Citation
The reactions of chloramine and dimethyl-chloramine with some arsine derivatives

Material Information

Title:
The reactions of chloramine and dimethyl-chloramine with some arsine derivatives
Creator:
Krannich, Larry Kent, 1942-
Publisher:
[s.n.]
Publication Date:
Language:
English
Physical Description:
ix, 164 leaves. : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Chloramines ( lcsh )
Organoarsenic ( lcsh )
Chemistry thesis Ph. D ( lcsh )
Dissertations, Academic -- Chemistry -- UF ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 158-160.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Larry Kent Krannich. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
021456451 ( ALEPH )
18066747 ( OCLC )

Downloads

This item has the following downloads:


Full Text
THE REACTIONS OF CHLORAMINE AND DIMETHYLCHLORAMINE WITH SOME ARSINE DERIVATIVES
By
LARRY KENT KRANNICH
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1968




ACKNOWLEDGMENTS
The author wishes to take this opportunity to
express his sincere gratitude to his research director, Dr. Harry H. Sisler. Although busy with many administrative responsibilities, Dr. Sisler was always available to consult with the author and to provide encouragement and valuable suggestions concerning this research. The manner In which he permitted the author to pursue and design the major portion of the research program was sincerely appreciated.
The author takes this opportunity to recognize the many valuable discussions concerning the progress of this research that he participated in with Dr. Kurt IUtvary, Charles Watkins, Ronald Highsmith, and other members of his research group. Their interests and suggestions were greatly appreciated and were of great value.
Finally, the author acknowledges the support of this research by the National Institutes of Health.




TABLE OF CONTENTS
Page
ACKNOWLDEGMENTS .. ii
LIST OF TABLES v
LIST OF FIGURES vii
CHAPTER
I. INTRODUCTION . 1
Bibliography 3
II. REACTIONS OF AMMONIA-FREE CHLORAMINE WITH
TRIALKYLARSINES AND PHENARSAZINES . . 4 Historical Background . . . . . 4
Experimenta --- . 6
Discussion . 36
Summary . . 42
Bibliography 43
III. REACTIONS OF AMMONIA, DEUTERATED AMMONIA,
AND AMMONIA-CHLORAMINE MIXTURES WITH
DIMETHYLCHLOROARSINE . . . . . 45
Historical Background . . . . . 45
Experimental . 48
Discussion . 59
Summary 61
Bibliography . 61
iii




Page
IV. REACTIONS OF AMMONIA-FREE CHLORAMINE,
DIMETHYLCHLORAMINE, AND CHLORINE WITH ARSINE
AND WITH PRIMARY AND SECONDARY ARSINES . 63 Historical Background . . . . . . 63
Experimental 71
Discussion 132
Summary 157
Bibliography 158
V. GENERAL CONCLUSIONS AND SUMMARY . . . 161 BIOGRAPHICAL SKETCH. . . . . . . . . 164
iv




LIST OF TABLES
Table Page
1. Infrared Absorption Data (cm- ) . . . 10
2. N. M. R. Spectral Data for HN(C6H4)2AsCH3 * 14 3. N. M. R. Spectral Data for [(CH3)3AsN(CH3)2C1 26
4. N. M. R. Spectral Data for [(CH3CH2) 3AsN(CH3)2]C1 27
5. N. M. R. Spectral Data for [(CH3CH 2CH 2) 3 AsN(CH 3 )21C1 .. .. .. .. .. 33
6. N. M. R. Spectral Data for (HN(C6 H4)2As(CH3)(NH2]cl . . . . . 36
7. Infrared Absorption Data (cm- 1) . . ... 51
8. N. M. R. Spectral Data for (CH3)2AsNH+Cl. . 54 9. Infrared Absorption Data (cm-) . . . . 73
10. N. M. R. Spectral Data for CH3AsH2 .... 75
11. N. M. R. Spectral Data for (CH3)2AsH . .... 80
12. N. M. R. Spectral Data for C6H5AsH2 . 83
13. N. M. R. Spectral Data for (C6H5)2AsH .... 85
14. Mass Spectral Data for (CH3As)5 . .. 95
15. Mass Spectral Data for (CH 3)2AsAs(CH3)2 .... 96
16. Mass Spectral Data for (C6H5As)6 . ........ 100
17. Mass Spectral Data for AsH3-NH2Cl Reaction
Sublimate 103
18. Mass Spectral Data for Red Solid at 1500C 0 112 19. Mass Spectral Data for Red Solid at 190C . 113
V




Table Page
20. Mass Spectral Data for (C6H5 )2AsAs(C6H5 )2 . 129 21. Summary of Chloramine and Dimethylchloramine
Reactions with Arsines Containing As-H Bonds 133 22. Relative Intensities of Various AsCl Ion
x
Species 136
23. Chemical Shifts of Hydrogen Nuclei in the As-H
Bonds of Various Arsines . . . . . 139
24. Summary of Chlorine Reactions with Arsines
Containing As-H Bonds 151
25. Average Chemical Shifts of Methyl Groups in
(CH As) in Various Solvents, at Room
Tem erasure 155
vi




LIST OF FIGURES
Figure Page
1. Infrared Spectrum of HN(C6H4)2AsCl (Mull) . 13 2. Infrared Spectrum of HN(C6H4)2AsCH3 (Mull) . 15
3. N. M. R. Spectrum of HN(C6H4)2AsCH . . . 16 4. Chloramine Generator ..... . . . . 17
5. Infrared Spectrum of
[(CH CH2)3AsNH21Cl (Mull) . . . . . 20
6. Infrared Spectrum of
[(CH3CH2CH2) 3AsNH2C1 (Mull) . . . . 22
7. Infrared Spectrum of
[(CH3)3AsN(CH 3)2]Cl (Mull) . . . . . 24
8. N. M. R. Spectrum of [(CH3)3AsN(CH3)21C1 . 25
9. Infrared Spectrum of
[(CH3CH2)3AsN(CH3)21C1 (Mull) . . . . 28 10. N. M. R. Spectrum of
(CH CH2 3AsN(CH )2 C1 ... 29
11. Infrared Spectrum of
[(CH 3CH2CH2)3AsN(CH3)21C1 (Mull) . . . 31 12. N. M. R. Spectrum of
[(CH3CH2CH2)3AsN(CH3)2]C1 32
13. Infrared Spectrum of
[HN(C6H4)2As(Cl)(NH2)]Cl (Mull) . . . . 35 14. Infrared Spectrum of
[HN(C6H4)2As(CH3)(NH2)]Cl (Mull) . . . 37 15. N. M. R. Spectrum of
[HN(C6H4)2As(CH3)(NH2) Cl . . . . . 38
16. Chloramination Reaction Flask . . . . 49
vii




Figure Page
17. Infrared Spectram of
(CH3)2AsNH3Cl (Mull) 53
18. Infrared Spectrum of
(CH3)2AsND3Cl- (Mull) 56
19. Infrared Spectrm of
[ (C 3)2AsN] 4 2HC1 (Mull) .. 58
20. Infrared Spectrum of CH3 AsH2 (gas) . . . 76 21. N. M. R. Spectrum of CH3 AsH 2 . . . . 77
22. Infrared Spectrum of (CH3)2AsH (Neat) . . 78 23. N. M. R. Spectrum of (CH3)2AsH . . . . 79
24. Infrared Spectrum of C6H5AsH2 (Neat) .... 81
25. N. M. R. Spectrum of C6H5AsH2 .. .. . 82
26. Reaction Flask Having a Sintered Glass Filter Funnel Base o 84
27. Infrared Spectrum of (C6H5)2AsH (Neat) . . 86
28. N. M. R. Spectrum of (C6H5)2AsH . . . . 87
29. Infrared Spectrum of (CH3As)5 (Neat) . . 89
30. N. M. R. Spectrum of (CH 3As)5$ Neat and at Various Sweep Widths ......... 90
31. N. M. R. Spectrum of (CH As)5-C6D6 Solution at Various Sweep Widths . . . . . . 91
32. Temperature Dependence of Chemical Shifts of Methyl Groups in (CH3As)5 (Neat) . . . 93 33. Temperature Dependence of Chemical Shifts of Methyl Groups in (CH3As)5 (C6D6 Solution) . 94 34. Infrared Spectrum of (CH3)2AsAs(CH3)2 (Neat) 97
35. N. M. R. Spectrum of (CH3)2AsAs(CH3)2 ..... 98
viii




Figure Page
36. Infrared Spectrum of (C6H5As)6 . . . . 101 37. N. M. R. Spectrum of Yellow Residue from AsH -C12 Reaction 106
38. N. M. R. Spectrum of CH AsH2-C12 Reaction Product Prior to Condensation . . . ... 111 39. Infrared Spectrum of As-N Product from (CH3)2AsH-NH2C1 Reaction . . . . . 117
40. N. M. R. Spectrum of As-N Product from (CH3)2AsH-NH2C1 Reaction . . . . . 118
41. Infrared Spectrum of (C6H5)2AsAs(C6 H5)2 (Mull). 127 42. N. M. R. Spectrum of (C6H5 )2AsAs(C6H5)2 . . 128 43. Burns and Waser's Structure of (CH As) from X-ray Studies 154
. .. ..o154
44. Other Possible Structure of (CH3As)5 . . 154
ix




CHAPTER I
INTRODUCTION
During the past 10 years, the chloraminations of
amines, phosphines, and arylarsines have been extensively investigated in our laboratories. Increased interest in the chloramination of the arsines resulted when the diphenylarsenonitrilic tetramer was obtained from the pyrolysis of the chloramination and amination products of diphenylchloroarsine. This demonstrated the possibility of obtaining previously unknown arsenonitrilic polymers by a reaction analogous to one of those used for obtaining phosphonitrilics.(1)
R2AsCl + NH 2C1 + 2NH 3 R2As(NH2)2Cl + NHCl R2AC .NH R2 1'4CI
2 R2As(NH 22]Cl -. [ As*'''N"'AsR2]Cl + NH 4Cl NH NH2k
2 2
2 [R2As-"N- AsR2] Cl [R2AsN)k + 2NH 4Cl
NH 2 NH2
2 2
In addition, many interesting arsenic-nitrogen intermediates were possible.
The chloramination of trialkyl arsines, other than
trimethyl-, had not been carried out, and thus the generality of the chloramination reaction with substituted arsines. had not been demonstrated.
R 3As + NH2Cl [R3AsNH2 Cl
No chloramination reactions had been reported where
1




2
dimethychloramine was used as the chloraminating agent. The preparation and investigation of the infrared and proton magnetic resonance spectra of a variety of aminoarsonium salts could give information concerning the relative extent of di-pIr bonding in the As-N bond.
The existence of 10-chloro- and 10-methyl-5,10dihydrophenarsazine, where there are two potential sites
- nitrogen and arsenic for electrophilic attack by chloramine, presented the opportunity to study selective chloramination. In these cases, no difficulties could be encountered with transamination reactions, since the nitrogen is not bonded to the arsenic.
The reaction of phosphine with dimethylchloramine, carried out by Ronald Highsmith, to give the unexpected reaction
2PH + 3(CH3)2NC1 -4 2P + 3(CH3)2NH2C1
indicated a new type of reaction for dimethylchloramine and possibly for chloramine. We were interested in the possibility of obtaining a new synthetic method for the preparation of compounds containing bonds between two atoms of a Group Va element from the reaction of chloramine and dimethylchloramine with primary and secondary phosphines and arsines. Since the As-H bond should be more labile than the P-H bond, we thought that there would be a greater ease of obtaining As-As bonded compounds by this reaction than of P-P bonded compounds.




3
Therefore, the purpose of this study wa's (1) to
expand the present knowledge of the reaction of chloramine with Group V bases to include the substituted arsines;
(2) to investigate the properties of dimethylchloramine relative to those of chloramine; (3) to study the chloramination of the phenarsazines to determine the site of the attack of chloramine; (4) to prepare some heretofore unknown dialkylarsenonitriles by the reaction of an ammonia-chloramine mixture with dialkylhaloarsines; and (5) to investigate the use of dimethylchloramine and chlboramine in synthesizing diarsines and cyclic polyarsines from reactions with primary and secondary arsines.
Bibliography
1. S. E. Frazier, Master's Thesis, University of Florida,
December, 1963.




CHAPTER II
REACTIONS OF AMMONIA-FREE CHLORAMINE WITH
TRIALKYLARSINES AND PHENARSAZINES Historical Background
The original interest of our laboratory in chloramine chemistry had its beginning in the early 1950's when much attention was being given to hydrazine as a high energy rocket propellant. The reaction of chloramine and liquid ammonia gave as one of its products hydrazine.
NH2C1 + 2NH3 -+ H2NNH2 + NH4Cl
The reaction immediately created interest, because of the potential it held for easy routes to the synthesis of hydrazine and substituted hydrazines in good yields. Sisler, Omietanski, and others(1) showed that the chloramination of primary and secondary amines gives good yields of the corresponding hydrazines. Chloramination of tertiary amines gave the corresponding hydrazinium salts, since the absence of an hydrogen atom attached to the quaternary nitrogen prevented any protolysis reaction.
The scope of chloramine reactions was eventually
expanded by reacting chloramine and dimethylchloramine with tertiary phosphines, aminophosphines, and hydrazinophosphines.(2,3,4,5,6,7) The generalized reaction may be written
4




RR'R"P + NH2C1 [RR'R"PNH2 Cl
RR'R"P + (CH3)2NC1 -, RR'R"PN(CH3)2CI
R = alkyl, aryl, amino, and/or hydrazino groups
In all cases the chloramination occurs at the most basic site the phosphorus atom yielding phosphonium salts. The only exception to this is the reaction of dimethylchloramine and tris-dimethylaminophosphine to give the tris-dimethylaminochlorophosphonium chloride.(6) One method of rationalizing the attack on the phosphorus atom is to assume that nitrogen to phosphorus pr-dwr back-bonding increases the Lewis basicity of the phosphorus atom and decreases that of the nitrogen.
The reactions of chloramine with triphenylarsine
and trimethylarsine have been shown (8) to give good yields of the respective aminoarsonium chlorides. Recently work carried out in this laboratory has shown that trialkyl and triarylstibines react with ammonia-free chloramine and an ammonia-chloramine mixture to produce compounds of the type R3Sb(Cl)]2NH.(9)
The reactions of various substituted amines with chloramine to give hydrazines and hydrazinium salts have been demonstrated to proceed by a SN2-type reaction mechanism with the chloramine molecules acting as the electrophiles.(10,11,12) The assumption has been made that these electrophiles react with other substituted Group V nucleophiles by the same SN2 reaction mechanism




6
to give products containing a N-N, N-P or N-As bond. The only previously found exceptions to this occur in the reactions of alkylchloramine with tris(dialkylamino)phosphines(6'13) and dialkylphosphonates. (14) Experimental
Manipulation of Materials.-Because of the high
reactivity of the arsines and arsonium salts toward oxygen and water and the volatility of the arsines, many of the reactions were carried out in an all glass, high vacuum line. All experimental purification work was conducted in a nitrogen atmosphere in a Vacuum Atmospheres Model HE-43 Dri-Lab equipped with a Model HE-93 Dri-Train. All of the arsenic-containing starting materials were stored and transferred inside the Dri-Lab.
The technique used for admitting anhydrous, ammoniafree chloramine into the vacuum line was as follows. A 250 ml two-necked, round-bottom flask equipped with a vacuum line adapter and a filter funnel was connected to a take-off on the vacuum line. The filter funnel had a vacuum-back stopcock and a fritted glass filter. The entire flask, up to the vacuum-back stopcock, was evacuated and flamed with a torch to insure removal of all entrapped oxygen and moisture. Powdered anhydrous copper sulfate was added to the filter funnel to give a two inch pad. The ammonia-chloramine solution, which had been stored over a considerable quantity of anhydrous copper




sulfate, was pipetted (under a nitrogen stream) onto the copper sulfate pad in the funnel. The round-bottom flask was then immersed in a Dewar of liquid nitrogen. The stopcock to the vacuum manifold was closed, the stopcock tothe filter funnel was then opened, and the chloramine-ether solution was admitted into the flask at a temperature of
-196C. After almost all the solution had passed from the filter funnel, the stopcock was closed to prevent the last few milliliters of ether solution and any oxygen or moisture from entering the round-bottom flask. The ether solution was warmed to room temperature, frozen out at
-196C, and degassed. This degassing procedure was repeated several times. The chloramine-ether solution was then distilled through the vacuum manifold into a flask having volume graduation marks and a serum-capped sidearm. The solution was warmed to room temperature, helium was admitted into the manifold and calibrated flask until atmospheric pressure was attained, and a sample of the solution was removed through the serum cap using an hypodermic syringe. The concentration of the chloramine solution was then determined by reacting sample aliquots with acidified potassium iodide solution and titrating the liberated iodine with standard sodium thiosulfate solution.
The dimethylchloramine solutions were transferred into a calibrated tube inside a nitrogen filled dry-bag. This tube and its contents were attached to the vacuum line and degassed several times before distilling the dimethyl-




8
chloramine solution into the reaction flask. The determination of the concentration of dimethylchloramine solution was carried out in a manner analogous to that for chloramine solutions.
Benzene, petroleum ether, hexane, and diethylether were all obtained as reagent grade materials and were dried and stored over calcium hydride. Trimethylaluminum, triethylaluminum, and tri-n-propylaluminum were obtained from Ethyl Corporation and were used as obtained. Arsenic trichloride, arsenic trioxide, and diphenylamine were obtained from J. T. Baker Chemical Company as reagent grade chemicals and were used as obtained. Methylmagnesium bromide was purchased from Pennisular Chemical Research and used as obtained.
Analyses.-Elemental analyses were done by Galbraith Microanalytical Laboratories, Knoxville, Tennessee, and Schwarzkopf Microanalytical Laboratory, Woodside, New York. Nitrogen analyses were carried out using the Kjeldahl method. Melting points were determined using the Thomas Hoover Capillary Melting Point Apparatus with an uncorrected thermometer.
Infrared Spectra.-Infrared spectra were recorded
on a Beckman IR-10 Spectrometer. The spectra were obtained on the solids in the form of Kel-F mulls using KBr plates for the range 2.5-7.5A and as Nujol mulls using CsI plates for the range 7.5-20/. A summary of the spectral bands not listed in the literature of materials used and




9
produced in this study is found in Table 1.
Nuclear Magnetic Resonance Spectra.-The proton magnetic resonance spectra were measured on a Varian Model A-60-A nmr spectrometer. Tetramethylsilane, chloroform, and dimethylsulfoxide were used as internal standards in the respective instances.
Preparation of Trialkylarsines.-Trimethylarsine,
triethylarsine, and tri-n-propylarsine were prepared by reacting arsenic trioxide with the respective trialkylaluminum compound according to the procedure described by Stamm and Breindel(15) and by the equations
2AlR + As20 -+ (R2As)20 + 2RA10
3 2 32 2
(R2As)20 + AlR3 2AsR3 + 3RA10
Preparation of 10-chloro-5,10-dihydrophenarsazine.10-chloro-5,10-dihydrophenarsazine was prepared by reacting arsenic trichloride with diphenylamine(16) according to the following equation
HC6H4\HN(C 6H5)2 + AsCl3 \ H /AsC1 + 2HC1 6H4
A mixture of diphenylamine (50.6 g, 0.3 mole), arsenic trichloride (54.1 g, 0.3 mole), and o-dichlorobenzene (100 ml) was boiled under reflux for 15 hours in a 250 ml roundbottom flask. As the reaction flask cooled to room temperature, large quantities of material crystallized. The crystals were washed with petroleum ether and recrystallized from boiling carbon tetrachloride. The bright yellow, crystalline product was dried in vacuo.




10
TABLE 1(a)
INFRARED ABSORPTION DATA (cm-)
HN(C6H 4)2AsCl, Mull
3340(vs), 3040(w), 1900(w), 1601(vs), 1580(w), 1570(vs), 1510(m), 1480(s), 1465(vs), 1438(s), 1360(w), 1325(w), 1278(w), 1255(w), 1232(m), 1160(m), 1130(m), i065(s), 1020(w), 895(w), 445(m), 750(vs), 740(vs), 715(s), 590(s), 440(s), 400(m).
HN(C6H4)2AsCH3, Mull
3380(vs), 3050(w), 2980(w), 2910(w), 1900(w), 1590(s), 1570(s), 1500(w), 1480(m), 1460(vs), 1440(s), 1410(m), 1322(s), 1280(w), 1255(s), 1230(m), 1160(m), 1130(w), 1120(w), 1060(w), 1030(w), 970(w), 930(w), 880(m), 860(w), 845(s), 835(m), 760(vs), 745(s), 722(vs), 670(w), 650(m), 595(w), 550(s), 535(s), 450(s).
(CH3CH 2) 3AsNH2] Cl, Mull
3140(vs), 2920(vs), 2870(vs), 2800(vs), 1485(w), 1440(s), 1410(m), 1380(m), 1260(s), 1240(w), 1090(s), 1040(s), 1020(s), 970(m), 800(m), 770(w), 750(m), 730(s), 690(s), 605(m), 545(m).
(cH 3cH2cH 2)3AsNH2 Cl, Mull
3190(vs), 3180(vs), 3140(vs), 3100(vs), 3090(vs), 3090(vs), 2950(vs), 2930(vs), 2870(vs), 1450(s), 1410(m), 1350(m), 1300(w), 1214(m), 1080(s), 1050(m), 840(m), 720(s), 695(s).
[(CH3 ) 3AsN(CH3)2] Cl, Mull
2990(vs), 2900(m), 2860(m), 2800(m), 1460(vw), 1452(w), 1448(w), 1420(w), 1312(w), 1305(w), 1273(m), 1250(w), 1160(s), 1105(m), 1055(s), 942(vs), 915(vs), 860(m), 795(w), 712(vw), 650(w), 642(m), 610(m), 568(vs), 354(vw), 320(m).
[(CH3CH2 3AsN(CH 3)2] Cl, Mull
2960(m), 2930(vs), 2805(m), 1460(s), 1382(w), 1312(w), 1250(m), 1150(s), 1040(m), 930(vs), 825(w), 794(w), 757(s), 605(m), 596(m), 360(w).




(CH3CH2CH2)3AsN(CH )2]Cl, Mull
2960(vs), 2920(vs), 2870(s), 2810(m), 1450(m), 1410(w), 1370(w), 1310(w), 1250(w), 1220(w), 1210(w), 1160(m), 1060(s), 1037(w), 940(vs), 838(w), 790(vs), 723(m), 645(vw), 595(vw), 345(vw).
C6H4 C1l
HN6 H4As/ C1, Mull
-CH \NH
3230(m), 3140(s), 3040(s), 3000(s), 2830(m), 1607(s), 1582(m), 1565(w), 1510(w), 1460(vs), 1439(s), 1405(s), 1362(w), 1322(2), 1275(w), 1245(m), 1240(sh), 1230(m), 1155(w), 1138(w), 1065(m), 1024(m), 967(m), 922(s), 890(w), 847(m), 830(w), 800(w), 742(vs), 720(m), 670(s), 640(w), 590(w), 570(w), 530(w), 440(s), 388(w), 340(s).
[ ,C6H4 OsHc 3
HN H As/ CH3 C1, Mull
CH NH
6 4 2
3240(s), 3150(vs), 3050(vs), 2980(vs), 2930(vs), 2850(s), 1610(vs), 1585(vs), 1575(s), 1520(s), 1470(vs), 1455(s), 1445(s), 1410(m), 1360(m), 1245(m), 1165(m), 1140(m), 1110(w), 1075(m), 1030(m), 860(s), 752(vs), 720(m), 690(w), 645(w), 620(w), 597(w), 535(w), 450(m), 390(w), 340(w).
(a)s, strong; m, medium; w, weak; v, very; sh, shoulder.




12
Mp 189-189.50C (Lit..(16) mp 191-192oC). The yield of 10-chloro-5,10-dihydrophenarsazine was 52 g or approximately 50% of theory based upon the amount of arsenic trichloride put into the reaction. The infrared spectrum is shown in Figure 1. No nmr spectrum was obtained because of the low solubility of the product in the common deuterated solvents.
Preparation of 10-methyl-5,10-dihydrophenarsazine.The methylsubstituted phenarsazine was prepared by the reaction of methylmagnesium bromide with 10-chloro-5,10dihydrophenarsazine(17) according to the following equation
/CH CH
IN ,AsC1 + CHMgBr -1 HN/ AsC + MgC1Br
C6H4 C6H4
To methylmagnesium bromide (0.2 mole) in diethylether in a 500 ml round-bottom flask equipped with a reflux condenser was added, in small quantities, 10-chloro-5,10dihydrophenarsazine (27.8 g, 0.1 mole). The addition was regulated so that no clogging of the condenser occurred. After all of the 10-chloro-5,10-dihydrophenarsazine was added, the reaction mixture was refluxed for 30 minutes. The reaction mixture was then added to a mixture of ice and sulfuric acid. The diethylether layer, containing the product, was separated from the aqueous layer by using a separatory funnel. The diethylether solution was then filtered and the filtrate was evaporated to give the crude product. The crude material was recrystallized from a boiling ethanol-water solution to give a 95% yield




3400 3200 V1400 1200 1000 800 600 400 cm1
Fig. 1.-Infrared Spectrum of HN(C 6H012 AsCl (Mull).




14
of white crystalline 10-methyl-5,10-dihydrophenarsazine melting at 106-1070C (Lit..(17) mp 106-1070C). The infrared spectrum is shown in Figure 2. The proton magnetic resonance spectrum (Figure 3) was obtained using d 6-DMSO as the solvent and the dimethylsulfoxide peak (-149 cps relative to the sodium salt of 3-trimethylsilylpropanesulfonic acid internal standard) as the internal reference. The approximate *r values and average chemical shifts are given in Table 2. The ratio of the area under
TABLE 2
C6H4
N. M. R. SPECTRAL DATA FOR HNC 6 AsCH C6H4/ 3
Group Chemical Shift, 1
NH 0.77
C6H4 2.77 (ave.)
CH3 8.95
the NH peak to those under the C6H4 peaks and to that under the CH3 peak is 1:8.05:3.04.
Preparation of Chloramine.-Chloramine was prepared
in a generator (Figure 4) by a completely anhydrous method developed by Nattair and Sisler. 18) The gaseous ammonia, nitrogen, and chlorine were metered in rotameters and mixed




a A I I I5
3400 3200 3000 1600 1400 1200 1000 800 600 400 cm
Fig. 2.-Infrared Spectrum of HN(C6H4)2AsCH3 (Mull).




NH C H1 CH3
Fig. 3.-N. M. R. Spectrumof HN(C 6H4)2AsCH3




t Glass Wool Plugs
-~ NH
Fig. 4.-Chlorainine Generator.




18
in a glass reaction tube. The approximate rate of flow of the gases was: NH3, 1.2 mole/hr; N2, 0.3 mble/hr; C12, 0.1 mole/hr. The approximate rate of production of chloramine was 0.1 mole/hr. The reaction proceeded according to the equation
C12(g) + 2NH NH2C1 + NH4C1(s)
A large excess of ammonia was used to prevent formation of NHC12 or NC13 and to reduce the tendency to form nitrogen. The gaseous mixture of chloramine and ammonia was passed through glass wool plugs placed in the reaction tube to filter out the finely divided ammonium chloride produced in the reaction and then bubbled into diethylether as soon as the mixture exited from the generator. This chloramineammonia solution was then used as the reagent for many chloramination reactions.
Because of experimental difficulties encountered in working with an ammonia-chloramine mixture on the vacuum line, all ethereal solutions obtained from the generator were rendered free of ammonia by passing the chloramine solution through a column of anhydrous copper sulfate. The anhydrous copper sulfate not only removed all ammonia, but also aided in the drying of the ethereal solution for later use on the vacuum line.
Preparation of Dimethylchloramine.-Dimethylchloramine
was prepared by a procedure analogous to the Raschig synthe(19)
sis of chloramine,19) in which dimethylamine hydrochloride is reacted with the hypochlorite ion in the form of Clorox.




19
OC1- + (CH3)2NH2C1 (CH3)2NC1 + H2Q + ClThe pure dimethylchloramine (bp 430C) distilled from the reaction mixture was then diluted with diethylether to give the reagent for the appropriate reaction.
The Reaction of Ammonia-free Chloramine with Triethylarsine.(CH3CH2)3As + NH2C1 [(CH3CH2)3AsNH2 Cl
Triethylarsine (5.3 mmole) was pipetted into the reaction flask inside the dry-box and then degassed on the vacuum line. Onto the arsine was condensed 71 ml of a 0.075 M chloramine (ammonia-free) solution in diethylether which had been previously degassed on the vacuum line. The reaction mixture was allowed to warm to room temperature and was vigorously stirred by means of a magnetic stirrer. As soon as the diethylether melted, a white solid formed. The reaction mixture was stirred overnight. All condensible materials were distilled from the reaction mixture. The remaining white solid was taken into the dry-box, washed with hexane, and dried in vacuo. Mp 630C with decomposition to a clear colorless liquid. Anal. Found: C, 33.83; H, 8.18; N, 6.50; As, 35.11; Cl, 16.57. Calculated for [(CH3CH2)3AsNH2]Cl: C, 33.75; H, 8.02; N, 6.56; As, 35.08; Cl, 16.60. The infrared spectrum is shown in Figure 5. Yield of aminotriethylarsonium chloride: 0.8 g (80% of theory based upon the amount of chloramine put into the reaction).




3200 3000 2800 1400 1200 1000 800 600 400 c
Fig. 5.-Infrared SpectruWi of ((CH 3CH2). 3AsNH2]C1 (Null).b




21
Reaction of Ammonia-free Chloramine with Tri-n-propylarsine.(CH3CH2CH2)3As + NH2C1 [(CH3CH2CH2) 3AsNH2]C
Tri-n-propylarsine (7.51 mmole) was pipetted into a reaction flask inside the dry-box. Onto the degassed arsine was condensed 7.65 mmole of a degassed, ammonia-free, ethereal chloramine solution. Vigorous stirring of the reaction mixture was maintained as room temperature was attained. As soon as the diethylether melted, a white solid formed. After several hours, all condensible materials were distilled from the reaction flask. The white solid remaining in the flask was taken into the dry-box and recrystallized from boiling benzene. Upon addition of diethylether to the benzene solution, a large quantity of a white needle-like, crystalline solid was obtained. The recrystallized material was dried in vacuo. Mp 92-940C with apparent decomposition. Anal. Found: C, 42.21; H, 9.29; N, 5.20; As, 29.08; C1, 13.97. Calculated for (CH3CH2CH2) 3AsNH2Cl: C, 42.28; H, 9.07; N, 5.48; As, 29.30; C1, 13.87. The infrared spectrum is shown in Figure 6. Yield of recrystallized aminotri-n-propylarsonium chloride: 0.77 g (40% of theory based upon tri-n-propylarsine used).
Reaction of Dimethylchloramine with Trimethylarsine.(CH3)3As + (CH3)2NCl [(CH3)3AsN(CH3)2]Cl The reaction was carried out on the vacuum line. Onto trimethylarsine (4.8 mmole) was condensed, at -1960C, 36 ml of a 1.02 M dimethylchloramine solution in diethylether.




II I I N I.. I I, 1 I .... I i3200 3000 2800 1400 1200 1000 800 600 400 cm
Fig. 6.-Infrared Spectrum of [(CH3CH2CH2)3AsNH21Cl (Mull).




23
The reaction mixture was then warmed to room temperature. No reaction was noted; immediately after the addition, the mixture was completely liquid. As soon as the vapor pressure above the solution was 5.5 mm/Hg, a white precipitate began to form. The reaction mixture was stirred for two hours as a large amount of white solid formed. When all condensible materials were distilled from the reaction flask, a white, solid residue remained. The white solid was taken into the dry-box, washed with hexane, and dried in vacuo. Mp 184-1860C with decomposition of the solid as indicated from the vigorous bubbling. The remaining solid was light orange-brown in color. The gas evolved during the melting process was basic toward moist litmus. Anal. Found: C, 30.12; H, 7.95; N, 5-35 (combustion nitrogen); As, 38.01; Cl, 18.67. Calculated for [(CH3)3AsN(CH3)2]Cl: C, 30.09; H, 7.58; N, 7.02; As, 37.54; Cl, 17.77. When a combustion nitrogen analysis Iscarried out on an arsenic-nitrogen compound, the results are always lower than expected because of the formation of highly condensed As-N species. The infrared spectrum is shown in Figure 7. The proton magnetic resonance spectrum (Figure 8) was run using CDC13 as the solvent and CHC13 (-436 cps relative to the tetramethylsilane internal standard) as the internal reference. Peak A refers to the (CH3)2N protons and peak B to the (CH3)2As protons. The approximate T values are listed in Table 3.




