The eight-membered metallacycles arising from the insertion of 1 equiv of alkyne into the Pd-C bond of ortho-metalated homoveratrylamine and phentermine can further react with alkenes to give two different types of mononuclear complexes depending on the nature of the olefin. When terminal alkenes (styrene and ethyl acrylate) are used, a mixture of the anti/syn η3-allyl Pd(II) complexes are isolated, which evolve slowly to the syn isomers by heating the mixtures appropriately. These η3-allyl Pd(II) complexes do not react with CO or weak bases, but when they are treated with a strong base, such as KOtBu, they afford Pd(0) and the functionalized starting phenethylamines containing a 1,3-butadienyl substituent in an ortho position. When 2-norbornene was used instead of terminal alkenes, the strained olefin inserts into the alkenyl Pd(II) complex to afford a 10-membered norbornyl palladium(II) complex, in which the new C,N-chelate ligand is coordinated to the metal through an additional double bond, occupying three coordination positions. The reactivity of these norbornyl complexes depends on the substituents on the inserted alkenyl fragment, and thus they can further react with (1) KOtBu, to give Pd(0) and a tetrahydroisoquinoline nucleus containing a tricyclo[3.2.1]octyl ring, or (2) CO and TlOTf, to afford Pd(0) and amino acid derivatives or the corresponding lactones arising from an intramolecular Michael addition of the CO2H group to the α,β-unsaturated ester moiety. Crystal structures of every type of compound have been determined by X-ray diffraction studies.
The eight-membered metallacycles arising from the insertion of 1 equiv of alkyne into the Pd-C bond of ortho-metalated homoveratrylamine and phentermine can further react with alkenes to give two different types of mononuclear complexes depending on the nature of the olefin. When terminal alkenes (styrene and ethyl acrylate) are used, a mixture of the anti/syn η3-allyl Pd(II) complexes are isolated, which evolve slowly to the syn isomers by heating the mixtures appropriately. These η3-allyl Pd(II) complexes do not react with CO or weak bases, but when they are treated with a strong base, such as KOtBu, they afford Pd(0) and the functionalized starting phenethylamines containing a 1,3-butadienyl substituent in an ortho position. When 2-norbornene was used instead of terminal alkenes, the strained olefin inserts into the alkenyl Pd(II) complex to afford a 10-membered norbornyl palladium(II) complex, in which the new C,N-chelate ligand is coordinated to the metal through an additional double bond, occupying three coordination positions. The reactivity of these norbornyl complexes depends on the substituents on the inserted alkenyl fragment, and thus they can further react with (1) KOtBu, to give Pd(0) and a tetrahydroisoquinoline nucleus containing a tricyclo[3.2.1]octyl ring, or (2) CO and TlOTf, to afford Pd(0) and amino acid derivatives or the corresponding lactones arising from an intramolecular Michael addition of the CO2H group to the α,β-unsaturated ester moiety. Crystal structures of every type of compound have been determined by X-ray diffraction studies.
Palladium is one of
the most versatile transition metals in organic
synthesis, given its well-known ability to catalyze the formation
of C–C and C–heteroatom bonds.[1,2] Nevertheless,
the extraordinarily high variability of applications of palladium
relies just on a few elementary steps inherent to its reactivity,
and among them, the migratory insertion reaction of an unsaturated
ligand into a Pd–C bond stands out.[3−5] Thus, a myriad
of methods to functionalize alkenes,[3,6] alkynes,[7] or allenes[8] through
Pd catalysis have been reported so far.Multicomponent reactions,
where several reagents assemble sequentially
giving rise to complex structures, are valuable protocols, since they
represent an effective modular approach to green organic synthesis.[9] Nevertheless, transition-metal-catalyzed processes
involving the use of several unsaturated coupling partners in the
reaction mixture are challenging fields due to the difficulty in controlling
the order of reactivity: that is, the control of regioselectivity
of the migratory insertion sequence for the different unsaturated
reagents.[10−12] Some successful examples of these types of reactions
are the copolymerization of alkenes and CO (Scheme a)[13] and the copolymerization
of olefins with and without polar groups,[14] among others.[15−17] Moreover, impressive progress has been made on intramolecular
cascade reactions where different unsaturated moieties are involved
(Scheme b).[5,18] In the latter case, however, the selectivity is usually determined
by the structure of the substrate itself.
Scheme 1
Examples of Reactions
Involving Sequential Migratory Insertion into
Pd–C Bonds[13,18c]
While catalytic conditions make difficult the aforementioned control
of the selectivity, stoichiometric reactions can offer the advantage
of stepwise insertion paths for the synthesis of valuable organic
products.[19−23]Our research group has previously reported the functionalization
of primary phenethylamines through the isolation of ortho-palladated intermediates.[24−26] These types of arylalkylamines
constitute the core of many relevant biomolecules, such as neurotransmitters
(i.e., dopamine), amino acids (i.e., phenylalanine), or marketed psychoactive
drugs. The catalytic functionalization of these scaffolds has proven
challenging, due to the strong coordination ability of the primary
amine group to Pd(II).[27] Hence, the study
of the stoichiometric functionalization of primary phenethylamines
can provide alternative routes to modify these interesting molecules.
For instance, the sequential insertion of alkynes and CO or of alkynes
and isocyanides into ortho-palladated derivatives
allowed us to synthesize medium-sized rings such as eight-membered
benzazocinones (Scheme ).[20,21]
Scheme 2
Synthesis of Benzazocinones through the
Sequential Insertion of Alkynes
and CO into the Pd–C Bond of Six-Membered Palladacycles
We have previously reported the synthesis and
isolation of stable
eight-membered C,N-palladacycles
arising from the insertion of one molecule of alkyne into the Pd–C
bond of ortho-palladated phentermine and homoveratrylamine
(Chart ).[20] We present here the results of a study on the
insertion of alkenes and alkenes/CO into these eight-membered alkenyl
palladacycles. We have studied the new organometallic species generated
upon each insertion, as well as the final functionalized organic products,
formed through depalladation of the final organopalladium compounds.
The overall processes rendering the final organopalladium complexes
involve the sequential insertion of (a) alkynes and alkenes or (b)
alkynes, alkenes, and CO into the Pd–C bond of the ortho-metalated derivatives from the primary phenethylamines.
Chart 1
Previously Reported Eight-Membered Alkenyl Palladacycles Used as
Starting Materials
Results and Discussion
Insertion
of Styrene or Ethyl Acrylate into the Pd–C
Bond
Synthesis and Structure of η3-Allyl Palladium(II)
Complexes
The reaction of the previously described alkenyl
palladacycle a or b, arising from the monoinsertion
of diphenylacetylene into the ortho-palladated homoveratrylamine
and phentermine derivative A or B, with
2 equiv of CH2=CHR′ (molar ratio Pd/alkene
= 1/1; R = Ph (1), CO2Et (2))
in CH2Cl2 at room temperature (16 h) gave a
mixture of two complexes, which were identified as the anti and syn isomers of the η3-allyl
Pd(II) complex 1 or 2 (Scheme ).
Scheme 3
Reactions of Eight-Membered
Alkenyl Palladacycles with Terminal Alkenes
The ratios correspond to the
isolated solids and were estimated by the integrals of the CH allylic
signals of both isomers in the 1H NMR spectra of the mixtures.
Reactions of Eight-Membered
Alkenyl Palladacycles with Terminal Alkenes
The ratios correspond to the
isolated solids and were estimated by the integrals of the CH allylic
signals of both isomers in the 1H NMR spectra of the mixtures.After workup, a mixture of η3-allyl Pd(II) complexes anti-/syn-1a (2.5/1) or anti-/syn-2a (5/1) was isolated.
Attempts to separate both isomers by fractional crystallization were
unsuccessful. However, single crystals of anti-1a or anti-2a, suitable for
X-ray diffraction studies, were obtained by slow diffusion of n-pentane into a solution of the mixture of anti-/syn-1a or anti-/syn-2a in CH2Cl2. The
ratio between isomers, for both 1a and 2a, practically did not change after 48 h in CDCl3 solution
at room temperature (by 1H NMR). However, the conversion
to the syn isomers was complete after heating CDCl3 solutions of the mixtures at 60 °C for 10 days (syn-1a) or 48 h (syn-2a; by 1H NMR; Figure and Table ).
Figure 1
1H NMR spectra (4.6–5.0 ppm) of complex 2a (CDCl3): (a) anti-enriched
mixture of complex 2a at room temperature (300 MHz) and
(b) after heating at 60 °C for 8 h (400 MHz), (c) 24 h (300 MHz)
and (d) 48 h (300 MHz).
Table 1
Isomerization
Ratios for anti-/syn-1 and anti-/syn-2
anti/syn ratio
temperature
time (h)
1a
1b
2a
2b
RT
0
2.5/1
2.5/1
5/1
1.25/1
12
2.5/1
4.2/1
16
1.25/1
24
2.5/1
4.2/1
48
2.5/1
2/1
3.3/1
1.25/1
60 °C
0
2.5/1
2.5/1
5/1
1.25/1
8
1.7/1
1/1
12
1/5
24
1.1/1
1/10
48
1/1.25
1/3.3
0/1
1/20
120
1/10
240
0/1
1H NMR spectra (4.6–5.0 ppm) of complex 2a (CDCl3): (a) anti-enriched
mixture of complex 2a at room temperature (300 MHz) and
(b) after heating at 60 °C for 8 h (400 MHz), (c) 24 h (300 MHz)
and (d) 48 h (300 MHz).Enriched mixtures of both anti and syn isomers of 1b and 2b could be obtained
due to the higher solubility of one of the two isomers in Et2O. When a solution of the mixture anti-/syn-1b or anti-/syn-2b (1b, ratio 1.25/1, 0.161 mmol; 2b, ratio 1.25/1, 0.260 mmol) in CH2Cl2 (1b, 3 mL; 2b, 2 mL) was treated with
Et2O (15 mL), a mixture enriched in the anti (1b) or syn (2b) isomer
precipitated (ratio anti/syn: 1b, 3/1; 2b: 1/3). When the mother liquors were
concentrated (1b: 3 mL; 2b: 5 mL) and treated
with n-pentane (1b, 20 mL) or cooled
to 0 °C in an ice bath (2b), a mixture enriched
in the syn (1b) or anti (2b) isomer precipitated (ratio anti/syn: 1b, 1/3; 2b: 3/1).
Almost spectroscopically pure samples of anti- and syn-1b and anti- and syn-2b could be obtained by growing single
crystals from the enriched mixtures.Since anti-1 and anti-2 complexes
practically did not isomerize to the syn-1 and syn-2 derivatives at room temperature
(Table ), both isomers
should be formed during the
reaction. We proposed that the formation of anti-/syn-1 or anti-/syn-2 could be envisioned by the mechanism
depicted in Scheme . Insertion of one molecule of the alkene into the Pd–C bond
of the starting palladacycle (a or b) would
afford the 10-membered alkyl palladacycle I. For this
intermediate, we propose that the R′ group is at the carbon
atom bonded to Pd(II), because (1) this is the regiochemistry that
explains the formation of the allyl moiety coordinated to Pd(II) in
complexes 1a,b and 2a,b and (2) this is the most frequent regioisomer found in the
insertion of electron-poor alkenes into the Pd–C bonds of neutral
complexes.[22,28] Intermediate I could
undergo β-hydride elimination to give the cisoid-(diene)PdH species II. Syn addition
of Pd–H to the alkene moiety with regioselectivity in contrast
with its previous elimination results in the formation of the η3-allyl complex anti-1 or anti-2 (Scheme ), which seems to be formed preferentially as the kinetic
product. On the other hand, the transoid-(diene)PdH
intermediate III could be generated from II upon decoordination and rotation of the tertiary olefin moiety,
prior to the Pd–H addition step, hence giving rise to the formation
of the syn isomer. The thermal isomerization of the
complexes anti-1 and anti-2 to the syn isomers at 60 °C
(Table ) proceeded
by the usual η3–η1–η3 rearrangement.
Scheme 4
Formation of anti and syn Isomers
of η3-Allyl Pd(II) Complexes
Complexes 1 and 2 were fully characterized
by elemental analyses and NMR and IR spectroscopic techniques. The 1H NMR spectra of all the allyl complexes showed the diastereotopic
nature of the hydrogen atoms of the NH2 and CH2 groups, as well as both of Me groups of the CMe2 moiety
for 1b and 2b. The most characteristic signal
was that corresponding to the methine group of the inserted fragment,
which appeared as a doublet of doublets in the range 4.85–5.16
ppm for the anti isomers and 4.75–4.84 ppm
for the syn isomers. The two coupling constants 3JHH (anti, 11.2
≤ 3JHH ≤ 12.3
Hz, average value 11.7 Hz; syn, 9.6 ≤ 3JHH ≤ 11.0 Hz, average
value 10.3 Hz; anti, 4.5 ≤ 3JHH ≤ 5.2 Hz, average value 4.7 Hz; syn, 3.7 ≤ 3JHH ≤ 4.2 Hz, average value 4.0 Hz) were slightly smaller for
the syn isomers than for the anti isomers.The crystal structures of the η3-allyl complexes anti-1a, anti-1b, syn-1b, anti-2a·CH2Cl2, anti-2b·CHCl3, and syn-2b were determined by X-ray diffraction (see the Supporting Information). The X-ray thermal ellipsoid
plots of anti-1a and syn-2b are depicted in Figures and 3, respectively.
In both cases, the palladium atoms adopted a square-planar geometry
(mean deviation from the plane Y–Pd(1)–X(1)–N(1)
0.0194 Å, where Y = centroid from C(7), C(8), and C(9) atoms).
The halogeno ligand and the amino group of the metalated fragment
were coordinated in cis positions. The other two
coordination sites were occupied by the allyl moiety, which was η3-bonded via C(7), C(8), and C(9), with C(7)–C(8)–C(9)
angles of 119.4 and 122.5° for anti-1a and syn-2b, respectively.
Figure 2
X-ray thermal
ellipsoid plot (50% probability) of anti-1a along with the labeling scheme. The hydrogen atoms
bonded to carbon have been omitted for clarity. Selected bond lengths
(Å) and angles (deg): Pd(1)–Br(1) = 2.4959(3), Pd(1)–N(1)
= 2.1124(15), Pd(1)–C(7) = 2.1295(15), Pd(1)–C(8) =
2.1047(15), Pd(1)–C(9) = 2.1181(16), C(7)–C(8) = 1.458(2),
C(8)–C(9) = 1.427(2); Br(1)–Pd(1)–N(1) = 92.82(4),
C(7)–C(8)–C(9) = 119.41(14), C(8)–C(9)–C(14)
= 131.14(14).
Figure 3
X-ray thermal ellipsoid plot (50% probability)
of syn-2b along with the labeling scheme.
The hydrogen atoms
bonded to carbon have been omitted for clarity. Selected bond lengths
(Å) and angles (deg): Pd(1)–Cl(1) = 2.3864(5), Pd(1)–N(1)
= 2.1257(18), Pd(1)–C(7) = 2.1198(19), Pd(1)–C(8) =
2.1400(19), Pd(1)–C(9) = 2.1167(19), C(7)–C(8) = 1.441(3),
C(8)–C(9) = 1.421(3); Cl(1)–Pd(1)–N(1) = 89.81(5),
C(7)–C(8)–C(9) = 117.38(18), C(8)–C(9)–C(14)
= 122.49(18).
