SYNTHESIS OF LABELLED
B.S., Sioux Falls College, 1983
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Science
Department of Chemistry
This thesis for the Master of Science degree by
has been approved for the
Tadesse, Seifu (M.S., Chemistry)
Opiate Receptor/ Agonist Interactions: Hofmann Elimination
in the synthesis of N-labelled Levophenol
Thesis directed by Assistant Professor Michael A. Mikita
The synthesis of N-labelled levorphenol is
required for the continuation of studies in this laboratory
directed at the opiate-receptor complex. Hofmann elimination
is a critical step in our strategy for its synthesis. Toward
this goal, this thesis examined the Hofmann eliminations of
opiate agonists levorphenol and codeine, and opiate models
tropine, piperdine, pyrrolidine, quinuclidine and
I would like to express my appreciation to my advisor,
Dr. Michael Mikita for providing me with the opportunity to be
involved with this project. Also for all the support, guidance
and patience he always showed me. I sincerely thank Dr. Larry
Anderson for all of his help with the word processor and his
computer without which, would have made this a much more
cumbersome task. I am also greatful to Dr. Robert Meglen and
the above-mentioned for sacrificing their time and effort to
read my thesis so thoroughly and offering their helpful
suggestions and criticisms.
In conclusion, I would like to thank my wife Karen for
her editing, proofreading, typing and support.
1.1 Research Statement..............................9
1.2 Method used to reach the goal..................10
II. RESULTS AND DISCUSION................................14
2.1 Protection of Hydroxyl Group.................. 15
2.2 Deprotection of Hydroxyl Group.................18
2.3 Preparation of Methiodide Salts................22
2.4 Preparation of Hydroxide Salts.................25
2.5 Hofmann Exhautive Methylation..................28
2.6 First Hofmann Degradation......................34
2.7 Preparation of Second Methiodide Salts.........38
2.8 Second Ammonium Hydroxide Salts................41
2.9 Second Hofmann Degradation.....................44
2.10 Hofmann Degradation of Codeine................48
2.11 Levorphenol Tartrate..........................52
2.12 Protection of Hydroxide Group of 3-Hydroxy
-N-methyl Morph inan.........................52
III. EXPERIMENTAL SECTION................................59
In 1974 B. Belleau and coworkers studied the
stereoelectronic effects of the basic nitrogen of morphinans.. From
the comparison of the x-ray structure of the morphinan and benzomor-
phan ring system, they conclude that the relative spatial orienta-
tion of the N lone-pair electron (with or without an attached
proton) in morphinans is of critical importance for productive
interaction with the opiate receptor They performed their studies
on compounds la, lb, normorphinan (Ila), and morphinan(lib). While
compounds la, lb, normorphinan, were completely inactive as anal-
gesics in the usual laboratory mouse tests, compound morphinan was
active. From comparison of the x-ray structure of la, lb, normor-
phinan, morphinan with the x-ray structure of the morphine Ilia, and
benzomorphan Illb ring, they concluded that the protonated axial N
lone-electron pairs of la, lb and normorphinan are projected toward'
the phenyl ring, while the corresponding lone pair in morphinan are
projected away from the phenyl ring.
la. X H
Ib. X OH
In the same year Belleau and coworkers reasoned that since
regiospecific orientation of the N lone pair of morphinans appeared
essential for analgesic activity, the protonated form as
hypothesized by Beckett, Casy, and Harper may be tentatively ruled
out as the active species. They offered the hypothesis that the N
electron pair of the free base may interact with an electrophilic
site where upon a stereospecific electron transfer leading to oxida-
tion of the N-methyl substituent may be operative. They referred to
such electronic phenomena at the receptor or enzyme level as "clasM
tic binding." They concluded that clastic binding, which is tan-
tamount to injecting electrons at the receptor level, may help
create interneuronal connections that would be otherwise forbidden.
This allows the multiplicity of biological effects associated with
several classes of narcotics.
In the same year they studied the activity of compounds III,
Ilia, Illb, IIIc. The replacement of the N-methyl hydrogens by
deuterium decreased the analgesic potency by a factor of 1.64. It
can be argued that if the.protonated form of levorphenol were the
active species, NCD^-levorphenol (being more basic) should have
increased rather than decreased activity relative to
NCH^-levorphenol. The isotopic effect on potency is not only absent
with the NCD^CH^ analog.but is actually reversed, the deuterioethyl
compound being 1.61 times more active than its protio counterpart.
Clearly, the convergent differences in the NCH^ and NCf^CH^ groups,
cannot serve to explain the divergencies in the potency ratio. The
only reasonable explanation for the reduced potency of NCD^ levor-
phenol relative to its protio analog is clastic binding on the
opiate receptor complex itself. That is, the opiate receptor complex
would possess an. intrinsic and specific oxidative N-demethylase
activity (clastic binding being characteristic of substrate interac-
tions with oxidative N-dealkylases) .
III R = CH3
Ilia R = CD3
Illb R = NCD2CH
IIIc R = NCHjCH.
From the study of D (-)-3~hydroxy-N, N-dimethylmorphananiura
iodide (IV) L( + )^3hyroxy->N, N-diraethy morphinaniuin iodide (V)
Opheim and Cox found the quaternary iodide of (VI) can exert typi-
cal stereospecific, naloxone reversible effects and, hence, must
interact with opiate receptors. This provides some support for the
belief that the cationic (protonated) form of the opiate drug is the
From this stand point, Horn and Rogers examined the x-ray
structure of 9 opiates which exhibit a range of pharmacological
activity and deduced that one of the reasons why the enkephalins and
related peptides possess morphine-like activity is because they have
a tyramine (VI) residue at the amino terminal position ("tyramine"
moiety of morphine (VII) is illustrated in dark).
They correlated the x-ray data for the agonists, antagonists
and agonist-antagonists and extracted a mean value of 4.6A (s.d =
0.1 A ) for the distance (A) of the N atom from the center of the
aromatic ring (Figure 1). From these common values of (A) Opheim et
al. suggested that the cationic (protonated) form of the opiate
drug is the active species and interacts with the receptor via ionic
association and that the distance of the cationic nitrogen atom
relative to the benzene ring is critical for the interaction.
In 1978 Schiotani et al. examined the x-ray structure of
the seven-member C-ring homologues (VIII), (IX), (X) and an Im-
position isomer (XI) of 6,7-benzomorphans. They found that compounds
VIII, IX, and X were analgesically as potent as morphine and XI as
potent as codeine. But the N lone-electron pairs of IX and X project
away from the benzene ring, where as, the corresponding electron
pairs of VIII and XI orient toward the benzene ring. These results
drew their attention to the possibility that compounds VIII and XI
may invert themselves at the receptor site taking a conformation in
which the N lone pair projects away from the benzene ring. Schiotani
et al. concluded that neither the orientation of the N lone-pair nor
the distance between the benzene ring and hydrogen.on a cationic
nitrogen of benzomorphan and morphinan analogues accounts for struc-
turally induced variations in their pharamacological properties.
