. f; p \ 9 . , r
.. kﬁxnkmuigﬁ...‘
. ﬁaan ..
bran. . a .
s t a a 2:»... Law,

3.5%. .«ﬁnw My??? ..
. n n3. . _ . . . . , . . . . «; ... new:

. can” #1:. irr. . . . . .. .. . , . . . '1
E35" .. .
.ﬁnuﬁkﬁ,

.
. I...
.

&A
t

S.
. 1.3%”...

{will
“It.“

it In...“

7.} ‘POI.I1IAu.!
“ﬂu c Eﬂnﬂurgm

3).".

I"!
.(5‘ a {‘1‘ .11..
lileetfi

i 1...!

l 9.11.! 3|.
.tlaxet’! tatsn'nblvl“.
tilt... :Vtxtlnn‘ivlt'l.‘
1...“...c.llo:tvtsllll
lhﬂ-i‘cAItlﬂﬂ.“
alizttéiw. :0;
[IF-tail,

.d.

tilt?
3.55.1

.i 4.401;. I.
“I. ‘ !\f.‘t ‘9l-"r\
.11. v H ul... :03 (min...

 

 

‘v’

 

 

.n.

 

 

 

 

 

gm é . a... ..,... .mp2...“ ._ ft

3: .

1...... .1...

 

 

 

LIBRARY
Minu-aui State
University

 

 

 

This is to certify that the
dissertation entitled

ORTHO [2+2] PHOTOCYCLOADDITION: APPLICATION
TOWARDS THE TOTAL SYNTHESIS OF 1-STERPURENE

presented by

Jason William Dahl

has been accepted towards fulﬁllment
of the requirements for the

 

 

 

 

Doctoral degree in Chemistry
4% b - LS, ..
Major rofessor’s Signature ‘1
S // 3/ (56/;
Date

MSU is an Afﬁrmative Action/Equal Opportunity Employer

- '--I--I-‘-o-l-l-l-I-0-l-l-I-I-o-o-O-l-o-l-o-o-l-o-

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:/Proleoc&Pres/ClRC/Dateoue.indd

 

 

ORTHO [2+2] PHOTOCYCLOADDITION: APPLICATION TOWARDS THE
TOTAL SYNTHESIS OF l-STERPURENE

By

Jason William Dahl

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Chemistry

2009

ABSTRACT
ORTHO [2+2] PHOTOCYCLOADDITION: APPLICATION TOWARDS THE
TOTAL SYNTHESIS OF l-STERPURENE
By

Jason William Dahl

Attempts to demonstrate the synthetic viability of the Wagner group’s ortho
[2+2] photocycloaddition and the resulting cascade of rearrangements were performed as
were attempts to expand the reaction’s synthetic scope. Sesquiterpene l-Sterpurene was
chosen as the synthesis target and the requisite alkene tethered acetophenone
photoprecursor was derived through synthetic means. The precursors were shown to
have abnormal room temperature characteristics in the 1H NMR spectroscopy. The
ketone chromophores underwent hydrogen abstraction from the adjacent alkyl tether and
closed to cyclobutenes fused to the arene. The ester chromophore variant does appear to
undergo the desired photoaddition and rearrangements, however the material has yet to
be obtained free of other similar compounds that are thought to be diastereomers/
regioisomers.

Attempts to alter the product from 4,6,5 fused ring systems to simple 4,6 were
approached using both temporary silicon tethers and intermolecular means. Neither
route was overly successful yet however useﬁIl information towards that end has been

observed in the preparation of the photoprecursor / reaction conditions

ACKNOWLEDGMENTS

I would like to thank the following for their support over this long process. I
would like to thank both my current and previous advisor, Peter J. Wagner and Robert
Maleczka, for their great advice and support when problems arose both within the lab
and outside the realm of normal grad school life. I would also like to thank my family
and the many ﬁiends I have made in graduate school for being there to help make sense
of the world. I’d like to extend a special thank you to Dr. William Wulff for rising
above his normal responsibilities in helping me ﬁnalize the thesis. Finally, I would like
to thank my wife Rachel, who through amazing patience has not killed me yet for

making her wait so long to settle down.

iii

TABLE OF CONTENTS

LIST OF FIGURES ........................................................................................................ v
LIST OF SCHEMES ..................................................................................................... vi
LIST OF SPECTRA .......................................................................................................... x
CHAPTER 1: INTRODUCTION AND REVIEW OF THE [2+2]
PHOTOCYCLOADDITION REACTION ...................................................................... 1
Discovery and Mechanism ................................................................................. 2
Regiochemistry and Stereochemistry ................................................................. 6
Useful Variations and Limitations ................................................................... 17
CHAPTER 2: SYNTHESIS OF l-STERPURENE .................................................... 21
Revising the Synthesis ........................................................................................ 39
Revising the Synthesis . . . Again ....................................................................... 47
What Remains to be Done .................................................................................. 66
CHAPTER 3: RELATED PHOTOCYCLOADDITION PROJECTS ...................... 67
Intramolecular [2+2] Photoaddition using Temporary Tethers .......................... 67
Intermolecular [2+2] photocycloaddition ........................................................... 74
Fixing the Meta Tether ....................................................................................... 80
Conclusion and Final Thoughts ........................................................................... 84
CHAPTER 4: EXPERIMENTALS ................................................................................ 85
APPENDIX ................................................................................................................... 144
REFERENCES ............................................................................................................. 155

iv

LIST OF FIGURES

Figure l: Illustrations of the possible alkene overlaps and the resulting biradicals ......... 7
Figure 2: Models for polarity directed addition .............................................................. 10
Figure 3: Lowest energy conformations of the ground state cyclooctatriene ................. 17
Figure 4: Examples of the Sterpurene family ................................................................. 21
Figure 5: Expected alkene additions ............................................................................... 26
Figure 6: The tentative product observed from the Barton ester chemistry ................... 39
Figure 7: The photochemical precursors synthesized ..................................................... 56
Figure 8: An observed byproduct ................................................................................... 57
Figure 9: The photochemical precursors for reference ................................................... 58

Figure 10: Proposed temporary tether products from their respective ortho and para
derived photoprecursors ................................................................................................... 67

Figure 11: An illustration of potential additions of unsymmetrical alkenes when using
methyl vinyl ether ............................................................................................................ 78

Figure 12: The triplet electron layout for the ortho, para, and meta examples .............. 81

LIST OF SCHEMES

Scheme 1: The discovered photochemical cycloaddition reaction ................................... 2
Scheme 2: Possible cyclooctatriene formation pathway ................................................... 2
Scheme 3: Proposed reaction pathway .............................................................................. 3

Scheme 4: The gem dimethyl derivative that allowed for detection of the initial

cycloaddition product 6 ..................................................................................................... 3
Scheme 5: Reaction mechanism, part 1 ......................................................................................... 4
Scheme 6: Reaction mechanism, part 2 ............................................................................ 5
Scheme 7: Observed regioselectivity directed by ortho substituents ................................ 6
Scheme 8: Observed regioselectivity derived from meta substituents ............................. 9

Scheme 9: Testing the regiochemical effects of polarity with Indanone chromaphores 10

Scheme 10: The effects of asymmetric tethers on relative stereochemistry ................... 11
Scheme 11: Possible pyramidalized radicals that can lead to closure to the cyclobutane12
Scheme 12: The potential facial selectivity brought about when using unsymmetic

tethers in systems with directed additions ....................................................................... 12
Scheme 13: Effects of chiral auxiliaries on the photocycloaddition ............................... 13
Scheme 14: Illustrating the alkene interactions with the auxiliary ................................. 14

Scheme 15: The proposed mechanism for the unexpected kinetic resolution of the
mixture of diastereomers derived from (7R)-(+) Camphorsultam (Sult) and racemic 43 a

......................................................................................................................................... 1 5
Scheme 16: Possible closures of the singlet biradical derived from para and ortho
acylated systems ............................................................................................................. 16
Scheme 17: The photochemical cycloaddition and rearrangement using non ketone
carbonyl groups ............................................................................................................... 18

Scheme 18: Stereochemical differences observed in 4 atom tether systems compared to
3 atom systems ................................................................................................................ 19

vi

Scheme 19: Intermolecular photocycloaddition performed by the Gilbert group .......... 20

Scheme 20: Biosynthetic pathway to l-Sterpurene ........................................................ 22
Scheme 21: Okamura’s Diels Alder synthesis of 1-Sterpurene ...................................... 23
Scheme 22: [2+2] Photocyclization and its selectivity rationale .................................... 24
Scheme 23: Undesired ring expansion resulting from direct reduction .......................... 25
Scheme 24: Key [2+2] Photocyloaddition and rearrangements ...................................... 25

Scheme 25: Friedel Crafts intramolecular cyclization and cleavage to form the desired
1,2,3 benzene ................................................................................................................... 27

Scheme 26: Regioselective halogenation for effective differentiation of the methyls ...28

Scheme 27: Ortho lithiation using a protected methyl as described by the Snickius

Group ............................................................................................................................... 28
Scheme 28: The initial retrosynthesis of Sterpurene by the Wagner group .................... 29
Scheme 29: Preparing for the intramolecular ring closure ............................................. 30

Scheme 30: Ring closure using an epoxide variant of lewis acid catalyzed Fridel Craﬂ
alkylation ......................................................................................................................... 31

Scheme 31: Dehydration of the alcohol with acidic and basic conditions ...................... 32

Scheme 32: Initial approaches to oxidative cleavage and preparing to protect the
benzylic ketone ................................................................................................................ 33

Scheme 33: Revised oxidative cleavage and the resulting reduction to a diol ............... 35

Scheme 34: ‘Protection’ of the alcohols to prepare for the conversion of the tosylate to
an iodide ............................................................................. , .............................................. 3 5

Scheme 35: The ‘failed’ oxidation attempts of the benzyllic alcohol ............................. 37

Scheme 36: The revised oxidation of the aldehyde prevented formation of the
unreactive hemiketal, leaving just the carboxylic acid and attachment of the Barton ester

.......................................................................................................................................... 38
Scheme 37: The partially reacted Barton ester expected byproduct ............................... 38
Scheme 38: Revised retrosynthesis for l-Sterpurene ..................................................... 4O

vii

Scheme 39: Differentiation of the methyls .................................................................... 41

Scheme 40: Generation of the primary alcohol .............................................................. 42
Scheme 41: An unexpected cyclization .......................................................................... 42
Scheme 42: Installing the halide in preparation for the Kumada coupling ..................... 43
Scheme 43: The failed attempts at the Kumada coupling ............................................... 44
Scheme 44: Installing of the alkene after the failure to obtain the material to perform a

successﬁil Kumada coupling ........................................................................................... 45
Scheme 45: Conversion of the methoxy to a triﬂate ....................................................... 46
Scheme 46: Failure to install a stananne in place of the triﬂate .................................... 47
Scheme 47: Bromination of 2,3 dimethylchlorobenzene ................................................ 48

Scheme 48: Creating the desired 1,2,3 substituted arene using ortho lithiation techniques
.......................................................................................................................................... 49

Scheme 49: The amide 144 proved remarkably robust to both reductive and non
reductive means of removal ............................................................................................. 49

Scheme 50: Cummin’s ortho lithiation of aldehydes in the presence of aryl chlorides ..50

Scheme 51: Installing the benzyl bromide ..................................................................... 52
Scheme 52: Performing the established chemistry on the new chlorine variant ........... 52
Scheme 53: The failed Stille coupling ............................................................................ 53
Scheme 54: The activation of the chlorine via a dissolving metal reaction .................... 54

Scheme 55: The Reike Magnesium activation of 157 and it’s quenching to make the
photochemical precursor 84 c .......................................................................................... 55

Scheme 56: The formation of the methyl photoprecursor 84 a ...................................... 57

Scheme 57: The photochemical [2+2] reaction and the resulting rearrangements
products including 88 c .................................................................................................... 63

Scheme 58: The product derived from the irradiated triﬂouromethyl ketone in acetone
65

Scheme 59: The remaining steps needed to ﬁnish l-Sterpurene .................................... 66

viii

Scheme 60: The proposed [2+2] cycloaddition and its likely rearrangements from an
ortho arrangement ............................................................................................................ 68

Scheme 61: The proposed [2+2] cycloaddition and its likely rearrangements resulting
from the para precursor .................................................................................................... 68

Scheme 62: The general retrosynthesis for the silicon tether formation ......................... 69

Scheme 63: Stepwise creation of the tether proved successful but difﬁcult to reproduce ..

.......................................................................................................................................... 71
Scheme 64: Attempts to acylate the bromide were unsuccessful ................................... 73
Scheme 65: The envisioned photocycloaddition and resulting rearrangement ............... 74
Scheme 66: An example of the Gilbert group’s success with similar [2+2]
photocycloaddition using acrylnitrile .............................................................................. 75
Scheme 67: The possible additions of an ethylene to an excited acetophenone ............. 77
Scheme 68: Reproduction of Scheme 8 from the introduction ........................................ 82
Scheme 69: The meta tethered linkage and its potential photocycloaddition and ring
opening products .............................................................................................................. 83
Scheme 70: The expected products from the singlet biradical closure ........................... 83

ix

LIST OF SPECTRA

Spectrum 1: 5-(2-methyl formate)-6-methylphenyl) 4,4-dimethylpentene 84 c in CDCl3

at 25 °C ............................................................................................................................ 59
Spectrum 2: 5-(2-methyl formate)-6-methylphenyl) 4,4-dimethylpentene 84 c in CDCl3

at -40 °C ........................................................................................................................... 60
Spectrum 3: Isolated materials from the irradiation of 84 c ........................................... 62

Spectrum 4: 1,4-diacylbenzene and 2,3 dihydrofuran prior to irradiation in deuterated
acetonitrile (300 MHz 1H NMR) ................................................................................... 145

Spectrum 5: 1,4-diacylbenzene and 2,3 dihydrofuran after 12 h of irradiation in
deuterated acetonitrile (300 MHz 1H NMR) ................................................................. 146

Spectrum 6: 1,4-diacylbenzene and 2,3 dihydrofuran after 12 h of irradiation in
deuterated chloroform (300 MHz 1H NMR) ................................................................. 147

Spectrum 7: p-thiomethoxyacetophenone and 2,3 dihydrofuran prior to irradiation in
deuterated acetonitrile (300 MHz H NMR) ................................................................. 148

Spectrum 8: p-thiomethyoxyacetophenone and 2,3 dihydrofuran after 12 h irradiation in
deuterated acetonitrile (300 MHz 1H NMR) ................................................................. 149

Spectrum 9: p-thiomethyoxyacetophenone and 2,3 dihydrofuran after 12 h irradiation in
deuterated chloroform (300 MHz 1H NMR) ................................................................. 150

Spectrum 10: a,a,a-triﬂouroacetophenone and 2,3 dihydrofuran before irradiation in
deuterated acetonitrile (300 MHz 1H NMR) ................................................................. 151

Spectrum 11: a,a,a-triﬂouroacetophenone and 2,3 dihydrofuran after 12 h irradiation in
deuterated acetonitrile (300 MHz 1H NMR) ................................................................. 152

Spectrum 12: a,a,a-triﬂouroacetophenone and 2,3 dihydrofuran after 12 h irradiation in
deuterated chloroform (300 MHz ‘H NMR) ................................................................. 153

Spectrum 13: Crude product from the irradiation of 84 b in acetone to generate 170 .154

Chapter 1: Introduction and Review of the [2+2] Photocycloaddition Reaction

The use of highly reactive materials and conditions is a staple of synthetic
organic chemistry. Unfortunately, high reactivity often comes with limitations to the
scope of the chemistry that can be performed — with the greatest limitations oﬁen
coming with the most reactive reagents/conditions. These limitations can range ﬁom
chemical issues such as regio/stereoselectivity and secondary reactions to more practical
aspects such as handling and disposal issues. As a result, the scientiﬁc community is
often looking for milder reactions while eschewing the less controllable or understood

reactions.

However, the study of the highly reactive materials and conditions often proves
to contain a wealth of interesting mechanistic information that is either overlooked in
their more mild counterparts or is necessary to create milder, more selective conditions.
Understanding these reactions is also important because the comparatively high energy
of these systems may allow for transformations that cannot be achieved otherwise. For
these reasons, high energy ﬁelds such as photochemistry prove to be interesting from

both a mechanistic and synthetic standpoint.

To this end, the Wagner group has been working for several decades towards a
better mechanistic understanding of several high energy reactions and has worked to
fully understand a previously overlooked chemical reaction. This reaction’s
“rediscove ” happened in the mid 1980’s when Keepyung Nahm observed an
intramolecular alkene addition to excited acylbenzenes (Scheme 1) while attempting to

measure the ability of a tethered alkene to quench the excited state.1 This observation

began a two decade long study into this addition, the highlights of which will be

discussed in this chapter.

8 O
O BU\¢O U
“Q . -———~ [0
.1.
0A) 0... ._
Li 0
1 2 3

Scheme 1: The discovered photochemical cycloaddition reaction

Discovery and Mechanism

As with many new reactions, the initial research focused on determining exactly
how the starting acylbenzene 1 was converted into the NMR observed cyclobutene 22
and the GC isolated cyclooctatriene 3.1 Extensive rearrangement of the starting material
was apparent for either product with no single reaction leading directly to either. The
cyclooctatriene, however, could be envisioned as being derived from a cyclobutane
intermediate 4 that results from a [2 +2] photocycloaddition of the alkene to the excited
benzene as shown below in Scheme 22 — a reaction with precedence in the literature
although most additions were 1,3 in nature. The then hypothetical cyclobutane product 4

could then be viewed as undergoing a thermal ring opening to form isolated 3.1

o
o o B"
“O '——+B“’ 03—. O

0/\)
O o
1 4 3

Scheme 2: Possible cyclooctatriene formation pathway

The light sensitive cyclooctatriene 3 was then demonstrated to be in equilibrium
with the cyclobutene 22, allowing the chromophore to be considered as an intermediate
in its production. With this information, a tentative reaction sequence was assigned
(Scheme 3) where the cyclobutane 4 is formed in a low steady state concentration and
rapidly opened to the triene 3, which itself is quickly converted to the cyclobutene 2
observed by spectroscopy.2 The existence of the cyclobutane intermediate 4 was
demostrated using a gem dimethyl substitution of the alkene terminus to destabilize the

cyclooctatriene 7, allowing cyclobutane 6 to be observable by NMR (Scheme 4).2

o
0 0 8” BUJO
hv hv ’-_/
B" A -—’ B" -—> -—~ E0
0 o A 0;“;
. o

1 4 3 2
Scheme 3: Proposed reaction pathway

0 O
O O
: ll hv : : f CE hv g
0 \ A
5 6 7 8

Scheme 4: The gem dimethyl derivative that allowed for detection of the initial
cycloaddition product 6.

With the missing step now established, the detailed mechanism is as follows:
Upon absorption of a photon of light by the chromophore 9, the oxygen’s non bonded
electron is excited to the it antibonding orbital, causing the 1: bond of the carbonyl to
cleave to a n,1r* Singlet biradical 10 as seen in Scheme 5. The singlet biradical then

undergoes a relatively rapid intersystem crossing (ISC) to a 7r,1t* triplet biradical ll.l

The relatively high rate of intersystem crossing is due to differences in how the each
radical is affected by the interaction of the electron’s spin with the spin of the
surrounding nuclei (spin orbit coupling) and hyperﬁne coupling.3 The differences in
how each radical is affected allows the normally 180° out of phase electrons (i.e. singlet)
to undergo spin rephrasing to mixed phases (1-179o phase difference) which leads to

them eventually being in phase or triplet in nature.3

 

 

 

 

O 00 111,11: * 0 31L?! 9:
0A) CA) CA)
9 10 11
o I i
O O
( A
o’ o
O
14 13 12

Scheme 5: Reaction Mechanism, part 1

The newly formed triplet biradicals repulse each other and separate as shown in
structure 11, with electron density at both the ipso and para (shown) positions relative to
the carbonyl. This migration creates a relatively electron deﬁcient aromatic ring, which
allows for an exciplex/charge transfer mechanism to operate in conjunction with
biradical partitioning to control the alkene’s stepwise addition to generate cyclobutane

13.4

The cyclobutane closure occurs when the alkyl radical’s orbital turns inwards
toward the former benzene ring. As the radical’s orbital begins to overlap with that of
the other site, the triplet is converted to a singlet biradical which then closes, resulting in

the cyclobutane 13.

Cyclobutane 13 is generally thermally unstable and opens reversibly to the
cyclooctatriene 14. This opening is accelerated by a ‘push/pull’ effect of the tether’s
oxygen donating electron density towards the carbonyl, helping to weaken the

cyclobutane’s bridgehead bond.2

As shown in Scheme 6, the reaction sequence continues once a second photon is
absorbed by the cyclooctatriene 14. While not explicitly stated, it is thought the
absorption of light by the carbonyl generates a n,1r* singlet biradical, followed by
intersystem crossing to generate a 1t,1t* triplet (15) that later undergoes a disrotary
electrocyclic ring closure to the observed cyclobutene ring 16. The cyclobutene is often

thermally unstable and can undergo ring opening to regenerate the cyclooctatriene 14.2

 

 

A
o 31m"
—* —' E0
0 o L“:
14 15 16

Scheme 6: Reaction mechanism, part 2.

Regiochemistry and Stereochemistry

With the mechanism known, work towards understanding many of the reaction’s
regiochemical and stereochemical aspects began in earnest. As both the addition of the
tethered alkene to the excited benzene (photocycloaddition) and the rearrangements can
be viewed as separate steps that fortuitously follow one another, the

regio/stereochemical traits of each step will be grouped together.

Since the discovery of this photocycloaddition, the initial photochemical addition
step has shown a relatively high degree of regiochemical control. This regiochemical
control is derived from either the interplay of the steric and electronic effects of groups

ortho to the tether or though polar effects of substituents meta to it.

o o o O
R a=OMe
..__. R _._._._. b=g3NH2
OM R C‘
0 ° 0 ° 0 d=Me
\k/

20 17 a-d 18 b-d 19 b-d

 

Scheme 7: Observed regioselectivity directed by ortho substituents. Compounds 19 b—d
and 20 are isolated materials. Selectivity is complete towards the side shown except for
the methoxy which is an 80:20 mixture favoring structure 20.

In most cases with a single group ortho to the tether, the initial addition favors
placement of the alkene towards the ortho group. The substituent’s electronic nature
does have an effect on this trend as electronegative groups tend to favor addition towards
themselves (19 b-d) while strong donors direct addition away (20) as shown above in

Scheme 7.5 Interestingly, weak donors such as alkyls (19 (I) favor addition towards

themselves.5 This suggests that competing effects may be determining the
regioselectivity — an initial exciplex which directs the alkene addition electronically, a

steric directing effect on the tether, and biradical partitioning. 5

The proposed exciplex, or excited complex, aligns the alkene donor such that it
will, barring other affects, overlap both the ipso carbon of the tether and the more
electronically deﬁcient side of the ring. This orientation allows for the most efﬁcient
charge transfer and establishes the observed bias for addition towards strong electron
withdrawing groups and away from strong donors.5 However, as the alkene approaches,
the exciplex directed approach may be interrupted by competing steric effects derived
from the tether’s (potentially) increasing interactions with the ortho substituents as

shown below in Figure 1.5

21 a 21 b 22 a 11
Favored Sterics Favored Electronics
Disfavored Disfavored
Electronics Sterics

Figure 1: Illustrations of the possible alkene overlaps and the resulting biradicals.
Please reference structure 11 for radical positioning in a general para distribution.

In the case of the methyl substituent, the exciplex directed addition is favored as
drawn in 21 h of Figure 1 as the donor is directed towards the less electronically rich
side of the ring. However, as the alkene approaches the ring, the tether is affected by an
increasing steric repulsion, twisting the tether away from the substituent and inhibiting

the alkene from achieving the overlap needed for the initial bond of the cyclobutane. As

a result, the product ﬁ'om the electronically favored side is actually disfavored overall.
The alternate approach 21 a is not inhibited by any steric interactions and can thus

become the preferred alignment.

Assuming the initial bond of the eventual cyclobutane forms from both
approaches in disproportionate amounts caused by the above interplay, the biradical
partitioning portion of the regioselective process begins. The generated 1,4 biradicals
22 a and 22 b can now: relax to starting material due to lack of a timely overlap caused
by steric interference; relax to starting material due to slow ring closure if there is poor
reactivity between the donor and acceptor; close to a cyclobutane on the electronically
favored side despite the tether’s steric repulsion inhibiting proper overlap (22 b); close
on the more sterically accessable, but electronically unfavorable side of the ring where
the radical will be donating into a relatively richer site (22 a). In the methyl substituent
case above, the last option prevails as the overlap is timely due to favorable sterics and
the closure is not greatly inhibited due to the accepting site being only moderately

electron rich.

While the interaction between steric and electronic effects is most readily
observable in the case of weak donors, this interaction exists for all donors and
acceptors. In the case of a more potent donor, thiomethoxy, the addition to the sterically
favored side is disfavored. This preference is due to the combination of the exciplex
favoring addition away from the thiomethoxy more strongly than in the methyl case
above and the biradical closure being slower on the side of the methoxy due to its
donating into a richer site - allowing relaxation to the ground state to compete more

effectively. As a result, the only observed product has the alkene adding to the aromatic

ring in the less sterically favored conformation similar to 22b followed by cyclobutane
formation. The competition between sterics and electronics is not pronounced when
using electronegative ortho groups as both effects are complimentary rather than

contrasting.

While ortho substituents have a very pronounced effect on the regiochemistry of
the addition, substituents meta to the tether also appear to have an effect on the
regiochemistry of the addition as well. However, unlike their ortho counterparts, they are
unable to have a direct steric inﬂuence on the tether and also appear to lack a conclusive
electronic affect as electron rich, alkyl, and electron deﬁcient substitutions result in
similar regioselectivity as shown in Scheme 8.‘5 Instead, meta directed selectivity is
based upon the steric effects of the substituent on the positioning of the acyl group and

the resulting polarity it brings to the molecule.6

0
o R o R R R
/AH£‘:L hv Moog :
0A A O\\' :
O 0 \...£

23 a-e 24 a-e 25 a-e 26 a-e
a=OMe a= >95%

b= F b= 85%

C: Me c= >95%

d= CF3 d= >95%

e= CN e= No Reaction

Scheme 8: Observed regioselectivity derived from meta substituents

In all cases, the acyl group is expected to position the carbonyl’s oxygen towards
the substituent to minimize its steric interactions, giving the molecule a deﬁned polarity.

The cycloaddition is expected to favor adding the alkene to the side the dipole is

pointing towards in order to reduce the polarity of the system during the addition as seen
in Figure 2.6 To demonstrate the importance of polarity and not the meta substituent
itself, excitation of tethered indanone 28 demonstrated addition away from the

methylene (29) and in agreement with the carbonyl’s set polarity (Scheme 9).6

   

27 a 27 b
Disfavored Favored

Figure 2: Models for polarity directed addition. The direction of the alkene’s electron
donation relative to the dipole is indicated by the arrows.

3531.};— RE:

28 29 30
Isolated

Scheme 9: Testing the regiochemical effects of polarity with Indanone chromaphores

When comparing the effects of substitution meta to the tether (polarity) to those
ortho to the tether (steric and electronic), the effects of polarity have been shown to
supersede those of the ortho group in some examples, but the conﬂicting directing

affects can also result in no observed reaction.6

Beyond the regiochemical balancing act that the alkene addition undergoes, the

reaction’s stereochemical control proves to be as similarly intriguing. Stereochemical

10

control may be introduced via one of two means; either a chiral substitution on the tether

or a chiral substitution on the aromatic ring.

The use of an unsyrnmetric tether such as one in Scheme 10 has been shown to
have low to moderate stereospeciﬁc effects on the ﬁnal product by directing the initial
photoaddition such that the tether’s larger group R1 is directed away from the forming
cyclobutane ring and the R2 substituent.7 The degree of the speciﬁcity, increases as the
steric bulk of R1 increases.7 For example, when the tether is substituted with a methyl,
as in 31 a, a 41% de is observed while replacing the methyl with an isopropyl results in a
67% de seen from 31 b. Increasing the steric bulk at the alkene (R2) as in 31 c also

increases the diastereospeciﬁcity.7

O O 0
«0—— 4— R
0M 0 -. >”' '. 2,, 2
H ’R1 0 ",R OJ
1

R1

 

 

31 32 33 34
a R1=Me R2=H a 41% de
b R1=1Pr R2=H b 67% de
c R1=Me R2=Me c 80% de

Scheme 10: The effects of asymmetric tethers on relative stereochemistry

The steric control introduced by the tether does not necessarily stop after the ﬁrst
of the cyclobutane’s bonds is formed. If the alkene is substituted with an R group as
shown below in Scheme 11, then a secondary set of steric interactions allows for
selective closure that positions the R away from the cyclohexadiene ring.7 Due to the
triplet nature of the biradical at this point in the mechanism, the initial stereochemistry of
the alkene is normally lost and the closure occurs after the biradical has achieved the

lowest energy conformation 35 b.7

11

 

35a 35b

Scheme 11: Possible pyramidalized radicals that can lead to closure to the cyclobutane.

As a corollary, one can theorize that since the direction of the alkene addition is
controlled in a molecule with an an othro substituted tether as shown below in Scheme
12, the steric interactions of the unsymmetric tether may be seen as effectively
controlling the addition’s facial speciﬁcity. In the example below, the addition should
occur towards the ketone while the steric interactions of the R group with the excited
phenyl ring will bias the addition to the upper face (37). This selectivity is due to the
steric repulsion between the R group and the ring shown in 39, which will twist the
tether out of alignment for efﬁcient biradical overlap, increasing the likelihood of
reversion to starting material.

fast
1% biradical
O

| closure
——-—'D

do 37
O Favored
slow
R biradical
\ 0 O O closure

1 z
36 K/PH

39 4O
Disfavored Disfavored

 

Scheme 12: The potential facial selectivity brought about when using unsymmetic
tethers in systems with directed additions.

While decent to good relative stereochemical control can be obtained when using
chiral tethers, chirality may also be induced in otherwise non chiral systems by using a
chiral auxiliary. AS one would expect, the degree of diastereospeciﬁcity is heavily
dependent on the chiral auxillary and can range from low to good selectivity with the
major diastereomer being generated when the alkene approaches from the face opposite
the auxiliary’s steric bulk.8 However, the exact mechanism for the transfer of

stereospeciﬁcity was not readily apparent.

o o o 0
Aux
1* ”Hi

41 a-d 42 a-d 43 a-d

 

Auxiliary DE
a None (OH) 0%
b H Menthol 15%

c (+) Camphorsultam 90%
d (-) 2.5-dimethylpyrrolidine 90%

 

Scheme 13: Effects of chiral auxiliaries on the photocycloaddition

Normally, one would expect to see differing steric inﬂuences on the cyclobutane
formation due to the facial bias brought about by the auxiliary. Using 45 shown below
as an example, the auxiliary can extend into the space of the reaction and affect both
steps of the addition.8 The initial cyclobutane bond formation may be affected indirectly
as the internal carbon of the alkene is not likely experiencing direct steric interactions
with the auxiliary.8 The terminus of the alkene, however, should be susceptible to the
auxiliary’s steric effects and may be held far enough away from the excited aromatic
ring to prevent the internal carbon from having efﬁcient overlap, slowing bond
formation on that face. Assuming the initial cyclobutane bond does form, the chiral

auxiliary is expected slow or halt the closure of the biradical by physically preventing

13

orbital overlap, increasing likelihood of reversion of the 1,4 biradical intermediates to

staring material 41.

o O 0 o
Aux ﬁ WA“ fr‘ilg‘fux
O o OK/K o O

44 41 45
Favored Sterics Disfavored Sterics

Scheme 14: Illustrating the alkene interactions with the auxiliary, the general direction
of which is illustrated by its bond to the carbonyl.

While the addition speciﬁcity is easily explained, the actual facial selectivity is
not. The chiral carboxylic acid derivative is expected to be perpendicular to the ring
according to molecular modeling and there appears to be no signiﬁcant energy
difference to favor one face over the other for the tether.8 With no real reason for tether

facial selectivity, the high degree of selectivity is unusual.

The lack of facial selectivity, when combined with the observation that the de’s
for camphorsultarn augmented 41 c increased as the reaction proceeds (30% do at 20%
completion versus 90% de at completion) suggests that another source of
diastereoselectivity may be present. To observe this other source, the Wagner group
created a mixture of diastereomers by reacting (7R)-(+) Camphorsultam with the
racemic cyclooctatriene carboxylic acid 43 a to produce 43 c and 47 shown in Scheme
15. When exposed to the photochemical conditions, this mixture of diastereomers saw
the conversion of 47 to 43 c with high selectivity. The Wagner group proposed that the
cyclooctatrienes are undergoing a ring opening to tetraene 46 and then selectively

reclosing to the lower energy conformer 43 c in a process similar to tetraene closures

l4

seen in literature.9 As a result of this unexpected kinetic resolution,8 good

stereospeciﬁcity can be observed in these systems.

Sult

 

Me

 

 

43 c 46 47

Scheme 15: The proposed mechanism for the unexpected kinetic resolution of the
mixture of diastereomers derived ﬁ'om (7R)-(+) Camphorsultam (Sult) and racemic 43 a.

While the regio and stereochemical issues surrounding the cyclobutane formation
often seem to involve an interesting interaction of electronic and steric effects, the
selective closure of the cyclooctatriene to a cyclobutene ring is somewhat more
straightforward. The regiochemistry of the closure appears to be directed by the ability
of the molecule to stabilize the triplet biradical through conjugationm’” In the para
substituted cases shown above in Scheme 15, the biradical 15 has its density localized
alpha to the carbonyl and the tether’s oxygen. This stabilization effectively places most
of the energy of the excited state in 2 of the 3 available double bonds and these bonds
undergo the disrotatory ring closure to produce 16 preferentially.10 Ortho substituted
examples are not as simple as all three alkenes are undergoing excitation as the carbonyl
and oxygen are placed at opposite ends as shown by 48 in Scheme 16. In these cases,
the cyclobutene closure towards the carbonyl 50 is expected to be the major product as

its carbonyl allows for better stabilization of the excitation energy.”

15

O
O O
15 48
Excited Para Derivative Excited Ortho Derivative

 

16 49 50 51
Major Minor Major Minor

Scheme 16: Possible closures of the triplet biradical derived from para and ortho
acylated systems.

When the oxygen linkage of the tether is replaced with a methylene, both closure
options are observed as the localization of the biradicals is no longer as well deﬁned.10
Fortunately for the synthesis discussed in the next chapter, good selectivity is expected
in the ortho derived system as only one double bond is positioned adjacent to the
carbonyl, allowing it to close in a manner similar to the linear fused ring structure 50 in

Scheme 16.

In both the ortho (48) and para (15) substituted tether systems shown in Scheme
16, the stereochemistry of the cyclobutene formation favors the cyclobutene being
positioned trans to the cyclopentane ring. The exact cause of this selective closure is
currently unknown, but it has been suggested that the direction of the closure is likely set

7,11

before irradiation. With this is mind, the selectivity may be explained by considering

16

that ground state conformer of the cyclooctatriene 14 would position the cyclopentane
ring on the top face in a pseudoequatorial arrangement relative to the cycloctatriene
while the diene pair that is expected to close is angled down and away as shown in
Figure 3. This preferred conformer then absorbs a photon of light and the ring ﬂattens
somewhat7, but retains the original conformational bias. The biradical then closes and
its rotation is determined by the remaining bias the excited cyclooctatreine has - in this
case opposite the cyclopentane. This explanation would account for the high selectivity
found in both ortho and para systems, which cannot be explained directly by sterics

itself.

 

14
Figure 3: Lowest energy conformations of the ground state cyclooctatriene. Markers
are included for clarity.
Useful Variations and Limitations

While the intricacies of the photocycloaddition and the resulting cascade of
rearrangements are interesting from an academic aspect, the reaction itself would be of
little practical use if it is not open to variations. To this end, the Wagner group has

found several variations to be viable in both the carbonyl and the tether.

17

Altering the carbonyl ﬁ‘om a ketone to either an ester or nitrile has been shown to
generate the expected addition and rearrangement products with varying degrees of
success. While the reaction is virtually unchanged in terms of products formed when
using esters,11 nitriles often Show reduced selectivity in the closure of the cyclooctatriene
to cyclobutenes 54 b and 55 when compared to ketone variants.11 All of the carbonyl
types are expected to generally produce the same products regardless of activation -

direct radiation or triplet sensitization.ll

(ZOO—*9: PO

52 a, b 53 a b 54 a, b
a) R=CN
b) R=COOEt

Scheme 17 : The photochemical cycloaddition and rearrangement using non ketone
carbonyl groups

While some modiﬁcation of the carbonyl can be performed, the tether may also
be altered in both its composition and its length. The alteration of the tether from the
benzene conjugated ether to either a non conjugated ethereal form or simple
hydrocarbon chains results in little to no change for the initial 2+2 addition.10 The loss
of the oxygen conjugation does result in a notably slowed rate of the cyclobutane ring
opening to the cyclooctatriene as the push/pull effect that was observed prior is no
longer applicable.10 Studies by the Gilbert group suggest that silanes may also be

acceptable tether components.12

18

 

14 55

Scheme 18: Stereochemical differences observed in 4 atom tether systems compared to
3 atom systems. Ortho derived systems show similar results to the para derived system
shown above.

Adjusting the tether length to 4 rather than the standard 3 does allow for the
addition to occur. The rate of the cycloaddition is roughly 2 orders of magnitude slower
for the initial addition and stereochemistry of the resulting rings is altered from the cis
formation of the cyclopentane ring to a trans cyclohexane ring as shown in Scheme 18
above.H The stereochemistry of the diene ring closure when using a 4 atom tether is cis
to the stereocenter that remains in the cyclooctatriene — opposite the results of the 5
member ring system. Other tether lengths failed to undergo addition on a detectable

scale.

Finally, in what may be considered one of the most potentially important variants
of the chemistry, an intermolecular version had been successfully attempted by the
Gilbert group.”16 However, the addition and eventual opening to a cyclooctatriene
were observed, the reaction is not very practical as extreme ratios of donor versus

acceptor must be used.

19

N
N? NQ MeO // NQ/\Q/\
I I —> I I I + +
\ N
0M8 0”” Q" \\ /OMe
N
57 58 59 60
Scheme 19: Intermolecular photocycloaddition performed by the Gilbert group. The

reaction was performed neat in acrylonitrile to generate a relative ratio of 2:7:1 of 58, 59,
and 60 respectively.

While the ability to undergo intermolecular addition is a very promising step
forward, it also greatly complicates product predictions. One would expect aspects such
as regiochemistry to be altered dramatically as the products can now be derived from
any site that has sufﬁcient radical density as demonstrated by the additions at both the
ether 58 and at the nitrile 59 in Scheme 19. One can theorize that the exciplex driven
alignment will still be observed for the additions. Furthermore, due to the no longer
‘ﬁxed’ site for the donor (i.e. the kinetically favored 4 position of the tether), the
regioselective addition of a non-symmetric donor and the resulting biradical partitioning
would need to be more fully considered, which is also illustrated in the scheme above by
58 and 59). Further stereochemical consideration of the non-symmetric donor’s addition
also needs to be addressed. All of these issues remain to be examined, but so far appear

to have been untouched, possibly due to the expected complexity involved.

Having covered in detail many of the aspects of the Wagner groups 2+2
photocycloaddition, the following chapters will look at work on demonstrating its
synthetic viablilty in the attempted total synthesis of Sterpurene as well as expanding on
its scope from the specialized 4,6,5 fused products to more synthetically attractive

alternatives.

20

Chapter 2: Synthesis of l-Sterpurene

While the photocycloaddition’s mechanism and its regiochemical/stereochemical
traits have been investigated, an example of the chemistry’s utility through total
synthesis remains unaddressed. While a total synthesis can be viewed as a formality, it
allows for a demonstration of the chemistry in systems that are often more complex than
those used to explore the basic characteristics of the reaction. Furthermore, any issues
that arise during the synthesis can help illustrate facets of the chemistry that either may
not have been brought to light in previous testing or were avoided in an effort to promote

the viability the chemistry.

   

1-Sterpurene Sterpurene-3.12.14 triol Sterpuric Acid Hydroxysterpuric Acid
61 62 63 64
1'2

t‘?’2‘i” “lo/‘5

5"6x ,8\9’ ‘14

1'3 7

Figure 4: Examples of the Sterpurene family.

In order to demonstrate the synthetic utility of its [2+2] photocycloaddition, the
Wagner group began searching for synthetic targets that would be amiable to the
relatively unique 4,6,5 fused ring system the chemistry produces. It found a family of
sesquiterpenes, Sterpurenes, that matched the basic structure — examples of which may
be observed above in Figure 4.17’18 l-Sterpurene 61 is believed to be the parent

molecule of the family.18

21

The Sterpurene family of terpenes is composed of metabolites that were isolated
from the fungus Chondrostereum purpureum by the Ayer group in the earlier 1980’s.17’
’8 This fungus tends to target various deciduous species and causes “silverleaf disease” -
a silvering of the host’s leaves by the separation of the epidermal cells from the palisade

cells, creating a waxy, reﬂective surface on the leaf which may result in host death.19

This family of harmful metabolites is believed to be derived naturally using an
extensive series of rearrangements starting from famesyl pyrophosphate 65.20 The
reaction sequence is thought to begin with an initial closure to a cycloundecatriene ring
which then undergoes an elaborate series of rearrangements to generate l-Sterpurene 61
as shown below in Scheme 2020’ 21 The ﬁrst synthesis of l-Sterpurene was performed
less than a year after its isolation by Murata, Ohtsuka, Shirahama, and Matsumoto

through emulation of the biosyntheitic route.22

4.
m —47% am #W
R 1 I +| I
66 67 68

65
R=pyrophosphate
:: +
«— .. Wm
H

E H

61 71 70 69
1 -Sterpurene

Scheme 20: Biosynthetic pathway to l-Sterpurene

Since the initial synthesis, several total syntheses have been performed. As one
might expect when looking at the molecule, the introduction of the cyclobutane ring in a

stereoselective manner is a fairly crucial reaction as fewer methods exist for 4,6 fused

22

ring junctions than the more common 5,6 ring system. Aside from an interesting quasi
Favorski reaction by the Harmata group,23 most groups either derived the carbon
skeleton from an already existing cyclobutene or introduced a cyclobutane ring through

an [2+2] enone photocyclization.

A notable example of the ﬁrst type was performed by the Okamura group to
demonstrate a stereospeciﬁc intramolecular vinyl allene Diels-Alder reaction (Scheme
21)2"’ 25 (This work was later emulated by Norbert Krause in his synthesis”). The
transference of central to axial chirality (73 to 74) and back (74 to 75) with high
speciﬁcity allowed for excellent control over the both the absolute and relative
stereochemical issues. The use of benzenesulfenyl chloride promotes the

rearrangement to the allene followed by an inverse demand Diels-Alder reaction in one

 

 

pot.
o
: *9” PhSCI C H\ Ph‘gm—jH
Q—josph __. f_
/ / /
72 73 74
SOPh

SOPh
H; i
61

76 75
1-Sterpurene

Scheme 21: Okamura’s Diels Alder synthesis of l-Sterpurene

While using a cyclobutene ring as the basis of the synthesis proved to be a viable

approach, most groups elected for a relatively standard [2+2] photochemical reaction

23

with an ethylene and an enone instead.'9’ 27‘3 1 While enone photocycloaddition can allow
for stereospeciﬁc additions through the use of chiral iminium ions, all of the groups
choose a standard enone cycloaddition route. In most cases, the addition proceeded
with good to excellent diastereoselectivity - with complete selectivity reported in the
case shown in Scheme 22.31 The observed selectivity can be explained using a model
proposed by Wiesner,32 who suggested that the carbonyl and the or carbon are trigonal in
nature while the [3 carbon is pyrimadalized - much like excited conjugated dienes. The

more stable pyrimadalization of the [3 carbon is expected to direct the facial selectivity.

 

311,71 *
o p o
1) hv ,9) [2+2] 0. H
/° 2) ISC /° CI
0 O O 0’
77 78 79

 

78 a 78 b
Favored Disfavored

Scheme 22: [2+2] Photocyclization and its selectivity rationale. An alkene is shown in
78 a and 78 b to help illustrate the conjugation with the oxygen radical suggested by
78.

One unexpected issue that was observed while reviewing the published
Sterpurene syntheses was the propensity of the cyclobutane ring to expand to a
cyclopentane during attempts to reduce an adjacent carbonyl to the requisite methyl as
shown in Scheme 23.31 Fortunately, other studies found that dicyclohexylcarbodiimide

(DCC) mediated couplings of the alcohol with methyl iodide to form the iodide followed

24

by subsequent lithium halogen exchange and quenching with water can sidestep this

issue.23

K

00

 

80 81 82 83

OTs

Scheme 23: Undesired ring expansion resulting from direct reduction. Relative ratios of
81, 82, and 83 are 3:1 :1 respectively.

Having brieﬂy touched on the strategies used in the previous total syntheses of
Sterpurine and the relevant problems that were observed, the Wagner group proposed a
distinct route based off of its intramolecular cycloaddition reaction. The group
envisioned having the photocycloaddition as a late stage transformation of the 1,2,3
trisubstituted arenes represented by 84 a-c into the corresponding core 4,6,5 fused ring

system of 88 a-c as shown in Scheme 24.

\ J
o
O ___. __.. C ____.
R R O O
R R

84 a-c 85 a-c 86 a-c 87 a-c 88 a-c

 

a) R= Me, b) R= CF3. c) R= OMe

Scheme 24: Key [2+2] Photocyloaddition and rearrangements.

The key step of this synthesis offers an excellent chance to observe the

interaction of exciplex/charge transfer, sterics, and biradical partitioning effects as the

25

selectivity is not as straightforward as in previous ortho substituted examples. The
regiochenristry of the initial [2+2] addition, both the exciplex and the steric directing
effects were expected to favor addition of the alkene towards the acyl group as in 89 b.
However, steric interactions of the tether with the methyl may twist the tether out of

alignment, resulting in a relatively slow closure to the cyclobutane ring 85.

311,71 *
\
\\
R. i: > *_ _’ R \
\é\ R @ﬁ
0*; o
O
893 89 89b
Disfavored Favored

a) R= Me. b) R= CF3, c) R= OMe

Figure 5: Expected alkene additions.

The later cyclooctatriene to cyclobutene closures (86 to 87and 88) are expected
to be slightly problematic as the excitation is no longer localized along 2 of the alkenes
as many of the previous oxygen tethered examples were. As a result, 2 products, 87 and
88, are likely with 88 expected as the major product due to the carbonyl being the
preferred stabilizing group. From a practical point of view, this regiochemical issue is
not overly vital so long as the products are separable as the undesired product 87 can
theoretically be reverted to the cyclooctatriene 86 then resubj ected to the photochemical

conditions to obtain more of the desired material 88.

Turning away from the reaction of interest, the primary issue of the synthesis was
perceived to be obtaining the desired 1,2,3 substituted arene in reasonable yields and

purity. A classic approach to this problem is to have a suitably functionalized precursor

26

undergo a ring closure followed opening of the newly formed ring to selectively
introduce the third substituent as shown below in Scheme 25. In this case, an
intramolcular Friedel Craft acylation (or a derivative thereof) can be used to bring about
the ring closure. Manipulation of the resulting carbonyl 91 to create either an enol ether
92 or a diketone (not shown) will allow for oxidative cleavage and realization of the
1,2,3 trisubsituted ring with no regiochemical issues. A potential downside of this
approach is its relative inefﬁciency in obtaining other carbonyl derivatives as
chromophores besides the one initially formed from the cleavage. While other
chromophores will require additional steps to obtain, this approach has extensive

precedence and chances of an insurmountable failure at this point are low.

0033333.

90 91 92 93

 

Scheme 25: Friedel Crafts intramolecular cyclization and cleavage to form the desired
1,2,3 benzene. The various R’s depend on the reaction conditions selected.

Alternatively, manipulation of inexpensive natural products that already possess
the desired 1,2,3 substitution pattern should be considered as well. A derivative of
commercially available 2,3 dimethylphenol or 1 halo 2,3 dimethylbenzene can be
envisioned as being made into the a usable substrate through the use of selective
halogenation of the 2 position’s methyl followed by elaboration for the alkyl tether.
With the tether in place, a palladium mediated addition of the carbonyl at the R can be
performed. This approach avoids the extra steps required to obtain derivatives as the

desired carbonyl group can be installed directly. However, separation of the initial

27

mixture of bromide regioisomers may be difﬁcult or not practical due to scale -

especially if the bromination’s selectivity is not overly high.

Br

EI—CCCI.

94 95 96
Ma'or Minor

Scheme 26: Regioselective halogenation for effective differentiation of the methyls. R=
halogen or an ether.

A third option for efﬁcient creation of the desired 1,2,3 arene is to use a directing
technique such as ortho lithiation to introduce one or two substituents in a controlled
manner. Ortho lithiation, while very potent, does have some issues regarding alkyl
groups adjacent to the directing group along with a fair amount of functional group
intolerance. Assuming the middle substituent is chosen as the directing group, the
removal of a proton from an adjacent methyl group will be an issue as its removal is
faster than an arene proton. This will force use of either a creative reaction order or a

protection sequence shown in Scheme 27 of a normally inert portion of the molecule.

(TMS)2CH o (TMS)20H o (TMS)20H o
drum "'bULi CKKNEQ RI dmaz
H - Li+ R
97 98 99

Schem3e327: Ortho lithiation using a protected methyl as described by the Snieckus
Group.
Considering these three options, the classic ring closure/ring opening approach

was deemed the most promising as few derivatives varying at the carbonyl derivatives

were planned.

28

\ I
H H
O
61 88a 84a

1—Sterpurene

? o
0/
60% C: (3;: 4: O
OH
102

104 103 101

Scheme 28: The initial retrosynthesis of Sterpurene by the Wagner group. A similar

route was planned for the triﬂuoroacyl version, while a haloforrn reaction on 84 a and
subsequent alkylation would generate the desired ester variant. (*) denotes protection.

With the approach to the 1,2,3 trisubstituted arene in place, one can now form the
remainder of the retrosynthesis. l-Sterpurine 61 was expected to be formed after
reduction of the photoproduct 88 a’s cyclobutene, haloforrn cleavage of the ketone’s
methyl to an acid, and subsequent reduction of the carbonyl to a methyl (Scheme 28).
The photochemical precursor 84 a was envisioned as being derived from a nickel
mediated Kumada coupling of a vinyl Grignard reagent with an alkyl iodide 100.34’35
The alkyl halide was to come from the subsequent reduction and halo genation of the

ester 101 while its benzyllic ketone was to be protected.

The carbonyls of 101 can be seen as coming from an oxidative cleavage of a
conjugated alkene, which is derived from the alcohol 102. The alcohol is expected from
a Lewis acid mediated Friedel Crafts addition to an epoxide 103. The epoxide variant
was chosen due to the reduced chance of possible regiochemical dehydration issues
when compared to products derived from a methyl ketone. The alcohol’s relative

stereochemistry to the methyl was not expected to be important as either cis or trans

29

isomers are expected to be capable of forming the desired alkene. The epoxide 103 is

to come from a standard epoxidation of oleﬁns with a peracid.

The mixture of alkenes used in the epoxidation was expected to be derived from
a Wittig reaction with an aldehyde 104. The aldehyde is from alkylation of
commercially available 2-methylbenzylbromide 105. To further improve the economy
of the synthesis, 2-methylbenzylbromide 105 can be formed via free radical bromination

of o-xylenes.

In a forward sense, the synthesis began with the slow addition of a dry
isobutylaldehyde and 2-methylbenzylbromide solution to a 65°C suspension of
powdered sodium hydroxide and tetrabutylammonium iodide in benzene as shown in
Scheme 29. The solid/liquid phase transfer catalyzed reaction produces the desired

aldehyde 104 in decent yields with no NMR detectable byproducts derived from addition

0 Ph3PEtBr
066% KOtBu mCPBA

NaOH 83% 97%
TBAI
66%

to the aldehyde 104.36
@F Br

Scheme 29: Preparing for the intramolecular ring closure

 

 

 

The puriﬁed aldehyde 104 is then added to orange ethyl phosphonium ylide to
form the alkene 106 in good yields. From the NMR of the crude material, there appears
to be complete selectivity for one isomer as the allylic methyl(s) show only 1 set of
doublets while 2 would normally be expected if both regioisomers are present. However

the peak pattern is more complex than a doublet with allylic splitting, so the possibility

30

of the other isomer being beneath it cannot be ruled out. The pattern for the vinyl
protons is relatively complex and also cannot be used to rule out the possibility of a

second isomer. The possible cis/trans mixture was taken to the next step.

The clear oil of alkene 106 was then treated with 3-chloroperbenzoic acid to
easily produce the epoxide 103 with good yields as a relatively clean oil. From the
spectra of the crude and puriﬁed epoxide, it became apparent that only one regioisomer
of the alkene existed — the cis alkene. For the purposes of the synthesis, the crude

epoxide is pure enough after the workup to proceed with the ring closure.

The ring closure was performed by adding concentrated titanium tetrachloride
(tin tetrachloride may be used as well) dropwise to a 0 °C solution of epoxide 103 in
dichloromethane to yield a golden solution that darkens with time (Scheme 30). The
reaction produced the desired alcohol 102 exclusively, which is then puriﬁed via

recrystallization in hot hexanes to yield a white solid.

0
TiCI4 \ ,3\|
M (E? 4\
OH / 6:

é
86% . a,

 

103 102
racemic

Scheme 30: Ring closure using an epoxide variant of lewis acid catalyzed Fridel Craft
alkylation.

While there were some initial concerns regarding the regioselectivity of the ring
closure to an alcohol, the reaction was performed according to precedent established by
the Taylor group.37’38 Their work demonstrated that the ring closure favors 6 member

ring formation while similar 5 member rings are relatively inert to simple phenyl rings.

31

Furthermore, they have demonstrated that terminal epoxides at the 6 position will not
undergo cyclization unless electron rich arenes are used - even then yields are low.
Efﬁcient cyclization requires the 6 position to be substituted for a reaction to occur, thus
the methyl and triﬂuoromethyl groups needed for 84 a and 84 b respectively should be

installed by this step.

The dehydration of the alcohol 102 to the alkene 107 proved to be slightly less
straight forward than expected as the normal acid techniques produced not only the
desired alkene, but an inseparable alkene 108 that is likely the product of methyl
migration after the generation of a carbocation. The ratio of desired material versus
byproduct could be improved by decreasing the reaction temperatures; however 108

could not be entirely avoided under acidic conditions.

TfOH
0” 77%

 

108
Tentative
Tf20
DMAP
81 -94%
1 09
Tentative

Scheme 31: Dehydration of the alcohol with acidic and basic conditions

The byproduct 108 was avoided by converting the alcohol 102 into a triﬂate in
dichloromethane at 0 °C in the presence of pyridine and dimethylaminopyridine

(DMAP), which then eliminated under basic conditions. The reaction must be

32

performed at 0 °C or colder and with triﬂic anhydride that is free of acid impurity
otherwise a new inseparable byproduct tentatively assigned 109 is observed in the proton

spectrum in amounts that are calculated to be up to 10% of the isolated yield.

Other leaving groups were attempted such as triﬂuoroacetate esters and tosylates,
but the elimination step failed to occur without addition of more potent bases. In
retrospect, ﬁrrther attempts with the more affordable leaving groups should have been
attempted in a more polar solvent such as tetrahydrofuran, since they do not have the

ethereal solvent polymerization issues triﬂic anhydride does.

One pot approaches to opening the alkene 107 to the benzyllic ketone and
carboxylic acid were attempted with uninspiring results. Ozonolysis followed by
oxidative workup provided the material, but in very low yields. Oxidative cleavage
using catalytic osmium tetraoxide and oxone (as seen below in Scheme 32) slowly
provided the acid 110 and hemiketal 111 in low yields in reactions that failed to go to

completion.39

 

 

0304 TMSCI
Oxone MeOH
26 - 50% 0 95%
OH

11

Scheme 32: Initial approaches to oxidative cleavage and preparing to protect the
benzylic ketone. Protection of the benzyllic ketone as a ketal failed.

The oxone/osmium tetraoxide reaction was looked at by both our group and by
Dr. Ben Travis in the Borhan group. In our hands, the reaction was incomplete after 1

week of stirring at room temperature even with additional catalyst loading up to 5 times

33

higher than generally suggested (40% of the starting material consumed of which 26% is
111). The hemiketal 111 was the only isolatable product related to the desired acid,
although there were several other undetermined products as well. Dr. Travis observed
similar difﬁculties in driving the reaction to completion, however he was able to obtain
the acid 110 in combination with the ketal 111 in higher yields that are reﬂected in the
range given. The ratio of these two products is variable and from work in our lab, the
acid appears to be dependent upon the workup/puriﬁcation as only 111 is observed if

aqueous lithium bromide washes are used to remove the dimethyl fonnamide solvent.

Both the acid and ketal products were used to produce a methyl ester 101 when
subjected to in situ generated acid in methanol. Subsequent attempts to protect the
benzyllic ketone of 101 as a ketal proved fruitless as only the starting material was
recovered. Other means of protection such as enol ethers were not tried as a uniform

route that could be apply to both the methyl and triﬂuoromethyl derivatives was desired.

While attempts to protect the ketone 101 as a ketal were being performed, other
methods to open the alkene of 107 were tried in an effort to improve overall yields. The
alkene 107 was oxidized using a catalytic amount of osmium tetraoxide with
trimethylamine n—oxide as a co-oxidant, producing the white solid cis diol 1 12 in decent
to good yields and purity after a reductive workup. The diol, although relatively pure,
has to be crystallized in hexanes before the next step, as a byproduct is inseparable from

the aldehyde product 113 by column after the oxidation.

34

 

 

 

0804 O OH
Me3NO @3 NalO4 LAH
00 o 0..
60-84% 0 H 92% 96%
107 112 113 114

Scheme 33: Revised oxidative cleavage and the resulting reduction to a diol.

The diol 112 is then oxidized to the benzyl ketone and aldehyde 113 via sodium
periodate. The resulting clear oil, once puriﬁed, must be stored under an oxygen free
environment as spontaneous oxidation of the aldehyde to the solid acid 110 occurs quite
readily. With this new ring cleaved product available, work towards protecting the
benzylic ketone of 113 as an ether rather than a ketal began. Reduction of both

carbonyls using lithium aluminum hydride in tetrahydrofuran yielded the diol 114 in

 

 

good yields and purity.
TBSOTf
0” TsCl T50 Hunig's 750
Pyridine Base
OH OH + 0
58% Yield not I
recorded
1 14 1 15 1 16 1 17

Scheme 34: ‘Protection’ of the alcohols to prepare for the conversion of the tosylate to
an iodide. The halide was expected to be converted into a vinyl group using a variant of
Kumada coupling as shown in Scheme 28 (page 29)

The primary alcohol of 114, which was envisioned as eventually being converted
into an iodide, was selectively ‘protected’ as a tosylate 115 while generating an ethereal
impurity 116 derived from intramolecular attack when initially attempted. When using
low temperature conditions (-30 °C) with slight excess of tosyl chloride, ether byproduct

116 was 20% of the reacted material (incomplete reaction, stopped after 30 h). When

35

the reaction conditions were changed to 6 equivalents of the tosyl chloride at rt
(complete reaction in ~ 8 - 10 h), the ether 116 was less than 5% of the material by crude
nmr. The protection of 115’s benzyl alcohol was attempted using dimethyl-tert-
butylsilyl triﬂate at -7 8 °C after no reaction was observed with dimethyl—tert-butylsilyl
chloride and imidazole at room temperature. However, the desired dimethyl-tert-
butylsilyl ether was not observed, instead an elimination of the benzylic alcohol occurred

to yield the alkene 117.

With the difﬁculty in protecting the benzyl alcohol, presumably due to steric
issues, concerns about the ability to oxidize the alcohol back up to the ketone arose
should the protection and Kumada coupling (or other chemistry to install the vinyl
group) ever occur. As a result, a small portion of benzyl alcohol 115 was oxidized to the
benzyllic ketone using either Swem, pyridinium chlorochromate, or Dess Martin
conditions. Each of the techniques yielded a clear, oily compound that appeared
overoxidized by ‘H NMR. The desired transformation of the benzyl alcohol to the
ketone had occurred, but a second oxidation on the benzylic methylene of the tether
seemed to have occurred as well (118). The spectroscopic evidence was contradictory as
the 13 C NMR and mass spectrum data suggested the desired ketone 119 had been
formed, but DEPT and 1H NMR data suggested otherwise as no protons were tied to the
central methylene carbon. The reaction was later shown to have made the desired
ketone 119, but at the time the reactions were deemed as failures due to the conﬂicting
data and the plan to oxidize the alcohol to a ketone following tether completion was

abandoned, effectively terminating the planned route.

36

 

Martin
OH 0 °' 0
Yield not
recorded
1 1 5 1 18 1 1 9

Scheme 35: The ‘failed’ oxidation attempts of the benzyllic alcohol. The reaction did
produce 119 exclusively. Slow rotation of the tether allowed the diastereotopic
methylene protons to appear as baseline at room temperature. They are observable at
lower temperatures in chloroform.

With the benzyllic ketone proving difﬁcult to protect and seemingly difﬁcult to
recover after reduction, the addition of the alkene to the tether needed to be rethought as
the planned method was incompatible with an exposed ketone. Ideally, a suitable
substitute would be relatively non nucleophilic and non electrophillic due to the nature
of the carbonyl’s carbon and oxygen atoms respectively. Towards this end, a free
radical addition of the remaining alkene was planned. Conversion of the previously
obtained carboxylic acid 110 to a Barton ester 120 followed by a free radical induced
decarboxylation and subsequent trapping of the radical by an allylic tin species was

expected to generate the photochemical precursor.

While a direct route to acid 110 was already known through Dr. Travis’s work
(page 33), yields of it were very low and the hemiketal ‘byproduct’ 11 1 was soon was
shown to be inert to the dicyclohexylcarbodiimide (DCC) coupling conditions needed to
generate the Barton ester 120. This required a new route to generate the acid -
preferably free of its unreactive counterpart as it would avoid dealing with equilibration
between the two forms. After several attempts using oxygen oxidations in CFC solvents,
which were ﬁnicky at best, pyridinium dichromate (PDC) was found to oxidize the

aldehyde 113 to the desired acid 110 as the sole product.

37

S

NaO.
Cl) 0 Adj 0 /
PDC \~ ,N I
OH O

0 o o s
72% DCC
54%

113 110 120

 

 

Scheme 36: The revised oxidation of the aldehyde prevented formation of the
unreactive hemiketal, leaving just the carboxylic acid and attachment of the Barton ester.

The acid was then coupled to the ester precursor using DCC to generate the
Barton ester 120 as a water, light, heat, and bulk silica (presumably due to the acid)
sensitive yellow solid. Efforts to purify the compound were often met with
decomposition of the bulk material with traces surviving for isolation. It was eventually
puriﬁed via Celite columns to remove the majority of the urea byproduct. The semi pure
material was then puriﬁed on TLC grade, binderless silica column under ﬂash conditions

with reduced decomposition.

O //
0.11 ' AIBN s N\
U —><-+
O S O /’ O
120 121

Scheme 37 : The partially reacted Barton ester expected byproduct.

 

With the ester 120 in hand and able to be produced in usable amounts; the
decarboxylation and radical trapping experiments were begun. One known potential
pitfall to the chemistry is a propensity of tertiary intermediate radicals formed from the

decarboylation to recombine with the nearby sulﬁrr radical to form a stable byproduct

38

that often required a second radical source to reinitiate the chain —- a habit weakly shared

with primary and secondary radicals according to Barton’s work.4O

Under the AIBN/reﬂux conditions used, the ester quickly underwent the
decarboxylation and from the tentative structure assigned, appeared to lose a hydrogen to
form an alkene from one of the germinal methyls as illustrated in Figure 6 below rather
than undergo the expected addition to 84 a or potential side reaction 121. Photochemical
activation yielded similar results. With the failure to produce the desired

photoprecursor, a new route was deemed necessary.

1 22
Tentative

Figure 6: The tentative product observed from the Barton ester chemistry.

Revising the synthesis

Looking back over the prior work, the majority of the issues revolve around an
inability to protect the benzylic ketone which severely limited synthetic options. Thus a
revised synthesis was developed where the carbonyl group was to be installed just prior
to the photochemistry to minimize the issues it may cause. The addition was to occur
after the elaboration of the alkyl chain so that established chemistry may be reused with

minor modiﬁcation and the previously planned coupling method may still be attempted.

39

The new synthetic approach then was based on using a commercially available
1,2,3 substituted molecule, 2,3-dimethylphenol 126. Phenol 127 was selected due its
ability to be converted into 2-methoxy-6-methy1benzylbromide 126 41 which may be
later elaborated into triﬂate 123 b. This triﬂate was expected to eventually undergo

transition metal mediated acylations to generate 84 a and other derivatives.

\ \
H H
m: __——:> O :—-__> 61>;
OR'
0
61 88a 84a

123 a,b
1-Sterpurene a) R'=Me
b) R'=Tf

(is: Cicz=éciie=ﬁ>ii

124 a.b
127 126 125 a) R=OH

b) R=l

Scheme 38: Revised retrosynthesis for l-sterpurene. The other desired chromaphore
derivatives (ester, triﬂuoromethyl) are to be added by converting the triﬂate to a stannate
then Stille coupling or conversion to an organolithium species and quenching to install
the acyl group. The triﬂouromethyl and ester derivatives were expected to be installed
in the same manner as the acyl group of 84 a

The synthesis begins with the protection of the phenol as a methyl ether 128 in
good yields using methyl iodide and potassium carbonate in a schlenk tube (Scheme 39).
The commercially available, room temperature melting solid was then brominated using
N-bromosuccimide (NBS) under free radical conditions to yield material that was
brominated at either the 2 (126) or 3 methyl in ratios ranging from 8:1 to 12:1
respectively, along with a small amount of starting material and 2,3 dibrominated

material. An additional byproduct was observed in spectra of crude material when the

40

reactions were not dilute enough suggesting bromination at the phenyl ring, but this
suspected byproduct was not isolated.42 Bulb to bulb distillation allowed for separation
of the monobrominated species from the non brominated and dibrominated material.
The monobromides do not keep well even when frozen and Should be used soon after

puriﬁcation.

 

Mel NBS
(:I/ N82003 d hv def
OH OMe OMe

79% 55%

127 128 126

Scheme 39: Differentiation of the methyls. The yield shown is for crystallization
isolated 126.

Initially, the mixture of the 2 mono bromides was carried into later reactions and
then separated with some difﬁculty. However, a serendipitous discovery occurred - the
desired bromide 126 can be crystallized away from the other regioisomer using hot

hexanes allowing for its isolation in moderate, but slightly variable yields.

With a source of pure 2-methoxy—6 methyl benzylbromide 126 in hand, adding
the alkyl chain to the 2 position was the ﬁrst priority. However, due to the relative
difﬁculty of getting the bromide compared to the previous synthesis (bought versus
made), the phase transfer catalyzed alkylation used initially was abandoned for a more
traditional enolate alkylation using methyl isobutyrate. Using lithium diisopropylarnide
(LDA) as the base, conversion of the bromide to the ester 129 occurred at low
temperatures cleanly and in good yield. After distillation, the clear oil was reduced

using lithium aluminum hydride to the desired alcohol 124 a.

41

 

 

o
H O OH
CC: °”° (1%”: ”“ CPS
OMe OMe OMe

LDA 93%
87%

126 129 1243

Scheme 40: Generation of the primary alcohol

Attempts to convert the alcohol 124 a into an iodide proved unsuccessful. In all
cases, it appeared that the halide was unable to successfully attack the carbon bearing the
activated alcohol. Using the triphenyl phosphine and iodide in tetrahydrofuran
conditions as an example, thin layer chromatography suggested that the attack of the
iodide was the unsuccessful step as one could monitor the consumption of triphenyl
phosphine to the iodo triphenyl phosphine salt and the subsequent consumption of the
salt by the alcohol. At this point the reaction ceased regardless of concentration or

temperature.

 

OH
OMe O

66%

124 a 130

Scheme 41: An unexpected cyclization.

While looking for suitable direct halogenations conditions, the alcohol 124 a and
thionyl chloride (800;) were dissolved in reﬂuxing toluene to see if halogenation was
possible under those conditions as the more expensive thionyl halides, such as bromides,
were not on hand. Unlike the previous reactions which halted midway through, a new

compound was obtained - however it was not the primary chloride. Instead an

42

intramolecular cyclization had occurred due to a nucleophilic attack of the phenol ether’s

oxygen on the carbon of the activated alcohol leaving group, generating 130.

This cyclic ether 130 was not considered interesting initially until the realization
that the ring offered not only an unexpected and selective form of cleaving the methyl
group, but could be used to simultaneously halogenate the methylene if an appropriate
reagent was used. To this end, the ring was opened by treating the ether with boron
tribromide in reﬂuxing dichloromethane for 3 days to generate the bromide 131. While
promising, the literature precedence for the upcoming Kumada coupling was done using
exclusively iodides.”35 Attempts to obtain the iodide by opening the ether using
trimethylsilyl iodide were unsuccessful as were later attempts to convert the bromide

132 to an iodide via Finklestein conditions.

Br TBSCI Br
0 OH OTBS

77% 69%

 

130 131 132

Scheme 42: Installing the halide in preparation for the Kumada coupling

The newly exposed phenol 131 was then protected using chloro dimethyl-tert-
butylsilane in the presence of imidazole in good yields. Having re-protected the phenol,
the Kumada coupling was deemed worth attempting with bromide 132 even though the
halide was not the desired iodine. The initial attempts at the coupling produced none of
the desired alkene, but did consume the starting material. The 1H NMR of the crude
product suggested the bromide had been reduced to the methyl 133 (tentative structure),

however this compound was not isolated.

43

 

Br CHZCHMgBr
OTBS OTBS

132 133
Tentative

Scheme 43: The failed attempts at the Kumada coupling.

Based on commentary that suggested reduction can occur more readily using the
nickel catalyst in tetrahydrofuran, the solvent was altered to diethyl ether in an effort to
slow this side reaction. However, a new source of vinyl bromide was needed as the
original material was a solution in tetrahydrofuran. The pure vinyl bromide gas was
acquired, condensed, dispensed, and the prerequisite Grignard reaction was attempted
using the previous conditions. However, vinyl magnesium bromide formation did not
occur. Switching the magnesium source from tunrings to magnesium anthracene and
later Reike magnesium in a schlenk tube did not result in the desired vinyllic Grignard

reagent.

Seeing as direct approaches to making vinyl magnesium bromide in diethyl either
were failing, a transmetalation approach was used. The starting bromide was converted
cleanly at -78 °C to the vinyl lithium using tert butyl lithium. The lithiated material was
added to freshly prepared anhydrous magnesium dibromide etherate to generate the
desired ‘Grignard’ precursor, which was then used in the coupling reaction. The
coupling attempts did appear to produce a small amount of a material that resembles the
desired material by 1H NMR spectroscopy of the crude product, but the trace amount

was not isolatable and the bulk of the material appeared to have undergone a reduction.

44

While attempting to work through the issues revolving around the formation of
the vinyl magnesium bromide, other approaches to installing the alkene were explored.
Having successfully used Witti g oleﬁnation on the aldehyde of the sterically similar 104
(Scheme 29 on page 30) in the original synthesis, a sequence of Wittig reactions were
planned to elongate the chain, reintroduce the aldehyde, and then have it undergo a ﬁnal
oleﬁnation. This sequence allows for complete control over the placement of the oleﬁn

as any mixtures were expected to be either difﬁcult to separate at best or inseparable at

 

 

 

worst.
Ph3P-

OH o CHZOCH3Br I °\ /°|

+
0MB 81 % OMe 75% OMe OMe
124a 125 134a 4:5 134b

\ Ph3PMeI Ox
tBuOK 3N HCI
OMe 93% OMe 90%
123a 135

Scheme 44: Installing of the alkene after the failure to obtain the material to perform a
successful Kumada coupling.

The alcohol 124 a, previously derived from the lithium aluminum hydride
reduction of an ester, was then oxidized to the aldehyde 125 using 2-iodoxybenzoic acid
(IBX) in reﬂuxing ethyl acetate. The clear oil aldehyde was then subjected to the deep
red ylide derived from methoxymethyl triphenylphosphonium iodide and potassium tert
butylalkoxide. The two, separable enol ethers 134 a and 134 b that result are hydrolyzed

to aldehyde 135 using dilute hydrochloric acid. This new aldehyde then undergoes a

45

second Wittig reaction with the yellow methyl tlriphenylphosphoniurn ylide to yield the

desired alkene 123 a in good yield.

With the side chain fully elaborated, the replacement of the methoxy group with
an acyl group stood as the ﬁnal task before the photochemistry. In order to achieve this,
the methyl protecting group needed to be removed, the phenol converted sequentially to
a triﬂate and then a stannane.43 The stannane will then allow a Stille coupling with
various acyl chlorides to occur, yielding several possible photochemical precursors.
Alternatively, conversion of the stananne to a lithium species and direct quenching may

be used as well.

The deprotection of the methyl ether of 122a to the phenol 136 proceeded
smoothly using lithium diphenyl phosphine.44 The newly exposed phenol was

converted to the triﬂate 122 b using triﬂic anhydride.

\ thPH \ TfZO \
n BuLi NaH
OMe OH OTf

93% 87%

 

 

1228 136 122b

Scheme 45: Conversion of the methoxy to a triﬂate.

With the triﬂate now in place, conversion of triﬂate 122 b to a stannane 137
using a palladium mediated distannane coupling was attempted and found to be
unsuccessful.43 Starting material was consistently recovered as the
tetralds(triphenylphosphine)palladium appeared to be unable to insert into the phenyl
triﬂate bond and instead generated palladium black over extended heating. More

thermally robust ligands such as 1,3-bis(diphenylphosphino)propane were able to slow

46

the death of the transition metal catalyst, but were unable to promote insertion.
Presumably steric interactions with the ortho tether blocked approach on one side while
the bulk of the triﬂate itself rendered the phenyl/triﬂate bond inaccessible via the other

side.

Pd(PPh)4
\ MeSSnz \
(1% ”°' (:02
__,(__..
OTf SnMea
122 b 1 37

Scheme 46: Failure to install a stananne in place of the triﬂate, presumably due to
inability to access the phenyl/triﬂate bond.

In an effort to salvage the material, means for the conversion of the phenol 136 to
a halide were sought, but ultimately rejected. The most promising conversion required
an excess of bromo triphenylphosphonium bromide used neat at >200°C on very
sterically accessible substrates, while the molecule in hand appeared to have pronounced
steric issues and a potentially labile oleﬁn. Previous steps also appeared to be
unfavorable for this conversion as well due to the presence of alkenes or oxygen

moieties. With this in mind, the need for a new placeholder was made clear.

Revising the Synthesis . . . Again

Having failed to convert the o-triﬂate to the desired stannane or observe any
evidence of insertion by the palladium species, the need for a more versatile and
sterically accessible placeholder was evident. The replacement needs to have greater

synthetic options in terms of activation in order to avoid repeating the previous synthetic

47

misstep of being locked into a particular reaction with the exception of the reaction of
interest. Thus it was clear that halides rather than pseudo halides would likely be the

best choice for the placeholder.

In the interest of returning to the previous point as soon as possible and reusing
much of the chemistry, attempts to brominate commercially available 2,3-
dimethylchlorobenzene 138 were performed analogous to the 2,3-dimethylanisole 128
(Scheme 39 on page 41) used the previous reaction sequence. Unfortunately, the
attempts to repeat the selective bromination with 2,3-dimethylchlorobenzene yielded
inseparable monobromides 139 and 140 with little selectivity. The mixture of bromides
was taken forward a couple steps, but the regioisomers proved to be difﬁcult to separate

by practical means even after elaboration, preventing effective reaction scaling.

Br

(if COOL
Cl Cl Cl

68%>

 

138 139 140

Scheme 47 : Bromination of 2,3 dimethylchlorobenzene proved to be too unselective to
use. A 5 to 4 ratio of 139 to 140 respectively was observed.

While attempts to elaborate the 1,2,3 substituted phenyl halides proved to be
unsuccessful from a practical aspect, alternate attempts to elaborate mono or
disubstituted phenyls using ortho lithiation conditions were met with relative success.
Using literature reactions developed by the Snieckus group,” benzyl chloride was
converted into N,N-diethyl-2 bromo-6- methylbenzamide 144 in a handful of steps with

the use of strategic protection.

48

0 1)sec-BuLiTM 0 1)sec-BuLi W5 0 Br2
(2)532 TMEDA (j/sz TMEDA dim 0014 (:{WEQ
2) TMSCI 2) Mel 67 (69)°/
64 (70)% 54)(91)%
141 142 143 144

Scheme 48: Creating the desired 1,2,3 substituted arene using ortho lithiation
techniques. This work was used directly from the Snieckus’ literature33 with no change.
Yields are observed yields, yields in parenthesis are literature.

The conversion of the amide 144 into any other functional group at this point
proved ﬁ'uitless. Saponiﬁcation and acid based hydrolysis of the amide failed while
reductions using super hydride or Schwartz’s reagent to yield the aldehyde 146 were also
unsuccessful. While reading about other variants of ortho lithiations, commentary
suggested that the particular amide used, while quite powerful as a directing group, was
also quite robust to almost all conditions especially when substituted on both Sides - with

the exception of intramolecular saponiﬁcation.

Br 0 Br 0 Br 0
(5? +0” CC +4” Oi
OH NEt2

145 144 146

Scheme 49: The amide 144 proved remarkably robust to both reductive and non
reductive means of removal.

The difﬁculty of removing the amide can be avoided by several methods such as
having a secondary rather than a tertiary amide for, the chemistry or by simply avoiding
the amide all together. In light of this second option, several interesting protecting
groups allow aldehydes to be masked and used as relatively weaker, but still effective
ortho directing groups. Of these protection schemes, the one example from the Comin

group proved particularly interesting.

49

Comin’s group demonstrated the insitu protection of several aldehydes as an
alpha amino lithium alkoxide using the lithium amide of N,N,N’-
trimethylethylenediamine. This newly installed diamine functionality was then used to
direct ortho lithiation. But what was particularly unusual was the ability to perform this
chemistry in the presence of a halogen as shown below with 2-chlorobenzaldehyde
147.45 After the eventual quenching of the newly generated aryl lithium, the amine is
removed via hydrolysis under acidic conditions during workup to yield 2-chloro-6-

methylbenzaldehyde 149 in good yields.

 

|
0 H. /\,N OH I . o o
d T \ dN/VN\ 1) n-BULI d &
_—.> 1 +
Cl n-BuLi Cl 2) M61 Cl CI
50-65%
147 148 149 150

Scheme 50: Comin’s ortho lithiation of aldehydes in the presence of aryl chlorides."s
The reproduction of this work however required altered conditions. The aldehyde’s
ratios were variable until the quench was altered. The alteration resulted in consistently
less than 7 % of 150 formed (via crude 1H NMR ratios).

When reproducing this work, the chemistry was productive but the conditions of
the reaction were dissimilar to those reported. While the times between each addition
needed to be lengthened to prevent side reactions such as addition to the aldehyde by n-
butyl lithium, the addition of iodomethane to the reaction mixture of the newly formed
aryl lithium and excess butyl lithium was cause for concern once the reaction was scaled
to greater than a few mmol levels. When rapid addition of methyl iodide to the reaction
was performed at -78 °C, a violent reaction occurred which liberated a great deal of gas,
often blowing the septum off or shattering damaged glassware. On the other hand, slow

addition of the methyl iodide allowed time for substantial deprotonation of the newly

50

installed methyl group and a subsequent second quench to form an ethyl group 150 in

variable yields.

In order to avoid both the loss of material due to ‘violent reaction conditions’ or
loss as the undesired ethyl derivative 150, the order of addition was re-examined.
Reversing the addition (having the lithiated species transferred into a solution of methyl
iodide) reduced the ethyl formation as there was no longer be an excess of any basic
species present when compared to the quenching agent, preventing the secondary
deprotonation. Furthermore, Slow addition of the lithiated solution allowed for control
of the gas production, allowing for a safer reaction. Comin’s reported 70% yields of the
2-chloro-6-methylbenzaldehyde 149 were never realized, but a usable 50-65% iS often
observed using this modiﬁcation assuming dry reagents and freshly puriﬁed 2-

chlorobenzaldehyde.

For synthetic purposes, a distillation may be done to acquire a mostly pure room
temperature melting solid of the desired aldehyde 149 with a trace of 6-ethyl-2-
chlorobenzaldehyde 150. Further puriﬁcation is not necessary for this synthesis so long

as the starting aldehyde 147 is not present.

With an affordable precursor to 2-chloro-6-methylbenzylbromide now in hand,
the aldehyde of 149 is reduced to alcohol 151 with either sodium borohydride or lithium
aluminum hydride in tetrahydrofuran and the resulting alcohol crystallized from hot
hexanes. Bromination of the alcohol at room temperature in dichloromethane yields 2-

chloro-6-methylbenzylbromide 139 in good yields.

51

 

 

0 NaBH4
(EL) or LAH 6:014'33'3
Cl Cl

61% 78%

149 1 51 1 39
Scheme 51: Installing the benzyl bromide.

(:68 How OWMM

 

 

 

 

 

LDA 92%
90%
139 152 153
13x
89%
P11313-
/° CHZOCH3erﬁ:I>< 0
81-94% on
155 a 155 b 154
4:5
HCI
90%
0\ Ph3PMe| \
M IBUOK
c1 90% or
156 157

Scheme 52: Performing the established chemistry on the new chlorine variant.

The elaboration of the alkyl chain then occurred smoothly using the previously
established chemistry shown above in Scheme 52 to grant a new chlorine derivative 157.
The materials were isolated almost exclusively through distillation using bulb to bulb
distillation to both improve scalability. Separation of the enol ethers 155 a and 155 b

for individual characterization requires distillation to separate them from the

52

triphenylphospine oxide byproduct which otherwise cospots with the trans regioisomer

when puriﬁed via column chromotography.

With the use of a chloride rather than a bromide or iodide in 157, a majority of
palladium based Stille chemistry (along with other couplings) was rendered inaccessible
with notable the exception of Fu’s tri tert-butyl phosphine derived palladium catalyst.46
This catalyst appeared unable to insert into the chlorine/arene bond when tested in the
presence of a vinyl stannane, presumably due to the tether’s interference with the large
ligand. Ethylene tri n-butyl tin was chosen due to its ready availability when compared
to 2-(2-methoxy)ethylene tri n-butyl starmane which needed to be synthesized from
methyl vinyl ether. Fortunately, other reactions were now possible such as Grignard and

dissolving metal reactions.

\ (t-butyl)3P \
Pd2(dba)3
(:02 _._.
Cl

VinylSnBU3 l
CsF
157 Dioxane 153

Scheme 53: The failed Stille coupling.
Initial attempts at Grignard reactions were met with recovery of the starting
material. Bulk magnesium shavings activated either with heat, iodine crystals, or

sonication proved to be unable to initiate the reaction with any success.

Dissolving metal reactions were then attempted using lithium 4,4’ di tert butyl
biphenyl (LiDBB) in tetrahydroﬁrran. While the reaction failed to produce the desired
product 84 c shown below in Scheme 55, it did react with the starting material which

was a much needed step in the right direction. According to the spectrum of the crude

53

product, the chlorine was replaced with a hydrogen (162) indicating activation of the
site. However, scrambling of the double bond from a terminal to an internal alkene
(106, but cis/trans regiochemistry is undeﬁned) was also observed. This could signify
that after the initial electron donation occurs, free radical hydrogen abstraction from
either the tether (161) or the solvent may be occurring before a second electron can be
donated from another equivalent of lithium 4,4’ di tert butyl biphenyl. This competing
pathway was not entirely unexpected due to the relative instability of phenyl radicals

compared to alkyl/ethereal radicals and the potential hydrogen donors present.

I I \ A
LIDBB LIDBB 0 CN
——x—> —x—> 0
CI ° - Li+

157 159 160 84 c

Solvent
H O
Tether
H 0
\ ‘3, \ |
I H .
+

H

162 161 162 106

 

 

 

Scheme 54: The activation of the chlorine via a dissolving metal reaction. Although the
reaction failed due to hydrogen abstract by the aryl radical, it signaled that a general
approach to making the photochemical precursor was near.

Encouraged by the consumption of the starting material, further attempts at
dissolving metal reactions were performed with lithium 4,4’ di tert butyl biphenyl, but
no new products were observed after lowering the reaction temperature in an attempt to

slow the hydrogen abstraction or increasing equivalents of lithium 4,4’ di tert butyl

54

biphenyl. Attempts to quench the reaction with deuterated water failed to show any
deuterium incorporation on the ring, supporting a rapid hydrogen abstraction theory.

Further testing of the theory using deuterated tetrahydrofuran was not performed.

Assuming the hydrogen abstraction theory is correct, one can envision alleviating
the issue by having a metal source that can donate a second electron rather than having
to rely on a second equivalent to come into position and donate before any side reactions
occur. A Reike magnesium reaction seemed ideal as the magnesium can be thought of
as extremely activated, ﬁnely divided metal47 such that it can be readily accessible in the
sterically demanding pocket the tether is thought to form around the halogen.
Furthermore, it can readily donate a second electron to the convert the phenyl radical

into the more stable phenyl anion, preventing or at least competing with hydrogen

 

 

abstraction.
\ MgC|2 \ \ i \
K0 \0 CN
cI ' Mg' - Mg“ 0
O\
157 163 164 84 C

Scheme 55: The Reike Magnesium activation of 157 and it’s quenching to make the
photochemical precursor 84 c. Quenching was also preformed with dimethyl
formarnide, and triﬂouroacetic anhydride.

Reike magnesium reactions, however, are not very well understood. The actual
reactive species, which is often simpliﬁed to Mgo, is an unknown amalgam of
pyrophoric salts that can have their reactivity dramatically altered by having extra salt

additives present at formation, creating a black box situation.47 This is complicated by

55

the special precautions needed to accurately measure and work with the metal before it is

melted and allowed to reduce magnesium chloride.

Initial attempts using freshly prepared Reike magnesium were unsuccessful as
the black, room temperature suspension either failed to react or reduced the chloride to a
hydrogen with no scrambling of the alkene. However, a premature addition of the
chloride to a still cooling suspension of Reike magnesium (Reike magnesium is prepared
in reﬂuxing tetrahydroﬁrran and normally cooled to room temperature or colder before
use) did yield some aldehyde when quenched with dry dimethylfonnarnide (DMF) along
with substantial reduced byproduct. (DMF was chosen for quenching as it was easy to
obtain as a pure, dry quenching material compared to the more reactive formats which
may degrade substantially with handling. The resulting aldehyde can be derivatized to
other carbonyl types readily.) Subsequent addition of the chloride 157 at reﬂux followed
by quenching by either DMF, triﬂouroacetic anhydride, methyl chloroformate, or methyl
cyanoforrnate (as shown above in Scheme 55) allowed for low yields of the

corresponding aldehyde, ketone, or ester.

\ \ \ \
O O O O
CF3 o\ H
84 a

84 b 84 c 165
Not obtained 24-42% 27% 25%
directly

Figure 7: The photochemical precursors synthesized. The methyl ketone 84 a is derived
from the aldehyde 164. Yields for the quenching are given below.

The synthesis of the methyl derivative 84 a was attempted using acetic

anhydride, however no addition was ever observed, only reduced material. Whether the

56

reactions failed due to inability to form the proper Grignard reagent (the reaction is
somewhat ﬁnicky and reduced material is always observed) or the resulting Grignard

reagent tends to act more basic than nucleophilic due to its steric environment is unclear.

The acyl variant was instead derived by treating the aldehyde 165 with methyl
magnesium bromide to generate the benzyl alcohol 166. This alcohol was oxidized to a
ketone 84 a cleanly using similar oxidative conditions to those that lead to the
misinterpreted reaction that stopped the original synthesis (IBX is a precursor to Dess

Martin Periodate).48

\ \ \
MeMgBr 113x
0 OH O
86% 75%
H
165 84 a

166

 

 

Scheme 56: The formation of the methyl photoprecursor 84 a.

\
O\
O

167
Tentative

Figure 8: An observed byproduct that while not fully isolated, supports the theory on
why the dissolving metal reactions are less than ideal due to hydrogen abstraction.

During the isolation of the various photochemical precursors, a new byproduct
167 (tentative structure via lH NMR) was observed when the Grignard is formed at
reﬂux and then quenched with methyl cyanoformate at either room temperature or

reﬂux. This material, which is normally found in the last 30% of the fractions

57

containing 84 c when puriﬁed by column chromatography and constitutes less than 5%
of total isolated 84 c, suggests that hydrogen abstraction from the tether followed by
anion formation on the tether is present in the reaction as previously suggested in
Scheme 54 (page 54). This byproduct does not appear to affect the following

photochemical step.

\ \ \
O O O
CF3 0\
84 a 84 b 84 c

Figure 9: The photochemical precursors for reference.

Before continuing with the synthesis, one must note that all of the photochemical
materials have a distinctive trait of producing rather unexpected spectroscopic results —
especially in regards to the proton spectra. While one would expect a nice crisp singlet
for the tether’s benzyllic methylene if they were ﬁeely rotating or a pair of doublets if
they are not, the actual spectra at room temperature produces neither. Instead the two
protons are broadened out to roughly baseline while the geminal methyls produce a
singlet and the allylic methylene produces the expected doublet which would suggest
free rotation. The “missing” protons do become a pair of sharp doublets at -40 °C while
the gem dimethyls and the allylic methylene ﬁnally show their diastereotopic nature as
well with splitting to 2 singlets and a multiplet respectively as shown on the following
pages with the spectra of 84 c at room temperature (page 59) and at -40 °C (page 60).
This trait also affects DEPT analysis as the coupling of the carbon with the broad peaks
of the benzyllic methylene often makes the carbon appear hydrogen free unless a great

excess of scans are run.

58

 

 

Li i1 12A-

 

Spectrum 1: 5-(2-methyl formate)-6-methylphenyl) 4,4-
dimethylpentene 84 c in CDC13 at 25 °C (500 MHz 1H

NMR).

59

.11..

A

 

 

 

3— 6.01

} 2.12
3— 3.06

} 3.09

} 2.00

}'- 0.98

:l— 1.02
3— 1.43

} 1.00

 

TIllIllllIlIllIllllIllllIlllFIllllIllllIllTIIllllIllIHIllllIllllIllIlIIlllIllllIllWIllT

8.00 7.50 7.00 6.50 6.00 5.50 5.00 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50

8.50

Spectrum 2: 5-(2-methyl formate)-6-methylpheny1) 4,4-
dimethylpentene 84 c in CDCl3 at -40 °C (500 MHz 1H

NMR).

 

 

84c

60

11

-31. ll ll 1 ll 1 . JI—

 

 

}- 5.79

I 2.06

]— 2.98
}— 0.97

} 0.95

1L 3.01

} 1.97

j' 0.97

HJHJLH
8'58

ITIITIIIIIIIITTIIIIIITTTIIIIIITIIIIllllIlIllIllrlIflllIlITlIllllrlllllllllIllllITllIIlllj

 

 

8. 00 7.50 7.00 6.50 6.00 5.50 5.00 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50

8.50

With a handful of photochemical precursors ﬁnally in hand, the photochemical
step can now be tested. Of the 3 options, the ester 84 c proved to be the only material to
produce anything resembling the desired material as determined by crude and puriﬁed
(but not isolated) IH NMR samples. While the ester was unreactive under direct
irradiation in acetonitrile, a 1/ 100 molar solution of ester 84 c sensitized by the acetone
solvent appears to have underwent the photoaddition and rearrangement (Scheme 57).
This is suggested by the distinctive new doublet and doublet of doublet of the
cyclobutene coupled with the broad singlet of the hydrogen at the 6,4 fused ring junction
along with the consumption of the aromatic ring as shown on the following page. The
reaction proceeds slowly with full conversion often requiring 7-9 days of irradiation.
The slow reaction is likely due to the methyl’s steric bulk twisting the tether such that
the alkene no longer has optimal overlap for the initial cyclobutane ring closure, thus

promoting relaxation to the starting material rather than addition.

61

 

 

 

 

 

 

 

a)
6.2.
9%
O o
I l-
2
>
.c
/ Q g
o 0
v
a)

 

Spectrum 3: Isolated materials from the irradiation of 84 c.
(Chloroform solvent, 300MHz)

62

2
O
a —o'
.23 —.‘3
g F
% :0.
N
l _
=1 e
_0.
(‘0
‘4 30.98 7
2%; :1—3.39 _
_o.
v
;o.
I!)
< _
O
:4 $9233 "'0'
I r—
_c>.
[x
I r
L.

 

 

 

\
+—“V Ctr «——A 0'
_._____> + _——>
O R
OMe OMe
R= COOCH3 O
84 c 85 c 168 86 c
2 possible
regioisomers
A hv
H '1. \x/
+ 5 + Meojm + MeO
MeO o MeO’§o o 0
88 c 169 170 171
Tentative

42% overall yield

Scheme 57: The photochemical [2+2] reaction and the resulting rearrangements is
thought to yield the expected ester 88 c along with several possible regio and
diastereoisomers shown above.

Isolation of cyclobutene 88 c has not been successful since yields have been low
(~30-40%) and there are at least 4 minor byproduct(s) that could not be removed by
standard column chromatography when using the ester’s methyl peak as a point of
reference. The majority of these byproducts are thought to be derived from the regio and
distereoselective closure to the cyclobutene ring from the cyclooctatriene 86 c as the
directing effects for the initial cyclobutane closures 85 c and 168 (discussed on pages 6-
8) are all favoring 85 c. The regioselectivity of the cyclobutene closure however is not
as clearly deﬁned because the biradical density is spread amongst the three alkenes and
should follow the selectivity discussed previously (page 15). The desired regioisomer
(represented by 88 c and 169) is thought to be the 2 major products as the 2 isolated

doublets at roughly 6 ppm would be expected from both sets of cyclobutene hydrogens

63

(which have roughly equivalent environments). The methyl ester of 88 c and 169
however should differentiate between the two based on pseudo 1,3 diaxial interactions of
the carbonyl with the chiral carbon shared by the cyclohexane/cyclopentane. If this is
correct, then the corresponding peaks of the other regioisomer (170 and 171) may not be
very easily observed due to resolution limitations preventing observation of the alkene

hydrogens since the signals for the ester methyl are weak.

Earlier attempts at the photochemistry of this compound had even greater
amounts of byproducts, however altering the standard reaction conditions from cooling
the reaction vessel with standing water to a constantly refreshed cooling bath seems to
have reduced the number of the minor materials formed. HPLC puriﬁcation has yet to

be attempted.

The triﬂouromethyl ketone variant 84 h was shown to be consumed faster than
the ester 84 c, however it does not appear to produce the desired reaction and instead
generates a new material. This material was initially quite curious as the aromatic region
appeared to have contracted, suggesting the ketone which caused the wider spread of
peaks was no longer present while the tethered alkene was untouched. Attempts at
isolating the reaction materials lead to the observance of a cyclobutane ring fused to the
aromatic ring 173 as illustrated in Scheme 58. This product is thought to be derived
from the hydrogen abstraction of the excited oxygen from the tether to form the enol

172, which can then undergo disrotary ring closure to produce the observed cyclobutanes

173.

64

\ / \

hv / hv
O \ OH ; OH
36% 61:3
CF3 CF3
84 b 172 173

Scheme 58: The product derived from the irradiated triﬂouromethyl ketone in acetone.
Product ratio is roughly 4 : 5. Similar results are observed in acetonitrile and methanol.

In one unreproducable reaction, the triﬂouroacyl photoprecursor 84 b in
deuterated acetone (1/100 molarity) does appear to have produced the desired
photochemical material when irradiated overnight as demonstrated by the formation of a
pair of doublets at 6 ppm suggesting cyclobutene hydrogens. This stands as the only
potentially successful reaction that has been performed with the ketones so far. It was
not isolatable. With the failures of the longer irradiated reactions, the possibility of the
photoproduct being destroyed and converted to the observed material cannot be
dismissed. However, other samples that were prepared under virtually identical

conditions and run for the same period of time have not shown the same product.

The methyl ketone variant 84 a received the least attention of the precursors due
to a combination of greater difﬁculty obtaining material and the ester variant appearing
to generate the desired photoreaction without the hydrogen abstraction issues that are
present in the ketones. However, 84 a appears to generate neither the desired
photochemical product nor the hydrogen abstraction product observed with the
triﬂuoromethyl variant. The two attempts have instead generated a polymeric material

with few peaks upﬁeld of 2 ppm (1H NMR) and no identiﬁable compounds.

65

What Remains to Be Done

H I H I H I H I
MeO O MeO O R
88 c

174 175 ab 61
a) R=OH 1-Sterpurene
b) R=l

Scheme 59: The remaining steps needed to ﬁnish Sterpurene.

Assuming that eventually all of the photochemical precursors yield the desired
photochemical products, the total synthesis would likely proceed with the ester variant of
the photochemical addition 88 c due to ease of reducing the carbonyl group to a single
methyl. Photoproduct 88 c would be submitted to a reduction of the cyclobutene using
palladium and 1,3 cyclohexadiene to selectively reduce the cyclobutene ring to the
cyclobutane 174 while preventing overreduction by limiting the amount of hydrogen
present. Lithium aluminum hydride reduction of the ester in 174 to an alcohol 175 a
followed by conversion to a halide 175 b using methyl iodide and
dicyclohexylcarbodiimide should prepare the molecule for the ﬁnal transformation. The
iodide is expected to be reduced to l-Sterpurene 61 by a lithium halogen exchange
followed by a water quench. These ﬁnal steps after the cycloaddition have not been
performed. Other incomplete projects related to the photochemistry of the Wagner

group are discussed in the next chapter.

66

Chapter 3: Related Photocycloaddition Projects

While the total synthesis of Sterpurene was intended to demonstrate the synthetic
utility of the Wagner group’s [2+2] photocycloaddition, other projects were started with
the goal of broadening the synthetic viability of the reaction. The primary goals in the
following projects were to either create a more common fused 4,6 ring structure rather
than the highly specialized 4,6,5 fused skeleton or to address the current regiochemical
limitations of the chemistry. These projects were not completed; however a brief

discussion of the observations and insights may be useful if the work is ever resumed.

Intramolecular [2+2] Photoaddition using Temporary Tethers

H R O
o-"'
o
R o .0
176 177

Figure 10: Proposed temporary tether products 176 and 177 from their respective ortho
and para derived photoprecursors.

Having the luxury of a large and relatively complete body of work pertaining to
the intramolecular cycloaddition on hand, it would be extremely useful to incorporate
this work in developing a method to create fused 4,6 ring systems with handles of known
relative stereochemistry for later manipulation. One of the more potentially useful
means of doing this would be through the use of a temporary tether that can be easily
discarded once its purpose has been ﬁilﬁlled. Ideally, this tether should be easy to

install, photochemically stable, and affordable.

67

Upon a review of the literature, silicon tethers derived from a dialkyl dihalide
silicon species appeared to be a plausible temporary tether due to ease of installation and
removal, broad commercial availability of the silane allowing for multiple variants, and
literature precedence for applications in photochemistry.49'53 However, the use of
ﬁmctional groups beyond alkoxys in forming an unsymmetric siloxane such as conjugate
acids with enol ethers are not well known in literature and their different sensitivities
may pose issues. The proposed photochemical reactions for the ortho and para version

are presented below in schemes 60 and 61 respectively.
O‘SLa
o hv H HF pyr 0
:Si:
OH
R

178 179 180 181

 

Scheme 60: The proposed [2+2] cycloaddition and its likely rearrangements should
result in 181 from an ortho arrangement. R is any alkyl group or ester and substitution
along the aromatic ring should be possible although some rate issues/competitions may
arise as a result.

9058'? ﬂbﬁ

182 183 184 185

 

\

—Si\’0

 

Scheme 61: The proposed [2+2] cycloaddition and its rearrangements resulting ﬁ'om the
para precursor. Ring expansion from 185 to a cyclooctadiene is expected to be a possible
complication and may require a neutral ﬂuoride source. R is any alkyl group or ester.

Using the para version for the retrosynthetic approach, the photochemical

precursor 182 was envisioned as being derived from a halogen substituted phenyl ring

68

that is already elaborated with the silicon tether 186. The acyl group was envisioned as
being added last to avoid complications with both the silicon dichloride reagent and the
silicon chloride intermediate 187 and also to avoid any complications brought about by
protecting group removal in the presence of the tether. The tether was seen as being
formed by treatment of phenol 188 with dichlorodimethylsilane and subsequent
quenching of the resulting siloxy chloride 187 by an appropriate carbonyl to generate the

3 atom tether normally needed for efﬁcient photoaddition.

This one pot approach to generating the tether has the downside of potentially
needing several equivalents of both the silane and carbonyl to bring about the desired
product, however the actual scale needed was unknown. Furthermore, the use of the
readily available and affordable dichlorodimethylsilane may restrict the chemical
tolerances of the tether. However, if this proves problematic, the issue may be addressed

by using bulkier alkyl such as isopropyls.

0‘ JO 0‘ ..0 0‘ «Cl OH
O R R R

Alk
182 186 a,b 187 a,b 188 a,b

a) R: H b) R=Br

Scheme 62: The general retrosynthesis for the silicon tether formation. Phenol was used
initially to get rough conditions for the halide.

While this approach seemed rather feasible on paper, this preparation of the
tether was not very practical. Substantial excess (10 equivalents) of

dichlorodimethylsilane was needed to consume unsubstituted phenol 188 a dissolved in

dichloromethane in a reasonable amounts of time and with minimal dimerization. While
this isn’t a major issue due to the cost of the dicholorodimethylsilane, this mandated
even greater amounts of carbonyl (30+ equivalents) for the quenching which limits the
application of this chemistry. Separation of the excessive amounts of byproduct from
the desired siloxane 186 a was possible through tedious distillations. The bromide
substituted phenol 188 b required even greater excesses due to the reduced electron
density at the phenol and thus it became apparent that a one pot route was not

synthetically viable.

Since the one pot solution was not very practical, the reaction sequence was
reexamined. Puriﬁcation of the siloxychloride intermediate 187 allows for reduced
amounts of the quenching carbonyl. Furthermore, converting the carbonyl to a lithium
enolate before addition to the siloxychloride 187 was expected to generate the tether 186
more rapidly and require as little as a single equivalent, potentially expanding the
chemistry’s application by allowing for the use of expensive or synthetically valuable
carbonyls. In order to prevent any byproducts derived from enolate formation from
hampering the reaction and to allow for control when using asymmetric ketones, the
lithium enolate was to be generated from the desired trimethylsilyl enol ether using

methyl lithium.54

The initial formation of monochloride 187 b from 4 bromophenol 188 b and
dichlorodimethylsilane proceeded smoothly as shown in Scheme 63. Special care was
taken with the crude product to ensure the majority of the excess dichlorodimethylsilane
is distilled off before ﬁltration of solids and concentration via rotary evaporation. Even

with this precaution, water rinses were needed to prevent the equipment’s rubber seals

70

from degrading. This issue may be avoided entirely by removing the ﬁltrate’s solvent via
distillation rather than rotary evaporation. The crude mixture was puriﬁed by bulb to

bulb distillation under reduced pressure to yield the unstable monochloride 187 b.

TMS\

 

 

605 9
0°
TEA Br
72%

188D 187b 186D

/ \
Br

MeLi 3'

Scheme 63: Stepwise creation of the tether proved successful, but difﬁcult to reproduce.

The monochloride 187 b must be used within days of preparation as an apparent
equilibrium exists that converts the product into the starting dichlorodimethylsilane and
the dimethyldi(4-bromophenoxy)silane even at low temperature. The equilibrium is
relatively slow (a couple percent conversion a day) so one night’s storage in the freezer
is not overly detrimental. The limits of this equilibrium process have not been observed.
Long term storage of the monochloride 187 b may be possible if stored as a ﬁ'ozen,

dilute benzene solution, but this has not been tested.

A slight excess of the desired silyl enol ether, derived from cyclopentanone, was
then treated with methyl lithimn in diethyl ether. The resulting lithium enolate was
transferred via canulla into a solution of silyl chloride 187 h at -78 °C and allowed to
gradually warm to room temperature. The crude product 186 b was then puriﬁed via

distillation.

Unlike the ortho lithiations of the total synthesis, the actual direction of addition

doesn’t seem to matter signiﬁcantly as silyl chloride addition to the enolate appears to

71

work as well. The resulting siloxane 186 b should avoid contact with water but it is
stable in it for short periods of time. Mildly acidic water such as saturated ammonium

chloride solutions must be avoided.

Attempts to re-obtain the completed tether 186 b for both yield and complete
characterization purposes have been disappointingly unsuccessful. While a material that
suggests the desired tether has been observed at the previously noted distillation ranges,
it has not been pure enough to conﬁdently assign the structure shown in Scheme 63. As
a result, the following tests which were done with the previously obtained, but not fully

characterized material, may be inaccurate.

While the tether showed less than the expected stability to even weak
nucleophiles such as water, attempts at lithium/halogen exchange were performed on
186 h followed by quenching with acyl chloride. The lithium/halogen exchange was
attempted based on a composite of information. Reports suggested that dialkyl siloxanes
were relatively stable to acid and base hydrolysis as long as there is no signiﬁcant ring
strain present.55 Trialkyl silanes show reduced stability to the same conditions, but have
been shown to be relatively stable to Grignard reagents and alkyl lithiums.56 This
suggested that perhaps the desired dialkyl siloxanes are also alkyl lithium stable
although information conﬁrming or disproving this hypothesis was not found in
literature. The observed half life for trialkyl silanes in the presence of alkyl lithiums
seemed enough to be able to both form and quench an aryl lithium species without
tremendous loss of yield as byproducts - assuming the sterics effects of the aryl lithium
is more akin to an n-butyl lithium than the methyl version in rates of addition and the

dialkyl siloxanes share the alkyl lithium stability of their trialkylated versions. The

72

tether failed to survive any attempts and no sign of acylation was observed in the crude

NMR. Grignard activation was also unsuccessful.

0‘ ’0 1) t-BULi 0‘ ,
O /Si\ /Si\
3‘ 0
Br 2) MeCOCI
186 b 189

Scheme 64: Attempts to acylate the bromide were unsuccessful.

Coupling reactions appeared to be a plausible solution to the issue, however the
necessary functional group precursor (stannane, borane, etc) was lacking if it was to be
reacted with an acyl chloride. The Stille reaction was chosen due to the likelihood of it
being tolerant of a fragile tether, however getting rid of the residual stannane byproducts
may be problematic. While the stannane can be placed on either the phenyl or on a
masked acyl group such as methyl 2-(trimethylstannane)vinyl ether, the high likelihood
of being unable to remove a protection group from the enol ether without damaging the

tether suggests that the tin should be on the arene.

At this point, the route was reworked to incorporate a trimethylstannane prior to
the tether formation and an unﬁ'uitful search for methods at installing the stannane into
the currently available material performed. However, attempts to make the stannane
substituted version were not performed upon the recommendation of the graduate
committee to focus on the l-sterpurene synthesis exclusively. If the photochemical
precursor had been made, it was expected to undergo addition according to the

guidelines set out by the Wagner group as illustrated below.

73

   

192

Scheme 65: The envisioned photocycloaddition and resulting rearrangement.
The products of the photochemistry are expected to undergo relatively rapid
cleavage of the silane using either dilute hydroﬂuoric acid and pyridine or

tetrabutylammonium ﬂuoride.

As a ﬁnal note, the original goal of using the most economical variant was
overzealous and a more conservative linkage should have been chosen. It is thought that
much of the tether’s weakness towards nucleophiles is likely the fault of the sterically
accessible silicon. Perhaps with a less accessible silicon, the direct conversion of the
bromide to a carbonyl via a lithium/halogen exchange and resulting quench will be
achieved and the photochemistry tested. Once the proof of principle is established,
studies to optimize the sequence for the more economical dichlorodimethylsilane can be

undertaken.

Intermolecular [2+2] photocycloaddition

While plans to form a fused 4,6 ring system were being looked at, a potentially
more difﬁcult but also more interesting variant can be obtained by removing the tether

entirely. Theoretically, if an intramolecular version of a reaction exists, an

74

intermolecular variant of the chemistry is potentially feasible with a few modiﬁcations.
These modiﬁcations often need to compensate for the reduced rate of reaction that is
observed in most cases. When looking at the photoaddition speciﬁcally, one would
expect several possible modiﬁcations such as increased reaction concentration, alteration
of donor/acceptor ratios, enhancing the difference between donor and acceptor to

improve odds of a reaction when they do collide, and solvent choices.

Work by the Gilbert group has demonstrated that the intermolecular version is
possible — albeit with very large excess of the donor and prolonged exposure.”'6 In their
most relevant work, they observed addition to both sites of radical density (at the
methoxy of 51 and nitrile of 52 in Scheme 66) in their system indicating that the

reaction’s regiochemistry is going to be an issue.

N
/
”\Q hv ”\\ : MeO / NW
—-> + +
\ /N
we OMe \\N \\
N OMe

50 51 52 53
I hv
OMe
OCN hv ch : :OMe
NC CN
1 93 194

Scheme 66: An example of the Gilbert group’s success with similar [2+2]
photocycloaddition using acrylnitrile. The initial products were isolated and 51
resubj ected to photochemical conditions to observe the ring opening 193.

75

Upon closer examination of the products, one will notice that not only is the site
of the addition an issue, but the regiochemistry of the alkene’s addition in 51 and 52 is
also different. This suggests that biradical partitioning present in the system is not
entirely analogous to excited enone systems as the examples in Scheme 63 are showing
addition on either end of the alkene. (This is assuming the initial addition and resulting
biradical formation occurs only at the sites of the most concentrated radical density, the
methoxy and the nitrile). In enone systems, the most stable radical of the alkene donor is
formed by addition to either site of the excited enone’s alkene. The product ratio is then
determined by the resulting biradical’s competition between ring closure and reversion

to starting material.

Taking these observations into account and applying them to the Wagner group’s
system, the reaction’s regiochemistry becomes quite uncertain. One can theorize that
each of the possible intermolecular reactions may to have their orientation initially
directed by an exciplex with addition occurring at the sites with the most radical density.
After the initial addition, the 1,4 biradical partitioning is likely the primary selective
process in determining the product ratios as the 1,4 biradical can now rotate freely upon
its newly formed bond and should favor donation into the more electron deﬁcient side as
before - assuming conditions are favorable for closure at all. Steric interactions are no
longer expected to play a signiﬁcant role in regioselectivity except with very bulky
donors. The importance of polarity will have in affecting the regioselectivity is

unknown, but it is currently expected to be difﬁcult to observe.

Looking more closely at the addition and biradical partitioning using a theoretical

example in Scheme 67, predicting the ability of each site to undergo initial radical

76

addition will require one to consider not only its ability to stabilize the excited radical
but also to consider whether the radical’s density is shared or not. In the example below,
three sites would be expected to have addition - at the acyl, ﬂuoride, and methoxy.
While the acyl is the best in stabilizing the radical density innately, the other two sites
are sharing radical density making them even less likely to undergo the addition due to
lack of radical density. Between those two conjugated sites, the ﬂuoride is expected to

get the majority of the additions.

O\\_ 40
' F "' ' F
+
OMe OMe
O 196 197
F / Minor Major

 

 

o
5
/—

   

200 201
Minor Ma'or

Scheme 67 : The possible additions of an ethylene to an excited acetophenone with the
expected preferred orientation of each site marked.

After the initial addition, the resulting biradical partitioning should then create a
major and minor product ratio based upon the ability of the 2 sides of the 1,4 biradical to
accept donation into the ring. This is shown above in Scheme 67 as the major and minor
for each site of addition. The 1,4 biradical ring closure rate differences are expected to
be large enough effects to prevent the minor products ﬁ'om being formed in any

substantial amount. Polarity effects might be observable in the case of the 200 and 201.

77

However, the complication of alkenes adding in a no longer ﬁxed manner adds
ﬁrrther regiochemical issues to the reaction. While this was previously held in check by
the tether, the intermolecular version may have rather severe practical limitations if the
donor alkene is either not symmetric or is not very well differentiated between each side
in such a way that one of the 1,4 biradicals is either too unstable to add (i.e. reverts to the
alkene before getting in position to close) or is too stable to effectively close, leaving the
other variant to close to the cyclobutane product. An example of possible additions can

be seen in the biradicals illustrated in Figure 11.

   

202 203
Figure 11: An illustration of potential additions of unsymmetrical alkenes when using
methyl vinyl ether that added alpha to the acyl group and is donating into the least
electron rich site.

With these things in mind, the initial study of the regiochemistry should be
performed with symmetric alkenes to help ease the identiﬁcation. The unsymmetric
radical partitioning can later be explored using particularly selective examples of the
arene. Whether the resulting products undergo ring expansion and rearrangement after
the addition is to be observed. Unfortunately, due an inability to obtain suitable
symmetric alkenes for screening, asymmetric alkenes l-hexene and 2,3-dihydrofuran
were used to attempt to ﬁnd conditions for addition while initially ignoring the actual

regiochemical issues they are expected to cause.

Establishing conditions for the reaction are needed as the previous work uses huge

excesses of the donor which makes the chemistry interesting academically but not

78

necessarily practical for synthesis. Thus far, work by other Wagner students has
provided no leads. Once conditions are found, appropriate symmetric alkenes were to be

purchased and more detailed studies begun.

The reaction screening began with small selection of acetophenones: acetophenone,
4-chloroacetophenone, 4-methoxyacetophenone, 4-thiomethoxyacetophenone, 1,4-
diacylbenzene, and a,or,or-triﬂouroacetophenone. Each acetophenone had roughly 0.1
mmol of material dissolved in 1 mL of a 1 M alkene/deuterated solvent solution for a
roughly 10 to 1 mixture of donor versus acceptor. The mixtures then had argon bubbled
through them for half an hour while cooled at 0 °C to prevent excessive evaporation. The
‘regassed’ samples then had their proton spectra taken before and after irradiation in
their original solvents. The irradiated samples were then concentrated to remove the
deuterated solvent and excess volatile alkene and their spectra were retaken in deuterated

chloroform.

The reactions of the aforementioned acetophenones with 1-hexene in both methanol
and acetonitrile were unsuccessful as no new alkenes were observed after irradiation
overnight. 4-chloroacetophenone was observed to make a polymer from the alkene.
2,3-dihydrofuran was tested next as it was expected to be a better donor than the
terminal alkene. Promising results were observed when the enol ether donor was
irradiated in acetonitrile with some of the acetophenones. 1,4 diacylbenzene, or,a,or-
triﬂouroacetophenone, and 4-thiomethoxyacetophenone all displayed signs of new
alkene formation and some arene consumption. Interestingly enough, 4-
methoxyacetophenone displayed no reaction under these same conditions and likely

needs to be retested to verify the result. Again, 4-chloroacetophenone was observed to

79

make a polymer from the alkene. Acetophenone did not appear to react with the donor
although some slight alteration of the spectra occurred. Spectra of 1,4-diacylbenzene, 4-
thiomethoxyacetophenone, and or,oc,or-triﬂouroacetophenone prior to and after

irradiation, as well as redissolved in chloroform, are included in the appendix on pages

148-150, 151-153, and 154-156 respectively.

Are these new alkenes peaks a representation of real products or breakdown
products? While the answer will be unknown until run on a larger scale and puriﬁed,
one can rationalize that they are not break down products from the 2,3-dihydrofuran as
not all samples containing the furan have them. Likewise, they are not breakdown
products of either the solvent or the respective acetophenone as they would have been
observed when using 1-hexene. The alkenes appear to be conjugated which precludes
photochemical meta addition (1,3 rather than 1,2) that is standard in arene systems that
lack a carbonyl. This is of course not conclusive and very open for interpretation. As a
result, an initial starting point for the intermolecular version has been established, but
placed on hold for other more promising projects such as the total synthesis of 1-

sterpurene and the photoaddition using a temporary silicon tether.

Fixing the meta tether

A ﬁnal relevant project that was planned, but not started, was an attempt to make
tethers meta to the acyl group viable in the [2+2] photoaddition. Unlike the ortho and
para version of the chemistry, the Wagner group observed that meta tethers failed to

undergo any addition and instead generated long lived triplet biradicals.4

80

O 31E,1t a 0 3112,11 * 0 311,1! *
O
0' “I
O O
204 11 205

Figure 12: The triplet electron layout for the ortho, para, and meta examples. Only meta
has radical density shared with the acyl stabilized site.

The reason for this difference is thought to be based on relative radical density at
the addition site. In the ortho and para examples of the photocycloaddition, the tether is
positioned such that the acyl group stabilized radical is not in resonance with the tether
and both sites have their own independent radical denisty. The meta position tether,
however, is conjugated and its radical density is shared with the acyl rather than having
its own separate radical density. This conjugation should lessen the chances that a
radical reaction will occur at the tether’s ipso carbon due to decreased radical density.
Evidence of this is seen in the relative rate reductions of the addition in the examples
where substitution meta to the tether exist.6 This competition for radical density
between the tethers alkoxy stabilized radical center and stabilized meta site results in a
reduced reaction rate. Furthermore, increases in the ability of the meta substituent to
stabilize a radical results in even greater reaction rate reductions and can effectively

prevent bond formation in the case of a nitrile substitution.

81

O
O R O R R
O/\/“\ O A o“ :
o 0 Li;

23 a-e 24 a-e 25 a-e 26 a-e
a) R=0Me a)= >95%
e) R= CN e) = No Reaction

Standard) R=H
Triplet lifetime for
a) 1.2 x107 s'1
e) 9 x106 s'1
Standard) 4 x107 s'1

Scheme 68: Reproduction of Scheme 8 from the introduction. Relevant structures are
retained while triplet lifetime information is added.

Combining both of these observations suggests that in the original meta system,
the electron density is split between the tether’s alkoxy and the carbonyl, with the
carbonyl having the lion’s share of the density, thus greatly slowing the initial closure to

the 1,4 biradical and promoting the existence of a long lived triplet.

In order to get around this issue, the radical density needs to be redirected back to
the site of the tether. Thus, an ester or amide linkage to the tether rather than an ether
may be useful. It should have similar electron withdrawing capability to allow for
greater electron density at the site when compared to the ether linkage, but the reaction is
still expected to be relatively sluggish as the radical density is now split between two
carbonyl sites roughly equally (An analogous situation that was shown to work in 23 a
(two competing ethers). The electron withdrawing nature of the ester/amide linkage
should improve radical donation into the site as well when compared to the ether

linkage.

82

O
hv ZS
o/ o/ O O/
O
0 0V 0 O O
206 207 208

Scheme 69: The meta tethered linkage and its potential photocycloaddition and ring
opening products.

If the photoaddition works, the opening of 207 to the cycloctatriene 208 should
be relatively slow - similar to the carbon tether examples as there is no push pull
mechanic. The closure of the cyclooctatriene 208 may favor closure with the ester
adjacent to the cyclobutene 210 due to preferential stabilization of the single biradical at

the ester, but other closures are likely to occur as well.

0
O o
O ———~“V
/ ‘— R
0 A R21
0 O O O O
208 209 210
R1: OMe
R2: 0

Scheme 70: The expected products from the singlet biradical closure. These products
are expected after cyclooctatriene 204 absorbs the second productive photon of the
reaction sequence.

While the meta tether may become potentially accessable, the requirement of a
second carbonyl will restrict its usefulness in synthetic situations. On the other hand, it

may prove to be yet another interesting method into fused 4,6 ring systems due to the

83

creation of a cleavable lactone/lactam. Either way, addressing the meta substitution

weakness of the photocycloaddition does seem possible and potentially rewarding.

Conclusion and Final Thoughts

While much of the work towards demonstrating the synthetic capability of the
photocycloaddition and expanding its usefulness have failed to yield the desired
chemical results, they have been extremely fruitful in regards to another goal — that of
teaching a ﬂedgling scientist the practical issues of how to approach and deal with
problems within the ﬁeld. Learning how to deal with practical issues such as
background research, when to halt failed routes, deciding which approaches to prioritize,
scaling and troubleshooting reactions are generally not taught in books and as a result
must be learned through observation and/or trial and error. While generally being forced
to resort to the later due to being the last of the Wagner group, being given the room to
try (and fail) rather than being simply given a direction was at times quite frustrating but
it should prove beneﬁcial as new skills and perceptions were obtained — resulting in a
more rounded researcher. Asking “why” and “how” is no longer just for mechanisms
and describing side reactions, but now an integral part of general execution of daily

tasks.

While there is still much to learn and still countless ways to error, I hope to use
what I have gotten here to bypass many of the most daunting obstacles as an independent

researcher and push into greater things to come.

84

Chapter 4: Experimentals

All reactions solvents were distilled prior to use unless otherwise noted.
Tetrahydrofuran and diethyl ether were distilled from sodium/benzophenone while
benzene, toluene, pyridine, diisopropyl amine, carbon tetrachloride, and
dichloromethane were distilled from calcium hydride. Iodomethane and dimethyl
formaldehyde were also distilled over calcium hydride and stored over molecular sieves.
Solvents for workup and column chromatography were used as received.

Potassium tert butoxide was used as obtained ﬁ'om Aldrich. All Wittig salts may
be purchased from Aldrich and used as is. The methoxymethyl derived Wittig salt
should be stored in a dessicator, other Wittig precursors are ﬁne with standard storage
techniques. All alkyl lithium species were titrated using 1,3 diphenylacetone p-
tosylhydrazone prior to use. All other chemicals were used as received.

Unless otherwise noted, all reaction glassware was oven dried in a 120 °C oven
prior to use.

With the exception of NBS bromination, all photochemical irradiations were
performed with a 450 W, medium pressure mercury arc lamp in a quartz immersion
well. NBS brominations were performed with a 200 W incandescent bulb. The sample
was positioned approximately 12” from the light source in both cases.

Chromotography was performed using silica gel 60 with a particle size of 0.040-
0.063mm (230-400 mesh). Bulb to bulb distillation was performed with a Buchi
kugelrohr and the air bath temperatures it reported are uncorrected. Melting point

temperatures are also uncorrected.

85

 

NBS
hV 31'

1 05
2-Methylbenzylbromide 105: To a dry 1L round bottom ﬂask ﬁtted with an air
condenser and containing a magnetic stir bar and a nitrogen atmosphere; o-xylenes (20.0
mL, 164 mmol, 1.07 equivalents) was added to a suspension of freshly powdered N-
bromosuccinimide (27.2662 g, 153.19 mmol, 1.0000 equivalent), azobisisobutyronitrile
(1.1531 g, 7.0221 mmol, 0.045839 equivalents) in carbon tetrachloride (800 mL). The
suspension was irradiated with a 200 W incandescent lamp for 6 h, during which time
the production and consumption of an auburn/tan color may occur. The reaction was
monitored by discontinuing stirring and observing whether the solid particulate ﬂoats or
sinks. Once all solid particulates ﬂoat, the reaction is complete. The resulting
suspension was then cooled to 0 °C and the solid succinimide byproduct removed via
ﬁltration. The solvent was then recovered via simple distillation and the crude product
was transferred to a 50 mL round bottom ﬂask to undergo bulb to bulb distillation (7O -
85 °C air bath temperature at 0.10 mm Hg) yielding 2-methylbenzylbromide 105 as a

clear, noxious oil in good yield. (21.3468 g, 115.35 mmol, 75%)

1H NMR samples match an authentic sample obtained from Aldrich. The protons may
be shifted slightly due to the chloroform residual peak being obscured. Spectral data for
105 1H NMR(CDC13, 300 MHz) 5 2.437 (s, 3H), 4.54 (s, 2H), 7.18 — 7.34 (m, 4H); l3c

NMR(CDC13, 75 MHz) 8 18.73, 32.35, 126.39, 128.96, 129.95, 130.79, 135.68, 137.25;

86

 

g. V (:4

NaOH

105 TBAI 104

3—(2-methylphenyl)-2,2-dimethylpropylaldehyde 104: To a dry 15 mL round bottom
ﬂask ﬁtted with a condenser and containing a magnetic stirbar and a nitrogen
atmosphere, a suspension of sodium hydroxide pellets (0.6602 g, 16.51 mmol, 1.007
equivalents) and tetrabutyl ammonium iodide (0.5903 g, 1.598 mmol, 0.09744
equivalents) were added to benzene (2.00 mL) and heated to reﬂux. To the reﬂuxing
solution, a solution of isobutylaldehyde (1 .50 mL, 16.4 mmol, 1.00 equivalent) and 2-
methylbenzylbromide 105 (2.60 mL, 19.5 mmol, 1.19 equivalents) was added dropwise
over 4 h via syringe pump then the resulting solution was allowed to stir overnight.
Normally the reaction is complete at this point (veriﬁed by TLC) however, occasionally
the sodium hydroxide pellets fuse to the reaction ﬂask walls and do not fully react.
Scraping them from the wall then resuming heating at reﬂux tends to ﬁx this issue.
(Note: if scale allows it, this reaction is probably best done with mechanical stirring).
Once the aldehyde is consumed, the reaction is cooled to rt and diluted with water. The
organic materials are extracted three times with diethyl ether. The combined organic
layers are then washed with 1N hydrochloric acid, 1N sodium hydroxide, water, and
brine before being dried over sodium sulfate. Concentration via rotary evaporation yields
a pungent crude product that is puriﬁed using silica gel column chromatography with 3%
ethyl acetate hexanes as an elutant. From the column, the aldehyde 104 is obtained as a
clear oil (often lightly yellow tinted if only puriﬁed by column chromatography once) in

moderate yields (1.9123 g, 10.850 mmol, 66%). Notes on scaling: due to inconsistent

87

access to syringe pumps, addition funnels were used with varying degrees of success due
to approximating the rate of addition. Too rapid or too slow of addition does appear to
be detrimental to yields with some examples dropping as low as 41%, but often yields

were observed in the 50-60% range.

Spectral data for 104 1H NMR (CDC13, 300 MHz) 8 1.07 (s, 6H), 2.28 (s, 3H), 2.82 (s,
2H), 6.99 - 7.16 (m, 4H), 9.58 (s, 1H); 13c NMR (CDC13, 75 MHz) 8 20.32, 21.51,
39.24, 47.65, 125.62, 126.63, 130.63, 130.95, 135.29, 136.75, 206.07; IR (thin ﬁlm)
3064w, 3021w, 2968s, 2932s, 2871m, 2808w, 2701w, 1726s, 1495m, 1469m, 1397w,
1378w, 1364w cm"; mass spectrum, m/z (% rel intensity) 176 M+ (3), 158 (46), 143

(27), 115 (14), 106 (51), 105 (100), 91 (43), 77 (54);

o Ph3PEtBr
W KOtBu

cis 5-(2-methylphenyl)—4,4-diemethylpent—2-ene 106: — In a dry 250 mL round bottom
ﬂask containing a magnetic stirbar and a nitrogen atmosphere, potassium tert-butoxide
(2.2426 g, 19.986 mmol, 1.0909 equivalent) was added to a rt suspension of ethyl
triphenylphosphonium bromide (8.0889 g, 21.788 mmol, 1.1893 equivalent) in
tetrahydrofuran (90 mL). The orange suspension was allowed to stir for 1 h then cooled
to -78 °C. The aldehyde 104 (3.2290 g, 18.321 mmol, 1.0000 equivalent) was then added
dr0pwise to the cooled solution then the dry ice/acetone bath was repacked. The
suspension was allowed to gradually warm to rt as the dry ice was consumed. The

suspension’s bright orange color paled to an ivory shade with a tint of the original

88

orange color. The reaction was quenched with water and extracted three times with
diethyl ether. The combined ether layers were washed with water and brine then dried
over sodium sulfate. Concentration of the dry ether solution via rotary evaporation
followed by silica gel column chromatography using hexanes as elutent
(dichloromethane is needed for loading) yields the oleﬁn as a clear oil in good yield
(2.8565 g, 15.169 mmol, 82%). Average yields range from low 80’s to 90%. The
identiﬁcation of the oleﬁn geometry was not possible at this stage and was derived from

the epoxide product 103.

Spectral data for 106 1H NMR(CDC13, 300 MHz) 8 1.13 (s, 6H), 1.59 (m, 3H), 2.32 (s,
3H), 2.70 (s, 2H), 5.26 - 5.39 (m, 2H), 7.05 - 7.17 (m, 4H); 13c NMR(CDC13, 75 MHz)
8 14.22, 20.62, 28.94, 38.61, 44.71, 123.14, 124.97, 125.87, 130.17, 131.42, 137.11,
137.65, 139.41; IR (thin ﬁhn) 3061m, 3007s, 2959s, 2870s, 1493s, 1470vs, 1410w,
1381s, 1362m, 1183m, 1140m cm“; mass spectrum, m/z (% rel intensity) 188 M+ (0.7),

173 (2), 106 (50), 105 (75), 83 (100), 55 (89);

 

1 06 1 03
cis 5-(2-methylphenyl)-4,4-diemethylpent-2-ene oxide 103: — In a dry 1L round
bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere, 77% meta
chloroperbenzoic acid (9.2163 g, 41 mmol, 1.5 equivalent) was added to a solution of
oleﬁn 106 (5.2998 g, 28.144 mmol, 1.0000 equivalent) and dichloromethane (500 mL).
The solution was allowed to stir for 2 h at rt then was diluted with water. The organic

material was extracted three times with dichloromethane, washed three times with IN

89

aqueous sodium hydroxide, once with water then brine. The dichloromethane solution
was dried over sodium sulfate and concentrated via rotary evaporation. The crude
product is often pure enough to continue to the next step, but may be further puriﬁed
using a short silica gel column with 3% diethyl ether/hexanes as elutant to yield the

epoxide 103 as a clear oil in excellent yields (5.5683 g, 27.254 mmol, 97%).

Spectral data for 103 ‘H NMR(CDC13, 300 MHz) 8 0.92 (s, 3H), 1.08 (s, 3H), 1.17 (d,
3H, J= 6.0 Hz), 2.33 (s, 3H), 2.68 (d, 1H, J= 13.8 Hz), 2.70 (d, 1H, J: 4.8 Hz), 2.77
(d, 1H, J= 13.5 Hz), 2.96 (qd, 1H, J= 6.0, 4.2 Hz), 7.15 — 7.08 (m, 4H); 13c: NMR
(CDCl;, 75 MHz) 8 14.16, 20.35, 23.19, 27.49, 37.03, 43.31, 54.98, 64.17, 125.170,
126.22, 130.40, 131.59, 136.54, 137.08; IR (thin ﬁlm) 306lw, 3019m, 2959s, 2872m,
1493s, 1458s, 1427m, 1398m, 13818, 1366m, 1140m cm'l; mass spectrum, m/z (% rel

intensity) 204 M+ (0.8), 186(5), 171 (17), 160 (36), 145 (52), 105 (83), 99 (84), 55

{El/QC TiCl4
OH

103 102

(100);

 

1,3,3,5-tetramethyl-1,2,3,4-tetrahydronaphthalen-Z-ol 102: - In a dry 5 00 mL round
bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere, titanium
tetrachloride (3.50 mL, 31.9 mmol, 1.28 equivalent) was added dropwise to a 0 °C
solution of epoxide 103 (5.1057 g, 24.990 mmol, 1.0000 equivalent) in dichloromethane
(330 mL). The reaction was allowed to stir at 0 °C for 2 h, upon which it became auburn

in color. Upon completion, the reaction solution was poured over a roughly equal

90

volume of ice. The organic material was then extracted out of the melt with
dichloromethane three times. The combined organic layers were washed with saturated
aqueous ammonium chloride, water, and brine before being dried over sodium sulfate.
The dichloromethane solution was then concentrated via rotary evaporation and the
crude product puriﬁed on a silica gel column using 4% ethyl acetate/hexanes as elutant
to yield the alcohol 102 as a white solid (4.3721 g, 21.399 mmol, 86%). Alternatively,
the product can be crystallized from hot hexanes in roughly identical yield (preferred
method). Be sure to crystallize slowly as crashing out the material traps signiﬁcant

byproducts.

Spectral data for 102 1H NMR(CDC13, 500 MHz) 8 0.90 (s, 3H), 1.16 (s, 3H), 1.29 (d,
1H, J = 6.0 Hz, is not observed when D20 is added), 1.43 (d, 3H, J = 7.0 Hz) 2.22 (s,
3H), 2.34 (d, 1H, J= 17.5 Hz), 2.60 (d, 1H, J= 17.5 Hz), 3.10 (qd, 1H, J: 7.0, 2.5 Hz),
3.42 (dd, 1H, J = 6.0, 3.5 Hz, becomes a (d, 1H, J = 3.5 Hz) when D20 is added), 7.01
(d, 1H, J = 7.5 Hz), 7.08 (t, 1H, J= 7.5 Hz), 7.17 (d, 1H, J= 8.0 Hz); 13C NMR (CDC13,
75 MHz) 5 16.94, 19.93, 25.51, 27.62, 33.97, 34.16, 36.02, 78.36, 125.11, 125.67,
127.54, 133.93, 136.32, 136.80; IR (KBr pellet) 3347s, 3066w, 2971s, 2939m, 2903s,
2868m, 1584w, 1473s, 1449m, 1381m, 1364m, 1339m, 1267w, 1120m, 1088m, 1053w,
1005m, 978s cm'l; mass spectrum, m/z (% rel intensity) 204 M+ (34), 186 (78), 171
(100), 156 (55), 143 (53), 133 (83), 132 (98), 117 (92), 115 (76), 91 (62); Melting point

=117—119°C.

91

Tf20

DMAP 00
OH

102 107

 

2,2,4,8-tetramethyl-l,2-dihydronaphthalene 107 — In a dry 500 mL round bottom
ﬂask containing a magnetic stir bar and a nitrogen atmosphere, triﬂic anhydride (5.20
mL, 30.9 mmol, 2.33 equivalents) was added dropwise to a rapidly stirring 0 °C solution
of N,N-dimethylaminopyridine (7.9413 g, 65.002 mmol, 4.8958 equivalent) and alcohol
102 (2.7126 g, 13.277 mmol, 1.0000 equivalent) in dichloromethane (150 mL). The
cloudy suspension was then allowed to stir overnight and to gradually warm to room
temperature as the ice melted. The reaction was diluted with water and the organic
material extracted three times with dichloromethane. The combined organic layers were
washed with 1N hydrochloric acid, saturated aqueous sodium bicarbonate, and brine.
The organic layer was dried over sodium sulfate and concentrated via rotary evaporation.
Puriﬁcation of the crude product by silica gel column chromatography using hexanes as

elutant yielded 107 as a clear oil in good yield (2.2540 g, 12.099 mmol, 91%).

Spectral data for 107 1H NMR (CDC13, 300 MHz) 81.02 (s, 6H), 2.03 (s, 3H), 2.28 (s,
3H), 2.60 (s, 2H) 5.57 (s, 1H), 7.02 — 7.13 (m, 3H); 130 NMR(CDC13, 75 MHz) 8 19.67,
28.19, 31.45, 39.34, 120.71,125.42, 128.76, 129.85, 133.74, 134.65, 135.11, 135.98; IR
(thin ﬁlm) 3069w, 3028w, 2953s, 2863s, 2822w, 1493s, 1584w, 1472m, 1460m, 1445m,
1377m, 1360m, 1329w cm"; mass spectrum, m/z (% rel intensity) 187 M+1 (10), 186

M+ (57), 171 (100), 156 (79), 141 (41), 128 (35), 115 (27);

92

 

0804 O
Oxone
oe 0“ - we
0 o
NaHCO3 OH

107 110 1 1 1
3-(2-Acetyl-6-methylphenyl)-2,2-dimethylpropanioc acid 110 and its hemilactone
111: - In a 100 mL round bottom ﬂask containing a magnetic stirbar and open to air,
osmium tetraoxide (0.400 mL of 4% weight in water, 0.06 mmol, 0.01 equivalents) was
added to a solution of alkene 107 (0.8144 g, 4.372 mmol, 1.000 equivalent) in
dimethylformamide (22.0 mL) and allowed to stir for 5 minutes. To the darkening
solution was added oxone (7.6832 g, 12.498 mmol, 2.859 equivalents) and sodium
bicarbonate (1.3751 g, 16.36 mmol, 3.747 equivalents) in one portion and the reaction
was allowed to stir for 5 days. Still incomplete (~40% conversion based on recovered
107), the reaction was quenched by addition of sodium sulﬁte (3.0567 g, 24.252 mmol,
5.547 equivalents) and the mixture was allowed to stir for one hour. 1 N hydrochloric
acid was added until all the salts were dissolved, roughly equal volume of ethyl acetate
was addend and the organic materials were extracted three times. The combined organic
layers were the washed with 1 N hydrochloric acid and brine then dried over sodium
sulfate. The dry ethyl acetate was then concentrated via rotary evaporation. The solid
crude product was then puriﬁed by silica gel column chromatography using 8% ethyl
acetate/hexanes as elutant until the product 111 (0.1058 g, 0.4516 mmol, 10% overall,
~26% of converted material) was obtained then a rapid gradient increase up to 35% ethyl
acetate/hexanes was used attempt to obtain 110, but it was not isolated (it was however

observed by TLC prior to column chromatography using a 1 10 sample obtained from Dr.

93

Ben Travis as he was able to isolate it). The hemilactone 111 was isolated as a solid in

low yields and may be crystallized in hexanes if trace impurities are present.
Spectral data for 110 may be found on page 102-103.

Spectral data for 111 lH NMR (CDC13, 300 MHz); 0.99 (s, 3H), 1.36 (s, 3H), 1.48 (s,
3H), 2.32 (s, 3H), 2.89 — 3.01 (m, 2H) 3.97 (s, 1H, Note: not observed after D20
experiment), 7.12 (d, 1H, J: 7.5 Hz), 7.20 (t, 1H, J= 7.5 Hz), 7.49 (d, 1H, J= 7.2 Hz);
13C NMR (CDC13, 75 MHz) 8; 19.77, 26.70, 27.04, 27.43, 38.00, 40.60, 75.94, 122.49,
127.09, 129.19, 131.68, 135.12, 139.69, 218.38; IR (thin ﬁlm) 3507s (Note: very sharp),
3075w, 2965m, 2872w, 1703s, 1468m, 1379m, 1256m, 1259m, 1225m, 1178m, 1099s,
1032s cm"; mass spectrum, m/z (% rel intensity); 218 (18), 200 (14), 175 (100), 132

(92), 104 (75); Melting point: 82.7 - 83.5 °C

TMSCI
MeOH

 

03-1

Methyl 3-(2-Acetyl—6-methylphenyl)-2,2-dimethylpropanioate 101: - In a 25 mL
round bottom ﬂask ﬁtted with a condenser, containing a magnetic stirbar and a nitrogen
atmosphere, trimethylsilyl chloride (0.133 mL, 1.04 mmol, 0.979 equivalents) was added
to a solution of acid 110 and hemilactone 111 (0.2489 g combined, ratio unrecorded,
1.062 mmol, 1.000 equivalent) in methanol (13.0 mL) and the resulting solution was
heated overnight at reﬂux. Upon completion, the reaction was cooled to room

temperature, diluted with water, and the organic layer extracted three times with ethyl

94

acetate. The combined organic layers were washed with saturated sodium bicarbonate,
water, and brine before being dried over sodium sulfate. The dry ethyl acetate solution
was then concentrated via rotary evaporation and the crude product puriﬁed on a silica
gel column using 10% ethyl acetate/hexanes as elutant. After puriﬁcation, a clear oil

101 was obtained in excellent yield (0.2501 g, 1.007 mmol, 95%).

Spectral data for 101 ‘H NMR(CDC13, 300 MHz) 8 1.07 (s, 6H), 2.29 (s, 3H), 2.56 (s,
3H), 3.39 (s, 2H, Note: broad peak), 3.59 (s, 3H), 7.14 (t, 1H, J: 7.5 Hz), 7.23 (d, 1H, J
= 6.9 Hz), 7.38 (d, 1H, J= 7.8 Hz); 13C NMR (CDC13, 75 MHz) 8 20.81, 25.06, 30.31,
35.73, 43.65, 51.77, 125.84, 126.05, 133.32, 135.13, 139.43, 141.27, 178.08, 203.90; IR
(thin ﬁlm) 3065w, 2975s, 2880m, 1730s,1688s, 1580m, 1474s, 1433s, 1387m, 1354s,
1263s, 1211s, 1140s, 1099s cm"; mass spectrum, m/z (% rel intensity) 248 M+ (1.3),

216 (28), 188 (22), 173 (83), 147 (100), 91 (73), 43 (77);

 

0804
cc 0;}
O H
107 112

1,3,3,5-tetramethyl-l,2,3,4-tetrahydronaphthalene-l,2-diol 112: - In a 250 mL round
bottom ﬂask containing a magnetic stirbar and open to the atmosphere, osmium
tetraoxide (0.600 mL of 4% weight in water, 0.09 mmol, 0.007 equivalents) was added
to a solution of tert-butyl alchohol (50.0 mL), water (15.0 mL), and alkene 107 (2.4615
g, 13.213 mmol, 1.0000 equivalent). Trimethylamine N-oxide dihydrate (2.2120 g,
19.903 mmol, 1.5063 equivalents) was added, a condenser ﬁtted, and the resulting

solution heated to 75 °C overnight. Upon completion, the reaction was allowed to cool

95

to rt and a solution of sodium bisulﬁte (58% ACS reagent grade, 2.4962 g) in water (50
mL) was added and allowed to stir for 1 hour. The reaction solution was then diluted
with water and the organic materials extracted three times with diethyl ether. The
combined organic layers were washed with water then brine, dried over sodium sulfate,
and concentrated via rotary evaporation. The resulting thick oil was then dissolved in
hexanes, resulting in a white precipitate crashing out. The precipitate is collected and
the product crystallized from hot hexanes to yield the diol 112 in decent (1.7423 g,
7.9084 mmol, 60%) to good yields (yields ranged from 60% to 84%). Minor amounts of
the diol may also be recovered from the mother liquors via silica gel column
chromatography (15% ethyl acetate/hexanes). Reaction note: The lower yields seem to
be due to poor reagent quality as all low yield runs were used old samples of osmium
tetraoxide and trimethylamine N-oxide dihydrate while the higher yielding runs used a
more recently opened containers suggesting either contamination or degradation of one

(or more) reagents.

Spectral data for 112 1H NMR(CDC13, 300 MHz) 8 0.88 (s, 3H), 1.15 (s, 3H), 1.57 (s,
3H), 2.24 (s, 3H), 2.28 (s, 1H, Note: is not observed after treatment with D20), 2.37 (d,
1H, J= 16.5 Hz), 2.63 (d, 1H J= 16.8 Hz), 2.88 (d, 1H J = 8.1 Hz, Note: is not observed
after treatment with D20), 3.44 (d, 1H, 1 = 7.8 Hz, after treatment with D20, becomes (s,
1H), 7.08 (d, 1H, J= 6.9 Hz), 7.16 (t, 1H, J: 7.5 Hz), 7.44 (d, 1H, J= 7.2 Hz); l3C
NMR(CDC13, 75 MHz) 8 20.04, 20.46, 28.45, 30.17, 34.24, 39.89, 71.73, 79.99, 125.13,
126.31, 129.17, 132.86, 135.93, 140.66; IR (thin ﬁlm) 3440sb, 33l8sb, 3069w, 3022w,
2991s, 2967s, 2941s, 2907s, 2876m, 1472m, 1460s, 1448s, 1401m, 1366s, 1252m,

1157m, 1132m, 1109m, 1082m, 1040s cm'l; mass spectrum, m/z (% rel intensity) 205

96

(M+ -15, 27), 202 (50), 187 (40), 133 (100), 132 (51), 115 (44), 105 (38), 91 (37);

Melting point: 114 — 116 °C

1’
(ij: Na|O4
O
O H

112 113

 

3-(2-acyl-6-methylphenyl)-2,2-dimethylpropana1 113: In a 250 mL round bottom
ﬂask containing a magnetic stirbar and a nitrogen atmosphere, sodium periodate (4.3777
g, 20.467 mmol, 2.0688 equivalents) was added in one portion to a solution of diol 112
(2.1796 g, 9.8933 mmol, 1.0000 equivalent), tetrahydrofuran (70 mL), and water (33
mL) and the resulting solution is allowed to stir overnight at rt. Within 15 minutes of
addition, white precipitate forms that will thicken over the course of the reaction. After
consumption of the diol, the reaction is diluted with water and the organic materials are
extracted three times with ethyl acetate. The combined organic layers are then washed
with water followed by brine, dried over sodium sulfate, and then concentrated via rotary
evaporation. The liquid crude product is then puriﬁed via silica column chromatography
using 8% ethyl acetate/hexanes as elutent. The resulting colorless oil (1.9875 g, 9.1049

mmol, 92%) is obtained in consistently good yields of greater than 90%.

Spectral data for 113 1H NMR(CDC13, 300 MHz) 8 0.99 (s, 6H), 2.31 (s, 3H), 2.54 (s,
3H), 3.30 (s, 2H, Note: broad peak), 7.17 (t, 1H, J= 7.5 Hz), 7.27 (d, 1H, J= 7.5 Hz),
7.45 (d, 1H, J: 7.5 Hz), 9.41 (s, 1H); 13c NMR (CDC13, 75 MHz) 8 21.52, 22.04, 30.49,
34.40, 48.16, 126.34, 127.12, 134.16, 135.26, 139.36, 140.35, 204.03, 205.74; [R (thin

ﬁlm) 3063m, 2969s, 2932s, 2872s, 2813m, 2718m, 1723s, 1684s, 1578m, 1470s, 1456s,

97

1426m 1399m, 1381m, 1354s, 12625, 1197w, 1174w, 1103m cm"; mass spectrum, m/z
(% rel intensity) 218 M+ (5), 200 (13), 175 (100), 152 (77), 147 (95), 133 (84), 115 (82),

91 (87), 77 (77), 43 (92);

LAH

 

113 114
3-(2-(ethan-1-ol)-6-methylphenyl)-2,2-dimethylpropanol 114: - In a dry 50 mL round

bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere, the keto/aldehyde
113 (0.5118 g, 2.344 mmol, 1.000 equivalent) was added dropwise over 15 minutes to a
0 °C suspension of lithium aluminum hydride (0.2717 mg, 7.159 mmol, 3.054
equivalents) in tetrahydrofuran (35.0 mL). The gray suspension was allowed to warm
gradually to rt as the ice melted while stirring over night. After observing consumption
of the starting material via TLC, the reaction was quenched by slow dropwise addition of
water. Upon quenching, the gray suspension turns white in color. The quenched
solution was diluted with water to a total volume roughly equal to the organic solvent
and then the organic material was extracted three times with diethyl ether. The
combined organic layers were washed twice with water, then once with brine, before
being dried over sodium sulfate. The suspended solids were ﬁltered away and the
ﬁltrate condensed via rotary evaporation to yield the crude product as either an oil or
white solid. (The oil often solidiﬁes upon standing to a white solid.) The crude product
is generally pure enough to be used in the next step, but it can be puriﬁed via silica gel
column chromatography with 35% ethyl acetate/hexanes as elutant to yield a clear oil in
excellent yields (0.5032 g, 2.263 mmol, 97%) with yields consistently at or above 85%.

This oil also solidiﬁes upon standing to a white solid.

98

Spectral data for 114 1H NMR (CDC13, 500 MHz, +50 °C) 8 0.90 (s, 3H), 0.94 (s, 3H),
1.41 (d, 3H, J = 6.0 Hz), 1.59 (s, 1H, note: broad peak), 1.91 (s, 1H, note: broad peak),
2.36 (s, 3H), 2.77 (d, 1H, J = 14.5 Hz, note: broad peak), 2.83 (d, 1H, J = 14.0 Hz), 3.41
- 3.36 (m, 2H), 5.34 (q, 1H, J: 6.5 Hz), 7.08 (d, 1H, J= 7.5 Hz), 7.15 (t, 1H, J= 7.5
Hz), 7.41 (d, 1H, J: 7.5 Hz); l3C NMR(CDC13, 125 MHz, +50 °C) 8 21.64, 24.36,
25.25, 34.86, 38.38, 66.51, 72.27, 123.18, 126.52, 129.88, 137.76 (2 sp2 carbons not
observed); IR (thin ﬁlm) 3351sb, 3067w, 2969s, 2874s, 1476s, 1466s, 1381m, 1365m,
1298m, 1277m, 1234, 1171w, 1090s, 10465 cm]; mass spectrum, m/z (% rel intensity)
204 M+ -H20 (23), 189 (69), 171 (84), 159 (95), 145 (63), 133 (72), 131 (90), 117 (97),

115 (100), 106 (74), 91 (76), 77 (69); The melting point was not taken.

0 0H
OMe LAH
O OH
1 01 1 14

3-(2-(ethan-l-ol)-6-methylphenyl)-2,2-dimethylpropanol 114 (alternative route): -
In a dry 25 mL round bottom ﬂask ﬁtted with magnetic stirring and containing a
nitrogen atmosphere, the ester 10] (0.1901 g, 0.7655 mmol, 1.000 equivalent) was added
dropwise to a suspension of lithium aluminum hydride (0.0678 g, 1.77 mmol, 2.31
equivalents) in tetrahydrofuran (10.0 mL). The gray suspension was allowed to stir over
night at rt. After observing consumption of the starting material via TLC, the reaction
was quenched by slow dropwise addition of water. Upon quenching, the gray
suspension turns white in color. The quenched solution was then diluted with water to a

total volume roughly equal to the organic solvent and then the organic material was

99

extracted three times with diethyl ether. The combined organic layers were then washed
with 1N hydrochloric acid, twice with water, followed by brine before being dried over
sodium sulfate. The solids were ﬁltered away and the ﬁltrate condensed via rotary
evaporation to yield the crude product as either an oil or white solid. The crude product
is then puriﬁed via column chromatography using a silica gel column with a solution of
35% ethyl acetate/hexanes as elutant to yield 114 as a clear oil in good yields (0.1408 g,

0.6333 mmol, 83%). The oil often solidiﬁes upon standing to a white solid.

Spectral data for 114 is given in the previous experimental.

OH TSCI 0T8
Pyridine
OH OH
1 15

114

 

3-(2-(ethan-1-ol)—6-methylphenyl)-2,2-dimethylpropanol Tosylate 115: - In a dry 25
mL round bottom ﬂask containing a magnetic stir bar and a nitrogen atmosphere, tosyl
chloride (0.5201 g, 2.728 mmol, 6.16 equivalents) was added to a rt solution of alcohol
114 (0.0984 g, 0.443 mmol, 1.00 equivalent) and pyridine (0.220 mL, 2.73 mmol, 6.16
equivalents) in dichloromethane (5.00 mL). The solution was allowed to stir for ~10 h.
After TLC veriﬁcation of alcohol consumption, the reaction was diluted with water and
the organic materials were extracted three times with dichloromethane. The combined
organic layers were washed with 1N hydrochloric acid, water, and brine before being
dried over sodium sulfate. The organic layer was then concentrated via rotary

evaporation and the crude material puriﬁed using silica gel column chromatography with

100

30% ethyl acetate/hexanes as elutant. After puriﬁcation, a thick oil 115 was obtained in

moderate yields (0.0963 g, 0.256 mmol, 58%)

Spectral data for 115 1H NMR(CDC13, 500 MHz, +50 °C) 8 0.88 (s, 3H), 0.90 (s, 3H),
1.36 (d, 3H, J = 5.5 Hz), 1.65 (s, 1H, note: broad peak), 2.28 (s, 3H), 2.44 (s, 3H), 2.74
(s, 2H, note: broad peak), 3.77 (d, 1H, J = 9.5 Hz), 3.82 (d, 1H, J = 9.5 Hz) 5.17 (q, 1H,
J= 6.5 Hz), 7.05 (d, 1H, J= 7.0 Hz), 7.14 (t, 1H, J= 7.5 Hz), 7.33 (d, 2H, J= 7.5 Hz),
7.40 (d, 1H, J= 7.5 Hz), 7.77 (d, 2H, J= 8.5 Hz); l3c NMR(CDC13, 125 MHz, +50 °C)
8 21.56, 24.22, 24.82, 34.47, 37.01, 66.53, 78.39, 123.20, 126.83, 127.94, 129.83,
129.96, 133.38, 137.67, 144.73 (multiple broadened peaks, 1 sp3 and 2 sp2 peaks not
observed); IR (thin ﬁhn) 3549mb, 3431mb, 3067w, 2972s, 2886m, 1599m, 1476m,
1466m, 1396m, 1358s, 1292m, ll90vs, ll76vs, 1097s, 1027m cm"; mass spectrum, m/z
(% rel intensity) 205 M+ -TsO (3.5), 204 (22), 189 (66), 171 (84), 159 (100), 131, (73),

77 (58);

0T3 O OTs OTs
Dess

Martin
OH 0 °' 0

 

115 118 119

3-(2-acyl-methylphenyl)-2,2-dimethylpropanol Tosylate 119: In a dry 25 mL round
bottom ﬂask containing a magnetic stir bar and a nitrogen atmosphere, a solution of Dess
Martin periodate (0.1412 g, 0.3329 mmol, 1.54 equivalents) in dichloromethane (6.00
mL) was added to a rt solution of alcohol 115 (0.0812 g, 0.216 mmol, 1.00 equivalent) in
dichloromethane (3.00 mL) and the resulting solution allowed to stir for 2 h. The

reaction was then diluted with a saturated solution of sodium thiosulﬁte and sodium

101

bicarbonate (1:1 mixture) and the organic materials were extracted three times with
dichloromethane. The combined organic layers were washed with water, brine, and
dried over sodium sulfate before being concentrated via rotary evaporation. Puriﬁcation
using silica gel column chromatography using 25% ethyl acetate/hexanes as elutent gave
a clear oil that was misidentiﬁed as the diketone 118 (yield not recorded for this ‘failed’

reaction). The material was later identiﬁed as the desired ketone 119.

Spectral data for 119 lH NMR (CDC13, 500 MHz, +25 0C) 8 0.76 (s, 6H) 2.31 (s, 3H),
2.43 (s, 3H), 2.52 (s, 3H), 3.69 (s, 2H), 7.14 (t, 1H, J= 7.5 Hz), 7.23 (d, 1H, J= 7.0
Hz), 7.33 (d, 2H, J= 8.0 Hz), 7.36 (d , 1H, J= 7.5 Hz) 7.76 (d, 2H, J= 8.0 Hz) (two
missing protons); 1H NMR (CDC13, 500 MHz, -40 °C) 8 0.71 (s, 3H), 0.72 (s, 3H), 2.29
(s, 3H), 2.44 (s, 3H), 2.55 (s, 3H), 2.61 (d, 1H, J: 13.5 Hz), 3.41 (d, 1H, J= 13.5 Hz),
3.65 — 3.59 (m, 2H), 7.15 (t, 1H, J= 7.5), 7.24 (d, 1H, J= 7.0 Hz), 7.34 — 7.37 (m, 3H),
7.75 (d, 2H, J= 8.0 Hz);13C NMR (CDCI3, 125 MHz, +25 °C) 8 21.19, 21.62, 23.67
(note: broad peak), 30.32, 33.79, 37.42, 78.49, 125.77, 125.95, 127.88, 129.80, 133.03,
133.53, 134.75, 139.54, 141.65, 144.67, 204.12; IR (thin ﬁlm) 3067w, 2971s, 2927s,
2884m, 1688vs, 1599, 1476s, 1456s, 1358vs, 1263vs, 1180vs, 1176vs, 1097s cm"; mass
spectrum, m/z (% rel intensity) 203 M+ -TsO (22), 202 (100), 187 (21), 173 (97), 157

(48), 145 (68), 129 (56), 115 (86);

102

 

o o
' PDC
OH
0 o
113 110

3-(2-Acetyl-6-methylphenyl)-2,2-dimethylpropanioc acid 110 - In a dry 25 mL round
bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere, pyridinium
dichromate (1.1510 g, 3.0595 mmol, 4.058 equivalent) was added to a solution of
aldehyde 113 (0.1646 g, 0.7540 mmol, 1.000 equivalent) in It dimethylforrnarnide (2.50
mL) and the resulting dark suspension was allowed to stir for 20 h. The suspension was
diluted with water and the organic materials extracted three times with diethyl ether.
The water layer was treated with a 1M aqueous lithium bromide solution and extracted
three times with diethyl ether. The combined ether layers were washed with a 1M
aqueous lithium bromide solution, water, and brine. The ether layer was then dried
using sodium sulfate, ﬁltered, and the ﬁltrate concentrated via rotary evaporation.
Column chromatography of the crude product using a silica gel colurrm chromatography
30% ethyl acetate/hexanes as elutant yields the desired acid 110 (0.1122 g, 0.4789

mmol, 72%) as a white solid.

Spectral data for 110 'H NMR(CDC13, 300 MHz) 8 1.10 (s, 6H), 2.35 (s, 3H), 2.56 (s,
3H), 3.45 (s, 2H, Note: broad peak), 7.16 (t, 1H, J= 7.5 Hz), 7.26 (d, 1H, J= 6.6 Hz),
7.40 (d, 1H, J= 7.5 Hz); 13C NMR (CDCl;, 75 MHz) 8 21.00, 24.79, 30.26, 35.45,
43.62, 125.97, 126.09, 133.49, 134.87, 139.63, 141.37, 184.18, 204.07; IR (thin ﬁlm)
2200 — 3500sb, 2992s, 1693s, 1575w, 1473m,1418m, 1383m, 1358m, 1292s, 1284s,
1163m cm"; mass spectrum, m/z (% rel intensity) 234 M+ (0.9), 216 (16), 188 (13), 173

(56), 160 (28), 147 (100), 115 (26), 91 (46); Melting point: 126 — 128 °c

103

S

NaO,
o N | o / |
\ N
OH 0’

O O S
DCC

 

110 120

2-mercaptopyridine-N-oxide-3-(2-(ethan-1-ol)-6-methylphenyl)-2,2-
dimethylpropionate 120: In a dry 25 mL round bottom ﬂask clad in aluminum foil and
containing a magnetic stir bar and nitrogen atmosphere, acid 110 (0.1943 g, 0.8293
mmol, 1.000 equivalent) was added in one portion to a suspension of anhydrous 2-
mercaptopyridine-N-oxide sodium salt (0.1578 g, 1.058 mmol, 1.275 equivalents), N,N’-
dicyclohexylcarbodiimide (0.2479 g, 1.201 mmol, 1.449 equivalents) in
dichloromethane (10.0 mL). The resulting suspension was allowed to stir at It for 3 d.
The yellow tinted suspension was then ﬁltered through Celite (rinsed with
dichloromethane until no yellow is observed on the Celite pad) and the ﬁltrate is
condensed via rotary evaporation. Avoid water workup as the product will decompose
rapidly. The thick, yellow, oily crude product is then puriﬁed via ﬂash chromatography
on a TLC grade, binderless silica gel column using 30% ethyl acetate hexanes as elutant.
The compound is readily visible as a bright yellow band on both TLC and column and
decomposes less rapidly on TLC grade silica than standard silica gel 60. The solvent
was removed under rt rotary evaporation then high vacuum (avoid heating the material)
to obtain a thick yellow oil in low to moderate yields (0.1542 g, 0.4490 mmol, 54%). In
order to improve puriﬁcation and prevent decomposition, reducing the amount of DCC
to 1 equivalent along with the acid may allow for just Celite column puriﬁcation, but

may extend reaction times. The product must be frozen for storage.

104

Spectral data for 120 1H NMR(CDC13, 300 MHz) 8 1.36 (s, 6H), 2.38 (s, 3H), 2.50 (s,
3H), 3.53 (s, 2H, Note: broad peak), 6.61 (td, 1H, J = 6.9, 1.8 Hz), 7.19 — 7.11 (m, 2H),
7.31 (d, 1H, J= 7.5 Hz), 7.50 (d, 1H, J= 7.5 Hz), 7.60 (dd, 1H, J: 8.7, 1.8 Hz), 7.90
(dd, 1H, J: 6.3, 1.8 Hz); 13C NMR(CDC13, 75 MHz) 8 21.40, 25.47, 30.01, 36.62,
43.29, 112.39, 126.31, 127.21, 133.41, 134.06, 135.34, 137.02, 138.33, 139.47, 172.93,
175.85, 203.58 (one sp2 carbon not observed); IR (thin ﬁlm) 31025, 3029s, 29785, 2930s,
2859m, 2510w, 2448w, 2249w, 2218w, 17925, l684sb, 1608s, 1578m, 15275, 14485,
14105, l391m, 13545, 12625, 1176s, 1134s, 10425, 10155 cm]; mass spectrum, m/z (%
rel intensity) the material appears to have broken down upon injection onto the GC,
multiple peaks were observed with no real evidence linking any of the peaks to the

desired product 120.

 

Mel
OH OMe
127 128

2,3-Dimethylanisole 128 - To a 100 mL schlenk tube containing a magnetic stir bar and
a nitrogen atmosphere, potassium carbonate (5.5133 g, 39.891 mmol, 1.0018
equivalents) was added to a solution of iodomethane (5.00 mL, 80.3 mmol, 2.02
equivalents), 2,3-dimethylphenol 127 (4.8647 g, 39.819 mmol, 1.0000 equivalent) and
acetone (40.0 mL). The resulting suspension was then heated in an oil bath at reﬂux for
48 h. The ﬂask was then allowed to cool to rt and its contents transferred into a 250 mL
round bottom ﬂask. The solvent was removed via rotary evaporation and the residue
dissolved in diethyl ether and water. The water layer was then extracted three times

with diethyl ether. The combined ether layers were washed twice with 1N sodium

105

hydroxide solution, once with water and brine before being dried over sodium sulfate.
The solvent was removed via rotary evaporation and the resulting crude product puriﬁed
via fractional distillation (atmospheric pressure, ~220 °C, potassium hydroxide is present
in the source pot to prevent contamination from remaining unreacted phenol). The
distillation yielded a clear oil that becomes a crystalline solid (4.2739 g, 31.382 mmol,
79%) that melts from body heat when handling at room temperature. This reaction is
often capricious and yields vary widely but often can be improved by adding more
equivalents of iodomethane if incomplete after 48 hours to obtain roughly the listed

yield.

The 1H NMR spectrum matched those of an authentic sample obtained from Aldrich.

Spectral data for 128 1H NMR(CDC13, 300 MHz) 8 2.15 (s, 3H), 2.27 (s, 3H), 3.81 (s,
3H), 6.72 (d, 1H, J= 8.4 Hz) 6.78 (d, 1H, J: 7.5 Hz) 7.06 (t, 1H, J= 7.8 Hz); 13C NMR

(CDCl3, 75 MHz) 8 11.51, 20.04, 55.50, 107.82, 122.19, 124.98, 125.76, 137.82, 157.50;

NBS
AIBN B,
OMe hv OMe

128 126

 

2-Methoxy-6-methylbenzylbromide 126 — To a dry 1L round bottom ﬂask ﬁtted with
an air condenser, containing a magnetic stir bar and a nitrogen atmosphere; 2,3-
dimethylanisole‘128 (8.1997g, 60.207 mmol, 1.0000 equivalent) was added to a
suspension of freshly powdered N-bromosuccinimide (10.7790 g, 60.559 mmol, 1.0058
equivalents), azobisisobutyronitrile (0.5424 g, 3.303 mmol, 0.05486 equivalents) in

carbon tetrachloride (800 mL). The suspension was irradiated with a 200 W

106

incandescent lamp for 6 h, during which time the production and consumption of an
auburn/tan color may occur. The reaction was monitored by discontinuing stirring and
observing whether the white solid succinimide particulate ﬂoats or sinks. Once all solid
particulates ﬂoat, the reaction is complete. The resulting suspension was then cooled to
0 °C and the solid succinimide byproduct removed via ﬁltration. The solvent was then
recovered via simple distillation and the remaining yellow/tan crude material was
transferred to a 50 mL round bottom ﬂask. Bulb to bulb distillation (75-125 °C air bath
temperature, 0.14 mm Hg) yielding 2-methoxy-6-methylbenzylbromide 126 and a
byproduct tentatively assigned 3-methyoxy-2-methylbenzylbromide in a 9:1 relative
ratio. The regioismomers are then dissolved in a minimal amount of hot hexanes and
allowed to recrystalize to yield white crystals of the desired regioisomer 126, which
were isolated via ﬁltration in moderate yield (7.1278 g, 33.139 mmol, 55%). Reaction
note: increasing the concentration of the photoreaction generates one additional

byproduct product which is thought to be the result of para bromination.

Spectral data for 126 1H NMR (CDC13, 300 MHz) 8 2.38 (s, 3H), 3.86 (s, 3H), 4.64 (s,
2H), 6.73 (d, 1H, J= 8.1 Hz) 6.78 (d, 1H, J= 7.5 Hz) 7.17 (t, 1H, J: 7.8 Hz); 13C NMR
(coch, 75 MHz) 8 18.77, 26.22, 55.77, 108.48, 122.76, 124.34, 129.45, 138.92,157.64;
IR (thin ﬁlm) 3069w, 3040w, 3000m, 2961s, 2938s, 2838s, 1599s, 1586vs, 1478vs,
1443vs, 1306m, 1267 vs, 1217vs, 1194m, 1143s, 1073vs cm'l; mass spectrum, m/z (%
rel intensity) 216 M+(81Br, 6), 214 M+ (7981', 6), 135 (100), 105 (92), 91 (50), 79 (33),

77 (26); The melting point was not taken.

107

1)LDA,

methyl- 0
OMe OMe

2)125

 

126 129

Methyl 3-(2-methoxy-6-methylphenyI)-2,2-dimethylpropanate 129: In a dry 100 mL
round bottom ﬂask containing a magnetic stir bar and a nitrogen atmosphere,
diisopropylamine (3.52 mL, 25.1 mmol, 1.19 equivalents) was added to tetrahydrofuran
(26.0 mL) and the solution was cooled to -78 °C. Once cooled, n-butyl lithium (1.34 M
solution in hexanes, 16.5 mL, 22.1 mmol, 1.05 equivalents) was added and allowed to
stir for 1 h at -78 °C. To the resulting solution of lithium amide was added dropwise
methyl isobutyrate (2.50 mL, 21.1 mmol, 1.00 equivalent) and then allowed to stir for l
more hour. To the -78 °C solution of lithium enolate, a solution of benzyl bromide 125
(4.6068 g, 21.418 mmol, 1.02 equivalents) in tetrahydrofuran (10.0 mL) was added
dropwise, the dry ice bath repacked, and the solution was allowed to stir overnight while
gradually warming as the dry ice was consumed. Upon completion, the reaction was
diluted with water and the organic materials extracted three times with diethyl ether.
The combined organic layers were washed with 1N hydrochloric acid, water, and brine;
dried over sodium sulfate; and concentrated via rotary evaporation. The resulting crude
product was distilled via bulb to bulb distillation (150 — 160 °C air bath temperature at
0.10 mm Hg) to yield the desired title ester as a clear, colorless oil in consistently good

yields (4.311 g, 18.24 mmol, 87%).

Spectral data for 129 1H NMR (CDC13, 300 MHz) 8 1.15 (s, 6H), 2.27 (s, 3H), 2.98 (s,

2H), 3.64 (s, 3H), 3.71 (s, 3H), 6.66 (d, 1H, J= 8.1 Hz), 6.76 (d, 1H, J= 7.8 Hz), 7.07 (t,

108

1H, J= 7.8 Hz); 13C NMR (CDC13, 75 MHz) 8 20.43, 25.40, 35.68, 43.46, 51.57, 55.01,
107.54, 122.73, 125.39, 126.81, 138.69, 158.32, 178.60; [R (thin ﬁlm) 3067w, 29505,
2877m, 2836 m, 17305, 1601w, 15845, 14725, 1387m, 1366w, 1331m, 12625, 11925,
11345 cm"; mass spectrum, m/z (% rel intensity) 236 M+ (54), 177 (32), 136 (58), 135

(100), 105 (92), 91 (69), 79 (61), 77 (63), 65 (34);

O OH
@088“ ”‘H 00%
OMe OMe

129 123a

 

3-(2-methoxy-6-methylphenyl)-2,2-dimethylpropanol 123 a - In a dry 250 mL round
bottom ﬂask containing a magnetic stir bar and a nitrogen atmosphere, the ester 129
(3.7512 g, 15.874 mmol, 1.0000 equivalent) is added dropwise to a room temperature
solution of lithium aluminum hydride (0.7716 g, 20.33 mmol, 1.281 equivalents) in
tetrahydrofuran (40.0 mL) and allowed to stir for 4 hours at rt. Upon completion, the
reaction was quenched by dropwise addition of water to the dark grey suspension. Once
quenched, the now light grey/off white suspension was diluted with water and the
organic materials extracted three times with diethyl ether. The combined ether layers
were then washed with 1 N hydrochloric acid, water, and brine before being dried over
sodium sulfate. The dry organic layer was ﬁltered and the ﬁltrate concentrated by rotary
evaporation. The crude product was puriﬁed by bulb to bulb distillation (155 - 170 °C air

bath at 0.09 mm Hg) to yield 123 a as a clear oil (3.0843 g, 14.807 mmol, 93%).

Spectral data for 123 a 1H NMR(CDC13, 300 MHz) 8 0.91 (s, 6H), 2.32 (s, 3H), 2.64 (s,

2H), 3.10 (s, 3H, Note: when treated with 1 drop of deuterated water, it becomes (5,

109

2H)), 3.82 (s, 3H) 7.73 (d, 1H, J= 8.4 Hz), 6.82 (d, 1H, J= 7.5 Hz), 7.09 (t, 1H, J= 7.8
Hz); l3C NMR(CDC13, 75 MHz) 8 20.95, 25.22, 32.84, 38.37, 55.71, 70.28, 108.02,
123.69, 126.25, 126.53, 139.44, 157.82; IR (thin ﬁlm) 3447sb, 3065m, 29538, 28705,
2836s, 1599m, 1582m, 14705b, 1383m, 1362m, 1325m, 12468, 1194m, 1171m, 11408,
10765, 10448 cm“; mass spectrum, m/z (% rel intensity); 208 M+ (70), 177 (12), 136

(86), 135 (100), 120 (58), 105 (91), 91 (71), 79 (53), 77 (63);

OH
OMe O
130

123a

 

3,3,5-Trimethylchroman 130— In a dry 100 mL round bottom ﬂask ﬁtted with a
condenser, containing a magnetic stining and a nitrogen atmosphere, thionyl chloride
(1.70 mL, 23.3 mmol, 1.20 equivalents) was added to a solution of alcohol 123 a (4.0288
g, 19.341 mmol, 1.0000 equivalent) in toluene (33 mL). The solution was then heated at
90 °C overnight. After conﬁrmation of the 123 a’s consumption by TLC, the solution
was cooled to room temperature, diluted with water, and the organic material extracted
three times with diethyl ether. The combined organic layers were washed with saturated
aqueous sodium bicarbonate, water, and brine before being dried over sodium sulfate
and concentrated via rotary evaporation. The resulting crude product was puriﬁed via
silica gel column chromatography using hexanes as elutant to yield the ether 130 as a

clear oil that solidiﬁes upon standing (2.2440 g, 12.732 mmol, 66%).

Spectral data for 130 1H NMR (CDC13, 300 MHz) 8 1.04 (s, 6H), 2.19 (s, 3H), 2.39 (s,

2H), 3.71 (s, 2H), 6.71 (d, 1H, J= 8.7 Hz), 6.74 (d, 1H, J= 7.5 Hz), 7.01 (t, 1H, J= 7.5

110

Hz); 13C NMR (CDC13, 75 MHz) 8 19.09, 25.27, 28.76, 36.96, 74.95, 114.13, 120.36,
121.83, 126.43, 137.77, 153.67; [R (thin ﬁlm) 3073w, 3033w, 2957s, 29248, 28708,
2834w, l604w, 1586s, 1474s, 1447m, 1393w, 1366w, 1329w, 1308m, 1281m, 12528,
12428, 1181m, 1084s, 10678 cm"; mass spectrum, m/z (% rel intensity) 176 M+ (100),
161 (36), 147 (91), 120 (82), 105 (61), 92 (86), 91 (92), 77 (67); Melting point: 32 — 33.5

°C

130 131
3 (2-Hydroxy-6-methylphenyI)-2,2-dimethylpropylbromide 131: - In a dry 50 mL
round bottom ﬂask clad in aluminum foil and containing a magnetic stir bar and a
nitrogen atmosphere, concentrated boron t1ibromide(0.95 mL, 10. mmol, 3.4
equivalents) was added to a solution of ether 130 (0.5082 g, 2.883 mmol, 1.000
equivalent) and dichloromethane (10.0 mL) at rt. The deep red solution was allowed to
stir for 48 h. Upon completion, the reaction was poured into a seperatory funnel and
diluted with water. The organic material was extraced three times with
dichloromethane. The combined organic layers were washed twice with water and once
with brine before being dried over sodium sulfate. The dichlormethane solution was
then concentrated via rotary evaporation and the resulting dark oil was puriﬁed using
silica gel column chromatography with 5% ethyl acetate/hexanes as elutant. The product
131 was obtained as a tan oil in moderate to good yields (0.5735 g, 2.230 mmol, 77%).
The product needs to be protected from light while stored. Yeilds ranged from mid 60 %

to the upper 70’s.

111

Spectral data for 131 1H NMR(CDC13, 300 MHz) 8 1.07 (s, 6H), 2.32 (8, 3H), 2.79 (s,
2H), 3.47 (s, 2H), 4.79 (s, 1H, note: broad peak), 6.58 (d, 1H, J = 8.1 Hz), 6.77 (d, 1H, J
= 7.5 Hz), 6.98 (t, 1H, J= 7.8 Hz), 13C NMR (CDC13, 75 MHz) 8 20.82, 26.15, 35.24,
38.11, 48.58, 112.93, 123.18, 123.57, 126.83, 139.51, 154.50; IR (thin layer) 3524sb,
3067m, 3029m, 29678, 29348, 28728, 1607m, 15828, 14648, 1427m, 13858, 1366s,
1296s, 12698, 11008, 10248 cm"; mass spectrum, m/z (% rel intensity) 258 M+(81Br,

10), 256 M+ (79Br, 10), 200 (4), 176 (8), 147 (12), 122 (37), 121 (100), 91 (50);

Br TBSCI Br
(EX lmidazole w
OH OTBS

131 132

 

3 (2-tert-butyldimethylsilyloxy-6-methylphenyl)-2,2-dimethylpropylhromide 132: -
In a 25 mL round bottom ﬂask containing a magentic stirring and a nitrogen atmosphere,
tert butyldimethylsilyl chloride (0.3288 g, 2.189 mmol, 1.888 equivalents) was added to
a rt solution of phenol 131 (0.2982 g, 1.159 mmol, 1.000 equivalent), imidazole (0.2557
g, 3.756 mmol, 3.241 equivalents) in dimethylformamide (3.00 mL) and the resulting
solution allowed was to stir overnight. Upon completion, the reaction was diluted with
water and the organic materials extracted three times with ethyl acetate. The combined
organic layers were washed twice with 1 M aqueous lithium bromide, then once with 1
N hydrochloric acid, water, and brine before being dried over sodium sulfate. The crude
product was puriﬁed using silica gel column chromatography with hexanes as elutant to

obtain 132 as a clear oil in decent yields (0.2970 g, 0.7996 mmol, 69%).

Spectral data for 132 1H NMR(CDC13, 300 MHz) 8 0.20 (s, 6H), 0.99 (s, 9H), 1.01 (s,

6H), 2.30 (s, 3H), 2.77 (s, 2H, note: broad peak), 3.43 (s, 2H), 6.64 (d, 1H, J = 7.5 Hz),

112

6.76 (d, 1H, J: 7.5 Hz), 6.96 (t, 1H, J= 7.8 Hz); l3C NMR(CDC13, 75 MHz) 8 -3.85,
18.43, 21.16, 25.97, 26.10, 35.74, 3837,4882, 116.31, 123.45, 126.27, 127.98, 139.11,
154.73; 1R (thin layer) 3065w, 3027w, 2957s, 2930s, 2884s, 2859s, 1595m, 1580s,
1464s, 1385m, 1364m, l327w, 1256s, 1179m, 1152m, 1138m, 1103m, 1036s cm"; mass,
spectrum, m/z (% rel intensity) 315 M+ -C4H9 (SlBr, 3), 313 M+ -C4H9 (79Br, 3), 259

(95), 257 (96), 235 (32), 179 (84), 177 (64), 163 (54), 149 (33), 105 (51), 73 (100);

OH (I)
OMe OMe

123 a 124

 

3-(2-methoxy-6-methylphenyl)-2,2-dimethylpropanal 124: — In a 100 mL round
bottom ﬂask ﬁtted with a water condenser, containing a magnetic stir bar and open to the
atmosphere, [BX ( 5.9051 g, 21.088 mmol, 2.1489 equivalents) was added to a solution
of alcohol 123 a (2.0441 g, 9.8133 mmol, 1.0000 equivalent) in ethyl acetate (46.0 mL).
The white suspension was then heated at reﬂux for 4 h. The reaction was then cooled to
rt, and the solid ﬁltered. The solid was rinsed three times with ethyl acetate and the
combined ethyl acetate layers were then washed with 1N sodium hydroxide, water, and
brine before being dried over sodium sulfate. The dried ethyl acetate solution was
concentrated via rotary evaporation and the crude product puriﬁed using a silica gel
column chromatography with 10% ethyl acetate/hexanes as elutant to yield the aldehyde

124 as a clear oil in good yield (1.6478, 7.9882 mmol, 81%)

Spectral data for 124 1H NMR(CDC13, 300 MHz) 8 1.02 (s, 6H), 2.29 (s, 3H), 2.85 (s,
2H), 3.69 (s, 3H), 6.65 (d, 1H, J= 8.1 Hz), 6.77 (d, 1H, J= 7.5 Hz), 7.07 (t, 1H, J= 7.8

Hz), 9.45 (s, 1H); l3C NMR(CDC13, 75 MHz) 8 20.50, 21.82, 34.27, 46.89, 54.65,

113

107.73, 122.88, 124.62, 127.02, 138.32, 157.34, 205.07; IR (thin layer) 3067m, 29658,
29318, 28708, 28388, 27135, 2780m, 2720m, 17255, 1601m, 1584s, 1472s, 1397m,
1363m, 1321m, 12608, 1156m, 1142m, 10808 cm"; mass spectrum, m/z (% rel intensity)

206 M+ (7), 135 (100), 105 (82), 91 (61), 77 (43);

 

t-BuOK
+
OMe OMe OMe
124 134 a 134 b

Methyl 4-(2-methoxy-6-methylphenyl)-2,2-dimethylbut-l-ene ethers 134 a (E) and
134 b (Z): - In a dry 100 mL round bottom ﬂask containing a magnetic stirbar and an
argon atmosphere, powdered potassium tert-butylalkoxide (0.6559 g, 5.845 mmol, 2.501
equivalents) was added in one portion to a suspension of methoxy methyl
triphenylphosphonium chloride (1.9960 g, 5.8225 mmol, 2.491 equivalents) in rt
tetrahydrofuran (20 mL). The resulting thick red suspension was allowed to stir for 1 h
then cooled to -78 °C, upon which the red color paled. Aldehyde 124 (0.4821 g, 2.337
mmol, 1.000 equivalent) was added dropwise causing the red color to lighten. The dry
ice bath was reﬁlled and the reaction allowed to stir overnight and gradually warm to rt
as the dry ice was consumed. Upon veriﬁcation that the room temperature suspension
had consumed all of the aldehyde via TLC, the reaction was quenched with water and
the organic material was extracted three times with diethyl ether. The combined ether
layers were washed with water, brine and dried over sodium sulfate. (NOTE: Exposure
to dilute acid during the workup will result in hydrolysis of the cis enol ether and
formation of the aldehyde 135 described below, the trans enol ether reacts much slower.)

After concentration via rotary evaporation, the mixture of enol ethers were puriﬁed by

114

silica gel chromatography using 1% ethyl acetate/hexanes as elutant to yield both enols

in good yield (0.4135 g, 1.764 mmol, 75%) in a roughly 5:4 mixture of cis to trans.

Spectral data for 134 a 1H NMR (CDC13, 300 MHz) 8 1.03 (s, 6H), 2.29 (s, 3H), 2.71 (s,
2H), 3.44 (s, 3H), 3.74 (s, 3H), 4.84 (d, 1H, J = 12.9 Hz), 6.08 (d, 1H, J = 13.2 Hz), 6.69
(d, 1H, J= 7.2 Hz), 6.76 (d, 1H, J= 7.5 Hz), 7.06 (t, 1H, J= 8.1 Hz); 13C NMR(CDC13,
75 MHz) 8 20.97, 28.42, 36.54, 38.70, 54.87, 55.76, 107.66, 114.65, 122.74, 126.25,
126.57, 138.94, 144.14, 158.38; IR (thin layer) cm'1 3071w, 2957s, 2869m, 2832m,
1669m, 16518, 15828, 14698, 1383w, 1364w, 1339w, 1269s, 1256s, 1209s, 1157s,
1119s, 1076s; mass spectrum, m/z (% rel intensity) M+ 234 (0.2), 219 (0.3), 135 (12),

105 (27), 100 (38), 99 (100), 91 (26);

Spectral data for 134 b 1H NMR (CDC13, 300 MHz) 8 1.11 (s, 6H), 2.34 (8, 3H), 2.84 (s,
2H), 3.49 (s, 3H), 3.74 (s, 3H), 4.27 (d, 1H, J = 7.2 Hz), 5.66 (d, 1H, J = 7.2 Hz), 6.69
(d, 1H, J= 8.1 Hz), 6.77 (d, 1H, J= 7.8 Hz), 7.05 (t, 1H, J= 7.8 Hz); 13C NMR (CDC13,
75 MHz) 8 20.94, 28.62, 37.47, 37.62, 54.89, 59.56, 107.65, 116.70, 122.69, 126.09,
127.187, 139.21, 144.67, 158.55; [R (thin layer) 3065w, 3022m, 29558, 29328, 2869m,
2834m, 1659m, 1582m, 14708, 1394w, 1377w, 1360w, 1321w, 12718, 1248m, 1208m,
11348, 11018, 10785 cm"; mass spectrum, m/z (% rel intensity) M+ 234 (0.3), 219 (0.6),

135 (23), 105 (47), 100 (65), 99 (100), 91 (43);

115

o\ o

| / 0‘
I 3N HCI
.1.
OMe OMe OMB

1348 134b 135

 

5-(2-Methyoxy-6—methylphenyl)-4,4-dimethylbutanal 135: In a 250 mL round
bottom ﬂask containing magnetic stir bar and open to the atmosphere, 3N hydrochloric
acid (34.0 mL) was added to a solution of enol ethers 134 a and 134 b (0.6272 g total,
2.677 mmol, 1.000 equivalent, ratio not recorded) in tetrahydrofuran (34.0 mL) and
allowed to stir for 2 h. After ensuring consumption of the trans isomer by TLC (the cis
regioisomer is consumed much more rapidly), the reaction is transferred to a seperatory
ﬁrnnel and the organic materials are extracted three times with ethyl acetate. The
combined ethyl acetate layers were then washed with water and brine before being dried
over sodium sulfate. The organic layer was concentrated via rotary evaporation. The
crude material 135 is generally pure enough to proceed to the next step but it may be
further puriﬁed using silica gel column chromatography with 5% ethyl acetate/hexanes

as elutant. This provided 135 as a clear oil in good yields (0.5348, 2.427 mmol, 91%).

Spectral data for 135 ‘H NMR (CDC13, 300 MHz) 8 1.07 (s, 6H), 2.24 (d, 2H, J= 3.0
Hz), 2.31 (s, 3H), 2.79 (s, 2H), 3.70 (s, 3H), 6.69 (d, 1H, J= 8.1 Hz), 6.79 (d, 1H, J=
7.5 Hz), 7.09 (t, 1H, J: 8.1 Hz), 9.83 (t, 1H, J= 3.0 Hz); 13(3 (CDClg, 75 MHz) 8 20.89,
28.29, 36.17, 37.11, 54.26, 54.71, 107.78, 123.05, 125.71, 126.78, 138.97, 157.99,
204.28; IR (thin ﬁlm) 3065w, 2959s, 2872s, 2836s, 2737w,1716s, 1599m, 1582s, 1472s,
1406w, 1387m, 1368m, 1329w, 12718, 12608, 12468, 1171m, 11428, 10808, 1047m cm'
‘; mass spectrum, m/z (% rel intensity) 220 M+ (12), 176 (7), 136 (23), 135 (100), 105

(68), 91 (22), 77 (20);

116

 

°\ Ph3PMel \
n-BuLi
OMe OMe

135 122 a
5-(2-Methyoxy-6-methylphenyl)-4,4-dimethylpent-l-ene 122 a: — In a dry 250 mL
round bottom ﬂask containing a magnetic stirbar and an argon atmosphere, n-butyl
lithium (8.80 mL of 2.5 M in hexanes, 22 mmol, 1.5 equivalents) was added dropwise to
the 0 °C suspension of methyl triphenylphosphonium iodide (10.4201 g, 25.778 mmol,
1.8117 equivalents) in tetrahydrofuran (160 mL). The resulting thick yellow suspension
was allowed to warm to rt and stir for 1 h. The ylide was then cooled to 0°C and the
aldehyde 135 (3.1345 g, 14.228 mmol, 1.0000 equivalent) was added dropwise over 5
minutes. The reaction mixture quickly paled in color and was allowed to stir overnight
while gradually warming to rt as the ice was consumed. The reaction was quenched
with water and the organic material was extracted three times with diethyl ether. The
combined ether layers were washed with water and brine then dried over sodium sulfate.
After concentration via rotary evaporation, the crude alkene was puriﬁed on a silica gel
column with hexanes as elutant to isolate alkene 122 a as a clear oil in good yield

(2.9089 g, 13.323 mmol, 94%).

Spectral data for 122 a 1H NMR(CDC13, 300 MHz) 8 0.84 (s, 6H), 2.07 (d, 2H, J= 7.5
Hz), 2.30 (s, 3H), 2.65 (s, 2H), 3.74 (s, 3H) 4.97 - 5.05 (m, 2H), 5.91 (ddt, 1H, J: 16.5,
10.5, 7.2 Hz), 6.69 (d, 1H, J= 8.1 Hz), 6.77 (d, 1H, J= 7.5 Hz), 7.06 (t, 1H, J= 7.8
Hz); 13C NMR (CDC13, 75 MHz) 8 20.94, 26.89, 36.72, 37.01, 48.07, 54.87, 107.65,
116.59, 122.79, 126.16, 127.03, 136.31, 138.95, 158.48; 1R (thin ﬁlm) 3072m, 29598,

29318, 2871m, 2834m, 1638m, 1599m, 15828, 14705, 14395, 1385m, 1363m, 1280m,

117

1261s, 1244s, 1071m, 1511m, 1092s cm"; mass spectrum, m/z (% rel intensity) 218 M+

(64), 177 (41), 136 (92), 135 (100), 121 (56), 105 (91), 91 (77), 79 (71), 77(73), 55 (80);

\ thPH \
n BuLi
OMe OH

122 a 136

 

5-(2-hydroxy-6-methylphenyl)-4,4-dimethylpent-l-ene 136: — In a dry 25 mL round
bottom ﬂask containing a magnetic stirbar and an argon atmosphere, n-butyl lithium
(3.00 mL of 2.5 M in hexanes, 7.5 mmol, 5.3 equivalents) was added dropwise to a 0 °C
solution of diphenylphosphine (1.20 mL, 6.90 mmol, 4.88 equivalents) in
tetrahydrofuran (8.40 mL). The resulting red solution was then warmed to rt for 45 min
then anisole 122 a (0.3087 g, 1.414 mmol, 1.000 equivalent) was added dropwise. The
reaction ﬂask was ﬁtted with a water condenser and heated to reﬂux for 48 h. The red
solution was diluted with 1N hydrochloric acid and extracted three times with diethyl
ether. The combined ether layers were washed with water and brine before being dried
over sodium sulfate. The ether layer was then concentrated via rotary evaporation and
the resulting crude product was puriﬁed on a silica gel colurrm with 5% ethyl
acetate/hexanes as elutant to yield the phenol 136 in excellent yields as a clear oil (0.290

g, 1.41 mmol, quantitative). Yields range from 93% to quantitative.

Spectral data for 136 1H NMR(CDC13, 300 MHz) 8 0.91 (s, 6H), 2.11 (d, 2H, J: 6.3
Hz), 2.30 (s, 3H), 2.62 (s, 2H), 4.64 (s, 1H), 5.01 — 5.08 (m, 2H), 5.91 (ddt, 1H, J= 16.8,
10.5, 7.5 Hz), 6.60 (d, 1H, J: 7.8 Hz), 6.75 (d, 1H, J= 7.5 Hz), 6.97 (t, 1H, J= 7.5
Hz); l3C NMR(CDC13, 75 MHz) 8 20.98, 26.93, 36.92, 37.31, 48.04, 112.88, 117.18,

123.02, 124.55, 126.45, 135.83, 139.47, 154.54; [R (thin ﬁlm) 35328b, 30715, 3029m,

118

2963s, 2873s, 1638m, 1608m, 15818, 14648, 13858, 13668, 12378, 1204m, 1148s,
1080m cm"; mass spectrum, m/z (% rel intensity) 204 M+ (40), 163 (14), 147 (10), 122

(86), 121 (100), 91 (52), 83 (55), 77 (62), 55 (92);

\ T120 \
NaH
OH OTf

136 122 b

 

5-(2-O-Triflate-6-methylphenyl)-4,4-dimethylpent-l-ene 122 b: - In a 25 mL round
bottom ﬂask containing a magnetic stirbar and nitrogen atmosphere, sodium hydride
(60% in mineral oil, 0.0445 g, 1.1 mmol, 2.3 equivalents) was added to a 0 °C solution of
phenol 136 (0.0996 g, 0.487 mmol, 1.00 equivalent) in dichloromethane (3.00 mL). The
resulting greenish grey solution was allowed to stir for 15 minutes then triﬂic anhydride
(0.095 mL, 0.56 mmol, 1.2 equivalents) was added. The green solution became yellow
and darkened to a tan while the reaction stirred at 0 °C for 2 h. After veriﬁcation that the
reaction was complete by TLC, the reaction was quenched with water and the organic
material extracted three times with dichloromethane. The ivory colored combined
organic layers were then washed with water and brine before being dried over sodium
sulfate. The resulting solution was concentrated via rotary evaporation and the resulting
crude product was puriﬁed using silica gel column chromatography with hexanes as
elutant (the crude material requires dichloromethane to load onto a column). This

yielded the triﬂate as a clear oil 122 b in good yield (0.1438 g, 0.4275 mmol, 88%).

Spectral data for 122 b 1H NMR (CDC13, 500 MHz) 8 0.86 (s, 6H), 2.09 (d, 2H, J = 7.5
Hz), 2.37 (s, 3H), 2.72 (s, 2H), 5.02 — 5.09 (m, 2H), 5.86 (ddt, 1H, J= 17.0, 10.5, 7.5

Hz), 7.11 (dd, 1H, J= 8.0, 2.0 Hz), 7.17 (t, 1H, J= 7.5 Hz), 7.18 (d, 1H, J= 7.5 Hz); 13C

119

NMR(CDC13, 125 MHz) 8 21.11, 26.42, 37.10, 37.95, 48.17, 117.60 , 118.55 (q, J=
318.1 Hz), 118.73, 127.02, 130.39, 131.10, 134.97, 140.99, 149.16; 1R (thin ﬁlm)
3077m, 3029w, 3005m, 2967s, 2880m, 1640m, 1611w, 1057w, 14708, 1420vs, 1389m,
1368m, 12508, 1215vs, 1181m, 1142vs, 1094m, 1075m cm"; mass spectrum, m/z (% rel
intensity) 336 M+ (1.5), 321 (1.5), 295 (15), 253 (47), 162 (77), 147 (41), 121 (47), 105

(42), 91 (75), 83 (98), 82 (85), 55 (100);

 

l
O H, N DU 0 O
(:6 TN \ C(kN/Vrh\ 1) n BuLi 6:” 65'
——-——> I +
Cl n BuLi CI 2) Mel Cl Cl
147 148 149 150

2-chloro—6—methylbenzaldehyde 149: In a dry 500 mL round bottom ﬂask containing a
magnetic stir bar and a nitrogen atmosphere, n-butyl lithium (1.58 M in hexanes, 47.3
mL, 74.7 mmol, 1.02 equivalents) was added in a rapid dropwise manner to a -78 °C
solution of N,N,N’- trimethylethylenediamine (10.2 mL, 78.8 mmol, 1.08 equivalents) in
tetrahydroﬁrran (200 mL). The resulting solution was allowed to stir for 1 h then
aldehyde 147 (8.20 mL, 72.9 mmol, 1.00 equivalent) was added dropwise and allowed to
stir for 3 h at -30 °C. During this time, a precipitate may be observed, but it doesn’t
appear to affect the yields. After formation of the intermediate 148, the round bottom
ﬂask was ﬁtted with an addition funnel and n-butyl lithium (1.58 M in hexanes, 150.
mL, 237 mmol, 3.25 equivalents) was added dropwise over ~0.5 h to the -30 °C
solution and then allowed to stir for 4 h. After 4 h, the aryl lithium was transferred
dropwise via cannula over approximately 1 h into a 1 L round bottom ﬂask containing a
-78 °C solution of methyl iodide (30.0 mL, 482 mmol, 6.61 equivalents) and

tetrahydrofuran (120 mL). The -78 °C dry ice/acetone bath was repacked and the

120

reaction was allowed to stir overnight, gradually warming to rt as the dry ice bath was
consumed. The yellow solution was then diluted with water and the organic material
was extracted three times with diethyl ether. Upon exposure to air, the organic layer will
darken to a deep red shade. The combined organic layers were washed with 1N
hydrochloric acid, water, saturated sodium thiosulfate, water, and brine before being
dried over sodium sulfate. The lightly yellow tinted solution was then concentrated via
rotary evaporation and the crude product puriﬁed using bulb to bulb distillation (0.09
mm Hg with a 100-130 °C air bath temperature) to yield the desired aldehyde 149 in
moderate yields (5.7344 g, 37.094 mmol, 51%) as a clear oil that can solidify upon
standing if the impurity 150 is in trace amounts. If any starting aldehyde 147 is present,
repeat the distillation as it will not be removed via recyrstalization after the next step —
unlike 150. The benzaldehyde 149 must be stored under an oxygen ﬁee environment to

prevent signiﬁcant conversion of the aldehyde to the benzoic acid.

Spectral data for 149 1H NMR(CDC13, 300 MHz) 8 2.56 (s, 3H), 7.14 (d, 1H, J = 7.5
Hz), 7.28 (dd, 1H, J: 8.1, 1.5 Hz), 7.33 (t, 1H, J= 7.8 Hz) 10.63 (s, 1H) ”C NMR
(CDCl;, 75 MHz) 8 21.27, 128.30, 130.70, 133.53, 139.05, 142.51, 192.48 (one 5p2
quaternary carbon is not observed); IR (thin ﬁhn) 30655, 29848, 29685, 27688 1698vs,
15918, 1564vs, 1455vs cm'l; mass spectrum, m/z (% rel intensity) 156 M+ (”CL 60),

154 M+(35C1, 88), 153 (100), 127 (44), 125 (85), 91 (85), 89 (98), 63 (89).

121

 

?
06 ”” 00°“
Cl CI
149 151

2-chloro—6-methylbenzalcohol 151: In a dry 500 mL round bottom ﬂask containing a
magentic stirbar and a nitrogen atmosphere, lithium aluminum hydride (1.4860 g, 39.16
mmol, 1.394 equivalents) was added to tetrahydrofuran (100 mL) to create a grey
suspension. To the rt solution was added aldehyde 149 (4.3419 g, 28.087 mmol, 1.0000
equivalent)(either as a neat oil or as a solution of aldehyde in minimal tetrahydrofuran if
solid) dropwise while observing the reaction ﬂask temperature, pausing addition until it
cools if the reaction ﬂask gets warm to the touch. (Note: for larger reactions, the ﬂask is
cooled to 0 °C prior to aldehyde addition). After addition, the suspension is allowed to
stir at rt for 4 h. Upon veriﬁcation of the consumption of the aldehyde by TLC, the
reaction is quenched by very slow dropwise addition of water to the suspension. Once all
bubbling has ceased, the white suspension is diluted with water and the organic material
is extracted three times using diethyl ether. The combined organic layers were washed
with 1N hydrochloric acid, water and brine and dried over sodium sulfate. The organic
layer is ﬁltered and the ﬁltrate is then concentrated via rotary evaporation to yield a
white solid that is puriﬁed by recrystallation in hexanes to recover the pure 151 white
solid in moderate yields (2.6852 g, 17.146 mmol, 61%). Additional product may be
obtained by concentrating and recrystalizing the mother liquor which often results in

cumulative yields in the high 70’s.

Spectral data for 151 1H NMR (CDC13, 300 MHz) 8 2.45 (s, 3H), 4.83 (s, 2H), 7.08 (d,

1H, J= 7.8 Hz), 7.12 (t, 1H, J= 7.5 Hz) 7.21 (d, 1H, J: 7.5 Hz); ”C NMR (CDC13, 75

122

MHz) 8 19.64, 59.54, 127.25, 128.99, 129.17, 134.78, 135.83, 139.60; IR (KBr Pellet)
32258b, 2924s, 2741m, 2614w 1595m, 1568m, 1453s, 1381m, 1319m, 1200m, 1173s,
1143s, 10138 cm"; mass spectrum, m/z (% rel intensity) 158 M+ (”CL 10), 156 M+

(”CL 32), 140 (67), 138 (100), 119 (29), 103 (88), 91 (97), 77 (98); Melting point: 83 —

00:" it 008'
CI Cl

151 139

84 °C

2-chloro—6-methylbenzylhromide 139 — In a dry 250 mL round bottom ﬂask containing
a magnetic stirbar and a nitrogen atmosphere, phosphorus tribromide (2.55 mL, 27.1
mmol, 1.10 equivalents) was added to a solution of benzyl alcohol 151 (3.8415 g, 24.529
mmol, 1.0000 equivalent) in dichloromethane (60 mL) and allowed to stir at room
temperature for up to 4 h. Upon veriﬁcation that the alcohol is consumed via TLC,
water is added dropwise until no further signs of reaction are observed, then the reaction
is further diluted with water and the organic materials extracted three times with
dichloromethane. The combined organic layers were washed with saturated aqueous
sodium bicarbonate, water, and brine; dried over magnesilun sulfate; and concentrated
via rotary evaporation to yield an off white pungent solid (Note: irritant). The solid is
then melted and distilled via bulb to bulb distillation to obtain a bright white solid in
good yield (4.1919 g, 19.097 mmol, 78 %). One note regarding bulb to bulb distillation:
while signs of the material coming over were observed at 110 °C air bath temperature at
0.09 mm Hg, a 135 °C air bath temperature was used to prevent condensed and solidiﬁed

product ﬁ'om clogging of the distillation apparatus.

123

Spectral data for 139 1H NMR (CDCl,, 300 MHz) 8 2.43 (s, 3H), 4.65 (s, 2H), 7.07 (d,
1H, J: 7.5 Hz), 7.12 (t, 1H, J= 7.8 Hz) 7.23 (d, 1H, J= 7.2 Hz); 13C NMR (CDC13, 75
MHz) 8 19.50, 28.16, 127.53, 129.13, 129.39, 133.50, 134.87, 139.45; IR (thin ﬁlm)
3063w, 3021w, 2998w, 2980w, 2955w, 2910w, 2865w, 1591s, 1568s, 1456s, 1445s,
1421m, 1379m, l221w, 1201w, 1179s, 1157w, 11018 cm'l; mass spectrum, m/z (% rel
intensity) 222 M+ (8‘Br3’C1, 4) 220 M+(8'Br35C1 and 7913870, 18); 218 M+ (79Br35Cl,

14), 141 (68), 139 (100), 103 (78), 77 (65); Melting point: 48 — 50 °C

 

1) LDA,
methyl- 0
def isobutyrate WY“
CI 2) 139 CI
1 39 1 52

Methyl 3-(2-chloro-6-methylphenyl)-2,2-dimethylpropionate 152: In a dry 100 mL
round bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere,
diisopropylamine (3.00 mL, 21.4 mmol, 1.17 equivalents) was added to tetrahydrofuran
(22.0 mL) and the solution was cooled to -78 °C. Once cooled, n-butyl lithium (14.1 mL
of a 1.36 M solution in hexanes, 19.2 mmol, 1.05 equivalents) was added and allowed to
stir for 1 h. To the resulting solution of lithium amide was added dropwise a solution of
methyl isobutyrate (2.14 mL, 18.6 mmol, 1.02 equivalents) in tetrahydrofuran (10.0 mL)
and then allowed to stir for 1 h. To the lithium enolate, a solution of benzyl bromide 139
(4.012 g , 18.28 mmol, 1.000 equivalent) in tetrahydrofuran (10.0 mL) was added
dropwise, the dry ice bath repacked, and the solution was allowed to stir overnight while
the dry ice was consumed and the reaction warmed to rt. After verifying the
consumption of the benzyl bromide via TLC, the reaction was diluted with water and the

organic materials extracted three times with diethyl ether. The combined organic layers

124

 

were washed with 1N hydrochloric acid, water, and brine; dried over sodium sulfate; and
concentrated via rotary evaporation. The resulting crude oil was distilled via bulb to
bulb distillation (air bath temperature of 145-160°C at 0.10 mm Hg) yielding the desired

title ester 152 as a clear, colorless oil in consistently good yields (3.9787, 16.527 mmol,

90%).

Spectral data for 152 1H NMR(CDC13, 300 MHz) 8 1.21 (s, 6H), 2.28 (s, 3H), 3.17 (s,
2H), 3.64 (s, 3H), 6.99 - 7.05 (m, 2H), 7.15 - 7.20 (m, 1H); ”C NMR (CDCl;, 75 MHz)
8 21.15, 25.54, 38.65, 43.71, 51.90, 127.15, 127.28, 128.98, 134.76, 136.03, 139.62,
178.21; IR (thin ﬁlm) 3061w, 29778, 29508, 2876m 1732vs, 1593w, 1568w, 1473s,
1452s, 1389 m, 1273m cm"; mass spectrum, m/z (% rel intensity) 205 M+ -Cl (87), 181

(20), 141 (75), 139 (100), 103 (48);

O OH
riots a. 0%
Cl Cl

152 153

 

3-(2-chloro-6-methylphenyl)-2,2-dimethylpropanol 153: — In a dry 250 mL round
bottom ﬂask containing a magnetic stirbar and a nitrogen atmosphere, methyl ester 152
(3.6171 g, 15.026 mmol, 1.0000 equivalent) was added dropwise into a room
temperature suspension of lithium aluminum hydride (0.7812 g, 20.58 mmol, 1.370
equivalents) in tetrahydrofuran (40 mL) The reaction was allowed to stir for
approximately 4 h. Upon verifying the consumption of the ester via TLC, the reaction
ﬂask was placed in a pyrex bowl (in case of bubbling over) and the reaction quenched by

addition of water added dropwise to the vigorously stirred solution until no fluther

125

 

hydrogen evolution occurred and the reaction mixture became white. The reaction was
then diluted with water and the organic material extracted three times with diethyl ether.
The combined ether layers were washed with 1H hydrochloric acid, followed by water
rinses until the ether layer is ﬁee of particulate (alternatively, the organic material may
be ﬁltered once dried) . The ether layer is then treated with brine, dried over sodium
sulfate, and concentrated via rotary evaporation. The crude product is then distilled via
bulb to bulb distillation (165 - 180 °C air bath temperature at 1.4 mm Hg) to yield a clear

oil 153 (2.9429 g, 13.835 mmol, 92%).

Spectral data for 153 1H NMR(CDC13, 300 MHz) 8 0.94 (s, 6H), 1.68 (s, 1H), 2.37 (s,
3H), 2.86 (s, 2H, Note: broad peak), 3.41 (s, 2H), 7.01 (t, 1H, J = 7.5 Hz) 7.06 (dd, 1H, J
= 7.2, 1.8 Hz), 7.20 (dd, 1H, J = 7.5, 1.8 Hz); ”’C NMR (CDCl;, 75 MHz) 8 21.66,
24.47, 36.69, 39.07, 72.23, 126.71, 127.34, 129.11, 135.82, 135.87, 138.88; IR (thin
ﬁlm) 3599s, 3399vb, 3059s, 2967s, 1591m, 1566s, 1453vs, 1391m, 1109s, 1042vs cm";

mass spectrum, m/z (% rel intensity) 214 M+ (”CL 4), 212 M+(35C1, 14), 181 (10), 140

OH (I)
Cl Cl

153 154

(100), 105 (91), 91 (28);

 

3-(2-chloro-6-methylphenyl)-2,2-dimethylpropanal 154: —NOTE: anhydrous
conditions are not required. In a 100 mL round bottom ﬂask containing a magnetic stir
bar and a nitrogen atmosphere, IBX (6.709 g, 23.95 mmol, 1.494 equivalents) was added

in a single portion to a solution of the alcohol 153 (3.4106 g, 16.033 mmol, 1.0000

126

equivalent) in ethyl acetate (60.0 mL). The ﬂask was ﬁtted with a water condenser and
heated at reﬂux for 4 h as a white suspension. Upon consumption of the starting
material as observed by TLC, the reaction was cooled to rt, the solid material ﬁltered off
and then the solid rinsed three times with ethyl acetate. The ﬁltrate was then washed
with 1N sodium hydroxide, water, and brine; dried over sodium sulfate; and
concentrated via rotary evaporation. The oily crude product was puriﬁed by bulb to bulb
distillation (135-150 °C air bath temperature at 0.08-0.09 mm Hg) to yield a colorless oil
in good yield (3.0177 g, 14.322 mmol, 89%) Spectroscopy note: While it appears to be
pure via other spectral data, the material appears to break down on GC injection (injector
port at 200 oC) into 4 signiﬁcant products with the largest component being the desired

material.

Spectral data for 154 1H NMR(CDC13, 300 MHz) 8 1.10 (s, 6H), 2.30 (s, 3H), 3.06 (s,
2H), 7.00-7.06 (m, 2H), 7.18 - 7.21 (m, 1H), 9.59 (s, 1H); 13C N1VfR(CDCl3, 75 MHz) 8
21.50, 22.11, 36.44, 47.61, 127.44, 127.53, 129.22, 133.96, 135.52, 139.34, 205.25; IR
(thin ﬁlm) 3061w, 2971vs, 2714m l724vs, 1593w, 1568m, 14528, l399w, 11908 11078
cm"; mass spectrum, m/z (% rel intensity) 212 M+ (”0, 0.5), 210 M+ (”CL 3), 192

(87), 177 (67), 139 (100), 115 (80), 103 (100), 91 (72), 77 (93);

 

t-BuOK
' CHZOCH3C| ' I
+
Cl Cl Cl
1 54 1 55 a 1 55 b

Methyl 4-(2-chlore-6-methylphenyl)-2,2-dimethylbut-1-ene ethers 155 a (E) and
155 b (Z)— In a dry 500 mL round bottom ﬂask containing a magnetic stirbar and an

argon atmosphere, powdered potassium tert-butylalkoxide (3.9643 g, 35.329 mmol,

127

2.1187 equivalents) was added in one portion to a suspension of methoxy methyl
triphenylphosphonium chloride (12.0393 g, 35.119 mmol, 2.1061 equivalents) in rt
tetrahydroﬁrran (250 mL). The resulting thick red suspension was allowed to stir for 1 h
then cooled to -78 °C, upon which the red color paled. A solution of the aldehyde 154
(3.5134 g, 16.675 mmol, 1.0000 equivalent) in tetrahydrofuran (15.0 mL) was added
dropwise over 5 minutes. The red color notably lightens soon after the addition. The
dry ice bath was reﬁlled and the reaction allowed to stir overnight as the dry ice was
consumed. Upon veriﬁcation that the room temperature suspension had consumed all of
the aldehyde via TLC (Note: If the reaction is still red tinted, generally the reaction is
complete. If the reaction is ivory, the ylide was consumed, but not necessarily all the
aldehyde as this Wittig salt is hydroscopic and stores poorly. If the reaction is not
complete, preparation of fresh ylide and addition via cannula at -78 °C is possible.) The
reaction was then quenched with water and the organic material was extracted three
times with diethyl ether. The combined ether layers were washed with water, brine and
dried over sodium sulfate. (NOTE: exposure to dilute acid during the workup will result
in hydrolysis of the cis enol ether and formation of the aldehyde 156 described below,
the trans enol ether reacts much slower and survives brief exposure.) After
concentration via rotary evaporation, the crude products were puriﬁed via bulb to bulb
distillation (165-200 °C air bath temperature at 0.06 mm Hg crude distillation, 165-175
°C air bath temperature at 0.06 mm Hg on the second distillation) to yield both enol
ethers in good yield (3.7571 g , 15.737 mmol, 94%) in a roughly 5:4 mixture of cis to

trans. The enols ethers may be further separated from each other for characterization

128

purposes by silica gel chromatography using hexanes as elutant. Yields ranged ﬁ'om

~85% to ~95%, often dependant on the distillation.

Spectral data for 155 a 1H NMR (CDC13, 300 MHz) 8 1.10 (s, 6H), 2.34 (s, 3H), 2.88 (s,
2H, Note: broad peak), 3.45 (s, 3H), 4.82 (d, 1H ,J= 12.9 Hz), 6.05 (d, 1H ,J= 12.9
Hz), 7.00 (t, 1H, J= 7.5 Hz), 7.04 (dd, 1H, J= 7.8, 2.1 Hz) 7.19 (dd, 1H, J= 7.5, 1.8
Hz); ”C NMR(CDC13, 75 MHz) 8 21.97, 28.78, 37.17, 42.28, 55.77, 113.74, 126.65,
127.37, 128.97, 135.82, 136.06, 139.85, 144.52; IR (thin ﬁlm) 3061w, 2959vs, 2907m,
2870m, 2830m, 1669m, 1651s, 1451s, 1385w, 1364m, 1213s, 1123s, 1096s cm"; mass
spectrum, m/z (% rel intensity) 225 M+ -CH3(37C1, 0.8), 223 M+ -CH3(35Cl, 2), 141 (18),

139 (33), 103, (31), 99 (100);

Spectral data for 155 b 1H NMR (CDC13, 300 MHz) 8 1.16 (s, 6H), 2.38 (s, 3H), 3.02 (s,
2H) 3.47 (s, 3H), 4.24 (d, 1H, J= 6.9 Hz), 5.66 (d, 1H, J= 6.9 Hz), 6.98 (t, 1H, J= 6.9
Hz), 7.03 (d, 1H, J: 6.3 Hz) 7.18 (d, 1H, J: 7.5 Hz); ”C NMR(CDC13, 75 MHz) 8
21.93, 29.15, 38.06, 40.85, 59.60, 116.01, 126.42, 127.22, 128.82, 136.22, 136.56,
140.13, 145.23; IR (thin ﬁlm) 3059w, 3023m, 2959s, 2934s, 2870m, 2820w, 1659s,
1453s, 1394m, 1285m, 1103vs cm"; mass spectrum, m/z (% rel intensity) 225 M+ -

CH3(37C1, 0.8), 223 M+ -CH3(35C1, 2), 141 (18), 139 (33), 103, (33), 99 (100);

 

OMe MeO O\
I I HCI
+
Cl Cl Cl
1 55 a 155 b 156

4-(2-chloro—6—methylphenyl)—3,3-dimethylbutanal 156: In a 500 mL round bottom

ﬂask containing a magnetic stirbar and open to the atmosphere, the mixture of enol

129

ethers 155 a and 155 b (2.4677 g, 10.336 mmol, 1.0000 equivalent, ratio not recorded)
was added to a rt solution of tetrahydrofuran (150 mL) and 3N hydrochloric acid (150
mL) and allowed to stir for 1.5 h. Upon veriﬁcation of consumption of the slower
reacting trans isomer via TLC, the solution was extracted three times with diethyl ether.
The combined organic layers were washed with saturated sodium bicarbonate, water,
and brine then dried over sodium sulfate. The solvent was removed by rotary
evaporation and the crude product was puriﬁed using bulb to bulb distillation (125 — 135
°C air bath temperature at 0.04 mm Hg). The aldehyde 156 was obtained as a clear oil in

good yield (2.0953g, 9.3236 mmol, 90 %)

Spectral data for 156 1H NMR (CDC13, 300 MHz) 8 1.13 (s, 6H), 2.36 (5, 3H), 2.42 (d,
2H, J = 3.0 Hz), 2.95 (s, 2H, Note: broad peak), 7.00 — 7.08 (m, 2H), 7.21 (dd, 1H, J =
7.2, 1.8 Hz), 9.83 (t, 1H, J= 2.7 Hz); 13C NMR (CDC13, 75 MHz) 8 21.89, 27.99, 37.22,
40.85, 55.72, 127.12, 127.61, 129.32, 134.99, 136.07, 139.78, 203.24; IR (thin ﬁlm)
3061w, 2963s, 2874s, 2831m, 2737m, 17218, 1592w, 1566w, 14705, 14518, 1389m,
1370m, 1290w, 1169m, llllm cm"; mass spectrum, m/z (% rel intensity) 226 M+ (”Cl,
0.4), 224 M+ (35Cl, 0.9), 180 (33), 150 (31), 142 (73), 141 (76), 140 (97), 139 (100), 105

(87), 103 (82), 85 (91), 57 (90);

0\ t-BuOK \
Cl Cl

1 56 1 57
5-(2-chloro-6-methylphenyl)-4,4-dimethylpent-l-ene 157 — In a dry 100 mL round

 

bottom ﬂask containing a magnetic stirbar and an argon atmosphere, potassium tert-

butylalkoxide (1.0560 g, 9.4109 mmol, 1.4003 equivalents) was added a it suspension

130

of methyl triphenylphosphonium iodide (3.7772 g, 9.3444 mmol, 1.3904 equivalents) in
tetrahydrofuran (50 mL). The resulting thick yellow suspension was then stirred for 1h.
The ylide was then cooled to -78 °C and a solution of the aldehyde 156 (1.5103 g,
6.7205 mmol, 1.0000 equivalent) in tetrahydrofuran (5.0 mL) was added dropwise. The
reaction quickly paled to a light yellow color. The dry ice bath was reﬁlled and the
reaction was allowed to stir overnight, gradually warming to rt as the dry ice was
consumed. Upon verifying the consumption of the starting aldehyde by TLC (similar
visual cues may be taken as in the methoxymethyl triphenylphosphine derived ylide used
for 155 a and 155 b), the reaction was quenched with water and the organic material was
extracted three times with diethyl ether. The combined ether layers were washed with
water, brine and dried over sodium sulfate. After concentration via rotary evaporation,
the crude product was puriﬁed by either bulb to bulb distillation (120-140 °C air bath
temperature at 0.06 mm Hg) or by silica gel column chromatography with hexanes as

elutant to isolate 157 as a clear oil in good yield (1.3472, 6.0480 mmol, 90%).

Spectral data for 157 1H NMR(CDC13, 300 MHz) 8 0.91 (s, 6H), 2.13 (d, 2H, J = 7.8
Hz), 2.34 (s, 3H), 2.82 (s, 2H, Note: broad peak), 4.99 - 5.08 (m, 2H), 5.89 (ddt, 1H, J =
16.9, 10.2, 7.5 Hz), 7.00 (t, 1H, J = 7.5 Hz), 7.05 (dd, 1H, J= 7.5, 1.8 Hz), 7.19 (dd, 1H,
J: 7.5, 1.8 Hz); l3C NMR(CDC13, 75 MHz) 8 21.92, 27.04, 37.48, 40.32, 48.53,
117.22, 126.55, 127.40, 129.02, 135.55, 136.21, 136.26, 139.80; 1R (thin ﬁlm) 3075m,
2963vs, 1640m, 1591w, 1568m, 14705, 1451vs, 1387m, 1365m cm"; mass spectrum,
m/z (% rel intensity) 224 M+ (”0, 0.8), 222 M+ (35Cl, 2.5), 181 (20), 140 (65), 139

(92), 82 (90), 55 (100);

131

 

\ 1) MgCl2 \
K0
c: 2) 157 O
H

3) DMF
157 165

5-(2-formyl-6-methylphenyl)-4,4—dimethylpent-l-ene 165 — Freshly cut potassium
metal (0.2521 g, 6.448 mmol, 4.816 equivalents) was prepared and measured under
octanes then transferred to a dry 25 mL round bottom ﬂask ﬁtted with a condenser and
containing a magnetic stirbar, a nitrogen atmosphere, and tetrahydrofuran (8.00 mL).
Powdered anhydrous magnesium chloride (0.3221 g, 3.383 mmol, 2.527 equivalents)
was then added and the resulting suspension was ﬂushed with nitrogen before being
heated to reﬂux. Within 30 m of heating, the reaction mixture undergoes signiﬁcant
darkening of its white particulate to very ﬁne dark gray color with a faint purplish tint.
Any particulate that accumulates on the walls must be swirled back into the solution.
After 3 h at reﬂux, the chloride 157 (0.2982 g, 1.339 mmol, 1.000 equivalent) is added
dropwise through the condenser to the hot solution and allowed to react at reﬂux over
4.5 h. After 4.5 h have passed, anhydrous dimethylformamide (0.80 mL, 10. mmol, 7.7
equivalents) was added and allowed to stir at reﬂux overnight. The dark solution was
then allowed to cool to rt then quenched by dropwise addition of 1N hydrochloric acid.
Acid was added until all solids are consumed. The now particulate free solution was
then diluted with water and the organic layer extracted three times with ethyl acetate.
The combined organic layers were washed twice with 1N hydrochloric acid, then once
with water and brine before being dried over sodium sulfate. The dry organic layer was

concentrated via rotary evaporation and the crude product puriﬁed on a silica gel column

132

 

using 2% ethyl acetate/hexanes as elutant to obtain 165 as a faintly yellow tinted oil in

low yields (0.0732 g, 0.338 mmol, 25 %).

