intramolecular diels-alder reactions of alkenylboranes a ... · in the first project,...
TRANSCRIPT
Intramolecular Diels-Alder Reactions of Alkenylboranes & A Hetero Diels-Alder Approach to
the Total Synthesis of Martinelline
Demy Lin
A thesis subrnitted in confomiity with the requirements
for the degree of Master's of Science
Graduate Department of Chemistry
University of Toronto
Toronto, Ontario, Canada
O Copyright Denny Lin, April, 1997.
395 Wellington Street 395, rue Wellington Ottawa ON K I A ON4 Ottawa ON K I A ON4 Canada Canada
Vour fi& Votre reférence
Our file Notre réfdrenco
The author has granted a non- L'auteur a accordé une licence non exclusive licence allowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or sel1 reproduire, prêter, distribuer ou copies of this thesis in microform, vendre des copies de cette thèse sous paper or electronic formats. la forme de microfiche/fih, de
reproduction sur papier ou sur format électronique.
The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation.
This thesis is a summary of research conducted since September 1995. It is
comprised of two distinct projects and contains three chapters: Introduction, Results and
Discussion, and Experimental.
In the first project, intramolecular Diels-Alder reactions of alkenylboranes were
investigated. A one-pot three sequence transformation which involved: (i) hydroboration of
a dienyne with dicyclohexylborane, (ii) thermal intramolecular Diels-Alder reaction in
benzene, containing 5 mol % of butylated hydroxytoluene, and (iii) subsequent oxidation by
Me3NO*H20 to an alcohol, was optimized. Al1 three substrates that were tested successfully
underwent the cycloaddition in moderate to good yield. In each case, a single diastereomer
corresponding to endo-addition, was forrned.
The second project was concerned with a hetero Diels-Alder approach to the total
synthesis of Martinelline. A three component coupling strategy was devised to construct the
pyrroloquinoline core of this natural product in a single step. In the presence of catalytic
dysprosium (III) triflate, N-acylated 2-pyrrolines, acting as dienophiles, underwent
cycloaddition with 2-azadienes fonned in situ frorn the condensation of substituted anilines
and benzaldehyde. Excellent yields were obtained, but poor diastereoselectivity was
observed. Deprotection of these adducts is also reported. In the absence of benzaldehyde,
the aniline and two equivalents of a 2-pyrroline under these conditions formed a cycloadduct
as a single diastereomer in good to excellent yield. NMR data suggested that the
diastereomer was the product of endo-addition. A cornparison of these data with those
reported in the first isolation of Martinelline, indicates that the relative stereochemistry of the
adduct is not the sarne.
To Mom, Dad and Ken,
Abstract ..................................................................................................................................................................................... ii List of Tables and Figures .................................................................................................................................... vi A bbreviations ....................................................................................................................................................................... vii
Acknowledgements ......................................... Aeknowledgements........................ ix
............................................................................................................................... Chapter 1 Introduction 10
1 . 1 Diels-Alder Reactions ............................................................................................................................... 11
1.1 . 1 General Aspects .................................................................................................................. 1
................... ......................................... 1.1.2 Intermolecular Diels-Alder Reactions ., 14
.................................................................. 1.1.2.1 Regio- and Stereoselectivity 14
............................................................ 1.1.2.2 Alkenylboranes as Dienophiles 15
........................................................................................ 1 . 1.2.2.1 Synthesis 15
..................... ................. 1 . 1.2.2.2 Regio- and Stereoselectivity .. 16 1.1.3 Intramolecular Diels-Alder Reactions ............................................................. 19
1.1.3.1 General Aspects ............................................................................................ 19
1.1.3.2 Regioselectivity ............................................................................................ 20
1.1.3.3 S tereoselectivity ........................................................................................... 1
.................... 1.1.3.3.1 Effect of Dienophile Activating Group 22 1.1.3.3.2 Steric Effects ............................................................................. 22
1.2 Martinelline and Related Compounds ........................................................................................... 24
1.2.1 Biological Properties .......................................................................................................... 24
1.2.2 Lewis Acid Catalysis of Diels-Alder Reactions ............................................. 25
1.2.2.1 Lanthanide (III) Salts as Lewis Acid Catalysts ...................... 26
........................................................ 1.2.3 Three Component Coupling Methodology 27
1.2.4 Enamines and 2-Azadienes in Diels-Alder Reactions .............................. 27
.............................................................................................. Chapter II Results & Discussion 34
............................................. II . 1 Intramolecular Diels-Alder Reactions of Alkenylboranes 35 ............................................................................................................................... . . II 1 1 Objectives 35
................................................................................................ . II 1.2 Syntheses of Substrates 35
.................................................................... II . 1.3 Alkync Hydroboration of Substrates 37
.................. ................................... II . 1.3.1 In Situ Formation of SHyTHF ... 39
..................................................................................... 11.1.4 Oxidation of IMDA Adducts 39
II . 1.5 Optimization of IMDA Reaction Conditions ............................................. 39
11.1.7 Stereochemistry of IMDA Adducts ........................................................................ 42
II . 1.8 Comparison of Stereochemistry with Other Relevant .................................................................................................................. IMDA Adducts
........................................................................................................................... . II 1.9 Conclusions 46 ............... II.2 A Hetero Diels-Alder Approach to the Total Synthesis of Martinelline 47
II.2.1 Objectives ................................................................................................................................ 47 11.2.2 Retrosynthetic Strategy .................................................................................................. 47
............................................ 11.2.3 Synthesis of N-Acylated 2, 3-Dihydropyrroies 49
.............................................................. I1.2.4 Three Component Coupling Reactions 50
11.2.4.1 Preliminary Work ....................................................................................... 50 ............................................. 11.2.4.2 Optimization of Reaction Conditions 5 1
11.2.5 Survey of Lanthanide (III) Triflates as Lewis Acid Catalysts ............ 52
11.2.6 Effect of Pyrroline Protecting Group ................................................................... 53
11.2.7 Non-Aromatic Aldehydes as Imine Precursors ............................................ 56
11.2.8 N-Acyl Dihydropyrroles as Irnine Precursors .............................................. 57
11.2.8.1 Optirnized Conditions ............................................................................... 58
11.2.8.2 Proposed Transition State Mode1 .................... .. ....................... 59
11.2.9 Deprotection of Cycloadducts .................................................................................. 60 11.2.9.1 Removal of Alloc Protected Adducts ........................................... 60
11.2.9.2 Removal of Cbz Protected Adducts .............................................. 61
.................................. 11.2.10 Cornparison of Stereochernistry with Martinelline 62 ......................................................................................................................... 11.2.1 1 Conclusions 65
....................................................................................................................... Chapter III Experimental 66 ............................................................................................................................ III . 1 General Experimental 67
111.2 Synthetic Preparations ......................................................................................................................... 68
Appendix A: Submitted Articles .................................................................................................................. 84 Appendix B: X-Ray Crystal Structure Data .................................................................................... 97
...................................................................................................... Appendix C: Selected Spectral Data 110
References ............................................................................................................................................................................... 138
Figure 1.1.1.1:
Figure 1.1.1 -2:
Figure 1.1.1.3:
Figure 1.1.2.1.1:
Figure 1.1.2.2.1.1:
Figure 1.1.2.2.2.1 :
Figure 1.1.2.2.2.2:
Figure 1.1.3.1.1:
Figure 1.1.3.3.1:
Figure 1.2.1:
Table II . 1 .5 . 1 :
Figure II . 1.7.1 :
Figure II . 1.7.2:
Figure II . 1.8.1:
Table 11.2.5.1:
Table 11.2.6.1 :
Figure 11.2.6.2:
Table 11.2.8.1.1:
Figure 11.2.8.2.1:
Figure 11.2.10.1 :
........ [4nS + 2ns] and [4na + 2nd Cycloaddition Transition States 1 1
........................................................ An IMDA approach to Endiandric Acids 12
An approach to taxol(3). utilizing inter- and intramolecular
.................................................................................................... Diels- Alder reactions 13
Endo- and exo- transition states of intermolecular
..................................................................................................... Diels-Alder reactions 14
......................................................................................... Routes to alkenylboranes 16
[4 + 21 Cycloadditions of vinyl-9-BBN with
.................................................................................................... representative dienes 17
Ab initio calculated transition states (6-3 lG*) of an intermolecular Diels-Alder reaction between an
alkenylborane and butadiene (hydrogen atoms
............................................................................... omitted, bond distances in A) 18
Types of tethers in lMDA reactions ................................................................... 19
Endo- and exo- transition states of bicyclo[4.4.0]
and [4.3.0] systems ...................................................................................................... 21
Martinelline (22) and Martinellinic Acid (23) ........................................... 24
Yields of IMDA adducts ............................. .... ............................................................ 40
NOE enhancements of IMDA cycloadducts .............................................. 42 ORTEP drawing of derivatized IMDA cycloadduct (64) ................... 43
Some comparative data of hydrindene cycloadducts ............................. 44
Comparison of Lewis acid; diastereomeric ratio and yields
of cycloadducts ................................................................................................................ 53
Effect of substituents on yield and diastereomeric ratio ...................... 54
......................................................................................... ORTEP drawing of (82a) 55 Yields of non-aromatic aldehyde derived cycloadducts ...................... 59
Proposed transition states for the formation of endo- and
exo-products of (84) and (87) ............................................................................. 60
COSY spectra of (90a) and (90b) ................................................................... 64
Ac Alloc
BBN
BHT Boc
Bz
"Bu CB Cbz
Chx
COSY A
d
D
DMAP DMF E
EI
Et
ether
EtOAc
FM0
h hfc
HMPA HOMO
IMDA i ~ r
IR
LUMO
m
M
Me
MeCN
acetyl
al1 y loxycarbony 1
9-borabicyclo[3.3.1 }nonane
butylated hydroxytoluene
tert-butoxycarbonyl
benzy 1
buty l catec holborane
carbobenzyloxy
c yclohexyl
correlated spectroscopy
heat
doublet
dimensional
dimeth yhninopyridine
dimethy lformamide
unspecified ester
electron impact
ethyl
diethyl ether
ethyl acetate
frontier molecular orbit al hours
3-(heptafluoropropylhydroxymethylene)-(+)-cmphorato
hexarnethylphosp horic triamide
highest occupied rnolecular orbital
intramolecular Diels-Alder reaction
isopropy 1
infra-red
lowest unoccupied molecular orbital
complex multiplet
molecular ion
methyl
acetonitrile
vii
min
Ms MS
NMR
NOE
P PBH Ph
psi
Ln
P l OTf
9 R
S
Sia
t
TBDMS
TFA THF
TLC TMS
X
minutes
melting point
methanesulphonyl
molecular sieves or mass spectrometry , depending on context
nuclear magnetic resonance
nuclear Overhauser effect
pentet
pinacolborane
phenyl
pounds per square inch
lanthanide
unspecified oxidation reaction
triflate; tnfluoromethansulphonyl
quartet
alkyl group
single t
siamy 1
triplet
tert-buty ldimethy lsily 1 trifluoroacetic acid
tetrahydrofuran
thin layer chromatography
trimethylsilyl
generic element
Prof. Rob Batey is well deserving of my first acknowledgement. He has taught me so much in so little time. His pleasant attitude and 'hip' style made the lab environment
extremely cornfortable. Deep and sincere thanks for tremendous guidance, supervision and
friendship, Rob.
The personalities of my lab mates have. Thanks to Bruce for good, 'excellent' and
'lucky' times and also for al1 those discussions about chernistry, life and other great stuff!
Special thnaks to Leah, the only female in the lab - 1 can't understand how you could've
lasted this long! And, to al1 my other lab mates (Bijan, Dave - hang in there, Zed (ie. Greasy
Zissi, Santha and Mustafa) - try to keep the al1 niters to a minimum!) good luck and thanks for everything. Appreciation goes out to the undergrads who've contributed to my thesis:
Rob, Christy and Andrew.
In these past 20 months, 1 have worked with interesting and entertaining people: the
Lautens crew - especially Tom and the late evening discussions; the JI3J group - Rich and
Mike for advice and lending me equipment; the Kluger group - Stew, Pete (looking pretty
huge these days!), Vittorio; and the McMillan people: Dom and Steve. Thanks, Prof. Andrew McMillan, for reading my thesis and correcting it so
prornptly. 1 would not have been able to complete this thesis without the assistance of the
technical staff: Nick Plavic, Dr. Alex Young and especially Drs. Tim Burrows and Alan
Lough. Tim was my persona1 NMR guy for just about 2 months - boy does he have
patience! Alan spent lots of time with my crystals, trying desperately to find one good enough.
Thanks to my familiy, Mom Dad and Ken, who have always pushed me to strive and
onIy settle for the best, for your love and caring. You have created the person 1 am today
and for that reason, 1 am dedicating my thesis to you.
And finally to my special star, Akiko, without whom I'd be insane or at least deeply
disturbed: my time with you has been incredible. My extreme gratitude for your support
and love.
Thanks and best wishes to all!
Chapter 1:
Introduction
1-1.1 General Aspects The Diels-Alder reaction is arguably the most popular method for ring formation
among synthetic organic chernists. In this pericyclic reaction, two reactive entities, the diene
and the dienophile, uni@ in a concerted mechanism to form a six membered ring. During the reaction, al1 four n-electrons of the diene interact with both sc-electrons of the dienophile,
producing two new o-bonds, in what is referred to as a [4 + 21 cycloaddition (Scheme
1.1.1.1). To enable it to react, the diene must be in the s-cisoid conformation.
Scheme I . l . l . l
In the Diels-Alder reaction, the two reactive entities, the diene and the dienophile,
must be complementary in electronic character. According to frontier molecular orbital
theory, the two molecular orbitals that will interact are the highest occupied molecular orbital
(HOMO) of one reactive entity and the lowest unoccupied molecular orbital (LUMO) of the
other.1 In a typical Diels-Alder reaction, the HOMO of the diene and the LUMO of the
dienophile interact. The reaction is highly dependent on the electronic nature of the HOMO
and LUMO: the lower the difference between the energies of these orbitals, the greater the
reactivity. To attenuate this gap in energy, electron rich substituents, which increase the
(H' LUMO
Lb LUMO
Figure 1.1.1.1: [4ns + 2 x s ] and [4xa + Zna] Cycloaddition transition states.
LUMO, are utilized. When the HOMO of the dienophile and the LUMO of the diene is the
dominant interaction, the reaction is termed an 'inverse electron demand' Diels-Alder reaction, such that electron rich dienophiles and electron poor dienes react best. In any case,
the HOMO and LUMO interact in a suprafacial manner, known as the 'cis-' rule (Figure 1.1.1. l), giving rise to products such as (la). Products of type (lb), derived directly from
antarafacial addition, are not observed in a Diels-Alder reaction.
In addition to the construction of one or more new rings, the regio- and
stereoselectivity of the Diels-Alder reaction can be controlled. For the syntheses of natural
products, particularly those possessing rings, these become obvious advantages. For
example, Nicolaou has synthesized endiandric acids (S), which are a class of secondary
metabolites of the plant Endiandru introrsa, using an intramolecular Diels-Alder (IMDA)
approach (Figure 1.1.1.2).*
Figure 1.1.1.2: An IMDA approach to Endiandric Acids.
More recently, in Nicolaou's total synthesis of taxol (3) both inter- and intramolecular [4 + 21 cycloadditions were employed to construct the two cyclohexane rings
(Figure 1.1.1.3).3 The functionality of the cyclohexene rings with the given stereochemistry
would be difficult to achieve through other means. The two units, (4) and (S), were further
rnanipulated and then linked together, forming the 8-membered macrocycle, B. Clearly, the Diels-Alder reaction has trernendous utility in organic chernistry.
However, it is the generation of reactive dienes and dienophiles that often represents the
greatest challenge for synthetic chernists. Discovering new reactive entities that undergo
these cycloadditions is therefore of utmost importance for the development of new synthetic
equivalents for the Diels-Alder reaction.
boronate cleavage
Figure 1.1.1.3: An approach to taxol (3), utitizing inter- and intramolecular Diels-Alder reactions.
1.1.2.1 Regio- and Stereoselectivity
- ' R2 "'DR' 'para-' 'meta-'
Scherne 1.1.2.1.1
Regioselectivity is an important consideration in the intermolecular Diels-Alder
reaction in which the diene and the dienophile are unsymmetrical.4 Generally, the
'ortho/para' rule, (which states that 'ortho' or 'para' substituted products on the cyclohexene
ring form in preference to 'meta' substituted products), predicts the major regioisomer of the
reaction (Scheme 1.1.2.1.1). When the diene or dienophile bears more than one substituent,
differences in directive effects of each substituent will dictate the regioselectivity of the
reaction. -
R2 - R1 endo-
L exo- -
51
("J ""R
6 endo-
51
O R
% exo-
Figure 1.1.2.1.1: Endo- and exo- transition states of intermolecular Diels-Alder reactions.
Alder rule, has been commonly used. This rule explains that, although stericaIly
disfavoured, endo-addition of the dienophile to the diene predominates over exo-addition,
the reason being enhanced secondary orbital overlap between the diene and dienophile in the
endo-case (Figure 1.1.2.1.1). However, the endo-rule applies more strictly to reactions
involving cyclic dienes (eg. cyclopentadiene) adding to cyclic dienophiles (eg. maleic
anhydride). Preference for endo-addition can be influenced by the presence of a Lewis acid.
1.1.2.2 Alkenylboranes as Dienophiles In 1963, Matteson reported the first exarnple of an alkenylboron species participating
in a Diels-Alder reaction. The dienophile, dibutyl ethyleneboronate, underwent
cycloaddition with cyclopentadiene and isoprene. Low endo/exo selectivity was observed in
the reaction with cyclopentadiene, but complete regioselectivity was reported with isoprene
(Scheme 1.1.2.2.1).5 Since this initial finding, alkenylboronic esters have also been shown
to undergo [2 + 21 and 1,3-dipolar [3 + 21 cycloaddition reactions.6~7
sole detected product
Scheme 1.1.2.2.1
Alkenylboron species are unique compounds due to their electronic properties and
their versàttility in conversion to other functional groups. As a consequence, the chernistry of
these compounds has been studied closely.8 It is their applicability, both in a reactive and
selective sense, towards the Diels-Alder [4 + 21 reaction that is of particular relevance in this
section.
1.1.2.2.1 Synthesis Several methods for the preparation of alkenylboranes have been reported, but some
of these generate a mixture of E- and 2-alkenylboranes. Two cornmon methods which avoid
this problem are transrnetallation and hydroboration. In transmetallation, a vinylic tin atom is
approach is capable of forming both the E- and 2- isomers (with respect to the boron
substituent), depending on the geometry of the precursor vinyl tin, while hydroboration of
alkynes can only generate the E-alkene. Various dialkyl alkenylboranes with Diels-Alder
reactivity, have been reported using both of these approaches (Figure 1.1.2.2.1.1). Due to
the instability of alkenylboranes, purification under standard means is impossible. It is
therefore, much more convenient to generate these reactive species in situ.
Transmetallation Hydroboration
Figure 1.1.2.2.1 .l: Routes to al kenylboranes.
Although transrnetallation is more versatile in ternis of the ability to forrn either E- or
Z-alkenylboranes, the toxicity of the requisite alkenyltin is a great disadvantage.
Hydroborating agents, are much less toxic, but issues of regio- and chemoselectivity of
alkyne hydroboration become a concern. Fortunately, there are severai hydroborating agents
which possess these qualities.8a
1.1.2.2.2 Regio- and Stereosetectivity In 1990, Singleton and Martinez reported for the first time, the high reactivity,
regioselectivity and endo-selectivity of vinyl-9-BBN in Diels-Alder reactions with various
1.3-dienes (Figure 1.1.2.2.2.1 ).9 They reasoned that the regio- and endo-selectivity
observed in the cycloadditions, was as expected on the basis of frontier molecular orbital
(FMO) analysis.
Diene lsolated Yield (%) Product(s) and Ratio
Figure 1.4.2.2.2.1 : 14 + 21 Cycloadditions of vinyl-9-BBN with representative dienes.
Following this initial finding, Singleton has deterrnined that alkenylboranes react
with electron rich, electron poor, and unactivated dienes, which has led to the designation
that alkenylboranes are omniphilic dienophiles.10 This is in contrast to simple resonance
analysis which predicts alkenylboranes should be electron poor alkenes and should react
preferentially with electron ric h dienes. Neither the reactivity nor regioselectivity of these
dienophiles can be explained by the poiarization in the resonance structure (6). F M 0
(i) vinyl-9-BBN
R 1 (ii) pl R - J"JOH +
OH
'para-' R = Ph, 'Bu, CI, Et0
Scheme 1.1.2.2.2.1
poor dienes (ie. R = Cl or E t0 in Scheme 1.1.2.2.2.1). For 2-substituted 1,3-butadienes,
the 'para-' regioisomer was formed in preference to the 'meta-' isorner. Only steric
congestion could explain these regiochemical results with vinyl-9-BBN. The reason for the ornniphilic nature of alkenylboranes is unclear. The belief that n-
electron withdrawal by boron activates alkenylboranes is probably not correct since the
electron rich, trivinylborane, which should deactivate the dienophile, was found to be more reactive.lOb An altemate explanation that a-electron withdrawal by the alkyl groups of the
alkenylboranes activates the dienophile, is also inconsistent with the observation that
vinylboronic esters have low reactivity.