IX R = R' = H
R = R' = CH3
Kometani and Schiotani.
also synthesized four other com-
pounds (XII-XV) having a fixed axial conformation of the aromatic
ring for the nitrogen-containing ring. The N-Ph distance values for
their weak analgesic molecules were found to be smaller than their
potent ones. They argued that this structure-analgesic activity
relationship supported a possible role of ionic association of
analgesics with the receptor and the importance of N-Ph distance for
stereochemically controlled binding of analgesics at the receptor
site. They concluded, A rather speculative but consistent explana-
tion of the Ni-Ph distance and activity data suggests itself if it is
assumed that the cationic nitrogen of homologues of 6,7-benzomorphan
interacts with the anionic site of the opiate receptor and the
aromatic ring interacts with the lipophilic site to form a drug-
receptor complex. The binding affinity could then be greatly af-
fected by the N-ph distance of the analgesics, and the optimum
distance for molecules having a fixed axial conformation of the
aromatic ring for the nitrogen containing ring might be 4.5 4.8A".
In 1977 Vera M. Kolb proposed a model of the opiate receptor
to explain structure- activity relationship of opiate agonists and
antagonists. She postulated that only one' conformation of the recep-
tor is needed for binding of both agonists and antagonists. There
are two different specifically fixed amine-binding sites in her
model: one agonist and one antagonist. The opiates undergo binding
to their amine-binding sites via the lone electron pair on nitrogen.
In this model, Kolb also suggested that the primary action of an-
tagonist on the receptor may be through the interaction of the
antagonist N'-lone pair and its.amine-binding site and not through
the lipophilic interaction of the N-alkyl chain with the specific
antagonist binding site.. .
The hypothesis for opiate-receptor interaction by Belleau et
al was recently analyzed by Vera M. Kolb et al. from a chemical
point of view and was found to.be chemically feasible. They looked
at a clastic binding process as the attraction of morphine to its
amine-binding site at the receptor, which could be achieved via
formation of a radical-anionsradical-cation (RARC) (charge-transfer)
complex between morphine and a hypothetical electron acceptor at the
receptor (scheme I).
1.1 Research Statment
The conflicting evidence for N protonation at the receptor
site and its central importance in receptor binding prompted us to
query as to whether N NMR signals from a N labelled opiate could
be used in an invitro environment of opiate receptors to obtain
direct evidence whether or not the nitrogen is protonated at the
1.2 Methods Used to Reach the Goal
The ultimate goal of the opiate research is to use N NMR
signals from a labelled opiate in an in-vitro environment of
solubilized opiate receptors to obtain direct evidence whether or
not the nitrogen is protonated at the receptor site.
The noninvasive nature of NMR makes it particularly well
suited for investigating interactions of biological molecules
without perturbing the chemistry. Recent successes in the
solubilization and purification of the opiate receptor, and recep-
tor subtype. in active form rat brain make attempts at this goal
The basic theory of Nuclear Magnetic Resonance (NMR) is
common to all experiments and all nuclei. The fundamental properties
involved are the nuclear spin and the magnetogyric ratio of the
nuclear magnetic moment. These properties are constant for a par-
ticular type of nucleus. N and .H both have a spin of 1/2 and so
have two energy levels in an applied field. The magnetogyric ratio
of N is smaller than that of .H by a factor of 9.87 and is op-
.. . .15
posite in sign. .
The energy of interaction involved in an NMR experiment is
proportional to these two properties, at a given applied field
strength. The intrinsic sensitivity of a nucleus depends on the
Boltzmann distribution resulting from the magnitude of the energy
differences between the nuclear Zeeman levels and, thus is largely
determined by the very identity of the nucleus. These factors, along
with a natural abundance of 0.36?, put the sensitivity of N rela-
tive to ?H at 3.8X10
The abundance problem can be overcome by introduction of an
N label into the opiate to be used in the experiment pertaining to
the ultimate goal. This thesis was to examine possible synthetic
strategies for the formation of a labelled opiate. The remaining,
intrinsic, insensitivity of N for NMR forms the basis for our
group correlation choosing Insensitive Nuclei Enhanced by Polariza-
tion Transfer (INEPT) as a possible tool for achieving the ultimate
Nuclei with large magnetogyric ratios (like .H) are
polarized to a greater extent in a given magnetic field, and thus,
are much more sensitive than those with low magnetogyric ratio, such
as N. In 1979, a new method was reported for enhancing the inten-
sity of NMR signals from nuclei of low magnetogyric ratio that are
scalar coupled to protons. The enhancement arises from the trans-
fer of nuclear spin polarization from the sensitive nucleus with
large Boltzmann population differences, to the weak nucleus using
scalar coupling to force the weak nucleus energy levels to follow
the inversion of the sensitive nucleus. Morris and Freeman adopted
the acronym INEPT for, "Insensitive Nuclei Enhanced by Polarization
Transfer", for this method.
The method was later demonstrated specifically for N
coupled to protons. The inherent insensitivity of N to NMR
spectroscopy makes its INEPT enhancement particularly striking. The
theoretical enhancement is equal to the ratio of the magnetogyric
ratios of the two coupled species which for N coupled to protons
This enhancement is a direct result of the presence of a
proton coupled to the nitrogen, and thus may provide straightforward
evidence of the presence or absence of protonation.
A number of enhancement by polarization transfer methods are
available that utilize some form of N-H coupling, including Nuclear
Overhauser Enhancement (NOE), Selective Population Inversion.
(SPI), and Selective Population Transfer (SPT). INEPT is not
subject to the possibility of signal nulling from partial NOE and is
much less dependent on precise knowledge of experimental conditions,
such the coupling constant and chemical shift values of either
nuclei, than SPI, SPT and other similar methods .. These considera-
tions make INEPT particularly suited for a system such as the
opiate-receptor work in which these conditions may be affected
during the course of the various experiments.
The use of N NMR and INEPT with model compounds that mimic
the proton exchange environment of an agonist nitrogen in an opiate-
receptor interaction was recently investigated in our laboratory
Due to the low sensitivity of N to NMR its theoretical INEPT
enhancement is tenfold. INEPT enhancement requires nitrogen-'proton
coupling. Evidence from these simple models suggests that a
protonated nitrogen of an opiate bound to its receptor may exhibit
an INEPT enhancement.
Protonated nitrogenous model compounds were reacted in a
system which mimicked the expected environment of the opiate
nitrogen in the opiate-receptor interaction. The model compounds
studied were l8^crown^6 ether in ammonium chloride solution and
quinuclidine in trifluoroacetic acid (TFA).
A solution of only ammonium chloride did not exhibit an
INEPT enhancement due to rapid proton exchange with the solvent.
When a sufficient amount of 18-crown$6 ether was added to the solu-
tion the N NMR spectrum of this new solution exhibited an INEPT
A study was also conducted on a solution of quinuclidine
free base in a trifluoroacetic acid to evaluate INEPT enhancements
from a nitrogen involved in an ion pair. The NMR spectrum of this
solution exhibited an INEPT enhancement indicating that the nitrogen
of quinuclidine was protonated and proton exchange with the solvent
was low Based upon these results an examination of INEPT enhance-
ments from an appropriate N-labeled agonist in a solution of
solubilized receptor sub-types may present a means of examining the
dynamics of the "fit" of an agonist at the receptor. So the N NMR
technique should allow us to identify the protonated from the free
base form of bound opiate. The sensitivity of the N nuclei, is
0.00101 relative to that of a proton in the same applied magnetic
field. This can be a problem when dealing with N NMR. The low
natural abundance of N can be largely overcome by isotopic enrich-
ment, which again is the inspiration for this thesis.
