9092
Acknowledgments.
We wish to thank C. Herndon
and J. 0. Frohliger, ibid., 36, 3480 (1971)). I t is of particular interest to note that the lowest energy electronic transitions were all slightly red shifted upon increasing the polarity of the solvent (cyclohexane to ethanol, dim,, 5-6 nm). This contrasts with the blue shift observed in a related methyl ester, (CH3)3SiC02CH3 (ref 4b). (b) This work was presented at the 164th National Meeting of the American Chemical Society, New York, N. Y . , 1972, Abstract INOR 133.
Williams, Jr., for helpful discussions and D. F . Shriver and B. G. Ramsey for enlightening comments on the manuscript prior to submission for publication. ~ l We are grateful t0 the InStitUtO de Fisica, Universidade de S5o- Paul0 for free computer time and to the Fundac50 de Amparo Pesquisa de Estado de sgO Paulo for financial support.
~
~
'
Reactions between Nitrosylpentaammineruthenium(I1) and Hydroxylamine or Hydrazine. Formation of (Dinitrogen oxide)pentaammineruthenium(II) F. Bottomley" and Janice R. Crawford Contribution from the Department of Chemistry, University of New Brunswick, Fredericton, New Brunswick, Canada. Received M a y 30, 1972 Abstract: [ R U ~ ~ ( N H ~ ) ~ NHO2I0Xand ~ . hydroxylamine formed [ R U ~ I ( N H ~ ) ~ N ~(X O]X = ~C1, Br, I). The products showed a strong infrared band in the 1160-cm-1 region (vl of N20) and a weak band in the 2250-cm-l region ( v 3 of N20). The bromide salt was face-centered cubic, a = 10.32 A. [RU~I(NH~)~NO]X~ and hydrazine hydrate formed, at room temperature, [RuII(NH~)~N~]X~ only. At -23' or below a mixture of [RuII(NH~)~N~]X~ and [ R U I ~ ( N H ~ ) ~ Nwas ~ O ]obtained. X~ The same mixture was obtained from reaction of OH- with a solution of [RUII(NH~)~NO]I~. H2O and NzH& at room temperature. The properties of [RuII(NH3)5N20]*+ are reported and mechanisms for the reactions of [Ru1I(NHJsNO]3+ with hydroxylamine or hydrazine suggested.
N
ucleophilic attack at the nitrosyl ligand has been plexes as they formed was a necessity (see Experimental demonstrated for a variety of ruthenium comSection). Addition of a precipitating anion to increase plexes. Particularly noteworthy are the reactions yields of the more soluble salts gave products contamiof [ R ~ I ~ C l ( d a s ) ~ N (das 0 ] ~ += o-phenylenebisdimethylnated with [ R U I I ( N H ~ ) ~ N and ~ ] ~ +[ R U ~ ~ ~ ( N H ~ ) ~ O H ] ~ + . arsine) with hydrazine, forming [ R ~ ~ T l N ~ ( d a s and ) ~ ] , The stability of the N 2 0 complexes was dependent on with azide ion, forming [ R ~ ~ T l N ~ ( d a sN) ~2 0] , and the anion (I- > Br- > Cl-). The iodide salt was stable N2.2 A similar reaction was also found for [RuIIClfor several weeks in dry air, but the chloride salt deWe have in(dipy)zNO]z+(dipy = 2,2'-dipyridine). composed in a few days. The rate of decomposition vestigated reactions of nitrosylpentaammineruthemay be an inverse function of the sample purity, since nium(II), [ R U ' ~ ( N H ~ ) ~ N O ] ~and + , ~present -~ here the the very soluble chloride