A STUDY OF UNCONJUGATED
CHROMOPHORIC INTERACTIONS RELATED
THOMAS WILLIAM BROOKS
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
The author wishes to express his deep appreciation to
Dr. G. B. Butler whose guidance and counsel during the execution
of this work were of inestimable value.
The author wishes also to express his gratitude to the
members of his advisory committee and his fellow graduate
students whose advice and criticism was a constant source of
inspiritation and encouragement.
Special thanks are due Mrs. Marie Eckart for her diligence
and conscientousness in the typing of this dissertation.
Finally, the author wishes to express his great appreciation
to his wife, Lucretia, whose forbearance and understanding during
months of neglect made this work much easier.
The financial support of this research by The American
Chemical Society Petroleum Research Fund is also gratefully
TABLE OF CONTENTS
ACKNOWLEDGMENTS .......... i
LIST OF TABLES .... .......... v
LIST OF FIGURES .......... ... v
I. INTRODUCTION . . . . . 1
A. History of Cyclopolymerization . 1
B. Non-conjugated Chromophoric
Interactions. . . . 11
II. A STUDY OF UNCONJUGATED
CHROMOPHORIC INTERACTIONS .... 19
A. Selection and Synthesis of Model
Compounds . . . 19
B. Results and Discussion . . 25
CS Experimental . . . . 37
III CYCLOPOLYMERIZATION INVOLVING
CUMULATIVE lI2 AND 1,4 ADDITION . 51
A. The Polymerization of
1 3, 8-Nonatriene . . . 51
B., Experimental . . 54
TABLE OF CONTENTS (Continued)
IV. SUMMARY ..... . . . 57
APPENDIX . . . . . . .. 59
BIBLIOGRAPHY .. ... ......... 6Z
BIOGRAPHICAL SKETCH . . . . . . 68
LIST OF TABLES
1. Olefins Found to Undergo Cyclopolymerization. 7
2. Selective Absorption of Butadiene Derivatives . 26
3. Selective Absorption of Styrene Derivatives . . 29
4. Selective Absorption of Methacrylate Derivatives
in Isooctane . . .. . . . 33
LIST OF FIGURES
1. Ultraviolet Absorption Spectra of cis and
trans-1,3,8-Nonatriene in Ethanol . . 27
2. Ultraviolet Absorption Spectra of Styrene
Derivatives in Ethanol . . . . 31
3. Ultraviolet Absorption Spectra of Methacrylate
Derivatives in Isooctane . .. . 34
A. History of Cyclopolymerization
Cyclopolymerization* or intramolecular-intermolecular
polymerization is a phenomenon observed with certain 1,5- and
1,6-dienes by which a polymer is formed having cyclic recurring
units. The process is Illustrated as follows.
* Cyclopolymerization is a somewhat ambiguous term for
that 31% of converted monomer should undergo some kind of
cyclization in the polymerization of diallyl phthalate. Simpson's
results are in fair agreement with the prediction and further work
has been pursued in this direction. 7 Oiwa and Ogatae subsequently
found that diallyl phthalate will yield a polymer by solution polymeri-
zation which is 81% cyclized.
The mechanistic implications found in earlier work were
not fully realized until Butler and Angelo9 made their dramatic
proposal that l,6-dienes could polymerize by an alternating
intramolecular-intermolecular process to afford soluble, fully
saturated polymers containing recurring six-member cyclic units.
The soluble polymers produced by polymerization of diallyl
quaternary ammonium salts were thus accounted for. A rigorous
structure study Butler, Crawshaw, and Miller10 substantiated
the cyclopolymerization hypothesis. The study involved degrading
poly-(diallylammonium bromide) I and poly-(diallyldimethylammonium
bromide) IV via the following schemes.
e e Ococi
H H CON
Butler and IngleyI found that diallyl quaternary ammonium
salts polymerized with free radical initiators to yield water-soluble,
non-crosslinked polymers. They found on the other hand that
monoallyl quaternary ammonium salts would not polymerize and
that the triallyl homologs polymerized to yield crosslinked products.
In view of Staudinger's hypothesis that non-conjugated dienes yield
exclusively crosslinked polymers, it was not expected that any
1,6-diene would yield a soluble polymer. The possibility of
cyclization occurring during the polymerization of a non-conjugated
diene had been recognized even before Butler's original work.
Walling3 had commented on the fact that observed gel-points for
some dienes should occur later than calculated by Stockmayer's4
equation because the equation failed to account for occasional
cyclization. Thus it was found that for the systems methyl methacrylate-
ethylene dimethacrylate and vinyl acetate-divinyl adipate, the ratios
of observed to calculated gel-points were frequently as high as
fifteen to twenty. Simpson, Holt, and Zeite5 -i their polymeri-
;atio.n studies of diallyl phthalate found that 40% of the units were
cyclized in the polymer. Award6 predicted by statistical calculations
intramolecular-intermolecular polymerization, but Sinco the term has
been allowed to slip into the literature and is less cumbersome than
alternative expressions, it will probably enjoy a wide usage except
with the terminological purists.
1. to hydrox.
Z, 2 thermal
N Br decomp.
SH3C CH3J n
H3C I CH3
The structure of III was supported by elemental analysis,
potcntiometric titration, infrared opectral analysis, and the fact
that heating above its miclting point yielded a sublimate of benzoic
acid plus a crosEllnked polymer residue. The structure of VI was
supported by similar data and the fact that thermal degradation of
the corresponding hydro:dde yielded trimethylamine and a crosllinhed
With the work discussed above providing the impetus, a
considerable research effort has been generated in the field of
cyclopolymerization. A polymer of diacrylylmethane has been
found by Jones11 to result from attempts to cause a Claicen
condensation of methyl vinyl ketone and ethyl acrylate. The
expected product apparently undergoes an anionic polymerization
to yield a soluble polymeric material having the properties of a
poly-(dihydroresorcinol). Marvel and Stille12 have demonstrated
that 1, 5-hexadiene and 1, 6-heptadiene yield soluble predominately
saturated polymers by the cyclic mechanism when polymerized with
a Ziegler catalyst. The hexadieno, however, was observed to give
a considerable amount of crosslinked material also. Field13 has
described the synthesis and polymerization of 2, 6-diphenyll1, 6-
heptadiene and has found that a soluble polymer may be derived from
this monomer by all four known modes of chain initiation. Schuller
and co-workers14 have reported copolymers of diallyl compounds
with such conventional vinyl monomers as acrylonitrile which they
claim have cyclic units arising from cyclization of the diene comer.
Soluble copolymers of diallylalkylamine oxides and conventional
vinyl monomers probably contain cyclic units also, though their
properties are not ascribed to this structure. 15 Quite recently the
formation of bicyclic recurring units by the cyclopolymerization
procsoo has been reported. Ball and Harvwood16 claim to have
obtained a polymer of 1, 4-dimethylenecyclohexane in which the
repeating units are norbornyl rings. Trifan and Hoglen17 have
obtained data which strongly suggest that certain triencs and
tctracnes such as triallyl and tetraallyl ammonium bromides may
polymerize in dilute solutions to yield bicyclic and tricyclic
analogs of the poly-(diallylammonium halides).
Table I lists all of the monomers reported up to the present
time which have been shown to undergo cyclopolymerization.
In addition to monomers of the type in Table 1, certain
1,4-dienec may be expected to cyclopolymerize in copolymer systems.
Butler,30 for instance, has found that divinyl ether yields a soluble
polymer when copolymerized with such monomers as maleic
In an effort to account for the unique tendency of the diene
systems under consideration to cyclopolymerize, it has been
proposed30 that a homoconjugative interaction occurs between the
unconjugated ethylenic bonds. A schematic representation of this
phenomenon is shown below.
OLEFINS FOUND TO UNDERGO CYCLOPOLYMERIZATION
N.AUIyl -N 2-chloroallylmorpholinium
2-Carbethoxybicyclo(2 2. 1)-2,5-
C rotylmethyl fumarate
Dia c rylyl methane
_ .... L .__ _._.~._~_~._ .. ..._~ __~.____
TABLE 1 (Continued)
Olefina System Reference
Diallylammronium chloride RI 10
Diallylbenzylamine oxide R 15
Diallylcyclopentamethylenesilane Z Z3
Diallylcyclotetramristhylenesilane Z 22
Dialyldiethylammonium bromide R 9,14
DiaUyldimethylammonium bromide R 9,10
Diallyldimethylsilane Z 23
Diallyldiphonylphosphonium bromide R 19
Diallyl phthalate R 5,8
Diallyldiphenylsilane R, Z 24
Diallylethylamine oxide R 15
Dialylmethylamine oxide R 15
bromide R 19
Diallylphcnylphoslphine o:idc R 20
bromide R 19
bromide R 19
Dimethallylamine oxide R 15
TABLE I (Continued)
Olefina System Reference
Dim ethallyleyclopentamethylenes lane
Dim ethallyldmeothyls iane
DimcthallylphoA phine o::ide
a, c -Dlrmethylenepimelonrit le
2, 5-Dimethyl-1, 5-hexadiene
2, 6-Diphenylol, 6-heptadiene
2, 5 Diphenyl-1, 5 -hexadiene
2, 7-Diphenyl-l, 7-octadiene
R, C, z
TABLE 1 (Continued)
Olefina System Reference
1, 6-Heptadiyne Z 25
1, 5-Hc:dicne Z 24
Mcthacrylic anhydride R 28
I yrceno C 67
aa'-Pimclic acid R 29
p,a'-Pimelic acid methyl ester R 29
aa'-Pimclic acid ethyl etecr R 29
Tetraallylammonium bromide R 17
Trial-ylammonium bromide R 17
3-Vinyl-1, 5 -hedadiene Z 17
a. It will be noted that monomers have been included
which do not yield exclusively linear saturated
b. R = Free radical A= Anionic
Z = Ziegler initiator C= Cationic
,rhile it aoeom presumptuous to suggest that ouch a
phenomenon would contribute greatly to the ground otate of a
molecule, it does not seem unlikely that it would have a stabilizing
influence on the excited state, thus, providing an energetically
favorable path from dienc to cyclic product. Thic subject will be
developed further in the following section of this paper.