RESULTS AND DISCUSSION
It is obvious from the structure of morphine, codeine and
Levorphenol utilizing a. N enriched precursor early in a synthesis
would be economically impractical. Our synthesis strategy is based
upon the introduction of the N label late in the. synthesis,
thereby requiring less of the costly label for similar amount of
stereospecific product. The best way for accomplishing this a is
reterosynthetic approach with removal of N from pure agonist. The
method of choice for accomplishing this removal was Hofmann elimina-
Because of the expense of codeine and morphine, it was
decided to perform Hofmann degradation on model compounds prior to
use of codiene or morphine. The compounds chosen were tropine,
piperidine, pyrrolidine, 2,5-dimethyl-'4;-(morpholinomethyl) phenol
hydrochloride monohydrate, 1-benzyl-4-hydroxypiperidine and
quinuclidine. These compounds were selected since they have the N>-
atom in the ring which is characteristic for morphine or codiene,
and were readily available.
The Hofmann degradation with morphine or codiene present
potential problems due to the reactivity of hydroxyl (OH) group. To
overcome that problem the protection of the OH group before the
Hofmann degradation is required. The model compound chosen for
studying protection and deprotection methods was tropine. Tropine
was selected since it has an OH functional group, N in a bicyclic
ring and, was less expensive than codiene and morphine.
2.1 Protection of Hydroxyl Group
The selection of a protective group is an important step in
synthetic methodology. When a chemical reaction is to be carried out
selectively at one reactive site in a multifunctional compound,
other reactive sites must be temporarily blocked. A protecting group
must fulfill a number of requirements
1) The protected derivative must be easily prepared in
2) The reagent used must react selectively without affecting
other portions of the molecule.
3) The derivative formed must be stable to the sequence of
chemical reactions undergone by the protected molecule.
4) The protective group must be selectively removed in good
yield by readily available, preferrably nontoxic reagents that do
not attack the regenerated functional group.
5) The protective group should form a crystalline derivative
that can be easily separated from side products associated with its
formation or cleavage.
6) The protective group should have a minimum of additional
functionality to avoid further sites of reaction.
Hydroxyl groups are present in a number of compounds of
biological and synthetic interest including nucleosides, car-
bohydrates, steroids, and the side chains of some amino acids.
Ethers, acetals and ketals (cleaved by mild acidic hydrolysis), and
esters (cleaved by basic hydrolysis) have all been utilized to
protect isolated hydroxyl groups. Benzyl type ethers, however can be
cleaved by hydrogenolysis, and silyl ethers by reaction with a
fluoride ion or by mild acidic hydrolysis.
For the protection of the hydroxyl group the method of E. J.
Corey and A. Venkateswarlu ultilizing dimethyl-tert-
butylsilylchloride looked promising. Dimethyl-tert-butylsilychloride
is stable under mild conditions and exceedingly effective for the
conversion of a variety of alcohols to dimethyl-tert-butylsilyl
ethers in high yield, when using imidazol as a catalyst and
dimethylformamide as solvent. Dimethyl-tert-butylsilyl ethers are
cleaved rapidly and mildly to alcohols by treatment with tetra-n-
butylammonium fluoride in tetrahydrofuran at room temperature.
Reaction of tropine (XVI) with dimethyl- tert-butylsilyl
chloride resulted in XVII. The infrared spectra of the product
(XVII) lacks the broad peak between 3100-3400 cm"\ and had a new
. -1 +
sharp peak around 1600-1700 cm .. The M ion peak at 255 m/e from
GC/MS is also indicative of the protection of the hydroxyl group.
Protecting reagent tetrahydropyran using pyridinium p-
toluenesulfonate as a catalyst was employed as an alternate.
Tetrahydropyranylation of hydroxyl groups has been recognized as a
useful method for the protection of alcohols^ .
Crystalline PPTS can easily be prepared from pyridine and P-
toluenesulfonic acid monohydrate as illustrated.
A solution of tropine (XVI) and dihydropyran in methylene
chloride containing pyridinium P-toluenesulfonate (PPTS) was reacted
together. The white crystalline product (XXII) had a M+ ion peak at
223, which confirms the protection of the hydroxyl group.
2.2 Deprotection of the Hydroxyl Group
The reagent we used for deprotecting the hydroxyl group or
cleaving the dimethyl -tert-butylsilyl. ethers was tetra-n-
butylammonium fluoride in tetrahydrofuran. Tetra-n^butylammonium
fluoride (TBAF) was less reactive in pure methanol solution probably
owing to the, solvation of fluoride ions. From the works of Ogilvie
et al. TBAF cleaves the silyl groups, regardless of their position
in nucleoside using tetrahydrofuran as a solvent.
To a stirred solution of XVII was added a solution of TBAF
in tetrahydrofuran. After work up, compound XXIII was recovered. The
infrared spectra of the new compound had a broad peak at 3100- 3400
cm ? (0-H stretch) which confirmed the formation of compound XXIII.
Hofmann elimination reaction requires a moderately strong
base, a Â£^hydrogen, and a positively charged nitrogen center. With
compounds like piperdine in which a hydrogen is attached to nitrogen
methylation is an essential first step in the Hofmann elimination
Methyl Piperdine (XXV) was prepared by adding methyl iodide
to a stirred solution of piperdine (XXI) in absolute ethanol. The
product, white-yellowish crystals, was recovered after work up.
Methyl Pyrrolidine (XXVII) was prepared by using the same
^ince 2,5-dimethyl-11-(morpholinomethyl) phenol was in
tartrate salt form the synthesis of the free base is an essential
step before preparing the methiodide salt.
2,S-dimethyl-^-Cmorpholinomethyl) phenol (XIX) was syn-
thesized by adding a solution of sodium hydroxide in methanol and
excess methyl iodide to a stirred solution of 2,5-dimethyl-4-
(morpholinomethyl) phenol monohydrate hydrochloric acid (XXVII) in
water. After work-up white crystals of (XXVIII) and (XXIX) were
Methyl 2,5~dimethyl-Il^(morpholinomethyl) phenol (XXX) was
synthesized by adding sodium hydroxide to a stirred solution of
(XXIX) in methanol. From the IR spectra the absence of the broad
peak around 3100-3200 cm ] confirmed the methylation of the hydroxyl
2.3 Preparation of Methiodide Salts
The previous works done on Hofmann degradation on different
compounds by different researchers gave us a wide range of promising
procedures to use. The methods we selected to use are those followed
by Kent E. Opheim et al. .
The quaternary methiodide salts of tropine (XVI), methyl
piperidine (XXV), methyl pyrrolidine (XXVII), methyl 2,5-dimethyl-4-
(morpholinomethyl) phenol (XXIX) and quinuclidine (XXXV) was syn-
thesized by adding excess methyl iodide to the stirred solution of
the starting materials in.absolute ethanol (see Table I).
REACTED PRODUCT X YIELD M.P
XVI XXXI as 335
XXVII XXXII 70
XXVIII XXXIII 75
XXX XXXIV 82 230
XXXV XXXVI 02 348
First Methyl Iodide Salt
2.4 Preparation of Hydroxide Salts
The synthesis of the quaternary hydroxide salt from the
quaternary iodide salt was prepared by using the improved technique
of Joseph Weinstock and V. Boekelheide Instead of using the the
conventional silver oxide method for converting the quaternary
ammonium iodide salt to its corresponding ammonium hydroxide this
technique employs a basic ion-exchange resin. In addition to being
quicker and easier to perform, it gives a cleaner product in higher
yield than does the silver oxide procedure. It'avoids the un-
desirable oxidation which occurs when sensitive compounds are
treated with silver oxide.
