The Ever-Elusive Tetra-ferf-butylethene (TTBE, 3,4-Di-ferf-butyl- 2,2,5,5-tetramethylhex-3-ene): Further insight on its preparation

Article abstract of DOI:10.1002/ejoc.200801292

Diketone 4 was prepared by hydrolysis of hexendiyne 5,the latter being available through McMurry coupling of acetyl- enic ketone 26. Upon treatment with dimethyltitanium dichlo- ride, 4 cyclizes into fully alkylated furan derivative 30;the still unknown t

Full text of DOI:10.1002/ejoc.200801292

FULL PAPER
DOI: 10.1002/ejoc.200801292
The Ever-Elusive Tetra-tert-butylethene (TTBE, 3,4-Di-tert-butyl-
2,2,5,5-tetramethylhex-3-ene): Further Insight on Its Preparation[‡]
Oliver Klein,[a] Henning Hopf,*[a] and Jörg Grunenberg[b]
Keywords: Diynes / Methylation / Titanium / Oxygen heterocycles / Cyclization
Diketone 4 was prepared by hydrolysis of hexendiyne 5, the
tyl-2,2,5,5-tetramethylhex-3-ene) was not produced. The
latter being available through McMurry coupling of acetyl-
structural properties of 4 and the mechanism of its cyclization
enic ketone 26. Upon treatment with dimethyltitanium dichlo- into 30 are discussed.
ride, 4 cyclizes into fully alkylated furan derivative 30; the
still unknown tetra-tert-butylethene (1, TTBE, 3,4-di-tert-bu-
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
three methyl groups of a tert-butyl substituent completely,
in an umbrella-like fashion (and mentally adds a carbon
atom), formally one of the sleekest groups in organic chem-
istry results: the ethynyl function. Performed four times on
1, tetraethynylethene (2, TEE) results. Although available
by the efforts of Diederich and coworkers,[7] we, several
years ago, tried to use its (formal) hydration product, tetra-
acetylethene (3), an easily available tetraketone,[8] to prepare
the title compound. Subjecting it to the well-established
method of Reetz for the conversion of methyl ketones into
tert-butyl moieties[9] disappointingly resulted in the genera-
tion of dioxabicyclo[3.3.0]octene derivative 6 only.[10]
Introduction
Looking at the synthetic accomplishments of natural[2]
and non-natural products chemistry,[3] it is surprising that
a simple and symmetrical hydrocarbon such as tetra-tert-
butylethene (TTBE, 1, 3,4-di-tert-butyl-2,2,5,5-tetrameth-
ylhex-3-ene) so far has escaped all efforts of preparation.[4]
According to recent DFT calculations, 1 should be a stable,
though strained (calculated strain energy ca. 93 kcal/mol)[5]
molecule; however, it is obviously not the energy content of
the final molecule that cannot be reached during its (at-
tempted) synthesis, but the steric resistance that so far no
synthetic pathway could overcome en route and which
forced respective intermediates to break away from the in-
tended synthetic course. Although the attempt described in
the present report adds to the still-growing list of failures
to synthesize 1,[6] we believe that our new effort deserves
publication, as it increases our knowledge about the prepa-
ration of highly hindered organic molecules.
Scheme 1 summarizes some of our strategic thoughts
concerning the preparation of 1. Because of the high strain
energy of the hydrocarbon (see above), it makes sense to
“smuggle” the strain-producing tert-butyl substituents into
the target molecule through smaller and hence less strain-
producing substituents. Although this “tied-back” ap-
proach has been previously unsuccessfully attempted sev-
eral times,[4] it could well be that in earlier attempts this
approach had not been taken far enough. If one “folds” the
Scheme 1. Some strategic considerations to prepare tetra-tert-bu-
tylethene (TTBE, 1).
[‡] Highly Hindered Olefins and Polyolefins, XIII. Part XII: Ref.[1a]
[a] Institut für Organische Chemie, Technische Universität
Braunschweig,
Clearly, to prevent the cyclization, the two vicinal acetyl
groups in the Z configuration have to be avoided, and
rather than employing 3 as a precursor, diethynyl derivative
5 and diketone 4 derived therefrom are better candidates
for the generation of 1. We hence decided to find a prepara-
tively satisfactory route to 4 [(3E)-3,4-di-tert-butylhex-3-en-
2,5-dione].
Hagenring 30, 38106 Braunschweig, Germany
Fax: +49-531-391-5388
E-mail: H.Hopf@tu-bs.de
[b] Institut für Organische Chemie, Technische Universität
Braunschweig, Arbeitsgruppe Computerchemie,
Hagenring 30, 38106 Braunschweig, Germany
Fax: +49-531-391-7744
E-mail: J. Grunenberg@tu-bs.de
Eur. J. Org. Chem. 2009, 2141–2148
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2141

O. Klein, H. Hopf, J. Grunenberg
FULL PAPER
Note that in all intermediates the final double bond is perature and could not be subjected to X-ray crystal struc-
always present during the complete synthetic sequence, ture analysis. However, it was shown by AM1 calculation
which avoids the introduction of unsaturation at a late stage that its two carbonyl groups were rotated out of the C–C
of the synthesis where the accessibility of the inner carbon double bond plane by ca 84°, causing severe steric hin-
atoms for any (attacking) reagent becomes increasingly dif- drance for any further nucleophilic attack (see below). That
ficult.
