- Email: [email protected]
Concerning the structure of alkyl aluminium chlorides
Notes
149
Concerning the structure of alkyl aluminium chlorides (Received 11 April 1960; in revised form 23 M a y 1960)
TH~ proton nuclear magnetic resonance spectra of methyl aluminium chlorides (Al~MenCle_n; N = 2, 4, 6) have recently been published, n,a~ The compounds AI~Me2C1, and AI2Me4CI~. individually in an inert solvent, show but one absorption separated by 0.3 p.p.m. At --60 ° (and at room temperature) a mixture of the two (25 ~o A12MeaCI, in 75 % AI~Me4CIt) exhibits but one resonance interpreted as indicating intermoloeular methyl exchange even at --60°.ta, 8~ The N M R spectrum of AltMe6 shows 2 lines at --75 °, these lines coalescing into one at the weighted mean of the two with increasing temperature and thus indicating exchange of the two hypothecated types of methyl groups, bridging and end. c8~ The above interpretations for the chlorides n~ are derived from the assumed methyl-bridged structures postulated from Raman spectra, ~4~(I) and (I[). C1
Me
\/
/\
A1
Me
C1
CI
\/ A1
and
/\ Me
Me
\/ /\
Me
C1
\/
A1
C1
AI
/\ Me
CI
(CI cis)
I
II
We wish to point out that the Raman results are compatible with the structures I' and II'. (The tetrachloride has generally been assumed to have methyls asoeiated with both aluminium atoms as in II or as in I with the bridged methyls replaced by chlorines.) Me
CI
\
/
\ /
A1+
Me
Me
A1-
CI
\
and
/ \ CI
Me
/
Me
C1
\/
A1+
Me
Al-
/ \ C1
I'
C1
II'
For I' then, magnetic equivalence of the methyl groups in the proposed structure for the diehloro compound can result from chloride exchange Me
CI
\ /
Me
\ /
AI +
Me
AIMe
C1
\ /
,
/ \ C1
Me ~
/ Me
A1-
/
Me
+A1
\
\
C1
Me
The actual position of the resonance lines 2 are consistent with this postulation; the "end"-methyl frequency for the hexamethyl compound, +2.14 p.p.m., (from CH~ in cyclohexane) more closely approximates the observed frequencies in the chlorides (1.75 p.p.m, dichloride; 1.45 p.p.m, tetrachloride) than does the "bridging"-methyl frequency, + 1-07 p.p.m. In this analysis, no allowance is made for the electron withdrawing influence of the chloride; an effect which would be expected to reduce the observed frequencies when comparing hexamethyl with the chlorides, c5) ID S. BROWNSTEIN,B. C. SMITH,G. ERLICHand A. W. LAUBENGAYER,J. Amer. Chem. Soc. 82, 1000 (1960). (2) N. MULLERand D. E. PRITCHARD,J. Amer. Chem. Soc. 82, 248 (1960). t3~ Not observing two peaks is not a priori evidence for exchange. c4~C. P. VANDERKELeNand M. A. HERMAN,Bull. Soe. Chim. Belg. 65, 362 (1956). All structures, I, II, I',II', can be represented as tetrahedral aluminium atoms joined on an edge. t~) B. P. DA]LEVand J. N. SHOOLERY,d. Amer. Chem. Soc. 77, 3977 (1955) ; M.P. GROENEWEGE,J. SMIDTand H. DE VRI~S[d. Amer. Chem. Soc. 82, 4426 (1960)] have extended the work cited in refs. 1 and 2 to investigations at lower temperatures and compared the results with those obtained on Ti alkyls. Their interpretation agrees in all respects with ours.