B. Non-conjugated Chromophoric Interactions
For many years it was believed a fundamental principle of
ultraviolet spectroscopy that chromophores separated by two or more
single bondo were additive in their absorption of light. 31 As the
literature on ultraviolet spectroscopy grcw, however, there began
to appear some notable c::ccptions to this principle. Ley, Wingchen,
and Dirking, 32, 33 for instance in their otudics of ultraviolet spectra
of hetones found that the absorption maximum of acetone could be
shifted to longer wavelengths by phenyl substitution. Thus they
observed that the maxima for a series of ketonce were in the order
acetone < phenylacetone < dibcnzylhetonc < acctophenone <
dc-o::ybcn-oin < bon=ophononc. A noticeable effect of unconjugatcd
groups on polyene absorption has been observed with certain
carotenoids. 3 Various other instances of apparent c::ccptions to
the additivity principle are extant in the early literature, though they
are few in number. 31
Attempts to interpret the non-additive ultraviolet absorption
of certain chrornophoric systems gave rise to two schools of
thought. Ingold, Shoppee, and Parekh35, 36 must be credited with
the first interpretation, even though it was based largely on chemical
rather than spectral observations. A peculiar lack of olcin
reactivity combined with a facility for cyclization of methyl and
ethylAl1 hexadiene-l:, 3:3,4:4, 6:6-octacarboxylatc was attributed
by the Ingold school to an interspacial interaction of the olefinic
bonds, or what they called a ring-chain mecomerism. A second
interpretation came about as a result of x-ray and ultraviolet
spectral studies on the same dienc by Bateman, Jeffrey, and
och. 37, 38 They found that the x-ray pattern for the ethyl ester
was incompatible with a cis conformation in the crystalline state
and on this basis concluded that Ingold's ring-chain mn nomorism
hypothotis was incorrect. They chose, instead, to interpret Ingold'o
chemical data and their own spectral and x-ray data in terms of a
hyperconjugative effect involving the entire carbon framework of
Draude and co-workers39 then found that 2,5-dihydro-
acetophenones exhibited a bathochromic shift of the carbonyl
aboorption band over that of 1-acetylcyclohexane. It was found,
furthermore, that the 2-substituted dihydroacetophenones did not
chow a shift. They chose to interpret this as a hyperconjugative
effect involving the methylones situated between the ethylcnic bonds
as illustrated below.
0 H 0 0
Oi II 1 'C II1
-C--CH3 -CH --CH
Thus, they were able to account for their own observations as well
as those of Ley et al. 32 33
To extend the hyperconjugative argument Braudc40
undertook an investigation of compounds of the type X- CHI Y and
X- CH2CH2- Y, where X and Y may represent like or unlike
chromophoric groups. Diphonylmethanc and bibonzyl, each of which
is a representative example of the two types, exhibited departures
from additivity in absorption of ultraviolet radiation. In light of
those results and in view of the fact that x-ray data for bibenzyl
ruled out any interspacial ring interaction in the crystalline state,
it was concluded that a hyperconjugative effect was responsible for
the non-additivity of the unconjugated chromophores.
In more recent years there has been a growing tendency for
theoreticians to favor the interspacial resonance interpretation.
A number of systems have been found which exhibit chemical and
physical properties indicative of such a phenomenon. Bartlott and
Lewis41 have observed that the ultraviolet absorption maxima of
trypticin are more intense by 3.3 and 5.5 times respectively and are
shifted bathochromically by nine and seven millimicrons from their
apparent equivalents in triphenylmethane. While it is impossible
to draw any cononical forms for trypticin which involve hyper-
conjugation, no less than one hundred eight forms may be drawn
involving transannular interaction of the aromatic rings. An even
more dramatic example of this phenomenon is furnished by the case
of bicyclo(2. 2. l)hcptadiene VII. WVinstein and Shatavolhyy42,43
have found that this compound adds one mole of such reagents as
bromine or the hydrogen halides to yield predominately saturated
products having a nortricyclenic structure. The reaction is shown
+ BrZ r Br
ca. 207% ca. UO',J
VII trans:cis ca. 2
Heato of hydrogenation experiments by Turner and co-workers4
have established that there is no ground state stabilizing interaction
of the rr orbitals in VtI. Wilcox, Winstein, and Mc'Millan,45 on the
other hand, have been able to establish that such interaction does
have a stabilizing effect on the excited state thus, accounting at
least partially, for the unique chemical behavior of the bicyclic
Spectral data have been reported by others46,47 which also
lend themselves well to interpretation in terms of interspacial
electronic interactions, particularly in rigid cyclic systems.
While most of the evidence for unconjugated electronic
interactions has come from studies of rigid cyclic systems, it does
not seem unreasonable to expect that such a phenomenon might be
operative in open-chain systems as well. Ingold's work in this
regard has already been cited and more recent work is equally
deserving of consideration. It has been shown in a variety of open-
chain compounds that a rr-electron system neighboring a developing
carboniurn ion center can e:ert a significant influence on the course
and the rate of a reaction. Simonetta and V'inotolin4 have noted a
* For leading references on this subject see ref-rence 44.
correlation between this phenonemon and the light absorption
properties of ketones such as VII, which exhibit a bathochromic
shift and an enhanced intensity of the carbonyl absorption frequency.
Cram and T'ilkinson,49 in an effort to understand the nature
of the electronic interactions between the aromatic rings of
paracyclophanes, studied compounds of the type IX.
It was found that those compounds in which n= 1, 23,4 would form
only the r ono(tricarbonylchromium) complex when reacted with
ihe:amcarbouylchromium. When the pa positions were substituted
wv'ith ethyl groups, however, electronic interaction was sufficiently
stcrically hindered for the Tr-syotems to appear independent in
With the data existing at the present time it is questionable
whether or not one would be justified in rejecting the hyperconjugative
interpretation for unconjagated chromophoric interactions. It is
significant, however, that this argument rests largely on the results
of x-ray experiments which measure molecular configuration in the
crystalline state only. Since spectral measurements were made in
solution, where the molecular configuration of an open-chain is not
constrained by crystal forces, one can only assume that one or
another configuration obtains. Consequently, one can only speculate
on molecular configuration in a radiation induced excited state,
particularly in the absence of confining steric effects ouch as found
in ortho-substituted biphenyls. 50 In view of Koch's51 observation
that 1,10-diphenyl-l, 9"decadiene is non-additive in absorption of
ultraviolet light and the work of Cram,49 total rejection of Ingold's37'38
postulated interopacial resonance interaction secms unjustified.
One can only conclude that while the whole problem lacks sufficient
data for an unequivocal answer, there is every indication that
interopacial Tr*orbital interactions can occur to produce unique
chemical and physical properties in certain systems.
A systematic investigation of Tr-bond interactions and the
attending chemical consequences in 6-dienoid cystome has never
boen undertaken, although Butler30 has pointed to number of
instances where data are suggestive of such a phenomenon. Aside
from the many cyclic polymers which have been characterized,
there is only one piece of physical evidence which gives direct
support to the postulated interactions of 6-dienes. Mikulasova
and Hvirik5 found the total activation energy for the polymerization
of diallyldimethylsilane to be nine klocalories per mole per double
bond lower than for allyltrimethylsilane. This represented a decrease
in the overall activation energy of some 30%.
A STUDY OF UNCONJUGATED
A. Selection and Synthesis of Model Compounds
The choice of model molecules with which to study
unconjugated chromophorie interactions such as those proposed by
Butler30 was governed by three major conslderations. The first
consideration was simply the commercial or synthetic availability
of the compounds. Secondly, it was necessary, because of
instrumental limitations, that the chromophores absorb ultraviolet
radiation at wavelengths above 190 millimicrons. Finally, it was
deemed necessary that the compounds studied be as closely related
as possible to system which are Inown to be capable of polymerizing
by the cyclic mechanism. With the above considerations in mind,
three series of compounds were studied spectrophotometrically in the
region of the ultraviolet spectrum extending from 190 to 270
The three series of compounds investigated in this work
were (1) butadiene derivatives, (2) styrene derivatives, and
(3) mcthacrylate derivatives. Synthesis of the butadine derivatives
and the styrene derivatives constituted a -major portion of the
c::perimental work required to gather the spectral data in this
investigation. The methacrylato derivatives, fortunately, were all
commercially available in high purity.