the conjugation in 11 is reduced is also indicated by the
electronic spectrum of the compound, which shows its ab-
sorption maximum at 222 nm (see the Experimental section
for other spectroscopic data). As these calculations show,
the orientation of the oxygen atoms of the dialdehyde is
very similar to the position of the two bromine substituents
in 9.[1b]
Results and Discussion
But-2-yn-1,4-diol Derivatives as Starting Materials for
(3E)-3,4-Di-tert-butylhex-3-en-2,5-dione (4)
As 11 was still lacking two carbon atoms for the prepara-
tion of 4, we tried to introduce them by the addition of
methyllithium (2 equiv.). Although reaction took place in
diethyl ether, we were unable to isolate a component from
the complex reaction mixture formed that showed spectro-
scopic data agreeing with structure 12. With methyl Grig-
nard, the results were only slightly better. After extensive
chromatography we could enrich at least a fraction that
gave the correct molecular ion peak for 12 (m/z = 228) and
displayed signals for the carbinyl protons in the expected
range (δ = 4.9 ppm, 3J = 6.2 Hz). However, after submitting
Our first attempt to prepare diketone 4 is summarized in
Scheme 2. Commercially available 1,4-dichlorobut-2-yne (7)
was converted into 2,3-di-tert-butylbuta-1,3-diene (8) by
treatment with tert-butylmagnesium bromide according to
a literature procedure in good yield.[11] When bromine was
added to hydrocarbon 8 at ice-bath temperature, a product
mixture was produced[12] from which 1,4-addition product
9 could be isolated. Although the yield of the process is
only fair (ca. 41%) it fulfills two important requirements:
the ultimately needed “interior” carbon–carbon double
bond is generated and the two bulky substituents are in a
trans orientation, as demonstrated earlier by X-ray struc-
tural analysis[1b] (Scheme 2).
this mixture to Swern oxidation, neither the 1H nor the 13
C
NMR spectrum showed any signs that 12 had indeed been
produced.
Obviously, we tried to introduce the methyl substituents
at too late a stage in the synthesis, when steric hindrance
was already significant; therefore, we decided to begin the
next route from a C6 precursor, hex-3-yn-2,5-diol (13), com-
mercially available as
a
mixture of diastereomers
(Scheme 3).
By treatment with either thionyl chloride (neat, –30 °C,
65% yield) or acetyl chloride (in N,N-diethylaniline, quanti-
tative yield) 13 was converted into the known dichlorides[15]
and diacetates,[16] respectively. tert-Butylation then took
place with the reagent prepared from cuprous chloride and
lithium bromide[17] to furnish a mixture of conjugated
dienes 15 and 16, as described previously.[18] The two hydro-
carbons were produced in a 3:2 ratio, with the E,E isomer
lacking completely.
Unfortunately, all bromination experiments were unsuc-
cessful and in no case (see Scheme 3 for variation of reac-
tion conditions) could desired 17 be isolated. Under the
first two variants,[19] the starting material was reisolated un-
changed, even when the temperature was raised to ambient
Scheme 2. 2,3-Di-tert-butylbuta-1,3-diene (8) as a possible interme-
diate to 1.
With the 8Ǟ9 conversion, the end of the dry spell of the temperature. For bromination in the presence of aluminum
synthetic sequence was not yet over. Hydrolysis of dibro- tribromide[20] and iron tribromide,[21] as well as with dibro-
mide 9 with the ion-exchange resin Amberlyst A 26, a rea- moisocyanuric acid (DIB),[22] more or less complete de-
gent which in its sodium carbonate form has been employed struction of the starting material was observed. In these ex-
very successfully for the hydrolysis of primary alkyl, allyl, periments, the color of the reaction mixture turned quickly
and benzyl halides,[13] furnished 10 in only 18% yield. For- from deep red to brown and finally to black, and the forma-
tunately, its Swern oxidation yielded dialdehyde 11 in good tion of hydrogen bromide was noted; no defined reaction
yield (73%) so that altogether 1–2 g of this crucial interme- products could be isolated. It is well known[23] that orthog-
diate were available from 9. The conversion of the latter onal dienes prefer substitution of olefinic hydrogen atoms
derivative into 11 by Kornblum reaction[14] gave the dialde- over the 1,2-addition to the alkene moiety. As far as oligo-
hyde directly, but in very disappointing yields of 5% only. mers from 15/16 are concerned, it could be that 17 is actu-
The dialdehyde is a slightly yellow, oily liquid at room tem- ally formed but – as a highly reactive halide – is converted
2142
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2141–2148

Further Insight on the Preparation of TTBE
Scheme 3. Further conjugated dienes as intermediates in the synthesis of 1.
into a triene under the reaction conditions by dehydrobro- ketone 20 in quantitative yield.[25] McMurry coupling
mination. This hydrocarbon could easily oligomerize in the (TiCl4, Zn) furnished a mixture of dimers 21 and 22 in poor
presence of hydrogen bromide and a Lewis acid.
yield (15–20%, ratio 2:1). That rearranged products such as
Be it as it may, after these further negative experiments 22 are produced in this process has been noted previously[24]
we terminated our experiments to prepare 4 from doubly and traced back to the presence of propargylic carbene in-
propargylic substrates and turned our attention to a termediates that can dimerize along different routes. Desi-
McMurry-type coupling reaction of suitably substituted lylation was performed by treatment with potassium car-
acetylenic ketones.
bonate and yielded hexendiynes 5 and 23. Because the sepa-
ration of this mixture was difficult, it was subjected directly
to hydration to yield 4 (37%) and 24 (36%), two ketones
that could be obtained in pure form easily by silica-gel col-
umn chromatography with pentane as eluent. Before the
physical and chemical properties of 4 are described, a vari-
tert-Butyl Ethynyl Ketones as Starting Materials for (3E)-
3,4-Di-tert-butylhex-3-en-2,5-dione (4)
In 1994 we reported on the preparation of various hex- ant of the above coupling will be presented that circumvents
3-en-1,5-diynes by McMurry coupling of appropriate alky- the generation of product mixtures; it is presented in
nones.[24] Among the hydrocarbons synthesized was di- Scheme 5.
acetylene 5 for which the pathway summarized in Scheme 4
was developed.
