Terminal Heck vinylations of chelating vinyl ethers

Article abstract of DOI:10.1002/adsc.200404184

Terminal chelation-controlled Heck vinylations of electron-rich amino-functionalized vinyl ethers were performed with high regioselectivity furnishing moderate to good isolated yields of the corresponding 1-alkoxy-1,3-butadienes. DFT calculations support

Full text of DOI:10.1002/adsc.200404184

FULL PAPERS
Terminal Heck Vinylations of Chelating Vinyl Ethers
Alexander Stadler, Henrik von Schenck, Karl S. A. Vallin, Mats Larhed,*
Anders Hallberg
Organic Pharmaceutical Chemistry, Department of Medicinal Chemistry, Uppsala Biomedical Centre, Uppsala University,
Box 574, SE-751 23 Uppsala, Sweden
Fax: (þ46)-18-471-4474, e-mail: mats@orgfarm.uu.se
Received: June 16, 2004; Accepted: September 27, 2004
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Terminal chelation-controlled Heck vinyla- Subsequent Diels–Alder reactions with dimethyl ace-
tions of electron-rich amino-functionalized vinyl tylenedicarboxylate under microwave irradiation re-
ethers were performed with high regioselectivity fur- sulted exclusively in partly aromatized bi- and tricy-
nishing moderate to good isolated yields of the corre- clic compounds by elimination of the aminoalkoxy
sponding 1-alkoxy-1,3-butadienes. DFT calculations group. Thus, the selected dimethylamino auxiliary
support an amine-palladium(II) coordination strength both controlled the regiochemistry in the palladium-
reactivity/selectivity rationale, where the dimethyl- catalyzed vinylation and was easily displaced in the
amino group was the preferred metal presenting func- aromatization process.
tionality. Controlled microwave heating effectively
accelerated these palladium-catalyzed reactions and Keywords: cycloaddition; DFT; Heck reaction; micro-
full conversion could be achieved within 30 minutes. waves; regioselectivity; vinyl ether
Introduction
this class of dienes appear rather toilsome,[14,15] more ef-
ficient procedures should be valuable.
A vast number of palladium-catalyzed Heck coupling
We herein report that the reaction of vinyl triflates
reactions with vinyl halides or pseudohalides have with selected amino-functionalized alkyl vinyl ethers
been described during the last three decades.[1–4] The vi- in the presence of a palladium catalyst gives linear 1-alk-
nylation generally occurs in the terminal b-position of oxy-1,3-butadienes of predominantly trans-configura-
electron-deficient alkenes, as in the case of the closely tion in high selectivity (Scheme 1). Complementary
related Heck arylation reactions. In contrast, the vinyla- DFT calculations provide a rationale for the experimen-
tion of electron-rich olefins can be performed under cat- tal results with regard to the reactivity and selectivity
ionic conditions to afford the electronically favored in- difference in the Heck coupling employing the investi-
ternal a-product, a branched 1,3-butadiene in high se- gated chelating olefins. Finally, microwave (MW) medi-
lectivity.[5–7] No direct Heck method has so far been re- ated Diels–Alder reactions were performed to verify the
ported to furnish linear 1,3-butadienes from electron- usefulness of the nitrogen-containing dienes as synthetic
rich olefins, e.g., enol ethers or enamides.
building blocks (Scheme 1).
We have previously developed a set of different auxil-
iary-controlled Heck arylations.[8,9] In these reactions
the specific introduction of an appropriately designed Results
palladium(II)-coordinating tertiary amino group into
the olefin allowed an efficient substrate presentation
Chelation-Controlled Heck Vinylation Reactions
and a high terminal regiocontrol.[10] Employing a cyclic
vinyl ether with a chiral N-methylpyrrolidine group,
highly diastereoselective b-arylations were realized.[11] We decided to initiate the investigation with three avail-
The presented knowledge suggested to us that chela- able two-carbon tethered b-amino substituted vinyl
tion-controlled b-vinylation of vinyl ethers should pro- ethers (1a–c, Scheme 2). The olefins were selected to
vide a new gateway into valuable 1-alkoxy-1,3-buta- permit evaluation of the capacity of the different met-
dienes. Since trans-alkoxydienes should be useful in al-coordinating amino groups in controlling the vinyl-
the synthesis of cyclohexene derivatives by Diels–Alder ation process. The choice of the proline derivative 1c
reactions[12,13] and since classical preparative routes to was triggered by the potential to study the influence
Adv. Synth. Catal. 2004, 346, 1773–1781
DOI: 10.1002/adsc.200404184
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1773

Alexander Stadler et al.
FULL PAPERS
Scheme 1. Regioselective terminal vinylations and subsequent microwave-mediated Diels–Alder/amino alcohol elimination
reactions.
