close

Вход

Забыли?

вход по аккаунту

?

Expanding the Scope and Utility of the Scientific Microwave Apparatus in Organic Synthesis: Reactive Gases, Scale-Up and New Methodology Development

код для вставкиСкачать
Expanding the Scope and Utility of the Scientific Microwave Apparatus in Organic Synthesis:
Reactive Gases, Scale-Up and New Methodology Development
Chad Michael Kormos, Ph. D.
University of Connecticut, 2010
Microwave heating has seen increasing use in synthetic organic chemistry ever since the first
published results using domestic microwave ovens in 1986. Commercially available scientific
microwave reactors offer increased reproducibility and safety for chemical reactions. This
dissertation covers a variety of methods developed using state-of-the-art scientific microwave
reactors. Phenols are prepared from aryl halides in high-temperature to near-critical water. In
the sealed vessel of the microwave reactor, reactive gases can be introduced in a facile manner.
Carbon monoxide is incorporated into aryl carboxylic acids or esters via a palladium-catalyzed
transformation of iodoarenes in water or simple alcohol solvents. Ethylene can be exploited to
produce unsymmetrically-substituted stilbenes via a two-step, one-pot palladium-catalyzed
Heck coupling with two iodo- or bromoarenes. In combination with an electrodeless discharge
lamp activated by the microwave field, oxygen is used as the terminal oxidant in the
transformation of 1,4-dihydropyridines into pyridines. Finally, the scalability of microwaveheated methods is tested, including the use of a prototype batch reactor capable of heating up
to 12 L reaction volume.
Expanding the Scope and Utility of the Scientific Microwave Apparatus in Organic Synthesis:
Reactive Gases, Scale-Up and New Methodology Development
Chad Michael Kormos
B. S., Worcester Polytechnic Institute, 2004
A Dissertation
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Doctor of Philosophy
at the
University of Connecticut
2010
UMI Number: 3447456
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMT
Dissertation Publishing
UMI 3447456
Copyright 2011 by ProQuest LLC.
All rights reserved. This edition of the work is protected against
unauthorized copying under Title 17, United States Code.
uest
ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346
APPROVAL PAGE
Doctor of Philosophy Dissertation
Expanding the Scope and Utility of the Scientific Microwave Apparatus in Organic Synthesis:
Reactive Gases, Scale-Up and New Methodology Development
Presented by
Chad Michael Kormos, B. S.
Major Advisor:
Associate Advisor
Associate Advisor
Associate Advisor
Associate Advisor
University of Connecticut
2010
Acknowledgements
First off, I'd like to apologize to everyone reading this. I recorded it on a cell phone after
driving almost ten hours, so I'm a bit stir crazy; you have to strike when inspiration hits.
I'd like to thank my Dad, a true and humble American hero, who I, like many, owe my
life. After all, he brought me into this world, and he never saw fit to take me back out. I must
also credit him for my problem-solving (read: jury-rigging) attitude; maybe I should have been
an engineer...
Mom, thanks for my sense of empathy, matched (in a scientist) only by Dr. Gary Jensen,
the big softy. If not for you, I'd not be nearly as well-rounded a human being. Sometimes I am
afraid I might be as crazy as I think you think I think you are.
To my brothers: without you guys, I don't think I'd have ever learned what it means to
be competitive. Remember that I've always had a head start of at least four years on you.
Between the three of us, we should be able to make an impact on the world, or at least leave a
new generation of Kormoses to it.
Jason: sometimes we think so alike, it really freaks me out. If not for the hundred or so
pounds, we could have been twins! I had a head start on you, too, by the way. I'm glad we had
the opportunity. This was a triumph. I'm making a note here: huge success.
I'd like to recognize Dr. Kirk Marat at the University of Manitoba who has produced a
beautiful piece of software in SpinWorks, available for free download and academic use, which
has saved me countless billed hours on the NMR over the years. If you haven't used it, go now!
I certainly have to thank the individuals and groups with whom I've been able to publish.
In order of increasing frequency of publication, then alphabetically by last name, they are:
ii
With one shared publication: Dr. Riina K. Arvela; Dr. Justin D. Fair; Jennifer L. Holcomb,
a former roommate; Rachel M. Hull; Mauro lannelli, Fabio Bergamelli and Stefano Paravisi
(collectively, "the Italians"), who took my carbonylation method to a whole new scale; Dr.s
Daniel P. Kennedy and Shawn C. Burdette, who I still believe gave me too much credit for a
simple hydrogenation; Tamera L. Mack; Dr. Cynthia M. McGowan; Lauren M. Stencel; Victoria A.
Williams.
With two shared publications:
Keri B. Avery, a very (ha ha, get it?) productive
(charmed?) REU student; Dr. Matthew D. Bowman, the one and only; William G. Devine, who
among his many notable credits can claim introducing me to the art and science of homebrewing; Dr. Jason R. Schmink, also a former roommate.
And finally, with twelve shared publications:
Dr. Nicholas E. Leadbeater. Unlike the
fleeting myriad of undergraduates, your presence persisted through it all.
And there's Tzip. Tzipporah, Tzippi, Skittles. The poor girl puts up with a lot. Seriously. I
mean, I nicknamed her after a candy so I could voice-dial her on my cell.
Look at me still talking when there's Science to do.
I've experiments to run. There is research to be done.
"Still Alive" from Portal
in
Table of Contents
Preface
1
List of Abbreviations
6
Introduction
8
Microwave radiation
Microwave-heating of organic reactions: the seminal papers
Modern microwave reactors
Advantages of using a microwave reactor
Reactions with gaseous reagents
Gas-loading capabilities of modern commercial microwave reactors
Published reports of microwave-heated reactions with gases
Increasing the scale of microwave-heated reactions
Phenols from iodo- and bromoarenes
Established methods for the preparation of phenols from aryl halides
Final optimization and substrate screening
Summary
Carbonylation of aryl iodides
Mechanism of palladium-catalyzed carbonylation
Carbonylation in water: hydroxycarbonylation
Carbonylation in alcohol: alkoxycarbonylation at high pressure
Carbonylation in alcohol alkoxycarbonlation at low Pco
Carbonylation in alcohol: large scale
Aminocarbonylative carbonylation
Summary
Oxidation of 1,4-dihydropyridines with molecular oxygen
Synthesis of 1,4-dihydropyridines
Oxidation of 1,4-dihydropyridines to pyridines
Photoexcitation by electrodeless discharge lamps (EDLs)
Lighting the EDL under microwave irradiation
Substrate screening with optimized oxidation conditions
Control experiments
Dimedone-derived 1,4-dihydropyridines
Summary
Heck reaction of ethylene: one-pot two-step synthesis of stilbenes
Aqueous conditions for the Heck reaction
Selective Heck reaction with ethylene
Extension to a two-step one-pot process
9
9
11
12
14
15
16
20
22
28
29
31
32
35
36
40
43
48
51
54
55
56
57
58
59
60
62
62
63
64
66
66
68
IV
Heck reaction of ethylene: one-pot two-step synthesis of stilbenes (continue
Reaction scale-up
Reactions with other low-boiling olefins
Summary
Reaction scale-up with Accelbeam prototype
Reaction development
Solvent heating
Suzuki reaction: preparation of 4-methoxybiphenyl
Heck reaction: preparation of 4'-methoxycinnamic acid
Thiouracil condensation and S-benzylation
Summary
Closing Remarks
Future work
Appendix I: Dielectric heating
Electrostatics: capacitors in equilibrium
Direct current: capacitors in a DC circuit
Alternating current: capacitors in an AC circuit
Microwaves and dielectric heating
Microwave heating and the bench-top chemist
Appendix II: Sealved-vessel reactions and kinetics
Heating solvent in a sealed vessel
Heating reactions in a sealed vessel
Appendix III: Tamoxifen
Williamson etherification
Sonogashira reaction with 1-butyne
Trans-bromination of an alkyne
Double Suzuki coupling with a vinyl dibromide
Double Suzuki coupling... the other way?
Appendix IV: Macrolactone support
Appendix V: Spectral Data
Phenols from iodo- and bromoarenes
Carbonylation of aryl iodides
Oxidation of 1,4-dihydropyridines with molecular oxygen
Heck reaction of ethylene: one-pot two-step synthesis of stilbenes
Reaction scale-up with the Accelbeam prototype
Appendix III: Tamoxifen
Appendix IV: Macrolactone support
General Experimental
Preface
To answer the obvious question first, the question that I will no doubt be hearing until
the day I retire, "No. I have not seen any evidence to suggest that microwave-heating provides
any sort of special kinetic boost to a reaction." I have expended months over the years, first
actually looking for the elusive "microwave effect" and then just reading published articles in
which the authors claim to have finally discovered it. Without fail, each "Eureka!" fails under
scrutiny of the experimental design, most often because the temperature at point A is
measured, while the reaction is occurring at point B. The original reports of a 'microwave effect'
frequently compared "apples to oranges" (to use the colloquial cliche), in that the reaction at
one temperature heated conventionally is compared to the reaction at a much higher
temperature in the microwave.
To answer the next most obvious question, "No, in the great majority of cases, the
chemistry reported herein was not repeated in the so-called conventional manner." Why? In
most cases it boils down ("concentrates," to the chemists) to safety. Most of these reactions are
conducted well above the atmospheric boiling point of their solvents, where a fair autogenic
pressure develops. Although there are pieces available that would allow it, conventional
glassware is not designed to operate under pressure. Even with the specialty labware, additional
protection is required. Then there is the added inconvenience of trying to reach the elevated
temperatures commonly accessed with microwave heating. I once walked in on a visiting
scientist in my lab attempting to measure a melting point in the vicinity of 250 °C in an oil bath.
There are certain ways things should not be done, particularly when there is an easier way to do
1
them. Since the scientific microwave reactor has become common place, trickling down as far as
the undergraduate laboratory experience,1 microwave heating has become that easier way.
As emphasized a number of times through the dissertation, the microwave is not the
best way for every process, nor is it entirely safe. Every chemist has those close call experiences,
emerging from them wiser and better prepared to avoid them in the future. When a glass tube
with 10-mL capacity undergoes an unanticipated pressure release at 200 psi, it will give you a
start, but when an 80-mL heavy-walled quartz tube capable of operating at 1160 psi, having
been filled with a flammable gas, undergoes a similar unanticipated pressure release... It is a
testament to the safety engineering of the manufacturers that I have never seen anyone
seriously injured in the lab.
One final thing I'd like to point out is my careful selection and use of the phrase
"microwave-heated" rather than "microwave-assisted,"2 "microwave-activated" or "microwaveinduced." I am uncomfortable with the term "assisted" because, like "activated," it seems to
suggest that the microwave (radiation) influences the course of a reaction directly, though it is
true that the microwave (reactor) "assists" the chemist in executing a particular set of
conditions. In order to avoid confusion, I also try to specify "microwave radiation," "microwave
irradiation" or "microwave reactor," whenever it is not completely clear to which I am referring.
The general format of this dissertation chronicles the development and substrate-scope
of a variety of transformations, with emphasis on those transformations that formed a coherent
"project" and were published. First, the preparation of phenols from aryl halides in hightemperature water is described from the inception to the final substrate screening. This is an
1
Leadbeater, N. E.; McGowan, C. M. Clean, Fast Organic Chemistry. CEM Publishing, 2006.
This appears to be the generally preferred term. For example, see: Tierney, J. P.; Lidstrom, P. Microwave
Assisted Organic Synthesis. Blackwell Publishing: Oxford, 2005.
2
2
interesting story because the transformation eventually pursued was initially a curious sideproduct in the preparation of a different product.
Next, I describe our foray into the realm of using reactive gases in microwave reactors.
Beginning with carbon monoxide, a few palladium-catalyzed carbonylation reactions are
examined, culminating in the development of the alkoxycarbonylation which utilizes a nearstoichiometric quantity of carbon monoxide. I like this particular transformation because at the
end of the reaction, virtually no carbon monoxide remains, so there is little to no hazard
involved in releasing the excess pressure. We had played with the idea of using
13
C-carbon
monoxide in this methodology, to limit the waste of the expensive isotopically-labeled reagent,
but ultimately this was never done, primarily because the cost was still prohibitive. I have no
doubt that the process would work efficiently.
For Dr. Neth, we have an interesting project that combined photochemistry with
microwave-heating. I see this project, the oxidation of 1,4-dihydropyridines (DHPs) using oxygen
gas as the terminal oxidant, as a proof-of-concept study, rather than anything particularly
synthetically useful. For one thing, the substrate scope was found to be narrow; some DHPs did
not oxidize at all under the conditions. Secondly, depending on substrate, two products were
formed.
The final submission into the collection of microwave-heated reactions of reactive gases
is something of which I am quite proud: the selective Heck reaction of ethylene. I rapidly
determined the challenge would be the "selective" part. To produce a symmetric stilbene
product from ethylene and two equivalents of an aryl halide was simple; pretty much any
reaction in water did the trick. Selectively reacting only one equivalent of the aryl halide was
considerably more difficult. Unfortunately, the styrene products (once I did succeed in
3
producing them) were not easily isolated, so 1 developed a two-step process whereby the
styrene was produced from ethylene, then reacted with a second aryl halide to produce an
unsymmetrically-substituted stilbene. For this project, I actually had a chance to use the HPLC
that I installed. It allowed me to quickly scan through a very large range of conditions to find
conditions to effect the Heck coupling of a styrene with an aryl halide in stoichiometric
combination; this is rarely done, as the olefin is almost always used in excess.
The final (regular) chapter of chemistry presents an overview of my part in a scale-up
project undertaken with Jason, assisted by some small-scale developmental work by Will
Devine, using a large-scale prototype microwave reactor. With a working reactor volume of 2-12
L, Jason and I were able to prepare significant quantities (kilograms in some cases) of a variety of
compounds.
This collection, however, is by no means comprehensive of the chemistry attempted. In
Appendix III, I present a thought-project in which Jason and I attempted to produce Tamoxifen,
the "gold standard" breast cancer medicine. Ultimately, the project didn't pan out, but Jason
and I were able to combine several of our own specialties into a multi-step sequence that
almost got us to the desired product.
4
So much chemistry has been left out completely: from things that worked, but had such
a narrow substrate scope to render them virtually useless, to things that out-right never worked
at all. Among the former, I looked at some carbonylative cyclizations including the PausonKhand synthesis of cyclopentenones, which actually lead to an investigation of another
cyclization of a diyne that produced a rather interesting bicyclo[3.3.0]octadienone, as shown
below:
'O
Co2(CO)8
Pd
Among the latter (reactions that never worked at all), reactions with nitrous oxide,
which proved to be extremely inert. Thermodynamically, it should be very reactive, but the
kinetics are unfavorable. This is well documented.3
I hope the survey that I present here is interesting and convinces you of a somewhat
cohesive doctoral research program!
3
For a review of some of the high-pressure, high-temperature processes that use nitrous oxide, see:
Parmon, V. N.; Panov, G. i.; Uriarte, A.; Noskov, A. S. Catalysis Today 2005,100,115-131.
5
List of Abbreviations
AC
alternating current
Ac
acetyl
atm
atmosphere
Ar
aryl
B:
base
Bn
benzyl
Bu
butyl
°C
degrees Celsius
cal
calorie
d
doublet
DABO
dihydro-alkoxy-benzyl-oxopyrimidine
dba
dibenzylideneacetone
DBU
l,8-diazabicyclo[5.4.0]undec-7-ene
dec.
decomposition
DC
direct current
DCB
dichlorobenzene
DHP
1,4-dihydropyridine
DMA
N,N-dimethylacetamide
DME
1,2-dimethoxyethane
DMF
N,N-dimethylformamide
DMSO
dimethylsulfoxide
(E)
EDL
Entogen
electrodeless discharge lamp
Et
ethyl
eq.
molar equivalent
ES-
electrospray ionization negative ion detection
ES+
electrospray ionization positive ion detection
F
Faraday
°F
Fahrenheit
FID
flame ionization detection
g
G
gram
GC
gigagas chromatography
HIV
human immunodeficiency virus
Hz
Hertz, cycles per second
i
iso
IR
infrared
J
joule
k
kilo-
K
Kelvin
KOtBu
potassium tert-butoxide
L
ligand, liter
In
natural logarithm
M
mega-, moles per liter
min.
minutes
m
meter, milli-, multiplet, meta-substituted
mol (%)
mole (percent)
m. p.
melting point
m/z
mass per charge ratio
n
nano-, normal
N
Normal
NMR
nuclear magnetic resonance
NNRTI
non-nucleoside reverse transcriptase inhibitor
OD
outer diameter
o-
ortho-substituted
p-
para-substituted
Pco
partial pressure of carbon monoxide
PEEK
polyether ether ketone thermoplastic
Ph
phenyl
ppm
parts per million
Pr
propyl
psi
pounds per square inch
PTFE
polytetrafluoroethylene (Teflon)
quant.
quantitative
R
generic hydrocarbon
s
seconds, singlet
S-DABO
dihydro-alkylthio-benzyl-oxopyrimidine
SNAr
nucleophilic aromatic substitution
t
tertiary, triplet
tan
tangent
TBAB
tetrabutylammonium bromide
THF
tetrahydrofuran
TOF
time of flight
tol
tolyl
UV
ultraviolet
W
Watt
X
halogen
(Z)
Zussamen
Introduction
Perusal of the chemical literature over the decades reveals a waxing and waning of
research "buzz words," betraying science as equally trendy as pop culture, albeit more esoteric.
When a "buzz word" graduates from the article title and abstract, only to be buried in the
experimental details, it can no longer be dismissed as trendy, and rather must be recognized as
an accepted member of the chemical lexicon. This has been the case for "microwave heating."
The use of microwave heating is becoming commonplace and has been widely
reviewed.4 A large variety of synthetic organic transformations have been successfully
accomplished under microwave heating;5 the breadth of variety is so great, in fact, it cannot
possibly be summarized in a concise manner: any reaction amenable to elevated temperature
will likely benefit from microwave heating. Reactions proceed more rapidly at higher
temperatures and the microwave reactor allows quick and safe access to those temperatures. A
recent review emphasizes how microwave heating can be used not only for general synthetic
transformations, but also for radical, organocatalytic, cycloaddition, metathesis, combinatorial
and solid-phase reactions (to name just a few).6
Even though microwave-heating has permeated nearly every facet of synthetic
chemistry, the subject still deserves an introduction from the basics and the beginning, a
chronicling of microwave radiation and irradiation as applied to the dielectric heating of
chemical reactions.
4
For an introduction to the basic principles and applications of microwave heating for organic sythesis,
see: Kappe, C. O. Angew. Chem. Int. Ed. 2004,43, 6250-6284. For an updated overview, see: Kappe, C. O.
Chem. Soc. Rev. 2008,37,1127-1139.
5
For a tabulated sample of some of the "older" microwave-heated reactions, see: Lidstrom, P.; Tierney,
J.; Wathey, B.; Westman, J. Tetrahedron 2001,57, 9225-9283.
6
Caddick, S.; Fitzmaurice, R. Tetrahedron 2009, 65, 3325-3355.
8
Microwave
radiation
Microwave radiation ranges in frequency from 300 MHz to 300 GHz, but the entire
range is not available for heating purposes. The electromagnetic spectrum is, after all, a valuable
commodity; various wireless communication technologies operate in this same region. In order
to minimize interference, particularly with commercial and military applications, microwave
heating is restricted by law to certain frequencies; the most commonly used being 2.45 GHz.
With a wavelength of 12.2 cm, these microwave photons possess only 1.01 J/mol, a quantity far
too small to break even a low-energy hydrogen bond.
In the gas phase, the absorption of a microwave photon corresponds to a quantized
excitation of molecular rotation. As the gas phase is condensed, intermolecular interactions
increase and broaden the quantized excitations. Those very intermolecular interactions are
responsible for converting the rotational energy into translational, heat energy. A single
frequency of 2.45 GHz is suitable for heating most (polar) solvents. Contrary to popular belief,
2.45 GHz does not correspond to the maximally efficient frequency for heating water. For a
more detailed discussion of the dielectric mechanism of microwave heating, refer to Appendix I.
Microwave-heating
of organic reactions: the seminal
papers
Two highly cited papers published in Tetrahedron Letters 1986 demonstrated the use of
domestic microwave ovens to heat organic reactions. The first published article,7 as of Feb. 15,
2010 and as reported by SciFinder Chemical Abstracts service, has been cited 665 times. A
number of reactions were heated in a sealed Teflon vessel in the microwave. Comparisons of
time required and yield were drawn versus conventionally heated reactions, and a ratio of
microwave rate to classical rate is explicitly reported, but no temperatures were measured for
7
Gedye, R.; Smith, F.; Westaway, K.; Ali, J.; Baldisera, L; Laberge, L; Rousell, J. Tetrahedron Lett. 1986, 27
(3), 279-282.
9
the microwave reactions. The authors explore an "inverse relationship between rate
enhancement and the boiling point of the solvent." A double reciprocal plot is presented
relating time to reach 65 % completion and the volume of the container, resulting in a
correlation coefficient of 0.996, "implying the reaction rate is directly proportional to the
pressure developed in the vessels." The authors do not speculate the pressure generated in a
sealed vessel may be proportional to temperature.
The second published report8 has been referenced less: only 457 citations had been
made as of the same date. The authors note the "potential for regulating the maximum
temperature desired by simple solvent choice," contrary to one of the benefits of the modern
microwave reactor.9 The need to monitor the temperature is recognized, but unable to use
traditional thermometers to record the temperature, they cleverly use a series of melting point
tubes containing an assortment of compounds (Table 1) to estimate the reaction temperature.
As each compound melted or decomposed, they were able to confirm that the reactor had
reached the indicated temperature.
Table 1. Compounds chosen as temperature indicators.
Compound
Benzophenone
Deoxybenzoin
Biphenyl
Vanillin
Benzil
Resorcinol
Benzoin
Adipicacid
Semicarbazide-HCI
Succinic acid
Melting point (° C)
48
55-56
69-72
81
95
109-110
134-136
152-167
177-178 (dec.)
187-190
Compound
(±)-tartaric acid
Phthalicacid
Caffeine
Furmaricacid
L-alanine
(±)-tyrosine
Potassium nitrite
Potassium nitrate
Lead nitrate
m.p.(°C)
205 (dec.)
210-211 (dec.)
236-237
299-300
315-317
325 (dec.)
361 (yellows)
370
400 (brown gas)
Giguere, R.J.; Bray, T.L.; Duncan, S.M. Tetrahedron Lett. 1986, 27(41), 4945-4948.
It is this author's belief that the greatest advantage to using a modern microwave reactor is the
capability to optimize the solvent and temperature of a reaction without regard for the solvent's boiling
point under atmospheric pressure.
9
10
For some reactions, the vessel was placed in a dish of vermiculite, which acted to absorb
the microwave energy and allowed even low absorbing solvents such as hexane to be readily
heated. This concept was to be lost for many years only to be reinvented as the silicon carbide
(or similar) heating element, a clever tool for heating reactions in a microwave reactor that do
not readily heat under microwave irradiation.
Modern microwave
reactors
These first reports of microwave heating of chemical reactions used domestic
microwave ovens, but the safety of such an application is dubious. Since then, a number of
commercial manufacturers of dedicated scientific microwave reactors have emerged, including
Anton Paar, Biotage, CEM and Milestone. Modern microwave reactors are designed with safety
in mind. Microwave leakage is minimized through engineering. Magnetic stirring is included.
Temperature and pressure sensors are integrated with software which allows control of the
reaction conditions. Whereas a domestic oven may power cycle (as in a "defrost" setting),
scientific reactors are typically capable of applying variable microwave power, for more precise
control and increased safety.
Small-scale reactors may feature a standing-wave monomode microwave field, which is
more focused than the mixed, multi-mode field of larger reactors and domestic ovens. Smallscale reactors typically allow reaction volumes between 0.5 and 100 mL and run one reaction at
a time. Robotically automated queue systems may be used to eliminate the need for direct
human intervention upon reaction completion, expediting reaction throughput. Infrared
measurement of the external temperature of the reaction vessel often suffices, although
frequent calibration of the IR sensor is mandatory. Internal, fiber optic temperature
measurement is preferred, but requires additional equipment. As solvents are heated above
11
their atmospheric boiling point, autogenic pressure arises in the vessel. Each vessel is approved
for use up to a given pressure, thus limiting the maximum temperature to which a particular
solvent can safely be heated. Most reactors therefore employ a mechanism for pressure
measurement, which can often provide a much more immediate glimpse into the reaction
conditions than the temperature measurement (delayed by the time taken to transfer the
internal heat to the exterior of the vessel for infrared sensing). Mechanical measurement of the
deflection of a rubber septum is the least invasive method of pressure measurement, but
pressure may also be recorded by a pressure transducer or through a hydraulic system.
Larger microwave reactors may feature multiple magnetrons rated for many more watts
than the monomode reactors, but because of the much larger reactor cavity volume, they have
a lower power density and are consequently slower to heat reaction mixtures. The microwave
field must be mixed, bouncing around until it finds a suitable absorber; over-heating of reactor
vessel components may occur as a result. Single reaction, large volume vessels may be used or a
rotor system may be implemented, allowing many reactions to be run at once. Although the
microwave field may be homogenized by mixing, care must be taken when using a rotor system:
if the microwave absorptivity differs significantly between vessels, dramatic heating differences
may occur.10 For safety, temperature control must be implemented through the highest
absorbing reaction vessel (internal monitoring of reaction temperature in a rotor system is
usually only supported for a single rotor position).
Advantages
of using a microwave
reactor
The primary advantage of using a microwave reactor is they are designed for sealedvessel conditions. A reaction heated in a sealed vessel can be heated to a temperature above
Leadbeater, N.E.; Schmink, J.R. Tetrahedron, 2007, 63, 6764-6773.
12
the atmospheric boiling point of the solvent. This allows the solvent and reaction temperature
to be optimized without regard for the normal boiling point of the solvent. If, for example, a
particular
reaction
mechanism
is
promoted
by
the
polarizable,
aprotic
nature
of
dichloromethane, the reaction can be run in dichloromethane. Even if the reactants are poorly
soluble at room temperature or reflux, the use of a particular solvent is not precluded because
solubility may improve considerably above the atmospheric boiling point.
The second advantage of running a reaction in a sealed vessel is kinetic; reactions
proceed more rapidly at higher temperature. Because reactions in sealed vessels can be heated
to higher temperatures than reactions open to the atmosphere, reactions in sealed vessels can
proceed more rapidly than those limited to the reflux temperature of the solvent. A more
extensive discussion of the advantages of running reactions in sealed vessels can be found in
Appendix II.
This is not to suggest there are no disadvantages nor occasions when a sealed vessel is
not desired. Some reactions benefit from driving off side-products such as water or other
volatiles, which would be maintained in the reaction mixture in a sealed vessel. Other reactions
may benefit from the vigorous mixing of solvent reflux. Not all reactions can be run at high
temperatures and would therefore certainly not benefit from heating above the boiling point.
Microwave reactors also introduce a number of new challenges and problems that may
be taken for granted otherwise. First, for the sake of safety, the vessel is not immediately
available for inspection. Color changes, formations of precipitate, and other visual cues are
therefore not available to the microwave chemist.11 Secondly, and on a related note, stirring is
not necessarily as strong in a microwave reactor as on a conventional stir plate. Unfortunately,
11
One work around for this problem is the integration of a digital camera. For an example of this strategy,
see: Bowman, M. D.; Leadbeater, N. E.; Barnard, T. M. Tetrahedron Lett. 2008, 49,195-198.
13
because the vessel is hidden from sight, the efficacy of stirring in a particular reaction cannot be
immediately assessed nor adjusted. As a result, particularly viscous reaction mixtures or
reactions forming significant quantities of precipitate may suffer.
Reactions
with gaseous
reagents
In order for a reaction with a gaseous reactant to occur in solution, the gas must first
dissolve in the solvent. Conventional wisdom tells us that the solubility of a gas increases with
decreasing temperature. Although this is often true, the dissolution of a gas actually follows the
same fundamental principles as any other solute.12 If the energy lost by opening a pocket in the
solvent is greater than the energy gained by filling said pocket with a molecule of the gas,
dissolution does not occur. Conversely, if the gas molecule has enough favorable interactions
with the solvent molecules lining the edge of the pocket to overcome the destabilization of
opening the pocket, dissolution occurs. Some gases exhibit quite complex temperature-solubility
relationships as a result. Sometimes the trend in solubility will even reverse over a given
temperature range. Hydrogen13 and helium,14 for example, are more soluble in warm benzene
than cold. Likewise, high pressure experiments with nitrogen15,16 and hydrogen17 have
demonstrated solubility minima at 70 and 55 °C, respectively, in water.
Sealed vessels allow the efficient use of gaseous reagents. Some reactions with gaseous
reagents can occur if the gas is bubbled through the solvent, but in general, it is preferred to
maintain an atmosphere of the gas above the reaction. In a sealed vessel designed for elevated
pressures, the solubility and effective concentration of the gaseous reagent can be increased by
12
Mysels, K.J. J. Chem. Ed. 1955,32 (8), 399.
Maxted, E.G.; Moon, C.H. Trans. Faraday Soc. 1936, 32, 769.
14
Lannung, A. J. Amer. Chem. Soc. 1930, 52, 68-80.
15
Wiebe, R.; Gaddy, V.L; Heins, C, Jr. J. Amer. Chem. Soc. 1933,55, 947-953.
16
Saddington, A.W.; Krase, N.W. J. Amer. Chem. Soc. 1934,56, 353-361.
17
Wiebe, R.; Gaddy, V.L. J. Amer. Chem. Soc. 1934, 56, 76-79.
13
14
increasing the pressure of the gas, ultimately increasing the reaction rate. This effect can be
exploited to counter the decreased solubility some gases suffer at increased temperature.
Gas-loading capabilities
of modern commercial
microwave
reactors
The compliment between the capabilities of microwave reactors designed for sealedvessel conditions and the requirements of reactions with gaseous reagents has been exploited a
number of times within the past decade. A review article has recognized the number of
publications in the field.18 Commercial microwave reactor manufacturers have also recognized
the market for equipment facilitating this application.
The Anton Paar Synthos 3000 was the first microwave reactor to offer a commerciallyavailable gas loading kit. The Synthos 3000 is a large-scale multimode system featuring two
variable-power 700 W magnetrons. There are several reactor rotors available, the most
interesting of which is the XQ80 rotor with its heavy-walled 80-mL quartz reaction vessels. Each
vessel is hermetically sealed with a Teflon and PEEK cap which includes an o-ringed vent port.
The vent port allows the introduction of gas through a detachable bayonet. Pressure is
measured across the entire rotor by a hydraulic system. Unfortunately, this system can only
record the highest pressure across all the loaded vessels. This is sufficient for safety control, but
does not enable the chemist to confirm the pressure of the loaded gas in any particular vessel.
Temperature is measured internally via a wireless gas-bulb probe placed in a single reaction
vessel and externally by an infrared sensor that reads each vessel approximately twice per
minute as the rotor turns. Both temperature sensors must be calibrated regularly. The heavywalled quartz vessels are approved for a maximum working pressure of 1160 psi and a
Petricci, E.; Taddei, M. Chimica Oggi 2007, 25 (3), 45-49.
15
maximum working temperature of 300 °C. For comparison, a pressure of 1160 psi is experienced
at a depth of 780 meters (0.49 miles) under the ocean surface.
The CEM Discovers system is a small-scale monomode reactor featuring a single 300 W
magnetron. The Discover system offers the chemist flexibility in terms of vessels and reaction
monitoring. The standard 10-mL borosilicate glass tubes can operate up to 200 °C and 300 psi
(equivalent to the pressure at 195 meters or 0.121 miles ocean depth). Depending on the set-up
chosen, temperature can be measured externally via an infrared sensor located at the base of
the reactor cavity or via an internal fiber optic probe. If the external temperature measurement
is selected, pressure is typically measured through the mechanical deflection of the vial-sealing
septum. The internal fiber optic probe, on the other hand, sits within an intrusive sapphire
thermowell that necessitates the use of a remote load-cell transducer to measure pressure. This
set-up can be easily modified to allow the introduction of gaseous reagents, with the added
bonus that the pressure can be monitored during addition.19 A formal gas-loading kit has also
become available from the manufacturer.