salt was difficult to obtain results of investigations of reactions of [ R U I I ( N H ~ ) ~ - pure. NO]3L with hydroxylamine and hydrazine which give The complexes had properties similar to [RuII(NH&(dinitrogen oxide)pentaammineruthenium(II), [RuIIN?O](BF& prepared from [ R U I ~ ( N H ~ ) ~2+H ~and O] (NH3)5N20]2+.NzO complexes of pentaamminerutheN20.12 The bromide and iodide salts evolved N 2 0 on nium(I1) were prepared previously from N 2 0 and oxidation with ceric(1V) and N 2 0 , N2, and H 2 0 on dry [RUII(NH~)~H~O]*+.~~-~? heating. N 2 0 , 8 2 2 ( + 5 % ; mean of six determinations) of the theoretical amount, was obtained on treatResults ment of the iodide salt with Ce4+. Treatment of the [ R u I I ( N H ~ ) ~ N O ]HzO X ~ . and hydroxylamine reacted complex prepared from [ R U I I ( N H ~ ) ~ ~ ~* N O ] Iand ~ HzO rapidly, at room temperature, forming pale yellow, 14NHzOH with Ce4+ gave 29N20 whose cracking patdiamagnetic [RU"(NH,)~N,O]X~(X = C1, Br, I). Betern in the mass spectrum showed almost exclusively cause of rapid decomposition, precipitation of the comI4NO+. It was concluded the product was [Ru"(NH3)j15N14NO]12(assuming N 2 0 is coordinated via ( 1 ) P. G. Douglas, K. D. Feltham, and H. G. Metzger, J . Amer. Chem. Soc., 93, 84 (1971); Chem. Commun., 889 (1970). nitrogen rather than oxygen (see below)). (2) P. G . Douglas and R. D . Feltham, J . Amer. Chem. Soc., 94, 5254 The infrared spectra of the complexes (Table I) (1972). Personal communication from Dr. R. D. Feltham. (3) F. J. Miller and T. J. Meyer, ibid., 93, 1294 (1971). showed a very strong band in the 1160-cm-' region (4) F. Bottomley and J. R. Crawford, J . Chem. Soc., Dalrorz Trans., (vl of N 2 0 ) and a very weak band in the 2250-cm-' 2145 (1972). region ( v 3 of N,O), as observed for [ R U ~ ~ ( N H ~ ) Z N Z O ] (5) F. Bottomley, ibid., in press. (6) F. Bottomley and J. R . Crawford, Chem. Commun., 200 (1971). (BF&.12 No bands were observed in the 700-400 (7) T. J. Meyer, J. B. Godwin, and N. Winterton, ibid., 872 (1970). cm-l region where metal-ammonia, metal-N20, or (8) J. B. Godwin and T. J. Meyer, Inorg. Chem., 10, 2150 (1971). v2 of N 2 0 might be expected. The frequencies of the (9) E. J. Baran and A. Miiller, Chem. Ber., 102, 3915 (1969). ( I O ) J. N. Armor and H. Taube, J . Amer. Chem. SOC.,91,6874 (1969). N 2 0 vibrations were anion dependent, and the intensity ( 1 1 ) J. N. Armor and H. Taube, Chem. Commun., 287 (1971). of v3(N20)was also markedly anion dependent (I- > (12) A. A. Diamantis and G. J. Sparrow, ibid., 819 (1970). Journal of the American Chemical Society 1 94:26
December 27, 1972
9093 Table I. Infrared Spectra of [RuII(NH&NzO]X~~
x = c1 3310 s, br 3180 s, br 2306 vw 2230 vw 1627 m 1273 s 1256 sh 1152 vs, br 798 m Q
w
-x X = Br 3305 3240 s, br 3180 2312 2236 vw 1612 1310 sh 1271 1157 792
I-----. 2 8 N ~ 0 15N14N0 Assignment =
3295 3233 3174 2352 2248 w 1618 1302 1273 1253 1175 779
Abbreviations: vs = very strong, s = strong, m = medium, weak, vw = very weak, sh = shoulder, br = broad.