Figure 1.1 2.2.2.2: Ab initio calculated transition states (6-31Gf) of an intermolecular Diels-Alder reaction between an alkenylborane and
butadiene (hydrogen atoms omitted, bond distances in A).
A [4 atom + 3 atom] transition state was proposed to rationalize the unusual reactivity
and selectivity of alkenylboranes (Figure 1.1.2.2.2.2).11 Through ab initio calculations, a
novel [4 + 31 endo-transition state in which the boron atom is within close proximity of Cl of the diene was determined; the bond length of C 1 -B is shorter than C 1 -Cg. In the exo-
variant, boron was found to play no role in the interaction with Cl and a typical [4 + 21
transition state was encountered. One important consequence of the [4 + 31 endo-transi tion
state is the accentuated effect of steric bulk on boron which explains the regiochemical
outcome in the reactions with vinyl-9-BBN.
1.1.3.1 General Aspects An IMDA reaction involves a substrate in which the diene is tethered to the
dienophile. In this way, there are two general classes of IMDA reactions, type 1 and type II, which differ only in the position of the linking chah (Figure 1.1.3.1.1).
TYPE 1:
Figure 1.1.3.1.1: Types of tethers in IMDA reactions.
In contrast to the interrnolecular reaction, two rings are foxmed simultaneously in the
intramolecular variant. For this reason and the ability to control the stereoc hemistry about
four centres, the IMDA reaction has been used in the synthesis of natural products, since the mid-1970s.12 For exarnple, (k)-pumiliotoxin C, (7), a physiologically active alkaloid found
in the skin of certain frogs, was synthesized with an IMDA reaction as the key step (Scheme
1.1.3.1.1).13
(i) 24S°C
(ii) MeOH, KF OTMS O H H H
Scheme 1.1.3.1.1
The IMDA has also been implemented in the syntheses of terpenoids. A route to
eremophilane and valencane sesquiterpenes was discovered using an intramolecular cycloaddition of (8) (Scherne 1.1.3.1.2). The key intermediate, (f )-(9), was then
transformed into sesquiterpenes such as eremoligenol (&)-(IO) and valencene (&)-(Il). 14
(10): eremoligenol (1 1): valencene
Scherne 1.1.3.1.2
Although one of the rings created by an M D A reaction is necessanly six-membered,
the size of the other ring is dependent on the length of the linking c h a h IMDA reactions are
not lirnited to just three or four atom tethers, as evidenced by the formation of macrocycles,
(12). (13) and (14), from (15).15 This IMDA reaction produced three different isomers
(Scheme 1.1.3.1.3).
1.1.3.2 Regioselectivity
(16)
Scheme 1.1.3.2.1
In conjunction with the formation of two rings, the IMDA reaction, like the
intermolecular variant, is capable of controlling up to 4 contiguous stereocentres. Unlike the
intermolecular reaction however, regiochemistry in the intramolecular reaction is of less
tethers, the length of the chain restricts the dienophile from forming regioisomers such as
(11) in a Type 1 cycloaddition. However, if the linking chah is too short (ie. n = O),
neither regioisomer forms.16 Only when the chain is long do regioisomers of type (16)
appear as products.
1.1.3.3 Stereoselectivity Stereoselectivity of a [4 + 21 cycloaddition is detennined by conformational, steric
and eIectronic effects of each possible transition state. IMDA reactions do not follow the
endo-rule discussed in Section 1.1.2.1.12 In systems that lead to [4.4.0] and l4.3.01 bicycles, two transition states (endo-/exo-) are possible, giving rise to trans- and cis-fused
products (Figure 1.1.3.3.1). For the [4.4.0] system, the transition state generally involves
the chair conformation of the linking chain.
- exo -
R trans
trans
Figure 1.1.3.3.1: Endo- and exo- transition states of bicyclo[4.4.0] and [4.3.0] systems.
Under thermal conditions, the geometry of a substituted dienophile in bicyclo[4.4.0]
systems, either tram- or cis-, (17) and (18) respectively, essentially plays no role in the
endo-/exo- product ratio (Scherne I. 1.3.3.1.1). 17 Secondary orbital interactions apparently
do not control stereochemistry. The completely unsubstituted 1,7,9-decatriene shows no
substantial preference for either diastereomer. l8
(1 7): R1 = COOMe, R2 = H (1 8): R1 = H, R2 =COOMe
Scheme
For the bicyclo[4.3.0] system, cis-fused rings predominate in the thermal cyclization
of unsubstituted trienes. 18 When substituted, however, trans-addition predominates. In the
[4.3.0] system (Scheme 1.1.3.3.1.2), endo-/exo-selectivity essentially has no dependence on
dienophile geometry, (19) and (U)).19
(19): RI = COOMe, R2 = H 72 28 (20): R1 = H, R2 =COOMe 67 33
Scheme 1.1.3.3.1.2
1.1.3.3.2 Steric Effects Steric bulk on the terminal end of the dienophile c m influence the stereoselectivity of
an IMDA reaction. For example, sulphonyl-substituted triene (21) produced just one
diastereoisomer, the exo-product, upon thermolysis (Scheme 1.1.3.3.2.1).*O Steric bulk
was explained to control the complete selectivity seen in the reaction. Here, the endo-
transition state is disfavoured due to the large sulfonyl substituent of the dienophile and only
the exo-product is recovered from the reaction.
exo
Scheme 1.1 A3 .2 . l
(23): Fi = H .
Figure 1.2.1: Martinelline (22) and Martinellinic Acid (23).
There are two species which comprise the tropical genus, Martinella (Bignoniaceae),
M. iquitosensis and M. obovata. The use of Martinella as eye medication by Amazon Indian
tribes has been reported.21 According to folklore, juice obtained from the bark of M.
obovata, when administered to the eye, has an immediate effect on inflammation and will
eventually cure conjunctivitis.
In 1995, Witherup and coworkers successfully isolated and characterized two
biologically active compounds from the roots of Martinella iquitosensis, Martinelline (22) and Martinellic acid, (23), (Figure 1.2.1).22 A comprehensive investigation of 1D and 2D
NMR data led to structure elucidation and assignment of relative stereochernistry about the
piperidine ring. An optical rotation was deterrnined for each compound, but the absolute
configurations of the three chiral centres are unknown.
1.2.1 Biological Properties
BiologicaI assays and in vitro antagonist studies were performed to determine the
properties of (22) and (23). Martinelline was found to inhibit the binding of radioligands to bradykinin Bl and B2, muscarinic and al-adrenergic receptors. As well, organ bath assays
with (22) indicated histaminergic receptor antagonism. Inflammation properties of (22) would be expected in vivo due to the inhibition of bradykinin and histarninergic receptor
systems, which mediate inflammatory reactions. In addition, bradykinin has potent nociceptive activity which suggests that antagonism of this receptor system would result in
(22), possessed antibiotic activity. The combined effects of analgesia, antibiotic activity and
control of inflammation provides the potential for (22) to have therapeutic value in the
treatment of ailments including conjunctivitis.
1.2.2 Lewis Acid Catalysis of Diels-Alder Reactions Much work in the area of Lewis acid catalysis of Diels-Alder reactions has resulted
since 1960, when Yates and Eaton first reported the acceleration of a Diels-Alder reaction by
a Lewis acid.23 They observed a rate acceleration of a Diels-Alder reaction between
anthracene and dienophiles such as dimethyl fumarate in the presence of aluminum
trichloride (Scheme 1.2.2.1). Without the addition of aluminum chloride, the reaction was
Scheme 1.2.2.1
not cornplete until after several days of refluxing dioxane. Lewis acids CO-ordinate to
compounds bearing electron-rich atoms such as nitrogen, oxygen or sulfur, which causes a net increase in electrophilicity. In a Diels-Alder reaction, this molecule is more often than
not, the dienophile. From CO-ordination by the Lewis acid, the energy of the LUMO of the
dienophile is lowered, thus decreasing the HOMO-LUMO energy gap of the diene and the
dienophile, respectively. This gives rise to an observed rate increase of the cycloaddition.
uncatalyzed 2.3 1 BF3.0Et2 exclusive
Scheme 1.2.2.2
Lewis acids can also affect the selectivity of Diels-Alder reactions. The addition of
BF3-OEt2 has been reported to completely reverse the regioselectivity from a thermal
1.2.2.2).24
Moreover, the addition of Lewis acids can influence the endo-lexo-selectivity of a
Diels-Alder reaction. For example, in the intramolecular reaction of dienyne (28), a
preference for the endo-products is observed under thermal, uncatalyzed conditions (Scheme
1.2.2.3). However, in the presence of the Lewis acid, AlClzEt, the cycloaddition of (28)
occurs at milder conditions and shows an even greater preference for formation of the endo-
produc t .25
EtOOC *
(28) endo- exo- 1 50°C 72 28 Et2AICI / 25OC >98 2
Sc heme 1.2.2.3
1.2.2.1 Lanthanide (III) Salts as Lewis Acid Catalysts One important characteristic of lanthanide (III) salts is their Lewis acidity. Compared
to AlCl3 and TiCl4, these synthetically useful compounds are milder Lewis acids and are
therefore compatible with acid sensitive molecules. Lanthanide salts have been found to
catalyze several types of reactions including Aldol condensations of silyl en01 ethers,
hydrocyanation of carbonyl compounds, polymerizations and cycloadditions.26 Lanthanide
shift reagents have been shown to catalyze Diels-Alder reactions, including the heteroatornic
variant? Modest selectivities are observed, but when the combination of a chiral diene and
the chiral lanthanide, (+)-Eu(hfc)3, is used, excellent diastereofacial selectivities result
(Scheme 1.2.2.1.1).28
O R *
+
Scheme 1.2.2.1 .f
26
Recently, Kobayashi has implemented a three component coupling approach in
lanthanide triflate catalyzed reactions. In one report, an overall Diels-Alder type
transformation was achieved using three separate components, an aldehyde, an aniline and
an electron rich dienophile (Scheme I.2.3.1).29 Complete endo-selectivity was observed
when cyciopentadiene acted as the dienophile. For vinyl ethers and thioethers, a mixture of
R1 = Ph, COPh, COOMe R2 = Hl p-MeO, O-Me0 R3 = SPh, OEt
Scheme 1.2.3.1
products was obtained. Multi-component couplings have many advantages, the greatest of
which is the reduction in the number of isolation/purification steps.
Kobayashi continues to investigate three component couplings and the Lewis acidity
of lanthanide triflates. For example, the generation of cycloadducts through this approach
has become amenable to combinatorial r n e t h o d ~ . ~ ~ In a Mannich-type three component
coupling, a 2-azadiene generated in situ from substituted aldehydes and amines, was
alkylated with TMS-CN in the presence of Yb(OTD3, (Scheme I.2.3.2).31
Yb(OTf)3 (5-1 0 mol %) HQP 1 + R2-NHZ + TMSCN H R I 4A MS
Scheme 1.2.3.2
1.2.4 Enamines and 2-Azadienes in Diels-Alder Reactions As a result of their electron rich nature, enamines react best with electron deficient
dienes in 'inverse electron dernand' Diels-Alder reactions. The first examples of enamines
participating in such reactions were reported by Berchtold in 1965. In this study, the
cycloadduct which either irnmediately eliminated to give conjugated dienes (29) and (30) or
were forced to elirninate upon treatment with methanolic hydrochloric acid (Scheme
Q COOMe A
COOMe
(30)
Scheme 1.2.4.1
3 COOMe COOMe
A (yN + - ) - I
/
Bn Bn
Scheme 1.2.4.2
As an extension in utility of this class of cycloadditions, quinolines and isoquinolines
have been synthesized from non-aromatic starting materials (Scheme 1.2.4.2).33 The
overall [4 + 21 cycloaddition was believed to go through a step-wise process involving
nucleophilic addition of the enamine to the electrophilic diene, not through a concerted Diels-
Alder process.
Enamines react regeospecifically with 2-(phenylsulfony1)- 1,3-dienes in an overall [4
+ 21 transformation (Scheme 1.2.4.3).34 The phenylsulfonyl dienes are synthetically
versatile in cycloadditions since they have been found to be reactive with both electron rich
deficient and electron rich dienophiles.3"
Scheme 1.2.4.3
Scheme 1.2.4.4
The first account of N-benzylidene aniline reacting as a diene in Diels-Alder reaction
was reported by Povarov (Scheme 1.2.4.4) and similar findings by Worth and Elsager
supported these results. Since then, much research on 2-azadienes as dienophiles has
followed.35 Cycloadducts produced by these dienes possess the tetrahydroquinoline
skeleton, a cornrnon moiety in bioIogically active natural products. In general, 2-azadienes
participate in inverse electron demand Diels-Aider reactions and are substituted with strongly
electron donating groups, thereby enhancing the reactivity towards electron rich alkenes.
The 2-azadiene derived from condensation of aniline and benzaldehyde has been shown to
undergo [4 + 21 cycloadditions with various electron rich species, under Lewis acidic
conditions. For example, vinyl thioethers, cyclopentene and cyclohexene acting as
dienophiles have been reported.35
In addition to intermolecular cycloadditions, 2-azadienes undergo intramolecular
reactions (Scheme I.2.4.5).36 The 2-azadiene, (31), generated from in situ aldehyde and
aniline condensation, cyclized with the tethered alkene, producing the tricycle, (32).
R1 = Me, H R2 = Me, H R3 = Me, H
Scheme 1.2.4.5
Isoquinolium salts have also been exposed for their high reactivity as 2-azadienes in
[4 + 21 cycloadditions.37
Inverse electron demand Diels-Alder reactions between enamines and 2-azadienes
were unknown prior to Nomura's publication in 1978. His group showed that under acidic
conditions, adducts corresponding to [4 + 21 cycloaddition could be isolated as single
diastereomers (Scheme I.2.4.6).38 Under neutral conditions, only alkylation products were
observed, but upon treatment of these products with acid, the cycloadducts (33), could be
generated. Accordingly, a stepwise or zwitterionic mechanism was proposed in which
alkylation of C=N of the diene preceded ring closure ont0 the aniline ring.
R1 = morpholino, Me R2 = Me, H
Scheme 1.2.4.6
The presence of an a-H on the enamine appears to influence the ring closure step of
the zwitterionic mechanism. Shono and CO-workers reported that the product of a reaction
between the iminium ion (34), and an enamine depended on whether the enarnine was
derived from an aldehyde or a ketone (Scheme L2.4.7).39 When the enamine (35) derived
from 1-butanal, was employed, a cycloadduct (36) was recovered. The cyclohexanone
derived enamine, (37), produced only the alkylation product, (38).
Scheme 1.2.4.7
Whether the cycloaddition between enarnines and 2-azadienes occurs via a concerted
or a zwittterionic mechanism continues to be debated. In one study, a concerted mechanism
was proposed for the BFyOEt2 catalyzed cycloaddition of enamine, (39) and other electron
rich alkenes, to 2-azadiene, (40), (Scheme 1.2.4.8).40 Three experimental observations led
to this proposal: (i) the bond stereochemistry of the azadiene was completely retained; (ii)
benzyl vinyl ether was reluctant to add to the isomer (41), even in a nucleophilic sense; (iii)
the side product, (44), of the cycloaddition between (40) and (42), which would be
expected to form in a step-wise mechanism, (nucleophilic addition to give (43), followed by
rapid desilylation to generate (44) upon reductive work-up), was not observed. The
existence of a step-wise mechanism, though not supported by the above data, was confirrned
in a separate study which successfully isolated the methanol quenched compound, (45),
(42) (43)
Scheme 1.2.4.8
concerted zwitterionic
H' H' Ph Ph Ph
(45)
Scheme 1 A 4 . 9
(Scheme 1.2.4.9). In this account, both the concerted and the zwitterionic mechanisms were
believed to be acting simultaneously; although the isolation of the intermediate compounds is
with solvent polarity was rationalized by the concerted mechanism. In either mechanism, the products can be regarded as those of a typical concerted Diels-Alder reaction.
Chapter II:
Results & Discussion
Alkenylboranes
11.1.1 Objectives The goal of this project was to develop and optimize a convenient 3 step, one-pot
procedure for the synthesis of bicyclo[4.3.0]alkenols. The 3 sequential steps involved were:
(i) hydroboration of an alkyne selectively over a conjugated diene, (ii) IMDA reaction of the
resultant alkenylborane and (iii) functionalization of the C-B bond (Scheme II. 1.1.1).
(47) (48) (49) Y = OH, NH2, Br etc
Scheme II.l.l.l
Hydroboration of an alkyne (46), generates an E-alkenylborane (47), which undergoes
thermal endo- or exo-cycloaddition. Functionalization of the resultant alkylborane (48), using standard techniques gives a route to substitued [4.3.0] bicycles (49). The net effect of
the alkenylborane is to act as a "masked" dienophile. In the event that the C-B bond is
oxidized, the alkenylborane becomes synthetically equivalent to an E-enol, a normally
unreactive dienophile. Ketenes are also inappropriate dienophiles since they undergo thermal [2 + 21 cycloadditions with 1,3-butadienes. But, a Diels-Alder reaction with an
alkenylborane followed by oxidation to a ketone accomplishes the same net transformation.
At the outset of this project, there were no examples of intramolecular Diels-Alder
cycloadditions of alkenylboranes. This project primarily dealt with the development of
methodology and it was imperative to synthesize stereochemically interesting substrates
rapidly, with ease and in relatively high yields.
11.1.2 Syntheses of Substrates In total, three substrates were synthesized for the purposes of generating their
corresponding IMDA adducts. Each substrate contained a three atom tether, connecting the
conjugated diene to the terminal alkyne. The simplest of the substrates, (SO), was formed
under standard Williamson ether synthesis conditions as depicted in Scheme II. 1.2.1. Thus,
I . I v a n , I nr, v u LU I .L. * J/ \b O H 2. = , , 0°C to r.t. 168% O
Br d
Scheme 11.1.2.1
trans,trans-2,4-hexadien- 101 was deprotonated with sodium hydride and then was alkylated
with propargyl bromide, providing the desired ether (50). Due to the availability of this
ether, most of the methodology was optimized using this substrate.
A more complex substrate, (51), was synthesized from dimedone as shown in
Scheme 11.1.2.2. Treatment of dimedone with methyl orthoformate followed by vinyl
magnesium bromide generated the conjugated diene system. Subsequent 1,2-reduction of
the resultant enone with NaBH4KeC13 and alkylation with propargyI bromide generated the
second substrate, (SI)?*
(i) HC(OMe)3, H2SO4 (cat.), MeOH / 96% (ii)
MgBr ' Et20, -78°C
(iii) NaBH4, Ceci3, MeOH / 58% for (ii) and (iii) (iv) NaH, -, , TH FI 55°C / 54%
Br
Scheme 11.1.2.2
To determine the effect of an al1 carbon tether and whether the one-pot procedure
would tolerate functional groups such as esters, substrate (52) was synthesized, using a
malonic ester procedure (Scheme II. 1.2.3). AIthough the conversion of alcohols to halides
c m generally be achieved through well established conditions, conversion of trans,trans-2,4-
hexadien-1-01 to its corresponding chloride can lead to scrambling of the chloride atom to
any one of 3 positions through allylic rearrangement. However, conditions developed by
Collington and ~ e y e r s 4 ~ in which allylic hydroxyl groups are converted to chloides without
rearrangement, was followed to convert the hexadienol to the corresponding chloride. In
this procedure, the alcohol is mesylated in situ and displaced by chloride ion. The resultant
crude chloride, (53), was subsequently used to alkylate deprotonated diethyl malonate in a
with sodium hydride and alkylated with propargyl bromide, to afford the final substrate
(52) .
(i i i)
(i) LiCI, MsCI, 2,6-lutidine, DMF (ii) NaH, CH2(COOEt)2 , THF, 0°C / 25% for (i) and (ii) 3 COOEt (iii) NaH, = , , THF 172% - COOEt
Br (=)
Scheme 11.1.2.3
11.1.3 Alkyne Hydroboration of Substrates Several hydroborating agents were considered for the first stage of the one-pot
procedure. Catecholborane (CB) and dirnesitylborane, both cornmercially available, were
screened for their potential to selectively hydroborate the alkyne and activate the dienophile
for the subsequent Diels-Alder reaction. For CB, only (55) was observed in the crude NMR spectrum. This was not entirely surprising since the conditions for hydroborating alkynes
with CB required heating, which likely facilitated the cycloaddition (Scheme II. 1.3.1).43
Dimesitylborane, the hydroborating agent that is most selective for alkynes, on the other
hand, did hydroborate the alkyne as observed in the NMR spectrum of an aliquot of the
Scheme ll.l.3.l
bulk of the rnesityl substituents prevented the triene from adopting a conformation suitable
for cycloaddition (Scheme II. 1.3.2). Another possible factor was the electron richness of
the 2 mesityl groups which may have played an electronic role, rendering the alkenylborane
unreactive.
