necessary axis at 4.74. If 6 is taken as the standard ash, this point, 4.74, in a case in which much was a t stake. Accordingly, is then connected with 4.4 on the lower part of the ash axis and the writer obtained fresh distillates, determined their the line cuts the left-hand axis again at 4.84, the cost per dry ton corrected for ash. This point is connected with 13,550 on respective den‘sities and refractometer readings, using, the inclined axis and the desired result, 0.1595,read at the in- however, in the lattsr work, glass beakers instead of tersection with the “Cost per million B. t. u.” axis. Proceeding the m-tal cup, in the belief t h a t evaporation would similarly with B, we obtain 0.1734 as the cost per million B. t. u. play a very small part in the case of a 2 5 per cent A‘is therefore the cheaper coal and the extra price of B is greater alcohol a t zoo C. The beakers were of course corked than justified by its better quality. Or, if it were desired to while they hung in the bath, and the corks were withknow what price should be charged for B to have the heat cost drawn only when alcohol and refractometer prism were equal to that of A, we would start with 0 . 1 5 9 5 , the cost of a both unquestionably a t 2 0 ’ C., t h a t is, about after half million heat units in A, and work backwards on the analysis of a n hour’s immersion in the bath. The readings gave B obtaining $4.95. The intermediate values obtained, if not interesting, need not be noted at all, the straight line being no evidence of the p r s e n c e of methyl alcohol. The merely pivoted over the point of intersection. The first original distillates were reexamined, this time using alinement solves Equation I , the second either 2 or 3, and 1 Unirevsity Press, Champaign, Illinois. 2 J. A m . Chem. Soc., 39 (1917), 2137. the third 4. 200
’
.
~
630
T H E J O U R N A L OF I N D U S T R I A L A N D EiVGINEERING C H E M I S T R Y
glass beakers and, again, no evidence of the presence of methyl alcohol was obtained. The only simple explanation of these differences is t h a t the temperature of the liquid in the metal cup was distinctly higher t h a n t h a t of the bath. Accordingly, the following experiments were made: The bath was brought to, and maintained a t , 20' C., according t o bath thermometer A (brass-jacketed, about I O in. long and graduated in tenths of degrees). Comparison of A with thermometer B (not jacketed, about 4 in. long and graduated in fifths of degrees), while the bulbs of both were in the bath, showed no noteworthy differences. I-An alcohol of about 2 2 . 7 5 per cent by weight was put in a glass beaker and also in the metal cup. Both these vessels were then fitted with corks and placed in the bath. After about 2 j min. thermometer B was inserted in each in turn, after the cork had been withdrawn and readings obtained as follows: Temperature of alcohol in glass beaker . . . 20.02' C. Temperature of alcohol in metal c u p . . , . . . 20.03' C . 11-Readings made with the refractometer dipping in the alcohol in the glass beaker were: 5j.78, j5.80, 53.79, 55.79, 55.78, jj.?g-Average, 55.788, which corresponds t o 2 2 . 7 5 per cent by weight. 111-Readings made with the alcohol in the metal cup, follom7ing Zeiss filling directions, were : 5 5.49, 55.49, 55.47, 55.49, 55.50, 55*48--Average, 5j.487, which corresponds t o 22,60 per cent by weight. IV-Readings made with the alcohol in the metal cup, the latter having been filled while off the refractometer and then clamped on, were: 55.55, 5 5 . 5 2 , 55.56, 55.54, 5 j.5s-Average, 55.543, which corresponds t o 2 2 . 6 2 per cent by weight. V-After these readings had been made, the metal cup was detached and the temperature of the alcohol contained in i t quickly read. Temperature of bath, thermometer A , . . . . . 20.0' C . Temperature of bath, thermometer B . . . . . . 20.0' C . Temperature of alcohol, thermometer B . . . 20.5' C. VI-Readings made with wat2r in a n open beaker were: 14.65, 14.67, 14.67, 14.66, 14.67, 14.66-Average, 14.663. VII-Readings made with water in the metal CUP, filled as directed by Zeiss, were: 14.57, 14.56, 14.57, 14.58, 14.57, 14.58-Average, 1 4 . 5 7 2 . VIII-After these readings had been made, the metal cup was detached, and the temperature of the water in it quickly determined. Temperature of water in bath.. . . . . . , , . . . . 20.0" C. Temperature of water in cup., .. . . . . . . . . . . 20.6' C . The temperature difference in V appeared t o be about o . ~ ' ,in VI11 about 0.6' C. I believe these are distinctly too great. I n the effort t o make the reading quickly, I do not believe time enough was given for the liquid and the thermometer t o come t o equilibrium. Besides, the metal cup was probably somewhat warmed in detaching from the refractomzter, t h e quantity of liquid was small, and the whole bulb of B was not immersed; I believe, therefore, t h a t about 0.4' or 045'
Vol.