=
drazine hydrate gave [ R U ~ ~ ( N H ~ ) ~l 5N ~only. ] C I ~A mixture of [RuII(NH~)~N#'+ and [ R U ' ~ ( N H ~ ) ~ N ~ O ] ~ + was obtained by two methods: (a) conducting the reaction between [ R U ' ~ ( N H ~ ) ~ N O ] X ~(X -H~ =OC1, Br, I) and hydrazine hydrate at temperatures from -40 to -23" and precipitating the mixed product within 30 sec, and (b) precipitating the products from solution as they formed by adding OH- to a saturated solution of N2H61zand [ R u ~ ~ ( N H ~ ) ~ NHzO O ] Iat~ .room temperature. When the low-temperature reaction was allowed to proceed for a longer period (approx 3 min) some [ R U ~ ~ ' ( N H ~ ) ~(X N ~=] C1, X ~I)~ was ~ obtained, in addition to the above products. The reaction between [Ru11(NH3)S15NO]13~ H 2 0 and 28N2H4,using method b gave a mixture containing [ R U ~ ~ ( N H ~ ) ~ and *~N~]I~ [ R u ~ ~ ( N H ~ ) ~ ~ ~as N ~determined ~ N O ] I ~from infrared spectroscopy (v3(N20)2220 cm-', vl(N20) 1165 cm-', and v(N2) 2075 cm-I and mass spectral analysis.
Br- > Cl-). Attempts to obtain Raman spectra of the complexes were unsuccessful because of decomposiDiscussion tion in the laser beam. The infrared spectrum of [ R U ~ ~ ( N H ~ ) ~ ~ had ~ N v3(N20) ~ ~ N O of ] Igreater ~ intensity (Dinitrogen oxide)pentaammineruthenium(II) Salts. than the analogous 28N20complex and Av3 was 25 The infrared spectra of the complexes in the NH3 ~ ) was ) . 15 cm-' (Av, = v3(28N20)- V ~ ( * ~ N ~Av, regions (Table I) were not unusual though no bands cm-l. The analogous values for gaseous 15N14N0 were observed between 700 and 400 cm-', where Ru-N are Av" = 21 cm-' and Avl = 15 cm-'.13 vibrations would be expected. The reason for this is The electronic spectra of air- or argon-saturated not clear. The N20 vibrations are of some interest. aqueous solutions of [ R U I ' ( N H ~ ) ~ N ~ Oshowed ]~+ one The v 1 vibration was shifted to lower energy on coorband, A,, 233 nm, E >lo3. Decomposition of such dination (gaseous N20, v1 = 1286 cm-' 1 3 ) and v j was solutions was rapid however. [ R U ~ ~ ~ ( N H ~ ) ~ O Hshifted ] * + to higher energy (v3 N 2 0 = 2224 cm-' In was produced in air-saturated solution and [RuIIaddition, marked changes in the relative intensity of (NHp)SH20]2+in argon-saturated solution; in both the bands were observed. For gaseous N 2 0 the relacases conversion being essentially quantitative within tive intensity falls in the order v 3 > v1 > v2 but for co4 min. Spectra of the bromide salt in N 2 0 saturated ordinated NzO, vl >> v 3 > v 2 (vz (589 cm-' in gaseous water taken within 1 min of mixing showed a band, N2OI3) was not observed for the complexes). FreA,, 234 nm, E 5.3 X lo3. Using the value of 7.0'O for quency shifts for v 1 and v 3 in the same direction as obthe equilibrium constant for the reaction served here were found in the infrared spectra of N 2 0 adsorbed on alkali-halide films, and on C Y - C ~ ~ O ~ , ' ~ [Ru"(NH~)~NzO]'+@ [ R U ~ ' ( N H ~ ) ~ H Z+ O ]NzO ~+ but in both cases the relative intensities were the same the measured extinction coefficient corresponds