Stereochemistry of the anionic oligomerization of tert-butyl vinyl ketone

MISSING IMAGE

Material Information

Title:
Stereochemistry of the anionic oligomerization of tert-butyl vinyl ketone
Physical Description:
ix, 130 leaves : ill. ; 28 cm.
Language:
English
Creator:
Bell, Bruce C., 1945-
Publication Date:

Subjects

Subjects / Keywords:
Stereoisomers   ( lcsh )
Ketones   ( lcsh )
Polymers   ( lcsh )
Polymerization   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1986.
Bibliography:
Includes bibliographical references (leaves 125-129).
Statement of Responsibility:
by Bruce C. Bell.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000926832
notis - AEN7532
oclc - 16016453
System ID:
AA00003803:00001


This item is only available as the following downloads:


Full Text










STEREOCHEMISTRY OF THE ANIONIC
OLIGOMERIZATION OF TERT-BUTYL VINYL KETONE







By

BRUCE C. BELL


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1986






















This is dedicated to the ones I love,

my wife Cecilia,

my daughters Kireina and Diana X.,

and my mother and father.

Their love kept me together

while I pulled this together.















ACKNOWLEDGEMENTS

I am gratefully indebted to the members of my supervisory committee: Dr.

George B. Butler, Dr. Wallace Brey, Dr. John F. Helling, and Dr. Christopher

Batich. Special thanks are due to Dr. Thieo E. Hogen-Esch for his guidance and

support.

To Dr. Brey and his enthusiastic assistant (to-be-Dr.) Jim Rocca, I extend

special thanks for their detailed high-field NMR work and valued advice.

And a gracious big thank you goes to for Dr. G. J. Palenik and Dr. Anna

Koziol for their successful efforts of determining trimer structure from X-ray data.

I am especially grateful to Dr. Roy King for his GC/MS work and his freely

offered sagacious bits(bytes) of NMR information.

Cheers and a toast go to the glassblowers Dick Mosier and Rudy Strohschein

who not only did fine work but brightened many a day with their wit.

I warmly thank a valued friend, Dr. Jan Lovy for his generously shared tips

and insights on obtaining better NMR spectra.

To Dr. Ken Wagener, Lorraine Williams and the many good people on the

'polymer floor', just thanks for being there. My life is richer for having known you

all.

To my dear wife, Cecilia, always by my side, I am totally grateful for her help

in all ways.
















TABLE OF CONTENTS


Page
ACKNOWLEDGEMENTS .................................. .iii

ABBREVIATIONS ........................................ vii

ABSTRACT ......................... ................... viii

CHAPTER


INTRODUCTION ................................

EXPERIMENTAL .. .............................

Preparation of Mannich Base of Pinacolone ...............

Preparation of t-Butyl Vinyl Ketone ....................

Preparation of t-Butyl Ethyl Ketone ................... .

Preparation of the Silyl Enol Ethers of t-BEK + t-BMK ......

Trimethylsilyl Enol Ether of t-Butyl Ethyl Ketone .......

Trimethylsilyl Enol Ether of t-Butyl Methyl Ketone .....

Drying and Dividing into Volumetric Ampoules ............

Preparation of 13C Labeled t-BEK .....................

Titration of Alkyllithium Solutions .................... .

Oligomerization of t-Butyl Vinyl Ketone ............... ..

Li-tBEK Initiated .............................

Li-tBMK Initiated Oligomerization ............... ......

Polymerization of t-Butyl Vinyl Ketone ............. ....

Group Transfer Polymerization ...................

Free Radical Polymerization......................

