Evaluation of color changes of indicators - Analytical Chemistry (ACS


Evaluation of color changes of indicators - Analytical Chemistry (ACS...

4 downloads 102 Views 512KB Size

1446

Anal. Chem. 1982, 5 4 , 1446-1449

Evaluation of Color Changes of Indicators J. Cacho,” C. Nedn, and L. Ruberte Department of Analytical Chemistry, Faculty of Sciences, University of Zaragoza, Spain

E. Rlvas Calcul Center, University of Zaragoza, Spain

An objective evaluation of the color transition of a visual indicator is completely necessary. Some studies in this direction have been made by several authors (1-5) and lately the Analytical Chemistry Division Commission has recommended practical analysts evaluate this area (6). However, there are few studies concerning this subject. MacAdam (7-10)carried out the main basic studies, and he proposed calculating the chromatic differences in terms of standard deviations. Reilley et al. (11, 12) combining these concepts with Beer’s law developed the complementary chromatic system and new parameters such as the one of dichromatism, in order to evaluate the color change. Kotrly, Vitras et al. (13) have introduced a new constant, based on Reilley’s article, which relates the color or optical concentration to the absorbance. They have studied the accuracy in the color specification (14) and have classified several indicators and inert dyestuffs, too (15-17). However, throughout all these articles, the color of the indicators is taken into account only a t the beginning and at the end of each titration, and the sharpness of transition is not considered at the equivalence point. This aspect has only been considered by Bhuchar et al. (18,19) for a few acid-base indicators, by means of the method called “specific color discrimination” based on MacAdam’s ellipses (8)in order to evaluate color changes of indicators. In spite of this there is not yet an objective chromatic transition classification of well-known indicators and there are not any studies about the sharpness of the chromatic transition in the case of metallochromic indicators. This fact can be due to the intrinsic difficulty of the method for the practical analysts. In this study a new procedure is described in order to evaluate the quality of the color change of the indicators. This procedure is based on the color or optical concentration, J, which was defined by Reilley (11). The method is simple and the absorbance values were treated by a lineal computer program in BASIC. The results obtained in the case of acid-base indicators are nearly similar to those obtained by Bhuchar (19). We have applied this method successfully to several metallochromic indicators used in the chelometric titrations of copper.

Buffer Solutions. These were prepared as described by Clark and Lubbs. Standard Cu Solution. A standard copper solution of 1000 ppm was prepared by dissolving electrolytical Cu. EDTA Solution. An EDTA 0.010 M solution was prepared. Procedure for the Acid-Base Indicators. Five milliliters of 2 X lo4 M indicator solution was taken and diluted with about 10 mL of water. The solution was buffered appropriately to obtain the desired pH, and the volume was finally made to 25 mL with distilled water. Procedure for the Metallochromic Indicators. In 25-mL calibrated flasks were poured 5 mL of the standard solution of Cu, 1 mL of an indicator solution, 5 mL of an appropriate buffer solution, and X mL of the EDTA solution. The volume was fiially made to 25 mL with distilled water. Absorption spectra of each of the solutions of the indicators mentioned above were taken at different pHs in the case of acid-base indicators, generally in 0.1 pH-unit steps and at different molar relations EDTA/Cu in the case of metallochromic indicators, generally in 0.01 EDTA/Cu unit steps at 10-nm intervals. BASIS OF THE METHOD To specify the color transition in a titration it is necessary to know the color variation that takes place through it. The color variation measurements can be taken in terms of J which represents the optical concentration for the whole system.

J = cE

(1)

where c is the analytical concentration of the indicator and

E is the effective absorptivity, which is related to the molar absorptivity as follows:

K being a constant for each wavelength.