X-ray thermal
ellipsoid plot (50% probability) of anti-1a along with the labeling scheme. The hydrogen atoms
bonded to carbon have been omitted for clarity. Selected bond lengths
(Å) and angles (deg): Pd(1)–Br(1) = 2.4959(3), Pd(1)–N(1)
= 2.1124(15), Pd(1)–C(7) = 2.1295(15), Pd(1)–C(8) =
2.1047(15), Pd(1)–C(9) = 2.1181(16), C(7)–C(8) = 1.458(2),
C(8)–C(9) = 1.427(2); Br(1)–Pd(1)–N(1) = 92.82(4),
C(7)–C(8)–C(9) = 119.41(14), C(8)–C(9)–C(14)
= 131.14(14).X-ray thermal ellipsoid plot (50% probability)
of syn-2b along with the labeling scheme.
The hydrogen atoms
bonded to carbon have been omitted for clarity. Selected bond lengths
(Å) and angles (deg): Pd(1)–Cl(1) = 2.3864(5), Pd(1)–N(1)
= 2.1257(18), Pd(1)–C(7) = 2.1198(19), Pd(1)–C(8) =
2.1400(19), Pd(1)–C(9) = 2.1167(19), C(7)–C(8) = 1.441(3),
C(8)–C(9) = 1.421(3); Cl(1)–Pd(1)–N(1) = 89.81(5),
C(7)–C(8)–C(9) = 117.38(18), C(8)–C(9)–C(14)
= 122.49(18).Some catalytic transformations
involving the sequential migratory
insertion of alkynes and alkenes into a Pd–C[11,12,17,29] or a Pd–H[15] bond have been reported in the literature. In
these processes, substituted 1,3-butadienes are formed. The mechanism
proposed for those catalytic transformations does not consider the
formation of η3-allyl intermediates. In our case,
however, the presence of a coordinating group, such as NH2, may favor the isomerization of the 1,3-butadiene to form a stabilized
allyl Pd(II) complex. As far as we are aware, this is the first stoichiometric
study where the sequential migratory insertion of an alkyne and an
alkene has been studied.
Reactivity of the η3-Allyl
Complexes
The nucleophilic attack to η3-allyl
Pd(II) intermediates
to give functionalized alkenes is a well-known strategy in organic
synthesis.[30] However, most of the times
the key η3-allyl Pd(II) species are generated through
the oxidative addition of allyl halides, pseudohalides or acetates
to Pd(0). We wondered if the η3-allyl complexes 1 and 2 could undergo the attack of a suitable
nucleophile to render an organic derivative. Hence, we performed a
range of reactions with several nucleophiles, such as KCN, [Tl(acac)], p-toluidine, and the phosphorus ylide Ph3PCHCOMe.
There was no reaction at room temperature with any of these nucleophiles.
When the reaction mixture was heated in refluxing toluene, complicated
mixtures were produced. The reaction of complex 1b and
dimethyl malonate in the presence of Cs2CO3 (CHCl3, room temperature) afforded an anti/syn mixture of isomers of a new complex containing the η3-allyl moiety and a coordinated malonate, whose structures
were not studied further. When a solution of this complex in CHCl3 was heated at 60 °C, again a complicated mixture was
obtained. In order to facilitate the nucleophilic attack, we performed
the reaction of the η3-allyl complex 1b with p-toluidine in the presence of TlOTf, which
would generate a cationic complex by replacing the Cl– ligand by OTf– in the coordination sphere of Pd(II).
In this case, the stable cationic η3-allyl complex 3b was isolated, which contained a coordinated p-toluidine ligand (Scheme ). This complex was stable in refluxing toluene and did not
decompose to the expected organic product.
Scheme 5
Reactions of η3-Allyl Complex 1b with p-Toluidine
The 1H NMR of complex 3b at room temperature
showed some broad signals and seemed to correspond to a mixture of
two isomers with a fluxional behavior (Figure ). Perhaps the most significant resonances
are those attributed to the Me groups: three sharp singlets at 1.45
(3 H), 1.42 (3 H), and 1.26 ppm (2.2 H) and two broad singlets at
2.28 (5.4 H) and 0.94 ppm (2.2 H). When the spectrum was measured
at −40 °C, the broad singlet below 1 ppm become sharper,
and the signal at 2.28 ppm split into two new singlets with the relative
intensities 3:2.2. According to these data, we assumed that both isomers, anti-/syn-3b, are formed in
a 1.33/1 ratio, one of which presented a fluxional behavior in solution.
The less stable isomer should be anti-3b due to steric hindrance of the toluidine ligand and the substituent
on the terminal allylic carbon, and for it, it was reasonable to suppose
that the neutral ligand could be coming in and out of the metal coordination
sphere.
Figure 4
1H NMR spectra (0.6–3.0 ppm) of complex 3b (400 MHz) in CDCl3 at 25 °C (bottom) and
−40 °C (top). The asterisk indicates the signal corresponding
to H2O. The blue and red circles correspond to the syn and anti isomers, respectively.
1H NMR spectra (0.6–3.0 ppm) of complex 3b (400 MHz) in CDCl3 at 25 °C (bottom) and
−40 °C (top). The asterisk indicates the signal corresponding
to H2O. The blue and red circles correspond to the syn and anti isomers, respectively.The crystal structure of the η3-allyl complex syn-3b·CH2Cl2 was
determined by X-ray diffraction (Figure ), showing that the phenyl and benzyl substituents
at the meso and terminal carbon atoms of the allylic unit occupied
mutually syn positions and indicating that this was,
in fact, the most stable isomer. For this complex there were two independent
molecules in the asymmetric unit (A and A′). For molecule A,
the palladium atom exhibited a square-planar geometry (mean deviation
from the plane: X–Pd(1)–N(1)–N(2) 0.0067 Å,
where X = centroid from C(7), C(8), and C(9) atoms). Both amino groups
were coordinated in cis positions. The other two
coordination sites were occupied by the allyl moiety, which was η3-bonded via C(7), C(8), and C(9), with a C(7)–C(8)–C(9)
angle of 116.9(2)°. The allyl plane, defined by atoms C(7), C(8),
and C(9), formed a dihedral angle of 118.1° with the Pd(II) coordination
plane. The aromatic rings formed angles of 71.3° (metalated ring),
65.5° (phenyl ring at C7), 71.9° (phenyl ring at C8), 73.3°
(phenyl ring at C14), and 92.4° (toluidine ring) with respect
to the Pd coordination plane, to avoid steric hindrance. Hydrogen
bond interactions were observed between the cationic palladium moiety
and the OTf– anion (see the Supporting Information).
Figure 5
Thermal ellipsoid plot (50% probability)
of the cation of one (A)
of the two independent molecules of the complex syn-3b·CH2Cl2 showing the labeling
scheme. The solvent molecule and the hydrogen atoms bonded to carbon
have been omitted for clarity. Selected bond lengths (Å) and
angles (deg) are given for both independent molecules (A and A′).
For A: Pd(1)–N(1) = 2.131(2), Pd(1)–N(2) = 2.161(2),
Pd(1)–C(7) = 2.100(2), Pd(1)–C(8) = 2.138(2), Pd(1)–C(9)
= 2.165(2), C(7)–C(8) = 1.454(3), C(8)–C(9) = 1.408(3);
N(1)–Pd(1)–N(2) = 90.26(9), C(7)–C(8)–C(9)
= 116.9(2), C(8)–C(9)–C(14) = 124.9(2). For A′:
Pd(1′)–N(1′) = 2.130(2), Pd(1′)–N(2′)
= 2.163(2), Pd(1′)–C(7′) = 2.117(2), Pd(1′)–C(8′)
= 2.134(2), Pd(1′)–C(9′) = 2.165(2), C(7′)–C(8′)
= 1.452(3), C(8′)–C(9′) = 1.413(3); N(1′)–Pd(1′)–N(2′)
= 88.94(8), C(7′)–C(8′)–C(9′) =
117.5(2), C(8)–C(9)–C(14) = 124.0(2).
Thermal ellipsoid plot (50% probability)
of the cation of one (A)
of the two independent molecules of the complex syn-3b·CH2Cl2 showing the labeling
scheme. The solvent molecule and the hydrogen atoms bonded to carbon
have been omitted for clarity. Selected bond lengths (Å) and
angles (deg) are given for both independent molecules (A and A′).
For A: Pd(1)–N(1) = 2.131(2), Pd(1)–N(2) = 2.161(2),
Pd(1)–C(7) = 2.100(2), Pd(1)–C(8) = 2.138(2), Pd(1)–C(9)
= 2.165(2), C(7)–C(8) = 1.454(3), C(8)–C(9) = 1.408(3);
N(1)–Pd(1)–N(2) = 90.26(9), C(7)–C(8)–C(9)
= 116.9(2), C(8)–C(9)–C(14) = 124.9(2). For A′:
Pd(1′)–N(1′) = 2.130(2), Pd(1′)–N(2′)
= 2.163(2), Pd(1′)–C(7′) = 2.117(2), Pd(1′)–C(8′)
= 2.134(2), Pd(1′)–C(9′) = 2.165(2), C(7′)–C(8′)
= 1.452(3), C(8′)–C(9′) = 1.413(3); N(1′)–Pd(1′)–N(2′)
= 88.94(8), C(7′)–C(8′)–C(9′) =
117.5(2), C(8)–C(9)–C(14) = 124.0(2).Additionally, we studied the behavior of the anti-/syn-1a,b complexes toward
a strong base. The reaction of these complexes with KOtBu in toluene under a N2 atmosphere afforded Pd(0) and
the functionalized phenethylamines containing a 1,3-butadienyl substituent
in an ortho position (4a,b; Scheme ). We did
not observe intramolecular cyclization arising from the possible nucleophilic
attack of the NH2 group to the η3-allyl
moiety. The compound 4a was a colorless solid and was
easily isolated, but 4b was obtained as a liquid. In
order to get a more easily isolable derivative, triflic acid was added
to a solution of compound 4b in diethyl ether (Scheme ). The corresponding
ammonium triflate 5b precipitated in the reaction mixture
and was obtained as a white powder in moderate yield (50%).
Scheme 6
Synthesis
of Functionalized Phenethylamine Derivatives Containing
1,3-Butadienyl Substituents
Miura et al.[12] described the intermolecular
three-component coupling of aryl iodides, diarylacetylenes, and alkenes
in the presence of palladium acetylacetonate and silver acetate as
catalysts, to give the corresponding 1:1:1 and 1:2:1 coupling products
(1,3-butadiene and 1,3,5-hexatriene derivatives, respectively). The
mechanism proposed for these catalytic transformations followed the
sequential steps analogous to those of the synthesis of compounds 4 reported here, (1) formation of an aryl Pd(II) complex (either
by oxidative addition or C–H activation), (2) alkyne insertion
into the Pd–C bond to form an alkenyl complex, (3) alkene insertion
into the Pd–C bond to form an alkyl intermediate, and (4) β-H
elimination, assisted by a base, to afford the final diene. In our
case, and due to the particular nature of the initial aryl group,
we have been able to isolate all of the organometallic intermediates
involved in the process.The crystal structure of the ammonium
triflate 5b·H2O was solved by X-ray diffraction
studies (Figure )
and showed a slightly distorted
planar diene skeleton (mean deviation from the plane: C(7)–C(8)–C(9)–C(1)–C(14)–C(21)–C(31)–C(41)
= 0.041 Å) with an E,E geometry.
Within the butadiene fragment, the average C–C double-bond
distance (C7–C8, C9–C14) was 1.35(2) Å, being slightly
longer than that corresponding to the most substituted unsaturated
unit, and the average single-bond value (C1–C7, C7–C41,
C8–C9, C8–C31, C14–C21) was 1.48(2) Å. The
aryl rings formed angles of 62.5, 20.8, 69.1, and 48.0° with
respect to the diene plane, adopting a helical conformation and releasing
steric hindrance.
Figure 6
Thermal ellipsoid plot (50% probability) of the cation
of compound 5b·H2O along with the labeling
scheme. The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
C(1)–C(7) = 1.503(2), C(7)–C(8) = 1.359(2), C(8)–C(9)
= 1.461(2), C(9)–C(14) = 1.341(2), C(14)–C(21) = 1.466(2),
C(8)–C(31) = 1.496(2), C(7)–C(41) = 1.487(2); C(1)–C(7)–C(41)
= 115.93(15), C(1)–C(7)–C(8) = 120.51(16), C(7)–C(8)–C(9)
= 120.85(16), C(8)–C(9)–C(14) = 127.20(17), C(9)–C(8)–C(31)
= 116.96(15).
Thermal ellipsoid plot (50% probability) of the cation
of compound 5b·H2O along with the labeling
scheme. The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
C(1)–C(7) = 1.503(2), C(7)–C(8) = 1.359(2), C(8)–C(9)
= 1.461(2), C(9)–C(14) = 1.341(2), C(14)–C(21) = 1.466(2),
C(8)–C(31) = 1.496(2), C(7)–C(41) = 1.487(2); C(1)–C(7)–C(41)
= 115.93(15), C(1)–C(7)–C(8) = 120.51(16), C(7)–C(8)–C(9)
= 120.85(16), C(8)–C(9)–C(14) = 127.20(17), C(9)–C(8)–C(31)
= 116.96(15).We also attempted to insert a
third unsaturated molecule into the
Pd–C bonds of the η3-allyl complexes by bubbling
CO through a solution of 1b in CH2Cl2 (room temperature, 4 h). Nevertheless, no reaction was observed,
probably because these complexes were too stable. A similar reaction,
with addition of TlOTf and use of THF as the solvent, led to a cationic
complex with no CO inserted.
Insertion of 2-Norbornene
into the Pd–C Bond
Synthesis and Structure of 10-Membered Norbornyl
Palladium(II)
Complexes
In contrast to styrene or ethyl acrylate, when
1 equiv of 2-norbornene was added to a solution of the alkenyl complex b (arising from insertion of one molecule of diphenylacetylene
into the Pd–C bond of palladacycle B; see Scheme for its structural
formula), no insertion reaction was observed, neither at room temperature
nor on heating the mixture to 65 °C in CHCl3. We also
attempted the insertion reactions using as starting materials the
eight-membered palladacycles c and d, arising
from the insertion of methyl phenylpropiolate and 1-phenyl-1-propyne,
respectively, into the Pd–C bond of complex B.
When palladacycle d was used, the reaction at room temperature
led to a new species that was tentatively assigned to the norbornene
coordination monomer species 6d (Scheme ), which could not be fully characterized,
since it evolved easily to the starting dimeric complex d when we tried to crystallize it. Nevertheless, when these reaction
mixtures (palladacycle c or d and 2-norbornene)
were heated to 65 °C, we could successfully isolate the complexes 7c,d, where the insertion of the 2-norbornene
moiety into the alkenyl Pd(II) complex had taken place (Scheme ). These complexes have a monomeric
nature, since the Pd center completes its coordination sphere with
the intramolecular olefin moiety.