The butadiene derivatives synthesized in the course of this
study were the cis and trans isomers of 1, 3, 8-nonatriene. The cis
icomer was obtained via the following sequence of reactions.
+ Br2 -
NH3 1 Cz2HOH
SAlc. KOH Dr
o OCzH5 O OCzH5
He I HO0
CH2= CHCHI2CHzI2M r
I CH2= CHCH=CHCHZCI
CHZ= CH(CIHZ))3CH =CHCH =CH2
The six step cynthosis of the intermediate pentadienyl
chloride wao first worked out by Woods and Lcderlc54 and latcr
improved upon by Crombie, Harper and Thompson. 55 The latter
authors assigned the trans configuration to the chloride, basing
their assignment largely on infrared data and the chemistry of
related compounds. It was presumed that while the pcrcurcor cyclic
acetal constrained the molecule into the c c configuration, a cis to
trans interconversion occurred during acid hydrolysis of the acetal
to the pentadienal.
On the basis of the evidence obtained in the present work,
the trans assignment for the pentadienyl chloride is believed to be
incorrect. The triene obtained by coupling of the pentadienyl
chloride with 3*butenylmagnesium bromide has definitely been
established to be a geometrical isomer of the trinoc obtained by the
WVittig56 reaction of 5-hexenyltriphenylphosphonium bromide with
acrolc.n. The boiling points, refractive indices and densities of
03 CHe (CIz) =CZ nu-Li 0P=CH(CH2)3CH =CH2
CH =II CHCHZCH CH2 CH=CHCH= CH
the tw/o triones are very similar and each takes up three moles of
hydrogen when hydrogenated over Adam's catalyst. The infrared
spectra of the two isomorrs are almost identical, vwith the most
significant difference being a strong band at 965 cm. present in
the material obtained via the T.ittig reaction and absent in the other.
The 965 cm"' band is known to be assignable to the C -H out-of-plane
deformation vibration for disubstituted ethylenic structures having
the trans configuration. 57 Furthermore, it is known that the Ttittig
reaction favors the formation of the trans olefin, whereas acid
hydrolysio of the cyclic acetal precursor to pentadlenal should favor
production of the cis olefin.56 It was established by gao-liquid
chromatography that the isomers were each produced in a
geometrically pure otatc. This appears to be the first instance
where the pure trans olefin has been isolated from a tVittig reaction.
A possible rationalization of this observation lies in the fact that the
conditions of low temperature and short condensation time under
which the triene was prepared in this work favor formation of the
most Idnctically stable product, v'hile the usual conditions of high
temperature and long reflux periods might allow time for some
equilibration between the cis and trans forms.
The nuclear magnetic resonance spectra of the two isomers
produced some anomalous data which should be noted here. In the
case of the cis isomer the peak area ratio for apz to sp3 bonded
hydrogen was found to be 4:3 in agreement with theory. The same
ratio for the trans isomer, however, was found to be 1:1. In the
latter case, a search for -CH3 resonance was unsuccessful, thus
ruling out the possibility of double-bond rearrangement to an internal
position. Cyclic structures are ruled out on the basis pf quantitative
hydrogenation results, leaving no other obvious choice than to assign
to this compound the structure of trans-1,3, 8-nonatriene.
A third method for producing 1,3, 8-nonatriene was tried
withe;t success, but the results are sufficiently interesting to be
included in this discussion; An attempt was made to produce the
desired triene by pyrolysis of 3-acetoxyl, 8-nonadiene at 450o0 The
pyrolysate turned out to be a three component mixture of hydrocarbons
which could not be separated with available distillation and
preparative gas-liquid chromatographic columns. The components
were separable on an analytical Carbowax ZOM-packed gas-liquid
chromatographic column, making it possible to rctimate the mole
percent composition of the mixture. Two of the components
represented 90% of the mixture and were in a ratio of approximately
2:3. The mixture analyzed for C9H14i gave a molecular weight
value of 122, and took up one mole of hydrogen when hydrogen-
ated over Adamis catalyst. A nuclear magnetic resonance
spectrum of the mixture gave a ratio of sp2 to sp3 bonded
hydrogen of 6:1 in good agreement with the quantitative hydrogenation
results. At this point the investigation of this rather unusual reaction
was discontinued, although from the data which was obtained it would
appear that the pyrolsate is a miatture of bicyclic compounds. The
two most likely compounds are bicyclo(3. 3. l)-*2nonene and
tetrahydroindan both of which could reasonably be formed by an
internal Diels*Alder reaction of 1,3,8-nonatriene in the vapor phase.
Syntheses of the styrene derivatives for this study were
accomplished via the following sequence of reactions.
CH=C- CH3 N-bromosuccinimide CH= C -CHBr
CHI=C-CH2Br + RPJM r --- CH= C-CHgR
R= CH3CHZCHI CH2=CHCH2-, CH2=C CH2-
Preparation of the intermediate 2*phenyl-*3-bromnopropene
was performed according to the method of Hatch and Patton8 and
was found to be contaminated with 0-bromo-ca-methylstyrene.
Pines, Alul and Kalobelsdi59 have studied this same reaction and
have described the isolation and identification of the impurity. In
this work no attempt was made to remove the p-bromo- a-methyl-
styrene because it involved a rather tedious and uneconomical process
using large scale column chromatography. The subsequent coupling
reactions with Grignard reagents using the impure allylic halide
were not hampered except for lowering of the yields. The final
products were obtained in good purity by careful fractionation of the
crude materials and selection of the fractions with the aid of
gas-liquid chromatography and index of refraction measurements.
B. Results and Discussion
Data for the ultraviolet absorption spectra of cis- and trans-
1, 3, 8-nonatriene are found in Table Z and the spectra are shown in
Figure 1. The calculated absorption maxima were determined
according to Woodward's Rule60 61 for the prediction of ultraviolet
absorption maxima of alkyl-substituted butadienes. Taking butadiene
as the standard with a maximum of Z17 millinicrons, the calculated
maxima are determined by adding five millimicrons per alkyl
substituent. Many examples of the excellent agreement between
calculated and observed values may be found in the monograph on
ultraviolet absorption spcctroscopy by GUlam and Stern. 62 Included
in Tablo 2 are several selected examples reported in the literature
which serve as comparators for the compounds studied in this work.
SELECTIVE ABSORPTION OF BUTADIENE DERIVATIVESa
Compound Calcd. Obs. x 104 Reference
Piperylene 222 223.5 2.30 63
Isoprene 222 222 2.39 64
Myrcene 222 224.5 1.70 65,66
ci 4, 3, 8-Nonatriene 222 225.5 2.03
trans-l3 3, 8-Nonatriene 222 227 2.10
a.~ All ~ dat were- deemie In ethano exep for
a. All data were determined in ethanol except for
isoprene which was determined in hexane.
b. Wavelengths are in millimicrons.
3. 8 -
210 220 230 240
Figure 1. Ultraviolet Absorption Spectra of cis
and trans-1,3, 8-Nonatriene in Ethanol
Myrcene and the cis and trans isomers of 1,3,8-nonatriene
all exhibit a distinct departure from VToodward's Rulc in their
absorption of ultraviolet radiation. In the case of myrcene the
departure is small and amounts to a bathochromic shift of only
2.5 millimicrons. It is significant, however, that Marvel and Hwa67
have reported polymerizing this compound by boron trifluoride
initiation to a soluble, apparently cyclic polymer. In the cases of
cis- and trans-l, 3,8-nonatriene the shifts amount to 3.5 and 5.0
millimicrons respectively. The trans isomer has been polymerized
in this work (see Chap. IIl) by a Ziegler catalyst initiator to yield
a polymeric product approximately half of which is soluble and
exhibits the properties expected for a cyclic polymer.
