US2847641A - Method for analyzing white fuming nitric acid - Google Patents

Description

l1g 1.2 1958 vJ. D. CLARK v 2,847,641

METHOD FOR ANALYZING WHITE FUMING NITRIC ACID IN VEN TOR.

J. D. CLARK Aug. 12, 1958 METHOD FOR ANALYZING WHITE FUMING NITRIC ACID Filed Dec. l, 1953 3 Sheets-Sheet 2 J. D. CLARK Aug. 12, 1495s' METHOD FOR ANALYZING WHITE FUMING NITRIC CID Filed Dec. 1, 1955 s sheets-smet :s

20a 25a aan 354 United States Patent() METHO/D FOR ANALYZING WHITE FU'MING NITRIC ACID yJohn D. Clark, Kinnelon, N. J.

Application December 1, "1953, Serial No, 395,626

3 Claims. (Cl. 324-30) (Granted under yTitle 35, U. S. Code(1952), sec. 266) The invention described herein may `be manufactured by or for the Government of the AUnited States of America for governmental purposes Without the payment of any royalties thereon or therefor.

This invention relates to a method and apparatus -for analyzing white fuming or anhydrous nitric acids, and especially relates to the lanalysis of the acids outside -the laboratory and by relatively unskilled operators.

As far as is known, no iield method for the analysis of white fuming nitric acid has been provided. The laboratory methods are unsuitable for eld work, re-

quiring standard solutions, fragile glassware and skilled manipulation.

One object of this invention is to provide a method and apparatus usable in the range of to 5% water .content and 0 to 2.5% NO2 content of the acid.

Another object of this invention is Vto maintain an accuracy up to 0.1% in VH2O content yand 0.1% in NO2 content.

Another object of this invention is to `provide analysis apparatus which is insensitive to such small amounts of metallic impurities as might be present in the acid.

Another object of this invention is to provide apparatus which is reasonably insensitive to outside conditions such as small variations in the ambient temperature.

Another object of this invention is that the apparatus be suitable for use by untrained personnel.

Another object of this invention is to ,provide .apparatus which is relatively compactand light `in weight but which, at the same'time, is rugged and durable.

Other objects and many of the Aattendant advantages of this invention will 'be readily appreciated' as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying `drawings wherein: p

Fig. l shows a preliminary chart relative to -the NQZ concentration of an acid, used in obtaining a calibration chart for H2O and NO2 concentrations.

Fig. 2 is a second preliminary chart illustrating the percentage of H2O vs. NO2.

Fig. 3 is a preliminary chart relative tothe H2O concentration.

Fig. 4 is a calibration chart upon which the readings on the dial are entered to obtain the H2O and NO2 con centrations.

Fig. 5 is a schematic `view of `a `bridge circuitused in the invention.

Fig. 6 is a front elevational view of the face of a dial used in conjunction with the circuit of "Fig, "1.

Fig. 7 is a calibration curve used Ato determine thedial angles in Fig. 2.

In order to analyze a `sample of white fuming nitric acid, two values must be determined, the percentage of H2O and the percentage of NGZ. The HNO3 may, if desired, be determined by difference, if metallicimpurities are absent or negligible in quantity.

-ln order to determine two independent variables, itis necessary to make two measurements. Sincethe 'classi- 2,847,641 Patented Aug. 12, 1958 cal titrimetric methods of analyzing acids are unsuitable for field use, it was therefore desirable and convenient, in this case, to measure two physical properties of the unknown acid. It was also necessary that these measurements be of Vthe same type, requiring the same apparatus. A `most suitable physical measurement is that of electrolyticconductivity since this is easilyand precisely meas- `urable and is relatively insensitive to temperature.

The conductivity of nitric acid is a sensitive `function .of the `water content. There is a minimum. in the conductivity curve in the region of 2.5% H20 and -a very broad maximum near H2O. Small quantities of .NO2 increase the conductivity in the region of 0-l-0% H2O to `an extent that varies with the H2O content. It

was, therefore, determined ,that three conductivities would be taken on a sample of acid with the same conductivity cell, one on the original acid and the other two on the original `diluted with water to dilerent and known diluwhich `would be treated as :the two quantities from which the two .unknowns (H2O and NO2) could be deduced. -As the three actual measurements are vreduced to two ratios, the constant of the conductivity cell cancels out and, in effect, an automatic calibration of the cell is :performed This procedure, likewise, tends `to reduce the temperature eilect since the three .conductivities might be expected to vary in the same direction with temperature, although not necessarily at the samerate.