Published reports of microwave-heated
reactions with gases
One of the earliest reports of a microwave-heated reaction with a gaseous reagent
demonstrated the "effect of pressure on microwave-enhanced Diels-Alder reactions," referring
to the pressure of the ethylene dienophile (Scheme l). 20 Optimization in dichlorobenzene
concluded increased temperature increased the reaction rate, although no benefit was noted
beyond 190 °C. The procedure conducted under elevated ethylene pressure was only marginally
more successful (in that the reaction could be heated higher than 190 °C without the retro-Diels-
19
20
Petricci, E.; Mann, A.; Schoenfelder, A.; Rota, A.; Taddei, M. Org. Lett. 2006,8 (17), 3725-3727.
Kaval, N.; Dehaen, W.; Kappe, CO.; Van der Eycken, E. Org. Biomol. Chem. 2004,2,154-156.
16
Alder reaction competing)21 than the same microwave-heated transformation conducted in a
vessel purged and sealed with only 15 psi ethylene.22 Grubbs metathesis reactions have also
been microwave-heated in a sealed vessel purged with ethylene.23
Ph
i
XX
_
CI^N^CI
1.10 bar ethylene, DCB
.
190°C,20min.
2 Na
° K H2°/THF
70 °C, 5 min.
....
9\ A
-Ph
H N Cl
O
85 % isolated
Scheme 1. Example of a microwave-heated Diels-Alder reaction at elevated ethylene reagent pressure.
Catalytic hydrogenation has been examined a number of times under microwaveheating. Heterogeneous palladium catalysts have been noted to retain increased activity over
longer time in the hydrogen-transfer hydrodechlorination of chlorobenzene.24 This was
exploited in a flow microwave reactor.25 Hydrogen gas was used instead of isopropanol as the
terminal reducing agent. Comparison between the microwave reactor and a conventionally
heated flow reactor is attempted at various temperatures. A microwave-heated flow reactor,
however, will have a temperature gradient across the length of the flow cell in the microwave
field. Temperature measurement was made at a single point along this path. As a result, it is
difficult to draw comparisons between conventional reactor efficiency and microwave reactor
efficiency
at
various
temperatures.
Interestingly,
the
product
selectivity
for
simple
hydrodechlorination is noted to be higher in the microwave reactor, whereas further reduction
of the aromatic was observed in the conventional reactor.
Hydrogenation with hydrogen and palladium on carbon has been used to effect
dearomatization, debenzylation and azide reductions in a custom polytetrafluoroethylene (PTFE)
21
Kappe,, O.C. Angew. Chem. Int. Ed. 2004, 43, 6250-6284.
Van der Eycken, E.; Appukkuttan, P.; Borggraeve, W.D.; Dehaen, W.; Dallinger, D.; Kappe, CO. J. Org.
Chem. 2002, 67, 7904-7907.
23
Castagnolo, D.; Renzulli, M.L.; Galletti, E.; Corelli, F.; Botta, M. Tetrahedron: Asymmetry 2005,16, 28932896.
24
Radoiu, M.T.; Calinescu, I.; Martin, D.I.; Calinescu, R. Res. Chem. Intermed. 2003,29 (1), 71-81.
25
Pillai, U.R.; Sahle-Demessie, E.; Varma, R.S. Green Chem. 2004, 6, 295-298.
22
17
vessel protected in a polyetheretherketone (PEEK) jacket in a multimode microwave reactor.
Elevated reaction temperatures allow the reductions to be completed in times dramatically
shorter and at hydrogen pressures lower than conventional methods.
CEM has developed and exploited their commercial gas-loading apparatus compatible
with the Discover system to effect the catalytic hydrogenation of a range of substrates.27
Palladium on carbon (1 mol %) under hydrogen (50 psi) allowed efficient reduction of alkenes,
imines, nitro groups and cleavage of the carbobenzyloxy protecting group. The key seems to
have been the selection of an appropriate solvent: ethyl acetate (2 mL) gave high yields of the
reduced products on the 0.5 mmol scale in 3-20 minutes at 80 °C.
Finally, hydrogenations of pyridines have been effected under hydrogen (120 psi) in
acetic acid (1.5 mL) with a pre-reduction/activation of the Pt0 2 catalyst (10 mol % at the 0.45
mmol scale).28 Repeated cycles of 20 minutes at 80 °C were used for sluggish substrates. The
reaction progress could be qualitatively monitored as hydrogen was consumed through the
pressure decrease. The H-Cube flow hydrogenation system was later used to effect similar
reductions under 435-1160 psi H2 (generated in-situ via electrolysis of water) at substrate
concentrations up to 0.2 M with a variety of different catalysts.29 The authors note that
optimization can be done "on-the-fly" and that the H-cube (non-microwave) reactor is safer
than batch hydrogenation conditions because only small quantities of hydrogen are ever
present, and particularly safer than microwave-heated batch hydrogenations in which selective
microwave coupling with the heterogeneous catalyst can occur.
Heller, E.; Lautenschlager, W.; Holzgrabe, U. Tetrahedron Lett. 2005,46,1247-1249.
Vanier, G. Synlett 2007,1,131-135.
Piras, L; Genesio, E.; Ghiron, C; Taddei, M. Synlett 2008, 8,1125-1128.
Irfan, M.; Petricci, E.; Glasnov, T.N.; Taddei, M.; Kappe, CO. Eur. J. Org. Chem. 2009,1327-1334.
18
A microwave-heated Sonogashira propynylation was exploited in the synthesis of
bromo-substituted arenyl diynes (Scheme 2). The diynes were then polymerized via alkyne
metathesis to yield dehydrobenzannulenes.30The propyne was loaded at 37 psi to a 0.113 mmol
scale reaction. The total reaction volume was approximately 1 mL in a 10 mL vial. This volume
and pressure equates to 0.9 mmol propyne. Even so, the microwave-heated procedure allowed
selective reaction with the iodide, retaining the bromide. Conventional reactions at lower
temperatures and longer reaction times suffered from decreased selectivity. A similar approach
was taken in the Sonogashira reaction between acetylene derivatives and resin-immobilized aryl
halides.31 It should be noted that propyne can be condensed to a liquid at -23.2 °C.
Br
,
YY —
B r
> \ ^ S
PdCI2(PPh3)2, Cul
Br
NEt3,DMF
110°C,3.75min.
Scheme 2. Sonogashira reaction of propyne in a sealed-tube microwave reactor.
More recently, cyclic carbonates have been formed via the zinc-catalyzed reaction of
epoxides with carbon dioxide in ionic liquids under microwave heating.32 The authors report
unique reaction kinetic parameters (activation energy and pre-exponential factor) under
microwave-heated conditions. "Solvent-free synthesis" (in ionic liquid media), however, is
notoriously difficult to maintain at a controlled temperature using microwave heating. No
heating profiles are reported. Additionally, the kinetic data from the microwave experiments
appears to have much less precision, even in the Arrhenius plot as presented. All this evidence
suggests that the temperature of the microwave-heated reactions were not controlled at the
desired temperature and instead fluctuated wildly above the programmed set point.
30
Milijanic, 0 . 1 ; Vollhardt, P.C.; Whitener, G.D. Synlett 2003,1, 29-34.
Erdelyi, M.; Gogoll, A. J. Org. Chem. 2003, 68, 6431-6434.
32
Ono, F.; Qiao, K.; Tomida, T.; Yokoyama, C. J. Mol. Catal. A: Chem. 2007,263, 223-226.
31
19
Increasing
the scale of microwave-heated
reactions
As microwaving heating is used more and more frequently for small-scale preparations,
a question arises: what is the maximum quantity that can be prepared using the same method?
A number of review articles consider this very question and address the concerns raised by
increasing the scale of a microwave-heated reaction.33,34 Current technology allows the scale-up
or scale-out of a microwave-heated reaction through larger batch reactors or flow reactors.
Proponents of flow reactors claim that because only a small quantity of the reaction
mixture is processed at a time, the overall safety is increased. Furthermore, the opportunity to
fix aberrations in the process is allowed through the use of in-stream monitoring, before an
entire lot is wasted. Unfortunately, flow processing presents a number of sizable challenges. It is
very hard to optimize a flow reactor, particularly one heated by microwave irradiation. With
increased residence time in the microwave field, the temperature of the mixture increases. As a
consequence, temperature increases along the length of the reactor until it reaches a maximum
as the mixture exits the microwave field. This variation in temperature means that a method
developed in a small-scale batch reactor often cannot be immediately translated to a flow
reactor at the same temperature. Candidate reactions for a flow reactor must have a short
reaction time, lest multiple passes through the reactor or very slow pumping rates become
necessary. Also, processing slurries or precipitated mixtures may complicate matters by clogging
entrance or exit ports. As a result, most real-world reactions do not easily translate to a flow
reactor.
Large microwave-heated batch reactors are criticized over safety concerns. A microwave
reactor, however, has the advantage that heating immediately ceases once microwave
33
Kremsner, J. M.; Stadler, A.; Kapper, C. O. Top. Curr. Chem. 2006, 266, 233-278.
34
Moseley, J. D.; Lenden, P.; Lockwood, M.; Ruda, K.; Sherlock, J.-P.; Thomson, A. D.; Gilday J. P. Org.
Process Res. Dev. 2008,12 (1), 30-40.
20
irradiation is stopped, whereas a conventional reactor would still have the latent heat present in
the heating apparatus. Safety concerns over exothermic reactions are therefore no different
than those of a conventional reactor. Microwave penetration is frequently criticized; since
microwave radiation penetrates only a few centimeters into a typical solvent, it is argued that a
temperature gradient would develop in the reactor. Again, this is no different from a traditional
reactor in which stirring is required because heat is applied from the vessel walls.
Questions of energy efficiency are, however, warranted. The conversion of electrical to
microwave energy is particularly inefficient, in the range of 50-70 %, compared to the 90 % or
better that can be claimed by conventional large-scale batch reactors.35 A full cost analysis is
necessary to determine which reactions would benefit enough from the decreased processing
time to overcome the additional cost of heating.
These numbers were gleaned from another excellent review on the scale-up of microwave-heated
processes: Strauss, C. R. Org. Process Res. Dev. 2009,13 (5), 915-923.
21
Phenols from iodoand bromoarenes
Reactions of interest to the organic chemist can be broadly divided into two categories:
those that form (or break) bonds between two carbon atoms and those that form (or break)
bonds between carbon and another atom that is not carbon. Bond forming reactions involving
aromatic sp2-hybridized carbon atoms are of interest because of the uniquely high energy
barriers involved. Catalysts that lower these barriers are frequently required. Heteroatom
arylation has been studied with great enthusiasm. The Ullmann ether synthesis was one of the
earliest methods to effect O-arylation36 of a phenol with an aryl halide. The traditional
conditions were harsh and required stoichiometric quantities of copper, while typically
providing only low yields of the diaryl ether. Recent developments have decreased the copper to
a catalytic loading and increased yields under more mild conditions (Scheme 3).
OH
t-Bu
Cu(OTf)2 (cat.)
Cs 2 C0 3 , ArC0 2 H
^ ^ / °
+
Br
^ ^
5 A molecular sieves
5 mol % EtOAc, toluene
t-Bu'
81 %
Scheme 3. Example Ullmann ether synthesis as developed by Buchwald, et al.
The copper-mediated Chan-Lam reaction (Scheme 4) allows coupling between a variety
of H-bearing heteroatoms with aryl boronic acids, stannanes or siloxanes.37,38 Yields are typically
moderate, but the reaction conditions are mild. It has broad applicability for a wide range of N-,
0-, and S-nucleophiles. Stoichiometric copper is usually required.
36
For review of diaryl ether synthesis, see: Sawyer, J. S. Tetrahedron 2000,56, 5045-5065.
Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.; Winters, M. P.; Chan, D. M. T.; Combs, A. Tetrahedron
Lett. 1998, 39, 2941-2944.
38
Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.; Averill, K. M.; Chan, D. M. T.; Combs, A. Synlett 2000, 5,
674-676.
37
22
,N^
+
o
/<55^B(OH) 2
Cu(OAc)2(1.5eq.)
pyridine (2.0 eq.)
^ \
I
°
4 A molecular sieves
CH2CI2, air, 2 days
2.0 eq.
63 % isolated
Scheme 4. Example of a Chan-Lam coupling reaction.
Buchwald-Hartwig cross-couplings rely on a variety of electron-rich phosphine ligands to
support the palladium-catalyzed coupling of amines with aryl halides or tosylates. Strong bases
are required. A wide variety of catalytic systems involving elaborate ligands have been
developed to provide excellent yields over a breadth of substrates. Early examples were limited
to aryl iodides (Scheme 5). The Buchwald-Hartwig reaction has been widely reviewed39 as the
method has been extended to aryl chlorides and a wide variety of heteroatomic nucleophiles.
N^
/ ^ / '
NaOt-Bu (2.8 eq.)
Pd2(dba)3 (1 mol %)
^
I
P(o-tol)3 (2 mol %)
dioxane, 100 °C
°
, „ „ , . , . ,
73 % isolated
2.4 eq.
Scheme 5. Early example of a Buchwald coupling with an aryl iodide and simple phosphine ligand.40
In special cases, bond-forming reactions involving sp2-hybridized carbon atoms can
occur in the absence of a catalyst, particularly when additional functionality stabilizes the highenergy
intermediates.41
For
example,
the
nucleophilic
aromatic
substitution
of
4-
bromoacetophenone, an electron-deficient aryl halide, by piperidine or pyrrolidine can occur in
high temperature water42 (Scheme 6) and has been explored in house.
39
For recent reviews, see: a) Hartwig, J. F. Ace. Chem. Res. 2008, 41 (11), 1534-1544. b) Schlummer, B.;
Scholz, U. Adv. Synth. Catal. 2004,346,1599-1626.
40
Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 1996, 61,1133-1135.
41
For a published example of this type of SNAr chemistry with activated fluoro-, chloro-, and bromoarenes
in microwave-heated DMSO, see: Li, F.; Wang, Q.; Ding, Z. Tao, F. Org. Lett. 2003,5 (12), 2169-2171.
For an early example of organic synthesis (of benzimidazoles) in high-temperature water, see: Dudd, L.
M.; Venardou, E.; Garcia-Verdugo, E.; Licence, P.; Blake, A. J.; Wilson, C; Poliakoff, M. Green Chem. 2003,
5, 187-192.
23
HN^n
H
20
0)n
Ui)n B r
Scheme 6. Aromatic substitution of an activated aryl-halide by a nitrogen nucleophile.
This reaction was explored at a variety of temperatures in the Anton Paar Synthos 3000
XQ-80 sealed quartz vessels. Reaction yields improved with two equivalents of the cyclic amine.
Because of the high temperatures involved, the total recovery (starting material plus product)
following the reaction was also tracked (Table 2).
Table 2. Total recovery of material and yield of the 4-amino-substituted acetophenone at various temperatures.
Temperature
200 °C
200 °C
250 °C
250 °C
300 "C
300 °C
Amine
10 mmol piperidine
20 mmol piperidine
10 mmol piperidine
20 mmol piperidine
20 mmol piperidine
20 mmol pyrrolidine
Yield
13%
33%
39%
58%
64%
80%
Recovery
85%
89%
87%
80%
74%
88%
Reactions were carried out as follows: 4-bromoacetophenone (10 mmol), water (10 mL), and amine were combined
and heated in a seal vessel over a 10 minute ramp to, followed by a 20 minute hold at the indicated temperature.
Yield and recovery were determined by 1H-NMR (CDCI3) with an internal standard (durene) following work up.
Although conversion significantly increased at higher temperatures, as a matter of
convenience, further investigation was undertaken in the CEM Discover system at 200 °C, the
limit of its operation (Table 3). Acyclic secondary amines were found to be unreactive.
Additionally, electron-rich substrates such as bromo- or iodotoluene and bromo- or iodoanisole
were found to be unreactive.
24
Table 3. Exploration of reaction scope of the aromatic substitution with secondary amines.
Substrate
X=
CI
Br
1
Yield
31%
76%
33%
Recovery
86%
quant.
quant.
L^NH
Br
39%
81%
n-Bu2NH
i-Bu2NH
Br
Br
Br
0%
0%
0%
quant.
90%
65%
1
0%
-
86%
87%
88%
91%
0%
0%
88%
95%
Nucleophile
OH
&
Xf
sr
0
H
CNH
-
^ N H
n-Bu2NH
i-Bu2NH
-
Reactions conditions: aryl halide (1 mmol), water (1 mL), and amine (2 mmol) were combined and heated in a seal
vessel over a 10 minute ramp to, followed by a 20 minute hold at 200 °C. Yield and recovery were determined by 1HNMR (CDCI3) with an internal standard (durene) following work up.
Reasoning that one nucleophile may act similar to another, the cyclic amine was
exchanged with phenol, but no reaction was observed. The addition of base to generate
phenoxide, a stronger nucleophile, did not yield any product. It was found, however, that the
addition of a catalytic quantity of electrolytic copper dust promoted substrate conversion. The
diaryl ether product was expected, but not observed. Instead, a new phenolic compound was
formed from the aryl halide substrate (Scheme 7).
not observed
Scheme 7. Conversion of aryl halide in high-temperature aqueous base with a copper catalyst.
As phenol was not incorporated in the product, it was excluded. Although sodium
carbonate was used, it was reasoned that the active nucleophile in water was hydroxide, so
25
sodium hydroxide was exchanged for the carbonate, dramatically improving the conversion. The
aromatic substitution was shown to be unique to water solvent; alkoxide nucleophiles in alcohol
solvent did not yield alkyl aryl ethers. The scope of the reaction was briefly explored at this time
(Table 4). Electron-deficient substrates were found to give better yields than electron-rich
substrates. Hydroxide was found to be more effective than carbonate. Significant decomposition
was evident with some substrates.
Table 4. Aromatic substitution with oxygen nucleophiles.
Substrate
Nucieophile
a"
1
.o 2.0 mmol
"OH
n-BuO
c
Base
Na2C03
2.5 mmol
Na2C03
4.0 mmol
Na2C03
4.0 mmol
Na2C03
NaOH
3.1 mmol
£T
,0-
NaOH
3.1 mmol
NaOH
3.1 mmol
NaOH
3.1 mmol
X=
Br
Catalyst
-
Yield
0%
Recovery
-
Br
-
<5%a
81%
Br
Cu(0) dust
18 % b
-
Br
Cu(0) dust
25%
85%
Br
CI
Br
1
Cu(0) dust
Cu(0) dust
Cu(0) dust
Cu(0) dust
0%
0%
65%
74%
86%
89%
80%
Br
Cu(0) dust
0%
63%
1
Cu(0) dust
12%
72%
2-Br
Cu(0) dust
30%
38%
4-Br
Cu(0) dust
70%
80%
-
Cu(0) dust
31%
81%
Substrate (1.0 mmol), water (1.0 mL), nucleophile/base, and catalyst (0.10 mmol) were heated to 200 °C for 20
minutes. a The product was not identified at this time. b The product formed was 4-hydroxyacetophenone, the
expected product was not observed. c Solvent changed to n-butanol.
A range of copper catalysts were tested. Copper in all three common oxidation states
effected conversion; at this time, metallic copper demonstrated the greatest activity (Table 5).
26
Table 5. Effect of copper catalyst on the aromatic substitution reaction of 4-bromoacetophenone with hydroxide.
Catalyst
Cu(OAc)2
Cu(N0 3 ) 2 a
Cul
Cu(0) dust
Yield
52%
34%
40%
65%
Recovery
90%
82%
86%
89%
NaOH
*•
H20
Substrate (1 mmol), water (1 mL), sodium hydroxide (3 mmol), and catalyst (0.10 mmol) were heated to 200 °C for 20
minutes, unless otherwise noted. a Catalyst and solvent were added in the form of 1 mL of an ICP solution
standardized at 1000 (ig Cu / 1 mL
A search of the literature revealed the C-N bond-forming reaction of primary or cyclic
secondary amines with aryl iodides and bromides could be effected in DMSO (9-36 hours at 6090 °C) by an amino acid, copper (I) iodide catalytic system.43 The method was later extended to
electron-rich anilines and heterocycles, albeit with longer reaction times.44
This copper (I) iodide catalyst with amino acid additive was tested for the promotion of
the Ullmann-type coupling of cyclic amines with activated aryl bromides in water, but
competition with the formation of the corresponding phenol was observed. In fact, the catalytic
system promoted both processes so well, the loading of the amino acid had to be decreased due
to competitive incorporation of proline (Scheme 8). The N-arylation of amino acids with copper
catalyst has been reported in DMA solvent.45 Considering the opportunities afforded by this
combination of reagents, the formation of the phenol product was pursued as the most unique.
Cul
proline
Br' -
"20 0 N ) n
V
^'n
C02H
Scheme 8. Competitive formation of multiple products in the Cul/proline catalyzed aromatic substitution.
1
Ma, D.; Cai, Q.; Zhang, H. Org. Lett. 2003,5, 2453-2457.
Zhang, H.; Cai, Q.; Ma, D. J. Org. Chem. 2005, 70, 5164-5173.
' Ma, D.; Zhang, Y.; Yao, J.; Wu, S.; Tao, F. J. Amer. Chem. Soc. 1998,120,12459-12467.
1
27
Established
methods for the preparation
of phenols from aryl
halides
Phenols have been prepared from electron-deficient aryl fluorides via SNAr substitution
with 2-butyn-l-ol followed by hydrolysis of the allenyl ether formed by KOtBu in DMSO.46 A
similar strategy reacts 2-(methylsulfonyl)-ethanol with the aryl fluoride in DMF with sodium
hydride.47 Direct dehalohydroxylation of some aryl halides can occur through an aryne
intermediate formed with potassium t-butoxide in dimethyl sulfoxide.48 A more common
approach prepares phenols indirectly from aryl halides through the oxidation of a boronate
intermediate.49
Palladium-mediated methods have also been explored. In a two-step process to the
phenol, the t-butyl aryl ether is generated in degassed toluene with 2-5 mol % Pd(dba)2 and a
ferrocene-based phosphine ligand. The ether is then hydrolyzed with triflic acid and
trifluoroacetic acid to give fair to moderate yields of the phenol from the 2-iodo-, 2-bromo-, and
2-chloro-l,4-dimethylbenzene substrates tested.50 A direct dehalohydroxylation method was
later developed that utilized a biphenyl-di-(f-butyl)phosphine with the palladium catalyst in
aqueous potassium hydroxide and dioxane.51 Excellent yields were achieved from a range of aryl
bromides and chlorides. Using tribasic potassium phosphate instead of potassium hydroxide
allowed direct access to symmetric aryl ethers.
Levin, J.I.; Du, M.T. Synthetic Commun. 2002,32 (9), 1401-1406.
Rogers, J. F.; Green, D. M. Tetrahedron Lett. 2002,43, 3585-3587.
48
Cram, D. J.; Day, A. C. J. Org. Chem. 1966, 31,1227-1232.
49
For examples of this strategy, see: a) Gotteland, J. P.; Halazy, S. Synlett 1995, 931-932. b) Halterman, R.
L; McEvoy, M. A. J. Am. Chem. Soc. 1990,112, 6690-6695. c) Skowronska-Ptasinska, M.; Aarts, V. M. L. J.;
Egberink, R. J. M.; Van Eerden, J.; Harkema, S.; Reinhoudt, D. N.J. Org. Chem. 1988,53, 5484-5491.
50
Mann, G.; Incarvito, C; Rheingold, A. L; Hartwig, J. F. J. Am. Chem. Soc. 1999,121, 3224-3225.
51
Anderson, K. W.; Ikawa, T.; Tundel, R. E.; Buchwald, S. I. J. Am. Chem. Soc. 2006,128,10694-10695.
47
28
Phenol has been generated from the hydrolysis of chlorobenzene in water/methanol
with zeolites at 215 °C for 4 hours.52 Early reports on the alkaline hydrolysis of chlorobenzene
indicated diphenyl ether was formed competitively and possibly as an intermediate; 4 molar
equivalents of hydroxide and 20 hours at 300 °C ensured the selective and high-yielding
formation of phenol.53 Copper metal and salts were reported to slightly accelerate the reaction.
Oxygenated phenylacetic acids have been accessed through the dehalohydroxylation of aryl
bromides in aqueous sodium hydroxide (2.4 M) with copper (II) sulfate catalyst (30 mol %) after
36 hours at 150 °C
54
Final optimization
and substrate
screening
Having established promising results with the copper (I) iodide and proline catalyst, the
reaction conditions were optimized for the conversion of 4-bromoacetophenone (Table 6).
Decreasing the loading of proline minimized the formation of the N-arylproline side product.
Additional time allowed for the complete consumption of the starting material.
Table 6. Final optimization of the catalyst for conversion of 4-bromoacetophenone.
Catalytic System
Cul/proline (1:1)
Cul/proline (2:1)
Time
20 min.
20 min.
Yield
60%
72%
Cul/proline (2:1)
30 min.
82%
O
NaOH
H,0
Substrate (1 mmol), water (1 mL), sodium hydroxide (3.1 mmol), and copper catalyst (0.10 mmol) were heated in a
sealed tube to 200 °C for the indicated time.
Early experiments had indicated certain substrates benefited from a further increase in
reaction temperature, but the CEM Discover is limited to a maximum of 300 psi at 200 °C in
water. The Anton Paar Synthos 3000, with the XQ-80 heavy-walled quartz vessels, allows
aqueous reactions to be heated to 300 °C and 1160 psi autogenic pressure. The final substrate
Kucera, M.; Kralik, M. Petroleum and Coal 1999, 41 (2), 96-98.
Meyer, K. H.; Bergius, F. Chem. Ber. 1914,47, 3155-3160.
' Weller, D. D.; Stirchak, E. P.; Yokoyama, A. J. Org. Chem. 1984, 49, 2061-2063.
1
29
screening was run at 200 °C in the CEM Discover and also in the Anton Paar Synthos 3000 at 300
°C (Table 7). Certain substrates slightly benefit from the increased reaction temperature.
Electron-rich aryl bromides, in contrast, only showed conversion at 300 °C. Yields were
moderate for most substrates, but excellent for the optimization substrate, 4-bromo- or
iodoacetophenone.
Table 7. Substrate screening with the Cul/proline catalyst.
Substrate
x=
Temperature
200 °C, 30 min.
300 °C, 0 s
200 °C, 30 min.
300 °C, 30 min.
200 °C, 30 min.
300 °C, 30 min.
200 °C, 30 min.
300 °C, 0 s
200 °C, 30 min.
300 °C, 30 min.
200 °C, 30 min.
300 °C, 0 s
200 °C, 30 min.
300 °C, 30 min.
Yield
99%
quant.
82%
94%
14%
44%
42%
39%
0%
30%
53%
65%
0%
30%
-
300 °C, 0 s
37%
-
200 °C, 30 min.
10%
i
tf
£T
/^Y0CH3
Br
CI
1
Br
1
Br
cc
JT
Reactions at 200 °C were run as follows in the CEM Discover: substrate (1 mmol), water (1 mL), sodium hydroxide
(3.1 mmol) and catalyst (0.10 mmol Cul, 0.05 mmol proline) were heated to 200 °C for 30 minutes. Reactions at 300
°C were run in the Anton Paar Synthos 3000 XQ-80 rotor as follows: substrate (5 mmol), water (10 mL), sodium
hydroxide (15 mmol) and catalyst (0.5 mmol Cul, 0.25 mmol proline) were heated to 300 °C over a 10 minute ramp
and held at that temperature for the period indicated before cooling. A 0 s reaction time indicates cooling
commenced as soon as the target temperature was achieved.
The preferred catalytic system is substrate dependent, as some substrates gave higher
yields previously using electrolytic copper dust. In order to achieve synthetically useful yields,
both temperature and time would have to be optimized for the given substrate. For many
30
substrates, higher temperatures were required for reactivity, but other substrates suffer from
decreased yields at higher temperature as decomposition occurs. Only robust substrates can be
expected to lead to desirable yields under the strongly basic conditions.
Summary
Phenols can be synthesized directly from aryl iodides and bromides in high-temperature
water using hydroxide salts and copper-catalysis. The method is extremely harsh; at 300 °C even
quartz glass is subject to corrosion by aqueous hydroxide. The method described in this chapter
has been published.55
Kormos, C. M.; Leadbeater, N. E. Tetrahedron 2006, 62, 4728-4732.
31
Carbonylation of
aryl iodides
Aryl halides are highly versatile starting materials for a very large number of
transformations. Among them, carbonylation is a powerful method for the preparation of
benzoic acid derivatives. Depending on the nucleophile used, benzoate esters or benzamides
can be formed. Under simultaneously reductive conditions, benzaldehydes can be prepared.
Early literature reports utilize carbon monoxide as the carbonyl source. For example,
aryl iodides can be efficiently carbonylated under 440 psi carbon monoxide in an autoclave using
bistriphenylphosphinepalladium(ll) chloride and molecular sieves (Scheme 9).56 This method
improved on an earlier strategy that employed tetraalkylureas as a mild base.57
PdCI2(PPh)3 (2 mol %)
4 A molecular sieves
Arl + CO + EtOH
«•
THF, 100 °C, 24 hours
ArC0 2 Et
62-97 %
(10 examples)
Scheme 9. Sample carbonylation method under autoclave pressures of carbon monoxide gas.
Later, surrogates for carbon monoxide became more popular, as carbon monoxide was
deemed too toxic and difficult to handle.58 Various compounds decompose to yield carbon
monoxide at elevated temperatures. Under certain conditions, these compounds can be used to
generate carbon monoxide or carbonyl equivalents in situ. For example, alkyl formates have
been used to carbonylate iodo- and bromoarenes, as well as tricarbonyl(chloroarene)chromium
complexes (Scheme 10).59
56
Urata, H.; Hu, N.-X.; Maekawa, H.; Fuchikami, T. Tetrahedron Lett. 1991,32 (36), 4733^1736.
Urata, H.; Maekawa, H.; Takahashi, S.; Fuchikami, T. J. Org. Chem. 1991, 56, 4320-4322.
58
The use of carbon monoxide surrogates has been reviewed: Morimoto, T.; Kakiuchi, K. Angew. Chem.
Int. Ed. 2004, 43, 5580-5588.
59
Carpentier, J.-F.; Castanet, Y.; Brocard, J.; Mortreux, A.; Petit, F. Tetrahedron Lett. 1991, 32 (36), 47054708.
57
32
Cr(CO)3
• H ^ v
>
^==*'
HC0 2 CH 3 (32 eq.)
PdCI2(PPh3)2 (0.5 mol %)
NaOCH 3 (1.4eq.)
15barN 2
2 hr., 80 °C
C|
'
84 %
Scheme 10. Methoxycarbonylation of tricarbonyl(chloroarene)chromium complex.
Metal carbonyl complexes, such as Mo(CO)6, have also been used as convenient sources
of carbon monoxide for small-scale reactions (Scheme l l ) . 6 0 The easily-weighed and handled
metal carbonyl can be added to a reaction vessel before it is sealed and heated, at which point it
decomposes
to
liberate
carbon
monoxide.
This
strategy
has
been
reported
for
aminocarbonylations under a variety of microwave-heated conditions.61,62,63,64 Unfortunately,
the metal carbonyl is used in excess, generating at least one molar equivalent of heavy metal
waste. Further, the metal carbonyl is itself toxic and prohibitively expensive for larger reactions.
H
2
^
N"^^
(1-5 eq.)
Mo(CO) 6 (1.0eq.)
DBU (3.0 eq.)