Scheme 2. Building blocks for terminal Heck vinylations.
of a chiral auxiliary in subsequent Diels–Alder reac- at the b-carbon of the vinyl ethers resulting in clear ter-
tions.
minal coupling products. Only the use of diethylamino
The reaction of vinyl triflates 2a–f with the chelating vinyl ether 1b produced small amounts of the internal
enol ethers 1a–c were performed on a 0.50-mmol scale coupling product with triflates 2c and 2d (b/a ꢀ 96:4).
under standard Heck reaction conditions (24 h oil- Notably, the corresponding methyl ketone, that should
bath heating at 608C), using either a catalytic amount have been formed after hydrolysis of the a-product,
of the tris(dibenzylideneacetone)dipalladium(0)/tri-t- was never detected. Furthermore, the terminal vinyla-
butylphosphonium tetrafluoroborate[16] system (Meth- tion turned out to be stereoselective with high E/Z ratios
od A, see Table 1) or palladium acetate and triphenyl- (Table 1). Importantly, and in contrast to the recent find-
phosphine as precatalytic combination (Method B, see ing in Heck arylation chemistry by Chandrasekhar,[18] vi-
Table 1). Since olefins 1a and 1c were not commercially nylation of non-chelating butyl vinyl ether using tri-
available (or very expensive), the excess of 1a–c was ethylamine as base was very sluggish and did not give
limited to 1.5 equivs. compared to the vinyl triflates. terminal selectivity employing PEG as solvent. Similar
The addition of an external base (triethylamine) showed to our previous investigation of palladium-catalyzed
no positive effect on the reaction outcome but instead couplings of alkyl vinyl ethers with vinyl triflates,[5]
produced more by-products. Thus, since the olefins high internal selectivity was achieved with butyl vinyl
and the products (3a–o) are tertiary amines, the experi- ether by using Method B after addition of 2 equivs. tri-
ments were conducted with no external base. All but one ethylamine. Butyl vinyl ether did not give any vinylation
vinylation of 1a and 1b furnished useful isolated yields product in the absence of an external base.
with Method A (54–74% of 3, except for product 3k), al-
In order to accelerate the reactions, the experiments
though reactions employing the less reactive olefin 1c have also been carried out in sealed vessels under con-
did not proceed at all under these conditions (Table 1). trolled microwave irradiation[19–21] using the identical
Switching to the well-documented Pd(OAc)2/PPh3 com- catalytic system as in Method B, but on a 0.20-mmol
bination[17] (Method B) generally afforded a slightly im- scale (Method C, see Table 2). Full conversion of 2 and
proved process with both 1a and 1b as coupling partners similar yields as with classic heating could be obtained
(56–92%). With Method B and an increased reaction with all three olefins in only 30 minutes reaction time
temperature (808C), even sluggish 1c produced dienes at 1208C. However, with this method the stereoselectiv-
3 m–o in modest yields (22–36%). As evident from Ta- ity was reduced to an average of E/Zꢀ70:30, whereas
ble 1, the reaction underwent strict vinylic substitution the regioselectivity remained unaffected, still resulting
1774
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1773–1781

Terminal Heck Vinylations of Chelating Vinyl Ethers
FULL PAPERS
Table 1. Chelation-controlled terminal vinylations of vinyl ethers with classic heating
Vinyl Ether
Triflate
Product
Method A[a] Isolated Yield
Method B[b] Isolated Yield
1
a
2a
3a
3b
54% E/Z¼90: 10
65% E/Z¼86: 14
2b
2c
56% E/Z¼94: 6
62% E/Z¼81: 19
74% E/Z¼89: 11
64% E/Z¼81: 19
3c
2d
3d
64% E/Z¼86: 14
68% E/Z¼87: 13
2e
2f
3e
3f
61% E/Z¼89: 11
73% E/Z¼40: 60
74% E/Z¼83: 17
56% E/Z¼90: 10
92% E/Z¼52: 48
63% E/Z¼92: 8
1
b
2a
3 g
2b
2c
3 h
3i
73% E/Z¼86: 14
70% E/Z¼89: 11
57% E/Z¼77: 23, b/a¼96: 4
50% E/Z¼72: 28, b/a¼96: 4
2d
3j
68% E/Z¼93: 7, b/a¼97: 3
58% E/Z¼76: 24, b/a¼94: 6
2e
2f
3k
3 l
<5%
56% E/Z¼57: 43
66% E/Z¼50: 50
58% E/Z¼40: 60
1c[c]
2b
2c
2d
3 m
3n
<5%
<5%
<5%
36% E/Z¼88: 12
22% E/Z¼ >99: 1
33% E/Z¼ >99: 1
3o
Reactions were conducted in sealed tubes with 0.50 mmol 2, 0.75 mmol 1 and 1.5 mL DMSO at 608Cfor 24 h under N 2. More
than>95% conversion of 2 and more than>95% purity of isolated products 3 (E/Z-mixture) by GC/MS and NMR.