Scheme 7
Sequential Insertion of Alkyne/2-Norbornene
into the Pd–C
Bond of Six-Membered Palladacycle B
The crystal structures of complexes 7c·CHCl3 and 7d·1/2CHCl3 were solved
by X-ray diffraction studies (see Figure and the Supporting Information), and both showed similar features. For the complex 7c·CHCl3, the atoms Cl(1), N(1), and C(1) together
with the midpoint of the C(3)–C(4) double bond form a square
plane around the palladium atom (mean deviation from the plane: Pd(1)–C(1)–N(1)–Cl(1)–X
0.0230 Å, where X = centroid from C(3) and C(4) atoms). The norbornene
unit adopted an exo conformation, arising from the syn addition of the Pd–C bond to the exo face of the olefin, as expected. The C(3)–C(4) double bond
forms an angle of 61.8° with the palladium coordination plane.
The methyl and phenyl substituents are mutually trans, which requires the isomerization of the first inserted alkyne.
This arrangement reduces the steric hindrance between the substituents
and is the normal behavior for di-inserted derivatives containing
cyclometalated amines.[31]
Figure 7
Thermal ellipsoid plot
(50% probability) of the complex 7c·CHCl3 along with the labeling scheme. The solvent
molecule and the hydrogen atoms bonded to carbon have been omitted
for clarity. Selected bond lengths (Å) and angles (deg): Pd(1)–N(1)
= 2.2185(19), Pd(1)–Cl(1) = 2.3589(5), Pd(1)–C(1) =
2.047(2), Pd(1)–C(3) = 2.159(2), Pd(1)–C(4) = 2.203(2),
Pd(1)–X = 2.065, C(1)–C(2) = 1.553(3), C(2)–C(3)
= 1.542(3), C(3)–C(4) = 1.403(3), C(4)–C(5) = 1.512(3);
N(1)–Pd(1)–Cl(1) = 88.03(5), Cl(1)–Pd(1)–C(1)
= 93.88(6), C(1)–Pd(1)–X = 76.7, X–Pd(1)–N(1)
= 101.3, C(2)–C(3)–C(4) = 117.17(17), C(3)–C(4)–C(5)
= 122.65(18). X represents the midpoint of the double bond C(3)–C(4).
Thermal ellipsoid plot
(50% probability) of the complex 7c·CHCl3 along with the labeling scheme. The solvent
molecule and the hydrogen atoms bonded to carbon have been omitted
for clarity. Selected bond lengths (Å) and angles (deg): Pd(1)–N(1)
= 2.2185(19), Pd(1)–Cl(1) = 2.3589(5), Pd(1)–C(1) =
2.047(2), Pd(1)–C(3) = 2.159(2), Pd(1)–C(4) = 2.203(2),
Pd(1)–X = 2.065, C(1)–C(2) = 1.553(3), C(2)–C(3)
= 1.542(3), C(3)–C(4) = 1.403(3), C(4)–C(5) = 1.512(3);
N(1)–Pd(1)–Cl(1) = 88.03(5), Cl(1)–Pd(1)–C(1)
= 93.88(6), C(1)–Pd(1)–X = 76.7, X–Pd(1)–N(1)
= 101.3, C(2)–C(3)–C(4) = 117.17(17), C(3)–C(4)–C(5)
= 122.65(18). X represents the midpoint of the double bond C(3)–C(4).We performed an analogous experiment reversing
the order of the
addition of the unsaturated reagents. That is, we studied the reactivity
of the previously reported norbornyl derivative e (arising
from the insertion of 2-norbornene into the six-membered palladacyle B) toward alkynes (Scheme ). A suspension of the norbornyl Pd(II) complex e in CHCl3 was heated to 65 °C in the presence
of methyl phenylpropiolate. The complex 7d, which arose
from a reverse insertion order of the reagent addition, was isolated
in 20% yield from the reaction mixture. In this reaction, palladacycle b was also formed. The formation of 7d could
be explained through the deinsertion of the 2-norbornene fragment
in e to give the starting palladacycle B, which in turn would undergo the sequential insertion of the alkyne
and 2-norbornene. The ability of this cyclic olefin to insert reversibly
into Pd–aryl bonds is well-known. This behavior has promoted
the use of norbornene as an essential ligand in the functionalization
of haloarenes: for instance, in the versatile Catellani type reaction.[2] We tried to apply this strategy to obtain the
diphenylacetylene/2-norbornene insertion derivative. Nevertheless,
the reaction of the norbornyl complex e with diphenylacetylene
gave rise mainly to the stable eight-membered palladacycle b, arising from monoinsertion of the alkyne (Scheme ). That is, the olefin did not insert into
the Pd–C bond of b, as noted above.
Scheme 8
Sequential
Insertion of Norbornene/Alkynes into the Pd–C Bond
of Palladacycle B
The synthesis of alkenyl palladacycles arising from the insertion
of one molecule of alkyne into the Pd–C bond of a starting
complex is not always a simple task, due to the possibility of di-
and tri-insertion processes. This aspect is especially relevant when
a very electrophilic alkyne, such as dimethyl acetylenedicarboxylate
(DMAD), is used. We have previously tried to isolate the palladacycle f arising from monoinsertion of DMAD into the Pd–C
bond of B; nevertheless, mixtures of mono-, di-, and
tri-inserted products were obtained (Scheme ). We thought that performing the reaction
of B and DMAD in the presence of 2-norbornene could drive
the reaction to the formation of a sequential DMAD/norbornene insertion
product, given the fact that norbornene seems to react readily with
other monoinserted alkyne derivatives. Indeed, when a solution of
the palladacycle B in CHCl3 was heated to
65 °C in the presence of 1 equiv of DMAD and 1 equiv of 2-norbornene,
the alkyne/alkene sequential insertion product 7f was
isolated in good yield (Scheme ). The crystal structure of complex 7f was also
solved by X-ray diffraction studies (see the Supporting Information) and showed features similar to those discussed
for 7c·CHCl3.
Scheme 9
Sequential Insertion
of Norbornene/DMAD into the Pd–C Bond
of Palladacycle B
Reactivity of the Norbornyl Complexes
We explored the
reactivity of the complex 7f toward KOtBu
in MeCN at 78 °C. From the reaction mixture, we could isolate
the tetrahydroisoquinoline derivative 8f by treatment
of the crude product with HOTf: that is, from the protonation of V (Scheme ). A possible path to explain the formation of 8f would
involve the intramolecular carbopalladation of the alkene moiety,
giving rise to a cyclopropyl ring. The new organometallic intermediate IV would then undergo a C–N coupling process with reductive
elimination of Pd(0). Several Pd-catalyzed procedures to promote the
cyclopropanation of 2-norbornene have been reported in the literature.[32] Some of these procedures rely on the migratory
insertion of norbornene into a Pd alkynyl[33] or Pd alkenyl[34] complex, and some organometallic
intermediates similar to complexes 7 have been proposed,
where the alkenyl moiety is coordinated intramolecularly to Pd.
Scheme 10
Reaction of Complex 7f with KOtBu
The crystal structure of 8f was
solved by X-ray diffraction
studies (Figure )
and showed the isoquinoline nucleus derived from phentermine substituted
at C1 with a methoxycarbonyl group and a tricyclo[3.2.1]octyl ring.
It is worth noting that highly strained hydrocarbons such as this
cyclopropane-containing norbornyl moiety have been tested as new highly
energetic materials for liquid-fueled propulsion systems.[35]
Figure 8
Thermal ellipsoid plot (50% probability) of the cation
of compound 8f along with the labeling scheme. The hydrogen
atoms bonded
to carbon have been omitted for clarity. Selected bond lengths (Å)
and angles (deg): N(1)–C(8) = 1.5233(17), N(1)–C(9)
= 1.5134(16), C(9)–C(10) = 1.5595(18), C(10)–C(13) =
1.5071(17), C(10)–C(18) = 1.5259(18), C(13)–C(18) =
1.5261(17); C(1)–C(9)–N(1) = 111.04(10), N(1)–C(9)–C(20)
= 105.93(10), C(20)–C(9)–C(10) = 110.53(10), C(10)–C(18)–C(13)
= 59.18(8), C(18)–C(13)–C(10) = 60.40(8), C(13)–C(10)–C(18)
= 60.42(8).
Thermal ellipsoid plot (50% probability) of the cation
of compound 8f along with the labeling scheme. The hydrogen
atoms bonded
to carbon have been omitted for clarity. Selected bond lengths (Å)
and angles (deg): N(1)–C(8) = 1.5233(17), N(1)–C(9)
= 1.5134(16), C(9)–C(10) = 1.5595(18), C(10)–C(13) =
1.5071(17), C(10)–C(18) = 1.5259(18), C(13)–C(18) =
1.5261(17); C(1)–C(9)–N(1) = 111.04(10), N(1)–C(9)–C(20)
= 105.93(10), C(20)–C(9)–C(10) = 110.53(10), C(10)–C(18)–C(13)
= 59.18(8), C(18)–C(13)–C(10) = 60.40(8), C(13)–C(10)–C(18)
= 60.42(8).We studied whether a further insertion
of an unsaturated molecule
could be carried out in the complexes 7c,d,f, obtained after alkyne/norbornene insertion. Complex 7f did not react when it was stirred in CHCl3 in
the presence of CO at room temperature for 5 h. When the reaction
mixture was heated to 65 °C, a complicated mixture was formed.
In order to facilitate the insertion of CO, we performed the reaction
in the presence of TlOTf in acetone, hence generating a cationic Pd(II)
intermediate in situ. In this case, the reaction
afforded the lactone 9f and Pd(0) (Scheme ). The formation of 9f could be explained as the result of the hydrolysis of the
acyl Pd(II) intermediate generated upon CO insertion into the Pd–C
bond present in 7f. Under those conditions, the carboxylic
acid derivative underwent an intramolecular Michael addition of the
CO2H group to the α,β-unsaturated ester moiety.
Scheme 11
Reactions of Complexes 7 with CO
When the cationic derivatives of complexes 7c,d (generated in situ) reacted with
CO at
room temperature, the amino acid derivatives 10c,d were obtained (Scheme ). In the case of 10d, intramolecular
cyclization to give 9d took place after heating at 65
°C in CHCl3 for 7 days. As expected, the derivative 10c was stable upon heating, since it was not activated for
the Michael addition step, as occurred in the cases of the alkenyl
complexes bearing a CO2Me substituent.The crystal
structure of 9f·Et2O has
been solved by X-ray diffraction studies (Figure ) and showed an ammonium salt derived from
the starting phentermine containing a bicyclo[2.2.1]heptane lactone
substituent. The hexahydro-4,7-methanoisobenzofuran-1(3H)-one core is a known compound that has attracted interest because
of its potential use as a prostaglandin/thromboxane receptor antagonist.[36] The most frequent routes for its synthesis involved
(a) the cycloaddition reaction of furan-2(5H)-one
and cyclopentadiene[37] and (b) the oxidation
of cis-2,3-bis(hydroxymethyl)bicyclo[2.2.1]heptane,
which could be performed in a stoichiometric way with standard organic
oxidants[38] or by catalysis, with the use
of enzymes[39] or transition metals.[40]
Figure 9
Thermal ellipsoid plot (50% probability) of the cation
of 9f·Et2O along with the labeling scheme.
The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
N(1)–C(8) = 1.5209(19), C(11)–C(12) = 1.522(2), C(11)–C(14)
= 1.557(2), C(14)–O(4) = 1.4423(18), C(17)–O(3) = 1.2010(19),
C(17)–O(4) = 1.3514(19), C(16)–C(17) = 1.499(2); C(1)–C(11)–C(12)
= 107.19(12), C(11)–C(14)–O(4) = 109.88(12), C(14)–O(4)–C(17)
= 111.86(11), O(4)–C(17)–C(16) = 111.77(13), C(17)–C(16)–C(15)
= 104.53(12), C(16)–C(15)–C(14) = 104.02(12), C(15)–C(14)–O(4)
= 106.07(11).
Thermal ellipsoid plot (50% probability) of the cation
of 9f·Et2O along with the labeling scheme.
The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
N(1)–C(8) = 1.5209(19), C(11)–C(12) = 1.522(2), C(11)–C(14)
= 1.557(2), C(14)–O(4) = 1.4423(18), C(17)–O(3) = 1.2010(19),
C(17)–O(4) = 1.3514(19), C(16)–C(17) = 1.499(2); C(1)–C(11)–C(12)
= 107.19(12), C(11)–C(14)–O(4) = 109.88(12), C(14)–O(4)–C(17)
= 111.86(11), O(4)–C(17)–C(16) = 111.77(13), C(17)–C(16)–C(15)
= 104.53(12), C(16)–C(15)–C(14) = 104.02(12), C(15)–C(14)–O(4)
= 106.07(11).The crystal structure of 10c has also been determined
by X-ray diffraction studies (Figure ), and it showed the ammonium salt derived from the
sequential insertion of alkyne/norbornene/CO into the initial phentermine
palladacycle. Both substituents (the carboxylate and the alkenyl groups)
at the norbornadienyl moiety were in an exo disposition.
Figure 10
Thermal
ellipsoid plot (50% probability) of the cation of 10c·H2O along with the labeling scheme. The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
N(1)–C(8) = 1.518(2), C(1)–C(11) = 1.504(3), C(11)–C(14)
= 1.342(3), C(14)–C(15) = 1.532(2), C(15)–C(16) = 1.582(2),
C(16)–C(17) = 1.506(3), C(17)–O(1) = 1.218(2), C(17)–O(2)
= 1.328(2); N(1)–C(8)–C(7) = 104.46(14), C(1)–(11)–C(14)
= 121.73(17), C(11)–C(14)–C(15) = 121.56(17), C(14)–C(15)–C(16)
= 118.26(15), C(15)–C(16)–C(17) = 116.58(15), C(16)–C(17)–O(1)
= 126.22(17), C(16)–C(17)–O(2) = 111.65(16).
Thermal
ellipsoid plot (50% probability) of the cation of 10c·H2O along with the labeling scheme. The
solvent molecule and the hydrogen atoms bonded to carbon have been
omitted for clarity. Selected bond lengths (Å) and angles (deg):
N(1)–C(8) = 1.518(2), C(1)–C(11) = 1.504(3), C(11)–C(14)
= 1.342(3), C(14)–C(15) = 1.532(2), C(15)–C(16) = 1.582(2),
C(16)–C(17) = 1.506(3), C(17)–O(1) = 1.218(2), C(17)–O(2)
= 1.328(2); N(1)–C(8)–C(7) = 104.46(14), C(1)–(11)–C(14)
= 121.73(17), C(11)–C(14)–C(15) = 121.56(17), C(14)–C(15)–C(16)
= 118.26(15), C(15)–C(16)–C(17) = 116.58(15), C(16)–C(17)–O(1)
= 126.22(17), C(16)–C(17)–O(2) = 111.65(16).
Conclusion
In summary, the stoichiometric
sequential insertion of alkynes
and alkenes into the Pd–C bond of metalated phenethylamines
has been studied. The result of these reactions depends on the nature
of the olefin used. The insertion of styrene or ethyl acrylate into
the eight-membered palladacycle arising from monoinsertion of diphenylacetylene
affords the η3-allyl species 1 or 2, via sequential β-H elimination/hydropalladation steps.
