Organometallic compounds of group 13, 55. Novel organoaluminum Lewis acids via selective aluminum-tin exchange processes - Electrophilic initiation by the aluminum halide and ensuing complexation by the resulting tin halide

Article abstract of DOI:10.1002/(sici)1099-0682(199901)1999:1<153::aid-ejic153>3.0.co;2-0

With the goals of preparing novel carbaluminating reagents and mono- and bidentate organoaluminum Lewis acids, the scope and limitations of synthesizing the requisite organoalanes by the aluminum-tin exchange between an aluminum halide and the appropriate organostannane have been examined in detail. The interactions of such tin precursors as 1,2- bis(trimethylstannyl)ethyne, allyltri-n-butyltin, benzyltrimethyltin, and 1,2-bis(trimethylstannyl)benzene and various aluminum chlorides of the type, R(n)AlCl(3-n) (R = Me, Et), gave selective aluminum-tin exchange at the sp- or sp²-hybridized carbon-tin bond and produced such organoalanes as allyl(methyl)aluminum chloride, benzylaluminum dichloride, 1,2- bis(diethylalumino)ethyne, 1,2-bis(dimethylalumino)benzene, 1,2- bis[chloro(methyl)aluminolbenzene, and 1,2-bis(dichloroalumino)benzene in high yield. A complicating factor was the tendency of the R₃SnCl by-product to complex with the resulting organoalane. In some cases, exemplified by allyl(methyl)aluminum chloride, such complexation did not interfere with the carbaluminating action of the reagent; in other cases, exemplified by 1,2- bis[chloro(methyl)-alumino]benzene, the R₃SnCl could be removed by means of π-bases and reduced pressures; and in still other structures, as with 1,2- bis(dichloroalumino)benzene, the tin chloride could not be dislodged at moderate temperatures. The structure and bonding in such tin halide-aluminum halide complexes in solution were investigated with the 1:1 adducts of AlCl₃ with Me₃SnCl and with nBu₃SnCl, respectively, by means of multinuclear NMR spectroscopy. Furthermore, an XRD of the solid complex, Me₃SnCl · AlCl₃, was carried out. Such complexes were shown to consist of 1:1 ion pairs of the type [R₃Sn⁺][AlCl₄^-] in dilute solution and of a chiral polymeric helix in the solid state wherein planar Me₃Sn⁺ units are linked to each other via bridging tetrachloroaluminate anions, Cl-(AlCl₂)-Cl^-. Treatment of such complexes with either benzyltriethylammonium chloride (one or two equivalents) or tetramethylphosphonium chloride leads to the displacement of Me₃SnCl and the formation of the expected ionic complexes. Finally, the importance of such novel reagents and chelating Lewis acids to organic synthesis and olefin polymerization is discussed and elucidated.

Full text of DOI:10.1002/(sici)1099-0682(199901)1999:1<153::aid-ejic153>3.0.co;2-0

FULL PAPER
Organometallic Compounds of Group 13, 55[°]
Novel Organoaluminum Lewis Acids via Selective Aluminum-Tin
Exchange Processes – Electrophilic Initiation by the Aluminum Halide
and Ensuing Complexation by the Resulting Tin Halide
John J. Eisch,*[a] Katrin Mackenzie,[a] Harald Windisch,[a] and Carl Krüger[b]
Dedicated to the memory of our most esteemed colleague, the late Professor Wilhelm P. Neumann, master of
organotin chemistry[ c]
Keywords: Organometallics / Aluminum / Tin / Metal-metal exchange / Electrophilic reagents / Lewis acid-base complexes
With the goals of preparing novel carbaluminating reagents
and mono- and bidentate organoaluminum Lewis acids, the
scope and limitations of synthesizing the requisite
organoalanes by the aluminum-tin exchange between an
aluminum halide and the appropriate organostannane have
been examined in detail. The interactions of such tin
precursors as 1,2-bis(trimethylstannyl)ethyne, allyltri-n-
alumino]benzene, the R3SnCl could be removed by means of
π-bases and reduced pressures; and in still other structures,
as with 1,2-bis(dichloroalumino)benzene, the tin chloride
could not be dislodged at moderate temperatures. The
structure and bonding in such tin halide–aluminum halide
complexes in solution were investigated with the 1:1 adducts
of AlCl3 with Me3SnCl and with nBu3SnCl, respectively, by
means of multinuclear NMR spectroscopy. Furthermore, an
XRD of the solid complex, Me3SnCl · AlCl3, was carried out.
butyltin,
stannyl)benzene and various aluminum chlorides of the type,
RnAlCl3–n (R Me, Et), gave selective aluminum-tin
benzyltrimethyltin,
and
1,2-bis(trimethyl-
=
Such complexes were shown to consist of 1:1 ion pairs of the
–
exchange at the sp- or sp2-hybridized carbon–tin bond and
produced such organoalanes as allyl(methyl)aluminum
chloride, benzylaluminum dichloride, 1,2-bis(diethylalumi-
no)ethyne, 1,2-bis(dimethylalumino)benzene, 1,2-bis[chloro-
(methyl)alumino]benzene, and 1,2-bis(dichloroalumino)-
type [R3Sn+][AlCl4
]
in dilute solution and of
a
chiral
polymeric helix in the solid state wherein planar Me3Sn+
units are linked to each other via bridging
tetrachloroaluminate anions, Cl–(AlCl2)–Cl–. Treatment of
such complexes with either benzyltriethylammonium
chloride (one or two equivalents) or tetramethylphosphonium
chloride leads to the displacement of Me3SnCl and the
formation of the expected ionic complexes. Finally, the
importance of such novel reagents and chelating Lewis acids
to organic synthesis and olefin polymerization is discussed
and elucidated.
benzene in high yield.
A complicating factor was the
tendency of the R3SnCl by-product to complex with the
resulting organoalane. In some cases, exemplified by
allyl(methyl)aluminum chloride, such complexation did not
interfere with the carbaluminating action of the reagent; in
other
cases,
exemplified
by
1,2-bis[chloro(methyl)-
In continuing investigations directed toward the synthesis
of novel Group 13 Lewis acids suitable as cocatalysts with
cyclopentadienyl transition metal derivatives for ZieglerϪ
Natta olefin polymerization catalysis, we have sought ways
of preparing unsymmetrical organometallics of Group 13
whose Lewis acidity might be enhanced either by the elec-
tron-withdrawing nature of the substituent E in 1 or by the
chelating effect of two proximally arrayed Lewis acid metal
centers M in 2. In the previous article of this series the
possible effect of such Lewis acidity enhancing factors in
selected organoboranes was examined.[1] This present study
has explored the Lewis acidity of analogous unsymmetrical
[°]
Part 54: J. J. Eisch, B. W. Kotowicz, Eur. J. Inorg. Chem. 1998, 761.