Pd(OAc) 2 (7mol%)
dioxane, 125 °C, 10min.
-
O
jj
rrV^^
H
J J
II
°
8oo/o(25mmolscale)
^
Scheme 11. Aminocarbonylation with Mo(CO)6 in the Biotage Advancer.
Interestingly, more and more recent reports use carbon monoxide gas. Butyl benzoates
were formed quantitatively from aryl tosylates and mesylates under carbon monoxide gas using
a commercially-available phosphine ligand (Scheme 12).65 This work improved upon a previous
method for the carbonylation of chloroarenes.66
60
Appukkuttan, P.; Axelsson, L; Van der Eycken, E.; Larhed, M. Tetrahedron Lett. 2008, 49, 5625-5628.
Georgsson, J.; Halberg, A.; Larhed, M.J. Comb. Chem. 2003, 5, 350-352.
62
Wu, X.; Larhed, M. Org. Lett. 2005, 7, 3327-3329.
63
Lagerlund, O.; Larhed, M.7. Comb. Chem. 2006, 8, 4-6.
64
Wu, X.; Ekegren, J. K.; Larhed, M. Organometallics 2006,25,1434-1439.
65
Munday, R. H.; Martinelli, J. R.; Buchwald, S. L. J. Amer. Chem. Soc. 2008,130 (9), 2754-2755.
66
Martinelli, J. R.; Clark, T. P.; Watson, D. A.; Munday, R. H.; Buchwald, S. L. Angew. Chem. Int. Ed. 2007,
46, 8460-8463.
61
33
(2.2 mol %)
Pd(OAc)2 (2 mol %)
OS0 2 R
t - Z v r ^
K 2 C0 3 (2 eq.)
CO (1 atm), n-BuOH (3 eq.)
4 A molecular sieves
toluene, 100 °C, 15 h
^^X02n-Bu
t-Bu^"^
Scheme 12. Sample palladium-catalyzed carbonylation of aryl tosylate or mesylate.
Very recently, palladium on carbon has been used in DMF solvent with DBU base and
130 psi carbon monoxide to effect the alkoxycarbonylation and aminocarbonylation of aryl
iodides with a variety of nucleophiles.67 The reactions were microwave-heated to 130 °C for 1020 minutes. Palladium on carbon for alkoxycarbonylation had been previously demonstrated in
benzene with triethylamine base.68 The same aminocarbonylations were previously disclosed
using bistriphenylphosphinepalladium(ll) chloride in THF solvent with diisopropylethylamine.69
Considerable attention has been paid to developing ever more complex catalysts for
carbonylation in order to expand substrate scope. Palladium-benzthiozole carbenes have been
used in ionic liquids to give excellent yields from bromo- and iodoarenes; electron-poor
chloroarenes reacted but gave low yields.70 Interestingly, there have been a number of reports
in which increasing complex catalysts and catalytic systems apply only to iodoarenes: a dimeric
palladium complex was developed and illustrated to give high yields of benzoate esters under
150 psi carbon monoxide using DBU as a base;71 iron and copper cocatalysts have been used to
accelerate the carbonylation of iodoarenes.72
67
Salvador'!, J.; Balducci, E.; Zaza, S.; Petricci, E.; Taddei, M. J. Org. Chem. 2010, 75 (6), 1841-1847.
Ramesh, C; Nakamura, R.; Kubota, Y.; Miwa, M.; Sugi, Y. Synthesis 2003,4, 501-504.
69
Cardullo, F.; Donati, D.; Merlo, G.; Paio, A.; Petricci, E.; Taddei, M. Synlett 2009,1, 47-50.
70
Calo, V.; Giannoccaro, P.; Nacci, A.; Monopoli, A.J. Organomet Chem. 2002, 645,152-157.
71
Ramesh, C; Kubota, Y.; Miwa, M.; Sugi, Y. Synthesis 2002,15, 2171-2173.
72
Satoh, T.; Kokubo, K.; Miura, M.; Nomura, M. Organometallics 1994,13, 4431-4436.
68
34
Mechanism
of palladium-catalyzed
carbonylation
The mechanism of palladium-catalyzed carbonylative reactions (Figure 1) may begin
with oxidative addition of the aryl halide to a palladium (0) species, generating a palladium (II)
complex. Ligand loss must precede the coordination of carbon monoxide at the cis-position of
the square-planar complex. A distinction should be drawn between insertion of carbon
monoxide into the palladium-carbon bond and migration of the alkyl group to the carbonyl; in
this case, migration of the alkyl group to the already coordinated carbon monoxide likely
occurs.73 Nucleophilic attack on the benzoylpalladium species may release the product, or a
reductive elimination of benzoyl iodide may occur.
Figure 1. A plausible mechanism for the palladium-catalyzed carbonylation of aryl iodides.
C=0 +
L
Ar "OR
ROH + B:
RO" + B+H
There is, however, evidence that alcoholysis must occur at palladium prior to the
reductive elimination which regenerates the palladium (0) catalyst.74 In the presence of excess
phosphine ligand, this alcoholysis is hindered, suggesting alkoxide must first coordinate to
palladium before the ester can be released (Scheme 13).
Anderson, G.K. Ace. Chem. Res. 1984,17, 67-74.
Ozawa, F.; Kawasaki, N.; Okamoto, H.; Yamamoto, T.; Yamamoto, A. Organometallics 1987, 6, 16401651.
35
T „ P
I—Pd"-X
i
Ar
ROH, B :
-*•
BH
T „ ,°
RO-Pd"^
i
Ar
+
"'"
-
O
JJ
RO^Ar
Scheme 13. Alcoholysis through a benzoylpalladium alkoxide, hindered by excess phosphine (L).
Evidence from IR spectroscopy has demonstrated that reductive elimination of benzoyl
iodide is possible in the absence of a nucleophile.75 Under strongly nucleophilic conditions the
release of mono-carbonylated ester product was favored; weaker nucleophiles tended to favor
the formation of the doubly-carbonylated ot-ketoester. High dielectric constant and high donor
number solvents, which favor the dissociation of iodide from the palladium complex, also
promoted the formation of ester over a-ketoester.
In contrast, the mechanism for aminocarbonylation may proceed through formation of a
carbamoylpalladium(ll) intermediate (Scheme 14).76 This mechanism is supported by the
formation of doubly-carbonylated products in the presence of amine nucleophiles (via the
reductive elimination of a benzoylcarbamoylpalladium (II) complex).77
T
Ar-Pd"-NHR 2
i
•
CO
„
B :
BH
°i
?
.0
L
NR2
+
T
°
Ar-Pd11^
i
NR2
Ar^^O
.
T.nP
L
°
11
Ar^NR2
O
^ . JL
Ar
NR2
MR2
o
Scheme 14. Alternate mechanistic pathway of in the palladium-catalyzed aminocarbonylation.
Carbonylation
in water:
hydroxycarbonylation
Given the variety of carbonylative methods that had been established in the literature
using microwave-heating, but with surrogates for carbon monoxide, interest turned toward
75
Yamashita, H.; Sakakura, T.; Kobayashi, T.-A., Tanaka, M. J. Mol. Cat. 1988, 48, 69-79.
Ren, W.; Yamane, M. J. Org. Chem. 2009, 74, 8332-8335.
77
For example, see: lizuka, M.; Kondo, Y. Chem. Commun. 2006,1739-1741.
76
36
developing conditions in the Anton Paar Synthos 3000 with the recently available gas-loading kit
that would effect the carbonylation with carbon monoxide gas. This would avoid the use of toxic
metal carbonyls and would circumvent the generation of heavy metal waste. Using simple
palladium salts as catalysts, the carbonylation of 4-iodoanisole in water was explored. Sodium
carbonate was initially used as the base, because of its established efficacy in the palladiumcatalyzed aqueous Suzuki reaction.78
Depending on substrate, the palladium catalyst of choice was either palladium acetate
(1 mol %) or palladium chloride from an atomic absorption standard solution (0.01 mol %). The
use of more elaborate palladium sources (1 mol % CataCXium® C, 87 %) had no benefit. The
addition of a phase-transfer agent such as tetrabutylammonium bromide (1 eq.) had a
deleterious effect (65 % yield). Decreasing the stoichiometry of the sodium carbonate similarly
decreased the yield (67 %). Increased reaction temperature (200 °C, 80 % yield) adversely
affected the yields of some substrates. After only 10 minutes at 165 °C or 20 minutes at 150 °C,
considerable starting material remained as evidenced by TLC analysis. Thus, the reaction
conditions optimized for 4-iodoanisole were:
substrate (2 mmol), water (10 ml_), sodium
carbonate (3.7 eq.), under carbon monoxide pressure (200 psi) heated to a maximum 165 °C for
a total of 20 minutes before cooling.
Given that the use of different simple palladium salts did not seem to affect the process,
palladium (II) acetate and palladium (II) chloride (in the form of the atomic absorption standard
solution) were used interchangeably depending on the desired catalyst loading. At higher
catalyst loadings, the acetate salt was easily weighed, whereas an inordinate volume of the
palladium chloride solution would have been required. At lower catalyst loadings, the acetate
salt could not be reproducibly weighed, but the chloride solution could be easily distributed with
78
Leadbeater, N. E.; Marco. M. J. Org. Chem. 2003, 68, 888-892.
37
a volumetric pipette. Interestingly, it was found ten-fold decreases in the palladium catalyst
loading had a relatively minor effect on overall yield, even without extending the reaction time
(Figure 2). This is likely an example of the diminishing return on catalyst loading described in the
literature as a result of aggregation and precipitation of colloidal palladium at high
concentration.
79
Figure 2. Logarithmic plot of catalyst loading and resultant yield of the hydroxycarbonylation.
I
0.0001
0.001
0.01
0.1
mol % Pd catalyst
A substrate screen at 1 mol % with palladium (II) acetate and at 0.01 mol % with a stock
palladium chloride solution was undertaken (Table 8). Most substrates benefited from increased
catalyst loading. Increased reaction time for the lower catalyst loading would be expected to
increase the yield, as the catalyst should remain soluble and active.
De Vries, J.G. Dalton Trans. 2006,421-429.
38
Table 8. Substrate screening for the hydroxycarbonylation of aryl iodides.
C0 2 H
R=
-H
4-CH3
2-CH3
4-N0 2
4-NH2
Product Yield
1 % Pd(OAc)2 0.01 % PdCI2
92%
80%
69%
80%
74%
70%
54%
60%
-
R=
4-Ac
4-OCH3
2-OCH3
4-F
4-CF3
Product Yield
1 % Pd(OAc)2 0.01 % PdCI2
82%
59%
86%
72%
68%
63%
77%
70%
59%
-
Substrate (2 mmol), sodium carbonate (7.4 mmol), water (10 ml_), and palladium catalyst were combined in a heavywalled quartz vessel, sealed, loaded onto the Anton Paar XQ-80 rotor and pressurized to 200 ps) with carbon
monoxide. The rotor was heated with an initial 800-1000 W until the reference vessel internal temperature reached
165 °C. After 20 minutes, the vessels were cooled to 50 °C.
A heteroaromatic iodide was also tested:
3-iodopyridine yielded only 15 % with
palladium acetate (1 mol %), possibly due to difficulty in isolating of the zwitterion from
aqueous medium. Aryl bromides showed no reactivity under these conditions, except for 4bromoiodobenzene, where terephthalic acid was generated in 86 % yield.
This interesting result called for exploration. A mixture of 4-bromoiodobenzene and 4bromoanisole produced only terephthalic acid as a product following carbonylation; 4bromoanisole remained. Similarly, a mixture of 4-bromoiodobenzene and 4-bromobenzoic acid
yielded a proportional mixture of terephthalic acid and 4-bromobenzoic acid following
carbonylation. These results indicated that 4-bromobenzoic acid is likely not released as an
intermediate on the path from 4-bromoiodobenzene to terephthalic acid. Furthermore, 4bromoiodobenzene does not in some way produce a palladium catalyst active toward aryl
bromides. Instead, it would seem that an aryl dipalladium species may be generated as first the
iodide, then the bromide undergo oxidative addition (Scheme 15). Further kinetic studies may
be warranted to explore this very interesting and unique substrate.
39
B,-^
BrJJ
* > ^
W * ^
Scheme 15. Proposed dipalladium arene pathway explaining the unique reactivity of 4-bromoiodobenzene.
Difficulties associated with using the Anton Paar rotor system limited the development
of this process. The pressure of carbon monoxide could not truly be optimized because a
minimum was required in order to reproducibly load the reaction vessels. Even then, vessels
occasionally failed to hold pressure during the run, a sure sign the carbonylation had also failed.
The rotor system only measures the highest pressure across all vessels; no individual
confirmation is available until the vessels are individually vented at the conclusion of the run.
Variability in yield was also noted depending on the selection of vessel for the internal
temperature probe. The external temperature of each vessel is monitored for safety and
indicated variation, although not extreme, that would be expected to affect the outcome.
Carbonylation
in alcohol: alkoxycarbonylation
at high
pressure
In order to expand the versatility of the carbonylation method and produce products
other than carboxylic acids, a solvent change would be necessary. With the original
hydroxycarbonylation conditions loosely based on an aqueous Suzuki reaction protocol and
conditions developed recently for the Suzuki reaction in aqueous ethanol,80 attention turned
toward ethanol as a solvent. If the catalytic cycle could be supported in pure ethanol, ethyl
esters would be produced.
Considering this solvent change from water, in which mineral bases are readily soluble,
the base was tested and selected first. Indeed, mineral bases such as sodium, potassium or
cesium carbonate gave modest yields of the desired ester from 4-iodoanisole, but with poor
reproducibility. Overall low recovery seemed to indicate some carboxylic acid may have been
80
Leadbeater, N. E.; Williams, V. A.; Barnard T. M.; Collins M. J. Org. Process Res. Dev. 2006,10, 833-837.
40
generated and lost in the aqueous work up. Triethylamine, as a soluble amine base, gave more
consistent
results,
but
only
slightly
more
attractive
yields. The
more
basic
1,8-
diazobicyclo[5.4.0]undec-7-ene (DBU) gave very high conversion to the desired ester in ethanol.
Anhydrous alcohols (200 proof in the case of ethanol) were used to avoid competitive formation
of benzoic acid derivatives.
Suitable reaction conditions were found at a lower pressure of carbon monoxide than
the carbonylation in water, partly because of the increased solubility of carbon monoxide in
alcohol solvents81 and almost certainly because of the increased level of confidence with which
the gas was loaded. A substrate screen was undertaken using the optimized conditions (Table 9).
Table 9. Screening of the alkoxycarbonylation of aryl iodides at high carbon monoxide pressure.
-I
*ir
R=
-H
-Ac
2-CH 3
4-CH 3
R'OH
EtOH
'-a
^
R'OH
Yield
91%
/-PrOH
90%
EtOH
99%
/'-PrOH
99%
EtOH
92%
/-PrOH
EtOH
.C0 2 R'
R=
4-OCH3
2-OCH3
R'OH
Yield
EtOH
90%
/-PrOH
90%
EtOH
95%
/-PrOH
76%
4-F
EtOH
91%
81%
4-CF3
EtOH
90%
94%
4-NO2
EtOH
89%
Substrate (1 mmol), DBU {1.1 mmol), alcohol solvent (10 mL), palladium (II) acetate (0.1 mol %) were combined and
sealed under carbon monoxide pressure (145 psi), then heated with an initial 1000 W to a maximum of 125 °C for a
total reaction time of 20 minutes before cooling. Yield was determined by 1H-NMR with an internal standard.
In contrast to the hydroxycarbonylation method, ethoxycarbonylation of 3-iodopyridine
gave an excellent yield (98 %). Alkoxycarbonylation of iodobenzene in rerr-butanol yielded only
17 % t-butyl benzoate. Again, 4-bromoiodobenzene yielded interesting results. Depending on
the conditions (primarily the choice of catalyst), diethyl terephthalate and ethyl 4bromobenzoate were formed in varying amounts (Table 10). Interestingly, decreasing the
Ohlin, C. A.; Dyson, P. J.; Laurenczy, G. Chem. Commun. 2 0 0 4 , 1 0 7 0 - 1 0 7 1 .
41
catalyst loading of palladium (II) chloride bistriphenylphospine increased the selectivity for ethyl
4-bromobenzoate. Although the diethyl terephthalate could be favored using palladium (II)
acetate as the catalyst, the results were variable depending on whether or not a stock solution
of catalyst was used as well as how and when the solution was prepared; overall conversion
remained moderate with a large quantity of the starting material remaining.
Table 10. Product distribution from the ethoxycarbonylation of 4-bromoiodobenzene.
I
Br
/^
^ X 0
BOH '
Conditions
1 mol % PdCI2(PPh3)2
0.1 mol % PdCI2(PPh3)2
0.1 mol % Pd(OAc)2 (3 trials)
BrXJ
2
E t
^<^C02Et
+
ethyl 4-bromobenzoate yield
85%
91%
(1)22%; (II) 46%; (III) 30%
utzJ^J
diethyl terephthalate yield
11%
4%
(1) 39 %; (II) 28 96; (III) 36 %
Substrate (1 mmol), DBU (1.1 eq.), ethanol (10 mL) and catalyst were combined and sealed under carbon monoxide
pressure (125 psi) then heated to 125 °C for a total reaction time of 20 minutes before cooling.
In the preparation of a stock solution of catalyst, simply dissolving palladium acetate in
ethanol was found to precipitate palladium black after several hours; DBU can be added to the
solution to increase the stability. Concentrated solutions of DBU, however, tend to decrease the
catalytic activity of the solution.
Although the substrate screening was conducted at the 1 mmol scale in 10 mL solvent,
the XQ-80 vessels actually require a greater reaction volume to immerse the internal
temperature probe. As a result, these reactions were monitored by the external IR sensor, as
the internal temperature probe was suspended above the reaction mixture. This was not noted
until scale-up in the same system was attempted and the internal temperature was noted to be
in disagreement with those previously recorded. Once the issue was recognized, the internal
temperature of the previously run reactions was estimated to be 125 °C based on the recorded
external temperature profiles, then confirmed by re-running a number of substrates to achieve
yields equivalent within experimental error.
42
Part of this temperature reassessment involved a screen of catalyst loading and
temperature (Table 11). Since the rotor system requires a minimum of four vessels, conversion
with triethylamine was also tested. Diminishing returns are seen upon decreasing the catalyst
loading below 0.1 mol % or increasing the reaction temperature above 125 °C, although the
reaction temperature could be decreased to 115 °C if the catalyst loading was concurrently
increased ten-fold.
Table 11. Demonstration of the optimum temperature and catalyst loading.
COoEt
EtOH
Pd(OAc)2
loading
1.0 mol %
0.1 mol %
0.01 mol %
Base (1.1 eq.)
Yield at 115 °C
DBU
DBU
NEt3
DBU
90%
86%
-
^O
Yield at 125 °C
(S.M. recovered)
91%
60 (23)%
12(75)%
Yield at 135 °C
(S.M. recovered)
75 (16)%
24(74)%
Substrate (1.5 mmol), base, catalyst, and ethanol (15 mL) under carbon monoxide (145 psi) were heated with an
initial 1000 W to the indicated temperature for a total reaction time of 20 minutes before cooling. Crude yield and
percent remaining starting material were calculated from the 1H-NMR using an internal standard.
Carbonylation
in alcohol: alkoxycarbonylation
at low Pco
Following the development of the alkoxycarbonylayion method under significant
pressures of carbon monoxide, attention was turned toward more challenging reactions with
gaseous reagents. In particular, hydroformylation with mixtures of carbon monoxide and
hydrogen was of interest. Although syngas mixtures may be purchased, separate hydrogen and
carbon monoxide cylinders provide additional flexibility in reaction optimization.
Unfortunately, the difficulty of reproducibly gas-loading the Synthos XQ-80 rotor would
only be compounded if two gases had to be sequentially added. For this reason and because of
the added flexibility afforded by running one reaction at a time, the CEM Discover was adapted
for the addition of gaseous reagents. The 80-mL vessel provides access ports for two Upchurch
43
Scientific standard fittings for 1/8" OD tubing as well as a temperature probe immersion well.
The original purpose of these fittings was to adapt the vessel for stop-flow batch processing.
Repurposing the ports for the introduction of gas required the use of a four-way valve to allow
switching gases and isolation of the vessel from the gas cylinders during the run (Figure 3).
Figure 3. Illustration of the 80-mL CEM set-up adapted for gas-loading. As shown, gas 1 is being loaded to the vessel.
Clockwise 90° rotation of the valve loads gas 2. Rotation 180° from the position shown isolates the reaction vessel and
puts gas cylinder 2 in-line with the plugged fourth position on the valve.
A similar set-up was described in the hydroformylation of alkenes with a rhodium
catalyst and XANTPHOS ligand in a toluene/ionic liquid mixture.82 A syngas mixture of carbon
monoxide and hydrogen (1:1) at 40 psi in the 80-mL CEM Discover vessel gave high yields of 3substituted propanals (0.64 mmol scale) in as little as 4 minutes at 110 °C. The solvent was a
toluene (4 ml_), ionic liquid mixture.
With an assembly for the addition of multiple gases, a hydroformylation reaction was
tested in the absence of phosphine ligands. Using 4-iodotoluene as a test substrate with DBU
and palladium (II) acetate in THF solvent, a trace quantity of 4-methylbenzaldehyde was
observed. Toluene was also evident, indicating hydrodehalogenation of the substrate. A variety
Petricci, E.; Mann, A.; Schoenfelder, A.; Rota, A.; Taddei, M. Org. Lett. 2006,8(17), 3725-3727.
44
of solvents and bases were screened looking for the formation of aldehyde product, but
hydrodehalogenation continued to be the favored process.
Carbon monoxide did not appear to be participating in the reaction across a wide range
of solvents. Manipulating the relative pressures of carbon monoxide and hydrogen had no
discernable effect. The previous success of incorporating carbon monoxide in alcohol solvents
lead to an attempt using the syngas mixture in a reaction using ethanol as the solvent.
Interestingly, the gaseous reactant selectivity switched and the ethyl ester was produced as the
only product. Interestingly, the reaction proceeded to full conversion under a much lower
partial pressure of carbon monoxide (only 20 psi) than previously used. Without the additional
pressure of hydrogen (130 psi), the crude reaction analysis indicated less than 5 % conversion.
Substituting nitrogen for hydrogen had the same effect, driving the ethoxycarbonylation to
completion. In fact, the vessel could be purged with a low pressure of carbon monoxide then
filled to only 100 psi with nitrogen, and the reaction would yield quantitative conversion to the
ethyl ester product.
It is important to note the difference between absolute and relative pressure. When the
vessel is purged repeatedly with a gas and released to the atmosphere, the pressure sensor will
register 0 psi (relative pressure), but the vessel will continue to contain 1 atm (nominally 14.7
psi) of the gaseous reagent (the absolute pressure). As a result, if the 80-mL vessel contains 15mL of reaction mixture and is purged with carbon monoxide, it will contain approximately 2.6
mmol carbon monoxide, disregarding any dissolved gas during the brief loading process.
/
(1 atm)(0.065 L) = n(0.0821
\
L * atm\
— - (298 K)
mol * K;
n = 2.6 mmol
45
If the vessel is not purged but loaded once to 15 psi, the result will be the same, even though
the pressure sensor will now register 15 psi, since the absolute partial pressure of carbon
monoxide is still just 1 atm. If the vessel is purged and then loaded to 15 psi, the vessel will
contain twice that amount, because the absolute pressure of carbon monoxide will be 2 atm.
The difference is trivial at higher gas loadings, especially if the stoichiometry of the gas is not of
concern. In order to load small quantities of a reactive gas precisely:
1. Purge the system with nitrogen. Load a small overpressure of nitrogen (~10 psi).
2. With the regulator already set above the noted pressure of nitrogen, switch to
the reactive gas and quickly load the precise quantity desired.
3. Switch the off the reactive gas as soon as the desired quantity is loaded to avoid
significant dissolution.
4. If desired, switch back to the nitrogen with the regulator already set above the
current pressure.
This process allows the user to safely purge undesired gases. Regulators are most precise within
a certain range, so loading 15 psi carbon monoxide from 10 to 25 psi seemed to be more
reproducible than trying to load from 0 to 15 psi. Regulators should always be set above the
pressure of the receiving vessel to avoid back contamination into the cylinder. An accurate
head-space volume is probably the most critical factor and the source of greatest error in
delivering precise quantities of a gas to a reaction.
Using these methods, the near-stoichiometric consumption of carbon monoxide in the
alkoxycarbonylation was explored. For convenience, the catalyst loading was increased to allow
direct weighing of the catalyst; this avoided the preparation of potentially unstable stock
solutions. As mentioned earlier, stock solutions tend to darken, then precipitate metallic
palladium after extended storage.
By maintaining the same loading of carbon monoxide (calculated to be 2.6 mmol) and
increasing the pressure of nitrogen and the reaction scale, the effect of the nitrogen
46
overpressure on the overall carbon monoxide consumption efficiency was probed (Table 12).
Increased partial pressure of nitrogen was found to increase the efficiency with which carbon
monoxide was consumed. The maximum quantity of product produced was just under 2.0 mmol
in agreement with the ideal gas law calculated dose of 2.6 mmol.
Table 12. Optimization of carbon monoxide utilization and yield under nitrogen overpressure.
•\
^
.C02Et
EtOH
Reaction Scale
1.5 mmol
2.0 mmol
2.5 mmol
3.0 mmol
Total Relative Pressure
Opsi
50psi
75 psi
100 psi
150 psi
100 psi
150 psi
100 psi
150 psi
100 psi
150 psi
Conversion
<5%
36%
75%
quant.
quant.
84%
94%
59%
79%
49%
62%
CO consumed
0.07 mmol
0.54 mmol
1.13 mmol
1.50 mmol
1.50 mmol
1.68 mmol
1.88 mmol
1.48 mmol
1.98 mmol
1.47 mmol
1.86 mmol
Substrate 4-iodotoluene, DBU (1.1 eq.), ethanol (200 proof, 15 mL) and palladium (II) acetate (0.5 mol %) were
combined sealed in an 80-mL vessel, purged with nitrogen, loaded with carbon monoxide (14 psi) then heated to 125
°C for 20 minutes before cooling. Conversion assessed by crude 1H-NMR analysis.
Based on these results, the 2.0 mmol scale under 150 psi total pressure was favored
because of the high efficiency with which carbon monoxide was consumed without sacrificing
reaction yield. A substrate screening was undertaken testing a variety of aryl iodide substrates in
two alcohol solvents (200 proof ethanol and anhydrous isopropanol) using these conditions
(Table 13). Electron-rich and electron-poor iodoarenes performed equally well. Substrates
bearing groups at the 2-position gave lower yields as steric factors slow the reaction. Yields were
lower for the isopropyl benzoates than the ethyl benzoates.
47
Table 13. Substrate screening of the near-stoichiometric carbon monoxide alkoxycarbonylation.
-a
-I
R=
4-OCH3
2-OCH3
4-CH3
2-CH3
4-Ac
-H
R'OH
EtOH
/-PrOH
EtOH
/-PrOH
EtOH
EtOH
/-PrOH
EtOH
EtOH
^
R'OH
X0 2 R'
<X
Conversion
99%
76%
71%
61%
95%
85%
79%
98%
95%
Isolated Yield
99%
76%
71%
55%
85%
74%
73%
97%
84%
Substrate (2.0 mmol), DBU (1.1 eq.), anhydrous alcohol solvent (15 mL) and palladium (II) acetate (0.5 mol %) were
combined and sealed in an 80-mL vessel, purged with nitrogen, loaded with carbon monoxide (14 psi), pressurized
with nitrogen (total relative pressure of 150 psi) then heated to 125 °C for 20 minutes before cooling. Products were
isolated by silica gel chromatography using an appropriate diethyl ether, petroleum ether mobile phase.
At this point, one additional useful technique evolved. In order to isolate the ester from
the alcohol solvent, diethyl ether was added to the crude reaction mixture. An acidic aqueous
wash with 1 N HCI eliminated almost all traces of DBU. A volume of petroleum ether equal to
the volume of diethyl ether would then be added. Two phases form as the remaining alcohol
solvent separates from the ether phase. An aqueous wash of the ether phase removes most of
the alcohol. The crude ester would then be dried over calcium chloride (which acts to absorb
small-chain alcohols as well as water), concentrated under vacuum, and isolated by silica gel
chromatography.
Carbonylation
in alcohol: large scale
The first attempt at scaling the alkoxycarbonylation reaction tested the limits of the XQ80 vessels in the Anton Paar Synthos 3000 (Table 14). With only 80 mL total volume available
per reaction, the limiting factor was thought to be the quantity of carbon monoxide that could
be loaded. For this reason, the total reaction volume was maintained around 16 mL, with a
carbon monoxide loading of 145 psi (26 mmol), while the reaction concentration was increased.
48
Table 14. Scale-up of alkoxycarbonylation in the Anton Paar Synthos 3000 XQ-80 rotor.
a'- &
Alcohol (ROH)
EtOH
/-PrOH
Scale
1.5 mmol
3.0 mmol
4.0 mmol
5.0 mmol
10 mmol
12.5 mmol a
32 mmol a
1.5 mmol
3.0 mmol
4.0 mmol
5.0 mmol
20 min.
78%
34%
82%
60%
25%
33%
Yield
30 min.
91%
93%
92%
85%
81%
60 % conv.
92%
83%
75%
64%
lodobenzene, DBU (1.1 eq.), palladium (II) acetate (0.1 mol %) and alcohol solvent (15 mL) were combined and sealed
under carbon monoxide pressure (145 psi), then heated with an initial 1000 W to a maximum 125 °C for the total
reaction time indicated before cooling. a Carbon monoxide pressure was increased (250 psi, 45 mmol).
In ethanol, the carbonylation proceeded smoothly up to the 5 mmol scale under 145 psi carbon
monoxide pressure; with eight vessels in the rotor, nearly 40 mmol could be produced in one
run. In other alcohol solvents, the yields decreased more rapidly as the scale was increased. It is
unknown whether this is due to overall slower reaction rate (in which case the yield could be
improved by further extending the run time) or if the catalyst falters. By increasing the pressure
of carbon monoxide (to 250 psi), and carefully controlling the addition of reagents to avoid
palladium acetate dissolving in concentrated solutions of DBU, the scale could be increased to
12.5 mmol per reaction vessel while maintaining 90 % conversion. Following work-up, ethyl
benzoate was isolated (12.2 g, 81 % on the 0.100 mol scale over eight vessels). It had been
thought high pressures of carbon monoxide would deactivate the catalyst, but the limiting
factor now appeared to be the excess of carbon monoxide available or the concentration of the
reaction mixture.
49
Ethyl benzoate was also produced from iodobenzene on a variety of reaction scales in
the Biotage Advancer (Table 15). The Advancer is a 1200 W batch microwave reactor featuring a
365 mL Teflon vessel. The volume was measured experimentally by filling the vessel with water
and closing the reactor down while allowing excess water to flow out. Mechanical stirring
ensures even heating and reaction homogeneity. Upon reaction completion, Flash Cooling™
rapidly cools the reaction mixture by adiabatic expansion into a large steel receiving container.
Nitrogen pressure ensures complete transfer of the reaction mixture. As a safety feature, Flash
Cooling is also activated in the event the system pressure exceeds the operational limit of 290
psi. This safety feature is disabled if the reaction method is paused (as revealed by one incident),
resulting in a potentially dangerous situation if the user intervenes during a pressure spike.
Carbon monoxide could be loaded through the inlet ports on the top of the reactor. At the
largest scale of 100 mmol, the pressure limit of the reactor was approached.