However, the insertion of 2-norbornene leads to norbornyl Pd(II) complexes
with an alkenyl moiety intramolecularly coordinated to the metal center.The treatment with a strong base of the organometallic complexes
obtained upon the sequential insertion of alkyne/alkene (1, 2, and 7) also led to different results.
While the η3-allyl complexes 1 and 2 afforded 1,3-butadienyl-substituted phenethylamines (4b and 5b), the norbornyl Pd derivative 7f gave a tetrahydroisoquinoline derivative upon cyclopropanation
of the norbornene fragment and C–N coupling (8f).The η3-allyl complexes 1 and 2 were unreactive toward CO insertion. In contrast, the norbornyl
derivatives 7c,d,f afforded
interesting amino acid or lactone derivatives upon CO insertion into
the Pd–C bond and subsequent hydrolysis, under the appropriate
conditions.Overall, we have shown that the isolation of the
organometallic
intermediates obtained upon stepwise insertion of alkynes and alkenes
into palladated phenethylamines allows the synthesis of functionalized
primary phenethylamines. These stoichiometric routes avoid the problems
associated with substrates that make difficult the development of
catalytic processes (such as primary alkylamines) and allow the control
of the regioselective insertion of the different unsaturated coupling
partners.
Experimental Section
Caution! Special precautions should be taken in handling
thallium(I) compounds, which are toxic.
General Procedures
Infrared spectra were recorded on
a PerkinElmer 16F-PC-FT spectrometer. C, H, N, and S analyses were
carried out with a LECO CHNS-932 microanalyzer. Conductance measurements
and melting point determinations were carried out as described elsewhere.[41] Unless stated otherwise, NMR spectra were recorded
in CDCl3 with Bruker Avance 300, 400, and 600 spectrometers.
Chemical shifts are referenced to TMS (1H and 13C{1H}). Signals in the 1H and 13C NMR spectra of all compounds were assigned with the help of APT,
HMQC, and HMBC experiments. High-resolution electrospray ionization
mass spectra (ESI-MS) were recorded on an Agilent 6220 Accurate-Mass
time-of-flight (TOF) LC/MS. Reactions were carried out at room temperature
without special precautions against moisture unless specified otherwise.
The groups C6H4 and C6H2 are denoted by Ar. Free and inserted 2-norbornene are denoted by
C7H10 (free), CH(C5H8)CH
(inserted), and nor (both, NMR signals).Styrene, ethyl acrylate,
2-norbornene, dimethyl acetylenedicarboxylate (DMAD), methyl phenylpropiolate, p-toluidine (Tol), KOtBu, and HOTf (HO3SCF3) were used as received from comertial sources. The
palladacycles [Pd{C,N-C6H2(CH2CH2NH2)-2-(OMe)2-4,5}(μ-Br)]2 (A),[26] [Pd{C,N-C6H4(CH2CMe2NH2)-2}(μ-Cl)]2 (B),[24] [Pd{C,N-C(Ph)=C(Ph)C6H2(CH2CH2NH2)-2-(OMe)2-4,5}(μ-Br)]2 (a), [Pd{C,N-C(Ph)=C(R)C6H4(CH2CMe2NH2)-2}(μ-Cl)]2 (R = Ph
(b), Me (c), CO2Me (d)),[20] and [Pd{C,N-CH(C5H8)CHC6H4(CH2CMe2NH2)-2}(μ-Cl)]2 (e)[19] were prepared as previously reported.
TlOTf was prepared by the reaction of Tl2CO3 and HO3SCF3 (1/2) in water and recrystallized
from acetone/Et2O. Chart gives the numbering schemes for the new organometallic
and organic derivatives.
Chart 2
Numbering Schemes for the New Palladium(II)
Complexes and the Organic
Derivatives
Synthesis of anti-/syn-1a
Styrene (23 μL,
0.202 mmol) was added to
a solution of palladacycle a (110 mg, 0.100 mmol) in
CH2Cl2 (10 mL), and the mixture was stirred
for 20 h. The resulting mixture was filtered through a plug of Celite,
the filtrate was concentrated to ca. 1 mL, and Et2O (15
mL) was added. The mixture was cooled to 0 °C, the resulting
suspension was filtered, and the pale yellow solid was air-dried to
give a mixture of anti-/syn-1a. Yield: 86 mg, 0.132 mmol, 66%. Mp: 143 °C. Anal.
Calcd for C32H32BrNO2Pd (648.936):
C, 59.23; H, 4.97; N, 2.16. Found: C, 59.32; H, 5.10; N, 2.19. IR
(cm–1): ν(NH) 3319 m, 3258 w, 3206 m, 3135
w. The NMR spectra of this complex showed two sets of signals in approximately
a 2.5/1 ratio (determined by 1H integration), corresponding
to the mixture anti-/syn-1a. Data for anti-1a (extracted from
the mixture) are as follows. 1H NMR (300.1 MHz): δ
1.63–1.70 (m, 1 H, NH2), 1.88 (m, 1 H, CH2Ar), 2.56 (br dd, 1 H, CH2Ar, 2JHH = 15.3, 3JHH = 7.8 Hz), 2.76–2.85 (m, partially obscured by the resonance
of MeO, 1 H, CH2Ph), 2.86 (s, 3 H, MeO), 2.97–3.02
(br d, 1 H, CH2N, 2JHH = 12.6 Hz), 3.22–3.35 (m, 2 H, 1 H of CH2N + 1
H of NH2), 2.38 (br dd, 1 H, CH2Ph, 2JHH = 17.4, 3JHH = 4.5 Hz), 3.87 (s, 3 H, MeO), 5.11 (dd, 1 H, CH, 3JHH = 12.3, 3JHH = 4.5 Hz), 6.47 (s, 1 H, H6), 6.64 (s, 1
H, H3), 6.80–7.27 (m, 15 H, Ph). 13C{1H} NMR (75.5 MHz): δ 35.2 (CH2Ar), 39.1 (CH2Ph), 42.9 (CH2N), 54.7 (MeO), 55.9 (MeO), 85.0
(CH), 93.8 (C7), 111.9 (CH6), 116.1 (CH3), 123.6 (C8), 125.8 (CH,
Ph), 126.8 (CH, Ph), 127.0 (CH, Ph), 127.3 (CH, Ph), 128.0 (CH, Ph),
128.4 (CH, Ph), 128.5 (CH, Ph), 128.6 (CH, Ph), 129.6 (CH, Ph), 132.2
(C1), 135.1 (i-C, Ph), 140.4 (i-C, Ph), 141.1 (i-C, Ph), 143.6 (C2), 146.4 (C5),
148.3 (C4).
anti/syn Isomerization of 1a
An NMR tube was charged
with a 2.5:1 mixture of anti-/syn-1a (20 mg) and CDCl3 (0.6 mL), and the solution
was heated at 60 °C. The
sample was checked by 1H NMR periodically, until complete
conversion. After 10 days at 60 °C, a spectroscopically pure
sample of syn-1a was obtained. Data
for syn-1a are as follows. 1H NMR (400.9 MHz): δ 1.61–1.67 (m, 1 H, NH2), 1.89 (br dd, 1 H, CH2Ar, 2JHH = 15.3, 3JHH = 10.8 Hz), 2.47 (br dd, 1 H, CH2Ar, 2JHH = 15.4, 3JHH = 8.0 Hz), 2.86–2.90 (m, 1 H, CH2Ph),
3.23 (dd, partially obscured by the resonance of NH2, 1
H, CH2N, 2JHH =
12.0, 3JHH = 8.0 Hz), 3.02
(br d, 1 H, NH2, 2JHH = 12.0 Hz), 3.42 (s, 3 H, MeO), 3.46 (br dd, 1 H, CH2Ph, 2JHH = 14.2, 3JHH = 3.5 Hz), 3.87 (s, 3 H, MeO), 4.80
(dd, 1 H, CH, 3JHH = 11.0, 3JHH = 3.7 Hz), 6.39 (s, 1 H, H6),
6.60 (s, 1 H, H3), 6.19 (br d, 1 H, Ph, 3JHH = 7.0 Hz), 6.80 (br d, 2 H, Ph, 3JHH = 7.0 Hz), 6.90–7.23 (m, 10 H, Ph),
7.40 (m, 1 H, Ph), 7.85 (br d, 1 H, Ph, 3JHH = 6.0 Hz). 13C{1H} NMR (75.5
MHz): δ 34.8 (CH2Ar), 35.8 (CH2Ph), 43.6
(CH2N), 55.6 (MeO), 60.0 (MeO), 84.5 (CH), 90.8 (C7), 110.0
(CH6), 116.2 (CH3), 124.0 (C8), 126.0 (CH, Ph), 127.0 (CH, Ph), 127.3
(CH, Ph), 127.7 (CH, Ph), 128.4 (CH, Ph), 128.5 (CH, Ph), 129.0 (CH,
Ph), 131.8 (C1), 133.2 (i-C, Ph), 133.7 (CH, Ph),
135.1 (i-C, Ph), 140.5 (i-C, Ph),
142.4 (C2), 147.5 (C5), 148.2 (C4).
Synthesis of anti-/syn-1b
Method A
Styrene
(52 μL, 0.453 mmol) was added
to a solution of palladacycle b (200 mg, 0.213 mmol)
in CH2Cl2 (10 mL). The solution was stirred
for 12 h, the solvent was concentrated to ca. 5 mL, Et2O (5 mL) was added, and the resulting suspension was filtered through
a plug of Celite. The filtrate was concentrated to ca. 1 mL, and n-pentane (20 mL) was added. The suspension was filtered,
and the pale yellow solid was washed with n-pentane
(2 × 5 mL) and air-dried to give a mixture of anti-/syn-1b (ratio ca. 1.25:1 by 1H NMR). Yield: 155 mg, 0.270 mmol, 63%. Dec pt: 252 °C. Anal.
Calcd for C32H32ClNPd (572.485): C, 67.14; H,
5.63; N, 2.44. Found: C, 67.19; H, 5.60; N, 2.39. IR (cm–1): ν(NH) 2295 w, 3238 w.
Method B
In a
different preparation, it was possible
to obtain two different crops by fractional crystallization, each
of them enriched in one of the isomers. Styrene (30 μL, 0.262
mmol) was added to a solution of palladacycle b (120
mg, 0.128 mmol) in CH2Cl2 (10 mL). The mixture
was stirred for 12 h and filtered through a plug of Celite. The filtrate
was concentrated to ca. 3 mL, and Et2O (15 mL) was added.
The resulting suspension was filtered, and the solid was air-dried
to afford 57 mg of an anti-enriched mixture of both
isomers (anti/syn = 3/1). The mother
liquors were concentrated to ca. 3 mL, and n-pentane
(20 mL) was added. The resulting suspension was filtered, and the
solid was air-dried to afford 25 mg of a syn-enriched
mixture of both isomers (anti/syn = 1/3). Almost spectroscopically pure samples of anti- and syn-1b could be obtained by growing
single crystals from the enriched mixtures. Data for anti-1b are as follows. 1H NMR (300.1 MHz): δ
1.34 (s, 6 H, CMe2), 1.86 (br d, 1 H, NH2, 2JHH = 10.8 Hz), 2.11 (d, 1 H,
CH2Ar, 2JHH = 15.0
Hz), 2.38 (br d, 1 H, CH2Ar, 2JHH = 14.7 Hz), 2.59 (dd, 1 H, CHPh, 2JHH = 16.5, 3JHH = 11.7 Hz),
3.18 (br d, 1 H, NH2, 2JHH = 10.8 Hz), 3.60 (dd, 1 H, CHPh, 2JHH =
16.5, 3JHH = 4.5 Hz), 5.16
(dd, 1 H, CH, 3JHH = 11.7, 3JHH = 4.5 Hz), 6.86–6.91
(m, 3 H, Ph + Ar), 7.04–7.34 (m, 16 H, Ph + Ar). 13C{1H} NMR (75.5 MHz): δ 27.8 (Me, CMe2), 36.0 (Me, CMe2), 39.0 (CH2Ph), 46.8 (CH2Ar), 52.3 (CMe2), 88.3 (CH), 89.4 (C7), 122.9 (C8), 125.9 (CH), 126.3 (CH), 126.7
(CH), 127.0 (CH), 127.5 (CH), 127.8 (CH, Ph), 127.9 (CH, Ph), 128.3
(br s, CH, Ph), 129.2 (CH), 129.4 (CH), 128.5 (CH, Ph), 132.3 (CH),
134.9 (CH3), 137.6 (C2), 139.8 (i-C, Ph), 140.4 (C1),
141.0 (i-C, Ph), 143.2 (i-C, Ph).
Data for syn-1b are as follows. 1H NMR (400.9 MHz): δ 1.29 (s, 3 H, Me, CMe2), 1.30 (s, 3 H, Me, CMe2), 1.85 (br d, 1 H, NH2, 2JHH = 10.0 Hz), 2.09 (d,
1 H, CH2Ar, 2JHH = 14.8 Hz), 2.28 (dd, 1 H, CH2Ar, 2JHH = 14.8, 4JHH = 1.2 Hz), 3.17 (dd, 1 H, CHPh, 2JHH =
14.8, 3JHH = 10.6 Hz), 3.29
(br d, 1 H, NH2, 2JHH = 10.4 Hz), 3.40 (dd, 1 H, CHPh, 2JHH = 15.2, 3JHH = 4.0 Hz), 4.84 (dd, 1 H,
CH, 3JHH = 10.4, 3JHH = 4.0 Hz), 6.50 (br s, 1 H, Ph),
6.87–6.91 (m, 4 H, Ph), 6.99–7.04 (m, 5 H, H3 + H6 +
3 H of Ph), 7.08 (td, 1 H, H5, 3JHH = 7.6, 4JHH = 1.6
Hz), 7.11–7.20 (m, 6 H, Ph), 7.22 (td, 1 H, H4, 3JHH = 7.2, 4JHH = 1.6 Hz), 7.36 (br s, 1 H, Ph), 7.84 (br s, 1 H, Ph). 13C{1H} NMR (100.8 MHz): δ 27.2 (Me, CMe2), 35.1 (CH2Ph), 36.1 (Me, CMe2), 46.3 (CH2Ar), 53.1 (CMe2), 84.7 (CH), 88.0 (C7), 124.5 (C8), 125.8 (CH, Ph), 126.2
(CH6), 126.9 (CH5), 127.1 (CH, Ph), 127.28 (CH4 or Ph), 127.29 (CH4
or Ph), 127.5 (CH, Ph), 127.9 (br s, CH, Ph), 128.21 (CH, Ph), 128.25
(CH, Ph), 128.4 (CH, Ph), 128.7 (CH, Ph), 131.4 (CH, Ph), 133.6 (br
s, CH, Ph), 134.9 (i-C, Ph), 135.1 (CH3), 138.4 (C2),
139.7 (i-C, Ph), 140.0 (C1), 142.2 (i-C, Ph).
anti/syn Isomerization of 1b
An NMR tube was charged
with a 2.5/1 mixture of anti-/syn-1b (20 mg) and CDCl3 (0.6 mL), and the solution
was heated at 60 °C. The
sample was checked by 1H NMR periodically, until no change
was observed. After 5 days at 60 °C, a 1/10 mixture of anti-/syn-1b was obtained.