Department of Chemistry, The State University of New York
at Binghamton,
[a]
Binghamton, New York 13902Ϫ6016, USA
[b]
[c]
Max-Planck-Institut für Kohlenforschung,
D-45470 Mülheim (Ruhr), Germany
Professor W. P. Neumann was born in Würzburg, Germany, on
October 29, 1926. Upon the completion of studies on pyrimidopy-
rimidines with Professor F. G. Fischer at the University of Würz-
burg in 1952, he was awarded the degree Dr. rer. nat. Thereafter,
he was postdoctoral research associate with Nobel Laureate Karl
Ziegler with whom he embarked on his lifelong study of organotin
chemistry. After habilitation on this theme at the University of
Gießen in 1960, he served on the same faculty as Außerplanmäßi-
ger Professor until 1968. He joined the faculty of the University
of Dortmund in 1969 as Ordinary Professor of Organic Chemistry
and until his formal retirement in 1992 he was chiefly responsible organoaluminum compounds in the well-established expec-
for developing the Organic Chemistry Institute of this new univer-
sity. His research and scholarly work on the organometallic chemi-
tation that such organoaluminum analogs would exhibit
greater Lewis acidities than their boron counterparts for
steric and electronic reasons:[2] the smaller covalent radius
of boron versus aluminum (B, 81 pm versus Al, 125 pm)
increasing the steric hindrance or F-strain upon com-
plexation (3) and any pπϪpπ back-bonding from ligands (E)
on the boron center, such as unshared electron pairs (E ϭ
X, RO, or R2N), decreasing the boron centerЈs Lewis acidity.
stry of tin and germanium and on organic intermediates, such as
free radicals and carbenes, were reported in over 250 articles and
reviews. At the time of his death on August 3, 1993 he was still at
the forefront of the organic chemistry of tin, a research area on
which he had composed a monograph as a young scientist (1970,
Wiley) and whose high relevance to the whole field of organic
chemistry his career in research has made manifest. We, his colle-
agues, will always fondly remember his great scientific zest and his
extraordinary human warmth.
Eur. J. Inorg. Chem. 1999, 153Ϫ162
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ1948/99/0101Ϫ0153 $ 17.50ϩ.50/0
153

J. J. Eisch, K. Mackenzie, H. Windisch, C. Krüger
FULL PAPER
of groups is achievable when the CϪSn bond is attached to
an sp- or sp2-hybridized group.[1] In addition, by combining
what has been established by Neumann in his pioneering
work on aluminum-tin exchanges[6] with what is known
about lithium-metal exchanges,[7] it was considered likely
that aluminum-tin exchanges would proceed selectively at
CϪSn bonds to allylic and benzylic groups as well. The
driving force for such exchanges appears to lead to the more
electropositive metal atom (Li, Al) being bonded to the
more electronegative carbon atom (sp-, sp2-, or allylic-
bonded organyl group) in the favored product.
As a preparative route to such unsymmetrical organoalu-
minum reagents, we have availed ourselves of the selective
exchange occurring between organotin compounds and
aluminum halides (Equation 1).[3] Such exchanges are po-
tentially reversible depending on the nature of the organyl
substituents and on experimental conditions. Where R, RЈ,
and RЈЈ are alkyl, the reaction proceeds to the left and pro-
duces tetraalkyltin compounds.[4] In the present case of
interest, the reaction has been found to lead selectively to
the right, producing 7, when R in 4 is an organyl group
more electronegative than alkyl, such as allyl, benzyl, aryl,
or 1-alkynyl. Furthermore, the extent of tin-aluminum ex-
change can in some instances be feasibly controlled by re-
moving one of the components of the equilibrium in Equa-
tion 1 from the reaction zone by volatization or com-
plexation, in accord with LeChatelierЈs principle.[5]
Allyl(methyl)aluminum Chloride (9)
This reagent could be readily generated from the interac-
tion of allyltri-n-butyltin with methylaluminum dichloride
at Ϫ78°C but the high autoreactivity of such allylic alanes
precluded its isolation at room temperature (Equation 3).
Hence, reagent 9 was generated at Ϫ78°C in the presence
of trimethyl(phenylethynyl)silane (10); subsequent hydro-
lytic work-up led to a high yield of (Z)-2-phenyl-1-trimeth-
ylsilyl-1,4-pentadiene (12). Work-up with D2O yielded 12
deuteriated at C-2. The exclusive formation of 12 establishes
that 9 had carboaluminated 10 with regio- and stereo-selec-
tivity to form precursor 11 (Equation 4).[8]
Because of the greater Lewis acidity of such organoalu-
minum derivatives, compared with their organoboron anal-
ogs, a serious complication occurs when utilizing such tin-
aluminum exchanges for preparing unsymmetrical orga-
noalanes (7, Equation 1): namely the organoalane 7 usually
complexes more or less firmly with the organotin halide by-
product 6 (Equation 2). Such complexation (8) generally
does not interfere in preparative tin-boron exchanges,[1]
most likely because of destabilizing F-strain in such com-
plexation (cf. 3). This complexation in tin-aluminum ex-
changes (8) has been of dual interest to us: not only did we
seek mild experimental conditions to disrupt such com-
plexation and to isolate 7 in satisfactory yield, but also we
wished to elucidate the structure and bonding in complex 8.
Benzylaluminum Dichloride (14)
The attempted exchange between benzyltrimethyltin and
methylaluminum dichloride to produce benzyl(meth-
yl)aluminum chloride (13, Equation 5) took an unexpected
course when the separation of 13 from its complex with 15
was undertaken. Under the conditions expected to lead to
the dissociation of the complex, 13 · 15, and the selective
volatilization of 15 (100°C in mesitylene), further tin-alumi-
Results
Preparation of Organoalanes by Aluminum؊Tin
Exchange Reactions
Previous studies on generating organoboranes by boron- num exchange occurred with the formation of pure 14 and
tin exchange processes had revealed that selective exchange the volatilization of tetramethyltin (Equation 6).[9]
154
Eur. J. Inorg. Chem. 1999, 153Ϫ162

Organometallic Compounds of Group 13, 55
FULL PAPER
1,2-Bis[chloro(methyl)alumino]benzene (19)
Similarly, the interaction of 17 with 2 equivalents of
methylaluminum dichloride in toluene at 25°C yielded the
analogous ortho-dichloro(methyl)alumino derivative 19,
which was even more strongly complexed with the Me3SnCl
by-product than was 18. Here the use of the π-base, toluene,
for the reaction aided the dissociation and removal of
Me3SnCl from 19. Evaporation of the toluene solvent and
repeated evaporation of several added portions of toluene
under vacuum served to chase off almost all of the remain-
ing Me3SnCl (Equation 9).[8]
1,2-Bis(diethylalumino)ethyne (16)
The interaction of 2 equivalents of diethylaluminum chlo-
ride with one equivalent of bis(trimethylstannyl)ethyne in
pentane at Ϫ30°C led to the deposition of 95% of the pale
yellow 16, which was highly pyrophoric and insoluble in
benzene or chloroform. The 1H-NMR spectrum of 16 from
the reaction mixture from Equation 7 in THF is consistent
with complexation of 16 with the ether solvent and the lib-
eration of free Me3SnCl: The triplet and quadruplet ethyl
signals of 16 are shifted downfield from those of unsolvated
Et3Al and the singlet due to free Me3SnCl is visible at δ ϭ
1.2.
1,2-Bis(dichloroalumino)benzene (20)
Finally, the interaction of 17 with 2 equivalents of alumi-
num trichloride in toluene suspension led to the aluminum-
tin exchange product, 20, but in this situation the Me3SnCl
by-product remained firmly complexed with 20 to form a
polymeric ion pair 21 (Scheme 1).