Subsequently, the hydroxide salts of compounds XXXI, XXXII,
XXXIII, XXXIV and XXVI was synthesized by adding ion-exchange resin
to the stirred solution of the methiodide salts (see Table II).
REACTED PRODUCT X YIELD
XXXVI XL I
First Hydroxide Salt
(_ ho) oov vai
( ho)001' vai
IRA wo <0H )
2.5 Hofmann Exhaustive Methylation
The conversion of an amine to an olefin by elimination of
the nitrogen atom and:an adjoining hydrogen atom is a useful
procedure for degradation and synthesis. The Hofmann exhaustive
methylation method has been used most often to bring about this
change, eventhough other methods such as the thermal decomposition
of amine oxides and the pyrolysis of amine phosphates or acetyl or
benzyl derivatives have often been employed to advantage.
Decomposition of a quaternary ammonium hydroxide with the
formation of an olefin, and water was reported by Hofmann in
1851^ . Since then it has become a routine step in the study of
alkaloids. Since a methyl group cannot be eliminated as an olefin,
cleavage must take place to free another group from the nitrogen
atom. If the original amine is heterocyclic, this cleavage gives
rise to a compound containing both an olefinic and a tertiary amino
group. Repetition of the procedure yields a diene and
trimethylamine. The degradation of N-methylpyrrolidine (XLII) may be
used to illustrate these steps.
In compounds like quinolizidine derivatives in which the
nitrogen atom is located at the bridgehead, three such steps would
be necessary to eliminate it as trimethylamine.
Thus the degradation not only introduces a new functional
group, the olefinic double bond, which allows further degradation,
but also the number of steps required to liberate the nitrogen atom
as trimethylamine is an indication of its location in the original
In order to describe these reaction products we used the
method used by Hofmann for nomenclature. According to the "methine"
system, the Hofmann product is called the methine or methine base of
the parent alkaloid; so IXL would be pyrrolidinemethine. The product
obtained by repeating the process of methylation and pyrolysis would
be the bis-methine and that obtained after three steps a tris-
methine. This nomenclature is used widely in naming degradation
products of morphine and its derivatives and some other alkaloids.
The decomposition of quaternary ammonium compounds was
described as belonging to that class of bimolecular elimination
reactions called E2 reaction by Highes, Ingold and Patel in 1933 .
Subsequent work has served to confirm the opinion that this is the
usual course of the reaction. The general requirements of the Hof-
mann elimination reaction suggest that a moderately'strong base, a
^-hydrogen atom, and a positively charged nitrogen center are in-
proceeds without1 difficulty in many compounds that do not have aÂ£-
hydrogen atom. These observations are in accord with either a con-
certed process (E2) or a stepwise reaction (E1cb, E1 elimination in
the conjugate base) in which the ^-hydrogen atom is removed first
forming a carbanion intermediate. Actually, as Ingold pointed out in
1933^ and as has been restated recently by Saunders and Williams'3 ,
these mechanisms may be taken as extremes which merge as the
lifetime of the carbanion is considered to become shorter in the
stepwise reaction or as the degree of carbon to hydrogen bond break-
ing in the transition state become greater in the concerted process.
volved since all of these are usually necessary. Elimination
R----C-----CRs + B <---------------->
R'zCcr2 + B <--------> R'gC
u z +
A choice between these mechanisms can not be made on the
basis of kinetic order, since both require second order behavior.
One of the requirements of the E2 mechanism is that the hydrogen
atom and the nitrogen group involved in the elimination process be
coplanar and in the trans conformation. In the transition state of
equation (1) the atoms H and NR"^ are trans oriented and the four
atoms H C C N are in the same plane (coplanar) and the
olefinic bond is partially formed. The transition state for the E2
process has olefinic characters and this influences its free energy.
The reaction is also sensitive to steric interactions and these two
factors, olefin stability and steric factors, determine the nature
of the product.
When the groups R B, or R1 are large and bulky, steric
interaction determines the course of the E2 process. The reaction
proceeds through the least crowded transition state where the base
abstracts the most accessible hydrogen atom, and the olefinic
product need not be the most highly substituted olefin. On the other
hand, when steric interactions are less important, the position of
the double bond is the determining factor, and in these cases
bimolecular elimination affords the most highly substituted alkene.
Elimination that yield the most stable olefin are called Saytzeff
eliminations, whereas those in which steric factors determine the
course of the reaction are called Hofmann eliminations.
The proportion of Hofmann elimination is also found to
increase with increasing branching in the alkyl group of the sub-
strate, and with increasing branching base. Most of the useful
applications of Hofmann elimination reactions have been with
alkaloids containing the nitrogen atom in a ring, usually five or
2.6 First Hofmann Degradation
Since compounds XXXVII, XXXVIII, XXXIX, XL and XLI are
heterocyclic and the N-atom is in the bridge they have to under go
two or three Hofmann degradations. Distillation of these compounds
resulted in the formation of compounds XLII, XLIII, XLIV, XLV and
XLVI (see table III).
REACTED PRODUCT 7. YIELD B.P. C
XXXVII XLII . 87
XXXVIII XLIII 57 28
XXXIX XLIV 55 28
XL XLV 87
XLI XLVI 73 no
First Hofmann Degradation
Compound XXXVII has a quaternary nitrogen and two Q-
hydrogens which are the least substituted. The base (OH ) can
abstract one of the ^hydrogens. Since the compound is cyclic and
needs two Hofmann eliminations to make the olefins it does not
matter which 0-hydrogen is going to be abstracted first. Because the
two 0-hydrogens are in the same enviroment.
This resulted in the formation of the first double bond. From the
infrared spectra of XLII the absorption around 1600 cm which is
absent in tropine (XVI) demonstrated the formation of the first
The N-atom in compound XXXVIII is located in the ring which
implies that to produce diolefins and trimethylamine it takes two
Hofmann degradations. From the infrared spectra of XLIII the peak
r ^ 1
around 1600 cm which was absent in piperidine showed the forma-
tion of the double bond. Also the M+ ion peak at 113 confirmed the
formation of product XLIII.
Compound XXXVIII has quaternary nitrogen atom and two (3-
hydrogens. So the base can abstract any of the two -hydrogens,
since we need two Hofmann eliminations. The base abstract the $-
hydrogen which resulted in the formation of the first double bond.
The infrared spectra peak at 1600 cm confirmed the formation of
the first double bond.
Compound XXXIX has two 0-hydrogen and a quaternary nitrogen.
The base can abstract one of the
formation of the double bond and produce XLIV. The infrared spectra
of compound XLIV which has an absorption at 1600 cm \*(-=c=c- stretch)
demonstrated the presence of the double bond, and the M ion peak at
99 m/e in GC/MS is confirmatory.
Compound XL has a quaternary nitrogen atom and two hydrogen.
The base can abstract one of the hydrogens, which results in the
formation of the double bond. From the infrared spectra of XLV the
peaks at 1600 and 1660 cm ? implied the formation of the double
bond, even though we know the aromatic group can contribute to the
peaks, the M+ ion peak at 243 m/e from GC/MS confirms the formation
of the double bond.
