Use of Wide-Range Indicators for Determination of pH - Analytical

---

INDUSTRIAL A N D ENGINEERING CHEMISTRY

July 15, 1931

% glycerol

=

weight carbon dioxide X 0.6975 X 100 weight sample

I n the apparatus the reflux condenser serves to remove chlorine gas or hydrochloric acid, the sulfuric acid and calcium chloride to dry the gas, and the ascarite tower to absorb the carbon dioxide. Notes-(1) &4nexcess of potassium dichromate must be present. One gram of this salt reacts with 0.134 gram of glycerol. ( 2 ) Tnis method is designed primarily for normal lubricating greases containing soaps, fats, free fatty acids, and mineral oil.

233

If water-soluble organic compounds other than glycerol, which cannot be precipitated with inorganic salts, are present, it cannot be used. (3) No glycerol is lost in this determination for, according to Lewkowitsch ( I ) , glycerol is not volatilized by boiling until the concentration is approximately 30 per cent. The concentration of glycerol in this determination is rarely over 2 per cent. (4) A determination requires about 3 hours.

Literature Cited (1) Lewkowitsch, J., “Chemical Technology and Analysis of Oils, Fats, and Waxes,” p. 251, Vol. I, Macmillan, 1922.

Use of Wide-Range Indicators for Determination of pH’ 1

F. R. McCrumb W. A. TAYLOR A N D Co., INC., BALTIMORE, MD.

It has been shown that accurate pH determinations tions, the pH of an indicator HERE is a simplicity cannot be made on slightly buffered materials with a solution is usually of little in the use of a single moment. However, many of wide-range indicator when the determinations are indicator to cover relacarried out with a single solution which may vary the solutions encountered in tively wide pH ranges that considerably in gH from the material b e i a tested. practice are o n l y s l i g h t l y appeals t o the practical buffered, and the reaction of So-called universal, utility, and range-finding indiminded. Consequently, cators should never be used on such materials as water, the indicator solution is one several such indicators have soil extracts, paper extracts, beater water, white water, of the most important factors been recommended for pH pure sugar liquors, sewage effluents, clay filtrates, to be considered. Quite a work and these are being sold brines, flotation feeds, evaporator water, air-condifew i n v e s t i g a t o r s h a v e under such names as univertioning water, laundry rinses, and, in short, any pointed this out. Birge and sal, range-finding, and utility Acree (1) first suggested the material that is not fairly well buffered, unless it is indicators. All are prepared known definitely that the pH of the indicator does not adjustment of the pH of the by dissolving a m i x t u r e of indicator in steps and treating vary appreciably from that of the sample. several indicator dyes in a Although fairly accurate results can be secured by the weakly buffered unknown suitable solvent. A number using a number of solutions of a wide-range indicator solution with indicator apof such mixtures have been proximately isohydric thereadjusted in steps of 1.0 pH according to the isohydric described in the literature (3). method, such a procedure offers no advantage over with to prevent p H errors. While several papers have Karracker (4, Marsh ( 5 ) , the use of several short-range indicators, with the appeared in which the use of disadvantage that a wide-range indicator is rarely and Pierre and Fudge (7) sugwide-range i n d i c a t o r s has gested that when w o r k i n g so sensitive as several short-range indicators used been a d v o c a t e d , there is a with soil extracts all indicator separately because the color changes are not so distinct. dearth of experimental data solutions be adjusted to their on the relative accuracv resulting from their use. It is the purpose of this paper to respective midpoints. Schlegel and Steuber (8)recommended present some results secured when attempts were made to a similar procedure with bromothymol blue when determining use a wide-range indicator to determine the pH of such ma- the pH of solutions of pure sugars. In determining the pH of paper, the extracts are invariably terials as are met with in practice. The indicator solution employed in this work is one that only slightly buffered and accurate colorimetric pH results has become quite popular as a universal or utility indicator. cannot be secured unless care is taken to adjust the pH of the The composition is methyl red, 0.02 per cent; bromothymol indicator solutions. The data of Wehmhoff (9) and of Mcblue, 0.04 per cent; thymol blue, 0.04 per cent; and phenol- Crumb and Kenny (6) are sufficient to demonstrate this. phthalein, 0.02 per cent. The solvent is 60 volumes of speRecently Fawcett and Acree (3) have published details on cially denatured (No. 30) alcohol diluted to 100 with distilled their isohydric-indicator method for determining the pH in water. The reaction can be adjusted by adding standard very dilute solutions and water. This method consists in sodium hydroxide solution. The color changes for such a the use of several solutions of each indicator adjusted to mixture are: pH 3.0, red; pH 4.0, orange-red; pH 5.0, orange; different pH values in 0.2 steps over the range of the indip H 6.0, yellow; pH 7.0, yellow-green; pH 8.0, green; pH 9.0, cator. By this procedure it is possible to use an indicator green-blue; pH 10, violet; pH 11.0, red-violet. Temporary solution having practically the same pH as the unknown color standards can be prepared from the buffer mixtures of solution. Clark and Lubs in intervals of 0.4 pH over the range 3.0 to These investigations have proved that the pH of slightly 10.0. The proportion of indicator solution to buffer mix- buffered solutions can be determined accurately only when ture is 0.25 cc. to 5 cc. This enables one to make readings the pH of the indicator solution is reasonably close to that of with a relative sensitivity of 0.2 pH. the unknown sample. For this reason Fawcett and Acree As soon as a wide-range indicator is applied to practical recommended that all indicator solutions, except those used solutions, the question arises as to the proper pH of the indi- to cover the strongly acid and alkaline ranges, be adjusted to cator solution. In determining the pH of well-buffered solu- the midpoint of the respective indicator and also to the lowest 1 Received January 30, 1931. and highest useful pH values for testing very weak buffers