3000 2800 1400 1200 1000 800 600 4
Fig. 7.-Infrared Spectrum of [(CH 3)3AsN(CH 3)2]C1 (Mull).




25
A B
Fig.- 8.-N. M. R-. Spectrum of [(CH 3)3 AsN(CH 3)2]O1.




26
TABLE 3
N. M. R. SPECTRAL DATA FOR [(CH3)3AsN(CH3)2]Cl
Group Chemical Shift,f
N(CH 3)2 7.37
As(CH3)3 7.68
The ratio of the.area under the (CH 3)2N peak to that under the (CH3)3As peak is 1:1.44. Yield of dimethylaminotrimethylarsonium chloride was 0.44 g (46% of theory based upon the trimethylarsine put into the reaction).
The Reaction of Dimethylchloramine with Triethylarsine.(CH3CH2)3As + (CH 3)2NC1 ~ [(CH3CH2)3AsN(CH 3)21C1 Triethylarsine (5 mmole) was added inside the dry-box to 25 ml of diethylether in a reaction flask equipped with a magnetic stirrer. The reaction flask and contents were degassed on the vacuum line. Onto the arsine was condensed an ether solution of dimethylchloramine (44 mmole). The reaction mixture was then warmed to room temperature. A white precipitate formed as the reaction solution warmed. The reaction was allowed to continue for six hours, during which time a large amount of white solid formed. All condensible materials were then distilled from the reaction flask. The white solid was taken into the dry-box, washed with diethylether several times, and dried in vacuo. The




27
melting point dataare as follows: 107-110 0C, solid darkens and becomes a jelly-like material; 128-1290C, material melts to a clear liquid. Anal. Found: C, 39.54; H, 8.75; N, 6.01; As, 30.83; C1, 14.96. Calculated for [(CH3CH2)3AsN(CH3)2]Cl: C, 39.76; H, 8.76; N, 5.80; As, 31.00; Cl, 14.67. The infrared spectrum is shown in Figure 9. The proton magnetic resonance spectrum (Figure 10) was measured using CDC1 as the solvent and CHC1 (-436 cps relative to the tetramethylsilane internal standard) as the internal reference. Peak A refers to the methylene and dimethylamino protons and peak B to the methyl protons. The approximate r values and average chemical shift values are listed in Table 4. The ratio TABLE 4
N. M. R. SPECTRAL DATA FOR [(CH3CH2) 3AsN(CH3)2]C1
Group Chemical Shift,
CH2 7.30 (ave.)
N(CH3)2 7.34
CH 8.82 (ave.)
of the areas under the CH3 peaks to those under the CH2 and N(CH3)2 peaks is 1:1.3. Yield of dimethylaminotriethylarsonium chloride was 0.48 g (40% of theory based upon the amount of triethylarsine put into the reaction).




3000 2800 14o10 00800 60040
Fig. 9.-Infrared Spectrum of [(CH 3CH)3 AsN(CH 3 )2Cl (Mull).
00




29
..A B
Fig. 10.-N. M. R. Spectrum of [(CH 3CH 2)3 AsN(CH 3)2]C1.




30
The Reaction of Dimethylchloramine with Tri-n-propylarsine.(CH 3CH2CH2) 3As + (CH3)2NCl [(CH3CH2CH2) 3AsN(CH3)2]Cl Inside the dry-box, tri-n-propylarsine (2.0 g, 9.8 mmole) was added to 80 ml of diethylether in a reaction flask. The flask was attached to the vacuum line and the solution was degassed. Onto this solution was condensed 11.1 ml of a 0.88 M dimethylchloramine solution (9.8 mmole) in diethylether. The reaction mixture was allowed to warm to room temperature and react for 48 hours. A white solid slowly began to precipitate. The volatile materials were distilled from the reaction flask, leaving a pale yellow residue. Purification of the product was carried out by dissolving the solid in warm benzene and precipitating the pale yellow solid by adding a small amount of hexane. The pale yellow material was washed with hexane and dried in vacuo. Mp 105-1070C. Anal. Found: C, 46.63; H, 9.69; N, 4.94; As, 26.33; Cl, 12.72. Calculated for [(CH3CH2CH2)3AsN(CH3 Cl: C, 46.54; H, 9.60; N, 4.94; As, 26.42; Cl, 12.50. The infrared spectrum is shown in Figure 11. The proton magnetic resonance spectrum (Figure 12) was measured using CDC13 as the solvent and tetramethylsilane as the internal reference. Peak A refers to the 4-CH2 protons, peak B to the N(CH3)2 protons, peak C to the /3-CH2 protons, and peak D to the CH3 protons. The approximate values and average chemical shifts are given in Table 5.




3000 2800 1400 1200 1000 800 600 400
Fig. 11.-Infrared Spectrum of [(CH 3CH2CH2) 3AsN(CH3)2]C1 (Mull).




B
A C D
Fig. 12.-N. M. R. Spectrum of [(CH C CH2CH2 3AsN(CH 3 2]Cl




33
TABLE 5
N. M. R. SPECTRAL DATA FOR [(CH3CH2CH2) 3AsN(CH3)2]Cl
Group Chemical Shift, T
4-CH2 7.08 (ave.)
N(CH3)2 7.18
3-CH2 8.25 (ave.)
CH3 8.88 (ave.)
The ratio of the areas under the 4-CH2 and N(CH3)2 peaks to that under the /-CH2 peaks was 1.96:1. The ratio of the areas under the 4-CH2 and N(CH3)2 peaks to that under the CH3 peaks was 1.23:1. Yield of dimethylaminotri-n-propylarsonium chloride was 1.76 g or 63% of theory based upon the amount of tri-n-propylarsine put into the reaction.
The Reaction of Ammonia-free Chloramine with 10-chloro-5,10-dihydrophenarsazine.CH 2 CH NH
HN( 6 4AsC1 + NH2Cl [HNCC6H/Asjl] Cl
64 624
Onto 10-chloro-5,10-dihydrophenarsazine (1.93 g, 6.98 mmole) was condensed 93 ml of 0.075 M chloramine (6.98 mmole) in a diethylether solution. The reaction mixture was then warmed to room temperature and was stirred for 4 hours. A yellow-green solid was formed during the reaction. All volatile materials were distilled from the reaction flask. The yellow-green product was washed with diethylether




34
and dried in vacuo. The melting point data are as follows: 2080C, compound turned dark green in color; 212oC, solid became a dark green viscous material; by 2500c, the material is black in color; 2630C, vigorous bubbling occurs. Anal. Found: C, 45.41; H, 4.26; N, 7.25; As, 20.49; C1, 18.04. Calculated for [HN(C6H4)2As(C1)(NH2) ]C.149.5%(CH3CH2 )20: C, 45.91; H, 4.39; N, 7.66; As, 20.49; C1, 19.39. The infrared spectrum is given in Figure 13. Yield of 10-methyl-10-amino-5,10-dihydrophenarsazonium chloride was 2.04 g or 89% of theory based upon the amount of chloramine put into the reaction.
The Reaction of Ammonia-free Chloramine with 10-methyl-5,10-dihydrophenarsazine.C H [HCH 4ACH3
HN 6 4AsCH + NH2C1 HN C6HA CH3]Cl
C6H4 6H NH2
To 10-methyl-5,10-dihydrophenarsazine (1.4 g, 5.4 mmole) dissolved in 50 ml of diethylether was added, with constant stirring, 70 ml of a 0.2 M ammonia-free chloramine solution in diethylether. A white solid formed immediately. The diethylether and excess chloramine were removed by filtering. The white solid was washed several times with diethylether and dried in vacuo. The melting point of the product gave the following results: 1400C, shrinkage; 1500C, white solid turned tan in color; 168-170 0C, solid expanded and turned bright yellow in color; 1760C, melted with decomposition. Anal. Found: C, 51.29; H, 5.28; N, 8.66;




3400 3000 1600 1400 1200. 1000 800 600 400cm
Fig. 13.-Infrared Spectrum of [HN(C6 H4)2 As(Cl)(NH 2)]Cl (Mull).




36
As, 23.65; Cl, 11.20. Calculated for [HN(C64 )2AsCH3(NH2)]Cl: C, 50.27; H, 4.57; N, 9.11; As, 24.27; Cl, 11.48. The infrared spectrum is given in Figure 14. The proton magnetic resonance spectrum (Figure 15) was obtained using d6-DMSO as the solvent with the dimethylsulfoxide peak (-149 cps relative to the sodium salt of 3-(trimethylsilyl)-propanesulfonic acid as internal standard) as the internal reference. The approximate values and average chemical shifts are given in Table 6. Yield of 10-methyl-10-amino-5,10-dihydrophenarsazonium chloride: 90% of theory.
TABLE 6
H As/CH3 C
N. M. R. SPECTRAL DATA FOR HN(C6H4)2As ] Cl
Group Chemical Shift,T
NH -0.80
C6H4 2.93 (ave.)
NH2 3.13
CH 7.97
Discussion
The products of the reactions of the trialkylarsines with chloramine and dimethylchloramine have been found, in all cases, to be the aminoarsonium salts. These amino-




I I- I VA
3200 3000 1600 1400 1200 1000 800 600 400
Fig. 14.-Infrared Spectrum of [IN(C 6H4)2 As(CH 3)(NH 2 )] Cl (Null)._




HI C6 H 4 NH 2 CH 3
Fig. 15.-N. M. R. Spectrum of [HN(C 6H 4)2 As(CH 3)(NH 2 ]C1.




39
arsonium salts are the direct analogs of the products of the corresponding amine and phosphine reactions with chloramine. If a tetracoordinate arsenic is assumed, then the structure [R3AsNR2]C1 would be the most probable. The unshared pair of electrons on the nitrogen atom would allow r bonding with the arsenic in an aminoarsonium salt. Assuming an SN2 reaction mechanism, the Cl group would be a better leaving group than a R2N group. The absence of a new infrared absorption peak in the region expected for an As-C1 stretching vibration to occur, 300 to 450 cm-1, further supports the assumption of the aminoarsonium chloride structure. (The P-Cl stretch occurs in the 440 to 580 cm-1(20) whereas the Sb-C1 stretch occurs below 300 cm 1. ) This absence of the As-Cl stretching band indicates that the chloramination products probably do not contain pentacoordinate arsenic.
In comparing the infrared spectra of the trialkylarsine starting materials with the respective aminoarsonium salt products, the infrared spectra of the products show, in each case, a new peak of medium to strong intensity in the 700 to 850 cm-1 region. This peak is probably due to the As-N stretching mode. This assignment is not unreasonable, since the absorption peak for the P-N stretching mode appears between 750 and 1000 cm-1(20) depending upon the substituents on the nitrogen.




40
The chloraminations of the triethyl- and tri-n-propylarsine proceed instantaneously in diethylether at a low temperature. Sisler and Stratton(8) observed that trimethylarsine reacts with chloramine instantaneously at a low temperature, but triphenylarsine does not react at a measurable rate at -79 C. Assuming that the chloramination reaction proceeds via an SN2 type of bimolecular reaction mechanism with chloramine acting as the electrophile, an increase in electron density on arsenic would increase the reaction rate. Therefore, the triethyl- and tri-n-propylarsine reactions support the postulate that one may correlate the variation in reaction rates with the higher electron density on the arsenic in the alkyl derivatives than in the phenyl derivatives.
The dimethylchloramine reactions do not occur
instantaneously at -79 0C as was found to be the case in the analogous chloramine reactions, but proceed slowly even at room temperature. In fact, 100% conversion of the arsine to the arsonium salt was never obtained, even when the dimethylchloramine was in a 10:1 molar excess. By analogy with chloramine, one might expect these reactions to proceed by an SN2 reaction mechanism with dimethylchloramine acting as the electrophile. The methyl groups on the nitrogen are electron donating, thereby increasing the electron density on the nitrogen atom and consequently decreasing its electrophilic character. Therefore, dimethylchloramine




41
should be a weaker electrophile than chloramine. Neglecting steric factors, its reactions, in comparison with those of chloramine, should proceed more slowly and at higher temperatures. This, indeed, appears to be the case.
The reactions of chloramine with 10-chloro- and
10-methyl-5,10-dihydrophenarsazine presented the opportunity to observe which of the two potential Lewis-base sites nitrogen and arsenic would be attacked by chloramine. The infrared spectra indicate that the chloramine molecule preferentially attacks the arsenic in both cases. As in the aminotrialkylarsonium salts, the infrared spectra of the phenarsazonium salts show the appearance of a new band in the 700 to 850 cm-1 region. This would indicate the existence of an As-N bond in the reaction product. Furthermore the chloramination products show no reducing properties (negative reaction toward iodate). If an hydrazinium salt had been formed by amination of the nitrogen atom, there should be a positive indication of reducing power of the chloramination product. This implies, neglecting any stereochemical arguments, that the arsenic atom is more susceptible to electrophilic attack than is nitrogen. The ability of the arsenic atom to utilize its vacant 4d-orbitals in forming a dir-pir bond with nitrogen may influence the formation of the aminoarsonium salt. Also the electron cloud of the arsenic atom in the 10-position should be more diffuse than that of the nitrogen in the 9-position. The arsenic atom is more readily polarized than the nitrogen




42
and a transition state, [As...NH2... Cl], can exist for a longer time when the chloramine molecule attacks the arsenic atom. The infrared spectrum of the methyl derivative shows no absorption peak for an As-Cl stretching mode. Therefore, pentacoordinate arsenic can probably be ruled out in the reaction product.
Summary
The reactions of chloramine and dimethylchloramine with trialkylarsine have resulted in the formation of heretofore unknown aminoarsonium salts. These results are directly analogous to those of the corresponding amine and phosphine reactions with chloramine. Therefore, the generality of the chloramination reaction has been extended to include the trialkylarsines.
Analogous chloramine and dimethylchloramine reactions tend to indicate that correlations may be made in the variation in reaction rates with the relative electrophilicities of the dimethylchloramine and chloramine molecules. As expected, chloramine appears to be a stronger electrophile than dimethylchloramine.
In the reactions with the 10-chloro- and 10-methyl5,10-dihydrophenarsazines, where there are two potential electron-donor sites for electrophilic attack by chloramine, the chloramine molecule preferentially attacks the arsenic site. This implies, neglecting any stereochemical arguments, that the arsenic atom is more susceptible to electrophilic attack than is the nitrogen atom.




43
Bibliography
1. G. M. Omietanski, A. D. Kelmers, R. W. Shellman,
and H. H. Sisler, J. Am. Chem. Soc., 78, 3874 (1956).
2. H. H. Sisler, A. Sarkis, H. S. Ahuja, R. J. Drago,
and N. L. Smith, J. Am. Chem. Soc., 81, 2982 (1959).
3. W. A. Hart and H. H. Sisler, Inorg. Chem., 3, 617 (1964).
4. D. F. Clemens and H. H. Sisler, ibid., 4, 1222 (1965). 5. S. R. Jain, W. S. Brey, Jr., and H. H. Sisler, ibid.,
6, 515 (1967).
6. S. R. Jain, L. K. Krannich, R. E. Highsmith, and
H. H. Sisler, ibid., 6, 1058 (1967).
7. H. H. Sisler and S. R. Jain, ibid., Z, 104 (1968).
8. H. H. Sisler and C. Stratton, ibid., 5, 2003 (1966).
9. R. L. NcKenney and H. H. Sisler, ibid., 69 1178 (1967). 10. J. W. Cahn and R. E. Powell, J. Am. Chem. Soc., 76,
2565 (1954).
11. R. S. Drago and H. H. Sisler, ibid., 77, 3191 (1955). 12. G. M. Omietanski and H. H. Sisler, ibid., 78, 1211
(1956).
13. D. B. Denney and S. M. Felton, Inorg. Chem., 1, 99
(1968).
14. K. A. Petrov and G. A. Sohol'ski, Zh. Obshch. Khim.,
26, 3377 (1956).
15. W. Stamm and A. Breindel, Angew. Chem., 76, 99 (1964). 16. H. Burton and C. S. Gibson, J. Chem. Soc., 450 (1926). 17. C. S. Gibson and J. D. A. Johnson, J. Chem. Soc.,
2518 (1931).
18. R. Mattair and H. H. Sisler, J. Am. Chem. Soc., 72,
1619 (1951).
19. A. Berg. Ann. Chim. Phys., 31, 319 (1894).




44
20. L. J. Bellamy, The Infrared Spectra of Complex Molecules,
John Wiley & Sons, Inc., New York, 1960, p 324.
21. R. L. McKenney, PhD Dissertation, University of Florida,
December 1965.




CHAPTER III
REACTIONS OF AMMONIA, DEUTERATED AMMONIA, AND
AMMONIA-CHLORAMINE MIXTURES WITH DIMETHYLCHLOROARSINE Historical Background
The reactions of dimethylchloroarsine with ammonia or in the presence of ammonia have not been reported in the literature. Mdritzer has reacted dimethylchloroarsine with dimethylamine to obtain the expected dimethylamino-substituted arsine.) (CH3)2AsCl + 2(CH3)2NH (CH3)2AsN(CH3)2 + (CH3)2NH2C1 The bis(trifluoromethyl)chloroarsine analog has been reacted with ammonia to give the aminoarsine and the condensed bis-arsineamine.(2)
3(CF3 )2AsC1 + 5NH3 (CF3)2AsNH2 + ((CF3 )2As]2NH + 3NH 4Cl
Dimethylchloroarsine has been reacted with methyl iodide to give tetramethylarsonium tri-iodide.(3)
(CH3)2AsCl + 3CH3I .- (CH3)4AsI3 + CH Cl
No successful attempts to use dimethylchloroarsine to prepare dimethyl-substituted arsenonitriles have been reported. Only one arsenonitrilic compound, [(C6Hs )2AsN4, has been prepared. Reichle(4) prepared this diphenylsubstituted compound by reacting diphenylchloroarsine with 45




46
lithium azide and pyrolyzing the immediate reaction product.
(C6H5)2AsC1 + LiN -4 (C6H5 2AsN3 + LiC1
4(C6H5 )2AsN3 ((C6H5)2AsN]4 + 4N2
(5)
Sisler and Stratton,5) on the other hand, have prepared this compound by the simultaneous chloramination and ammonolysis of diphenylchloroarsine.
4(C6H5)2AsCl + 8NH3 + 4NH2C1 -, [(C6H5 )2AsN]4 + 4NH4Cl
This is the only known arsenic analog of the well-known phosphonitrilic compounds.
The phosphonitrilic chlorides, [C12PN]n, have been
known for many years. The earliest known method for their preparation involved the ammonolysis of phosphorus pentachloride.(6) The reaction product is a mixture of cyclic and linear polymers which may be separated by fractional crystallization and distillation. The trimeric and tetrameric phosphonitrilic chlorides can be further reacted to form many halogen, pseudohalogen, alkyl, aryl, alkoxy, aryloxy, N-substituted primary and secondary amine, hydroxy, and other derivatives.
Various phosphonitrilic polymers have been easily
prepared by the reaction of an ammonia-chloramine mixture with di-substituted chlorophosphines or tetra-substituted diphosphines.(7,8,9)
4R2PC1 + 8NH3 + 4NH2Cl -[ R2PN4 + 4NH Cl 2R2PPR2 + 4NH3 + 6NH 2Cl [R2PN]4 + 6NH4Cl




47
Interest in cyclic phosphonitriles arises from the
stability of these compounds and from the speculation that the phosphorus-nitrogen rings have aromatic character. The thermal stability of some of these compounds have made them of special interest in inorganic polymer chemistry. From a theoretical standpoint, structural investigations of the phosphonitrilic polymers are of great interest. Use has been made of infrared and ultraviolet spectroscopy, nuclear magnetic resonance spectroscopy, and electron and X-ray diffraction in these structural studies. Early structural investigations established the fact of i-bonding in the ring on the basis of observed bond lengths. The trimeric phosphonitrilic chloride has a planar, six-membered cyclic structure.
Cl Cl Cl Cl
S\= / P\
N N N N
I II II I
C1--P P-Cl Cl-P P--C1
Cl N Cl Cl N Cl
Theoretical studies have shown that even though the structures are formally analogous to the Kekule' structures of benzene there are very essential differences. In the phosphonitrilics use may be made of both p- and d-orbitals in their-system, whereas in benzene the carbon atoms can utilize only p-orbitals in forming the ir-system. Consequently, the bonding in the P-N system is complex, because of the variety of d-orbitals available and the symmetry differences of the p- and d-orbitals.




48
The first arsenic analog of the phosphonitrilics
prepared by Reichle, Sisler, and Stratton was assumed to have a structure similar to that of the corresponding phosphonitrile. The As=N vibration was assigned to the
-1
most intense infrared absorption band occurring at 943 cm The physical properties and the infrared spectrum indicated that the bonding in the arsenonitrile was similar to that of the phosphonitrile viz., a 2pir-4dir bond superimposed on the sigma bond system.
For all the above reasons we decided to investigate the reactions of dimethylchloroarsine with chloramine, ammonia, and mixtures of the two in order to determine whether or not dimethylarsenonitriles can be prepared.
Experimental
Manipulation of Materials.-The experimental techniques used were the same as those described on pages 6-8.
The material to be chloraminated on the generator
was added to the reaction flask (Figure 16) equipped with a large cold finger, an adapter for connection to the generator, and an adapter for attachment to the vacuum line. The flask was then attached to the generator, the connecting tubes were flushed with dry-nitrogen, the stopcocks on the reaction flask were opened, and the effluent gases from the generator were bubbled through the reaction solution for the desired length of time. The cold finger was always filled with dry-ice to minimize the




49
Fig. 16.-Chloramination Reaction Flask.




50
loss of solvent and reactants through evaporation. The nitrogen passed from the reaction flask through the stopcock and into a continually-flowing nitrogen stream.
Dimethylarsinic acid and magnesium nitride were obtained from K & K Laboratories and was used as obtained. Sodium hypophosphite was obtained from J. T. Baker Company and was used as obtained. Gaseous ammonia and chlorine were obtained in high purity grade from Matheson Scientific, Inc., and used as obtained. Deuterium oxide and hexadeuterodimethylsulfoxide were obtained from Stohler Isotope Chemicals and used after being dried over type 3A molecular sieve.
Analyses.-Elemental analyses were done by Schwarzkopf and Galbraith laboratories.
Infrared Spectra.-Infrared spectra were recorded
on a Beckman IR-10 Spectrometer. A summary of the spectral bands of the materials produced in this study is found in Table 7.
Nuclear Magnetic Resonance Spectra.-The proton
magnetic resonance spectra were measured as described on page 9.
The Preparation of Dimethylchloroarsine.-Dimethylchloroarsine was prepared by the reaction of dimethylarsinic acid with sodium hypophosphite and hydrochloric acid according to the procedure described by Steinkopf and Mieg. (10)




51
TABLE 7(a)
INFRARED ABSORPTION DATA (cm-1 (CH )2AsNH;CI Null 3120(s), 3040-2900(s,bd), 2260(w), 2210(m), 1750(w), 1435(w), 1405(vs), 1255(m), 1230(w), 1135(m), 902(s), 872(s), 842(s), 800(w), 755(w,sh), 732(m), 670(m), 590(s), 545(vs), 315(w), 262(w).
(CH )2AsND+Cl-, Mull 2990(m), 2910(m), 2310(vs), 2250(vs), 2180(s), 1660(w,bd), 1405(m), 1255(s), 1230(vw), 1120(m), 1060(vs), 905(s), 890(s), 860(s), 838(m), 810(m), 732(m), 670(w), 600(m), 590(s), 580(s), 510o(vs).
[(CH3 )2AsN ]42HC1, Mull
2980(s), 2960(s), 2940(s), 2900(s), 2540(m), 1420(w,sh), 1405(m), 1380(w,sh), 1268(m), 1000(s), 900(s), 865(m), 830(m), 800(s), 720(mpd), 665(w), 640(m), 620(m), 600(w), 528(m), 305(m), 295(w,sh).
(a)s, strong; m, medium; w, weak; v, very; sh, shoulder, bd, broad.




52
The Preparation of Deuterated Ammonia.-Deuterated
ammonia was prepared by the reaction of magnesium nitride with deuterium oxide. The deuterated ammonia was passed through a column of magnesium nitride to insure the removal of all water vapor and condensed into a flask on the vacuum line.
The Preparation of Chloramine.-Chloramine was prepared by the procedure described on pages 14-18.(11)
The Reaction of Ammonia with Dimethylchloroarsine.(CH3)2AsC1 + NH -+ (CH )2AsNH+C1323 3 2 3
Approximately 25 ml of ammonia at -790C from a gas cylinder was dried over sodium in a flask on the vacuum line. Dimethylchloroarsine (2 ml, 22 mmole) was added to a reaction flask containing 25 ml of diethylether. Dry ammonia (0.46 ml at -790C, 22 mmole) was distilled into the reaction flask. The reaction mixture was warmed to room temperature. As soon as the ether melted, a white solid formed. After a short time, the volatile materials were distilled from the reaction flask. The white solid product was vacuum sublimed at 600C and 10 mm/Hg. The product did not melt, but sublimed. Anal. Found: C, 15.48; H, 5.98; N, 8.45; As, 47.37; Cl, 22.18. Calculated for (CH3)2AsNH+Cl-: C, 15.25; H, 5.76; N, 8.90; As, 47.58; Cl, 22.51. The infrared spectrum is shown in Figure 17. The proton magnetic resonance spectrum was measured using hexadeuterodimethylsulfoxide as the solvent and the dimethylsulfoxide peak (-149 cps relative to the




2600 2400 1400 1200 1000 800 600 400
Fig. 17.-Infrared Spectrum of (CH3)2AsNH +Cl (Mull).




54
sodium salt of 3-(trimethylsilyl)-propanesulfionic acid as the internal standard) as the internal reference. The spectrum consisted of two single peaks. The 1 values are given in Table 8. The ratio of the area under the (CH3)2As TABLE 8
N. M. R. SPECTRAL DATA FOR (CH3)2AsNH ClGroup Chemical Shift,f
NH3 2.68
(CH3)2As 8.78
peak to that under the NH3 peak is 1.8:1. Yield of dimethylarsinoammonium chloride: 2.67 g (80% of theory based upon the dimethylchloroarsine used).
Reaction of Deuterated Ammonia with Dimethylchloroarsine.(CH3)2AsC1 + ND -s (CH3)2AsND Cl
3 3 2 3
Dry deuterated ammonia from the reaction of deuterium oxide and magnesium nitride was transferred into a calibrated tube on the vacuum line until a volume of 0.46 ml at -790C was obtained (22 mmole). Dimethylchloroarsine (2 ml, 22 mmole) was added to a flask containing 25 ml of diethylether. The reaction flask was attached to the vacuum line and degassed. The deuterated ammonia was distilled into the reaction flask. The reaction flask




55
was warmed to room temperature. When the diethylether melted, a white solid formed immediately. After a short time, the volatile materials were distilled from the reaction flask. The remaining white solid was purified by vacuum sublimation at 600C and 10 mm/Hg. Anal. Found: C, 15.20; N, 8.67; As, 47.15; C1, 21.35. Calculated for (CH3)2AsNDC1 -: C, 14.97; N, 8.73; As, 46.69; Cl, 22.09. The infrared spectrum is shown in Figure 18. The yield of (CH3)2AsND Cl was 2.7 g (80% of theory based upon the dimethylchloroarsine used).
Reaction of An Ammonia-Chloramine Mixture with Dimethylchloroarsine.4(CH3)2AsCl + 6NH3 + 4NH2C1 [(CH3)2AsN]4 2HC1 +
6NH4C1
Dimethylchloroarsine (2.1 g, 15 mmole) was dissolved in 50 ml of 20-400C boiling petroleum ether in a reaction flask (Figure 16). This addition was carried out inside the dry-box. The flask was attached to the chloramine generator. The effluent gases from the generator were bubbled through the solution for 25 minutes (42 mmole of chloramine). A white solid precipitated immediately. After 6 minutes, the white solid appeared to dissolve and to form a dense oil immiscible with petroleum ether. After the chloramination process was completed, the reaction flask was warmed to room temperature, under a stream of dry nitrogen. As the ammonia evolved, the immiscible lower layer slowly turned to a white solid.




3000 2600 2400 1400 1200 1000 800 6004,0'cFig. 18.-Infrared Spectrum of (CH ) AsND +C]7 (mull).
3 2 3




57
The reaction flask was attached to the vacuum line and all condensible materials were distilled from the reaction flask. The remaining tacky, white solid was taken into the dry-box and extracted with acetonitrile. The infrared spectrum of the remaining white solid was identical with that of ammonium chloride.
A white crystalline solid was obtained when the
acetonitrile solution was cooled or added to diethylether. The infrared spectrum of this white solid is shown in Figure 19. Mp 217-2180C with decomposition. A proton magnetic resonance spectrum was obtained using deuterium oxide as the solvent and the sodium salt of 3-(trimethylsilyl)propane sulfonic acid as the internal standard. A single proton resonancewas observed at 8.23T and attributed to the (CH3)2As group. The infrared spectrum, nmr spectrum, and analysis, in addition to the results of the analogous
(7)
dimethylchlorophosphine reaction, suggest that the product is the dihydrochloride of the dimethylarsenonitrile tetramer. The nmr spectrum of this dihydrochloride of the arsenonitrile tetramer should show two different types of methyl protons if there is no rapid exchange of the NH protons. Since the spectrum was obtained using deuterium oxide as the solvent, rapid proton exchange is very possible. This may explain the observance of the single methyl resonance. Anal. Found: C, 17.68; H, 4.70; N, 10.45; As, 54.62; C1, 13.27. Calculated for (CH )2AsN]4*2HC1:




3000 2600 1400 1200 1000 800 60040
Fig. .19.-Infrared Spectrum of [(CH 3)2 AsN]4 e21IC1 (Mull).
00




59
C, 17.49; H, 4.78; N, 10.21; As, 54.60; Cl, 12.92. Yield of the product: 1.53 g of recrystallized product (75% of theory based upon the amount of dimethylchloroarsine used).
Attempts to remove the HC1 attached to the arsenonitrile by extracting the product with triethylamine were unsuccessful. A quantitative recovery of triethylamine and tetramer was always obtained.
Discussion
The results of this study show that dimethylchloroarsine readily reacts with ammonia in a 1:1 mole ratio to form the heretofore unknown arsinoammonium chloride and with an ammonia-chloramine mixture to give the dihydrochloride salt of the tetrameric dimethyl-substituted arsenonitrile. This establishes the first known alkyl-substituted arsenonitrile and opens up a new area of As-N chemistry. The results also extend the generality of the ammoniachloramine reaction with dialkylhalophosphines(7'8) to include the dialkylhaloarsines. It appears now, therefore, that chloramine can be an important synthetic intermediate in the preparation of arsenonitriles.
In contrast to the analogous dimethylchlorophosphine reaction where pyrolysis of the chloramination product, (CH 3)2P(NH2)2]C, must be carried out to obtain the phosphonitrilic tetramer, the arsenonitrilic tetramer is obtained directly in the chloramination process of the




60
dimethylchloroarsines. This suggests that diaminodimethylarsonium chloride is unstable at room temperature in the presence of ammonia and chloramine. The alkyl-substituted arsenonitriles appear to be very stable and much easier to obtain synthetically than the corresponding phosphonitriles. The reaction of other dialkylchloroarsines should be carried out to completely establish this method of preparing these arsenic-nitrogen polymers.
No broad, intense infrared absorption peak was observed at 943 cm-1 to correspond with the As=N assignment made by Reichle(4) in the diphenyl derivative. Broad intense peaks are observed at 1000 cm-1, 900 cm-1, and 800 cm-1. An assignment of one of these bands to the As=N vibration in the dimethyl derivative would be speculation. A shift to lower energy for this vibrational mode would be expected when the substituents on the arsenic are changed from phenyl to methyl. This change would decrease any effective dir-ptr back-bonding existing in the As=N bond due to d-orbital expansion(12), and thereby weaken this bond.
The inability to remove the HC1 from the tetramer indicates that this compound is probably quite basic. The analogous (CH3)2PC1 reaction with ammonia-chloramine does not give the hydrochloride salt. This possibly suggests that there is less di-pir bonding between the arsenic and nitrogen atoms than between the phosphorus and nitrogen atoms. Consequently the f-bonding electrons




61
may be localized more on the nitrogen atoms in the As'-'N bond. A weaker di-pw interaction in the arsenonitriles is expected, since the larger covalent radius of arsenic and the diffuseness of the vacant 4d-orbitals on the arsenic should create a less effective overlap of these orbitals with the 2p-orbital on the nitrogen than in the case of phosphorus.
Summary
This study has demonstrated the use of chloramine
as an important synthetic intermediate in the preparation of the heretofore unknown dialkyl-substituted arsenonitriles. The preparation of this arsenonitrilic tetramer in addition to that previously reported of ((C6H5)2AsN]4 has opened a new area of arsenic-nitrogen chemistry in which there exists many research possibilities.
The existence of the dihydrochloride and the inability to remove this HC1 from the tetramer suggests that the nitrogen atoms in the As-N ring may be more basic and that the dr-pW bonding in the As-N ring may be weaker than in the analogous phosphonitrilic tetramer.
Bibliography
1. K. Midritzer, Ber., 92, 2637 (1959).
2. W. R. Cullen and H. J. Emelius, J. Chem. Soc., 372 (1959).
3. A. E. Goddard, A Textbook of Inorganic Chemistry,
Vol. XI, Charles Griffin and Co., London, England,
1938, p 28.