Compound XLI has a quaternary nitrogen atom and two least
substituted hydrogen atoms, which implies the elimination follows
that of Hofmann type. The base can abstract one of the hydrogens,
which results in the formation of the first double bond. The M+ ion
peak at 125 m/e from the GC/MC confirms the formation of the first
2.7 Preparation of Second Methiodode Salt
The preparation of the second methiodide salts of compounds
XLII, XLIII, XLIV, XLV and XLVI followed the same techniques we used
for preparing the first one. Compounds XLVII, XLVIII, XLIX, L and LI
were thus synthesized, (see table IV).
REACTED PRODUCT .Z'YIELD M.P. C
XLII XLVI I 65
XLIII XLVIII 79
XLIV XLIX 70
XLV L 52 152
XLVI LI 73 190
Second Methyl Iodide Salt
2.8 Second Ammonium Hydroxide Salts
Similarly the second ammonium hydroxide salts of compounds
XLVII, XLVIII, XLIX, L and LI were synthesized by the same technique
discussed earlier (see table V).
REACTED PRODUCT X YIELD
XLVII LII - 58
XLVIII LI 11 45
XLIX LIV 53
L LV 57
LI LVI 55
Second Hydroxide Salt
IRA JUD <0H")
IRA 400 (OH )
[______| f + IRA 400COH )
2.9 Second Hofmann Degradation
In Hofmann elimination reactions of heterocyclic compounds,
we assumed the trans elimination, formation of conjugate olefins
when possible, and preferential loss of hydrogen from methyl group
in competition with other alkyl groups.
The reaction mechanism is that of an E2 which was described
for the first Hofmann degradation. Distillation of compounds LII,
LIII, LIV, LV and LVI resulted in the formation of compounds LVII,
LVIII, LIX, LX and LXI with two new double bonds (see table VI).
REACTED PRODUCT Z YIELD B.P. C
LII LVII 58
LIII LVIII 18
LIV LIX IS
LV LX 52 88
LVI LXI 57 98
Second Hofmann Degradation
Compound LII has only one least substituted 0-hydrogen. The
base (OH ) can abstract this j-hydrogen which results in the forma-
tion of the second double bond. From the infrared spectra the ab-
sorption at 1600 cm (hc=c^ stretch) and 1650 cm (-c=c^ stretch)
the formation of the conjugate olefin, can be inferred.
Compound LIII has only one ^-hydrogen left. The base can
abstract this j-hydrogen which results in the formation of the
second double bond. From the reports of Hofmann degradation done on
similar compounds, the product assumed formed was 1,3-pentadiene
instead of 1,4^pentadiene. This was because of the allylic shift,
since a conjugate double bond is more stable based on the ther->
modynamic consideration. The formation the double bond was confirmed
by the property of the product which is gaseous at room temperature.
LIV has only one 0-hydrogen remaining. The base abstracts
this ^hydrogen Which results in the formation LIX. LIX was a
gaseous at room temperature.
Compound LV also has only one t-hydrogen which can
abstracted by the base. The peaks at 1600 cm (^c=c- stretch) and
1650 cm ? (-c=c- stretch) confirm the formation of the second double
Compound LVI has two $-hydrogens in the same environment.
The base can abstract any one of the two ^.-hydrogens since two more
steps are required to get the olefin. The M ion peak at 139 m/e
from the GC/MS confirmed the formation of the second double bond.
2.10 Hofmann Degradation of Codiene
Preparing the methiodide salt
The methiodide salt of codiene was prepared by the same
method used for model compounds. We recovered a white crystal of
LXIII which was insoluble in absolute ethanol, methanol, diethyl
ether and acetone. It dissolved in water and N.N^dimethylformamide
The synthesis of the ammonium hydroxide salt, of LXIII was
done by the same technique used for the model compounds. We
recovered a cloudy oil of compound LXIV.
LXIV has three (3-hydrogens, but the two @-hydrogens are the
least substituted. One of the (^hydrogens is more acidic than the
rest. The base will abstract the acidic ^-hydrogen first which
results in the formation of LXV. The GC/MS of LXV shows The M+ ion
peak at 313.15 m/e which confirms the formation of the first double
Second Methiodide Salt
Using the same technique used for the model compounds, the
second methiodide salt of LXV was synthesized.
The hydroxide salt of LXVI was synthesized using the same
methods employed for the model compounds.
Decomposition of LXVII produced LXVIII. The ^-hydrogen of
compound LXVII was abstracted by the OH which resulted in the
formation of the second double bond. The M ion peak at 270.15 m/e
in the GC/MS confirmed the formation LXVIII.
2.11 Levorphenol Tartrate
Before synthesizing the methiodide salt of levorphenol
tartrate the free base has to be separated from the salt. The levor-
phenol tartrate was reacted with sodium hydroxide from which was
recovered a white crystal LXX. The M+ ion peak at 257 m/e from GC/MS
confirmed the formation of LXX.
2.12 Protection of Hydroxide Group of 3~Hydroxide-N-Methyl Morphinan
Levorphenol.was reacted with acetyl chloride. After work up
a crude oil of LXXI was recovered. The M ion peak at 299 m/e from
GC/MS confirmed the protection of the hydroxide group and formation
Methiodide Salt (LXXII)
The methiodide salt of LXXI was prepared by the same method
used for the other model compounds. A brownish crude oil of LXXII
was recovered, which gave yellow precipitate with silver nitrate.
Hydroxide Salt (LXXIII)
The hydroxide salt of LXXII was prepared using the same
technique used for the other model compounds. A brownish crude oil
of LXXIII was recovered.
Decomposition of LXXIII produced LXXIV. LXXIV exhibited a M+
ion peak at 271 m/e from the GC/MS which confirmed the formation of
the first double bond.
Using the same technique used for model compounds the second
methiodide salt of LXXIV was synthesized and recovered LXXV. LXXV
gave a yellow precipitate with silver nitrate.
The hydroxide salt of LXXV was also synthesized by the
method used for the first hydroxide salt.
Decompostion of LXVI produced LXVII. The M+ ion peak at 226
m/e from the GC/MS confirmed the formation of LXVII.
The methodology established in this thesis for the Hofmann
elimination of nitrogen from opiates and their models is an impor-
tant and integral step toward the-ultimate synthesis of N-labelled
levorphenol. For this total synthesis the Hofmann elimination is
essential, as illustrated in Fig. II. Efforts are continuing in our
laboratory toward this goal.
T uH txcnonge
5. Repeat 2-4
2. CIV NoOH ,
Melting points were determined on a Mel-Temp apparatus and
are uncorrected for pressure. Infrared spectra were determined on a
Perkin-Elmer 237B spectrophotometer using potassium bromide pellets.
. H and C NMR spectra were measured with an IBM/80 FT-NMR Fourier
transform spectrometer using Me^Si as an internal standard. GC-MS
analyses were done with a Hewlett Packard Model 5790A gas
chromatograph equipped with a Hewlett Packard Model 5970A mass
selective detector. Analytical and preparative TLC's were done on
commercial plates of silica gel (60 F25M, DARMSTADT). And spots were
revealed by their UV emission. Column chromatography was done on a
silica gel (100-200 mesh) Dihydrophyran, dichloromethane and DMF
were distilled and stored over 5A molecular sieves. The rest of the
solvents were used without further purification. All of the organic
compounds utilized in this study were commercial products. The anion
exchange resin (OH form ) was washed several times with absolute
ethanol before use. All glassware was dried thoroughly in a drying
oven prior to use.