Scheme 11.1.3.2
The dialkyl alkenylborane, vinyl-9-BBN, had been previously detennined to be a
reactive dien0~hile.9 However, the corresponding hydroborating agent, 9-BBN, which
would be used to generate an alkenylborane, is not selective for alkynes and was therefore
unsuitable for our methodology.*a Instead, other dialkylboranes were considered for the
hydroboration 1 activation of the dienophile. In particular, disiamylborane, Sia2BH, and
dicyclohexylborane, Chx2BH, were the most promising of the hydroborating agents, based
on their selectivity and reactivity for alkynesm45 In separate reactions, each reagent was made
in situ under normal conditions before the introduction of dieneyne (50).8a Following
hydroboration (2-3 hours), solvent was added (toluene or benzene) to dilute the reaction to
0.05 M in substrate and the reaction was refluxed for 17-20 hours. Oxidation of the
trialkylborane with NaOH 1 H202, and purification by colurnn chromatography resulted in
the isolation of a single diastereomer of the desired bicyclo[4.3.0]alkenol, but in poor yields
(both Sia2BH and Chx2BH generated less than 20% of products). Of the two hydroborating
agents, Chx2BH, a thick white precipitate in THF, gave more consistent results since the
observation of this precipitate confirmed the presence of active Chx2BH. The addition of
substrate to Chx2BH caused the precipitate to dissolve. The quality of SiazBH, on the other
hand, could not be so readily determined since it remained a colourless solution in THF. When Sia2BH was used, a broad range of yields, (0% to 20%), were obtained, with the
unhydroborated product as the only other identifiable side product. As a consequence of
these practicd advantages, Chx2BH was chosen as the hydroborating agent in our three step
transformation.
THF, 0°C 3 NaBH4 + 4 BF3-0Et2 - 4 BH3.THF + 3 NaBF4-t- 4 EtaO
Equation 11.1.3.1.1
Commercial borane-THF complex has a limited shelf life and its concentration in
solution is diffïcult to determine.46 Preparation of borane-THF can be achieved by treating
NaB H4 with BF3-OEt2 in THF (Equation II. 1.3.1.1). Our approac h was to generate a
known quantity of BH3.THF, add 2 equivalents of 2-methyl-2-butene or cyclohexene, and
combine this mixture with the substrate, al1 in one pot. Unfortunately, this failed as it led to
incomplete and unselective hydroboration (possibly originating from other hydroborating
species with formulae such as RBH2) of trial substrates, hexyne and 3-cyclohexyl-1-
propyne as observed in crude NMR spectra. Instead, a commercial solution of BHySMe2
complex became the preferred source of borane due to its increased stability.8a
11.1.4 Oxidation of IMDA Adducts Oxidation methods which maintain the configuration of the C-B bond were sought
for the functionalization of the trialkylborane intermediate adducts to alcohols. The most
traditional method, treatment with alkaline peroxide, would be effective for the oxygen
tethered adducts, but less nucleophilic conditions were desired for the diester substrate (52), to avoid ester hydrolysis. Two oxidants, NaB03 and Me3N0.2H20, are known to oxidize
trialkylboranes under neutral conditions.47 To compare these 2 oxidants, a solution
containing trialkylborane precursor to adduct (56) was divided into 2 equal portions, one of
which was oxidized with NaBO3 and the other with Me3N0-2H20. Although both
produced the desired alcohol, slightly better yields were obtained with Me3NO.2H20.
11.1.5 Optimization of IMDA Reaction Conditions The type of solvent, the temperature and duration of heat were varied in the
optimization of the Diels-Alder step. Toluene, THF and benzene, al1 freshly distilled, were
added to separate reactions, diluting the alkenylborane, derived from (50), to 0.05 M.
Varying the temperature of the solvent, it was found that refluxing benzene generated the
highest conversion to the cycloadduct.
While productive strides at this point were made in the hydroboration, oxidation and
IMDA reaction conditions, the best yield of cycloadduct was only 20%. No other products
could neither be isolated nor identified. Because crude mass yields were consistently lower
the trialkylborane Ieading to decomposition to volatile compounds was suspected. The effect
of adding a radical inhibitor was consequently exarnined. In fact, when a srnaIl quantity (5
mol 96) of butylated hydroxytoluene (BHT), dissolved in benzene, was added to the
alkenylborane and was heated, a profound increase in yield was obtained (70%).
The final optimized conditions for Our one-pot transformation employed Chx2BH for
the hydroboration step, 5 mol % BHT in benzene for the cycloaddition step and
Me3NO-2H20 as the oxidant. The substrates, (51) and (52), were then subjected to these
conditions and both gave rise to their corresponding adducts, (57) and (58) respectively, as
single diastereomers in moderate yields (Table II. 1 S. 1).
Substrate Cycloadduct Isolated Yield (%)
(52)
E = COOEt .. -. -
Table 11.1.5.1: Yields of IMDA adducts.
An alternate hydroborating agent, pinacolborane, PBH, was considered due to the potential for isolating the alkenylborane by column chromatography. The large pinacol
auxillary protects the alkenylborane from oxidation or hydrolysis, even in the presence of
silica ge1.48 This was particular attractive since the alkenylborane could be isolated and the
difficulties in the hydroboration step would then be avoided. Hydroboration with PBH is
chemo- and regioselective, generating the Z-alkenylborane almost exclusively. The first
report of PBH as a hydroborating agent notes that 2 equivalents of PBH must be employed
for complete hydroboration of alkynes. For our purposes, this was not adequate since Our
substrates contained more than one potential site of hydroboration. Work reported later by
Srebnik showed that these hydroborations can be achieved with stoichiometric PBH in
presence of Schwartz' Reagent, HZrCplCl (Scheme 11. 1.6.1).49 Yields reported under
these catalytic conditions were excellent as was the preference for forming the Z- alkenylborane of terminal alkynes.
HZrCp2Cl (cat.) y $ - H + H*R1 0. x R 1
CH2CI2
Scheme 11.1.6.1
Substrate (50) was hydroborated under Srebnik's conditions and the resultant
alkenylborane was isolated. Several attempts were made to improve the yields, using only
freshly made Schwartz' Reagent, but the greatest yield of the alkenylborane, (59), was only
19%? Particularly puzzling was that 2 equivalents of PBH, not the stoichiometric amount,
were used to obtain this yield. When the reaction was run in the absence of catalyst, a yield
of 9% was obtained, thereby confirming the enhanced, albeit slight, yield provided by
Schwartz' reagent. Srebnik did not atternpt his methodology on olefins and whether this
would affect the chemoselectivity of the PBH / Schwartz' Reagent cornbination is unknown.
Upon heating the alkenylborane, (59), in benzene (following degassing via freeze-thaw
cycles) in an NMR tube, an IMDA adduct formed (Scheme II. 1.6.2), but attempts to purify
it by silica column chromatography failed. Because of the poor yield of alkenylborane and
the inability to purify the IMDA adduct, this aspect of the project was not studied further.
Nevertheless, this remains a potentid route to stable, isolable dialkyl pinacolboronates.
2 eq. PBH,
HZrCp2CI (cat.), CH2CI2
PhH, A ____t x o
?-8
Scherne 11.1.6.2
11.1.7 Stereochemistry of IMDA Adducts According to the 1D and 2D NMR spectra of each pure product, it was evident that
only one diastereomer was formed in the three-stage transformation. Nuclear Overhauser
effect difference (NOE) experiments were performed to determine relative stereochernistry within each adduct. Tabulated in Figure II. 1.7.1 are the percentage enhancements of
protons, which measure the relative proximity of protons, between protons of interest.
NO€ 1 rradiated Enhancement
NOE l rradiated Enhancement
NOE lrradiated En hancement
Figure 11.1.7.1 : NOE enhancements of IMDA cycloadducts.
For each molecule, the data suggest trans-ring junctions between the 2 newly forrned
rings. In each case, none of the protons in the new bridge exhibited NOE's, indicating a
(56), was provided by X-ray analysis. Many derivatives of benzoic acids (60), (61), (62),
and (63) were synthesized, but only derivative, (64). generated a suitable crystal (Scheme II. 1.7.1). The ORTEP drawing (Figure II. 1.7.2) clearly depicts the trans-fusion of the 5-
and 6-membered ri11~s.5~
02N-O 0 H 9 DCC, *
DMAP (0.3 eq), CH2C12 O
Scheme 11.1.7.1
Figure 11.1.7.2: ORTEP drawing of derivatized IMDA cycloadduct (64).
The formation of trans-fused rings in adducts (56), (57) and (58) results from enda- addition in the IMDA step and is therefore consistent with the proposed [4 + 33 transition
state mode1 discussed in section 1.1.2.2.2. Considering steric congestion alone, this
geometry is sornewhat surprising since the cyclohexane rings on the boron atom are so
bulky. It is certain that the boron substituent on the dienophile has exceptional electronic
effects, which result in complete endo-selectivity.
"exo-" transition state &-
Ref. X RI R2 trans- : cis-
(65a) 52a CH2 H H 1 : 3 (ab) 52b CH2 COOCH3 H 6 : 4 (65c) 52b CH2 COOCH3 CPr 72 : 28 (65d) 5 2 ~ CH2 NO2 H 89 : 11 (65e) 52d O H H 1 : 3
(65f) 52d O CH=CHCH3 H 4 : 1
Figure 11.1.8.1 : Some comparative data of hydrindene-type cycloadducts.
In Figure II. 1.8.1, trans-:cis- selectivity data are presented for analogous products of
thermal IMDA reactions.52 The simplest case, (65a), in which there is no terminal
substitution on either the dienophile or diene, shows a preference for the cis fused
cycloadduct. However, substitution at the terminal end of the dienophile as in (65b), (65c)
and (65d), forming a 2-alkene, leads to the formation of trans-fused rings. Substitution at
both termini, as in (65c), shows even greater selectivity for the trans-isomer than
substitution on the dienophile alone, (65b). For the oxygen tethered substrates, (65e) and
(650, sirnilar selectivities for the trans-fused bicycles are observed with substitution at Ri. Although there is a lack of appropriate comparative data (ie. IMDA adducts with substitution
bear substituents on either or both termini, there is an inherent preference for endo-addition. In our exarnples, it appears that the effect of the boron substituent is two-fold: (i) the endo-
isomer is preferentially formed; (ii) the conditions needed to generate IMDA adducts are rnilder than those used to forrn (65a-f).
3 (ii) 7S°C, 2.5h S (iii) NaB03.4H20
(66)
Scheme ll.l.8.l
Since the initiation of this project, Singleton and Lee have reported the only other
example of an intramolecular Diels-Alder reaction of an alkenylb0rane.5~ In their exarnple, a
single diastereomer, the trans-decalin derivative, (66), was obtained (Scheme II. 1.8.1) and
this selectivity was attributed to the [4 + 31 transition state. Indeed, molecular calculations
support this hypothesis.52 However, other 1,7,9-decatriene systems bearing substituents at
the 8-position, also show selectivity for the formation of trans-fused decalins (Scheme
II. 1 .8.2). In each of these examples, only one diastereomer, the tram-fused product, is observed in the reaction.54y12b Although this does not contradict the idea of a [4 + 31
95% (trans-fused)
TMS >95% (trans-fused)
Scheme 11.1.8.2
8-position is the primary controlling factor; that is, the diastereoselectivity in the product (66) is chiefly due to the nature of the 8-methyl-1,7,9-decatriene skeleton of the E- alkenylborane, not the electronic effects of boron.
II. 1.9 Conclusions The goal of developing a one-pot, three sequence transformation, involving
hydroboration of dienynes followed by intrarnolecular Diels-Alder reaction and oxidation,
has been achieved. Hydroboration of a terminal alkyne generates a reactive dienophile which undergoes an IMDA reaction under thermal conditions. Subsequent oxidation provides
alcohols, which in Our case were bicyclo[4.3.0]alkenols. Optimized conditions employ
dicyclohexylborane as the hydroborating agent, BHT (5 mol %) in benzene as the refluxing
solvent and Me3NO-2H20 as the oxidant. The conditions tolerate an oxygen atom in the
tethering chain and the ester functional group. Yields of the three substrates are moderate to good, but complete stereoselectivity in the transformation is observed. In fact, al1 of the
cycIoadducts, (56), (57) and (58), derive from endo-addition of the alkenylborane in the
IMDA step.
It is clear that this methodoiogy is a promising, new stereoselective route to
hydrindene-type structures. Still, there are many aspects of the reactivity of alkenylboranes
that remain unexplored. To fully expose their synthetic utility, the chemistry of
alkenylboranes will continue to be the subject of future research in Our group.
Synthesis of Martinelline
11.2.1 Objectives The primary goal of this project was to develop an efficient and high yielding route to
Martinelline, (22), and Martinellinic acid, (23). The greatest synthetic challenge of
Martinelline is the pyrroloquinoline core with the designated relative stereochemistry. Our
proposal t'o synthesize this scaffold was to react a 2-azadiene with an enamine in an overall
Diels-Alder transformation. To effect the cycloaddition, Lewis acids, particularly lanthanide
(III) triflates, would be investigated. Moreover, we proposed the possibility of employing a three component coupling, in which three reactive species come together and react, to
generate the pyrroloquinoline core in a single step. This represents the greatest obstacle to an
elegant, but efficient, first total synthesis of Martinelline.
11.2.2 Retrosynthetic Strategy The general retrosynthetic approach to the total synthesis of MartinelIine is outiined in
Scheme 1.2.2.1.55 ~isconnection of the six-membered ring of (22) and (23) as shown
ieads to the possibiiity of an inverse electron demand Diels-Alder reaction between an
enamine (67) and a 2-azadiene, which would likely require Lewis acid catalysis. To
generate the diene, a cornrnercially available 4-aminobenzoate ester, (68), can be condensed
with the aldehyde, (69). This can be achieved either through a two-transformation, imine
formation and then cycloaddition, or a one-pot reaction, in which the imine (2-azadiene) is
forrned in situ. In any case, to synthesize fragments (67) and (69), 2-pyrroline (70) and 4-
aminobutanal, (71), could be alkylated with the synthon, (72). Should the cycloaddition
between (67), (68) and (69) prove to be problematic or hindered by the guanidino side
chains, the pyrroloquinoline core could be pre-formed from benzoate (68), enamine, (701,
and aldehyde (71); the side chains could be installed afterwards by alkylation with (72).
Access to other Martinelline related compounds (ie. R = H, Me) would seemingly be trivial
through transesterification.
To install the guanidine segments, numerous conditions are available. For example,
reaction of the free amines with arninoiminomethansulfonic acid (73) followed by treatment
with prenyl bromide, (74), gives the desired moiety (Scheme 1.2.2.2).~~
n 14 + 21 (one- or two-pot)
/
Scheme 11.2.2.1
K (i) H03S NH2 Br (73) NH (74)
R-NH2 * '*NKNH2
(ii) OH- H base H H
Scheme 11.2.2.2
Scheme 11.2.3.1
0
The two most most obvious means of generating dihydropyrrole are through the
reduction of pyrrole or the oxidation of pyrrolidine. The reduction of pyrrole requires
forcing conditions since pyrrole is aromatic, but the greatest obstacle is that the
monohydrogenated pyrrole is more susceptible to further reduction than the parent pyrrole.
Alternatively, 2-pyrroline could be generated by oxidizing pyrrolidine. Indeed, pyrrolidine
can be oxidized under mild conditions to give the 2,3-dihydropyrrole, but this compound is
most stable as the corresponding 1-pyrroline (Scheme 11.2.3.1) which is in turn in
equilibrium with the trimer, ( 7 3 . 5 7 This equilibrium is shifted, favouring the monomer,
when CO-distilled with THF. In a receiving flask cooled to -780C, the monomer can then be
acylated with an acid chloride as indicated in Scheme 11.2.3.1.58 Both Cbz and ~ 1 1 0 ~ 5 9
( i ) A, THF (75)
(ii) CICOOR, NEt3, THF, -78OC / Ca. 25%
protected enamines, (76) and (77) respectively, were prepared in this manner.
The corresponding acid chloride for the Boc protecting group is exteremly unstable,
not commercially available and extremely difficult to prepare?O Attempts were made to
generate the Boc protected enamine (78) from Boc20. Due to the difference in electrophilic
character between anhydrides and their acid chlorides, DMAP (10 mol %) was added to
further activate the anhydride to nucleophilic substitution. Unfortunately, d l attempts failed,
showing no desired product in the crude NMR spectra. A recent report by Dieter and
beginning with protected pyrrolidines (Scheme 11.2.3.2).61 Indeed, the desired
enecarbarnate was produced in two steps, though in a low overall yield.
(78)
Scheme 11.2.3.2
Each of the three protected N-acyl 2,3-dihydropyrroles, (76), (77) and (78), can be
deprotected under a variety of conditions. The Cbz protecting group is easily removed by
hydrogenation with H2 and a Pd catalyst. Alloc-carbamates are deprotected by Pd(PPh3)4,
while Boc-carbarnates are deprotected under acidic conditions (e.g. T F A ) . ~ ~ Having easy
access to these carbarnates was important in tenns of the flexibility of the deprotection stage
in Our synthesis.
11.2.4 Three Component Coupling Reactions
11.2.4.1 Preliminary Work The first successful three component coupling reaction performed in Our research
group was perfomed with benzaldehyde, aniline and methyl carbamate analogue of (76) (present in two-fold excess), in an overall isolated yield of 73% of two diastereomers, (79), (Scheme II.2.4.1.1).63 In cornparison, when the same transformation was performed in
two steps, (ie. the imine of benzaldehyde and aniline was prefonned, isolated and subjected
to the dienophile), the yield of the diastereomers, (SO), after the two steps was 42%.
Y b(OTf)3 (cat.)
(79): R = COOMe (80): R = Cbz
Scheme ll.2.4.l.l
Limited rotation of the carbamate protecting group gave rise to rotomers which complicated
the NMR spectrum causing structure identification to become impossible. Attempts to
conditions, and thus, the Cbz protected analogue, (76), was selected in the proceeding
studies. Later however, it was found that the observation of rotomers could be avoided by
conducting NMR at elevated temperatures (40°C).
Assuming a step-wise mechanism, the formation of (79) and (80) can be easily
rationalized (Scheme 11.2.4.1.2). The first step involves the nucleophilic addition of the
protected enamine, to Lewis acid CO-ordinated benzylidene aniline. In the second step,
electrophilic aromatic substitution of the iminium ion is followed by re-aromatization, which
generates the observed diastereomers.
nucteophilic *
addition
electrophilic aromatic
substitution
rearomatization
Scheme 11.2.4.1.2
11.2.4.2 Optimization of Reaction Conditions As expected, the Cbz protected dihydropyrrole (76) underwent the three component
coupling reaction under the same conditions used to form @O), in 42% yield. As an
altemate dessicant to MgSO4, which is difficult to remove from the reaction, 4A molecular
sieves were used. No noticeable difference in yield was obtained (38%). Acetonitrile
proved to be a suitable solvent for the reaction; in dichloromethane, the reaction produced an
inseparable mixture of products as evidenced by TLC. Of the three components in the reaction, the protected dihydropyrrole was the least
readily available and efforts were made to decrease the quantity of (76) needed in the
reaction. When 1.7 to 1.8 equivalents each of aniline and benzaldehyde were used, a yield
point, the initial optimized conditions were: 1.7 to 1.8 equivalents each of benzaldehyde and
aniline, 10 mol % of lanthanide triflate, 4A MS in MeCN at 0-40C for 20 hours. The need
for dessicant in the three component coupling was later determined to be unnecessary (see
Section 11.2.5).
11.2.5 Survey of Lanthanide (III) Triflates as Lewis Acid Catalysts The lanthanide (III) triflates, La(OTf)3, are diverse in size and electronics, leading to
differences in Lewis acidity.*6 Al1 of the lanthanide (III) triflates are available comrnercially,
although some (eg. thulium, lutetium and scandium) are more expensive. In this survey, the
less expensive lanthanide triflates, and consequently, those which would have the potential
for industrial scale applications, were exarnined. To compare their ability to catalyze the
three component coupling, each Lewis acid was used to catalyze the unification of
benzaldehyde, aniline and Cbz-protected enamine (76). In Table 11.2.5.1, the yield of
cycloadducts (80a) and @Ob), obtained as a pair of diastereomers, is listed along with the
diastereomeric ratio, for each Lewis acid. This ratio was determined by NMR analysis of a
purified mixture of diastereomers. Two peaks, one from each of the diastereomers, were
conveniently isolated from each other allowing accurate integrals to be measured.
The data contained in the table are somewhat surprising. For example, Yb(OTf)3,
which was the most effective lanthanide Lewis acid at catalyzing the aldol of silyl en01
e t h e r ~ , 6 ~ showed the lowest isolated yields of the two diastereomers. Dysprosium triflate,
which was almost as effective at catalyzing aldol reactions as Yb(OTQ3, produced the
highest yield, 94%. The control experirnent in which no Lewis acid was added to the
reaction gave the expected result of no observed cycloadduct. Only the imine and the
dihydropyrrole were recovered from the reaction. According to the ratios of diastereomers
for each Lewis acid, there appeared to be no selectivity of one diastereomer over the other,
and most of the ratios are within the experimental error of NMR (5%). A second control
experiment was perfonned to determine the effect of the molecular sieves on the reaction.