Synthesis of anti-/syn-2a
Ethyl acrylate (40 μL, 0.367 mmol) was added
to a solution of palladacycle a (200 mg, 0.183 mmol)
in CH2Cl2 (10 mL), and the mixture was stirred
for 48 h. The resulting mixture was filtered through a plug of Celite,
the filtrate was concentrated to ca. 1 mL, and Et2O (20
mL) was added. The suspension was filtered, and the pale yellow solid
was washed with Et2O (2 × 5 mL) and air-dried to give
an anti-/syn-2a mixture
(171 mg; ratio ca. 5/1 by 1H NMR). The filtrate was concentrated
to ca. 1 mL, and n-pentane (20 mL) was added. The
resulting suspension was filtered, and the solid was air-dried to
afford a anti-/syn-2a mixture (30 mg; ratio ca. 1/5 by 1H NMR). Yield: 201
mg, 0.312 mmol, 85%. Mp: 169 °C. Anal. Calcd for C29H32BrNO4Pd (644.901): C, 54.01; H, 5.00; N,
2.17. Found: C, 54.25; H, 5.00; N, 2.49. IR (cm–1): ν(NH) 3293 w, 3214 br; ν(CO) 1727 s, 1711 s. Data
for anti-2a are as follows. 1H NMR (400.9 MHz): δ 1.19 (t, 3 H, MeCH2, 3JHH = 7.2 Hz), 1.61
(br d, 1 H, NH2, 2JHH = 12.0 Hz), 1.90 (dd, 1 H, CH2Ar, 2JHH = 10.8, 3JHH = 4.8 Hz), 2.35 (dd, 1 H, CHCO2Et, 2JHH = 17.3, 3JHH = 11.6
Hz), 2.58 (br dd, 1 H, CH2Ar, 2JHH = 15.6, 4JHH = 6.8 Hz), 2.93 (m, 1 H, CH2N), 3.26–3.33 (m,
3 H, 1 H of NH2 + 1 H of CH2N + 1 H of CHCO2Et), 3.86 (s,
3 H, MeO), 3.91 (s, 3 H, MeO), 4.01 (m, 1 H, CH2O), 4.15
(m, 1 H, CH2O), 4.85 (dd, 1 H, CH, 3JHH = 11.6, 3JHH = 4.8 Hz), 6.68 (s, 1 H, H3), 6.87 (s, 1 H, H6), 6.99–7.10
(m, 8 H, Ph), 7.11–7.25 (m, 2 H, Ph). 13C{1H} NMR (100.8 MHz): δ 14.1 (MeCH2), 35.1 (CH2Ar), 39.4 (CH2CO2Et), 42.9 (CH2N), 56.0 (MeO), 56.0 (MeO),
60.8 (CH2O), 79.1 (CH), 93.8 (C7), 112.5 (CH6), 116.4 (CH3),
123.9 (C8), 126.8 (CH), 127.2 (CH), 127.3 (CH), 127.9 (CH), 128.0
(CH), 129.6 (CH), 132.1 (C1), 132.6 (C2), 140.7 (i-C, Ph), 143.1 (i-C, Ph), 147.0 (C5), 148.6 (C4),
170.0 (CO).
anti/syn Isomerization of 2a
An NMR tube was charged
with a 5/1 anti-/syn-2a mixture (20 mg) and CDCl3 (0.6 mL), and the solution
was heated at 60 °C. The
sample was checked by 1H NMR periodically, until complete
conversion. After 48 h at 60 °C, a spectroscopically pure sample
of syn-2a was obtained. Data for syn-2a are as follows. 1H NMR (300.1
MHz): δ 1.10 (t, 3 H, MeCH2, 3JHH = 7.2 Hz), 1.67 (br d, 1 H,
NH2, 2JHH = 10.8
Hz), 1.88 (dd, 1 H, CH2Ar, 2JHH = 15.6, 3JHH = 10.8 Hz), 2.49 (dd, 1 H, CH2Ar, 2JHH = 15.3, 3JHH = 6.6 Hz), 2.80 (dd, 1 H, CHCO2Et, 2JHH = 18.0, 3JHH = 10.2
Hz), 2.90 (m, partially obscured by the resonance of CHCO2Et, 1 H, CH2N), 3.25–3.32 (m, 2 H, 1 H of CH2N + 1 H of CHCO2Et), 3.37 (br
d, 1 H, NH2, 2JHH = 9.9 Hz), 3.89 (s, 3 H, MeO), 3.93–3.96 (m, partially obscured
by the resonance of MeO, 1 H, CH2O), 3.96–4.05 (m,
partially obscured by the resonance of MeO, 1 H, CH2O),
3.99 (s, 3 H, MeO), 4.75 (dd, 1 H, CH, 3JHH = 10.2, 3JHH = 4.2 Hz), 6.63 (s, 1 H, H3), 6.87–6.91 (m, partially obscured
by the resonance of H6, 2 H, Ph), 6.93 (s, 1 H, H6), 7.01–7.17
(m, 6 H, Ph), 7.27–7.34 (m, 1 H, Ph), 7.65 (br d, 1 H, Ph, 3JHH = 6.0 Hz). 13C{1H} NMR (75.5 MHz): δ 14.1 (MeCH2), 34.7 (CH2Ar), 34.9 (CH2CO2Et), 43.7 (CH2N), 56.0 (MeO), 56.2
(MeO), 60.4 (CH2O), 75.2 (CH), 92.2 (C7), 110.1 (CH6),
116.2 (CH3), 125.6 (C8), 127.1 (CH), 127.3 (CH), 127.6 (CH), 127.8
(CH), 128.8 (CH), 131.5 (C1), 133.0 (CH), 133.1 (C2), 134.6 (i-C, Ph), 142.2 (i-C, Ph), 147.8 (C5),
148.4 (C4), 170.8 (CO).
Synthesis of anti-/syn-2b
Ethyl acrylate (50 μL,
0.460 mmol) was
added to a solution of palladacycle b (200 mg, 0.213
mmol) in CH2Cl2 (10 mL). The solution was stirred
for 12 h and then concentrated to ca. 5 mL. Et2O (5 mL)
was added, and the resulting suspension was filtered through a plug
of Celite. The filtrate was concentrated to ca. 1 mL, and n-pentane (20 mL) was added. The suspension was filtered,
and the pale yellow solid was washed with n-pentane
(2 × 5 mL) and air-dried to give as anti-/syn- mixture (ratio ca. 1.25/1 by 1H NMR). Yield:
197 mg, 0.346 mmol, 81%. Mp: 127 °C. Anal. Calcd for C29H32ClNO2Pd (568.451): C, 61.27; H, 5.67; N,
2.46. Found: C, 61.38; H, 5.42; N, 2.46. IR (cm–1): ν(NH) 3293 w, 3214 br; ν(CO) 1727 s, 1711 s.
Method
B
In a different preparation, it was possible
to obtain two different crops by fractional crystallization, each
of them enriched in one of the isomers. Ethyl acrylate (40 μL,
0.262 mmol) was added to a solution of palladacycle b (150 mg, 0.160 mmol) in CH2Cl2 (10 mL). The
mixture was stirred for 12 h and then filtered through a plug of Celite.
The filtrate was concentrated to ca. 2 mL, and Et2O (15
mL) was added. The resulting suspension was filtered, and the solid
was washed with Et2O (2 × 5 mL) and air-dried to afford
58 mg of a syn-enriched mixture of both isomers (anti/syn = 1/3). The mother liquors were
concentrated to ca. 5 mL and cooled to 0 °C. A suspension formed,
which was filtered, and the solid was air-dried to afford 60 mg of
an anti-enriched mixture of both isomers (anti/syn = 3/1). Almost spectroscopically
pure samples of anti- and syn-2b could be obtained by growing single crystals from the enriched
mixtures. Data for anti-2b are as follows. 1H NMR (400.9 MHz): δ 1.22 (t, 3 H, MeCH2, 3JHH = 7.2
Hz), 1.27 (s, 3 H, Me, CMe2), 1.32 (s, 3 H, Me, CMe2), 1.81 (br d, 1 H, NH2, 2JHH = 10.0 Hz), 2.09 (d, 1 H, CH2Ar, 2JHH = 14.8 Hz), 2.21 (dd, 1 H, CHCH, 2JHH = 17.2, 3JHH = 11.2 Hz), 2.36 (dd, 1 H, CH2Ar, 2JHH = 14.8, 4JHH = 1.6 Hz), 3.13 (br d, partially obscured by the resonance
of CHCH, 1 H, NH2, 2JHH = 9.6 Hz), 3.16
(dd, 1 H, CHCH, 2JHH = 17.2, 3JHH = 5.2 Hz), 4.11 (m, 2 H, CH2O),
4.94 (dd, 1 H, CH, 3JHH = 11.2, 3JHH = 5.2 Hz), 6.86–6.90
(m, 2 H, Ph), 7.03–7.11 (m, 6 H, 5 H of Ph + H6), 7.13 (td,
partially obscured by the resonance of H6, 1 H, H4, 3JHH = 7.6, 4JHH = 1.2 Hz), 7.21 (m, 2 H, o-H, Ph), 7.27
(m, partially obscured by the resonance of CHCl3, 1 H, p-H, Ph), 7.30 (dd, 1 H, H3, 3JHH = 7.6, 4JHH =
1.2 Hz), 7.34 (td, 1 H, H5, 3JHH = 7.6, 4JHH = 1.6 Hz). 13C{1H} NMR (100.8 MHz): δ 14.2 (MeCH2), 27.7 (Me, CMe2), 36.0 (Me, CMe2), 39.0 (CH2CH), 46.6 (CH2Ar), 52.3 (CMe2), 60.7 (CH2O), 81.0 (CH), 90.0 (C7), 123.5 (C8), 126.5 (CH5), 126.8 (p-CH, Ph), 127.1 (o-CH, Ph), 127.6 (m-CH, Ph), 127.9 (m-CH, Ph), 128.2 (CH4),
128.3 (m-CH, Ph), 128.9 (CH6), 129.1 (o-CH, Ph), 129.5 (o-CH, Ph), 132.2 (p-CH, Ph), 135.0 (CH3), 137.8 (C2), 140.49 (C1), 140.55 (i-C, Ph), 142.8 (i-C, Ph), 169.4 (CO). Data for syn-2b are as follows. 1H NMR (300.1
MHz): δ 1.16 (t, 3 H, MeCH2, 3JHH = 7.2 Hz), 1.28 (s, 3 H, Me,
CMe2), 1.30 (s, 3 H, Me, CMe2), 1.82 (br d,
1 H, NH2, 2JHH =
9.9 Hz), 2.09 (d, 1 H, CH2Ar, 2JHH = 14.7 Hz), 2.29 (dd, 1 H, CH2Ar, 2JHH = 14.4, 4JHH = 1.8 Hz), 2.76 (dd, 1 H, CHCH, 2JHH =
18.3, 3JHH = 9.9 Hz), 3.17
(dd, 1 H, CHCH, 2JHH = 18.3, 3JHH = 4.2 Hz), 3.22 (br d, partially obscured
by the resonance of CHCH, 1 H, NH2, 2JHH = 9.3 Hz), 4.01 (m, 2 H, CH2O), 4.79 (dd, 1 H, CH, 3JHH = 9.6, 3JHH = 4.2 Hz), 6.87–7.36 (m, 13 H, Ph
+ Ar), 7.65 (br s, 1 H, CH, Ph or Ar). 13C{1H} NMR (75.5 MHz): δ 14.1 (MeCH2), 27.2 (Me, CMe2), 34.4 (CH2CH), 36.1 (Me, CMe2), 46.3 (CH2Ar), 53.4 (CMe2), 60.5 (CH2O), 76.7 (CH), 88.9
(C7), 125.5 (C8), 126.6 (CH), 127.1 (CH), 127.3 (CH), 127.5 (CH),
127.6 (CH), 127.9 (CH), 128.3 (CH), 128.5 (br s, CH), 128.8 (CH),
129.5 (CH), 131.2 (CH), 133.0 (br s, CH), 134.5 (i-C, Ph), 135.1 (CH3), 138.3 (C2), 139.8 (C1), 142.0 (i-C, Ph), 170.4 (CO).
anti/syn Isomerization of 2b
An NMR tube was charged
with a 1.25/1 anti-/syn-1b mixture (20 mg)
and CDCl3 (0.6 mL), and the solution was heated at 60 °C.
The sample was checked by 1H NMR periodically, until no
change was observed. After 48 days at 60 °C, a 1/20 anti-/syn-2b mixture was obtained.