Proceeding now to the styrene derivatives, it can be seen
from Table 3 that two effects seem to be operative in their light
absorption properties. A marked steric effect is observed when
comparing styrene and a-methylstyrene in that a hypsochromic
shift of the so-called K-band for styrene is produced when the a
position is substituted with a methyl group. When the bulk of the
group is increased by extension to n-pentyl the magnitude of the
shift is correspondingly increased. The nature of this steric effect
is fully discussed in the literature,50 6Z so it Is sufficient to say here
that it arises because of the tendency for the bulky a. substituents to
SELECTIVE ABSORPTION OF STYRENE DERIVATIVES
E x 104
a. Wavelengths are in millimicrons.
force the olefinic carbon atoms out of coplanarity with the aromatic
The second effect is more subtle in that it manifests itself
only in the intensities of the maxima and not in their location. In
going from 2-phenyl-l-hexene to 2-phenyl-l, 5-hexadiene and
2, 5-dlphenyl-l,5-hexadiene, the intensities of the K-band increase
when the spectra are measured in ethanoLI In the non-polar solvent,
isooctane, there is almost perfect additivity. Since this trend in
the intensities is roughly in the same order as would be expected for
the tendencies of these models to cyclopolymerize, it is felt that
the data does hold some significance. Since the magnitude of the
absorption intensities are, at least to a degree, dependent upon the
polarity of the excited state,68 it seems conceivable that any effect
which tends to enhance this polarity (such as an interspacial
interaction of the absorbing chromophore with a neighboring
chromophore) might also enhance the intensity. That this effect is
observed in a solvent with a high dielectric constant and not in
isooctane might be expected if the effect arise purely from the
polar condition of the excited state.
The spectra obtained for the styrene derivatives prepared
in this work are presented in Figure 2.
2-Phenyl-1, 5-hexadiene -----
230 240 250 260
Figure 2. Ultraviolet Absorption Spectra of Styrene
Derivatives in Ethanol
In the methacrylate series it is again seen that the spectral
data follow the same trend that would be expected for the tendencies
of the models to undergo cyclopolymerization. A bathochromic
shift of the maximum with methacrylic anhydride is observed and a
slight enhancement of the extinction coefficients is seen to occur
in going from the monoolefin to the diolefins. The spectral data for
the methacrylate models are shown in Table 4 and the spectra are
presented in Figure 3,
It is felt that the spectral data shown and discussed above
are significant evidence for the unconjugated electronic interactions
proposed by Butler30 as an explanation for the propensity of certain
unconjugated diolefins to undergo cyclopolymerization. The
bathochromic shifts and enhanced absorption intensities all follow
a pattern which would be expected for such electronic effects.
The data may also be taken as further evidence against the hyper*
conjugative interpretation of non-additive light absorption discussed
in Chapter I. If one compares the data for piperylene with that for
1,3,8-nonatriene, for example, the hyperconjugation theory fails
to correctly predict which model will absorb at the longer wave-
length. In the cases of the methacrylate and styrene models, where
hyperconjugative influences are more or less constant, marked
deviations from additivity are still observed.
SELECTIVE ABSORPTION OF' METHACRYLATE DEi'JVATIVE3
Compound ax.x x10
Ethyl nethacrylate 206 0, 7Z
Allyl methacylate Z05 0 85,
-lethallyl methacylate 204. 5 0 95
1.1ethacrylic anhydride ZO 1.50
a. Wavelengths are in mnillmicrons,
3. 7 / \
// \ \
3. / 3\
I I -I I
200 210 220 230
Figure 3. Ultraviolet Absorption Spectra of
Methacrylate Derivatives in Isooctane
In the styrene serilo of models the departures from additivity
are admittedly small and it may be argued that too much significance
is being attached to them. If the data for this series is taken by
itself, such an argument might seem just, but in view of the data
for the other models it is felt that the results are significant in that
they fit the general trend seen in all of the spectral results. The
weakness of the effect in the styrenes may arise partially from the
fact that interaction is between 1,5 situated olefinic bonds rather
than 1,6, The phenomenon of cyclopolymerization is likewise more
pronounced when the olefinic bonds are 1,6.
All things thus considered, the data discussed above are
taken to mean that in diene systems where the double bonds are
situated 1,5 or 1,6 with respect to one another, the Tr systems may
interact intramolecularly. The chemical consequences of this
interaction and the attending stabilizing effect are revealed in the
phenomenon of cyclopolymerisation. In view of Winstein's42,43,45
studies in non-classical carbonium ion systems it is believed that
this effect is either peculiar to the excited (or activated) state of
the reacting molecules or to the radical or ionic transient
intermediates produced during the propagation step of the polymeri-
zation. An illustration of these two possibilities is shown below for
a radical propagated reaction.
In the first case the propagation may be considered as
occurring via a stepwise process with intramolecular reaction favored
over the intermolecular, thereby leading to ring closure and hence
the cyclic recurring unit, In the second case the propagation may be
considered as a concerted mechanism leading to cyclic product
completely exclusive of an intermolecular competing step. No data are
available at the present time with which to distinguish between these
Source and Purification of Materials. Ethylene oxide, allyl
bromide, vinylmagnesium chloride, 5-hexen-1-olS and triphenyl
phosphorus were obtained from Peninsular Chemrnesearch,
Incorporated. The allyl bromide was distilled before use and the
vinylmagnesium chloride, which was obtained as an approximately
3.5 molar solution in tetrahydrofuran, was analyzed for concentration
by the double titration method.
Dihydropyran was obtained from the Chemicalo Division
of The Quaker Oats Company and was used as received.
Tetrahydropy ran- 2-methanol and ac rolein vierc obtained
from Carbide and Carbon Chemicals Company. The tetrahydropyran-
2-methanol was distilled before use and the acrolein was used as
Acetyl chloride, a-methylstyrene, and N-brornosuccinimide
were obtained from the Distillation Products Industries division of
Eastman Kodak Company and were used as received.
n-Propyl bromide was obtained from Columbia Organic
Chemicals Company and was distilled before use.
Allyl methacrylate, ethyl methacrylate, and methacrylic
anhydride wvere obtained from Monomer-Polymer Laboratories,
Chemicals Division, Borden Company. Gas chromatography indicated
these compounds to be sufficiently pure for ultraviolet spectral
Equipment and Data. Temperatures reported in this paper
are uncorrected and are recorded in degrees centigrade.
Infrared data were obtained with a Perldn-Elmer Infracord
Double-beam Infrared Recording Spectrophotometer or a Perkin-Elmer
Model 21 Double-beam Infrared Recording Spectrophotometor.
Ultraviolet data were obtained with a Bausch and Lomb
Spectronic 505 Double-beam Recording Ultraviolet-visible
Spectrophotometer equipped with an IP28 photomultiplier and an
air-cooled hydrogen lamp-tungsten lamp combination light source.
Gas-liquid chromatographic analyses were made with a
Wilkens Aerograph Model A-10O-C Gas Chromatographic Instrument
using helium for the eluent gas. Unless otherwise indicated, gas
chro-matographic analyses were made on a five foot column packed
with 20j Silicone GE SF-96 on fire brick. (Reference to this
technique is abbreviated G. L. C. in this paper.)
Nuclear magnetic resonance data were obtained with a
Varian V-4302 High Resolution INuclear Magnetic Resonance
Elemental analyses were performed by Galbraith Laboratories,
Synthesis of 1-Bromo-3-butene. 3*Buten-l-ol was
prepared in yields of 60-70% from vinylmagnesium chloride and
ethylene oxide according to the method of Ramsden, t al. 70 and
converted to the bromide in 55% yield by the general method of
Gaubert, Linstead, and Rydon.71 B.P. 96-99, nZ 1.4652, lit. b.p.
98.5-990, n0 1.4621.
Synthesis of 2,4-Pentadienyl Chloride. 2,4-Pentadienyl
chloride was synthesized in six steps by the method of Woods and
Lederle54 as modified by Crombie, Harper, and Thompson. 55
B.P. 46-470/54 mm., n12 1.4951, lit. 55 b.p. 80-820/240 mm.,
Synthesis of cio-l,3, 8-Honatriene. The Grignard reagent
of l-bromo-3-butenc was prepared in the usual way by adding a
solution of 20 g. (0.15 mole) of the halide in 35 ml. of absolute
ethyl ether to a mixture of 3,5 g. (0.14 atoms) magnccium. sand
and 25 ml. of absolute ethyl ether. A solution of 10.0 g. (0.10 mole)
2,4-pcntadienyl chloride in 25 ml.! of absolute other wvas then added
dropwise to the stirred Grignard solution at such a rate as to cause
gentle refluxing of the ether. Shortly after addition of the chloride
was completed the reaction mixture became a slurry and was heated
to a gentle reflux for 30 min. longer. The mixture was then cooled
in an ice water bath and decomposed by careful addition of chilled
dilute hydrochloric acid until all of the solid material was dissolved.
The ethereal layer was separated the aqueous layer extracted with
25 ml. of ether and the two ethereal portions combined. The ether
solution of crude product was washed with 10. sodium bicarbonate
solution and distilled water and dried over anhydrous sodium sulfate.
The ether was removed by distillation and the residual liquid was
distilled at reduced pressure through a 23 plate pinning band column.
The fraction collected at 65-670/41 mm. was refractionated to
yield 3.0 g. (25%) of the desired product b. p. 65-670/41 mm.,
n20 1.4694, D20 0.891. The compound, which was found to be 99 -t
pure by G. L.C. analysis, took up 3.18 moles of hydrogen per mole
of sample. The infrared spectrum had the following absorption
bands assignable to the proposed structure:57 1640 cm. "
(C=eC, non-conj.), 1000 cm.'1 ( -CH=CHZ), 913, 900 cm."1
( -CH=CH2), An N. M. R. spectrum of the compound gave an
integrated peak area ratio for sp2 to sp3 hydrogen of 4:3. Data on
the ultraviolet spectrum are found in Table 2 and the spectrum in
Anal. Calcd. for C9H14: C, G88.45; H, 11.55.