1,2-Bis(dimethylalumino)benzene (18)
The interaction of 1,2-bis(trimethylstannyl)benzene (17)
with 2 equivalents of dimethylaluminum chloride in hexane
at 25°C produced the ortho-dimethylalumino derivative 18
as its 1:1 and 1:2 complexes with Me3SnCl (Equation 8), as
evidenced by Me3Sn group singlets in the δ ϭ 0.02 to Ϫ0.02
1
region in the H NMR, as well as by extra aromatic dou-
blets in the δ ϭ 7.0Ϫ8.0 region. Such Me3SnCl was largely
removed by warming under reduced pressure and by the
distillative removal of added heptane at ordinary pressure
to yield essentially free 18.
Scheme 1
The presence of the trichloroaluminate anion is consist-
ent with a 27Al-NMR resonance signal at δ ϭ 105, as is the
presence of a planar trimethylstannyl cation supported by
the low-field 119Sn-NMR peak at δ ϭ 357.5. (More struc-
tural information on such ion-pairs follows below). Ther-
mal dissociation of 21 under high vacuum or by use of π-
basic solvents under 100°C has thus far proved to be inef-
fectual in removing the complexed Me3SnCl.[8]
Eur. J. Inorg. Chem. 1999, 153Ϫ162
155

J. J. Eisch, K. Mackenzie, H. Windisch, C. Krüger
FULL PAPER
Scheme 2
enhancing the coordination and hence the shielding of the
nBu3Snϩ cation (27 in Equation 10).
Complexation of Organotin Halides with
Aluminum Halides
Trimethyltin Chloride with
1,2-Bis(dichloroalumino)benzene in a 2:1 Ratio (21)
Although Me3SnCl is difficult to remove from complex
21, the trimethylstannyl cation can readily be displaced
either singly or dually by the benzyltriethylammonium ion
(22 and 23, Scheme 2) or by the tetramethylphosphonium
ion (24). In all cases, the 27Al-NMR spectra display a reso-
nance peak in the δ ϭ 103 region; in 22 and 24 the 119Sn-
NMR spectra have a peak in the δ ϭ 317Ϫ320 region; and
X-ray Crystallographic Structure of [Me3SnCl·
AlCl3]n, Complex 25, and Its Formulation as
Polymeric Trimethyltin Cationic-Tetrachloroaluminate
Anionic Pairs
1
in each complex the appropriate H- and 13C-NMR peaks
corroborate the ionic structures and compositions.
As shown in Figure 1 [Me3SnCl·AlCl3]n (25) consists in
the solid state of trigonal bipyramidal trimethyltin frag-
ments and tetrahedral coordinated tetrachloroaluminate
fragments forming a chiral polymeric helix.[13] The three
Complexes of Trimethyltin Chloride with Aluminum
Chloride (Complex 25) and of Tri-n-butyltin Chloride
with Aluminum Chloride (Complex 26)
These 1:1 complexes were first described by Neumann, equatorial methyl groups bonded to the tin atoms are com-
˚
Schick, and Köster almost 35 years ago and upon the basis pletely planar, the SnϪC bond lengths (average, 2.109 A,
of the relatively high electrical conductivity of such com- compare Tables 1Ϫ3) are within the usual range for
plexes and their formula weight in dilute solution, they were SnϪCH3 bonds.[14Ϫ22] The two axial positions are occupied
depicted as ion pairs. Above 1 mol/L concentrations higher by chlorine atoms of the tetrachloroaluminate fragments.
degrees of association (2Ϫ6) became significant.[9] To add The SnϪCl distances are much longer than usual,[16,17,20,21]
˚
to our insight into the structure of complexes 25 and 26, a especially more than 0.30 A longer than the SnϪCl bond
multinuclear NMR study was undertaken in solution. in Me3SnCl (2.430 A).[16][17] These values show that 25 can
˚
Noteworthy are the following observations: 1) the 27Al- best be understood as a contact ion pair of trimethyltin
NMR signals occur at δ ϭ 104Ϫ105, almost identical with cations and tetrachloroaluminate anions, as was already
the reported peak for AlCl4 ;
Ϫ [10] 2) the 119Sn-NMR peaks indicated by the results of the NMR-spectroscopic measure-
are displayed at δ ϭ 358 and 380, respectively, consistent ments. Therefore it would be more accurate to write an
with the presence of a planar R3Snϩ cation;[11][12] and 3) ionic formula like [Me3Snϩ][AlCl4Ϫ] for compound 25.
the 119Sn-NMR signal of complex 26 shows a significant Furthermore it is worth mentioning that the SnϪCl dis-
upfield shift with increasing π-basicity of the medium: δ ϭ tances alternate, each tin atom [except Sn5 with equal dis-
˚
˚
380 ([D6]benzene), 375 (toluene), 356 (mesitylene), and 338 tances of 2.790(4) A] has a short (average, 2.772 A) and a
˚
(saturated solution of hexamethylbenzene in pentane). This long bond length (average, 2.832 A) to a chlorine atom. The
last observation is consistent with such increasing π-basicity ClϪSnϪCl angles are on an average 172.9° instead of 180°
156
Eur. J. Inorg. Chem. 1999, 153Ϫ162

Organometallic Compounds of Group 13, 55
FULL PAPER
Figure 1. Structure of the polymeric chiral helix of [Me3SnCl·AlCl3] (25)
Table 1. Selected bond lengths and angles of [Me3SnCl·AlCl3] (25) bis(trimethylstannyl)ethyne (28) and 1,2-bis(trimethylstan-
nyl)benzene (17), attack by the aluminum chloride,
Minimum Maximum Average
RЈnAlCl3Ϫn (29), occurs cleanly and exclusively at the car-
value
value
value
bonϪtin bond of higher s-character (Equations 7Ϫ9 and
Scheme 1). This preference for aluminum-tin exchange at
such sites accords with a fostered electrophilic attack of 29
at such sites through transition states similar to 30. Attack
of 29 at the α-carbon atom of the CϪSn bond is favored
by the σ-bond hyperconjugation stabilizing the positive
charge at the β-carbon atom.[25]
˚
Bond lengths [A]
SnϪC
2.02(2)
2.15(2)
2.109
2.772
2.832
2.090
2.171
SnϪCl (short)
SnϪCl (long)
SnϪCl terminal
SnϪCl bridging
Bond angles [°]
CϪSnϪC
2.738(4)
2.800(4)
2.065(8)
2.143(7)
2.792(5)
2.887(4)
2.109(6)
2.205(6)
116.4(7)
168.1(1)
123.6(7)
175.7(1)
116.8(3)
106.6(2)
122.1(2)
120.0
172.9
114.9
104.8
115.8
ClϪSnϪCl
Cl(terminal)ϪAlϪCl(terminal) 114.0(3)
Cl(bridging)ϪAlϪCl(bridging) 103.4(2)
SnϪClϪAl
111.1(2)
as one would expect for an ideal trigonal bipyramidal con-
figuration.
The preferential reactivity of allylic or benzylic CϪSn
bonds over alkyl CϪSn bonds can be explained by similar
electrophilic attack by RЈnAlCl3Ϫn (29) (Equations 4 and 5)
except that such attack occurs at the γ-carbon atom with
rearrangement (31, Equation 11). Here again, the favored
transition state 31 is dually stabilized by σ-bond hypercon-
jugation of the β-carbon atomЈs positive charge.
The tetrachloroaluminate fragments show different dis-
tances from the aluminum atoms to its chlorine atoms (av-
˚
erage, 2.090 A) and to the bridging chlorine atoms (average,
˚
˚
2.171 A), but both bonds are only 0.04 A shorter or longer
[23]
˚
than those found for NaAlCl4 (2.13 A).