Table 15. Biotage Advancer scale-up of the ethoxycarbonylation of iodobenzene.
cr
COoEt
Scale (Concentration)
50 mmol (0.23 M)
75 mmol (0.36 M)
100 mmol (0.49 M)
Catalyst
1 mol % Pd(OAc)2
0.5 mol % Pd(OAc)2
0.1 mol % Pd(OAc)2
Carbon monoxide
200 psi (84 mmol)
200 psi (86 mmol)
250 psi (110 mmol)
Yield
88.3 %
87.0 %
85.8 %
Iodobenzene, DBU (1.1 eq.), ethanol, and palladium acetate were combined in the Teflon vessel, sealed and
pressurized, then heated to 125 °C for 30 minutes before flash cooling. Product isolated by aqueous acidic work-up.
Each of the carbonylation reactions in the Biotage Advancer, aided by powerful mechanical
stirring, indicated quantitative conversion by ^-NMR of the crude reaction mixtures. These
were concentrated under vacuum prior to an acidic aqueous work up, dried over calcium
chloride (which in addition to its desiccating ability will also absorb any residual alcohol), and
finally concentrated under vacuum. Isolated yields may have been higher if the crude reaction
mixture had been distilled.
50
Finally, the carbonylation method has been tested in the Milestone UltraCLAVE, a largescale microwave reactor featuring a 2 L vessel contained within a 3 L isolation chamber. Prior to
a run, the isolation chamber is pressurized with nitrogen, or in this case, with mixtures of
nitrogen and reactive gas, up to 730 psi. The nitrogen overpressure allows mixtures to be heated
above the atmospheric boiling point without significant loss to the vapor phase. The reaction of
iodobenzene in ethanol yielded 80 % ethyl benzoate at the 1 mol scale under 390 psi carbon
monoxide (an 8 % excess).83 The UltraCLAVE also has the flexibility to run a series of substrates
in individual vessels within the isolation chamber. Six substrates were reacted simultaneously,
each on the 50 mmol scale, again under an 8 % excess of carbon monoxide with an overpressure
of nitrogen (Table 16). Even though all six vessels are open to the same isolation chamber, the
overpressure of nitrogen prevents cross-contamination.
Table 16. Milestone UltraCLAVE 6-vessel test run.
R-nR=
-H
4-CH3
2-CH3
«o
C0 2 Et
A
EtOH
Conversion
>99%
99%
97%
R=
4-Ac
4-OCH3
2-OCH3
Conversion
> 99 %
94%
91%
lodoarene (50 mmol), DBU (1.1 eq), ethanol (90 mL) and palladium (II) acetate (0.1 mol %) were combined in a 150 mL
vessel which was placed in a rack within the isolation chamber. The chamber was sealed then pressurized with carbon
monoxide (120 psi) and nitrogen (up to 730 psi total). Reactions were heated to 125 °C for 30 minutes before cooling.
Aminocarbonylative
carbonylation
As an interesting addendum to the hydroxycarbonylative and alkoxycarbonylative
methods, a rarely-reported double carbonylation has also been observed. As with the
preparation of phenols from aryl halides, this was not the intended product, but an interesting
side product of a different reaction that was then explored in some additional detail.
lannelli, M.; Bergamelli, F.; Kormos, CM.; Paravisi, S.; Leadbeater, N.E. Org. Process Res. Dev. 2009, 13
(3), 634-637.
51
Experiments began with the solvent-free, copper-free Sonogashira
reaction in
piperidine.84 It was found that the reaction was more reproducible and generally higher yielding
with the addition of a trace quantity of copper (I) iodide and was more controllable in
acetonitrile solvent. The neat reaction heated extremely rapidly and frequently overheated
under even low-wattage microwave irradiation. An attempt was made to effect the
carbonylative Sonogashira reaction under these conditions. Instead, the Sonogashira product
predominated with small amounts of the aminodicarbonylated
product (Scheme 16).
Interestingly, neither the aminomonocarbonylated product nor the carbonylative Sonogashira
coupling product were observed.
CO(150psi)
PdCI2(PPh3)2, Cul
I
+
O
^ ^
\ ^
piperidine, CH3CN
Scheme 16. Mixture of undesired products from the attempted carbonylative Sonogashira reaction.
Even though the reaction in acetonitrile is more controllable than neat reaction, in this
case, an exotherm occurred as the reaction reached the desired temperature of 70 °C carrying
the mixture to 100 °C. After 10 minutes, the reaction was cooled and a silica gel column was run
to separate the mixture of products apparent by thin layer chromatography.
The surprisingly up-field ^ - N M R chemical shifts for the aminodicarbonylated product
produced with piperidine lead to the product initially being misidentified as monocarbonylated.
Switching to benzyl amine (Scheme 17) yields greater differentiation between the mono- and
dicarbonylated products in the 1H-NMR, making analysis of the reaction mixture more facile.
84
Leadbeater, N. E.; Tominack, B. J. Tetrahedron Lett. 2003,44, 8653-8656.
52
Mass spectrometry also confirmed the dicarbonylated product: the MS/ESI + indicated M + l ,
M+Na, and fragments peaks corresponding to fragmentation between the carbonyls, in
accordance with
literature
reports.85 The preference for dicarbonylation with
amine
nucleophiles has been reported with the use of a recyclable ionic liquid/palladium catalyst
system.86 The rate of dicarbonylation is much lower with alcohol nucleophiles, typically requiring
very high pressures of carbon monoxide (440 psi) to achieve synthetically useful yields.87
I
I
150psiCO
3 eq. BnNH2
0.6 mol% PdCI2(PPh3)2
1.0 M in CH3CN
O
I
Scheme 17. Aminocarbonylation and aminodicarbonylation of bromoanisole with benzylamine.
The aminocarbonylation proceeds in the absence of the alkyne and copper catalyst. The
carbonylation process occurs considerably slower than the Sonogashira coupling. After 10
minutes at 70 °C, a very low conversion exclusively to the dicarbonylated product was evident.
After 4.5 hours at 70 °C, a mixture of products was observed: 18 % dicarbonylated product and
31 % monocarbonylated product. At 100 °C, the monocarbonylated product could be isolated in
good yield (89 % conversion by 1H-NMR, 78 % isolated following a recrystallization).
It would be very interesting to see selective formation of either mono- or
dicarbonylated products based solely on the manipulation of carbon monoxide pressure and
temperature. Future work may explore this possibility. The conditions developed here for
carbonylation in acetonitrile with bistriphenylphosphinepalladium(ll) chloride have also been,
used to effect lactonization via an intramolecular alkoxycarbonylation.
85
Ozawa, F.; Soyama, H.; Yanagihara, H.; Aoyama, I.; Takino, H.; Izawa, K.; Yamamoto, T.; Yamamoto, A. J.
Amer. Chem. Soc. 1985,107 (11), 3235-3245.
86
Mizushima, E.; Hayashi, T.; Tanaka, M. Green Chemistry 2001,3, 76-79.
87
Yamashita, H.; Sakakura, T.; Kobayashi, T.-A., Tanaka, M. J. Mol. Cat. 1988,48, 69-79.
53
Summary
Palladium-catalyzed carbonylations of aryl iodides can occur in water or alcohol solvents
using simple palladium salts without the addition of any additional ligands. Carbon monoxide
gas consumption can be optimized such that a near-stoichiometric quantity can be employed.
Many of the methods and reactions described in this chapter have been published.88'89,90,91,92
Kormos, C. M.; Leadbeater, N. E. Synlett 2006,11,1663-1666.
Kormos, C. M.; Leadbeater, N. E. Org. Biomol. Chem. 2007, 5, 65-68.
90
Kormos, C. M.; Leadbeater, N. E. Syn/err2007,13, 2006-2010.
91
Bowman, M. D.; Holcomb, J. L; Kormos, C. M.; Leadbeater, N. E. Org. Process Res. Dev. 2008,12, 41-57.
92
Bowman, M. D.; Schmink, J. R.; McGowan, C. M.; Kormos, C. M.; Leadbeater, N. E. Org. Process Res. Dev.
2008,12, 1078-1088.
89
54
Oxidation of
1,4-dihydropyridines
with molecular oxygen
Multicomponent reactions are a powerful class in which three or more reactants can be
combined in a single pot to selectively produce one product, rapidly building molecular
complexity. Examples include the Passerini93 and Ugi94 reactions (Scheme 18) which utilize an
isocyanide as one of the starting materials and can be used to produce peptide-like products.
Ht
+
v
R1
N+
+
+
D 2^ D 3
R2 R3
O
JJ
R4_C 2H
°
"
Rl
R4
N"V°V
A
H R3
R2
n
O R
+
3
R -NH2
+
4
R -C02H
-
R
3
^M^N/^
R 4
Scheme 18. Generic reaction schemes for the Passerini and Ugi reactions, respectively.
The Biginelli pyridimidone synthesis,95 in contrast, condenses an aldehyde with a 1,3dicarbonyl compound and urea, to yield products such as Monastrol (Scheme 19), which has
been found to interrupt cellular mitosis.96
0
1
*0>
^^O
^^/OH
\ \ 1
* V
NH
+
2
H2^S
CHO
Scheme 19. Biginelli synthesis of Monastrol, a potent inhibitor of mitotic kinesin Eg5.
93
For general information, see: Li, J. J. Passerini Reaction. In Name Reactions, 4th ed. Springer-Verlag
Berlin Heidelberg, 2009, pp 415-416.
For an example application of the Ugi reaction to prepare a range of proline peptidomimetics, see:
Nenajdenko, V. G.; Gulevich, A. V.; Balenkova, E. S. Tetrahedron, 2006, 62, 5922-5930.
95
For a review, see: Kappe, C. O. Ace. Chem. Res. 2000, 33, 879-888.
96
Dallinger, D.; Kappe, C. O. Nature Protocols 2007,2, 317-321.
55
Multicomponent reactions such as these can be used to rapidly generate libraries of
complex molecules from a relatively small number of starting materials.
Synthesis
of
1,4-dihydropyridines
Another multicomponent reaction, the Hantzsch dihydropyridine (DHP) synthesis
(Scheme 20) is the condensation of ammonia with two equivalents of a 1,3-dicarbonyl
compound and one equivalent of an aldehyde.97 The product, sharing structural features with
the reduced forms of the key biological coenzymes nicotinamide adenine dinucleotide and
nicotinamide adenine dinucleotide phosphate, can often have biological activity.98 The synthesis
has been adapted for microwave-heating in a sealed vessel.99 Aqueous conditions are possible in
a sealed vessel, but the mixture becomes biphasic at lower temperatures prohibiting translation
to a scalable open-vessel reflux. The addition of ethanol to the mixture makes the mixture
homogeneous, improving the yield at lower temperatures.100
R3
NH
3
+
RA.R2
Rl
+
RJ
R3
—
YY
R^N'V
H
Scheme 20. Hantzsch synthesis of 1,4-dihydropyridines derivatives.
A series of 1,4-dihydropyridines were synthesized by this method (Figure 4). Two
additional substrates were purchased. Some of the 1,4-dihydropyridines are easily isolated by
crystallization; others were purified by silica gel chromatography. DHPs were an attractive
candidate for a clean oxidation process using a pressure of oxygen gas under microwaveheating.
Hantzsch, A. Justus Liebigs Ann. Chem. 1882, 215 (1), 1-82.
For a review, see: Reddy, G. M.; Shiradkar, M.; Chakravarthy, A. K. Curr. Org. Chem. 2007,11, 847-852.
99
Ohberg, L; Westman, J. Synlett 2001,8,1296-1298.
100
Bowman, M.D.; Holcomb, J.L.; Kormos, C M . ; Leadbeater, N.E.; Williams, V.A. Org. Process Res. Dev.
2008,12, 41-57.
98
56
Figure 4. Collection of prepared 1,4-dihydropyridines.
Oxidation of 1,4-dihydropyridines
to
pyridines
With the driving force of aromatization, the oxidation of 1,4-dihydropyridines to
pyridines is a facile process. In fact, the aromatization was observed even under reductive
"hydrogen-transfer" conditions in the presence of a palladium on charcoal catalyst.101 The
oxidation with palladium on charcoal was found to require a catalytic quantity of acetic acid;
better yields for some substrates were achieved with potassium permanganate.
Computational analysis indicates DHPs assume a boat-like conformation in the gas
phase (Figure 5).102 The 4-substituent preferentially occupies the pseudo-axial position, putting
the 4-H in the pseudo-equatorial position.
Kamal, A.; Ahmad, M.; Mohd, N.; Hamid, A. M. Bull. Chem. Soc. Jpn. 1964,37(5), 610-612.
Memarian, H. R.; Abdoli-Senejani, M.; Tangestaninejad, S. J. Iran. Chem. Soc. 2006, 3 (3), 285-292.
57
Figure 5. Conformational preference of 1,4-dihydropyridines.
Eto2c^^
wy/
-C0 2 Et
^
Eto2c
^
An ene reaction with singlet oxygen is prohibited by this conformational preference. The
other conformation, however, may be assumed in solution or at elevated temperatures, which
would allow the ene reaction to occur between DHP and excited oxygen.
Photo excitation
by electrodeless
discharge lamps (EDLs)
The application of microwave irradiation to power the emission of UV light from
electrodeless discharge lamps (EDLs) has been reviewed.103 An EDL is comprised of an evacuated
tube containing a small quantity of fill gas, such as mercury vapor. Electrons accelerated in the
applied microwave field promote the fill gas atoms to an excited state from which UV emission
occurs.
Direct excitation of triplet oxygen to singlet oxygen is forbidden; a sensitizer dye is used
to allow its production. For example, methylene blue has been used as a sensitizer for the
production of singlet oxygen with a 400 W VIAOX lamp (k > 450 nm) for the oxidation of
dihydropyridines.104 Given that dihydropyridines have a strong absorption at 235 nm which is
not present in the corresponding oxidized pyridine,105 it was reasoned that singlet oxygen might
be generated with the dihydropyridine as both reductant and sensitizer irradiated by a mercury
vapor EDL (Amax = 252 nm106). Because the active oxidizing species would be generated in close
proximity to the reductant, an additional rate benefit may be expected.
103
Horikoshi, S.; Serpone, N. J. Photochem. Photobiol., C2009,10, 96-110.
Memarian, H. R.; Abdoli-Senejani, M.; Tangestaninejad, S.J. Iran. Chem. Soc. 2006, 3 (3), 285-292.
105
Fasani, E.; Albini, A.; Mella, M. Tetrahedron 2008, 64, 3190-3196.
106
Horikoshi, S.; Miura, T.; Kajitani, M.; Serpone, N. Photochem. Photobiol. Sci. 2008, 7, 303-310.
104
58
Alternatively, the EDL emission at 252 nm, corresponding to 113 kcal/mol, could
homolytically cleave the N-H bond of a 1,4-dihydropyridine. The bond dissociation energy for
diethyl 2,6-dimethyl-4-phenyl-l,4-dihydropyridine-3/5-dicarboxylate has been calculated to be
90.8 kcal/mol (Scheme 21).107 This could be the initiation step in a radical propagated oxidation
mechanism with oxygen as the terminal oxidant.
.-.,_.
OEt
+ 90.8
kcal/mol
^
Et0
ji
Scheme 21. Homolytic N-H bond cleavage of DHPs.
With multiple possible mechanisms, the oxidation of DHPs to pyridines under oxygen
pressure and ultraviolet irradiation by an EDL was an attractive transformation exploiting the
capabilities of a pressurized reactor under microwave irradiation.
Lighting
the EDL under microwave
irradiation
The first challenge to overcome is the ignition of the EDL. Polar solvents were found to
inhibit the ignition because they absorb the majority of the incident microwave irradiation. For
this reason, less absorbing solvents had to be selected. Unfortunately, the 1,4-dihydropyridines
suffer from poor solubility in nonpolar solvents. A compromise is achieved in acetonitrile, which
dissolves the substrate, particularly at elevated temperature, but also allows transmittance of
microwave energy to the EDL.
Keeping the EDL lit is a second challenge. The EDL generates heat in addition to the
dielectric heating of the solvent. As the programmed temperature is approached, the
microwave system decreases the microwave energy to maintain the desired temperature.
107
Cheng, J.-P.; Lu, Y.; Zhu, X.-Q.; Sun, Y.; Bi, F.; He, J. J. Org. Chem. 2000, 65, 3853-3857.
59
Below a certain threshold microwave wattage, the EDL will extinguish. This threshold is lower
than the wattage required for the initial ignition, such that if the EDL extinguishes, it will not
reignite. The only means to maintain the microwave intensity required is to simultaneously cool
the reactor. This is achieved by blowing pressurized air over the surface of the reaction vessel. It
is also critical that the microwave irradiation be continuous and variable rather than pulsed.
Otherwise, the EDL also pulses, while never reaching full intensity. The complete experimental
set-up is illustrated below (Figure 6).
Figure 6. Illustration of the reaction set-up for the oxidation of DHPs under oxygen gas and UV irradiation in a
microwave reactor. The 4-way valve as shown is loading oxygen to the reaction vessel. To run, the valve would be
rotated 90° clockwise so that the reaction pressure is monitored.
reaction vessel inside Jacket
Substrate
screening
with optimized
oxidation
conditions
Under oxygen pressure (145 psi) and continuous irradiation by the EDL, diethyl 2,6dimethyl-4-phenyl-l,4-dihydropyridine-3,5-dicarboxylate
(40
mM
in
acetonitrile)
cleanly
oxidized to diethyl 2,6-dimethyl-4-phenylpyridine-3,5-dicarboxylate in 30 minutes at 150 °C.
Interestingly, when other substrates were tested under these conditions (Table 17), a second
product was detected.
60
Table 17. Product distribution of the oxidation of various DHPs.
R3
R
R
2\^\.R
R
R.
2
R
i " "N'
H " 11
Ri =
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
R
'r^f*R,
A
Rf
N
R-,
Product Distribution
4-R 3 :4-H
100:0
3:97
28:72
100:0
90:10
n.a.
R3 =
R2 =
-C02Et
-C02Et
-C02Et
-C02NHPh
-Ac
-C02t-Bu
2
-Ph
-cyclohexyl
-n-hexyl
-Ph
-CH3
-H
DHP (0.3 mmol) in acetonitrile (7.5 mL) was heated to 150 °C for 30 minutes under 145 psi initial pressure of oxygen
in the presence of the EDL. A maximum of 600 W was used for the initial heating and lamp ignition; maintaining lamp
ignition required external cooling of the vessel and that the maximum power to be set to 250 W for the duration at
the target temperature. Conversion as estimated by 1H-NMR was found to be > 95 % in all cases. Product distribution
was estimated by H-NMR and confirmed by MS/ES+.
When the 4-substiutent of the DHP was varied from the phenyl substituent of the test
substrate, substituent loss was detected, yielding 4-H-pyridines as a second product. Over a
range of substituents, it was found that the pattern of the product distribution was consistent
with radical loss of the substituent. In this way, when the substituent was phenyl, no phenyl loss
was detected, but as the substituent was varied to methyl, n-hexyl, and cyclohexyl, the 4-Hproduct increased until it was nearly exclusive (Figure 7).
CH3
97 % loss
28 % loss
10% loss
100%
retention
Figure 7. Radical fragment stability as correlated with 4-substituent loss.
This substituent loss has been reported previously under a variety of oxidation
conditions. Photochemical oxidation with oxygen in the presence of a variety of sensitizers lead
to a mixture of 4-substituted- and 4-H-pyridines depending on both the sensitizer used and the
nature of the 4-substituent. In particular, the 4-(2-furfuryl)-DHPs were found to exclusively form
61
the 4-H-pyridine product.108 The same phenomenon was observed with nitrous acid oxidation.109
The mechanism of the oxidation in this case was concluded to involve loss of a carbocation.
Control
experiments
In order to confirm the efficacy of the combination of the EDL irradiation and oxygen
pressure, control experiments were conducted removing each of these variables. A sample of
diethyl
2,6-dimethyl-4-phenyl-l,4-dihydropyridine-3,5-dicarboxylate
(100
mg
in
7.5
mL
acetonitrile) heated to 150 °C for 30 minutes under 145 psi oxygen indicated 17 % oxidation
without the EDL. A sample similarly treated under 145 psi nitrogen in the presence of the
illuminated EDL indicated 20 % oxidation by ^-NMR; the vessel had not been purged of air, so
some oxygen was present. These results indicated both light from the EDL and oxygen are
required for the rapid, high-yielding oxidation.
A previous study of the photochemical oxidation of DHPs similarly found although the
oxidation occurs in the absence of oxidation (under argon atmosphere) the process is expedited
under oxygen (85 % isolated yield after 4 hours compared to 81 % after 9.5 hours for a model
substrate).110'111
Dimedone-derived
1,4-dihydropyridines
The 1,4-dihydropyridines derived from 5,5-dimethyl-l,3-cyclohexanedione (dimedone)
did not undergo oxidation under these conditions (Scheme 22). Additional ring strain introduced
as the central pyridine ring becomes planar may prevent the oxidation from occurring.
108
Memarian, H. R.; Abdoli-Senejani, M.; Tangestaninejad, S. J. Iran. Chem. Soc. 2006, 3 (3), 285-292.
Loev, B.; Snader, K. M. J. Org. Chem. 1965, 30,1914-1916.
110
Memarian, H. R.; Sadeghi, M. M.; Momeni, A. R.; Dopp, D. Monatsh. Chem. 2002,133, 661-667.
111
Memarian, H. R.; Bagheri, M.; Dopp, D. Monatsh. Chem. 2004,135, 833-838.
62
hu
02
Scheme 22. Dimedone-derived DHPs did not oxidize under these conditions.
Summary
A microwave reactor may be used in conjunction with an electrodeless discharge lamp
to effect photochemical reactions. A range of 1,4-dihydropyridines, under a pressure of oxygen
and irradiation from an EDL, cleanly oxidize to pyridines. Retention of the 4-substituent during
the oxidation depends on the nature of the substituent. The method of oxidation described in
this chapter has been published.112
Kormos, C. M.; Hull, R. M.; Leadbeater, N. E. Aust. J. Chem. 2009, 62, 51-57.
63
Heck reaction of ethylene:
one-pot two-step
synthesis of stilbenes
The palladium-catalyzed Heck olefination is a powerful carbon-carbon bond-forming
transformation, allowing substitution of a vinylic hydrogen. The catalytic cycle (Figure 8)
regenerates the olefin and releases the product through p-hydride elimination.
Figure 8. Plausible mechanism for the palladium-catalyzed Heck olefination reaction.
^-R'
R
^
oxidative
insertion
B+H + X" - ^
D
^Pd'X
L L
^ z
PdL2
/
R
^ P d '
R'
X
palladium(O) I
regenerated \
migration
X
L - P d - L p-hydride R
elimination.
,H
B=^ •-v^'Vi^^R'
product released
The Heck reaction is used in the production of a diverse collection of fine chemicals
(Figure 9), such as the herbicide Prosulfuron™; a UV-B sunscreen agent (2-ethylhexyl pmethoxycinnamate) and the pain reliever and anti-inflammatory agent, Naproxen™.113
Figure 9. Key intermediates produced industrially via the Heck reaction.
O
(2 steps from Prosulfuron)
2-ethylhexyl p-methoxycinnamate
2 _ n f
J ^
from^apraxen)6"6
De Vries, J. G. Can. J. Chem. 2001, 79,1086-1092.
64
The use of ethylene in the Heck reaction is rare, despite Heck's assertion that ethylene is
more reactive than methyl acrylate or styrene.114 In the few isolated reports of reactions with
ethylene, very high pressures (290-730 psi) are employed.115 As with carbonylation methods,
surrogates are frequently preferred. Vinyltrialkylsilanes or vinyltrialkoxysilanes may be used.116
Vinyl acetate can be used in a Suzuki coupling with an aryl boronic acid to yield the same
product as a Heck reaction between ethylene and an aryl halide (Scheme 23).117
X
+
„
ii —
< ^ V ^
/OAc
U ~— (
+
(HO)2B
Scheme 23. Heck reaction with ethylene and Suzuki reaction with vinyl acetate to yield styrene derivatives.
Further Heck reaction of a styrene derivative can produce a stilbene. A variety of
electron-rich stilbenes have been shown to have biological activity. Resveratrol (Figure 10), for
example, is an antioxidant found in plants, and has been shown to extend the lifespan of
Saccharomyces cerevisiae (brewer's yeast).118 Combretastatin A4 (Figure 10) is a powerful
inhibitor of tubulin polymerization, which is crucial for tumor vasculature growth.119
Figure 10. Resveratrol (left) and Combretastatin A4 (right), two biologically-active stilbenes.
HO
114
Heck, R. F. J. Amer. Chem. Soc. 1971,93 (25), 6896-6901.
For isolated reports of Heck reactions with ethylene, see: a) Reetz, M. T.; Lohmer, G.; Schwickardi, R.
Angew. Chem. Int. Ed. 1998,37 (4), 481-483. b) Feiring, A. E.; Wonchoba, E. R. J. Fluorine Chem.2000,105,
129-135. c) Furstner, A.; Thiel, O. R.; Kindler, N.; Bartkowska, B. J. Org. Chem. 2000, 65, 7990-7995. d)
Detert, H.; Sadovski, O.; Sugiono, E. J. Phys. Org. Chem. 2004,17, 1046-1050. e) Southard, G. E.; Murray,
G. M. J. Org. Chem. 2005, 70, 9036-9039.
116
For an example, see: Jeffery, T.; Ferber, B. Tetrahedron Lett. 2003,44,193-197.
117
Lindh, J.; Savmarker, J.; Nilsson, P.; Sjoberg, P. J. R.; Larhed, M. Chem. Eur. J. 2009,15 (18), 4630-4636.
118
Howitz, K. T.; Bitterman, K. J.; Cohen, H. Y.; Lamming, D. W.; Lavu, S.; Wood, J. G.; Zipkin, R. E.; Chung,
P.; Kisielewski, A.; Zhang, L.-L; Scherer, B.; Sinclair, D. A. Nature 2003,425,191-196.
119
Tron, G. C; Pirali, T.; Sorba, G.; Pagliai, F.; Busacca, S.; Genazzani, A. A. J. Med. Chem. 2006, 49,1-12.
115
65
Aqueous conditions for the Heck
reaction
Heck couplings between aryl bromides and activated olefins such as styrene or acrylic
acid can be conducted in aqueous media. An excess of potassium carbonate combined with
trace levels of palladium (II) chloride (10 ppm or less in the water) effect the reaction in high
yield over 20 minutes at 175 °C. Key to the transformation is the use of a phase-transfer agent,
tetrabutylammonium bromide.120 These conditions served as a starting point for the exploration
of the reaction with ethylene.
Selective Heck reaction with
ethylene
Aqueous conditions for the Heck reaction have a number of advantages. The reaction is
fast and the catalyst loading is very low. No phosphine ligands for the palladium are required.
Unfortunately, the autogenic pressure of the reaction at 175 °C is already very high, leaving very
little room for the addition of a gaseous reactant. With the high rate of reaction and low
concentration of ethylene, due both to low pressure and low solubility in water, the styrene
intermediate could never be isolated; instead, the styrene initially formed immediately reacted
with a second equivalent of the aryl halide to yield the symmetric stilbene.121
The rather daunting task of controlling the reaction with ethylene presented itself. If the
conditions for the Heck reaction were too favorable, the reaction of the styrene would occur as
the ethylene was consumed. In order to compete with the styrene, a high concentration of
ethylene would be required from start to finish. A solvent with favorably high ethylene solubility
under a high pressure of ethylene would therefore be required. The consumption of ethylene
Arvela, R.K.; Leadbeater, N.E. J. Org. Chem. 2005, 70,1786-1790.
This product has been noted in previous attempts to use ethylene in a Heck reaction. For example, see:
Grasa, G. A.; Singh, R.; Stevens, E. D.; Nolan, S. P.J. Organomet. Chem. 2003, 687, 269-279.
121
66
should also be slow on the timescale of ethylene dissolution, such that the reaction mixture
would remain saturated with ethylene throughout.
It was clear a solvent change would be necessary. A higher boiling solvent would
decrease the autogenic pressure and allow a greater pressure of ethylene to be used. The use of
an organic solvent would have the additional benefit of increased ethylene solubility. These two
factors would combine to allow the aryl halide to react competitively with ethylene over the
initially-formed styrene product. Dimethylformamide was chosen as a high-boiling, organic
solvent. Mixtures of DMF and water maintained good reactivity, but, as with the reaction in
pure water, the symmetric stilbene was favored.
Working in pure DMF required a re-optimization of the reaction conditions. Inexpensive
mineral bases suffer from poor solubility, so an amine base was used. Low-boiling amines,
contributing significantly to the autogenic pressure at elevated temperature, limited the
maximum ethylene loading, already established as key for favoring the isolation of styrene
products. Diisopropylethylamine (Hunig's base) required the addition of phosphine ligands to
support the catalysis. Tributylamine proved effective without phosphine addition.
With an excess of tributylamine in DMF, aryl iodides could be converted to the
corresponding styrenes with low loading of simple palladium sources in the absence of
phosphine ligands. The scope of this reaction was briefly explored (Table 18).
Table 18. Exploration of substrate scope for the ethenylation of aryl iodides.
CH2CH2 (g)
PdCI2
R
NBu3, DMF
R
R=
-H
-Ac
-CH3
-OCH3
Conversion
82%
88%
91%
91%
Aryl iodide (1 mmol), DMF (1.5 mL), tributylamine (0.25 mL), palladium chloride (20 u± of a 1 mg Pd/1 mL 5 wt. % HCI
atomic absorption standard solution), and ethylene (150 psi) were heated in a 10-mL tube to 125 °C for 60 minutes.
Conversion of the crude mixture was determined by H-NMR in d6-DMSO.
67
The low loading of the palladium catalyst (0.02 mol %) and moderate temperature was
necessary to slow the rate of reaction such that the dissolution of ethylene did not limit the
conversion to the desired product.
Extension to a two-step one-pot
process
Having achieved the selective formation of styrene derivatives, but somewhat
disappointed in the isolated yields of the volatile aromatic product from DMF solvent, a process
was desired that could react the crude styrene to generate an interesting, more easily isolated
product. Already primed for the formation of symmetric stilbenes, a method to generate
unsymmetrical stilbenes seemed both attractive and easily within reach.
An excess of olefin could not be used so a series of trials were performed with the
objective of developing a high-yielding Heck coupling between styrene and an aryl halide in
stoichiometric combination. The addition of water, various bases, and different palladium
sources were tested. Temperature and time were constant considerations. The catalytic system
invariably faltered before the process had reached acceptable conversion, until the palladacycle
trans-di(u.-acetato)bis[o-(di-o-tolylphosphino)benzyl]dipalladium
(II), marketed by Strem as
CataCXium® C (Figure 11), was combined with potassium carbonate. The method extended to
the use of aryl bromides as well as the iodides. The use of this palladacyle122 in DMF to effect the
Heck reaction of aryl bromides has been demonstrated.123 The same catalyst has been used in
water under microwave heating to aminocarbonylate aryl bromides with Mo(CO)6.124
Dupont, J.; Consorti, C. S.; Spencer, J. Chem. Rev. 2005,105, 2527-2571.
For examples, see: a) Tietze, L F.; Nobel, T.; Spescha, M. J. Amer. Chem. Soc. 1998,120, 8971-8977. b)
Tietze, L. F.; Schirok, H. J. Amer. Chem. Soc. 1999,121,10264-10269.
124
Wu, X.; Larhed, M. Org. Lett. 2005, 7 (15), 3327-3329.
123
68
Figure 11. CataCXium18 C palladacycle catalyst.
\\
^ ^
II
|
K^P'
o-tol
/
PcLp-
?
Pd
v
..6 i7°- to1
o-f
V t o l
°-tol
X
Unsymmetrical stilbene products could now be produced in two steps from an aryl
iodide followed by an aryl bromide. This method was used to generate a range of products
(Table 19).
Table 19. Preparation of unsymmetrical stilbenes from aryl iodide and aryl bromide.
