A novel approach towards intermolecular stabilization of para-quinone methides. First complexation of the elusive, simplest quinone methide, 4-methylene-2,5-cyclohexadien-1-one

Article abstract of DOI:10.1002/(SICI)1521-3765(20000204)6:3<454::AID-CHEM454>3.0.CO;2-Q

A novel approach towards the intermolecular stabilization of "simple" (i.e. methylene-unsubstituted) p-quinone methides (QMs) by their coordination to a transition-metal center is described. 4-Bromomethyl phenols, protected by a silyl group, were employed as the QM precursors and cis-chelating diphosphine Pd⁰ complexes were chosen as the metal precursors, since they have strong back-bonding interactions with the electron-poor QM moiety. Removal of the silyl protecting-group from the corresponding [LPd(benzyl)Br] complex (L=bisphosphine) with fluoride results in the spontaneous rearrangement of the unobserved zwitterionic Pd^II complex into the QM-Pd⁰ complex. The feasibility of this approach was demonstrated in the synthesis of the structurally characterized Pd⁰ complex of BHT-QM (4), a biologically relevant metabolite of 2,6-di-tert-butyl-p-cresol, and the synthesis of the complex of 4-methylene-2,5-cyclohexadien-1-one (11), the simplest, and so far unobserved QM molecule. These complexes exhibit a remarkable thermal stability and do not react with alcohol or water. In both cases, the use of an appropriate incoming ligand allowed the release of the coordinated QM into the reaction media in which it was effectively trapped by added nucleophiles.

Full text of DOI:10.1002/(SICI)1521-3765(20000204)6:3<454::AID-CHEM454>3.0.CO;2-Q

FULL PAPER
A Novel Approach Towards Intermolecular Stabilization of para-Quinone
Methides. First Complexation of the Elusive, Simplest Quinone Methide,
4-Methylene-2,5-cyclohexadien-1-one
Oded Rabin, Arkadi Vigalok, and David Milstein*[a]
Abstract: A novel approach towards
the intermolecular stabilization of ªsim-
pleº (i.e. methylene-unsubstituted) p-
quinone methides (QMs) by their coor-
dination to a transition-metal center is
described. 4-Bromomethyl phenols, pro-
tected by a silyl group, were employed as
the QM precursors and cis-chelating
diphosphine Pd0 complexes were chosen
as the metal precursors, since they have
strong back-bonding interactions with
the electron-poor QM moiety. Removal
of the silyl protecting-group from the
corresponding [LPd(benzyl)Br] com-
plex (L ˆ bisphosphine) with fluoride
results in the spontaneous rearrange-
ment of the unobserved zwitterionic PdII
complex into the QM ± Pd0 complex.
The feasibility of this approach was
demonstrated in the synthesis of the
structurally characterized Pd0 complex
of BHT-QM (4), a biologically relevant
metabolite of 2,6-di-tert-butyl-p-cresol,
and the synthesis of the complex of
4-methylene-2,5-cyclohexadien-1-one
(11), the simplest, and so far unobserved
QM molecule. These complexes exhibit
a remarkable thermal stability and do
not react with alcohol or water. In both
cases, the use of an appropriate incom-
ing ligand allowed the release of the
coordinated QM into the reaction media
in which it was effectively trapped by
added nucleophiles.
Keywords: ligand exchange reac-
tions
´ palladium ´ P ligands ´
quinone methides ´ synthesis design
Introduction
stabilizing substituents on the methylene group were report-
ed.[2] By means of an approach often applied to the
preparation of the closely related metal ± xylylene complexes,
Ir complexes of ortho-QMs, in which the metal center is
coordinated in an h4 fashion to the two endo-cyclic double
bonds, were recently prepared.[3] The quinonoid ring in the
resulting complexes exhibits a large distortion from planarity;
this is reflected in the reactivity of the QM moiety. We have
already reported the characterization of two Rh complexes of
simple QMs, in which the QM moiety is a part of a bis-
chelating PCP-type ligand system.[4] In none of the reported
metal complexes could the QM species be removed from the
coordination sphere of the metal.
QMs, especially the simple ones (i.e. those which do not
have substituents on the exocyclic methylene group), are
highly reactive organic intermediates, which are often very
difficult to isolate in a pure form as they rapidly polymerize
when their dilute solutions are concentrated.[5] Reactions with
the medium or self-condensations to give the corresponding
phenols are evidently driven by aromatization. Despite this
instability, reactive QMs have been extensively studied on
account of their importance in many chemical and biochem-
ical processes.[6] For example, the biosynthesis of the natural
polymers melanin and lignin involves para-QM intermedi-
ates.[7] The mechanism of action of several antitumor drugs is
believed to proceed via the formation of QMs.[8] Various QMs
act as alkylating agents and form covalent bonds with
Otherwise inaccessible reactive molecules or molecular frag-
ments can often be stabilized by coordination to metal
centers[1] to provide unique opportunities for their character-
ization by spectroscopic methods and the elucidation of their
structure by X-ray crystallography. Furthermore, under
appropriate conditions these species may be chemically
modified or displaced from the metal. Carbanions, carbenes,
ylides, xylylenes, cyclobutadienes, and benzyne are some
examples of elusive organic transients, the metal complexes of
which have been investigated. Surprisingly, metal complexes
of quinone methides (QMs) have so far attracted scarce
attention. OsII complexes of ortho-QMs that contained
O
R1
[a] Prof. Dr. D. Milstein, O. Rabin, A. Vigalok
Department of Organic Chemistry, The Weizmann Institute of Science
Rehovot 76100 (Israel)
Fax : (‡ 972)8-9344142
E-mail: comilst@wiccmail.weizmann.ac.il
454
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0454 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 3

454±462
functional groups of biopolymers.[9] Elegant syntheses of
natural products that involve transient para-QM intermedi-
ates have also been developed.[10] Moreover, studies with
alkyl-substituted phenols revealed that QMs are responsible
for the toxicity of these compounds.[11] Metabolic oxidation of
these phenols results in the formation of QM derivatives
which are attacked by cellular nucleophiles. Most of the
toxicology-related studies have involved the use of 2,6-di-tert-
butyl-4-methylphenol (butylated hydroxytoluene, BHT) and
its derivatives.[12] This compound is widely used as an
industrial antioxidant to prevent deterioration of food prod-
ucts, as a radical scavenger, and as an oxygen-reducing agent.
Thus, the metabolism of this compound and the toxicological
effects of its metabolites, primarily the quinone methide
derivative BHT-QM, are of great interest. Yet, the study of
the reactivity of QMs in biotic systems is seriously obstructed
by the instability of these compounds and their incompati-
bility with protic media.
We report here on a novel general approach towards the
generation, characterization, and reactivity studies of metal
complexes of QM. The complexation of simple quinone
methides and their controlled release have been accomplished
for the first time. Specifically, we report on the generation,
stabilization, and properties, including the controlled dis-
placement, of BHT-QM and the even more reactive, so far
elusive, unsubstituted QM analogue. Part of this work has
already been communicated.[13]
metal precursor
O
O
L
PG
PG
Br
M
Br
L
(A)
(B)
_
- PG
- Br
O
L
O
L
L
M
M
+
(C)
(D)
L
PG = protecting group
Scheme 1. Synthetic strategy to produce h2 methylene-coordinated, p-
quinone methide-metal complexes.
Synthesis of the BHT-QM palladium complex: The precursor
selected for BHT-QM was the corresponding trimethylsilyl
(TMS) ether of 4-bromomethyl-2,6-di-tert-butylphenol (1,
Scheme 2). The phenolic function of the commercially
O
O Si
Br
BHT-QM
Scheme 2. BHT-QM and its precursor 1.