Also the
ClϪAlϪCl angles are slightly different. The angles includ-
ing the terminal chlorine atoms are on average 114.9°,
whereas the ClϪAlϪCl angles for the bridging chlorine
atoms are on average 104.8° compared to 109.5° for an
ideal tetrahedral configuration. These structural differences
at the tetrachloroaluminate fragments in this particular
complex can be explained by the interactions of the bridg-
ing chlorine atoms with the trimethyltin fragments.
Discussion
Selectivity of Aluminum-Tin Exchange
With organotin derivatives having both sp- or sp2-hy-
bridized and sp3-hybridized carbonϪtin bonds, such as 1,2-
Eur. J. Inorg. Chem. 1999, 153Ϫ162
157

J. J. Eisch, K. Mackenzie, H. Windisch, C. Krüger
FULL PAPER
Applicability of this mechanism to the reaction of benzyl- no)benzene (18) and 1,2-bis[chloro(methyl)alumino]ben-
trimethyltin (3) with MeAlCl2 (Equation 5) would require zene (19) are clearly weaker Lewis acids than 20 because
that the initial reaction product 33 would have to undergo the complexed Me3SnCl can be removed by use of high
facile rearrangement to 13 (Scheme 3). Transformation of vacuum and the addition of π-basic arenes (Equations 8
33 to 13 would involve a 1,3-suprafacial sigmatropic re- and 9).
arrangement which should be thermally allowed via the
However, there is one structural feature of compounds
available p-orbital on the sp2-hybridized aluminum atom 18, 19, and 20 that make them especially appealing as novel
in 33.[25]
Lewis acids, and this is their potential as structurally rigid,
chelating Lewis acids in ZieglerϪNatta olefin polymeri-
zation. This attractive prospect motivates our continuing ef-
forts to prepare 19 and 20 free of any complexation with
Lewis bases. In the final section our recent progress in this
research is outlined.
Novel Applications of Unsymmetrical Organoalanes
in Organic Synthesis and Polymerization
Such aluminum-tin exchange processes constitute a most
useful selective route to vinylic, allylic, acetylenic, aryl, and
benzylic organoalanes in their unsolvated form. Illustrative
are the preparations depicted in Equations 3, 5, 7, 18, and
19. Since the requisite organotin precursor reacts readily
with RAlCl2 or R2AlCl in alkanes, toluene or dichloro-
methane, the organoalane can be generated free of any co-
ordinated ether solvent and despite its auto-association, in
its more electrophilic form. The weakly coordinated R3SnX
by-product can often be removed by a combination of π-
Scheme 3
Lewis Acidity of Aluminum Halides towards
Trialkyltin Chlorides and the Structure of the
Resulting Lewis Complexes
From the evidence on the spectral properties of com- basic hydrocarbons and high vacuum. In some cases, as
plexes 25 and 26, R3SnCl·AlCl3, in solution (R ϭ Me, Bu), with allyl(methyl)aluminum chloride (9 in Equation 3), the
it is clear that AlCl3 can heterolyze the tinϪchlorine bond presence of the R3SnX does not interfere with the carbalu-
completely to form the ion pair, [R3Sn]ϩ[AlCl4]Ϫ. In dilute minating action of the organoalane.
solutions of π-basic hydrocarbons such as arenes, the trialk-
Since many organotin derivatives can be prepared by
yltin cation is probably solvated by arenes and exists as a Grignard and organolithium reactions or by hydrostanna-
simple solvated ion pair with the tetrachloroaluminate tions, subsequent aluminum-tin exchanges permit access to
anion, as corroborated by 119Sn- and 27Al-NMR spec- the more reactive organoalanes (Scheme 4).[26] Typical is
troscopy. The decreasing chemical shift of the 119Sn signal the stereoselective generation of organoalane 35 from or-
of 26 as the π-basicity of the arene solvent increases, from ganotin precursor 34. Reagent 35 can carbaluminate
δ ϭ 380 in [D6]benzene to 375 in toluene to 356 in mesityl- alkynes with selective syn addition (36),[27] undergoes car-
ene to 338 in hexamethylbenzene dissolved in pentane, is boxylation via an ate complexation with retention of con-
consistent with an increasing complexation of the Bu3Snϩ figuration (37)[28][29] and yields the corresponding allyl de-
with the arene π-electron cloud and thus an increasing dia- rivative (38) by ate-complex-assisted allylation.[30] These
magnetic shielding of the tin cation by the π-electrons. In selective reactions at the vinylic CϪAl bonds, rather than
more concentrated solutions (> 0.01 mol/L) and especially at the methyl CϪAl bonds, highlight the great utility of
in the solid state, planar R3Sn cations are linked to each such vinylalanes in organic synthesis.
other by AlCl4 anions through ClϪ(AlCl2)ϪCl bridges
Finally, the first successful synthesis of the structurally
(Figure 1), whereby the bridging SnϪCl separations are rigid, chelating Lewis acids of types 39 and 40 from the
some 30 pm longer (12%) than the SnϪCl bond in the orig- corresponding organotin precursors has led to their evalu-
inal Me3SnCl (at δ ϭ 243). Such relatively weak coordi- ation as potential cocatalysts with transition metal metal-
nation by the two chlorine atoms means that the planar locenes for the polymerization of olefins. Indeed, such co-
R3Sn unit is essentially still cationic.
catalysts, when employed with titanocene methyl chloride,
The stability of the complex 21 between 1,2-bis(dichlo- Cp2Ti(Me)Cl, in a 1:8 Ti/Al ratio, have been found decid-
roalumino)benzene (20) and 2 equivalents of Me3SnCl dem- edly superior to other aluminum cocatalysts such as Me-
onstrates that an arylaluminum dichloride is still a strong AlCl2 and even MAO, in polyethylene productivity numbers
enough Lewis acid to heterolyze Me3SnCl completely. Pre- at comparable Ti/Al ratios. Productivity numbers (PN) of
sumably the structure of viscous 21 consists of intramolecu- 2000Ϫ3000 g of polyethylene per g Ti ϫ per h ϫ per atm
lar and intermolecular bridging of Me3Snϩ units between are attained with 39 and 40, compared with PN values of
Ϫ
arene AlCl3 sites (Scheme 1). The 1,2-bis(dimethylalumi- 800 for MAO and 200 for MeAlCl2.[31] The greater cocata-
158
Eur. J. Inorg. Chem. 1999, 153Ϫ162

Organometallic Compounds of Group 13, 55
FULL PAPER
Scheme 4
CaH2; and ethers from sodium/benzophenone combinations. Deut-
eriated solvents (Cambridge Scientific) for NMR measurements of
oxygen- and moisture-sensitive substances were vacuum-trans-
ferred to the Schlenk glass vessels and stored over Molecular Sieves
lytic efficacy of 39 and 40 may well stem from the greater
binding of the chloride ion through their chelating ability
(41b) and thus the cleaner generation and separation of the
solvent-separated cation 41a, which has been established as
the active site in the metallocene polymerization of olef-
ins.[32Ϫ34]
˚
(4 A) under argon. Gas chromatographic analysis (GC) was per-
formed with a Hewlett-Packard Chromatograph, model 5880A,
with helium as the carrier gas. The 1H- (360.1 MHz) and 13C- (90.6
MHz) NMR spectra were recorded with a Bruker AM 360 spec-
trometer and the 27Al- (78.2 MHz) and 119Sn- (111.9 MHz) NMR
spectra with a Bruker AC 300 spectrometer. Solid-state NMR spec-
tra were recorded with a Bruker AC 300 spectrometer at 75.4 (13C),
78.2 (27Al) and 111.9 MHz (119Sn). The 1H- and 13C-NMR chemi-
cal shifts are referred to resonances of the solvent [D6]benzene. For
27Al-NMR spectra aqueous AlCl3 (δ ϭ 0) and for 119Sn-NMR
spectra Me3SnCl (16.4% in C6D6: δ ϭ 159.7[28]) were used as exter-
nal reference standards.