Protection of the Hydroxyl Group
To a stirred solution of tropine (XVI) (0.5 gm, 0.0035 mol)
in N,N^dimethylformamide (2 mL) was added dimethyl-tert-butylsilyl
chloride (0.64027 gm, 0.00428 mol). The reaction mixture was stirred
under nitrogen gas. After two hours the color of the solution
changed from light yellow to light*-greenish. From comparison of the
infrared spectra of tropine (XVI) and the infrared spectra of the
solution (XVII), the absence of the broad peak between 3100 3400
cm and appearance of a sharp peak around 1700 -1600 cm in the
solution mixture confirmed the protection the hydroxyl group. The
M+] ion peak at 255 m/e of XVII (0.122) from the GC/MS shows the
protection of the hydroxyl group.
P-Pyridium Toluenesulfonate (PPTS) (XX)
P-toluenesulfonic acid monohydrate (XVIII) (5.70 gm, 30
mmol) was added to pyridine (XIX), (12.1 mL, 150 mmol). After stir-
ring for twenty minutes the pyridine was removed with rotary
evaporator on a water bath at about 60 c to afford a quantitative
yield of PPTS (XX) as slightly hygroscopic colorless crystals.
Recrystallization from acetone gave the pure salt (6.80 gm)(90$)
melting point 107-108 *C.
Protectoin of OH Group of Tropine
A solution of tropine (XVI) (0.00218 moles, 0.308 gm) and
dihydropyran (XXI) (0.252 gm, 0.00299 moles, 0.2733 mL) in methylene
chloride (14 mL, freshly redistilled and dried over molecular
sieves) containing pyridinium p-toluenesulfonate (PPTS) (o.05 gin,
0.000199 mol) was stirred overnight at room temperature. Then the
solution was diluted with ether, at which point the solution turned
to a milky color from colorless. It was then washed once with half-
saturated brine (NaCl solution) (25 mL) to remove the catalyst. The
ether solution was dried over anhydrous magnesium sulfate (MgSO^).
After removing the ether via rotary evaporator, a white crystal with
an oily residue of XXVI (0.7*1 gm, 82$) was recovered. From the M+
ion peak at 223 m/e (0.896) the protection of the the hydroxyl group
To a stirred solution of 2,5-dimethyl-4-(morpholinomethyl)
phenol (XXIX) (3.5 gram, 0.015780 mol) in methanol (40 mL) were
added sodium hydroxide (0.5 gram 0.0125 moles) and stirred for one
half hour at room temperature. The resulting solution was put in
water bath (60-70 *C) for one half hour and cooled down to room
temperature. To the cooled mixture was added methyl iodide (9.5 gram
0.0669 mol) dropwise. The resulting yellow solution was refluxed for
an hour. Evaporation of the solvent furnished a white yellow oily
crystal (XXX). From the infrared spectra (solution cell) the absence
of the broad peak at 3100-3400 cm .(O-H stretch) confirmed the
methylation of the compound.
Deprotection of Hydroxyl Group
A solution TBAF (0.846 gm,0.00268 mol) in tetrahydrofuran
3.7 mL freshly distilled and dried over molecular seive (10.6 ml/g
of XVII) was added dropwise to the flask containing stirred solution
of XVII (0.347 gm, 0.0013^ mol). After an hour the solution mixture
was diluted with ethyl acetate, washed with water and dried over
magnesium sulfate. After evaporation the solvent were recovered
XXIII (75%, 0.075 gm,0.0005319 mol) white-brownish crystals (oily
2,5-Dimethyl-4-(Morpholinomethyl) Phenol.L- (XXIX)
A solution of 2,5-dimethyl^4-(morpholinomethyl) phenol
hydrochloride monohydrate (5 gram,0.01829 mol) in distilled water
(50 mL) containing sodium hydroxide (0.8 gram, 0.02 mol) was stirred
for half hour at room temperature. The resulting solution was washed
with diethyl ether (150 mL). The ether extract was dried over mag-
nesium sulfate and evaporated to yield 2,5-Dimethyl -4-
(morpholinomethyl) phenol (XXIX) (3.5 gram,87.5%) a white crystals,
melting point 78-81 #C.
Methiodide Salt (XXXIV)
A solution of 3.5 gram (0.015780 mol) Of (XXX) and methyl
iodide (9.5 gram, 0.0669 mol) of methyl iodide in 50 mL of absolute
ethanol was stirred under reflux for 16 hours. The resulting reac-
tion mixture was concentrated under vacuum to give 2.87 gram,0.00765
mol (82$) of white yellow crystals (XXXIV) which decomposed at
Hydroxide Salt (XL)
To a stirred solution of 2.87 gram,0.00766 mol of XXXIV in
50 mL of absolute ethanol was added 32' gram of anion exchange resin
(OH form). The reaction mixture was stirred overnight and filtered
by gravity. The filterate was concentrated under reduced pressure
and the resulting residual oil (XL) was distilled at 110-115 cC to
give 1.9 gram (67%) of viscous colorless oil (XLV), which was con-
firmed by its sharp absorption at 1600 cm ^ (-c=c- stretch) in
Second Methiodide Salt (L)
The second methiodide salt was prepared by treating (XLV)
with excess of methyliodide in absolute ethanol (25 mL) at room
temperature for one hour followed by boiling under reflux overnight.
Removal of the solvent gave a white solid of (L) (1.653
Second Hydroxide Salt (LV)
To a stirred solution of L (1.653 gmf) in 25 mL of absolute
methanol was added 32 gram of anion exchange resin (OH form). After
the reaction mixture was stirred overnight at room temperature, it
was filtered and the filterate was concentrated to yield an oil (LV)
which was distilled at 110->115 CC to give a viscous colorless oil
(LX) (0.859 gm, 52%). The infrared spectrum (solution cell) at 1600
cm (-c=c* stretch) and 1650 cm (-c=c- stretch) confirmed the
formation of the double bond.
Tropine Methiodide Salt (XXXI)
A solution of tropine (XVI) (5 gm, 0.0354 mol) in absolute
methanol (25 mL) was placed on an ice bath to which methyl iodide
(5.05 gm, 0.0354 mol) was added dropwise to the solution. After
three hours the solvent was evaporated to yield (8.6gm 0.0303 mol)
(89?) of white crystals of (XXXI) which decomposes at 333335 "'C.
XXXI gave a yellow precipitate with silver nitrate.
Tropine Hydroxide Salt (XXXVII)
To a stirred solution of (XXXI) (8.6 gram, 0.03038 mol) in
absolute ethanol (25 mL) was added 24 gram of anion exchange resin
(OH form). After the mixture was stirred overnight at room tempera-
ture it was filtered and the filtrate was concentrated and the
resulting residual oil (XXXVII) was distilled at 78-80 'C to give a
viscous oil (XLII) (5.7 gm, 0.0367 mol) (67?), which exhibited an IR
(solution cell) 1600 cm ? (-c=c-= stretch).