62
4. W. T. Reichle, Tetrahedron Letters, 51 (1962). 5. H. H. Sisler and C. Stratton, Inorg. Chem., 1,
2003 (1966).
6. H. N. Stokes, Am. Chem. J., 1 275 (1895); 18, 629
(1896); 19, 782 (1897); 20, ~70 (1898).
7. S. E. Frazier, Master's Thesis, University of Florida,
December, 1963.
8. H. H. Sisler and S. E. Frazier, Inorg. Chem., 4,
1204 (1965).
9. S. E. Frazier, Doctoral Dissertation, University of
Florida, December, 1965.
10. W. Steinkopf and W. Mieg, Ber., 52, 1016 (1920). 11. R. Mattair and H. H. Sisler, J. Am. Chem. Soc., 73,
1619 (1952).
12. H. R. Allcock, Chem. & Eng. News, 46, 73 (1968).




CHAPTER IV
REACTIONS OF AMMONIA-FREE CHLORAMINE, DIMETHYLCHLORAMINE, AND CHLORINE WITH ARSINE AND SOME PRIMARY
AND SECONDARY ARSINES Historical Background
The reactions of arsine and the primary and
secondary arsines have been the subject of intensive investigations over the last seventy years. Arsine
(1) t(chord )
has been reacted with the halogens), arsenic trichloride(2)
(3) (3) phosphorus trichloride, and phosphorus pentachloride.3) The reaction with the halogens gives in all cases elemental arsenic and haloarsines.
X2 + AsH3(excess) As + haloarsines
X2(excess) + AsH --* haloarsines X = Cl, Br
The reaction with arsenic trichloride gives elemental arsenic
AsH + AsC1 2As + 3HC1
The reaction of arsine with phosphorus trichloride gives As2H2, while with phosphorus pentachloride it gives elemental arsenic.
AsH + PC15 PC13 + 2HC1 + As + 1H2
3 5 322
When As2H2 is reacted with bromine and iodine, elemental arsenic and arsenic triiodide are obtained,
63




64
respectively.(4)
As2H2 + Br2 -. 2HBr + 2As
As2H2 + 412 2HI + 2AsI
Dehn and co-workers have investigated the reactions
of the halogens with methylarsine to obtain arsenic-halogen containing species.(5) When gaseous methylarsine is passed through a solution of iodine, methyldiiodoarsine is obtained.
CH3AsH2 + 212 -. CH3AsI2 + 2HI
There was evidence of intermediate compounds which were thought to be CH AsHI*HI and CH AsHI. The overall reaction is described by the equations
CH AsH2 12 -. CH AsHI*HI
CH AsHI*HI -4 CH AsHI + HI
CH AsHI + 12 -~ CH AsI 2. HI CH AsI2 + HI If the arsine is in excess, the product is a brown amorphous mass, the composition of which was not elucidated.
Using a solution of bromine in carbondisulfide, arsenic tribromide is obtained.(5)
CH3AsH2 + 3Br2 -. AsBr + 2HBr + CH Br
The haloarsines are obtained probably because the halogen was kept in excess in these reactions.
The reaction of methylarsine with methyliodide yields the tetramethylarsonium iodide.(5)
CH3AsH2 + 3CHI (CH3)4AsI + 2HI




65
The reaction of methylarsine with methylarsenic
oxide gives a good yield of cyclo-pentamethylpentaarsine(6)
10CH3AsH2 + 5CH3As0 4 3(CH 3As)5 + 5H20
which can exist in two forms a yellow liquid and a red solid. The red solid is formed, by heating the yellow form to 100 0C. This red solid then melts at 180 0C to give the yellow liquid again.
The cyclo-pentamethylpentaarsine is commonly prepared by the action of a mild reducing agent such as 50% hypophosphorus acid on sodium methylarsonate.'8'
Dehn and Wilcox have studied the reactions of the halogens, haloarsines, and alkylhalides with dimethylarsine.(10) Chlorine in excess reacts with dimethylarsine to give dimethylchloroarsine and methyldichloroarsine as products.
(CH3)2AsH + C12 (CH )2AsCl + HC1 (CH3)2AsC1 + C12 -. CH3AsC12 + CH 3C1
If the dimethylarsine is maintained in excess, the products are elemental arsenic and a black solid polymer. The reaction is very vigorous.
Bromine in excess reacts with dimethylarsine to give (CH 3)2AsBr*HBr and (CH3)2AsBr in good yields.(10)
(CH3)2AsH + Br2 -; (CH3)2AsBr*HBr
(CH3)2AsH-HBr (CH3)2AsBr + H2
(CH3)2AsH + HBr (CH3)2AsBr + H2




66
Iodine also reacts with dimethylarsine to form only (CH3)2AsI*HI.
(CH 3)2AsH + 12 -. (CH3)2AsI*HI
This compound in the presence of water gives dimethyliodoarsine.
Dehn(10'11) has shown that dimethylarsine reacts with dimethylchloroarsine to give tetramethyldiarsine.
(CH3)2AsH + (CH3)2AsC1 -4 (CH3)2AsAs(CH3)2 + HC1 This reaction has been used to make the unsymmetrical tetraalkyldiarsines when R2AsH and R2AsC1 are used as starting materials.(11)
He has also shown that dimethylarsine will react with arsenic trichloride.(10)
(CH 3)2AsH + AsCl3 (CH3)2AsC1 + (CH3As)x + 2HC1
Dehn found that dimethylarsine reacts with many alkyliodides to give the arsonium iodides.(10)
(CH3)2AsH + 2R'I -4 R'(CH3)2AsI + HI
Tetramethyldiarsine (Cadet's liquid, Cacodyl,
Alkarsin) is one of the oldest known alkylarsenicals and was first prepared by L. C. Cadet de Gassicourt in the 1700's(12) by the heating of a mixture of equal parts by weight of arsenic trioxide and potassium acetate in a glass retort. In addition to its preparation by the reaction of dimethylarsine with dimethylchloroarsine, it may be obtained by the action of dimethylchloroarsine on zinc dust.(13)




67
2(CH3)2AsCl + Zn -+ (CH3)2AsAs(CH3)2 + ZnC12
F. F. Blicke(~14,15,16,17) W. M. D ( n(5,10) and
their co-workers have studied the reactions of phenylarsine with bromine, iodine, and substituted haloarsines. Dehn reacted phenylarsine with excess bromine to obtain arsenic tribromide and bromobenzene as the principle products.(5)
C6 H5AsH2 + Br2 C6H5AsHBr*HBr C6H5AsHBr + Br2 ~ C H 5AsBr2 HBr
C6H5AsBr2 + Br2 -~C H5Br + AsBr3
The intermediate compounds were postulated, but were not specifically shown by analytical results to exist.
Blicke and co-workers have shown that iodine reacts with phenylarsine to form diphenyldiiododiarsine and arsenobenzene, depending upon the reaction conditions.(17)
2C6H5AsH2 + 312 -4 C6H5(I)AsAs(I)C6H5 + 4HI
6C6H5AsH2 + 2 (C6H5As)6 + 12HI
The diiododiarsine may be prepared by the reaction of
(15)
phenyldiiodoarsine with phenylarsine.(15)
3C6H5AsI2 + C6H5AsH2 -4 2C6H5(I)AsAs(I)C H5 + 2HI
Blicke has also shown that phenyldichloroarsine
reacts with diphenylarsine and phenylarsine to give arsenobenzene, diphenylchloroarsine, and tetraphenyldiarsine as products.(15)
12C6 H5AsC12 + 12(C6H5)2AsH 2(C6H5As)6 +
12(C6H5)2AsC1 + 12HC1




68
6c6I15 AsC12 + 12 (C6 H5)2 ASH -~(C6H 5As)6,. +
2(C 6 H5 )2AsAs(C6H 5)2 + 12HCl 3C6 H 5AsCl2 + 3C6 5 AsH 2 -4 (C6IH5 AS)6 + 6HCl
Similarly phenyldilodoarsine reacts with phenylarsine to form arsenobenzene and the diphenyldiiododiarsine of which no chlorine analog is known.(15)
3C6H 5As12 + 3C6H 5AsH2 -4 C6 5A) + 6HI
3C6 H 5AsI 2 + C 6H 5ASH 2 ~2C 6H 5(I)AsAs(I)C 6H 5 + 2HI
The diphenyliiododiarsine is evidently the intermediate in the reaction of iodine with excess phenylarsine, since
(14)
it will react with phenylarsine to give the arsenobenzene.
2C 6H 5 ()AsAs(I)C6 H 5 + 2C 6 H5AsH 2 -4 (C6 H 5As)6 + 4iiI
The diphenyldiiododiarsine will also react with diphenylarsine to give arsenobenzene as the product. (14) 6c6H 5 (I)AsAs(I)C6Hi5 + 6(c6H 5)2 AsH ,(C 6H 5As )6 + 6iiI + 3(C6H 5)2 AsAs(C6H 5)2
Phenylarsine will react with diphenylchloroarsine and diphenyliodoarsine to give as one of the reaction products arsenobenzene. (15)
12(C6H 5)2AsCl + 6c6IH5 AsH2 ~(C 6H 5As)6 + 12HCl 6(c6H 5)2 AsAs(C6 5 ) 2
12(C6H 5)2AsI + 6c6H 5AsHi2 ~(C 6H 5As)6 + 12HI 6(C6H 5)2AsAs(C6H 5)2
When phenylarsine is oxidized very slowly by exposure of an ether solution of the arsine to the atmosphere, some arsenobenzene is obtainedi(5)




69
6C6 H5AsH2 + 302 (C6H5As)6 + 6H20
Palmer and Adams(18) have shown that phenylarsine reacts with aromatic aldehydes in the presence of glacial acetic acid to give good yields of arsenobenzene.
6C6 H5AsH2 + 6RCHO -* (C6H5As)6 + 6RCH20H Arsenobenzene is also readily prepared by the
reaction of phenylarsine with phenylarsenic oxide.(19)
3C6H5AsH2 + 3C6H5AsO (C6 H5As)6 + 6H20
Phenylarsine has been reacted with methyl and ethyl iodides at 1200C to give the respective phenyltrialkylarsonium iodides.(20)
C6H5AsH2 + 3RI [C6H5AsR3 ]I + 2HI
Similar types of reactions as described above have
been carried out using diphenylarsine. When diphenylarsine (10,17)
in excess is reacted with iodine,(117) tetraphenyldiarsine is obtained
2(C6Hs 5)2AsH + 12 -* (C6H5)2AsAs(C6 H5)2 + 2HI
If the iodine is maintained in excess, 17 diphenyliodoarsine is obtained
(C6H5 )2AsH + 12 ~ (C6 H5)2AsI + HI
This indicates that tetraphenyldiarsine is first formed and then reacts with more iodine to give the iodoarsine. The reactions of the tetraaryldiarsines with chlorine to give
(21)
diarylhaloarsines is well known.(21)
(C61 5)2AsAs(C6H5)2 + C12 -~ 2(C6H5)2AsC1




70
The reaction of diphenylarsine with bromine in
excess gives both diphenylbromoarsine and diphenylarsenic tribromide.(10)
(C6H5 )2AsH + Br2 (C6H5 )2AsBr + HBr
(C6H5 )2AsBr + Br2 -0 (C6H5)2AsBr3
Diphenylarsine reacts with many arsenic-halogen compounds to give tetraphenyldiarsine as one of the products.(15,17)
6C6H5AsCl2 + 12(C6H5)2AsH (C6H5As)6 +
6(C6H5 )2AsAs(C6H5 )2 + 12HC1 (C6H5)2AsC1 + (C6H5)2AsH (C6H5)2AsAs(C6H5)2 + HC1
(C6H5)2AsI + (C6H5)2AsH ~ (C6H5)2AsAs(C6H5)2 + HI
(C6H5)3AsC12 + 2(C6H5 )2AsH (C6H5 )2AsAs(C6H5)2 + (C6H5)3As + 2HC1
OH
(C6H5)3As \ + 2(C6H5)2AsH (C6H5)2AsAs(C6H5)2 +
Cl (C6 H5 )3As + HC1 + H20
Blicke and Smith have shown that tetraphenyldiarsine may be obtained from the reaction of diphenylchloroarsine with mercury.(22)
2(C6H5)2AsCl + Hg -. (C6HS )2AsAs(C6H5)2 + HgCl2
The tetraphenyldiarsine is also the principal product from the reaction of diphenylarsine with [(C6 H5)2As] 20 or [(C6 H5)2As]2S.(17)
No mention is made in the literature of the reaction of chloramine or dimethylchloramine with arsine, a primary arsine, or a secondary arsine. Only one reference is made to the reaction of a substituted chloramine with a




71
secondary phosphine, (C H5)2PH.(23) In this case none of the products were elucidated. The authors thought that a mixture of amino- and chlorophosphines was present in the pasty reaction residue. The use of an excess of the substituted chloramine in the reaction caused secondary side reactions to occur which accounted for the inability to isolate pure reaction products.
Experimental
Manipulation of Materials.-The experimental techniques used were the same as those described on pages 6-8.
Triethylamine and acetonitrile were obtained as reagent grade materials and were dried and stored over calcium hydride. Aluminum arsenide, sodium methylarsenate, and dimethylarsinic acid were obtained from K & K Laboratories and were used as obtained. Benzenearsonic acid and triphenylarsine were obtained from Eastman Organic Chemicals and were used as obtained. Benzaldehyde, zinc dust, mercuric chloride, and sodium hypophosphite were obtained from J. T. Baker Company and were used as obtained. Lithium aluminum hydride was purchased from City Chemical Company and used as obtained. Hypophosphorus acid was obtained from Matheson Coleman and Bell and was used as obtained. Gaseous chlorine was obtained as high purity grade from Matheson Scientific, Inc.,and used as obtained. Deuterochloroform and hexadeuterobenzene were obtained from Stohler Isotope Chemicals and used as obtained.




72
Analyses.-Elemental analyses were done by Schwarzkopf and Galbraith laboratories as described on page 8.
Infrared Spectra.-Infrared Spectra were recorded
on a Beckman IR-10 Spectrometer. The spectra were obtained of the neat liquids between KBr plates for the range
2.5-7.5.a and between CsI plates for the range 7.5-20#. The spectra of gases were obtained using a gas cell having KBr windows. A summary of the spectral bands of the materials used and produced in this study is found in Table 9.
Nuclear Magnetic Resonance Spectra.-The proton
magnetic resonance spectra were measured as described on page 9.
Mass Spectra.-The mass spectra were obtained on a Hitachi Perkin-Elmer RMU-6E Mass Spectrometer at an ionizing voltage of 70 ev.
The Preparation of Arsine.-Arsine was prepared in
the vacuum line by the acid-hydrolysis of aluminum arsenide. An aqueous 20% sulfuric acid solution was added dropwise to the solid arsenide. The arsine formed was fractionally distilled and stored on the vacuum line. The infrared spectrum of the gas and the vapor pressure measurements on the liquid were identical with those reported for pure arsine. (24)
The Preparation of Methylarsine.-Methylarsine was prepared by the reduction of sodium methylarsonate using




73
TABLE 9(a)
INFRARED ABSORPTION DATA (cm-1
CH3AsH2, Gas
3020(s), 2960(s), 2938(s), 2860(m), 2850(m), 2840(m), 2090(vs), 1440(m,bd), 920(vs,bd), 675(w,bd), 570(s), 560(s), 548(s).
(CH3)2AsH, Neat
3220(w), 2990(vs), 2950(vs), 2870(vs), 2610(w), 2500(vw), 2430(vw), 2280(w), 2160(vw), 2075(vs), 1990(m), 1450(s,bd), 1436(s), 1395(vs), 1383(vs), 1360(s), 1350(s), 1295(m), 1130(vs), 1980(s,sh), 1070(s), 914(s), 850(m), 837(m), 650(w), 582(s), 570(s), 488(vw), 430(m).
C6 H5AsH2, Neat
3075(m), 3060(m), 3020(m), 3002(m), 2070(s), 1586(m), 1482(s), 1438(s), 1327(w), 1300(vw), 1080(w), 1060(w), 1020(m), 996(m), 956(m), 700(s,bd), 430(m).
(C6H5)2AsH, Neat
3075(s), 3060(s), 3030(m), 3020(m), 3005(m), 2060(s), 1958(w), 1873(w), 1815(w), 1760(w), 1584(w), 1482(s), 1435(s), 1380(w), 1328(m), 1300(m), 1260(w), 1180(w), 1153(w), 1080(m), 1072(m), 1062(m), 1020(s), 996(s), 980(w), 960(w), 900(w), 840(w), 784(s), 725(s), 695(s), 658(m), 610(vw), 465(s), 430(m).
(CH 3As)5, Neat
2975(m), 2905(s), 2800(w), 2450(w), 1408(s), 1365(m,sh), 1236(s), 1090(w,bd), 820(s), 550(s).
(CH3)2AsAs(CH 3)2, Neat
2980(s), 2910(s), 2810(m), 2460(w), 1810(w), 1412(s), 1245(s), 885(s), 820(s), 575(s), 565(s,sh).
(C6H5)2AsAs(C6H5)2, Mull
3062(w), 3045(w), 1575(w), 1476(m), 1428(s), 1302(w), 1180(w), 1150(w), 1072(w), 1068(m), 1016(m), 990(m), 900(w), 730(s,sh), 722(s), 684(s), 460(m), 445(m), 308(m).




74
(C6H5As)6, Mull
3070(w), 1570(w), 1472(m), 1430(m), 1310(w), 1294(w), 1059(w), 1015(w), 991(w), 732(s), 688(m), 460(m).
C3 H9.8As1.4NC11.1, Mull 3100(m,bd), 2972(s), 2905(s), 2820(m,bd), 1450(w), 1405(m), 1315(w), 1270(w), 1090(w), 1000(m,bd), 980(m,bd), 900(s), 805(s), 730(s), 650(w), 640(w), 595(w), 420(m,bd), 350(m).
(a)s, strong; m, medium; w, weak; v, very; sh, shoulder; bd, broad.




75
hydrochloric acid and a zinc-mercury amalgam according to the procedure described by Dehn.(25) The infrared spectrum is given in Figure 20. The proton magnetic resonance spectrum (Figure 21) was obtained using CDC13 as the solvent and tetramethylsilane as the internal standard. Peak A refers to the AsH2 protons and peak B to the methyl protons. The approximate r values and average chemical shifts are given in Table 10.
TABLE 10
N. M. R. SPECTRAL DATA FOR CH AsH2
Group Chemical Shift,
CH3 9.02 (ave.)
AsH2 7.93 (ave.)
The ratio of the area under the AsH2 peak to that under the CH3 peak is 1:1.54.
The Preparation of Dimethylarsine.-Dimethylarsine
was prepared by reducing dimethylarsinic acid in the presence of a zinc/mercury amalgam and hydrochloric acid according to the procedure described by Dehn and Wilcox.(26) The infrared spectrum is shown in Figure 22. The proton magnetic resonance spectrum (Figure 23) was obtained using CDC1 as the solvent and tetramethylsilane as the internal
3




3000 2800 2200 2000 1400 1200 1000 800 600 cm
Fig. 20.-Infrared Spectrum of CH3AsH2 (Gas).
0




j A B Ms
Fig. 21.-N. M. R. Spectrum of CH AsH
3 2*




3000 2800 2000 1600 1400 1200 1000 800 600 cm
Fig. 22.-Infrared Spectrum of (CH 3)2AsH (Neat).
cO




A B TMS
Fig. 23--N. M. R. Spectrum of (CH 3 )2 AsH.




80
standard. Peak A refers to the AsH proton and Peak B to the methyl protons. The approximate 7 values and average chemical shifts are given in Table 11.
TABLE 11
N. M. R. SPECTRAL DATA FOR (CH 3)2AsH
Group Chemical Shift, '
CH 8.98 (ave.)
AsH 7.57 (ave.)
The Preparation of Phenylarsine.-Phenylarsine was prepared by reducing benzenearsonic acid by means of a zinc/mercury amalgam and hydrochloric acid according to the procedure described by Goddard.(20) The infrared spectrum is shown in Figure 24. The proton magnetic resonance spectrum (Figure 25) was obtained using CDC13 as the solvent and tetramethylsilane as the internal standard. Peak A refers to the phenyl protons and peak B to the AsH2 protons. The *r values and average chemical shifts are given in Table 12. The ratio of the areas of the C6H4 peaks to that of the AsH2 peak is 2.51:1.
The Preparation of Diphenylchloroarsine.-Diphenylchloroarsine was prepared by the action of arsenic trichloride on boiling triphenylarsine according to the following




3200 3000 2200 2000 1600 1400 1200 1000 800 60
Fig. 24. Infrared Spectrum of C6 H 5AsH2 (Neat).
65 2 Neat)




A B
Fig. 25--N. M. R. Spectrum of C H 5 AsH 2
00




83
TABLE 12
N. M. R. SPECTRAL DATA FOR C H5AsH2
Group Chemical Shift, T
C6H5 2.70 (ave.)
AsH2 6.44
equations and the procedure described by Goddard.(27)
(C6H5)3As + 2AsC1l 3C6H5AsC12
2(C 6H5 3As + AsC13 3(C6H5)2AsC1
C H5AsCl2 + (C6H5)3As 2(C6H5)AsCl
The Preparation of Diphenylarsine.-Diphenylarsine
was prepared by the reduction of diphenylchloroarsine with lithium aluminum hydride according to the procedure described by Wiberg and M8dritzer.(28) Since a modification was made in the procedure described in this reference, the method of preparation will be described in detail. To a 500 ml, three-necked, round-bottom flask equipped with a mechanical stirrer, a pressurized dropping funnel, a reflux condenser, and a sintered-glass filter funnel attached to the base of the flask (Figure 26) was added 150 ml of diethylether and 8.34 g of lithium aluminum hydride (0.22 mole). The flask was cooled to -760C and 30 g of diphenylchloroarsine (0.11 mole) in 50 ml of diethylether was slowly added to the lithium aluminum hydride solution under a nitrogen stream. A very vigorous reaction occurred. When all the




Fig. 26.-Reaction Flask Having a Sintered Glass
Filter Funnel Base.




85
arsine had been added, the reaction flask was armed to room temperature. In order to decompose the alane-diphenylarsine complex, 25 ml of gaseous oxygen-free water was added to the reaction flask. The reaction mixture was filtered through the sintered-glass filter funnel in order to have minimal contact of the ethereal solution with oxygen. The ethereal solution was dried over calcium chloride and the diethylether distilled from the flask in a nitrogen atmosphere. The liquid remaining in the flask was fractionally distilled at a reduced pressure to give diphenylarsine boiling at 170-1720C at 22 mm/Hg (Lit..(29) bp 1740C at 25 mm/Hg). The infrared spectrum is shown in Figure 27. The proton magnetic resonance spectrum (Figure 28) was obtained using CDC13 as the solvent and tetramethylsilane as the internal standard. Peak A refers to the phenyl protons and peak B to the As-H proton. The approximate r values and average chemical shifts are given in Table 13. The ratio of the area of the As-H peak to those of the C6H5 peaks is 1:10.6.
TABLE 13
N. M. R. SPECTRAL DATA FOR (C6H5)2AsH
Group Chemical Shift, r
C6H5 2.75 (ave.)
AsH 5.12




3000 2000 1600 1400 1200 1000 800 600 cm
Fig. 27.-Infrared Spectrum of (C6H5)2AsH (Neat).




A B
Fig. 28.-N. M. R. Spectrum of (C 6 H )2 AsH.
CO




88
Yield of diphenylarsine: 9.44 g (40% of theory based upon the amount of diphenylchloroarsine used).
The Preparation of Cyclo-pentamethylpentaarsine.Cyclo-pentamethylpentaarsine was prepared by the reaction of sodium methylarsonate and hypophosphorous acid according to the procedure described by Palmer and Scott.(8) The infrared spectrum is shown in Figure 29. The proton magnetic resonance spectrum (Figure 30) of the pure liquid cyclo-pentamethylpentaarsine was obtained at sweep widths of 500 cps, 100 cps, and 50 cps using tetramethylsilane as the internal standard. The spectrum shows three peaks in the approximate area ratios of 2:2:1; corresponding to the three magnetically non-equivalent types of methyl groups present in the compound. The average chemical shift was
8.36T.
A proton magnetic resonance spectrum was obtained using CDC1 as the solvent and tetramethylsilane as the internal standard. The spectrum was identical with that of the neat liquid. The average chemical shift was 8.351.
A proton magnetic resonance spectrum was obtained using CC14 as the solvent and tetramethylsilane as the internal standard. The spectrum was identical with that of the neat liquid. The average chemical shift was 8.35.
A proton magnetic resonance spectrum (Figure 31) was obtained at sweep widths of 500 cps, 100 cps, and 50 cps using C6D6 as the solvent and tetramethylsilane as the internal standard. The spectrum shows three peaks in




34
I I t r y I I I I I I I I I I I I 1
3000 2600 1400 1200 1000 800 600 400 cm-1
Fig. 29.-Infrared Spectrum of (CH3As)5 (Neat).




50 cps 100 cps 500 cps
Fig. 30.-N. M. R. Spectrum of (CH3As)5, Neat and at Various Sweep Widths.




50 ops 100 cps 500 cps
Fig. 31.-N. M. R. Spectrum of (CIIAs)5--C 6D6 Solution
at Various Sweep Widths.




Full Text

PAGE 1

THE REACTIONS OF CHLORAMINE AND DIMETHYLCHLORAMINE WITH SOME ARSINE DERIVATIVES By LARRY KENT KRANNICH A DISSERTATIO N PHESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE HEQUIREMENTS FOP. THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1968

PAGE 2

ACK N OWLEDGMENTS The author wishes to take this opportunity to express his sincere gratitude to his research director, Dr. Harry H. Sisler. Although b~sy with many administrative responsibilities, Dr. Sisler was always available to consult with the author and to provide enc o uragement and valuable suggestions con c erning this research. The manner in which he permitted the author to pursue and design the major portion of the research program was sincerely appreciated. The author takes this opportunity to recognize the many valuable discussions concerning the progress of this research that he participated in with Dr. Kurt Utvary, Charles Watkins, Ronald Highsmith, and other members of his research group. Their interests and su gg estions were greatly appreciated and were of great value. Finally, the author acknowledges the support of this research by the National Institutes of Health. ii

PAGE 3

ACKNOWLDEGMENTS. LIST OF TABLES LIST OF FIGURES. CHAPTER I. I N TRODUCTION Bibliography TABLE OF CONTENTS . . . . . . . II. REACTIONS OF AMMONIA-FREE CHLORAMINE WITH ) Page ii V vii 1 3 TRIALKYLARSINES AND PHENARSAZINES. 4 III. Historical Background Experimen tar Discussion Summary Bibliography REACTIONS OF AMMON IA, DEUTERATED AMMONIA, AND AMMONIA-CHLORAMINE MIXTURES WITH DIMETHYLCHLOROARSINE ......... Historical Background Experimental Discussion Summary Bibliography iii 4 6 36 42 43 45 45 48 59 61 61

PAGE 4

IV. REACTIONS OF AMMONIA-FREE CHLORAMINE, DIMETHYLCHLORAMINE, AND CHLORINE WITH ARSINE Page AND WITH PRIMARY AND SECONDARY ARSINES 63 Historical Background. Experimental . . v. Discussion . . . Summary Bibliography GENERAL CONCLUSIONS AND SUMMARY. BIOGRAPHICAL SKErcH iv . 63 71 132 157 158 161 164

PAGE 5

LI S T OF TABLES Table Page 1. I n f ra red Ab sor p t i o n D a t a (cm1 ) . . 10 2 N M R Spect ral Data for HN ( C 6 H 4 ) 2 A sCH 3 14 3 N M R Spe c t r al D a ta f or [( CH 3 ) 3 A s N ( CH 3 ) 2 ] Cl 26 4. N M. R Spect ral D ata for [( CH 3 CH 2) 3AsN ( CH 3)2] c 1 . . . 27 5. N M R Spectra l Da ta for [( cH3CH2CH2 ) 3AsN ( CH 3)2]c1 . . 33 6 N M R Spect r a l D ata for [ HN ( C 6 H 4) 2As ( CH 3)( NH2 )]c1 . . 36 7. Infr a r e d A b s or p ti o n Data (cm1 ) . . 51 8 N M R. Spect r a l D a t a for ( CH 3)2 AsN H;c1-. 54 9 In fr a r ed Absa r p ti on Dat a (cm1 ) . 73 10. N M R S p ect r a l Da t a for C H 3 A sH 2 . 75 11. N M R Spec tr a l Da t a for ( CH 3 ) 2 AsH . 80 1 2 N M R Spect r al Data f o r c 6 H 5 AsH 2 . 83 1 3. N M R S p ect r a l Data fo r (c 6 H 5 ) 2 A sH 85 14. Ma s s Spe c t r a l Data for ( CH 3 A s) 5 . . 95 1 5. Ma ss Spec tr a l D a t a fo r ( cH 3 ) 2 A s A s(C H 3 ) 2 . 96 16. Ma ss Spectra l Data for ( C 6 H 5 A s) 6 . 100 1 7. Mas s Sp e ct r a l Dat a f or A s H 3 NH 2 C l Re action S ublimate . . . 103 1 8 Mas s Sp e c tra l D at a for Re d S olid at 150c 112 19. M a ss S p e ctr a l Data for R e d S olid at 190c 113 V

PAGE 6

Table Page 20. Mass Spectral Data for (c 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 129 21. 22. 23. 24. 25. Summary of Chloramine and Dimethylchloramine Reactions with Arsines Containing As-H Bonds Relative Intensities of Various AsCl~ Ion Species Chemical Shifts of Hydrogen Nuclei in the As-H Bonds of Various Arsines Summary of Chlorine Reactions with Arsines Containing As-H Bonds Average Chemical Shifts of Methyl Groups in (CH 1 As)~ in Various Solvents, at Room Temperature vi 133 136 139 151 155

PAGE 7

LIST OF FIGURES Figure Page 1. Infrared Spectrum of HN ( C 6 H 4 ) 2 AsCl (Mull) 13 2. Infrared Spectrum of HN(C 6 H 4 ) 2 AsCH 3 (Mull) 15 3. N M. R. Spectrum of H N (c 6 H 4 ) 2 AsCHJ . 16 4. Chloramine Generator . . . 17 5. Infrared Spectrum of ((CH3 CH2)3AsNH2 ]c1 (Mull) . . 20 6. Infrared Spectrum of ((c H 3 cH 2 cH 2 ) 3 AsNH 2 ] Cl (Mull) . . 22 7. Infrared Spectrum of [( CH3)3AsN(CH 3)2] c1 (Mull) . . 24 8. N. M. R. Spectrum of ((CH3)3As N(CH3 )2Jc1 25 9. Infrared Spectrum of ((c H3CH2 )3 AsN(CH3 )2] Cl (Mull) . . 28 10. N. M. R. Spectrum of ((CH3C H2 )3 AsN(CH 3)2]c1 . . . 29 11. Infrared Spectrum of ((CH3C H2CH2 )3As N(CH3 )2]c1 ( Mull ) . 31 12. N. M. R. Spectrum of ((CH3CH2CH2)3 AsN(CH3 )2]c1 . . 32 13. Infrared Spectrum of [ HN(C 6 H 4 ) 2 As(Cl) ( NH 2 )] Cl (Mull) . . 35 14. Infrared Spectrum of [H N(C 6 H 4 ) 2 As(CH 3 ) ( NH 2 )] Cl (Mull) . 37 15. N. M. R. Spectrum of [H N (C 6 H 4 ) 2 As(C H 3 ) ( NH 2 )] Cl . . 38 16. Chloramination Reaction Flask . . 49 vii