Applying the acylation with 19 to the TIPS-protected
acetylene 25 [TIPS = tris(isopropylsilyl)][26] furnished acetyl-
This route, optimized in the present study, begins with enic ketone 26, again in excellent yield. However, when this
the Friedel–Crafts acylation of bis(trimethylsilyl)ethyne (18) was reductively dimerized, only desired endiyne 27 was pro-
with pivaloyl chloride (19) to yield protected acetylenic duced. The structure of this important intermediate was de-
Scheme 4. McMurry coupling of acetylenic ketones.
Eur. J. Org. Chem. 2009, 2141–2148
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2143

O. Klein, H. Hopf, J. Grunenberg
FULL PAPER
Because unique assignment of individual conformers on
the exclusive basis of the above NMR experiment is not
trivial, we resorted to theoretical methods. A combination
of force-field simulations (conformational analysis) and
quantum chemical calculations at the DFT level of theory
was performed to assess the stability of the two conformers
(see the Experimental Section for details). The result is
shown in pictorial form in Figure 1.
Scheme 5. (3E)-3,4-di-tert-butylhex-3-en-2,5-dione (4) by a specific
McMurry coupling.
rived from its spectroscopic data (see Experimental Sec-
tion). It was confirmed qualitatively by X-ray structure de-
termination, but was beset by disorder problems, which pre-
vented satisfactory refinement.
Deprotection under harsher conditions than those above
yielded hydrocarbon 5 in good yield (81%), and when desi-
lylation and hydration were carried out in a one-pot reac-
tion diketone 4 was produced in 39% yield. Using this im-
proved route, 4, a slightly yellow solid (m.p. 98 °C), was
obtained in 50 to 100 mg lots, enough for further experi-
mentation.
Figure 1. Optimized gas-phase structures for both rotamers of 4 at
the M05–2X/tz level of theory (see text). Both conformers, the C2
symmetric (left) and the Ci symmetric (right), exhibit approximately
the same energy. The acetyl functions are rotated ca. 85° out of the
plane defined by the central, slightly elongated double bond
[1.338 Å C2 rotamer; 1.340 Å Ci rotamer; experimental data[28] of
the Ci rotamer: central double bond: 1.344 (3) Å, rotation of acetyl
function: 78.71(6)°].
The Structure of (3E)-3,4-Di-tert-butylhex-3-en-2,5-dione
(4)
The M05–2X functional in combination with a polarized
triple zeta basis set [6-311+G(d,p)][27] was used, because we
are dealing with many noncovalent intramolecular interac-
tions. It turned out that in the gas phase both isomers have
nearly the same energy with a very tiny preference
(0.05 kcal/mol) in favor of the Ci symmetric (transoid; Fig-
ure 1, right) conformer. Whereas this should lead to a 1:1
mixture in the gas phase (assumed that the barrier of in-
terconversion is not too high), any slightly polar medium
should favor the C2 symmetric conformer, which has quite
a substantial dipole moment of 4.3 debye computed at the
same level of theory as mentioned above (Figure 1, left). A
tiny difference in the energies of solvation for both rotamers
of ca. 0.5 kcal/mol would be enough to explain the mea-
sured 2.8:1 equilibrium. Furthermore, we must include en-
tropic contributions. As the C2 symmetric conformer is chi-
ral, both enantiomers contribute to the total free energy.
The entropy of racemization again favors the C2 symmetric
conformer by ca. 0.41 kcal/mol at room temperature (∆G
= –RTln2). We hence propose that the major conformer in
CDCl3 is the syn conformer (syn-4).
Structure in Solution
The structure (conformation) of 4 in deuteriochloroform
is revealed by its NMR spectra. In both the proton and the
carbon spectra all signals are doubled, giving a first hint
that 4 exists in two conformations. From the integration of
the proton signals a conformer ratio of 2.8:1 is derived.
That, in fact, we are dealing with rotamers, and not with
constitutional isomers, is shown by the same connectivity
in the C,H-correlation as well as the COLOC spectra. The
protons of the tert-butyl groups (conformer I: δ = 1.15 ppm;
conformer II: δ = 1.17 ppm) display couplings with the car-
bon atoms to which they are directly bound (conformer I:
δ = 30.9 ppm; conformer II: δ = 34.1 ppm). Furthermore,
the COLOC spectrum shows cross peaks of these hydrogen
atoms through two bonds to the quaternary carbon atoms
of the (CH3)3C substituents (conformer I: δ = 34.6 ppm;
conformer II: δ = 34.1 ppm, respectively) and through three
bonds to the carbon atoms of the C–C double bond (con-
former I: δ = 144.7 ppm; conformer II: δ = 145.5 ppm). The
hydrogen atoms of the acetyl groups (conformer I: δ =
2.397 ppm; conformer: δ = 2.404 ppm) show direct coup-
In the solid state 4 prefers the anti conformation as re-
lings to the corresponding carbon atoms (C,H correlation, ported by us in an earlier publication.[28] The acetyl func-
COLOC; conformer I: δ = 35.0 ppm; conformer II: δ = tions are rotated ca. 80° out of the plane defined by the
34.8 ppm) and in addition cross peaks through two bonds central, slightly elongated double bond [1.344(3) Å]. To
with the signals of the carbonyl group (conformer I: δ = avoid steric repulsion, the C–C(tert-butyl) single bonds are
208.0 ppm; conformer: δ = 208.4 ppm).
markedly longer [1.555(2) Å] than in noncrowded ethylenes.
