1675
V O L U M E 24, NO. 10, O C T O B E R 1 9 5 2 proxiniately 1 hour a t room temperature before spraying. Irrigations were performed in covered glass jars 12 inches in outside diameter and 24 inches in height. A commercially available type of rack (Berkeley Chromatography Division, University Apparatus Co.) holding two glass troughs was employed. The jars were kept in a constant temperature room a t 30' C. Indications are that spent sulfite liquor contains, in addition to the sugars reported here, other carbohydrate components. The investigation is being continued. LITERATURE CITED
(1) Consden, R., Gordon, A. H., and Martin, A. J. P., Biochem. J . , 38, 224 (1944).
(2) Green, J.
iV.,personal communication.
(3) Gustafsson, C., Sundman, J., Pettersson, S., and Lindh, T., Paper and Timber ( F i n l a n d ) , 33, 300 (1951). (4 ) Hagglund, E., "Chemistry of Wood," New York, Academic Press, 1951. ( 5 ) Hough, L., Jones, J. K. N., and Wadman, W. H., J . Chein. Soc., 1950,1703. (6) Jermyn, M. A,, and Isherwood, F. ii., Biochem. J.. 44, 402 11949). \_._.,_
( 7 1) Ma, R. hl., and Fontaine, T. D., Science, 110, 232 (1949). (8)I Mulvany, P. K., Agar, H. D., Peniston, Q. P., and JIcCarthy, J. L., J . A m . Chem. Soc., 73, 1255 (1951). (9) Partridpe. S. hl.. with a note by Kestall, R. G., Biochern. J., 42, 238 (1948). (10) Sundman, *J., Paper and Timher ( F i n l a n d ) , 32B, 267 (1950) (11) Williams, R. J., and Kirby, H., Science, 107, 481 (1948). R E C E I V Efor D review February 6 , 1932. Accepted -4ugust 6 , 1932.
Absorption Characteristics of the Dithizone Mixed Color System A Theoretical Treatment ROBERT G.MILKEY Geological SurGey, Lnited States Department of the Interior, Washington, D. C.
T IS common procedure to determine trace amounts of lead by means of dithizone extraction. A solution containing the lead is adjusted to a pH of 9 to 10.5, and various complexing agents are added to prevent the extraction of other reacting element? that may be present. The lead is then estracted with a dilute solution of dithizone in chloroform, and the optical density of the dithizone solution is determined with a spectrophotometer a t an appropriate wave length of light. T h r lead content of the solution is determined from a standard cmve, expressed in terms of the optical density of solution versus micrograms of lead. The question arises whether the standard curve has a constant or a varying slope. Conformance to Beer's law in one part of the curve (dilute solutions) does not ensure a straight-line function throughout the range of the curve. Orice the standard curve has been established, the analyst will discover that it does not necessarily remain dependable for an?. appreciable length of time. Alteration of the various components within the analysis can result in an almost daily change of position of the curve. This lack of dependability of the curve is a matter of some importance, as thr qtandard curve is the basis of the analysis. The explanation for these, and for other characteristics that affect the slope and shape of the curve, lies in an analysis of the factor- that enter into the derivation of the standard curve.
If one analyzes the system mathematical1~-,the resulting equations make for an understanding of the disturbances introduced by these factors and of their relative importance. DERIVATION
ilssume a divalent metal in aqueous solution, estracted with a dilute solution of dithizone in chloroform. Me++
+ 2Dz --+;Ile(Dz)? + 2H'
The optical density of the resulting organic solution is equal to density of dithizone present plus density of metal-dithizonate formed. Assuming conformance to Beer's law: Density = kiclL
where k is the extinction coefficient, c is the concentration of the absorbing constituent, and L is the lengt,h of light path in the solution; subscript 1 refers to dithizone and subscript 2 refers to dithizonate. Let T' = unit weight metal per milliliter of organic solution Let C = original concentration of dithizonr solution in milligrams per milliliter Add 1 unit weight of metal. Dithizone equivalent of metal = KIT' Dithizonate equivalent of metal = K2T' Density of solution = density of dithizone density of dithizonate
+
=
klClL
FirPt of all, it is necessary to define some of the factors that determine the standard curve: 1. The presence of a mixed-color system. The density of the dithizone solution is the result of two absorbing systems-the absorption due to the lead dithizonate formed in the extraction and, in addition, the absorption due to the dithizone left in the solution after the reaction. 2. The characteristics of the dithizone solution used. (a) Changes in concentration from one dithizone solution to the next. I t is difficult to prepare dithizone solutions that are identical in concentration. (P) Degree of instability of the solution, whirh is especially significant in sunlight and warm weather. (c) The presence of impurities in the dithizone used. 3. ('ontamination from apparatus, reagents, and the atmosphere. T h r questions then arise: To what degree does a change in any of these factors affect the usability of the curve? If these factors are susceptible to appreciable change, will it be necessary to make almost day-to-day rechecks to be sure that the curve still applies?
+ k*c*L
+ k*K,T.'L = k,(C - 2K1V)L + 2kZKgVL
= ki(C DISCUSSION
+ k2c2L
-
K1V)L
Add 2 unit weights of metal.