Found: C, 88.26; H, 11.39.
Synthesis of l-Bromo-5-hexene. l-Bromo-5-hexene was
prepared essentially as described by Butler and Price72 with the
following modifications. The 2-chloromethyl tetrahydropyran was
prepared according to the procedure of Crombie metal. 73 and the
5-hexen-l-ol was converted to the alkenyl bromide by the general
method of Gaubert, Linstead, and Rydon. 71 (5-Hexene-l-ol
purchased from Peninsular ChemResearch, Xncorporated was also
used in this synthesis.) With these modifications the yields
reported by Butler and Price were improved by approximately
two-fold (50-60%) and the general method was considerably
simplified. The product had b. p. 460/14 mm., n3 1.4640, lit.
b.p. 76-780/45 mm., n5 1.4630.
Synthesis of 5-Hexenyltriphenylphosphonium Bromide. *
A 500 ml. resin kettle was charged with 105. 6 g. (0. 65 mole)
l-bromo-5-hexene, 170 g. (0.65 mole) triphenyl phosphorus, and
200 ml. of anhydrous xylene and the resulting solution was stirred
and refluxed for 10 hr. The mixture was then allowed to cool to
room temperature at which time a lower viscous layer of liquid
developed and became semi-solid on standing. The supernatent
liquid was poured from the kettle and the vessel was evacuated with
a water aspirator and heated to drive off most of the remaining
xylcne. The kettle was then evacuated with a vacuum pump and
heated to remove the last traces of xylene after which the crystalline
residue was pulverized, washed with benzene and vacuum dried in a
desiccator over phosphorus pentoxide. The yield of crystalline
phosphonium salt was 237 g. (86%) and had a m. p. 160-162.
The infrared spectrum of the salt agreed very closely with that of
allyltriphenylphosphonium bromide, a known compound.56
Anal. Calcd. for C24H26BrP: C, 67.88; H, 6.17; Br, 18.84;
P, 7.09. Found: C, 67.33; H, 6.06; Br, 18.91; P, 7.34.
Synthesis of trans-1, 3,8-Nonatricne. The procedure for
the preparation of this compound is essentially that devised by
Hauser and Raymond. 74 A stirred slurry of 42.4 g. (0. 10 mole)
5-hexenyltriphenylphosphonium bromide in 200 ml. of absolute
ethyl ether in a dry nitrogen atmosphere was treated with 85 ml.
of an approximately 15% solution of n-butyl lithium in hexane. The
resulting deep orange solution of the triphenylphosphorus-5-
hexenylide was cooled to 100 by means of an ice water bath and a
solution of 6.7 g. (0.12 mole) acrolein in 25 ml. of absolute ethyl
ether was added over a 30 sec. period followed 15 sec. later by
100 ml. of 2 molar ammonium chloride and finally 100 ml. of
water. Addition of the acrolcin caused discharge of the orange
color and formation of a fine white precipitate. Upon addition of
the ammonium chloride solution and the water, the precipitate
went into the aqueous layer as a gummy v'hite mass. The liquid
layers were poured away from the solid, separated and the aqueous
layer was extracted with 50 ml. of ether. The solid phaeo of
phosphine oxide and lithium bromide was washed with 50 ml. of
ether and the washings combined with the original organic layer and
the ether extract. The combined cthercal portions wore then dried
over anhydrous sodium sulfate, filtered and inhibited with 0. 2 g.
of hydroquinone. The ether and hexane were distilled through a 30
cm. vacuum-jacketed Vigreux column until the pot temperature rose
to 700. The remaining liquid was then distilled through a 23 plate
spinning band column under reduced pressure to yield 2.0 g. of a
clear, colorless liquid b.p. 68-690/43 mm., 20 1.4721, D20 0.771.
G. L. C. analysis of the product revealed only one
component and quantitative hydrogenation resulted in a hydrogen
uptake of 3.00 moles per mole of sample. The infrared spectrum of
the material showed the following bands assignable to the proposed
structure:57 1640 cm.,' (C=C, non-conj.), 1595 crn.'1 (C=C C=C),
999 cm."1 ( -CH=CH2), 965 cm.1" ( =C -, trans), 900 cm. '
( -CH=CH2). An N.M.R. spectrum of the compcu'd gave an
integrated peak area ratio for sp2 to Ep3 hydrogen of 1:1. G. L. C.
analysis of a mixture of this triene and that prepared by the Grignard
reaction showed two narrowly spaced peaks. Data on the ultraviolet
spectrum are found in Table 2 and the spectrum in Fig. 1.
Anal. Calcd. for C9H14: C, 88.45; H, 11.55.
Found: C, 00.36 H, 11.77.
Synthesis of 1, 8-Nonadien-3-ol. The procedure for the
preparation of 1, 8-nonadien-3-ol is essentially that devised by
Nazarov and Kakhniashvil75 for the reactica of Grignard reagents
with a, B* unsaturated carbonyl compounds. The Grignard reagent
of 1-bromo-5-hexene was prepared in the usual manner by the
addition of a solution of 325 g. (2. 0 moles) of l-bromo-5-hc::cno in
300 ml. of absolute ethyl ether to 73 g. (3.0 g. atoms) of magnesium
turnings in 500 ml. of absolute ethyl cther. V;'hen addition of the
halide was completed and spontaneous refluwdng of the ether ceased,
the mixture was heated to a gentle reflux for two hr. longer. A
solution of 95 g. ( 17 moles) acrolein in 950 ml. absolute ethyl
ether was then added dropwise to the Grignard solution over a
period of 10 hr. The resulting mi:ture was stirred for approximately
10 hr. at room temperature, cooled in an ice vrtcr bath and
decomposed carefully with just enough dilute hydrochloric acid to
dissolve all colids. The ethereal layer was separated, the organic
layer extracted once with 100 ml. of ether and the combined ethereal
portions dried over anhydrous magnesium sulfate. The cther was
then removed under vacuum in a flash evaporator and the residual
liquid fractionated through a 30 cmo vacuurn-jacheted Virgre : column
to yield 134 g. (56%) of the alcohol, b.p. 83-860/10 mm., n3 1.4545.
G. L C. analysis showed the material to be 99+ pure and the
infrared spectrum agreed with the expected structure.
Anal. Calcd. for C9H160: C, 77.09; H, 11.50.
Found: C, 77.30; H, 11.69.
Synthesis of 3-Acctoxy-l, 8-nonadieno. A solution of
134 g. (0.96 mole) 1,8-nonadien-3-ol, 103 g. (1.30 moleo)
anhydrous pyridino, and 300 ml. of absolute ethyl ether was
maintained at 00 by an ice vrater bath and to it was added dropwise
94 g. (1.2 moles) of acetyl chloride. The resulting slurry was
stirred for 9-10 hr. at room temperature and then poured over
cracked ice. The ether layer wan separated and the aqueous layer
extracted with ether, The combined ether layer and ether extract
vwac w'ashcd with dilute hydrochloric acid, and saturated sodium
bicarbonate solution and dried over anhydrous sodium culfate. The
theor was removed under vacuum with a flash evaporator and the
residual liquid distilled through a 30 cm. vacuum-jacklted
Vigreu:: column at reduced pressure. The yield of ester was 142 g.
(81%), b.p. 62-640/1 mm.. nD 1.4405. The product was found
by G, L.C. analysis to be 99+% pure and the infrared spectrum
agreed with the expected structure.
Anal. Calcd. for C IH 102: C, 72Z49; H, 9.96.
Found: C, 72.51; H, 10.13.
Pyrolysis of 3-Acetoxy-l, 0-nonadienc. The apparatus
employed for ester pyrolysio is essentially that described by Bailey
and King.76 In a typical run 36.4 g. (0. 20 mole) of 3-acetoxy-1,8-
nonadicnc was dropped at the rate of 1.0 g. per min. into the pyrolysis
tube at 450 + 5 with an external nitrogen flow rate of 40 ml. per
min. When all of the ester had been discharged from the addition
funnel the furnace was allowed to cool to room temperature and the
cracking tube washed down with ethyl ether. The pyrolysate was
then recovered from the cold traps, combined with the ether
washings, and washed free of acid with 10% sodium bicarbonate
solution and finally with distilled water. The ethereal pyrolysate was
dried over anhydrous sodium sulfate, the ether removed by distillation,
and the residual hydrocarbon mixture distilled under reduced pressure
through a 23 plate spinning band column. A 12. 8 g. fraction b. p.