T

AYALY TICAL EDITION

234

and water. Some commercial firms are now supplying indicator solutions so adjusted in Pyrex bottles. This question would be of theoretical interest only if it were not that so many of the solutions found in practice are very slightly buffered. Among the most common are the majority of waters, both natural and treated, soil extracts, pure sugar liquors (pan liquors), paper extracts, many beater waters, white water, sewage effluents, clay filtrates, many brines, some flotation feeds, evaporator water, condenser water, air-conditioning water, laundry rinses, etc. It is very important that the colorimetric method be applied in such cases because, owing to low buffer capacity and low conductivity, the pH can rarely be satisfactorily determined by an electrometric method. It is in the case of such materials that one immediately gets into difficulty when a wide-range indicator is used. Obviously adjustment to the midpoint, pH 7.0, means little if the solution which is being tested is decidedly acid or alkaline. I n the case of a short-range indicator like bromothymol blue, if adjustment is made to the midpoint pH 6.8, the greatest variation between indicator and unknown that will be considered is 0.8 pH. If the variation were greater, the sample would be considered as falling outside the range of bromothymol blue and the reading would be disregarded. However, with a wide-range indicator, if one were testing a solution with a pH of 3.0 to 4.0 or 10.0 t o 11.0, by using an indicator solution adjusted to pH 7.0 the variation might be 3 or 4 entire pH units. Where the indicator solution may have a lower or higher pH, this variation between indicator and unknown may be even greater than 4 units. It is apparent that such difference must result in incorrect pH readings, except in the case of fairly highly buffered solutions.

Tahle I-Determination

of pH of Water

WIDE-RANGE INDICATOR WITH

SHORT-RANGE INDICATORS5 . 2 5

SAMPLE

PH 5.2 75 . 62 6.8 6.6 6.6 7.2 6.4 6.4 6.4 7.2 10.0 9.4 9.2 9.0 9.4 9.2

9.10

variation Max.