to a as for gaseous N 2 0 . The bands were assigned to NzO ~+. value of E of 1.9 X lo4 for [ R U I I ( N H ~ ) ~ N ~ O ]Armor adsorbed via the oxygen atom.l8,lg Bands for which and Taube obtained a value of (1.7 f. 0.2) X IO4 at both v1 and v 3 increased over the gaseous 'N20 values 238 nm by extrapolation from the above equilibrium. lo were assigned to N 2 0 adsorbed on a-Cr203 cia the The reason for the difference in A,, between this work nitrogen atom.Ig In the present work foul- modes of and the literature10s12is not clear. coordination of N 2 0 may be considered: N bonded, X-Ray powder photographs of the bromide salt 0 bonded, and N,O-dimer formation' (all involving (attempts to obtain single crystals were unsuccessful perpendicular N 2 0 coordination) and parallel coordidue to decomposition in solution) showed this salt to nation from the N 2 0 T bonds. The crystal symmetry effectively rules out N,O-dimer formation, and, while not be face-centered cubic, a = 10.32 A (uncalibrated film). conclusive, strongly favors a linear RuNNO group over The calculated density for four molecules per unit cell the bent RuONN or parallel situations. The linear is 2.36 and the observed 2.26 g/cm3 (by flotation in RuNNO group implies the N=N+-0canonical form CHC13.CHBr3). The cell dimension was found to b: predominates in coordinated N20, and this it; consistent the same as for [ R U ' I ( N H ~ ) ~ N ~(cf., ] B ~a~ = 10.41 A with the frequency shifts on coordination. From the from a single ~ r y s t a l ' ~for ) , which the calculated density infrared spectra it appears unlikely there is extensive 10.32 A) is 2.26 and observed 2.21 g/cm3. In(a = w-electron donation from ruthenium to N 2 0 , and this tensities of the powder lines from the N 2 0 and N2 complexes were closely similar (14 lines observed). (15) A. D. Allen, F. Bottomley, R. 0. Harris, V. P. Reinsalu, and [ R u " ( N H ~ ) ~ N ~ O reacted ] B ~ ~ with O2 and HBr giving C. V. Senoff, J . Amer. Chem. SOC.,89, 5595 (1967). an essentially quantitative yield of [ R U I " ( N H ~ ) ~ B ~ ] B ~ ~(16) . Borod'ko, et al.,17 quote values of v(N2) for [Ru1I(NH3)~1(N1~N]2098 ~ ~and N ] 2094 I ~ cm-', respectively. In With ammonia, in an argon atmosphere, [RUI'(NH&]~+ I2 and [ R u ~ ~ ( N H ~ ) ~ ~ofS N the present work we cannot be certain which nitrogen is attached to was obtained by isolation of the iodide salt (64 yield); ruthenium. The large difference between the value of' v(Nz) found no other product could be isolated from the reaction. here and that quoted by Borod'ko, et al., is probably due to the presence of [Ru"(NH~)~N~O]I~. At room temperature [ R u ' ~ ( N H ~ ) ~ N O ] C and I ~ hy(13) G. M. Begun and W. H. Fletcher, J . Chem. Phys., 28,414 (1958). (14) F. Bottomley and S. C. Nyburg, Acta Crysfallogr.,Secr. B, 24, 1289 (1968).
(17) Yu. G. Borod'ko, A. IC. Shilova, and A. E. Shilov, Russ. J . Phys. Chem., 44, 349 (1970). (18) Y . ICozirovski and M. Folman, Trans. Faraday S O C . ,65, 244 ( 1969). (19) A. Zecchina, L. Cerruti, and E. Borello, J . C a r d , 25, 55 (1972).