Preparation of Aluminum-Tin Complexes with Trialkyltin Chlorides
Complex 25 of Trimethyltin Chloride and Aluminum Trichloride in
a 1:1 Ratio: To a suspension of 1.94 g (14.5 mmol) of aluminum
trichloride in 20 mL of toluene was added 7.95 mL (7.95 mmol) of
a 1.0  solution of trimethyltin chloride in toluene via a syringe at
room temperature. The resulting suspension was stirred for 3 d,
filtered, and the solvent removed under reduced pressure. The re-
maining yellow crystals were washed with 10 mL of pentane, the
pentane layer was decanted and the product dried in vacuo. In this
way 2.55 g (7.7 mmol) of the crude product was obtained (97%).
For further purification the crystals of 25 were again dissolved in
10 mL of toluene, the solution was filtered and cooled to Ϫ78°C.
After standing overnight the solvent was decanted at a temperature
of Ϫ78°C and the crystalline product was washed at room tempera-
ture with a mixture of 8 mL of pentane and 4 mL of toluene and
finally dried in vacuo. Ϫ 1H NMR (C6D6): δ ϭ 0.88 (s, SnCH3,
2JSn-H ϭ 56.2/58.5 Hz). Ϫ 13C NMR (C6D6): δ ϭ 5.7 (SnCH3,
1JSn-C ϭ 334.4/350.1 Hz). Ϫ 27Al NMR (C6D6): δ ϭ 104.6 (W1/2 ϭ
960 Hz). Ϫ 119Sn NMR (C6D6): δ ϭ 342.0.
Experimental Section
General Techniques: All preparations and purification procedures
involving manipulations with oxygen- and moisture-sensitive re-
agents were carried out under anhydrous, oxygen-free argon or ni-
trogen employing Schlenk techniques.[35] Those procedures which
did not require a protective atmosphere but involved working with
toxic chemicals (such as boron and tin compounds) were carried
out in an exhaust hood evacuating air at 100 cubic ft/min. For
the operations under reduced pressure the required vacuum was
provided either by a rotary vacuum oil pump (down to 0.1 Torr)
or by a water pump (down to 10 Torr) and toxic volatiles were
condensed in cold traps. Commercially available solvents of reagent
grade were purified according to the recommended methods.[35]
Solvents used in reactions were always freshly distilled under argon
or nitrogen from the appropriate drying agents: aliphatic hydro-
carbons from LiAlH4; toluene from sodium; dichloromethane from
Complex 26 of Tri-n-butyltin Chloride and Aluminum Trichloride in
a 1:1 Ratio: To a suspension of 1.94 g (14.5 mmol) of aluminum
trichloride in 20 mL of pentane was added 1.5 mL (5.5 mmol) of
Eur. J. Inorg. Chem. 1999, 153Ϫ162
159

J. J. Eisch, K. Mackenzie, H. Windisch, C. Krüger
FULL PAPER
tributyltin chloride via a syringe at room temperature. The resulting
suspension was stirred for 2 d, filtered, and the solid was extracted
three times with 10 mL of pentane. The solvent of the combined
extracts and filtrate was removed in vacuo to leave 2.38 g of a
yellow oil which was identified as the pure complex 26 by NMR
and the residue was then heated in vacuum up to 40°C in order to
remove all volatile materials. The remaining yellow residue was
treated with C6D6 which caused again the formation of two layers,
which were individually investigated by NMR spectroscopy. The
spectra of the lower layer showed all signals corresponding to the
reaction product, namely bis(benzyltriethylammonium) 1,2-phenyl-
1
spectroscopy (94%). Ϫ H NMR (C6D6): δ ϭ 0.86 (t, CH3, 3 H),
1.29 (sext, CH2, 2 H), 1.72 (m, CH2, 4 H). Ϫ 13C NMR (C6D6): enebis(trichloroaluminate) (23). Ϫ 1H NMR (360.1 MHz, C6D6,
1
δ ϭ 13.6 (CH3), 25.7 (CH2, JSn-C ϭ 266.0/278.3 Hz), 27.0 (CH2,
25°C): δ ϭ 0.75 (t, CH3CH2N), 2.28 (q, CH3CH2N), 3.46 (s,
C6H5CH2N), 6.92Ϫ6.95 (m, C6H5), 6.98Ϫ7.10 (m, C6H4),
7.22Ϫ7.26 (m, C6H5). Ϫ 13C NMR (90.6 MHz, C6D6, 25°C): δ ϭ
7.6 (CH3CH2N), 52.5 (CH3CH2N), 60.5 (C6H5CH2N), 125.5
(C6H4), 125.9 (C6H5), 128.5 (C6H4), 129.2 (C6H4), 129.7 (C6H5),
131.0 (C6H5), 132.2 (C6H5). Ϫ 27Al NMR (78.2 MHz, C6D6,
25°C): δ ϭ 103.5 (W1/2 ϭ 14 Hz).
3
2JSn-C ϭ 75.9 Hz), 28.0 (CH2, JSn-C ϭ 27.7 Hz). Ϫ 27Al NMR
(C6D6): δ ϭ 103.7 (W1/2 ϭ 3000 Hz). Ϫ 119Sn NMR (C6D6): δ ϭ
380.4.
Complex 21 of Trimethyltin Chloride and 1,2-Bis(dichloroalumino)-
benzene (20) in a 2:1 Ratio: To a suspension of 4.43 g (33.2 mmol)
of aluminum trichloride in 25 mL of toluene was added 2.03 g (5.0
mmol) of 1,2-bis(trimethylstannyl)benzene[1] via a syringe at room
temperature. The resulting yellow reaction mixture was stirred for
3 d and then filtered. The solvent was evaporated from the filtrate.
In order to remove all volatile material the residue was heated in
vacuum up to 50°C. The resulting yellow oil was washed with 10
Displacement of One Equivalent of Me3SnCl by Tetramethylphos-
phonium Chloride. ؊Complex 24: To 152 mg (1.2 mmol) of tetram-
ethylphosphonium chloride was added 4 mL of a 0.3  solution of
complex 21 (1.2 mmol) at room temperature. Stirring the reaction
mixture overnight resulted in the formation of two layers: the upper
mL of pentane, the pentane layer was decanted and the residual oil layer was a clear yellow solution and the lower layer was a golden-
dried in vacuum. In this way 3.25 g (4.8 mmol) of the pale yellow colored oil. After evaporating the solvent, the residue was heated
oil of 21 was obtained in 97% yield. Ϫ 1H NMR (360.1 MHz,
C6D6, 25°C): δ ϭ 0.86 (s, 6 H, SnCH3, JSn-H ϭ 58.1 Hz), 6.98 (d,
in vacuum up to 40°C in order to remove all volatile materials.