63-67/25 mm. was collected in addition to 6. 3 g. of uncracked ester
for a material balance of 70%. G.L. C. analysis with a 10 ft. 1/4 in.
diameter column packed with 20% Carbowax 20M on fire brick revealed
the product to be a mixture of two components present in an integrated
peak area ratio of 2:3. Upon quantitative hydrogenation the mixture
took up 1. 13 moles of hydrogen per mole of sample based on the
assumption that the two components are isomers. A cryoscopic
molecular weight determination in benzene gave a value of 122 vs.
a calculated value of 122. The nuclear magnetic resonance spectrum
showed the ratio of sp2 hydrogen to sp3 hydrogen to be 1:6.
Anal. Calcd. for C9Ht4: C, 88.45; HI, 11.55,
Found: C, 88.59; H, 11.54.
Attempts to carry out the pyrolysis at lower temperatures
resulted in smaller conversions of the ester and greater complexity
of the pyrolysate.
Synthesis of 3-Bromo-2-phenylpropcne. Preparation
of 3-bromo-2-phenylpropene was accomplished via the
N-bromosuccininnide bromination of a-methylstyrene according
to the procedure of Hatch and Patton. 58 The product had the
physical properties reported by these authors, but was contair.inated
with B-bromo-* -methylstyrene as was found by Pines, et al. 59
in their studies of this reaction. G. L. C. analysis with available
columns failed to resolve the two halides so an estimation of purity
was not possible. The impure product was, however, successfully
used to prepare desired model compounds in low yields. The
properties of the impure product used in this work were b. p.
89-910/5 mm.. n 1.5869, lit.58 b.p. 900/5 mm., n .0 1.5369.
Synthesis of 2-Phenyl- -hex3eno. The Grignard reagent
of n-propyl bromide was prepared in the usual manner from 27.6 g.
(0.23 mole) of n-propyl bromide in 30 ml. of absolute ethyl ether
and 1 g. (0. 21 g. atom) of magnesium turnings in 45 ml. of
absolute ethyl ether. A solution of 29.6 g. (0.15 mole)
3.bromo-Z-phenylpropene in 30 ml. of absolute ethyl ether was then
added dropwise to the Grignard solution at a rate such as to cause
gentle refluxing of the ether. When addition of the allylic halide
was complete the resulting mixture was refluxed overnight, cooled
in an ice water bath and decomposed by cautiously adding dilute
hydrochloric acid. The organic layer was separated, the aqueous
layer extracted with ether and combined organic layer and ether
extract washed free of acid with l07' sodium bicarbonate. The
ethereal product was then washed vith water, dried over anhydrous
sodium sulfate and the ether removed under vacuum with a flash
evaporator. The residual liquid was then distilled under reduced
pressure through a 23 plate spinning band column to yield five
fractions boiling from 830/8 mm. to 900/8 mm. totaling 12 g.
(50T7) in weight. Each of the fractions was found by G. L. C. and
index of refraction measurements to be slightly contaminated with
P-bromo- a-methylstyrene. Redistillation of the middle fractions
yielded a 1.0 g. fraction of analytical purity, b.p. 89-900/10 mm.,
n 1.5254. The ultraviolet spectrum is shown in Fig. 2 and other
data in Table 3.
Anal. Calcd. for C12H16: C, 89.94; H, 10.06.
Found: C, 89.96; H, 10.10.
Synthesis of 2-Phenyl-1, *5hexadiene. The preparation of
2-phonyl-1, 5-hexadiene was accomplished according to the method
of Pines, et al. 59 The product was obtained in analytical purity as
described for Z-phenyl-l-hexene, and had a b. p. 89-900/9 mm.,
nj0 1.5361, lit.59 b.p. 1040/10 mm., n0 1.5314. 17hile the rather
large discrepencies in physical properties noted here raise a
question as to the purity of the sample, G. L. C. analyois as well
as infrared and elemental analysis indicate the sample prepared in
this work to be of high purity. The ultraviolet spectrum is shown
in Fig. 2 and other data in Table 3.
Anal. Calcd. for C1H14: C, 91.03; H, 8.92.
Found: C, 90.91; H, 8.68.
Synthesis of 2,5-Diphenyl-1,5-hexadienc. A solution of
59 g. (0.3 mole) 3-bromo-2-phcnylpropen in 150 ml. absolute
ethyl ether was added dropwise to a stirred mixture of 5.8 g.
(0.15 g. atoms) magnesium turnings in 50 ml. of absolute ethyl
other over a period of 2 hr. The resulting mixture was then treated
with 0.9 g. of anhydrous cobalt chloride77 causing a mildly exothermic
reaction which kept the ether refluxing for approximately 30 min.
The resulting black-brown mixture was then refluxed for two hr.
after which it wao cooled in an ice water bath and decomposed with
dilute hydrochloric acid. The organic layer was csparated and the
aqueous layer extracted twice with 25 ml. portions of ethyl ether. The
combined organic layer and ether extracts were washed with 10%
sodium bicarbonate and water and dried over anhydrous sodium sulfate.
The ether was then removed under vacuum in a flash evaporator and
the dark brown residual oil distilled at reduced pressure. After a
forerun of volatile materials, a fraction came over at 150-160/1 mm.
%vhich crystallized upon cooling. The solid material, 2* 8 g. (8.3%)
was crystallized four times from methanol to afford a white crystalline
product which had a m. p. 48.5-49. 0, lit. 53 m. p. 51.0-51.30. The
infrared spectrum of the product agreed with that reported by
Marvel and Gall53 and G. L. C. analysis showed only one peak. The
ultraviolet spectrum is shown in Fig. 2 and other data in Table 3.
Anal. Calcd. for C18H18: C, 92.26; H, 7.74.
Found: C, 92.42; H, 7.94.
Synthesis of Methallyl Methacylate. Methallyl
methacrylate is commercially available at relatively low prices,
and was synthesized in this work merely to expedite the study. The
ester was prepared from methacrylic anhydride and methallyl alcohol
in the usual manner and was fractionated through a 23 plate spinning
band column to obtain an analytically pure sample, b. p. 580/15 mm.,
nD 1.4415. lit. b.p. 630/17 mm., n0 1.4400. The ultraviolet
spectrum is shown in Fig. 3 and other data in Table 4.
CUMULATIVE 1, AND 1,4 ADDITION
A. The Polymerization of 1,3,8-Nonatriene
1,3, 8-Nonatriene is a monomer which is unique in that it
is potentially capable of undergoing cyclopolymerization by a
cumulative 1,2 and 1,4 addition during the propagation stop or
cstps. Marvel and Hwa67 have already reported observing ouch a
phenomenon in the boron trifluoride initiated polymerization of
myrcene which is shown below.
+ I I %C.
--C-C C C
IIB- I I
BF3-C- C C
As it turns out the resulting polymer has the same structure
as one reported by Robert. and Day which was produced by the
aluminum chloride initiated polymerization of p-pinene. The
cationically initiated polymerization of B-pinene which was studied
by the latter authors is shown below.
HAlC13 A1C13--- H2C
A1C13 ---- 3
CH2 C AIC13- CHZ C
C n C
The polymerization of 1,3,8-nonatriene was successfully
accomplished in the case of the trans isomer by employing the
Ziegler-type initiator system, titanium tetrachloride-triethylaluminum.
Two attempts to polymerize the cis isomer under conditions
equivalent to those employed for the trans form resulted in
approximately 10% conversion to insoluble polymer. Polymer-
ization of the trans isomer afforded a 20% conversion to
polymer approximately half of which was benzene-soluble.
No attempt was made to learn the optimum conditions for carrying
out the polymerizations. A comparison of molecular models of the
cis and trans isomers of 1,3, 8-nonatriene does not give any
indication as to why the trans form should polymerize any more
readily than the cis. There seems to be no reason why under
appropriate conditions the cis isomer could not give a good conversion
The soluble portion of the material obtained by polymerizing
1,3, 8-nonatriene has an infrared absorption spectrum which clearly
agrees with the structure which would be expected for a cyclic
product. The overall reaction is shown below.
S Al(CzHs ,)3
A band at 1650 cm. in the infrared spectrum is assignable
to the residual double bond and a strong band at 965 cm. "I is
assignable to the trans configuration about that bond. 57 molecular
models of the recurring units show that the trans configuration gives
the least amount of steric crowding in the system and allows complete
freedom of rotation about the carbon-carbon bonds adjacent to the
ring, while the cis form produces just the opposite condition. In the
cis form the rings must fold back on one another for ring closure to
occur in adjacent units producing a highly crowded condition which
would probably act to inhibit ring formation.
The absence of terminal unsaturation in the soluble polymer
was evidenced by the absence of an infrared absorption band in the
1000 cm. region and only a weak band in the 900 cm. -1 region.
These bands which are highly characteristic of terminal unsaturation 57
were present in the monomer as well as in all of the other terminal
olefins prepared in this work.
Source and Purification of Materials. Titanium
tetrachloride was obtained from Peninsular ChemResearch,
Incorporated and was distilled before use.
Triethyl aluminum was obtained as a 25% solution in
heptane from Hercules Powder Company and was used as received.
The preparation and purification of 1, 3, 8-nonatriene is
described in the experimental section of Chapter II,
Solvents and other materials were obtained from stock.