150
Notes
The structures involving methyl bridges are intuitively less satisfying than the halide-bridged structures. The ion pair structures are in accord with the Raman spectra, (i.e. symmetry for II' Csv; Ta), the numbers and polarizations of lines predicted agreeing with experiment quite as well as for the methyl-bridged structures but without the rather strange assumption that only the cis isomer (I) is found. (4) The time of observation for the Raman spectra, essentially the time for a vibration, is sufficiently short so that the structures observed correspond to the "ion-pair" structures, while the observation time for the NMR experiments is sufficiently longer so that the chloride exchange could cause but one line for the methyls. A mechanism for exchange of methyls in the mixture of I' and II' would be simple metathesis. Methyl exchange in hexamethyl could proceed through the equivocable ethane-like structure that was concluded from electron diffraction~e) on this substance. This "no electron bond" intermediate will be described further elsewhere. (7) The chemistry of the alkyl aluminium corhpounds lends credence to the halide-bridged structures.cs) As the electron-donating character of the alkyl group decreases, the stabilities of the compounds decrease and conversely. A rather critical observation is that compounds AIsR,~CIs_ n are reported only for n = 0, 2, 4, 6; even in the hydrides polymerization occurs to the extent necessary for formation of AIR2+ units. ¢8) The lack of compounds of n : 1, 3 and 5 can be reasonably ascribed to the processes 2AIsMeCI6 = A12MeiCll + AI~C16 2AlsMe.C18 = AI,,Me,tCI. + Al,,MesCl,i 2AI2MesCI = Al2Me~Ch + AljMe6 with the driving force for these reactions possibly derived from the stability of the AlMet + unit, and the stability conferred by double chloride bridges, except, of course, for the hexamethyl. The only gas phase dipole moment measurement for the compounds AIaMe,~CI~__,is for the tetrachloride: a dipole moment of 1.9 D was found, an impossible situation on the basis of the methylbridged structures but quite reasonable with halide bridges. Unfortunately, as the authors have indicated, the clarity of the result is not conducive to the drawing of firm inferences. ¢9) The electron-diffraction results of BROCKWAY and DAVIDSON(e) favour the halide-bridged structures, although here too, equivocation is possible. The totality of the evidence seems more compatible with the halide-bridged structures than with the methyl bridged structures. R. E. GLICK A. ZWlCr~L
Department o f Chemistry The Florida State University Tallahassee, Florida (s) L. O. BROCKWAYand N. R. DAVIDSON,J. Amer. Chem. Soc. 63, 3287 (1941). (7) A. ZWJCKELand R. E. GLIeK, to be published. ~e) G. E. COATES,Organo-Metallic Compounds p. 7288. Methuen, London (1956). (9) R. H. WlSWALLand C. P. SMYTVl,J. Chem. Phys. 9, 352 (1941).
Some studies in the inorganic chemistry of the reaction between uranyl fluoride and chlorine trifluoride (Received 25 May 1960) Tim fluorination of uranium tetrafluoride with chlorine trifluoride to yield uranium hexafluoride has been discussed by LAnATON(~). Another process by which uranium bexafluoride may be made involves the treatment of uranyl fuoride with chlorine trifluoride, and an attempt has been made to define the inorganic chemistry of the reaction by establishing the nature of the gaseous products. (1) V. Y. LAnATON,Y. Inorg. NucL Chem. 10, 86 (1959).
149
Concerning the structure of alkyl aluminium chlorides (Received 11 April 1960; in revised form 23 M a y 1960)
TH~ proton nuclear magnetic resonance spectra of methyl aluminium chlorides (Al~MenCle_n; N = 2, 4, 6) have recently been published, n,a~ The compounds AI~Me2C1, and AI2Me4CI~. individually in an inert solvent, show but one absorption separated by 0.3 p.p.m. At --60 ° (and at room temperature) a mixture of the two (25 ~o A12MeaCI, in 75 % AI~Me4CIt) exhibits but one resonance interpreted as indicating intermoloeular methyl exchange even at --60°.ta, 8~ The N M R spectrum of AltMe6 shows 2 lines at --75 °, these lines coalescing into one at the weighted mean of the two with increasing temperature and thus indicating exchange of the two hypothecated types of methyl groups, bridging and end. c8~ The above interpretations for the chlorides n~ are derived from the assumed methyl-bridged structures postulated from Raman spectra, ~4~(I) and (I[). C1
Me
\/
/\
A1
Me
C1
CI
\/ A1
and
/\ Me
Me
\/ /\
Me
C1
\/
A1
C1
AI
/\ Me
CI
(CI cis)
I
II
We wish to point out that the Raman results are compatible with the structures I' and II'. (The tetrachloride has generally been assumed to have methyls asoeiated with both aluminium atoms as in II or as in I with the bridged methyls replaced by chlorines.) Me
CI
\
/
\ /
A1+
Me
Me
A1-
CI
\
and
/ \ CI
Me
/
Me
C1
\/
A1+
Me
Al-
/ \ C1
I'
C1
II'
For I' then, magnetic equivalence of the methyl groups in the proposed structure for the diehloro compound can result from chloride exchange Me
CI
\ /
Me
\ /
AI +
Me
AIMe
C1
\ /
,
/ \ C1
Me ~
/ Me
A1-
/
Me
+A1
\
\
C1
Me
The actual position of the resonance lines 2 are consistent with this postulation; the "end"-methyl frequency for the hexamethyl compound, +2.14 p.p.m., (from CH~ in cyclohexane) more closely approximates the observed frequencies in the chlorides (1.75 p.p.m, dichloride; 1.45 p.p.m, tetrachloride) than does the "bridging"-methyl frequency, + 1-07 p.p.m. In this analysis, no allowance is made for the electron withdrawing influence of the chloride; an effect which would be expected to reduce the observed frequencies when comparing hexamethyl with the chlorides, c5) ID S. BROWNSTEIN,B. C. SMITH,G. ERLICHand A. W. LAUBENGAYER,J. Amer. Chem. Soc. 82, 1000 (1960). (2) N. MULLERand D. E. PRITCHARD,J. Amer. Chem. Soc. 82, 248 (1960). t3~ Not observing two peaks is not a priori evidence for exchange. c4~C. P. VANDERKELeNand M. A. HERMAN,Bull. Soe. Chim. Belg. 65, 362 (1956). All structures, I, II, I',II', can be represented as tetrahedral aluminium atoms joined on an edge. t~) B. P. DA]LEVand J. N. SHOOLERY,d. Amer. Chem. Soc. 77, 3977 (1955) ; M.P. GROENEWEGE,J. SMIDTand H. DE VRI~S[d. Amer. Chem. Soc. 82, 4426 (1960)] have extended the work cited in refs. 1 and 2 to investigations at lower temperatures and compared the results with those obtained on Ti alkyls. Their interpretation agrees in all respects with ours.