When the chromatic conditions of a solution through a titration are modified, either from changes in the pH in the case of acid-base indicators or from changes in the pM, in the case of metallochromic indicators, the J values change. Color differences between two points of the titration can be calculated in terms of J difference units. In order to evaluate quantitatively the perceptibility of this difference, that is to say, the sensitivity of an indicator, one should consider the variation of J for small constant pH or pM differences, such as 0.1 pH or pM units. But as this is difficult to achieve experimentally, we followed a more conEXPERIMENTAL SECTION venient procedure. According to this procedure the variation A Pye Unicam spectrophotometer, Model SP8-100, with 1-cm of J between two close pH or pM values is calculated and the silica cells; an Orion Research pH meter, Model No 901, and a rate AJlApH is obtained in the case of acid-base indicators Computer Digital PDP-11/55 were used. and the rate AJ/ApM in the case of metallochromic indicators. Indicator Solution. Generally, Merck indicators were used. When the value of this relation is maximum we have ob(MeTDAN) The 2-(5-methyl-1,3,4-thiadiazolylazo-l)-2-naphthol (20), 2-(5-methyl-1,3,4-thiadiazolylazo-6)-3,4-dimethylphenol tained the equivalence point that will correspond to the pH (MeTDADMF) (21), and 5-(1,2,3,4-tetrazolylazo-6)-2,4-di- or pM value at which value this maximum takes place. methylphenol (TeADMF) (22) were synthesized by the method If this procedure is applied to several indicators a scale of described by Goerdeler (23). their sensitivity in terms of maxima AJlApH or AJ/ApM A2X M solution of each of the indicators in acetone was values can be obtained. However, this does not represent an used for the phthalein indicators. For the sulfonephthalein, a accurate judgement of the quality of an indicator, since we 2X M aqueous solution was used. must consider, too, the sharpness of the peak a t the equivaA 4.5 X lo4 M ethanolic solution in the case of Murexide and lence point, that is to say, the pH or pM interval in which the 9 X M ethanolic solutions in case of l-(pyridyl-2’-azo-2peak is reached. The rapidity of color changes can be shown naphthol) (PAN), MeTDAN, MeTDADMF, and TeADMF were by the half bandwidth in pH or pM units. Consequently the used. 0003-2700/82/0354-1446$01.25/0

0 1982 American Chemical Society

ANALYTICAL CHEMISTRY, VOL. 54, NO. 8, JULY 1982

1447

Table I. pII and J Valies for the Color Changes of Some Acid-Base Indicators cresolphthalein

thymolphthalein

PH 8.28 8.50 8.88 9.05 9.11 9.15 9.18 9.40 9.72 9.87 11.58

pH 10.:12 10.47 10.52 10.65 10.130 10.90 10.!)7 11.02 11.115

J

0.0207 0.0457 0.2114 0.4686 0.5223 0.5744 0.6361 0.9429 1.2653 1.3185 1.4287

phenolphthalein

-

J 0.1016 0.1887 0.2877 0.4646 0.6609 0.7400 0.8321 0.8862 0.9683

IPH 9.62 10.05 10.25 101.32 2 10.37 10.65 10.85 11.88

J

0.0563 0.1570 0.2193 0.2477 0.2644 0.3571 0.4144 0.4485

phenol red PH 6.48 6.92 7.65 7.75 7.90 8.20 8.25 8.40 8.90 9.10 10.20

cresol red

J 0.6980 0.7554 0.9319 1.0067 1.0214 1.1496 1.1585 1.1693 1.2554 1.2704 1.3138

PH 7.00 7.35 7.45 7.60 7.90 8.00 8.15 8.35 8.68 9.05

blue thymol PH J

J

0.7422 0.8660 0.8942 1.0287 1.1520 1.1897 1.2394 1.3840 1.4422 1.5037

7.90 8.17 8.45 8.67 8.80 8.95 9.00 9.15 9.23 9.35 9.50

0.6239 0.6468 0.6783 0.7722 0.8542 0.8891 0.9320 0.9738 1.0327 1.0579 1.1281

congo red PH 4.82 4.98 5.07 5.15 5.20 5.52 5.82 6.02

J

0.9944 1.0312 1.0719 1.0902 1.0997 1.1306 1.1537 1.1598

H SCD Values at Maximum Color Change of Phthalein and Sulfonephthalein Indicators Table 11. A J / A ~ and half half bandwidth bandwidth of change of change range of PHmx ofSCDin pH,, of AJ/A pH indicator color change (SCD) SCD,, pH units ( A J ~ A P H )(AJ~APH),, in pH units phenolph tha1ei.n o-cresolphthalein