PH 6.6 7.0 7.8 7.6 7.4 7.4 7.8 7.2 7.2 7.2 7.6 10.0 9.6 9.4 9.2 9.6 9.4

PH 8.6 8.8 8.4 8.4 8.5 8.6 8.4 8.4 8.4 8.4 8.2 10.0 9.8 9.6 9.6 9.8 9.6 9.4 9.4 6.4 6.8 4.8 7.4 7.0 6.0

PH 3.4 3.2 1.2 1.6 1.9 2.0 1.2 2.0

9.0 8.8 6.0 6.6 4.6 7.0 6.6 5.6

8.0 5.6 6.4 4.2 6.4 6.2 6.2 a

Distilled.

b Drinking.

Boiler,

0

P H OR:

7.45

8.6

2.0

2.0 1.0 0.0 0.4 0.4 0.6 0.4 0.4 0.8 1.4 0.8 0.4 0.6 1.0 0.8 0.8

d Natural.

PAPER EXTRACTS-some pH determinations were made on samples of paper following the extraction method. The extractions were made by gently boiling a mixture of 5 grams of finely divided paper and 250 cc. of distilled water for 1hour. The total weight of the flask containing the mixture was determined before boiling, and the loss of weight was replaced with boiled distilled water. The mixtures were not filtered, the supernatant extract being withdrawn by means of a pipet for pH determinations. The results are given in Table 11. of pH of Paper Extracts

Table 11-Determination

Experimental Procedure

I n order to determine the magnitude of the errors that may be encountered, several series of pH determinations were carried out by using three separate solutions of the widerange indicator adjusted to the pH values 5.25,7.45, and 9-10. These values were determined both by a colorimetric method and by a quinhydrone electrode. In addition, pH determinations were made on each sample by using solutions of shortrange indicators adjusted to the respective midpoints. The color standards used were prepared from the buffer mixtures of Clark and Lubs in intervals of 0.2 pH. These were all checked electrometrically. The short-range indicators employed were the sulfonphthaleins of Slagle, White and Acree, Clark and Lubs, and Cohen, together with methyl red and a proprietary indicator, phthalein blue. Samples were collected and kept in Pyrex vessels. All tests were made in 5-02. test tubes with a diameter of 11.5 mm. Precaution was taken to avoid variations due to the action of the carbon dioxide of the air. I n every case the indicator solution was added to the empty tube and the measured sample pipetted into the tube so that the sample was delivered below the surface as recommended by Fawcett and Acree. Final mixing was secured by gentle stirring with the tip of the pipet. The form of tabulating the results is uniform throughout the paper and a brief explanation may be useful. In one column are given the pH values as determined by a shortrange indicator, in the following columns are found the pH values as determined when wide-range indicator solutions adjusted to different pH values are used, in the last column is found the difference in pH between the results secured with a wide-range indicator solution adjusted to 5.25 and one adjusted to 9.10. NATURAL AND TREATED WATERS-The first series of tests was made on a number of samples of natural and treated waters. The results are shown inTable I.

Vol. 3, No. 3

SAMPLE

WIDE-RANGE INDICATOR WITH PH OF:

SHORT-RANGR lNDICAToRs

PH 4.2 4.1 4.6 4.7 5.9

1Q 2Q 3b 40 5d

5 25

7.45

9.10

fiH 4.4 4.4 4.6 5.0 6.0

PH 4.6 4.7 5.2 5.8 6.6

PH

Max, variation

PH 0.4 0.6 1.0 1.5 2.0

4.8 5.0 5.6 6.5 8.0

~

a

Letter,

b Yellow copy.

C

Newsprint.

d Kraft wrapping.

SOILExTRAcTs-several soils were extracted by using 20 grams of soil and 80 cc. of distilled water. After thorough mixing in Pyrex bottles, the extractions were permitted to settle for 30 minutes and the pH of the supernatant extract determined, The results are shown in Table 111. of pH of Soil Extracts

T a b l e 111-Determination

WIDE-RANGE INDICATOR SAMPLE

SHORT-RANGE INDICAToRs 6.25 PH

9H la

,"e

4c

4.8

F,d

8.0

a

Sandy loam.

b

t

5.0 7.6

Heavy loam.

