Mem. Fac. Educ., Shimane Univ. (Nat. Sci.), Vol. 10. pp. 29 34. December 1976.

An ESR Study of Radicals Produced in 7'-Irradiated Ammonia-Boron Trifluoride, NH3BF3, and ND3BF3

Kunihisa SOGABE*

Abstract : ESR spectra of radicals produced in ammonia-boron trifluoride, NH3BF3, and ND8BF8 subjected to r-irradiated at 77 K were observed. From analyses of these spectra, the radicals were identified as a BF3-0r a NH8BF3-and a ･NHBF2. The latter radical is mterpreted to be prepared by a loss of a hydrogen from NH2BF2, which was produced by the decomposition of the NH3BF8･ ESR parameters of the ･NHBF2 radical were aF=26 G, aB=36G, aN=260 G, and aH=41 G, respectively. These results show that a dative bond of the NH3BF3 differs from that of the NH8BH3 in the bond strengh, although both bonds are stronger than a P-B dative bond in phoshine-boranes under ?'-irradiation

Introductiom that a radical produced by T-irradiation of diborane, B2H6, at 77 K was not BH3 radical Boron halides or boranes in which boron formed by a symmetrical cleavage of a B-B atorn is short of two electrons having complete bond but a 'B2H5 by an extraction of one of valence shell could produce many interesting two bridged hydrogens. Lovas and Johnson,9) addition products (adducts) with all molecules also, established that N-B bond in amino- containing an unshared electron pair. In par- fluoro borane, NH2BF2, produced during the ticular, ammonia-boron trifluoride (NH3BF3) gas phase reaction of NH3 and BF3 is fairly or ammonia-borane (NH3BH3) were recognized strong. These observations may suggest an to possess a nitrogen-boron dative bond and existence of an alternative nature to that a staggered conLiguration with C3v symmetry proposed by Armstrong and Perkins5) on by intensive studies on X-ray diffractions, 1'2) the N-B dative bond in the NH3BF3 adduct. NMR,3) IR spectra,4) or theoretical calculatio- Ammonia-borane is known to be isoelectric ns.5,6) and to possess the same configuration with Armstrong and Perkins5) recently reported C2H6 molecule, which is irradiated by T-rays that boron trifluoride would be transL0rmed to produce an ethyl radical. Nevertheless, the a planar configuration (D3h) into a pyramidal isoelectronic amine borane radicals are rec- form (C8v) on the NH3BF8 adduct formation and ieved little attention. The unpaired electron at this time, O . 23 electron would be transferred serves a measure of variations for the electron from NH3 to BF3' These adducts, therefore, distribution in an interesting radical and ESR seems to involve similar a weak dative bond parameters reflect the actual electronic envir- with phosphine-borane adducts in which a onments within this adduct. The NH3BF3, phosphorus - boron bond with incomplete together with the NH3 BH3, rs therefore sharing of electron pair were easily broken investigated in this paper for identifications to produce phosphoranyl radicals under T- of radicals and an evaluation of a nature on irradiation.7) the N-B dative bond. However, Hasegawa8) more recently reported Ex perimental * Laboratry of Chemistry, Faculty of Education, University of Shimane, IVlatue, 690 Japan. Ammonia- boron trifluoride ad duct was ・○ ヂ

、。．・

． 。 ．、、 ①雲－昌曇自98甲雨石O冒冒く冒勺8毛◎ξωH8唱吋串ち壱εω串ω国

’9 ●

θ’ ‘

一（、 α ） ／ at77K ・ ’ ● ■ 〆 ） ・1 ’