PAGE 8

Figure Page 17. Infrared Sec~rum of (CH 3 ) 2 As N H 3 c1 (Mull) . . . 53 18. Infrared S]j2ec~rum of (cH 3 ) 2 AsND 3 c1 (Mull) . . . 56 19. Infrared SpectrQm of ((cH 3 ) 2 As N ] 4 HCl (Mull) . . 58 20. Infrared Spectrum of CH 3 AsH 2 (gas) . 76 21. N. M. R. Spectrum of CH 3 AsH 2 . 77 22. Infrared Spectrum of (CH 3 ) 2 AsH (Neat) . 78 23. N. M. R. Spectrum of (cH 3 ) 2 AsH . . 79 24. Infrared Spectrum of c 6 H 5 AsH 2 (Neat) . 81 25. N. M. R. Spectrum of c 6 H 5 AsH 2 . . 82 26. Reaction Flask Having a Sintered Glass Filter Funnel Base . . . 84 27. Infrared Spectrum of (C 6 H 5 ) 2 AsH (Neat) 86 28. N. M. R. Spectrum of (c 6 H 5 ) 2 AsH . . 87 29. Infrared Spectrum of (cH 3 As) 5 (Neat) 89 30. N. M. R. Spectrum of (CH 3 As) 5 Neat and at Various Sweep Widths . . . 90 31. N. M. R. Spectrum of (cH 3 As) 5 -c 6 D 6 Solution at Various Sweep Widths 91 32. Temperature Dependence of Chemical Shifts of Methyl Groups in (CH 3 As) 5 (Neat) . 93 33. Temperature Dependence of Chemical Shifts of Methyl Groups in (CH 3 As) 5 (c 6 D 6 Solution) 94 34. Infrared Spectrum of (CH 3 ) 2 AsAs(cH 3 ) 2 (Neat) 97 35. N. M. R. Spectrum of (cH 3 ) 2 AsAs(CH 3 ) 2 . 98 viii

PAGE 9

Fi g ure P a ge 3 6 I n fr a red Spectru m of ( c 6 H 5 As) 6 . 101 37. N. M R S p ec tr um o f Y ello w Re sid u e from A s H 3 c 1 2 Reac tion . . . 106 38 N M R Spe c tr~m o f CH~A s H 2 c 1 2 Reac tion Pro d u c t P ri o r to Conden at ion . 111 3 9 I nf r a red Spect r u m o f A sN P r o duct from ( cH 3 ) 2 AsH NH 2 C l Reac t i on . . 117 4 0. N M R Spec trum o f As N P roduct from ( cH 3 ) 2 A s H NH 2 C l Reac tion . . 118 41. I nf r ared Spe c trum of ( c 6 H 5 ) 2 AsA s( c 6 H 5 ) 2 ( Mu ll). 127 4 2 N M R Spe ctru m o f (c 6 H 5 ) 2 AsA s (c 6 H 5 ) 2 . 128 43 Burns and Wa s er s Stru c tu re of ( CH 3 As ) 5 from X ray Stud ies . . . . 154 4 4. O the r P o ssibl e Struct ure of ( CH 3 A s) 5 . 154 ix

PAGE 10

CHAPTER I INTRODUCTION During the past 10 years, the chloraminations of amines, phosphines, and arylarsines have been extensively investigated in our laboratories. Increased interest in the chloramination of the arsines resulted when the diphenyl arsenonitrilic tetramer was obtained from the pyrolysis of the chloramination and amination products of diphenyl chloroarsine. This demonstrated the possibility ~f obtaining previously unknown arsenonitrilic polymers by a reaction analogous to one of those used for obtaining phosphonitrilics. (l) R 2 AsCl + NH 2 Cl 2 [R 2 As (NH 2 )' 2 ] Cl + 2NHJ -+ [R 2 As (NH 2 ) 2 ) Cl + NH 4 Cl [R As~N~AsR ]c1 + NH Cl 2 NH NH 2 4 2 2 2 [R 2 As~ N ~AsR 2 ] Cl -+ [R 2 AsN) 4 + 2NH 4 c1 NH 2 NH 2 In addition, many interesting arsenic-nitrogen intermediates were possible. The chloramination of trialkyl arsines, other than trimethyl-, had not been carried out, and thus the generality of the chloramination reaction with substituted arsines had not been demonstrated. RJAs + NH 2 Cl [R 3 AsNH 2 ] Cl No chloramination reactions had been reported where 1

PAGE 11

dimethychloramine was used as the chloraminati~g agent. The preparation and investigation of the infr a red and proton magnetic resonance spectra of a variety of aminoarsonium salts could give information concerning the relative extent of d~-p~ bonding in the As-N bond. The existence of 10-chloroand 10-methyl-5,10dihydrophen a rsazine, where there are two potential sites nitrogen and arsenic for electrophilic attack by chloramine, presented the opportunity to study selective chloramination. In these cases, no difficulties could be encountered with transamination reactions, since the nitrogen is not bonded to the arsenic. The reaction of phosphine with dimethylchloramine, carried out by Ronald Highsmith, to give the unexpected reaction + -t 2P + indicated a new type of reaction for dimethylchloramine and possibly for chlorarnine. We were interested in the possibility of obtaining a new synthetic method for the preparation of compounds containing bonds between two atoms of a Group Va element from the reaction of chloramine and dimethylchloramine with primary and secondary phosphines and arsines. Since the As-H bond should be more labile than the P-H bond, we thought that there would be a greater ease of obtaining As-As bonded compounds by this reaction than of P-P bonded compounds. 2

PAGE 12

Therefore, the purpose of this study wa~ (1) to expand the present knowledge of the reaction of chloramine with ~roup V bases to include the substituted arsines; (2) to investigate the properties of dimethylchloramine rela~ tive to those of chlora.mine; (3) to study the chloramination of the phenarsazines to determine the site of the attack of chloramine; (4) to prepare some heretofore unknown dialkyl arsenonitriles by the reaction of an ammonia-chloramine mixture with dialkylhaloarsines; and (5) to investigate the use of dimethylchloramine and chloramine in synthesizing diarsines and cyclic polyarsines from reactions with primary and secondary arsines. Bibliography 1. S. E. Frazier, Master's Thesis, University of Florida, December, 1963. 3

PAGE 13

CHAPTER II REACTIONS OF AMMONIA-FREE CHLORAMINE WITH TRIALKYLARSINES AND PHENARSAZINES Historical Background The original interest of our laboratory in chloramine chemistry had its beginning in the early 195O's when much attention was being given to hydrazine as a high energy rocket propellant. The reaction of chloramine and liquid ammonia gave as one of its products hydrazine. NH 2 Cl + 2NHJ H 2 NNH 2 + NH 4 Cl The reaction immediately created interest, because of the potential it held for easy routes to the synthesis of hydrazine and substituted hydrazines in good yields. Sisler, Omietanski, and others(l) showed that the chloramina tion of primary and secondary amines gives good yields of the corresponding hydrazines. Chloramination of tertiary amines gave the corresponding hydrazinium salts, since the absence of an hydrogen atom attached to the quaternary nitrogen prevented any protolysis reaction. The scope of chloramine reactions was eventually expanded by reacting chloramine and dimethylchloramine with tertiary phosphines, aminophosphines, and hydrazino phosphines. ( 2 ,J, 4 ,5, 6 ,?) The generalized reaction may be written 4

PAGE 14

RR'R"P + NH 2 Cl RR'R"P + (CH 3 ) 2 NC1 -+ [RR' R"PNH 2 ] Cl -+ {RR'R"PN(CH 3 ))c1 R = alkyl, aryl, amino, and/or hydrazino groups In all cases the chloramination occurs at the most basic site the phosphorus atom yielding phosphonium salts. The only exception to this is the reaction of dimethyl chloramine and tris-dimethylaminophosphine to give the tris-dimethylaminochlorophosphonium chloride.( 6 ) One method of rationalizing the attack on the phosphorus atom is to assume that nitrogen to phosphorus prr-d~ back-bonding increases the Lewis basicity of the phosphorus atom and decreases that of the nitrogen. The reactions of chloramine with triphenylarsine and trimethylarsine have been shown ( 8 ) to give good yields of the respective aminoarsonium chlorides. Recently work carried out in this laboratory has shown that trialkyl and triarylstibines react with ammonia-free chloramine and an ammonia-chloramine mixture to produce compounds of the type lR 3 Sb(C1)] 2 NH. ( 9 ) The reactions of various substituted amines with chloramine to give hydrazines and hydrazinium salts h~ve been demon~trated to proceed by a SN 2 -type reaction mechanism with the chloramine molecules acting as the (10 11 12) electroph1les. The assumption has been made that these electrophiles react with other substituted Group V nucleophiles by the same SN 2 reaction mechanism 5

PAGE 15

to give products containing a N-N, N-P or N -As bond. The only previously found exceptions to this occur in the reactions of alkylchloramine with tris(dialkylamino)phos phines(6,l3) and dialkylphosphonates. (l 4 ) Experimental Manipulation of Materials.-Because of the high reactivity of the arsines and arsonium salts toward oxygen and water and the volatility of the arsines, many of the reactions were carried out in an all glass, high vacuum line. All experimental purification work was conducted in a nitrogen atmosphere in a Vacuum Atmospheres Model HE-43 Dri-Lab equipped with a Model HE-93 Dri-Train. All of the arsenic-containing starting materials were stored and transferred inside the Dri-Lab. The technique used for admitting anhydrous, ammonia free chloramine into the vacuum line was as follows. A 250 ml two-necked, round-bottom flask equipped with a vacuum line adapter and a filter funnel was connected to a take-off on the vacuum line. The filter funnel had a vacuum-back stopcock and a fritted glass filter. The entire flask, up to the vacuum-back stopcock, was evacuated and flamed with a torch to insure removal of all entrapped oxygen and moisture. Powdered anhydrous copper sulfate was added to the filter funnel to give a two inch pad. The ammonia-chloramine solution, which had been stored over a considerable quantity of anhydrous copper 6

PAGE 16

sulfate, w a s pipetted (under a nitrogen str ea m) onto the copper sulfate pad in the funnel. The r ou nd-bottom flask was then immersed in a Dewar of liq u id nitrogen. The stopcock to the vacuum manifold wa s closed, the stopcock to the filter funnel was then ope n ed, and the chloramine-ether solution was admitted into th e flask at a temperature of -196c. After almost all the solution had passed from the filter funnel, the s topcock was closed to prevent the last few milliliters of ether solution and any oxygen or moisture from entering the round-bottom flask. The ether solution was warmed to room temperature, frozen out at -196c, and degassed. This degassing procedure was repeated several times. The chloramine-ether solution was then distilled through the vacuum manifold into a flask having volume graduation marks and a ser u m-capped sidearm. The solution was warmed to room temperature, helium was admitted into the manifold and calibrated flask until atmospheric pressure was attained, and a sample of the solution was removed through the serum cap using an hypodermic syringe. The concentration of the chloramine solution was then determined by reacting sample aliquots with acidified potas s ium iodide solution and titrating the liberated iodine with standard sodium thiosulfate solution. The dimethylchloramine solutions were transferred into a calibrated tube inside a nitrogen filled dry-bag. This tube and its contents were attached to the vacuum line and degassed several times before distilling the dimethyl7

PAGE 17

chloramine solution into the reaction flask .. The determination of the concentration of dimethylchloramine solution was carried out in a manner analogous to that for chloramine solutions. Benzene, petroleum ether, hexane, and diethylether were all obtained as reagent grade materials and were dried and stored over calcium hydride. Trimethylaluminum, triethylaluminurn, and tri-n-propylaluminum were obtained from Ethyl Corporation and were used as obtained. Arsenic trichloride, arsenic trioxide, and diphenylamine were obtained from J. T. Baker Chemical Company as reagent grade chemicals and were used as obtained. Methyl magnesium bromide was purchased from Pennisular Chemical Research and used as obtained. 8 Analyses.-Elemental analyses were done by Galbraith Microanalytical Laboratories, Knoxville, Tennessee, and Schwarzkopf Microanalytical Laboratory, Woodside, New York. Nitrogen analyses were carried out using the Kjeldahl method. Melting points were determined using the Thomas Hoover Capillary Melting Point Apparatus with an uncorrected thermometer. Infrared Spectra.-Infrared spectra were recorded on a Beckman IR-10 Spectrometer. The spectra were obtained on the solids in the form of Kel-F mulls using KBr plates for the range 2.5-7-5)' and as Nujol mulls using CsI plates for the range 7.5-20p. A summary of the spectral bands not listed in the literature of materials used and

PAGE 18

produced in this study is found in Table 1. N uclear Ma gnetic Resonance Spectra.-The proton magnetic resonance spectra were measured on a Varian Model A-60-A nmr spectrometer. Tetramethylsilane, chloro form, and dimethylsulfoxide were used as internal standards in the respective instances. Pre p aration of Trialkylarsi n es.-Trimethylarsine, triethylarsine, and tri-n-propylarsine w ere prepared by reacting arsenic trioxide with the respective trialkyl aluminum compound according to the procedure described by Stamm and Breindel(l5) and by the equations ..... + + 2RA10 JRAl0 Preparation of 10-chloro-5,10-dihydrophenarsazine.10-chloro-5,10-dihydrophenarsazine was prepared by reacting arsenic trichloride with diphenylamine(l 6 ) according to the following equation + C H -+ HN( 6 ~AsCl C6H4 A mixture of diphenyl a mine (50.6 g, 0.J + 2HC1 mole), arsenic 9 trichloride (54.1 g, 0.J mole), and o-dichlorobenzene (100 ml) was boiled under reflux for 15 hours in a 250 ml round bottom flask. As the reaction flask cooled to room temperature, large quantities of material crystallized. The crystals were washed with petroleum ether and recrys tallized from boiling carbon tetrachloride. The bright yellow, crystalline product was dried in vacuo.

PAGE 19

TABLE l(a) (cm -1) INFRARED ABSORPTION DATA HN(C 6 H 4 ) 2 AsCl, Mull 3340(vs), J040(w), 1900(w), 1601(vs), 1580(w), 157o(vs), 1510(m), 1480(s), 1465(vs), 14J8(s), 1J60(w), 1J25(w), 1278(w), 1255(w), 12J2(m), 1160(m), 11JO(m), 1065(s), 1020(w), 895(w), 445(m), 750(vs), 740(vs), 715(s), 590(s), 440(s), 400(m). HN(C 6 H 4 ) 2 AscH 3 Mull 3380(vs), J050(w), 2980(w), 2910(w), 1900(w), 1590(s), 1570(s), 1500(w), 1480(m), 1460(vs), 1440(s), 1410(m), 1J22(s), 1280(w), 1255(s), 12JO(m), 1160(m), 11JO(w), 11 2 0(w), 1060(w), 10JO(w), 970(w), 9JO(w), 88 0(m), 860(w), 845 (s), 8J5 (m), 760(vs), 745(s), 722(vs), 670(w), 650(m), 595(w), 550(s), 535(s), 450(s). ~CH3CH2)3As NH2 ]c1, Mull J140(vs), 2920(vs), 2870(vs), 2800(vs), 1485(w), 1440(s), 1410(m), 1J80(m), 1260(s), 1240(w), 1090(s), 1040(s), 1020(s), 970(m), SOO(m), 770(w), 750(m), 7JO(s), 690(s), 605(m), 545(m). ((CH3CH2CH2)3AsNH2]c1, Mull 319o(vs), J180(vs), J140(vs), JlOO(vs), J090(vs), J090(vs), 2950(vs), 29JO(vs), 2870(vs), 1450(s), 1410(m), 1J50(m), 1JOO(w), 1214(m), 1080(s), 10-SO(m), 840(m), 720(s), 695(s). [( CHJ) 3 As N ( CHJ) 2 ] Cl, Mull 2990(vs), 2900(m), 2860(m), 2800(m), 1460(vw), 1452(w) ,10 1448 ( w) 14 2 O ( w ) 1312 ( w) 13 O 5 ( w) 12 7 3 ( m ) 12 5 O ( w) 11 6 0 ( s ) 1105(m), 1055(s), 942(vs), 915(vs), 86 0(m), 795(w), 712(vw), 65o(w), 642(m), 610(m), 568(vs), J54(vw), J20(m). [( CHJ CH 2 ) 3 As N ( CHJ) 2 ] Cl, Mull 2960(m), 29JO(vs), 2805(m), 1460(s), 1J82(w), 1J12(w), 1250(m), 1150(s), 1040(m), 9JO(vs), 825(w), 794(w), 757(s), 605(m), 596(m), 360(w).

PAGE 20

[(cH3CH2CH2)3AsN(CH3)2]c1, Mull 296o(vs), 292o(vs), 2870(s), 2s1o(m), 1450(m), 1410(w), 137o(w), 131o(w), 1250(w), 1220(w), 1210(w), 1160(m), 1060(s), 1037(w), 940(vs), 838(w), 790(vs), 723(m), 645(vw), 595(vw), 345(vw). [ HN(C 6 H1>As~Cl ] Cl, Mull c 6 H 4 NH 2 323o(m), 314o(s), 304o(s), J000(s), 2830(m), 1607(s), 11 1582(m), 1565(w), 1510(w), 1460(vs), 1439(s), 1405(s), 1362(w), 1322(2), 1275(w), 1245(m), 1240(sh), 1230(m), 1155(w), 1138(w), 1065(m), 1024(m), 967(m), 922(s), 8 90 (w), 847 (m ), 830(w), 800(w), 742(vs), 720(m), 670(s), 640(w), 590(w), 570(w), 530(w), 440(s), 388(w), 340(s). [ C H CH l HN/ 6 4 'As/ 3 Cl Mull / c 6 H 4 NH 2 324o(s), J150(vs), J0S0(vs), 2980(vs), 293o(vs), 2850(s), 1610(vs), 1585(vs), 1575(s), 1520(s), 1470(vs), 1455(s), 1445(s), 1410(m), 1J60(m), 1245(m), 1165(m), 1140(m), 1110(w), 1075(m), 10J0(m), 860(s), 752(vs), 720(m), 69o(w), 645(w), 620(w), 597(w), 535(w), 450(m), 390(w), J40(w). (a)s, strong; m, medium; w, weak; v, very; sh, shoulder.

PAGE 21

Mp 189-189.5c (Lit.~ 16 ) mp 191-192C). Tpe yield of 10-chloro-5,10-dihydrophenarsazine was 52 g or approximately 50% of theory based upon the amount of arsenic trichloride put into the reaction. The infrared spectrum is shown in Figure 1. No nmr spectrum was obtained because of the low solubility of the product in the common deuterated solvents. Preparation of 10-methyl-5,10-dihydrophenarsazine. The methylsubstituted phenarsazine was prepared by the reaction of methylmagnesium bromide with 10-chloro-5,10dihydrophenarsazine(l?) according to the follo w ing equation C H HN/ 6 4 'AsCl 'c H / 6 4 + + MgClBr To methylmagnesium bromide (0.2 mole) in diethylether in a 500 ml round-bottom flask equipped with a reflux cond enser was added, in small quantities, 10-chloro-5,10dihydrophenarsazine (27.8 g, 0.1 mole). The addition was regulated so that no clogging of the condenser occurred. After all of the 10-chloro-5,10-dihydrophenarsazine was added, the reaction mixture was refluxed for JO minutes. The reaction mixture was then added to a mixture of ice artd sulfuric acid. The diethylether layer, containing the product, was separated from the aqueous layer by using a separatory funnel. The diethylether solution was then filtered and the filtrate was evaporated to give the crude product. The crude material was recrystallized from a boiling ethanol-water solution to give a 95% yield 12

PAGE 22

J400 3200 1400 1200 1000 800 Fig. 1.-Infrared Spectrum of HN (C 6 H 4 ) 2 AsCl (Mull). 600 400 -1 cm

PAGE 23

of white crystalline 10-methyl-5,10-dihydrophenarsazine melting at 106-107c (Lit.! 17 ) mp 106-107c). The infrared spectrum is shown in Figure 2. The proton m a gnetic resonance spectrum (Figure J) was obtained using d 6 -DMSO as the solvent and the dimethylsulfoxide peak (-149 cps relative to the sodium salt of J-trimethylsilyl propanesulfonic acid internal standard) as the internal reference. The approximate~ values and average chemical shifts are given in Table 2. The ratio of the area under TABLE 2 C H N M. R. SPECTRAL DATA FOR H N ( 6 4 ;AsCHJ C6H4 Group NH Chemical Shift,~ 0.77 2.77 (ave.) 8.95 the NH peak to those under the c 6 H 4 pe a ks and to that under the CHJ peak is 1:8.05:J.04. Preparation of Chloramine.-Chloramine was prepared in a generator (Figure 4) by a completely anhydrous method developed by Mattair and Sisler~lS) The gaseous ammonia, nitrogen, and chlorine were metered in rotameters and mixed 14

PAGE 24

\ 3400 3200 3000 1600 1400 1200 1000 800 600 400 cm1

PAGE 26

. Glass Wool Plu g s Fig. 4.-Chloramine Generator.

PAGE 27

in a glass reaction tube. The approximate rate of flow of the gases was: NH 3 1.2 m~le/hr; N 2 0.J mble/hr; c1 2 0.1 mole/hr. The approximate rate of production of chloramine was 0.1 mole/hr. The reaction proceeded according to the equation + + A large excess of ammonia was used to prevent formation of NHC1 2 or Nc1 3 and to reduce the tendency to form nitrogen. The gaseous mixture of chloramine and ammonia was passed through glass wool plugs placed in the reaction tube to filter out the finely divided ammonium chloride produced in the reaction and then bubbled into diethylether as soon as the mixture exited from the generator. This chloramine ammonia solution was then used as the reagent for many chloramination reactions. Because of experimental difficulties encountered in working with an ammonia-chloramine mixture on the vacuum line, all ethereal solutions obtained from the generator were rendered free of ammonia by passing the chloramine solution through a column of anhydrous copper sulfate. The anhydrous copper sulfate not only removed all ammonia, but also aided in the drying of the ethereal solution for later use on the vacuum line. 18 Preparation of Dirnethylchloramine.-Dimethylchloramine was prepared by a procedure analogous to the Raschig synthe sis of chloramine; 19 ) in which dimethylamine hydrochloride is reacted with the hypochlorite ion in the form of Clorox.

PAGE 28

OCl + (CH 3 ) 2 NH 2 Cl --t (CH 3 ) 2 NC1 + H 2 Q + Cl The pure dimethylchloramine (bp 43c) distilled from the reaction mixture was then diluted with diethylether to give the reagent for the appropriate reaction. The Reaction of Ammonia-free Chloramine with Triethylarsine.+ Triethylarsine (5.3 mmole) was pipetted into the reaction flask inside the dry-box and then degassed on the vacuum line. Onto the a rsine was condensed 71 ml of a 0.075 M chloramine (ammo~ia-free) solution in diethylether which had been previously degassed on the vacuum line~ The reaction mixture was allowed to warm to room temperature and was vigorously stirred by means of a magnetic stirrer. As soon as the diethylether melted, a white solid formed. The reaction mixture was stirred overnight. All condensible materials were distilled from the reaction mixture. The remaining white solid was taken into the dry-box, washed with hexane, and dried in vacuo. Mp 63c with decomposition to a clear colorless liquid. Anal. Found: C, 33.83; H, 8.18; N, 6.50; As, 35.11; Cl, 16.57. Calculated for ~CH3CH2)3AsNH2]c1: c, 33-75; H, 8.02; N, 6.56; As, 35.08; Cl, 16.60. The infrared spectrum is shown in Figure 5. Yield of aminotriethylarsonium chloride: 0.8 g (80% of theory based upon the amount of chloramine put into the reaction). 19

PAGE 29

J200 JOOO 2800 Fig. 1400 1200 1000 800 600 5.-Infrared Spectrum of [(cH 3 cH 2 ) 3 AsNH 2 )c1 (Mull). 400 -1 cm N 0

PAGE 30

21 Re a ction of Am mo n ia-free Chlor a rnine with T ri-n-propylarsine.(CH3CH2CH2)3As + NH2Cl [(CH3CH2CH2)3AsNH2]c1 Tri-n-propylarsine (7.51 mmole) was pipetted into a reaction flask inside the dry-box. Onto the degassed arsine was condensed 7.65 mmole of a de g assed, ammonia-free, ethereal chloramine solution. Vigorous stirring of the reaction mixture was maintained as room temperature was attained. As soon as the diethylether melted, a white solid formed. After several hours, all condensible materials were distilled from the reaction flask. The white solid remaining in the flask was taken into the dry-box and recrystallized from boiling benzene. Upon addition of diethylether to the benzene solution, a large quantity of a white needle-like, crystalline solid was obtained. The recrystallized material was dried in v a cuo. Mp 92-94c with apparent decomposition. Anal. Found: C, 42.21; H, 9.29; N, 5.20; As, 29.08; Cl, 13.97. Calculated for ~CHJC H 2 CH 2 ) 3 As N H 2 )c1: C, 42.28; H, 9.07; N, 5.48; As, 29.30; Cl, 13~87. The infrared spectrum is shown in Figure 6. Yield of recrystallized aminotri-n-propylarsonium chloride: 0.77 g (40 % of theory based upon tri-n-propylarsine used). Reaction of Dimethylchlor am ine w ith Trimethylarsine.+ The reaction was carried out on the vacuum line. Onto trimethylarsine (4.8 mmole) was condensed, at -19 6 c, 36 ml of a 1.02 M dimethylchloramine solution in diethylether.

PAGE 31

3200 3000 2800 1400 1200 1000 800 600 400 Fig. 6.-Infrared Spectrum of ((cH 3 CH 2 CH 2 ) 3 AsNH 2 ]c1 (Mull). -1 cm N N

PAGE 32

The reaction mixture was then warmed to room ~emperature. No reaction was noted; immediately after the addition, the mixture was completely liquid. As soon as the vapor pressure above the solution was 5.5 mm/Hg, a white precipitate began to form. The reaction mixture was stirred for two hours as a large amount of white solid formed. When all condensible materials were distilled from the reaction flask, a white, solid residue remained. The white solid was taken into the dry-box, washed with hexane, and dried in vacuo. Mp 184-186c with decomposition of the solid as indicated from the vigorous bubbling. The remaining solid was light orange-brown in color. The gas evolved during the melting process was basic toward moist litmus. Anal. Found: C, 30.12; H, 7-95; N, 5.35 (combustion nitrogen); As, 38.0l; Cl, 18.67. Calculated for [(CH3)3AsN(CH3)2]c1: c, 30.09; H, 7.58; N, 7.02; As, 37.54; Cl, 17.77. When a combustion nitrogen analysis IB carried out on an arsenic-nitrogen compound, the results are always lower than expected because of the formation of highly condensed As-N species. The infrared spectrum is shown in Figure 7. The proton magnetic resonance spectrum (Figure 8) was run using CDc1 3 as the solvent and CHc1 3 (-436 cps relative to the tetramethylsilane internal standard) as the internal reference. Peak A refers to the (CH 3 ) 2 N protons and peak B to the (cH 3 ) 2 As protons. The approximate~ values are listed in Table 3. 23

PAGE 33

JOOO 2800 1400 1200 1000 800 600 Fig. 7.-Infrared Spectrum of [(cH 3 ) 3 AsN (CH 3 ) 2 ]c1 (Mull). 400 -1 cm

PAGE 34

25 A Fig. 8.-N. M. R~ Spectrum of [( cH 3 ) 3 A s N (C H 3 ) 2 ]cl.

PAGE 35

TABLE 3 N. M. R. SPECTRAL DATA FOR ((CH3)3AsN(CH3)2]c1 Group Chemical Shift, 'T 7.37 7.68 The ratio of the area under the (CH 3 ) 2 N peak to that under the (CH 3 ) 3 As peak is 1:1.44. Yield of dimethylamino trimethylarsonium chloride was 0.44 g (46 % of theory based upon the trimethylarsine put into the reaction). The Reaction of Dimethylchloramine w ith Triethyl arsine.+ ...... Triethylarsine (5 mmole) was added inside the dry-box to 25 ml of diethylether in a reaction flask equipped with a magnetic stirrer. The reaction flask and contents were degassed on the vacuum line. Onto the arsine was 26 condensed an ether solution of dimethylchloramine (44 mmole). The reaction mixture was then warmed to room temperature. A white precipitate formed as the reaction solution warmed. The reaction was allo w ed to continue for six hours, during which time a large amount of white solid formed. All condensible materials were then distilled from the reaction flask. The white solid was taken into the dry-box, washed with diethylether several times, and dried in vacuo. The

PAGE 36

melting point dataare as follows: 0 107-110 c,, solid darkens and becomes a jelly-like material; 128-129c, material melts to a clear liquid. Anal. Found: C, 39.54; H, 8.75; N, 6.01; As, 30.83; Cl, 14.96. Calculated for [(cH3CH2)3AsN(CH3)2]c1: c, 39.76; H, 8.76; N 5.80; As, 31.00; Cl, 14.67. The infrared spectrum is shown in-Figure 9. The proton magnetic resonance spectr~m (Figure 10) was measured using CDc1 3 as the solvent and CHC1 3 (-436 cps relative to the tetramethylsilane internal standard) as the internal reference. Peak A refers to the methylene and dimethylamino protons and peak B to the methyl protons. The approximate~ values and average chemical shift values are listed in Table 4. The ratio TABLE 4 N. M. R. SPECTRAL DATA FOR [(CH3CH2)3AsN(CH3)2]c1 Group Chemical Shift, CH 2 7.30 (ave,.) N(CH 3 ) 2 7.34 CHJ 8.82 (ave.) of the areas under the CH 3 peaks to those under the CH 2 and N(CH 3 ) 2 peaks is 1:1.3. Yield of dimethylamino triethylarsonium chloride was 0.48 g (40% of theory based upon the amount of triethylarsine put into the reaction). 27

PAGE 37

3000 2800 1400 1200 1000 8 00 600 Fig. 9.-Infrared Spectrum of [(C H JC H 2 ) 3 As N (C H J) 2 ]c1 (Mull). 400 -1 cm l\) 0)

PAGE 38

29 A B Fig. 10.N M. R. Spectrum of ~c H 3 c H 2 ) 3 As N (C H 3 ) 2 ]c1.

PAGE 39

The Reaction of Dimethylchloramine with Tri-n-propyl arsine.+ 30 Inside the dry-box, tri-n-propylarsine (2.0 g, 9.8 mmole) was added to 80 ml of diethylether in a reaction flask. The flask was attached to the vacuum line and the solution was degassed. Onto this solution was condensed 11.1 ml of a 0.88 M dimethylchloramine solution (9.8 mmole) in diethyl ether. The reaction mixture was allowed to warm to room temperature and react for 48 hours. A white solid slowly began to precipitate. The volatile materials were distilled from the reaction flask, leaving a pale yellow residue. Purification of the product was carried out by dissolving the solid in warm benzene and precipitating the pale yellow solid by adding a small amount of hexane. The pale yellow material was washed with hexane and dried in vacuo. Mp 105-107c. Anal. Found: C, 46.63; H, 9.69; N, 4.94; As, 26.33; Cl, 12.72. Calculated for [(cH 3 CH 2 CH 2 ) 3 AsN(CH 3 ~Cl: C, 46.54; H, 9.60; N, 4.94; As, 26.42; Cl, 12.50. The infrared spectrum is shown in Figure 11. The proton magnetic resonance spectrum (Figure 12) was measured us ing CDc1 3 as the solvent and tetramethylsilane as the internal reference. Peak A refers to the ~-CH 2 protons, peak B to the N(CH 3 ) 2 protons, peak C to the ~-CH 2 protons, and peak D to the CH 3 protons. The approximate 1' values and average chemical shifts are given in Table 5.