1
Results and Discussion
available 2,6-di-tert-butyl-4-methylphenol (BHT) was silyl-
ated by refluxing with hexamethyldisilazane in DMF.[15] The
resulting silyl ether was brominated with NBS in CCl4 under
intense illumination to give 1. Compound 1 is a white powder
which can be stored in the solid state at room temperature for
weeks with only slight decomposition. The benzylic CH2Br
General considerations: Our synthetic strategy is based on the
notion that the desired product, the h2 methylene-coordinat-
ed, p-quinone methide-metal complex, can be also regarded
as a zwitterionic h1-methylene-p-phenoxy metal complex
(Figure 1). The outline of the synthesis is depicted in
1
group in 1 gives rise to a singlet at d ˆ 4.49 in the H NMR
spectrum and a singlet in the 13C NMR spectrum at d ˆ 35.27.
Initially, we tried to oxidatively add the benzyl bromide 1 to
chelated diphosphine Pd0 complexes. Reaction of
[(dppp)Pd(dba)][16] (dba ˆ dibenzylidene acetone; dppp ˆ
1,3-bis(diphenylphosphino)propane)) with three equivalents
of 1 at room temperature in acetone resulted in oxidative
addition; however, the final product could not be separated
from DBA by extraction with common organic solvents. The
separation from DBA appeared to be crucial in later stages
(vide infra). The complex [(dippp)Pd(dba)] (dippp ˆ 1,3-
bis(diisopropylphosphino)propane) did not react with three
equivalents of 1 under the same conditions, probably because
of the tighter binding of DBA in this electron-rich complex.
When the mixture was heated at 908C, the product from
the oxidative addition was obtained in low yield (ꢀ10%). The
major product was [(dippp)PdBr2],[17] as detected by 31P NMR
spectroscopy. We attributed this reactivity to electron-transfer
processes that involve the electron-rich zero-valent metal
center and the benzyl bromide. Because of these drawbacks,
we decided to apply a slightly different approach that utilized
PdII precursors. It was demonstrated that benzyl bromide
oxidatively adds to [(tmeda)PdMe2] (TMEDA ˆ N,N,N',N'-
-
O
O
+
M
M
quinone methide form
zwitter-ionic form
Figure 1. Quinone methide form and the zwitterionic form of p-quinone
methide in the metal complex.
Scheme 1. The oxidative addition of a QM precursor A to a
metal complex would result in complex B. This complex is
then modified to generate the zwitterionic complex C, which
is expected to rearrange to the final QM metal complex D.
Since 4-(bromomethyl)phenols, the QM precursors, are
unstable and tend to eliminate HBr to form the corresponding
unstable QMs, the phenolic group was protected by a silyl
group. cis-Chelating diphosphine Pd complexes were chosen
as the metal precursors, since they are expected to have strong
backbonding interactions with the electron-poor QM moiety
in the final product D, which enhances the stability of the
complex.[14]
Chem. Eur. J. 2000, 6, No. 3
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0455 $ 17.50+.50/0
455

D. Milstein et al.
FULL PAPER
Me
N
Me
Me
O
Si
79%
+
CH3CH3
Pd
N
Br
Me
2
Me
Me
Me
N
+
Pd
O Si
Br
N
Me
Me
Me
1
Me
N
Me
Me
21%
Me
Br
O
Si
+
Pd
N
Me
Scheme 3. Reaction of 1 with [(tmeda)PdMe2].
tetramethyl-1,2-ethanediamine) to give, after reductive elim-
ination of ethane, [(tmeda)Pd(Br)(CH2Ph)] in good yield.[18]
Exchange of tmeda by an appropriate bisphosphine was
expected to be a facile step.
product was observed. The 31P NMR spectrum showed a
singlet at d ˆ 1.56 and a broad signal at 21.70. Apparently,
the small size of the phosphine ligand allows more than one
molecule to interact with the metal center.
Indeed, reaction of 1 with [(tmeda)PdMe2] resulted in a
mixture of the benzyl complex 2 (79%) and [(tmeda)Pd-
Me(Br)] (21%) (Scheme 3). Complex 2 was readily extracted
with diethyl ether and was fully characterized by NMR
spectroscopy. The CH2 group bound to the metal center gives
rise to a singlet at d ˆ 2.84 in the 1H NMR spectrum and to a
singlet at d ˆ 14.84 in the 13C NMR spectrum. The two amine
groups of the tmeda ligand are chemically nonequivalent and
show four signals in both 1H and 13C NMR spectra. The benzyl
bromide complex 2 was treated with a series of chelating
bisphosphine ligands in order to exchange the tmeda ligand.
The most efficient ligand-exchange reaction was obtained with
dppe (bis(diphenylphosphino)ethane) (Scheme 4), and the
conditions were optimized to achieve the quantitative and clean
Remarkably, when 3 was treated with one equivalent of
nBu4NF in THF, the quinone methide complex
[(dppe)Pd(BHT-QM)] (4) was obtained in 91% yield
(Scheme 5). The ammonium phenoxy intermediate formed
by the cleavage of the Si O bond, spontaneously dispropor-
tionated to give 4 and nBu4NBr. Alternatively, when 3 was
(nBu)4N+
-
O
O
-
(nBu)4N+
-Me3SiF
F
Ph2
P
3
Ph2
P
-(nBu)4NBr
Pd
P
Ph2
Pd
Br
P
Ph2
4
Scheme 5. Formation of the quinone methide ± metal complex 4 with
nBu4NF.
Me
N
Ph
P
Me
Ph
Ph
O Si
O Si
THF, dppe
-30°C
Pd
Pd
N
Me
P
Ph
Br
Br
treated with one equivalent of silver trifluoromethanesulfo-
nate (triflate, AgOTf) in wet THF in the dark, complexes 5
and 6 were formed (Scheme 6) as evidenced by 31P and
1H NMR analyses. Complex 5 is completely hydrolyzed into 6
within 48 hours at room temperature. Complexes 5 and 6
possess similar NMR features (see Experimental Section),
except that the 1H NMR spectrum of 6 lacks the pattern
characteristic of the silyl protecting-group. Addition of an
equimolar amount of base (tBuOK) to a solution of 6 in THF
resulted in the clean formation of 4 (Scheme 6).
Me
3
2
Scheme 4. Exchange reaction of the tmeda ligand with dppe.
formation of the bisphosphine benzyl complex 3. The 31P
NMR spectrum of 3 in C6D6 exhibits two doublets of equal
intensities at d ˆ 30.12 and 52.90 [J(P,P) ˆ 40.6 Hz], corre-
sponding to the two nonequivalent phosphorus atoms of the
chelating ligand. The benzylic protons appear in the 1H NMR
spectrum as a doublet of doublets at d ˆ 3.94 [J(P,H) ˆ 10.5
and 5.4 Hz] as a result of coupling to the phosphines. Good
exchange results were also obtained with dippp, dppp, and
dtbpp (1,3-bis(di-tert-butyl-phosphino)propane), but other
minor products, which appeared as singlets in the 31P NMR
spectra, were also formed. When dmpe (1,2-bis(dimethyl-
phosphino)ethane) was added to 2, no desired exchange
The 31P NMR spectrum of 4 in C6D6 shows two sharp
doublets at d ˆ 29.07 and 37.49 [J(P,P) ˆ 14.2 Hz], indicative
of two nonequivalent phosphorus nuclei in a mutual cis
configuration. The upfield 13C chemical shift of the carbons of
the coordinated double bond and their coupling to the two
nonequivalent phosphines, d(CH2) ˆ 51.34 [d, J(P,C) ˆ
ˆ
30.8 Hz], and d(C CH2) ˆ 82.27 [dd, J(P,C) ˆ 12.5 and
456
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0456 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 3

Quinone Methides
454±462
Selected bond distances and bond angles are given in
Table 1. The Pd atom lies in a distorted trigonal-planar
environment, coordinated to the two phosphorus atoms and
the exocyclic double bond of the BHT-QM moiety. The
Ph
P
Ph
Ph
AgOTf
O
Si
3
Pd
P
OTf
Table 1. Selected bond lengths [Š] and bond angles [8] for the crystal
structure of 4.
Ph
5
Pd1 C1
Pd1 C2
C1 C2
C2 C3
C3 C4
C4 C5
C5 C6
C6 C7
C2 C7
C5 O5
C4 C8
2.088(2)
2.208(2)
1.437(2)
1.443(3)
1.362(2)
1.474(3)
1.474(3)
1.364(3)
1.441(2)
1.251(2)
1.539(3)
P2-Pd1-P3
C1-Pd1-C2
C2-C1-Pd1
C1-C2-Pd1
C1-C2-C3
C7-C2-C3
C2-C3-C4
C3-C4-C5
C3-C4-C8
C4-C5-C6
C4-C5-O5
86.73(2)
38.97(6)
75.06(10)
65.97(10)
122.0(2)
116.7(2)
123.3(2)
118.6(2)
122.5(2)
118.1(2)
120.9(2)
H2O
Ph
Ph
Ph
OH
P
tBuOK
Pd
4
P
OTf
Ph
coordinated double bond is notably elongated (1.437(2) Š)
relative to that of free olefins, as a result of substantial
backdonation from the metal. The quinonoid character of the
ligand is reflected in the alternation of the ring bonds,
whereby the C3 C4 and C6 C7 bond lengths of 1.362(2) Š
and 1.364(3) Š, respectively, are noticeably shorter than the
rest of the carbon ± carbon bonds of the ring (cf. C2 C3
1.443(3) Š and C4 C5 1.474(3) Š). The carbonyl C5 O5
bond length of 1.251(2) Š is in the range reported for other
quinonoid compounds.[20] It is noteworthy that the strong
backbonding from the metal center to the QM moiety results
in the loss of planarity around C2; the C1 C2 exocyclic bond
is displaced out of the ring plane by 10.788 away from the
palladium atom. A similar observation is reported for the
structurally relevant [(PMe3)2Pd(pentamethylfulvene)],
where the coordinated exocyclic double bond is displaced
from the plane of the fulvene ring by 10.818.[21, 22]
Complex 4 is a thermally stable compound and can be
stored under a dinitrogen atmosphere for months. Remark-
ably, it is also stable towards
6
Scheme 6. Alternative formation of the quinone methide ± metal complex
4 with AgOTf and subsequent hydrolysis.
4.8 Hz], as well as the doublet of doublets that originates from
1
the benzylic protons in the H NMR spectrum at d ˆ 3.40
[J(P,H) ˆ 7.2 and 3.8 Hz], indicate that coordination takes
place through the exocyclic double bond. The carbonyl carbon
is also coupled to the phosphorus atoms and appears as a
broad doublet at d ˆ 183.96, which is in the region observed
for other 2,5-cyclohexadienones and quinones.[19] The carbon-
yl group of the QM moiety exhibits a characteristic, strong IR
1
absorption band at nÄ ˆ 1598 cm , similar to the one reported
for PCP pincer-type QM RhI complexes.[4]
Slow diffusion of pentane into a solution of 4 in diethyl
ether resulted in orange crystals that were subjected to an
X-ray diffraction study, which has already been communicat-
ed.[13] A view of a molecule of 4 is shown in Figure 2.
gentle heating (558C) in ben-
zene or even in wet MeOH,
which clearly indicates that
there is no spontaneous disso-
ciation of the QM moiety from
Pd. Free BHT-QM would have
reacted immediately with the
media were there any equilibria
present involving the [P2Pd]
fragment and the QM moiety.
Exposure of a benzene solution
of 4 to air resulted in the total
decomposition with quantita-
tive formation of dppe oxide
within 2 h. This oxidation pro-
cess is apparently catalyzed by
the metal,[23] since only partial
oxidation of dppe in air was
observed after heating a solu-
tion of dppe at 808C for
Figure 2. Crystal structure of 4.
8 hours. Substitution of the
Chem. Eur. J. 2000, 6, No. 3
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0457 $ 17.50+.50/0
457

D. Milstein et al.
FULL PAPER
BHT-QM fragment from the metal center in 4 could not be
achieved by the use of hard donor ligands, such as acetonitrile
or pyridine; complex 4 remained unchanged. In contrast,
reaction of 4 with the electron-deficient alkene dibenzyli-
deneacetone (DBA) or with diphenylacetylene (DPA) resulted
in the clean formation of the corresponding [P2Pd(dba)] and
[P2Pd(dpa)] complexes. The unstable free BHT-QM was
detected in a C6D6 reaction solution by 1H NMR spectroscopy
immediately after its release. When the same experiments
were performed in methanol as the solvent, the free QM was
trapped immediately with the formation of the 1,6-Michael-
type adduct, 2,6-di-tert-butyl-4-methoxymethylphenol, which
was detected by 1H NMR spectroscopy and GC-MS
(Scheme 7). Similarly, when CD3OD was used as the solvent,
a CD3O group was incorporated into the organic product.
Thus, the controlled release of free QM from the metal into
solution, in which it is effectively trapped by nucleophiles, was
achieved for the first time.
Si
O
Si
O
OH
Si
Cl
NBS
Br
7
Scheme 8. Synthesis of 4-(bromomethyl)phenyl (thexyldimethyl)silyl
ether (7).
O Si
7
Br
Me
N
Me
Me
O Si
8
Pd
Br
N
Me
O
OH
DBA
MeOH
4
-[(dppe)Pd(dba)]
dppe
dtbpp
OMe
BHT-QM
Ph
P
Ph
Ph
Scheme 7. Release of BHT-QM which is immediately trapped by MeOH.
P
P
O Si
O Si
Pd
Br
Pd
Br
P
Ph
9
10
Synthesis of the s-QM palladium complex: Isolation of the
simplest (that is, unsubstituted) para-quinone methide (s-
QM) can be regarded as the ultimate goal for a study which
deals with stabilization of quinone methides, since it is
considered as one of the most reactive members of this family
of compounds. Indeed, this species reacts immediately upon
formation and has never been characterized.[24] We decided to
utilize the strategy outlined above to try to isolate a stable
s-QM palladium complex. As in the case of BHT-QM, the
trialkylsilyl ether of 4-(bromomethyl)phenol was chosen as
the precursor for this quinone methide. The corresponding
4-(bromomethyl)phenyl (trimethyl)silyl ether was synthe-
sized by silylation of para-cresol with Me3SiCl followed by
bromination of the product. Unfortunately, this precursor, a
transparent liquid substance, rapidly decomposed to give a
purple film. Presumably, hydrolysis by traces of moisture
results in the unstable 4-(bromomethyl)phenol, which under-
goes HBr elimination and formation of the highly reactive
s-QM. In order to increase the stability of the QM precursor,
we decided to use bulkier silyl protecting-groups.[25] Reaction
of p-cresol with thexyldimethylchlorosilane (thexyl ˆ 1,1,2-
trimethylpropyl) and subsequent bromination by NBS gave
4-(bromomethyl)phenyl (thexyldimethyl)silyl ether (7)
(Scheme 8). Neat benzyl bromide 7 also showed extensive
decomposition under ambient conditions, although at a slower
rate than that of the TMS analogue. Therefore, 7 was
generated in a benzene solution (ꢀ0.2m) and was kept at
308C under nitrogen. The benzene was removed under
vacuum just before 7 was used.
(nBu)4N+ F-
(nBu)4N+ F-
?
P
Pd
P
O
11
Scheme 9. Formation of the novel Pd0 complex of s-QM (11).
8 and the corresponding diphosphine Pd-benzyl complexes
(vide infra) undergo slow decomposition at room temper-
ature, probably as a result of hydrolysis of the silyl protecting-
group. The 1H NMR spectrum of 8 in C6D6 shows a singlet at
d ˆ 3.36 for the benzylic protons and two doublets for the
aromatic protons [d ˆ 6.77 and 7.83, J(H,H) ˆ 8.5 Hz]. Treat-
ment of 8 with dppe (1 equiv) afforded the bisphosphine
palladium complex 9. As with the related complex 3, the 31P
NMR spectrum of 9 in [D6]benzene consists of two doublets
characteristic of nonequivalent phosphine groups coordinated
1
in cis positions. The H NMR spectrum shows a doublet of
doublets for the aromatic protons of the benzyl group, one of
which shows a small splitting as a result of coupling with 31P.
The benzylic protons appear as a doublet of doublets at d ˆ
3.93 [J(P,H) ˆ 10.6 Hz and 5.0 Hz].
Reaction of [(tmeda)PdMe2] with 7 gave the corresponding
benzyl bromide adduct 8 in 35% yield (Scheme 9). Complex
Reaction of 9 with nBu4NF led to the formation of a new
bisphosphine palladium complex. The 31P{1H} NMR spectrum
458
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0458 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 3

Quinone Methides
454±462
indicates the presence of an AB spin system [d ˆ 38.62 and
56.42, J(P,P) ˆ 24.6 Hz]. The signal of the benzylic protons is
strongly shifted upfield to d ˆ 1.21 and is also split by coupling
to the two nonequivalent phosphorus atoms [dd, J(P,H) ˆ 8.0
and 4.0 Hz]. No signals of the thexyldimethylsilyl group are
observed. The IR spectrum of this compound does not contain
can be attributed to the steric hindrance introduced by the
tert-butyl groups on the phosphorus atoms. These groups
block the approach of a new ligand to the metal, which is a
requirement for the associative ligand-substitution pathway.
When a solution of 11 in methanol was kept under a CO
atmosphere for 12 h; discoloration of the orange solution was
observed, and the 31P{1H} NMR spectrum indicated the
complete disappearance of the s-QM complex 11. GC-MS
analysis of the mixture revealed the presence of 4-methoxy-
methyl phenol, the product from the addition of MeOH to
s-QM. When ethanol was used as the solvent, the ethoxy
analogue was detected; this indicates that the s-QM release
can be achieved by the use of an appropriate ligand and the
released QM species is effectively trapped in protic media to
give the 1,6-Michael-type addition products.
1
any sharp absorption bands in the range n ˆ 1570 ± 1670 cm ,
which indicates the absence of a carbonyl group. While the
nature of this complex is not yet clear, its spectroscopic
features show no similarities to those obtained for the
[(dppe)Pd(BHT-QM)] complex. Since the only difference
between 9 and 3 is the presence of two bulky tert-butyl groups
on the QM moiety, the difference in the products obtained
from the reactions of these complexes with nBu4NF may be
the result of the availability of the phenolate group for
coordination or for further reaction in the case of 9.
In order to avoid the formation of nonquinonoid products
by the interaction of the phenolate group with the metal
center, steric hindrance was introduced into the system by the
use of the dtbpp ligand instead of dppe. Thus, addition of
dtbpp (1 equiv) to a THF solution of 8 gave the benzyl
complex 10 in 52% yield. The 31P{1H} NMR spectrum of 10
exhibits two doublets, corresponding to the phosphorus atoms
of dtbpp coordinated in cis positions. The benzylic protons
appear as a broad doublet at dH ˆ 3.95 [J(P,H) ˆ 8.1 Hz] and
one of the doublets of the aromatic AB spin system is
broadened as a result of coupling to the phosphorus atoms.
Addition of nBu4NF to a THF solution of 10 results in the
removal of the silyl protecting-group and the clean formation
of the novel Pd0 complex of s-QM 11 (Scheme 9). This
complex shows two doublets in the 31P{1H} NMR spectrum for
its two nonequivalent phosphorus atoms coupled to each
other with a relatively low coupling constant of J ˆ 16.1 Hz.
Its 1H NMR spectrum exhibits, in addition to the signals from
the dtbpp ligand, a doublet of doublets at d ˆ 2.76, which
corresponds to the exocyclic CH2 group that has shifted
upfield. A similar shift is observed for the ring protons, which
resonate at d ˆ 6.58 and 6.91. The latter signal is coupled to
one phosphorus atom, and appears as doublet of doublets.
The 13C{1H} NMR spectrum gives rise to a broad singlet at d ˆ
184.36, attributed to the quinonoid carbonyl group. The
exocyclic carbon appears as a doublet at d ˆ 46.89 [J(P,C) ˆ
33.3 Hz]. The IR spectrum of 11 shows a strong absorption at
Conclusions
We have developed a general approach towards the synthesis
of simple p-quinone methide metal complexes. These highly
reactive organic transients are generated in the coordination
sphere of the metal and are stabilized by coordination of the
exocyclic double bond of the quinone methide moiety to an
electron-rich palladium(0) center. This coordination results in
a remarkable increase in stability and inertness towards
alcohol and water. Under appropriate conditions, the QM
moiety can be released and trapped. The generality and
efficiency of the method was demonstrated by the generation,
stabilization, full characterization, and controlled release of
the biologically relevant quinone methide BHT-QM and of
the simplest, and so far elusive, representative of the quinone
methide family, 4-methylene-2,5-cyclohexadien-1-one.
Experimental Section
General procedures: All operations with air- and moisture-sensitive
compounds were performed in
a nitrogen-filled glovebox (Vacuum
Atmospheres with an MO-40 purifier). All solvents were of reagent grade
or better. Pentane, diethyl ether, benzene, and THF were distilled over
sodium/benzophenone ketyl. Acetone was dried by filtration through
neutral alumina. All solvents were degassed and stored under high-purity
nitrogen after distillation. All deuterated solvents (Aldrich) were stored
under high-purity nitrogen on molecular sieves (3 Š). Commercially
available reagents were used as received. [(tmeda)PdMe2] was prepared
as previously reported.[26]
1
1
nÄ ˆ 1587 cm (with a shoulder at 1608 cm ) that originates
ˆ
from the C O stretching mode of the conjugated carbonyl.
Complex 11 is thermally stable in solution and in the solid
state. A sample kept under nitrogen atmosphere for two
months showed no substantial decomposition. A benzene
solution of 11 (3 mg, 1 mL) slowly reacted with ambient
dioxygen to yield the bisphosphine oxide (approximate
halflife time 2 days). Remarkably, the s-QM moiety in
complex 11 is stable towards alcohols, whereas the free
s-QM has no measurable life time in such media. Moreover,
no displacement of the s-QM from the complex was achieved
by addition of a large excess of DBA or methyl acrylate to an
ethanol solution of 11, even upon heating for 1 h at 558C. The
observation that the s-QM could not be released from the
complex [(dtbpp)Pd(s-QM)] under conditions which allowed
the displacement of BHT-QM from its dppe-Pd complex 4,
1H, 31P, and 13C NMR spectra were recorded on a Bruker DPX250
spectrometer. 1H and 13C chemical shifts are reported in ppm downfield
from TMS and referenced to the residual solvent h1 ([D]chloroform
7.24 ppm, [D6]acetone 2.04 ppm, [D6]benzene 7.15 ppm) and all-d solvent
peaks (chloroform 77.00 ppm, acetone 29.8 ppm, benzene 128.00 ppm),
respectively. 31P chemical shifts are reported in ppm downfield from H3PO4
and referenced to an external 85% phosphoric acid sample. All measure-
ments were performed at 218C. Infrared spectra were recorded on a FT-IR
Â
Â
Nicolet PROTEGE460 spectrophotometer. IR samples were prepared in
the glovebox as films on NaCl plates. The elemental analyses were
performed at the Hebrew University of Jerusalem (Israel).
Synthesis of [4-(bromomethyl)-2,6-di-tert-butylphenoxy)]trimethylsilane
(1)
Synthesis of the intermediate (2,6-di-tert-butyl-4-methylphenoxy)trimethyl-
silane:
A solution containing 2,6-di-tert-butyl-4-methylphenol (5.2 g,
0.023 mol) and 1,1,1,3,3,3-hexamethyldisilazane (15 mL, 0.071 mol) in
Chem. Eur. J. 2000, 6, No. 3
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0459 $ 17.50+.50/0
459

D. Milstein et al.
FULL PAPER
DMF (35 mL) was refluxed under argon atmosphere for 20 h. The solution
turned blue as the reflux started. Upon concentration by distillation under
reduced pressure, a white precipitate was formed that was collected on a
sinter, washed with DMF, and dried under vacuum to give the pure
trimethylsilyl ether of 2,6-di-tert-butyl-4-methylphenol[15] (3.2 g, 47%), as
two sets of signals corresponding to the triflate complex with the TMS
protecting-group on the benzylic group (5) and the triflate complex with an
unprotected phenol (6). The benzene solution was kept at room temper-
ature and complete conversion of 5 to 6 was observed within 48 h (4 mg,
ꢀ50%). Addition of equimolar tBuOK to a THF solution of 6 resulted in
the clean formation of 4, as observed by 31P and 1H NMR spectroscopy.
1
identified by 1H NMR. H NMR (CDCl3): d ˆ 0.38 (s, 9H; SiMe3), 1.38 (s,
Compound 5: 31P{1H} NMR (C6D6): d ˆ 42.66 (d, J(P,P) ˆ 40.4 Hz), 52.95
(d, J(P,P) ˆ 40.4 Hz); 1H NMR (C6D6): d ˆ 0.25 (s, 9H; SiMe3), 1.25 (s,
18H; tBu), 2.04 (m, 2H; PCH2CH2P), 2.25 (m, 2H; PCH2CH2P), 3.34 (m,
2H; CH2Pd), 6.93 (d, J(P,H) ˆ 5.2 Hz, 2H; Ar-H), 7.08 ± 7.74 (m, dppe).
18H; tBu), 2.24 (s, 3H; CH3), 7.02 (s, 2H; Ar-H).
Synthesis of 1 from the intermediate phenyl silyl ether: N-Bromosuccin-
imide (400 mg, 2.2 mmol) and azobisisobutyronitrile (10 mg) were added to
a CCl4 solution (30 mL) of the intermediate phenyl silyl ether (500 mg,
1.7 mmol). The mixture was irradiated at close range with a 100 W lamp,
which initiated the reflux. After 20 ± 30 min the mixture was cooled and the
floating imine was filtered off. The solvent was removed on a rotary
evaporator and the residue was extracted with hexane. The dilute hexane
solution (10 mL) was cooled to 48C for 6 h and the byproducts precipitated.
From the concentrated supernatant liquid (ꢀ4 mL) 1 crystallized (310 mg,
49%) and was isolated by decantation. 1H NMR (CDCl3): d ˆ 0.41 (s, 9H;
SiMe3), 1.40 (s, 18H; tBu), 4.49 (s, 2H; CH2Br), 7.26 (s, 2H; Ar-H); 13C {1H}
NMR (CDCl3): d ˆ 3.97 (SiMe3), 31.12 (tBu), 35.13 (tBu), 35.27 (CH2Br),
126.81 (Ar-H), 129.09 (Ar-C), 141.15 (Ar-C), 153.41 (Ar-O), the assignment
was confirmed by a DEPT experiment; C18H31BrOSi: C 58.21, H 8.41;
found C 58.30, H 8.16.
Compound 6: 31P{1H} NMR (C6D6): d ˆ 40.19 (d, J(P,P) ˆ 40.0 Hz), 52.75
(d, J(P,P) ˆ 40.0 Hz); 1H NMR (C6D6): d ˆ 1.18 (s, 18H; tBu), 1.93 (m, 2H;
PCH2CH2P), 2.16 (m, 2H; PCH2CH2P), 3.34 (d, J(P,H) ˆ 9.3 Hz, 2H;
CH2Pd), 6.85 (d, J(P,H) ˆ 4.5 Hz, 2H; Ar-H), 7.04 ± 7.10 (m, 4H; dppe),
7.21 ± 7.30 (m, 12H; dppe), 7.53 (dd, J(P,H) ˆ 12.4 Hz, J(H,H) ˆ 8.1 Hz, 4H;
dppe).
X-ray crystal structure determination of 4: Orange crystals of 4, suitable for
a single crystal X-ray diffraction study, were obtained by slow diffusion
at room temperature of pentane into a diethyl ether/pentane solution
(ꢀ3:1 v/v) of 4.
Crystal data: C41H46OP2Pd, orange prisms, 0.3 Â 0.3 Â 0.3 mm3, monoclinic,
P2(1)/n (No. 14), a ˆ 9.553(2), b ˆ 28.122(6), c ˆ 13.774(3) Š, b ˆ
107.52(3)8, from 25 reflections, Tˆ 110 K, Vˆ 3528.7(13) Š3, Z ˆ 4, Fw ˆ
Synthesis of 2: A cold ( 308C) solution of 1 (57 mg, 0.154 mmol) in
acetone (2 mL) was added to a cold solution of [(tmeda)PdMe2] (36 mg,
0.143 mmol) in acetone (2 mL). After standing for 15 min at 308C, the
solvent was evaporated in vacuo, the residue was washed with pentane
(4 mL), and 2 was extracted with diethyl ether (10 mL). Evaporation of the
solvent gave pure 2 (67 mg, 79%). 1H NMR ([D6]acetone): d ˆ 0.37 (s, 9H;
SiMe3), 1.39 (s, 18H; tBu), 2.43 (s, 6H; NMe2), 2.49 (s, 6H; NMe2), 2.50 (t,
J(H,H) ˆ 5.5 Hz, 2H; NCH2CH2N), 2.79 (t, J(H,H) ˆ 5.5 Hz, 2H;
NCH2CH2N), 2.84 (s, 2H; CH2Pd), 7.55 (s, 2H; Ar-H), the 1H NMR of
this complex in CDCl3 is similar to the spectrum reported for [(tmeda)-
(PhCH2)PdBr];[18] 13C{1H} NMR ([D6]acetone): d ˆ 3.70 (SiMe3), 14.84
(CH2Pd), 32.00 (tBu), 35.49 (tBu), 48.65 (NMe2), 49.66 (NMe2), 57.82
(NCH2CH2N), 63.79 (NCH2CH2N), 127.87 (Ar-H), 140.11 (Ar-C), 140.32
1
723.12, 1calcd ˆ 1.4361 gcm 3, m ˆ 0.648 mm
.
Data collection and treatment: Rigaku AFC5R four-circle diffractometer,
MoKa, graphite monochromator (l ˆ 0.71073 Š), 14156 reflections collect-
ed, 1.45 ꢁ q ꢁ 27.508, 12 ꢁ h ꢁ 7, 0 ꢁ k ꢁ 36, 17 ꢁ l ꢁ 17, w-scan method,
scan width ˆ 1.48, scan speed 128min 1, typical half-height peak width ˆ
0.258, 3 standards were collected 72 times each, with a 4% change in
intensity, 8109 independent reflections (Rint ˆ 0.0560).
Solution and refinement: Structure solved by direct methods (SHELXS-
96). Full-matrix least-squares refinement based on F 2 (SHELXS-93).
Idealized hydrogens were placed and refined in a riding mode, 412 param-
eters with 0 restraints, final R1 ˆ 0.0290 (based on F 2) for data with I > 2sI,
and R1 ˆ 0.0336 for all data based on all 8104 reflections, goodness-of-fit on
F 2 ˆ 1.061, largest electron density ˆ 0.972 eŠ 3. Full details are described
in the Supporting Information provided with ref. [13].
(Ar-C), 149.72 (Ar-O), the assignment was confirmed by
experiment.
a DEPT
Synthesis of 3: A solution of bis(diphenylphosphino)ethane (dppe; 35 mg,
0.088 mmol) in THF (1 mL) was added to a solution of 2 (52 mg,
0.088 mmol) in THF (2 mL) at 308C. After 30 min the solvent was
removed in vacuo and the residue was washed with pentane (4 mL).
Extraction of the residue with diethyl ether (7 mL) and evaporation of the
solvent gave pure 3 (70 mg, 91%). 31P{1H} NMR (C6D6): d ˆ 30.12 (d,
J(P,P) ˆ 40.6 Hz), 52.90 (d, J(P,P) ˆ 40.6 Hz); 1H NMR (C6D6): d ˆ 0.40 (s,
9H; SiMe3), 1.45 (s, 18H; tBu), 1.50 (m overlapping with signal of tBu, 2H;
PCH2CH2P), 1.91 (dtd, J(P,H) ˆ 27.0 Hz, J(H,H) ˆ 8.4 Hz, J(P,H) ˆ 5.2 Hz,
2H; PCH2CH2P), 3.94 (dd, J(P,H) ˆ 10.5 Hz and 5.4 Hz, 2H; CH2Pd),
6.97 ± 7.10 (m, 12H; dppe), 7.31 ± 7.40 (m, 4H; dppe), 7.75 (d, J(P,H) ˆ
1.9 Hz, 2H; Ar-H), 7.91 ± 7.99 (m, 4H; dppe).
Displacement and trapping of BHT-QM: A solution of dibenzylideneacet-
one (DBA) (3 mg, 0.013 mmol) in C6D6 (1 mL) was added to a solution of
4 (7 mg, 0.010 mmol) in C6D6 (1 mL) at room temperature. Based on
31P{1H} NMR integration, ꢀ90% conversion of
4
to the known[16]
[(dppe)Pd(DBA)] was observed within 10 min. Concomitant formation
of free BHT-QM was detected by 1H NMR spectroscopy. Excess MeOH
(ca. 2 mL) was added and the volatiles were removed in vacuo. Extraction
with pentane (4 mL) gave 2,6-di-tert-butyl-4-methoxymethylphenol, which
was identified by means of GC-MS analysis and 1H NMR spectroscopy.
1
Yields are quantitative based on H NMR integration.
BHT-QM: 1H NMR (C6D6): d ˆ 1.37 (s, 18H; tBu), 5.20 (s, 2H; CH2), 6.77
(s, 2H; C-H).[5e]
Synthesis of 4 by the treatment of 3 with TBAF: Tetra-n-butylammonium
fluoride trihydrate (TBAF, 1 equiv, 25 mg, 0.080 mmol) was added to a
solution of 3 (70 mg, 0.080 mmol) in THF (2 mL). The mixture was kept at
room temperature for 15 min, then the solvent was removed in vacuo. The
residue was washed with pentane (4 mL) and 4 was extracted with diethyl
ether (7 mL). Evaporation of the solvent gave pure 4 (53 mg, 91%). 31P{1H}
NMR (C6D6): d ˆ 29.07 (d, J(P,P) ˆ 14.2 Hz), 37.49 (d, J(P,P) ˆ 14.2 Hz);
1H NMR (C6D6): d ˆ 1.61 (s, 18H; tBu), 1.83 (m, 2H; PCH2CH2P), 1.91 (m,
2H; PCH2CH2P), 3.40 (dd, J(P,H) ˆ 7.1 Hz and 3.8 Hz, 2H; exocyclic CH2),
6.93 ± 7.55 (22H; Ar-H and QM-ring C-H); 13C{1H} NMR (C6D6): d ˆ 25.53
(dd, J(P,C) ˆ 26.1, 16.3 Hz; PCH2CH2P), 26.60 (dd, J(P,C) ˆ 26.5, 16.8 Hz;
PCH2CH2P), 51.34 (d, 30.7 Hz, exocyclic CH2), 82.28 (dd, J(P,C) ˆ 12.5,
2,6-Di-tert-butyl-4-methoxymethylphenol: 1H NMR (C6D6): d ˆ 1.37 (s,
18H; tBu), 3.21 (s, 3H; OMe), 4.33 (s, 2H; CH2O), 4.94 (s, 1H; OH), 7.31 (s,
‡
‡
2H; Ar-H); GC-MS (EI): m/z: 250 [M] , 235 [M CH3] , 219 [M
‡
OCH3] .[27, 28]
Reaction of 4 with air: A solution of 4 (ꢀ5 mg) in C6D6 (1 mL) was exposed
to air for 2 h at room temperature. 1H and 31P{1H} NMR revealed the
quantitative formation of bis(diphenylphosphinoxide)ethane (dppe-oxide).
Dppe-oxide: 31P{1H} NMR (C6D6): d ˆ 31.17 (s); 1H NMR (C6D6): d ˆ 2.69
(d, 4H; J(P,H) ˆ 2.5 Hz, PCH2CH2P), 6.97 (m, 12H), 7.71 (m, 8H).[29]
Synthesis of 7: Thexyldimethylchlorosilane (6 mL, 30 mmol) was added to
a solution of p-cresol (2.5 g, 23 mmol) and imidazole (3.9 g, 57 mmol) in
DMF (10 mL). The solution was stirred at room temperature for 15 h.
Hexane (20 mL) was added to create a two-phase system. The hexane
phase was washed with water (10 mL) and dried with Na2SO4. The hexane
was removed on a rotary evaporator at 308C to give the pure phenyl silyl
ether (5.66 g, 98%). 1H NMR (CDCl3): d ˆ 0.21 (s, 6H; SiMe2), 0.95 (s, 6H;
CMe2), 0.95 (d, J(H,H) ˆ 6.8 Hz, 6H; (CH3)2CH), 1.74 (sept, J(H,H) ˆ
6.8 Hz, 1H; (CH3)2CH), 2.28 (s, 3H; Ar-CH3), 6.72 (d, J(H,H) ˆ 8.4 Hz,
2H; Ar-H), 7.02 (d, J(H,H) ˆ 8.4 Hz, 2H; Ar-H); C15H26OSi: C 71.94,
H 10.46; found C 70.80, H 10.25.
ˆ
4.8 Hz, C CH2), 125 ± 142 (Ar and QM ring), 183.96 (brd, J(P,C) ˆ 3.7 Hz;
ˆ
C O); the assignment was confirmed by a DEPT and a C-H correlation
experiment; IR(neat): nÄ ˆ 1598 cm (s); C41H46OP2Pd: C 68.10, H 6.41;
1
found C 68.55, H 7.05.
Synthesis of 4 by the treatment of 3 with AgOTf: Silver triflate (3 mg,
0.012 mmol) was added to a solution of 3 (10 mg, 0.011 mmol) in THF
(2 mL) in the dark. The solid was filtered off and the solvent was removed
in vacuo. The residue was washed with pentane (4 mL) and extracted with
1
benzene (6 mL). 31P and H NMR analysis of the benzene extract showed
460
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0460 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 3

Quinone Methides
454±462
N-Bromosuccinimide (370 mg, 2.0 mmol) and azobisisobutyronitrile
(10 mg) were added to a solution of the phenyl silyl ether (410 mg,
1.6 mmol) in benzene (30 mL). The mixture was irradiated under a nitrogen
atmosphere at close range with a 100 W lamp, which initiated reflux. After
20 min the mixture was cooled, the solvent was evaporated with a rotor
evapotator at 308C until ꢀ7 mL remained, and the floating imine was
filtered off. Based on integration of the 1H NMR signals (with a program to
suppress the benzene signal), the reaction mixture was found to contain
75% benzyl bromide 7 and 25% of the unreacted material. Upon removal
of the solvent and exposure to air, liquid 7 hydrolized rapidly to result in a
purple film. Therefore, the benzene solution of the benzyl bromide 7
(ꢀ0.2m) was degassed and introduced into the glovebox. Aliquots of this
solution were concentrated under vacuum and dissolved in acetone
immediately before use. 1H NMR (CDCl3): d ˆ 0.22 (s, 6H; SiMe2), 0.93
(d, J(H,H) ˆ 6.9 Hz, 6H; (CH3)2CH), 0.94 (s, 6H; CMe2), 1.72 (sept,
J(H,H) ˆ 6.9 Hz, 1H; (CH3)2CH), 4.48 (s, 2H; Ar-CH2-Br), 6.77 (d,
J(H,H) ˆ 8.6 Hz, 2H; Ar-H), 7.25 (d, J(H,H) ˆ 8.6 Hz, 2H; Ar-H).
13C{1H} NMR (CDCl3): d ˆ 2.44 (SiMe2), 18.54 (Me2CH), 20.11 (Me2C),
25.04 (Me2C), 34.02 (CH2Br), 34.11 (Me2CH), 115.65 (Ar-C), 120.36 (Ar-
H), 130.37 (Ar-H), 155.74 (Ar-O); the assignment was confirmed by DEPT
and C-H correlation experiments.
6.9 Hz, 1H; (CH3)2CH), 3.95 (brd, J(P,H) ˆ 8.1 Hz, 2H; CH2Pd), 6.81 (d,
J(H,H) ˆ 8.6 Hz, 2H; Ar-H), 7.81 (brd, J(H,H) ˆ 8.2 Hz, 2H; Ar-H).
Synthesis of 11: A cold ( 308C) solution of nBu4NF ´ 3H2O (7 mg,
0.022 mmol) in THF (1 mL) was added to a cold solution of 10 (17 mg,
0.022 mmol) in THF (1 mL). The mixture was kept for at 308C 15 min.
The solution was evaporate under vacuum and washed with pentane
(4 mL). Extraction of the residue with diethyl ether (7 mL) and removal of
the solvent afforded spectroscopically pure 11 (12 mg, 100%). 31P{1H}
NMR (THF): d ˆ 38.63 (d, J(P,P) ˆ 16.1 Hz), 52.29 (d, J(P,P) ˆ 16.1 Hz);
1H NMR (C6D6): d ˆ 0.85 (d, J(P,H) ˆ 12.7 Hz, 18H; tBu), 0.92 (d, J(P,H) ˆ
12.6 Hz, 18H; tBu), 1.50 (m, coupled to 31P, 4H; dtbpp), 2.76 (dd, J(P,H) ˆ
6.0, 4.0 Hz, 2H; exocyclic CH2), 6.58 (d, J(H,H) ˆ 9.2 Hz, 2H; QM ring),
6.91 (dd, J(P,H) ˆ 2.5 Hz, J(H,H) ˆ 9.2 Hz, 2H; QM ring); 13C{1H} NMR
(C6D6): d ˆ 21.57 (dd, J(P,C) ˆ 9.5, 1.5 Hz; PCH2CH2CH2P), 21.79 (dd,
J(P,C) ˆ 8.6, 3.3 Hz; PCH2CH2CH2P), 23.70 (dd, J(P,C) ˆ 5.2 Hz, 4.4 Hz;
PCH2CH2CH2P), 29.73 (d, J(P,C) ˆ 7.6 Hz, tBu), 29.98 (d, J(P,C) ˆ 6.2 Hz;
tBu), 35.68 (d, J(P,C) ˆ 6.7 Hz, tBu), 36.29 (dd, J(P,C) ˆ 7.0, 1.5 Hz; tBu),
46.89 (d, 33.3 Hz, exocyclic CH2), 125.10 (s, QM ring), 143.50 (s, QM ring),
ˆ
184.36 (s, C O); the assignment was confirmed by a DEPT and a C-H
correlation experiment; IR(film): nÄ ˆ 1587 (s), 1608 cm 1 (m).
Synthesis of 8: A cold solution ( 308C) of 7 (1.5 mL, ꢀ0.3 mmol) in
acetone (1 mL) was added to a cold solution of [(tmeda)PdMe2] (47 mg,
0.186 mmol) in acetone (1 mL). After standing for 15 min at 308C, the
solvent was evaporated in vacuo, the residue was washed with pentane
(4 mL) and extracted with diethyl ether (7 mL). Evaporation of the ether
gave 8 (36 mg, 35%). Extraction of the residue with benzene (4 mL)
afforded 10 mg of the known[26] [(tmeda)Pd(CH3)Br] complex (17%).
1H NMR (C6D6): d ˆ 0.18 (s, 6H; SiMe2), 0.95 (s, 6H; CMe2), 0.97 (d,
J(H,H) ˆ 6.8 Hz, 6H; (CH3)2CH), 1.42 (m, 2H; NCH2CH2N), 1.51 (m, 2H;
NCH2CH2N), 1.73 (sept, J(H,H) ˆ 6.8 Hz, 1H; (CH3)2CH), 1.79 (s, 6H;
NMe2), 2.24 (s, 6H; NMe2), 3.36 (s, 2H; CH2Pd), 6.77 (d, J(H,H) ˆ 8.5 Hz,
2H; Ar-H), 7.83 (d, J(H,H) ˆ 8.5 Hz, 2H; Ar-H); 13C{1H} NMR (C6D6):
d ˆ 2.30 (SiMe2), 14.87 (CH2Pd), 18.81 (Me2CH), 20.42 (Me2C), 25.27
(Me2C), 34.53 (Me2CH), 48.53 (br, NMe2), 49.29 (br, NMe2), 56.94 (br,
NCH2CH2N), 62.70 (br, NCH2CH2N), 120.16 (Ar-H), 130.94 (Ar-H), 141.32
(Ar-C), 152.21 (Ar-O); The assignment was confirmed by a DEPT
experiment.
Acknowledgments
This work was supported by the U.S.±Israel Binational Science Foundation,
Jerusalem (Israel) and by the MINERVA Foundation, Munich (Germany).
We thank Dr L. J. W. Shimon for performing the X-ray structural study.
D.M. is the holder of the Israel Matz professorial chair of organic chemistry.
[1] The Chemistry of the Metal ± Carbon Bond: The Structure, Prepara-
tion, Thermochemistry and Characterization of Organometallic Com-
pounds, Vol. 1 (Eds.: F. R. Hartley, S. Patai) Wiley, NY, 1982.
[2] M. E. Kopach, W. D. Harman, J. Am. Chem. Soc. 1994, 116, 6581.
[3] H. Amouri, Y. Besace, J. Le Bras, J. Vaissermann, J. Am. Chem. Soc.
1998, 120, 6171.
Synthesis of 9: A cold ( 308C) solution of dppe (7 mg, 0.018 mmol) in
THF (1 mL) was added to a cold solution of 8 (10 mg, 0.018 mmol) in THF
(1 mL). After standing at 308C for 15 min, the solvent was removed
under vacuum. The residue was washed with pentane (4 mL), and extracted
into diethyl ether (12 mL) with the assistance of mechanical shaking.
Removal of the solvent in vacuo gave 9 (13 mg, 86%). 31P{1H} NMR
(C6D6): d ˆ 32.21 (d, J(P,P) ˆ 40.2 Hz), 54.63 (d, J(P,P) ˆ 40.2 Hz); 1H NMR
(C6D6): d ˆ 0.13 (s, 6H; SiMe2), 0.95 (s, 6H; CMe2), 0.97 (d, J(H,H) ˆ
6.9 Hz, 6H; (CH3)2CH), 1.52 (m, 2H; PCH2CH2P), 1.70 (sept, J(H,H) ˆ
6.9 Hz, 1H; (CH3)2CH), 1.87 (m, 2H; PCH2CH2P), 3.93 (dd, J(P,H) ˆ 10.6,
5.0 Hz, 2H; CH2Pd), 6.56 (d, J(H,H) ˆ 8.4 Hz, 2H; Ar-H), 6.96 ± 7.03 (m,
12H; dppe), 7.38 (dd, J(H,H) ˆ 8.4 Hz, weakly coupled to 31P, 2H; Ar-H),
7.27 ± 7.36 (m, 4H; dppe), 7.86 ± 7.94 (m, 4H; dppe).
[4] a) A. Vigalok, D. Milstein, J. Am. Chem. Soc. 1997, 119, 7873; b) A.
Vigalok, L. J. W. Shimon, D. Milstein, J. Am. Chem. Soc. 1998, 120,
477.
[5] For reviews, see: a) A. A. Volodꢁkin, V. V. Ershov, Russ. Chem. Rev.
1988, 57, 336; b) H.-U. Wagner, R. Gompper, in The Chemistry of
Quinonoid Compounds, Vol. 1 (Ed.: S. Patai), Wiley, NY, 1974,
p. 1145; c) A. B. Turner, Quart. Rev. 1964, 18, 347. See also: d) L. J.
Filar, S. Winstein, Tetrahedron Lett. 1960, 25, 9; e) L. K. Dyall, S.
Winstein, J. Am. Chem. Soc. 1972, 94, 2196.
[6] For a review, see: D. C. Thompson, J. A. Thompson, M. Sugumaran, P.
Moldeus, Chem.-Biol. Interact. 1993, 86, 129.
[7] S. M. Shevchenko, A. G. Apushkinskii, Russ. Chem. Rev. 1992, 61, 105.
[8] H. W. Moore, R. Czerniak, Med. Res. Rev. 1981, 1, 249.
[9] a) M. G. Peter, Angew. Chem. 1989, 101, 572; Angew. Chem. Int. Ed.
Engl. 1989, 28, 555; b) T. Li, Q. Zeng, S. E. Rokita, Bioconjugate
Chem. 1994, 5, 497.
[10] a) A. A. L. Gunatilaka, in Progress in the Chemistry of Organic
Natural Products, Vol. 67, Springer, Wien, NY, 1996, p. 1; b) G. Prota,
in Progress in the Chemistry of Organic Natural Products, Vol. 64,
Springer, Wien, NY, 1995, p. 93; c) D. A. Bolon, J. Org. Chem. 1970,
35, 3666; d) S. R. Angle, M. S. Louie, Tetrahedron Lett. 1993, 34, 4751;
e) J. P. Marino, S. L. Dax, J. Org. Chem. 1984, 49, 3671.
[11] a) T. E. Gram, Pharmacol. Rev. 1988, 49, 297; b) D. C. Thompson, Y.-
N. Cha, M. A. Trush, J. Biol. Chem. 1989, 264, 3957; c) K. Z. Guyton, P.
Bhan, P. Kuppusamy, J. L. Zweier, M. A. Trush, T. W. Kensler, Proc.
Natl. Acad. Sci. USA 1991, 88, 946; d) J. L. Bolton, S. B. Turnipseed,
J. A. Thompson, Chem.-Biol. Interact. 1997, 107, 185.
[12] a) J. L. Bolton, J. C. Y. Le Blanc, K. W. M. Siu, Biol. Mass Spec. 1993,
22, 666; b) K. Z. Guyton, P. M. Dolan, T. W. Kensler, Carcinogenesis
1994, 15, 817; c) K. Yamamoto, S. Kato, K. Tajima, T. Mizutani, Biol.
Pharm. Bull. 1997, 20, 571; d) P. G. McCracken, J. L. Bolton, G. R. J.
Thatcher, J. Org. Chem. 1997, 62, 1820; e) M. Reed, D. C. Thompson,
Chem. Res. Toxicol. 1997, 10, 1109; f) M. A. Lewis, D. G. Yoerg, J. L.
Reaction of 9 with nBu4NF: A cold ( 308C) solution of 9 (13 mg,
0.016 mmol) in THF (2 mL) and
a cold solution of TBAF (5 mg,
0.016 mmol) in THF (1 mL) were mixed and kept at 308C for 25 min.
The solvent was evaporated under vacuum and the residue was extracted
with diethyl ether (8 mL). The solution was analyzed by means of 31P{1H},
1H, and 1H{31P} NMR spectroscopy, which revealed that it contained 43%
[(dppe)2Pd] and 57% of a new complex, the characteristic NMR signals of
which are: 31P{1H} NMR (C6D6): d ˆ 38.62 (d, J(P,P) ˆ 24.6 Hz), 56.42 (d,
1
J(P,P) ˆ 24.6 Hz); H NMR (C6D6): d ˆ 1.21 (dd, J(P,H) ˆ 8.0, 4.0 Hz, 2H;
benzylic position), 1.69 (m, 2H; PCH2CH2P), 1.87 (m, 2H; PCH2CH2P).
Synthesis of 10: A cold ( 308C) solution of 8 (36 mg, 0.065 mmol) in THF
(1 mL) was mixed with a cold solution of bis(di-tert-butylphosphino)pro-
pane (dtbpp) (24 mg, 0.066 mmol) in THF (1 mL), and was kept for 20 min
at 308C. The solution was dried under vacuum, washed with diethyl ether
(4 mL), and extracted with benzene (6 mL). Removal of the solvent gave 10
(26 mg, 52%). 31P{1H} NMR (C6D6): d ˆ 30.30 (d, J(P,P) ˆ 49.4 Hz), 49.55
(d, J(P,P) ˆ 49.4 Hz); 1H NMR (C6D6): d ˆ 0.21 (s, 6H; SiMe2), 0.95 (s, 6H;
CMe2), 0.96 (d, J(H,H) ˆ 6.9 Hz, 6H; (CH3)2CH), 1.18 (d, J(P,H) ˆ 12.6 Hz,
18H; tBu), 1.20 (d, J(P,H) ˆ 13.0 Hz, 18H; tBu), 1.70 (sept, J(H,H)ˆ
Chem. Eur. J. 2000, 6, No. 3
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0461 $ 17.50+.50/0
461

D. Milstein et al.
FULL PAPER
Bolton, J. A. Thompson, Chem. Res. Toxicol. 1996, 9, 1368; g) G.
Hocman, Int. J. Biochem. 1988, 20, 639.
[13] O. Rabin, A. Vigalok, D. Milstein, J. Am. Chem. Soc. 1998, 120, 7119.
[14] a) S. D. Ittel, J. A. Ibers, Adv. Organomet. Chem. 1976, 14, 33; b) J. H.
Nelson, H. B. Jonassen, Coord. Chem. Rev. 1971, 6, 27; c) J. Silvestre,
T. A. Albright, J. Am. Chem. Soc. 1985, 107, 6829.
[15] S. Friedman, M. L. Kaufman, I. Wender, J. Org. Chem. 1962, 27, 664.
[16] W. A. Herrmann, W. R. Thiel, C. Broûmer, K. Öfele, T. Priermeier, J.
Organomet. Chem. 1993, 461, 51.
Y. T. Struchkov, Organometallics 1993, 12, 991; c) R. Benn, P. Betz, R.
Goddard, P. W. Jolly, N. Kokel, C. Kruger, I. Topalovic, Z. Natur-
forsch. Teil B 1991, 46, 1395, and ref. [16].
[23] G. Wilke, H. Schott, P. Heimbach, Angew. Chem. 1967, 79, 62; Angew.
Chem. Int. Ed. Engl. 1967, 6, 92.
[24] For transient generation of simple QMs under laser flash photolysis
conditions see: a) L. Diao, C. Yang, P. Wan, J. Am. Chem. Soc. 1995,
117, 5369; b) P. Wan, B. Barker, L. Diao, M. Fischer, Y. Shi, C. Yang,
Can. J. Chem. 1996, 74, 465.
[17] M. Portnoy, F. Frolow, D. Milstein, Organometallics 1991, 10, 3960.
[18] W. de Graaf, J. Boersma, G. van Koten, Organometallics 1990, 9, 1479.
[19] C. J. Pouchert, J. Behnke, The Aldrich Library of 13C and 1H FT NMR
Spectra Vol. 1, 1st ed., Aldrich, 1993.
[20] D. Rabinovich, M. J. Schmidt, J. Chem. Soc. B 1967, 127.
[21] H. Werner, G. T. Crisp, P. W. Jolly, H.-J. Kraus, C. Kruger, Organo-
metallics 1983, 2, 1369.
[25] a) H. Wetter, K. Oertle, Tetrahedron Lett. 1985, 26, 5515; b) T. W.;
Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis, 2nd
ed., Wiley, NY, 1991.
[26] W. de Graaf, J. Boersma, W. J. J. Smeets, A. L. Spek, G. van Koten,
Organometallics 1989, 8, 2907.
[27] S. Satish, P. A. Ganeshpure, Proc. Indian Acad. Sci. Chem. Sci. 1986,
96, 59.
[22] For other examples of structurally characterized [P2Pd(olefin)]
complexes, see: a) M. Kranenburg, J. G. P. Delis, P. C. J. Kamer,
P. W. N. M. van Leeuwen, K. Vrieze, N. Veldman, A. L. Spek, K.
Goubitz, J. Fraanje, J. Chem. Soc. Dalton Trans. 1997, 1839; b) V. V.
Bashilov, P. V. Petrovskii, V. I. Sokolov, S. V. Lindeman, I. A. Guzey,
[28] R. S. Macomber, J. Org. Chem. 1982, 47, 2481.
[29] A. M. Aguilar, J. Beisler, J. Org. Chem. 1964, 29, 1660.
Received: July 22, 1999 [F1931]
462
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0603-0462 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 3