The three conductivities are taken upon the Afollowing samples, Vor samples made up in the same proportions: (A) the original acid, (B) cc. of the original to which have been added 5 cc. of H2O, and `(C) 75 cc. of `H20 to whichhave been added 25 cc. of the original unknownacid.

rllheiinishedcalibration was visualized as augridwwhose coordinates would correspond to the two conductivity yratios and B- crossed by two sets of contour lines corresponding to the .H2O and NO2 content. lIt was found empirically `that the best spread of these contours would -be obtained when t lthe coordinates were log A-A and log lyi (where 4M corresponds to measured conductivity). .In practice, the conversion from resistance to conductivity was omitted and the coordinates finally chosen were AR measured R specific and an Industrial Instruments Co. RC-1 conductivity bridge. Later, the RC-l 4bridge was replaced with `a log and log C ibridge designed for the purpose, which is described later `in this application. The sample to be measured was Aheld `in a test tube just large enough to enclose the dip cell. It was found that with this cell the measured samples need be only 40 cc. in volume.

140 samples of acid were used to cover the field of -5% H2O and 0-2.5% NO2. The unit operation involved a series of ten samples, made up as follows: Two 500 cc. samples of nitric acid X and Y were prepared,

Two curves were now drawn:

log RA/RB vs. NO2, and log RB/RC vs. NO2 These are shown iii chart 2 illustrated in Fig. 1. From with fairly similar water content, but with, in one case, 5 the SmOOthed Curves and from the NOT-H20 line an little or no NO2, and in the other 2.5% or 2,6%. The interim table for the series was constructed as follows: ten samples were then made up from these as follows: Table Il Sample No. cc. X cc. Y NO: 10g R24/RB log Ra/Rc Percent 0 100 10 90 0. 327 1. 325 4. 45 80 0. 309 1. 312 4. 45 70 o. 294 1. 29s 4. 46 55 0. 280 1. 256 4. 47 45 0.267 i. 274 4. 4s 30 0. 256 1. 262 4. 4s 20 0. 245 1. 251 4. 49 i0 o. 255 1. 241 4. 50 0 0. 225 1. 25o 4. 50 0. 216 1. 220 4. 51 0. 207 1. 211 4. 52 l 0. 19s i. 202 4. 52 The exact proportions of X and Y in any sample were g-gg g unimportant, but it was convenient to have the ten spread 01176 11176 .[55 fairly uniformly over the range. 0- 169 1- 158 4- 55 o. 163 1. 160 4. 56 Each of the ten samples was now diluted appropriate- 0,157 1,152 4.57 1y. 0. 157 1. 152 4. 57 0.145 1.137 4. 5s (a) Some 40 cc. were set aside 1n a 50 cc. iodine 0.139 1.129 4 59 flask for the A subsample. 0-134 1-122 L60 0.128 1.115 4.61 (b) 50 cc. were run into a similar flask with an auto- 0.124 1.108 4.6i matic vacuum pipette and 2.5 cc. of H2O were care- 0-119 1-101 M2 0.115 1. 094 4. 63 fully run in below the surface by means of a burette with a long capillary tip. This was the B subsample.

(c) 30 cc. of H2O were run into a third flask, and 10 cc. of the acid were added with the automatic pipette. This was the C subsample.

All of the volume measurements were made with reasonable care, but the most critical volume was the 2.5 cc. of H2O added to B, since a very small error in the amount of water added will lead to a considerable variation in the conductivity.

The samples were then mixed by swirling, and placed in a 25 water bath to equilibrate. The resistance measurements were then made (at the same temperature) and the log RA/RB and log RB/RC parameters were computed and tabulated.

The A subsample was then analyzed for H2O and NO2 by highly rened classilical methods. The NO2 content is determined directly, and is probably accurate to 0.01%, but the precision of the determination of the water content by difference, approaches 0.03% (standard deviation), and the water content, therefore, cannot be taken ydirectly from the analytical results.