Equipment and Data. Intrinsic viscosity measurements
were made with a Cannon-Ubbelohde semi-micro dilution viscometer.
Other information on equipment and data are given in the
experimental section of Chapter Il.
Attempted Polymerizations of cis-1, 3, 8-Nonatricne. A
small serum capped vial was charged by means of a micro*
hypodermic syringe with 0. 0028 g. (0.07 millimrle) of titanium
tetrachloride and 0.08 ml. of heptane containing 0.013 g. (0. 11
millimole) of triethyl aluminum in a dry-box under a dry nitrogen
atmosphere. The vial was further charged with 1.0 mL of high
purity hexane and 0.61 g. (5 millimoles) of the tricnc, slaken vwll,
and allowed to stand at 250 in the dry-box for a period of 72 hr. The
mixture was then poured into methanol, filtered on a scintered glass
dic and the residue washed with methanol. The residue was a
tough plastic substance which could not be ground up. The material
was placed in a soxhlet extractor and extracted for two days with
hot benzene. The bcnzene extract was then poured into methanol
causing less than a milligram of colid to be precipitated. The solid
which was recovered from the e::traction thimble and dried appeared
unchanged. A second attempt to polymerise thio monomer at a
concentration of 10% in he:rane gave essentially the same results.
The crosslinkcd material which was insoluble in several organic
solvents was obtained in the amount of 0.05 g. (8.5%) and was not
investigated any further.
One attempt was made to polymerize this mnonomer in bulk
with 0. 1% azo-bis-iabutyronitrilo and was not successful after one
week at 80.
Polymerization of tran1,, 3, 8-Nonatriene. A ocrevcap vial
vwa charged by means of a micro-hypodermic Gyrineg with 0.056 g.
(0. 15 millimole) titanium tetrachioride, 0. 08 ml. containing 0. 013 g.
(0. 11 millimole) of triethyl aluminum and finally with 3. 6 ml. of
high purity hex:ane and 1.0 g. (8 millimoles) of trans_-1,3,8-
nonatrienc. All of the operations were carried out in a dry-box under
an atmosphere of dry nitrogen.. The mixture was shaken well and
allowed to stand for 40 hr. at 250 after which it was decomposed by
pouring into methanol, The gclatinous material was filtered off,
washed with methanol and allowed to stand in dry benzena for 10 hr.
vith occasional vigorous agitation. The remaining undissolved
polymer was then filtered off and the filtrate poured into methanol to
afford 0.1 g. of clear colorless tacky polymer which after a second
precipitation from benzene into methanol had a flow temperature of
50-550 and an intrinsic viscosity of 0, 15. The insoluble material,
which was a soft rubbery substance, weighed 0. 13 g. for a total
conversion of 23%. An infrared spectrum of the soluble polymer
(run as a film pressed out from the melt) had the following absorption
bands. 2950 cm. *1, strong, broad; 1650 cm.*l weak; 1450 cm.1
strong; 1380 cm. ", strong; 1070 cm.*", weak; 965 crm.- strong;
910 cm. ", weak.
Anal. Calcd. for C9H14: C, 88.45; H, 11,55.
Found: C, 83.86; H, 9.81; Ash, 1.8 (?).
Three series of compounds which are derivatives of
(1) butadiene, (2) styrene, and (3) methacrylic acid have been
studied spectrophotometrically in the region of the ultraviolet
spectrum extending from 190 to 270 millimicrons. The methods of
synthesis and properties of the butadienes and styrenes have been
described in detail.
The butadiene derivatives which were studied were the
cis and trans isomers of 1,3, 8-nonatriene. The ultraviolet
absorption maxima for these two isomers were shifted bathochromically
from the position which would be calculated by Woodward's Rule60 61
for predicting the maxima of acyclic alkyl-substituted butadienes.
The styrene derivatives studied were 2-phenyl-l-hexene (1),
2-phenyl-, 5-hexadiene (II), and 2, 5-diphenyl-1, 5-hexadiene (III).
The absorption maxima for these compounds vere essentially
identical, although in a polar solvent the absorption intensities
increased in the order I < I
The compounds studied in the methacrylate series were the
ethyl, allyl, and methallyl esters of methacrylic acid and finally
methacrylic anhydride. In the case of the esters again, the
absorption maxima are in essentially the same location but the
intensities for the diolefins are greater than those for the monoolefins.
The anhydride, on the other hand, exhibits a band which is shifted
bathochromically some five millimicrons from those of the esters.
These data are taken as evidence for unconjugated interactions
of the 1,6 and possibly the 1,5 ethylenic bonds in the excited state.
Such interactions account at least partially for the tendency of
certain 1,5 and 1,6 dienes to undergo polymerization by the
intra-intermolecular mechanism leading to linear saturated polymers. 30
The trans isomer of 3, GC-nonatriene was polymerized vith
a Zieglcr-type initiator to yield a polymer half of which was soluble.
The infrared spectrum of this polymer was compatible with the
structure which would be expected to result frcm cumulative 1,2
and 1,4 additions during the propagation steps of the polymerization.
Additional Data on 1-Alkenyl Substituted Butadienes
In an effort to determine the extent of unconjugated
chromophoric interactions in 1,5-diolefinic systems, 1,3,7-
octatriene was synthesized. The method employed for this
synthesis has yielded what appears to be the pure cis isomer.
The ultraviolet absorption maximum occurs at 225 millimicrons
with an C value of 21,000. These data fit into the general trend
of the data on butadienes discussed in the text (Chap. II, Sec. B).
As would be expected, the 1,5 interaction in 1,3,7-octriene is
somewhat weaker than a 1,6 interaction and the absorption maximum
is correspondingly closer to tdhe value which would be predicted by
The synthesis and identification of cis-1, 3, 7-octatriene
has resulted in some rather surprising data which should be noted
here. This compound was prepared via the Wittig reaction between
triphenylphosphorusallytide and 4-pentenal, It will be recalled that
trans 1,3, 8-nonatricne was prepared by the reaction of
triphenylphosphorus-5-hexenylide with acrollen. It would thus
appear that by choosing the appropriate aldehyde and ylide one may
synthesize in geometrically pure form either the cis or the trans
isomers of 1-substituted butadienes, If the trans isomer is desired
acrolein would be employed, while if the ciJ isomer is desired the
allylide would be chosen. Bohlmann and Mannhardt80 have reported
observing a somewhat similar instance in their application of this
reaction to the synthesis of tridecal, 3,5, lltetraen-7, 9*dlyne.
WVhere acrolein was used the pure cis configuration about the
3-ethylenic bond was obtained and when the allylide was used a
mixture of the corresponding cis and trans isomers was observed.
No explanation was offered for these rather surprising results.
In this work the cis configuration was assigned to
1, 3 7-octatriene on the basis of infrared data. The strong
absorption band at 965 cm. 1 assignable to the trans configuration
is absent from the infrared spectrum of this material. Furthermore,
the spectrum bears A striking resemblance to that of cis 4 l3,8-
nonatriene the configuration of which has been established in prior
work (see Chap. II, Sec. A). Gas chromatography and infrared
analysis both indicate the material to be pure and free of the trans
isomer. The physical properties of cis *1, 3, 7-octatriene are
b.p. 115-117, n0 1 4594.
Obviously further experiments will be necessary in order
to validate the findings of this study and establish the Generality of
these phenomena. In the event that these reactions prove to be
general for the synthcsir of butadienes vAth 1-alkyl or 1-alkenyl
substituents, a convenient method for the preparation of geometrically
pure cis and trans butadienes vill be available.
1. G. B. Butler and F. T. Ingley, J. Am. Chem. Soc., 3,
2. H. Staudinger and Vt. Heuer, Ber., 67, 1159 (1934).
3. C. Walling, J. Am. Chem. Soc., 67, 441 (1945).
4. W. H. Stocmaayer, J. Chem. Phys., 12, 125 (1944).
5. W. Simpson, T. Holt, and R. Zeite, J. Polymer Sci.,
10, 489 (1953).
6. R. N. Haward, ibid., 14, 533 (1954).
7. R. N. Haward and 1W. Simpson, ibid., 8, 440 (1955).
8. N. Oiwa and J. Ogata, J. Chem. Soc. Japan, 79, 5771 (1957)*
9. G. B. Butler and R1. J. Angelo, J. Am. Chem. Soc., 79,
10. G. B. Butler, A. Cravwshaw, and W. L. Miller, ibid. 80,
11. J. F. Jones, J. Polymer Sci. 33, 7 (1953).
12. C. S. Marvel and J. K. Still, J. Am. Chem. Soc., 80,
13. N. D. Field, J. Org. Chem,., 25, 1006 (1960).
14. W. H. Schuller, J. A. Price, S. T. Moore, and W.. M. Thomas,
J. Chem. Eng. Data, 4, 273 (1959).
15. J. A. Price, U.S. Pat. 3,871,229, Jan. 27, 1959 to American
16. L. E. Ball and H. J. Ha rvood, Paper presented at the 139th
r.eeting of The American Chemical Society, St. Louis, Missouri,
17. D. S. Trifan and J. J. Hoglen, J. Am. Chem. Soc., 83,
18. M. D. Barnett, A. Crawshaw and G. B. Butler, J. Am. Chem.
Soc., 81, 5946 (1959).