Surprisingly, no discernable differences in yield were found. As a result, the optimized
conditions for three component couplings were modified and dessicant was no longer
involved in the reaction. In this way, the reaction became easier to manipulate and to work-
up since no filtration was needed to remove the molecular sieves.
. . Q C bz.
0 NH2 H@ MeCN, 4 A MS,
O 0-4OC, 2Oh
Ln(OTf)3 (hyd rous) (80a) : (Wb) lsolated Yield (%)
no Lewis acid starting material only
WOT93 51 :49 75
Pr(OT93 52 : 48 84
Nd(OTf)3 51 : 49 79
Gd(OTf)3 52 : 48 90
Eu(OT93 57 : 43 74
Er(OT93 52 : 48 83
Y b(OTf)3 50 : 50 €8
Yb(OTf)3 (anhydrous) 60 : 40 83
DY (Om3 51 :49 94
DY(OT~)~ (no MS) 46 : 54 91
Table 11.2.5.1: Cornparison of Lewis acid, diastereomeric ratio and yields of cycloadducts.
11.2.6 Effect of Pyrroline Protecting Group Under the optimized conditions with Dy(OTf)3 as the Lewis acid, the Alloc protected
dihydropyrrole, (77) was coupled to benzaldehyde and aniline. As expected, the yield
obtained was cornparatively high (93%) with that of the Cbz analogue (91%) and no
diastereoselectivity was observed (1 : 1 mixture of diastereomers, (81a) and (8 1 b)) . Displayed in Table 11.2.6.1 are the yields and the diastereomeric ratios of the three
component coupling products derived from benzaldehyde. In the case for methyl 4-
aminobenzoate, a substantially lower yield was obtained with some diastereoselectivity.
Ln(OTQ3 (10 mol %), - MeCN, 0-4OC, 20h
RI Ri! (a) : (b) lsolated Yield (%) C bz.
(80) : H Cbz 51:4ga 91 MeOOC,p&
(81): H Alloc 50: 50a a
(82): COOMe Cbz 70:30b 61C
Determined by NMR integration b Determined by isolated masses
Unoptimized
Table 11.2.6.1: Effect of substituents on yield and diastereomeric ratio.
Representing another 13% of the product mixture was the oxidized product, ( 8 2 ~ ) .
Apparently, this product was formed by oxidation of either diastereomer (82a) or (82b),
since pure NMR samples (in CDC13) of either diastereomer when Ieft standing showed the
presence of (82c). One plausible mechanism which involves oxidation of the quinoline ring
system followed by an electrocyclic reaction is shown in Scheme 11.2.6.1. A driving force
(82a) or (82b) (82~)
E = COOMe
Scheme 11.2.6.1
for the electrocyclic reaction is the re-conjugation of the phenyl ring with the aniline ring.
For the oxidation step, a radical type process is possible, but how and why it is occurring
- . - - . - - - - - - -
has a key role.
Diastereomers (82a) and (82b) were easily distinguished upon inspection of the
aromatic region of the corresponding lH NMR spectra. For (82a), isolated as colourless
crystals, stereochemical assignment was provided by X-ray crystallographic data (Figure
11.2.6.3). By comparing the the aliphatic region of the lH NMR spectra, the cycloadducts
(80a) and @la) were determined to possess the same reIative stereochernistry as (82a), the
product of endo-addition. Using the same approach, cycloadducts @Ob), (81b) and (82b)
were identified as exo-isomers. In addition to the similarities in NMR spectra, al1 of the
endo-isomers were isolated as solids, and the exo-isomers as oils.
Figure 11.2.6.2: ORTEP drawing of (82a).
Work until this point utilized benzaldehyde as one of the three components used to
forrn the imine. This is because of the notorious instability of irnines generated from
aliphatic aldehydes. The added p-electron delocalization of benzaldehyde stabilizes the irnine
so that it becomes isolable.65 In Our strategy of forming a conjugated imine in situ and
subsequently reacting it with a dienophile, this difficulty in irnine isolation is avoided. Thus,
in principle at least, the aliphatic aldehyde can be used as a precursor to products containing
an aliphatic chah (Scheme II.2.7.1).
Cbz.
Conditions; Lr1(0Tf)~ (0.1 eq.), MeCN, 0-4*C, 20h
x Cbz.
(84)
Scheme 11.2.7.1
The first of the non- aromatic aldehydes whose suitability towards our met hodology
was tested, was ethyl glyoxylate. Generated from the periodic acid cleavage of diethyl
tartrate,66 this aldehyde is known to form stable irnines with aniline.6' A two component
approach to the synthesis of (83) was attempted with the preformed imine. Although traces
of cycloadduct did appear in the crude NMR spectrum, purification was problematic and no
further experiments using this aldehyde were performed. Further studies remain to be
undertaken to determine whether ethyl glyoxylate would be useful in our three component
coupling methodology.
In the interests of simplifying the diastereomeric product mixture, forrnaldehyde was
a logical choice as the aldehyde. Without any substituents on the aldehyde, only
enantiomeric products could result from the reaction. A formaldehyde equivalent, p-
formaldehyde, was used in a three component coupling, but no cycloadduct matching the
chromatography was tentatively designated on the basis of NMR, as a product of some
polymerization process. This was later found to be an incorrect assessment, (section LI.2.8).
A simple aliphatic aldehyde bearing no substituents or functionality was tested.
Valeraldehyde (pentanal), freshly distilled, was combined with aniline and the protected
dihydropyrrole, (62), stirred under the optimized conditions (0.1 eq Dy(OTf)3, MeCN, O-
4"C, 20 hours). Following purification, analysis of the product by lH NMR indicated that
no terminal methyl group of the valeraldehyde was present and thus, it was concluded that
none of the expected cycloadduct was formed. hstead, the product was determined to be the
the same as that of the unidentified product from the three component coupling with
forrnaldehyde. Clearly, the aliphatic aldehydes, fonnaldehyde and valeraldehyde, in both
experiments were not incorporated into the product, otherwise different NMR spectra from
these two different experirnents would have been obtained. A structure, (84), was proposed
through careful analysis of NMR and mass spectra.
11.2.8 N-Acyl Dihydropyrroles as Imine Precursors Following the observations of three component couplings of non-aromatic aldehydes
(section II.2.7), an experiment in which two equivalents of the dihydropyrrole (76) and
aniline were stirred together under the cycloaddition conditions, showed that the same
product as in the reactions with non-aromatic aldehydes, was formed. Mass spectral data
further supported the presence of (84). A mechanism to explain its formation must involve
the ring opening of the N-acyl-dihydropyrrole as in the proposed mechanism depicted in
Figure 11.2.8.1. In this mechanism, ring opening of N-acyl dihydropyrrole, (76), occurs by
the formation of the iminium ion (85), nucleophilic attack by aniline, followed by
deprotonation and ring cleavage as shown in the intermediate (86). After the eventual
formation of the 2-azadiene, "[4 + 21 cyclization" fashions the observed product (84). The
core structure, including the propylamine side chain, of Martinelline type compounds, (22)
and (23), is thus conveniently synthesized in a single step. Using the protected pyrrole, (67), it can be envisaged that the biosynthesis of Martinelhe also follows this mechanism.
The observation of just one spot by TLC was curious, since a pair of diastereomers,
which had been observed in each of the 3 component couplings with benzaldehyde, would
be expected to be distinguishable by TLC. 1H NMR spectra failed to reveal whether a pair
of diastereomers was isolated in the case of (84), due to the complexity of the aromatic
region and the broad polymeric-like peaks scattered throughout. In addition, several
C bz.
C bz.
(84)
Scheme 11.2.8.1
attempts to isolate individual diastereomers by chromatography failed. The product was then
tentatively identified as a single diastereomer.
In any case, the cycloadduct resulting from these conditions smoothly furnishes the
pyrroloquinoline core with the desired substitution of the propylarnine side chain, as in
Martinelline and related compounds. Installation of the guanidine groups with the
accompanying side chahs would complete the construction of this natural product.
11.2.8.1 Optimized Conditions In the interests of synthesizing Martinelline, an aniline functionalized in the 4-
position of the aromatic ring was sought to provide the required functionality of the
pyrroloquinoline core. The aniline, methyl 4-aminobenzoate, was subjected to the three
component coupling conditions for 48 hours, after which a single product, (87), was
isolated (Table 11.2.8.1.1). Unlike the case for (84), comparing the IH NMR spectrum of
(87) with the corresponding benzaldehyde analogues, (82a) and (82 b), the relative
stereochernistry of the pyrroloquinoline core could be assigned. From the similarities of the
splitting patterns of the aniline ring with the erzdo-diastereomer, (82a), (87) was identified
FI 1 lsolated Yield (%)
(84): H 73 (unoptimized)
(87): COOMe 92
Table 11.2.8.1.1: Yields of non-aromatic aldehyde derived cycloadducts.
to be the endo-diastereomer. No splitting pattern for the exo-diastereomer, (82b), was
observed. In addition to these similarities, the difficulty in cleaving the protecting groups of
(84) and (87) (Section II.2.9.2), correlates well with the endo-derivatives, (80a) and
(82a). This further supports the belief that adducts, (84) and (87), are single diastereomers
with endo-type stereochemistry.
11.2.8.2 Proposed Transition State Mode1 Two transition states, proposed in Figure 11.2.8.2.1, attempt to justify the complete
diastereoselectivity found in the reactions with N-acyl dihydropymles as imine precursors.
In transition state (88), which leads to exo-product, there are strong steric interactions
between the aliphatic chain of the diene and the hydrogen atoms located on the pyrrole ring.
In an alternate transition state, (89), these interactions are avoided and the result is the
formation of the endo-product, the sole product of the reaction. These steric considerations
would produce little or no diastereoselectivity in the case of benzilidene aniline since al1
atoms lie in the same plane. This is consistent with the experimentally observed results
tabulated in Figure 11.2.5.1. One additional feature that distinguishes the endo- and exo-
transition states is the similatity of the exo-transition state, (88) with that of the typical,
concerted Diels-Alder reaction. In contrast, the endo-transition state, (89), must involve
rotation about the newly formed bond before cyclization to the aniline ring is possible.
Further studies, especially those involving substituents in the pyrrole ring and the aliphatic
chain, are needed to determine the vaiidity of these proposed transition states.
Cbz.
Figure 11.2.8.2.1 : Proposed transition States for the formation of endo- and exo-products of (84) and (87).
11.2.9 Deprotection of Cycloadducts
11.2.9.1 Removal of Alloc Protected Adducts
Scheme 11.2.9.1.1
Alloc-protected amines are removed by Pd(PPh3)4 in the presence of nucleophilic
species such as carboxylates (Scheme II.2.9.1.1).62 The allyl portion of the carbarnate is cleaved by the formation of a x-allyl paladium complex and subsequent decarboxylation
liberates the nitrogen anion, which is protonated upon workup or by an acidic species in the
reaction.
The Alloc-protected endo-adduct, @la) , was successfully deprotected using
Pd(PPh3)4 (catalytic) and isobutyric acid (stoichiornetric) in CH2C12, albeit in 8% isolated
Despite this result, the deprotections of Alloc-protected adducts were not studied further as
the deprotection of Cbz-protected adducts proved to be cleaner and higher yielding.
However, the deprotection of Alloc-protected adducts via paladium chemistry still remans an
attractive alternative.
11.2.9.2 Removal of Cbz frotected Adducts
Scheme 11.2.9.2.1
Numerous hydrogenolysis conditions were examined for the removal of the Cbz
protected carbamates (Scheme 11.2.9.2.1).62 Standard conditions, such as 10% PWC in
THF or CH2Cl2, H2 balloon, produced inconsistent results and for the most part, showed
no reaction. When the source of palladium was changed to Pd black, the desired deprotected
product, (90b), of the exo-adduct, ($Ob), was isolated in 73% yield, after chromatography.
Under the same conditions, the endo-adduct, (80a), failed to deprotect, as only starting
material was recovered.
As an alternate source of Hz, a 10% aqueous fomiic acid solution together with 10%
Pd/C smoothly deprotected d l , but one of the Cbz protected adducts, to generate (90a),
(90b) and (91b) in moderate to good yield (Scheme II.2.9.2.2a). This one adduct, (82a),
underwent deprotection, but the product, (91a), could not be isolated with sufficient purity.
Interestingly, both endo-products were generated in low yield, which may have been the
result of some cornrnon feature to the stereochemistry of the core.
Using these conditions, the deprotections of adducts, (84) and (87), derived from
two equivalents of protected pyrrole, were atternpted. The adduct, (84), appeared to
undergo complete deprotection according to crude lH NMR, but the product could not be
successfully purified, by column chrornatography or by recrystallization as the HCl salt. Its
methyl ester derivative, (87), was also subjected to the deprotection conditions, but analysis
of the crude lH NMR spectrum clearly indicated that only one of the two Cbz groups had
been cleaved. Moreover, upon close inspection of the aromatic region of the spectrum, it
endeadduct (90a): Ri = H 56% (91a): RI = COOMe <10°/~
Conditions: 10% HCOOH solution,
H . 10% PLVC, THF, 1 -6h
H -R1&Ri H exo-adduct (90b): fll = H 81% (91 b): Ri = COOMe #!A
Scheme 11.2.9.2.2a
HP (1 000 psi) M e 0
Pd/C, EtOH
H H
Scheme 11.2.9.2.2 b
was concluded that it was the primary amine that had been liberated. One final set of
conditions was examined, high pressure H2 (1000 psi) in the presence of 10% PdIC
(Scheme 11.2.9.2.2b). Stirring (87) under H2 for several days, the resulting diamine, (92),
was isolated as the HCl salt, providing the deprotected product, contarninated however with
inseparable impurities (possibly a mixture of the dihydrochloride and monohydrochloride
salts). This prelirninary result rnust be optimized.
11.2.10 Comparison of Stereochemistry with Martinelline
The relative stereochemistry of the pyrroloquinoline core in Martinelline was
determined by ID and 2D NMR experiments.22 The most compelling evidence was
provided by NOE experiments. A strong NOE observed between H9, and Hia, indicated
that Hgb and the side chain are on the same face of the piperidine ring. The absence of a
because of the Karplus angle betwen these protons.
The cycloadducts, (84) and (87), derived from two equivalents of dihydropyrrole,
do not appear to be consistent with the stereochemistry reported for (22) and (23). To
generate this stereochernistry, a forma1 exo-addition must occur in our [4 + 21 cycloaddition. This is not the case for our observed endo-products, (84) and (87). Formation of
Martinelline then, appears to be impossible under Our conditions, since the wrong
diastereomer is formed.
The belief that the absence of correlation between H3, and H4 in (93) would arise
from anti-stereochemistry seems plausible at first glance. However, when a mode1 is
constructed, these protons, with syn-orientation, have a dihedral angle close to 60° and
minimal correlation between these protons would be expected. When these protons have an
anti-relationship, the dihedral angle approaches 180" and accordingly, a large correlation
would be expected in the COSY spectmm. This rationalization was fully supported by the
examination of COSY spectra of deprotected cycIoadducts, (90a) and (90b), (Figure
11.2.10.1). Not surprisingly, the COSY spectrum of (90a), the apparent endo-adduct,
actually displays a srnaller correlation between H3, and H4 than that of the exo-adduct,
(90b). Thus, the absence of this correlation is not a suitable determinant of an anti-
relationship between H3, and H4.
Clearly, NMR studies are needed to shed light on the stereochemical relationship
about the piperidine ring in (92) and (93). It is conceivable that the stereochemistry of
Martinelline was incorrectly assigned, but in order to determine this, cornparisons between
the NMR data (especially NOE's) of (93) and the deprotected adduct (92) are needed. If
this is not the case, then Our approach may not be appropriate in synthesizing Martinelline.
Instead, modifications to the direct three component coupling strategy will be required.
Figure 11.2.1 0.1 : COSY spectra of (90a) and (90b).
A three component coupling strategy fonning a pyrroloquinoline skeleton rnay be a
viable and convenient route to the total synthesis of Martinelline and related compounds, (22) and (23). In this approach, an aldehyde and an aniline condense to generate a 2- azadiene which undergoes a formal Diels-Alder transformation with an enamine. The
optimal conditions for this reaction involves the Lewis acidic lanthanide (III) triflate and the
solvent, acetonitrile. For the mode1 study with benzaldehyde, the Lewis acid that provided
the greatest yield of cycloadduct was Dy(OTf)3; no diastereoselectivity was observed. Non-
aromatic aldehydes, such as ethyl glyoxylate and valeroaldehyde, did not form the desired
adduct. Three component coupling conditions utilizing two equivalents of Cbz-protected
dihydropyrrole, generated in excellent yield cycloadducts as single diastereomers with the
desired functionality as in Martinelline, including the C4-propylamine side chain. These
diastereomers appear to be the products of endo-addition. Removal of protecting groups
from cycioadducts proved to be non-trivial, especially in case of the endo-adducts. For the
adducts containing the Cq-propylamine side chain, prelirninary results show that both Cbz
protecting groups can be removed, but only under forcing conditions.
There is a possibility that the stereochemistry of Martinelline was incorrectly assigned
and that the true diastereomer is the al1 syn-adduct. Further NMR analysis is necessary,
however, to confirrn this suspicion. If the stereochemistry is indeed incorrect, then our three
component coupling strategy generates the desired diastereomer. If, on the other hand, our
suspicion proves to be false, other means to install the C4-propylamine side chain with the
desired stereochemistry must be investigated.
Chapter III:
Experimental
Al1 1H and 1 3 ~ NMR spectra were obtained on a 400 1 100 MHz or 500 1125 MHz Varian Unity spectrometer in CDCl3 (referenced at 6 7.24 and 77.00 ppm for 'H and 13c, respectively), C6D6 (referenced at 6 7.15 and 127.00 ppm for I H and 13c, respectively) or
D20 (referenced at 6 4.63 ppm for 1~ NMR). Features of peaks in the 1H NMR spectra are
labelled in brackets proceeding each chernical shift in the following order: integration,
multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, p = pentet, m = complex multiplet), coupling constant, identity of proton (if possible). FT-IR spectra were recorded
on a Perkin-Elmer Spectrum 1000 or a BOMEM-Hartmann and Braun, Michelson Series spectrometer, with samples loaded as films on NaCl plates or as pellets of KBr. Low
resolution mass spectra were recorded on a Bell and Howell2 1-490 spectrometer, and high resolution spectra were recorded on an AEI MS3074 spectrometer.
Reaction solvents were distilled pnor to use, under inert atmosphere unless otherwise
stated. Diethyl ether (ether), THF, benzene and toluene were distilled from sodium wire 1
benzophenone. Dichloromethane and DMF were distilIed from CaH2. Acetonitrile was
dned over 4 A molecular sieves overnight and stored under Nz. Al1 reagents, unless
otherwise noted, were purchased from Aldrich Chernical Company, Fisher Scientific Limited
or BDH.
Flash column chromatography on silica gel (60 A, 230-400 rnesh, obtained from
Whatman Company or Toronto Research Chernicals, Inc.) was perforrned with distilled
hexanes, distilled ethyl acetate, reagent grade dichloromethane and spectrograde acetone
(Caledon). Analytical thn-layer chromatography (TLC), was performed on pre-coated silica
gel plates, (Alugram SIL GNV2s4 purchased from Rose Scientific Limited), visualized with
a UV254 larnp (Spectroline, Longlife Filter) and stained with 20% phosphomolybdic acid in
ethanol (commercially available), ceric molybdate (17.3 g MoO3, 14.4 mL concentrated
NH40H in 48 rnL H20, slowly added to 7.6 g of (NH4)2Ce(S04)4.2H20, in 100 mL of
50% H2S04, then diluted to 500 mL with H20) or iodine. Solvent systems associated with
Rf values and chromatography are reported as volumetric ratios.
To sodium hydride (2.27 g, 60% oil dispersion, 56 mmol), washed free of oil with
hexanes, was added THF (30 mL). Cooling to O°C, trans,trans-2,4-hexadien-1-01 (5.0 g,
5 1 rnmol) was added slowly, forming a rnirky brown mixture which was allowed to warm to
room temperature. Propargyl bromide (6.3 mL, 80% in toluene, 74 mmol) was added
slowly at 0°C. After warming to room temperature and observing the disappearance of
starting material by TLC, distilled water (20 mL) was added to quench the reaction. The reaction mixture was extracted with ether (3 x 25 mL) and the combined organic layers were
washed with water (50 rnL) and brine (50 mL). After removal of solvent, in vacuo, vacuum
distillation (0.15 rnmHg, bp 4 1 4 ° C ) and further purification by flash chromatography (O to
20% EtOAchexanes), the title compound was isolated as a colourless oil(4.70 g, 68%). Rf = 0.10 (10% EtOAdhexanes); IR (neat) u 3300,3022.2935,29 15,2853, 1662, 144 1,
1357, 1266, 108 1, 988, 926 cm-'; lH NMR (200 MHz, CDC13) 6 6.10 (2H, m), 5.62
(2H, m), 4.08 (2H, d, J = 2.4 Hz, 2 x H7), 4.03 (2H, d, J = 6.6 Hz, 2 x Hg), 2.40 (lH,
m, Hg) 1.7 1 (3H, d, J = 7.0 Hz, 3 x H 1); MS (EI) m/e (rel intensity) 136 (2 1, M-H+), 12 1
(1 l), 106 (60), 91 (lOO), 81 (391, 79 (48), 77 (33), 69 (99), 67 (32), 65 (28), 73 (13), 55 (13), 53 (39).
Lithium chloride (1.52 g, 52 rnmol), dried by heating under vacuum for several
hours, was dissolved in DMF (20 mL). To this solution was added a solution of trans,trans-
2,4-hexadien- 1-01 (2.0 g, 20 rnmol) in 2,6-lutidine (4.8 mL, 41 rnmol). After stining for 20
minutes at O°C, a white slurry formed and methanesulfonyl chloride, (2.4 mL, 17 mmol),
was added. The reaction mixture was stirred overnight, allowing to warm to room
temperature. Distilled water (50 r d ) was added to quench the reaction and the mixture was
extracted with ether (5 x 25 mL). The combined organic layers were washed sequentially
with saturated CuS04 solution (2 x 25 mL), (to remove traces of lutidine), water, saturated
bicarbonate solution, water and then brine. After drying with anhydrous MgS04, the
solvent was removed in vacuo, producing a yellow liquid (2.31 g), which was used
unpurified to alkylate diethyl malonate.
2-(tram, trans-Hexa-2,4-dieny1)-malonic acid diethyl ester, (54):
To a solution of sodium hydride (3.67 g, 60% dispersion in oil, 92 mrnol), washed
free of oil with hexanes, in THF (40 mL), was added diethyl malonate (15.5 mi,, t 11
rnmol), at -15°C. After 1 hour of stirring, the resultant pale brown solution was transferred
via cannula to a flask, cooled to -15OC, containing the crude chloride, (53), (2.31 g, 20
mmol) and THF (60 rnL). This reaction was stirred for 1 hour before warming to room
temperature and stirring for 2 additional hours. Water (50 mL) was added to quench the
reaction and the aqueous layer was extracted with ether (3 x 30 mL). The organic layers
were combined and then washed with saturated bicarbonate solution, water and brine.
Concentration in vncuo followed by vacuum distillation (8 rnmHg, bp 1 17-122°C) and flash
chromatography afforded the product as a colourless oil ( 1.17 g, 24% from tram, trans-2,4-
hexandien- 1-01).
6.00 (2H, m), 5.35-5.66 (2H, rn), 4.15 (4H, q, COOCH_2CH3), 3.34 (lH, t, H7), 2.60 (2H, t, 2 x Hi), 1.69 (3H, d, 3 x H6), 1.22 (6H, t, COOCH2CI&); 13C NMR (50 MHz, CDC13) 8 168.86 (COOCH2CH3), 133.18, 130.99, 128.56, 126.09, 61.29, 52.06, 31.77,
17.95, 14.03 (COOCHgH3).
2-(trans,trans-Hexa-2,4-dienyl)-2-(prop-2-ynyl)-malonic acid diethyl ester,
(52):
To a flask cooled to O°C, containing a solution of sodium hydride (0.33 g, 60%
dispersion in oil, 8.3 mmol), washed free of oil with hexanes, in THF (12 rnL) was added a
solution of the mono-alkylated malonate, (48), (1.17 g, 4.9 rnmol) in THF (5 mL). After
stirring for 2 hours at room temperature, a orangehrown mixture forrned. Propargyl
brornide (80% in toluene, 17 rnmol) was added at 0°C and the reaction was stirred ovemight
at room temperature. The reaction was then quenched with water and after aqueous workup,
and solvent removal in vacuo, a pale brown liquid was isolated. Flash chromatography
(10% EtOAchexanes) provided the title compound as a pale yellow oil(0.97, 72%). Rf = 0.62 (30% EtOAchexanes); 1H NMR (200 MHz, CDC13) 8 6.02 (2H, m), 5.61 (IH,
m), 5.30 (lH, m), 4.19 (4H, q, J = 7.2 Hz, COOC&CH3), 2.78 (4H, m, 2 x H i , 2 x Hg),
2.00 ( lH, rn, Hlo), 1.69 (3H, d, J = 7.0 Hz, 3 x Hg), 1.22 (6H, t, J = 7.1 Hz, COOCH2CH3); 13C NMR (50 MHz, CDC13) 6 169.67 (COOCH2CH3), 135.06, 13 1 .OS,
128.89, 123.54, 78.94, 71.30, 61.55, 56.89, 35.16, 22.54, 17.95, 14.02.
General Procedure for the Construction of Bicyclo[4.3.0]alkenoIs, (56)-(58):
Borane dimethyl sulfide cornplex, BHySMe2 (2.0 M solution in THF, 2.0 rnmol),
was transferred to a flame dried round bottom flask equipped with a magnetic stirrer, reflux condenser and N2 inlet. Cyclohexene (405 PL, 4.00 rnrnol) was added through the top of
as a white precipitate. After 2 h of stirring at room temperature, the dieneyne (1.9 mrnol)
was added dropwise through the top of the condenser, causing the precipitate to dissolve.
After stirring for an additional 2 h, butylated hydroxytoluene (BHT; 22 mg, 0.1 mmol) in
benzene (50 mL) was added to the reaction and the reaction mixture was heated at reflux for
17-20 h. After cooling to room temperature, trimethylamine-N-oxide (0.72 g, 6.5 mrnol)
was added and the reaction was reheated to reflux for 24 h. Upon cooling to room
temperature, distilled water (20 m.) was added and the reaction mixture was heated to 60°C
for 30 min to ensure cornpiete hydrolysis. The aqueous layer was extracted with ether (4 x
10 r d ) , and the combined organic layers were washed with water (20 rnL) and brine (20
mL), dried with anhydrous MgS04 and concentrated in vacuo. Purification by flash column
chromatography (silica, EtOAchexanes) afforded the desired bicyclo[4.3.0]alkenoIs.
Obtained in 70% yield; colourless oil; Rf = 0.10 (10% EtOAchexanes); IR (neat) u 3418,
3021, 2933, 2869, 1632, 1455, 1392, 1372, 1325, 1161, 1117, 1102, 1070, 1040, 1017, 98 1, 960, 88 1, 806, 729, 68 1 cm- l ; lH NMR (400 MHz, CDCl3) 6 5.63 (2H, m, H6, H7).
4.06 (3H, m, Hl, H3, H4), 3.58 ( lH, dd, J = 11.0, 7.4 HZ, H39, 3.38 (lH, dd, J = 11.7,
7.0 Hz, Hy), 2.67 (lH, m, Hs), 2.48 (lH, m, H7,), 2.06 (IH, dq, J = 11.2, 6.9 HZ, H3,), 1.71 (lH, s, OH), 1.04 (3H, d, J = 6.9 Hz, 3 x Hg); 13C NMR (100 MHz, CDC13) S 134.94, 122.77, 72.59, 70.64 (Ci), 70.00 (C3), 45.22, 45.09, 37.55 (Cs), 14.53 (Cs); MS
(EI) d e (rel intensity) 153 (42, M-H+), 139 (37), 123 (42), 109 (40), 93 (54), 81 (100), 67
(49), 57 (78); HRMS (EI) rde calcd (M-H+) 153.09 15, found 153.09 1 1.
Obtained in 35 % yield; colourless oil; Rf= 0.29 (60 % EtOAchexanes); IR (neat) u 3409,
2951, 2942, 2864, 1663, 1463, 1364, 1191, 1091, 1038, 1018, 892 cm-l; I H NMR (400
MHz, CDC13) 6 5.20 (lH, m, Hs), 4.20 (lH, ddd, J = 12.7, 9.0, 4.3 Hz, Hg,), 4.10
(2H, m, H2, H3), 3.51 (lH, dd, J = 11.0, 7.7 Hz, Hz'), 2.71 ( lH, m, H4), 2.44 (IH, rn,
Hgb), 2.02 (2H, m, H2,, Hql), 1.84 (2H, s, 2 x Hg), 1.53 (lH, ddd, J = 12.7, 4.5, 1.1 HZ, Hg), 0.99 (4H, m, Hg., 3 x Hg), 0.89 (3H, s, 3 x Hia); l 3 ~ NMR (100 MHz, CDC13) 6
136.85 (Cs,), 119.54 (Cs), 74.90 (Cg,), 70.70 (C2), 70.38 (Cg), 48.82 (C2& 45.97 (Csb),
43.45 (Cs) , 43.19 (C6), 38.30 (C4), 32.14, 3 1.82, 30.72; MS (EI) m/e (rei intensity) 208 (54, M+), 177 (40), 152 (100), 135 (47); HRMS, d e calcd (M+) 208.1463, found
208.1467.
(3aR*,4S*,5S*,7aR*)-7-Hydro~y-6-methyl-l~3~3a,6~7~7a-hexahydro-
indene-2,S-dicarboxylic acid diethyl ester, (58):
% COOEt
- COOEt 8 ~ c o o i t 1 COOEt
Obtained in 56% yield; colourless oil, Rf= 0.50 (40% EtOAchexanes); IR (neat) u 3453,
2977, 2931, 1726, 1452, 1368, 1262, 1174, 1106, 1046, 960, 858, 810, 729 cm-1; I H
NMR (400 MHz, CaD6) 6 5.53 (lH, m, Hs), 5.41 ( lH, m, H4), 3.98 (4H, m, CH2CH3),
3.69 (lH, dd, J = 11.0, 6.6 Hz, H3), 2.99 (IH, dd, J = 12.6, 6.5 Hz, Hl), 2.82 ( l H , dd, J
= 12.9, 6.6 Hz, Hy), 2.40 ( lH, m, H6), 2.20 (lH, m, H7,), 2.00 (lH, t, J = 12.0 Hz,
Hll), 1.89 (lH, t, J = 12.7 Hz, Hg), 1.87 (IH, m, H3,), 0.95 (3H, d, J = 7.3 Hz, 3 x Hg),
0.94 (3H, t, J = 6.6 Hz, CH2Cj&), 0.89 (3H, t, J = 7.2 Hz, CHIC&); 1 3 ~ NMR (100 MHz, C6D6) 8 134.33 (COOEt), 126.59 (COOEt), 74.22 (C7), 61.42 (CHZCH~), 61.41
(CH2CH3), 59-19 (C2), 45.55 (C3,), 45.38 ( C T ~ ) , 39.1 1 (Cg), 37.84 (C3), 37.55 (CI),
(32), 173 (56), 149 (27), 131 (75), 91 (27), 81 (26); HRMS (EI) d e calcd (M+) 296.1624, found 296.1636.
To a 5 M solution of pinacolborane (440 PL, 2.2 rnmol)$g was added dienyne (50)
(128 mg, 0.942 mmol) at O"C under a nitrogen atmosphere. This solution was then
transferred via cannula to a flask, cooled to O°C, containing Schwartz' Reagent,
HZrCp2C1>0 (16 mg, 0.062 mrnol). The reaction mixture was wrapped in tin foi1 to protect
from light, and allowed to warm to ambient temperature ovemight. After stirring for an additional day, the solvent was removed in vacuo. Purification by colurnn chromatography
(2-10 % EtOAchexanes), generated the title cornpound (46 mg, 19%)) as a colourless oil. Rf = 0.44 (20 % EtOAchexanes); IR (film) II 2977,293 1, 1725, 1695, 1645, 1446, 1349,
1144, 971, 849 cm-l; 1H NMR (CDCL3) 6 6.98 (lH, dt, J = 18.0, 4.5 Hz, Hl), 6.14 (2H,
m, H2,Hs), 5.94 (lH, m, H6), 5.50 (2H, m, H7, Hg), 3.86 (2H, dd, J = 1.8, 5.4 Hz, 2 x
H3), 3.81 (ZH, d, J = 5.8 Hz, 2 x H4), 1.53 (3H, d, J = 6.6 Hz, 3 x Hg), 1.08 (12H, s, 3 x
Hi2, 3 x Hi3, 3 x H14, 3 x H15); 13C NMR (100 MHz, Cg&) 8 150.41 (Cl), 132.68,
131.62, 129.1 1, 127.76, 127.59, 83.1 1, 71.62, 70.79, 24.98, 18.15; MS (EI) rn/e (rel
intensity) 249 (16), 194, (9), 149 (10)- 97 (53 , 81 (83), 69 (100); HRMS (EI) m/e calcd
(M') 264.1897, found 264.1899.
hexahydro-isobenzofuran-4-y1 ester, (64):
To the alcohol, (56), (53 mg, 0'.34 mmol) in CH2Cl2 (3.0 mL) was added
dicyclohexylcarbodiimide (DCC; 1 12 mg, 0.54 mmol), dimethylaminopyridine (DMAP; 12
mg, 0.098 mmol) and p-nitrobenzoic acid (80 mg, 0.48 mmol). Stirring overnight gave a yellow opaque mixture to which Hz0 (5 d) was added. The aqueous layer was extracted
with ether (3 x 10 mL) and the combined organic layers washed with brine (25 mL). The
organic layer was dried with anhydrous MgS04, filtered and concentrated. Purification by
flash chromatography (10 % to 20 % EtOAchexanes) gave the p-nitrobenzoate ester as a
pale yellow solid (85 mg, 82 96). The solid was recrystallized in a mixture of ethedpentane
(60/40). Large pale yellow crystals grew slowly over several days in an erlenmeyer flask
covered with a needle pierced septum. mp = 120-122 OC; for X-ray data, see Appendix A; Rf = 0.23 (20 % EtOAchexanes); IR (KBr pellet) u 2972,2867, 1727, 1610, 1532, 1350, 1276, 1109, 1023,880,742, 718 cm-
1 ; I H NMR (400 MHz, CDC13) 6 8.29 (2H, m, H12, 8.18 (2H, m. Hll , HI^), 5.76
(lH, m, Hg), 5.64 (IH, m, H7), 5.38 (lH, dd, J = 11.8, 7.0 Hz, H4), 4.11 ( lH, t, J = 7.3
HZ, H3), 4.04 ( lH, t, J = 7.4 HZ, Hi), 3.60 (lH, dd, J = 11.0, 7.3 Hz, H3*), 3.46 ( lH, dd, J = 11.4, 7.0 Hz, Hl#), 3.06 (lH, m, Hs), 2.68 (lH, m, H7,), 2.39 (lH, dq, J = 11.2, 6.8 Hz, H3,), 1.06 (3H, d, J = 7.0 Hz, 3 x Ha); 13C NMR (100 MHz, CDC13) 6 163.9 1
(Cl3), 150.55, 135.26, 134.08, 130.62, 123.59, 123.96, 76.59, 70.51 (Cl), 69.54 (C3), 44.74, 43.20, 35.08 (C5), 15.46 (Cs); MS (EI) d e (rel intensity) 150 (100), 136 (38), 106
(47), 104 (42), 9 1 (34); HRMS (EI) d e calcd (M+) 303.1 107, found 303.1101.
A solution of unpurified 1-pyrroline trimer (5.0 g, 24 mrno1)57 in THF (250 mL)
was azeotropically distilled under atmospheric pressure into a flask chilled to -78OC.58
Triethylamine (80 mmol) and allyl chloroformate (75 mmol) were added to the reaction.
After warming to room temperature overnight, the reaction mixture was filtered through a
sintered funnel and concentrated in vacuo. Vacuum distillation (0.75 mrnHg, bp 68-7 1 OC)
using a Vigreux colurnn afforded the title compound (3.0 g, 20 mmol), in a yield of 27 %,
based on the trimer. Al1 NMR spectra were consistent with the reported values?
General Procedure for the Construction of 4-Phenyl- 2,3,3a,4,5,9b-hexahydropyrrolo[3,2-c]quinolnes, (80)-(82):
Syntheses of these compounds were performed in non-flame dried glassware, but
under a nitrogen atmosphere. Al1 additions were done at 0 " ~ . To a solution of benzaldehyde (100 PL, 0.98 rnmol) in MeCN (0.5 rnL) was added the aniline (0.98 mrnol). After rnixing
for 2 minutes, the solution was transferred via cannula to a flask containing dysprosium (III)
triflate, Dy(OTf)3 (32 mg, 0.052 mmol) suspension in MeCN (2 rd). Additional MeCN
(0.5 mL) was added to complete the transfer. A solution of enamine (0.705 mmol) in MeCN
(0.5 rnL) was then transferred via cannula into the above mixture with an additional portion
of MeCN (0.5 mL) to cornplete the transfer. Once outfitted with a N2 balloon, the reaction
was stirred for 20 h in a cold room ( 4 " ~ ) . Distilled water (5 rnL) was added to the reaction
and the aqueous layer was extracted with CH2C12 (4 x 5 mL). The combined organic layers
were dried with anhydrous MgS04 and concentrated in vacuo leaving a colourless or pale
yellow viscous oil. Purification by flash chromatography (siiica, EtOAchexanes) provided
two diastereorners of the desired pyrroloquinolines whose ratio was determined by IH NMR. Complete separation of the diastereomers was achieved using a gradua1 gradient of
ethyl acetate/hexanes.
benzyl ester, (80a) and (80b):
Obtained a 5 1 :49 mixture of diastereomers (80a) and (80b) in combined yield of 9 1 %.
Initially, both diastereomers were isolated as colourless materials, but upon exposure to
normal lighting, discolouration to brown results.
(3aS*,4S*,9bS*)-, (80a): al1 syn diastereomer White solid; mp 149- 15OOC (EtOAchexanes); Rf = 0.47 (30% EtOAchexanes); IR (film) u
3343, 3031, 2953, 1693, 1606, 1586, 1483, 1452, 1412, 1358, 1327, 1301, 1263, 1210,
1175, 1105, 1085, 1029, 910, 754, 700, 666 cm-]; 1H NMR (500 MHz, CDC13), rotomers, S 7.70 (0.6H, d, J = 7.8 Hz, Hg), 7.28-7.48 (10H, COOC&C&, Ph), 7.15
(O-lH, t, J = 7.6 Hz), 7.06 (lH, m), 6.78 (0.6H, t, J = 7.1 Hz), 6.67 (0.4H, t, J = 7.8
Hz), 6.58 (IH, t, J = 7.2 Hz, Hg), 5.46 (0.6H, d, J = 7.3 Hz), 5.38 (0.4H, d, J = 6.9 Hz),
5.16-5.32 (lH, m), 5.19 (lH, d, J = 3.0 Hz), 4.71 (IH, d, J = 2.2 Hz), 3.92 (0.3H, s,
Hs), 3.90 (0.7H, s , Hs), 3.40 (2H, m, 2 x Hz), 2.55 ( lH , m, H3,), 2.40 (lH, m, H3), 1.55 (lH, m, H39; 1 3 ~ NMR (125 MHz, CDCl3), rotomers, 6 156.58, 155.39, 143.67,
141.74, 141.64, 137.00, 136.65, 130.59, 129.89, 129.24, 128.62, 128.52, 128.41,
128.19, 128.03, 127.82, 127.75, 127.68, 126.54, 122.42, 121.14, 119.07, 118.70,
115.06, 114.75, 68.13, 66.83, 57.00, 56.71, 56.53, 56.32, 45.05, 44.88, 44.43, 22.74,
21.81; MS (EI) m/e 384 (44, Mf), 293 (37), 249 (100), 220 (23), 206 (44), 91 (87);
HRMS (EI) d e calcd 384.1838, found 384.1827.
(3aS*,4R*,9bS*)-, (80b): anti diastereomer Colourless oil; Rf = 0.37 (30% EtOAcIhexanes); IR (film) u 3363, 3029, 2951, 2890,
1691, 1607, 1497, 1451, 1414, 1359, 1320, 1299, 1262, 1212, 1157, 1108, 1080, 1067, 1029, 1002, 91 1, 750, 699, 607 cm-l; l H NMR (500 MHz, CDCl3), rotorners, 6 7.56
(OSH, m, Hg), 7.20-7.40 (10.5H, m, COOCH2C6&, Ph, Hg), 7.07 (lH, t, J = 7.2 Hz,
H g ) , 6.69 (lH, m, H7), 6.57 (lH, dd, J = 8.0, 1.0 Hz, Hg), 5.06-5.34 (2H, m, COOC&c6H5), 4.92 (lH, m), 4.37 (lH, d, J = 3.2 Hz), 4.30 (IH, s, Hg), 3.59 (lH, m,
H2), 3.44 ( IH, m, H29, 2.60 (1 H, m, H3,), 2.10 (ZH, m, 2 x H3); 3~ NMR (500 MHz, CDC13), rotomers, 6 144.78, 142.59, 136.85, 130.03, 128.70, 128.39, 128.25, 127.86,
28.1 1; MS (El) m/e 384 (25, M+), 293 (30), 249 (44), 218 (32) , 2U6 (IUU), Y 1 (./3);
HRMS (EI) d e calcd 384.1838, found 384.1847.
4-Phenyl-2,3,3a,4,5,9b-hexahydropyrrolo[3,2-c]quinoline-l-carboxylic acid
allyl ester, @la) and (81b):
Alloc
Obtained a 50:SO mixture of diastereomers (81a) and (81b) in combined yield of 92%.
(3aS*,4S*,9bS*)-, (81a): al1 syn diastereomer
White solid; mp 169-170"~ (EtOAdhexanes); Rf = 0.45 (30% EtOAchexanes); IR (KBr pellet) u 3321, 2968, 2910, 1666, 1609, 1489, 1444, 1407, 1332, 1263, 1104, 996, 938,
768, 75 1, 701 cm-1; IH NMR (500 MHz, CDC13), rotomers, 6 7.62 (0.5H, d, J = 7.8 Hz,
Hg), 7.35-7.43 (5.5H, m, Hg, Ph), 7.31 (2H, t, J=7 .6Hz , Ph), 7.25 (lH, t, J = 7 . 3 H z ,
Ph), 7.00 (lH, t, J = 7.6 Hz, Hs), 6.69 (lH, m, H7), 6.51 (lH, d, J = 8.3 HZ, Hg), 5.93
(lH, m, COOCH2CH=CH2), 5.1 1-5.40 (3H, m, Hgb, COOCH2CH=C&), 4.66 (2H, d, J
= 6.3 HZ, COOC&CH=CH2), 4.58 (lH, d, J = 2.9, H4), 3.86 (lH, d, J = 7.6 Hz, Hs),
3.32 (2H, m, H2, H29, 2.49 (lH, m, H3b), 2.13 (lH, m, H3), 1.50 (lH, m, H39; 13C NMR (125 MHz, CDCl3) rotomers, 8 156.44, 143.62, 141.72, 133.18, 132.98, 130.57,
129.88, 128.62, 128.12, 128.02, 127.76, 126.56, 122.41, 119.01, 118.71, 117.69,
117.05, 114.74, 66.02, 65.78, 58.92, 56.90, 56.68, 56.49, 56.31, 45.07, 44.98, 44.86,
44.39, 22.75, 21.79; MS (EI) m/e 334 (50, M+), 249 (100), 220 (29), 206 (56), 130 (22),
9 1 (33); HRMS (EI) d e calcd (M+) 334.168 1, found 334.1696.
(3aS*,4R*,9bS*)-, (81b): anti diastereomer Colourless oil; Rj = 0.36 (30% EtOAcIhexanes); IR (film) u 3362, 3057, 3025, 2743,
2890, 1690, 1607, 1497, 1407, 1320, 1299, 1263, 1 109, 995, 934, 750, 701 cm-]; 1H NMR (400 MHz, CDCI3) 6. 7.18-7.58 (6H, m, Ph, Hg), 7.07 ( 1 H, dt, J = 8.0, 1 .O Hz,
Hg), 6.70 (IH, dt, J = 7.3, 1.2 Hz, H7), 6.37 (lH, dd, J = 8.0, 1.1 Hz, H6), 5.87 (lH, m, COOCH2CN=CH2), 5.30 (lH, dd, J = 17.2, 1.8 Hz, COOCH2CH=CI-&), 5.19 (IH, dd, J
= 10.5, 1.3 Hz, COOCH2CH=CI&), 4.88 (lH, m), 4.62 (2H, rn), 4.30 (lH, d, J = 3.3
m, 2 x H3); 13C NMR (125 MHz, CDC13) 6 144.82, 142.55, 133.12, 130.07, 128.70,
128.25, 127.26, 125.80, 120.76, 117.84, 117.31, 113.51, 65.83, 55.83, 53.10, 44.77,
44.00; MS (EI) d e 334 (30, Mf), 249 (47), 220 (27), 206 (100), 91 (26); HRMS (EI) m/e
(M+) calcd 334.168 1, found 334.1684.
acid 1-benzyl ester 8-methyl ester, (82a) and (82b):
Cbz. "
The two diastereomers, (82a) and (82b) were obtained in a combined yield of 61%
and in a 5050 ratio. As a side product, irnine, (82c), represented another 14%.
(3aS*,4S*,9bS*)-, (82a): al1 syn diastereomer
Colourless crystals; mp = 178- 18 1 " ~ (30% EtOAchexanes); for X-ray data, see Appendix A; Rf = 0.37 (5% EtOAdCH2C12); IR (film) u 3354, 2949, 1700, 1609, 1251, 1495,
1417, 1359, 1301, 1195, 1108, 990, 769, 699 cm-1; lH NMR (500 MHz, CDC13), rotomers, 8 8.31 and 8.21 ( lH, s, Hg), 7.74 (lH, d, J = 8.3 Hz, H7), 7.50 ( lH, d, J = 7.4
Hz) 7.26-7.43 (9H, m, COOCH2C6&, Ph) 6.54 (lH, dd, J = 8.3, 3.2 Hz, Hg), 5.43
(0.5H, d, J = 7.0 HZ), 5.18-5.34 (2.5H, m, COOC&C6H5), 4.77 (lH, dd, J = 1 1.2, 2.6
Hz), 4.33 (lH, d, J = 12.7 Hz), 3.82 (3H, d, J = 8.3 Hz, COOC&), 3.28-3.48 (2H, m, 2
x H2), 2.54 (lH, rn, H3a), 2.06 (lH, h, J = 10.8 HZ, H3), 1.56 (lH, h, J = 6.6 HZ, H31); I3c NMR (125 MHz, CDC13) rotomers, 6 167.14, 166.95, 156.57, 155.53, 147.54,
147.46, 140.92, 140.77, 137.02, 132,59, 132.27, 130.06, 128.80, 128.65, 128.52,
128.43, 128.35, 128.12, 128.06, 127.86, 127.80, 126.44, 121.21, 120.49, 120.21,
114.13, 67.59, 67.01, 56.55, 56.32, 56.12, 56.05, 51.62, 51.55, 45.12, 44.94, 44.41,
43.88, 23.13, 22.19; MS (EI) rn/e 442 (20, M+), 351 (47), 307 (64), 265 (28), 264 (100),
9 1 (55); HRMS (EI) d e (Mf) cakd 442.1893, found 442.188 1.
Colourless oil; Rf = 0.10 (1% EtOAdCH2C12); IR (film) u 3342, 3032,2950, 1700, 1610,
1514, 1415, 1358, 1301, 1241, 1206, 1101,994,911,770,734, 702 cm-'; 1H NMR (500 MHz, CDC13, 40°C), rotomers, 8 8.3 1 and 8.2 1 (lH, s, Hg), 7.74 ( IH, d, J = 8.3 Hz, H7),
7.50 (lH, d, J = 7.4 Hz), 7.23-7.41 (10H, rn, COOCH2C6&, Ph), 6.55 and 6.54 (lH, d,
J = 8.3 HZ, Hg), 5.43 (OSH, d, J = 7.0Hz), 5.33 (OSW, d, J = 7.1 Hz), 5.16-5.30 (3H,
m, Hgb, COOC&C6H5), 4.77 (lH, dd, J = 11.2, 2.5 Hz), 4.34 and 4.31 (lH, s, Hs), 3.82 and 3.81 (3H, s, Me), 3.29-3.49 (2H, m, 2 x H2), 2.53 (lH, m, H3,), 2.06 (lH, m, H3), 1.56 (LH, m, H31); 1 3 ~ NMR (125 MHz, CDCl3) rotomers, 6 167.14, 166.95,
156.57, 155.53, 147.54, 147.46, 146.86, 140.92, 140.77, 137.02, 132.59, 132.27,
130.06, 128.80, 128.65, 128.52, 128.43, 128.35, 128.12, 128.06, 127.86, 127.80,
126.44, 121.21, 120.49, 120.21, 114.13, 67.59, 67.01, 56.55, 56.32, 56.12, 56.05,
5 1.62, 5 1.55, 45.12, 44.94, 44.41, 43.88, 23.13, 22.19; MS (EI) m/e 442 (25, M+), 308
(32), 307 (100), 276 (20), 264 (22), 91 (60); HRMS (EI) d e (M+) calcd 442.1893, found
442.1 894.
5-f2-benzylideneamino-5-methoxycarbonyl-phenyl)-2,3-dihydropyrro~e-l-
carboxylic acid benzyl ester, (82c): Colourless oil; Rf = 0.09 (5% EtOAdCH2Cl2); IR (film) u 3345, 3032,2951, 172 1, 1625,
1531, 1447, 1260, 1188, 1100, 1012,924, 847, 773, 755, 733, 701 cm-1; 1H NMR (500 MHz, CDC13), 6 8.53 (lH, s, H12), 8.25 (lH, dd, J = 8.8, 1.7 Hz), 8.10-8.16 (2H, m),
7.40-7.54 (5H, m), 7.22-7.33 (5H, m), 5.00 (2H, s, COOCH2C6H5), 4.5 1 (lH, rn, H4), 3.98 (3H, s, COOMe), 3.3 1 (2H, m, 2 x H2), 3.03 (2H, rn, 2 x H3); l3C NMR (1 25 MHz, CDC13), 6 166.53, 162.58, 156.06, 148.31, 139.91, 137.89, 136.33,. 130.96, 130.15,
129.42, 128.78, 128.59, 128.54, 128.48, 128.41, 128.11, 128.05, 127.95, 126.44,
66.61, 52.28, 40.95, 33.14; MS (EI) / d e 440 (5, M+), 349 (21), 276 (100), 275 (18), 232
(181, 217 (32), 216 (31), 91 (64), 83 (19); HRMS (EI) m/e (M+) calcd 440.1736, found
440.1723.
General Procedure for the Construction of 4-(3-Aminopropy1)- 2,3,3a,4,5,9b-hexahydropyrrolo[3,2-c]quinolnes, (84) and (87):
To a suspension of Dy(OTf)3 (0.240 mmol) in MeCN (5 mL), pre-cooIed in an ice
water bath, was added via cannula, a solution of the aniline (1.01 mmol) in MeCN (5 mL),
with additional MeCN (5 rnL) to complete the transfer. After stirring for 2 minutes, a
to the reaction mixture, with an additional portion of MeCN (5 mL) . The reaction was then
stored in a cold room (0-40C) and was allowed to stir for 2 days. Distilled water (20 rnL) was added to the reaction. The mixture was extracted with CH2C12 (4 x 20 rnL) and was
dried with anhydrous MgS04. Concentration in vacuo and purification by column
chromatography (silica, EtOAclhexanes) afforded the desired compounds as single
dias tereomers.
(3aS*,4R*,9bS*)-4-(3-Carbobenzyloxy-aminopropyl)-2,3,3a,4,5,9b-
hexahydropyrrolo[3,2-clquinoline-1-carboxylic acid 1-benzyl ester, (84):
Obtained in 73% yield; white solid; mp = 160 - 162°C (EtOAc); Rf = 0.08 (30% EtOAcIhexanes); IR (film) u 3346, 3032, 2937, 1696, 1607, 1586, 1496, 1454, 1414,
1358, 1301, 1256, 1178, 10100, 1028, 915, 752, 698, 609 cm-'; iH NMR (500 MHz, CDC13), rotomers, 6 7.62 (OSH, d, J = 7.7 Hz, Hg), 7.28-7.44 (10.5H, rn, 2 x
COOCH2CtjH5, Hg), 7.00 (lH, q, J = 7.6 Hz, Hg), 6.69 and 6.60 (lH, t, J = 7. lHz, H7), 5.07-5.30 (5H, rn, 2 x COOCH2C6H5, Hgb), 4.85 (lH, m, H5), 3.14-3.75 (6H, m, Hi3, 2
x H12, 1 x H4,2 x H2), 2.36 (lH, m, H3,), 1.77-2.10 (2H, m, 2 x H3), 1.41-1.63 (4H, rn, 2 x Hl 1, 2 x Hia); 13c NMR (125 MHz, CDC13) rotomers, 6 156.44, 156.37, 143.34,
136.95, 136.58, 136.39, 130.44, 129.75, 128.50, 128.41, 128.12, 128.01, 127.96,
127.90, 127.82, 127.67, 122.53, 121.2, 118.67, 118.30, 114.29, 67.09, 66.81, 66.73,
56.50, 56.23, 51.52, 51.14, 50.83, 44.90, 41.65, 41.20, 40.93, 40.66, 32.22, 31.27,
27.63, 26.5 1, 22.10, 2 1.16; MS (EI) m/e 499 (36, M+), 39 1 (39), 364 (20), 32 1 (27), 307
(49), 300 (35), 256 (78), 130 (27), 108 (43 , 107 (33 , 91 (100), 79 (53), 77 (36); HRMS
(EI) d e (M+) calcd 499.247 1, found 499.246 1.
hexahydropyrrolo[3,2-c]quinoline-l,8-dicarboxylic acid 1-benzyl
methyl ester, (87):
ester 8-
Obtained in 92% yield; colouriess oil; Rf = 0.18 (50% EtOAcIhexanes); IR (film) u 3353,
3033,2948, 1701, 1610, 1517, 1435, 1358, 1282, 1244, 1100, 1028,915, 831,770,736, 698 cm-1; 1H NMR (500 MHz, CDCl3), rotomers, 6 8.23 and 8.13 (lH, s, Hg), 7.67 (lH,
d, J = 8.3 Hz, H7), 7.48 (IH, d, J = 7.1 Hz), 7.25-7.40 (IOH, m, COOC&C6&), 6.41
(lH, d, J = 8.3 Hz, Hg), 5.02-5.32 (5H, COOC&C6H5, Hs), 4.11 (lH, m, H13), 3.79
and 3.78 (3H, s and s, Me), 3.17-3.60 (SH, m, 2 x H2, H4, 2 x H12), 2.36 ( lH, m, H3,),
1.82 (2H, rn, 2 x H3), 1.4-1.68 (4H, 2 x Hlo, 2 x Hi 1); I3c NMR (125 MHz, CDCl3) rotomers, 6 167.13 and 166.94 (Cg), 156.44 (Cg), 155.38, 147.23, 147.15, 137.00,
136.53, 136.44, 132.44, 132.11, 129.92, 128.48, 128.39, 128.25, 128.13, 128.05,
127.98, 127.81, 127.75, 121.14, 120.13, 119.93, 119.61, 113.53 (c6), 67.48, 66.93,
66.73, 56.08, 55.87, 5 1.43 (Me), 50.65, 50.5 1, 44.86, 44.69, 41.23 and 40.67 (C3,),
30.93, 26.39, 22.20 and 21 -35 (C3); MS (EI) m/e 558 (16, M+), 422 (17), 3 14 (44), 108
(27), 107 (221, 91 (100), 79 (33), 77 (21); HRMS (EI) d e (M+) calcd 557.2526, found 557.2522.
General Procedure for the Deprotection of Cbz-Protected Pyrroloquinolines: Synthesis of 2,3,3a,4,5,9b-Hexahydro-1H- pyrrolo[3,2-clquinolines, (90a), (90b) and (91b):
Palladium on carbon (IO%, 12 mg) was added to a stirred solution of the Cbz-
protected pyrroloquinoline (mmol) in THF (2 mL). Formic acid (aqueous solution 10%)
was added in portions (1 mL) every 2 h until no starting material was present by TLC. The
P X was removed from the reaction mixture by filtration through a pasteur pipette plugged
with Cotton and filled with celite. NaOH (2 M aqueous solution, 10 rnL) was added, causing
the colourless solution to become rnilky white. After extraction with CH2C12 (4 x 5 mL), the
combined organic layers were dried with anhydrous MgS04 and concentrated in vacuo. The
the desired deprotected pyrroloquinolines.
(3aS*,4S*,9bS*)-4-Phenyl-2,3,3a,4,5,9b-hexahydro-lH-pyrrolo[3,2-
clquinoline, (90a):
Obtained in 56% yield; colourless oil; Rf = 0.14 (1% TENacetone); IR (film) u 3361,
3237, 3026, 2964, 2873, 1609, 1587, 148, 1452, 1364, 1316, 1296, 1264, 1156, 1 L 16, 1069, 103 1, 994, 910, 754, 703 cm-'; l H NMR (500 MHz, CDCl3) 6 7.43 (2H, d, J = 7.0
Hz, Ph), 7.36 (2H, t, J = 7.6 Hz, Ph), 7.28 (2H, rn, Ph, Hg), 7.04 (lH, t, J = 7.6 Hz, Hg), 6.79 (lH, t, J = 7.5 HZ, H7), 6.59 (IH, d, J = 7.8 HZ, H6), 4.61 (IH, d, J = 2.9 Hz,
Hgb), 4.53 (lH, d, J = 8.1 Hz, H4), 3.82 (lH, s, Hs), 2.90 (lH, m, Hz), 2.76 (lH, q, J =
8.1 Hz, H29, 2.68 (lH, m, H3,), 2.24 (lH, s, Hi), 1.93 (lH, m, H3), 1.38 (lH, rn, Hy); 13c NMR (125 MHz, CDC13) 6 145.33, 142.67, 129.22, 128.51, 127.52, 127.36, 126.47,
125.34, 1 19.30, 1 15.08, 58.07, 57.95, 46.82, 45.46, 25.28; MS (EI) m/e (rel intensity)
250 (67, M+), 220 (62), 207 (100), 144 (41), 130 (88), 117, (48), 91 (46), 77 (28); HRMS
(EI) d e calcd (M+) 250.1470, found 250.1459.
Obt ained in 80% yield; colourless oil; Rf = 0.26 (1% TENacetone); IR (film)
1608, 1586, 1482, 1409, 1352, 1264, 1051,754,701 cm-1; 1H NMR (500 MHz, CDC13) 6
7.38 (6H, m, Ph, Hg), 7.07 (lH, t, J = 7.2 Hz, Hg), 6.76 (lH, t, J = 7.3 Hz, H7), 6.59
10.8 Hz, Hgb), 3.1 1 (lH, m, H2), 2.88 ( lH, q, J = 9.1 Hz, Hz'), 2.38 (lH, m, H3,), 2.08 (IH, s, Hl), 1.83 (IH, m, H3), 1.52 (lH, m Hy); 13C NMR (125 MHz, CDCl3) 6 145.24,
141.96, 130.71, 128.58, 128.24, 127.96, 122.28, 118.17, 114.62, 58.84 (C4), 58.20
(Cgb), 44.55 (C2), 43.81 (C3a), 29.31 (C3); MS (ET) d e (rel intensity) 250 (14, M+), 232 (14), 220 (35), 207 (28), 206 ( 100), 128 ( 1 1 ), 9 1 ( 15), 77 (1 1); HRMS (EI) d e calcd (M+)
250.1470, found 250.1470.
(3aS*, 4R*, 9bS*)-4-Pheny1-2,3,3a,4,5,9b-hexahydro-lH-pyrrolo[3,2- clquinoline-8-carboxylic acid methyl ester, (91b):
Obtained in 8 1 % yield; colourless oil; Rf = 0.33 (1 % TEAiacetone); IR (film) u 335 1,3030,
2949, 2875, 1703, 1612, 1514, 1454, 1436, 1336, 1288, 1254, 1196, 1131, 1099, 1056, 833, 77 1, 735, 703 cm-l ; 1H NMR (500MHz, CDCl3) 6 8.04 (lH, d, J = 1.9 Hz, Hg),
7.74 (lH, dd, J = 8.3, 1.9 Hz, H7), 7.36 (5H, m, Ph), 6.55 (lH, d, J = 8.3, Hg), 4.52
(lH, s, Hs), 3.96 (lH, d, J = 5.9 Hz, H4), 3.83 (3H, s, Me), 3.78 (lH, d, J = 11.0 Hz,
Hgb), 3.12 ( lH, m, H2), 2.94 ( lH, m, H29, 2.34 ( lH, m, H3), 1.91 (2H, m, Hl, H3b),
1.56 (lH, m. H39; l 3 ~ NMR (125 MHz, CDCI3) 6 167.18, 148.98, 141.34, 132.85,
129.96, 128.75, 128.22, 128.17, 121.15, 119.20, 113.79, 58.35, 57.60, 51.52, 44.18,
42.9 1, 28.78; MS (EI) d e (rel intensity) 308 (13, M+), 290 (1 8), 278 (41), 264 (LOO), 229
(8), 205 (7), 115 (7), 91 (9); HRMS (EI) d e calcd (M+) 308.1525, found 308.1519.
Appendix A:
Submitted Articles
Acta Crystallographica C, submitted March 1997.
Tetrahedron Letters, accepted April 1 997.
-. . - - - - - ,-.-- .-.-..-. ..-... ...-.....- aUt.y.CJJ.Van V L L C U U L L U ~ I L ~ L C I ~ p a r d i ~ l t x c r s ( ~ . g . CL. y in t~ionoclltiac space groups), full formatting of reference l ist: and the placement of r he experiniental text . Note also t hat cotiipoiiiid name identifiers are derived from the arbitrary data block names within the CIF: these are linble to hc cliaiigeti. Please indicate any necessary changes for implementation by Acta editorial staff.
Acta Cryst. (1997). C53, 000-000
(3aR*,49,5 SC,7aR*)-4-nitrobenzoic acid 5-met hyl-1,3,3,4,5,7a-hexa-
hydro-isobenzofuran-4-yl ester
ROBERT A. BATEY, DENNY LIN AND ALAN J . LOUGH
Department of Chernidry: Unzver .dy of Toronto, Toronto, h t a r i o , Canada:
M5S 9H6. E-mail: alough@alchemy. ch,ern.utoronto.ca
Abstract
The crystal structure deterniinat ion of tlie title coiiipoiinrl (3aR*,4S*,5S,7aRS)-
4-ni trobenzoic acicl 5-inetliyl - 1.3,3,4,8.7a-lirxaliydro- isobei1zofuraii-4-yl ester.
CI6Hl7NO5, establislies the relative stereot:lieiiiistry. Tlie iimlecule c-oiitains a
furanyl ring wit h a twist conforinatioxi f~iscrl t o an iixisat urat ed six-~iieriilwrerl
ring with a twist-chair coxiforination. Tliere are weak iiiterr~ialec:iilar C-H- - -0
interactions, with distances C- - -0[1-z. 0.6+g. 0.5-z] 3.254 (1) A and Ha . .O [l-2.
O.5+ y, 0.5-4 2.41 A.
Comment
Recent stuclies have identifiecl alkenylboranes as reactivr clieiiopliiles in Diels-
Alder reactions (Matteson: 1995). The potential to coiitrol tlie relativv stemri-
chemistry of ai least 3 new st ereocentres in t lir intraiiiolec~iilar variant: a i r 1 t lit.
synthetic utility of the C-B l~olicl in the c~yrloaclcliiçts liaw proiiipterl ils t o
examine tliis reactioxi. During tlie course of oiir stiidies the first exaiiiplr of t his
an alkenylborane. generatecl in situ by se1~c.t ive liyclrolmrat ion of a r.lit.riyne:
undergoes an intra~nolecular Diels-Alcler reactioii after wliicli the C-B Imrid
in t lie cycloaddiict is t ransfor~iiecl. In an int raiiiolecular react ion: t wo ~iiodes
of cycloaddition ( e n d o / exo) are possilde. In the case of dienyii~ (1): a siiigle
diastereomer (2): formed as a viscous oil, wliicli was clerivatizecl as tlie c-rys-
talline p-nitrobenzoate ester to cletermiiie the relative stereoclie~iiistry ul,oiit t lit.
6-inembered ring.
The crys t al strircture de termina tioxi est aldislies the cliastereoiiier as t h
product of endo addit ion in t lie int rainoleciilar Diels-Alcler react ion. A seaidi
of the Cambridge Structural Database (Allen et al., 1979) rcvealecl that tliere
are only three other structures (refcocles; DAHDUL, FUMZIW a~icl JEWRUY)
containing similar trans-fused six ancl five-iiiernlierecl rings as in (3): h i t (3)
is the first compouiicil reportecl which cont ailis an 1.insiihstitiitec1 fiiranyl group
fusecl to a cyclohexene ring sys tem.
The five-meiiibered furanyl ring is in a twist roiiforiiiatioii. C6 is 0.412 (3)
above and C7 is 0.301 (3) A 11elow tlie plane foriiied by the tliree atonis CS: CO
and 0 3 . The twist conformation of tlie fiiraiiyl groiip iii (3) is coiisiste~~t witIi
the conformation of the moleciile iii t lie st riict iire of t et raliyclrofiiraii wliicli wns
determined at 103K and 148K 11y Luges Sc Biisliiiiaiiii, 1953. The fiiranyl groiip
in (3) has similar bond lengt 11s and angles to tliose in tet raliydroftiraii aiid the
only exception is the magnitucle of the angles CG-Cg-03 and C7-CS-O3 in
(3): which are 104.63 (9)" ancl 104.32 ( 9 ) O respec-tively: coilipareil to 101.4 ( 4 ) O nt
103K and 106.7 (4)" at 14SK for tet raliyclrofiiraii ( t lie t et raliytlrofirraii iiioletwlr
l-ias cryst allograpliic two-folcl syniniet ry ).
The six-membered ring of the fusecl rings systeni in (3) has a twist-chair
conformation. Atonls C3, C4: C5 aiicl C6 fornl a least-squares plane (witli
maximum deviation of 0.008 (1) A for C4) and C7 is 0.644 (2) A I~clow the plane
while C2 is 0.139 (2) A above tlie plane. Tlie six alid five-iiieiiil~rrril rings are
trans-fused along tlie C6-Cï I~oixl (sre Fig. 1).
- ., A
c
close to t lie plane of the benzene ring, the largest tlevia tion beiiig 0.135 (2 ) A fins
01. The fused rings system is rotatect out of the plane of the iiitrol~enzoat e est es
group by an angle of 58.0 (1)".
In (3) furanyl O atoms are irivolvecl in iatermolecular close contacts of the
type C-H- -O. Tliese interactioiis occur between nol le cul es relatecl l ~ y 2] scwrew
axes, to form infinite chains of molecriles (see Fig. 2). Tlie relavent clis t anws
are C l 3 -*03(1-2, 0.5+y, 0.5-i] 3.254 (1) aiicl H13A - - 0 3 [1-s, -O.8+ y, 0 . 5 ~ 1
2.41 a.
Experimental
Crystal data
CisHz7NOs M, = 303.31
Monoclinic
P2 1 /c
a = 8.0485 (9) À b = 14.301 (2) À
c = 13.2703 (14) A [j = 91.769 (7)"
v = 1526.7 (3) A3
2 = 4
D, = 1.320 Mg ni-'
Ce11 paranieters froili 43 reflet: t ions
8 = 5.25-25.00"
p = 0.099 min-'
T = 213 (2) K Flat iieeclle
0.49 x 0.45 x 0.36 nini
Pale yellow '
Crys t al source: see t ext
Dm not measured
Mo Ka radiation
A = 0.71073 A
Absorption correct ion: none
4686 measured reflec t ions
4405 independent reflections
3152 reflections witli
I > 244 Rint = 0.0195
Ref inement
Refinement on
R[P > 2o(P)] = 0.0433
wR(F2) = 0.1288
S = 1.045
4405 reflections
201 parameters
H atoms riding w=1/[o2(F:) + (O.O71OP)* +
0.0556 P] wliere P = (F: + 2 F 2 ) / 3
X : = 0 + 2 0
1 = -1s + 1s
3 st aliclartl reflect ions
every 97' reflert ions
intemi ty clecay: less t han 2%
(d/a),,, = 0.001
ApI,lax = 0.251 e À-" 4pI,,in = -0.210 e A-" Extinction correct ion: SHELXL Extiiictioii coefficient: 0.0027 (14)
Scat teriiig fart ors froiii International
Tizb1e.q for Crys ta l lopaphy (Vol. C )
Y
0.74125 ( 5 )
0.7132S (7)
0.50303 (6)
1.2308S (S)
1.19499 (8)
0.79776 (8)
0.64308 (7)
0.6277s (8)
0.52714 (9)
0.46160 (8)
0.48425 (7)
0.58793 (1)
0.59168 (5)
0.44064 (S)
0.69417 (10)
OS9604 (S)
0.92341 (S)
1.0154'1 (8)
l.OïSO6 (8)
1.05345 (9)
0.96151 (9)
1.17553 ( 8 )
2
0.43078 (6)
0.56236 (S)
0.34971 (7)
0.51066 (12)
0.37049 (11)
0.49350 (9)
0.42684 (3)
0.38504 (9)
0.35631 (10)
0.35097 (9)
0.38084 (9)
0.36386 (8)
0.35599 (10)
0.32773 (10)
0.30187' (10)
0.47835 (5)
0.39160 (9)
0.37970 (10)
0.455S9 (10)
0.54211 (10)
0.55336 (9)
0.44456 (11)
1.517 ( 2 ) .
1.5266 (14)
1.516 (2) 100.63 (9)
120.39 (O)
107.91 (S)
1oo.ss (9)
104.63 (9)
104.32 (9)
Table 3. Hydrogen-bonding g e o n e t r y (A; O )
D-H- --A D-H He. .A Dm. - A D-H. - A
C 1 3 H 1 3 A - . -03' 0.94 2.41 3.254 (2) 140
Synunetry codes: (i) 1 - z: 3 + y: ) - s.
(SO xng, 0.4Smmol) in CH2C12 (3.0 1111). After stirriiig overnight: HiO ( 5 1111)
was aclded and the aqueoits layer was extractecl witli CH2& ( 3 x Ci i d ) . The
organic layer was t hen waslied wit 11 Ixiiie (10 1111): cirierl witli anhyrlroiis kIgS04
ancl concentrated in V ~ L C U O . Flash coluixiii clironiat ograpliy provitlecl t he t i t Ie
compound as a pale yellow solicl (55 mg, 82%). Pale yellow crystals g r ~ w fsorii a
solution of ether:pentane (3:2) over 3 ciays at 295 K.
Data collection: Siemens XSCANS, 1994. Cr11 refirieiiieiit : Sienieiis XS CANS.
1994. Data reduction: Siemens XSCANS, 1994. Prograin(s) iised to solve striir-
t ure: Siemens SHELXTL, 1994. Program(s) iisecl to sefiiie stririi t iise: Sienielis
SHELX TLS 1994. Molecular graphies: Sie~iiexis SHELXTL, 1994. Software used
to prepare material for piiblicatioii: Siemens SHELXTL. 1994.
This work was siipportecl by the NSERC Canada aiicl the University of
Toronto.
Lopies inay be ol~tailiecl t liroiigh -1 lit? ~\/laiiagiiig Ecii tex. lilterllat ional Ci ilion oI
Crystallograplly: 5 -4l)l)ey Square: C1iestc.r CH1 ?HU1 Eiiglaiicl.
References
Allen, F. H., Bellarcl. S., Brice, M. D.. Cartwright, B. -4.. Do~iljlerlay~ A.. Higgs,
H.: Hurnmelink, T.: Huxiiiiieli~ik-Peters: B. G.: Iieiiiiarrl, O., Motherwell: W. D.
S.: Rodgers, J . R. & Watson: D. G. (1979). Acta Cryat. B35, 2331-2339.
Batey, R. A., Lin: D.: Haylioe. C. L. S. & 1Vo1ig; A. (1097). Tetrahedron Lett.
Subrnit ted.
Luger, P. & Buschmailn 3. (1983). Angew. Chem. Int. Ed. Engl. 22. 410.
Mat teson, D. S. ( 1995). "S tereoclirectecl Syrit liesis wit 11 Orgaiiol~orrtiies' : (-11. 8:
Spriiiger-Verlag Berlin Heiclell~erg: New York. Shelclrick, G. M. (1994). S H E L X T L I P C Version 5. 0: Users Maiiiial, Sieiiieiis
Aiialytical X-ray Instriiinents Iiic., Mactisoii, Wiscolisiii. USA.
Siemens (1994). XSCANS Versioil 3. 1: Tecliiiid Reference Manual, Sieineiis
Aiialytical X-ray Iiistrtinients Iiic:., blarlisoii: \Visc:oiisiii, USA.
Singleton, D. A. & Lee, Y. K. (1995). Tetrahedron Lett. 36, 3473-3476.
Fig. 1 View of ~iioleciile wit h a toiiiic la1,elliiig sclieine. Displacenient ellipsoids are
clrawii at tlie 30% probability levrl aiid liydrogeii atniiis are tlrawn as sniall
splieres.
Intramolecular Diels-Alder Reactions of Alkenylboranes - A Stereoselective Route to Functionalized Bicyclo[4.3.0]nonenes.
Robert A. ~atey,*l Denny Lin, Andrew Wong2 and Christina L. S. Hayhoe.2
Department of Chemistry, Lash Miller Laboratories, 80 St. George Street, University of Toronto, Toronto, Ontario, MSS 3H6, CANADA.
Key words:
A bstract:
Intramolecular Diels-Nder Reaction, Alkenylborane, Hydroboration, Masked Dienophile, E-En01 Dienophile Equivalent.
A one-pot procedure for the synthesis of bicyclo[4.3.0]alkenols is described. Alkenylboranes, forrned in situ by hydroboration, undergo stereoselective intramolecular Diels-Alder reactions to give alkylboranes, which upon oxidation, yield the products corresponding to endo addition.
The intrarnolecular Diels-Alder reaction is a powerful method for the formation of bicyclic and polycyclic molecules.3 The major difficulties associated with this methodology are the formation of the requisite precursors, and the potential limitations inherent in the steric and electronic requirements of the dienophile and diene, and the resultant effect upon the stereoselectivity and rate of cycloaddition. As part of a wider interest in the use of boron substituents in controlling rea~ t iv i ty ,~ we have investigated the utility of alkenylboranes in intramolecular Diels-Alder reactions.5 The general strategy (Scherne 1) that we envisaged employs a three-stage "one-pot" reaction, in which an enyne 1 is used as a precursor for the cycloaddition substrate alkenylborane 2. Conversion of 2 to the cycloadduct borane 3 under thermal conditions, is followed by transformation to the desired adducts 4 or 5, which contain three new contiguous stereocentres. For example, straightforward oxidation would yield alcohol4. Such an approach is attractive, since the boron substituents can in principle be used to control the diastereo- and enantioselectivity of addition, and the C-B bond in the cycloadduct 3 can be
oxidation
or C-B bond conversion
Scheme 1
5 (FG = NR2, SR, CR3 etc.)
I Y U k L b V l . l i > .
Intermolecular Diels-Alder reactions using alkenylboron species have been known for over thirty years, the first example reported by Matteson using alkenylboronic esters.' More recently, Singleton and CO-workers have firmly established alkenylboron compounds as practical dienophiles in intermolecular r eac t i~ns .~ During the initial stages of Our study, Singleton reported the first example of an intramolecular Diels-Alder reaction using an alkenylborane, in which a trans-fused bicyclo[4.4.0]deceno1 10 (Figure 1) was formed, using either hydroboration or transmetallation to form the requisite alkenylborane.9 In contrast, Our work has focussed on the formation of bicyclo[4.3.0]nonenols 4 (n=l, X=O or CR2), through a "one-pot" procedure involving (i) hydroboration, (ii) cycloaddition, and (iii) oxidation (Scheme 1).
Three substrates 6a-c were chosen to test this methodology, but initial optimization was carried out on the readily available propargyl ether 6al0 (Table 1). The choice of hydroborating agent is critical for the success of this procedure, since chemoselective hydroboration of the alkyne in the presence of the diene is necessary, and the alkenylborane must also be a good dienophile. Several hydroborating agents that are selective for alkynes over alkenes are known,6 and four of these were screened: dicyclohexylborane, disiamylborane, catecholborane and dimesitylborane. Dicyclohexylborane and disiamylborane both led to the formation of product 7a. Initial studies were complicated by varying yields of products. However, the addition of a small quantity (5 mol%) of butylated hydroxytoluene (BHT), as a radical inhibitor, prior to heating the reaction led to increased yields of cycloadducts 7a-c. In general, dicyclohexylborane gave better yields than disiamylborane, and experimentally was easier to use, since the success of the initial hydroboration step could be followed by the dissolution of the dicyclohexylborane. Optirnized conditions for the Diels-Alder step occuned on refluxing the alkenylborane in a benzene / THF solvent mixture overnight. Several oxidation procedures were screened, but oxidation by trimethylamine-N-oxide gave the best results, and had the added advantage that functional groups such as esters, which are normally labile to the traditional alkaline hydrogen peroxide work-up, were unaffected. Thus, under optirnized conditions," 7a was formed as a single diastereomer.12 Catecholboranel3 did not undergo selective hydroboration of 6a at 80°C, and the only isolable product was that derived from
Table 1. Conversion of dienynes 6 into cycloadducts 7, using dicyclohexylborane.~ 1
- -
Substrate C ycloadduc t Isolated Yield
EtOOC EtOOC 6 k 7c
--ii--.ivr.u, r v u u i i i b a.. w w . r i vu".," *.A *irv rurrrrurruri v r "irrp*, iuuiirvi- iii-u--i-,-----i. . - --- . - --- ---------- J ----- (Table 1). NOE experiments revealed that the stereochemistry of each of the cycloadducts 7a-7c contained a trans-fused ring junction between the newly formed rings. In the case of 7a, the relative stereochemistry was proven by an X-ray crystal structure of the corresponding p-nitrobenzoate ester.
Ref. X RI R2 tram : cis
\ - - I l a 14a CH2 H H 1 : 3 -
8 9 I l b 14b CH2 COOCH3 H 6 : 4 - -
"endo" transition state "exo" transition &te I l c 14b CH2 COOCH3 ;Pr 72 : 28 11d 14c CH2 NO2 H 8 9 : 11 I l e 146 O H H 1 : 3 I l f 14d O CH=CHCH3 H 4 : 1 I l g 14a (CH2)2 H H 48 : 52
10 11 I l h 14e (CH2)2 COOCH3 H 51 : 49
O H Ili t4f CH20 COOEt H 3 : 2
Figure 1: Some Comparative Data for Thermal Intrarnolecular Diels-Alder Cycloadditions
Reaction via the "endo" transition state 8 leads to the formation of trans-fused adducts, whereas reaction via the "exo" transition state 9 would yield the cis-fused adducts (Figure 1). However, thermal intrarnolecular Diels-Alder reactions are known not to follow the "endo rule", with non-bonded interactions in the tethering chain having profound effects on the stereoseIectivity.3 Substituted nonatrienes l lb -d , analogs of 6a and 6c, show a preference for formation of the trans-fused-[4.3.0]nonenes under thermal conditions (Figure l).14 Interestingly, the oxygen tethered substrate l le yields rnainly cis-fused adducts, whereas l l f yields mainly trans-fused adducts.l5 Singleton has proposed using ab initio calculations that a [4+3] transition state occurs for the intermolecular reaction of alkenylboranes with dienes, in which there is substantial bonding character between the boron atom and the terrninal position of the diene in the endo-transition state.16 Sirnilar interactions were also calculated to be important in the corresponding intramolecular cases leading to the formation of trans- fused-[4.4.0]decenes, such as 10 (Figure l).9 By comparison, the cycloaddition of simple substituted decatrienes l lg- i results in poor diastereoselectivity, but the addition of a methyl group at the 3-position of the diene (i.e. 3-methyl- l,3,9-decatriene) '7 leads to exclusive formation of trans-fused bicyclo[4.4.0]decenes (cf. l l g ) . Thus, in the case of compound the methyl group may have a strong influence on the observed stereoselectivity. The current work demonstrates the formation of trans-fused bicyclo[4.3.0]adducts 7a-7c,18 and is consistent with reaction via a [4+3] transition state, although this remains speculative because of a lack of appropriate comparative experimental data for thermal intramolecular Diels-Alder reactions. In any case, the reactivity of 6a-c is notably higher than trienes in similar reported thermal cycloadditions (e.g. l la- i) , l4 demonstrating the activating effect of a boron substituent on the dienophile in thermal intramolecular Diels-Alder reactions. Also, the observed stereoselectivity is comparable to that achieved in Lewis acid catalyzed intrarnolecular Diels-Alder reactions, although the reactivity is lower than for Lewis acid activated dienophiles.,
In summary, we have developed methodology for the synthesis of trans-fusedC4.3.01 bicycloalkenols, using a three-step hydroboration-cycloaddition-oxidation strategy. Optimized conditions employ dicyclohexylborane for the hydroboration step, with BHT as a radical inhibitor for the thermal cycloaddition step. The boron substituent both activates the cycloaddition and is a synthetic equivalent for an E-en01 as a dienophile. The observed stereoselectivity corroborates the [4+3] transition state mode1 for the cycloadditions of alkeny lboranes. Further studies in the application of organoboranes to cycloaddition chemistry are now in progress in Our laboratory and will be reported in due course.
References and Notes: ( 1) E-mail: rbatey @chem.utoronto.ca. (2) Undergraduate Research Participants. (3) For reviews on intramolecular Diels-Alder reactions, see: (a) Roush, W. R. in Cornprehensive Organic Synthesis, Trost, B. M., Fleming, I., Paquette, L. A., Eds.; Pergamon: Oxford, 1991, Vol. 5, 5 13-550; (b) Carruthers, W. Cycloaddition Reactions in Organic Synthesis, Pergamon: Oxford, 1990; (c) Craig, D. Clzem Soc. Rev. 1987, 16, 187-238; (d) Fallis A. G. Can. J. Chem. 1984,62, 183-234. (4) Batey, R. A.; Pedram, B.; Yong, K.; Baquer, G. Tetrahedron Lett. 1996,37, 6847-6850. (5) This work was presented at the 7th Ontario-Quebec Minisymposium, University of Waterloo, Ontario, October 1996, "Recent Advances in Intrarnolecular Diels-Alder Cycloaddition Chemistry"; and at the Canadian Society for Chemistry Meeting, Mernorial University, St. John's, Newfoundland, June 1996, "Intramolecular Diels-Alder Reactions of Alkenylboranes." (6) (a) Matteson, D. S. Stereodirected Synthesis with Organoboranes, Springer-Verlag: Berlin, 1995; (b) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents, Academic: London, 1988. (7) Matteson, D. S.; Waldbillig, J. O. J. Org. Chem. 1963,28, 366-369. (8) (a) Singleton, D. A.; Martinez, J. P. J. Am. Chem. Soc. 1990,112, 7423-7424; (b) Singleton, D. A.; Martinez, J. P.; Ndip, G. M. J. Org. Chem. 1992,57, 5768-5771; (c) Singleton, D. A.; Martinez, J. P.; Watson, J. V. Tetrahedron Lett. 1992,33, 1017- 1020. (9) Singleton, D. A.; Lee, Y.-K. Tetrahedron Lett. 1995,36, 3473-3476. (10) Compounds 6a and 6b were synthesized by propargylation of the corresponding sodium alkoxides. Compound 6c was synthesized by standard malonate alkylation techniques. (1 1) Dieneyne 6a-c (1.9 rnrnol) was added dropwise to a suspension of dicyclohexylborane (2.0 mm01 in 4 ml of THF). After stirring for 2 h, BHT (O. 1 mrnol) in freshly distilled benzene (50 mL) was added to the reaction and the reaction mixture was heated at reflux for f 7-20 h. After cooling to room temperature, trimethylamine-N-oxide dihydrate (6.5 rnmol) was added and the reaction was reheated to reflux for 24 h. Upon cooling to room temperature, distilled water (20 mL) was added and the reaction mixture was heated to 60 OC for 30 min. Aqueous work-up and purification by flash colurnn chromatography (silica gel, EtOAc / hexanes) afforded the desired bicyclo[4.3.0]alkenols 7a-c. In each case only a single diastereomer was observed by NMR and t.1.c. (12) Selected spectral data for compound 7a: IR (neat) u 34 18, 302 1 2933, 2869, 1632, 1455, 1 160, 1 117, 1102, 1070, 1040, 1016,981,881, 867,806,729 cm-1; 1H NMR (400 MHz, CDC13) 6 5.63 (2H, m), 4.06 (3H, m), 3.58 ( lH, dd, J = 11.0, 7.4 Hz), 3.38 (lH, dd, J = 11.7, 7.0 Hz), 2.67 (lH, m), 2.48 ( lH, rn), 2.06 ( lH , dq, J = 11.2, 6.9 Hz), 1.71 ( lH, s, OH), 1.04 (3H, d, J = 6.9 Hz); 13C NMR (100 MHz, CDC13) 8 134.94, 122.77, 72.59, 70.64, 70.00, 45.22, 45.09, 37.55, 14.53; HRMS (EI) m/e calcd. (M-H+) 153.0915, found 153.0911. (13) Brown, H. C.; Gupta, S. K. J. Am. Chem. Soc. 1975,97, 5249-5255. (14) (a) Lin, Y.-T.; Houk, K. N. Tetrahedron Lett. 1985,26, 2269-2272; (b) Roush, W. R.; Gillis, H. R.; Ko, A. 1. J. Ain. Chern. Soc. 1982, 104, 2269-2283; (c) Kurth, M. J.; O'Brien, M. J.; Hope, H.; Yanuck, M. J. Org. Chern. 1985, 50, 2626-2632; (d) Hertel, R.; Mattay, J.; Runsink, J. J. Am. Chem. Soc. 1991, 113, 657-665; (e) Roush, W. R.; Hall, S. E. J. Ain. Chent. Soc. 1981, 103, 5200-52 1 1 ; (f) Boeckman, Jr., R. K.; Dernko, D. M. J. Org. Chern. 1982,47, 1789-1792. ( 15) The selectivity observed in the cycloaddition of l l f , is presumably complicated by considerable diradical character for the transition States. (16) Singleton, D. A. J. Am. Chem. Soc. 1992, 114, 6563-6564. (17) Chou, T.-S.; Tso H.-H., Chang, L.-J. J. Cliern. Soc., Chern. Conzmrin. 1985, 236-237. (18) Trans-fused products similar to that of tricycle 7b are also formed from the allylated analog of 6b, see: Burke, S. D.; Smith Strickland, S. M.; Powner, T. H. J. Org. Clrein. 1983, 48, 454-459.
Appendix B:
X-Ray Crystal Structure Data
xaentlrxataon coae
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system
Space group
Unit ce11 dimensions
Volume, 2
Denuity (calculated)
Absorption coefficient
P(000)
Crystal size
8 range for data collection
Limiting indices
Rsflections collected
Independent reflections
Absorption correction
Refinamsat method
Data / restraints / parameters
Goodness-of-fit on F 2
Final R indices [1>2a(I)]
R indices (al1 data)
Extinction coefficient
Largest diff. peak and hole
SYDaU
C H NO5 16 17
303.31
213(2) K
0.71073 A Monoclinic
p 2 p
e = 8.0485(9) A alpha = 90° b = 14.301(2) A bsta = 91.769(7)~ c = l3.2703(14) A gamma = 90°
is26.7(3) A ~ , 4
1.320 Mg/- 3
0.099 mm-'
640
0.49 x 0.45 x 0.36 mm
2.53 ta 30.00~
O 5 h 5 11, O d k I 20, -18 S 1 S 18
4686
4405 (Rint = 0.0195)
None
Full-matrix least-squares on F 2
4405 / O / 201
1.045
R1 = 0.0433, wR2 = 0.1212
R1 = 0,0607, wR2 = 0.1288
O.OOZ7 (14)
0.251 and -0.219 c ~ A - ~
one t h i r d of the trace of the orthogonalized Uij tensor.
Spmnietry transformations used to generate equivalent atoms:
-2n' [ (ta )'ull + .-. + 2hka b U12 ]
Identification code
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system
Space group
Unit ce11 dimensions
Volume, 2
Density (calculated)
Absorption coefficient
F(000)
Crystal size
e range for data collection
Limiting indices
Reflections collected
Independent reflections
Absorption correction
Refinement method
Data / restraints / parameters
Goodness-of -f it on F 2
Final R indices [I>2u(I)]
R indices (al1 data)
Absolute structure paraaeter
Largest diff. peak and hole
Orthorhombic
P21212
a = 11 .O687(14) A alpha = 90° b = 25.113(3) A beta = 90°
c = 7.8901(8) A gamma = 90°
2193.2(5) A ~ . 4 1.340 Mg/m 3
0.090 mm-'
3622 (Rint = 0.0000)
None
Full-matrix least-squares on F 2
2.6(12)
0.411 and -0.366 e ~ - ~
one t h i r d of the trace of the orthogonalized Uij tensor.
one third of the trace of the orthogonalized C i i j tensor.
Symetry transformations used to generate equivalent atoms:
L = L I u
-2n [ (ha ) Ull + . .. + 2hka b U12 ]
Appendix C:
Selected Spectral Data
STRflDWD PROTOII PRRARETERS Dinny Lin 8123 DLE39E
SMPLE da t e nar 18 97 iolvmnt CDC13 111i -xP
RCQUISITlON s l r q 499.856 Ln n i r t 1 .892 nP 38272 au BI00. e Ib no t uaid br 32 tpur 58 Pu 6.7 dl 0 tof see. e n t 16 c t 16 a lo tk n p i n n o t urmd
F M 5 11 n 1 n n dp ha nn
DISPLAY .P -0.1 UP 4998.4 Y. 162 .C e
C u 258 td b r i i 4 . a - l a 1129.61
r f l 4633.9 r r p 3619.8 LL 1 1 na 1.888 n i ph
DEC. L UT d t r q , 499.855 dn Hl dput 1 e do1 B dm nnn d i i 6
dm f 2BQ damq drmr 1 .O homo U t u p 25.8
MC2 d I rq2 B dn2 d p r 2 1 do12 B d i 2 n d i i 2 c dmf2 29e d.mq8 d r m d 1 .B homo2 n
PROCESSIIG ut1 1 Io proc I t 1 n 65536 math f
Cbz,
SI~INDARD PROTON PARAnEIERS
Soluent: CDC13 Taip. 35.8 C / 398.1 K UMITY-588 'unlty588'
PULSE SEOUENCE Pulse 38.1 digraar k q . tiis 1.892 aac u d t h 8888.8 nz 16 repatitions
OBSERVE Hf. 499.8528661 M z D A T I PROCESSINC FT rlzs 65536 Tota l Clma 1 ilnuta
Cbz, N?
c&,I;œ"am ? C a ! œ * - = u œ q c 'D % = - . Y I 2 !v œ
c g - rn * rn ; !?
3 * N
a z VI W
V1 N N
a 2 W
N N N O
5 ;= ~<+34s: r+r : . - r : : g ;;gws:4::am z m p N t z g : : s r * O a:: 5 : p m 4 : ; : J -
I E ln
w $ G C U m m Y
N r m Yl - U -ln
= . a m 5 qt m . . y' = ; ;
# - . Y . m 4 5 3 2 d x - 4 g 2 ~ h 2 +~i: a' u - - s ~ ~ ~ ~ ~ ~ ~ ~ X ~ E ~ := O:,, = : 8 m 4 L I f w * . v r w L . s W a n . Z - A C ~ 2:: J " ~ ~ T 2 2 8 L - Z
œ . C C fi 1 œ.a .r. 2 - ; i L ; t 5t2;;g J D i ' z : 2 5 ; 5 2 m m : : q s g z=L8;::gg'""a"0 IL C
Fleming, 1. Frontier Orbitals and Organic Chernical Reactions, Wiley-Interscience,
Avon, 1976. Nicolaou, K. C.; Petasis, N. A.; Zipkin, R. E.; Uenishi, J. J. Am. Chem. Soc.,
1982, 104, 5555-5557. Nicolaou, K. C.; Yang, 2.; Liu, J. J.; Ueno, H.; Nanteromet, P. G.; Guy, R. K.;
Claiborne, C. F.; Renaud, J.; Couladouros, E. A.; Paulvannan, K.; Sorensen, E. J.
Nature, 1994,367, 630-634. For reviews of intermolecular Diels-Alder reactions, see: (a) Oppolzer, W. in
Comprehensive Organic Synthesis, Trost, B. M., Fleming, I., Paquette, L. A.,
Eds.; Pergarnon: Oxford, 1991, Vol. 5, 3 16-399; (b) Sauer, J. Angew. Chem. Znt.
Ed. Engl., 1967, 6, 16-33.
Matteson, D. S.; Waldillig, J. O. J. Org. Chem., 1963,28, 366-369.
Wallace, R. H.; Zong, K. K. Tetrahedron Lett., 1992,33, 6941-6944.
Hollis Jr., G. W.; Lappenbusch, W. C.; Everberg, K. A.; Woleben, C. M.
Tetrahedron Lett., 1993,34, 75 17-7520.
(a) Pelter, A.; Smith, K.; Brown, H. C. Best Synthetic Methods: Borane Reagents,
Acadernic Press, London, 1988; (b) Matteson, D. S. Stereodirected Synthesis with
Organoboranes, Springer, Berlin, 1995.
Singleton, D. A.; Martinez, J. P. J. Am. Chem. Soc., 1990, 112, 7423-7424.
(a) Singleton, D. A.; Martinez, J. P.; Watson, J. V. Tetrahedron Lett., 1992, 33,
1017-1020; (b) Singleton, D. A.; Martinez, J. P.; Watson, J. V.; Ndip, G. M.
Tetrahedron, 1992,48, 583 1-5838.
Singleton, D. A. J. Am. Chem. Soc., 1992,114, 6563-6564.
See reviews on IMDA reactions: (a) Carruthers, W. Cycloaddition Renctions in
Organic Synthesis, Pergamon: Oxford, 1990; (b) Roush, W. R. in Conzprehensive
Organic Synthesis, Trost, B. M., Fleming, I., Paquette, L. A., Eds.; Pergamon:
Oxford, 1991, Vol. 5, 513-550; (c) Brieger, G.; Bennett, J. N. Chem. Rev., 1980,
80, 63-97; (d) Fallis, A. G. Can. J. Clzem., 1984,62, 183-234; (e) Craig, D. Chem. Soc. Rev., 1987,16, 187-238.
Oppolzer, W.; Fehr, C.; Warneke, J. Helv. Clzirn. Acta, 1977,60, 48-58.
Naf, F.; Decorzant, R.; Thommen, W. Helv. Chim. Acta, 1982,65, 22 12-2223.
Corey, E. J.; Petrzilka, M. Tetrahedron, Lett., 1975, 30, 2537-2540.
House, H. O.; Cronin, T. H. J. Org. Chern., 1965,30, 106 1-1070.
Roush, W. R.; Gillis, P.J. Org. Chem., 1982,47, 4825-4829.
Lin, Y. -T.; Houk, K. N. Tetrahedron Lett., 1985,26, 2269-2272.
Roush, W. R.; Gillis, H. F.; Ko, A. 1. J. Am. Chem. Soc., 1982, 104, 2269-
2283.
Craig, D.; Fischer, D. A.; Kemal, O.; Plessner, T. Tetrahedron Lett., 1988,29,
6369-6372.
Gentry, A.; Cook, K. J. Ethnopharmacol., 1984, 11, 337-343.
Withemp, K. M.; Ransom, R. W.; Graham, A. C.; Bernard, A. M.; Salvatore, M. J.; Lurnma, W. C.; Anderson, P. S.; Pitzenberger, G. M.; Varga, S. L. J. Am.
Chem. Soc., 1995, 11 7, 6682-6685.
Yates, P.; Eaton, P. J. Am. Chem. Soc., 1960,82, 4436-4437.
Trost, B. M.; Ippen, J.; Vladuchick, W. C. J. Am. Chem. Soc., 1977,99, 81 16-
8118.
Wu, T. -C.; Houk, K. N. Tetrahedron Lett., 1985,26, 2293-2296.
For a review of the applications of lanthanides, see: (a) Imarnoto, T. Lanthanides in
Organic Synthesis , Academic Press, London, 1994; (b) Molander, G. A. Chem.
Rev., 1992,92, 29-68; (c) Kagan, H. B.; Namy, J. L. Tetrahedron, 1986,42,
6573-66 14.
(a) Ishar, M. P. S.; Wali, A.; Gandhi, R. P. J. Chem. Soc. Perkin Trnns. 1, 1990,
2185-2192; (b) Bednarski, M.; Danishefsky, S. J. Am. Chem. Soc., 1983,105,
3716-3717; (c) Bednarski, M.; Danishefsky, S. J. Am. Chem. Soc., 1983,105,
6968-6969.
Bednarski, M.; Danishefsky, S. J. Am. Chem. Soc., 1986, 108, 7060-7067.
Kobayashi, S.; Ishitani, H.; Nagayama, S. Chem. Lett., 1995, 423-424.
Kobayashi, S.; Nagayama, S. J. Am. Chem. Soc., 1996,118, 8977-8978.
Kobayashi, S.; Ishitani, W.; Ueno, M. Synlett, 1997, 115-1 16.
Berchtold, G. A.; Ciabattoni, J.; Tunick, A. A. J. Org. Chem., 1965,30, 3679-
3682.
Danishefsky, S.; Cavanaugh, R. J. Org. Chem., 1968,33, 2959-2962.
(a) Backvall, J. -E.; Juntunen, S. K. J. Am. Chem. Soc., 1987,109, 6396-6403;
(b) Padwa, A.; Harrison, B.; Murphree, S. S.; Yeske, P. E. J. Org. Chem., 1989,
54,4232-4235.
For reviews on Diels-Alder reactions of azadienes, see: (a) Povarov, L. S. Rrrssiarz
Chem Rev., 1969, 36, 656-670; (b) Boger, D. L. Tetrahedron, 1983,39, 2869-
2939; (c) Kametani, T.; Hibino, S. Adv. in Heterocyclic Clzem., 1987,42, 245-
333; (d) Sandhu, J. S.; Sain, B. Heterocycles, 1987,26,777-818; (e) Waldmann,
H. Synthesis, 1994, 535-55 1.
Laschat, S.; Lauterwein, J. J. Org. Chem., 1993,58, 2856-2861.
(a) Gisby, G. P.; Sammes, P. G; Watt, R. A. J. Chem. Soc. Perkin Trans. 1, 1982,
249-255; (b) Day, F. H.; Bradsher, C. K.; Chen, T. -K. J. Org. Chem., 1975,40,
1195-1 198; (c) Chen, T. -K.; Bradsher, C. K. J. Org. Chem., 1979,44, 4680-
4683.
Nomura, Y.; Kimura, M.; Takeuchi, Y.; Tomoda, S. Chem. Lett., 1978, 267-270.
Shono, T.; Matsumura, Y.; Inoue, K.; Ohmizu, H.; Kashimura, S. J. Am. Chem.
SOC., 1982,104, 5753-5757.
Cheng, Y. -S.; Ho, E.; Mariano, P. S.; Ammon, H. L. J. Org. Chem., 1985,50,
5678-5686.
Spectral data for this compound (CAS [106144-74-51) have been reported.
Hayakawa, K.; Kazuyoshi, A.; Shiro, M.; Kanematsu, K. J. Am. Chern. Soc.,
1989,111, 53 12-5320.
Collington, E. W.; Meyers, A. 1. J. Org. C'hem., 1971,36, 3044-3045.
Brown, H. C.; Gupta, S. K. J. Am. Chem. Soc., 1975,97, 5249-5255.
Pelter, A.; Singaram, S.; Brown, H. C. Tetrahedron Lert., 1983,24, 1433-1436.
Zweifel, G.; Ayyangar, N. R.; Brown, H. C. J. Am. Chem. Soc., 1963,85, 2072-
2076.
Brown, H. C. Organic Syntheses Via Boranes, John Wiley & Sons, 1974.
NaB03 oxidation of trialkylboranes: (a) Kabalka, G. W.; Shoup, T. M.; Goudgaon,
N. M. J. Org. Chem., 1989,54, 5930-5933; Me3NO oxidation of trialkylboranes:
(b) Kabalka, G. W.; Slayden, S. W. J. Organornet. Chem., 1977, 12.5, 273-280.
Tucker, C. E.; Davidson, J.; Knochel, P. J. Org. Chem., 1992,57, 3482-3485.
Pereira, S.; Srebnik, M. Organornetallics, 1995,14, 3 127-3 128.
For a convenient preparation of Schwartz' Reagent, see: Buchwald, S. L.; LaMaire,
S. J.; Nielsen, R. B.; Watson, B. T.; King, S. M. Tetrahedrorz Letr., 1987,28,
3895-3898.
Batey, R. A.; Lin, D.; Lough, A. J. Acta. Crystallographica C., 1997, subrnitted.
(a) Lin, Y. -T.; Houk, K. N. Tetrahedron Lett., 1985,26, 2269-2272; (b) Roush,
W. R.; Gillis, H. R.; Ko, A. 1. J. Am. Chem. Soc., 1982, 104, 2269-2283; (c)
Kurth, M. J.; O'Brien, M. J.; Hope, H.; Yanuck, M. J. Org. Chem., 1985, 50, 2626-2632; (d) Hertel, R.; Mattay, J.; Runsink, J. J. Am. Cheni. Soc., 1991, 113,
657-665.
Singleton, D. A.; Lee, Y. -K. Tetrahedron Lett., 1995,36, 3473-3476.
Wilson, S. R.; Mao, D. T. J. Am. Chem. Soc., 1978,100, 6289-6291. For a review of the syntheses of 1,2,3,4-tetrahydroquinolines, see: Katritzky, A. R.; Rachwal, S.; Rachwal, B. Tetrahedron, 1996,52, 1503 1- 15070.
Crombie, L.; Jarrett, S. R. M. J. Chem. Soc. Perkin. Trans. 1, 1992, 3179.
Nomura, Y.; Ogawa, K.; Takeuchi, Y.; Tomoda, S. Chem. Lett., 1977, 693-696.
Kiaus, G. A.; Neuenschwander, K. J. Org. Chem., 1981,46, 4791-4792. Nagasaka, T.; Tamano, H.; Maekawa, T.; Hamaguchi, F. Heterocycles, 1987,26,
6 17-624.
(a) Choppin, A. R.; Rogers, J. W. J. Am. Chem. Soc., 1948, 70, 2967; (b)
Vaughn Jr., J. R.; Osato, R. L. J. Am. Chern. Soc., 1952, 74, 676-678.
Dieter, R. K.; Sharma, R. R. J. Org. Chem., 1996,61, 4 180-4 184.
Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis, Wiley-
Interscience, New York, 199 1.
Srnyj, R.; Batey, R. A. An Approach Towards the Synthesis of Martinelline
Utilizing Hetero Diels-Alder Methodology, Undergraduate Thesis, University of
Toronto, 1996.
Kobayashi, S.; Hachiya, 1. J. Org. Chem., 1994,59, 3590-3596.
Sandhu, J. S.; Sain, B. Heterocycles, 1987,26, 777-818.
Kelly, T. R.; Schmidt, T. E.; Haggerty, J. G. Synthesis, 1972, 544-555. Borrione, E.; Prato, M.; Scorrano, G.; Stivanello, M. J. Heterocyclic Chem., 1988,
25, 183 1-1835.
II ,.&
r= L u L U Y IlIll& 11111A 11111-
APPLIED 4 I M G E . Inc - = 1653 East Main Street - -. , Rochester, NY 14609 USA -- -- - - Phone: 71 61482-0300 -- -- - - Fax: 71 61288-5989