The assayed H2O content of the series was therefore plotted against the NO2. Under the conditions under which the specimens were made up (all ten from varying proportions of two large samples) it can be demonstrated that the H2O content must be a linear function of the NO2. The best line was therefore drawn through the 20 points (2 assays on each of 10 samples) by the method of least squares, and the H2O values accepted were those taken from the line and not from the points. Under these circumstances, the standard deviation from the true value of water content was less than 0.01%. The tabulation was as follows:

The process was then repeated for the next series, and so on until the field was covered. Fig. 2 shows a chart 4 illustrating the coverage attained, each line correspending to a series of ten samples. (Minus values of H2O correspond to acid containing excess N205.) From the whole series of interim tables, an isonox table was constructed, taking one point from each series. A typical isonox table is shown in Table Ill, for the isonox corresponding to zero NO2 content.

Table IIL-[sonos: 0.0

Series log RA /RB 10g Ria/Rg Percent o0, E. 799 1 715 0. 65 9. nos 1. 704 0. 6a 0. 118 1. 603 l. 79 0. 132 1. 594 1. 79 0. 2151 1. 511 2. G2 0. 292 1. 482 2. 89 0. 333 l. 359 4. 09 0. 322 1. 437 3. 32 0. 327 l. 325 4. 45 0.310 1. 279 ."n 09 Q 276 l. 494 2. 7i) 2. 675 l.. 732 0. 31 2. 987 1.653 1.18 0. 4.17 i. 771 0. 29

These values of log RA/RB and RB/RC vs. H2O were then plotted, smoooth curves 5 and 6 drawn through them, as illustrated in chart 8 in Fig. 3, and the iinal log RA/RB and log RB/RC vs. H2O values taken from the curves.

These values were then plotted upon the final calibration chart, and the isonox drawn in. This was repeated for the other isonoxes; the intersecting isodamps constructed, and the chart completed, as shown at 10 in Fig. 4. It will be noted that on the completed chart the coordinates are labelled l-l-A-B and l-C rather than log R A/RB and log RB/RC. This is because the special bridge built for the purpose is calibrated in logarithm of ohms rather than in ohms in order to reduce the amount of computation necessary, and in order to get more uniformly spaced dial graduations. The resistance of the A and B samples is between 100 and 1000 ohms, and that of the C is about 14 ohms, using the cell described. Only the mantissa of the logarithm d is on the dial, so that l-l-A-B is defined as l plus reading A minus reading B, and B-C is simply read ing B minus reading C. This avoids negative values for the coordinates and makes the use of the chart a com pletely mechanical process.

The chart was calibrated at C., and it was necessary to determine the effect of the temperature at which the readings were made upon the calibration. Readings were made at various temperatures between 20 and C. on several samples Whose composition varied Widely within the limits of the eld. It was found that since the temperature coeflicient of conductivity Varied with the composition, no simple temperature correction could be applied, but that the coeihcient was sufficiently small that a variation of 22 C. from 25 would lead to an error of less than 0.1% in H2O or NO2.

From the values, therefore, of the three resistances, the H2O and N02 content may be determined from the chart.

lt is possible to use an ordinary conductivity bridge to malte the measurements, but the calculation of m RC is inconvenient and may lead to error. ln order to satisfactorily use the method of this invention it is necessary to have a bridge with the following characteristics:

(l) It should be capable of measuring any resistance that would be encountered in the test procedure, but should have no superfluous ranges; (2) The dial readings should be applicable to the calibration chart with a minimum of intervening computation; (3) The sensi-- tivity (to variations in resistance) should correspond to the closeness of reading possible with the chart; (4) The dial graduations on the instrument should be as uniformly spaced as possible to facilitate interpolation by an unskilled operator, and there should be only one set of graduations on the dial; and (5) The bridge should be as small, simple in operation and rugged as is consistent with the other qualities.

The bridge, illustrated at 12 in Fig. 5, eliminates the need to calculate log and log log TE- and RB resistance Ito in series with the end resistors ZZ and 24 which are themselves connected in series with a potentiometer 26 connected to the conductivity cell through an ampliiier The bridge is designed to operate on 110 volt, 60 cycle current, a though a battery operated bridge could be substituted, if it were considered desirable. The variable capacitance 3@ is provided to sharpen the balance point of the instrument, since with cycle current the imaginary component of the current is appreciable.

Since the a and b resistances run in the order of hundreds of ohms, and the c resistance is usually near 14 or l5 ohms, the bridge is provided with the two resistance ranges 14 and lr6, the resistance 14 being the low range and the resistance 16 being the high range. By proper choice of Athe resistances in the bridge, it is possible to calibrate the bridge so that the dial 32 illustrated in Fig. 6, connected thereto reads directly in the mantissa of the logarithm of the measured resistances, the dial graduations, as indicated by the curve in chart l34., illustrated in Fig. 7, being nearly uniformly spaced.

