November. 1924
INDUSTRIAL A N D ENGINEERIhTG CHEMISTRY
1189
A Convenient Battery Power-Line Circuit for t h e Potentiometer’ By Alan Leighton BUREAUO F DAIRYING, DGPARTMENT O F AGRICULTURE, WASHINGTON, D. C.
LTHOUGH the potentiometer is an excellent piece of apparatus for the accurate measurement of minute electrical pressures, it has certain inconveniences arising from the changing voltage of its battery on continuous discharge. As would be expected, a freshly charged battery gives the most trouble. Moreover, the battery must be charged frequently, and for this purpose it should be removed from the potentiometer circuit, as it is not considered good practice to instal a double-throw switch which will permit it to be charged in position.2 I n short, a potentiometer system requires constant care. The author makes potentiometric measurements only occasionally and has found that in the press of other work his battery is frequently neglected. I n order to make certain that hit; potentiometer system was always in working condition the following circuits were devised : The potentiometer (Fig. 1) is connected through suitable resistances to the main 220-volt power line of the laboratory. A 2-volt battery is then “floated” across the terminals of the p~tentiometer.~This battery serves as a voltage regulator and is on continuous charge a t low rate. The battery will, of course, pick up the potentiometer load if the main line circuit is broken. The current, about 0.075 ampere, is led from the power line through one 25-wattJ 220-volt lamp. It then goes to the potentiometer terminal, where it divides, about 0.02 ampere passing through the potentiometer circuits and the remaining 0.056 ampere through the battery. This is just enough current to keep the 160-ampere hour, 17-plate, Westinghouse 2volt battery a t charge. After 3 months’ operation the specific gravity of the acid in the battery was 1.255 (full charge was 1.3). This means, then, that a t the 0.055-ampere charging rate just sufficient current is passing through the battery to overcome its internal discharge a t about half full charge. The battery is then constantly on the so-called “flat” of the discharge curve, the best place for good service on a potentiometer circuit. The two shunts come together again a t the other potentiometer terminal, when the current is again led through another 25-watt lamp and thence returned to the power line. The resistance of the potentiometer circuit is more than 100 ohms. We may consider the internal resistance of the battery as being of the magnitude of 0.01 ohm. This means that any voltage fluctuation in the current passing through the potentiometer system will be distributed a t the rate of 1 part in the potentiometer to 10,000 parts in the battery. If the aiithor’s calculations are correct, an error of only 0.0005 per cent in the potentiometer reading will thus be introduced by the normal 5 per cent voItage fluctuation which might occur in the power line between the time the potentiometer
A
1 Presented under the title “A Convenient Potentiometer Set-up’’ before the Division of Physical and Inorganic Chemistry a t the 67th Meeting of the American Chemical Society, Washington, D . C., April 21 t o 26, 1924. 2 Clark, “The Determination of Hydrogen Ions,” p 226. a The idea of “floating” a battery across the terminals of a potentiometer is apparently not new [see White, P h y s . Rev , 23, 447 (1906) I, although this paper had not come to the author’s attention a t the time the plan here outlined was put into operation. Dr. White, however, took his power from another storage battery.
is balanced against the standard cell and the time the reading is made. This error is too small to detect. The greatest error encountered would come from the complete cessation of the power-line current between the time the potentiometer is balanced and the time the reading is made. This error, with the present set-up and charge rate, has been shown to be about 0.06 per cent. This represents greater accuracy than is necessary in any of the author’s work, although it may be too great an error for some kinds of work. Of course, this error would not usually be encountered, since in good potentiometric practice it is customary to check up the potentiometer balance before and after each reading. (It will be apparent that no effort is being made to hold the battery potential absolutely constant from day to day by guarding against temperature changes of the battery, but rather to insure that there will be no voltage variation during the time that a reading is being made.) The magnitude of the possible error is interesting, however. It will be greater, the greater the resistance of the battery circuit and the greater the charging rate. E. Hill Turnoch, Jr., assistant chief engineer a t the Westinghouse Battery Company, suggests that the employment of a battery of lower capacity than that described above, with the resulting diminished charging rate, would minimize this error; also that a system might be worked out wherein the battery would be charged a t a lower rate while the potentiometer was in use, this rate to be boosted slightly when the system was idle. ’
A
IA c
I
FIG.I-WIRING DIAGRAM A A , 220-volt power-line terminals E , 2-volt battery P , Potentiometer circuit L, Lamps, 25 watt, 220 volt
Let us now consider the question of shielding. It so happens that the terminals of the laboratory 220-volt power circuit are each a t a potential of 110 volts with the ground; that is, the ground potential is intermediate between that of the terminals. The external resistances of the potentiometer circuit have been so balanced that the potential of the whole potentiometer system is so close to that of the ground that no harm can come to the instruments by an accidental ground on any part of the potentiometer circuit. This low instrument potential with respect to the ground also facilitates shieldingthat is, protection against those stray currents which may induce insidious errors in the voltage measurements. Another advantage is gained, in that if a switch is so arranged that the shield plates can themselves be grounded, the presence of any actual ground in the potentiometer circuit can immediately be detected by disconnecting the e. m. f. terminals separately and pressing the galvanometer key. If there is a ground the galvanometer coil will be deflected.