2144
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2141–2148

Further Insight on the Preparation of TTBE
Scheme 6. Dimethylation of (3E)-3,4-di-tert-butylhex-3-en-2,5-dione (4).
Cyclization of (3E)-3,4-Di-tert-butylhex-3-en-2,5-dione (4)
also observed in the GC–MS of the product mixture pro-
duced when dibromodi-tert-butylmethane was treated with
sodium or potassium.[4,29]
Application of the quaternization method of Reetz[9] to
4 produced a complex mixture from which we could isolate
by silica-gel column chromatography only one defined
product: fully alkylated furan derivative 30, formed in 34%
yield as a colorless, not very stable oil (Scheme 6).
Conclusions
Although the conditions for preparing 1 from 4 were, in
our opinion, ideal (i.e., trans orientation of the two acetyl
functions and an excellent, general method for producing
tert-butyl groups from acetyl ketones[9]), our substrate pre-
ferred to cyclize to furan derivative 30, evidently a conse-
quence of the reduction of the double bond barrier in inter-
mediate 28/29. Because this abolishment is unavoidable in
this approach to 1 we are terminating our efforts to reach
this synthetic goal by alkylation of 3 or 4 or related com-
pounds.
The structure of 30 follows from its spectroscopic data
(see Experimental Section) and in particular from its NMR
spectral data {1H NMR: δ = 1.43 (s, 18 H, tert-butyl),
2.34 ppm (s, 6 H, CH3). 13C NMR: δ = 143.8 (s, C-2), 129.1
(s, C-3), 33.7 [q, C(CH3)3], 31.8 [s, C(CH3)3], 17.6 ppm (q,
CH3)} and the mass spectrum with a molecular ion peak at
m/z = 208 (42%). These signals were found to change with
time when 30 was kept at room temperature for extended
periods of time, accompanied by a color change from color-
less via red to finally black.
We propose that 30 is generated from 4 by the steps illus-
trated in Scheme 6. The cyclization begins, as discussed in
the literature,[9] with the formation of adduct 28 of dimethyl
dichlorotitanium and 4. In this complex the central double
bonds between the olefinic carbon atoms is replaced by an
allylic system, thus allowing rotation between C-3 and C-4.
Whether a coplanar arrangement as indicated in 29 ↔ 32
is really achieved is doubtful, as it puts the two bulky sub-
stituents into the least favorable orientation. However, even
if only partial rotation, that is, into an orthogonal arrange-
ment of the two tert-butyl moieties, is reached, it brings the
uncomplexed oxygen atom close to the developed positive
charge, allowing cyclization as illustrated by the closure of
32 to 31. In the final step, 31Ǟ30, one oxygen atom of the
substrate is transferred to the titanium atom.
Column chromatography also yielded a minute amount
of an (impure) fraction that displayed a singlet at δ =
1.25 ppm in its proton spectrum, and a quartet in the 13C
spectrum at δ = 29.6 ppm; 13C signals in the olefinic region
could not be detected. Although in the mass spectrum a
signal was registered with the correct mass for 1 [m/z =
252 (4.3%)], we consider our evidence as too weak to claim
Experimental Section
General: Chromatography: TLC: Polygram Sil G/UV 254 (Mach-
erey–Nagel); CC: Kieselgel 60 (70–230 mesh, Merck) and alumi-
num oxide (neutral, activity III–IV, Woelm). M.p. Kofler hot stage
(uncorr.). 1H and 13C NMR: in deuteriochlorform (int. TMS),
Bruker AC 200 (200.1 and 50.3 MHz, resp.), Bruker AM 400 (400.1
and 100.6 MHz). IR: KBr pellets or film (neat), Nicolet 320 FT-
IR. UV/Vis: if not noted otherwise in acetonitrile, Beckman UV
5230. MS: EI at 70 eV, Finnigan MAT 8430. GC–MS: fused silica
capillary column, Carlo–Erba HRGC 5160/Finnigan MAT 4515
(EI, 40 eV).
Computational Details: The Merck Molecular Force Field
(MMFF) implemented in the MacroModel software package
(MacroModel 9.0; Maestro 8.0 interface) was used in a first step in
order to scan the conformational space by a Monte Carlo torsional
sampling procedure. The resulting two stable conformers were fur-
ther optimized by quantum chemical methods. We employed the
M05–2X meta hybrid functional and a 6-311+G(d,p) basis set im-
plemented in the Gaussian03 program, characterizing the two sta-
tionary points as minima by frequency calculations.
Bromine Addition to 8: To a solution of 8 (75.0 g, 0.45 mol) in car-
generation of the title compound. A peak of this mass was bon tetrachloride (250 mL) at 0 °C was added bromine (72.3 g,
Eur. J. Org. Chem. 2009, 2141–2148
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2145

O. Klein, H. Hopf, J. Grunenberg
FULL PAPER
23.3 mL, 0.452 mol) in carbon tetrachloride (100 mL) keeping the
temperature at 0 °C. After stirring for 1 h at this temperature, the
solvent was removed in vacuo; compound 9 precipitated and was
removed by vacuum filtration. After washing with cold petroleum
ether, adduct 9 (60.4 g, 41%) was obtained as slightly yellow need-
les. M.p. 92–93 °C. 1H NMR (200.1 MHz): δ = 1.42 [s, 18 H, (CH3)3-
After careful extraction with ether the organic phases were com-
bined, dried (sodium sulfate), and the solvents were removed in
vacuo. The remaining oil was purified by column chromatography
(silica gel) to yield a mixture of 15 and 16 (9.3 g, 44%). Further
purification by silica gel chromatography provided two main frac-
tions: fraction 1, 3.2 g, consisting largely of 16, and fraction 2, 5.9 g
C], 4.1–4.9 (m, 4 H, CH2) ppm. 13C NMR (50.3 MHz): δ = 31.3 of a mixture of 15 (39%) and 16 (61%). Data for 15: 1H NMR
3
(s, Cq), 32.0 (q, CH3), 34.8 (t, CH2), 110.8 (s, olefinic C) ppm. IR
(200.1 MHz): δ = 1.09 [s, 18 H, (CH3)3C], 1.47 (d, J = 6.74 Hz, 6
(KBr): ν = 2965 (s), 1210 (s), 634 (s) cm–1. MS: m/z (%) = 326 (1.4)
H, CH3), 5.58 (q, J = 6.74 Hz, 2 H, =CH–) ppm. Data for 16: H
NMR (200.1 MHz): δ = 1.02 [s, 9 H, (CH3)3C], 1.12 [s, 9 H, (CH3)3-
3
1
˜
[M+], 247 (1.1), 165 (4.5), 83 (100), 57 (46).
C] 1.62 (d, J = 6.80 Hz, 3 H, CH3), 1.82 (d, 3J = 6.60 Hz, 3 H,
3
Preparation of 10
3
3
CH3) 4.98 (q, J = 6.80 Hz, 1 H, =CH–), 5.40 (q, J = 6.60 Hz, 1
H, =CH–) ppm. All other spectroscopic data as well as the prepara-
tion of 15 and 16 from diacetates 14b[16] are given in ref.[18] For
the attempted bromine addition to 15 and 16, see the Results and
Discussion.
Preparation of Amberlyst A-26 in the Na(CO3)– Form: Amberlyst
A-26 (chloride form, 34 g) was placed in a chromatography column
and sodium carbonate solution (1 , 2 L) was slowly passed
through. Subsequently the ion exchange resin was washed with
methanol, acetone, and diethyl ether and dried for 2 h under high
vacuum. Yield: 17 g A-26 in the sodium carbonate form.
Hydration of the Mixture of 5 and 23: To the reagent mixture pre-
pared from water (10 mL), mercuric sulfate (90 mg, 0.3 mmol), and
concentrated sulfuric acid (0.2 mL) was added at 60 °C a solution
of hexendiynes 5 (91 mg, 0.48 mmol) and 23 (58 mg, 0.23 mmol) in
THF (5 mL), prepared from 20 as described in the literature.[24]
After stirring at 60 °C for 3 h, the mixture was cooled to room
temperature, water (20 mL) was added, and the product mixture
isolated by careful extraction with diethyl ether. The organic phases
were combined, dried (sodium sulfate), and the solvent was re-
moved in vacuo. The remaining yellow solid was separated by col-
umn chromatography (silica gel, pentane/diethyl ether, 10:1) and
two fractions were isolated. Several recrystallizations (pentane/
methanol/dichloromethane) of fraction 1 afforded 4 (40.0 mg, 37%)
as pale-yellow needles. Fraction 2 afforded 24 (20.0 mg, 36%) as a
colorless oil. Data for 4: M.p. 98 °C. 1H and 13C NMR: see Results
Hydrolysis: To a suspension of A 26 (ca. 100 g) in THF (250 mL)
was added 9 (12.3 g, 38 mmol), and the mixture was stirred vigor-
ously for 6 h at room temperature. The resin was removed by fil-
tration and washed carefully with dichloromethane and methanol.
The combined organic washings were concentrated to ca. 50 mL,
and the solution was placed into an ice box for crystallization to
yield 10 (2.2 g, 18%) as colorless plates. M.p. 152 °C. 1H NMR
(400.1 MHz): δ = 1.33 [s, 18 H, (CH3)3C], 4.39 (m, 4 H, CH2) ppm.
13C NMR (100.6 MHz): δ = 32.77 (q, CH3), 36.56 (s, Cq), 60.85 (t,
CH ), 147.49 (s, –C=) ppm. IR (KBr): ν = 3286 (br. s), 2873 (s)
˜
2
1624 (m), 1399 (s), 1367 (s) cm–1. MS (70 eV, CI, NH3+): m/z (%)
= 200 (64) [M+], 183 (100), 144 (26), 58 (8).
Swern Oxidation of 10: A solution of oxalyl chloride (2.45 g,
1.68 mL, 19.3 mmol) in anhydrous THF (40 mL) was cooled to
–60 °C and a solution of DMSO (3.2 g, 2.98 mL, 42 mmol) in anhy-
drous dichloromethane (40 mL) was added within 15 min. After
stirring for 10 min, a solution of 10 (1.40 g, 7 mmol) in dry dichlo-
romethane (100 mL) was added over 40 min. When the addition
was complete the mixture was stirred for additional 10 min, fol-
lowed by the addition of triethylamine (19.9 g, 24.7 mL, 0.177 mol).
After stirring for 15 min, the mixture was warmed to room tem-
perature, water was added, and the organic phase carefully washed
with water. The dried (sodium sulfate) organic phase was concen-
trated to a few mL and the remainder was purified/separated by
column chromatography on alumina (petroleum ether/diethyl ether,
9:1) to yield 11 (1.0 g, 73%) as a yellow oil. 1H NMR (400.1 MHz):
δ = 1.17 [s, 18 H, (CH3)3C], 10.28 (s, 2 H, CHO) ppm. 13C NMR
(100.6 MHz): δ = 31.24 (q, CH3), 36.71 (s, Cq), 146.77 (s, –C=),
and Discussion. IR (KBr): ν = 2967 (s), 1686 (s), 1656 (m), 1470
˜
(m), 1372 (s), 1365 (s) cm–1. UV: λ (log ε) = 218 nm (3.21, sh.).
MS: m/z (%) = 224 (12) [M+], 209 (4), 181 (100), 167 (24), 125 (32),
57 (18), 43 (71). C14H24O2 (224.34): calcd. C 72.74, H 10.77; found
C 72.54, H 10.60. HRMS: calcd. for C14H24O2 224.1776; found
224.1770. Data for 24: 1H NMR (400.1 MHz): δ = 0.14 [s, 9 H,
(CH3)3Si], 1.26 [s,
9 H, (CH3)3C], 1.33 [s, 9 H, (CH3)3-
C], 2.24 (s, 3 H, CH3) ppm. 13C NMR (100.6 MHz): δ = 207.7 (s,
C-2), 168.0 (s, C-3), 119.4 (s, C-4), 113.3 (s, C-6), 79.4 (s, C-5), 35.
9 (s, Cq), 33.7 (q, CH3), 30.6 (q, CH3), 29.0 (q, CH3), 28.6 (s, Cq),
–0.4 ppm. [(CH ) Si]. IR (film): ν = 2968 (s), 1699 (s), 1656 (m),
˜
3 3
1471 (s), 1377 (s), 1363 (s) cm–1. UV: λ (log ε) = 244 nm (3.89).
MS: m/z (%) = 281 (12) [M+], 221 (74), 73 (100), 57 (22). HRMS:
calcd. for C17H33OSi 281.2301; found 281.2318.
1-Triisopropylsilyl-4,4-dimethyl-pent-1-yn-3-one (26): To a suspen-
sion of aluminum trichloride (1.35 g, 10.1 mmol) in pentane
(15 mL) was added at 0 °C under an atmosphere of nitrogen a solu-
tion of 25 (2.14 g, 8.4 mmol)[26] and 19 (1.07 g, 1.09 mL, 8.8 mmol)
in pentane (10 mL). After stirring for 45 min at 0 °C and 1 h at
room temperature, the reaction mixture was poured onto ice (6 g).
The organic phase was separated, and the aqueous phase was ex-
tracted several times with pentane. The combined organic phases
were washed with saturated. hydrogen carbonate solution and dried
(sodium sulfate). After solvent removal, the remaining brown oil
(2.96 g) was purified by silica gel column chromatography (pen-
tane/diethyl ether, 30:1) to afford 26 (2.05 g, 91%) as a slightly yel-
200.21 (d, CHO) ppm. IR (KBr): ν = 2968 (s), 2927 (s), 1752 (s),
˜
1592 (m) cm–1. UV: λ (log ε) = 222 nm (3.81). MS: m/z (%) = 196
(10) [M+], 195 (64), 167 (26), 139 (54), 57 (100). For the attempted
methylation of 11, see Results and Discussion.
tert-Butylation of 2,5-Dichloro-hex-3-yne (14a, Mixture of Isomers):
To a suspension of magnesium (16.4 g, 0.68 mmol) in anhydrous
THF (40 mL) was added a few drops of tert-butyl chloride, fol-
lowed by a few drops of 1,2-dibromoethane for activation. After
the reaction was started by slight warming, the main fraction of
tert-butyl chloride (73 mL, 62.5 g, 0.68 mmol) in THF (180 mL)
was added at such as rate as to keep the mixture under gentle re-
flux. To complete the Grignard formation, the solution was stirred
for 12 h at room temperature, and the now-clear, black solution
was transferred into a dropping funnel. At –35 °C, the Grignard
reagent was added to a mixture of cuprous bromide (0.385 g,
2.68 mmol) and 2,5-dichloro-hex-3-yne.[15] After stirring for an ad-
ditional 3 h at 0 °C, the mixture was hydrolyzed under ice cooling.
1
low oil. H NMR (400.1 MHz): δ = 1.05–1.18 [m, 21 H, (C3H7)3-
Si], 1.22 [s, 9 H, (CH3)3Si] ppm. 13C NMR (100.6 MHz): δ = 193.7
(s, C-3), 102.3 (s, C-2), 96.6 (s, C-1), 44.5 (s, Cq), 26.0 [q, (CH3)3-
Si], 18.4 (q) and 11.0 [d, (C H ) Si] ppm. IR (film): ν = 2963 (s),
˜
3
7 3
2144 (m), 1672 (s), 1477 (s), 1392 (s), 1367 (s) cm–1. UV: λ (log ε)
= 234 nm (3.79). MS: m/z (%) = 265 (8) [M+], 251 (100), 223 (58),
2146
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2141–2148

Further Insight on the Preparation of TTBE
209 (40), 181 (18), 85 (26), 57 (12), 43 (24). C16H30OSi (266.50): Some spectroscopic data not contradicting 1 as being part of this
calcd. C 72.11, H 11.35; found C 72.11, H 11.41.
mixture are discussed in the Results and Discussion section. Data
for 30: 1H NMR (400.1 MHz): δ = 1.43 [s, 18 H, (CH3)3C], 2.34 (s,
6 H, CH3) ppm. 13C NMR (100.6 MHz): δ = 143.8 (s, C-2), 129.1
(s, C-3), 33.7 [q, (CH3)3C], 31.8 [s, (CH3)3C], 17.6 (q, CH3) ppm.
McMurry Coupling of 26: To a mixture of titanium tetrachloride
(1.13 g, 0.65 mL, 6.0 mmol) and anhydrous tetrahydrofuran
(15 mL) was added under an atmosphere of nitrogen and ice cool-
ing zinc dust (714 mg, 10.9 mmol) and anhydrous pyridine (370 mg,
0.38 mL, 4.7 mmol). A solution of ketone 26 (1.25 g, 4.7 mmol) in
anhydrous THF (8 mL) was added, and the reaction mixture was
stirred for 30 min at 0 °C followed by further stirring at room tem-
perature for 2 h. For workup, the reaction mixture was hydrolyzed
(10% aq. potassium carbonate solution, 25 mL), the black precipi-
tate was removed by filtration through a Büchner funnel and
washed with pentane (100 mL). The precipitate was dissolved with
dilute hydrochloric acid, and the resulting aqueous phase was thor-
oughly extracted with diethyl ether. After neutralization with hy-
drogen carbonate solution, the organic phases were united and
dried (sodium sulfate). The solvent was removed in vacuo and the
resulting yellow oil (1.13 g) was purified by silica gel column
chromatography to afford 27 (0.73 g, 62%) as colorless needles.
M.p. 48 °C. 1H NMR (400.1 MHz): δ = 1.10 [m, 42 H, (C3H7)3Si],
1.39 [s, 18 H, (CH3)3C] ppm. 13C NMR (100.6 MHz): δ = 138.7 (s,
C-3), 108.8 (s, C-2), 106.8 (s, C-1), 35.7 (s, Cq), 30.0 [q, (CH3)3C],
IR (film): ν = 2957 (s), 1649 (w), 1470 (s), 1379 (s), 1364 (s), 1051
˜
(s) cm–1. UV: λ (log ε) = 266 (2.52), 216 nm (3.38, sh.). MS: m/z
(%) = 208 (42) [M+], 193 (100), 151 (20), 57 (25).
[1] a) R. Hänel, H. Hopf, P. G. Jones, Acta Crystallogr., Sect. E
2006, 62, o2095–o2096; b) Part XI: H. Hopf, P. G. Jones, R.
Hänel, P. Bubenitschek, Acta Crystallogr., Sect. E 2006, 62,
o1480–o1482; c) Part X: H. Hopf, C. Mlynek, D. Klein, M.
Traetteberg, P. Bakken, Eur. J. Org. Chem. 2001, 1385–1391.
[2] a) K. C. Nicolaou, E. J. Sorensen, Classics in Total Synthesis –
Targets, Strategies, Methods, Wiley-VCH, Weinheim, 1996; b)
K. C. Nicolaou, S. A. Snyder, Classics in Total Synthesis II –
More Targets, Strategies, Methods, Wiley-VCH, Weinheim,
2003.
[3] H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH,
Weinheim, 2000.
[4] Most recent summary: D. Lenoir, C. Wattenbach, J. F. Lieb-
man, Struct. Chem. 2006, 17, 419–422.
18.6 (q) and 11.5 [d, (C H ) Si] ppm. IR (KBr): ν = 2957 (s), 2127
˜
3
7 3
[5] H. M. Sulzbach, E. Bolton, D. Lenoir, P. v. R. Schleyer, H. F.
Schaefer III, J. Am. Chem. Soc. 1996, 118, 9908–9914.
[6] D. Lenoir, unpublished results; private communication.
(m), 1463 (s), 1392 (s), 1362 (s) cm–1. UV: λ (log ε) = 312 (4.38),
300 (4.42), 296 (4.42), 288 nm (4.34, sh.). MS: m/z (%) = 500 (40)
[M+], 457 (100), 157 (20). HRMS: calcd. for C32H60Si2 500.4234; [7] Y. Rubin, C. B. Knobler, F. Diederich, Angew. Chem. 1991, 103,
708–710; Angew. Chem. Int. Ed. Engl. 1991, 30, 698–700.
[8] a) R. G. Charles, Org. Synth. 1963, 4, 869–871; b) A. M. Celli,
L. R. Lampariello, S. Chimichi, R. Nesi, M. Scotton, Can. J.
Chem. 1982, 60, 1327–1332.
[9] M. T. Reetz, J. Westermannm, S. H. Kyung, Chem. Ber. 1985,
118, 1050–1057; cf.M. T. Reetz, Organotitanium Reagents in
Organic Synthesis, 1986, Springer, Heidelberg, pp. 210–214.
[10] J. Dannheim, W. Grahn, H. Hopf, C. Parrodi, Chem. Ber. 1987,
120, 871–872.
found 500.4218. C32H60Si2 (500.92): calcd. C 76.72, H 12.07; found
C 76.78, H 12.01.
Desilylation of 27: To a solution of tetrabutylammonium fluoride
(TBAF) in THF (8 mL, 8.0 mmol) was added water (0.2 mL,
200 mg, 11.1 mmol) and diacetylene 27 (0.951 g, 1.90 mmol). The
mixture was stirred for 44 h at room temperature, diethyl ether was
added, and the organic phase was washed carefully with water. Af-
ter drying (sodium sulfate), the solvent was removed in vacuo, and
the residue obtained was purified by column chromatography (sil-
ica gel, pentane) to afford 5 (0.290 g, 81%) identified by spectral
comparison with the authentic material.[24]
[11] L. Brandsma, J. Meijer, H. D. Verkruijsse, J. Chem. Soc., Chem.
Commun. 1980, 922–923; cf. O. Klein, Diploma Thesis,
Braunschweig, 1997.
[12] H. J. Backer, Recl. Trav. Chim. Pays-Bas 1939, 58, 643–661;
cf.H. J. Backer, J. Strating, Recl. Trav. Chim. Pays-Bas 1937,
56, 1069–1092; R. Hänel, PhD Dissertation, Braunschweig,
1996. For a recent investigation into the preparation of 9 by
bromine addition to 8, see: R. Bianchini, C. Chiappe, D.
Lenoir, B. Meyer, Gazz. Chim. Ital. 1995, 125, 453–458.
[13] G. Cardillo, M. Orena, G. Porzi, S. Sandri, Synthesis 1981,
793–794.
Desilylation/Hydration of 27: As described above, 27 (0.150 g,
0.30 mmol) was desilylated; however, to the reaction mixture were
added a few crystals of mercuric sulfate and concentrated sulfuric
acid (0.2 mL). After additional water (10 mL) was added, the mix-
ture was heated to reflux for 3.5 h. Extractive workup resulted in
a yellow solution (0.171 g), which by column chromatography on
silica gel yielded 4 (26 mg, 39%), identical in its spectroscopic prop-
erties with the data discussed above and in the main section.
[14] N. Kornblum, W. J. Jones, G. J. Anderson, J. Am. Chem. Soc.
1959, 81, 4113–4114.
[15] A. J. Jones, P. J. Stiles, Tetrahedron Lett. 1977, 18, 1965–1968.
[16] G. Dupont, R. Dulou, D. Lefort, Bull. Soc. Chim. Fr. 1951,
755–777.
Methylation of 4 with Dimethyltitanium Dichloride: A solution of
dimethylzinc in toluene (2 , 3.93 mL, 7.87 mmol) was injected into
anhydrous dichloromethane (10 mL) at –12 °C under an atmo-
sphere of nitrogen. To this solution was added titanium tetrachlo-
ride (0.87 mL, 1.49 g, 7.87 mmol) at such a rate as to hold the tem-
perature at –12 °C. After stirring for 30 min the reaction mixture
was cooled to –50 °C and a solution of 4 (55 mg, 0.25 mmol) in
anhydrous dichloromethane (10 mL) was added. The mixture was
stirred overnight while its temperature increased to ambient tem-
perature. For workup, the product mixture was poured into ice
water (250 mL), the aqueous phase was thoroughly extracted with
diethyl ether, and the combined organic fractions were dried with
sodium sulfate. After solvent removal by rotary evaporation the
yellow residue was purified by column chromatography (silica gel,
pentane) to provide two fractions: the first fraction afforded 30
(17 mg, 34%) as a colorless oil and a second fraction afforded a
minute amount (Ͻ5 mg) of sample as a mixture of components.
[17]
T. L. McDonald, D. R. Reagan, R. S. Brinkmeyer, J. Org.
Chem. 1980, 45, 4740–4747.
[18]
[19]
H. Hopf, H. Lipka, Chem. Ber. 1991, 124, 2075–2084.
Bromination with CuBr2 in acetonitrile: W. C. Baird, J. H. Sur-
ridge, M. Buza, J. Org. Chem. 1971, 36, 3324–3330.
D. E. Pearson, H. W. Pope, W. W. Hargrove, W. E. Stamper, J.
Org. Chem. 1958, 23, 1412–1414.
[20]
[21]
[22]
[23]
H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1969, 91, 3534–
3543.
Autorenkollektiv, Organikum, 19th ed., Deutscher Verlag der
Wissenschaften, Berlin, 1993, pp. 328–331.
a) H. Hopf, R. Hänel, P. G. Jones, P. Bubenitschek, Angew.
Chem. 1994, 106, 1444–1445; Angew. Chem. Int. Ed. Engl. 1994,
33, 1369–1370; b) H. Hopf, R. Hänel, M. Traetteberg, Nachr.
Chem. Techn. Lab. 1994, 42, 856–862 and references cited
therein.
Eur. J. Org. Chem. 2009, 2141–2148
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2147

O. Klein, H. Hopf, J. Grunenberg
FULL PAPER
[24]
[27] Y. Zhao, N. E. Schultz, D. G. Truhlar, J. Chem. Theor. Comput.
2006, 2, 364–382.
[28] O. Klein, I. Dix, H. Hopf, P. G. Jones, Acta Crystallogr., Sect.
C 1999, 55, 2078–2080.
[29] A. Krebs, W. Born, B. Kaletta, W.-U. Nickel, W. Rüger, Tetra-
hedron Lett. 1983, 24, 4821–4824.
W. R. Roth, H. Hopf, C. Horn, Chem. Ber. 1994, 127, 1781–
1795.
[25] Initially we performed the McMurry coupling with the unpro-
tected ketone and obtained a complex, inseparable product
mixture from which the expected dienyne could not be isolated.
[26] C. J. Helal, P. A. Magriotis, E. J. Corey, J. Am. Chem. Soc.
1996, 118, 10938–10939.
Received: December 29, 2008
Published Online: March 13, 2009
2148
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2141–2148