CH2CH2
R=
-OCH3
-CH3
-Ac
ArBr
4-bromoacetophenone
4-bromotoluene
1-bromonaphthalene
4-bromoanisole
4-bromoanisole
ArBr
Ar
Isolated Yield (Purity)
41 % (89 %)
57 % (92 %)
45 % (88 %)
35 % (92 %)
38 % (93 %)
Aryl iodide (1 mmol), tributylamine (0.25 mL), DMF (1.5 mL), palladium chloride (20 u.L of a 1 mg Pd/1 mL 5 wt. % HCI
atomic absorption standard solution), and ethylene (150 psi) were heated in a 10-mL tube to 125 °C for 60 minutes.
Upon cooling, the excess ethylene pressure was released; aryl bromide (1 mmol), potassium carbonate (1 mmol), and
CataCXium® C (2.3 mg) were added to the crude mixture and heated to 175 °C for 15 minutes. Unsymmetrical
stilbene products were extracted from water with diethyl ether then recrystallized from ethanol.
Interestingly, when 1-vinylnaphthalene was prepared from 1-iodonaphthalene and
ethylene, then reacted with 4-bromoanisole, the major product was distinct (by 1H-NMR) from
the product reported to form from the reaction of 4-iodoanisole and ethylene followed by 1bromonapthalene. The distinct product could not be identified due to difficult in isolation, but it
is suspected that the polymerization of 1-vinylnaphthalene may have out-competed the Heck
coupling with 4-bromoanisole.
It was later found the use of CataCXium® C could be extended to the reaction with
ethylene (albeit at higher temperature), allowing aryl bromides to be used in both steps. A
series of products were generated via this method (Table 20). After two steps and the sequential
69
addition of two aryl bromides, the unsymmetrical stilbene was purified by recrystallization in
ethanol to give moderate yields.
Table 20. Preparation of unsymmetrical stilbenes from two aryl bromides.
Br
Rlh
R=
4-OCH3
4-Ac
4-OCH3
4-CH3
4-OCH3
2-OCH3
4-CH3
4-Ac
4-NH2
2-CH3
ArBr
NBu3, K 2 C0 3
»•
cataCXium© C
DMF
CH 2 CH 2 (g)
NBu3, K 2 C0 3
cataCXium© C
DMF
Conversion to Styrene
(Isolated Yield)
81 % (67 %)
73 % (36 %)
73 % (not isolated)
83 % (69 %)
78 % (44 %)
95 % (not isolated)
Ar
Stilbene Yield
(Purity)
64 % (89 %)
6 % (75 %)
55 % (91 %)
61 % (91 %)
59 % (88 %)
19 % (95 %)
72 % (91 %)
13 % (95 %)
34 % (95%)
-
ArBr
4-bromoacetophenone
4-bromoanisole
4-bromotoluene
4-bromoanisole
2-bromoanisole
4-bromoanisole
4-bromoacetophenone
4-bromotoluene
4-bromoanisole
-
Aryl bromide (1 mmol), anhydrous potassium carbonate (1 mmol), tributylamine (0.25 mL), DMF (1.5 ml_),
CataCXium® C (2.3 mg), and ethylene (150 psi) were heated in a 10-mL tube to 150 °C for 60 minutes. Styrene
derivatives could then tested for conversion by crude 1H-NMR in d6-DMSO or isolated by silica gel chromatography.
To generate unsymmetrical stilbenes directly, the second aryl bromide (1 mmol), potassium carbonate (1 mmol), and
CataCXium® C (2.3 mg) were added to the crude mixture which was then heated to 175 °C for 15 minutes.
The
ethenylation
of
1-bromonaphthalene
yielded
66
% of
the
desired
1-
vinylnaphthalene following silica gel separation. Ethenylation of 4-chlorotoluene proved
unsuccessful, indicating only 5 % conversion to the desired styrene derivative after 1 hour. The
attempted ethenylation of phenyl tosylate, crystallized from a mixture of phenol and tosyl
chloride in pyridine, generated no product with palladium chloride at 125 °C nor with
CataCXium© C at 150 °C.
In this two-step double-Heck process to unsymmetrical stilbenes, the same final product
can be generated two ways. It is apparent from Table 20 (with the two paths grouped together
by shading for 4 different products), the more reactive aryl halide should be reserved for the
second step. The first reaction with ethylene was optimized to occur slowly over 1 hour; a more
70
reactive substrate would be more likely to over-react in the first stage. Additionally, any leftover aryl halide from the first stage could compete in the second stage with the newly added
aryl halide.
Reaction
scale-up
Interestingly, when scale-up of the Heck reaction with ethylene was attempted in the
Biotage Advancer, the seals proved to be permeable to ethylene. Since the atmosphere of
ethylene could not be maintained, the scale-up effort was discontinued.
Reactions
with other low-boiling
olefins
With the means to ethenylate iodo- and bromoarenes established, interest turned
toward other small olefins. Initial experiments with 1-octene indicated a mixture of products
was formed, but the reactivity of cis-2-butene was explored. Again a number of products were
formed (Table 21).
Table 21. Heck reaction of 4-bromotoluene with cis-2-butene (condensed into tube).
R
'Br
+
|f
CataXium C
NBu3, K 2 C0 3
DMF
Conditions
175 °C, 15 minutes
175 °C, 15 minutes, 100 psi nitrogen
(E)-Product
10%
7%
Exo-Product
10%
19%
(Z)-Product
54%
37%
Condensed cis-2-butene (150 mg) was added to a cold mixture of 4-bromotoluene (1 mmol), DMF (1.5 mL),
tributylamine (0.25 mL), potassium carbonate (1 mmol), and CataCXium® C (2.3 mg). Conversion as indicated was
estimated based on the 1H-NMR integration of the vinyl protons relative to the tolyl methyl.
The crude products were easily identified by 1H-NMR. Interestingly, the product distribution
changed when the reaction was run under nitrogen pressure. The selectivity for the major
product is explained by a c/s-addition / c/'s-elimination mechanism, as noted by Heck for a
71
variety of arylpalladium reactions with olefins (Figure 12).125 Decreased selectivity has also been
noted at elevated temperatures,126 presumably as palladium-mediated isomerizations occur.
Figure 12. Mechanism of the formation of the major product from the Heck reaction with cis-2-butene.
Selectivity would not be a factor with a symmetric olefin such as isobutylene. Slow
addition of t-butanol over 2.5 hours to hot phosphoric acid produced isobutylene gas by acidcatalyzed dehydration using the illustrated experimental set-up (Figure 13).127 The isobutylene
was condensed and transferred to a 10-mL microwave tube sealed with a crimp cap septum and
stored in the refrigerator.
Figure 13. Experimental set-up for the production and isolation of isobutylene.
18mLf-BuOH
added dropwise
isobutylene gas (4.9 g)
—*• condensed with
dry ice cold finger
10mLH 3 PO 4
at150°C
126
Heck, R.F. J. Amer. Chem. Soc. 1969, 91 (24), 6707-6714.
Heck, R.F.; Nolley, J.P., Jr. J. Org. Chem. 1972, 37 (14), 2320-2322.
127
This procedure was based on the published method for the preparation of ethylene and propylene:
Newth, G. S.J. Chem. Soc. 1901, 79, 915-917.
72
Test reactions indicated good selectivity for a single olefin product (Table 22), but the
products were difficult to isolate from the DMF solvent and any remaining starting material.
Furthermore, hydrodehalogenation was prominent. Since synthetically useful yields of product
were not attained, efforts were discontinued.
Table 22. Heck reaction with isobutylene.
•,
R=
-Ac
-OCH3
-CH3
Yield
84 mg
124 mg
107 mg
CataXium C
^.
^^
^
Notes
5% ethyl acetate in petroleum ether; acetophenone evident
100% petroleum ether; not separated from aryl bromide
100% petroleum ether; mixture of product and toluene
Reaction conditions: substrate (1 mmol), isobutylene (150 mg), 0.5 mol% palladium catalyst, 0.25 mLtributylamine, 1
mmol potassium carbonate, and 1.5 mL DMF were heated to 150 °C for 60 minutes. Extraction with diethyl ether and
aqueous washing was followed by silica gel chromatography of the concentrated residue.
Summary
Under carefully controlled conditions, ethylene gas can competitively undergo the Heck
reaction with aryl iodides and bromides to form styrene derivatives selectively. These styrene
derivatives can be reacted, without prior isolation and in stoichiometric combination with a
second aryl bromide to yield unsymmetrically-substituted stilbene products. Many of the
methods described in this chapter have been published.128
Kormos, C. M.; Leadbeater, N. E.J. Org. Chem. 2008, 73, 3854-3858.
73
Reaction scale-up with the
Accelbeam prototype
The scale-up of microwave-heated reactions has been an area of continued interest.
Reactions could be scaled-out by running multiples examples across the positions of a rotor such
as that of the Anton Paar Synthos 3000. Reactions can also be sequentially processed in a smallscale reactor using a stop-flow approach.129 Multimode microwaving of large single-pot
reactions were often limited to lower pressures than the glass vial counterparts due to the
Teflon construction of the vessel. Dedicated large scale reactors have become available that
allow access to the same range of conditions as glass vials. The Milestone Advancer,130 for
example, has a 365 mL reaction vessel that can operate up to 320 psi. With an overhead
mechanical stirrer, even viscous reaction mixtures that are troublesome in a small-scale reactor
can be processed.
The Milestone UltraCLAVE131 microwave reactor features a 2 L reaction vessel contained
within a 3 L isolation chamber. The isolation chamber is pressurized with inert gas prior to a run;
prohibiting the reaction mixture from boiling out of the vessel. This reactor is uniquely able to
operate at pressures up to 2900 psi.
The Accelbeam prototype microwave reactor, representing the next step in the scale-up
of batch microwave reactors, is comprised of a 304L stainless steel isolation chamber (rated for
365 psi and 250 °F) inside of which lies a glass plate suspended above a triad of microwave
129
For an example of this strategy applied to Suzuki and Heck couplings, see: Arvela, R. K.; Leadbeater, N.
E.; Collins, M. J. Tetrahedron 2005, 61, 9349-9355.
130
For a description of the apparatus and examples of its application, see: Bowman, M. D.; Schmink, J. R.;
McGowan, C. M.; Kormos, C. M.; Leadbeater, N. E. Org. Process Res. Dev. 2008,12,1078-1088.
131
For a description of the apparatus and an example of its application, see: lannelli, M.; Bergamelli, F.;
Kormos, C M . ; Paravisi, S.; Leadbeater, N.E. Org. Process Res. Dev. 2009,13 (3), 634-637.
74
magnetron waveguides featuring a proprietary design. Reaction mixtures are contained in a
large (5-14 L) evaporating dish supported on the glass plate. A Teflon lid helps contain the
reaction mixture and provides a bearing for the shaft of the stir blade, although stirring
performance is considerably improved by adding a divot in the dish. The Teflon lid also supports
the fiber optic temperature probe. Three water-cooled magnetrons are situated externally on
the underside of the steel isolation chamber and are powered by three remote power supplies
rated to 3 kW each. The recommended maximum output of each magnetron is 2500 W, for a
total of 7.5 kW microwave power.
In order to operate the instrument, cooling water must be run through each of the
magnetrons and the overhead stir motor. The isolation chamber is sealed by an O-ring and held
closed by six bolts spaced around the hatch circumference. The isolation chamber can then be
pressurized with inert gas (usually 280 psi nitrogen). The stir motor speed is controlled by the
software that also records the pressure and temperature profile of the reaction.
With its minimum working volume of 2 L, the Accelbeam prototype microwave reactor
matches the maximum working volume of the Milestone UltraCLAVE, and represents the next
step in the scale-up of microwave-heated batch reactions.
Reaction
development
The goal of the Accelbeam project was to demonstrate the direct scalability of smallscale microwave-heated methods. Ideally, the survey of scaled reaction methods should
demonstrate the versatility of the Accelbeam reactor toward both homogeneous and
heterogeneous reaction mixtures and a wide range of reaction classes. As a matter of
practicality, preference was given for reactions that would yield easily isolable products from
readily available and inexpensive starting materials.
75
Many of the reactions initially considered were eventually discarded. For example, the
synthesis of celecoxib was considered (Scheme 24). The indicated Claisen condensation was
reportedly run in a microwave reactor at 160 °C for 10 minutes.132 Test reactions revealed a
strong exotherm upon the addition of the ethyl trifluoroacetate. Analysis of the crude reaction
mixture using ^-NMR revealed complete conversion upon addition in an ice bath.
O
O
°
NaH
FaC^OEt
DME
OH
^A^c
a
II
CF3
1
H2NO,S
O
OH
I,
^^NHNHz
•HCI
H2N'\
EtOAc,H20
N~N
Scheme 24. Proposed synthesis of celecoxib.
The synthesis of aryl hydrazines has been reported with ascorbic acid reduction of the
diazonium intermediate (Scheme 25).133The diazonium salt is prepared in the usual fashion using
sodium nitrite and aqueous acid.
/^.NH
fi ^ T
o^S, y
H2N'"Q
]
^^N2+C|\\ ^ T
2
NaNC
NaNU2
HCI
""'
»
,, „ q
. Sv , II
2N'"b
H
J
ascorbic acid
HC| A
MU A
'
„
.^
^^NHNH
H M
,, k ,o, sv ; II
H N
2 ''b
2
J -HCi
Scheme 25. Unsuccessful preparation of the aryl hydrazines for the synthesis of celecoxib.
The ascorbic acid reduction proved incompatible with the sulfonamide, so the reduction was
eventually conducted with tin (II) chloride in the usual manner using stannous chloride in
aqueous acid, resulting in a yield of 91 % over two steps. The pyrazole condensation reaction
had been optimized for the correct isomer in ethyl acetate and water mixtures at reflux;134
heating above atmospheric reflux to expedite the condensation would likely only sacrifice
132
Humphries, P.S.; Finefield, J.M. Tetrahedron Lett. 2006,47, 2443-2446.
Norris, T.; Bezze, C; Franz, S.Z.; Stivanello, M. Org. Process Res. Dev. 2009, 23 (2), 354-357.
134
Reddy, A.R.; Sampath, A.; Goverdhan, G.; Yakambaram, B.; Mukkanti, K.; Reddy, P.P. Org. Process Res.
Dev. 2009,13 (1), 98-101.
133
76
selectivity. On the 1 mmol scale, celecoxib was isolated in 48% yield by crystallization, but the
synthetic route was deemed unsuitable for the Accelbeam reactor.
Solvent
heating
Before undertaking synthetic chemistry, a series of solvents were heated to learn the
operation of the instrument. Each solvent (4.0 L) in the series was heated to 150 °C under
nitrogen (280 psi) using the maximum recommended power of the instrument (7.5 kW). The
sample of solvent was mechanically stirred and the temperature was monitored via a fiber-optic
probe inserted directly into the reaction mixture. The heating profile was recorded on an
attached computer. Upon reaching 150 °C, the application of microwave power was ceased and
the hot solvent could be ejected from the reaction chamber (assisted by the nitrogen pressure)
up through a Teflon dip tube leading to a stainless steel exit port controlled by a needle valve
and out through a stainless steel counter-flow heat exchanger coil and finally into a 5-gallon
HDPE collection vessel. The heat exchanger proved extremely effective, rapidly cooling the
exiting superheated liquid to approximately 40 °C. The entire 4-L contents of the reactor could
be ejected and cooled in this manner in a matter of a few minutes. It was suspected and later
demonstrated that the heat-exchanger is too effective, causing precipitation in the cooling
reaction mixtures and clogging in the exit lines.
It has long been noted that certain solvents heat more rapidly than others under
microwave irradiation. A number of formulas have been derived relating the dielectric
properties of a solvent to the microwave heating efficiency and rate in order to explain why
solvents like ethanol and water, with strong dielectric properties, typically heat faster than less
polar solvents. The results of heating the range of solvents in the Accelbeam were therefore
initially quite surprising and confusing (Figure 14).
77
Figure 14. Overlay of the full power Accelbeam 4-L solvent heating profiles (time from 30 °C to 150 °C in parenthesis).
175
2-butanone (4:13)
dichloromethane (4:16)
acetonitrile (4:21)
tetrahydrofuran (4:28)
ethyl acetate (4:54)
ethanol (5:39)
water (9:24)
0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00
time (minutes:seconds)
Across the range of solvents tested, water and ethanol were in fact the slowest to heat.
In order to reach 150 °C, water took more than twice as long as less polar solvents such as
methyl ethyl ketone, dichloromethane, acetonitrile, or tetrahydrofuran. Among the organic
solvents (i. e. those other than water), ethanol proved to heat the slowest, again in contrast to
the heating rates observed during small-scale microwave heating tests. In an attempt to explain
this result, various properties of the solvents were considered. Ultimately, the only property to
correlate to the pattern in heating was the total heat capacity of the solvent sample, or the heat
energy calculated to heat the 4-L volume of solvent from 30-150 °C (Qcaic) (Table 23). With such
a strong correlation (R2 = 0.9569, Figure 15) between the calculated heat energy requirement
and the time to heat (given that the energy was supplied at constant rate of 7.5 kW or 7.5 kJ/s),
one must conclude that each solvent converts the applied microwave field to heat equally well.
If the heating efficiency (defined as the percentage of the "applied" energy converted to heat) is
calculated, it is clear that this is more or less true. Water actually again emerges as marginally
more efficient than the less polar solvents. It is important to note that this number is calculated
78
based on the energy reported by the power supplies; given that magnetron efficiency is typically
only about 50-70 %,135 it is not surprising that the mean heating efficiency is 41 %.
Table 23. Collected solvent properties included calculated energy to heat 4-L by 120 °C (Qcaic)-
Time to
heat from
30-150 °C
4:12.8
4:15.6
4:21.0
4:28.1
4:53.7
5:39.1
9:24.1
Solvent
2-butanone
dichloromethane
acetonitrile
THF
ethyl acetate
ethanol
water
Heat Capacity
(Jmol'K 1 )
Density
(g/mL)
dielectric
constant
Qcalc
158.4
102.3
91.7
123
170
112
75.3
0.805
1.33
0.786
0.889
0.897
0.789
1.00
18.5
9.08
37.5
7.52
6.02
24.3
78.54
849
767
843
728
831
924
2008
averag
(kJ)
heating
efficiency
44.7 %
39.9 %
43.1 %
36.2 %
37.7 %
36.3 %
47.5 %
41 ±4.4
%
Figure 15. Correlation between time to heat solvent sample 120 °C and calculated energy requirement.
11:00
10:00
JJ
9:00
o
m
,
R2 = 0.9569
8:00
7:00
S so
4-.
V
to *;
51
E
6:00
5:00
4:00
3:00
2:00
1:00
0:00
500
1,000
1,500
2,000
2,500
Q ^ to heat 4-L solvent 120 °C (kl)
With a larger sample size and greater cross-sectional area to absorb the incident
microwave energy, even low absorbing solvents heat as efficiently as ethanol. Ethanol may
absorb the majority of the incident microwave energy within the first few centimeters of
1
Strauss, C. R. Org. Process Res. Dev. 2009,13 (5), 915-923.
79
penetration; even heating of the load occurs through stirring. A less polar solvent accesses a
greater cross-sectional area for the microwave absorption. In other words, the microwave
energy may penetrate further through a less polar solvent, but the end result is the same: all
the microwave energy is eventually absorbed and converted to heat.
Two additional solvents were tested but not included: toluene and ethylene glycol, in
that order. Within moments of attempting to heat toluene, a pressure jump was noted and the
run was discontinued. When the reactor chamber was opened, a fine layer of carbon covered
the interior. Presumably, the pressure jump was caused by the concomitant generation of
hydrogen gas as toluene was electrolyzed into carbon black. The heating of ethylene glycol was
tested next. With the reactor fully pressurized and running at full power, a faint pop was heard,
followed by a much less faint pop as the reactor pressure blew out through a brass microwave
waveguide on the underside of the instrument, opposite the operators. It was determined the
toluene run (or the cleaning) had compromised the feed-through seal. After the seal gave way,
the brass was not strong enough to contain the reactor pressure. The seals have since been
redesigned to shed rather than trap liquids collected at base of the reactor. The waveguide has
also been replaced with a stronger aluminum design.
Toluene was not part of the originally planned heating test; it was only included after
the results with progressively less polar solvents never revealed a limit. Given the result of
attempting to heat toluene, however, it is safe to say there is a limit with regard to which the
dielectric properties do not contribute to overall heating efficiency on the large scale. It is
possible that with an even greater cross-sectional area, toluene could be dielectrically heated,
albeit with very deep microwave penetration through the sample. The heating of pure toluene,
however, represents a very unlikely synthetic scenario in which there are no polar reactants,
salts, etc., which could also serve to effect dielectric heating in the sample.
80
Suzuki reaction: preparation
of
4-methoxybiphenyl
The aqueous ethanol Suzuki methodology has proven time and again to be a reliable
and robust transformation, requiring very low catalytic loading of palladium. The conditions for
this reaction were first established in 2005.136 A number of improvements in several respects
have led to the currently preferred conditions,137 which utilize hydroxide bases in ethanol and
water solvent to rapidly effect the coupling of aryl bromides with aryl boronic acids with very
high catalytic turnover numbers from simple palladium salts. Since the reaction mixture is
homogeneous at the reaction temperature, the methodology is fairly insensitive to stirring.
Since the catalytic cycle is so fast, temperature variations a few degrees in either direction rarely
affect the end result. As the reaction mixture is cooled, the product frequently forms a solidcrust
easily isolable from the rest of the reaction mixture. For these reasons, the Suzuki reaction
between 4-bromoanisole and phenyl boronic acid was selected for scale-up in the Accelbeam
reactor (Scheme 26).
/ ^ /
\
1
0
J
B r
/^/B(OH)2
+
8 J
^-^
0.0004 mol % PdCI2
2 eq. NaOH
H20/ethanol
MW, 150 °C, 5min
^
O'
Scheme 26. Aqueous Suzuki coupling of 4-bromoanisole and phenyl boronic acid.
In order to ensure homogeneity, the 4-bromoanisole (4.00 moles) was diluted with
ethanol to 4.0 L before it was added to the reactor dish. A 10 % excess of phenyl boronic acid
dissolved in aqueous sodium hydroxide (2.1 M, 3.8 L) was then added. A commercially available
palladium chloride solution (1.6 mL, 1.001 mg Pd/mL) was diluted to 0.20 L and added last.
Heating to the desired temperature of 150 °C took over 13 minutes. After 5 minutes, the
136
Arvela, R. K.; Leadbeater, N. E.; Collins, M. J., Jr. Tetrahedron 2005, 61, 9349-9355.
In chronological order: a) Leadbeater, N. E.; Williams, V. A.; Barnard T. M.; Collins M. J. Org. Process
Res. Dev. 2006,10, 833-837. b) Bowman, M. D.; Holcomb, J. L; Kormos, C. M.; Leadbeater, N. E.; Williams,
V. A. Org. Process Res. Dev. 2008,12, 41-57. c) Bowman, M. D.; Schmink, J. R.; McGowan, C. M.; Kormos,
C. M.; Leadbeater, N. E. Org. Process Res. Dev. 2008,12,1078-1088.
137
81
reaction mixture was ejected directly into the 5-gallon HDPE receiving container containing ice.
Since this protocol notoriously produces a solid product directly from the cooling reaction
mixture, it was prudent not use the heat-exchanger which would have quickly clogged. After
filtration, an aqueous wash, and drying, 4-methoxybiphenyl was isolated in 91.8 % yield. At
0.0004 mol % palladium loading, the catalytic turnover number for this reaction exceeded
240,000.
Heck reaction: preparation
of 4'-methoxycinnamic
acid
Along the same vein as the Suzuki methodology, the aqueous Heck reaction (Scheme 27)
was the next reaction scaled up in the Accelbeam reactor. The conditions for this reaction have
likewise been tested and iteratively developed since 2005.138 Unlike the Suzuki reaction, the
Heck reaction proves a little more sensitive to reaction conditions: the addition of TBAB as a
phase-transfer agent solubilizes all the reagents, but the loading of TBAB can be decreased if
powerful mixing compensates for its homogenizing effect. The reaction requires a higher
temperature for a longer time. Pressure accumulates through the course of the reaction due to
the hydrolysis of the methyl ester and decomposition of potassium carbonate. As a result, the
Heck reaction is often a good test of the upper limits, both in temperature and pressure, for a
microwave reactor. By varying the TBAB loading, the stirring efficacy can also be probed.
^V6'
^ Jk J
°
0.002 mol % PdCI2
0
+
X
f|
"
0CH
3
3.7eq. K2CQ3
H20/TBAB *"
MW, 175 °C, 15min
°
^
°
Scheme 27. Heck coupling of 4-bromoanisole with methyl acrylate.
Chronologically: Arvela, R. K.; Leadbeater,
Leadbeater, N. E.; Collins, M. J. Tetrahedron
Kormos, C. M.; Leadbeater, N. E.; Williams, V.
D.; Schmink, J. R.; McGowan, C. M.; Kormos,
1078-1088.
N. E. J. Org. Chem. 2005, 70, 1786-1790. b) Arvela, R. K.;
2005, 61, 9349-9355. c) Bowman, M. D.; Holcomb, J. L;
A. Org. Process Res. Dev. 2008, 12, 41-57. d) Bowman, M.
C. M.; Leadbeater, N. E. Org. Process Res. Dev. 2008, 12,
82
On the 2 mole scale, 4-bromoanisole and methyl acrylate (2 eq.) were combined in the
reactor dish with TBAB (1 eq.) and an aqueous solution of potassium carbonate (3.7 M, 2.0 L).
The palladium catalyst (4 mL of a 1.006 mg Pd/mL solution of palladium (II) chloride) was diluted
and added to the reactor to bring the total volume to 4 L. The heating ramp was more
aggressive than the Suzuki example, taking less than 12 minutes to reach 175 °C. As the reaction
progressed, the reaction was allowed to cool slightly (to 165 °C by the end of the 15 minute
reaction time) to maintain the pressure below 350 psi. The mixture was then ejected into the 5gallon HDPE receiving container containing cold water. The process was repeated decreasing the
TBAB loading each time. At 0.5 eq. TBAB, the yield remained 95 %, but decreased to 65 % if the
TBAB was decreased to only 0.25 eq. This result is consistent with that in the Biotage Advancer
(92 % on the 0.1 mole scale), where an overhead paddle stirrer allows the TBAB loading to be
decreased to a similar degree.
Thiouracil condensation
and
S-benzylation
Dihydro-alkoxy-benzyl-oxopyrimidines
(DABOs)
and
dihydro-alkylthio-benzyl-
oxopyrimidines (S-DABOs) have been demonstrated to be nonnucleoside reverse transcriptase
inhibitors (NNRTIs) of HIV-1 reverse transcriptase, a key enzyme in the infection pathway of the
HIV retrovirus. The thiouracil core of S-DABOs has been synthesized via the condensation of
thiourea with a 1,3-ketoester in refluxing ethanol under basic conditions.139 Selective Salkylation of thiouracil derivatives has been reported under microwave heating.140,141
Mai, A.; Artico, M.; Sbardella, G.; Massa, S.; Novellino, E.; Greco, G.; Loi, A.G.; Tramontano, E.;
Marongiu, M.E.; Colla, P.L. J. Med. Chem. 1999,42 (4), 619-627.
140
Petricci, E.; Mugnaini, C; Radi, M.; Corelli, F.; Botta, M.J. Org. Chem. 2004, 69, 7880-7887.
141
Manetti, F., et al. J. Med. Chem. 2005, 48 (25), 8000-8008.
83
S
fl
H2N^NH2
0
+
0
II
II
^ ^ ^ O E t
1 eq. KOH
y
*•
ethanol
^f
HN..NH
T
S
Scheme 28. Preparation of an S-DABO from thiourea and ethyl acetoacetate.
The thiuracil condensation (Scheme 28) was optimized in a CEM Discover on the 2.0
mmol scale in a 10-mL reaction tube. Ethyl acetoacetate (0.5 M in ethanol) condensed with
thiourea (1.2 eq.) in the presence of potassium hydroxide (1 eq.) at 125 °C for 20 minutes
yielding approximately 70 % of the desired product. Upon cooling, when the reaction could be
viewed, the product had formed a sponge in the tube. It was presumed that the product
precipitated upon cooling, but scale-up in the Accelbeam revealed that the product forms the
sponge even at 125 °C. After 20 minutes at 125 °C, the vent was opened and clean solvent
ejected into the receiving container, leaving the dry sponge of product behind in the reactor
dish. The sponge (determined to be the potassium salt of the product) dissolved easily in water.
The product was then isolated by precipitation upon acidification. With the added mixing of the
mechanical stirrer, the reaction yield increased to 94.6 % on the 4 mole scale. Presumably, the
yield faltered on the small scale when the sponge began to prohibit magnetic stirring.
This reaction was conducted in the Accelbeam on three scales as the vessels became
available. On the two mole scale, ethyl acetoacetate (2.0 mol), thiourea (2.6 mol), and
potassium hydroxide (2.0 mol) were dissolved in ethanol, up to a total volume of 4 L. The
reaction was loaded into the reactor chamber, which was sealed and pressurized to 280 psi with
nitrogen. The reaction was heated with 6 kW microwave power to 125 °C as recorded by the
internal fiber optic probe and maintained at that temperature for 25 minutes. Heating was
stopped, the solvent ejected into a collection receptacle, leaving the spongy product behind.
Purification gave a 90 % isolated yield of the desired product. On the 4 mole scale, there were
no significant deviations. All masses and volumes were doubled; the product yield was similar
84
(95 %). In the largest vessel, the reaction was attempted at the 5.5 mole scale, with a total
volume of 11 L. As the reaction reached 100 °C, the stirring motor began to falter, so heating
was stopped. Within 2 minutes, the stirring stopped completely, so the run was discontinued.
The second step in the sequence, the S-alkylation (Scheme 29), was optimized on the small scale
in the CEM Discover based on literature precedent.142'143 In DMF solvent, the reaction at 100 °C
yields 70 % of the desired S-protected thiouracil in 20 minutes. On the 3.4 mole scale in the
Accelbeam, the reaction of 6-methylthiouracil (3.42 mol) with benzyl chloride (1 eq.) and
potassium carbonate (1 eq.) in DMF (5.8 L) gave S-benzyl 6-methylthiouracil in 64 % yield.
_
r H
°YY
HN
YNH
1 eq. BnCI
3
1 eq. K 2 C0 3
—^~*
n
r M
°YY
HN N
r
S
SBn
Scheme 29. Selective S-benzylation of 6-methylthiouracil.
Deoxychlorination with phosphorus oxychloride was followed by a SNAr substitution
reaction to yield the final product in the 4-step synthetic pathway (Scheme 30). On the small
scale, the chlorination was effected in 10 minutes at 120 °C but an exotherm was noted upon
the combination of the reagents, so at the large scale the reaction was refluxed 3 hours using a
heating mantle. The SNAr reaction was effected in 10 minutes at 150 °C. Comparable yields were
achieved at small and large scale.
CH,3
T
SBn
„ .t - N..
pnrL/F
POCI
3/Et3N
Ck^ ^^ ^^^ o Xn H
3
Y l
o l
T
SBn
1 eq. AcOH
H
„1eq. aniline
,„
^ ^
dioxane
^ ^
yJ M . ^
X Hn 3
^f
.HC|
SBn
Scheme 30. Final two steps in the 4-step synthetic pathway.
For the four step synthetic sequence, the overall yield on the small scale using the CEM
Discover was 39 %. At the large scale in the Accelbeam reactor, the yield over 4 steps was 38 %.
142
143
Petricci, E.; Mugnaini, C; Radi, M.; Corelli, F.; Botta, M. J. Org. Chem. 2004, 69, 7880-7887.
Manetti, F., et al. J. Med. Chem. 2005,48 (25), 8000-8008.
85
Whereas the initial condensation benefited from mechanical stirring in the Accelbeam, the
cumbersome isolation in the later steps decreased the yields as compared to the small scale.