Addition of C6D6 resulted in the formation of the two layers which
were individually investigated by NMR spectroscopy. The spectra
of the upper layer showed only signals corresponding to Me3SnCl,
2
2 H, C6H4, JH-H ϭ 7.8 Hz), 7.08 (d, 2 H, C6H4, JH-H ϭ 7.8 Hz).
1
Ϫ
13C NMR (90.6 MHz, C6D6, 25°C): δ ϭ 5.9 (SnCH3, JSn-C
ϭ
346.3/331.7 Hz), 125.6 (C6H4), 128.6 (C6H4), 129.4 (C6H4). Ϫ 27Al while in the spectra of the lower layer the signals of the reaction
NMR (78.2 MHz, C6D6, 25°C): δ ϭ 105 (W1/2 ϭ 20000 Hz). Ϫ product were found, namely the 1:1 complex of tetramethylphos-
119Sn NMR (111.9 MHz, C6D6, 25°C): δ ϭ 357.5.
phonium
trichloro(2-dichloroaluminophenyl)aluminate
with
Me3SnCl (24). Ϫ 1H NMR (360.1 MHz, C6D6, 25°C): δ ϭ 0.82 (d,
Displacement of Trimethyltin Chloride from Complex 21
2
2
CH3P, JP-H ϭ 14.1 Hz), 1.04 (s, CH3Sn, JSn-H ϭ 58.8 Hz),
6.96Ϫ7.22 (m, C6H4). Ϫ 13C NMR (90.6 MHz, C6D6, 25°C): δ ϭ
5.9 (CH3Sn, 1JSn-C ϭ 345.0/358.2 Hz), 9.3 (CH3P, 1JP-C ϭ 56.0 Hz),
130.1 (C6H4), 134.3 (C6H4), 139.6 (C6H4). Ϫ 27Al NMR (78.2
MHz, C6D6, 25°C): δ ϭ 103 (W1/2 ϭ 1500 Hz). Ϫ 13P NMR (121.4
MHz, C6D6, 25°C): δ ϭ 21.4. Ϫ 119Sn NMR (111.9 MHz, C6D6,
25°C): δ ϭ 317.0 (W1/2 ϭ 30 Hz).
Displacement of One Equivalent of Me3SnCl by Benzyltriethylam-
monium Chloride. Ϫ Complex 22: To 211 mg (0.93 mmol) of
benzyltriethylammonium chloride was added 3.1 mL of a 0.3  of
solution of the 2:1 complex of Me3SnCl and 1,2-bis(dichloroalumi-
no)benzene (21) (0.93 mmol) at room temperature. Stirring of the
reaction mixture overnight resulted in the formation of two layers.
While the upper layer was a clear yellow solution, the lower layer
was a brown oil. After removal of solvent, the residue was heated
in vacuum up to 40°C in order to remove all volatile material.
Addition of C6D6 resulted again in the formation of the two layers
Aluminum-Tin Exchange Reactions
1,2-Bis(diethylalumino)ethyne (16): To a stirred solution of 0.80 g
(2.3 mmol) of 1,2-bis(trimethylstannyl)ethyne (28), dissolved in 20
which were separately investigated by NMR spectroscopy. The mL of dry pentane and cooled to Ϫ30°C, was added 5.0 mmol
spectra of the upper layer showed no signals corresponding to
starting materials or to an adduct of bis(dichloroalumino)benzene
with benzyltriethylammonium chloride, but did exhibit signals for
Me3SnCl. In the spectra of the lower layer signals corresponding
to the presence of the benzyltriethylammonium cation and consist-
ent with the formation of the complex 22 were observed. Ϫ 1H
NMR (360.1 MHz, C6D6, 25°C): δ ϭ 0.65 (t, CH3CH2N), 1.08 (s,
SnCH3), 2.13 (q, CH3CH2N), 3.27 (s, C6H5CH2N), 6.80Ϫ6.90 (m,
C6H4), 7.14Ϫ7.25 (m, C6H4). Ϫ 13C NMR (90.6 MHz, C6D6,
of diethylaluminum chloride in 30 mL of hexane. A pale yellow
precipitate appeared almost immediately. The reaction suspension
was stirred and brought to room temperature over 60 min. After
removal of the volatiles in vacuo, the residue was found to be insol-
uble in alkanes, benzene or CDCl3 but readily dissolved in THF or
ether. The 1H-NMR spectrum in THF showed the absence of 28
and the presence of free Me3SnCl. Furthermore, in [D8]THF the
broaded triplet and quadruplet of the ethyl groups of 16 appeared
at δ ϭ 1.3 and 0.8, consistent with coordination of THF, rather
25°C): δ ϭ 5.9 (SnCH3), 7.5 (CH3CH2N), 52.4 (CH3CH2N), 60.5 than Me3SnCl, at aluminum.
(C6H5CH2N), 125.6 (C6H5), 128.5 (C6H4), 129.3 (C6H4), 129.8
(C6H5), 131.3 (C6H5), 132.0 (C6H5).
1,2-Bis(dimethylalumino)benzene (18): To a solution of 6.0 g (14.9
mmol) of 1,2-bis(trimethylstannyl)benzene (17)[1] in 20 mL of hep-
Displacement of Two Equivalents of Me3SnCl by Benzyltriethylam-
tane was added 2.76 g (29.8 mmol) of pure dimethylaluminum chlo-
monium Chloride. ؊ Complex 23: To a suspension of 340 mg (1.50 ride by syringe. The resulting solution was stirred at room tempera-
mmol) of benzyltriethylammonium chloride in 2 mL of toluene was ture for 72 h and thereupon the solvent and the Me3SnCl by-prod-
added 2.5 mL of a 0.3  solution (0.75 mmol) of the 2:1 complex uct were evaporated in vacuo without heating. The residue was
of Me3SnCl and 1,2-bis(dichloroalumino)benzene (21) in toluene
at room temperature. Stirring of the reaction mixture overnight re-
sulted in the formation of two layers. While the upper layer was a
pale yellow solution, the lower layer was a bright yellow oil. The
separated upper layer was shown by NMR spectroscopy to contain
only Me3SnCl. The separated lower layer was washed with pentane
redissolved in heptane and volatiles again removed under reduced
pressure; this addition and evaporation of heptane was repeated to
remove all traces of residual Me3SnCl. The residual viscous pale
yellow oil was essentially pure 1,2-bis(dimethylalumino)benzene
(18) in 89% yield, as evidenced by spectral data. Ϫ 1H NMR (360.1
MHz, C6D6, 25°C): δ ϭ Ϫ0.41 (br., 6 H, AlCH3), Ϫ0.32 (br., 6 H,
160
Eur. J. Inorg. Chem. 1999, 153Ϫ162

Organometallic Compounds of Group 13, 55
FULL PAPER
AlCH3), 6.96 (dd, 2 H, C6H4), 7.96 (2 H, dd, C6H4). Ϫ 13C NMR
(90.6 MHz, C6D6, 25°C): δ ϭ Ϫ7.7 (AlCH3), 6.4 (br., AlCH3),
129.7 (C6H4), 144.9 (C6H4), 160.3 (C6H4). Ϫ 27Al NMR (78.2
they belong to the triclinic crystal system and the space group of
Pl, No. 1. The absorption corrections were made by the ψ-scan
method. Observed reflections were defined as those reflections with
MHz, C6D6, 15°C): δ ϭ 174 (W1/2 ϭ 3200 Hz); (Ϫ8°C): δ ϭ 179 I Ն 2σ(I). Only these were used in the solution and refinement of
(W1/2 ϭ 5200 Hz).
the structure (F refinement). For refinement based on F2 all reflec-
tions were used. Refinement was done by least-squares; the quan-
tity minimized was Σ w(ԽYoԽ Ϫ ԽYcԽ)2, with Y ϭ F or F2. Compu-
tations were done using VAX, Sun4, and Silicon Graphics com-
puters. In addition to several locally written programs, the follow-
ing software was used: S. L. Lawton, R. A. Jacobson, TRACER
(cell reduction), United States Energy Commission, Report IS-
1141, Iowa State University, USA, 1965. P. Coppens, L. Leisero-
witz, D. Rabinovich, DATAP (data reduction), Acta Crystallogr.