Second Methiodide Salt (XLVII)
To a stirred solution of 5.7 gram, 0.0367 mol of XLII in
absolute methanol (25 mL) an excess of methyliodide was added. The
resulting solution was stirred for an hour, then refluxed for two
hours at 92-94 "C. Removal of the solvent produced a yellowish
crystals of (XLVII) (4.8 gram, 0.0163 mol, 85?) which decomposed at
Second Hydroxide Salt (LII)
A solution of XLVII (4.8 gram, 0.0163 mol) in absolute c
ethanol (25 mL) and 36 gram of anion exchange resin (OH form) was
stirred overnight at room temperature. The resulting solution mix;*
ture was filtered and the filtrate was concentrated. The resulting
residual oil (LII) was distilled at 98-105 !>C to give a viscous
colorless oil of (LVII) (2.6 gm, 0.0236 mol, (56?). From IR spectrum
(solution cell) the peak at 1600 cm (-c=c- stretch) and 1650 (<*
c=c- stretch) confirmed the formation of the conjugate double bond.
Methiodide Salt (XXXII)
The dimethyl piperidine ammonium iodide salt was prepared by'
adding excess methyl iodide (4.04 gm, 2.84 mL) to a solution of
methyl piperidine (XXVII) (2.90 gm, 0.0292 mol) in absolute ethanol
(25 ml). The solution mixture was refluxed for two hours. After the
removal of the solvent white-yellowish crystal (.2.26 gm, 0.009347
mol) (78?) (XXXII), were recovered which decomposed around 112 aC.
Hydroxide Salt (XXXVIII)
To a stirred solution of (2.26 gm, 0.009347 mol) (XXXII) in
25 mL of absolute methanol was added 16 gm of anion exchange resin
(OH form). The reaction mixture was stirred over night and flit-
tered. The filtrate was concentrated under reduced pressure and the
resulting residual oil (XXXVII) was distilled at 80 9C for two
hours. (1.4 gm, 0.01238 mol) of XLII was recovered. IR (solution
cell) 1600 cm ^ (-c=c- stretch) and mass spectra 113 m/e from the
GC/MS confirmed the formation the first double bond.
Second Methiodide Salt (XLVIII)
The second methiodide salt was prepared by adding excess
methyl iodide to a stirred solution of XLIII (2.00 gmf 0.0176 mol)
in absolute methanol (25 mL) in ice bath. The resulting solution was
refluxed for two hours. Removal of the solvent gave white-yellowish
crystals of XLVIII (1.8 gm, 0.014 mol) (79%) which decomposes at 135
*C and gave a yellow precipitate with silver nitrate.
Second Hydroxide Salt (LIII)
To a stirred solution of XLVIII (1.8 gmf 0.014 mol) in 25 mL
of absolute ethanol was added 20 gm of anion exchange resin (OH
form). After the reaction mixture was stirred overnight at room
temperature, it was filtered and the filtrate was concentrated. The
resulting residual oil (LIII) was distilled at 4 mmHg and 25-35 :C.
The clear oil collected in an ice bath was gaseous at room tempera-
Methiodide Salt (XXXIII)
The dimethyl pyrrolidine ammonium iodide salt was prepared
by adding excess methyl iodide (5.989 gm, 0.0428 mol) to the stirred
solution of methyl pyrrolidine (3 gm, 0.0428 mol) in absolute
methanol (25 mL) in an ice bath. The resulting solution was refluxed
for two hours at 55 C. After the evaporation of the solvent XXXIII
(2.25 gm, 0.0105 mol) (75%), which decomposed at 235 cC was
Hydroxide Salt (XXXIX)
The basic ion-exchange resin was washed several times with
absolute ethanol before use. To a stirred solution of XXXIII (2.10
gm,0.00985 mol) was added (4-equivalent) the washed anion-exchange
resin. The resulting mixture was allowed to react overnight at room
temperature by stirring. After filtering the resulting solution was
concentrated and distilled at 4 mmHg and 20-25 ,:C which produced a
clear oil XLIV (1.3 gm, 0.0185 mol, 65%). From IR spectrum (solution
cell) the absorption at 1600 cm ] (-c=c- stretch) confimed the the
formation of XLIV.
Second Methiodide Salt (XLIX)
The second methiodide salt was prepared by treating (XLIV)
(1.0 gm, 0.0142 mol) in 25 mL absolute methanol with excess methyl
iodide at room temperature. The resulting solution was refluxed over
night. Removal of the solvent gave white-yellowish solid which
decomposed at 233 *C (0.7 gm,0.00377 mol, 70$).
Second Hydroxide Salt (LIV)
To a stirred solution of XLIX (0.7 gm, 0.0037 mol) in ab-
solute ethanol (25 mL) was added 10 gm of anion exchange resin.
After the reaction mixture was stirred overnight at room temperture,
it was filtered and the filtrate was concentrated. The resulting
residual oil (LIV) was distilled at 4 mmHg and 20-25 'C to give a
viscous colorless oil in ice bath, which was gaseous at room tem-
Methiodide Salt of Codiene
A solution of codiene (3.10 gram, 0.01035 mol ) in absolute
ethanol (25 mL) was placed in an ice bath and methyliodide (1.788
gram 0.0125 mol, 0.641439 mL) was added dropwise with continuous
stirring. After three hours the solvent was evaporated to yield
LXIII (3.86 gram, 0.00848 mol 79.09$) white crystals which decompose
at 139-142 'C. The crystals were insoluble in absolute methanol,
ethanol, diethylether and acetone. It disolved in water and N,N-
To a stirred solution of LXIII (3.5 gram, 0.00769 mol) in
DMF (25 mL) was added 28 gram of anion exchange resin (OH form).
After the resulting mixture was stirred overnight at room tempera-
ture it was filtered and the mother liquor was distilled at 55-59 C
and 10 mmHg for an hour at which point a browinsh oil of LXXIV was
collected. Cold finger distillation of LXIV at 98-110 C for four
hours at 3 mmHg produced white brownish crystals of LXV (2.1 gram,
0.006709 mol 60$), which has a M+ peak at 313 m/e (1.0) on GC/MS.
Second Methiodide Salt
A solution of LXV (2.00 gram, 0.00638 mol )in absolute
ethanol (25 mL) was stirred at room temperature to which three fold
excess methyl iodide was added dropwise. The resulting solution was
refluxed for an hour and cooled to room temperature. The reaction
mixture was concentrated under vacuum to give 4.19 gram (0.00928
mol) of gummy crystals. Recrystallization of the gummy crystals from
absolute methanol gave us 1.56 gram of white brown crystals of
LXVI, which precipitate with silver nitrate solution.
Second Hydroxide Salt
To a stirred solution of LXVI (1.5 gram, 0.00329 mol) in 25
mL of DMF was added 20 gram of anion exchange resin (OH form). The
reaction mixture was stirred overnight at room temperature and
filtered by gravity. The mother liquor was distilled at 32 'C and 3
mmHg and recovered a brownish oil of LXVII (0.71 gram) which was
soluble in absolute methanol. Distillation of LXVII at 85-90 ''C and
3 mmHg gave us LXVIII ( 0.50 gram, 0.00185 mol 67%). From GC/MS the
M+ peak at 270 m/e confirmed the formation of LXVIII.