Bottomley, Cruwford
(Dinitrogen oxide)pentuummineruthenium(II)
9094
is borne out by the ease of displacement of N,O in solution. Mechanism of the Reaction between [Ru11(NH&N0]3+ and Hydroxylamine. [ R U I I ( N H ~ ) ~ N O ]and ~ + [RuII(NH3)aN02]+coexist in equilibrium in aqueous alkaline solution. Reductions of NO2- generally proceed through equilibrium amounts of NO+ ions. For example, the reaction between nitrous acid and hydroxylamine HNOz
+ NHzOH --+- NzO + 2H20
involves, as the first step, nitrosation of hydroxylamine by NO+ giving O N N H 2 0 H+. 2o By analogy the reaction of [Ru1I(NH3);N0l3+ with N H 2 0 H may be formulated as
+
[RuII(NH~);NO]~+ NHzOH
+
0
[Ru"(NH,);NNHzOH]~' + [ R u " ( N H ~ ) ~ N O N H O H ] ~H+ ~
+
[Ru"(NH3)b(NONHOH)12+ may be compared to the intermediate d2jHjFe(C0)2(CONHNH2) in the reaction of N2H4 and T - C ~ H ~ F ~ ( C O ) This, ~ + . by loss of NH3 and rearrangement, formed r-CjHbFe(C0)2N C 0 . 2 1 Such a rearrangement (with loss of H 2 0 ) can be written for [Ru"(NH&(NONHOH)]~+, but clearly predicts [ R U I ~ ( N H ~ ) ~ ~ ~and N OI4NHzOH ]~+ will give [Ru'I(NH~)~'~N~~NO]*+
dinated ligand, but whether NO+, N3-, or N,O was involved is unknown. Experimental Section Ruthenium trichloride hydrate was obtained from Johnson, Matthey and Mallory, Montreal. It was converted into [RuIII(NH3)6l3+z 3 and then to [ R U ~ ~ ( N H ~ ) ~ N O ] Xby ~ .the H Zmethods O~~ cited. l6NOwas purchased from Merck Sharp and Dohme. NH5OHBr, NH30HI, NzH6BrZ,and N z H ~ Iwere z prepared by recrystallizing NHsOHCl or N2H6ClZ from the appropriate hydrohalic acid. All other materials were reagent grade. (Dinitrogen oxide)pentaammineruthenium(II) Dihalide, [RIP(NH&NZO]X~ (X = C1, Br, I). To a solution of NH,OHBr (1.0 g) and [ R U ~ ~ ( N H ~ ) ~ N O ] B(0.094 ~ ~ . Hg)Z O in water (9 ml) were added 5 pellets (1.0 g) NaOH. A pale yellow precipitate of [ R U ~ ~ ( N H ~ ) ~ NzO]Br* formed as the NaOH dissolved, and as soon as NaOH dissolution was complete the precipitate was removed by filtration, washed with alcohol and ether, and dried Cz L;CICUOover P z O ~yield , 0.057 g, 74%. The chloride and iodide salts were prepared using the appropriate nitrosyl and hydroxylamine salts. Yields: chloride salt, 29%; iodide, 83 %. Analyses are collected in Table 11. Table 11. Analytical Data for [RuII(NH~)sNzO]XZ ---Calcd, H N
%--X
---Found, %--H N X
[RU"(NH,)SNZO]C~Z5.02 32.55 23.54 5.10 31 90 22.92 [RUII(NH~)SNZO]BTZ 3.87 25.13 40.97 3,87 25.87 41.04 [&~"(NHa)aNz0]Iz 3.12 20.26 52.43 3.14 20.84 52.10
+
[ R u " ( N H ~ ) ~ ~ ~ N O N H O+ H ] ~ [+R U " ( N H ~ ) ~ ' ~ N ~ ~ N O ]HZO Z+
Reaction of [ R U ~ ~ ( N H ~ ) ~ N with Z O ] Oz-HBr. B~~ [RulI(NH& NzOIB~Z (0.0105 g) was added t o an 0 2 saturated solution of HBr whereas the major product was [ R U ~ I ( N H ~ ) ~ ~ ~ N ~ * N(48%, O ] ~1+ml) . in water (20 ml). The mixture was gently heated, giving a bright yellow solution, which was diluted to 100 ml. Hughes and StedmanZ0suggest ONNHOH may give, 398 nm. B 1809; The electronic spectrum of the solution ,,A,( by proton transfer, hyponitrous acid, HON=NOH. 398 nm, t 192OZ5)indicated a 94% yield of [RulI1(NHB);lit., , ,A This corresponds (with loss of a proton) to [RuIIBr]Brz. Evaporation of the solution gave orange crystals of (NH3),N(-0)=NOH]+, which may be compared to [ R U ~ ~ ~ ( N H ~ )whose ~ B ~ ]spectra B ~ ~ , were in good agreement with the l i t e r a t ~ r e . A~ i~d~. ~ Calcd ~ for Br3H15NSRu: N, 16.44. [ R U ~ ~ C ~ ( ~ ~ S ) , N ( - O ) = N Nbelieved H ~ ] , to be intermeFound: N, 16.34. diate in the formation of [ R ~ ~ ' C l N ~ ( d afrom s ) ~ ] [RuIIReaction of [Ru11(NH3)5Nz01Brz with NH,. [ R U ~ ~ ( N H ~ ) ; N ~ O ] B ~ ~ C l ( d a ~ ) ~ N 0 ]and ~ + N2H4. [ R U I ' ( N H ~ ) ~ N ~ Oprob]~+ (0.0307 g) was added (with stirring) to degassed 0.880 NH3 ( 5 ml). ably is formed by an analogous route The resultant yellow solution was diluted to 100 ml with water. The electronic spectrum showed a band at 275 nm (lit. for [Rurl0 , , ,A 275 nm, e 6242i); intrusion of a high-energy tail (NH3)6I2+, into this region made accurate determination of an extinction co[RU"(NH~)S'~N=NOH]++ [ R U " ( N H ~ ) ~ ~ ~ N ~ * N OOH]~+ efficient impossible. Addition of potassium iodide to the solution gave [Rur1(NH3)6]I~ (0.0230 g, 64%), whose infrared spectrum was Attempts to isolate an intermediate by reaction of in good agreement with the literature.26 Aim[. Calcd for H1&[Ru'I(NH&NO]~+ with CH3NHOH, NH?OCH3, or N6Ru: N, 18.39. Found: N, 18.30. CH3NHOCH3 were unsuccessful. In each case the Reaction of [ R U ~ ~ ( N H ~ ) ~with N OHydrazine. ]~+ (a) With Hydraonly products were those expected for attack by OHzine Hydrate at Room Temperature. [ R u ~ ~ ( N H , ) ~ N O.HzO ]C~, on [ R U I I ( N H ~ ) ~ N O ] ~ + . ~ (0.124 g) was dissolved in water (2 ml) and hydrazine hydrate (85 %, Mechanism of the Reaction between [ R U I ~ ( N H ~ ) ~ N O ] 5~ ml) + added. The solution was stirred for 5 min and NHaCl added until precipitation of [RUII(NH~)~NZ]C~Z was complete. The preand Hydrazine. Using similar arguments to those for cipitate was collected by filtration, purified by reprecipitating twice the hydroxylamine reaction, the probable first interfrom water, washed with alcohol and ether, and dried ijz cucuo over mediate for the hydrazine reaction is [RuI1(NH3),NPz05, yield 0.045 g, 46%. The spectral properties of the product were in good agreement with the literature.l5 A d . Calcd for ONHNH2I2+. This may lose NH3 and rearrange to C12H15NiRu: H, 5.30; N, 34.38. Found: H, 5.61; N,34.12. [ R U I ' ( N H ~ ) ~ N ~or, O ] ~by~ proton loss, give [RuIIHydrazine Hydrate at -23". To a solution of (NH3)aN(-O)=NNH2]+ and then [ R u ~ ~ ( N H ~ ) ~ N ~hydrazine ](b) ~ +With , hydrate @Sir,, 4.5 ml) which had been cooled to -23" [ R U ~ ~ ( N H ~ ) ~ Nwas ~ O observed, ]*+ and, as predicted by (CCL-liquid nitrogen) was added rapidly an ice-cold solution of [ R U ~ ~ ( N H , ) ~ N O ] C(0.28 ~ ~ ~g) H~ inOwater (10 ml). The solution the rearrangement mechanism, [ R U I I ( N H ~l5N0I3+ )~ was stirred for 10 sec and NH4C1 added until precipitation was and *8N2H, gave [ R u I ~ ( N H & ' ~ N ~ ~ N O ][Rul'~+. The product was removed by filtration, washed with (NH3)5N3]+was not observed, though [ R U ~ I ~ ( N H ~ ) ~ N , ] complete. *alcohol + and ether, and dried in cucuo over P2OS,yield 0.19 g. The
+
was. The latter could arise from oxidation of [RuII(NH3)jN3]+ by hydrazine.22 [ R U " ( N H ~ ) ~ ~ ~ was N~]~+ (23) A. D. Allen, F. Bottomley, R. 0. Harris, V. P. Reinsalu, and obtained from [ R U I I ( N H ~l5N0l3+ )~ and 28N2H4. This C . V. Senoff, Itiorg. Sun., 12, 2 (1970). (24) J. N. Armor, H. A. Scheidegger, and H. Taube, J . Amer. Chem. indicates the source of [RU'I(NH~)~N?]~+ was a coor(20) M. N. Hughes and G. Stedman, J . Chenr. Soc., 2824 (1963). (21) R. J. Angelici and L Busetto, J . Amer. Chem. Sac., 91, 3197 (1969). (22) F. Bottomle), Can. J . Chem., 48, 351 (1970).