alkaline alkaline

10.26 9.00

45 71

0.4 0.4

thymolphthalein

a1kaline

phenol red

a1kaline

cresol red

alkaline

thymol blue congo red

alkaline

10.50 10.90 7.68 8.32 7.90 8.70 8.90 5.05

35 30 35 16 23 19 27 17

0.5 0.4 0.85 0.85 0.90 0.80 0.90 0.72

10.32 9.18 9.05 10.52 10.97 7.75 8.2 7.60 8.40 9.0 5.07

0.405 2.06 1.51 1.58 1.36 0.75 0.43 0.90 0.57 0.96 0.45

0.95 0.4 0.2 0.35 0.3 0.2 0.55 0.30 0.60 0.75 0.35

20

z return

16

I,

i

2

12

1 Input N a r n e a Data 2 D ~ s p l a yof data 3 Write N a m e IS. Data 4 Process€ 5 Obtention ot 0. J,E C, Y.P.@j etc 6 Writeall 7 Absorbance tp Trasrnitancc 8 End ot wotlng 9 Data IO print spoter 10 Loop f o r € of Data for columns

08

I 1 Product of S f o r aproplate coefficients etc 12 Wrile all

5

Flgure 1. Line flow of the

RICANER

program.

sensitivity and the rapidity of color change will define the quality of an indicator. The Calculation of J . To calculate the J value, we have worked on a fully lineal computer program written in BASIC called RICANER. From the absorbance values a t 10-nm intervals alone and using the RICANER program we can calculate the color concentration J besides the complementary clolor coordinates (Qr),

6

7

8

9

10

11

12

pH

Flgure 2. Change of optical concentration of phthalein and sulfonephthalein indicators with pH: (- -X--) phenolphthalein; (- -A--) cresolphthalein; (- -0--) thymolphthalein; (-X-) phenol red; (-A-)cresol red; (-0-) thymol blue; (-El-) congo red.

the true color coordinates (Pr), the dichromatism numerical values (D), and the luminosity (Y). The line flow of the program is shown in Figure 1.

RESULTS AND DISCUSSION Acid-Base Indicators. The J values through a titration for each indicator studied are shown in Table I.

1448

ANALYTICAL CHEMISTRY, VOL. 54, NO. 8, JULY 1982

Table 111. L/M Ratio and J Values for the Color Changes of Some Metallochromic Indicators PAN pH 5 L/M J

MeTDAN pH 5 L/M J

Murexide pH 10 L/M J

0,909 0.935 0.961 0.981 0.994 1.000 1.007 1.013

0.909 0.935 0.961 0.981 0.994 1.000 1.007 1.039

0.922 0.955 0.987 1.000 1.007 1.013 1.025

0.8173 0.8174 0.8273 0.8205 0.8234 0.7568 0.7210 0.6833

0.8926 0.8773 0.8844 0.8538 0.7659 0.6656 0.6197 0.6140

MOLAR RELATION EDTA/c

"

Figure 3. Progress of color concentration vs. t h e ratio EDTA/Cu.