[a]
3 mol % Pd2(dba)3 ·CHCl, 10 mol % t-Bu3PHBF4.
1 mol % Pd(OAc)2, 3 mol % PPh3.
The reactions with 1c were performed at 808Cfor 48 h.
3
[b]
[c]
Adv. Synth. Catal. 2004, 346, 1773–1781
asc.wiley-vch.de
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1775

Alexander Stadler et al.
FULL PAPERS
Table 2. Results of microwave-accelerated Heck vinylations.
Diels–Alder Reactions
Entry
Product
Isolated Yield
E/Z ratio
After purification, the E/Z-mixtures of 3a–d were treat-
ed with the electron-poor dienophile dimethyl acetyl-
enedicarboxylate (DMAD) in order to obtain the corre-
sponding Diels–Alder adducts (Scheme 3). The thermal
reactions (8 h at room temperature or 4 h at 608C, re-
spectively) turned out to be rather sluggish, furnishing
significant amounts of side products. However, the mi-
crowave protocol (60 min at 1208C, Scheme 3) proceed-
ed reasonably well with complete conversions of the
E/Z-diene mixtures, but yielding the unexpected tetra-
hydronaphthalene and dihydrophenanthrene deriva-
tives 5a–d. The aromatic compounds 5a–d were ob-
tained in moderate to good yields after simple purifica-
tion of the crude mixture by flash chromatography.
The formation of the resulting products 5 could be ex-
plained by immediate elimination of 2-dimethylamino-
ethanol from the primarily formed, intermediate
Diels–Alder adducts 4 (Scheme 3).[12] In fact, perform-
ing the microwave-mediated Diels–Alder reaction
with 3b under reductive conditions (by addition of
3 equivs. hydroquinone) did not lead to the aromatic
compound 5b but rather to intermediate 4b. In this reac-
tion, non-conjugated 4b was observed by GC/MS in an
82:18 ratio to the competing Diels–Alder adduct of al-
koxybutadiene 3b and hydroquinone.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
59%
60%
51%
47%
44%
46%[a]
65%
64%
51%
45%
50%[a]
55%[a]
30%
28%
30%
80: 20
81: 19
60: 40
77: 23
68: 32
44: 56
74: 26
75: 25
57: 43
75: 25
93: 7
36: 64
75: 25
99: 1
77: 23
Reactions were conducted in sealed tubes with 0.20 mmol 2,
0.30 mmol 1 and 0.6 mL DMSO at 1208Cfor 30 min. More
than>95% conversion of 2 and more than>95% purity of
products 3 (E/Z-mixture) by GC/MS and NMR.
[a] Sluggish reaction with non-complete conversion of 2, yield
determined from GC/MS using 2,3-dimethylnaphthalene
as internal standard.
in exclusively terminal vinylations to form the desired 1-
Full aromatization of the compounds 5c and 5d was re-
alkoxy-1,3-butadienes 3 (Table 2). Contrary to the ex- alized under microwave conditions by treatment with
periments with classic heating, compounds 3i and 3j 7 equivs. of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
showed no accompanying internal coupling product un- (DDQ) for 60 minutes at 1208C, furnishing the phenan-
der microwave heating.
threne derivatives 6 (Scheme 4).[24]
Photochemistry
Discussion
In the presented Heck syntheses of 1-alkoxy-1,3-dienes
chelation-control was successfully exploited to control
the regioselectivity while the geometrical selectivity re-
Mechanism
mained partly insufficient. Photoisomerization[22,23] of In a chelation-controlled terminal Heck vinylation proc-
dihydronaphthalene derivative 3c was therefore investi- ess the reaction route exemplified in Scheme 5 appears
gated as a model reaction with the aim to improve the reasonable.[8,25,26] Starting from vinyl triflates, cationic
E/Z ratio. The photoisomerization of the 1-alkoxy-1,3- intermediates should prevail from the oxidative addi-
diene 3c was carried out in dry DMSO using a very dilute tion complex A and onwards.[6,27–29] The chelated ring
sample (1 mM) and focused 280 nm irradiation (lmax for size controls the direction of insertion into p-complex
cis-3c¼281.5 nm). After 2 h, the selected regioisomeric B and provides the s-alkylpalladium complex C with
mixture was converted from E/Z¼60:40 to E/Z¼ high b-selectivity. This intermediate palladacycle collap-
90:10. Longerirradiation times didnot improve the geo- ses via syn-b-elimination to give the terminal product 3.
metrical ratio, as a side reaction occurred, forming a
In this scenario, the palladium-directing amino group
non-identified by-product in amounts increasing with must, a) function as a good ligand for the vinylpalladi-
time. Changing the solvent did not improve the method, um(II) complex to enable an adequate presentation of
as both MeCN and MeOH promoted the unexpected the double bond and to control the regiochemistry in
side reaction directly from the beginning of the UV irra- the insertion, and b) provide reversible coordination
diation.
