ctr
tometric
er
tion of Cyana
A. BUYSKE and VINCENT DOWNING Experimental Therapeofics Research, lederle laboratories, American Cyanamid Co., Pearl River, N. Y.
A spectrophotometric method for e quantitative determination of cyanamide in complex mixtures hes been developed. This procedure measures the absorbance a t 530 rnp of the red i s formed when complex that cyanamide and pentacyanoamrnineferrate interact at pH 10.5 in carbonate buffer. The lower limit of sensitivity i s about 1 pg. Recoveri blood, urine, and soil are 9 better. Contaminants or additives often found in cyanamide preparations do not interfere with the determination.
to studies on the pharmacology of calcium cyanamide and on the mechanism of action of this compound in the treatment of alcoholism, a simple, sensitive, and specific method of determination was desirable. I n the diverse uses of calcium cyanamide as a chemical intermediate as well as in its application in agriculture as a fertilizer, the methods for determination most widely used involved the formation of an insoluble silver salt complex, fol!owed by n total nitrogen determination (1) or a back-titration of excess silver ion with potassium thiocyanate (6, 8, 10). Recently a spot test procedure (7) and an efficient paper chromatographic method (9) for the separation and qualitative detection of microgram quantities of cyanamide have been described. Since none of these procedures had the specificity or sensitivity required for the quantitative determination of the compound in complex mixtures, an alternative method was sought. A promising lead was suggested in the work of Buchanan and Barsky (4), who studied the fleeting red color which was occasionally observed when ironcontaining ores were treated with a solution of crude sodium cyanide. They established that this color was due to an interaction of calcium cyanamide, which was present as an impurity, with a complex of ferrocyanide ion. The work described herein ie an adaptation of the observations reported by these earlier workers, and their extension t o a simple spectrophotometric method of the determination for cyanamide in various mixtures. RELIMINARY
E X ~ € R I M ~ ~ ~ A ~
Apparatus.
1
e
A Beckman 1Model B
ANALYTICAL CHEMISTRY
spectrophotometer with Corex cells of 1.002-em. light path was used for t h e cyanamide determinations. All measurements of p H were done with a Leeds &- Northrup Model 7664 pH meter. eagents. 4 standard solution of cyanamide was prepared b y dissolving 20 mg. of powdered reagent grade calcium cyanamide in 20 ml. of 0.1N hydrochloric acid. Sufficient 1.ON sodium hydroxide was then added t o increase the pH t o 7 and water was added to yield a total solution volume of 100 mi. This solution contained 200 pg. of calcium cyanamide or 105 pg. of cyanamide per ml. A buffer solution was prepared by the slow addition of concentrated hydrochloric acid to a 0.2M solution of sodium carbonate until the p H was lowered to 10.5. An aqueous solution of 0.2M trisodium pentacyanoanimineferrate, N a r [Pe(CN)6 NHa] (K a n d K Laboratories, Inc., Long Island City, N. Y . ) , was prepared. Preparation of Standard Curve. T h e standard calcium cyanamide solution was added in 0.1-, b.2-, 0.3-, 0.4-, and 0.5-ml. amounts to test tubes coniaining 2.0 ml. of buffer and 0.2 nil. of 0.2144 Naa[Fe(CN)j "31. Water mas added to bring the total volume t o 3.5 nil. A tube containing 2 ml. of buffer, 0.2 ml. of Nas[Fe(CK)b N?3], and 1.3 ml. of vater but no cvanamide provided the reagent blank- solution. This reagent solution had a n absorbance of less than 0.05 when read a t 530 rnp against distilled water. After the contents had been mixed, each tube was allowed to stand for 45 minutes under the normal conditions of laboratory light and temperature and its absorbance a t 530 mp determined against the reagent black. A plot of absorbance LIS. concentration yielded a straight line which intersected the origin. This graph served as a standard curve for the determination of cyanamide in the concentration range of 0 to 52.5 fig. per sample tube. Determination in Soil. One hundred milliliters of O.1N hydrochloric acid was added t o 30 grams of soil, t h e mixture was stirred vigorously for 30 minutes and filtered, and sodium hydroxide was added t o a pH of 7. This solution was then assayed directly and the concentration of cyanamide found by a comparison with the standard curve. Determination in Blood. Whole dog blood was collected in a glassstoppered tube containing sodium fluoride as anticoagulant. An equal volume of a solution of 10% trichloro-
acetic acid in water -was added to the blood and the tube shaken vigorously. After standing for 20 minutes, t h e sample was centrifuged at 2100 r.p.m. for 10 minutes. The supernatant liquids were filtered through Whatman No. 2 filter paper and the filtrate was adiusted to oH 7 before analysis. Determination in Urine. Urine was assayed directly, provided i t was not too heavily pigmented. An alternative procedure was to adjust the urine t o p H 2 with concentrated hydrochloric acid and t o extract with 10 volumes of ethyl acetate which previously had been equilibrated with 0.01N hydrochloric acid. The ethyl acetate was removed by in vacuo distillation on a rotary evaporator, and the residue in the flask dissolved in the buffer solution for direct analysis. RESULTS
The absorption spectra for the red complex of the pentacyanoammineferrate with a solution of calcium cyanamide are shown in Figure 1 Because cyanamide is relatively unstable and not readily available, the calcium complex x a s used as a source for free cyanamide. This was justified, because solutions of freshly prepared cyanamide containing no calcium and a solution of calcium cyanamide produced identical absorption spectra when each reacted with the pentacyanoammineferrate reagent. When calculated on the basis Of cyanamide equivalents, an absorptivity of 75 was obtained for both these preparations. A study of the conditions for the reaction (Table I) resulted in the selection of p H 10.5, a reagent volume of 0.2 ml., and a reaction time of 46 minutes as an optimum combination for the maximurn development of color. For each determination in Table I, 40 pg. of calcium cyanamide was assayed in a buffer volume of 2.0 ml. and a total volume of all. ingredients of 3.5 ml. per tube. When a 0.2-ml. cell with a light path of 1.0 em. was used and the volumes of the buffer, reagent, and sample were appropriately reduced, the lower limit of the sensitivity of the method was less than 1 pg. per determination. Compounds that are associated with cyanamide as a result of degradation and polymerization or are present ae
ulated (4) to result from the following mechanism . [Fe(CN)o]- 4
$ [Fe(CN)a] + CN-3
7
[Fe(CN)a]-3
(1)
+ HzNCN
[Fe(CN)J&NCN]- 8 (2) Deep red color Atmospheric oxygen was thought to play no role in the formation of the complex. Baudisch (3) confirmed the photosensitivity but attributed the reaction to result as follows:
Figure 1. Absorption spectra of the product of the reaction of cyanamide with pentacyanoammineferrate in alkaline aqueous solution
[Fe(CN)a]- 4 contaminants or additives in special preparations include urea, ammonia, dicyandiamide, melamine, guanidine, cyanide, and citric acid. Table 11 shows the absorbance at 530 mp when solutions containing these compounds with and without calcium cyanamide are determined by the method described. The compounds tested were present a t a concentration five times higher than that of calcium cyanamide or almost ten times higher than the concentration of cyanamide equivalents. The detectable absorbance noted for dicyandiamide may be due to a small impurity of cyanamide. When present in concentrations equal to that of cyanamide, none of the compounds in Table 11 caused any inhibition or increase in the development of the red complex. Khen solutions of 5 to 40 pg. of calcium cyanamide per ml. of whole heparenized blood or urine were prepared and determined by the method described in the procedure section, recoveries varied from 80 to 90% and from 78 to 82%, respectively. If a second extraction of urine with 10 volumes of fresh ethyl acetate was done and the procedure followed from this point on as described, recoveries were 93% or better. The reproducibility of the determination of aliquots of the same sample was to & L O % . The recoveries from soil that contained 40 and 200 pg. of calcium cyanamide per gram of sample were 95 and 98%. respectively. However, if the amount of cyanamide present is below 20 pg. per gram of soil and the acid filtrate is highly pigmented, it may be difficult to determine directly. I n such cases an ethyl acetate extract similar to that used for the urine would perhaps be useful. When the paper chromatographic system described by Milks and Janes (9) was used in our laboratory, 5 to 10 pg. of cyanamide produced a clearly visible red color when the paper was sprayed with a solution of 0.2M N ~ ~ [ F ~ ( C N ) ~ dissolved N H Z ] in the p H 10.5 carbonate buffer.
Table I. Conditions for Optimum Color Development of the Reaction of Cyanamide with Pentacyanoammineferrate
Absorbance at, 530 ?Lip in
pH 5.0 7.0 8.5
9.5 10.0 10.5 11.0 10.5
0 . 2 ~ Presence Reaction T\’aa[Feof Time, (CN)hNHg], 40 pg. Min. M1. CaCT\‘a 30 0.2 0.08 30 0.15 0.2 30 0.32 0.2 0.45 0.2 30 0.2 30 0.50 30 0.2 0.52 30 0.2 0.50 5 0.2 0.38 10 0.2 0.48 0.50 20 0.2 0.52 30 0.2 0.52 0.2 60 0.50 300 0.2 30 0.48 0.05 0.10 0.50 0.52 0.15 0.52 0.20
DISCUSSION
Buchanan and Barsky (4) and Baudisch (3) reported that ferrocyanide solutions formed a red complex with cyanamide. This reaction was shown to be light-dependent and was post-
Table II. Effect of Various Compounds on Development of the CyanamidePentacyanaammineferrate Colored Complex
Calcium cyanamide Cyanamide Urea Ammonia Dicyandiamide Melamine Guanidine Cyanide Citric acid
Sesayed,
Absorbance at 530 M p Plus
”&%:
Alone CaCNz
40
21 200 200 200 200 200 200 200
40 fig.
0.51 . . . 0.51 -0.02 0.50 -0.01 0.50 0.07 0.58 -0.02 0.50 0.05 0.52 -0.02 0.50 0 . 0 0 0.51
+
HzOlight [Fe(CN)sOH]-4
+ HCN
(3)
The pentacyano-hydroxyl complex then was thought to undergo autoxidation in air to produce a n intermediate which reacts with cyanamide to give a red complex. It was confirmed in our laboratories that aqueous solutions of ferrocyanide (but not ferricyanide) would react with cyanamide, but i t was laecessary first to activate ferrocyanide by a 24to 36-hour aeration before use. I n the presence of cyanamide, a t an alkaline pH, these freshly prepared solutions developed a red color only when exposed to strong light. A useful quantitative method of analysis was based on these observations. However, the inconrenience and the difficulty of reproducing the reagent left something to be desired. The resemblance of the pentacyanoammineferrate, Ka3[Fe(CN),NH,], to the intermediate postulated in Equation 3 and the availability of this compound prompted the attempt to substitute it for the light- and air-activated ferrocyanide. This work resulted in the quantitative assay procedure as described. After the completion of this work, Fearon’s description (6) of a qualitative test for guanidines, urea, and thiourea came to the attention of the authors. This worker studied a variety of compounds containing an amidine or related functional groups and listed cyanamide as slowly yielding an orange-red color when in contact n i t h a solution of freshly prepared pentacyanoanimineferrate. Although no effort was made to adapt this finding to a quantitative method of analysis, Fearon must be credited as among the first to indicate the potential utility of pentacyanoammineferrate as an analytical reagent. I n view of the requirement for light and air before ferrocyanide will react with cyanamide, the effects of these two variables were studied on the development of color when the pentacyanoammineferrate reagent was used, Solutions of all the reactants were prepared in water that previously had been boiled and cooled under nitrogen VOL. 32, NO. 13, DECEMBER 1960
1799
to reduce the oxygen content. Thebe solutions were mixed and the reaction was allowed to take place in total darkness. The absorption spectra, the intensity of color, and the speed at vr-hich the maximum intensity was reached were compared with solutions saturated with oxygen and developed in direct light. There was no detectable difference in any of these samples and neither light nor air was required for the interaction of cyanamide 15 ith this particuiar reagent. Sodium nitroprusside, Naz[Fe(Cn’)~-
NO], will also react with cyanamide to
produce a color (4) with an absorption maximum a t 530 mp. The xork of Raudisch (8, 3) and Fearon (6) suggests that this reaction may in fact be due to the photocatalyzed partial hydrolysis of nitroprusside to form pentacyanoaquoferrate, [Fe(CN)&Tz01-3, which in turn reacts with
cyanamide. This aquo analog of pentacyanoammineferrate differs only in the substitution of a molecule of mater for one of ammonia and has been reported (6) to have the same general reactivity as that of the ammino analog. There has appeared in the literature no report on the use of sodiuni nitroprusside for the quantitative analysis of cyanamide. In our own laboratory, various combinations of pH, buffers, concentration of nitroprusside, and illumination were employed, but the optimum sensitivity obtained when nitroprusside was used to determine cyanamide was only 25y0 of that obtained with pentacyanoammineferrate. ACKNOWLEDGMENT
The authors are indebted to E. A. Peets and Allan Shurr for technical
assistance and to E. H. Dearborn for encouragement and continued interest during the course of this work. LITERATURE CITED
(1) ASSOC.Offic. Agr. Chemists, “Methods of Analysis,” 7th ed., p. 15, 1950. (2) Baudisch, O., Ber. 62, 2706 (1929).
(3) Ibid., 68, 769 (1935). (4) Buchanan, G. H., Barsky, G., 2. angew. Chem. 44,383 (1931). (5) Capitani, C., Gambelli, G., Chinz. e ind. (Milan)35,890 (1953). ( 6 ) Fearon, W. R., Analyst 71, 562 (1916).
(7) Feigl, F., Gentil, V., Mikrochim.
Acta 1959, 44, (8) Inaba, H., Japan Analyst 3, 196 (19.54). (9) Milks, J. E., Janes, R. H., ANAL. CHEW28,846 (1956). (10) Sato, M.,Sato, J., Fujesaiva, T., J . Electrochem. SOC. Japan 22, 370 (1954). \ - - - - I
RECEIVEDfor review April 7 , 1960. Accepted July 25, 1960,
Chlorate I rn p w rities
n I EUGENE A. BURNS1 Jet Propulsion laboratory, California lnsfitute o f Technology, Pasadena, Calif.
A rapid, accurate spectrophotometric method for the determination of small amounts of chlorate impurities in ammonium perchlorate is based on the production of a colored haloquinone b y reaction of chlorate with benzidine in a hydrochloric acid medium. The effect of temperature and time of color development on the intensity of the colred habquinone has been investiuted. A method for the determination of trace amounts of reducing impurities [sulfite) or oxidizing impurities [chlorine) of reagent grade hydrochloric acid is reported.
applied to rocket-grade ammonium perchlorate, which requires better than 10% relative accuracy. Accordingly, the benzidine-chlorine test (1) has been investigated for applicability and adaptability to a spectrophotometric method which would be suitable for use by semitechnical personnel. The spectrophotometric procedure is based upon the color produced when chlorate ion is converted to chlorine: C108-
+ 5C1- + 6H+
+
3C12 -I- 3HzO
(1)
The chlorine subsequently oxidizes benzidine to 8, yellow haloquinone (1) many years the procurement and quality control of ammonium perchlorate for employment as a solidpropellant oxidizer have lacked a rapid, accurate, and sensitive analytical method for the determination of chlorate impurities. Because of the low chlorate content, conventional “wetchemistry” techniques (3) are inadequate. A colorimetric method sensitive to low chlorate levels is currently in use (g), but is unsatisfactory when OR
e
ANALYTICAL CHEMISTRY
However, because of favorable reaction rates, the experimental parameters can be adjusted so that a desirable quantitative method is obtained. This method is not selective for chlorate; corrections must be made for the presence of oxidizing agents capable of oxidizing chloride to chlorine-e.g., bromate. APPARATUS
A Beckman Model DK-2 ratiorecording spectrophotometer was employed. If the room temperature is subject to variations greater than &t5O C., a constant temperature bath is necessary. REAGENTS
NH2 +C1-
The method is complicated by decomposition of the haloquinone:
NHnfC1-
-+
decay product^
Bemidine Solution, 0.4yG. (y./v.). Dissolve 2.0 grams of benzidine in 2.5 ml. of reagent grade hydrochloric acid and 100 ml. of distilled water, heating 1 Present address, Propulsion Department, Poulter Laboratories, Stanford Research Institute, Menlo Park, Calif.