Journal ofthe American Chemical Society
Sac., 90, 5928 (1968).
(25) H . Hartmann and C. Buschbeck, Z . P h j , ~ Chem. . ( F r u ~ k f u raf m Muin), 11, 120 (1957). (26) A. D. Allen and C. V. Senoff, Can. J . Chem., 45, 1337 (1967). (27) 7. J. Meyer and H. Taube, Inorg. Chem., 7, 2369 (1968).
1 94:26 1 December 27, 1972
9095 iodide and bromide salts were prepared by addition of NH4Br or K I instead of NH,Cl. The products were shown by spectroscopy { vl(NzO), 1155 (Cl- salt), 1160 (Br-), 1175 cm-I (I-); v3(N20), 2235 (Br-), 2250 cm-I (I-), unobservable for C1- salt; ~ ( N z )2100 , (Cl-), 2110 (Br-), 2120 cm-1 (1-);16 vl(NzO) and ~ ( N zwere ) of comparable intensity, A,, 221 ([RuI1(NH3)jN~l2+) and 235 nm ([RUII(NH~)~N~O]~ the + ) ;latter band decreased rapidly and a new band appeared at 299 nm, due to [Rurrr(NH&OHlZ+28 and analysis (Table 111) to be mixtures of [RuI1(NH3)sNz]Xzand [RurrTable m. Analytical Data for Mixed [RUI’NH~)~NZO]X”RU’’(NH~)~N~]XI Product -Calcd, H
2N
X
-Found, H
z--N
X
[R~”(NHa)sNzlClz 5.30 34.38 24.86 33.80 23.55 IRuIYNHIhN21Brz 4.04 26.21 42,72 3.98 25.87 40,42 -,- ~~ R u I I ( N H ~ ) ~ N ~3.23 ] I ~ 20.95 54.22 3.23 21 ,62 52.70 5 The calculated percentages for [RuII(NH3)sNzO]X2are given in Table 11.
(NH3),NzO]Xz, with the latter complex comprising, at maximum, 7 0 z of the product. By a similar method, but stirring the cold solution for approximately 3 min before precipitating, brick red (Cl-) or purple (I-) salts were obtained. These salts, which showed paramagnetism, became pale yellow on setting aside. The infrared spectra showed a (28) J. A. Broomhead and L. A. P. Kane-Maguire, ibid., 8, 2124 (1969).
band at 2020 cm-’ in addition to those of [RuI1(NH3)sN~]X2and [Rur1(NH3)jN~O]X2.The electronic spectrum showed a band at 460 nm which decreased with time. These properties indicated the presence of [ R U ~ ~ ~ ( N H ~ ) ~ N ~ ] ~ + . ~ ~ (c) With N 2 H P and OH- at Room Temperature. To a solu~ . H Zg) O and N2H612(1.O g) in water tion of [ R U ~ ~ ( N H ~ ) ~ N O ] I(0.089 (7 ml) were added four pellets (0.8 g) of NaOH. The precipitate which formed on dissolution of the NaOH was removed by filtration, washed with alcohol and ether, and dried in uacuo over PzOS, yield 0.040 g. The product was a mixture of [RU~~(NH&NZO]IS and [Ru1I(NH3):N~]I2in similar proportion to that obtained by method b above. Electronic spectra were measured on a Hitachi-Perkin-Elmer EPS-3T instrument; infrared spectra (as Nujol or hexachlorobutadiene mulls between KBr plates) on a Beckman 1R 12 instrument. Mass spectra were obtained, at ionizing voltages of 30 and 70 V, on a Hitachi-Perkin-Elmer RMU-6D instrument. Magnetic moments were by the Gouy method. Microanalyses were by A. Bernhardt, West Germany, and Chemalytics, Tempe, Ariz.