Synthesis
of anti-/syn-3b
TlOTf (56 mg, 0.158 mmol) was added to a suspension
of complex 1b (90 mg, 0.157 mmol) in acetone (10 mL),
and the resulting mixture was stirred for 2 h. The solvent was removed,
and CH2Cl2 (15 mL) was added. The suspension
was filtered through a plug of Celite, p-toluidine
(17 mg, 0.159 mmol) was added to the filtrate, and the mixture was
stirred for another 30 min. The solvent was concentrated to ca. 1
mL, and n-pentane was added. The suspension was filtered,
and the pale yellow solid was washed with cold n-pentane
(2 × 5 mL) and air-dried to give an anti-/syn-3b mixture (ratio ca. 1/1.33 by 1H NMR). Yield: 60.0 mg, 0.076 mmol, 48%. Mp: 120 °C. ΛM (Ω–1 mol–1 cm2): 106 (c = 2.7 × 10–4 M). Anal. Calcd for C40H41F3N2O3PdS (793.237): C, 60.57; H, 5.21; N, 3.53; S,
4.04. Found: C, 60.38; H, 5.17; N, 3.30; S, 4.04. IR (cm–1): ν(NH) 3295 w, 3251 m, 3159 w. Data for anti-3b (minor isomer, extracted from the mixture) are as
follows. δ 0.86 (s, 3 H, Me, CMe2), 1.27 (s, 3 H,
Me, CMe2), 1.92 (br d, 1 H, NH2, 2JHH = 12.6 Hz), 1.98 (d, 1 H, CH2, 2JHH = 15.2 Hz),
2.20–2.29 (m, 1 H, CH2, overlapped with one H of
CH2 of the major isomer), 2.29 (s, 3 H, Me, Tol), 2.44–2.55
(m, 1 H, CH2, overlapped with one H of CH2 of
the major isomer), 2.73 (dd, 1 H, CH2, 2JHH = 14.8, 3JHH = 9.2 Hz), 4.13 (br d, 1 H, NH2, 2JHH = 11.2 Hz), 4.74 (dd, 1 H, CH, 3JHH = 8.8, 3JHH = 6.0 Hz), 5.33 (br d, 1 H, NH2, 2JHH = 11.2 Hz), 5.54 (br
d, 1 H, NH2, 2JHH = 10.4 Hz), 6.32–7.55 (m, 23 H, Ph + Ar, overlapped with
the aromatic protons of the major isomer). Data for syn-3b (major isomer, extracted from the mixture) are as
follows. 1H NMR (400.9 MHz, −40 °C): δ
1.45 (s, 3 H, Me, CMe2), 1.48 (s, 3 H, Me, CMe2), 1.70 (br d, 1 H, NH2, 2JHH = 10.8 Hz), 2.06 (d, 1 H, CH2, 2JHH = 15.2 Hz), 2.20–2.29 (m,
1 H, CH2, overlapped with one H of CH2 of the
minor isomer), 2.33 (s, 3 H, Me, Tol), 2.44–2.55 (m, 1 H, CH2, overlapped with one H of CH2 of the minor isomer),
2.81 (dd, 1 H, CH2Ar, 2JHH = 16.4, 3JHH = 4.0
Hz), 3.92 (d, 1 H, NH2, 2JHH = 9.6 Hz), 4.24 (dd, 1 H, CH, 3JHH = 10.0, 3JHH = 4.0 Hz), 5.02 (br d, 1 H, NH2, 2JHH = 11.2 Hz), 6.30 (br d, 1 H, NH2, 2JHH = 10.0 Hz), 6.32–7.55
(m, 23 H, Ph + Ar, overlapped with the aromatic protons of the minor
isomer). Data for both isomers are as follows. 13C{1H} NMR (100.8 MHz, −40 °C): δ 20.7 (Me,
Tol, major), 20.8 (Me, Tol, minor), 25.9 (Me, CMe2, minor),
27.4 (Me, CMe2, major), 33.8 (CH2, minor), 34.4
(Me, CMe2, major), 34.8 (Me, CMe2, minor), 41.4
(CH2, major), 46.5 (CH2, minor), 47.2 (CH2, major), 52.6 (CMe2, major),
53.0 (CMe2, minor), 84.6 (CH, minor),
86.6 (C, major), 87.1 (C, minor), 93.7 (CH, major), 188.8 (CH, major),
119.8 (CH, minor), 119.9 (q, CF3, 1JCF = 318.8 Hz), 124.1 (C, major), 126.3 (CH), 126.4 (CH),
126.5 (CH), 126.7, 126.8 (C, minor), 127.1 (CH), 127.2 (CH), 127.3
(CH), 127.4 (CH), 127.6 (CH), 127.7 (CH), 127.8 (CH), 127.8 (CH),
127.9 (CH), 128.0 (CH), 128.1 (CH), 128.1 (CH), 128.4 (CH), 128.6
(CH), 128.7 (CH), 129.0 (CH), 129.5 (CH), 129.7 (CH), 129.8 (CH),
130.0 (CH), 131.5 (CH), 131.9 (CH), 132.7 (CH), 133.1 (C, major),
133.6 (C, minor), 134.1 (C, minor), 134.9 (CH), 135.0 (CH), 137.7
(C, major), 138.2 (C, minor), 138.5 (C, minor), 138.9 (C, minor),
139.0 (C, major), 139.4 (C, minor), 139.6 (C, major), 139.7 (C, major),
139.9 (C, major), 140.8 (C, minor), 141.2 (C, major).
Synthesis
of 4a·1/2H2O
In a Carius tube,
KOtBu (190 mg, 1.55 mmol) was added
to a solution of anti-/syn-1a (100 mg, 0.154 mmol) in dry toluene (10 mL), under a nitrogen
atmosphere. The mixture was heated at 100 °C for 12 h. Decomposition
to metallic palladium was observed. The solvent was removed, and CH2Cl2 (20 mL) was added to the residue. The resulting
suspension was filtered through a plug of Celite, the filtrate was
concentrated to ca. 1 mL, n-pentane (20 mL) was added,
and the mixture was cooled to 0 °C. The resulting suspension
was filtered, and the solid was air-dried to give 4a·1/2H2O as a colorless solid. Yield: 55 mg, 0.117 mmol,
76%. Mp: 129 °C. Anal. Calcd for C32H31NO2·1/2H2O (470.611): C, 81.67; H, 6.85;
N, 2.97. Found: C, 81.84; H, 6.60; N, 3.05. ESI-HRMS (m/z): exact mass calcd for C32H32NO2, 462.2433 [(M + H)+]; found, 462.2433.
IR (cm–1): ν(NH) 3375 w. 1H NMR
(300.1 MHz): δ 1.22 (br s, 3 H, NH2 + 1 H of H2O), 2.50–2.61 (m, 2 H, CH2Ar), 2.63–2.70
(m, 1 H, CH2N), 2.70–2.83 (m, 1 H, CH2N), 3.86 (s, 3 H, MeO), 3.95 (s, 3 H, MeO), 6.23 (d, 1 H, CH = , 3JHH = 15.9 Hz), 6.80 (s, 1 H,
H3), 6.81 (s, 1 H, H6), 6.86–6.90 (m, 2 H, Ph), 6.93 (d, 1
H, CH = , 3JHH = 15.9 Hz),
6.97–7.00 (m, 3 H, Ph), 7.12–7.24 (m, 7 H, Ph), 7.27–7.35
(m, 3 H, Ph). 13C{1H} NMR(100.8 MHz): δ
37.4 (CH2Ar), 42.8 (CH2N), 55.8 (MeO), 56.1
(MeO), 111.9 (CH3), 112.6 (CH6), 126.3 (CH, Ph), 126.4 (CH, Ph), 126.9
(CH, Ph), 127.3 (CH, Ph), 127.4 (CH, Ph), 128.1 (CH, Ph), 128.5 (CH,
Ph), 130.5 (CH, Ph), 130.9 (CH = ), 131.2 (C2), 131.3 (CH, Ph), 132.3
(CH = ), 133.8 (C1), 137.5 (i-C, Ph), 139.7 (i-C, Ph), 139.7 (i-C, Ph), 141.3 (C7),
141.8 (C8), 147.0 (C4), 148.2 (C5).
Synthesis of 5b·H2O
In
a Carius tube, KOtBu (250 mg, 2.04 mmol) was added to a
solution of anti-/syn-1b (120 mg, 0.209 mmol) in dry toluene (10 mL), under a nitrogen atmosphere.
The mixture was heated at 100 °C for 12 h. Decomposition to metallic
palladium was observed. The solvent was removed, and Et2O (20 mL) was added to the residue. The suspension was filtered through
a plug of Celite, and the solvent was removed from the filtrate to
give crude 4b as an oily residue, which was was characterized
by 1H NMR. Data for 4b are as follows. 1H NMR (400.9 MHz): δ 0.99 (br s, 2 H, NH2), 1.04 (s, 3 H, Me, CMe2), 1.09 (s, 3 H, Me, CMe2), 2.34, 2.54 (AB system, 2 H, CH2Ar, 2JAB = 13.6 Hz), 6.20 (d, 1 H, CH = , 3JHH = 15.6 Hz), 6.81–6.85
(m, 2 H), 6.96–7.00 (m, 3 H), 7.03 (d, 1 H, CH=, 3JHH = 16.0 Hz), 7.10–7.47
(m, 14 H). Crude 4b was dissolved in CH2Cl2 (5 mL), HOTf (0.05 mL, 0.565 mmol) was added, and the resulting
solution was stirred for 30 min. The solvent was concentrated to ca.
1 mL, Et2O (20 mL) was added, and the mixture was cooled
to 0 °C. The resulting suspension was filtered, and the solid
was washed with Et2O (2 × 5 mL) and air-dried to give
the salt 5b·H2O as a colorless solid.
Yield: 63 mg, 0.105 mmol, 50%. Mp: 131 °C. ΛM (Ω–1 mol–1 cm2): 108.5 (c = 5.09 × 10–4 M). Anal. Calcd for C33H32F3NO3S·H2O (597.696): C, 66.31; H, 5.73; N, 2.34;
S, 5.36. Found: C, 66.02; H, 5.59; N, 2.43; S, 5.59. ESI-HRMS (m/z): exact mass calcd for C32H32N, 430.2529 [(M – CF3SO3)+]; found, 430.2533. IR (cm–1): ν(NH)
3479 w. 1H NMR (400.9 MHz): δ 1.23 (s, 3 H, Me, CMe2), 1.25 (s, 3 H, Me, CMe2), 2.55, 2.62 (AB system,
2 H, CH2Ar, 2JAB = 14.0 Hz), 3.00 (br s, 2 H, H2O), 6.22 (d, 1 H, CH=, 3JHH = 16.0 Hz), 6.75–6.95
(m, 4 H, Ph), 6.94–6.97 (m, 2 H, Ph), 6.99 (d, 1 H, CH = , 3JHH = 16.0 Hz), 7.09–7.21
(m, 6 H, Ph), 7.22–7.27 (m, 4 H, 3 H of Ph + H3), 7.33 (td,
1 H, H4, 3JHH = 7.6, 4JHH = 1.2 Hz), 7.38 (td, 1 H, H5, 3JHH = 7.2, 4JHH = 1.2 Hz), 7.50 (dd, 1 H, H6, 3JHH = 7.2, 4JHH = 1.2 Hz), 7.70 (br s, 3 H, NH3). 13C{1H} NMR (100.8 MHz): δ 24.7 (Me, CMe2), 26.4 (Me, CMe2), 42.4 (CH2Ar), 57.0 (CMe2), 119.7 (q, CF3, 1JCF = 318.7 Hz), 126.5 (CH, Ph), 127.0
(CH, Ph), 127.3 (CH5), 127.6 (CH, Ph), 128.1 (CH, Ph), 128.5 (CH,
Ph), 128.6 (CH4), 130.3 (CH, Ph), 130.6 (CH = ), 130.9 (CH, Ph), 132.2
(CH3), 133.4 (CH = ), 133.9 (C2), 134.2 (CH6), 137.3 (i-C, Ph), 139.4 (i-C, Ph), 140.5 (C8), 141.26 (i-C, Ph), 141.34 (C7), 142.2 (C1).
Synthesis of [Pd{C,N-C(Ph)=C(CO2Me)C6H4CH2CMe2NH2-2}Cl(C7H10)] (6d)
2-Norbornene (C7H10; 25 mg, 0.265
mmol) was added to a solution of palladacycle d (65 mg,
0.072 mmol) in CH2Cl2 (10 mL), and the yellow
solution was stirred for 2 h. The solvent was removed under vacuum
at room temperature (to prevent the evolution of the complex), and
Et2O (10 mL) was added. The resulting suspension was filtered,
and the solid was washed with Et2O (2 × 5 mL) and
air-dried to afford complex 6d as a colorless solid.
Yield: 58 mg, 0.106 mmol, 73%. Mp: 182 °C dec. Anal. Calcd for
C27H32ClNO2Pd (544.429): C, 59.56;
H, 5.92; N, 2.57. Found: C, 59.37; H, 6.01; N, 2.41. IR (cm–1): ν(NH) 3323 w, 3267 w; n(CO) 1725 s. 1H NMR (400.9 MHz): δ 0.66 (br d, 1 H, CH, nor, 3JHH = 8.0 Hz), 0.97 (br d, 1 H,
CH2, nor, 2JHH =
9.6 Hz), 1.04 (br m, 1 H, CH2, nor), 1.11 (br m, 1 H, CH2, nor), 1.35 (s, 3 H, Me, CMe2), 1.38 (m, partially
obscured by the resonance of Me group, 2 H, CH2, nor),
1.40 (s, 3 H, Me, CMe2), 1.79 (br s, 1 H, CH, nor), 2.05
(br d, 1 H, CH2, nor, 2JHH = 10.0 Hz), 2.49 (d, 1 H, CH2Ar, 2JHH = 15.6 Hz), 2.67 (d, 1 H, CH2Ar, 2JHH = 15.2 Hz),
3.08 (br s, 1 H, CH, nor), 3.26 (s, 3 H, MeO), 4.36 (d, 1 H, CH, nor, 3JHH = 7.6 Hz), 7.16 (dd, 1 H,
H3, 3JHH = 7.6, 4JHH = 1.2 Hz), 7.23–7.35 (m, 6
H, Ar + Ph), 7.83 (d, 1 H, H6, 3JHH = 7.2 Hz), 7.95 (m, 1 H, o-H, Ph). The 1H resonance corresponding to the NH2 group was
not observed. 13C{1H} NMR (75.5 MHz): δ
18.0 (CH, nor), 27.6 (Me, CMe2), 27.8 (CH2,
nor), 28.9 (CH2, nor), 36.5 (Me, CMe2), 36.9
(CH2, nor), 40.4 (CH, nor), 40.6 (CH, nor), 47.1 (CH2Ar), 51.1 (CH, nor), 52.7 (MeO), 54.9 (CMe2), 99.8 (C7), 123.9 (CH6), 126.8 (CH, Ph), 127.0 (CH, Ph),
127.3 (CH, Ph), 127.4 (CH, Ph), 127.5 (CH5), 127.9 (CH4), 134.0 (C1),
134.1 (o-CH, Ph), 135.1 (CH3), 137.2 (i-C or C8), 138.4 (C2), 170.0 (CO).
Synthesis of [Pd{C,N-CH(C5H8)CHC(Ph)=C(Me)C6H4CH2CMe2NH2-2}Cl]·CHCl3 (7c·CHCl3)
In a Carius
tube, 2-norbornene
(100 mg, 1.06 mmol) was added to a solution of palladacycle c (250 mg, 0.308 mmol) in CHCl3 (15 mL), and the
mixture was heated at 65 °C for 12 h. The resulting solution
was filtered through a plug of Celite, the solvent was removed from
the filtrate, and Et2O (20 mL) was added. The resulting
suspension was filtered, and the solid was washed with Et2O (2 × 5 mL) and air-dried to give complex 7c·CHCl3 as a yellow solid. Yield: 258 mg, 0.416 mmol, 67%. Dec pt:
223 °C. Anal. Calcd for C26H32ClNPd·CHCl3 (619.796): C, 52.32; H, 5.37; N, 2.26. Found: C, 52.57; H,
5.48; N, 2.29. IR (cm–1): ν(NH) 3313 w, 3250
w. 1H NMR (300.1 MHz): δ 0.82 (br d, 1 H, NH2, 2JHH = 9.6 Hz), 0.95
(br d, 1 H, CH2, nor, 2JHH = 10.2 Hz), 1.07 (br d, 1 H, Hα, 3JHH = 7.5 Hz), 1.12 (s, 3 H, Me, CMe2), 1.20–1.27 (m, 2 H, CH2, nor), 1.37–1.50
(m, 2 H, CH2, nor), 1.64 (s, 3 H, Me, CMe2),
2.05 (br d, 1 H, CH2, nor, 2JHH = 9.9 Hz), 2.14 (br s, 1 H, CH, nor), 2.20 (br d, 1
H, NH2, 2JHH = 9.6
Hz), 2.35 (s, 3 H, MeC=), 2.57 (d, 1 H, CH2Ar, 2JHH = 12.9 Hz), 3.09 (br s, 1
H, CH, nor), 3.65 (d, 1 H, Hβ, 3JHH = 7.8 Hz), 3.96 (d, 1 H, CH2Ar, 2JHH = 12.9 Hz), 6.67 (d, 1 H, H6, 3JHH = 7.8 Hz), 6.76 (m, 1 H, H5),
7.01 (m, 2 H, H4 + H3), 7.18–7.25 (br m, partially obscured
by the resonance of CHCl3, 3 H, p-H + m-H, Ph), 2.26 (s, 1 H, CHCl3), 7.42 (br d, 2
H, o-H, Ph, 3JHH = 7.8 Hz). 13C{1H} NMR (75.5 MHz): δ
21.5 (CHα), 27.9 (MeC=), 28.5 (CH2, nor), 28.9 (CH2, nor), 30.7 (Me, CMe2), 30.9 (Me, CMe2), 37.3 (CH2, nor), 40.0 (CH,
nor), 41.1 (CH, nor), 48.1 (CH2Ar), 52.3 (CMe2), 54.6 (CHβ), 77.8 (C8), 107.7 (C7), 126.3 (CH5),
126.7 (CH4), 126.8 (p-CH, Ph), 131.0 (br s, CH6 or o-CH of Ph), 131.1 (CH6 or o-CH of Ph),
132.7 (CH3), 135.8 (C2), 138.6 (i-C, Ph), 142.6 (C1).