SHORT-RAXGE INDICAToRs

1= 2a 3a 4b 56 (L.

After sour.

5.5 4.2 5.0 7.7

7 6

b After suds.

9.10

Max.

OH 6.4 6.8 6.8 6.4

PH 7.4 8.0 8.0 5.6 8.4

PH 1.6 2.0 2.0 0.6 0.8

Medium loam.

d Heavy soil.

of p H of Laundry R i n s e s

Table IV-Determination

I

WIDE-RANGE ~

IXDICATOR WITH

5 25

7.45

9.10

5 4 5 7

5 5 5 7 7

6 0 5 4 5 8 8.0

4 8 0 2

7 0

PH OF:

7.45

8.0 0

WITH

6 0 4 6 6

8 0

PH OF'

variation Max,

0 0 0 0

6 6 8 8

10

INDUSTRIAL AND ENGINEERING CHEMISTRY

July 15, 1931

LAUNDRY RINSES-since pH control is now being applied to laundry operations, a few tests were made on samples of rinse waters. The results are shown in Table IV. DILUTEALUMSoLuTIoNs-The next series of tests was carried out on dilute solutions of paper-maker’s alum. The results are given in Table V.

sHORT-WIDE-RANGEI N D I C A T O R WITH PH OF: SAMPLECONCENTRA-RANGE Max. TION INDICATORS 5.25 7.45 9 . 1 0 variation P. Q. m. 8000 40 80 800 8

of D i l u t i n g S o d i u m Bicarbonate S o l u t i o n WIDE-RANGE INDICATOR WITH PH OF:

SAMPLE DILUTIONlNDICAToRs

5.25

7.45

Max. variation

9.10

of pH of D i l u t e A l u m S o l u t i o n s

Table V-Determination

1 2 3 4 5

T a t l e IX-Effect

235

QH

pH 4.0 4.8 4.8 4 4 5.0

3.8 4.6 4.4 4 0 5.0

9H 4 0 5 2 5.2 4.6 6.8

9H 4.0 5.0 5.0 4.6 5.6

QH 0.0 0.4 0.4 0.2 1 8

SUGAR SOLUTIONS-In the refining of sugar, as the purity of the liquors increases, they become correspondingly less buffered. Accordingly, pH determinations on highly purified char-filtered liquors should be made by following all the precautions necessary to secure accuracy with unbuffered solutions. This is also important in beet sugar manufacture where pH control is used extensively on the pan liquors. The type of results which may be expected on such liquors is shown in Table VI.

I

Table X-Effect

of D i l u t i n g S o d i u m Sesquicarbonate S o l u t i o n WIDE-RANGE IXDICATOR

SAXPLE DILUTION INDICATORS 5,25

1 2 3 4 5

None 1:lO

1:lOO 1:lOOO 1:10000

8.4 6.4

9.0 6.4 5.4

WITH

7 45

10.2 10.2 9.2 7.0 6.6

PH

OF:

Max. variation

10.2 10.2 9.6 8.6 8.4

0.0

0.0 0.6 2.2 3.0

Summary of Results

The readings given in the tables tend t o show the magnitude of the errors that may be encountered when attempts are made t o determine the pH of many practical solutions with a single wide-range indicator. It is apparent that so-called universal or utility indicators covering a wide pH range can be used with safety only on wellT a b l e VI-Determination of pH of Purified Sugar S o l u t i o n s buffered materials. Unfortunately in testing solutions of 1 WIDE-RANGEINDICATOR WITH PH OF: unknown nature, there is no simple or rapid way of deterSAMPLE SHORT-RANGE Max. INDICATORS mining the buffer capacity. Consequently, pH control under 5.25 7.45 9 . 1 0 variation such circumstances becomes somewhat uncertain. There being no way of checking results, the operator may make 6.0 6.8 7.2 1.2 serious errors. i.2 i.6 i.2 i.0 21 In order to avoid the effect of variation in pH between the 3 6.5 6.0 6.6 7.2 1.2 4 5.8 5.4 5.8 6.2 0.8 indicator and the unknown solution, a number of different solutions of the wide-range indicator might be adjusted in EFFECT OF DILUTING BUFFEREDSoLuTIoNs-In order to steps of 1.0 between pH 3.0 and 11.0. By application of the study the effect of dilution observed by Fawcett and Acree isohydric-indicator technic with these nine indicator solutions, on the pH readings secured with a wide-range indicator at it is possible to determine the pH of weakly buffered solutions variable reactions, several well-buff ered solutions were di- with a fair degree of accuracy. Three wide-range indicator luted with distilled water (pH 5.8) and the pH values deter- solutions adjusted at pH 3.0, 7.0, and 11.0 can be used for mined as before. The buffer solutions used were 0.05 M po- preliminary field or factory tests. Intermediate pH adjusttassium acid phthalate, a 0.05 M phosphate mixture, and 1.0 ments of the indicator can be made up by using different per cent solutions of sodium bicarbonate and sodium sesqui- cdculated ratios of drops of the pH 3.0,7.0, and 11.0 solutions carbonate. The results are shown in Tables VII, VIII, IX, for the final tests. For more accurate work, however, such a procedure offers no advantage over the use of several shortand X. range indicators, with the disadvantage that a mixed indicator is rarely so sensitive as several indicators used separately Table VII-Effect of D i l u t i n g P o t a s s i u m Acid P h t h a l a t e Solution because the color changes are not so distinct. Since single-test unadjusted wide-range indicators may 1 WIDE-RANGEI N D I C A T O R WITH PH OF: 5 , 25 yield such erroneous results with slightly buffered materials, SAXPLE DILUTIONSHORT-RANGE/ Max. this practically prohibits their use in many fields. It is 7.45 variation unfortunate that the use of such unadjusted indicators has I QH QH QH pH been encouraged because it may not only came economic loss 1 None 4.0 4.0 4.0 4 0 0.0 but may make the user lose confidence in pH control. 4.2 4.4 4.6 0.4 2 1:lO

I

!?

3 4

1:lOO 1:lOOO 1:10000

R

Table VIII-Effect

I

4.6 5.0 5 2

4 8 4 8

4.8 5.8 6.4

5.2 6.6 7.4

0.6 1.6 2.2

of D i l u t i n g P h o s p h a t e Buffer S o l u t i o n WIDE-RANGE INDICATOR WITH PH OF:

SAMPLEDILUTIONs

1 2 3 4 5

None 1:lO 1:lOO

1:lOOO 1:lOOOO

$ 6.9 6.9 6.7 6.5 6.4

5 . 25

~

~

~

Max. variation

7.45

7.0

1 ili

6.8 6.8

.._

8.0 8.6

0.0 1.8 3.4

Literature Cited (1) Birge and Acree, J . A m Chem. Soc., 41, 1037, footnote (1919). (2) Clark, “The Determination of Hydrogen Ions,” 3rd ed., pp. 96-98, Williams & Wilkins, 1928. and Acree, J . B a d , 17, 163 (1929); IND.ENG.CHEM., Anal. ~(3) Fawcett Ed., 2, 78 (1930). (4) Karracker, Soil Sczence, 15, 473 (1923). (5) Marsh, Science, 59, 216 (1924). (6) McCrumb and Kenny, J . Soc. Chem. Ind., 49, 425T, 427T (1930). (7) Pierre and Fudge, J . A m . Chem. Soc., SO, 1254 (1928). (8) Schlegel and Steuber, IND.ENG. CHEM.,19,631 (1927). (9) Wehmhoff, U. S. Govt. Printing Office Repts., Tech. Bull. 8 (1930); I b i d , 11 (1930).