1965, 18, 1035. G. M. Sheldrick, SHELXS-86 (crystal structure
determination), Crystallographic Computing 3 (Eds:. G. M. Sheld-
rick, C. Krüger, R. Goddard), Clarendon Press, Oxford, 1985, dp.
175. G. M. Sheldrick, SHELXL-93 (least-squares refinement), Uni-
versity of Göttingen, 1993. E. Egert, G. M. Sheldrick, PATSEE
(structure solution), Acta Crystallogr. 1985, A41, 262. P. Main, S.
J. Fiske, S. E. Hull, L. Lessinger, G. Germain, J.-P. Declercq, M. M.
Woolfson, MULTAN80 (structure solution), University of York,
England and Louvain, Belgium, 1980. W. R. Busing, K. O. Martin,
H. A. Levy, GFMLX, a highly modified version of ORFLS (full-
matrix least-squares refinement), Report ORNL-TM-305, Oak
Ridge National Laboratory, Tennessee, USA, 1962; H. D. Flack,
(enantiomorph-polarity estimation), Acta Crystallogr. 1983, A39,
876. P. Roberts, G. M. Sheldrick, XANADU (calculation of best
planes, torsion angles and idealized hydrogen atom positions), Uni-
versity of Cambridge, England, 1976. R. E. Davis, D. R. Harris,
DAESD (calculation of distances and angles), Roswell Park Mem.
Inst., USA, 1970. V. Schomaker, K. N. Trueblood, RIGID (rigid-
body analysis) Acta Crystallogr. 1968, B24, 63. C. K. Johnson, OR-
TEP (thermal ellipsoid plot program), Report ORNL-5138, Oak
Ridge National Lab., Tennessee, USA, 1976. E. Keller, SCHAKAL
(molecular drawing), Chem. Unserer Zeit 1986, 20, 178. B. W. van
der Waal, FSYN (slant plane Fourier syntheses), Technical Univer-
sity of Enschede, Netherlands, 1975. SYBYL (molecular calcu-
lations), Tripos Associates, Inc., St. Louis. INSIGHT/DISCOVER,
Biosym Technologies, San Diego, USA. Crystallographic data (ex-
cluding structure factors) for the structure reported in this paper
have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC-103346. Copies of
the data can be obtained free of charge on application to CCDC,
12 Union Road, Cambridge CB2 1EZ, UK [Fax: int. code ϩ
44(1223)336-033; E-mail: deposit@ccdc.cam.ac.uk]
1,2-Bis[chloro(methyl)alumino]benzene (19): To a solution of 6.0 g
(14.9 mmol) of 1,2-bis(trimethylstannyl)benzene in 20 mL of tolu-
ene were added 29.8 mmol of a solution of methylaluminum di-
chloride in hexane at room temperature. After stirring for 72 h, the
solvents were evaporated in vacuo. In order to remove the more
tightly complexed Me3SnCl, the procedure of adding 20 mL of
toluene to the reaction residue and evaporating all volatiles at 25°C
in high vacuum had to be performed four times. The residue thus
obtained was a pale yellow waxy substance (85% yield) whose
NMR spectra corroborate that it is rather pure 19, existing as a
1
bridging isomeric mixture. Ϫ H NMR (360.1 MHz, C6D6, 25°C):
δ ϭ 0.25 (br., AlCH3, 2 H), 0.8 (br., AlCH3, 4 H), 6.8Ϫ7.3 (m, br.,
C6H4, 2 H), 7.4Ϫ7.9 (m, br., C6H4, 2 H). Ϫ 13C NMR (90.6 MHz,
C6D6, 25°C): δ ϭ Ϫ6.6 (AlCH3), 4.3 (br., AlCH3), 129.5 (br.,
C6H4), 137.0 (C6H4), 150.4 (C6H4). Ϫ 27Al NMR (78.2 MHz,
C6D6, 25°C): δ ϭ 101 (W1/2 ϭ 5000 Hz).
Benzylaluminum Dichloride (14):[36] A solution of 7.6 g (30.0 mmol)
of benzyl(trimethyl)tin in 200 mL of mesitylene was treated with
30 mL of a 1.0  solution of methylaluminum dichloride in hexane
and the resulting clear solution was stirred at 20°C for 15 h and
then heated at 100Ϫ110°C for 3 h. The mesitylene was slowly eva-
porated at 100°C under a low pressure and caught in a cold trap.
1
By H- and 13C-NMR spectroscopy the mesitylene was shown to
contain tetramethyltin but not trimethyltin chloride. Spectral analy-
sis of the residual viscous oil showed it to be pure benzylaluminum
dichloride (14), free of any organotin by-product. Ϫ 1H NMR
(C6D6): δ ϭ 2.15 (s, 2 H), 6.71 (br. s, 2 H), 6.95 (t, 1 H), 7.10 (t, 2
H). Ϫ 13C NMR (C6D6): δ ϭ 29 (CH2 by DEPT), 123.6, 127.8,
127.86, 127.93.
Allyl(methyl)aluminum Chloride (9):[37] To a solution of 1.87 mL
(6.0 mmol) of allyltri-n-butyltin and 1.13 mL (5.8 mmol) of tri-
methyl(phenylethynyl)silane (10) in 6.0 mL of toluene, cooled to
Ϫ78°C, was added dropwise 6.1 mL of a 1.0  solution of MeAlCl2
in hexane. The reaction mixture was slowly brought to 0°C over 4
h. Hydrolytic work-up with aqueous NaOH and ether extraction
of the organic product allowed the isolation of a mixture of tri-n-
butyltin chloride and (Z)-2-phenyl-1-trimethylsilyl-1,4-pentadiene;
no starting tin compound or 10 remained. The adduct 12 was
cleanly separated from the nBu3SnCl by column chromatography
1
on silica gel with hexane eluent (90% yield). Ϫ H NMR (CDCl3):
δ ϭ Ϫ0.15 (s, 9 H), 3.15 (d, 2 H), 5.00 (m, H), 5.59 (s, H), 5.84 (m,
H), 7.15Ϫ7.35 (m, 5 H).[7]
116.41, 126.97, 127.75, 127.98, 135.91, 144.06, 157.43. Ϫ When a
similar reaction mixture was worked up with NaOD in D2O, the
1H singlet at δ ϭ 5.59 was absent in the spectrum of 12.