The 13C resonance corresponding to m-CH
of Ph was overlapped with the signal of CH6 or o-CH.
Synthesis of [Pd{C,N-CH(C5H8)CHC(Ph)=C(CO2Me)C6H4CH2CMe2NH2-2}Cl]·1/2CHCl3 (7d·1/2CHCl3)
In a
Carius tube, methyl phenylpropiolate
(48 μL, 0.324 mmol) was added to a solution of palladacycle e (120 mg, 0.156 mmol) in CHCl3 (10 mL), and the
mixture was heated at 65 °C for 2 h. The yellow solution was
concentrated to ca. 1 mL, and Et2O was added (20 mL). The
suspension was filtered, and the solid was washed with Et2O (2 × 5 mL) and air-dried to give complex 7d·1/2CHCl3 as a pale yellow solid. Yield: 37 mg, 0.061 mmol, 20%. The
filtrate was contentrated to ca. 2 mL, and n-pentane
(20 mL) was added. The suspension was filtered, and the yellow solid
was air-dried to give a ca. 1/0.7 mixture of complex 7d and palladacycle d (80 mg).In a
Carius tube, 2-norbornene (50 mg, 0.531
mmol) was added to a solution of palladacycle d (200
mg, 0.223 mmol) in CHCl3 (15 mL), and the mixture was heated
at 65 °C for 4 h. The resulting solution was filtered through
a plug of Celite, the filtrate was concentrated to ca. 1 mL, and Et2O (20 mL) was added. The suspension was filtered, and the
solid was washed with Et2O (2 × 5 mL) and air-dried
to give complex 7d·1/2CHCl3 as a pale
yellow solid. Yield: 198 mg, 0.328 mmol, 73%. Dec pt: 204 °C.
Anal. Calcd for C27H32ClNO2Pd·1/2CHCl3 (604.117): C, 54.68; H, 5.42; N, 2.32. Found: C, 54.64; H,
5.42; N, 2.29. IR (cm–1): ν(NH) 3320 w, 3266
w; ν(CO) 1716 s. 1H NMR (400.9 MHz): δ 0.83
(br d, 1 H, NH2, 2JHH = 9.6 Hz), 0.95 (br d, 1 H, CH2, nor, 2JHH = 10.0 Hz), 1.05 (br d, 1 H, CH2, nor, 2JHH = 10.4 Hz), 1.13
(s, 3 H, Me, CMe2), 1.19–1.23 (m, 2 H, Hα
+ 1 H of CH2 of nor), 1.35–1.45 (m, 2 H, CH2, nor), 1.64 (s, 3 H, Me, CMe2), 1.93 (d, 1 H,
CH2, nor, 2JHH =
9.6 Hz), 2.15 (br s, 1 H, CH, nor), 2.37 (br d, 1 H, NH2, 2JHH = 9.6 Hz), 2.45 (d,
1 H, CH2Ar, 2JHH = 13.2 Hz), 3.18 (br s, 1 H, CH, nor), 3.61 (d, 1 H, Hβ, 3JHH = 8.0 Hz), 3.76 (s, 3 H, MeO),
4.35 (d, 1 H, CH2Ar, 2JHH = 12.8 Hz), 6.83 (m, 1 H, H5), 6.88 (br d, 1 H, H6, 3JHH = 7.2 Hz), 7.01 (br d, 1 H,
H3, 3JHH = 7.2 Hz), 7.08 (td,
1 H, H4, 3JHH = 7.6, 4JHH = 1.2 Hz), 7.28 (br m, partially
obscured by the resonance of CHCl3, 3 H, p-H + m-H, Ph), 7.48 (br s, 2 H, o-H, Ph). 13C{1H} NMR (100.8 MHz): δ 23.8
(CHα), 28.3 (CH2, nor), 28.4 (CH2, nor),
30.7 (Me, CMe2), 30.75 (Me, CMe2), 37.0 (CH2, nor), 39.8 (CH, nor), 41.2 (CH, nor), 47.4 (CH2Ar), 52.2 (CMe2), 53.3 (MeO), 54.5 (CHβ),
101.5 (C7), 126.6 (CH5), 127.5 (p-CH, Ph), 127.7
(br s, m-CH, Ph), 127.9 (CH4), 130.5 (o-CH, Ph), 132.5 (CH3), 133.2 (CH6), 135.1 (C1), 136.2 (i-C, Ph), 137.6 (C2), 169.1 (CO). The resonance corresponding to C8
was not observed.
Synthesis of [Pd{C,N-CH(C5H8)CHC(CO2Me)=C(CO2Me)C6H4CH2CMe2NH2-2}Cl] (7f)
In a Carius tube, dimethyl
acetylenedicarboxylate
(82 μL, 0.667 mmol) was added to a suspension of palladacycle e (240 mg, 0.312 mmol) in CHCl3 (15 mL), and the
mixture was heated at 65 °C for 8 h. The resulting solution was
filtered through a plug of Celite, the filtrate was concentrated to
ca. 1 mL, and Et2O (20 mL) was added. The suspension was
filtered, and the solid was washed with Et2O (2 ×
5 mL) and air-dried to give the complex 7f as a bright
yellow solid. Yield: 247 mg, 0.469 mmol, 75%.In a
Carius tube, a solution of dimethyl acetylenedicarboxylate
(65 μL, 0.529 mmol) and 2-norbornene (50 mg, 0.531 mmol) in
CHCl3 (5 mL) was added to a solution of palladacycle B (150 mg, 0.312 mmol) in CHCl3 (15 mL), and the
mixture was heated at 65 °C for 8 h. The resulting solution was
filtered through a plug of Celite, the filtrate was concentrated to
ca. 2 mL, and Et2O (20 mL) was added. The suspension was
filtered, and the solid was washed with Et2O (2 ×
5 mL) and air-dried to give complex 7f as a bright yellow
solid. Yield: 90 mg, 0.171 mmol, 33%. Dec pt: 188 °C. Anal. Calcd
for C23H30ClNO4Pd (526.368): C, 52.48;
H, 5.74; N, 2.66. Found: C, 52.21; H, 6.11; N, 2.67. IR (cm–1): ν(NH) 3326 m, 3267 m; ν(CO) 1731 s, 1716 s. 1H NMR (400.9 MHz): δ 1.08 (br d, partially obscured by the
resonance of Me, 1 H, Hα, 3JHH = 8.0 Hz), 1.10 (s, 3 H, Me, CMe2), 1.11 (m,
partially obscured by the resonance of Me, 1 H, CH2, nor),
1.21 (m, 1 H, CH2, nor), 1.33 (br d, 1 H, CH2, nor, 2JHH = 10.4 Hz), 1.44
(m, 1 H, CH2, nor), 1.51 (s, 3 H, Me, CMe2),
1.58 (m, 1 H, CH2, nor), 1.62 (br d, 1 H, NH2, 2JHH = 10.4 Hz), 2.07 (br
d, 1 H, NH2, 2JHH = 10.4 Hz), 2.26 (br d, 1 H, CH2, nor, 2JHH = 10.4 Hz), 2.48 (dd, 1 H, CH2Ar, 2JHH = 13.2, 4JHH = 1.2 Hz), 2.70 (br d, 1 H, CH, nor, 3JHH = 3.6 Hz), 3.21 (br d, 1 H,
CH, nor, 3JHH = 4.0 Hz), 3.60
(d, partially obscured by the resonance of MeO, 1 H, Hβ, 3JHH = 8.4 Hz), 3.61 (s, 3 H, MeO),
3.75 (s, 3 H, MeO), 4.12 (d, 1 H, CH2Ar, 2JHH = 13.2 Hz), 7.13 (d, 1 H, H3, 3JHH = 7.6 Hz), 7.20 (m, 2 H, H6 + H5),
7.28–7.33 (m, 1 H, H4). 13C{1H} NMR (100.8
MHz): δ 25.6 (CHα), 27.6 (CH2, nor), 28.3 (CH2, nor), 30.4 (Me, CMe2), 30.6 (Me, CMe2), 38.1 (CH2, nor), 39.2 (CH, nor), 42.4 (CH, nor), 47.7
(CH2Ar), 52.2 (CMe2), 52.5
(MeO), 53.3 (MeO), 53.4 (CHβ), 68.3 (C8), 102.9 (C7), 127.4
(CH5), 128.9 (CH4), 130.4 (CH6), 133.1 (CH3), 135.2 (C1), 137.2 (C2),
166.9 (CO), 167.7 (CO).
Synthesis of 8f
In a Carius tube, KOtBu (50 mg, 0.445 mmol)
was added to a solution of complex 7f (120 mg, 0.228
mmol) in dry CH3CN, under an
N2 atmosphere (10 mL), and the mixture was heated at 78
°C for 3 days. Decomposition to metallic palladium was observed.
The solvent was removed, n-pentane (20 mL) was added,
and the mixture was filtered through a plug of Celite. The solvent
was removed from the filtrate, the residue was dissolved in Et2O (10 mL), and HOTf (0.01 mL, 0.113 mmol) was added. The resulting
suspension was filtered, and the solid was washed with Et2O (2 × 2 mL) and air-dried to give compound 8f as
a colorless solid. Yield: 23 mg, 0.043 mmol, 19%. Mp: 197 °C.
ΛM (Ω–1 mol–1 cm2): 104.0 (c = 3.82 × 10–4 M). Anal. Calcd for C24H30F3NO7S (533.566): C, 54.02; H, 5.67; N, 2.62, S;
6.00. Found: C, 53.88; H, 6.09; N, 2.69; S: 5.97. The hydrogen content
found in the elemental analysis was slightly outside the accepted
range (6.09 vs 5.67%; Δ = 0.42). This could be attributed to
the presence of small traces of Et2O in the sample (see
the Supporting Information), which remained
in spite of the compound being dried under vacuum. ESI-HRMS (m/z): exact mass calcd for C23H30NO4, 384.2169 [(M – CF3SO3)+]; found, 384.2174. IR (cm–1): ν(CO) 1740 (s), 1683 (s). 1H NMR (300.1 MHz):
δ 0.57 (br d, 1 H, CH2, nor, 2JHH = 12.6 Hz), 0.84 (d, 1 H, CH2,
nor, 2JHH = 12.3 Hz), 1.28
(s, 3 H, Me, CMe2), 1.29–1.55 (m, 4 H, CH2, nor), 1.56 (d, partially obscured by the resonance of CH2 of nor, 1 H, Hα or Hβ, 3JHH = 7.2 Hz), 1.83 (s, 3 H, Me, CMe2), 2.01
(d, 1 H, Hα or Hβ, 3JHH = 7.5 Hz), 2.62 (br s, 1 H, CH, nor), 2.75 (d, partially
obscured by the resonance of CH of nor, 1 H, CH2Ar, 2JHH = 16.8 Hz), 2.76 (br s, 1
H, CH, nor), 3.02 (d, 1 H, CH2Ar, 2JHH = 17.1 Hz), 3.10 (s, 3 H, MeO), 3.90 (s, 3 H, MeO),
7.07 (m, 1 H, H5), 7.37 (m, 2 H, H7 + H6), 7.85 (m, 1 H, H8), 7.99
(br d, 1 H, NH2, 2JHH = 11.4 Hz), 9.62 (br d, 1 H, NH2, 2JHH = 11.4 Hz). 13C{1H}
NMR (75.5 MHz): δ 22.4 (Me, CMe2), 27.3 (CHα
or CHβ), 28.5 (Me, CMe2), 28.6 (CH2, nor),
29.0 (CH2, nor), 29.6 (CHα or CHβ), 30.1 (CH2, nor), 35.5 (CH, nor), 35.7 (C(CO2Me)CH), 37.6 (CH, nor), 40.0 (CH2Ar), 52.6 (MeO), 54.6
(MeO), 57.8 (CMe2), 69.7 (C1), 120.4 (q,
CF3SO3, 1JCF = 319.5 Hz), 126.2 (C8a), 126.7 (CH7), 128.5 (CH5), 129.7 (CH6),
130.9 (C4a), 131.1 (CH8), 165.3 (CO), 174.1 (CO).
Synthesis
of 9f
TlOTf (84 mg, 0.237 mmol)
was added to a solution of complex 7f (125 mg, 0.237
mmol) in acetone (15 mL), and the mixture was stirred at room temperature
for 12 h under an CO atmosphere, using a toy balloon. Decomposition
to metallic palladium was observed. The suspension was filtered through
a plug of Celite, the filtrate was concentrated to ca. 1 mL, and Et2O (20 mL) was added. The suspension was filtered, and the
solid was washed with Et2O (2 × 2 mL) and air-dried
to give the compound 9f as a colorless solid. Yield:
103 mg, 0.177 mmol, 75%. Mp: 223 °C. ΛM (Ω–1 mol–1 cm2): 101.4 (c = 5.04 × 10–4 M). Anal. Calcd for
C25H32F3NO9S (579.592):
C, 51.80; H, 5.56; N, 2.41, S; 5.53. Found: C, 51.56; H, 5.87; N,
2.50; S: 5.37. ESI-HRMS (m/z): exact
mass calcd for C24H32NO6, 430.2230
[(M – CF3SO3)+]; found, 430.2225.
IR (cm–1): ν(CO) 1771 vs, 1733 vs, 1626 s. 1H NMR (400.9 MHz, acetone-d6):
δ 1.10 (m, 1 H, CH2, nor), 1.21 (m, 1 H, CH2, nor), 1.31 (m, 2 H, CH2, nor), 1.48 (s, 3 H, Me, CMe2), 1.54 (m, 2 H, CH2, nor), 1.58 (s, 3 H, Me, CMe2), 2.28 (br s, 1 H, CH, nor), 2.50 (br s, 1 H, CH, nor), 2.73
(d, 1 H, Hα, 3JHH = 8.0
Hz), 2.88 (d, 1 H, Hβ, 3JHH = 8.0 Hz), 3.10 (d, 1 H, CH2Ar, 2JHH = 14.4 Hz), 3.51 (s, 3 H, MeO), 3.66 (s, 3 H, MeO),
3.69 (d, 1 H, CH2Ar, 2JHH = 14.4 Hz), 4.99 (s, 1 H, H7), 7.26–7.33 (m, 2 H,
H4 + H5), 7.37 (m, 1 H, H3), 7.63 (m, 1 H, H6), 7.70 (br s, 3 H, NH3). The resonance corresponding to the OH group was not observed. 13C{1H} NMR (100.8 MHz, acetone-d6): δ 25.3 (Me, CMe2), 26.9 (Me, CMe2), 27.6 (CH2, nor), 28.5 (CH2, nor),
35.1 (CH2, nor), 40.9 (CH, nor), 41.1 (CH, nor), 41.7 (CH2Ar), 51.1 (CHα), 51.3 (CHβ), 52.2 (MeO), 53.7
(MeO), 53.0 (CHAr), 57.4 (CMe2), 90.2
(C8), 128.2 (CH5), 128.7 (CH4), 131.4 (CH6), 133.1 (CH3), 134.5 (C1),
135.0 (C2), 166.3 (CO2Me), 171.2 (CO2Me), 177.6 (CO2H).