’“

、 ） 瓜 〆 ．。ノ ↓

‘■ ’ 1 ‘ ）

い ノ at77K

Fig. 1 ESR spectra of T-irradiated (a) polycrystallme of NH8BF3 (full line) 8 and ND3BF3 (dotted line) (b) single crystal of NH8BF3- Kunihisa SOGABE 31 prepared by the reaction between ammonia ones were annealed completely at ca. 170 K. and boron trifluoride etherate BF3 O (C2H5)2 A residual signal was in accord with that at -780C. ND3BF3 was prepared in a similar observed when the sample was irradiated at way by the use of deutro ammonia, ND8 room temperature. This signal is interpreted Deutro ammonia gas was made synthetically to result from a BF3- or NH3BF3- radical by the reaction of a freshly prepared mag- since these radicals may be produced by a nesium nitride with heavy water. NH3BH3 Was decomposition of this adduct under T-irradi- prepared from of NH4Cl and LiBH4 and was ation but discussion is put away in this separated from LiCl by filteration at - 780C. Io) paper. All obtained adducts were washed with An unstable radical species is of interest anhydrous ether and dried in vacuo for ca. and a detailed discussion will be developed 24hr. Single crystal of NH3BF3 Was prepared here. The central position of the outer two from anhydrous acetone. BF3 gas (99.70/0) was signals is placed at lower field than DDPH obtained commercially and purified before by ca. 20 G. This may be caused by effects use by a trap-to-trap fractionation. D20 of a second-order perturbation which are (99.50/0) and LiBH4 obtained from Merck often observed in the cases of a nucleus with Sharp & Dohme of Canada. Ether and acetone a nuclear spin more than I=1. In this case, were distilled and stored over sodium metal a B nucleus (1=3/2), a N nucleus (1=1), two A conventional high-vacuum line employing equivalent F nuclei (21=1), or two . equivalent gleaseless stopcocks was used to prevent H nuclei (21=1) are involved but the B nucleus decomposition of boron compounds is excluded because of being unable to find Sample was introduced in an ESR tube and up a quartet with a total separation of ca. 520 sealed off in vacuo. Then sample was irra- G in this spectrum. This spectrum is, there- diated with 60Co T-rays at 77 K for ca. 4.0hr fore, consisted of a triplet in which a central at dose rate of 106 rad/h. ESR spectra were component (MI =0) disappeared behind ~- measured at 77 K with no warm up and at radical with a relative intensity various temperature with standared variable An ESR spectrum of a r-irradiated ND3BF3 temperture attachments. The purity of the shown in Fig. I (b) was observed to determine sample was checked by means of powder the nuclear showing the effects of the second- X-ray diffraction and mass spectrometer.n) order perturbation. In addition to the central For comparison, X-ray diffraction powder intense signal due to the 7t-radical, the two pattern and ESR spectrum on ammonium components being subdivided to a quintet boro fluoride, NH4BF4 were observed were observed with a slight change in line width, while the coupling of the triplet Res~alts and Discussionu remained unchaged. It became clear to exclude An ESR spectrum of the T-irradiated NH3- the splitting due to a hydrogen nuclear from BF3 in polycrystalline observed at 77 K with the coupling of the triplet but to include in no warm up is shown in Fig. 1. The spectrum that of the quintet. An exchange of a hydro- generally consisted of three components, in gen to a deutrium resulted in only change which relative weak two outer components in line width, for which the auther deduced with symmetric quintet feauters have a separ- that a coupling for a hydrogen must be ca ation of ca. 5_90 G and a relative intense 40 G. Accordingly, the triplet are considered inner one has a total linewidth of ca. 200 G to be formed by either a nitrogen or two By raising temperature of the sample, the equivalent fluorine nuclei mner component was present to ca. 400 K To further identify this radical species, an without a change of features, while the outer ESR spectrum of a T-irradiated single crystal 32 ESR Study of Radicals Produced in Ammonia-Boron Trifluoride

H// a o 0 0 0 o 0 0 0 o o o 、o 0 Qo 0 o o 00 o o o 0 o 0 0 o o 0 ◎ 0 00 ◎ 0 0 0 o o 0⑰ 0 o 0 o 0 o ③0 o 00 0 ③ o o 0 000 0 o 0 o 0 0 000 0 o 000 o ◎ 0 o 0 ⑬ o o 0 o Hllb 0 ⑰ o 0 0 o o 0 o ③ o o ooo◎ 0000 0 000 ◎ o o◎ ◎o◎ o 0 o 0 q 0 0 o 0 00 o 0 0 000 o o o o o o00 0 0o 0 0 0 b 0 ◎ 00 0 00 00 0 0 0 00 0o o 0 0 0 0 0 00 0 0 0 0 oo③ o o 0◎ 0 ◎O000 0 ◎ ◎ 0 0 0 0 0 00 o o o 0 0 000000 0 o oo 0 ◎00 H// a -4 o

-8 o o H F4(GAllss) o +8 o Fig. 2 Angular dependeces about b axis of T-irradiated single crystal of NH8BF8 The parts only in the lower field side are shown of NH3BF3 Was similarly observed. This spe- Furthermore, figure shows to involve of a ctrum obtained at the direction of being quartet of three kinds of doublet and this parallel to b axis of single crystal is shown quartet is assumed to arise from a B nucleus in Fig. I (c). Angular dependences about b with a nuclear spin of I=3/2. A residual axis were measured and low field component nuclear is a nitrogen with a nuclear spin of only in the triplet is shown in Fig. 2. Ana- I=1 and the triplet having separation ca. 260 lyses of other two orthogonal axes were G is therefore considered to arise from the N unsuccessful. In Fig. 2, however, three nuclear. kinds of doublet were found. Two doublets From the analyses of the spectrum, the are composed of two nuclei having a nuclear radical obtaind here is interpreted to be spin of I=1/2, with a relative anisotropy, ･NHBF2 produced by a loss of a hydrogen while one doublet is attributed to a nuclear atom from NH2BF2 Which might be formed having a nuclear spin of I=1/2, with no by an elimination of HF from NH3BF8 under anisotropy and the coupling of ca. 40 G. The T-irradiation coupling of the hydrogen nuclear is ca. 40 G The s character of the nitrogen nuclear in and then the doublet with no anisotropy is the 'NHBF2 radical is therefore calculated to identified to arise from the hydrogen nuclear. be 470/0 . This result is particularly surprising Two doublets with anisotropy are therefore in view of the fact that this ra,dical contains determined to result from two fluorine nuclei. a sp hybrid on nitrogen. To judge the CY) CY) DDPH~