To a well stirred solution of Levorphenol tartrate (5.35 gm,
0.06815 mol) in 10 mL of absolute ethanol was added sodium hydroxide
(2.726 gm, 0.0685 mol) in ethanol (7 mL). To the resulting mixture
was added 50 mL of ethyl ether. The solution has a pH of 11. After
washing the resulting solution with ethyl ether (200 mL) and drying
over MgSO^, evaporation of the solvent yielded (3.1127 gm, 0.01211
mol 92?) of white crystals. From the GC/MS the M+ ion peak at 257
m/e (.967) confirmed the separation of the free base from the salt.
Acylation of Hydroxide Group (LXXI)
A solution of Levorphenol (5.00 gm, 0.019 mol) in dioxane 25
mL containing sodium hydroxide (2 gm, 0.05 mol) and
methyltrialkylammonium chloride was stirred at room temperature. To
the resulting mixture was added a solution of acetyl chloride (11
mL) in dioxane (10 ml) dropwise over 30 minutes. The mixture was
filtered and washed with dioxane. After evaporation of the solvent a
crude oil of (LXXI) (3.112 gm, 0.0122 mol) was recovered. The M+
peak at 299 m/e (1.0) from GC/MS confirmed the formation of LXXI.
Methiodide Salt (LXXII)
A solution of LXXI (3.112 gm, 0.0122 mol) and 3.8 mL ( 8.6
gm, 0.06074 mol) of methyl iodide in 25 mL absolute ethanol in ice
bath was stirred for an hour. The resulting mixture was concentrated
under vacuum to give 4.6 gm, 0.0115 mol (87?) of the methiodide salt
LXXII, which decomposes at 220- 225 C.
Hydroxide Salt (LXXIII)
To a stirred solution of LXXII (4.5 gm, 0.0112 mol) in 30 mL
of absolute methanol was added 40 gm of anion exchange resin. The
reaction mixture was stirred over night and filtered by gravity. The
filtrate was concentrated under vacuum and the resulting residual
oil (LXIII) was distilled -at 135-155 C to give a white crystals of
LXXIV (2.88 gm, 0.0106 mol), which has an M+ peak at 271.
Second Methiodide Salt (LXXV)
The second methiodide salt of LXXIV was prepared by adding
excess methyl iodide (4 mL, 9.08 gm, 0.0639 mol) to a stirred solu-
tion (2.80 gm, 0.0103 mol) in 25 mL absolute ethanol. The resulting
solution was refluxed for two hours. Removal of the solvent gave a
yellowish gummy (2.15 gm, 0.00503 mol, 76?) crystals LXXV, which
gave a yellow precipitate with silver nitrate.
Second Hydroxide Salt (LXVI)
To a stirred solution of LXXV (2.10 gm, 0.00508 mol) in
absolute methanol (25 mL) was added 24 gm of anion exchange resin.
After the reaction mixture was stirred overnight at room tempera-
ture, it was filtered and the filtrate was concentrated under
vacuum. The resulting residual oil (LXXVI) was distilled at 120 C.
in an oil bath to give a gummy residue (LXVII) (1.5 gm, 0.00066 mol)
which has a mass spectrum m/e 226 (0.09705). Also a second product
which has a mass spectrum m/e 299 (1.0) was recovered.
1. B. Belleau, T. Lonway, F. Ahmed and A. Hardy J. Med.
Chem. 17, 907 (1974).
2. F. Ahmed and A. Hardy, Acta. Cryst. B31, 2919 (1975).
3. B. Belleau and F. Morgan, J. Med. Chem. 17, 908
4. A. Beckett, A. Casy and N. Harper, J. Pharm. Phar
macol 8, 874 (1956).
5. B. Belleau, Stud. Biophys., 4, 95 (1967).
6. B. Belleau, Isotop. Exp. Pharmacol., Lect. Int.
Conf., 1964, 469 (1965).
7. K. Opheim and B. Cox, J. Med. Chem. 19, 857 (1976).
8. A. Horn and J. Rodgers, J. Pharm. Pharmac. 29, 257
9. S. Shiotani, T. Kometani, Y. Litaka and I. Itai, J.
Med. Chem. 21, 153 (1978).
10. T. Kometani and S. Shiotani, J. Med. Chem. 21, 1105
(1978) . ..............
11. Vera M. Kolb, J. Pharm. Sci. 67, 999 (1978).
12. Vera M. Kolb, J. Pharm. Sci. 73, 715 (1984).
13. Bidlack, J., and L. Abood Life. Sci. 27, 231 (1980).
14. Chow, T., and R. Zukin., Mol. Pharm. 24, 203 (1983).
15. Abraham, R., and P.Loftus, "In Proton and Carbon -13
NMR Spectroscopy." Heyden, London.
16. Martin, G., M. martin, and J. Gousnard, "In 15N NMR
Spectroscopy." Springer-Verlag, Berlin (1981).
17. Morris, G., and R. Freeman, J. Am.Chem. Soc. 103
18. Morris, G., J. Am. Chem. Soc. 102, 428 (1980).
19. Pachler, K., and P. Wessels, J. Mag. Res. 12,
337 (1973). ...
20. Jacobsen, H., S. Linde, and S. Sorenson, J. Mag.
Res. 15, 385 (1974).
21. Garber, A., "IBM Spectroscopy Application Note 6"
22. Schilling, H. Kevin, Masters Thesis University
of Colorado, 1986.
23. C. B. Resse, In "Protective Groups in Organic
Chemistry", J. F. McComie, Ed. Plenum Press.
24. Nasaaki Miyashita, Akira Yoshikoshi, and
Paul A. Grieco, J. Org. Chem., 42,
25. E. J. Corey, A. Venkateswarlu, J. Am. Chem.
Soc., 94, 6190 (1972).
26. C. B. Resse, In "Protective Groups in Organic
Chemistry", J. F. McComie, Ed. Plenum press.
London 1973. Chapter 3. L. F. Fieser and M.
Fierser, "Reagents for Organic Synthesis".
Vol. 1, Wiley. New YorK, N.Y. 1967. Pp 256.
27. A. Debal, T. Cuvigny, and M. Larcheveque,
Synthesis, 391 (1976).
28. E. J. Corey and R. H. Wollenberg, J. Org.
Chem., 40, 2265 (1975).
29. P. E. Sonnet, Synth. Commmun. 6, 21 (1976).
30. J. H. Van Boom, J. D. M. Herscheid, and C.
B. Resse, Synthesis, 169 (1973).
31. H. Alper and L. Drinkes, Synthesis, 81
32. Nasaaki Miyashita, Akira Yoshikoshi,
and Paul A. Grieco, J. Org. Chem., 42,
33. Kelvin K. Ogilvie, Serge L. Beauage,
Aria L. Schifman, Nicole Y. Theriault,
Krishan L. Sadana, Can. J. Chem., 56-,
34. Joseph Weinstock and V. Boekelheide, J. Am.
Chem. Soc., 75, 2546*50 (1953).
35. Hofmann, Ann., 78, 253 (1851).
36. Hofmann, Ann., 78, 11 (1851).
37. Hughes, Ingold, and Patel, J. Chem. Soc.,.1933,
526. ' '
38. Saunders and Williams, J. AM. Chem. Soc., 79,
39. Thodora W. Greene, "Protecting Groups in Organic
Synthesis", 10*87 (1981). By John Wiley and Sons.
40. Arno Liberies, "Theoretical Organic Chemistry",
330^-331 (1968), The Macmillan Company.
41. Peter Sykes,- "A Guidebook to Mechanism in Organic
Chemistry", 250-253 (1981) Longman, London and