Summary
A large-scale prototype microwave reactor has been tested. The reactor allows reactions
on the 2-12 L scale. Initially, a sequence of solvents were heated to reveal that given sufficient
cross-sectional area, even those solvents considered by many to be poor microwave-absorbers
can be effectively heated. A series of reactions were run in order to test various capabilities of
the reactor. The reactions were microwave-heated methods either taken directly from smallscale published reports or optimized in-house on the small scale. The reaction conditions were
maintained as the reactions were scaled and run in the prototype. The prototype successfully
demonstrated the linear scalability of these microwave-heated reactions. The work described in
this chapter has been published.144
144
Schmink, J. R.; Kormos, C. M.; Devine, W. G.; Leadbeater, N. E. Org. Process Res. Dev. 2010,14, 205214.
86
Closing Remarks
The work of a chemist is never done. A reaction can always be optimized just a bit more.
Future
work
With regard to the nucleophilic substitution of electron-deficient aryl halides such as 4-
bromoacetophenone, a catalytic quantity of secondary amine could form the acetophenone
imine, which may allow substitution of the halide with nucleophiles other than the secondary
amine.
In the formation of phenols, amino acids other than proline should be explored as
ligands for the copper catalyst. Because proline resembles pyrrolidine, one would expect the
competitive amine Ullmann coupling product. A different amino acid could favorably support
the catalysis without competitively reacting with the substrate.
The reactivity of 4-bromoiodobenzene deserves more study. Kinetic studies may reveal
the mechanism by which its unusual reactivity occurs.
The aminocarbonylation in acetonitrile could be explored further. The production of 1,2dicarbonyl compounds quite unusual. If conditions could be found to selectively form one
product cleanly, a useful methodology may be had.
A substantial amount of work was devoted to trying to selectively utilize small-chain
olefins other than ethylene in the Heck reaction. A more selective catalyst may be required. In
cases where the products were numerous, a lower temperature may allow one product to be
more strongly favored. Unfortunately, hydrodehalogenation appeared to also be a competing
process. Given that there are a number of other means to cleanly produce the same products,
there may be little interest in extending the gaseous Heck reaction methodology.
87
Appendix I:
Dielectric heating
Electrostatics:
capacitors
in
equilibrium
The dielectric mechanism of microwave heating is understood and described in terms
originating in the description of electrical capacitors.145 A capacitor can be visualized as two
parallel plates separated at some distance by a medium. When a voltage is applied across them,
the plates develop and store charge (Figure 16).
yr
I I I I I
+;+ + +
1/
1
II
Figure 16. Illustration of a capacitor in a DC circuit.
The charge (Q) will be proportional to the voltage (V) applied as well as the capacitance (C)
which defines the capacitor.
Q = V•C
or, equivalent^
C= -
The capacitance is proportional to the area (A) of the plates over which the charge is distributed,
and inversely proportional to the distance (d) between the plates. A greater plate area allows
distribution of the charge such that the opposition of like-charges is decreased. Greater distance
between the plates disfavors the capacitance due to the energy associated with the separation
Craig, D.Q.M. Dielectric Analysis of Pharmaceutical Systems, Taylor and Francis: Bristol, PA, 1995.
88
of the charges. The permittivity (e) or dielectric constant of the medium describes the influence
of the nature of the medium on the stabilization of the charge separation.
For convenience, the permittivity relative to a vacuum (e0 ~ 8.85410"
F/m) is commonly used.
Polar or polarizable molecules are able to align such as to oppose the charges building on the
plates of the capacitor, stabilizing the further build up of charge. Thus, media comprised of polar
or polarizable molecules have large dielectric constants (e > e0). Non-polar media provide little
stabilization relative to a vacuum (e = e0).
Direct
current:
capacitors
in a DC circuit
If a capacitor and resistor (a light bulb, for example) are hooked up to a voltage source
such as a battery, current will flow through the resistor (temporarily lighting the bulb), until the
charge built up in the capacitor yields a voltage equal to that applied by the battery (Figure 17).
i—Hh—m/—i
capacitor
resistor
battery
I
|l
1
Figure 17. Illustration of a capacitor and a resistor in a DC circuit.
A resistor acts according to Ohm's law:
V = I •R
or, equivalently
v
R= -
The current (/) is proportional to the applied voltage (I/) and inversely proportional to the
resistance (/?) in the circuit.
89
Alternating
current:
capacitors
in an AC circuit
tt
alternating voitage
Figure 18. Illustration of a capacitor in an AC circuit.
In an alternating current circuit (Figure 18), the voltage varies according to a sine wave,
such that:
V = V0 sin cot
The current (/) through a capacitor, equivalent to the change in charge over time, can be shown
to be 90° out-of-phase with the applied voltage:
dQ d(V-C)
d smart
I = —r- =
7-— = V0 • C
— = a) • C • Vn cos cot
dt
dt
dt
The maximum current (/0) in the capacitor is therefore proportional not only to the peak voltage
(V0) and the capacitance (C) but also the frequency (co) of the alternating voltage.
/ 0 = co • C • V0
Under alternating voltage, capacitors can therefore be shown to demonstrate a property
analogous to resistance.
_v
co • C = 0— = Xc
analogous to
R= -
This opposition to the alternating current created by the capacitor is known as the reactance
(Xc). As a result of this reactance, the voltages measured across a capacitor and a resistor in
series will differ. Because the reactance and resistance occur 90° out-of-phase, the total
90
opposition to current flow in the circuit, known as the impedance (Z), is illustrated by a phase
diagram (Figure 19).
Figure 19. Phase diagrams illustrating the analogy between impedance (Z) and overall voltage (VT).
The voltage across the capacitor (l/c) and the voltage across the resistor (VR) are similarly related
to the total voltage of the system (VT). The phase angle (<5) describes the behavior of the
resultant voltage relative to the orthogonal relationship previously established between voltage
and current.
Microwaves
and dielectric
heating
In order to understand the mechanism of dielectric heating, the behavior of solvent
molecules under applied voltage must be considered. In an applied electric field, molecules will
align their dipole with the polarity of the field, storing energy like a capacitor stores current. The
dielectric constant describes this ability of the bulk medium (comprised of molecules possessing
either a permanent or induced dipole moment) to store charge (i.e. align) in the presence of an
externally applied electric field, relative to the permittivity of vacuum (e' = 1). If the polarity of
the applied field is alternating, such as is the case for electromagnetic waves, molecules rotate
in an attempt to maintain the alignment of their molecular dipoles with the alternating field. At
low frequencies, on the order of 106 Hz, the molecular realignment is able to match the
alternating field and the bulk medium continues to behave as an ideal dielectric. The dielectric
constant at these frequencies is at a maximum. At frequencies greater than 1012 Hz, the field is
alternating so fast as to have virtually no effect on the orientation of the molecules and their
91
molecular dipoles, and the dielectric constant describing the relative permittivity falls; no charge
is stored because the molecular dipoles never align with the rapidly alternating field. The bulk
medium at these high frequencies behaves as an insulator. At frequencies between 106 and 1012
Hz, the dielectric constant exhibits frequency dependence. The medium behaves as a non-ideal
dielectric because the orientation of the molecular dipoles is affected but cannot sync their reorientation with the frequency of the alternating electric field. Resistive heating occurs as the
stored energy of the aligned molecular dipoles fights back and forth. On a molecular level,
rotational excitations of one molecule transfer through molecular collisions to nearby molecules
as translational kinetic energy, or heat. The dielectric loss (e") of a medium describes the extent
to which this resistive heating occurs at a given frequency.
a.
b.
c.
Figure 20. Phase diagram relationship between dielectric constant, loss, and permittivity.
In an ideal dielectric, the current (I) is orthogonal to the applied voltage (E) (Figure 20 a.). In an
alternating electric field at low frequency, the stored charge in the system remains orthogonal
to the applied field, such that I = e' (Figure 20 b.). At higher frequency, dielectric loss (e") occurs
in opposition to the applied field such that the total relative permittivity (e*) has a component
in-phase with the electric field (E) (Figure 20 c ) . The loss angle (6) relative to the orthogonal
orientation of an ideal dielectric is defined as:
tan 6 = e" / e'
92
For a series of solvents with similar dielectric constants (e'), this loss tangent is commonly used
as an estimate of the relative ease with which they heat in a microwave field.
As stated previously, both the dielectric constant and the dielectric loss are dependent
on the frequency of the applied electric field. Measurement of dielectric properties is a
complicated process and requires specialized equipment depending on the frequency range to
be investigated. Measured values for water have been collected from a variety of sources and
plotted (Figure 21) to illustrate the frequency dependence.146147
Figure 21. Frequency dependence of the dielectric values for water at 25 C.
80 -r
70 -60 -50 -40
30 +
20
10 +
0
1E+7
- • — dielectric
constant
•
dielectric loss
• i
i i i 1111
1E+8
microwave frequency (Hz)
It can be seen that at low frequency, very little energy is lost as heat (the dielectric loss
approaches zero). As frequency is increased, more energy is lost as heat, until the maximum
dielectric loss is reached at around 18 GHz. Beyond 18 GHz, as the dielectric constant which
describes the ability of water to interact at all with the alternating field drops, so too does the
146
Kaatze, U. J. Chem. Eng. Data 1989,34, 371-374.
' Mattar, K.E.; Buckmaster, H.A.J. Phys. D:Appl. Phys. 1990,23,1454-1457.
93
dielectric loss. At the maximal dielectric loss, one can imagine the frequency of the field
synchronizing with the relaxation time of the rotational excited states.
Water has a very high dielectric constant (£' = 77.4 at 25 °C) but because the dipoles of
the tiny water molecules can very nearly match their re-orientation with a field alternating at
2.45 GHz, the dielectric loss of water is relatively low (e" = 9.48 at 25 °C). Ethanol on the other
hand, with a lower dielectric constant (e' = 24.3 at 25 °C), heats as well or faster than water
because the relaxation time is greater resulting in a greater dielectric loss (e" = 22.9 at 25 °C).
Microwave heating and the bench-top
chemist
Understanding the principles behind dielectric heating allows the scientist to make
justifications about real-world heating phenomenon, but the practical aspects are what affect
day-to-day research. A series of solvents were heated in order to rank their practical, relative
microwave absorptivity and heating efficiency (Table 24). In each case, 2 mL of solvent was
heated for 60 seconds with a constant 25 W as reported by the CEM Discover. The external
cooling was disabled so that the maximum temperature achieved could be reliably recorded by
the IR sensor. Since each sample started at room temperature, the energy converted to heat
was calculated as the product of the heat capacity, volume of solvent, and change in
temperature. The solvents were ordered from least to most efficient. Ethanol and
dichloromethane were the only two solvents that approached their respective boiling points,
where the heat of vaporization becomes a significant factor.
94
Table 24. Practical heating of solvents in a scientific microwave.
Solvent
toluene
dichloromethane
1,4-dioxane
acetonitrile
1,2-dichloroethane
chlorobenzene
o-dichlorobenzene
ethanol
water
Boiling
point
(°C)
110-111
39.8-40
100-102
81-82
83
132
178-180
78
100
Heat
capacity
(JmL'^K1)
1.46
1.60
1.74148
1.76
1.63
1.48
1.51150
1.92
4.18
Dielectric
constant (e0)
tan 6
2.38 at 25°C
0.040
0.042
9.08 at 20°C
9.84149 at 30 °C
0.062
37.5 at 20 °C
0.127
10.5 at 20 °C
5.69 at 293 K 0.101
2.21 151 at25°C 0.280
0.941
24.3 at 25 °C
78.5 at 25 °C
0.123
Max.
temp.
achieved
33 °C
35 °C
40 °C
43 °C
46 °C
50 °C
65 °C
77 °C
81 °C
Heat
Energy
23 J
32 J
52 J
63 J
68 J
74 J
120 J
200 J
470 J
Murthy, N.M.; Subrahmanyam, S.V. Indian J. PureAppl. Phys. 1979,17(9), 620-622.
' Sastry, N.V.; Dave, P.N. Proc. Indian Acad. Sci., Chem. Sci. 1997,109, 211.
' Narbutt, J. Z. Elektrochem. 1918, 24, 339-342.
1
Kinart, C M . ; Kinart, W.J.; Cwiklinska, A. J. Chem. Eng. Data 2002,47, 23.
95
Appendix II:
Sealed-vessel reactions
and kinetics
Heating solvent in a sealed
vessel
The boiling point (T) of a liquid varies with pressure (P) according to the ClausiusClapeyron equation:
d ln(P)
dT
=
AHvap
RT2
where AHvap is the enthalpy of vaporization and R is gas constant. Trouton's rule suggests that
the entropy of vaporization is approximately constant across a wide range of solvents:
_ ^Hvav _
j
u>vap - r
~ b b Imol • K
The entropy of vaporization of water is high, 109 J/mol-K, while that of methane is low, 73
J/molK. The combination of Trouton's rule and the integrated Clausius-Clapeyron equation
allows the estimation of the boiling point of a liquid at any pressure give a single temperaturepressure data point. This relationship has been elegantly executed as a Java applet calculator.152
A solvent heated in a sealed vessel will develop autogenic pressure. The consequence of
this autogenic pressure and the Clausius-Claperyon relationship is that, given enough liquid in a
sealed container, solvents can be easily heated above their atmospheric boiling point, a
condition often erroneously described as superheating. True superheating occurs in a liquid
carefully heated without disturbance or a nucleation point; eventually, a superheated solvent
will boil vigorously and rapidly cool to the normal boiling point.
152
Goodman, J.M.; Kirby, P.D.; Haustedt, L.O. TetrahedronLett. 2000,41, 9879-9882.
96
In an open container, heat applied initially causes an increase in temperature, until the
vapor pressure reaches the external pressure and boiling occurs. Further heat does not increase
the temperature, but rather vaporizes the liquid. In a sealed vessel, heat applied increases the
temperature and the vapor pressure. As the vapor pressure of the liquid, and thus the external
pressure on the liquid, increases, so too does the boiling point of the liquid. But by the time the
liquid has reached that theoretical boiling point, additional vapor pressure has developed. This
can continue until the ever increasing density of the vapor phase reaches that of the ever
decreasingly dense liquid phase at the critical point.
Heating reactions
in a sealed
vessel
As a general rule-of-thumb, an increase of 10 °C doubles the rate of a reaction; a
decrease of 10 °C will slow the rate of a reaction by one-half.
k = Ae~EalRT
Analyzing the Arrhenius equation, however, reveals there is also a dependence on the activation
energy (Ea) of the reaction. Ignoring the temperature dependence of the pre-exponential factor
(A), the effect of a 10 °C temperature increase on a theoretical series of reactions of varying
activation energy has been illustrated below (Figure 22).
97
Figure 22. Factor increase in rate constant given a 10 °C increase in temperature for various activation energies.
-100
-50
0
50
100
150
200
250
300
Reference initial reaction temperature (°C)
Reactions with very low activation energy experience dramatic temperature-dependent rate
changes at a very low initial temperatures, and the effect of increasing temperature is
diminished at higher temperature. Reactions with very high activation energy continue to
benefit from further increasing temperature. It can be seen that reactions of intermediate
activation energy (2 6 -2 7 kJ/mol) do experience an approximate doubling of the rate constant per
every 10 °C increase over the range 0-200 °C.
This effect of increased rate at increased temperature is further illustrated by
considering a theoretical reaction with an activation energy of 75 kJ/mol (Figure 23). The
reaction is initially run in refluxing ethanol (boiling point 78 °C). In a sealed vessel at 90°C, the
rate constant is doubled. For each additional 10 °C, the rate is again approximately doubled. At
150°C, a very reasonable temperature for an ethanoic reaction in a microwave-heated sealedvessel reactor, the reaction proceeds at a rate more than 80 times that at reflux.
98
Figure 23. Factor rate constant increase for a theoretical reaction (E3 = 75 kJ/mol) initially run in refluxing ethanol.
80 ]
,
70 C
/
60 -
/
o
50
/
40 -i
to
ai
/
30
/
20 -•
; /
10 <
o .4
80
^
t
_T_Z
90
100
^
^
^
!
1
110
120
( —.—,
130
140
1
150
New reaction temperature (°C)
It is worth repeating that these rate increases are calculated purely from the Arrhenius
equation, and are a consequence of the widely-recognized Maxwell-Boltzmann distribution of
molecules with a particular energy at a particular temperature. In summary, reactions in a
sealed vessel can be heated to a higher temperature and, as such, react faster because more
molecules possess the energy required to react. Hence, sealed vessel reactions possess the
potential to react faster than their reflux-limited counterparts.
99
Appendix III:
Tamoxifen
Tetrasubstituted olefins can be prepared by a large number of methods including olefin
metathesis and phosphorus ylide chemistry, but the selective formation of a single isomer is
often challenging. The challenges and solutions to this problem have been reviewed.153
Tamoxifen, the "gold standard" in the treatment and prevention of estrogen-receptorpositive breast cancer, is among the most important tetrasubstituted olefins ever synthesized.
Originally developed as an orally-available contraceptive, tamoxifen was a magnificent failure,
exhibiting little contraceptive efficacy in humans. Non-steroidal anti-estrogens like tamoxifen
were recognized to have applications as anticancer drugs, but most research in the area was
focusing on nonspecific cytotoxic chemotherapy, so it was overlooked for many years.
Eventually, the combined efficacy and mild side-effect profile of tamoxifen was recognized and
it became one of the most widely used anticancer drugs.154
The industrial scale synthesis of (Z)-tamoxifen as described in the U. S. patent involves
removal of the (E) isomers in the final step (Scheme 31).155 The described synthesis results in a
tertiary alcohol via the Grignard addition of phenyl magnesium bromide to a substituted
acetophenone. Upon acid-catalyzed dehydration, a mixture of isomers is produced. The desired
isomer can be isolated by fractional crystallization of the citrate salt. As a result, a large quantity
of the produced material is wasted late in the synthesis, prompting a plethora of stereoselective
syntheses in the chemical literature.
Flynn, A.B.; Ogilvie, W.W. Chem. Rev. 2007,107, 4698-4745.
Jordan, V.C. Nature Reviews 2003, 2, 205-213.
Harper, M.J.; Richardson, D.N.; Wapole, A.L Alkene Derivatives. U. S. Patent 4,536,516. Aug. 20,1985.
100
C0 2 H
^S ' CC°2H
HO^f
HCI
separation
EtOH, A
fractional
crystallization
CO,H
Scheme 31. Final steps in the industrial process to (Z)-tamoxifen.
An attractive "single-step" palladium-catalyzed synthesis of tetra-substituted olefins via
a three component coupling has been published and exploited in the synthesis of Tamoxifen
(Scheme 32).156 Several steps are invested in the synthesis of the aryl boronic acid, however,
which is used in large excess due to the competitive formation of the Suzuki coupling byproduct.
3eq.
3eq.
-O
1%PdCI2(PhCN)2
3 eq. K 2 C0 3
>%
B(OH)2
\0
I
\
2:1 DMF:H20
45°C, 24 h
68% yield, > 95% pure
removed by column
chromatography
Scheme 32. Larock "single-step" synthesis of Tamoxifen.
A synthesis of Tamoxifen was envisioned that would be regioselective, relying on a
highly efficient and rapid Suzuki coupling to yield the desired isomer of the final product
(Scheme 33). The regiochemistry was to be set using a previously reported trans-bromination of
an alkyne. A palladium-catalyzed Sonogashira coupling with a gaseous reagent provided the
alkyne. Williamson etherification installed the 2-(dimethylamino)ethyl ether.
Zhou, C ; Larock, R. C. J. Org. Chem. 2005, 70 (10), 3765-3777.
101
I
Suzuki
)
Sonogashira
OH
>
X
B(OH)2
Scheme 33. Envisioned retrosynthetic analysis of Tamoxifen.
Williamson
etherification
The first step in the envisioned synthesis of Tamoxifen was a Williamson etherification
of a 4-halophenol (Scheme 34). In an eight step synthesis of Tamoxifen with 36 % overall yield,
there is literature precedent to reflux the potassium phenolate with dimethylaminoethyl
chloride in toluene for 12 hours.157 The substitution reaction in acetone can be much more
expeditious when conducted at elevated temperatures.158 Experiments with 4-bromophenol,
cheaper and more readily available than 4-iodophenol, lead to conditions to produce the
desired ether in high yield.
OH
CI'
,NL
CX
KOH
acetone
~N'
|
Scheme 34. Williamson etherification of 4-iodophenol.
Ultimately, the process was found to be stirring-limited to the 5 mmol scale by the
precipitation of potassium chloride (in the CEM Discover). A free-base procedure was developed
to prepare a solution of dimethylaminoethyl chloride in acetone. The amine hydrochloride (10
mmol) was shaken in acetone (20 mL) with potassium hydroxide (10 mmol, assuming 15 %
water content) until the potassium hydroxide dissolved. Potassium chloride was filtered from
the resulting suspension under positive pressure to yield a solution of the free amine. This
solution was combined with 4-iodophenol (5 mmol) and potassium hydroxide (5 mmol) in the
Shimizu, K.; Takimoto, M.; M o r i , M.; Sato, Y. Synlett 2006,18, 3182-3184.
' Leadbeater, N.E.; Schmink, J.R. Tetrahedron 2007, 63, 6764-6773.
102
80-mL Discover vessel. The reaction was heated to 100 "C for 30 minutes, cooled, filtered and
evaporated to yield the crude aryl ether in quantitative conversion by ^-NMR. If the scale was
increased, the precipitation of potassium chloride during the run prevented efficient stirring and
the conversion decreased.
Sonogashira
reaction with
1-butyne
The Sonogashira reaction is a palladium-mediated coupling between a terminal alkyne
and an aryl or vinyl halide. Effective catalysts, conditions, and applications of the Sonogashira
reaction have been reviewed.159 There is literature precedent for the Sonogashira coupling on
an iodoarene containing the dimethylaminoethyl ether in a seven step synthesis of Tamoxifen
with an overall yield of 35.5 %.160
The Sonogashira method with 1-butyne was developed (Table 25) starting with copperfree conditions in piperidine using bistriphenylphosphine palladium (II) chloride. The addition of
copper(l) iodide was found to increased reproducibility. Acetonitrile was added as a solvent
when the solvent-free reaction was found to exotherm. Deactivated aryl iodides were found to
react slowly with phenylacetylene at room temperature. Simple non-conjugated alkynes were
found to be reactive only at elevated temperature. Temperatures above 70 °C caused the
palladium catalyst to precipitate as palladium black.
159
160
Chinchilla, R.; Najera, C. Chem. Rev. 2007,107, 874-922.
Tessier, P. E.; Penwell, A. J.; Souza, F. E. S.; Fallis, A. G. Org. Lett. 2003,5, 2989-2992.
103
Table 25. Development of conditions for the Sonogashira reaction with butyne.
R'
R
Aryl Iodide
Alkyne
J^
cx J^
0"
.-0"
^y
R
^
^JO
<0^
VX
-R'
Notes
Conditions
0.5 mol % PdCI2(PPh3)3
lmol%Cul
Piperidine
High conversion
Solvent-free, r.t.
Exotherm
0.5 mol % PdCI2(PPh3)3
1 mol % Cul
piperidine, CH3CN
High conversion
Slower rate at r.t.
0.5 mol % PdCI2(PPh3)3
1 mol % Cul
piperidine, CH3CN
Very slow reaction at r.t.
High conversion at 70 °C
0.5 mol % PdCI2(PPh3)3
1 mol % Cul
piperidine, CH3CN
No conversion at r.t.
High conversion at 70 °C
The crude iodide from the Williamson etherification could be used directly in the
Sonogashira coupling (Scheme 35). Piperidine (1.5 mL) and acetonitrile (2.5 mL) were added to
the residue from the 5 mmol reaction described previously and transferred to a 10-mL tube.
Butyne (1.08 g) was condensed into the solution in a dry ice and acetonitrile bath. Finally,
bistriphenylphosphinepalladium (II) chloride (16.6 mg, 0.5 mol %) and copper (I) iodide (10.6 mg,
1 mol %) were added. The vessel was sealed and pressurized with nitrogen (100 psi), then
heated to 70 °C for 30 minutes.Upon cooling, the pressure was released. As the vessel was
removed from the microwave, the reaction mixture began to crystallize and degas, expanding
out of the tube. Crude 1H-NMR analysis indicated quantitative conversion.
Pd, Cu
,N.
piperidine
CH3CN
Scheme 35. Sonogashira reaction with 1-butyne
104
Trans-bromination
of an alkyne
A literature procedure for the dibromination of alkynes in phase-transfer agent reports
that the (E)-selectivity arises from the viscosity of the ionic liquid, however, the bromide salt
yields high-selectivity, whereas the hexafluorophosphate salt lead to isomeric mixtures.161 The
bromide anion can be invoked as a participant in the bromination to better explain the
selectivity.162 In the absence of phase-transfer agents in cold ethanol with lithium bromide, we
found high selectivity for the (E)-brominated product.163
The 2D-NOESY spectrum provided evidence for the trans-bromination. A correlation
could be clearly observed between the -OCH2- and the nearby aromatic protons, but no
correlation was seen for the vinyl -CH2- and the aromatic protons. Had the product been cisbrominated, a correlation would have been expected.
Double Suzuki coupling with a vinyl
dibromide
Although Suzuki couplings with 3-bromostyrenes are well-know, there is very little
precedent for couplings with a-bromostyrenes, and even fewer reports of dual couplings on ct,3dibromoethylenes. One report of 1,2-dibromoethylene reacting with phenyl boronic acid
indicated a 20 % yield of products in a 36 : 64 ratio of stilbene (the desired double Suzuki
product) to biphenyl (from the homo-coupled phenylboronic acid).164
The model study with the (E)-2,3-dibromo-3-phenylprop-2-ene indicated a trace of the
desired double Suzuki coupled product. Unfortunately, oxidative insertion of palladium into
either carbon-bromide bond results in rapid elimination to generate the alkyne (Scheme 36).
Chiappe, C; Capraro, D.; Conte, V.; Pieraccini, D. Org. Lett. 2001,3,1061-1063.
Berthelot, J. Can. J. Chem. 1986, 64, 603-607.
Schmink, J.R. Ph.D. thesis, 2010.
Organ, M.G.; Ghasemi, H.; Valente, C. Tetrahedron 2004, 60, 9453-9461.
105
Br.
^Br
Scheme 36. Palladium-catalyzed elimination of the vinyl dibromide to regenerate the alkyne.
Double Suzuki coupling... the other way?
Lithium-halogen exchange was attempted on the dibromide. If the dilithium species
could be formed, perhaps then the diboronic ester could be prepared (Scheme 37). Slow
addition of a solution of the dibromide substrate (5 ml_, 0.2 M in 3:2 ethenpentane) to a
solution of t-butyllithium (1.26 mL of a 3.50 M solution in heptane, diluted in 5 mL 3:2
ethenpentane ) at -78 °C, followed by a methanol quench, lead exclusively to the eliminated
alkyne product, indicating no dilithium had be formed.
(RO) 2 B.
.N
Br
-*
"-~
/N
" "^
(T^T
^Li
*
/N
B(OR)2
Scheme 37. Alternative double-Suzuki synthetic route toward Tamoxifen.
Although the initial studies with a test substrate had indicated traces of the doubleSuzuki coupled product, Tamoxifen was never observed. Given that he challenges of attempting
Suzuki couplings on vicinyl olefinic halogens had already been reported, this should not have
come as a surprise. Literature reports of Negishi couplings on 1,2-dibromoethylene show much
more promising results,165166167 but the coupling with phenylzinc chloride was not attempted.
' Carpita, A.; Rossi, R. Tetrahedron Lett, 1986, 27 (36), 4351-4354.
' Andreini, B. P.; Benetti, M.; Carpita, A.; Rossi, R. Tetrahedron, 1987, 43 (20), 4591-4600.
' Koga, N.; Matsumura, M.; Noro, M.; Iwamura, H. Chem. Lett, 1991,8,1357-1360.
106
Appendix IV:
Macrolactone support
The following reactions were executed in support of a palladium-catalyzed carbonylative
lactonization project. The conditions for the lactonizations were based on conditions developed
from an attempted carbonylative Sonogashira reaction as described in latter part of the
carbonylation chapter, which was, in fact, developed during the Tamoxifen project.
Manganese (IV) oxide was prepared by literature procedure from manganese sulfate
and potassium permanganate.168 The yield was greater than quantitative (by mass), because the
material was not dried, so as to retain the highest activity.
Triethyl ammonium formate (75.1 g) was isolated as the constant-boiling middle
fraction (58 °C under oil pump vacuum) from a formic acid (88 %, 50 mL) and triethylamine (65
mL) mixture.
2-iodobenzyl alcohol (Scheme 38, step 1) was prepared as follows: trimethylborane (20
mL) was added slowly to a solution of 2-iodobenzoic acid (15.9 g, 64 mmol) in anhydrous
tetrahydrafuran (40 mL) at 0 °C under nitrogen and stirred for 20 minutes. Borane dimethyl
sulfide complex (7 mL, ~10 M) was added drop-wise over one hour at 0 °C. The solution was
then allowed to warm to room temperature over 1 hr. Methanol was slowly added until no
further bubbling was observed. The solution was concentrated under vacuum. Additional
methanol was added and evaporated until a constant mass of 2-iodobenzyl alcohol (14.93 g, 99
% isolated yield) was reached. This procedure was based on a literature report.169
Vogel, A. I. Vogel's Textbook of practical organic chemistry, 5 ed. Harlow, England: Longman Group,
1989, p. 445.
169
Lane, C. F.; Myatt, H. L; Daniels, J.; Hopps, H. B. J. Org. Chem. 1974, 39 (20), 3052-3054.
107
1.BH3, Me2S
2. Mn0 2
/-^/l
^v^o
Scheme 38. Preparation of 2-iodobenzyl alcohol in two steps.
2-iodobenzaldehyde (Scheme 38, step 2) was prepared as follows:
manganese (IV)
oxide (25 g) was added to a solution of 2-iodobenzyl alcohol (5.20 g) in chloroform (130 mL) and
stirred for 24 hr. The mixture was centrifuged, the supernatant was decanted, filtered through
Celite and concentrated to yield 2-iodobenzaldehyde (3.79 g, 73 % isolated yield). This
procedure was based on a literature report.170
O
+
O
Scheme 39. Preparation of two iodoaryl acids from Meldrum's acid and o-iodobenzaldehyde.
3-(2-iodophenyl)-propanoic acid (Scheme 39) was prepared as follows: A solution of 2iodobenzaldehyde (3.79 g) in triethylammonium formate was heated to 100 °C. Meldrum's acid
(2.30 g) was added. Vigorous bubbling immediately occurred. The solution was maintained at
100 °C for 75 minutes then was cooled. Ice (15 g) was added then the pH was adjusted to 1 with
6 N HCI. The product 3-(2-iodophenyl)-propanoic acid (2.43 g) crystallized and was isolated by
filtration. The remaining crude product was extracted with dichloromethane and evaporated
onto silica gel (2 g) and loaded onto a silica gel column (10 g). Remaining aldehyde was eluted
with 5 % EtOAc/petroleum ether. Additional 3-(2-iodophenyl)-propanoic acid (0.70 g) was eluted
with 20 % EtOAc/petroleum ether. Methanol eluted a second product, 3-(2-iodophenyl)-
170
Olivera, R.; SanMartin, R.; Domfnguez, E.; Solans, X.; Urtiaga, M. K.; Arriortua, M. \.;J. Org. Chem. 2000,
65 (20), 6398-6411.
108
pentanedioic acid (0.54 g). This procedure was adapted from a literature report, so that it could
be monitored via ReactlR. 171
C0 2 H
^ / \ / C H
2
O H
Scheme 40. Preparation of 3-(2-iodophenyl)-propanol and proposed carbonylative lactonization.