(a) at 7 7K

~

r~I~

O (/)O (b)

c:l c,) at 77K

S:F: ~f ~f l

(c)

at 77K

Fig. 3 ESR spectra of T-irradiated polycrystalline of (a) NH3BF3, (b) NH8BH8, and (c) NH4BF4 34 ESR Study of Radicals Produced in Ammonia-Boron Trifluoride

correctness of such an assignment of this This radical is of very interesting for the radical species, ESR spectra of T-irradiated radical having the sp hybrid ,on the nitrogen NEl:3BH3 and NH4BF4 in polycrystalline shown but is still not completely elucidated. Accor- in Fig. 3 were observed. Signal attributed to dingly, many experiments on single crystal the 'NHBF2 radical were also observed in of ND3BF3 and NH3BH3, together with the case of NH4BF4, but no signal attributed NH3BF3, and INDO calculations on another to a 'NHBH2 One was observed in NH3BH3 possible form of the 'NHBF. radical will be From these results, a NH3BH3- or other 7r- performed to perfectly identify this radical radical is interpreted to be mainly produced ACK.NO WLEDGMENT : Present auther is by T-irradiation of the NH8BH3 and to exhibit deeply grateful to Dr. A. Hasegawa and the si*anal shown in Fig. 3 (b). The NH3BH3- K. Takeda of Hiroshima University, and radical was postulated by Claxton et. al. ,12) Professer M. Miura of Okayama Science who more recently reported the formation College for thier valuable discussions and of the 'NMe2BH3 radical from stable adduct, continued cooperation NMe3BH3' The NH3BF3 adduct first dec.om- poses to NH2BF2 and then the 'NHBF2 R,eferences radical is produced by the extraction of the l) J. L. Hoard, S. Geller, and W. M. Cashin : hydrogen atom from the NH2BF2 under T- Acta. Cryst., 4, 396 (1951). 2) W. J. Jenkins : J. Amer. Chem. Soc., 78, 5500 irradiation, while no decomposition of the (1956) R. W. Parry, G. Kodama, and D . R I~EI3BH3 adduct proceeds to a NH2BH2' Schultz : J. Amer. Chem. Soc.. 80, 24 (1958) ; A. P. Caron, J. L. Ragle, M. E. Yorke, and H. The aminodifluoro borane was known to Yen : J. Phys. Chem., 72, 556 (1968). possess a planar structure9) and in order to 3) C. W. Heitsch : Inorg. Chem., 4, 1019 (1965) approach the experimental data, INDO-UHF J. B. Leane and R. E. Richards : Spectrochimica Acta., 10, 154 (1957). calculations on various planar configurations 4) J. Goubeau and H. Mitschelen : Z. Physik of the ' NHBF2 radical were carried out by Chem., 14, 61 (1958). 5) D. R. Armstrong and P. G. Perkins : Inorg using structure parameters reported by Lovas Chim. Acta., 10, 77 (1974) and Johnson.9) The s character of the N 6) A. A. Frost : Theoret. Chim. Acta 18, 156 nuclear caluculated by INDO-UHF method ( 1970). 7) K. Sogabe : J. Sci. Hiroshima Univ., Ser. A, 39, are 120/0 at the most in spite of many atte- 11 (1975). mpts and are made no comparison with those 8) A. Hasegawa and J. Sohma : Mol. Phys., 20, 118 (1972). estimated from the ESR spectrum 9) F. J. Lovas and D. R. Johnson : J. Chem. Phys., 59, 2347 (1973). F .. ' H s character of 12 olo 10) S. G. Shore and R. W. Parry : J. Amer. Chem. Soc., 80, 8 (1958). F /B p characterNi* of 81 olo ll) P. A. Barfield, M. F. Lappert, and J. Lee : J. Chem. Soc., A, 554, (1968) E. F. Rothgery, H A. Mcgee, Jr., and S. Pusatcioglu : Inorg Chem., 14, 2236 (1975). This radical is assumed to possess a stru- 12) T. A. Claxton, S. A. Claxton, S. A. Fieldhouse, R. E. Overill and M. C. R cture described below and so on, although Symons : Mol. Phys., 29, 1453 (1975) calculations are still not performed.

F, B/ ~~~~Ne

F/ ¥.+ ~H /