150
Notes
The structures involving methyl bridges are intuitively less satisfying than the halide-bridged structures. The ion pair structures are in accord with the Raman spectra, (i.e. symmetry for II' Csv; Ta), the numbers and polarizations of lines predicted agreeing with experiment quite as well as for the methyl-bridged structures but without the rather strange assumption that only the cis isomer (I) is found. (4) The time of observation for the Raman spectra, essentially the time for a vibration, is sufficiently short so that the structures observed correspond to the "ion-pair" structures, while the observation time for the NMR experiments is sufficiently longer so that the chloride exchange could cause but one line for the methyls. A mechanism for exchange of methyls in the mixture of I' and II' would be simple metathesis. Methyl exchange in hexamethyl could proceed through the equivocable ethane-like structure that was concluded from electron diffraction~e) on this substance. This "no electron bond" intermediate will be described further elsewhere. (7) The chemistry of the alkyl aluminium corhpounds lends credence to the halide-bridged structures.cs) As the electron-donating character of the alkyl group decreases, the stabilities of the compounds decrease and conversely. A rather critical observation is that compounds AIsR,~CIs_ n are reported only for n = 0, 2, 4, 6; even in the hydrides polymerization occurs to the extent necessary for formation of AIR2+ units. ¢8) The lack of compounds of n : 1, 3 and 5 can be reasonably ascribed to the processes 2AIsMeCI6 = A12MeiCll + AI~C16 2AlsMe.C18 = AI,,Me,tCI. + Al,,MesCl,i 2AI2MesCI = Al2Me~Ch + AljMe6 with the driving force for these reactions possibly derived from the stability of the AlMet + unit, and the stability conferred by double chloride bridges, except, of course, for the hexamethyl. The only gas phase dipole moment measurement for the compounds AIaMe,~CI~__,is for the tetrachloride: a dipole moment of 1.9 D was found, an impossible situation on the basis of the methylbridged structures but quite reasonable with halide bridges. Unfortunately, as the authors have indicated, the clarity of the result is not conducive to the drawing of firm inferences. ¢9) The electron-diffraction results of BROCKWAY and DAVIDSON(e) favour the halide-bridged structures, although here too, equivocation is possible. The totality of the evidence seems more compatible with the halide-bridged structures than with the methyl bridged structures. R. E. GLICK A. ZWlCr~L
Department o f Chemistry The Florida State University Tallahassee, Florida (s) L. O. BROCKWAYand N. R. DAVIDSON,J. Amer. Chem. Soc. 63, 3287 (1941). (7) A. ZWJCKELand R. E. GLIeK, to be published. ~e) G. E. COATES,Organo-Metallic Compounds p. 7288. Methuen, London (1956). (9) R. H. WlSWALLand C. P. SMYTVl,J. Chem. Phys. 9, 352 (1941).
Some studies in the inorganic chemistry of the reaction between uranyl fluoride and chlorine trifluoride (Received 25 May 1960) Tim fluorination of uranium tetrafluoride with chlorine trifluoride to yield uranium hexafluoride has been discussed by LAnATON(~). Another process by which uranium bexafluoride may be made involves the treatment of uranyl fuoride with chlorine trifluoride, and an attempt has been made to define the inorganic chemistry of the reaction by establishing the nature of the gaseous products. (1) V. Y. LAnATON,Y. Inorg. NucL Chem. 10, 86 (1959).