Analysis of Naphthalenesulfonic Acids and Naphthalene' Supplementary Report By W. S. Calcott, F. L. English, and F. B. Downing R. I. DU PONTDE NEMOURS Rr C o . , WILMINOTON, DEL.
N A recent paper by the authors upon the "Analysis of NaphthalenesuIfonic Acids and Naphthalene,"2 the detailed procedure of analysis and preparation of reagents for the analysis of naphthalenesulfonic acids was accidentally omitted when the article was printed. The methods, which consist in determining the naphthalene carbon present by oxidation of the naphthalene residue to phthalic anhydride, and the sulfonic acid sulfur present by isolation of the soluble barium sulfonates with subsequent gravimetric determinations of barium, are given here.
I
REAGENTS VANADICACID SOLUTION-TOprepare 1 liter of the oxidizing solution, transfer 63 grams of pure ammonium metavanadate (NHdVOs) to a 2.5-liter beaker and add 220 CC. of distilled water. T o this mixture add gradually with agitation 780 cc. of concentrated sulfuric acid. Cool the solution to room temperature and preserve it in a glassstoppered bottle. BARIUM CARBONATE-The barium carbonate used in this determination of sulfonic acid sulfur must be free from all soluble salts. Satisfactory material may be obtained by boiling 200 to 300 grams of pure precipitated barium carbona t e with 1 liter of distilled water, filtering, and washing the barium carbonate with hot water. ANALYTICAL PROCEDURE DETERMINATION O F NAPHTHALENE CARBON (Gio)-If SUIfites are present in the sample it should be first boiled in acid solution to eliminate SOZ. Pipet a 25-cc. aliquot of the sample containing 0.25 to 0.30 gram of the naphthalenesulfonic acid into a 200-cc. Erlenmeyer flask and add a t room temperature 90 cc. of the vanadic acid reagent. Insert a thermometer in the flask and heat the mixture to 120" C. in not less than 10 minutes. Maintain the temperature of the oxidation a t 120" C. * 2" for exactly 15 minutes, then immediately pour the solution into 200 cc. of cold water in a 750-cc. Erlenmeyer flask. Rinse the small Erlenmeyer with cold water, cool the drowned mixture to room temperature, then transfer it to a 500-cc. volumetric flask and dilute to the mark with distilled water. Titrate 150-cc. aliquots of the solution in 200 cc. of water a t 70" to 80" C. with 0.1 N permanganate solution. The permanganate may be run in rapidly until the solution changes i n color from green to greenish yellow. Then add the permanganate in 2 to 3-drop portions until an end point permanent for 20 seconds is obtained. A blank determination should be run upon the oxidizing solution using 25 cc. of distilled water in place of the sample 1
Received September 25, 1924. 16, 27 (1924).
* THISJOURNAL,
aliquot. Reagent made from good quality ammonium metavanadate will have a blank of not more than 1.0 cc. of 0.1 N permanganate. Calculation: (Cc. KMnOa-blank) X normality o f KMnOd X 0.667 = Weiaht of samole i n aliauot titrated Per cent naphthalene carbon
DETERMINATION OF SULFONICACID SULFUR-Pipet a 100-cc. aliquot of the sample (having the same concentration as the aliquot used for the naphthalene carbon determination) into a 1-liter beaker, add 400 cc. of hot distilled water and 5 grams of washed barium carbonate. Cover the beaker and boil the mixture for 30 to 40 minutes, then filter off the barium sulfate and excess carbonate and wash the cake with cold water. Dilute the filtrate and washings to 800 cc. in a liter beaker, heat to boiling, and precipitate the barium by adding 40 CC. of 5 per cent sulfuric acid (by volume). Digest the solution until precipitation is complete, then filter off, ignite, and weigh the barium sulfate. Run a blank determination on the barium carbonate using 5 grams of the carbonate and 500 cc. of distilled water. Treat this determination exactly like the determination of the sample. Calculation: The degree of sulfonation is then expressed by the following formula, where a result of 1.00 indicates monosulfonation, 2.00 complete disulfonation, and 3.00 complete trisulfonation.
DISCUSSION OF THE ANALYTICAL PROCEDURE The procedure for naphthalene carbon applied to ordinary mixtures of sulfonic acids is considered accurate within 1 per cent of the amount present. The analysis is open to the objection that other oxidizable material present will be determined and calculated as naphthalene carbon. I n practice, however, this objection has been found more theoretical than actual; the conditions of temperature and high concentration of oleum in the process of sulfonation operate to eliminate easily oxidizable organic materials, and it should be noted in this respect that each mol of naphthalene residue requires 18 equivalents of oxygen, a ratio which causes the presence of inorganic oxidizable impurities to introduce relatively small errors. The analysis for degree of sulfonation where the estimation of sulfonic acid sulfur is included is considered accurate within 2 per cent. This method is not applicable to the salts of the sulfonic acids or to mixtures containing salts such as sodium sulfate.