of the different palladium(II) species to allow catalytic
turnover (Scheme 5). Thus, the nitrogen-palladium
bond strength will be a compromise between these two
contrasting demands. The results from the vinylation ex-
1776
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1773–1781

Terminal Heck Vinylations of Chelating Vinyl Ethers
FULL PAPERS
Scheme 3. Microwave-mediated synthesis of tetrahydronaphthalene and dihydrophenanthrene derivatives via Diels–Alder in-
termediates. The reactions were conducted in sealed vials with 0.20 mmol 3, 0.40 mmol DMAD and 0.5 mL MeCN at 1208C
for 60 min. More than>95% conversion of 3 (E/Z-mixture) and more than>95% purity of isolated products 5 by GC/MS.
Scheme 4. Microwave-mediated aromatization to afford phenanthrene derivatives 6c and 6d. The reactions were conducted in
sealed vials with 0.10 mmol 5, 0.70 mmol DDQ and 1.5 mL MeCN at 1208Cfor 60 min. More than >95% conversion of 5 and
more than>95% purity of products 6 by GC/MS.
periments demonstrate that the dimethylamino- and the 1.5 equivs. each of two different vinyl ethers 1a–c using
N-methylated pyrrolidine frameworks in 1a and 1c were Methods B and C.
comparable regarding regiocontrol, while the diethyl-
(i) Dimethylaminoethyl vinyl ether (1a) vs. diethylami-
amino auxiliary did not prove sufficiently effective noethyl vinyl ether (1b): GC/MS analysis of the product
(see Table 1, triflates 2c and 2d). When comparing the mixture showed clearly that the formation of dimethyl
reactivity of olefins 1a–c, the vinylation of the chiral derivative 3b was preferred to 3 h in this experiment.
1c was found to be very sluggish under all investigated On conducting the reaction with classic heating (608C,
reaction conditions. The reduced terminal selectivity 24 h) 3b was found to be predominant compared with
with 1b indicates a relatively weak metal-nitrogen bind- 3 h (3b/3h¼88:12). As 3b was isolated in comparable
ing and the low yields with pyrrolidine derivative 1c sug- yield to the single experiment in Table 1 (50%), no sig-
gest an enhanced stability of intermediates A, B or C nificant inhibition of the reaction could be assessed. Un-
(Scheme 5).
der microwave conditions (1208C, 30 min) the reaction
also showed high conversion, although 3b was preferred
only in a 75:25 ratio.
Competitive Experiments
(ii) Dimethylaminoethyl vinyl ether (1a) vs. (S)-N-
methyl-2-vinyloxymethyl pyrrolidine (1c): In the experi-
Further evidence for the role of the N-methylpyrroli- ment at 808Caccording to Method B the pyrrolidine de-
dine functionality as the most strongly palladium(II) rivative 1c inhibited the Heck process. Thus, only traces
binding auxiliary was provided in a series of competitive of 3m could be detected and the yield of 3b also drasti-
experiments with the model cyclohexenyl triflate 2b. cally decreased (10%). Under high-density microwave
The reactions were performed with 1 equiv. 2b and irradiation, 30 min at 1608Cwas needed to observe
Adv. Synth. Catal. 2004, 346, 1773–1781
asc.wiley-vch.de
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1777

Alexander Stadler et al.
FULL PAPERS
Scheme 5. Reaction route for chelation-controlled Heck vinylations.
any dimethyl product (3b, <10%). Still, only partial con-
version was achieved as significant amounts of the start-
ing triflate 2b could be recovered. Increasing the tem-
perature further led to considerable decomposition of
both 1a and 1c.
(iii) Diethylaminoethyl vinyl ether (1b) vs. (S)-N-
methyl-2-vinyloxymethylpyrrolidine (1c): These final
experiments furnished similar results as described under
(ii). The pyrrolidine ether 1c inhibited the reaction and
only traces of the desired coupling products were detect-
ed, regardless of the condition used.
Scheme 6. General model system for investigating the N-
Pd(II) coordination strength in vinylpalladium(II) complexes.
These experimental results indicate a rate-limiting ef-
fect of the pyrrolidine ether 1c in the vinylation process.
If no other vinyl ether is present in the catalytic cycle, 1c
acts as the olefinic reagent, sluggishly affording the cor-
responding diene in moderate yields. In competition
with the other species, 1a and 1b, the pyrrolidine ether
is responsible for the slow reaction and decreasing yields
of the desired dimethyl and diethylamino products due
to strong coordination/chelation of active palladium,
Scheme 7. Chelating and dissociated p-complexes of used ni-
trogen auxiliaries.
rendering the regenerationof the active catalyst rate-de- lized tertiary amine groups (Nx) were removed from the
termining. In the case where 1a and 1b are in competi- p-complex B0 (Scheme 6). The N-Pd(II) coordination
tion, results suggest a relatively strong N-Pd(II) coordi- strength, DE, was calculated as the energy difference be-
nation for 1a, however, without adverse effects on the tween the dissociation products (B0’ and Nx) and start-
overall reactivity.
ing B0. Results are reported in Table 3, Entries 1–3.