Density of solution n unit weights of metal. Density of solution = kl(C - nKIV)L nk2KzVL = klCL - nklKIVL nk2K2VL = nVL(k2Kz - klK1) klCL which is, thus, the equation of the standard curve. When n = zero units: Density = (0 units) X [VL(k2K2- k1Kl)I klC' I, =k,CL The slope of the density curve-that is, the rate of change of the density with change in unit weights of metal-is the differential of t'he density expression: Slope = d(density)/dn = L (k2K*- kl K,) Add
++ +
+
v
IYTERPRETATIOR-
The curve of density versus weight of metal estracted is a straight line, as t,he slope is a constant and the original density. or concentration of dithizone, C, appears only in the expression that defines the zero inkrcept. Then the concentration of the dithizone solution used in the extraction has no effect on the dope
ANALYTICAL CHEMISTRY
1676 or shape of the standard curve. Changes in day-to-day concentration of the solution merely shift vertically the position of the curve. The same effect will obtain from the presence in the dithizone, in constant amount, of elements that react with dithizone under the same conditions and thus contribute to the density. Such elements may be originally present in the dithiaone or may be introduced as contaminants into the various reagent solutions that are used in the analysis. I n practice a blank can be run with each batch of samples, and the density of the blank sample can be used t o determine the zero intercept of the curve. From this zero intercept the line with the standard slope (once the slope has been determined) can then be drawn, and the values for the samples can be read directly. Curves obtained a t different intervals of time will be a series of parallel lines with varying zero intercepts. The controlling factor in determining the standard slope is in the optics of the spectrophotometer. An occasional check point may be made to ascertain the consistency of the spectrophotometer. Other than that, the analyst may safely eliminate the time-consuming, repetitious rechecks of the standard curve that are usually recommended. C4LCULATION OF SLOPE
Proceeding from the above theoretical expression of the standard curve, it is possible t o calculate numerical values for the terms and t o derive a theoretical value for the slope of the curve. Let lead be the metal to be extracted, and assume that the ketotautomerism of the dithizonate is formed: K1 = dithizone equivalent of lead Equivalent weight of lead = (207.2)/2 Equivalent weight of dithizone = 256.06
x
K1 = 21-[ 286'06 =
[#]
x
Slope of standard curve
=
('+)
Standard slope =
V L (k,KZ
- klK1)
X 1 [ ( 8 4 . 4 3717 )(3 ~)
=
(24.25) (5$2)]
E18l [(292.13) - (59.89)]
=
15
(232.24) = 0.0185 unit density per microgram of lead
Over a period of several months, the slopes of eight standard curves were determined experimentally for use in the determination of trace amounts of lead. These lead analyses were made as part of a study of carnotite ores for purposes of age determination of minerals (1, d ) . Derived from points from 0 to 20 micrograms of lead, the values of the slopes obtained were 0.0155, 0.0155, 0.0159, 0.0163, 0.0164, 0.0164, 0.0166, and 0.0166, with zero intercepts that varied between a density of 0.107 and 0.204 The arithmetical average of the slopes of these curves is 0.01615, or 0.0162 unit density per microgram of lead. The difference between the experimental and theoretical values is 4.3 % of the experimental-a value that is within the realm of accuracy of such colorimetric analyses. I n the derivation of the numerical value of the standard slope, it was assumed that the lead dithizonate equivalent weight n'as 717.3, which corresponds to the keto form of the compound 1
/N----N
1
S=C
c=s
Ph
(1%-eightof lead)
(weight of lead)
KZ = lead dithizonate equivalent of lead Equivalent weight of lead = 207.2 Equivalent weight of lead dithizonate = 717.3
rather than 561.39, which corresponds to the enol form: /Ce&
//S-N
C -S-Pb
\
\\-=S,'
Let V = the unit weight of metal ml. solution)
V
=
(1 microgram of lead)/(l6-
E OolL (milligrams of lead)/(milliliters of dithizone) 15
L
= length of light path in solution. k, = extinction coefficsient of dithizone. To obtain this constant, a dithizone solution in chloroform Has prepared from purified dithizone, a t a concentration of 8 mg. per liter. The density was determined a t a wave length of 520 mm. and slit width of 0.03 mm., with a light path of 1 cm. (Coreu cell). The density was 0.194. Therefore, 0.194 = k1ClL
0.194 = k , (S/lOOO) k1
x
1
= - = 24.25 cm./mg.
8
To obtain this coefficient, 20 ml. of a solution that contained 1 mg. of lead per milliliter was adjusted to a pH of approximately 9.5 and shaken with 15 ml. of the solution of dithizone. The density of the red dithizonate solution a t a wave length of 520 mm. and slit width of 0.03 mm. was determined to be 0.946. Therefore, 0.946 = k2C2L 717.3
k2
= extinction coefficient of lead dithizonate.
k --
(':)(%) -
=
84.43 cm./mg.
C '& The close correlation of the experimental and theoretical results substantiates this initial assumption. The results also indicate that neither a change in the partiti011 coefficient of dithizone between the chloroform and aqueous phases, which might arise with the variation of total lead extracted, nor the incidence of molecule and ion association within the solution, has enough of an effect to alter significantly the shape of the standard curve. Over its entire length the system follows Beer's law. It is thus possible, when the metal-dithizonate to be formed is lead, t o calculate a check curve for the experimentally determined standard curve. This calculated curve is most helpful when the range of lead to be determined is 1 t o 2 micrograms or less; for n i t h such dilute solutions contamination of the sample is particularly troublesome, and the points determining the curve could assume very random positions. The analyst, in determining a standard curve for use in such lead determinations, may, by reference to the calculated check curve, make a judicious selection of the points that obviously represent contaminated samples LITERATURE CITED
(1) Milkey, R. G., U. S.Geol. Survey, Trace Elements Investigations, Rept. 245 (1952). (2) Stieff, L. R., Stern, T.IT., and Milkey, R. G., Ibid., 268 (1952). RECEIVEDfor review April 7, 1952. Accepted August 13, 1952. Publication authorized b y the Director, U. S. Geological Surrey. Washington, D. C.