Figure 2 shows the progress of color concentration values of the different indicators with change of pH. These values at maximum color change are given in Table 11. If the Qx, Qy values of each indicator through the titration are plotted on a complementary diagram CIE, the path is nearly straight in the case of phenol red, congo red, cresol red, and thymol blue, whereas the path is crooked in the case of thymolphthalein and cresolphthalein. The transition chromaticity curves in the case of phenol red, thymol blue, and cresol red pass close to the achromatic point. Therefore the J values decrease the nearer the gray point is and a later increase of J values will show the change to another color field on the CIE diagram. If this happens, two peaks in AJ/ApH vs. pH graph must appear as is shown in Figure 2. When the chromatic transition changes its direction and the path is crooked, a peak must appear for each curving. This fact is related to dichromatism. If the number of experimental points taken to define the transition is not enough, it may not be possible to tell the two peaks because a single wider peak may be obtained which includes the two sharper ones. Two examples of this are shown in Figure 2. In the case of thymol blue only a few points experimentally obtained have been plotted and the line connecting these points gives a single wide peak. In the same way, in the case of cresolphthalein

0.2732 0.2941 0.2693 0.2289 0.2180 0.2074 0.2006

MeTDADMF pH 6 LIM J 0.927 0.954 0.980 1.000 1.013 1.020 1.060

0.6506 0.5971 0.5467 0.4662 0.4206 0.4116 0.4088

TeADMF pH 5 L/M J 0.955 0.976 0.583 0.990 1.000 1.005 1.132

0.3792 0.3404 0.2272 0.2931 0.1601 0.1506 0.1512

two peaks appear instead of the three theoretic ones. The two peaks in the SCD method are attributed by Bhuchar to the presence of impurities. This is only true if such impurities produce changes in the color indicator and consequently produce changes in the color transition. The exact knowledge of pH at each peak allows us to know the pH change through a titration and to predict the end point. Table I1 gives at each peak the pH, the height, and the half bandwidth values for the studied indicators. By observing these data the quality scale for these indicators in the alkaline region must be as follows: cresolphthalein > thymolphthalein > cresol red > thymol blue > red phenol > congo red > phenolphthalein. The quality judgement of these indicators is similar to the judgement proposed by Bhuchar; the only discrepancy is the case of the phenolphthalein. This difference may be due to low purity of phenolphthalein used. This fact makes clear the value of the proposed method and it has the advantage of not needing the use of the gll, gI2,and gZ2coefficients, which are usually affected by errors of approximation. Metallochromic Indicators. The J values through a titration for each indicator studied are shown in Table 111. The same considerations proposed for the acid-base indicators can be applied to the metallochromic indicators with some reservations due to the different titration process. In a complexometric titration the end point will always correspond to the last peak reached, since from a ratio EDTA/Cu = 1 no important color change occurs. Figure 3 represents the progress of color concentration vs. the ratio EDTA/Cu. The ratio EDTA/Cu instead of pM has been chosen to show that the greatest variation of color transition corresponds exactly to the end point of the titration. Besides, it has to be considered that when the experimental conditions of a complexometric titration change, for example, because of an alternation of pH or a change in the concentration of a complexing agent, such as NH3, the pM is modified, and this will result in a different value of AJ/ApM for the same indicator in the titration of the same amount of metal. From Figure 3 we can extract the conclusion that the best of the studied indicators for the direct titration of Cu with EDTA is TeADMF followed by MeTDAN and, at a distance, by PAN. We must emphasize the presence of a very sharp peak, previous to the equivalence point, which is due to a nearly total loss of color for EDTA/Cu = 0.990. This makes the location of the end point in the titration much more accurate, with a lower possibility of visual errors.

LITERATURE CITED (1) Fortune, W. B.; Mellon, M. G. J . Am. Chem. SOC. 1938, 6 0 , 2607. (2) Woods, J. T.; Mellon, M. G. J . Phys. Chem. 1914, 45, 313. (3) Van Wyk, J. J.; Clark, W. M. J . Am. Chem. SOC. 1947, 69, 1296. (4) King, J. Ana/yst (London) 1952, 7 7 , 742. (5) Kotrly, S. S b . Ved. Pr., Vys. S k . Chemickotechonol. Pardubice 1988, 1 4 , 165. (6) I.U.P.A.C. Analytlcal Chemistry Division Commlsion on Analytical Reactions and Reagents Pure Appl. Chem. 1979, 5 l ( b ) , 1357,

1359-1365. (7) MacAdam, D. L. J . O p t . SOC.Am. 1942, 3 2 , 2 .