3-(2-iodophenyl)-propanol (Scheme 40) was prepared as follows: To a tetrahydrafuran
(14 mL) solution of 3-(2-iodophenyl)-propanoic acid (6.21 g, 22.5 mmol) in an ice bath was
slowly added trimethylborane (7.0 mL). After stirring for 20 minutes, borane dimethyl sulfide
complex (2.5 mL, ~10 M) was added drop-wise as bubbling ceased. Work-up and isolation were
as per 2-iodobenzyl alcohol. The product 3-(2-iodophenyl)-propanol was isolated (5.52g, 94 %)
as a low melting solid. The second product isolated from the reaction of Meldrum's acid with oiodobenzaldehyde in triethylammonium formate is also proposed as a substrate for successive
reduction to the diol and carbonylative lactonization (Scheme 41).
,CH2OH
I
^
"I
Scheme 41. Proposed reduction of, 3-(2-iodophenyl)-pentanedioic acid and carbonylative lactonization.
The carbonylative lactonization reaction optimization was done with 3-(2-iodophenyl)propanol, to yield the seven-membered lactone, but there was interest in increasing the ring
size to test the limits of the method. The substrate for an eight-membered lactone was
prepared (Scheme 42). The first three steps of the sequence were reproduced from a literature
report.172 The next two steps were adapted from a separate report.173 A one gram sample of the
crude iodide product (4.31 g total) was purified by silica gel chromatography (40 g, 10 % ethyl
Toth, G.; Kover, K.E. Synthetic Commun. 1995,25 (19), 3067-3074.
Nieduzak, T. R.; Boyer, F. E. Synthetic Comm. 1996, 26(18), 3443-3452.
1
Mulbaier, M.; Giannis, A. Angew. Chem. Int. Ed. 2001,40 (23), 4393-4394.
1
109
ether with 5 % acetic acid in petroleum ether, Rf = 0.77, 0.82 g isolated). The reduction to 4-(2iodophenyl)-l-butanol proceeded as described previously for 2-iodobenzyl alcohol.
C0 2 H
NOH
92%
79%
80%
H,N
C0 2 H
CH2OH
72%
52%
from silica
Scheme 42. Sequence to prepare 4-(2-iodophenyl)-l-butanol for the proposed lactonization.
A second series of substrates were investigated. Beginning with o-iodophenol, a
Williamson etherification with ethyl bromoacetate174 (isolated yield of 73 % following silica
chromatography with 5 % ethyl acetate in petroleum ether) can be followed by a reduction of
the ester175 to the terminal alcohol (41 % isolated) (Scheme 43).
0.
OEt
OH
THF, K2C03
-
^
S -I
C
°2Et
DMS
I
OH
Scheme 43. Williamson synthesis of an iodoaryl ether substrate for the lactonization.
The one-carbon homologated case produced no desired product from the attempted
etherification (Scheme 44). Elimination of ethyl acrylate likely occurs under the basic reaction
conditions.
I
Br
O
OEt
OH
THF, K 2 C0 3
^C02Et
OEt
not observed
Scheme 44. Failed Williamson etherification.
1
Portscheller, J. L; Malinakova, H. C. Org. Lett. 2002,4 (21), 3679-3681.
Brown, H. C; Choi, Y. M. Synthesis 1981, 6,439-440.
110
Additional substrates have been prepared alongside the carbonylative lactonizations by
a separate chemist. Initial findings indicate that seven-membered lactones can be readily
formed by the method. Substrates for eight-membered or larger lactones yield multiple
products, which may be dimers or various other oligomers.
Ill
Appendix V:
Spectral Data
Phenols from iodo- and
bromoarenes
N-(4-acetylphenyl)pyrrolidine.
O
*H-NMR (CDCI3, 500 MHz): 6 7.85 (d, 2H, J=8.8 Hz), 6.48 (2, 2H, J=8.8 Hz),
3.31 (t, 4H, J=6.1 Hz), 2.49 (s, 3H), 2.01 (t, 4H, J=6.1 Hz).
^
O
N-(4-acetylphenyl)piperdine.
O
f l ] ^
A ^ J
*H-NMR (CDCI3, 500 MHz): 6 7.82 (d, 2H, J=8.8 Hz), 6.80 (d, 2H, J=8.8 Hz),
3.31 (t, 4H, J=4.9 Hz), 2.46 (s, 3H), 1.62 (m, 6H).
N-(4-nitrophenyl)pyrrolidine.
,N0 2
/ ^ H
1
HH-NMR (CDCI3, 300 MHz): 6 8.07 (d, 2H, J=9.4 Hz), 6.43 (d, 2H, J=9.3 Hz),
3.38 (t, 4H, J=6.5 Hz), 2.07 (t, 4H, J=6.4 Hz).
^ ^
N-(4-nitrophenyl)piperidine
N0 2
.
*N' ^ ^
^-NMR (CDCI3, 300 MHz): 6 8.08 (d, 2H, J=9.4 Hz), 6.78 (d, 2H, J=9.4
Hz), 3.49-3.38 (m, 4H), 1.74-1.64 (m, 6H).
4'-hydroxyacetophenone.
O
^-NMR (CDCI3, 400 MHz): 6 8.28 (bs, 1H), 7.90 (d, 2H, J=8.8 Hz), 6.93 (d, 2H,
J=8.8Hz), 2.56 (s, 3H).
HO'
^
112
p-cresol
CH 3
^f'
HH-NMR (CDCI3, 400 MHz): 6 6.97 (d, 2H, J=7.8 Hz), 6.71 (d, 2H, J=8.0 Hz),
HCT ^
2.25 (s,3H)
2-nitrophenol.
a
2
(CDCI3, 300 MHz): 5 8.11 (d, 1H, J=9.0 Hz), 7.58 (t, 1H, J=9.0 Hz), 7.16 (d,
1H, J=9.0 Hz), 6.99 (t, 1H, J=7.5 Hz).
HH-NMR
4-nitrophenol.
X
H-NMR (CDCI3, 500 MHz): 6 8.21 (d, 2H, J=9.1 Hz), 6.96 (d, 2H, J=9.1 Hz),
6.07 (bs, 1H).
N0 2
13
C-NMR: 6 161.4, 141.7, 126.3, 115.7.
|R (dry film): 3315 (broad), 1587,1320,1165,1110 cm"1.
HO" ^ ^
TOF/MS/ES-: expected 139.0197, found 138.0194.
4-methoxyphenol.
.OCH3
XT
X
H-NMR (CDCI3, 400 MHz): 6 6.75 (s, 4H), 3.72 (s, 3H).
2-methoxyphenol.
^ w O C H
[I
T
3
*H-NMR (CDCI3, 400 MHz): 6 6.94-6.69 (m, 4 H), 385 (s,3H).
113
Carbonylation
of aryl
iodides
benzoic acid.
„C0 2 H
iH.NMR
(CDC|3/ 4 0 0 M H z
) : 5 8 03 (bs, 1H), 7.98 (d, 2H, J=7.4 Hz), 7.47 (t, 1H,
J=7.4 Hz), 7.35 (t, 2H, J=7.4 Hz).
4-methyI benzoic acid.
-C0 2 H
i H . N M R ( CDC | 3/ 400 MHz): 6 7.89 (d, 2H, J=8.1 Hz), 7.79 (bs, 1H), 7.23 (d, 2H,
J=7.9 Hz), 2.39 (s, 3H).
2-methylbenzoic acid.
a
C
°2H
^-NMR (CDCI3, 400 MHz): 6 11.06 (bs, 1H), 8.07 (dd, 1H, Ja=7.9 Hz, Jb=1.6 Hz),
7.45 (td, 1H, Ja=7.4 Hz, Jb=1.4 Hz), 7.32-7.22 (m, 2H), 2.67 (s, 3H).
4-nitrobenzoic acid.
X02H
1
H-NMR (dg-DMSO, 400 MHz): 6 8.32 (d, 2H, J=8.7 Hz), 8.17 (d, 2H, J=8.8
Hz).
02N
4-aminobenzoic acid.
X02H
u
M
. „ ~
,
*H-NMR (d6-DMSO, 500 MHz): 6 12.1 (bs, 1H), 7.63 (d, 2H, J=8.5 Hz), 6.55
(d, 2H, J=8.4 Hz), 5.72 (bs, 2H).
M2N
4-acetyI benzoic acid.
^ ^ C 0
Jj
J
2
H
^-NMR (CDCI3, 400 MHz): 5 8.08 (d, 2H, J=8.4 Hz), 7.97 (d, 2H, J=8.2 Hz),
2.61 (3H, s).
Y ^ ^
O
4-methoxybenzoic acid.
\
^ ^ ,C0 2 H
f ^ f
J l ^
t
X
H-NMR (CDCI3, 500 MHz): 6 10.4 (bs, 1H), 8.03 (d, 2H, J=8.9 Hz), 6.94 (d,
2H, J=8.9 Hz, 3.87 (s, 3H).
2-methoxybenzoic acid.
a
C0 2 H
OCH3
^H-NMR (CDCI3, 400 MHz): 6 9.78 (bs, 1H), 8.13 (dd, 1H, Ja=7.8 Hz, Jb=1.8 Hz),
7.55 (td, 1H, Ja=7.8 Hz, Jb=1.8 Hz), 7.10 (td, 1H, Ja=7.5 Hz, Jb=0.9 Hz), 7.05 (d, 1H,
J=8.4 Hz), 4.04 (s, 3H).
114
4-fluorobenzoic acid.
^-NMR (de-DMSO, 400 MHz): 6 7.31 (t, 2H, J=8.8 Hz), 8.01 (dd, 2H, Ja=8.5
Hz, Jb=5.8 Hz), 13.0 (bs, 1H).
^ ^ /
C
° 2
H
13
C-NMR: 6 116.1 (d, J=22.0 Hz), 127.8 (d, J=2.7 Hz), 132.6 (d, J=9.4 Hz),
J I J
165.3 (d, J=259 Hz), 166.4,166.8.
IR (dry film): 2900 (broad), 1672,1428,1293,1159, 854 cm"1.
TOF/MS/ES-: expected 139.0201, found 139.0205.
4-(trifluoromethyl)benzoic acid.
X02H
'H-NMR (CDCI3, 400 MHz): 5 8.17 (d, 2H, J=8.2 Hz), 7.70 (d, 2H, J=8.2 Hz).
F3C"
nicotinic acid.
^^/C0
II
2
H
iH_NMR
J
N
(CDC|^ 4Q0 M H Z
gg
J.
i g
^
1H)/ 8
79 (d#
1H#
j
= 3
3 Hz)/
8
32 (dt, 1H,
Ja=7.9 Hz, Jb=1.4 Hz), 7.49 (dd, 1H, Ja=7.7 Hz, Jb=4.8 Hz).
terephthalic acid.
^^.C0
^-NMR (d6-DMSO, 400 MHz): 6 8.05 (s, 4H).
i3c.NMR. g
1671/13A9/1299
H
2
H O cr-"-^
IR
(drv f i l m ) : 2827 (broad)' 2 5 5 5 < 1 6 8 2 - 1 4 2 5 ' 1 2 8 3
cm_1
-
TOF/MS/ES-: expected 165.0193, found 165.0187.
ethyl benzoate.
1
i,,^ ^ r ^1
CO Et
" ^ 22
1H
" N M R ( C D C | 3 ' 400 MHz):
6 8.05 (d, 2H, J=7.1 Hz), 7.54 (t, 1H, J=7.5 Hz), 7.43
(t, 2H,J=7.4Hz),4.38(q, 2H, J=7.1
Hz), 1.40 (t, 3H,J=7.1Hz).
13
^ ^
C-NMR: 6 166.7,132.9,130.6, 129.6, 128.4, 61.0, 14.4.
isopropyl benzoate.
C0 2 iPr
i H . N M R ( CDC | 3j 3 0 0
MHz
) . g g 02 (d, 2H, J=7.7 Hz), 7.51 (t, 1H J=7.4 Hz), 7.40
(t, 2H, J=7.6 Hz), 5.24 (heptet, 1H, J=6.3 Hz), 1.35 (d, 6H, J=6.3 Hz).
ethyl 4-acetylbenzoate.
X02Et
^-NMR (CDCI3, 400 MHz): 6 8.11 (2H, d, J=8.2 Hz), 7.99 (d, 2H, J=8.2 Hz),
4.40 (q, 2H, J=7.1 Hz), 2.64 (s, 3H), 1.41 (t, 3H, J=7.1 Hz).
13
C-NMR: 6 197.7,165.8,140.3, 134.4,129.9, 128.3, 61.6, 27.0, 14.4.
IR (dry film): 2993,1705, 1682,1278,1129, 1014, 855 cm'1.
GC/MS: m/z (% base peak): 192 (26), 177 (100), 149 (38), 147 (46).
115
isopropyl 4-acetylbenzoate.
,C0 2 iPr
^-NMR (CDCI3, 300 MHz): 6 8.08, (d, 2H, J=8.5 Hz), 7.96, (d, 2H, J=8.4
Hz), 5.23 (heptet, 1H, J=6.3 Hz), 2.15 (s, 3H), 1.35 (d, 6H, J=6.3 Hz).
ethyl 2-methylbenzoate.
X
XQ2Et
H-NMR (CDCI3, 400 MHz): 6 7.94 (dd, 1H, Ja=8.3 Hz, Jb=1.6 Hz), 7.42-7.38 (m,
I H ) , 7.28-7.24 (m, 2H), 4.39 (q, 2H, J=7.1 Hz), 2.64 (s, 3H), 1.42 (t, 3H, J=7.1
Hz).
13
C-NMR: 6 167.7,140.0, 131.8, 131.7, 130.5,130.0, 125.7, 60.7, 21.7, 14.4.
isopropyl 2-methylbenzoate.
2
X02iPr
H-NMR (CDCI3, 400 MHz): 6 7.91 (dd, IH, Ja=8.4 Hz, Jb=1.5 Hz), 7.43-7.39 (m,
I H ) , 7.29-7.25 (m, 2H), 5.28 (heptet, IH, J=6.3 Hz), 2.62 (s, 3H), 1.40 (d, 6H,
J=6.3 Hz).
13
C-NMR: 6 167.5,139.9, 131.8,131.7, 130.7,130.5, 125.8, 68.3, 22.13, 21.8.
ethyl 4-methylbenzoate.
1 H
CO Et
"NMR
(CDCI3, 400 MHz): 6 7.93 (d, 2H, J=8.2 Hz), 7.19 (d, 2H, J=8.4 Hz),
4.34 (q, 2H,J=7.1Hz), 2.36 (s, 3H), 1.37 (t, 3H,J=7.2Hz).
2
13
C-NMR: 6 166.5, 143.2, 129.5, 128.9,127.7, 60.6, 21.4,14.2.
ethyl 4-methoxybenzoate.
COoEt
XX
H-NMR (CDCI3, 400 MHz): 6 7.95 (d, 2H, J=8.8 Hz), 6.85 (d, 2H, J=8.8 Hz),
4.30 (q, 2H, J=7.2 Hz), 3.78 (s, 3H), 1.33 (t, 3H, J=7.1 Hz).
13
C-NMR: 6 166.2, 163.2, 131.4,122.9,113.5, 60.5, 55.2, 14.3.
isopropyl 4-methoxybenzoate.
C0 2 iPr
^-NMR (CDCI3, 400 MHz): 6 7.99 (d, 2H, J=8.9 Hz), 6.90 (d, 2H, J=8.9
Hz), 5.23 (heptet, I H , J=6.2 Hz), 3.83 (s, 3H), 1.35 (d, 6H, J=6.3 Hz).
13,
C-NMR: 6 165.8, 163.2,131.5, 123.4, 113.5, 67.9, 55.3, 22.0.
ethyl 2-methoxybenzoate.
a
X
H-NMR (CDCI3, 400 MHz): 6 7.78 (dd, IH, Ja=7.9 Hz, Jb=1.8 Hz), 7.47-7.42 (m,
C0 2 Et
1H)( 6 99_6
95 ( m / 2H), 4.35 (q, 2H, J=7.1 Hz), 3.89 (s, 3H), 1.37 (t, 3H, J=7.1
13
C-NMR: 6 166.2, 159.1,133.3, 131.4,120.5,120.1,112.1, 60.8, 56.0,14.3.
116
isopropyl 2-methoxybenzoate.
^-NMR (CDCI3, 400 MHz): 6 7.77 (dd, 1H, Ja=7.9 Hz, Jb=1.8 Hz), 7.49-7.44 (m,
1 H ) ; 701-6.97 (m, 2H), 5.27 (heptet, 1H, J=6.3 Hz), 3.92 (s, 3H), 1.38 (d, 6H,
J=6.2 Hz).
C0 2 iPr
O
13
C-NMR: 6 165.8,159.2, 133.2,131.4, 121.1, 120.2, 112.2, 68.2, 56.1, 22.0.
ethyl 4-fluorobenzoate.
S
2
V'
^-NMR (CDCI3, 300 MHz): 6 8.06 (dd, 2H, Ja=8.9 Hz, Jb=5.5 Hz), 7.07 (t, 2H,
Ja=8.7 Hz), 4.37 (q, 2H, 7.1 Hz), 1.39 (t, 3H, J=7.1 Hz)
ethyl 4-(trifluormethyl)benzoate.
^ .
.C0 2 Et
(I ^ f
,
X
\ T
H-NMR (CDCI3, 300 MHz): 6 8.16 (d, 2H, J=8.4 Hz), 7.70 (d, 2H, J=8.5 Hz),
,J<^>
F,(T
4.42 (q, 2H, J=7.2 Hz), 1.42 (t, 3H, J=7.2 Hz).
ethyl 4-nitrobenzoate
X02Et
1
H-NMR (CDCI3, 300 MHz): 6 8.23 (d, 2H, J=9.0 Hz), 8.15 (d, 2H, J=9.0
Hz), 4.38 (q, 2H, J=7.1 Hz), 1.38 (t, 3H, J=7.1 Hz).
„ ^„
02N
^-^
nicotinic acid ethyl ester.
/^.C02Et
(j
T
^N
Vl-NMR (CDCI3, 300 MHz): 6 9.17 (d, 1H, J=2.1 Hz), 8.73 (dd, 1H, Ja=4.9 Hz,
Jb=1.7 Hz), 8.33 (dt, 1H, Ja=7.9 Hz, Jb=1.9 Hz), 7.45 (ddt, 1H, Ja=8.0 Hz, Jb=4.9
Hz, Jc=0.7 Hz) 4.42 (q, 2H, J=7.1 Hz), 1.41 (t, 3H, J= 7.1 Hz).
t-butyl benzoate.
C02tBu
i H . N M R (CDC\3, 400 MHz): 6 7.97 (d, 2H, J=7.7 Hz), 7.51 (t, 1H, J=7.2 Hz), 7.41
(t, 2H,J=7.1Hz), 1.58 (s, 9H).
diethyl terephthalate.
rj^Y"
2
X
^
H-NMR (CDCI3, 400 MHz): 6 8.13 (s, 4H), 4.40 (q, 2H,J=7.1Hz), 1.42 (t,
B02CXJ
3H,J=7.1Hz).
ethyl 4-bromobenzoate.
X02Et
_
^-NMR (CDCI3, 400 MHz): 5 7.92 (d, 2H, J=8.4 Hz), 7.58 d, 2H, J=8.5 Hz),
4.40 (q, 2H, J=7.1 Hz), 1.42 (t, 3H, J=7.1 Hz).
Br'
117
4-methoxydiphenylacetylene.
X
H-NMR (CDCI3/ 300 MHz): 6 5.57-7.50 (m, 2H), 7.50 (d, 2H, J=9.0
Hz), 7.40-7.31 (m, 3H), 6.90 (d, 2H, J=8.9 Hz), 3.85 (s, 3H).
l-(4-methoxyphenyl)-2-(piperidin-l-yl)ethane-l,2-dione.
O
N^ J
*H-NMR (CDCI3, 300 MHz): 6 7.85 (d, 2H, J=8.7 Hz), 6.87 (d, 2H,
J=8.7 Hz), 3.84 (s, 3H), 3.36-3.15 (m, 4H), 1.81-1.48 (m, 6H).
N-benzyl-4-methoxybenzamide.
*H-NMR (CDCI3, 300 MHz): 6 7.77 (d, 2H, J=8.9 Hz), 7.41-7.21 (m, 5H),
6.92 (d, 2H, J=8.9 Hz), 6.45 (bs, 1H), 4.63 (d, 2H, J=5.7 Hz), 3.86 (s, 3H).
13
C-NMR: 6 166.9, 162.2, 138.4, 128.8, 128.7, 127.9, 127.5, 127.4, 113.8,
55.4, 44.1.
IR (dry film): 3254 (broad), 1630,1605,1249,1177,1029, 842 cm"1.
N-benzyl-2-(4-methoxyphenyl)-2-oxoacetamide.
*H-NMR (CDCI3, 500 MHz): 6 8.48 (d, 2H, J=9.0 Hz), 7.46 (bs, 1H),
7.41-7.32 (m, 5H), 6.98 (d, 2H, J=9.1 Hz), 4.59 (d, 2H, J=6.1 Hz), 3.92
(s, 3H).
13
C-NMR (100 MHz): 6 185.8, 164.9, 162.5, 137.5, 134.1, 129.0,
128.0, 127.9, 126.6, 114.0, 55.7, 43.5.
TOF/MS/ES+: 270.1065 (M+l), 136.0458, 135.0263.
118
Oxidation of 1,4-dihydropyridines
with molecular
oxygen
diethyl 2,6-dimethyl-4-phenyl-l,4-dihydro-3,5-pyridinedicarboxylate.
^ - N M R (CDCI3, 500 MHz): 6 1.23-1.26 (m, 6H), 2.34 (s, 6H), 4.084.15 (m, 4H), 5.01 (s, 1H), 5.75 (bs; 1H), 7.14 (t, 1H, J=6.7 Hz), 7.23 (t,
2H, J=7.5 Hz), 7.30 (d, 2H, J=8.0 Hz).
O
EtO
13
OEt
C-NMR: 6 14.3, 19.6, 39.6, 59.7, 104.2, 126.1, 127.8, 128.0, 143.9,
147.8,167.7.
IR (dry film): 2981 (broad), 1686,1648,1552,1486,1297,1208 cm"1.
TOF/MS/ES+: m/z expected 330.1700, found 330.1705.
diethyl 2,6-dimethyl-4-cyclohexyl-l,4-dihydro-3,5-pyridinedicarboxylate.
EtO
OEt
^-NMR (CDCI3, 500 MHz): 6 0.93-1.30 (m, 6H), 1.32 (t, 6H, J=7.1 Hz),
1.56-1.80 (m, 5H), 2.32 (s, 6H), 3.95 (d, 1H, J=5.7 Hz), 4.13-4.27 (m,
4H), 5.61 (bs, 1H).
13
C-NMR: 6 14.4,19.5, 26.6, 26.7, 28.8, 38.4, 45.8, 59.6, 102.0,144.4,
168.7.
3,3,6,6-tetramethyl-9-phenyl-2,4,5,7,9,10-hexahydro-l,8-acridinedione.
*H-NMR (CDCI3, 400 MHz): 6 1.10 (s, 6H), 1.24 (s, 6H), 2.29-2.49 (m,
8H), 5.54 (s, 1H), 7.09 (d, 2H, J=7.8 Hz), 7.17 (t, 1H, J=7.1 Hz), 7.26 (m,
2H), 11.90 (bs, 1H).
13,
C-NMR: 6 27.4, 29.7, 31.4, 32.8, 46.5, 47.1, 115.6, 125.8, 126.8,
128.2,138.1, 190.4.
3,3,6,6-tetramethyl-9-cyclohexyl-2,4,5,7,9,10-hexahydro-l,8-acridinedione.
^-NMR (CDCI3, 300 MHz): 6 0.76 (q, 2H, J=10.9 Hz), 1.07 (s, 6H), 1.09
(s, 6H), 1.16-1.35 (m, 3H), 1.58-1.75 (m, 5H), 2.20-2.40 (m, 8H), 2.492.68 (m, 1H), 3.60 (d, 1H, J = l l . l Hz), 11.5 (bs, 1H).
3,3,6,6-tetramethyl-9-isopropyl-2,4,5,7,9,10-hexahydro-l,8-acridinedione.
oV
o
X
H-NMR (CDCI3, 300 MHz): 6 0.85 (d, 6H, J=6.8 Hz), 1.07 (s, 6H), 1.10
(s, 6H), 2.24-2.38 (m, 8H), 2.93-2.98 (m, 1H), 3.48 (d, 1H, J = l l . l Hz),
11.5 (bs, 1H).
119
diethyl 2,6-dimethyl-4-hexyl-l,4-dihydro-3,5-pyridinedicarboxylate.
EtO
OEt
^-NMR (CDCI3, 300 MHz): 6 0.86 (t, 3H, J=7.0 Hz), 1.15-1.35 (m,
14H), 2.30 (s, 6H), 3.94 (t, 1H, J=5.8 Hz), 4.04-4.29 (m, 4H), 5.53 (bs,
1H).
2,6-dimethyl-4-phenyl-l,4-dihydropyridine-3,5-dicarboxyanilide.
O
Ph
PhHNT ^
^-NMR (d6-DMS0, 400 MHz): 6 2.09 (s, 6H), 5.10 (s, 1H), 6.96 (t,
2H, J=7.3 Hz), 7.08 (t, 1H, J=6.8 Hz), 7.18-7.24 (m, 8H), 7.54 (d,
4H, J=7.9 Hz), 8.05 (s,lH), 9.28 (s,2H).
O
^T
^NHPh
13
C-NMR: 6 17.3, 42.0,105.7, 119.4, 122.6, 125.9,127.1, 128.0,
128.4, 137.8,139.5,167.4.
diethyl 2,6-dimethyl-4-phenyl-3,5-pyridinedicarboxylate.
'H-NMR (CDCI3, 500 MHz): 6 0.91 (t, 6H, J=7.1 Hz), 2.62 (s, 6H), 4.01
(q, 4H, J=7.2 Hz), 7.24-7.38 (m, 5H).
13
EtO
OEt
C-NMR: 6 14.3, 19.6, 39.6, 59.7, 104.2, 126.1, 127.8, 128.0, 143.9,
147.8,167.7.
IR (dry film): 1649,1487,1298,1209,1090,1019 cm"1.
TOF/MS/ES+: m/z expected 328.1543, found 328.1554.
diethyl 2,6-dimethyl-3,5-pyridinedicarboxylate.
9
9
*H-NMR (CDCI3, 500 MHz): 6 1.41 (t, 6H, J=7.1 Hz), 2.85 (s, 6H), 4.40
(q 4 H J = ? 2 HZ) 8 6 9 ( 1 H S)-
EtO^\|^Y"Et
/ ^
N
'
'
'
' '
'
13
^ \
C-NMR: 6 14.3, 24.4, 61.5,123.4,141.2, 162.1, 165.8.
diethyl 2,6-dimethyl-4-hexyl-3,5-pyridinedicarboxylate.
^-NMR (CDCI3, 300 MHz): 6 0.86 (m, 3H), 1.40 (t, 3H, J=7.1 Hz), 1.101.40 (m, 6H), 2.83 (s, 6H), 3.98 (t, 2H, J=6.6 Hz), 4.38 (q, 4H, J=7.1 Hz).
OEt
EtO
C-NMR: 6 13.9,14.2, 22.2, 22.4, 24.5, 27.3, 28.0, 61.5, 123.2,141.1,
162.1,165.8.
2,6-dimethyl-4-phenylpyridine-3,5-dicarboxyanilide.
O
PhHN
Ph
O
NHPh
^-NMR (d6-DMS0, 400 MHz): 6 2.53 (s, 6H), 7.00-7.41 (m, 15H),
10.23 (s, 2H).
120
B.S-diacetyl-ZAS-trimethyl-pyridine.
°
'
9
^-NMRfCDCI^OOMHz): 6 2.06 (s,3H), 2.43 (s,6H), 2.48 (s,6H).
13,
C-NMR: 6 16.1, 21.8, 32.2,135.9,151.7,175.6, 205.5.
di-tertbutyl 2,6-dimethyl-3,5-pyridinedicarboxylate.
O
O
i
*H-NMR (CDCI3, 300 MHz): 6 1.57 (s, 18H), 2.78 (s, 6H), 8.52
1:
'
N
V-_MMB-
C-NMR: 6 24.5, 28.2, 82.3, 124.9,141.0, 161.0,165.2.
^
121
Heck reaction of ethylene:
one-pot two-step synthesis
of
stilbenes
styrene.
*H-NMR (CDCI3,300 MHz): 6 7.47-7.24 (m, 5H), 6.75 (dd, 1H, Ja=17.6 Hz, Jb=10.9
Hz), 5.78 (dd, 1H, Ja=17.6 Hz, Jb=1.0 Hz), 5.28 (dd, 1H, Ja=10.9 Hz, Jb=1.0 Hz).
4'-vinylacetophenone.
*H-NMR (CDCI3, 300 MHz): 6 7.83 (d, 2H, J=8.4 Hz), 7.39 (d, 2H, J=8.4 Hz),
6.66 (dd, 1H, Ja=17.6 Hz, Jb=10.9 Hz), 5.78 (dd, 1H, Ja=17.6 Hz, J b =l.l Hz), 5.31
(dd, 1H, Ja=10.9 Hz, J b =l.l Hz), 2.50 (s, 3H).
0
13
C-NMR: 6 197.6,142.2,136.4, 136.0,128.8,126.4,116.8, 26.7.
4-methylstyrene.
*H-NMR (CDCI3,300 MHz): 6 7.22 (d, 2H, J=8.1 Hz), 7.05 (d, 2H, J=7.9 Hz), 6.61 (dd, 1H,
Ja=17.6 Hz, Jb=10.7 Hz), 5.61 (dd, 1H, Ja=17.6 Hz, Jb=0.9 Hz), 5.10 (dd, 1H, Ja=10.9 Hz,
Jb=0.8 Hz), 2.25 (s, 3H).
13
C-NMR: 6 137.7,135.0, 136.9, 129.3, 126.3,112.9, 21.3.
4-vinylanisole.
*H-NMR (CDCI3, 300 MHz): 6 7.23 (d, 2H, J=8.6 Hz), 6.75 (d, 2H, J=8.8 Hz), 6.56
(dd, 1H, Ja=17.6 Hz, Jb=10.9 Hz), 5.50 (dd, 1H, Ja=17.6 Hz, Jb=1.0 Hz), 5.02 (dd,
1H, Ja=10.9 Hz, Jb=1.0 Hz), 3.68 (s, 3H).
13
C-NMR: 6 159.5,136.3, 130.5, 127.5,114.0,111.6, 55.3.
4'-acetyl-4-methoxystilbene.
*H-NMR (CDCI3, 500 MHz): 6 7.83 (d, 2H, J=8.4 Hz), 7.44 (d, 2H,
J=8.4 Hz), 7.37 (d, 2H, J=8.8 Hz), 7.07 (d, 1H, J=16.3 Hz), 6.88 (d,
1H, J=16.3 Hz), 6.82 (d, 2H, J=8.8 Hz), 3.73 (s, 3H), 2.49 (s, 3H).
13
C-NMR: 6 197.6, 160.0, 142.5, 135.7, 131.1, 129.6, 129.0,
128.2,126.3, 125.4,114.4, 55.4, 26.6.
IR (dry film): 1258,1672,1597,1258,1176,1030, 967 cm"1.
GC/MS: m/z (% base peak): 252 (100), 237 (67), 165 (33).
4-methoxy-4'-methylstilbene.
^
*H-NMR (CDCI3, 300 MHz): 6 7.45 (d, 2H, J=8.7 Hz), 7.40 (d, 2H,
J=8.0 Hz), 7.16 (d, 2H, J=7.9 Hz), 7.03 (d, 1H, J=16.3 Hz), 6.96 (d,
1H, J=11.9 Hz), 6.90 (d, 2H, J=8.7 Hz), 3.84 (s, 3H), 2.36 (s, 3H).