Spectral data for 165 1H NMR (CDC13, 300 MHz, 25 0C) 8 0.79 (s, 6H), 2.09 (d, 2H, J =
7.5 Hz), 2.39 (s, 3H), 5.00 — 5.10 (m, 2H), 5.87 (ddt, 1H, J= 16.8, 10.2, 7.5 Hz) 7.24 (t,
1H, J= 7.5 Hz), 7.39 (d, 1H, J= 7.2 Hz), 7.73 (d, 1H, J= 7.8 Hz), 10.31 (s, 1H) (two
missing protons); 1H NMR (CDC13, 500 MHz, - 40 °C) 8 0.76 (s, 6H), 2.07 - 2.09 (m,
2H), 2.38 (8, 3H), 2.80 (d, 1H, J= 14.0 Hz), 3.28 (d, 1H, J= 13.5 Hz), 5.03 - 5.09 (m,
2H), 5.82 -— 5.90 (m, 1H), 7.25 (t, 1H, J: 7.5 Hz), 7.41 (d, 1H, J = 7.5 Hz), 7.75 (d,
1H, J = 7.5 Hz), 10.31 (s, 1H); l3C NMR(CDC13, 75 MHz, 25 °C) 8 20.84, 26.39, 36.74,
36.77, 48.34, 117.78, 126.10, 127.45, 134.94, 135.93, 136.11, 139.11, 140.65, 192.56;
IR (thin ﬁlm) 3073m, 3004m, 29635, 28788, 2753w, 2726w, 1690vs, 15908, 1472m,
1385m, 1365w, 12358 cm"; mass spectrum, m/z (% rel intensity) 216 M+ (1.6), 175

(17), 157 (30), 134 (100), 133 (76), 105 (34);

\ 1) MgCl2 \ \
4..
cu 2) 157 ° °
OMe
3) Methyl

157 Chloro— 84 c 167
forrnate

 

5-(2-(methyl formate)-6-methylphenyl)-4,4-dimethylpent-l-ene 84 c: - Freshly cut
potassium metal (0.2247 g, 5.747 mmol, 4.372 equivalents) was prepared and measured
under octanes then transferred to a dry 25 mL round bottom ﬂask ﬁtted with a condenser
and containing a magnetic stirbar, a nitrogen atmosphere, and tetrahydrofuran (8.00 mL).
Powdered anhydrous magnesium chloride (0.2951 g, 3.099 mmol, 2.358 equivalents)

was then added and the resulting suspension was ﬂushed with nitrogen then heated to

133

reﬂux. Within 30 m of heating, the reaction undergoes signiﬁcant darkening of its white
particulate to very ﬁne dark gray color with a purplish tint. Any particulate that
accumulates on the walls must be swirled back into the solution. After 3 h at reﬂux, the
chloride 157 (0.2928 g, 1.314 mmol, 1.000 equivalent) is added dropwise through the
condenser to the hot solution and allowed to react at reﬂux over 4.5 h. After 4.5 h have
passed, methyl chloroforrnate (4.00 mL, 52.0 mmol, 39.6 equivalents) was added and
allowed to stir at reﬂux overnight. The dark solution was allowed to cool to rt and then
quenched by dropwise addition of 1N hydrochloric acid. Acid was added until all solids
are consumed. The now solid free solution was diluted with water and the organic layer
extracted three times with ethyl acetate. The combined organic layers were washed
twice with 1N hydrochloric acid, then once with water and brine before being dried over
sodium sulfate. The dry organic layer was then concentrated via rotary evaporation and
the crude product puriﬁed by silica gel column chromatography using 3% ethyl
acetate/hexanes as elutant to obtain 84 c as a clear oil in low yields (0.0876 g, 0.356
mmol, 27 %) which contains a minor impurity 167 as about 5% of the isolated total

isolated 84 c (166 structure is tentative).

Spectral data for 84 c 1H NMR (CDC13, 300 MHz, 25 °C) 8 0.73 (s, 6H), 2.02 (d, 2H, J =
7.5 Hz), 2.36 (s, 3H), 3.83 (s, 3H), 4.96 - 5.06 (m, 2H), 5.84 (ddt, 1H, J= 16.8, 10.2, 4.5
Hz), 7.11 (t, 1H, J= 7.5 Hz), 7.26 (d, 1H, J= 7.5 Hz), 7.49 (d, 1H, J= 7.8 Hz) (two
protons missing); 1H NMR (CDCl;;, 500 MHz, - 40 °C) 8 .069 (s, 3H), 0.71 (s, 3H), 1.96
— 2.05 (m, 2H), 2.34 (s, 3H), 2.64 (d, 1H, J= 13.5 Hz), 3.48 (d, 1H, J= 13.5 Hz), 3.83
(s, 3H), 5.00 (dd, 1H, J= 17.0, 2.5 Hz), 5.04 (dd, 1H, J= 10.0, 2.5 Hz), 5.79 — 5.88 (m,

1H), 7.12 (t, 1H, J: 7.5 Hz), 7.26 (d, 1H, J: 7.5 Hz), 7.57 (d, 1H, J: 7.5 Hz); ”C

134

NMR (CDC13, 75 MHz, 25 0C) 8 21.50, 25.89, 36.82, 38.09, 48.15, 51.94, 117.21,
125.34, 127.40, 133.42, 133.46, 135.39, 137.84, 139.03, 170.47; [R (thin ﬁlm) 3075m,
3002m, 29638, 29518, 2876m, 2840w, 17258, 1640w, 1591w, 1471m, 1458m, 1435m,
1385w, 1366w, 12798, 12718, 1248m, 1199m ,11218 cm"; mass spectrum, m/z (% rel

intensity) 246 M+ (5), 205 (34), 173 (78), 164 (91), 145 (79), 132 (100), 105 (77);

\ 1) MgCl2 \
K0
cr 2) 157 °

 

CF3
3) triflouro-
157 acetic 84 b
anhydnde

5-(2-triflouroacyl-6—methylphenyI)-4,4-dimethylpent-l-ene 84 b — Freshly cut
potassium metal (0.1668 g, 4.266 mmol, 2.244 equivalents) was prepared and measured
under octanes then transferred to a dry 25 mL round bottom ﬂask ﬁtted with a condenser
and containing a magenetic stirbar, nitrogen atmosphere, and tetrahydrofuran (8.00 mL).
Powdered anhydrous magnesium chloride (0.2127 g, 2.233 mmol, 1.175 equivalents)
was then added, the reaction mixture ﬂushed with nitrogen, and the resulting suspension
heated to reﬂux. Within 30 m of heating, the reaction undergoes signiﬁcant darkening
of its white particulate to a very ﬁne dark gray color with a purplish tint. Any particulate
that accumulates on the walls must be swirled back into the solution. After 3 h at reﬂux,
the chloride 157 (0.4234 g, 1.901 mmol, 1.000 equivalent) is added dropwise through
the condenser to the hot solution and allowed to react at reﬂux over 4.5 h. After 4.5 h
have passed, triﬂouroacetic anhydride (5.00 mL, 36.0 mmol, 18.9 equivalents) was
added and allowed to stir at reﬂux overnight. The dark solution was cooled to rt and then

quenched by dropwise addition of 1N sodium hydroxide. The solution was then poured

135

 

into a seperatory funnel containing 1N hydrochloric acid. Ethyl acetate was added to
achieve a roughly 1:1 water /organic solution, then the organic materials extracted out
three times using ethyl acetate. The combined organic layers were washed with 1N
hydrochloric acid, water, and brine then dried over sodium sulfate. The dry organic
layer was concentrated via rotary evaporation and the crude product puriﬁed on a silica
gel column using hexanes as elutant to obtain 84 b as a faintly yellow tinted oil in low

yields (0.2279 g, 0.8016 mmol, 42%). Yields range from 24% to 42%.

Spectral data for 84 b 1H NMR (CDC13, 300 MHz, 25 °C) 8 0.71 (s, 6H), 2.00 (d, 2H, J =
7.5 Hz), 2.39 (s, 3H), 4.97 - 5.08 (m, 2H), 5.81 (ddt, 1H, J= 16.8, 10.2, 7.5 Hz), 7.22 (t,
1H, J= 7.8 Hz), 7.39 (d, 1H, J= 7.5 Hz), 7.52 (d, 1H, J= 7.8 Hz) (two missing protons);
1H NMR (CDC13, 500 MHz, - 40 OC) 8 0.63 (s, 6H), 1.94 — 2.03 (m, 2H), 2.37 (s, 3H),
2.65 (d, 1H, J: 14.0 Hz), 3.45 (d, 1H, J= 14.0 Hz), 5.01 (d, 1H, J= 16.5 Hz), 5.06 (d,
1H, J= 10.5 Hz), 5.76 — 5.85 (m, 1H), 7.22 (t, 1H, J= 7.5 Hz), 7.39 (d, 1H, J= 7.5 Hz),
7.52 (d, 1H, J= 7.5 Hz); ”C NMR(CDC13, 75 MHz, 25 °C) 8 21.50, 25.83 (broad),
36.77, 36.90, 48.14, 116.17 (q, J= 292.7 Hz), 117.62, 122.02, 125.38, 126.41 - 126.28
(broad peak), 133.10, 134.85, 135.63, 139.38, 140.35, 185.70 (q, J= 33.0 Hz); IR (thin
ﬁlm) 3077m, 3005w, 2967s, 2878m, 1725vs, l640w, 1580m, 1473m, 1458m, 1385m,
1368m, 1314m, 12008, 11728, 11488 cm"; mass spectrum, m/z (% rel intensity) 284 M+

(0.4), 266 (1.1), 243 (29), 202 (75), 133 (82), 103 (50), 83 (100);

136

MeMgBr
O OH

H
165 166

 

5-(2-(ethan-l-ol)-6-methylphenyl)-4,4-dimethylpent-1-ene 166 - To a dry 15 mL
round bottom ﬂask containing a magnetic stirring and a nitrogen atmosphere, a solution
of methyl magnesium bromide (3.0 M in diethyl ether, 0.500 mL, 1.5 mmol, 12
equivalents) was added to a solution of aldehyde 165 (0.0273 g, 0.126 mmol, 1.00
equivalent) in tetrahydrofuran (2.00 mL) and allowed to stir overnight. (The large
excess of methyl magnesium bromide was due to having difﬁculty transferring the
aldehyde to the reaction ﬂask in a measured, anhydrous manner. Thus excess Grignard
reagent was used to overcome the expected water vapor that was likely introduced.
Larger scales will likely not need such an excess. Larger scales also will probably need
cooling before addition.) The solution gradually became milky white as the reaction
proceeded. Upon veriﬁcation that the aldehyde was consumed, the reaction was diluted
by 1N hydrochloric acid added dropwise initially until any signs of a reaction ceased,
then the remainder added all at once. The organic materials were extracted three times
with ethyl acetate and then washed with water and brine. The combined organic layers
were dried over sodium sulfate, concentrated via rotary evaporation, and the product
puriﬁed using silica gel column chromatography with 8% ethyl acetate/hexanes as
elutant. After puriﬁcation, the alcohol 166 was obtained as a clear oil in good yields

(0.0252 g, 0.108 mmol, 86%).

Spectral data for 166 1H NMR(CDC13, 500 MHz, +50 °C) 8 0.86 (s, 3H), 0.89 (s, 3H),

1.41 (d, 3H, J= 5.0 Hz), 1.54 (s, 1H, Note: broad peak), 2.12 (d, 2H, J: 7.5Hz), 2.34 (s,

137

3H), 2.69 — 2.77 (m, 2H), 5.02 - 5.09 (m, 2H), 5.28 (q, 1H, J = 6.0 Hz), 5.89 (ddt, 1H, J
= 17.5, 10.5, 7.0 Hz), 7.08 (d, 1H, J= 7.0 Hz), 7.15 (t, 1H, J= 7.5 Hz), 7.41 (d, 1H, J =
7.5 Hz); ”C NMR (CDCl;, 125 MHz, +50 °C) 8 21.80, 26.87, 27.09, 36.50, 38.01,
48.70, 66.83 (note: broad peak), 117.41, 122.92, 126.39, 129.91, 135.34, 137.86; (2 8p2
carbons not observed) IR (thin ﬁlm) 3368 vb, 3072m, 2969s, 2879m, 1638m, 14728,
13858, 13658, 1291m, 1262m, 1233m, 1169m, 10948, 10658 cm"; mass spectrum, m/z
(% rel intensity) 232 M+ (0.4), 217 (8), 214(5), 188 (6), 173 (23), 157 (13), 143 (13),

133 (45), 132 (95), 131 (80), 117 (49), 91 (70), 55 (100)

\ \
IBX
OH O
166 84 a

5-(2-acyl-6-methylphenyl)-4,4-dimethylpent-l-ene 84 a: In a 15 mL round bottom

 

ﬂask ﬁtted with a condenser and magnetic stirbar, and open to the atmosphere, 2-
iodoxybenzoic acid (0.0574 g, 0.205 mmol, 1.89 equivalents) was added to a solution of
alcohol 166 (0.0252 mg, 0.108 mmol, 1.00 equivalent) in ethyl acetate (3 mL) and was
allowed to stir at reﬂux for 5 h. Upon verifying the consumption of 166 by TLC, the
reaction mixture was cooled to rt and the suspension ﬁltered using a ﬁitted ﬁmnel
(rinsed three times with ethyl acetate). The ﬁltrate was washed twice with 1N sodium
hydroxide then once with water and brine before being dried over sodium sulfate. The
organic layer was concentrated via rotary evaporation and the crude product puriﬁed via
a short silica gel column using 3% ethyl acetate/hexanes as elutant. After puriﬁcation,

the ketone 84 a was obtained as an oil (0.0189 g, 0.0820 mmol, 76%).

138

Spectral data for 84 a 1H NMR(CDC13, 500 MHz, +25 °C) 8 0.73 (s, 6H), 2.00 (d, 2H, J
= 7.0 Hz), 2.35 (s, 3H), 2.56 (s, 3H), 4.97 - 5.05 (m, 2H), 5.82 (ddt, 1H, J= 16.5, 10.0,
8.0 Hz), 7.12 (t, 1H, J: 7.5 Hz), 7.23 (d, 2H, J= 8.0 Hz), 7.33 (d, 2H, J= 8.0 Hz); ‘H
NMR (CDC13, 500 MHz, - 40 °C) 8 0.69 (8, 3H), 0.70 (s, 3H), 1.94 — 2.03 (m, 2H), 2.33
(s, 3H), 2.58 — 2.60 (m, 4H), 3.45 (d, 1H, J= 14.0 Hz), 4.99 (d, 1H, J = 17.0 Hz), 5.03
(dd, 1H, J = 10.0, 2.5 Hz), 5.77 - 5.86 (m, 1H), 7.13 (t, 1H, J = 7.5 Hz), 7.23 (d, 1H, J
= 6.0 Hz), 7.34 (d, 1H, J = 7.5 Hz); 13C NMR (CDC13, 125 MHz, +25 0C) 8 21.44,
26.18, 30.39, 36.89, 37.30, 48.10, 117.28, 125.22, 125.37, 133.12, 135.30, 136.36,
139.58, 142.27, 204.60; [R (thin ﬁhn) 3073w, 3003w, 2964s, 2874m, 169lvs, l639w,
1473m, 1436w, 1383w, 12638 cm"; mass spectrum, m/z (% rel intensity) 230 M+ (0.3),

215(22), 189 (17), 171 (51), 148 (90), 133 (100), 105 (37), 55 (77);

 

hv H
O
OMe MeO 0
84 c 88 c
Tentative

Methyl 5,5,7-trimethyl-3,3a,4,5,6,7a-hexahydro-2aH—cyclobuta[f]indene-Za-
carboxylate 88 c: - In a new borosilicate test tube, ester 84 a (0.0264 g, 0.107 mmol,
1.00 equivalent) was diluted into deuterated acetone (10.0 mL) and the opening closed
with a septum. The solution was cooled to 0 °C then had argon bubbling through the
solution for 30 m to displace dissolved oxygen. The reaction was then allowed to warm
to rt then positioned vertically in a 250 mL beaker with a constant ﬂow of cool water.
The reaction container and cooling bath was irradiated with a 450 W medium pressure

Hg lamp for 7 days. After the irradiation was complete, the solution was transferred to a

139

round bottom ﬂask and concentrated via rotary evaporation. The resulting crude
material was initially puriﬁed of its hexanes insoluble material with a short silica gel
column with 5% ethyl acetate/hexanes as elutant (loaded with minimal dichloromethane)
and the all of the collected material concentrated to yield a semi puriﬁed oil. The
product was then further puriﬁed using a second silica gel column with 2% ethyl
acetate/hexanes as elutant to yield a clear oil in low yields (0.0112 g, 0.0455 mmol,
42%). The oil, while a single spot by TLC, appears to contain multiple materials which
are theorized to be possible regioisomers/diastereomers however this has yet to be

conclusively shown.

Spectral data for 88 c: Due to issues obtaining a pure sample, the puriﬁed but still not

pure 1H NMR spectrum is included on page 62.

\ \
hv
0 . OH

CF3 CF3

84 b 173
3-Methyl-2-(2-methylpent-4-en-2-yl)-1-(triﬂuoromethyl)-1,2-
dihydrocyclobutabenzen-l-ol 173 : - In a new borosilicate test tube,
triﬂouroaetophenone 84 b (0.0557 g, 0.196 mmol, 1.00 equivalent) was diluted into
deuterated acetone (20.0 mL) and the opening closed with a septum. The solution was
cooled to 0 °C then had argon bubbling through the solution for 45 m to displace
dissolved oxygen. The reaction was then allowed to warm to rt then positioned
vertically in a 250 mL beaker with a constant ﬂow of cool water. The reaction container

and cooling bath was irradiated with a 450 W medium pressure Hg lamp for 3 days.

140

After the irradiation was complete, the solution was transferred to a round bottom ﬂask
and concentrated via rotary evaporation. The crude product was puriﬁed using a second
silica gel column with 3.5% ethyl acetate/hexanes as elutant to yield a clear oil in low
yields (0.0201 g, 0.0707 mmol, 36%). The oil, while a single spot by TLC, appears to
contain two materials which thought to be a mixture of diastereomers (crude ratio is 5:4).
The major diastereomer was isolated ﬁom the ﬁrst ﬁaction of the column. Attempts to

isolate the minor diastereomer have been unsuccessﬁrl.

Spectral data for 173 (major diastereomer) lH NMR(CDC13, 500 MHz, +25 °C) 8 1.09
(s, 6H), 2.23 — 2.31 (m, 2H), 2.33 (s, 3H), 2.67 (s, 1H), 3.73 (s, 1H) 5.02 (d, 1H, J= 17
Hz), 5.06 (d, 1H, J= 10.5 Hz) 5.87 (ddt, 1H, J= 17, Hz, J= 9.5 Hz, J: 8.0 Hz) 7.07 (d,

1H, J= 7.0 Hz), 7.17 (d, 1H, J= 7.0 Hz), 7.20 (t, 1H, J= 8.0 Hz);
Due to the loss of the minor isomer the crude is included in the appendix on page 157.

The following IR and MS data was collected from a roughly 1:3 mixture of minor/major
diastereomers. IR (thin ﬁlm) 3478sb, 3074m, 29618, 29288, 28748, 1640m, 1614w,
1478m, 1464m, 1416w, 1389m, 1368m, 1316m, 1260m, 11598 cm"; mass spectrum, m/z
(% rel intensity) 284 M+ (0.4), 269 (0.9), 266 (0.9), 243 (23), 225 (15), 205 (17), 195

(30), 175 (19), 156 (17), 129 (25), 128 (24), 115 (31), 83 (78), 55 (100);

0 TMSCI OTMS
<5 TEA

Trimethylsilyl cyclopentenol ether: - In a dry 100 mL schlenk tube containing a

 

magnetic stirbar and a nitrogen atmosphere, trimethylsilyl chloride (7.60 mL, 59.5

mmol, 1.10 equivalent) was added to a solution of cyclopentanone (4.80 mL, 54.3 mmol,

141

1.00 equivalents) and triethylarnine ( 10.0 mL, 71.7 mmol, 1.32 equivalents) in
dimethylformamide (20.0 mL). The resulting cloudy suspension was then closed and
heated in a 90 °C oil bath for 48 h. During this time, the reaction will become a dark red
in color while a solid layer may form on the surface. This layer should be broken up and
mixed back into the solution. After 48 h, the reaction is allowed to cool to rt, then
poured into a separatory funnel and diluted with diethyl ether. Water is then added and
the organic materials rapidly extracted three times with diethyl ether. The combined
ether layers were then washed twice with l M lithium bromide, then once with saturated
ammonium chloride and brine before being dried over sodium sulfate. Ammonium
chloride must be used as the acid source for amine removal otherwise the product
decomposes rapidly. The dried solution is then concentrated via rotary evaporation. 1H
NMR spectroscopy of the crude material suggests little to no decomposition during
workup (often, no cyclopentanone is observed) but the extractions are not efﬁcient
according to the crude yield. The crude product is then distilled at atmospheric pressure
using a packed column distillation with the bulk of the product distilling at 150 - 155 °C
although earlier fractions that start coming over at 140 °C often are pure as well.

Distillation yielded the desired enol ether as a clear oil (4.1117 g, 26.306 mmol, 48%).

Spectral data for Trimethylsilyl cyclopentenol ether: 1H NMR (CDC13, 300 MHz) 8
0.183 (s, 9H), 1.829 (pentet, 2H, J= 7.2 Hz), 2.20 — 2.27 (m, 4H), 4.60 (s, 1); ”C

(CDCl3, 75 MHz) 8 0.02, 21.28, 28.72, 33.48, 102.11, 154.97;

142

 

0H MeZSiCIZ 0. .0:
Br Br

TEA

184D 183b

p-Bromophenoxydimethylsilyl Chloride 187 b: In a dry 500 mL round bottom ﬂask
equipped with a condenser and containing a stirbar and a nitrogen atmosphere, 4-
bromophenol (11.6772 g, 67.490 mmol, 1.0000 equivalent) to a rt solution of
dichlorodimethylsilane (48.0 mL, 399 mmol, 5.91 equivalents) and triethylamine (9.40
mL, 67.5 mmol, 1.00 equivalents) in dichloromethane (200 mL). The resulting cloudy
solution was then heated at reﬂux for 3 d. The condenser was then replaced with a short
path distillation head and the solvent/excess dichlorodimethylsilane distilled away to
leave mixture of solid and oil. Hexanes was then added to the remaining mixture and the
solid broken into free ﬂowing particles before ﬁltering the solid rapidly under a blanket
of nitrogen gas. The solid was rinsed once with hexanes. The ﬁltrate was then put into
the original 500 mL round bottom ﬂask and the solvent again removed by distillation,
resulting in the crude product as an oil. The crude product was then transferred to a 50
mL round bottom ﬂask and puriﬁed via bulb to bulb distillation (85 — 120 °C air bath
temperature at 0.15 mm Hg) to yield the mono chloride 183 b as a clear oil in good

yields (12.9109g, 48.608 mmol, 72%).

Spectral data for 183 b 1H NMR(CDC13, 300 MHz) 8 0.058 (s, 6H), 6.82 — 6.85 (m,
2H), 7.34 -737 (m, 2H) ”C NMR (CDC13, 75 MHz) 8 2.29, 115.28, 121.82, 132.52,
152.44; mass spectrum, m/z (% rel intensity) 252 M+ -Me (37C181Br, 3), 250 M+ -Me
(37C179Br and 35C18‘13r, 45), 248 M+ -Me (35C17913r, 39), 233 (8), 153 (37), 91 (15), 79

(24), 77 (100);

143

APPENDIX

144

 

‘_.
.‘

 

 

 

 

 

 

 

Spectrum 4: 1,4 diacylbenzene and 2,3 dihydrofuran prior to
irradiation in deuterated acetonitrile (300 MHz 1H NMR)

145

 

 

 

 

 

 

 

 

 

Spectrum 5: 1,4 diacylbenzene and 2,3 dihydrofuran after 12 h
of irradiation in deuterated acetonitrile (300 MHz 'H NMR)

146

 

 

 

Jim JLm-_J

 

 

 

 

 

Spectrum 6: 1,4 diacylbenzene and 2,3 dihydrofuran after 12 h
of irradiation in deuterated chloroform (300 MHz IH NMR)

147

T

 

 

 

 

 

 

 

 

25
j.

1

 

Spectrum 7: p thiomethoxyacetophenone and 2,3 dihydrofuran
prior to irradiation in deuterated acetonitrile (300 MHz 1H NMR)

 

 

148

9.0

 

 

 

 

 

 

Spectrum 8: p-thiomethyoxyacetophenone and 2,3 dihydroﬁrran
after 12 h irradiation in deuterated acetonitrile (300 MHz 1H NMR)

149

 

 

 

 

 

 

 

 

Spectrum 9: p-thiomethyoxyacetophenone and 2,3 dihydrofuran
after 12 h irradiation in deuterated chloroform (300 MHz 1H NMR)

150

 

 

1.0

T

l

11TI7
2.0

l

 

 

 

 

 

 

 

 

 

Spectrum 10: 01,01,111 triﬂouroacetophenone and 2,3 dihydrofuran
before irradiation in deuterated acetonitrile (300 MHz H NMR)

151

4L

 

l I

 

 

 

 

 

 

 

 

 

Spectrum 11: 01,11,111 triﬂouroacetophenone and 2,3 dihydrofuran
after 12 h irradiation in deuterated acetonitrile (300 MHz 1H NMR)

152

 

 

 

 

 

 

 

 

 

Spectrum 12: 11,01,111 triﬂouroacetophenone and 2,3 dihydroﬁrran
after 12 h irradiation in deuterated chloroform (300 MHz 1H NMR)

 

 

153

a .22 1WWM
17 IITTlIllliI l

 

 

 

 

 

 

 

 

 

 

 

i w WMJWL;

/
I
0 on
LI.
"'0 m
[x
‘-
>
..C
/ O at)
‘5 .0

Spectrum 13: Crude product from the irradiation of 84 b in acetone
to generate 173. Product ratio taken from the cyclobutane H to be
roughly 4:5. (chloroform solvent, 300 MHz)

 

154

3: 6:35

3?— 3:33

3: 6:93 b

} 3.54

r 2.18

} 6.39 _

 

REFERENCES

1.) Wagner, P. J .; Nahm, K. J. Am. Chem. Soc. 1987, 109, 4404-4405.
2.) Wagner, P. J .; Nahm, K. J. Am. Chem. Soc. 1987, 109, 6528-6530.

3.) Nicholas J. Turro, Modern Molecular Photochemistry, Benjamin/Cummings
Publishing Co, 1978

4.) Wagner, P.J.; Sakamoto, M. .J. Am. Chem. Soc. 1989, 111, 8723-8725.

5.) Wagner, P. J .; Sakamoto, M.; Madkour, A. E. Am. Chem. Soc. 1992, 114, 7298-7299.
6.) Smart, R. P.; Wagner, P. J. Tetrahedron Lett. 1995, 36, 5131-5134.

7.) Wagner, P. J .; Cheng, K. Tetrahedron Lett. 1993, 34, 907-910.

8.) Wagner, P. J.; McMahon, K. Am. Chem. Soc. 1994, 116, 10827-10828.

9.) Chapman, 0. L.; Borden, G. W.; King R. W.; Winkler B. J. Am. Chem. Soc. 1964, 86,
2660-2663.

10.) Wagner, P. J .; Alehashem, H. Tetrahedron Lett. 1993, 34, 911-914.
11.) Wagner, P. J .; Smart, R. P. Tetrahedron Lett. 1995, 36, 5135-5138.

12.) Amey, D. M.; Gilbert, A.; Jones, D. T. J. Chem. Soc., Perkin Trans. 2 1998, 213-
218.

13.) Al-Jalal, N.; Drew, M. G. 3.; Gilbert, A. J. Chem. Soc., Chem. Commun. 1985, 85-
86.

14.) Gilbert, A.; Heath, P. Tetrahedron Lett. 1987, 28, 5909-5912.

15.) Al-jalal, N.; Gilbert, A.; Heath, P. Tetrahedron 1988, 44, 1449-1459.

16.) Gilbert, A.; Yianni, P. Tetrahedron Lett 1982, 23, 255-256.

17.) Ayer, W. A. Saeedi-Ghomi, M. H. Tetrahedron Lett. 1981, 22, 2071-2074.
18.) Ayer, W. A. Saeedi-Ghomi, M. H. Can. J. Chem. 1981, 59, 2536-2538.

19.) Strunz, G. M.; Bethell, R.; Dumas, M. T.; Boyonoski, N. Can. J. Chem. 1997, 75,
742-753.

155

 

20.) Ayer, W. A.; Nakashima, T. T.; Saeedi-Ghomi, M. H. Can. J. Chem. 1984, 29, 531-
533.

21.) Abell, C.; Leech, A. P. Tetrahedron Lett. 1987, 28, 4887-4890.

22.) Murata, Y; Ohtsuka, T.; Shirahama, H.; Matsumoto, T. Tetrahedron Lett. 1981, 22,
43 13-43 14.

23.) Harmata, M.; Bohnert, G. J. Org. Lett. 2003, 5, 59-61.
24.) Gibbs, R. A.; Okamura, W. H. J. Am. Chem. Soc. 1988, 110, 4062-4063.

25.) Gibbs, R. A.; Bartels, K.; Lee, R. W. K.; Okamura, W. H. J. Am. Chem. Soc. 1989,
111, 3717-3725.

26.) Krause, N. Liebigs Ann. Chem. 1993, 521-525.
27.) Moens, L.; Baizer, M. M.; Little, D. J. Org. Chem. 1986, 51, 4497-4498.
28.) Zhao, S.-K.; Helquist, P. J. Org. Chem. 1990, 55, 5820-5821.

29.) Birkenes, O. J .; Hansen, T. V.; M’dachi, S.; Skattebel, L.; Stenstrom, Y. Acta. Chem.
Scand. 1988, 52, 806-812.

30.) Ishii, S.; Zhao, S.; Mehta, G.; Knors, C. J .; Helquist, P. J. Org. Chem. 2001, 66,
3449-3458.

31.) Mehta, G.; Sreenivas, K. Tetrahedron Lett. 2002, 43, 703-706.

32.) Weisner, K. Tetrahedron 1975, 31, 1655.

33.) Mills. R. J .; Taylor, N. J .; Snieckus, V. Org. Chem. 1989, 54, 4372-4385.
34.) Park, K.; Yuan K.; Scott, W. J. J Org. Chem. 1993, 58, 4866-4870.

35.) Watson, A. T.; Park, K.; Wiemer, D. F.; Scott, w. J. J Org. Chem. 1995, 60, 5102-
5106.

36.) Purohit, V. G.; Subramanian, R. Chemistry and Industry. 1978, I6, 731- 732.

37.) Taylor, S. K.; Hockerman, G. H.; Karrick, G. L.; Lyle, S. B.; Schramm, S. B. J.
Org. Chem. 1983, 48, 2449-2452.

38.) Taylor, S. K.; Davisson, M. E.; Hissom, B. R.; Brown, S. L.; Pristach, H. A.;
Schramm, S. B.; Harvey, S. M. J. Org. Chem. 1987, 52, 425-429.

156

39.) Travis, B. R.; Narayan, R. S.; Borhan, B. J. Am. Chem. Soc. 2002, 124, 3824-3825.
40.) Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron 1985, 41, 3901-3924.
41.) Ghera, E.; Plemenitas, A.; Ben-David, Y. Synthesis 1984, 504 — 506.

42.) Gruter, G-J. M.; Akkerman, O. S.; Bickelhaupt, F. J. Org. Chem. 1994, 59, 4473-
4481.

43.) Wulff W. D.; Peterson, G. A.; Bauta, W. E.; Chan, K.-S.; Faron, K. L.; Gilbertson, S.
R.; Kaesler, R. W.; Yang, D. C.; Murray, C. K. J. Org. Chem. 1986, 51, 277-279.

44.) Ireland, R.E.; Walba, D.M. Org. Synth., Collect. Vol. VI, 1988, 567

45.) Comins, D. L.; Brown, J. D. J. Org. Chem. 1984, 49, 1078-1083.

46.) Fu reference

47.) Reike, R. D.; Bales, S. E. J. Am. Chem. Soc. 1974, 96, 1775-1781.

48.) Frigerio, M.; Santagostino, M.; Sputore, S. J. Org. Chem. 1999, 64, 4537-4538.
49.) Ward, S. C.; Fleming, S. A. J. Org. Chem. 1994, 59, 6476-6479.

50.) Bradford, C. L.; Fleming, S. A; Ward, S. C. Tetrahedron Lett. 1995, 36, 4189-4192.
51.) Shea, K. J.; Gautheir, D. R. Tetrahedron Lett. 1994, 35, 7311-7314.

52.) Crimmins, M. T.; Guise, L. E. Tetrahedron Lett. 1994, 35, 1657-1660.

53.) Blakemore, D. C.; Gilbert, A. Tetrahedron Lett. 1995, 36, 2307-2310.

54.) House, H. 0.; Gall, M.; Olmstead, H. D. J. Org. Chem. 1971, 36, 2361-2371.
55.) Corey, E. J .; Hopkins P. B. Tetrahedron Lett. 1982, 23, 4871-4874.

56.) Comprehensive Organic Synthesis Trost, B.M.; Fleming, 1.; Eds. Pergamon:
Oxford1991; Vol 6, pp 631.

157

 

   

M1111111111111[111111111111111115s

479