Anal. Chem. 1982, 5 4 , 1449-1450 MacAdam, D. L. J. Opt. SOC.Am. 1942, 32,247. MacAdam, D. L. J. Opt. SOC.Am. 1943, 32, 675. MacAdam, D. L. J. Opf. SOC.Am. 1943, 33, 18. Reilley, C. N.; Fiaschka, H. A.; Laurent, S.;Laurent, B. Anal. Chem. 1960, 32, 1218. Reilley, C. N.; Smith, tillls M. Anal. Chem. 1960, 32, 1233. Vytras, K.; Kotrly, S.;Vondriskova, E.; Vorac, B. Collect. Czech. Chem. Commun . 1970, 35 3379. Kotrly, S.;Vitras, K. Talanta 1971, 78, 253. Kotrly, S.; Vytras, K. S6. VedPr ., Vys. Sk. Chemickotechnol. Pardublce 1989, 79, 21. Kotrly, S.;Vltras, K.; Oharek, J.; Vondrusko'va, E. S b . Ved Pr., Vys. Sk. Chemickotechnol. Pardublce 1970, 2Zm19.

1449

(17) Vitras, K.; Vytrasova, J.; Kotrly, S. Talanta 1975, 22, 529. (18) Bhuchar, V. M.; Das, S.R. J. Opt. SOC.Am. 1964, 5 4 , 817. (19) Bhuchar, V. M.; Kukreja, V. P.; Das. S. R. Anal. Chem. 1971, 43, 1847. (20) Cacho, J.; Nerln, C.Afinidad 1981, 374, 345. (21) Cacho, J.; Nerin, C.; Calvo, A.; Ruberte, L. R . SOC. E s p . Fis. Quim. Reun. Bienal, 78th 1980. (22) Nerln, C.; Cacho, J. R . SOC. Fls. Quim. Reun. Bienal, 78th 1980. (23) Goerdeler, J.; Galinke, J. Chem. Ber. 1980, 9 3 , 397.

~

RECEIVED for review December 21,1981. Accepted March 22, 1982.

Colorimetric Determination of 2,3-Butanediol R. A. Speckman' and E. B. Collins" Department of Food Science and TechnologJv, University of California, Davis. California 956 16

2,3-Butanediol (2,3-butylene glycol) is a metabolic end product of many microorganisms. Current analyses involve either cleavage of the diol to acetaldehyde ( I ) or oxidation of the diol to acetoin mid diacetyl followed by determination of the amount of oxidant ( 2 )or products (3). Both methods involve reactions that are extremely susceptible to slight changes in experimental conditions and difficult to control. A method we devised depends on strict adherence to a rigid experimental protocol ithat must be performed in an exhaust hood in the dark and requires red light-resistant tubes, rubber septa, addition of reagent with a syringe, and precise control of pH, heating, and reaction time ( 4 ) . The purpose of this study was t o develop a simple colorimetric procedure for quantitative determination of the microgram amounts of 2,3-butanediol €ound in biological test Bystems.