19. D. L. Skinner, Ph.D. Dissertation, University of Florida,
20, K. D. Berlin and G. B. Butler, J. Am. Chem. Soc., 82,
21. N. Pappas, E. L. Buhle, and P. Grahame, Paper presented
at the 139th roeeting of The American Chemical Society,
St. Louis, Missouri, March, 1961.
22. R. W. Staclmnan, Ph.D. Dissertation, University of Florida,
23. G. B. Butler and R. 1,. Stacknan, J. Org. Chem., 25,
24. C. S. Marvel and R. D. Vest, J. Am. Chem. Soc., 1,
25. J. K. Still and D. A. Frey, ibid., 83, 1697 (1961).
26. A. Crawshaw and G. B. Butler, ibid., 80, 5464 (1958).
27. J, F. Jones, Italian Pat. 563,941, June 7, 1957, to
G. F. Goodrich Company.
28. J. F. Jones, J. Polymer Sci., 33, 15 (1958).
29. C. S. Marvel and R. D. Vost, J. Am. Chem. Soc., 79,
30. GC B. Butler, Paper presented before The International
Symposium on Macromolecular Chemistry, Moscow, U.S. S. R.,
31. E. A. Braude, Ann. Rep., 42, 105 (1945).
32. H. Ley and H. Wingchen, Bar., 67B, 501 (1934).
33. H. Ley and H. Dirking, ibid., 67B, 1331 (1934).
34. L. Zechmeistcr, Chem. Reviews, 34, 267 (1944).
35. C. Ingoldand C. W. Shoppee, J. Chem. Soc., 365 (1928).
36. C. K. Ingold, M. M. Parcold, and C. W. Shoppee, ibid.,
37. L. Bateman and G. A. Jeffrey, ibid., 211 (1945).
30. L. Bateman and H. P. Koch, ibid., 216 (1945).
39. E. A. Braude, E. R. H. Jones, F. Sondheimer, and
J. B. Toogood, ibid., 607 (1949).
40. E. A. Braude, ibid., 1902 (1949).
41, P, D. Bartlett and E. S. Lewis, J. Am. Chem. Soc., 72,
42. S. Winstein and M. Shatavsky, Chem. Ind., 56 (1956).
43. S. Winstein, J. Am. Chem. Soc., 83, 1516 (1961).
44. R. B. Turner, W. R. Meador, and R. E. 1Vinkler, ibid., 79,
45. C. F. Wilcox, S. Winstein, and W. G. McMillan, ibid., 82,
46. P. D. Bartlett and E. S. Levio, ibid., 78, 2473 (1956).
47. J. Meinwald, S. L. Emerman, N. C. Yantz, and G. Buchi,
ibid., 77, 4401 (1955).
48. M. Simonetta and S. Vinstein, ibid., 76, 18 (1954).
49. D J. Cram and D. I. Willdnson, ibid., 82, 572 (1960).
50. M. S. Newman, Ed., Steric Effects in Organic Chemistry,
J. Wiley and Sons, Inc. New York, N.Y., 1956, p. 479
(L. L, Ingraham).
51. J. P. Koch, J. Chem. Soc., 1123 (1948).
52. 0.: Mikulacova and A. Hvirik, Chemische Zvesti, 11,
53. C. S. Marvel and E. J. Gall, J. Org. Chem., 25, 1784 (1960).
54. G. F. Woods and H. F. Lederle, J. Am. Chem. Soc., 73,
55. L. Cromble, S. Harper, and D. Thompson, J. Chem. Soc.,
56. G. Vittig and U. Scholkopf, Chem. Ber., 97, 1318 (1954).
57. L. J. Bellamy, The Infrared Spectra of Complex Molecules,
J. W7iley and Sons, Inc., New York, N.Y., 1954.
58. L. F. Hatch and T. L. Patton, J. Am. Chem. Soc., 76,
59. H. Pines, H. Alul, and M. Kalobeisld, J. Org. Chem., 22,
60. R. B. Woodward, J. Am. Chem. Soc., 63, 1123 (1941).
61. R. B. Woodward, ibid., 64, 72, 76 (1942).
62. A. E. Gillam and E. S. Stern, An Introduction to Electronic
Absorption Spectroscopy in Organic Chemistry, Edward Arnold
(Publishers) Ltd., London, 1958, p. 98.
63. H. Booker, L. K. Evans, and A. E. Gillarn, J. Chem. Soc.,
64. A. Smakula, Angew. Chem., 47, 657 (1934).
65, R. T. O'Connor and L. A. Goldblatt, Analyt. Chem., 26,
66. E. D. Parker and L. A. Goldblatt, J. Am. Chem. Soc., 72,
67. C. S. Marvel and C L. Hwa J. Polymer Sc. 45, 25 (1960).
68. L. N. Ferguson, Chem. Reviews, 43, 394 (1948).
69. C. G. Overberger and D. Tanner, J. Am. Chem. Soc., 77,
70. H. E. Ramsden, J. R. Leebrick, S. D. Rosenberg, E. H, Miller,
J. J. Walburn, A. E. Balint, and R. Cserr, 3. Org. Chem., 22,
71. P. G. Gaubert, R. P. Linstead, and H. N. Rydon, J. Chem. Soc.,
72. G. B. Butler and G. C. Price, J. Org. Chem., 24, 1092 (1959).
73. L. Cromble, S. H. Harper, F. C. Newman, D. Thompson, and
R. J. D. Smith, J. Chem. Soc., 126 (1956).
74. C. F. Hauser and M. A. Raymond, Private Communication.
75. I. N. Nazarov and A. I. Kakhniashvilt, Sbornik Statei
Obehichel Khim., 2, 919 (1954).
76. W. J. Bailey and C. Iring, J. Am. Chem. Soc., 77, 75 (1955).
77. A. R. Pray, Inorg. Synth., 5 153 (1957).
78. C. E. Rehberg, C. H. Fisher, and L. T. Smith,
J. Am. Chem. Soc., 65, 1003 (1945).
79. W. J. Roberts and A. R. Day, J. Am. Chem. Soc., 72,
80. F. Bohlmann and M. J. Mannhardt, Chem. Ber., 88,
Thomas W. Brooks was born on November 11, 1933 in
Wyandotte, Michigan. He attended public schools in both Michigan
and Florida and was graduated from Manatee County High School,
Bradenton, Florida in 1952. He attended The Citadel, The Military
College of South Carolina, from which he was graduated in 1956
with the degree of Bachelor of Science in Chemistry. In the fall
of 1956 he entered The University of Florida from which he received
the degree of Master of Science in 1959. He has continued his
graduate study at The University of Florida up to the present time.
During his stay at this institution he has held the various positions
of graduate assistant, research assistant, and predoctoral research
The author is a member of Alpha Chi Sigma Chemical
Professional Fraternity and The American Chemical Society,
The author is married to the former Flora Lucretia Smith
and is the father of one child, Lucretia Collette.
This dissertation was prepared under the direction. of the
chairman of the candidate's supervisory committee and has been
approved by all members of that committee. It was submitted to
the Dean of the College of Arts and Sciences and to the Graduate
Council, and was approved as partial fulfillment of the requirements
for the degree of Doctor of Philosophy.
August 12, 1961
Dean, College of Arts dnd Sciences
Dean, Graduate School
C. c. O A
Page 2 of 2
Internet Distribution Consent Agreement
In reference to the following dissertation:
AUTHOR: Brooks, Thomas
TITLE: A study of unconjugated chromophoric interactions related to
cyclopolymerization. (record number: 421844)
PUBLICATION DATE: 1961
I, Dr. Thomas W. Brooks as copyright holder for the aforementioned
dissertation, hereby grant specific and limited archive and distribution rights to the Board of Trustees of
the University of Florida and its agents. I authorize the University of Florida to digitize and distribute
the dissertation described above for nonprofit, educational purposes via the Internet or successive
This is a non-exclusive grant of permissions for specific off-line and on-line uses for an indefinite term.
Off-line uses shall be limited to those specifically allowed by "Fair Use" as prescribed by the terms of
United States copyright legislation (cf, Title 17, U.S. Code) as well as to the maintenance and
preservation of a digital archive copy. Digitization allows the University of Florida or its scanning
vendor to generate image- and text-based versions as appropriate and to provide and enhance access
using search software.
This grant of permissions prohibits use of the digitized versions for commercial use or profit.
Signature of Copyright Holder
Thomas W. Brooks
Printed or Typed Name of Copyright Holder/Licensee
Personal information blurred
Date of Signature
Please print, sign and returi-to:
Cathleen Martyniak --
UF Dissertation Project
University of Florida Libraries
P.O. Box 117008
Gainesville, FL 32611-7008