Acknowledgments. We wish to thank Dr. L. T. Trembath, Department of Geology, University of New Brunswick, for assistance with the X-ray powder photographs, Dr. G . A. Ozin of the University of Toronto for some of the Raman experiments, and Mrs. G. Aarts for the mass spectra. Financial assistance from the National Research Council of Canada and the University of New Brunswick Research Fund and a generous loan of ruthenium salts by Johnson, Matthey and Mallory, Montreal, are gratefully acknowledged.
Ground States of Molecules. XIX.’ Carbene and Its Reactions’ Nicolae Bodor,a Michael J. S. Dewar,* and John S. Wasson
Contribution from the Department of Chemistry, The University of Texas at Austin, Austin, Texas 78712. Received April 15, 1972 Abstract: MINDO/Z calculations are reported for singlet and triplet carbene and for their reactions with methane and ethylene. The energies of the various states of carbene have been calculated as a function of HCH bond angle. The results lead to the prediction that for small bond angles the singlet state becomes progressively favored. This prediction has been tested by calculations for cyclopropylidene, cyclopropenylidene, and vinylidene.
T
he divalent carbon compounds known variously as methylenes or carbenes have long fascinated organic chemists,4 and numerous theoretical studies of them have been reported.5 It is now recognized that compounds of this kind can exist in both singlet and triplet states, and carbene itself has been shown@-8 to have a triplet ground state, in agreement with recent ab initio SCF calculation^.^ (1) Part XVIII: N . Bodor, M. J. S. Dewar, and D. H . Lo, J . Amer. Chem. SOC.,94,5303 (1972).
(2) This work was supported by the Air Force Office of Scientific Research through Contract F44620-71-C-0119 and the Robert A. Welch Foundation through Grant F-126. (3) Robert A. Welch Postdoctoral Fellow. (4) P. P. Gaspar and G . S. Hammond in “Carbene Chemistry,” W. Kirmse, Ed., Academic Press, New York, N. Y., 1964, p, 235. (5) For recent ab initio calculations and references to earlier work, see (a) J. F. Harrison and L. C. Allen, J. Amer. Chem. SOC.,91, 807 (1969); (b) C. F . Bender and H . F . Schaeffer 111, [bid., 92, 4984 (1970); (c) W. A . Lathan, W. J. Hehre, and J. A . Pople, ibid., 93, 808 (1971); (d) J. F. Harrison, ibid., 93, 4112 (1971); (e) S. Y. Chu, A. K. 0. Siu, and E. F. Hayes, ibid., 94,2969 (1972). (6) G. Herzberg, Proc. Roy. Soc., Ser. A,262,291(1961).
The reactions of singlet and triplet carbene have also aroused much interest, particular attention being paid to the mechanism and stereochemistry of insertion and addition reactions. Here the only detailed calculations are those recently reported by Hoffmann, e t al.;$ these, however, were carried out by a procedure (“extended Huckel” method) which is known to give very poor estimates of molecular energies. Here we report calculations for singlet and triplet carbene and their addition and insertion reactions, using a procedure (MIND0/2lo) which has been shown to (7) E. Wasserman, W. A. Yager, and V. J. Kuck, Chem. Phys. Lett., 7, 409 (1970); E. Wasserman, V. J. Kuck, R. S. Hutton, and W. A. Yager, J . Amer. Chem. SOC., 92,749 1(1970). (8) R . A. Bernheim, H . W. Bernard, P. S. Wang, L. S . Wood, and P. S. Skel1,J. Chem. Phys., 53,1280 (1970). (9) R. C. Dobson, D. M. Hayes, and R. Hoffmann, J . Amer. Chem. SOC.,93,6188 (1971). (10) (a) M. J. S. Dewar and E. Haselbach, ibid., 92, 590 (1970); (b) N. Bodor, M. J. S. Dewar, A . Harget, and E. Haselbach, ibid.,92, 3854 (1970).
Bodor, Dewar, Wasson
Carbene and Its Reactions