The calculated order of increasing DE is in agreement
with experimental results, where the coordination
strength has been evaluated for a series of amines coor-
dinating to the cationic complex Pd(dmpe)Me, where
Computational Investigations
To provide further insight into the experimental differ- dmpe¼1,2-bis(dimethylphosphino)ethane.[30]
ence of the studied enol ethers, we decided to investigate
Increasingly detailed models more closely represent-
the palladium(II)-nitrogen coordination strength in the ed the p-complexes of the real experimental systems.
relevant chelated species using computational tools. In a second set of calculations, the coordination strength
Thus, the coordination strengths of a number of nitrogen of the actual nitrogen auxiliary was determined in the
containing auxiliaries were calculated using density chelated p-complex B (Scheme 7). While the amino-
functional theory at the B3LYP level. Computational substituted vinyl ethers were fully represented, the
details are provided below.
phosphine ligand was first kept minimal with respect
A set of square planar cationic vinylpalladium(II) to size, as PH3, (Table 3, Entries 4–6). Steric effects
complexes of increasing size were used to model the ex- due to the phosphine ligand were later included by re-
perimental systems. Initially, alkylated mimics of the uti- placing PH3 with PPh3 (Entries 7–9).
1778
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1773–1781

Terminal Heck Vinylations of Chelating Vinyl Ethers
FULL PAPERS
Table 3. Calculated N-Pd(II) coordination strengths for uti-
lized N-auxiliaries.
gy, influencing the respective ligands ability to function
in the catalytic process.
In agreement with the competitive experiments, the
pyrrolidine ether 1c shows the highest ligand bond
strength towards Pd(II), therefore withdrawing the cat-
alyst from the reaction cycle and leading to significantly
lowered yields. Comparing the dimethyl and the diethyl
vinyl ethers, the dimethyl derivative 1a has obviously the
optimum ligand bond strength for proceeding the cata-
lyst cycle, as observed in the competitive experiments.
Energetics favor the coordination of 1a to Pd, leading
to a displacement of diethyl olefin 1b from the metal
and consequently to higher yields of the vinylated di-
methyl product. Switching to the question of prepara-
tive usefulness in the investigated direct synthesis of 1-
alkoxy-1,3-butadienes, 1a is also the most attractive
choice. Compared with 1b both reactivity and regiose-
lectivity are superior (Table 1). Due to the fairly weak
N-Pd(II) coordination of 1b (DE¼4.1), it is quite rea-
sonable that a fraction of the olefins could undergo reac-
tion without the directing influence of the amine. The
electronically favored a-product would then be formed,
as was the case for 1b reacting with triflates 2c and 2d
(Table 1).
Entry Nx
1
PR3 DE [kcal/mol] dPd [Š]
ꢂ
N
PH3 28.6
2.29
2
3
PH3 25.0
PH3 23.2
PH3 22.7
2.32
2.32
2.29
4
1c
5
6
7
1a
1b
1c
PH3 18.9
PH3 17.1
PPh3 15.1
2.32
2.33
2.31
8
9
1a
1b
PPh3 10.7
PPh3 4.1
2.35
2.38
Conclusion
The successively more detailed models allowed for de-
convolution of components affecting the N-Pd(II) coor-
dination strength. In the first ligand set (Entries 1–3),
DE was unaffected by palladacycle strain and by the sig-
nificant steric effects introduced by PPh3. A certain ster-
ic component of the total coordination energy is still ex-
pected, even for these reduced systems. An estimate of
the steric component of DE was calculated by compar-
ing the energy of free Nx in the geometry optimized
for B0, with the energy of relaxed Nx. The energy loss
due to geometrical rearrangement upon coordination
We have been able to perform highly selective terminal
vinylations of vinyl ethers by utilizing palladium(II)-co-
ordinating amino substituents. Furthermore, the scope
of the method was extended by acceleration of the reac-
tions with controlled high-density microwave irradia-
tion. Theoretical DFT calculations assisted in explaining
the experimental outcome for the different palladi-
um(II)-coordinating nitrogen auxiliaries. Finally, the
synthesized dimethylaminoethoxy-1,3-butadienes were
investigated as substrates in microwave-mediated
Diels–Alder reactions.
’
ofNx to PdB0 was calculated to1.6 kcal/mol for N-meth-
ylpyrrolidine, 1.9 kcal/mol for NMe3 and 3.3 kcal/mol
for NEt3. These results imply that the proline moiety of-
fered the most favorable electronic interactions with
palladium while NEt3 was the sterically most demand-
ing, once again confirmed by experiments.[30] Introduc-
ing the effect of palladacycle ring strain in the second li-
gand set (Entries 4–6) revealed a systematic weakening
Experimental Section
General
Microwave heating was conducted in a single mode cavity
of the N-Pd coordination strength. Comparing Entries 1 (SmithSynthesizer), operating at 2.45 GHz, with a built-in
magnetic stirrer, on-line IR and pressure sensors. Flash column
chromatography (FCC) was generally carried out using com-
mercially available silica gel 60, employing ethyl acetate/hex-
ane 1:9 (þ1% triethylamine) as eluent, except where other-
wise indicated.
with 4, 2 with 5 and 3 with 6, gives differences in DE of
5.9, 6.1 and 6.1 kcal/mol, respectively. In the third ligand
set (Entries 7–9), replacement of PH3 with PPh3 weak-
ened the nitrogen coordination even further with the
relatively bulky diethyl vinyl ether 1b being most affect-
ed. Comparing Entries 4 with 7, 5 with 8 and 6 with 9,
gives differences in DE of 7.6, 8.2 and 13.0 kcal/mol, re-
spectively. This progressive influence due to phosphine
size, in combination with the effects described above,
1H NMR spectra were recorded in CDCl3 at 270 and
400 MHz, respectively. Mass spectra were recorded on a GC/
MS equipped with a CP-SIL 8 CB (30 mꢁ0.25 mm) capillary
column, utilizing electron impact at an ionizing energy of
70 eV. Additionally LC/MS electrospray analyses were per-
leads to N-Pd(II) coordinations well separated in ener- formed using a Chromolith Performance RP-18e column
Adv. Synth. Catal. 2004, 346, 1773–1781
asc.wiley-vch.de
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1779

Alexander Stadler et al.
FULL PAPERS
(4.6ꢁ100 mm) and an MeCN/water gradient. Elemental anal-
yses were performed byAnalytische Laboratorien, Prof. Dr. H.
Malissa and G. Reuter GmbH. The instability of the isolated
compounds in some cases prevented accurate elemental anal-
ysis. The high resolution MS analyses were performed by Einar
Nilsson, University of Lund.
Photochemical reactions were carried out on DMSO solu-
tions under N2 gas flow using an Oriel 1000 W Xe ARClight
source and a 280 nm Oriel UV filter. The emitted light density
was determined using a UVenhanced Si photodiode (5.8 mm2)
attached to a current meter.
and dissolved in dry DMSO (1.5 mL). Pd2(dba)3 ·CHCl
3
(3 mol %) and (t-Bu3PH)BF4 (Fu salt, 10 mol %) were added,
the vessel was flushed with nitrogen, sealed and inserted into
a heating block. The mixture was kept at 608Cfor 24 h under
slight stirring. Thereafter the reaction was quenched with 1
M NaOH (5 mL) and extracted with diethyl ether (2ꢁ
10 mL). The combined organic layers were washed with brine
(2ꢁ5 mL) and dried with MgSO4 overnight. After filtration,
evaporation of the solvent afforded the desired terminal cou-
pling products as slightly yellow to orange oily liquids. Purifica-
tion of the crude products was performed by FCC using ethyl
acetate/hexane (1:9þ1% triethylamine) as eluent.
Method B: The reactions were carried out as above, using
Pd(OAc)2 (1 mol %) and PPh3 (3 mol % ) as precatalytic sys-
tem.
Materials
(t-Bu3PH)BF4 was purchased from Strem Chemicals, DMSO
(ꢃ99.5%, over molecular sieves) was obtained from Fluka,
all other chemicals, including 1b, were purchased from Merck
and used without further purification. Dimethylaminoethyl vi-
nyl ether 1a was prepared according the literature procedure.[8]
Method C: 0.20 mmol vinyl triflate 2 and 1.5 equivs.
(0.30 mmol) of the corresponding vinyl ether 1 were placed
in a 2-mL microwave reaction vessel and dissolved in 0.6 mL
of dry DMSO. Pd(OAc)2 (1 mol %) and PPh3 (3 mol % )
were added, the vessel was flushed with nitrogen, sealed and in-
serted into the microwave cavity and irradiated at 1208Cfor
30 min, subsequently followed by air-jet cooling. Work-up
was carried out as in Method A, yielding the products in com-
parable yields to the thermal runs. The b-vinylation products
[as (E)/(Z) mixtures, unless separated] exhibited spectral and
analytical properties as summarized in the Supplementary In-
formation.
(S)-N-Methyl-2-vinyloxymethylpyrrolidine (1c)
A 50-mL thick-walled glass tube was equipped with a stirring
bar and charged with S-(ꢂ)-N-methyl-2-hydroxymethylpyrro-
lidine (1.73 g, 15 mmol) dissolved in ethyl vinyl ether (28 mL).
Pd(OAc)2 (128 mg, 3.5 mol %) and 2,2’-bipyridyl (81 mg,
3.5 mol %) were added, the vessel was flushed with nitrogen,
sealed and placed in an appropriate heating block. The mixture
was refluxed (T¼508C) for at least 72 h, then diluted with pen-
tane (40 mL) and washed with NaOH (0.1 M) and water (2ꢁ
20 mL each). The organic layer was dried with K2CO3 and fi-
nally evaporated to obtain 1c as an orange liquid; yield:
850 mg (6 mmol, 40%).
Photoisomerization
The dihydronaphthalene derivative 3c (0.5 mg, 2.0 mmol, E/
Z¼60:40) was dissolved in dry DMSO (2 mL) and placed in
a conventional UV vessel (d¼1 cm). The vial was flushed
with nitrogen and sealed with a septum. During irradiation
N2 was bubbled through the solution. Measuring a full UV
spectrum of the sample, the absorption maxima were deter-
mined to 304.5 nm for the trans-compound and 281.5 nm for
the cis-isomer. The sample was placed in the focus of a
280 nm optical lens and irradiated for several hours to trans-
form the mixture into the trans-isomer (E) of 3c (after 2 h E/
Z¼90:10). Aliquot samples have been taken every hour and
analyzed by GC/MS.
The corresponding triflates have been prepared according
to literature protocols.[31,32]
N-Benzyl-1,2,3,6-tetrahydropyridine 4-triflate (2e);
Representative Procedure
N-Benzylpiperidin-4-one (378 mg, 2.0 mmol) was dissolved in
dry THF (10 mL) and cooled to ꢂ788C. A 1 M solution of
LHMDS in THF (2.4 mL, 1.2 equivs.) was added dropwise un-
der N2 flush. After addition, the mixture was stirred at ꢂ788C
for approx. 20 min, thereafter 1.2 equivs. of PhN(Tf)2 (857 mg,
2.4 mmol) were added at once. The cooling bath was removed,
allowing the mixture to warm up to room temperature under
stirring for 1 h, then it was refluxed for additional 2 hours. Fi-
nally the reaction was quenched with water (10 mL), the mix-
ture was extracted with EtOAc (3ꢁ10 mL), washed with
NH4Cl and brine (10 mL each) and dried with MgSO4. Evapo-
ration of the solvent afforded 827 mg dark orange highly vis-
cous liquid. The crude product was purified by FCC, using hex-
ane/EtOAc (9:1) as eluent, obtaining the desired triflate 2e as
colorless viscous liquid; yield: 460 mg (72%).
Diels–Alder Reactions, General Procedure
0.20 mmol of diene 3 (E/Z mixtures, see Table 1) was placed in
a 2-mL microwave reaction vial and dissolved in MeCN
(0.5 mL).
2 equivs. of dimethyl acetylenedicarboxylate
(DMAD, 0.40 mmol) were added, the vial was flushed with ni-
trogen, sealed and inserted into the microwave cavity. The re-
action mixture was irradiated at 1208Cfor 60 min, subsequent-
ly followed by air-jet cooling. The solvent was removed under
vacuum and the reminder purified by FCC, using ethyl acetate/
hexane (9:1)þ1% triethylamine as eluent.
Aromatization Procedure of the
Dihydrophenanthrene Derivatives
Heck Couplings, General Procedures
Method A: Vinyl triflate 2 (0.50 mmol) and 1.5 equivs. of vinyl
ether 1 (0.75 mmol) were placed in a thick-walled glass vessel
0.10 mmol dihydrophenanthrene 5 wasplacedin a2-mL micro-
wave reaction vial and dissolved in MeCN (1 mL). 7 equivs. of
1780
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1773–1781

Terminal Heck Vinylations of Chelating Vinyl Ethers
FULL PAPERS
2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ, 0.7 mmol)
were added, the vial was flushed with nitrogen, sealed and in-
serted into the microwave cavity and irradiated at 1208Cfor
60 min, subsequently followed by air-jet cooling. The solvent
was removed under vacuum and the crude 6 purified by FCC,
using ethylacetate/hexane (9:1)þ1% triethylamine as eluent.
[8] C. M. Andersson, J. Larsson, A. Hallberg, J. Org. Chem.
1990, 55, 5757.
[9] M. Larhed, C. M. Andersson, A. Hallberg, Acta Chem.
Scand. 1993, 47, 212.
[10] P. Nilsson, M. Larhed, A. Hallberg, J. Am. Chem. Soc.
2001, 123, 8217.
[11] P. Nilsson, M. Larhed, A. Hallberg, J. Am. Chem. Soc.
2003, 125, 3430.
[12] T. N. Maksimova, V. B. Mochalin, B. V. Unkovskii,
Khim. Geterotsikl. Soedin. 1980, 273.
[13] M. Hayashi, K. Tsukada, H. Kawabata, C. Lamberth,
Tetrahedron 1999, 55, 12287.
[14] H. A. M. Jacobs, M. H. Berg, L. Brandsma, J. F. Arens,
Recl. Trav. Chim. Pays-Bas 1965, 84, 1113.
[15] J. Zhang, Y. Zhang, Y. Zhang, J. W. Herndon, Tetrahe-
dron 2003, 59, 5609.
[16] M. R. Netherton, G. C. Fu, Org. Lett. 2001, 3, 4295.
[17] R. F. Heck, Acc. Chem. Res. 1979, 12, 146.
[18] S. Chandrasekhar, C. Narsihmulu, S. S. Sultana, N. R.
Reddy, Org. Lett. 2002, 4, 4399.
Computational Details
Geometries and energies were fully optimized using the gradi-
ent-corrected hybrid density functional method B3LYP.[33] We
used a basis set of double-z valence quality labeled lacvp in the
Jaguar 4.0 program (Jaguar 4.0, Schrodinger, Inc., Portland, Or-
egon, 2000). For Pd, the core electrons were replaced by a rel-
ativistic electron core potential (ECP) developed by Hay and
Wadt.[34] After geometry optimization, the energies were recal-
culated at the B3LYP level using a valence triple-z quality basis
set labeled lacv3p** in the Jaguar 4.0 program.
[19] C. R. Strauss, R. W. Trainor, Aust. J. Chem. 1995, 48,
1665.
Acknowledgements
[20] P. Lidstrçm, J. Tierney, B. Wathey, J. Westman, Tetrahe-
dron 2001, 57, 9225.
[21] M. Larhed, C. Moberg, A. Hallberg, Acc. Chem. Res.
2002, 35, 717.
[22] D. H. Waldeck, Chem. Rev. 1991, 91, 415.
[23] T. Arai, K. Tokumaru, Chem. Rev. 1993, 93, 23.
[24] M. Adeva, H. Sahagun, E. Caballero, R. P. L. de Clairac,
M. Medarde, F. Tome, J. Org. Chem. 2000, 65, 3387.
[25] N. D. Duezo, J. C. de la Rosa, J. Priego, I. Alonso, J. C.
Carretero, Chem. Eur. J. 2001, 7, 3890.
[26] M. Larhed, C. M. Andersson, A. Hallberg, Tetrahedron
1994, 50, 285.
[27] W. Cabri, I. Candiani, Acc. Chem. Res. 1995, 28, 2.
[28] A. Jutand, A. Mosleh, Organometallics 1995, 14, 1810.
[29] J. M. Brown, K. K. Hii, Angew. Chem., Int. Ed. Engl.
1996, 35, 657.
[30] A. L. Seligson, W. C. Trogler, J. Am. Chem. Soc. 1991,
113, 2520.
[31] A. Arcadi, A. Burini, S. Cacchi, M. Delmastro, F. Ma-
rinelli, B. R. Pietroni, J. Org. Chem. 1992, 57, 976.
[32] K. Pal, Synthesis 1995, 1485.
We acknowledge the financial support from the Swedish Re-
search Council and from Knut and Alice Wallenbergꢀs Founda-
tion. We also thank Biotage AB for providing us with the Smith-
Synthesizer. Paralleldatorcentrum (PDC) at the Royal Institute
of Technology is acknowledged for providing computer facili-
´ ´
´
ties. Finally, we would like to thank Dr. Mate Erdelyi for reward-
ing advices in the photochemical reactions and Dr. Kristofer
Olofsson for intellectual contributions to this project.
References
[1] R. F. Heck, Org. React. 1982, 27, 345.
[2] A. De Meijere, F. E. Meyer, Angew. Chem., Int. Ed.
Engl. 1994, 33, 2379.
[3] I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100,
3009.
[4] M. Larhed, A. Hallberg, in: Handbook of Organopalladi-
um Chemistry for Organic Synthesis, Vol. 1, (Ed.: E.-i
Negishi), Wiley & Sons Inc., New York, 2002, pp. 1133.
[5] C. M. Andersson, A. Hallberg, J. Org. Chem. 1989, 54,
1502.
[33] P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J.
Frisch, J. Phys. Chem. 1994, 98, 11623.
[34] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[6] K. S. A. Vallin, M. Larhed, K. Johansson, A. Hallberg, J.
Org. Chem. 2000, 65, 4537.
[7] K. S. A. Vallin, Q. S. Zhang, M. Larhed, D. P. Curran, A.
Hallberg, J. Org. Chem. 2003, 68, 6639.
Adv. Synth. Catal. 2004, 346, 1773–1781
asc.wiley-vch.de
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1781