Synthesis
of 10c
In a Carius tube, TlOTf
(72 mg, 0.204 mmol) was added to a suspension of the complex 7c·CHCl3 (100 mg, 0.161 mmol) in acetone (15
mL) and the mixture was stirred for 15 min. CO was bubbled through
the suspension for 3 min, the pressure of CO was increased to 1 atm,
the tube was sealed, and the mixture was stirred at room temperature
for 20 h. Decomposition to metallic palladium was observed. The suspension
was filtered through a plug of Celite, the filtrate was concentrated
to ca. 1 mL, and Et2O (20 mL) was added. The suspension
was filtered, the solvent was removed from the filtrate, and the solid
residue was stirred in n-pentane (20 mL). The resulting
suspension was filtered, and the solid was washed with n-pentane (2 × 5 mL) and air-dried to give the compound 10c as a colorless solid. Yield: 84 mg, 0.152 mmol, 94%. Mp:
160 °C. ΛM (Ω–1 mol–1 cm2): 117.1 (c = 5.00
× 10–4 M). Compound 10c was hygroscopic,
and no satisfactory elemental analysis could be obtained. ESI-HRMS
(m/z): exact mass calcd for C27H34NO2, 404.2590 [(M – CF3SO3)+]; found, 404.2589. IR (cm–1): ν(NH) 3452 br; ν(CO) 1704 s. 1H NMR (400.9 MHz): δ 0.98 (br d, 1 H, CH2, nor, 2JHH = 10.4 Hz), 1.07 (s, 3 H,
Me, CMe2), 1.29 (m, 2 H, CH2, nor), 1.42 (m,
partially obscured by the resonance of Me group, 1 H, CH2, nor), 1.45 (s, 3 H, Me, CMe2), 1.60 (m, 1 H, CH2, nor), 1.81 (br s, 1 H, CH, nor), 1.83 (br d, partially obscured
by the resonance of CH of nor, 1 H, CH2, nor), 2.19 (s,
3 H, MeC=), 2.22 (d, 1 H, CH2Ar, 2JHH = 13.6 Hz), 2.60 (br d, 1 H, CH, nor, 3JHH = 4.0 Hz), 3.08 (d, 1 H, Hα, 3JHH = 10.0 Hz), 3.22 (d, 1 H,
Hβ, 3JHH = 9.6 Hz), 3.52
(d, 1 H, CH2Ar, 2JHH = 13.2 Hz), 6.58 (br s, 1 H), 6.77 (d, 1 H, H3, 3JHH = 7.6 Hz), 6.90–6.99 (m, 4 H), 7.06–7.12
(m, 6 H), 7.33 (br s, 1 H, CO2H). 13C{1H} NMR (100.8 MHz): δ 22.9 (MeC=),
24.1 (Me, CMe2), 26.0 (Me, CMe2), 29.0 (CH2, nor), 30.9 (CH2, nor), 36.8 (CH2,
nor), 39.7 (CH, nor), 40.3 (CH, nor), 41.5 (CH2Ar), 50.8
(CHβ), 54.1 (CHα), 57.3 (CMe2), 125.9 (CH4 or p-CH of Ph), 126.0 (CH4 or p-CH of Ph), 126.6 (CH5), 126.9 (br s, o-CH + m-CH, Ph), 129.1 (CH6), 129.4 (CH3), 131.3
(C2), 134.4 (C7), 140.1 (i-C, Ph), 140.3 (C8), 146.0
(C1), 177.6 (CO).
Synthesis of 10d and 9d
In
a Carius tube, TlOTf (50 mg, 0.141 mmol) was added to a solution of
complex 7d·1/2CHCl3 (75 mg, 0.124 mmol)
in acetone (15 mL), and the mixture was stirred for 2 h. CO was bubbled
through the suspension for 2 min, the pressure of CO was increased
to 1 atm, the tube was sealed, and the mixture was stirred at room
temperature for 15 h. Decomposition to metallic palladium was observed.
The suspension was filtered through a plug of Celite, the solvent
was removed from the filtrate, and Et2O (20 mL) was added.
The mixture was filtered through a plug of Celite, the filtrate was
concentrated to ca. 1 mL, and n-pentane (20 mL) was
added. The resulting suspension was filtered, and the solid was washed
with n-pentane (2 × 5 mL) and air-dried to give
the compound 10d as a colorless solid. Yield: 60 mg,
0.100 mmol, 81%. Mp: 122 °C. ΛM (Ω–1 mol–1 cm2): 78.3 (c = 7.4 × 10–4 M). Compound 10d was hygroscopic, and no satisfactory elemental analysis
could be obtained. ESI-HRMS (m/z): exact mass calcd for C28H34NO4, 448.2482 [(M – CF3SO3)+]; found: 448.2484. IR (cm–1): ν(NH) 3486
br; ν(CO) 1715 vs, 1626 s. 1H NMR (300.1 MHz): δ
1.02 (br d, 1 H, CH2, nor, 2JHH = 9.9 Hz), 1.1 (s, 3 H, Me, CMe2), 1.20–1.32
(m, 2 H, CH2, nor), 1.36–1.47 (m, 2 H, CH2, nor), 1.51 (s, 3 H, Me, CMe2), 1.79 (br d, 1 H, CH,
nor, 3JHH = 2.7 Hz), 1.83 (br
d, 1 H, CH2, nor, 2JHH = 9.9 Hz), 2.19 (d, 1 H, CH2Ar, 2JHH = 14.1 Hz), 2.44 (br d, 1 H, CH, nor, 3JHH = 3.6 Hz), 3.21 (d, 1 H, Hα, 3JHH = 9.3 Hz), 3.45 (d, 1 H, CH2Ar, 2JHH = 13.8 Hz), 3.57
(d, 1 H, Hβ, 3JHH = 8.4
Hz), 3.79 (s, 3 H, OMe), 6.59 (br s, 1 H), 6.80 (d, 1 H, H3, 3JHH = 7.8 Hz), 6.90–7.09
(m, 6 H), 7.11–7.20 (m, 3 H), 7.40 (br s, 1 H, CO2H). 13C{1H} NMR (75.5 MHz): δ 23.9 (Me,
CMe2), 26.8 (Me, CMe2), 29.5 (CH2, nor), 30.3 (CH2, nor), 36.6 (CH2, nor), 39.7
(CH, nor), 41.3 (CH, nor), 42.1 (CH2Ar), 50.9 (CHβ),
52.6 (OMe), 55.2 (CHα), 56.8 (CMe2), 118.3 (C), 121.4 (C), 126.7 (CH), 126.9 (CH5), 128.2 (CH), 130.2
(C7 or C8), 131.1 (CH), 131.4 (CH3), 132.1 (C1 or C2), 138.3 (C1 or
C2), 138.6 (C), 156.3 (C7 or C8), 169.5 (CO2Me), 177.7 (CO2H).In a Carius tube, a solution
of amino acid 10d (45 mg, 0.075 mmol) in CHCl3 (2 mL) was heated at 65 °C for 7 days. The solvent was removed
from the resulting white suspension to ca. 0.5 mL, and Et2O (10 mL) was added. The suspension was filtered, and the solid was
washed with Et2O (2 × 3 mL) and air-dried to afford
the cyclic lactone 9d as a colorless solid. Yield: 30
mg, 0.048 mmol, 64%. Mp: 251 °C. ΛM (Ω–1 mol–1 cm2): 88.8 (c = 4.1 × 10–4 M). Anal. Calcd for
C29H34F3NO7S (597.643):
C, 58.28; H, 5.73; N, 2.34, S; 5.37. Found: C, 58.15; H, 5.56; N,
2.12; S: 5.42. ESI-HRMS (m/z): exact
mass calcd for C28H34NO4, 448.2482
[(M – CF3SO3)+]; found, 448.2493.
IR (cm–1): ν(NH) 3167 m; ν(CO) 1749
s, 1733 s. 1H NMR (400.9 MHz, DMSO-d6): δ 0.69 (m, 2 H, CH2, nor), 0.91–0.94
(m, 2 H, 1 H of CH2 + 1 H of CH2, nor), 1.19
(s, 4 H, Me of CMe2 + CH of nor), 1.23 (s, 3 H, Me, CMe2), 1.26–1.28 (m, 2 H, 1 H of CH2 + 1 H of
CH2, nor), 1.69 (s, 1 H, CH, nor), 2.17 (s, 1 H, CH, nor),
2.19 (s, 1 H, CH, nor), 2.88 (d, 1 H, CH2Ar, 2JHH = 14.0 Hz), 3.21 (s, 3 H, MeO), 3.41
(d, 1 H, CH2Ar, 2JHH = 13.6 Hz), 4.86 (s, 1 H, H7), 7.28 (br t, 1 H, H5, 3JHH = 7.2 Hz), 7.32–7.50 (m, 6
H, H3 + H4 + CH of Ph), 7.51–7.60 (br s, 1 H, CH, Ph), 7.63
(d, 1 H, H6, 3JHH = 8.0 H),
7.87 (br s, 3 H, NH3). The resonance corresponding to the
OH group was not observed. 13C{1H} NMR (100.8
MHz, DMSO-d6): δ 24.8 (Me, CMe2), 24.9 (Me, CMe2), 27.4 (CH2, nor),
27.8 (CH2, nor), 33.5 (CH2, nor), 39.7 (CH,
nor), 39.8 (CH, nor), 41.0 (CH2Ar), 49.5 (CH, nor), 50.4
(CH, nor), 51.5 (MeO), 54.7 (CMe2), 55.6
(CH7), 89.7 (C8), 127.2 (CH5), 127.6 (CH, Ph), 127.8 (br s, CH, Ph),
128.4 (CH4), 131.1 (CH6), 131.5 (C1), 132.2 (CH3), 135.1 (C2), 138.4
(i-C, Ph), 169.4 (CO2Me), 177.4 (CO2H).
Single-Crystal X-ray Structure
Determinations
Single
crystals were obtained as follows: complex anti-1a, slow diffusion of n-pentane into a solution
of a 2.5/1 mixture of anti/syn-1a in CH2Cl2; complex anti-1b, slow diffusion of Et2O into a solution
of a 3/1 mixture of anti-/syn-1b in CHCl3; complex syn-1b, slow diffusion of n-pentane into a solution
of a 1/3 mixture of anti-/syn-1b in CHCl3; complex anti-2a·CH2Cl2, slow diffusion of n-pentane into a solution of a 5/1 mixture of anti-/syn-2a in CH2Cl2; complex syn-2a, slow diffusion of n-pentane into a solution of the complex in CHCl3, complex anti-2b·CHCl3, slow diffusion of n-pentane into a solution of
a 3/1 mixture of anti-/syn-2b in CHCl3; complex syn-2b: slow diffusion of n-pentane into a solution
of a 1/3 mixture of anti-/syn-2b in CHCl3; complex syn-3b·CH2Cl2: ,slow diffusion of n-pentane into a solution of anti-/syn-3b in CH2Cl2; compound 5b·H2O, slow diffusion of n-pentane into a solution of the complex in CH2Cl2, compounds 7c·CHCl3, 7d·1/2CHCl3, 7f, and 8f,
slow diffusion of n-pentane into a solution of the
corresponding compound in CHCl3; compound 9d, slow diffusion of Et2O into a solution of the compound
in acetone; compound 9f·Et2O, slow diffusion
of Et2O into a solution of 9f in acetone;
compound 10c·H2O, slow diffusion of n-pentane into a solution of 10c in CHCl3.
Data Collection
Crystals suitable for X-ray diffraction
were mounted in inert oil on a glass fiber and transferred to a Bruker
diffractometer. Data were recorded at 100(2) K, using graphite-monochromated
Mo Kα radiation (λ = 0.71073 Å) and the ω-scan
mode. Multiscan absorption corrections were applied for all complexes.
Structure
Solution and Refinements
Crystal structures
were solved by Patterson (syn-1b, anti-2b·CHCl3, and 7c·CHCl3) or direct methods (anti-1a, anti-1b, anti-2a·CH2Cl2, syn-2a, syn-2b, syn-3b·CH2Cl2, 6b·H2O, 7d·1/2CHCl3, 7f, 8f, 9d, 9f·Et2O, and 10c·H2O), and all non-hydrogen
atoms were refined anisotropically on F2 using the program SHELXL-2018/3.[42] Hydrogen
atoms were refined as follows: complex anti-1a, NH2, free with SADI; ordered methyl, rigid
group; all others, riding; compounds anti-1b, syn-1b, anti-2a·CH2Cl2, syn-2a, syn-2b, 7c·CHCl3, 7d·1/2CHCl3, 7f, 8f, and 9d, NH2 or
NH3, free with SADI; methyl, rigid group; all others, riding;
complex anti-2b·CHCl3, NH2, free with DFIX; methyl, rigid group; all others,
riding; compounds syn-3b·CH2Cl2 and 9f·Et2O, NH2 or NH3, free; methyl, rigid group; all others,
riding; compound 6b·H2O, NH3 and H2O, free with SADI; methyl, rigid group; all others,
riding; compound 10c·H2O, NH3, free with SADI; H2O (crystallization water), free with
DFIX; methyl, rigid group; all others, riding.
Special Features
For anti-1a, one methyl group was disordered
over two positions, with a ca.
95/5 occupancy distribution. For syn-1b, one phenyl ring was disordered over two positions, with a ca. 57/43
occupancy distribution. For syn-2a,
the CO2CH2CH3 fragment was disordered
over two positions, with a ca. 75/25 occupancy distribution. The structure
was refined as an inversion twin. Absolute structure (Flack) parameter:[43] 0.027(8). For anti-2b·CHCl3, the chloroform molecule was disordered over
two positions, with a ca. 81/19 occupancy distribution. For 7d·1/2CHCl3, the crystallization solvent (half
a molecule of chloroform), located on a crystallographic inversion
center, was disordered over two positions, with a ca. 62/38 occupancy
distribution.Relevant crystallographic data, details of the
refinements, and details (including symmetry operators) of hydrogen
bonds for the compounds anti-1a, anti-1b, syn-1b, anti-2a·CH2Cl2, syn-2a, anti-2b·CHCl3, syn-2b, syn-3b·CH2Cl2, 6b·H2O, 7c·CHCl3, 7d·1/2CHCl3, 7f, 8f, 9d, 9f·Et2O, and 10c·H2O are given in the Supporting Information.