EXPERIMENTAL SECTION 2,3-Butanediolwas obtained from K & K Laboratories,Jamaica, NY. It was distilled and the fraction at 181 "C was collected and used for preparing standrud solutions. All other organic chemicals were obtained from Aldrich Chemicals, Milwaukee, WI. All inorganic chemicals weire reagent grade. Spectrophotometric measurements were made with a Beckman spectrophometer, Model DB. The procedure we developed for determining 2,3-butanediol requires the following reagents: ethylene glycol, 0.1 M H,I06, 4% (w/v) CuS04, a standard solution of 2,3-butanediol (10 yg/mL), and 1.5% (w/v) p-hydroxydiphenyl in 0.5% NaOH. The p hydroxydiphenyl solutioin should be stored in a brown bottle and is stable up to 1month ,st room temperature. For determining the diol, pipet 1.0 mL of the unknown, adjusted to pH 7.0 (5) and containing 0 to 10 yg 2,3-butanediol/mL, into a test tube and add 4 mL of water. Run a reagent blank and several 2,3-butanediol standards (0-10 yg/mL) parallel. Add 1.0 mL of the periodic acid solution to each test tube, mix, and let stand at rmm temperature for 30 min. Add 2 drops of ethylene glycol to each tube, mix well, and let stand for 5 min. 14dd 0.05 mL of the CuS04solution, mix, add 0.1 mL of the p-hydroxydiphenyl reagent, mix, and incubate for 30 min at 30 "C in a thermostatically controlled water bath. Shake the tubes occasionally during the incubation. Place the tubes in a bath of boiling water for 90 s to destroy the excess p-hydroxydiphenyl and then cool them to room temperature. Determine absorbancy ai, 570 nm using water passed through the periodate reaction as the internal standard. Compare the absorbancy values for unknowns to the standard curve. RESULT13 AND DISCUSSION Investigations with Vanadium Pentaoxide. Diols from which ketones can be formed are cleaved readily by vanadium Present address: Department of Food !3cience, University of Illinois, Urbana, IL 61801. 0003-2700/82/0354-1449$01.25/0

(6), but 2,3-butanediol is oxidized to diacetyl (7), a diketone that can be determined readily with the Westerfeld procedure (8). Solutions (0.01-0.30 M) of vanadium pentaoxide (V205), a stable powder that can be stored indefinitely and used as needed, were prepared in 1 M H2S04(7) and used for testing solutions of 2,3-butanadiol (0-1000 hg/mL). Following oxidation of the diol, with heating from 1 to 30 min in a bath of boiling water, we attempted to test for diacetyl with the Westerfeld procedure. A white precipitate formed upon addition of the a-naphthol, a problem we had encountered earlier in attempts to analyze acidic column effluents. Oxidized solutions of the diol were adjusted to neutrality by dropwise addition of 40% NaOH, and analyses again were attempted. Upon addition of the a-naphthol, the color of solutions became murky brown. We observed that solutions containing the larger amounts of 2,3-butanediol became somewhat green during heating and assumed that the green color resulted from formation of a mixture of pervanadyl ions (yellow) and vanadyl ions (blue). Nevertheless, measurements of absorbancy a t 200-700 nm failed to give a linear relationship between green color and original concentration of 2,3-butanediol. Cleavage of 2,3-Butanediol to Acetaldehyde with Periodic Acid. Periodic acid cleaves stoichiometrically many organic compounds containing vicinal hydroxyl groups, carbonyl groups, or either of these and an amino group (6). Desnuelle and Naudet ( I ) used periodic acid to cleave 2,3butanediol to acetaldehyde. Analysis of the acetaldehyde, however, depends on spectrophotometric measurement of maximal absorbancy of a transient blue color that appears during the reaction, and the method is useful for determining only large amounts of the diol (50-500 yg). Nevertheless, we found their method ( I ) to yield a stoichiometric relationship between 2,3-butanediol and the acetaldehyde produced. Choice of Reagent for Determining Acetaldehyde. The procedure for determining lactic acid by Barker and Summerson (9) actually determines acetaldehyde, employs p hydroxydiphenyl (PP) as reagent, and is very sensitive (0-10 yg/mL). The method of Sawicki et al. (IO),adapted for use in milk systems by Lindsay and Day ( I I ) , employs 3methyl-2-benzothiazolone hydrazone (MBT), which reacts with acetaldehyde in acidic aqueous solutions to give a blue cationic dye. N-Hydroxybenzenesulfonamide (HBS) was used in a method by Ismail and Wolford (12) to measure acetaldehyde in orange juice and dates. We tested and found each of these reagents and methods satisfactory for determining known concentrations of acetaldehyde (0-50 yg/mL), but none gave satisfactory results when used to measure 0-50 yg of acetaldehyde/mL derived by cleavage of 2,3-butanediol with 0 1982 American Chemical Society