13
C-NMR: 6 159.3, 137.2, 135.0, 129.5, 127.7, 127.4, 126.7,
126.3,114.3, 55.5, 21.4.
122
(E)-l-(4-methoxyphenyl)-2-(l-naphthyl)ethene.
OCH,
*H-NMR (CDCI3, 400 MHz): 6 8.22 (dd, 1H, Ja=8.0 Hz, Jb=0.9 Hz), 7.86 (dd, 1H,
Ja=7.4 Hz Jb=2.0 Hz), 7.78 (d, 1H, J=9.2 Hz), 7.75 (d, 1H, J=16.2 Hz), 7.73 (d, 1H,
J=7.1 Hz), 7.57-7.45 (m, 3H), 7.55 (d, 2H, J=8.6 Hz), 7.11 (d, 1H, J=16.0 Hz), 6.95 (d,
2H,J=8.9Hz), 3.86 (s, 3H).
13
C-NMR (75 MHz): 6 159.5, 135.3, 133.8, 131.4, 131.3, 130.5, 128.6, 127.9,
127.7,126.0, 125.8,125.7,123.8,123.7, 123.4,114.2, 55.4.
2,4'-dimethoxymethoxystilbene.
Vl-NMR (CDCI3, 500 MHz): 6 7.47 (d, 1H, J=7.6 Hz), 7.37 (d, 2H, J=8.5
Hz), 7.26 (d, 1H, J=16.4 Hz), 7.12 (t. 1H, J=7.6 Hz), 6.97 (d, 1H, J=16.5
Hz), 6.85 (t, 1H, J=7.5 Hz), 6.78 (d, 2H, J=8.3 Hz), 3.77 (s, 3H), 3.71 (s,
3H).
13
I
C-NMR: 6 159.3, 158.9, 130.9, 128.8, 128.4, 127.9, 126.9, 126.3,
121.5,120.9, 114.2, 111.0, 55.6, 55.4.
4-acetyl-4'-methylstilbene.
O
*H-NMR (CDCI3, 500 MHz): 6 7.85 (d, 2H, J=8.3 Hz), 7.47 (d, 2H,
J=8.3 Hz), 7.34 (d, 2H, J=8.0 Hz), 7.10 (d, 1H, J=16.7 Hz), 7.10 (d,
2H, J=7.3 Hz), 6.98 (d, 1H, J=16.3 Hz), 2.50 (s, 3H), 2.28 (s, 3H).
13
C-NMR: 6 197.6, 142.3, 138.5, 135.9, 134.0, 131.5, 129.6,
129.0, 126.9, 126.5,126.5, 26.7, 21.4.
4'-amino-4-methoxystilbene.
,NH 2
^-NMR (CDCI3, 500 MHz): 6 7.32 (d, 2H, J=8.6 Hz), 7.22 (d,
2H, J=8.5 Hz), 6.79 (d, 4H, J=7.8 Hz), 6.57 (d, 2H, J=8.4 Hz),
3.73 (s, 3H), 3.62 (bs, 2H).
13
C-NMR: 6 158.9, 145.9, 130.9, 128.5, 127.6, 127.4, 126.8,
124.8,115.4, 114.2, 55.4.
2-vinylanisole.
//
\
\
/
^ - N M R (CDCI3,300 MHz): 6 7.34 (dd, 1H, Ja=7.6 Hz, Jb=1.7 Hz), 7.05 (td, 1H, Ja=7.8 Hz,
Jb=1.7 Hz), 6.95 (dd, 1H, Ja=17.8 Hz, J b = l l . l Hz), 6.80 (t, 1H, J=7.5 Hz), 6.71 (d, 1H,
J=8.2 Hz), 5.61 (dd, 1H, Ja=17.8 Hz, Jb=1.6 Hz), 5.13 (dd, 1H, Ja=11.2 Hz, Jb=1.6 Hz), 3.67
f
\
(s,3H).
13,
C-NMR: 6 156.8,131.8, 128.9, 126.8,126.6,120.7,114.4, 110.9, 55.4.
123
4-vinylaniline.
*H-NMR (CDCI3,300 MHz): 6 7.24 (d, 2H, J=8.5 Hz), 6.63 (d, 2H, J=8.5 Hz), 6.63
(dd, 1H, Ja=17.5 Hz, Jb=11.0 Hz), 5.56 (d, 1H, J=17.5 Hz), 5.06 (d, 1H, J=11.0 Hz),
3.69 (bs, 2H).
NH
2
13
C-NMR: 6 146.3,136.6, 128.4, 127.4,115.1, 110.1.
2-methylstyrene.
2
H-NMR (CDCI3, 300 MHz): 6 7.47-7.42 (m, 1H), 7.19-7.09 (m, 3H), 6.91 (dd, 1H,
Ja=17.5 Hz, Jb=10.9 Hz), 5.61 (dd, 1H, Ja=17.4 Hz, Jb=1.4 Hz), 5.26 (dd, 1H, Ja=11.0
Hz,Jb=1.4Hz), 2.32 (s, 3H).
1-vinylnaphthalene.
*H-NMR (CDCI3,300 MHz): 6 8.21 (d, 1H, J=8.1 Hz), 7.94 (dd, 1H, Ja=6.2 Hz, Jb=2.7
Hz), 7.88 (d, 1H, 8.2 Hz), 7.72 (d, 1H, J=7.1 Hz), 7.63-7.52 (m, 4H), 5.89 (dd, 1H,
Ja=17.3 Hz, Jb=1.6 Hz), 5.57 (dd, 1H, J^IO.9 Hz, J2=1.6 Hz).
13
C-NMR: 6 135.7,134.5, 133.7, 131.2,128.6,128.2,126.2, 125.9, 125.8, 123.9,
123.8, 117.2.
(E)-2-(4-tolyl)-2-butene.
*H-NMR (CDCI3, 300 MHz): 6 7.45-7.02 (m, 4H), 5.86 (qd, 1H, Ja=6.8 Hz,
Jb=1.2 Hz), 2.36 (2, 3H), 2.04 (s, 3H), 1.82 (d, 3H, J=6.8 Hz).
2-(4-tolyl)-l-butene.
^-NMR (CDCI3, 300 MHz): 6 7.45-7.02 (m, 4H), 5.28 (d, 1H, J=0.5 Hz), 5.05
(d, 1H, J=1.4 Hz), 2.53 (q, 2H, J=8.0 Hz), 2.33 (s, 3H), 1.13 (t, 3H, J=7.4 Hz).
(Z)-2-(4-tolyl)-2-butene.
^-NMR (CDCI3, 300 MHz): 6 7.45-7.02 (m, 4H), 5.57 (qd, 1H, Ja=6.8 Hz, Jb=1.3
Hz), 2.38 (s, 6H), 1.64 (dd, 3H, Ja=7.0 Hz, Jb=1.4 Hz).
isobutylene.
X
H-NMR (CDCI3, 300 MHz): 6 4.68 (s, 2H), 1.75 (s, 6H).
13
C-NMR: 6 142.3,110.4, 24.0.
124
l-(4-acetylphenyl)-2-methyl-l-propene.
O
^-NMR (CDCI3,300 MHz): 6 7.92 (d, 2H, J=8.2 Hz), 7.31 (d, 2H, J=8.2 Hz),
6.30 (s, 1H), 2.60 (s, 3H), 1.95 (s, 3H), 1.91 (s, 3H).
13
C-NMR: 6 197.7, 143.7, 138.3, 134.5, 128.7, 128.2, 124.5, 27.1, 26.5,
19.6.
l-(4-methoxyphenyl)-2-methyl-l-propene.
Vl-NMR (CDCI3, 300 MHz): 6 7.21 (d, 2H, J=8.5 Hz), 6.91 (d, 2H, J=8.5 Hz),
6.26 (s, 1H), 3.84 (s, 3H), 1.94 (s, 3H), 1.90 (s 3H).
13
^°
C-NMR: 6 157.7,133.9,131.3, 129.8,124.6,113.5, 55.2, 26.9,19.4.
l-(4-tolyl )-2-methyl-l-propene.
X
H-NMR (CDCI3, 300 MHz): 6 7.23-7.15 (m, 4H), 6.31 (s, 1H), 2.41 (s, 3H),
1.97 (s, 3H), 1.93 (s, 3H).
13,
C-NMR: 6 135.8, 135.3, 134.7, 128.8, 128.7, 125.0, 26.9, 21.2, 19.4.
125
Reaction scale-up with the Accelbeam
prototype
4,4,4-trifluoro-l-(4-methylphenyl)-l,3-butanedione.
O
II
r
^
OH
1^
j
^-NMFMCDCIfcSOOMHz): 6 15.3 (bs, 1H), 7.83 (d, 2H, J=8.1 Hz), 7.29
(d, 2H, J=8.0 Hz), 6.55 (s, 1H), 2.44 (s, 3H).
CF 3 i3c NMR .
^
6 1 8 6 5^ 1 7 6 7
^
J = 3 6 4 Hz)) 1 4 5 5) 1 3 0 0j 129J>
127J>
117.4 (q, J=284 Hz), 91.8 (q, J=1.7 Hz), 21.5.
4-sulfonamidophenyl hydrazine hydrochloride.
,NHNH2«HCI
^-NMR (d6-DMSO, 300 MHz): 6 10.4 (bs 3H), 8.86 (bs, 1H),
7.71 (d, 2H, J=8.7 Hz), 7.19 (bs, 2H), 7.04 (d, 2H, J=8.7 Hz).
13
H 2 N0 2 S
C-NMR (500 MHz): 6 148.7, 136.4,127.4, 113.9.
4-[5-(4-methylphenyl)-3-(trifluoromethyl)-lH-pyrazol-l-yl]benzenesulfonamide (celecoxib).
H 2 N0 2 S
X
H-NMR (CDCI3,400 MHz): 6 7.85 (d, 2H, J=8.5 Hz), 7.41 (d, 2H,
J=8.5 Hz), 7.16 (d, 2H, J=8.0 Hz), 7.10 (d, 2H, J=8.1 Hz), 6.74 (s,
1H), 5.67 (bs, 2H), 2.35 (s, 3H).
13
C-NMR (d6-DMSO, 500 MHz): 6 145.7, 144.5, 142.7 (apparent
d, J=37.7 Hz), 141.6, 139.6, 129.9, 129.3, 127.3, 126.5, 125.8,
121.8 (apparent d, J=269 Hz), 106.6, 21.3.
(E)-4-methoxycinnamic acid.
C
v
°2H
O
^-NMR (d6-DMSO, 300 MHz): 6 12.23 (bs, 1H), 7.61 (d, 2H, J=8.8
Hz), 7.56 (d, 1H, J=16 Hz), 6.95 (d, 2H, J=8.7 Hz), 6.38 (d, 1H, J=16
Hz), 3.77 (s, 3H).
13
C-NMR: 6 168.3,161.4, 144.2, 130.3,127.3, 117.0,114.8, 55.7.
6-methylthiouracil.
^-NMR (d6-DMSO, 300 MHz): 6 12.25 (bs, 2H), 5.68 (s, 1 H), 2.06 (s, 3H).
13,
C-NMR: 6 176.3, 161.4,153.6, 104.1,18.5.
S-benzyl-6-methylthiouracil.
'
" ' ""
NH A
J
H-NMR (d6-DMSO, 300 MHz) (tautomeric mixture): 6 7.43-7.24 (m, 5 H),
5.99 (s, 0.86 H), 5.39 (s, 0.14 H), 4.38 (s 2H), 2.33 (s, 0.4 H), 2.21 (s, 2.6 H).
13
C-NMR: 6 164.5,164.4,137.9, 129.5,128.9,128.8,127.7, 34.1, 23.6.
126
Appendix III: Tamoxifen
N-[2-(4-iodophenoxy)ethyl]-N,N-dimethylamine.
f|
j
l ^ \ ^
^1
^-NMR (CDCI3 300 MHz): 6 7.54 (d, 2H, J=8.9 Hz), 6.69 (d, 2H, J=8.9 Hz),
^ N ^ 4.02 (t,2H,J=5.7 Hz), 2.70 (t,2H,J=5.7 Hz), 2.32 (s,6H).
N-[2-(4-(l-butynl)phenoxy)ethyl]-N,N-dimethylamine.
s
*H-NMR (CDCI3 400 MHz): 6 7.33 (d, 2H, J=8.8 Hz), 6.84 (d, 2H,
J=8.8 Hz), 4.07 (t, 2H, J=5.7 Hz), 2.73 (t, 2H, J=5.7 Hz), 2.42 (q, 2H,
J=7.5 Hz), 2.35 (s, 6H), 1.24 (t, 3H, J=7.5 Hz).
N-[2-(4-[(E)-l,2-dibromo-l-butenyl]phenoxy)ethyl]-N,N-dimethylamine.
Bi\
/
NL
Br
^-NMR (CDCI3 300 MHz): 6 7.32 (d, 2H, J=8.7 Hz), 6.91 (d, 2H, J=8.7
Hz), 4.15 (t, 2H, J=5.6 Hz), 2.88-2.81 (m, 4H), 2.42 (s, 6H), 1.24 (t, 3H,
J=7.4 Hz).
"O"
127
Appendix IV: Macrolactone
support
2-iodobenzyl alcohol.
H-i-NMR (CDCI3, 300 MHz): 6 7.83 (d, 1H, J=7.8 Hz), 7.46 (d, 1H, J=7.5 Hz), 7.37
(t, 1H, J=7.4 Hz), 7.00 (t, 1H, J=7.4 Hz), 4.66 (s, 2H), 2.41 (bs, 1H).
XH 2 OH
2-iodobenzaldehyde
X
H-NMR (CDCI3, 300 MHz): 6 10.02 (s, 1H), 7.91 (d, 1H, J-7.8 Hz) 7.83 (d, 1H,
J=7.5 Hz), 7.43 (t, 1H, J=7.4 Hz), 7.25 (t, 1H, J=7.5 Hz).
CHO
3-(2-iodophenyl)-propanoic acid.
r
^^r^^-^'
C 0
2
H
Hi-NMR (CDCI3, 500 MHz): 6 11.7 (bs, 1H), 7.86 (d, 1H, J=7.8 Hz), 7.367.26 (m, 2H), 6.95 (td, 1H, Ja=7.9 Hz, Jb=2.1 Hz), 3.10 (t, 2H, J=7.9 Hz), 2.73
(t,2H,J=7.9Hz).
13
C-NMR: 6 179.0, 142.6,139.7,129.5, 128.6,128.4, 100.3, 35.6, 34.2.
3-(2-iodophenyl)-pentandioic acid.
x
X02H
CQ2H
H-NMR (d6-DMSO, 500 MHz): 6 7.87 (d, 1H, J=7.9 Hz), 7.38-7.27 (m, 2H),
6.94 (td, 1H, Ja=7.5 Hz, Jb=2.0 Hz), 3.98 (pentet, 1H, J=7.4 Hz), 2.71 (d,
24H,J=7.4Hz).
13
C-NMR: 6 174.1,144.8, 139.9, 128.5,128.4, 126.6,100.8, 41.7, 38.7.
3-(2-iodophenyl)-l-propanol.
,CH2OH
'H-NMR (CDCI3, 500 MHz): 6 7.84 (dd, 1H, Ja=7.9 Hz, Jb=1.2 Hz), 7.3323 ( m / 2H), 6.91 (td, 1H, Ja=7.9 Hz, Jb=2.0 Hz), 3.74 (t, 2H, J=6.4 Hz),
2.85 (td, 2H, Ja=7.7 Hz, Jb=1.8 Hz), 1.94-1.86 (m, 2H), 1.71 (bs, 1H).
7
13
C-NMR: 6 144.4,139.5, 129.5, 128.4,127.8,100.6, 62.1, 37.0, 33.1.
1-tetralone oxime.
N'°H
^ - N M R (CDCI3, 300MHz): 6 9.68 (bs, 1H), 7.93 (d, 1 H , 7.5 Hz), 7.37-7.14 (m,
3H), 2.89 (t, 2H, J=6.6 Hz), 2.81 (t, 2H, J=5.9 Hz), 1.93 (pentet, 2H, J=6.2 Hz).
13
C-NMR: 6 155.4, 139.9, 130.5, 129.2,128.7,126.5,124.1, 29.9, 23.9, 21.3.
2,3,4,5-tetrahydro-lH-benzazepin-2-one.
X
H
N-^
0
H-NMR (CDCI3, 300 MHz): 6 7.86 (bs, 1H), 7.30-7.21 (m, 2H), 7.15 (td,
1H, Ja=7.6 Hz, Jb=1.2 Hz), 7.00 (d, 1H, 7.8 Hz), 2.83 (t, 2H, J=7.1 Hz), 2.38
(t, 2H, J=6.9 Hz), 2.26 (pentet, 2H, J=7.3 Hz).
13
C-NMR: 6 137.8,134.4, 129.9,127.5,125.7, 121.8, 32.7, 30.4, 28.4.
128
4-(2-aminophenyl)-butyric acid hydrochloride.
C0 2 H
•HCI
||
1|H NMR
(de-DMSO, 400 MHz): 6 10.2 (bs), 7.44 (d, 1H, J=7.0 Hz), 7.36-7.26 (m,
3H), 2.69 (t, 2H, J=8.0 Hz), 2.30 (t, 2H, J=7.5 Hz), 1.82 (pentet, 2H, J=7.9 Hz).
13
C-NMR: 6 174.8, 135.9,131.2, 130.8,128.6, 127.8,124.2, 34.0, 29.7, 25.5.
H2N " "
4-(2-iodophenyl)-butyric acid.
C0
H
1 2
X
H-NMR (CDCI3, 300 MHz): 6 11.1 (bs, 1H), 7.83 (d, 1H, J-8.0 Hz), 7.36-7.19 (m,
2H), 6.91 (t, 1H, J=7.4 Hz), 2.81 (t, 2H, J=7.7 Hz), 2.46 (t, 2H, J=7.4 Hz), 1.98
(pentet, 2H, J=7.4 Hz).
13
C-NMR: 6 179.7,143.8,139.6, 129.5,128.4, 128.0,100.5, 39.8, 33.3, 25.0.
4-(2-iodophenyl)-l-butanol
CH2OH
*H-NMR (CDCI3, 500 MHz): 6 7.83 (d, 1H, 7.9 Hz), 7.29 (t, 1H, J=7.4 Hz), 7.24 (d,
1H, J=7.5 Hz), 6.90 (t, 1H, J=7.8 Hz), 3.73 (t, 2H, J=5.8 Hz), 2.77 (t, 2H, J=7.4 Hz),
1.74-1.67 (m, 4H).
13
C-NMR: 6 144.8, 139.5, 129.4, 128.3,127.7,100.6, 62.7, 40.5, 32.3, 26.4.
ethyl (2-iodophenoxy)-acetate-.
^-NMR (CDCI3, 300 MHz): 6 7.81 (d, 1H, J=7.5 Hz), 7.35-7.24 (m, 1H),
6.82-6.70 (m, 2H), 4.70 (s, 2H), 4.29 (q, 2H, J=7.2 Hz), 1.31 (t, 3H, J=7.1
Hz).
X>
C0 2 Et
13
C-NMR:
6 168.3, 156.8, 139.8, 129.4, 123.6, 112.5, 86.5, 66.5, 61.4,
14.1.
2-(2-iodophenoxy)-ethanol.
1
QH
*H-NMR (CDCI3, 500 MHz): 6 7.80 (dd, 1H, Ja=7.8 Hz, Jb=1.5 Hz), 7.32 (t,
1H, J=7.8 Hz), 6.87 (d, 1H, J=8.2 Hz), 6.77 (t, 1H, J=7.7 Hz), 4.17 (t, 2H,
j=4.5 Hz), 4.01 (t, 2H, J=4.6 Hz), 2.22 (bs, 1H).
13
C-NMR: 6 157.0,139.4, 129.6,123.2, 112.8, 87.0, 70.8, 61.3.
129
General Experimental
All solvents and chemicals were purchased from Aldrich, J.T. Baker, or Fisher Scientific
and used as received without purification. Phenylboronic acid was purchased from Optima
Chemical Group, Douglas, GA and used as received. For automated flash chromatography, prepacked columns of 60 mesh silica gel available from Biotage or Luknova were employed in 25,
40, or 50 g sizes, as appropriate. All reactions described were prepared in air. All NMR were
recorded using 300 MHz, 400 MHz, or 500 MHz Bruker NMR spectrometers as indicated in the
Spectral Data.
Preparation
of phenols from aryl halides with copper
dust.
In a 10-mL glass tube, aryl halide (1.0 mmol), electrolytic copper dust (6 mg, 0.1 mmol),
and NaOH (1.0 mL, 3.1 M aqueous) are combined with a stir bar. The tube is sealed then heated
with a low to moderate microwave power to 200 °C for 20 minutes. Once cooled, the reaction is
acidified to pH 5-7 hydrochloric acid (2 N). The product is extracted from the aqueous with ethyl
acetate (3 x 15 mL). The combined organics are dried over anhydrous magnesium sulfate and
concentrated in vacuo.
Phenols from aryl halides with copper(l)
iodide and proline at 200 °C.
In a 10-mL glass tube, aryl halide (1.0 mmol), copper(l) iodide (19 mg, 0.1 mmol), proline
(5 mg, 0.05 mmol) and NaOH (1.0 mL, 3.1 M aqueous) are combined with a stir bar. The tube is
sealed then heated with a low to moderate microwave power to 200 °C for 30 minutes. Once
cooled, the reaction is acidified to pH 5-7 with hydrochloric acid (2 N). The product is extracted
from the aqueous with ethyl acetate (3 x 15 mL). The combined organics are dried over
anhydrous magnesium sulfate and concentrated in vacuo.
130
Phenols from aryl halides with copper(I) iodide and proline at 300 °C.
In a 80-mL glass tube, aryl halide (5.0 mmol), copper(l) iodide (95 mg, 0.5 mmol), proline
(25 mg, 0.25 mmol) and NaOH (10. mL, 1.6 M aqueous) are combined with a stir bar. The tube is
sealed, loaded on to a rotor with at least 3 other similarly prepared reactions, then heated with
a full microwave power to 300 °C (or 1160 psi). Depending on the substrates, the reaction would
either be cooled immediately once 300 °C was reached, or held at that temperature for 30
minutes. Once cooled, the reaction is acidified to pH 5-7 with hydrochloric acid (2 N). The
product is extracted from the aqueous with ethyl acetate (3 x 15 mL). The combined organics
are dried over anhydrous magnesium sulfate and concentrated in vacuo.
Hydroxycarbonylation
of aryl
iodides.
To a dry 80-mL Anton Paar Synthos 3000 XQ-80 vessel was placed iodoarene (2.0 mmol),
sodium carbonate (0.785 g, 7.4 mmol), palladium(ll) acetate (4.5 mg, 1 mol %), and water (10
mL). The vessel was sealed and loaded onto the rotor with at least three other similarly
prepared reactions. Each vessel was then loaded with 200 psi carbon monoxide. The reactions
were then heated to 165 °C as indicated by an internal probe placed in a reference vessel. After
20 minutes at the desired temperature, the reactions are cooled and vented. The reaction
mixture is acidified to pH 1-3. Products could be collected by filtration, or extracted into ethyl
ether (3 x 15 mL) then dried over magnesium sulfate and concentrated in vacuo.
Alkoxycarbonylation
of aryl
iodides.
To a dry 80-mL Anton Paar Synthos 3000 XQ-80 vessel was placed iodoarene (1.0 mmol),
alcohol solvent (10 mL), DBU (1.1 mmol) and palladium(ll) acetate (0.1 mol %). The vessel was
sealed and loaded onto the rotor with at least three other similarly prepared reactions. Each
vessel was then loaded with 145 psi carbon monoxide. The reactions were then heated to 125 °C
131
as indicated by an internal probe placed in a reference vessel. After 20 minutes at the desired
temperature, the reactions are cooled and vented. Products can be isolated best using the workup procedure as described below.
Alkoxycarbonylation
with low partial pressure
of carbon
monoxide.
To a dry 80-mL CEM Discover vessel was added palladium acetate (2.2 mg, 0.01 mmol),
iodoarene (2.0 mmol), anhydrous ethanol (15.0 mL), and finally l,8-diazabicyclo[5.4.0]undec-7ene (335.0 mg, 2.2 mmol). The vessel was sealed in the Discover apparatus, at which time 50 psi
nitrogen was introduced with the pressure sensor closed. The line to the nitrogen regulator was
closed and the pressure was vented to atmospheric through the Discover pressure sensor. This
process was repeated four more times, then the pressure sensor was closed and the line was
switched from the nitrogen tank to the carbon monoxide tank. Carbon monoxide was slowly
introduced until the pressure reading by the CEM apparatus was 14 psi. The regulator line was
closed, switched back to nitrogen, opened, and the total pressure was slowly increased to 150
psi as measured by the CEM apparatus. The regulator line was then closed and the reaction was
promptly ramped to 125 "C using 150 W microwave power with stirring for a total reaction time
of 20 minutes. The reaction was cooled to below 50 °C, at which time the remaining pressure
was vented through the pressure sensing device. Diethyl ether (50 mL) was added and the
organic was washed with brine (20 mL) and 1 N hydrochloric acid (20 mL). Petroleum ether (50
mL) was added to the organic layer and the resulting solution was washed with water (10 mL)
twice. The combined aqueous washings were extracted once with petroleum ether: ethyl ether
(1 : 1, 20 mL). The combined organics were dried over magnesium sulfate and evaporated in
vacuo. The resulting oil can be purified by silica gel chromatography using petroleum ether to
elute any remaining iodoarene, then petroleum ether: ethyl ether (85 :15) to elute the product.
The combined fractions were evaporated in vacuo to yield the product ester.
132
Oxidation of
1,4-dihydropyridines.
To a quartz reaction vessel was added a 1,4-dihydropyridine ( 0.316mmol), acetonitrile
(7.5mL), and the EDL The vessel was sealed with the screw-top with ports for gas-addition and
fiber optic temperature measurement. A pressure of 145 psi oxygen was introduced, then
vented to the atmosphere. This process was repeated three times. A final charge of 145 psi
oxygen was introduced. The vessel was then sealed, the temperature probe inserted, and the
vessel placed in the microwave cavity. The reaction was heated to 150 °C with a maximum of
600 W for the first minute of heating, then the maximum was decreased to 250 W. An external
pressurized-air line, fit into the vessel jacket, cooled the reaction enough to require more than
100 W to maintain the desired temperature of 150 °C. After 20 minutes, the reaction was cooled
and the remaining pressure was carefully vented. The solvent was removed in vacuo to yield the
crude pyridine product.
Preparation
of styrenes from aryl
iodides.
In a 10-mL microwave tube were placed iodoarene (1 mmol), anhydrous potassium
carbonate (1 mmol, 138 mg), dimethylformamide (1.5 mL), and tributylamine (0.25 mL). The
mixture wasstirred with a magnetic stir bar as palladium ICP standard (20 u± of a 1000 ppm
solution in 5% HCI, 0.02 mol% Pd) was added. Using the gas-loading manifold, a pressure of 150
psi ethene as set at the cylinder regulator was introduced into the vessel by slowly turning the
switch from "Run" to "Load", then back to "Run", isolating the reaction vessel. A second brief
switch back to "Load" was sometimes necessary for the manifold analog pressure gauge to read
150 psi. The reaction mixture was heated to 125 °C with stirring using an initial microwave
power of 50 W and held at that temperature for 60 minutes. The reaction mixture was cooled to
40 °C, and the remaining pressure was carefully vented. This crude mixture can be worked up
133
for isolation of the product by silica gel chromatography as described in the following
procedure, or carried on for the synthesis of a stilbene derivative.
Preparation
of styrenes from aryI
bromides.
In a 10 mL microwave tube were placed bromoarene (1.00 mmol), anhydrous potassium
carbonate (1.00 mmol, 138 mg), CataCXium C (2.3 mg, 0.50 mol% Pd), dimethylformamide (1.5
mL), and tributylamine (0.25 mL, 1.05 mmol). The mixture was stirred briefly, then the vessel
was sealed and loaded with ethylene gas (150 psi) using the method outlined in procedure 1.
The reaction mixture was heated to 150 °C using an initial microwave power of 50 W and held at
that temperature for 60 minutes. The reaction mixture was cooled to 40 °C, and the remaining
pressure was carefully vented. This crude mixture can be carried on directly to a stilbene
synthesis. Otherwise, the reaction was poured into diethyl ether (50 mL) and water (25 mL). The
organic layer was washed with HCI (2 N, 15 mL). The evaporated organic residue is purified by
silica gel chromatography (40 g, 5% ethyl acetate in hexanes) to isolate the styrene derivative.
Preparation
of stilbenes from
styrenes.
To the crude, cooled mixture from one of the two previous procedures was added
anhydrous potassium carbonate (1.00 mmol, 138 mg), CataCXium C (2.3 mg, 0.5 mol % Pd), and
a bromoarene (1.00 mmol). The reaction mixture was heated to 175 °C with stirring using an
initial microwave power of 5-25 W and held at that temperature for 15 minutes. The presence
of iodide greatly increased the microwave absorptivity of the sample, so when the styrene was
prepared from an aryl iodide, careful monitoring was required to prevent over-heating. The
reaction mixture was cooled to 40 °C and poured into ethyl ether (50 mL) and water (25 mL).
The organic layer was washed with 2 M HCI (2 x 15 mL). The evaporated organic layer yields a
solid which can be recrystallized from ethanol: water to yield the stilbene product.
134
Preparation
of 4'-methoxycinnamic
acid in the Accelbeam
protype
In the 5 L vessel, 4-bromoanisole (250.0 mL, 2.0 mol) and methyl acrylate (360 mL, 4.0
mol) were combined with tetrabuylammonium bromide (322 g, 1.0 mol) and an aqueous
solution of potassium carbonate (3.7 M, 2.0 L). Finally, a solution of palladium chloride (1.006
mg Pd/mL, 4 mL) was diluted in water (1 L) and added to the reaction mixture. The chamber was
sealed and pressurized to 280psi with nitrogen before heating to 175 °C until the pressure
approached the working limit of the reactor (~10 minutes at 350 psi). The reaction was allowed
to cool to 165 °C over the next five minutes before the reaction was ejected from the reactor
into cold water (2 L). The solution was acidified to pH 2 with HCI (concentrated). The solid was
filtered under vacuum, washed with water (2 L), and dried to yield the title compound (95 %).
Preparation
of 6-methyl-2-thiouraciI
in the Accelbeam
prototype
Ethyl acetoacetate (2.00 mol, 253 mL) and thiourea (2.60 mol, 200. g) were combined
with potassium hydroxide (> 2 mol, 132 g) dissolved in ethanol (4.0 L). The reaction mixture was
heated to 125 °C under 300 psi N2 and held at this temperature for 25 minutes. Ejection of the
reactor contents into water (2 L) left behind a spongy solid in the vessel, which was dissolved in
the ejected liquids. Sulfuric acid (58 mL) added slowly with mixing precipitated the product and
adjusted the supernatant to pH 6.8. The precipitate was collected by vacuum filtration, washed
with water (1 L) and dried overnight in an oven at 70 °C to yield 6-methyl-2-thiouracil (257 g,
90% yield) as a white powder.
S-benzylation
of 6-methyI-2-thiouraciI
in the Accelbeam
prototype
A suspension of 6-methyl-2-thiouracil (486 g, 3.42 mol) and potassium carbonate (472 g,
3.42 mol) was prepared in N,N-dimethylformamide (6.8 L). Benzyl chloride (394 mL, 3.42 mol)
was added just before heating to 100°C for 30 min under 100 psi nitrogen. The reaction mixture
135
was ejected into water (2 L), acidified with concentrated hydrochloric acid and allowed to cool
overnight. The resulting solids were filtered, washed with water and dried to yield S-benzyl-6methyl-2-thioruacil (510 g, 64 %).
136
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 067 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа