Mikrochimica Acta [Wien] 1978 II, 309--314

MIKROCHIMICA ACTA 9 by Springer-Verlag 1978

Lawrence Livermore Laboratory, University of California Livermore, California 94550, U. S. A.

The Nonaqueous Microdetermination of Nitroguanidine* By

Walter Selig With 1 Figure

(Received March 30, 1978) Nitroguanidine (NQ) is a c o m p o u n d used in propellants. According to its crystal structure 1 it would be better named "guanyl nitramine" : H2N\ C 9 =N-N02 H2N/ Methods for the determination of nitrogen in N Q were evaluated 20 years ago by Fauth and Stalcup 2. Only one non-aqueous titrimetric m e t h o d for N Q , by De Vries et al. 3, is found in the literature. These authors determined N Q derivatives (1) as bases in N,N-dimethylformamide (DMF) and ethylenediamine (EDA) with sodium * Work performed under the auspices of the U.S. Department of Energy under contract No. W-7405-Eng-48. Notice: "This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the United States Department of Energy, nor any of their employees, nor any of their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes any legal liability of responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately-owned rights." Reference to a company or product names does not imply approval or recommendation of the product by the University of California or the U. S. Department of Energy to the exclusion of others that may be suitable.

0026-3672/78/7802/0309/$ 01.20

310

W. Selig:

methoxide, and (2) as acids in trifluoroacetic acid with perchloric acid in trifluoroacetic acid. In this work we have reinvestigated the nonaqueous titrimetric methods presented by De Vries et al. a. We have developed a simple micromethod for the determination of N Q in acetic anhydride, by titration with perchloric acid in glacial acetic acid. Nitramines such as 1,3,5-trinitro-l,3,5-triazacyclohexane (RDX) and 1,3,5,7-tetranitro1,3,5,7-tetrazacyclooctane (HMX) do not interfere.

Experimental The titrant was approximately 0.05N perchloric acid in glacial acetic acid. It was prepared by adding 4.25 ml of 70% perchloric acid and 20 ml of acetic anhydride to 500 ml of glacial acetic acid, and diluting to 1 liter with glacial acetic acid. The titrant was standardized against potassium acid phthalate or sodium acetate trihydrate, dissolved in glacial acetic acid. N Q and NQ-containing composites were dissolved in acetic anhydride (Aldrich, 99 + %). Aliquots containing up to 5 mg of N Q per 10 ml were diluted to 50 ml with acetic anhydride prior to titration. The titration system was controlled by a Tektronix 4051 graphics system. Details of its operation have been previously described 4,5. Emf's were monitored by a glass indicator electrode and a ceramic fiber junction calomel reference electrode (Beckman No. 39402), in which the salt bridge was a saturated solution of tetramethylammonium chloride in methanol. (A platinum electrode may be substituted for the glass electrode.) Stirring was provided by a magnetic stirrer. The stirring motor was separated from the titration vessel by a water cooling-plate and an aluminum plate connected to ground. Titration endpoints were calculated according to Savitsky and Golay 6. A convolute was used for a third-order second derivative using 25 points. The zero-crossing was found later by linear interpolation near the sign change. Titrations were performed at 23 + 10 C. Discussion and Results Titrimetric methods for the determination of N Q are based on its reduction with buffered titanous chloride 7 or with ferrous ion in concentrated sulfuric acid 8-1~ Our interest in nonaqueous titrimetry prompted us to reinvestigate the only method reported in the

The Nonaqueous Microdeterrninationof Nitroguanidine

311

literature, by De Vries et al. 3. These workers studied the acidic and basic character of nitroguanidines and related compounds. Nitroguanidine derivatives were titrated as acids in DMF or EDA with sodium methoxide; however, no results were presented for N Q alone. Indeed Aubertein and Pascal 1~ stated that the titration 'with sodium methoxide in DMF could not be applied to NQ. In addition, recent studies by Bissett and Levasseur 11 show that while N Q is fairly stable in water and dilute acid, it rapidly decomposes at pH > 10. Titration of NQ in DMF vs. sodium methoxide, using an antimony/calomel electrode couple according to De Vries et al. 3, yielded small endpoint breaks, of approximately 40 inV. Aliquots of the same stock solution did not yield reproducible results. The apparent equivalent weight, calculated from the sample weight and the meq of titrant used, varied from 70--80. Additional titrations were made with 0.05N tetrabutylammonium hydroxide using a glass/modified calomel electrode system. The solvents were DMF, pyridine, dimethylsulfoxide and 1,1,3,3-tetramethylguanidine. Although good titration curves were obtained, they were not reproducible; the apparent equivalent weight varied from 45 to about 80. This corroborates the studies of Bisset and Levasseur 11 and confirms that nitroguanidine cannot be titrated as an acid. The moieties titrated are probably acidic fragments resulting from the decomposition of NQ in the basic medium. De Vries et al. 3 titrated NQ as a base in trifluoroacetic acid; the titrant was perchloric acid in trifluoroacetic acid. Two platinum electrodes were used for endpoint indication. According to these workers NQ yielded smooth titration curves with sharp endpoint breaks, although they stated that the recovery values were not highly accurate. No data or titration curves were presented. Our preliminary experiments with NQ in trifluoroacetic acid, using perchloric acid in glacial acetic acid, yielded very shallow titration curves with small endpoint breaks, of about 30 inV. A glass/modified calomel electrode couple was used. The 2 platinum electrodes recommended by De Vries et al. did not produce any endpoint break. This, in addition to the difficulties in working with trifluoroacetic acid which requires exclusion of moisture as well as carbon dioxide, led us to abandon further experiments with this solvent. Streuli 12 and Wimer 13 studied acetic anhydride as a solvent for the titration of weak bases. A number of compounds which do not exhibit basic properties in water, glacial acetic acid, and acetonitrile could be quantitatively titrated in the anhydride. Our preliminary experiments showed that acetic anhydride is a satisfactory solvent

312

W. Seiig:

for the titration of NQ with perchloric acid in glacial acetic acid. Its advantages over trifluoroacetic acids are as follows: (1) steeper titration curves and larger endpoint breaks are obtained, (2) the solvent is not nearly as hygroscopic as trifluoroacetic acid; no special precautions for the exclusion of moisture and carbon dioxide are required, (3) the solvent is considerably cheaper than trifluoroacetic acid. A typical titration curve for N Q in acetic anhydride vs. perchloric acid in acetic acid is shown in Fig. 1. The mean endpoint break is

Titration of 9~79 mg of NO in

f

800 --

600

0

~

I 0.8

,

I

,

1.6 Titrant volume, m[

I

2.4

~

I 3.2

Fig. 1. Titration curve of nitroguanidine in acetic anhydride

approximately 75 mV. The precision of the method for 2 to 10 mg of NQ is shown in Table I. Table I. Statistics for the Microdetermination of Nitroguanidine in Acetic Anhydride Taken, mg NQ

Number of replicates

Std. deviation, % absolute

9.8 4.6 2.2

6 6 8

0.27 0.52 0.30

Because of the limited solubility of the commonly used standards (acid potassium phthalate, and sodium acetate trihydrate) in acetic anhydride, standardization in glacial acetic acid is recommended. Alternatively, for the determination of NQ in explosive composites,

The Nonaqueous Microdetermination of Nitroguanidine

313

one can standardize the titrant vs. the lot of N Q used in the composite to be analyzed. The solubility of N Q in acetic acid is sufficient for the solution of 0.05 mM ( ~ 5 rag) per 10 ml. Solution can be speeded up by ultrasonic agitation. Titrations are also feasible if N Q is dissolved in glacial acetic acid and sufficient acetic anhydride is added for an anhydride: acid ratio of >4 : 1. This increases the amount of N Q that can be dissolved. However, as the amount of acetic acid increases, the steepness and magnitude of the endpoint break decreases. In acetic acid alone, as mentioned by De Vries et al. 3, N Q cannot be titrated. The steepness of the titration curves and magnitude of the endpoint break could probably be increased by using a titrant of perchloric acid in anhydride. This titrant, however, becomes quickly discolored and, unlike the very stable perchloric acid in acetic acid, decomposes gradually. Table II. Analysis o5 NQ in Composites Composite

mg sample, 10 ml aliquot

Number of replicates

NQ, mean %

Standard deviation

HMX/NQ/Estanea (51.4/44.4/4.2) HMX/NQ/Vitonb

12.2

6

43.72

0.12

10.3

4

48.67

0.11

(47.5/47.5/5)

a A polyester urethane (B. F. Goodrich Chemical Co., Cleveland, OH). b A copolymer of hexafluoropropylene and vinylidene fluoride (E. I. du Pont de Nemours and Co., Wilmington, Del.). Results for the analysis of N Q in two propellant composites are presented in Table II. Both composites contained approximately 5% of a plastic bonding agent, and equal amounts of N Q and H M X . Neither the plastics nor H M X exhibited any acidic behavior. Thus N Q can be determined in their presence without prior separation. Acknowledgement The author wishes to express his sincere appreciation to Ted Sofios.

Summary The Nonaqueous Microdeterrnination o[ Nitroguanidine Nitroguanidine can be determined as a base by nonaqueous titration in acetic anhydride. The titrant is perchlorJc acid in glacial acetic acid. Endpoints are determined potentiometrically by a glass

314

W. Selig: The Nonaqueous Microdetermination of Nitroguanidine

(or platinum)/modified calomel electrode system. Nitramines such as H M X and R D X do not interfere.

Zusammenfassung Nitroguanidin i/it~t sich als Base in Essigsfiureanhydridl6sung mit Perchlorsiiure, gel6st in Eisessig, titrieren. Der Endpunkt wird potentiometrisch mit Hilfe einer Glas-(oder Platin-)Elektrode gegen eine modifizierte Kalomelelektrode gemessen. Nitramine wie HMX und RDX st6ren nicht.

References 1 j. H. Bryden, L. A. Burkhardt, and E. W. Hughes, Acta Cryst. 9, 573 (1956). M. I. Fauth and H. Stalcup, Analyt. Chemistry 30, 1670 (1958). 3 j. E. De Vries, S. Schiff, and E. St. Clair Gantz, Analyt. Chemistry 27, 1814 (1955). 4 A. M. Kray, Lawrence Livermore Laboratory, Rept. UCRL-79938 (1977). 5 K . A. Gertz, Lawrence Livermore Laboratory, Rept. UCRL-79937 (1977). 6 A. Savitsky and M. J. E. Golay, Analyt. Chemistry 36, 1627 (1964). 7 M. Roth and R. F. Wegman, Analyt. Chemistry 30, 2036 (1958). 8 C. Frejacques and M. Leclercq, Mere. Poudres 28, 39 (1956). 9 S. Sandi and G. Flanquart, Chim. analytique 39, 20 (1957). 10 R. Aubertein and H. Pascal, Mere. Poudres 40, 113 (1958). 11 F.H. Bissett and L.A. Levasseur, U.S. Army Natick Research and Development Command, Rept. TR-76/47 (1976). 12 C. A. Streuli, Analyt. Chemistry 30, 997 (1958). 13 D. C. Wimer, Analyt. Chemistry 30, 77 (1958).

Correspondence and reprints: Dr. W. Selig, Lawrence Livermore Laboratory, University of California, Livermore, CA 94550, U. S. A.