Since only logarithms are being dealt with here it is d necessary only to subtract one reading from another in other to obtain logarithm values to be used on the chart 10.

'in practice, a tube, representing the A value, is filled with the unknown acid to approximately the same level as the B and C tubes. The B and C samples are first stirred, then all three tubes are capped and placed in a water bath lled with water at approximately 25 C. (if). The bridge is then turned on and allowed to Warm up while the samples come to temperature. With the bridge on the high range, with resistance 16 connected, the cell is immersed in the A sample, the bridge balanced, and the A reading recorded. With the bridge still on the high range, the process is repeated on the B sample. It is not necessary to rinse the cell between the A and B readings. The bridge range is then shifted to the low range with resistance 14- connected, the cell is dipped into the water of the water bath, allowed to drain a moment and then immersed in the C sample, after which the bridge is balanced and the reading recorded, taking care to take the reading from the C scale on the chart 10.

It should be noted that the A-B range on the chart is provided in the form of 1-l-A-B'- This is donc because it is desirable to keep this value positive at all times and this would not be the case in the f1-B range if the value of B is larger than the value of A.

Obviously' many modifications and variations of the present invention are possible in the light of the above teachings. it is, therefore, to be understood that within the scope of the appended claims the invention may be practiced otherwise than as specifically described.

What is claimed is:

l. The method of analyzing white fuming nitric acid consisting in dividing the acid into three parts, one part constituting the original acid, a second part constituting a rst concentration of the acid in water, and a third part constituting a second concentration of the acid in water, bringing each of said parts to the same temperature, measuring the electrical resistance of each part, determining the ratio of the value of the electrical resistance of the first part to the value of the electrical resistance of the second part, determining the ratio of the value of the electrical resistance of the second part to the value of the electrical resistance of the third part, and, thereafter, evaluating these ratios according to a previously calibrated chart having one series of curves representing the percentage of water in the acid and having another series of curves representing the percentage of nitric dioxide in the acid.

2. The method of analyzing a liquid for a plurality of contaminants consisting of dividing said liquid into a plurality of parts at least equal to the number of contaminants analyzed for, diluting each of said parts except one part with unequal amounts of Water to thereby obtain a plurality of dilutions, and measuring the electrical conductivity of said plurality of dilutions.

3. The method of analyzing white fuming nitric acid for nitric dioxide and water consisting of dividing said acid into a plurality of parts, diluting each of said parts except one part with unequal amounts of water so as to form a plurality of dilutions, and measuring the electrical conductivity of said plurality of dilutions.

References Cited in the tile of this patent UNITED STATES PATENTS 2,364,898 Hassler Dec. 12, 1944 2,367,949 Langer lan. 23, 1945 2,599,413 Reichertz June 3, 1952 2,680,834 Burns et al. June 8, 1954

Claims (1)

1. THE METHOD OF ANALYZING WHITE FUMING NITRIC ACID CONSISTING IN DIVIDING THE ACID INTO THREE PARTS, ONE PART CONSTITUTING THE ORIGINAL ACID, A SECOND PART CONSTITUTISNG A FIRST CONCENTRATION OF THE ACID IN WATER,S AND A THIRD PART CONSTITUTING A SECOND CONCENTRATION OF THE ACID IN WATER, BRINGING EACH OF SAID PARTS TO THE SAME TEMPERATURE, MEANSURING THE ELECTRICAL RESISTANCE OF EACH PART, DETERMINING THE RATIO OF THE VALUE OF THE ELECTRICAL RESISTANCE OF THE FIRST PART TO THE VALUE OF THE ELECTRICA RESISTANCE OF THE SECOND PART, DETERMINING THE RATIO OF THE VALUE OF THE ELECTRICAL RESISTANCE OF THE SECOND PART TO THE VALUE OF THE ELECTRICAL RESISTANCE OF THE THIRD PART, AND, THEREAFTER, EVALUATING THESE RATIOS ACCORDING TO A PREVIOUSLY CALIBRATED CHART HAVING ONE SERIES OF CURVES REPRESENTING THE PERCENTAGE OF W ATER IN THE ACID AND HAVING ANOTHER SERIES OF CURVES REPRESENTING THE PERCENTAGE OF NITRIC DIOXIDE IN THE ACID.

Images