IO,
KO.8
would be nearer the true difference. The effect of a ' C. difference in temperature was tried. IX-Readings of alcohol in a n open beaker a t 20.5' C. were: jj.40, jj.41, 55.40, 55.41, 55.41, 5 j.q~-Average, 55.407, which corresponds t o 2 2 . 5 5 per cent by weight. It thus appears t h a t t h e low readings obtained when the closed metal cup is used are due, a t least chiefly, t o a difference in temperature between the water of the bath and the liquid in the cup, and t h a t a similar difference in temperature does not exist when a glass beaker is used instead of the metal cup. It now became a matter of interest t o learn under what conditions the d a t a were secured upon which Leach and Lythgoe based their method for the detection and estimation of methyl alcohol in the presence of ethyl. Inquiry of Dr. Lythgoe brought word t h a t glass beakers only had been used in their work. The next question was, What is the cause of this difference of temperature in the contents of the metal cup, according as i t is attached or not attached t o the refractometer? I was informed by Dr. W. J. A. Bliss t h a t , a t the Johns Hopkins physical laboratory, Dr. Pfund had found t h a t , in standardizing a dipping refractometer, complete accord in the readings could be secured only when the temperature of the room was close t o t h a t of the bath. This suggested t h a t heat was conducted by the metallic parts of the instrument and of the cup t o the contents of the cup. Accordingly, Dr. Bliss and I made readings of the two thermometers when the bath was only I' t o 1.5' colder than the air. No effort was made t o keep the bath a t a fixed temperature; B's readings in the water of the bath averagzd 0.07' higher t h a n A's. I n the liquid in the metal cup, immediately aftdr detaching it from the refractometer, B's readings averaged 0.10 O higher than A's readings in the bath, which indicates t h a t the contents of the cup wzre, under these conditions, only about 0.03' warmer t h a n the surrounding water of the bath. Later, I cooled the bath to about ~ j while ' the B, hanging alongside the room temperature was 27'. cup, with the bottom of its bulb about a half inch above the water of the bath, read 2 2 ' . When placed side-by-side in the water of the bath, A read 15.00', B, 14.92'. (Probably A's brass jacket was keeping its rzadings somewhat higher t h a n B's.) The temperature of the bath was slowly rising: when A read I j . I 5 ', B, placed in the alcohol in the cup just after i t was detached, gave a reading of 15.65', quickly falling t o 15.50". These rather rough experiments seem t o bear out the conclusion t h a t heat is conducted from the air to the contents of the metal cup through the metal parts of t h e instrument, and t h a t the difference in temperature between the contents of the metal cup and the water of the bath is roughly proportional to the difference between the air temperature and t h a t of the bath. 0 .j
LABORATORY O F THE
STATE OF MARYLAND DEPARTMENT BALTIMORE,
OF
MARYLAND
HEALTH
Aug., 1918
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 DECANTING
A DEVICE TO INSURE TIGHT CONNECTIONS BETWEEN GLASS AND RUBBER TUBING
By H. TILLISCH Received January 10, 1918
The decanting of liquids from residues is generally a very tedious operation. T h e time required for the separation can be much abbreviated without mechanical means, if t h e t u b e simply is held in an inThe liquid clined position, e. g., a t an angle of 4 j ’. will then-form a channel in t h e upper part of the tube, while t h e residue will go along t h e under part t o the bottom. I n this way the two currents, upwards and-downwards, will be separated from each other.
63 1
By C. C. KIPLINGER Received April 17, 1918
I n gas analysis trouble is experienced frequently in t h e attempt t o make tight connections between glass and rubber tubing. Experience has shown t h a t this is accomplished best by wrapping a single t u r n of wire about the joint and twisting tightly. However, there are two objections t o this method. The wire tends, if twisted tightly, t o cut t h e rubber, and if the rubber tubing is appreciably over-size, the tubing is compressed or pinched near the twisted portion of the wire, frequently making a small channel through which leakage occurs. The device heredith described overcomes these difficulties, permits the use of over-size rubber tubing, and insures gas- and water-tight joints. It has been used throughout the year with Liebig condensers and gas apparatus and has given complete satisfaction. A is a piece of stout wire bent in U form of such size t h a t the limbs of the U will just slip over both tubes. A loop of stout cord is tied about t h e connection, the wire U is slipped through this loop as shown in dotted lines, the cord now twisted, using c==:==: t h e wire as a lever, and as --A soon as t h e joint is tight, the U is turned as shown a t A. Cord is better than the usual copper wire for this purpose in t h a t the former distributes the force more uniformly throughout its length. A further advantage of this mode of attachment lies in the ease with which it may be dismantled, requiring as it does no pliers or other tools for this purpose. I====-
-..
344 HARRISON AVENUE LEXINGTON, KENTUCKY
A SIMPLE AND ENTIRELY ADJUSTABLE RACK FOR KJELDAHL DIGESTION FLASKS 0
The decanting operation can thus be finished in onethird of t h e time required by the usual method of using vertical tubes. It is common practice t o use narrow inclined tubes, etc., for obtaining a rapid decanting in liquids or for separating dust from air. The same principle can, as shown, be of use i n the 1abo;atory. The common tube holders ought t o be slightly modified for easy decanting in inclined tubes. The modification is suggested in t h e diagram. AARHUS,DENMARK
By FRANK E. RICE Received iMarch 11, 1918
The apparatus here described can be made by any pipe fitter from standard pipe, and unions, and without any specially prepared parts. I t will be found t o cost much less t h a n similar equipment on the market. I t takes up but little space when in use, and its great flexibility in adjustment makes easily possible still further contraction when it is not being used. A A’ is a n iron pipe in which are mounted burners, a , each with a stopcock. At t h e ends of this pipe are found stopcocks, b , for gas intake. This line is adjustable up and down on standard B B’, which is in turn adjustable forward and back on support C C’. An iron rod, D D’, is adjustable up and down on standard E E’, which is also in t u r n adjustable for-