Ϫ
13C NMR (CDCl3): δ ϭ 0.0, 46.76,
Acknowledgments
This research has received its principal financial support from the
former Schering AG of Bergkamen, Germany and its successor,
Witco GmbH. We wish to thank the respective directors of re-
search, Dr. Peter Borner and Dr. Wolfram Uzick, for moral sup-
port and helpful technical discussions. The final research efforts
were supported by funding from the National Science Foundation
and a Senior Scientist Award to the principal investigator from the
Alexander von Humboldt Foundation. Valuable technical assist-
ance was provided by Dr. Boguslaw Kotowicz and Dr. Jürgen
Weber, as well as by Ms. Leza Luchetta.
X-ray Crystallographic Study of Trimethylstannyl Tetrachloroalu-
minate, Me3SnAlCl4 (25). ؊ Crystal Mounting and Data Collection:
A pale yellow-brown crystal of 25, 0.25 ϫ 0.53 ϫ 0.42 mm in size,
was mounted with a perfluorinated polyether oil on the tip of a
glass fiber on the goniometer head and cooled immediately by an
argon current to Ϫ173°C. Intensity data were collected with an
Enraf-Nonius CAD4 diffractometer by a coupled ω-2θ scan tech-
nique with speeds varying from 1.0 to 10.0°/min depending on the
standard deviation to intensity ratio of a preliminary 10°/min scan.
The radiation employed was Mo-Kα with a graphite monochroma-
tor. The crystals of molecular weight 332.6 g molϪ1 have 6 mol-
[1]
J. J. Eisch, B. W. Kotowicz, Eur. J. Inorg. Chem. 1998, 761. In
this article the authors availed themselves of the pioneering re-
search insights of the Nöth group into the course of boron-tin
ecules per unit cell and have a calculated density of 1.97 g cmϪ3
;
Eur. J. Inorg. Chem. 1999, 153Ϫ162
161

J. J. Eisch, K. Mackenzie, H. Windisch, C. Krüger
FULL PAPER
[16]
[17]
exchange processes: H. Nöth, H. Vahrenkamp, J. Organomet.
Chem. 1968, 11, 399.
G. E. Coates, K. Wade, Organometallic Compounds, 3rd ed., vol.
2, Methuen, London, 1967, pp. 304Ϫ319.
H. Lehmkuhl, K. Ziegler, Methoden Org. Chem. ( Houben-
Weyl) , vol. XIII/4, 1970, pp. 264Ϫ266.
W. P. Neumann, The Organic Chemistry of Tin, Wiley-Intersci-
ence, 1970, pp. 20Ϫ32.
M. B. Hossain, J. L. Lefferts, K. C. Molloy, D. van der Helm,
J. J. Zuckerman, Inorg. Chim. Acta 1979, 36, 1409.
J. L. Lefferts, K. C. Molloy, M. B. Hossain, D. van der Helm,
J. J. Zuckermann, J. Organomet. Chem.1982, 240, 349.
E. O. Schlemper, D. Britton, Inorg. Chem. 1966, 5, 507.
R. A. Forder, G. M. Sheldrick, J. Organomet. Chem. 1970, 21,
115.
[2]
[18]
[19]
[3]
[4]
[20]
[21]
[22]
B. Beagley, K. McAloon, J. M. Freeman, Acta Crystallogr.
1974, B30, 444.
[5]
The chemical consequences of the principle of Le Chatelier on
A. G. Davies, H. J. Milledge, D. C. Puxley, P. J. Smith, J. Chem.
Soc. A 1970, 2862.
chemical reactions at equilibrium, when subjected to changes in
temperature, pressure and/or concentration of a reaction part-
ner, are clearly and succinctly analyzed in the textbook: N. Wi-
berg, Holleman-Wiberg: Lehrbuch der Anorganischen Chemie,
91Ϫ100th ed., de Gruyter, Berlin, 1985, pp. 202Ϫ215.
W. P. Neumann, R. Schick, R. Köster, Angew. Chem. Int. Ed.
1964, 3, 385.
R. A. Forder, G. M. Sheldrick, J. Organomet. Chem. 1970, 22,
611.
[23]
[24]
N. C. Baenziger, Acta Crystallogr. 1951, 4, 216.
W. Hanstein, H. J. Berwin, T. G. Traylor, J. Am. Chem. Soc.
1970, 92, 829.
[6]
[25]
R. B. Woodward, R. Hoffmann, The Conservation of Orbital
Symmetry, Verlag Chemie, Weinheim, 1970, pp. 114Ϫ140.
J. J. Eisch, J. E. Galle, unpublished studies.
J. J. Eisch, C. K. Hordis, J. Am. Chem. Soc. 1971, 93, 2974.
J. J. Eisch, M. W. Foxton, J. Organomet. Chem. 1968, 11, P7.
G. Zweifel, R. B. Steele, J. Am. Chem. Soc. 1967, 89, 2754.
J. J. Eisch, G. A. Damasevitz, J. Org. Chem. 1976, 41, 2214.
J. J. Eisch, W. Uzick, K. Mackenzie, S. Gürtzgen, R. Rieger, U.
S. Patent 5,726,332, Mar. 10, 1988 (Chem. Abstr. 1998, 128,
192778r).
J. J. Eisch, A. M. Piotrowski, S. K. Brownstein, E. J. Gabe, F.
L. Lee, J. Am. Chem. Soc. 1985, 107, 7219.
J. J. Eisch, K. R. Caldwell, S. Werner, C. Krüger, Organometal-
lics 1991, 10, 3417.
[7]
U. Schöllkopf, Methoden Org. Chem. ( Houben-Weyl) , vol. XIII/
1, 1970, pp. 127Ϫ133.
[26]
[27]
[28]
[29]
[30]
[31]
[8]
Product 12 has also been the exclusive adduct obtained when
10 is allowed to react with η3-allylbis(cyclopentadienyl)titanium
and then treated with H2O or D2O: J. J. Eisch, M. P. Boleslaw-
ski, J. Organomet. Chem. 1987, 334, C1.
[9]
This method of removing the most volatile component, here
Me4Sn, from such aluminum-tin exchange equilibria has great
promise for the synthesis of a variety of uncomplexed RAlCl2
reagents from RSnMe3 and MeAlCl2, such as 20.
H. Nöth, R. Rürlander, P. Wolfgardt, Z. Naturforsch. 1982,
37B, 29.
T. Birchall, V. Manivannan, J. Chem. Soc., Dalton Trans.
1985, 2671.
J. B. Lambert, B. Kuhlmann, J. Chem. Soc., Chem. Commun.
1992, 931.
[32]
[33]
[34]
[35]
[36]
[37]
[10]
[11]
J. J. Eisch, S. I. Pombrik, G. X. Zheng, Organometallics 1993,
12, 3856.
J. J. Eisch, Organometallic Syntheses, vol. 2, Academic Press,
New York, 1981, pp. 1Ϫ84.
[12]
[13]
The helical sinuosity in the structure of solid 25 has prompted
Dr. Jürgen Weber conducted this experiment and established
the unexpected aluminum-tin exchange depicted in Equation 6.
An undergraduate research assistant in our research group,
Leza Luchetta, was first in realizing this novel generation of the
allylating agent 9.
one of us (C. K.) to view such a structure as a “DNA strand
for tin soldiers”.
[14]
I. Lange, J. Krahl, P. G. Jones, A. Blaschette, J. Organomet.
Chem. 1994, 474, 97.
I. Lange, D. Henschel, A. Wirth, J. Krahl, A. Blaschette, J.
Organomet. Chem. 1995, 503, 155.
[15]
Received August 31, 1998
[I98296]
162
Eur. J. Inorg. Chem. 1999, 153Ϫ162