PAGE 40

3000 2800 Fig. 1400 1200 11.~Infrared Spectrum of 1000 8 00 600 400 [( C H 3 c H 2 C H 2 ) 3 As N ( C H 3 ) 2 ] Cl (Mull) -1 cm

PAGE 42

TABLE 5 N. M. R. SPECTRAL DATA FOR [(CH3CH2CH2)3AsN(CH3)2]c1 Group Chemical Shift, T ci(-CH 2 7.08 (ave.) N(CH 3 ) 2 7.18 /3-CH 2 8.25 (ave. ) CH 3 8.88 (ave.) The ratio of the areas under the ~-CH 2 and N(CH 3 ) 2 peaks to that under the fl-cH 2 peaks was 1.96:1. The ratio of the areas under the ~-CH 2 and N(CH 3 ) 2 peaks to that under 33 the CH 3 peaks was 1.23: 1. Yield of dimethylaminotri-n-propyl arsonium chloride was 1.76 g or 63% of theory based upon the amount of tri-n-propylarsine put into the reaction. The R eaction of Am monia-free Chloramine with 10-chloro-5,10-dihydrophenars a zine.Onto 10-chloro-5,10-dihydrophenarsazine (1.93 g, 6.98 mmole) was condensed 93 ml of 0.075 M chloramine (6.98 mmole) in a diethylether solution. The reaction mixture was then warmed to room temperature and was stirred for 4 hours. A yello~-green solid was foraed during the reaction. All volatile materials were distilled from the reaction flask. The yellow-green product was washed with diethylether

PAGE 43

and dried in vacuo. The melting point data are as follows: 208c, compound turned dark green in color; 212c, solid 34 0 became a dark green viscous material; by 250 C, the material is black in color; 263c, vigorous b~bbling occurs. Anal. Found: C, 45.41; H, 4.26; N, 7.25; As, 20.49; Cl, 18.04. Calculated for (HN(C 6 H 4 ) 2 As(Cl)( NH 2 )] Cl.5% (CH3CH2)20: C, 45.91; H, 4.39; N, 7.66; A s, 20.49; Cl, 19.39. The infrared spectrum is given in Figure 13. Yield of 10-methyl-10-amino-5,10-dihydrophenarsazonium chloride was 2.04 g or 89% of theory based upon the amo unt of chloramine put into the rea~tion. The Reaction of Ammonia-free Chlora~ine with 10-methyl-5,10-dihydrophenarsazine.+ NH 2 Cl [HN~C 6 H)As~CHJ] Cl c 6 H 4 NH 2 C H HN"' 6 4 'AsCH 'c H/ J 6 4 To 10-methyl-5,10-dihydrophenarsazine (1.4 g, 5.4 m~ole) dissolved in 50 ml of diethylether was added, with constant stirring, 70 ml of a 0.2 M ammonia-free chloramine solution in diethylether. A white solid formed immediately. The diethylether and excess chloramine were removed by filtering. The white solid was washed several times with diethylether and dried in vacuo. The melting point of the product gave the following results: 14o 0 c, shrinkage; 150c, white solid turned tan in color; 168-170c, solid expanded and turned bright yellow in color; 176c, melted with decomposition. Anal. Found: C, 51.29; H, 5.28; N, 8.66;

PAGE 44

3400 JOOO Fig. II 1600 1400 1200 1000 800 600 1J.-Infrared Spectrum of [HN(C 6 H 4 ) 2 As(Cl)(NH 2 ))Cl (Mull). 400 -1 cm

PAGE 45

36 As, 23.65; Cl, 11.20. Calculated for [H N (C 6 E 4 ) 2 AsCH 3 ( N H 2 )]Cl: C, 50.27; H, 4.57; N, 9.11; As, 24.27; Cl, 11.48. The infrared spectrum is given in Figure 14. The proton magnetic resonance spectrum (Figure 15) was obtained using d 6 -DMSO as the solvent with the dimethylsulfoxide peak (-149 cps relative to the sodium salt of 3-(trimethylsilyl)-propane sulfonic acid as internal standard) as the internal reference. The approximate~ values and average chemical shifts are given in Table 6. Yield of 10-methyl-10-amino-5,10-dihydro phenarsazonium chloride: 90% of theory. TABLE 6 Group Chemical Shift,T NH -0.80 C6H4 2.93 (ave.) NH 2 3.13 CH 3 7.97 Discussion The products of the reactions of the trialkylarsines with chloramine and dimethylchloramine have been found, in all cases, to be the aminoarsonium salts. These amino

PAGE 46

3200 3000 1600 1400 1200 1000 800 600 Fig. 14.-Infrared Spectrum of (HN(c 6 rr 4 ) 2 As(CH 3 )(NH 2 )] Cl (Mull). 400 :..1 cm

PAGE 47

C6H4 NH 2 CHJ Fig. 15.-N. M. R. Spectrum of [HN(C6H4)2As(CHJ)(NH2)Jc1.

PAGE 48

arsonium salts are the direct analogs of the products of the corresponding amine and phosphine reactions with chloramine. If a tetracoordinate arsenic is assumed, then the structure [R 3 AsNR 2 ]c1 would be the most probable. The unshared pair of electrons on the nitrogen atom would allow rr bonding with the arsenic in an aminoarsonium salt. Assuming an SN 2 reaction mechanism, the Cl group would be a better leaving group than a R 2 N group. The absence of a new infrared absorption peak in the region expected for an stretching vibration to occur, JOO to 450 -1 cm further supports the assumption of the aminoarsonium chloride As-Cl 39 structure. (The P-Cl stretch occurs in the 440 to 580 cm-l( 2 0) whereas the Sb-Cl stretch occurs below JOO cm-\( 2 l) ) This absence of the As-Cl stretching band indicates that the chloramination products probably do not contain penta coordinate arsenic. In comparing the infrared spectra of the trialkylarsine starting materials with the respective aminoarsonium salt products, the infrared spectra of the products show, in each case, a new peak of medium to strong intensity in 8 -1 the 700 to 50 cm region. This peak is probably due to the As-N stretching mode. This assignment is not unreasonable, since the abso rption peak for the P-N stretching mode appears between 750 and 1000 cm1 ( 2 0) depending upon the substituents on the nitrogen.

PAGE 49

40 The chloraminations of the triethyland tri-n-propyl arsine proceed i n stant a neously in diethylether at a low temperature. Sisler and Stratton(S) observed that trimethyl arsine reacts with chloramine instantaneously at a low temperature, but triphenylarsine does not react at a 0 measurable rate at -79 C. Assuming that the chloramination reaction proceeds via an SN 2 type of bimolecular reaction mechanism with chloramine acting as the electrophile, an increase in electron density on arsenic would increase the reaction rate. Therefore, the triethyland tri-n-propyl arsine reactions support the postulate that one may correlate the variation in reaction rates with the higher electron density on the arsenic in the alkyl derivatives than in the phenyl derivatives. The dimethylchloramine reactions do not occur instantaneously at -79c as was found to be the case in the analogous chloramine reactions, but proceed slowly even at room temperature. In fact, 100% conversion of the arsine to the arsonium salt was never obtained, even when the dimethylchloramine was in a 10:1 molar excess. By analogy with chloramine, one might expect these reactions to proceed by an SN 2 reaction mechanism with dimethylchloramine acting as the electrophile. The methyl groups on the nitrogen are electron donating, thereby increasing the electron density on the nitrogen atom and consequently decreasing its electrophilic character. Therefore, dimethylchloramine

PAGE 50

41 should be a weaker electrophile than chloramine. Neglecting steric factors, its reactions, in comparison with those of chloramine, should proceed more slowly and at higher temperatures. This, indeed, appears to be the case. The reactions of chloramine with 10-chloroand 10-methyl-5,10-dihydrophenarsazine presented the opportunity to observe which of the two potential Lewis-base sites nitrogen and arsenic would be attacked by chloramine. The infrared spectra indicate that the chloramine molecule preferentially attacks the arsenic in both cases. As in the aminotrialkylarsonium salts, the infrared spectra of the phenarsazonium salts show the appearance of a new band in the 700 to 850 cm-l region. This would indicate the existence of an As-N bond in the reaction product. Further more the chloramination products show no reducing properties (negative reaction toward iodate). If an hydrazinium salt had been formed by amination of the nitrogen atom, there should be a positive indication of reducing power of the chloramination product. This implies, neglecting any stereochemical arguments, that the arsenic atom is more susceptible to electrophilic attack than is nitrogen. The ability of the arsenic atom to utilize its vacant 4d-orbitals in forming a d~-p~ bond with nitrogen may influence the formation of the aminoarsonium salt. Also the electron cloud of the arsenic atom in the 10-position should be more diffuse than that of the nitrogen in the 9-position. The arsenic atom is more readily polarized than the nitrogen

PAGE 51

and a transition state, NH 2 ca~ exist for a longer time when the chloramine molecule attacks the arsenic atom. The infrared spectrum of the methyl deriva tive shows no absorption peak for an As-Cl stretching mode. Therefore, pent aco ordinate arsenic can probably be ruled out in the reaction product. Summary The reactions of chloramine and dimethylchloramine with trialkylarsine have resulted in the formation of heretofore unknown am in oa rsonium salts. These results are directly analogous to those of the corresponding amine and phosphine reactions with chloramine. Therefore, the generality of the chloramination reaction has been extended to include the trialkylarsines. Analogous chloramine and dimethylchloramine reactions tend to indicate that correlations may be made in the variation in re a ction rates with the relative electro philicities of the dimethylchloramine and chloramine molecules. As expected, chloramine appears to be a stronger electrophile than dimethylchloramine. 42 In the re ac tions with the 10-chloroand 10-methyl5,10-dihydrophenarsazines, where there are two potential electron-donor sites for electrophilic attack by chloramine, the chloramine molecule preferentially attacks the arsenic site. This implies, neglecting any stereochemical arguments, that the arsenic atom is more susceptible to electrophilic attack than is the nitrogen atom.

PAGE 52

Bibliogr.:J2.hy 1. G. M. Omietanski, A. D. Kelmers, R. W. Shellman, and H. H. Sisler, J. Am. Chem. Soc., .'ZJ1, 3874 (1956). 2. H. H. Sisler, A. Sarkis, H. S. Ahuja, R. J. Drago, and N. L. Smith, J. Am. Chem. Soc., ._, 2982 (1959). w. A. Hart and H. H. Sisler, Inorg. Chem., J.' 617 (1964). 3. 4. 5. D. F. Clemens and H. H. Sisler, ibid., ~' 1222 (1965). s. R. Jain, w. s. Brey, Jr. and H. H. Sisler, ibid., 2.' 515 (1967). 6. s. R. Jain, L. K. Krannich, R. E. Highsmith and 7. 8. 9. H. H. Sisler, ibid.,_, 1058 (1967). H. H. Sisler and s. R. Jain, ibid. Z, H. H. Sisler and c. Stratton, ibid., 2' R. L. McKenney and H. H. Sisler, ibid. 104 (1968). 2003 (1966). 2.' 1178 (1967). 10. J. W. Cahn and R. E. Powell, J. Am. Chem. Soc., 7., 2565 (1954). 11. R. S. Drago and H. H. Sisler, ibid., 11., 3191 (1955). 12. G. M. Omietanski and H. H. Sisler, ibid., .2.._, 1211 ( 19 56). 13. D. B. Denney and S. M. Felton, Inorg. Chem., Z, 99 (1968). 14. K. A. Petrov and G. A. Sohol'ski, Zh. Obshch. Khim., 26, 3377 (1956). 15. W. Stamm and A. Breindel, Angew. Chem., Z, 99 (1964). 16. H. Burton and C. S. Gibson, J. Chem. Soc., 450 (1926). 17. C. S. Gibson and J. D. A. Johnson, J. Chem. Soc., 2518 (1931). 18. R. Mattair and H. H. Sisler, J. Am. Chem. Soc., 1.J., 1619 (1951). 19. A. Berg. Ann. Chim. Phys., J., 319 (1894).

PAGE 53

44 20. L. J. Bellamy The I n frared Spectra of Comp l ex Mo lecules, John W iley & Sons Inc., New York 1960, p 324. 21. R. L. McKenney, PhD Dissertation, University of Florida, December 1965.

PAGE 54

CHAPTER III REACTIONS OF AMMONIA, DEUTERATED AMMONIA, AND A M MO N IA-CHLORAMINE MIXTURES WITH DIMETHYLCHLOROARSINE Historical Background The reactions of dimethylchloroarsine with ammonia or in the presence of ammonia have not been reported in the literature. M~dritzer has reacted dimethylchloro arsine with dimethylamine to obtain the expected dimethyl amino-substituted arsine. (l) + + The bis(trifluoromethyl)chloroarsine analog has been reacted with ammonia to give the aminoarsine and the condensed bis-arsineamine. ( 2 ) + + [( CF J) 2 As] 2 NH + JNH 4 c1 Dimethylchloroarsine has been reacted with methyl iodide to give tetramethylarsonium tri-iodide. (J) + + No successful attempts to use dimethylchloroarsine to prepare dimethyl-substituted arsenonitriles have been reported. Only one arsenonitrilic compound, [(c 6 H 5 ) 2 AsN] 4 has been prepared. Reichle( 4 ) prepared this diphenyl substituted compound by reacting diphenylchloroarsine with 45

PAGE 55

lithium azide and pyrolyzing the immediate reaction product. (c 6 H 5 ) 2 AsCl + LiNJ 4(c 6 H 5 ) 2 AsNJ Sisler and Stratton(5) (C 6 H 5 ) 2 AsNJ + LiCl ~c 6 H 5 ) 2 AsN] 4 + 4N 2 on the other hand, have prepared this compound by the simultaneous chloramination and ammonolysis of diphenylchloroarsine. + + [(c 6 H 5 ) 2 AsN) 4 + 4NH 4 c1 This is the only known arsenic analog of the well-known phosphonitrilic compounds. The phosphonitrilic chlorides, [c1 2 P N ]n' have been known for many years. The earliest known method for their preparation involved the ammonolysis of phosphorus penta chloride. ( 6 ) The reaction product is a mixture of cyclic and linear polymers which may be separated by fractional crystallization and distillation. The trimeric and tetrameric phosphonitrilic chlorides can be further reacted to form many halogen, pseudohalogen, alkyl, aryl, alkoxy, aryloxy, N-substituted primary and secondary amine, hydroxy, and other derivatives. Various phosphonitrilic polymers have been easily prepared by the reaction of an ammonia-chloramine mixture with di-substituted chlorophosphines or tetra-substituted diphosphines. (?,S, 9 ) + + + + [R 2 PN] 4 [R 2 PN] 4 + + 46

PAGE 56

Interest in cyclic phosphonitriles arises from the stability of these compounds and from the speculation that the phosphorus-nitrogen rings have aromatic character. The thermal stability of some of these compounds have made them of special interest in inor g anic polymer chemistry. From a theoretical standpoint, structural investigations of the phosphonitrilic polymers are of g reat interest. Use has been made of infrared and ultraviolet spectroscopy, nuclear magnetic resonance spectroscopy, and electron and X-ray diffraction in these structural studies. Early structural investigations established the fact of ~-bonding in the ring on the basis of observed bond lengths. The trimeric phosphonitrilic chloride has a planar, six-me m bered cyclic structure. Cl" /Cl p /" N N I II Cl-P P-Cl I'\/\ Cl N Cl Cl" /Cl /p' N N II I Cl-P P-Cl I" I\ Cl N Cl Theoretical studies have shown that even though the structures are formally a na logous to the Kekule' structures of benzene there are very essential differences. In the phosphonitrilics use may be made of both pand d-orbitals in the 11'-system, whereas in benzene the carbon atoms can utilize only p-orbitals in forming the Tr-system. Consequently, the bonding in the P-N system is complex, because of the variety of d-orbitals available and the symmetry differences of the pand d-orbitals. 47

PAGE 57

48 The fir s t arsenic analog of the phosphQnitrilics prepared by Reichle, Sisler, and Stratton was assumed to have a structure similar to that of the corresponding phosphonitrile. The As=N vibration was assigned to the most intense infrared absorption band occurring at 943 cm1 The physical properties and the infrared spectrum indicated that the bonding in the arsenonit rile was similar to that of the phosphonitrile viz., a 2prr-4d~ bond superimposed on the sigma bond system. For all the above reasons we decided to investigate the reactions of dimethylchloroarsine with chloramine, ammonia and mixtures of the two in order to determine whether or not dimethylarsenonitriles can be prepared. Experimental Manipulation of M aterials .-T he experimental techniqu es used were the same as those described on pages 6-8. The mater ial to be chloraminated on the generato r was added to the reaction flask (Figure 16) equipped with a large cold finger, an adapter for connection to the generato r, and an adapter for attachment to the vacuum line. The flask was then attached to the generator the connecting tubes were flushed with dry-nitrogen, the stopcocks on the reaction flask were opened, and the effluent gases from the gene ra to r were bubbled through the reaction solution for the desired length of time. The cold finger was always filled with dry-ice to minimize the

PAGE 58

49 Fig. 16.-Chloramination Reaction Flask.

PAGE 59

loss of solvent and reactants through evaporation. The nitrogen passed from the reaction flask through the stopcock and into a continually-flowing nitrogen stream. 50 Dimethylarsinic acid and magnesium nitride were obtained from K & K Laboratories and was used as obtained. Sodium hypophosphite was obtained from J. T. B a ker Company and was used as obtained. Gaseous ammonia and chlorine were obtained in high purity grade from Matheson Scientific, Inc., and used as obtained. Deuterium oxide and hexadeuterodimethylsulfoxide were obtained from Stohler Isotope Chemicals and used after being dried over type JA molecular sieve. Analyses.-Elemental analyses were done by Schwarzkopf and Galbraith laboratories. Infrared Spectra.-Infrared spectra were recorded on a Beckman IR-10 Spectrometer. A summary of the spectral bands of the materials produced in this study is found in Table 7. Nuclear Magnetic Resonance Spectra.-The proton magnetic resonance spectra were measured as described on page 9. The Preparation of Dimethylchloroarsine.-Dimethyl chloroarsine was prepared by the reaction of dimethylarsinic acid with sodium hypophosphite and hydrochloric acid according to the procedure described by Steinkopf and Mi (10) eg.

PAGE 60

TABLE 7(a) INFRARED ABSORPTION DATA (cm1 ) (CH3)2AsNH;c1-, Mull J12O(s), JO4O-29OO(s,bd), 226O(w), 221O(m), 175o(w), 14J5(w), 14O5(vs), 1255(m), 12JO(w), 11J5(m), 9O2(s), 872(s), 842(s), 8OO(w), 755(w,sh), 7J2(m), 67O(m), 59O(s), 545(vs), J15(w), 262(w). (CH3)2AsND;c1-, Mull 299O(m), 291O(m), 2J1O(vs), 225O(vs), 2180(s), 166O(w,bd), 14O5(m), 1255(s), 12JO(vw), 112O(m), 1O6O(vs), 9O5(s), 89O(s), 86O(s), 8J8(m), 810(m), 732(m), 67O(w), 6OO(m), 59O(s), 58O(s), 51O(vs). ((cH 3 ) 2 AsN] 4 HCl, Mull 298O(s), 296O(s), 294O(s), 29OO(s), 254O(m), 142O(w,sh), 14O5(m), 1J8O(w,sh), 1268(m), 1OOO(s), 9OO(s), 865(m), 8JO(m), 80O(s), 72O(mpd), 665(w), 64O(m), 62O(m), 6OO(w), 528(m), JO5(m), 295(w,sh). (a)s, strong; m, medium; w, weak; v, very; sh, shoulder, bd, broad. 51

PAGE 61

The Preparation of Deuterated Ammon ia.~Deuterated ammonia was prepared by the reaction of magnesium nitride with deuterium oxide. The deuterated ammonia was passed through a column of magnesium nitride to insure the removal of all water vapor and condensed into a flask on the vacuum line. 52 The Preparation of Chloramine.-Chloramine was prepared by the procedure described on p~ges 14-18. (ll) The Reaction of Ammonia with Dimethylchloroarsine.+ -+ ( CHJ) 2 AsNH; Cl Approxi ma tely 25 ml of ammonia at -79C from a gas cylinder was dried over sodium in a flask on the vacuum line. Dimethylchloroarsine (2 ml, 22 mmole) was added to a reaction flask containing 25 ml of diethylether. Dry ammonia (0.46 ml at -79C, 22 mmole) was distilled into the reaction flask. The reaction mixture was warmed to room temperature. As soon as the ether melted, a white solid formed. After a short time, the volatile materials were distilled from the reaction flask. The white solid 6 0 -4 / product was vacuum sublimed at 0 C and 10 mm Hg. The product did not melt, but sublimed. Anal. Found: C, 15.48; H, 5.98; N, 8.45; As, 47.37; Cl, 22.18. Calculated for (cH 3 ) 2 AsNH;Cl-: C, 15.25; H, 5.76; N 1 8.90; As, 47.58; Cl, 22.51. The infrared spectrum is shown in Figure 17. The proton magnetic resonance spectrum was measured using hexadeuterodimethylsulfoxide as the solvent and the dimethylsulfoxide peak (-149 cps relative to the

PAGE 62

3000 2600 2400 1400 1200 1000 8 00 600 Fig. 17.-Infrared Spectrum of (cH 3 ) 2 As N H;c1(Mull). 400 -1 cm

PAGE 63

sodium salt of 3-(trim et hylsilyl)-propanesul(onic acid as the internal standard) as the internal reference. The spectrum consisted of two single peaks. The~ values are given in Table 8. The ratio of the area under the (cH 3 ) 2 As TABLE 8 N. M. R. SPECTRAL DATA FOR (CH3)2AsNH;c1Group Chemical Shift, -r 2.68 8.78 peak to that under the NHJ peak is 1.8:1. Yield of dimethylarsinoammonium chloride: 2.67 g (80 % of theory based upon the dimethylchloroarsine used). Reaction of Deuterated Ammonia w ith Dimethylchloro arsine.(CHJ)2AsCl + NDJ (CH3)2AsND;c1Dry deuterated ammonia from the reaction of deuterium oxide and magnesium nitride was transferred into a calibrated tube on the vacuum line until a volume of 0.46 ml at -79c was obtained (22 mmole). Dimethylchloroarsine (2 ml, 22 mmole) was added to a flask containing 25 ml of diethylether. The reaction flask was attached to the vacuum line and degassed. The deuterated ammonia was distilled into the reaction flask. The reaction flask 54

PAGE 64

was warmed to room temperature. When the di~thylether melted, a white solid formed im m ediately. After a short time, the volatile materials were distilled from the reaction flask. The remaining white solid was purified by vacuum sublimation at 6o 0 c and 104 mm/Hg. Anal. Found: C, 15.20; N, 8.67; As, 47.15; Cl, 21.J5. Calculated for (CH 3 ) 2 As N D;Cl-: C, 14.97; N, 8.73; As, 46.69; Cl, 22.09. The infrared spectrum is shown in Figure 18. The yield of (cH 3 ) 2 AsND;c1was 2.7 g (80 % of theory based upon the dimethylchlor oa rsine used). Reaction of An Ammonia-Chloramine Mixture with Dimethylchloroarsine.+ + 4NH 2 Cl ..... ~cH 3 ) 2 AsN] 4 HCl 6NH 4 c1 Dimethylchloroarsine (2.1 g, 15 mmole) was dissolved in 50 ml of 20-4o 0 c boiling petroleum ether in a reaction flask (Figure 16). This addition was carried out inside the dry-box. The flask was attached to the chloramine generator. The effluent gases from the generator were bubbled through the solution for 25 minutes (42 mmole of chloramine). A white solid precipitated immediately. After 6 minutes, the white solid appeared to dissolve and to form a dense oil immiscible with petroleum ether. After the chloramination process was completed, the reaction flask was warmed to room temperature, under a stream of dry nitrogen. As the ammonia evolved, the immiscible lower layer slowly turned to a white solid. + 55

PAGE 65

3000 2600 2400 1400 1200 1000 800 600 Fig. 18.-Infrared Spectrum of (cH 3 ) 2 AsND;c1(Mull). 400 -1 cm

PAGE 66

The reaction flask was attached to the vacuum line and all condensible mater ials were distilled from the reaction flask. The remaining tacky, white solid was taken into the dry-box and extracted with acetonitrile. The infrared spectrum of the remaining white solid was identical with that of amm onium chloride. A white crystalline solid was obta i ned when the acetonitrile solution was cooled or added to diethylether. The infrared spectrum of this white solid is shown in Fi gur e 19. Mp 217-218c with decomposition. A proton magnetic reson a nce spectrum was obtained using deuterium 57 oxide as the solvent and the sodium salt of J-(trimethylsilyl) propane sulfonic acid as the internal standard. A single proton reso nan ce was observed at 8.23~ and attributed to the (cH 3 ) 2 As group. The infrared spectrum, nmr spectrum, and ana lysis, in addition to the results of the analogous dimethylchloroph osph ine reactionf 7 ) suggest that the product is the dihydrochloride of the dimethylarsenonitrile tetramer. The nm r spectrum of this dihydrochloride of the arsenonitrile tetramer should show two different types of methyl protons if there is no rapid exchange of the NH protons. Since the spectrum was obtained using deuterium oxide as the solvent, rapid proton exchange is very possible. This may explain the observance of the single methyl resonance. Anal. Found: C, 17.68; H, 4.70; N, 10.45; As, 54.62; Cl, 13.27. Calculated for ~cH 3 ) 2 AsN] 4 HCl:

PAGE 67

.._ _________ ~ I ~ : __ ...._ _,._ __ ._ _._ __ .__ _,._ __ ..__ _._ __. _________ """:: 1 :-)000 2600 1400 1 2 00 1000 8 00 6 00 400 cm Fig. 19. -In f rared Spectrum of [( C H J) 2 As N ] 4 2 H C1 (Mull).

PAGE 68

C, 17.49; H, 4.78; N, 10.21; As, 54.60; Cl, 12.92. Yield of the product: 1.53 g of recrystallized product (75 % of theory based upon the amount of dimethylchloroarsine used). Attempts to remove the HCl att a ched to the arseno nitrile by extracting the product with triethylamine were unsuccessful. A quantitative recovery of triethylamine and tetramer was always obtained. Discussion 59 The results of this study show that dimethylchloro arsine readily reacts with ammonia in a 1:1 mole ratio to form the heretofore unknown arsinoammo n ium chloride and with an a m monia-chlor a mine mixture to give the dihydrochloride salt of the tetrameric dimethyl-substituted arsenorittrile. This establishes the first known alkyl-substituted arsenonitrile and opens up a new area of AsN chemistry. The results also extend the g e nerality of the ammonia c~loramine reaction with dialkylhalophosphines(?,B) to include the dialkylhaloarsines. It appears now, therefore, that chloramine can be an important synthetic intermediate in the preparation of arsenonitriles. In co ntrast to the ana l o g ous dimethylchlorophosphine reaction where pyrolysis of the chlor a mination product, ((cH 3 ) 2 P(NH 2 ) 2 ]c1, must be carried out to obtain the phosphonitrilic tetramer, the arsenonitrilic tetramer is obtained directly in the chloramination process of the

PAGE 69

dimethylchloroarsines. This sug g ests that diaminodimethyl arsonium chloride is unstable at room te m perature in the presence of ammonia and chloramine. The al k yl-substituted arsenonitriles appear to be very stable and much easier 60 to obtain synthetically than the correspo n ding phospho nitriles. The reaction of other dialkylchloroarsines should be carried out to completely establish this method of preparing these arsenic-nitrogen polymers. N o broad, intense infrared absorption peak was observed at 943 cm1 to correspond with the As=N assi g nment made by Reichle( 4 ) in the diphenyl derivative. Broad intense pe a ks are observed at 1000 cm1 900 cm1 -1 and 800 cm An assignment of one of these bands to the As= N vibration in the dimethyl derivative would be speculation. A shift to lower energy for this vibrational mode would be expected when the substituents on the arsenic are changed from phenyl to methyl. This change would decrease any effective d~-prr back-bonding existing in the As=N bond due to d-orbital expansion( 12 ), and thereby weaken this bond. The inability to remove the H Cl fr o m the tetramer indicates that this compound is probably quite basic. The analogous (cH 3 ) 2 PCl reaction with ammonia-chloramine does not give the hydrochloride salt. T his possibly suggests that there is less d~-prr bonding between the arsenic and nitrogen atoms than between the phosphorus and nitrogen atoms. Consequently the ~-bonding electrons

PAGE 70

may be localized more on the nitrogen atoms in the As!..!...!.N bond. A weaker d~-p~ interaction in the arsenonitriles is expected, since the larger covalent radius of arsenic and the diffuseness of the vacant 4d-orbitals on the arsenic should create a less effective overlap of these orbitals with the 2p-orbital on the nitrogen than in the case of phosphorus. Summary This study has demonstrated the use of chloramine as an important synthetic intermediate in the preparation 61 of the heretofore unknown dialkyl-substituted arsenonitriles. The preparation of this arsenonitrilic tetramer in addition to that previously reported of [(c 6 H 5 ) 2 AsN] 4 has opened a new area of arsenic-nitrogen chemistry in which there exists many research possibilities. The existence of the dihydrochloride and the inability to remove this HCl from the tetramer suggests that the nitrogen atoms in the As-N ring may be more basic and that the dff-p~ bonding in the As-N ring may be weaker than in the analogous phosphonitrilic tetramer. Bibliography 1. K. M~dritzer, Ber., .2.?., 2637 (1959). 2. W. R. Cullen and H.J. Emelius, J. Chem. Soc., 372 (1959). 3. A. E. Goddard, A Textbook of Inorganic Chemistry, Vol. XI, Charles Griffin and Co., London, England, 1938, p 28.

PAGE 71

4. 5. 6. 7. 8. 9. 10. 11. 12. W. T. Reichle, Tetrahedron Letters, 51 (1962). H. H. Sisler and C. Stratton, Inorg. Chem., 2, 2003 (1966). H. N. Stokes, Am. Chem. J., 1.2., 275 (1895); 18, 629 (1896); 12., 782 (1897); 20, ?7+0 (1898). s. E. Frazier, Master's Thesis, University of Florida, December, 1963. H. H. Sisler and S. E. Frazier, Inorg. Chem.,~' 1204 (1965). S. E. Frazier, Doctoral Dissertation, University of Florida, December, 1965. W. Steinkopf and W. Mieg, Ber., ..2.J., 1016 (1920). R. Matta ir and H. H. Sisler, J. Am. Chem. Soc., U, 1619 (1952). H. R. Allcock, Chem. & Eng. News, 46, 73 (1968). 62

PAGE 72

CHAPTER IV REACTIONS OF AMMONIA -F REE CHLORAMINE, DIMETHYLCHLORAMINE, AND CHLORI NE WITH ARSINE A N D SO M E PRIMAR Y AND SECO N DARY ARSINES Historical Backgroun d The reactions of arsine and the primary and secondary arsines have been the subject of intensive investigations over the last seventy years. Arsin e has been reacted with the halogens } 1 ) arsenic trichloride! 2 ) phosphorus trichloridef3) and phosphor~s pentachloridef3) The reaction with the halogens gives in all cases elemental arsenic and haloarsines. X2 + AsHJ(excess) X 2(excess) + As + haloarsines -. haloarsines X = Cl, Br The reaction with arsenic trichloride gives elemental arsenic + 2As + JHCl The reaction of arsine with phosphorus trichloride gives As 2 H 2 while w ith phosphorus pentachloride it gives elemental arsenic. + PClS + 2HC1 + As When As 2 H 2 is reacted with bromine and iodine, elemental arsenic and arsenic triiodide are obtained, 63

PAGE 73

respectively. ( 4 ) --t 2HBr + 2As -+ 2HI + Dehn and co-workers have investigated the reactions of the halogens with methylarsine to obtain arsenic-halogen containing species. ( 5 ) When gaseous methylarsine is passed through a solution of iodine, methyldiiodoarsine is obtained. + + 2HI There was evidence of intermediate compounds which were thought to be CHJAsHIHI and CH 3 AsHI. The overall reaction is described by the equations + CH 3 AsH 2 CH 3 AsHIHI CH 3 AsHI + I 2 I 2 -+ CH 3 AsHI HI -+ CH 3 AsHI + HI -t CH 3 AsI 2 HI -+ CH 3 AsI 2 + HI If the arsine is in excess, the product is a brown amorphous mass, the composition of which was not elucidated. Using a solution of bromine in carbondisulfide, arsenic tribromide is obtained. ( 5 ) + --+ AsBr 3 + 2HBr + The haloarsines are obtained probably because the halogen was kept in excess in these reactions. The reaction of methylarsine with methyliodide yields the tetramethylarsonium iodide.( 5 ) + + 2HI 64

PAGE 74

The reaction of methylarsine with meth~larsenic oxide gives a good yield of cyclo-pentamethylpentaarsine( 6 ) + + which can exist in two forms a yellow liquid and a red solid. The red solid is forme~ by heating the 0 yellow form to 100 C. This red solid then melts at 180c to give the yellow liquid again. The cyclo-pentamethylpentaarsine is commonly prepared by the action of a mild reducing agent such as 50% hypophosphorus acid on sodium methylarsonate. (?,B, 9 ) Dehn and Wilcox have studied the reactions of the halogens, haloarsines, and alkylhalides with dimethyl arsine. (lO) Chlorine in excess reacts with dimethylarsine to give dimethylchloroarsine and methyldichloroarsine as products. If the dimethylarsine is maintained in excess, the products are elemental arsenic and a black solid polymer. The reaction is very vigorous. Bromine in excess reacts with dimethylarsine to give (cH 3 ) 2 AsBrHBr and (CH 3 ) 2 AsBr in good yields. (lO) (CH 3 ) 2 AsH + Br 2 __. (cH 3 ) 2 AsBrHBr ( CHJ) 2 AsH HBr ( CHJ) 2 AsBr + H 2 ( CHJ) 2 AsH + HBr __. ( CHJ) 2 AsBr + H 2 65

PAGE 75

Iodine also reacts with dimethylarsine to form only (CH 3 ) 2 AsIHI. (CH 3 ) 2 AsH + This compound in the presence of water gives dimethyliodoarsine. Dehn(lO,ll) has shown that dimethylarsine reacts with dimethylchloroarsine to give tetramethyldiarsine. + + HCl This reaction has been used to make the unsymmetrical tetraalkyldiarsines when R 2 AsH and R 2 AsCl are used as starting materials. (ll) He has also shown that dimethylarsine will react with arsenic trichloride. (lO) + + Dehn found that dimethylarsine reacts with many alkyliodides to give the arsonium iodides. (lO) + 2R'I + HI 2HC1 Tetramethyldiarsine (Cadet's liquid, Cacodyl, Alkarsin) is one of the oldest known alkylarsenicals and was first prepared by L. C. Cadet de Gassicourt in the 1700's( 12 ) by the heating of a mixture of equal parts by weight of arsenic trioxide and potassium acetate in a glass retort. In addition to its preparation by the reaction of dimethylarsine with dimethylchloroarsine, it may be obtained by the action of dimethylchloroarsine on zinc dust. (lJ) 66

PAGE 76

2 ( CHJ) 2 AsCl + Zn ( CHJ) 2 AsAs ( CHJ) 2 ZnC1 2 F. F. Blicke! 14 ,l 5 ,l 6 ,l?) W. M. Dehn! 5 ,lO) and their co-workers have studied the reactions of phenylarsine with bromine, iodine, and substituted haloarsines. Dehn reacted phenylarsine with excess bromine to obtain arsenic tribromide and bromobenzene as the principle products. ( 5 ) c 6 H 5 AsH 2 + Br 2 c 6 H 5 AsHBrHBr c 6 H 5 AsHBr + Br 2 c 6 H 5 AsBr 2 HBr c 6 H 5 AsBr 2 + Br 2 c 6 H 5 Br + AsBrJ The intermediate compounds were postulated, but were not specifically shown by analytical results to exist. Blicke and co-workers have shown that iodine reacts with phenylarsine to form diphenyldiiododiarsine and arsenobenzene, depending upon the reaction conditions. (l?) 2c 6 H 5 AsH 2 + 6c 6 H 5 AsH 2 + + 4HI 12HI The diiododiarsine may be prepared by the reaction of phenyldiiodoarsine with phenylarstne. (l5) + + 2HI Blicke has also shown that phenyldichloroarsine reacts with diphenylarsine and phenylarsine to give arseno benzene, diphenylchloroarsine, and tetraphenyldiarsine as products. (l 5 ) + 12 ( c 6 H 5 ) 2 AsH 2 ( c 6 H 5 As) 6 12(C6H 5 ) 2 AsCl + + 12HC1

PAGE 77

+ + 12(c 6 H 5 ) 2 AsH -, (C 6 H 5 As)fr + 2(C 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 + 12HC1 Similarly phenyldiiodoarsine reacts with phenylarsine to form arsenobenzene and the diphenyldiiododiarsine of which no chlorine analog is known. (l5) 3c 6 H 5 AsI 2 + JC 6 H 5 AsH 2 ( c 6 H 5 As) 6 + 6HI 3c 6 H 5 AsI 2 + c 6 H 5 AsH 2 2c 6 H 5 (I)AsAs(I)c 6 H 5 + 2HI The diphenyldiiododiarsine is evidently the intermediate in the reaction of iodine with excess phenylarsine, since 68 it will react with phenylarsine to give the arsenobenzene. (l 4 ) 2c 6 H 5 (I)AsAs(I)c 6 H 5 + 2c 6 H 5 AsH 2 _. (c 6 H 5 As) 6 + 4HI The diphenyldiiododiarsine will also react with diphenylarsine to give arsenobenzene as the product.(l 4 ) 6c 6 H 5 (I)AsAs(I)C 6 H 5 + 6(c 6 H 5 ) 2 AsH -, (c 6 H 5 As) 6 + 6HI + J(C 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 Phenylarsine will react with diphenylchloroarsine and diphenyliodoarsine to give as one of the reaction products arsenobenzene. (l5) 12 ( c 6 H 5 ) 2 AsCl + 6c 6 H 5 AsH 2 _,. ( c 6 H 5 As) 6 + 12HC1 6(c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 + 12HI 6(c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 When phenylarsine is oxidized very slowly by exposure of an ether solution of the arsine to the atmosphere, some arsenobenzene is obtained. (5)

PAGE 78

6c 6 H 5 AsH 2 + 30 2 (c 6 H 5 As) 6 + 6H 2 0 Palmer and Adams(lS) have shown that phenylarsine reacts with aromatic aldehydes in the presence of glacial acetic acid to give good yields of arsenobenzene. + 6RCHO + Arsenobenzene is also readily prepared by the reaction of phenylarsine with phenylarsenic oxide. (l 9 ) + + Phenylarsine has been reacted with methyl and ethyl iodides at 120c to give the respective phenyltrialkyl. d"d (20) arson1um 10 1 es. + JRI + 2HI Similar types of reactions as described above have been carried out using diphenylarsine. When diphenylarsine in excess is reacted with iodine;lO,l?) tetraphenyldiarsine is obtained 2 ( c 6 H 5 ) 2 AsH + I 2 _. ( c 6 H 5 ) 2 AsAs ( c 6 H 5 ) 2 + 2HI If the iodine is maintained in excess}l?) diphenyliodoarsine is obtained (c 6 H 5 ) 2 AsH + I 2 _. (c 6 H 5 ) 2 AsI + HI This indicates that tetraphenyldiarsine is first formed and then reacts with more iodine to give the iodoarsine. The reactions of the tetraaryldiarsines with chlorine to give diarylhaloarsines is well known. ( 2 l) +

PAGE 79

The reaction of diphenylarsine with bromine in excess gives both diphenylbromoarsine and diphenylarsenic tribromide. (lO) ( c 6 H 5 ) 2 AsH + Br 2 ( c 6 H 5 ) 2 AsBr + HBr (c 6 H 5 ) 2 AsBr + Br 2 (c 6 H 5 ) 2 AsBrJ Diphenylarsine reacts with many arsenic-halogen compounds to give tetraphenyldiarsine as one of the products. (lS,l 7 ) (c 6 H 5 ) 2 AsCl (c 6 H 5 ) 2 AsI + + + 12(C 6 H 5 ) 2 AsH (C 6 H 5 As) 6 + 6(c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 + 12HC1 (C 6 H 5 ) 2 AsH (C 6 H 5 ) 2 AsH --+ (c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 + HCl (C 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 + HI + + 2(C 6 H 5 ) 2 AsH --+ (C 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 + (C 6 H 5 ) 3 As + 2HC1 2 ( c 6 H 5 ) 2 AsH ( c 6 H 5 ) 2 AsAs ( c 6 H 5 ) 2 + (C6H 5 ) 3 As + HCl + H 2 0 Blicke and Smith have shown that tetraphenyldiarsine may be obtained from the reaction of diphenylchloroarsine with mercury. ( 22 ) 2 ( c 6 H 5 ) 2 AsCl + Hg ( c 6 H 5 ) 2 AsAs ( c 6 H 5 ) 2 + HgC1 2 The tetraphenyldiarsine is also the principal product from the reaction of diphenylarsine with [(c 6 H 5 ) 2 As] 2 o or [(c 6 H 5 ) 2 As] 2 s. (l7) No mention is made in the literature of the reaction of chloramine or dimethylchloramine with arsine, a primary arsine, or a secondary arsine. Only one reference is made to the reaction of a substituted chloramine with a 70

PAGE 80

secondary phosphine, (c 6 H 5 ) 2 PH.( 1 J) In this case none of the products were elucidated. The authors thought that a mixture of aminoand chlorophosphines was present in the pasty reaction residue. The use of an excess of the substituted chloramine in the reaction caused secondary side reactions to occur which accounted for the inability to isolate pure reaction products. Experimental Man ipulation of Mat erials.-The experimental techniques used were the same as those described on pages 6-8. Triethylamine and acetonitrile were obtained as reagent grade materials and were dried and stored over calcium hydride. Aluminum arsenide, sodium methylarsenate, and dimethylarsinic acid were obtained from K & K Laboratories and were used as obtained. Benzenearsonic acid and triphenylarsine were obtained from Eastman Organic Chemicals and were used as obtained. Benzaldehyde, zinc dust, mercuric chloride, and sodium hypophosphite were 71 obtained from J. T. Baker Company and were used as obtained. Lithium aluminum hydride was purchased from City Chemical Company and used as obtained. Hypophosphorus acid was obtained from Matheson Coleman and Bell and was used as obtained. Gaseous chlorine was obtained as high purity grade from Matheson Scientific, Inc.,and used as obtained. Deutero chloroform and hexadeuterobenzene were obtained from Stohler Isotope Chemicals and used as obtained.

PAGE 81

Analyses.-Elemental analyses were done by Schwarzkopf and Galbraith laboratories as described on page 8. Infrared Spectra.-Infrared Spectra were recorded on a Beckman IR-10 Spectrometer. The spectra were obtained of the neat liquids between KBr plates for the range 2.5-7-5p and between CsI plates for the range 7-5-20)'. The spectra of gases were obtained using a gas cell having KBr windows. A summary of the spectral bands of the materials used and produced in this study is found in Table 9. Nuclear Magnetic Resonance Spectra.-The proton magnetic resonance spectra were measured as described on page 9. Mass Spectra.-The mass spectra were obtained on a Hitachi Perkin-Elmer RMU-6E Mass Spectrometer at an ionizing voltage of 70 ev. The Preparation of Arsine.-Arsine was prepared in the vacuum line by the acid-hydrolysis of aluminum arsenide. An aqueous 20% sulfuric acid solution was added dropwise to the solid arsenide. The arsine formed was fractionally distilled and stored on the vacuum line. The infrared spectrum of the gas and the vapor pressure measurements on the liquid were identical with those reported for pure arsine. ( 24 ) The Preparation of Methylarsine.-Methylarsine was prepared by the reduction of sodium methylarsonate using 72

PAGE 82

TABLE 9(a) (cm -1) I N FRARED ABSORPTION DATA CH 3 AsH 2 Gas 73 J020(s), 2960(s), 2938(s), 2860(m), 2850(m), 2840(m), 2090(vs), 1440(m,bd), 920(vs,bd), 675(w,bd), 570(s), 56o(s), 548(s). (cH 3 ) 2 AsH, Neat J220(w), 2990(vs), 2950(vs), 2870(vs), 2610(w), 2500(vw), 24J0(vw), 2280(w), 2160(vw), 2075(vs), 1990(m), 1450(s,bd), 14J6(s), 1J95(vs), 1J8J(vs), 1J60(s), 1J50(s), 1295(m), 11J0(vs), 1980(s,sh), 1070(s), 914(s), 8 50(m), 8J?(m), 650(w), 582(s), 570(s), 488(vw), 4J0(m). c 6 H 5 AsH 2 Neat J075(m), J060(m), J020(m), J002(m), 2070(s), 1586(m), 1482(s), 14J8(s), 1J27(w), 1J00(vw), 1080(w), 1060(w), 1020(m), 996(m), 956(m), 700(s,bd), 4J0(m). (C 6 H 5 ) 2 AsH, Neat J075(s), J060(s), J0J0(m), J020(m), J005(m), 2060(s), 1958(w), 187J(w), 1815(w), 176o(w), 1584(w), 1482(s), 14J5(s), 1J80(w), 1J28(m), 1J00(m), 1260(w), 1180(w), 115J(w), 1080(m), 1072(m), 1062(m), 1020(s), 996(s), 980(w), 960(w), 900(w), 840(w), 784(s), 725(s), 695(s), 658(m), 610(vw), 465(s), 4J0(m). (CH 3 As) 5 Neat 2975(m), 2905(s), 2800(w), 2450(w), 1408(s), 1J65(m,sh), 12J6(s), 1090(w,bd), 820(s), 550(s). (CH 3 ) 2 AsAs(CH 3 ) 2 Neat 2980(s), 2910(s), 2810(m), 2460(w), 1810(w), 1412(s), 1245(s), 885(s), 820(s), 575(s), 565(s,sh). (C 6 H 5 ) 2 As A s(c 6 H 5 ) 2 Mull J062(w), J045(w), 1575(w), 1476(m), 1428(s), 1J02(w), 1180(w), 1150(w), 1072(w), 1068(m), 1016(m), 990(m), 900(w), 7J0(s,sh), 722( s), 684(s), 460(m), 445(m), J08(m).

PAGE 83

(c 6 H 5 As) 6 Mull 307o(w), 1570(w), 1472(m), 14JO(m), 1J10(w), 1294(w), 1059(w), 1015(w), 991(w), 732(s), 688(m), 460(m). c 3 H 9 8 As 1 4 Nc1 1 1 Mull J100(m,bd), 2972(s), 2905(s), 2820(m,bd), 1450(w), 1405(m), 1J15(w), 1270(w), 1090(w), 1000(m,bd), 980(m,bd), 900(s), 805(s), 7JO(s), 650(w), 640(w), 595(w), 420(m,bd), J50(m). (a)s, strong; m, medium; w, weak; v, very; sh, shoulder; bd, broad. 74

PAGE 84

hydrochloric acid and a zinc-mercury amalgam according to the procedure described by Dehn. ( 25 ) The infrared spectrum is given in Figure 20. The proton magnetic resonance spectrum (Figure 21) was obtained using CDc1 3 as the solvent and tetramethylsilane as the internal standard. Peak A refers to the AsH 2 protons and peak B to the methyl protons. The approximate 'T' values and average chemical shifts are given in Table 10. TABLE 10 N. M. R. SPECTRAL DATA FOR CH 3 AsH 2 Group Chemical Shift, -r 9.02 (ave.) 7. 9 3 (ave. ) The ratio of the area under the AsH 2 peak to that under the cH 3 peak is 1:1.54. The Preparation of Dimethylarsine.-Dimethylarsine was prepared by reducing dimethylarsinic acid in the presence of a zinc/mercury amalgam and hydrochloric acid according to the procedure described by Dehn and Wilcox.( 26 ) The infrared spectrum is shown in Figure 22. The proton magnetic resonance spectrum (Figure 23) was obtained using CDc1 3 as the solvent and tetramethylsilane as the internal 75

PAGE 85

J000 2800 2200 2000 1400 1200 1000 Fig. 20.-Infrared Spectrum of CH 3 AsH 2 (Gas). 800 600 -1 cm

PAGE 86

.. B Fig. 21.N M. R. Spectrum of C H 3 AsH 2

PAGE 87

3000 2800 2000 1600 1400 1200 1000 Fig. 22.-Infrared Spectrum of (CH 3 ) 2 AsH (Neat). 800 600 -1 cm

PAGE 88

A B TMS Fig. 2J.N M. R. Spectrum of (CH 3 ) 2 AsH.

PAGE 89

standard. Peak A refers to the AsH proton a~d_Peak B to the methyl p~otons. The approximate T values and average chemical shifts are given in Table 11. TABLE 11 N. M. R. SPECTRAL DATA FOR (CH 3 ) 2 AsH =====-=============:===-=-=-=-=-=-==-=-=--=7.:--=--=-=--==---___-_____--_ -_ -_ -_ -_-_ -_ -_ -_ -_ -_ -__:-_ -Group Chemical Shift, T 8.93 (ave.) 7. 57 (ave.) The Preparation of Phenylarsine.-Phenylarsine was prepared by reducing benzenearsonic acid by means of a zinc/mercury amalgam and hydrochloric acid according to the procedure described by Goddard. ( 2 o) The infrared spectrum is shown in Figure 24. The proton magnetic resonance spectrum (Figure 25) was obtained using CDc1 3 as the solvent and tetramethylsilane as the internal standard. Peak A refers to the phenyl protons and peak B to the AsH 2 protons. The~ values and average chemical shifts are given in Table 12. The ratio of the areas of the c 6 H 4 peaks to that of the AsH 2 peak is 2.51:1. 80 The Preparation of Diphenylchloroarsine.-Diphenyl chloroarsine was prepared by the action of arsenic trichloride on boiling triphenylarsine according to the following

PAGE 90

3200 3000 2200 2000 Fig. 1600 1400 1200 1000 8 00 24. Infrared Spectrum of c 6 H 5 AsH 2 ( N eat). 600 -1 cm

PAGE 91

A B Fig. 25.N M. R. Spectrum of c 6 H 5 As H 2 (X) l\)

PAGE 92

Group TABLE 12 Chemical Shift, T 2.70 (ave.) 6.44 equations and the procedure described by Goddard. ( 27 ) (C 6 H 5 ) 3 As 2(C 6 H 5 ) 3 As + + 2AsClJ JC 6 H 5 AsC1 2 + AsClJ J ( c 6 H 5 ) 2 AsCl (C 6 H 5 ) 3 As 2(C 6 H 5 )AsCl The Preparation of Diphenylarsine.-Dip ~ enylarsine was prepared by the reduction of diphenylchloroarsine with lithium aluminum hydride according to the procedure described by Wiberg and MBdritzer.( 2 S) Since a modification was made in the procedure described in this reference, the method of preparation will be described in detail. To a 500 ml, three-necked, round-bottom flask equipped with a mechanical stirrer, a pressurized dropping funnel, a reflux condenser, and a sintered-glass filter funnel attached to the base of the flask (Figure 26) W :3.S added 150 ml of diethylether and 8.34 g of lithium aluminum hydride (0.22 mole). The flask was cooled to -76c and JO g of diphenylchloroarsine (0.11 mole) in 50 ml of diethylether w as slowly added to the lithium aluminum hydride solution under a nitrogen stream. A very vigorous reaction occurred. When all the 83

PAGE 93

Fig. 26.-Reaction Flask Having a Sintered Glass Filter Funnel Base. 84

PAGE 94

arsine had been added, the reaction flask was ~armed to room temperature. In order to decompose the alane-diphenyl arsine complex, 25 ml of gaseous oxygen-free water was added to the reaction flask. The reaction mixture was filtered through the sintered-glass filter funnel in order to have minimal contact of the ethereal solution with oxygen. The ethereal solution was dried over calcium chloride and the diethylether distilled from the flask in a nitrogen atmosphere. The liquid remaining in the flask was fractionally distilled at a reduced pressure to give diphenylarsine boiling at 170-172c at 22 mm/Hg (Lit./ 29 ) 85 bp 174c at 25 mm/Hg). The infrared spectrum is shown in Figure 27. The proton magnetic resonance spectrum (Figure 28) was obtained using CDc1 3 as the solvent and tetrarnethylsilane as the internal standard. Peak A refers to the phenyl protons and peak B to the As-H proton. The approximate~ values and average chemical shifts are given in Table 13. The ratio of the area of the As-H peak to those of the c 6 H 5 peaks is 1:10.6. TABLE 13 N. M. R. SPECTRAL DATA FOR (C6H 5 ) 2 AsH Group Chemical Shift, 2.75 (ave.) 5.12

PAGE 95

JOOO 2000 1600 1400 1200 1000 800 Fig. 27.-Infrared Spectrum of (c 6 H 5 ) 2 AsH (Neat). 600 -1 cm

PAGE 96

l r A B Fig. 28.-N. M. R. Spectrum of (C 6 H 5 ) 2 As H

PAGE 97

Yield of diphenylarsine: 9.44 g (40 % of theory based upon the amount of diphenylchloroarsine used). The Preparation of Cyclo-pentamethylpentaarsine. Cyclo-pentamethylpentaarsine was prepared by the reaction of sodium methylarsonate and hypophosphorous acid according to the procedure described by Palmer and Scott. ( 3 ) The infrared spectrum is shown in Figure 29. The proton magnetic resonance spectrum (Figure JO) of the pure liquid cyclo-pentamethylpentaarsine was obtained at sweep widths of 500 cps, 100 cps, and 50 cps using tetramethylsilane as the internal standard. The spectrum shows three peaks in the approximate area ratios of 2:2:1; corresponding to the three magnetically non-equivalent types of methyl groups present in the compound. The average chemical shift was 8.J6T. A proton magnetic resonance spectrum was obtained using CDc1 3 as the solvent and tetramethylsilane as the internal standard. The spectrum was identical with that of the neat liquid. The average chemical shift was 8.35~. A proton magnetic resonance spectrum was obtained using cc1 4 as the solvent and tetramethylsilane as the internal standard. The spectrum was identical with that of the neat liquid. The average chemical shift was 8.35-,. A proton magnetic resonance spectrum (Figure Jl) was obtained at sweep widths of 500 cps, 100 cps, and 50 cps using c 6 D 6 as the solvent and tetramethylsilane as the internal standard. The spectrum shows three peaks in 88

PAGE 98

300 0 2600 1400 1200 1000 8 00 Fig. 29.-Infrared Spectrum of (c H 3 As) 5 ( N eat). 600 400 -1 cm

PAGE 99

50 cps 100 cps 500 cps Fig. JO.N ~ M. R. Spectrum of (cH 3 As) 5 N eat and at Various Sweep Widths. 0

PAGE 100

50 cps 100 cps Fig. 31.-N. M. R. Spectrum of (C IL 3 As) 5 c 6 n 6 Solution at Various Sweep Widths. 500 cps

PAGE 101

the area ratios of 2:1:2; coTresponding to th~ three magnetically non-equivalent types of methyl groups present in the cyclo-pentamethylpentaarsine. The average chemical shift was 8. 48 7'. A temperature-dependent nmr study of the neat liquid was carried out over the temperature range of -20c to 180c. The average chemical shift was unchanged with the change in temperature. The most upfield and dowofield methyl peaks each moved together 0.5 cps over the temperature range of 4o 0 c to 180c. 0 Below 20 C, very broad and ill-defined methyl peaks resulted, probably because of the increased viscosity of the liquid. 0 At -20 C, the width of the signal was 75 cps. The results are given in Figure 32 where the chemical shifts of the three methyl peaks relative to the midpoint of the two outer peaks are plotted against the temperature. A temperature-dependent nmr study of the c 6 n 6 -cyclo-pentamethylpentaarsine solution was carried out over the temperature range of -2o 0 c to 90c to determine the effect of temperature upon the relative positions of the methyl proton peaks and to study the presence of any inversion of the methyl groups about the ring. The results are shown in Figure 33 where the chemical shifts of the three methyl peaks relative to the midpoint of the two outer peaks are plotted against the temper.ature. As the temperature was lowered, the average chemical shift value increased 0. 7 5 '1' or moved 5 cps 92

PAGE 102

----. ~00 () ...__., ~-_ ... ___________ .,_, _________ __ ----~------. v-~-~ <,..., M ..c: 9 8 Ci) ,--j en () ,,-j .-.,------0----------':9 ___ _......_,._.-----....-------s Q) 96 ..c: 0 40 60 8 0 100 120 140 160 180 Temperature (oC) Fig. J2.-Temperature Dependence of Chemical Shifts of Methyl Groups in (C H 3 As) 5 ( N eat). \.,0

PAGE 103

102 ,,-... 100 r C/) P-< 0 98 ---~-------~ ----------0 _.,.o --___, --1 0 0 10 20 JO 40 5 0 0 Temperature ( C ) 6 0 70 8 0 Fig JJ .Temperature Depe nde n c e of Chemical Sh ifts of Methyl Groups in (C H 3 As) 5 (c 6 D 6 Solution ). 90

PAGE 104

upfield. The most upfield and downfield meth~l peaks each moved approximately 1.3 cps toward the midpoint as the temperature increased 110 degrees. A mass spectrum of the cyclo-pentamethylpentaarsine was obtained at an ionizing voltage of 70 ev and an inlet temperature of 150c. The mass spectral data are given in Table 14. m/e 90 103 105 163 165 195 225 255 270 435 450 TABLE 14 MASS SPECTRAL DATA FOR (CH 3 As) 5 Intensity 8 10 35 10 12 10 19 36 36 9 15 Relative Intensity 23 28 97 28 33 28 53 100 100 25 42 Assignment CH 3 As+ (CH 3 ) 2 As+ CH 3 As 2 + (CH 3 ) 3 As 2 + As 3 + (CH 3 ) 2 As 3 + (CH 3 ) 3 As 3 + (CH 3 ) 4 As 5 + (cH 3 As) 5 + 95

PAGE 105

The Preparation of Dimethylchloro a rsine.-Dimethyl chloroarsine was prepared by the reaction of dimethylarsinic acid with sodium hypophosphite and hydrochloric acid according to the procedure described by Steinkopf and Mieg. (JO) The Preparation of Tetramethyldiarsine.-Tetramethyl diarsine was prepared by the direct action of dimethylchloro arsine on zinc dust as described by Bunsen. (lJ) The infrared spectrum is shown in Figure J4. The proton magnetic resonance spectrum (Figure J5) was measured using CDClJ as the solvent and tetramethylsilane as the internal standard. The chemical shift of the single methyl peak varied from 8. 78 ,to 8. 86,-, depending upon the concentration of tetramethyldiarsine present. A mass spectrum was obtained using an ionizing voltage of 70 ev. The mass spectral data are given in Table 15. m/e 27 29 41 42 4J TABLE 15 MASS SPECTRAL DATA FOR (CH 3 ) 2 AsAs(CH 3 ) 2 Intensity 21 63 16 19 JO Relative Intensity 28 SJ 21 25 39 Assignment

PAGE 106

3000 2600 1400 1200 1000 8 00 6 00 -1 cm

PAGE 107

TMS

PAGE 108

Table 15 (cont'd) 75 10 13 As+ 88 16 21 89 58 76 90 36 47 CH 3 As+ 91 11 14 CH 3 AsH+ 101 50 66 103 76 100 105 76 100 ( CH 3 ) 2 As + 149 38 50 150 74 97 As 2 + 162 13 17 163 61 80 164 24 32 165 74 97 CH 3 As 2 + 167 22 29 180 70 92 (CH 3 ) 2 As 2 + 181 28 37 (CH 3 ) 2 As 2 H+ 182 13 17 (CH 3 ) 2 As 2 H 2 + 195 off scale >100 (CH 3 ) 2 AsAs(CH 3 )+ 210 off scale >100 (CH 3 ) 2 AsAs(CH 3 ) 2 + The Preparation of Arsenobenzene.-Arsenobenzene was prepared by the reaction of phenylarsine with benzaldehyde according to the procedure described by Palmer and Adams. (l 8 ) The infrared spectrum is shown in 99

PAGE 109

Figure 36. The low solubility of arsenobenzene in the common deuterated solvents resulted in the inability to obtain a proton magnetic resonance spectrum of this compound. A mass spectrum of the arsenobenzene was obtained using an ionizing voltage of 70 ev and an inlet temperature 0 of 195 C. The mass spectral data are given in Table 16. m/e 77 152 TABLE 16 MASS SPECTRAL DATA FOR (C6H 5 As)6 Intensity 25 14 Relative Intensity 100 Assignment The Preparation of Chloramine.-Chloramine was prepared by the procedure described on pages 14-18. The Preparation of Dimethylchloramine.-Dimethylchlor amine was prepared by the procedure described on pages 18-19. The Reaction of Ammonia-free Chloramine with Arsine.+ 2As + 100 Ammonia-free chloramine (100 ml of a 0.047 M solution in diethylether, 4.7 mmole) was condensed onto arsine (3.13 mmole) in a reaction flask on the vacuum line. When the reaction

PAGE 110

3000 1600 1400 1200 1000 800 Fig. 36.-Infrared Spectrum of_ (C 6 H 5 As) 6 (Mull). 600 -1 cm

PAGE 111

mixture was warmed to room temperature, a white solid precipitated. After 2 hours, the white solid began to turn light tan in color. For the next hour, the color of the solid slowly darkened until a very dark brown solid resulted and a silver-colored mirror forilled on the inside glass surface of the reaction flask. N o u n reacted chloramine remained. When a sample of t h e reaction product was extracted with boiling water, a w ater-insoluble black powder and a w ater-soluble material were obtained. A white solid was obt a ined upon evaporation of the aqueous solution from the extraction. The infrared spectrum of the vacuum sublimed white solid was identical with that of pure ammonium chloride. When a sample of the reaction product w a s heated in a vacuum sublimation apparatus at 110c and 104 mm/Hg, a white sublimate containing traces of a yellow material was obtained on the cold finger. Arsine was evolved during the sublimation procedure and was identified by its infrared spectrum. A black solid was obtained as the residue. Resublimation of the yellow-white sublimate always gave a yellow-white sublimate, arsine as a gas, and a brown-black residue. An infrared spectrum of the sublimate showed that the majority of the material was ammonium chloride. Anal. of the black solid residue. Found: H, O.Jl; N, 0.01; As, 88.99; Cl, 2.89. Calculated for elemental arsenic: As, 100.0. 102

PAGE 112

A mass spectrum of the sublimate was ohtained at an ionizing voltage of 70 ev, an inlet temperature of 150c, and a m/e range of 10-500. The mass spectral data are given in Table 17. TABLE 17 MASS SPECTRAL DATA FOR AsH 3 -NH 2 Cl REACTION SUBLIMATE m/e Intensity 15 18 16 77 17 80 36 42 76 20 110 27 112 10 145 77 147 74 149 22 154 3.2 155 3.6 180 62 182 60 184 24 186 3.2 Relative Intensity 23 96 100 53 25 34 13 96 93 28 4 5 78 75 JO 4 Assignment NH+ NH 2 + NH 3 + HCl+ AsH+ AsCl+ AsCl(iso)+ AsC1 2 + AsClCl(iso)+ AsC1 2 (iso)+ H 2 AsAsH 2 + H 5 As 2 + Asc1 3 + AsC1 2 Cl(iso)+ AsC1Cl 2 (iso)+ Asc1 3 (iso)+ 103

PAGE 113

Table 17 (cont'd) 189 191 4.1 5.2 5 7 Assuming that the black powder was elemental arsenic, a 77 % yield of arsenic was obtained based upon the amount of chloramine put into the reaction. The Reaction of Dimethylchloramine with Arsine .+ 2As + Dimethylchloramine (0.51 ml, 6.5 mmole ) in 80 ml of diethyl ether was condensed onto arsine (4.33 mmole ) in a reaction flask attached to the vacuum line. When the reaction 0 mixture was warmed to 0 C, a black solid precipitated immediately and a silver-colored mirror formed on the inside glass surface of the reaction flask. A finely divided black powder and a chloroform-soluble material were obtained when the reaction product was extracted with chloroform. Addition of the chloroform solution to diethylether gave a white precipitate. The melting point (169-171c) and infrared and proton magnetic resonance spectral data of the white precipitate were identical with those for pure dimethylamine hydrochloride. Anal. of the black powder. Found: C, 1.66; H, 0.J2; As, 93.11. Calculated for elemental arsenic: As, 100.0. Assuming that the black powder is elemental arsenic, a yield of 66% of theory based upon the amount of dimethylchloramine 104

PAGE 114

put into the reaction was obtained. The above analyses support this ass umption. The Reaction of Chlorine with Arsine in the Presence of Triethylamine.+ + 2As + To a reaction flask inside the dry-box was added 100 ml of diethylether. The flask and contents were then degassed on the vacuum line. Onto the diethylether was distilled 105 5 ml of degassed and dried triethylamine. Onto the amine -ether solution wa s condensed 654 ml of ars ine (8 mmole) at 225 mm/Hg and 296K. The solution was warme d and mixed thoroughly. Onto this was condensed 654 ml of gaseous chlorine (8 mmole) at 225 mm/Hg and 296K. The reaction mixture was th e n warmed to room temperature and stirred continually. As the solution warmed, the reaction mixture continually darkened in color (pale yellow, yellow, yellow-brown, brown, dark brown). The r2action mixture was stirred for 21 hours at ro o m temperature. After 21 hours all condensible materials were distilled from the reaction flask. The remaining dark-brown solid residue was taken into the dry-box. The solid was washed three times with 5 ml portions of diethylether. The yellow-colored ether wa shing solutions were evaporated; leaving a yellow residue. A proton magnetic resonance spectrum (Figure 37) of the yellow residue was obtained using c 6 n 6 as the solvent and tetramethylsilane as the internal standard. Two pe ak s having the 1"" values 8.56r

PAGE 116

and 9.78~ were obtained. These peaks were a~tributed to hydrogen atoms attached directly to arsenic in some lower condensed arsenic hydrides. The dark-brown, solid, reaction product was placed in a vacuum sublimation apparatus and heated to 100c at 10-3 mm/Hg. A large quantity of white sublimate and a black-brown residue were obtained. The sublimate was recrystallized from acetonitrile to give white crystals having a melting point (252-254c) and infrared spectrum identical to those of pure triethylamine hydrochloride. Yield of triethylamine hydrochloride: 0.41 g (19% of theory based upon the amount of chlorine put into the reaction). Assuming that the black-brown residue from the sublimation process was ele m ental arsenic, a yield of 0.08 g (20 % of theory based upon the amount of chlorine put into the reaction) was obtained. The analyses support this assumption. Anal. Found: C, 2.94; H, 0.49; N, 0.28; As, 92.10. The Reaction of Ammonia-free Chloramine with Methylarsine.5 CH 3 AsH 2 + 5NH 2 Cl ( CH 3 As) 5 + 5NH 4 Cl Methylarsine (654 ml at 225 mm/Hg and 2 96K, 8 mmole) was condensed into a reaction flask equipped with a magnetic stirring bar. Onto this was distilled JO ml of a 0.134 M ammonia-free chloramine solution to which was added 107

PAGE 117

70 ml of diethylether. As soon as the diethyrether melted, a white solid began to form. The reaction mixture was stirred for 16 hours at room temperature. The solution wa s somewhat yellow-orange in color with a trace of red-brown solid on the glass at the surface of the reaction solution. The highly volatile materials were distilled off the reaction mixture. The remaining yellow-orange solid was taken into the dry-box and extracted with diethylether to give a white solid and a bright-yellow solution. The white solid was dried in vacuo. The bright-yellow diethyl ether washings were collected and all the diethylether was distilled from the solution. The remaining material was a bright-yellow, highly-viscous liquid. The infrared and proton magnetic resonance spectra of this yellow liquid were identical with those of pure cyclo-pentamethyl pentaarsine (Figure 29 and Figure 30). Yield of cyclo-pentamethylpentaarsine: 0. 33 g ( 94 % of theory based upon the chloramine put into the reaction). The infrared spectrum of the ether-insoluble white solid was identical with that of pure ammonium chloride. Yield of ammonium chloride: 0.25 g (100 % of theory based upon the amount of chloramine put into the reaction). 108 The Reaction of Dimethylchloramine with Methylarsine.+ ..... + Diethylether (50 ml) was pipetted into the reaction flask inside the dry-b~x. The flask was then attached to the vacuum line and the solution degassed. Onto the diethylether

PAGE 118

was condensed methylarsine (656 ml at 340 mm/H~ and 296K, 12.1 mmole). The diethylether solution was warmed to obtain an ether-methylarsine solution. Onto this solution was condensed 7.5 ml of a 0.72 M dimethylchloramine (5.33 m~ole) solution in diethylether. The reaction was warmed to room temperature. 0 Below O C a very large quantity of white solid was formed. The reaction mixture was stirred for 24 hours. The reaction flask contained a large quantity of a pale-yellow solid. In the uppe r portion of the flask there was a small quantity of a red solid. All the highly volatile materials were distilled from the reaction flask. The remaining tacky, pale-yellow solid 109 was extracted several times with small portions of acetonitrile to give a clear colorless solution and an immiscible yellow oil containing traces of a red solid. The clear colorless solution was added to diethylether to give a white solid having a melting point (169-170c), infrared spectrum, and proton magnetic resonance spectrum identical with those of pure dimethylamine hydrochloride. Yield of dimethylamine hydrochloride: 0.42 g (96 % of theory based upon the amount of dimethylchloramine put into the reaction). Diethylether was added to the immiscible oil to give a clear yellow solution and a trace amount of red solid. The diethylether was distilled from the solution. The residue was a bright-yellow, highly-viscous liquid having infrared and proton magnetic resonance spectra

PAGE 119

identic a l with those of cyclo-pentamethylpentqarsine (Fi gu re 29 and Figure JO). Yield of cyclo-pentamethyl pentaarsine: 0.40 g (85 % of theory based upon the amount of dimethylchloramine put into the reaction). The Reaction of Chlorine with Methylarsine in the Presence of Triethylamine.+ + + red solid Onto 100 ml of diethylether in a reaction flask attached 110 to the vacu um line was distilled 10 ml of dried triethylamine and methylarsine (1137 ml at 200 mm/Hg ani 296K, 12.J mmole). Onto this solution was condensed, with constant stirring at -76c, gas eous chlorine (653 ml at 175 mm/Hg and 296c, 6.2 mmo le). The clear colorless solution immediately turned to a yellow color and a pale-yellow solid formed. The reaction mixture was stirred for one hour at -76c as the solution gradually darkened in color. The highly volatile components we re then distilled from the flask. The orange residue was washed several times with diethylether. Upon distilling the diethylether from the washings, a yellow-brown oil was obtained which immediately be ga n to darken in color. A proton magnetic resonance spectrum (Figure JS) was obtained using c 6 D 6 as the solvent and tetramethylsilane as the internal standard. The spectrum showed in the region of -110 cps to -40 cps the presence of at least 14 different types of magnetically non-equivalent protons. This is the region for cH 3 As and AsH proton peaks. This oil eventually

PAGE 121

changed to a brick-red solid. The orange residue was extracted with 50 ml portions of boiling acetonitrile until no white precipitate resulted when the acetonitrile solution was added to diethylether. An acetonitrile-insoluble, brick-red solid (I) was obtained and was dried in vacuo. The melting point data for the brick-red solid were as follows: 161-164c, solid changed from a red to a deep purple color with some apparent decomposition; up to 254c, a liquid which readily oxidized in the air and had an arsine odor distilled from the purple solid. The infrared spectrum of this solid was completely void of absorption peaks. No proton magnetic resonance spectrum could be obtained due to the insolubility of the red solid. Mass spectra of the red solid were obtained at temperatures below and above its deco~position point. The mass spectral data are given in Table 18 and Table 19. m/e 91 150 197 255 TABLE 18 MASS SPEC'I'RAL DATA FOR RED SOLID AT 150c Intensity 24 43 10 6 Relative Intensity 56 100 23 14 Assignment 112

PAGE 122

Table 18 (cont'd) 270 290 304 397 7 12 14 14 16 28 33 33 TABLE 19 MASS SPECTRAL DATA FOR RED SO LID AT 190c m/ e Intensity 75 10 89 8 90 26 91 10 103 24 105 77 150 19 151 8 163 22 165 24 195 8 225 40 255 73 270 82 Relative Intensity 12 10 32 12 29 94 23 10 27 29 10 49 89 100 Ass ign me nt As+ CH 3 As+ (CH 3 ) 2 As+ As 2 + CH 3 As 2 + (CH3)3As2+ As 3 + (CH 3 ) 2 As 3 + (CH 3 ) 3 As 3 + 113

PAGE 123

Table 19 (cont'd) 300 450 16 10 20 12 (CH 3 ) 5 As 3 + (CH 3 As) 5 + 114 A comparison of the mass spectral data of the red solid 1 i sted in Table 19 with th ose obtained of ( CH 3 As) 5 ( Table 14) shows that the pentamethylpentaarsine is formed during the heating of the red solid. Anal. of (I). Found: C, 6.66; H, 1.49; As, 89.47; Cl, 0.05. This gives the empirical formula CH 2 67 As 2 14 for (I). Yield of (I): 0.2 g. A destructive distillation of (I) gave a purpleblack solid residue and a small amount of yellow distillate. The proton magnetic resonance spectrum of the yellow distillate was identical with that of pure cyclo-pentamethyl pentaarsine. No yield data w e re obtained for the cyclo-pentamethylpentaarsine. The acetonitrile solutions were added to diethylether to give a white solid having a melting point (253-254c) and an infrared spectrum identical with those of pure triethylamine hydrochloride. Yield of triethylamine hydrochloride: 1.0 g. The Reaction of Ammonia-free Chloramine with Dimethylarsine.2 ( CH 3 ) 2 AsH + NH 2 Cl -+ ( CHJ) 2 AsAs ( CH 3 ) 2 + NH 4 Cl Into the reaction flask on the vacuum line was distilled 28 ml of dimethylarsine (32 mmole). Onto this was

PAGE 124

condensed 87 ml of a 0.026 M chloramine (2.22 ~mole ) solution in diethylether. The reaction mixture was warmed to room temperature and stirred vigorously for 10 hours. A considerable amount of white solid formed immediately. All highly volatile materials were distilled from the reaction flask at -23c (carbon tetrachloride-liquid nitrogen slush) on the vacuum line. This distillate contained the excess dimethylarsine. From the reaction flask at room temperature was distilled a liquid of low volatility. The infrared spectrum of the liquid was identical with that of tetramethyldiarsine (Figure 34). Yield of tetramethyldiarsine: 0.44 g ( 96 % of theory based upon the amount of chloramine put into the reaction). The solid remaining in the reaction flask was orange-yellow in color. This solid was extracted with warm tetrachloroethane to give an insoluble white solid and a clear solution. The infrared spectrum of the white solid was identical with that of ammonium chloride. Yield of ammonium chloride: 0.12 g (100 % of theory based upon the amount of chloramine put into the reaction). Approximately 0.04 g of an orange-colored solid was obtained when the tetrachloroethane solution was added to diethylether. When the reaction was repeated using a dimethylarsine to chloramine ratio of 13.7 mmole:5.1 mmole, only a 10% yield of tetramethyldiarsine was obtained. The major product (0.14 g) was the tetrachloroethane 115

PAGE 125

116 soluble, yellow-orange colored solid. The melting _point data for the solid were: 70c, solid began t~ contract; 102c, solid melted with vigorous bubbling; by 150c, bright-orange liquid. The infrared spectrum is shown in Figure 39. The proton magnetic resonance spectrum (Figure 40) was obtained using d 6 -DMSO as the solvent and the sodium salt of 3-(trimethylsilyl)-propanesulfonic acid as the internal standard. Peak A (8.22~) refers to the methyl protons attached to arsenic, while peak B (2.53~) could refer to NH 2 protons. Both the infrared and nnr spectra are very comparable to those obtained for 8cH 3 ) 2 AsN] 4 HCl (Figure 19). Anal. Found: C, 18.31; H, 5.01; N 7.21; As, 51.92; Cl, 20.3. This gives the empirical formula c 3 H 9 8 As 1 4 Ncl 1 .~. The Reaction of an Ammonia-Chloramine Mixture with Dimethylarsine.+ + ~cH 3 ) 2 As N ] 4 HCl + 6NH 4 c1 To a reaction flask equipped with a large cold finger, an adapter for attaching the flask to the chloramine generator, and an adapter for connecting to the vacuum line (Fi gure 16) was added 100 ml of 20-4o 0 c boiling petroleum ether. The flask and contents were degassed on the vacuum line and 2.3 ml of dimethylarsine (26 mmole) was condensed onto the petroleum ether. The reaction flask and contents were attached to the chloramine generator and the effluent gases from the generator were bubbled through the petroleum ether solution for 45 minutes (approximately 75 mmole of

PAGE 126

300 0 2600 1400 1200 1000 800 600 400 -1 cm Fig. 39.-Infrared Spectrum of As-N Product from (cH 3 ) 2 As H -NH 2 Cl Reaction. (Mull)

PAGE 127

A Fig. 40.-N. M. R. Spectrum of As-N Product from (c H 3 ) 2 As H NH 2 Cl Reaction.

PAGE 128

chloramine) while the temperature of the reac~ion solution was maintained at -76c to minimize the loss of the volatile dimethylarsine. A white solid formed immediately. After a few minutes, the white solid disappeared and a clear, colorless dense layer formed which was immiscible with the petroleum ether. After the chloramination was completed, the reaction mixture was warmed to room temperature under a stream of dry nitrogen. As the ammonia evolved, the immiscible lower layer slow ly turned to a white solid. The reaction flask was attached to the vacuum line and al l condensible materials were distilled from the reaction flask. The r emaining tacky, white solid was taken into the dry-box and extracted with tetrachloro ethane. The infrared spectrum of the remaining white solid was identical with that of ammonium chloride. The tetrachloroethane solutions were added to diethylether. The resulting white precipitate was recrystallized from acetonitri le. The infrared spectrum of this white solid was similar to that of ((cH 3 ) 2 As N ] 4 HCl -1 0 except for an absorption peak at 970 cm Mp 170-185 C with decomposition. (Melting point previously observed for ((cH 3 ) 2 As N ] 4 HC l was 217-218c with decomposition.) This sample possibly contains a mixture of some other arsenic nitrogen compoun1s as a minor impurity. This would account for the broad and low melting point and the additional -1 infrared absorption peak at 970 cm The product could 119 consist of an equimolar mixture of ~cH 3 ) 2 As N ] 4 HCl and r

PAGE 129

((cH 3 ) 2 AsN] 4 in addition to ((cH 3 ) 2 As N ] 4 HC1._ The nmr spectrum in deuterium oxide and the analysis of such a mixture would be identical with those of ((cH 3 ) 2 AsN] 4 HCl, while the infrared spectrum and melting point would be different. Anal. Found: C, 17.67; H, 4.70; N, 10.26; As, 54.56; Cl, 13.37. Calculated for [(cH 3 ) 2 As N ] 4 HCl: C, 17.49; H, 4.78; N, 10.21; As, 54.60; Cl, 12.91. Yield of [(cH 3 ) 2 AsN] 4 HCl: 1.1 g (36 % of theory based upon the amount of dimethylarsine put into the reaction). 120 The Reaction of Dimethylchloramine w i th Dimethylarsine.+ (CH 3 ) 2 NH 2 Cl Dimethylarsine (4.0 ml, 45 mmole) was distilled into a reaction flask on the vacuum line. Onto this was condensed 80 ml of a 0.17 M dimethylchloramine (14 mmole) solution in diethylether. The reaction mixture was warmed to o 0 c and stirred at this temperature for 1 hours. Within JO minutes a white solid formed. The reaction flask was then warmed to room temperature and stirred for 24 hours. All the volatile materials were distilled from the reaction flask through a series of traps at -196c and -79c. The least volatile components were trapped at -79c as a white solid. The material obtained at -79c was purified by fractional distillation on the vacuum line to give a solid melting between -10c and o 0 c and having infrared and proton magnetic resonance spectra identical to those of tetramethyldiarsine (Figure 34 and

PAGE 130

Figure 35). Yield of tetramethyldiarsine: 1.0 g (45% of theory based upon the amount of dimethylchloramine put into the reaction). The solid remaining in the reaction flask was recrystallized from acetonitrile to give white needles having a melting point (169.5-171c), an infrared spectrum, and a proton magn etic resonance spectrum identical with those for dimethylamine hydrochloride. Yield of dimethyl amine hydrochloride: 0.52 g (46 % of theory based upon the amount of di methy lchlora m ine put into the reaction). The Reaction of Chlorine with Dimethylarsine in the Presence of Triethylamine.+ + + 2(CH 3 cH 2 ) 3 NHCl Onto 70 ml of diethylether in a reaction flask on the vacuum line was distilled 3.3 ml of dimethylarsine (37.6 mmole) and 10 ml of triethylamine. The temperature of the reaction vessel was maintained at -76c and the solution was vigorously stirred while gaseous chlorine (655 ml at 280 mm/Hg and 296K, 9.9 mmole) was condensed into the reaction mixture. A white solid was formed immediately as the chlorine came into contact with the ether solution. The reaction mixture was warmed to room temperature and stirred for 2 hours. The highly volatile components were then distilled from the reaction flask on the vacuum line. The least volatile materials 121

PAGE 131

were trapped in a U-tube trap at -23c and we~e further purified by fractional distillation to give pure tetramethyldiarsine (confirm ed by infr a red spectroscopy). Yield of tetramethyldiarsine: 1.02 g (49 % of theory based upon the amo unt of chlorine put into the reaction). The white solid remaining in the r eact ion flask was confirmed by melting point and infrared spectral data to be triethylamine hydrochloride. Yield of triethylamine hydrochloride: 1.9 8 g (73 % of theory based upon the amount of chlorine put into the reaction). The Reaction of Ammonia -free Chloramine with Phenylar.sine.6c6H5AsH2 + 6NH2Cl ..... ( C6H5As) 6 + 6NH4 Cl Phenylarsine (1.5 g, 10 mmole) was dissolved in 25 ml of diethylether in a be ake r inside a nitrogen-filled glove bag To this solution was added, with stirring, 37.3 ml of a 0.134 M am:n Jnia-free chloramine (5 mmole) solution in diethylether. The r ea ction proceeded i mmediate ly with the for mat ion of a pale-yellow solid. The reaction mixture was filtered and the solid was washed several times with diethylether and dried in vacuo. Extraction of the solid with boiling benzene yielded a pale-yellow solid upon cooling of the benzene solution. A b enzene -insoluble white solid was also obtained. The infrar ed spectrum of the benzene recryst a llized material was identical with that obtained for pure arsenobenzene (Fi gur e 36). Mp 215-216c (Li teraturel 31 ) melting point varies from 195c to 212c). 122

PAGE 132

Yield: 0.30 g of recrystallized solid (39 % oz theory based upon the amount of chloramine put into the reaction). 123 Infrared spectroscopy showed that the benze ne -insoluble white solid was ammonium chloride. Yield of ammonium chloride: 0.25 g (93 % of theory based upon the amount of chloramine put into the reaction). The Reaction of Dimethylchloramine with Phenylarsine.+ + To 50 ml of diethylether containing 1.65 g of phe nyla rsine (11 mmo le) was added, with stirring, 5 ml of a 0.71 M dimethylchloramine solution (3.55 m~ole) in diethylether. A pale-yellow precipitate formed immediately. The reaction mixture was stirred for 2 hours and then filtered. The pale yellow solid obtained was washed several tim e s with diethylether. The solid was extracted several times with boiling benzene to give a benzene-insoluble white solid an d a benze ne solution from which a pale-yellow solid crystallized upon cooling. The infrared spectrum was identical with that of pure arsenobenzene (Figure 36). Mp 222-223c. Yield of arsenobenzene: 0.48 g of crude material (90 % of theory b as ed upon the amount of dimethyl chloramine put into the reaction). The benzene-insoluble white solid was recrystallized from boiling acetonitrile to give white needles. The melting point (169-171c), infr a red spectrum, and proton magnetic resonance spectrum were identical with those of pure dimethylamine hydrochloride. Yield of dimethylamine

PAGE 133

hydrochloride: 0.2 6 g (90% of theory based upon the amount of dimethylchlor am ine put into the reaction). The Reaction of Chlorine with Phenylarsine in the Presence of Triethylamine.+ + + 12(CHJCH 2 ) 3 NHC1 Phenylarsine (1.53 g, 10 mmole) and 75 ml of diethylether we re added to a reaction flask inside the dry-box. The solution was then degassed on the vacuum line. Onto the solution was distilled 5 ml of triethylamine. Gaseo us chlorine (654 ml at 225 mm/Hg and 296K, 8 mmo le) wa s then condensed onto this reaction solution. The reaction mixture was warmed to room temperature. At the melting point of diethylether, a yellow solid began to form. Below o 0 c, the reaction solution contained such a quantity of pale-yellow solid that stirring W3.S very inefficient and a very bright-yellow materia l formed on the surface of the reaction solution. This bright-yellow material could indicate futther chlorination of the initial products. A web-like mater ial formed above the surface of the re act ion mixture. A white so lid was also formed in the vapors above the solution. A fter one hour, all condensible materials were distilled from the reaction flask. The bright-yellow solid remaining in the flask was extracted with bJiling benzene to give a bright yellow insoluble solid and a benzene solution from which 124

PAGE 134

crystallized a pale-yellow solid upon cooling~ The infrared spectrum of this pale-yellow solid was identical with that of arsenobenzene. Mp 214-216c. Yield of arsenobenzene: 0.8 g (73% of theory based upon the amount of chlorine put into the reaction). The remaining bright-yellow solid was extracted with acetonitrile to give an insoluble bright-yellow solid and an acetonitrile solution. Upon addition of diethylether to the acetonitrile solution, a large quantity of a white precipitate resulted. The melting point (252-254c), infrared spectrum, and proton magnetic resonance spectrum of this white solid were identical with those for pure triethylamine hydrochloride. Yield of triethylamine hydrochloride: 2.0 g (96 % of theory based upon the amount of chlorine put into the reaction). An infrared spectrum of the acetonitrile-insoluble, bright-yellow solid showed only the peaks for arsenobenzene. The broad melting point range (211-220C) indicated the presence of an impurity in this suspected arsenobenzene. Upon addition of silver nitrate to a nitric acid solution of this solid, trace amounts of a white precipitate resulted. This indicated the presence of chlorine in the yellow solid, probably arising from further chlorina tion of the arsenobenzene formed. The impurity must be less than 10% by weight of this yellow solid or peaks other than those assigned to arsenobenzene would be observed in the infrared spectrum. 125

PAGE 135

The Reaction of Am:no~ia-free Chloramine. with Diphenylarsine .2(C 6 H 5 ) 2 AsH + NH 2 Cl (c 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 + NH 4 c1 Onto 2.J g of diphenylarsine (10 mmole ) in 10 ml of diethyl ether in a reaction flask on the vacuum line was condensed 70 ml of a O. 053 M amm-Jnia -free chloramine solution (3.7 mmole) The reaction mixture was warmed to room temperature and stirred for 3 hours. A white solid formed when the diethylether melted. The condensible materials were distilled from the reaction product The solid was taken into the dry-box and washed with boiling benzene to give an insoluble white solid and a clea~ colorless benzene solution. When the benzene solution was evaporated to dryness, a pale -yellow solid was obtained. The infrared spectrum (Figure 41) was identical with that of pure tetraphenyldiarsine. Mp 127-129c (Lit./ 22 ) 0 0 ) 1 JO C, 1JO 5 C The proton magnetic resonance 0 mp 120-125 C; spectrum (Figure 42) was obtained of the tetraphenyldiarsine using c 6 D 6 as the solvent and tetramethylsilane as the internal standard. The average 1' values of the two types of phenyl peaks were 2. 54 1' and 2. 96 -r. A mass spectrum of tetraphenyldiarsine was obtained using an inlet temperature of 120c. The mass spectral data are given in Table 20. 126

PAGE 136

3000 1600 1400 1200 1000 8 00 600 400 -1 cm p N --..J

PAGE 137

._. N co

PAGE 138

m/e Intensity 51 16 77 14 151 18 152 38 154 103 155 17 229 20 231 14 266 14 306 10 458 6 TABLE 20 Relative Intensity 16 14 17 37 100 17 19 14 14 10 6 Assignment c 6 H 5 As+ c 6 H 5 AsH 2 + (c 6 H 5 ) 2 As+ (C 6 H 5 ) 2 AsH 2 + (c 6 H 5 ) 3 As+ (C6 H5 )4As2+ The Reaction of Dimethylchloramine w ith Diphenyl arsine.2(C 6 H 5 ) 2 AsH + (CH 3 ) 2 NC1 -+ (c 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 + (CH 3 ) 2 NH 2 Cl Onto 2.3 g of diphenylarsine (10 mmole) in a reaction flask on the vacuum line was condensed 100 ml of a 0.0336 M dimethylchloramine solution (3.36 mmole) in diethylether. The reaction was warmed to room temperature and stirred 129

PAGE 139

overnight. A white solid formed below o 0 c. All the highly volatile materials were then distilled from the reaction flask on the vacuum line. The remaining off-white colored solid was washed with 20-4o 0 c boiling petroleum ether to remove the excess diphenylarsine. The solid was then extracted with benzene to give a benzene-insoluble white solid and a benzene solution. When the benzene solution was evaporated to dryness an off-white-colored solid was obtained. The infrared and proton magnetic resonance spectra were identical with those of tetraphenyldiarsine. Mp 128-1J0C (Lit/ 22 ) mp 120-125c, 130c, 0 130.5 C). Yield of tetraphenyldiarsine: 1.Jl g (87% of theory based upon the dimethylchloramine put into the reaction). The melting point, infrared spectrum, and proton magnetic resonance spectrum of the benzene-insoluble solid were identical with those of dimethylamine hydro chloride. Yield of dimethylamine hydrochloride: 0.27 g (100% of theory based upon the amount of chloramine put into the reaction). The Reaction of Chlorine with Diphenylarsine in the Presence of Triethy_lamine.+ + -+ (C 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 2(CHJCH 2 ) 3 NHC1 Diphenylarsine (2.J g, 10 mmole) was added to JO ml of diethylether in a reaction flask inside the dry-box. The reaction flask and contents were degassed on the vacuum lJO +

PAGE 140

line. Onto the diethylether-diphenylarsine so~ution was distilled 2.8 ml of triethylamine. Onto this solution was condensed, with stirring, gaseous chlorine (654 ml at 225 m~/Hg and 296K, 8 mmole). At the melting point of the diethylether, a white solid began to form. When the temperature of the reaction solution was approximately 0 O C, a very large quantity of pale-yellow solid was present. After one hour, all condensible materials were distilled from the reaction flask. The tacky solid when extracted three times with 25 ml portions of benzene yielded a benzene solution and an insoluble wh ite solid. The benzene solution was evaporated to dryness to give a tack~ off-white-colored solid. This solid was washed with 20 ml of acetoni trile in a Buchner funnel. The sol id remaining was very powdery and pale yellow in color. The melting point, infrared spectrum, and proton magnetic resonance spectrum were identical with those obtained for tetraphenyldiarsine. Yield of tetraphenyldiarsine: 1.45 g (63% of theory based upon the amount of diphenylarsine put into the reaction). The acetonitrile filtrate was bri gh t yellow in color and upon evaporation yielded a tacky, off-white colored solid. This material contained some tetraphenyl diarsine in addition to a small amount of liquid. No complete separation could be made of the components present in this material, because of the high solubility of the diarsine in the liquid. The acetonitrile solution had a 131

PAGE 141

very sharp odor characteristic of substituted ~hloroarsines. The benzene-insoluble solid was dissolved in JO ml of warm acetonitrile. When the acetonitrile solution was added to diethylether, a large quantity of wh ite crystals we re obtained. The me lting point and infrared spectrum were identical w ith those of triethylamine hydrochloride. Yield of triethylamine hydrochloride: 1.J g (100 % of theory based upon the amount of diphenylarsine put into the reaction). Discussion The Chloramine and Dimethylchloramine Reactions.The results of this study demonstrate the importance of chloramine and dimethylchloramine as synthetic intermediate in converting arsine primary arsinesi and secondary arsines to elemental arsenic, cyclic polyarsines, and diarsines, re spec ti vely. In all c ases w~1ere the arsine was kept in major stoichiometric excess, essentially a qua~titative yield of the respective arsenic-to-arsenic bonded compound was obtained. A summary of the chloramine and dimethyl chloramine re actions with arsines containing As-H bonds is given in Table 21. 132 All of the previous reported reactions using chloramine and dimethylchloramine have demonstrated their use as oxidizing agents with Group V bases. They have been reported to perform an electrophilic attack on the most basic site, thereby participating in an SN 2 type of bimolecular reaction

PAGE 142

TABLE 21 SUMMARY OF CHLORAMINE AND DIMETHYLCHLORAMINE REACTIONS WITH ARSINES CONTAINING As-H BONDS zRJ AsH + x{R 2 NC1 (RJ As) + XZR'NH Cl -X X -X Z 2 2 2 RJ-xAsH R 2 NC1 Mole Ratio (RJ As) Yield R 2 NH 2 Cl Yield Other Yield X -x z Products AsHJ NH 2 Cl 1: 1.5 As 77 % N H 4 Cl 77 % AsHJ (CH 3 ) 2 NC1 1: 1.5 As 66 % (CH 3 ) 2 NH 2 Cl 66 % CH 3 AsH 2 NH 2 Cl 2:1 (CH 3 As) 5 94 % NH 4 Cl 100 % CH 3 AsH 2 (CH 3 ) 2 NC1 2.J:1 (CH 3 As) 5 85 % (CH 3 ) 2 NH 2 Cl 96 % (CH 3 ) 2 AsH NH 2 Cl 14.5:1 (CH 3 ) 2 AsAs(CH 3 ) 2 96 % NH 4 Cl 100 % (cH 3 ) 2 AsH NH 2 Cl 2.7:1 (CH 3 ) 2 AsAs(CH 3 ) 2 10 % NH 4 c1 100 % [ ( CH 3 ) 2 AsN] 4 11 8. 6% (cH 3 ) 2 AsH NH 2 Cl + NHJ 1:28.8 none NH 4 Cl 100% [( CHJ) 2 AsN] 4 11 36% (CH 3 ) 2 AsH (CH 3 ) 2 NC1 J.2:1 (CH 3 ) 2 AsAs(CH 3 ) 2 45 % (CH 3 ) 2 NH 2 Cl 46 % c 6 H 5 AsH 2 NH 2 Cl 2:1 (c 6 H 5 As) 6 39 % NH 4 Cl 93 % c 6 H 5 AsH 2 (CH 3 ) 2 NC1 J.1:1 (c 6 H 5 As) 6 90 % (CH 3 ) 2 NH 2 Cl 90 % (C6H 5 ) 2 AsH NH 2 Cl 2.7:1 (c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 92 % NH 4 c1 100 % (c 6 H 5 ) 2 AsH (CH 3 ) 2 NC1 2.9:1 (c 6 H 5 ) 2 AsAs(c 6 H 5 ) 2 87 % (CH 3 ) 2 NH 2 Cl 100 % f--' 'v.) 'v.)

PAGE 143

mechanism with ultimate breakage of the N-Cl bond to form hydrazinium, aminophosphonium, or aminoarsonium salts. (32,33,34,35,36,37,38,39) + [ R M : : Cl] 3 R' l [R 3 MNR 2 ] Cl R = alkyl, aryl; R' = H, CH 3 ; M = P, As The reactions of chloramine and dimethylchloramine with arsines containing As-H bonds zR 3 AsH -X X + xzR'NCl 2 2 -+ (R 3 As) -x z + xzR'NH Cl 2 2 2 demonstrate that the chloramine molecules do not react according to the previously supposed reaction mechanism for the trisubstituted phosphines and arsines. Considering the above mechanism, the initial products expected from the arsine, primary arsine, and secondary arsine reactions would be of the type [HR 2 AsNR 2 ]c1. These species would possibly not be stable because of the lability of the As-H bond. Even if decomposition occurred, at least some type of As-N compound would be expected. Breakage of the As-N bani would not be expected since previous experimental results with trialkylor triaryl-substi tuted phosphines and arsines have always given products, even after secondary reactions have occurred, 134 in which an As-N bond is maintained. When the arsenic hydrogen compound is maintained in an excess, the products of the chloramination reaction definitely contain no As-N linkages.

PAGE 144

The analytical results on the black soltd formed and the mass spectrum (Table 17) of the sublimate from the arsine-chloramine reaction show a complete lack of As-N containing species. The mass spectrum shows the presence of ammonium chloride, arsenic hydrides, and chloroarsines in the sublimate. The chloroarsines could result from the interaction of HCl with the arsenk hydrides in the mass spectrometer at a temperature of 150c or could be inter mediate reaction products. The relative intensities of the various Asel: ion species are approximately those expected from theory if only the percentage natural abundance of c1 35 (75%) and c1 37 (25%) are taken into consideration. The results are shown in Table 22. The results of dimethylarsine-chlor am i ne reactions indicate that the mole ratio of dimethylarsine to chloramine is the determining factor for the type of products obtained (Table 21). When the ratio is 14.5:1, the tetramethyldi arsine is the principal arsenic containing product. As the ratio of chloramine is increased, the yield of the diarsine decreases an d another arsenic-containing material results. This species c onta ins not only C, H, and As, but also N and Cl. The infr are d spectrum of the ne~ chloramina tion product shows many of the peaks characteristic of the [(cH 3 ) 2 AsN] 4 HCl spectrum (Figure 19), although the analyses do not conform to that of the arsenonitrilic 135

PAGE 145

TABLE 22 RELATIVE INTENSITIES OF VARIOUS AsCl+ ION SPECIES X m/e 110 112 145 147 149 180 182 184 186 Relative Intensity Found 100 37 100 96 29 100 96 39 5 Relative Intensity Theory 100 30 100 67 11 100 100 33 4 Ass i gnmen t AsCl+ AsCl(iso)+ AsC1 2 + AsClCl(iso )+ AsC1 2 (iso)+ Asc1 3 + AsC1 2 Cl(iso)+ AsC1Cl 2 (iso )+ Asc1 3 (iso)+ Therefore, analytical results and the infrared spectrum suggest that the chloramination product contains As-N 136 bonds. When dimethylarsine was chlora~inated on the generator, in the presence of am m onia and excess chloramine, the only arsenic-containing product recovered was the arsenonitrile, [(cH 3 ) 2 AsN ] 4 2HCl. Arsenic-nitrogen compounds are formed only when chloramine is in a higher concentration or in excess in the reaction. These species may result from the chlorarnination reaction with the diarsine in a manner analogous to that with the diphosphine. ( 4 o) Therefore,

PAGE 146

by analogy the overall reactions occurring in the presence of excess chloramine can be postu l ated to be 4(cH 3 ) 2 AsH + 2NH 2 C l 2(CH 3 ) 2 AsAs (CH 3 ) 2 + 2NH 4 c1 2(CH 3 ) 2 AsAs ( CH 3 ) 2 + 2NH 2 C l 2(CH 3 ) 2 AsNH 2 + 2 ( CH 3 ) 2 AsCl 2(CH 3 ) 2 AsCl + 2NHJ 2(CH 3 ) 2 AsNH 2 + 2HC 1 2HC1 + 2NHJ 2NH 4 Cl 4 ( CHJ ) 2 AsNH 2 + 4NH 2 Cl _. 4 [ ( CHJ ) 2 As ( NH 2 ) 2 ] Cl 4 [ ( CHJ ) 2 As ( NH 2 ) 2 ) Cl 2 [ ( CH ) As ~~ N ~ As ( CH ) ] Cl + 3 2NH NH 3 2 2 2 2 N H 4 Cl 2 [ (CH3 ) 2As ~~~ N ~ As ( CH3 )2] c1 NH 2 NH 2 (( cH 3 ) 2 AsN ] 4 + 2NH 4 c1 ---4 ( CH J)2 AsH + 8NH2Cl + 4NH3 (( cH3 )2 As N ]4 + 4NH 4Cl When the analogous reaction is carried out in the abs ence of arn~onia the following species would be expected to be present i n the reaction mixture in add ition to (( cH 3 ) 2 As ~ N ~ As ( CH 3 ) 2 ]c1. These additional speci es Cl Cl would result from the reaction s [ ( CHJ ) 2 AsNH 2 ] Cl Cl The emp irical formula of the product of the dimethyl arsine chloramine ( ammonia -f ree ) reaction probably is representative of a mixture consisting of [( cH 3 ) 2 AsN ] 4 HC l, 137

PAGE 147

condensed species. This WQUld explain the observed ratio of N : As :Cl of 1:1.4:1.1 instead of the expected 1:1:0 or 1:1:0.5 for [(cH 3 ) 2 As N ] 4 or (( cH 3 ) 2 AsN ] 4 HCl, respectively. The results of this study dem)nstrate that removal of the hydrogen from the arsenic in the course of chlora m ination does occur. The reaction of arsine with chloramine gives a white precipitate which during 24 hours time, appears to u nd ergo further reaction to give elementa l arsenic, higher arsenic hydrides, and a mm onium chloride. The reactions of the phenyl substituted arsines with chlor amine and dimethylchloramine occur very rapidly at room temperature and go to completion within a few minutes. The lability of the As-H bo~d is a function of the extent to which the hydrogen nucleus is shielded by the e lectrons in its im~ediate environment. The more deshielded the hydrogen nucleus, the more labile the As-H bond. The relative shielding of the hydrogen nuclei in the various arsines is sh)wa by the chemical shifts of these nuclei in their respective proton magnetic resonance spectra. These chemical shifts are given in Table 2.3. This suggests that the As-H bond in diphenylarsine is the most labile. Its reactions with chloramine and dimethylchloramine appear to proceed the fastest in comparison w ith those of the other arsines. 1.38

PAGE 148

TABLE 23 CHEMICAL SHIFTS OF HYDROGEN NUCLEI IN THE As-H BONDS OF VARIOUS ARSINES Arsine Chemical Shift,~ AsH 3 8.78 CH 3 AsH 2 7.93 (ave.) (cH 3 ) 2 AsH 7.57 (ave.) c 6 H 5 AsH 2 6.44 (C 6 H 5 ) 2 AsH 5.12 The reaction of dimethylchloramine with arsine to give elemental ar3enic occurs at a visibly faster rate at a given temperature and comparable concentration than does the analo g ous chloramine reaction. These results are the reverse of those observed in the reactions of chloramine and dimethylchlor a mine with the trialkylarsi n es. Since dimethylchl oramine is known to be a poorer electrophile 139 than chloramine, this inconsistency in qualitative reaction rate observations suggests that the trialkylarsine and arsine reactions proceed by two different mechanis m s or that, in one of the reactions another rate-determining step occurs. The same observations are made for the primary and secondary arsine reactions. The possible mech ~ nisms are discussed below. In the trialkylarsine reactions, the first step and

PAGE 149

also the rate determining step may be the for mat ion of the tetracoordinate transition state or a pentacoordinate arsenic intermediate. Path 1: The arsenic performs a nucleophilic attack on the chloramine molecule + to give tetracoordinate arsenic. The tetracoordinate transition state dissociates to give the respective aminoarso n ium chloride. [R3As ... NR2 ... c1] [R3AsNR2] Cl Path 2: The arsine reacts with the chlora m ine molecule + (I) (II) to give an initial product containing pentacoordinate arsenic. The pentacoordinate intermediate dissociates to give the respective aminoarsonium chloride. No experimental evidence has been presented to eliminate this possible reaction mechanism for the trisubstituted phos phines and arsines. In the case of the primary and secondary arsines, there are several possible mechanisms. Path J: The arsenic performs a nucleophilic attack on the chloramine molecule H >AsH + R2NC1 [>As. NR2 c1] to give a tetracoordinate arsenic species. The species 140

PAGE 150

then dissociates to give an aminoarsonium chloride. H [> A s NR 2 ... Cl] The end-product of this reaction w~uld probably contain an arsenic-nitrogen bond. Breakage of the As-N bond would not be expected from previous experimental results. Similar reactions of trialkyl or triaryl-substituted phosp~ines and arsines with chloramine, even after condensation reactions have occurred, have always given products in which an As-N bond is ma intained. Since no arsenic-nitrogen compounds a re obtained in this study, this reaction path appears to be of minor importance. Path 4: The arsine reacts with the chloramine molecule to give a pentacoordinate arsenic intermediate. [>HAs(NcR12l > AsH + R 2 NC1 (III) (a) This intermediate may ionize to give the aminoarsonium chloride. H -+ [> AsNR 2 ] Cl The end-product again would probably contain an arsenic nitrogen bond. No arsenic-nitrogen compounds are obtained in this study. This path, therefore appears to be of minor importance. (b) The intermediate may undergo a disproportionation reaction. The hydrogen bonded to the arsenic migrates as a proton to the chlorine, followed by the breakage of the As-Cl bond to give an aminoarsine and 141

PAGE 151

hydrogen chloride. This aminoarsine then reac-ts with the primary or secondary arsine to give an arsenic arsen ic bonded compound and the respective amine. The hydrogen chloride reacts with the amine to g ive the respective amine hydrochloride. )As( [ NB'] H Cl >As NR 2 + HCl > AsNR 2 + > AsH -. >A sAs < + HCl + R'NH 2 R'NH 2 (IV) (V) This mechanism provides an explanation for obtaining the products observed in the reactions carried out in this study. Reaction (IV) is doubtful, since the hydrogen bonded to the arsen ic in the pentacoordinate intermediate should migrate to the strongest Lewis base attached to the arsenic. The strongest Lewis base in this case should be the -N(CH 3 ) 2 or -NH 2 group The reaction of an aminoars ine with a primary or secondary arsine to g iv e an arsenic-arseni c bJnded compound has not been reported in the literature or attempted in this laboratory. Consequently the poss ibility that react ion ( V) occurs is uncertain. (c) The pentacoordinate intermediate may undergo a disproportionation reaction in which the hydrogen bonded to the arsenic migrates as a proton to the amino g r oup followed by the breaking of the As-N bond to give a ch loroarsine and the respective amine. This chloroarsine then reacts with the primary or secondary arsine to give hydrogen chloride and an arsenic-arsenic bonded compound. The hydrogen chloride 142

PAGE 152

reacts with the a mine to give the respective qmine hydr chlo ride. [> ~<:2 1 -+ [> A s < ::R 2] > AsC l + R NH 2 (VI) 1 4 J '> As C l + > AsH [> AsAs (+ C l -1 -+ > AsAs < + HC l (VII) H HC l + R NH 2 -+ R 2 NH 2 C l The hydro g en b o nde d to the a r sen i c shou l d m i gra t e to t he strongest Lew i s base attached to the arsine v iz. t he N ( CH 3 ) 2 o r NH 2 group Therefore it wou l d seem re a sona ble that re a ction ( V I) may occu r. The re ac tion s R 2 AsC l 6c 6 H 5 As I 2 + + R 2 AsAsR 2 + -+ ( c 6 H 5 As ) 6 HC l + 12H I are very we ll estab l ished in the literature as prev i ous ly discussed Othe r react i ons i nvolv i ng ha l oars i nes a nd hyd r ogen su b st i t u te d ars i nes t o f o r m arsen i c a rsen ic ( J 1 0 14 1 5) bonded compounds a re we ll known T he r efor e r eact i on ( V II) i s we ll estab li shed Of t he mechanism d i scussed Path 4 ( b ) and Path 4 (c) best expla i n the experimenta l resu l ts Path 4 (c) is f avore d because o f the ab i l i ty to f i nd known react i on s to support portions 0f this postulated mechan i s m. Whe n a pr i mary or secondary arsine is r eacted w i t h c h l o r amine an d dimethylchlor a m i ne the react i on r ate ma y not depe n1 upon the rat e of react i on (I I I). S in c e th e phe n y lsubstitute d arsine react i ons occur a t a v i s ibly f aste r react i on r ate than th o se of the othe r As H compo u n d s

PAGE 153

144 this suggests that the rate determining step does not occur in reaction (III) formation of the pentacoordinate intermediate. (Note: This is demonstrated by the reaction of arsine with chloramine. A white solid forms immediately which, during 24 hours time, appears to slowly underg o further reaction to give elemental arsenic, arsenic hydrides, and ammon ium chloride.) Proton transfer is known to be a very r apid process. Therefore this step should not be rate determining. The rate determining step possibly is reaction (VII). This suggests that the rate a t which >AsH is used depends upon the reaction >AsCl + >AsH >AsAs( + HCl A thorough study of the reaction rates of the various arsine reactions with chloramine and dimethylchloramine should be carried out. This proposed mechanism Path 4(c) is shown below for the arsine, primary arsine, and secondary arsine reactions. Mechanism applied to arsine reaction: NR' AsHJ + R 2 NC1 -+ H 3 As( 2 Cl l H 2 AsCl + R NH 2 H 2 AsCl + AsHJ -+ [H2AsAsH;c1 -1 H H 2 AsAsH 2 + HCl HCl + R'NH 2 -+ R 2 NH 2 Cl

PAGE 154

+ HC l + + R NH 2 l H 2 AsAsH + R 2 NH Cl [ H2AsAsAsH;c1 -J H l Further analogous steps give higher arsenic hydrides, both linear and branched chains These arsenic hydrides decompose in the presence of heat and light to elemental arsen i c arsines and hyd r ogen arsenic hydrides R T l i gh t As + + Mechanism applied to primary arsine reactions : l l RAsAsR + HC l H H HC l + R NH 2 R 2 NH 2 C l + arseni c hydrides 14 5

PAGE 155

RAsAsR R 2 NC 1 [ R (RZ ] + -+ RAsAs H H Cl l R RAsAs ( + R NH H C l 2 R [ BAsAs~sH+C l -1 RAsAs ( + RAsH 2 H Cl H R H l RAsAsA s H + HC l H R R HC l + R NH 2 -+ R 2 NH 2 Cl Furthe r analogous steps give species o f the typ e RAs ( AsR ) AsR The thermodynam i c and kinetic consi d eratio ns H xC l o f the possible products deter~ i ne the v alue of x T h e conde n sed ch l oroarsine the n unde r goes an i nt r amo l ec ul a r reac t ion to g iv e the complete l y a r sen i c arseni c bond ed st r uctu r e RAs ( AsR ) AsR -t ( RAs ) + H Cl H xC l y Mechanism app l ied to the se c ondary arsin e react i on s: l + R NH 2 146

PAGE 156

HCl + R'NH 2 1 R 2 AsAsR 2 R 2 NH 2 Cl + HCl This proposed mechan ism is supported by the following. (1) Many pentacoordinate species of phospho rQs are known to exist and. to be thermally stable. Some pentacoordinate arsenic compounds are known but they are relatively few in numbe r. Since the arsenic covalent radius is l arge r than that of phosphorus, the momentary existence of a pentacoordinate arsenic species is very likely. (2) The pentacoordinate arsenic in termediate wou ld be expected to be thermodynamically unstable with respect to the chloroarsine and amine ( ammon ia or dimethylamine). Entropy factors alone would favor the reaction I NR l R 2 As( 2 H Cl + R'NH 2 The total number of particles in solution increases and one of the new species is a gas in this reaction. Also the amine formed can later react with hydro gen chloride to form the amine hydrochloride which is precipitated from the reaction. (J) Proton m i gration or transfer from the arsenic atom to the NR 2 group is very likely, since the strongest base or nucleophile attached to the arsenic is the NR 2 group, proton migration is knowa to be a very rapid process, and the nitrogen is more basic to 1 mrd protons 147

PAGE 157

than is arsenic. (4) Sin~e the reaction solu~ion contains a very large excess of the respective A s-H compound, the chloroarsine formed has a decidedly greater probability of encountering the As-H compound than the a~ine. (5) Reactions involving haloarsines and primary a n d second a ry arsines to 148 f b d d d are well known.(J,lO,l 4 ,l5) orm arsenic-arsenic one compoun s Therefore a similar reaction where R = H would be expected to occur. (6) The hydrogen chloride formed in a dilute solution of the arsenic-arsenic bonded compounds would react with the strongest base now present in the solution the amine to give the amine hydrochloride. (?) The species H 2 As(AsH)xAsH 2 and other higher hydrides formed would readily condense to elemental arsenic, evolve arsine, and give hydrogen at room temperature and in the presence of light. This is a known type of reaction for higher arsenic hydrides. ( 4 l) This reaction mechanism not only explains the results obtained from the current study, but easily explains the results of all chloramination reactio~s involving arsines and phosphines. Two possible paths of reaction are available after the for m ation of the pent a coordinate intermediate. Path I Path II + R'NNR' 2 2

PAGE 158

Whichever reaction path prevails depends upon the overall thermodyna m ics and kinetics of each competing reaction. In the trialkyl and triaryl phosphines and arsines, Path I appears to be the more favored. This possibly results from the added stability of the aminophospho nium or a~ino arsonium salt resulting from eff ectiv e d~-p-oback-bonding between the Group V atom an d the nitrogen atom. Reaction Path II explains the results of the reaction of the tris(dimethylamino)phosphine reaction w ith dimethylchlor amine(39) to give the tris(dimethyla~ino)chlorophosphonium chloride instead of the tetrakis(dimethylamino)phosphonium chloride. In the ana logous arsenic case, where the covalent radius of the arsenic is greater, the tetrakis(dimethylamino) arsonium chloride is obtained. The steric factors and lattice energy effects ma y make for ma tion of the tris(dimethylamino)chlorophosphonium chloride more favored than that of the tetrakis(dimethylamino)phosphonium chloride. Also the competition of the 4 lone pairs of electrons on the nitrogen atoms for the vacant Jd-orbitals on the phosphorus in a tetrakissa lt may decrease the added stability that should otherwise result from more effective d~-p~ back-bonding if there were fewer nitrogens bonded to the phosphorus as in the chlorophosphonium chloride. The absence of available low ene rgy orbitals on the nitrogen to form a pentacoordinate nitrogen intermediate, indicates that the proposed re ac tion mechanism would not 149

PAGE 159

hold for the chloramination of the amines. Instead the SN 2 type of bimolecular reaction mechanism whe re chloramine attacks the amine nitrogen with ultimate br eakag e of the N-Cl bond to form the hydrazinium salts explains these results best. The Chlorine Reactions.-The results of the reactions of arsine, primary arsines, and secondary arsines with chlorine demonstrate that the proposed reaction mechan ism also holds when an halo gen is substituted for chloramine. Th is study also shows that chlorine may be used in addition to chloramine, dimethylchloramine, bromine, and iodine as a synthetic intermediate in obtainingdiarsines and cyclic polyarsines from hydrogen-arsenic bonded compounds. Triethylamine was used as a hydrogen chloride acceptor in a manner analogous to the amine liberated in the chloramine reactions. A summary of these reactions is given in Table 24. In all cases arsenic-arsenic bonded products were obtained in substantial yields. In some instances, traces of arsenic-chlorine containing species were obtained. These compounds were probably the result of the observed localized reactions with gaseous chlorine occurring on the surface of the reaction solution. The results of this study do not agree with those reported by Dehn and co-workers. ( 5 ,lo) Their reactions of chlorine with primary and secondary arsines always yielded compounds having arsenic-chlorine bonds. However, it is clear that such products may have resulted because chlorine 150

PAGE 160

RJ AsH -X X AsHJ CH 3 AsH 2 (CH 3 ) 2 AsH c 6 H 5 AsH 2 (c 6 H 5 ) 2 AsH TABLE 24 SUMMARY OF CHLORINE REACTIONS WITH ARSINES CONTAINING As-H BONDS Mole Ratio RJ AsH :Cl 2 -X X 1: 1 2:1 J.8:1 1. 3: 1 1.J: 1 (RJ As) -x z As (CH 3 As) 5 (cH 3 ) 2 AsAs(CH 3 ) 2 (C 6 H 5 As) 6 (c 6 H 5 ) 2 AsAs(C 6 H 5 ) 2 Yield 20 % not obtained 49% 73% 63% xz(CHJCH 2 ) 3 NHC1 (CHJCH 2 ) 3 NHC1 Other Products Yield 19% higher arsenic hydrides 58 % CH2.67As2.14 73 % none 96% As-Cl materials (trace) 100 % As-Cl materials (trace)

PAGE 161

was in excess and no hydrogen chloride acceptor was present in these reaction. Consequently, the products were actually those expected from the chlorination of the diarsines and cyclic polyarsines. This is shown by the various reactions carried by B licke(l?) and Dehn(5) using iodine and phenylarsine in varying mole ratios 6c 6 H 5 AsH 2 + 612 (c 6 H 5 As) 6 + 12H1 2c 6 H 5 AsH 2 + 312 c 6 H 5 (1)AsAs(1)C 6 H 5 + 4H1 c 6 H 5 AsH 2 + 212 c 6 H 5 As1 2 + 2H1 The arsenic arsenic bonded reaction product obtained from the methylarsine -chlorine reaction was not the cyclo-pentamethylpentaarsine as expected, but a brick-red solid having the empirical formula CH 2 67 As 2 14 The high arsenic-to-carbon ratio implies that during the reaction a loss of methyl groups from the arsenic o~curred and that a highly condensed arsenic-arsenic bonded product resulted. That a highly condensed material formed is indicated by the inability of the compound to absorb li gh t in the infrared region. The condensed arsenic compound is undoubtedly a complex species, since prior to its final condensation its proton magnetic resonance spectrum (Figure 38) showed the presence of at least 14 magnetically nonequivalent methyl groups or protons. This material probably contains some of the red solid form of the cyclo-pentamethylpentaarsine~ 6 ) At 161-164c the brick-red solid decomposes to give a purple black solid and a liquid shown by nmr spectroscopy and mass 152

PAGE 162

spectrometry to be the cyclo-pentamethylpentaqrsine. The literature( 6 ) states that the red solid form of the cyclic polyarsine melts at 1Bo 0 c. An impurity present would probably lower the melting point. Therefore the dimethylarsine-chlorine reaction yields some cyclo-penta methylpentaarsine and other non-chlorine containing arsen ic compounds. N M R Spectra l Study of Cyclo-pentametb,ylpenta arsine.-The nm r spectra obtained of cyclo-pentamethylpenta arsine as a neat liquid and in deuterated chloroform 153 carbon tetrachloride, and hexadeuterobenzene solutions all show the existence of three type of magnetica lly nonequ ivalent methyl groups having the area ratios 2:2:1 in the compound. These studies indica te that the structure (Fi gure 43) of the compound in the liquid state is possibly the same as that observed by Burns and Waser( 42 ) in their x-ray crystal studies. One other structure also agrees with the nmr data (Fi gure 44). A molecular model of these structures shows that steric considerations favor the Burns and Waser structure. The average chemical shift of the methyl reson ance in various solvents is g iven in Table 25. The ave rage chemical shift observed is independent of the solvent when deuterated chloroform and carbon tetra chloride are used. When hexadeuterobenzene is used as the solvent, the average chemical shift of the methyl groups shifts upfield approximately o.13~ or 8 cps relative to

PAGE 163

Fig. 4J.B urns and W a ser's S truc t ure of (c H 3 As) 5 from X -r a y S tudies. Fig. 44.-0ther Possible S tructure of (c H 3 As) 5 154

PAGE 164

TABLE 25 AVERAGE CHEMICAL SHIFT OF METHYL GROUPS IN (CH 3 As) 5 IN VARIOUS SOLVENTS, AT ROOM TEMPERATURE Solvent Average Chemical Shift T none 8.36 CDClJ 8.35 cc1 4 8.35 C6D6 8.48 that observed in the previous solvents. This implies that 155 the methyl groups on the arsenic ring are more highly shielded in the benzene case. The nmr spectra of the neat sample, chloroform solution, and carbon tetrachloride solution are all identical in appearance. The area ratios of the three peaks, proceeding from the most downfield methyl to the most upfield methyl, is 2 :2:1. These results, in addition to the upfield shift when the solvent is changed to the hexadeuterobenzene, indicate that there exists an interaction between the benzene molecule and the cyclo -p entamethylpenta arsine molecule. A statement regarding the actual type of association occurring would be only speculative, althoug h an interaction of the ~-electron cloud in the benzene

PAGE 165

molecule with the vacant 4d-orbitals of the arsenic atoms is not unlikely. A model of the polyarsine shows that not all the arsenic atoms can be equal ly affected by the benzene molecule in any type of solvent-solute int eraction Possibly a uv or visible spectroscopy study would indicate the nature of this association. The res~lts of the temperature dependent nmr study of the neat polyarsine sample (Figure 32) indicate that since the average chemical shift is unchanged, there are no solvent-solute interactions or self-association. The most upfield and downfield methyl groups tend to move toward the midpoint very slightly (each moves approximately 0.5 cps) over the 140 degree temperature range studied. This slight temperature dependence possibly results from the electronic environment of the methyls becoming more nearly equal through possible inversion about the ring. The results of the temperature dependent nmr study of the hexadeuterobenzene solution of the polyarsine (Figure 33) indicate that, since the average chemical shift moves upfield 5 cps wi th a temperature decrease of 110c, there is solvent-solute interaction. Th e most upfield and downfield methyl group peaks each move approximate ly 1.3 cps toward the m idpoint over the tempera ture range studied. The midd le methyl group peak moves upfield 0.75 cps. The order of the area ratios of the methyl peaks remains 2:1:2 throughout the study. Figure 33 156

PAGE 166

shows that there is a tendency toward a change in the order of the peaks from 2 : 1 : 2 to 2 : 2 :1, as the temperature is increased, but within the temperature range studied this complete change does not occur. This may indicate that the tendency toward solvent-solute interaction decreases with increasing temperature Summar y Th is study de m onstrates a new type of reaction for chloramine and dimethylchloramine and shows their importance as synthetic reagents in converting secondar y and primary arsines to diarsines and cyclic polyarsines In all cases where the arsine is kept in large stoichiometric excess essentially a quantitative yield of the respective arsenic-arsenic bonded compound results. The proposed mechanism explains the experimental resul ts of these reactions involving the formation of arsenic arseni c bonds. This mechanism also explains the results of al l previous chloramine and dimethylchloramine reactions wi th tri-substituted phosphines and arsines 157 As the mole ratio of chloramine to dimethylarsine is i ncreased arsenic nitrogen compounds begin to form. Whe n chloramine actually is in excess and ammon i a is present, the dihydrochloride of the dimethylarsenonitrilic tetrame r results. This suggests that chloramine reacts with the tetramethyl

PAGE 167

diarsine in a manner analogous to that with t~tramethyl diphosphine. In contrast with the res ults previously reported in the literature for chlorine in excess, chlorine not in excess reacts with primary and secondary arsines to give the respective arsenic-arsenic bonded compounds as stable products. The results of similar reactions, discussed in the literature, are those resulting from a complete halogena tion of the primary or secondary arsine The temperature dependent nmr study of the cyclo pentamethylpentaarsine indicates that the structure of this compound is the same in both the liquid and the solid state. The magnetic environment of the methy ls become more equal through possible inversion of the ring. This study also shows that there is a definite solvent-solute interaction in a hexadeuterobenzene solution of the polyarsine. This interaction is temperature dependent. Bibliography 1. P. Pascal, ed ., Nouveau Traite' de Che m ie Min~ra le, Masson et Cie Paris, 1958, p 111. 2. J. V. Janowsky, Ber., Q, 219 (1 8 73). J. P. Pascal, .9.12. cit., p 95. 4. P. Pascal, ibi., p 96. 5. W. M. Dehn, Am. Chem. J., JJ., 101 (1905). 6. E.G. Rochow, D. T. Hurd, a'1d R. N. Lewis, The Chemistry of Organometallic Compounds John Wiley and Sons Inc., New York, 1957, p 210. 158

PAGE 168

7. M V. Auger Compt r end ., 1..J.., 1705 (1904). 8. C. S Palmer and A. B Scott, J. Am Chem Soc., 22,, 536 (1928). 9. A. J. Qu ick and R Adam s, J. Am Chem Soc ., 44, 809 (1922). 10. W. M. Dehn and B. B Wilcox Am. Chem J., J..5., 1 (1 906 ). 11. W. M Dehn Am. Chem J., 40, 123 (1908). 12 G T Morgan Organic Col1,Pounds of Arsenic and Antimony Longman's, Green and Co. New York 191 8 p 3-6. 13. 14. 15. 16. R. F. 315 F. F. w Bunsen F. Bl i cke (1933). F. B li cke F. Blicke Ann ., 42, and L. D. and L. D and J. F. 14 (1842). Powers, J. Am Chem Soc ., 2.2., Powers, ibi.9:., 2'.' 3353 (1932). Oneto, ibid., ..li.' 749 (1 9 35). 17. F. F. B li cke R A Patelski, and L D. Powers, ibid., 5.5., 1161 (1933). 18. 19. 20. 21. 22 23. 24. 25. 26. C. S. Palmer and R. Adams, ibid., 44 1356 (1922). E. G Rochow D. T. Hurd R N. Lewis, 2.12. cit., p 211. A. E. Goddar d, A Textbook of Inorganic Chemistrr Vol. XI Charles Griffin and Co ., London, England 1930, pp 62 -63. G. T. Morgan 2.12. cit., p 76. F. F. Bl i cke and F. D Sm ith, J. Am Chem. Soc., 21, 2272 (1929). K. A. Petrov V A Parshina, B A Orlov, and G. M Tsypina, J. Gen. Chem USSR ( Eng Trans.), J.,g_, 3944 (1962). R. C. Weast, Ed., Handbook of Chemistry and Physics, 45th edition, Chemical Rubber Co ., Cleveland Ohio, 196 4 p D-95, W. M. Dehn, Am. Qiem. J. pp 1 20 -122. W. M. Dehn and B B. Wilcox, 2.12. cit., p 45. 159

PAGE 169

27 A E Goddard QP. cit., pp 53-54. 28. E Wiberg and K MBdritze r, Z Naturforsch ., 1 2 b, No 2 127-128 (1 957 )~ 29. A. E Goddard cit>, p 67. JO. W Steinkopf and W Mieg Ber ., 21,, 1016 (1920). J1. A E Goddard ~i~-, p 339. 32. H H. Sis l e r, A Sar~is H S Ahuja R J. Drago and N L. Smith J. Am Chem Soc ., 81 2982 (1959). 33. w. A Hart an d H. H Sis ler, Inorg Chern ., J.' 317 (19 64 ). 34. D. F Clemens and H H Sisler ibid., !' 1222 (1 96 5). 35 s R J ain w. s Brey Jr. and H H. Sisle r, ibid., 2.' 515 (1 967 ). 36. H H Sisler and s R Jain, ibid. 7.' 104 (1968)~ 37. H J. Vetter and H NBth z Anorg Allgem Chem., JJ.Q, 233 (1 964 ). 38. H. H S i s l e r and c. Stratton Inorg. Chem., .5.' 200J (1 966 ). 39. S R Jain, L. K. Krannich R E Highsmith and H. H Sisle r, _i bj_d ., _, 105 3 ( 1967). 40. s E. Frazier Doctoral Dissertation, University of Florida, December 1965. 160 41. J. W Mellor A Comp rehensive Treatise on Inorgan ic an d Theoretical Che m istry Vo l. IX, Longman's, G r een an d Co., 1929 p 50. 42. J. H Burns and J Waser J Am Chem Soc ., Z.2, 8 59 (19.57).

PAGE 170

CHAPTER V GENERAL CONCLUSIONS AND SUMMARY In general, this study has shown that (1) chloramine and dimethylchloramine react with trialkylarsines to give the expected aminoarsonium chlorides; (2) qualitatively chloramine is a stronger electrophile than dimethylchloramine; (J) chloramine preferentially attacks the arsenic in its reactions with the phenarsazines; (4) an ammonia-chlora~ine mixture reacting with dimethylchloroarsine gives a substantial yield of the dihydrochloride of the dimethylarsenonitrile tetramer; (5) an ammonia-chloramine mixture reacts with dimethylarsine to give a hydrochloride salt of the dimethylarsenonitrile tetramer; and (6) chloramine and dimethylchloramine react with primary and secondary arsines to provide a novel synthesis of cyclic polyarsines and diarsines in high yields. The preparation of the dimethylarsenonitrile tetramer from the reaction ofan ammonia-chloramine mixture with dimethylchloroarsine and dimethylarsine has provided the first known synthetic method for obtaining alkyl arseno nitriles. In contrast to the analogous dimethylchloro phosphine reaction where pyrolysis of the chloramination product, [(cH 3 ) 2 P(NH 2 ) 2 ]c1, must be carried out to obtain 161

PAGE 171

the phosphonitrilic tetramer, the arsenonitrilic tetramer is obtained directly in the chloramination process of the dimethylchloroarsine. The inability to remove the HCl from the tetramer suggests that this compound is probably quite basic. The analogous (CH 3 ) 2 PC1 reaction with ammoni chloramine mixture does not give the hydrochloride salt. This possibly suggests that there is l ess &t-p~ bonding between the arsenic and nitrogen atoms than between the phosphorus and nitrogen atoms. Consequently the ~-bonding electrons may be localized more on the nitrogen atoms in the As.!...!...!..N bond. The reactions of chloramine and dimethylchloramine with the primary and secondary arsines demonstrate a new type of reaction for the chloramine molecules. This indicates their importance as synthetic intermediates in 6onverting secondary and primary arsines to diarsines and cyclic polyarsines, respectively. As a result of this study a mechanism was proposed whic h explains the experimental results of these arsine re act io ns and of all previous substituted phosp~ine and arsine reactions with chloramine and di methy lchlor amine The first step in this mechanism is the attack of chloramine on the arsenic to give a pentacoordinate arsenic intermediate. R 3 As + RZNCl --+ hAs(:: 2 1 162

PAGE 172

The path of the reaction now depends upon the nature of the arsine : (1) if a primary or secondary arsine, the pentacoordinate intermediate decomposes to the chloroarsine and the amine The chloroarsine reacts with the excess arsenic -hydro gen compound to g ive the arsenic -arsenic bonded compounds. + + R'NH 2 + HCl (2) if a trisubstituted ar3ine the pentacoordinate intermediate could decompose into either the aminoarsoniu m chloride or the chloroarsonium chloride. + R'NNR' 2 2 The thermodynamics, kinetics, and steric considerations of the respective reactions would determine the appropria te reaction. The temperature dependent nrnr study of the cyclo pentamethylpent a arsine indicated that the structure of this co~pound is the same in both the li quid and the solid state. The magnetic environment of the methyl became more equal through possible inversion of the ring with increased temperature. This study also indicated that there was a definite solvent-solute interaction in a hexadeuterobenzene solution of the polyarsine. This interaction was temperature dependent. 163

PAGE 173

BIOGRAPHICAL SKETCH Larry Kent Krannich was born September 5, 1942 at Pekin, Illinois. He was graduated from Pekin Community High School in Pekin, Illinois in June, 1960. He received the Bachelor of Science degree with a major in mathematics from Illinois State Normal University in June, 1963. From June, 1963, until April, 1965, he worked as a graduate teaching assistant while pursuing his graduate degree. In June, 1965, he received the M aster of Science degree with a major in Chemistry from Illinois Stat~ University. In April, 1965, he enrolled as a graduate stu dent at the University of Florida to pursue the Doctorate of Philosophy degree in chemistry. Mr. Krannich is a member of the American Chemical Society. 164

PAGE 174

This dissertation was prepared under the direction of the chairman of the candidate's supervisory committee and has been approved by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. June, 1968 Dean, Coll Sciences Dean, Graduate School Supervisory Committee: Ch~