iv


1

6

6

7

9

10

13

14

15

17

18

19

19

24

25

25

25


1

2












Anionic Polymerization in Hexane ................. 26

Butadiene-t-BVK Black Copolymer ................. 27

Epimerizations ................................... 27

Partial Epimerization ........................... 28

Total Epimerization ............................ 29

Instrumental Analyses .............................. 30

Gas Chromatography ........................... 30

Preparative Liquid Chromatography ................. 31

NMR Spectroscopy ............................ 32

Infrared Spectroscopy .................... ...... 33

Gas Chromatography / Mass Spectrometry ............ 33

X-Ray Diffraction Study of Crystalline Heterotactic Trimer. 33

3 IDENTIFICATION OF OLIGOMER STEREOISOMERS ...... 35

Dimer. ........... ............................. 35

Trim er ........................................ 41

Tetram er ................................. ..... 54

End Methyl Group 13C NMR Assignments ........... .... 65

4 OLIGOMERIZATION STEREOCHEMISTRY ............. 70

Methylation Kinetics and Stereochemistry ................ 70

Stereochemistry of Vinyl Addition in THF .............. .. 76

Vinyl Addition in Hexane ........................... 82

Thermal History of Living Oligomer Solution .............. 83

Polymer Stereochemistry ........................... 87

5 STRUCTURE OF ENOLATES ....................... 90













Li-tBEK Initiator ... ....... ........... ... ........ 90

Trapped 'Living Oligomers' .. ...... ................. 107

6 CONFORMATIONAL ANALYSIS ....... ........... 112

t-Butyl Vinyl Ketone ............ .. .............. 112

Oligomers ..................................... 116

Total Epimerizations ..... ...................... 120

Dim er .................... ................. 120

Trim er ................................... 121

Tetramer ........................... ....... 123

REFERENCES ....................... ......... 125

BIOGRAPHICAL SKETCH ............ ............. 130


























vi















ABBREVIATIONS


t-BVK tert-butyl vinyl ketone (4,4-dimethyl-l-penten-3-one)

t -BEK tert-butyl ethyl ketone (2,2-dimethyl-3-pentanone)

t -BMK tert-butyl methyl ketone (pinacolone) (3,3-dimethyl-

2-butanone)

Li tBEK lithium enolate of t-BEK

Li tBMK lithium enolate of t-BMK

Si tBEK trimethylsilyl enol ether of t-BEK

Si tBMK trimethylsilyl enol ether of t-BMK

THF tetrahydrofuran

2VPy 2-vinyl pyridine

DMSO dimethylsulfoxide

LDA lithium diisopropylamide

APT attached proton test, a 13C-( 1H} NMR technique

GTP group transfer polymerization (or polymer thereof)

GC gas chromatograph, with wall-coated capillary columns

LC liquid chromatograph, medium pressure preparative scale


vii


















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


STEREOCHEMISTRY OF THE ANIONIC OLIGOMERIZATION
OF TERT-BUTYL VINYL KETONE

By

Bruce C. Bell

August, 1986

Chairman: Thieo E. Hogen-Esch
Major Department: Chemistry

The lithium enolate of t-butyl ethyl ketone, derived from the corresponding

silyl enol ether, was used to initiate oligomerization of t-butyl vinyl ketone in THF.

Only linear oligomers, formed by 1,2-vinyl addition without side-reactions, were

found and were separated by preparative liquid chromatography.

All stereoisomers of dimer, trimer and tetramer were identified unambiguously

using capillary gas chromatography, 1H NMR and 13C NMR. Stereochemical

assignments were facilitated by 13C labelling (at the initial and terminal positions)

and by determination of the stereoisomer distribution of base catalyzed

epimerization under kinetic and thermodynamic control as well as a comparison with

distributions calculated by conformational analysis. Also experiments in which

solutions of 'living' oligomers were divided and terminated separately (by

methylation and protonation) proved to be important identification aids.


viii












The structure of crystalline heterotactic trimer was determined by X-ray

diffraction studies. The (mr) tBVK trimer was found to be in the gtgg

conformation.

The initiator and living' dimer were determined unequivocally to be present as

only the (Z) geometrical enolate isomers. Also 'living' trimer and tetramer were

deduced to be only (Z) isomers from the distributions of Me3SiCl trapped

stereoisomers.
A 13C NMR study of a series of t-butyl ethyl enols showed a linear correlation

between chemical shift and electronegativity. Calculations based on spin-lattice

relaxation measurements indicated the initiator to be a dimeric aggregate in THF.

The kinetics of CH3I methylation of initiator was found to be second order with

respect to initiator for the first half-life.

The conformational equilibrium of t-butyl vinyl ketone was analyzed by 1H

NMR, and solvent and temperature effects noted. The s-cis conformer predominates

though the s-trans form increases in concentration in polar solvents like THF with

decreasing temperature.

The microstructure of polymers made by free-radical, group transfer and

anionic polymerizations in different solvents could only be analyzed qualitatively.

















CHAPTER 1

INTRODUCTION
Using anionic polymerization techniques, macromolecules of controlled

stereochemistry and narrow molecular weight distribution may be synthesized.

Such polymers behave more predictably than those resulting from other

polymerization methods offering less control. Moreover, unique and well defined

macromolecular architectures may often be achieved anionically.

In order to better understand the factors affecting stereochemical control in

anionic polymerization, many workers have endeavoured to investigate structure and

mechanism of oligomerization under anionic conditions. (What is meant by

oligomerization here is a degree of polymerization (DP) of, commonly, two to five

units.) These oligomer studies have dealt with a variety of monomers:

vinylpyridines,1-4 vinyl sulfoxides,5,6 styrenes,7-10 acrylates,11-16 dienes17-18

and vinyl ketones. 19-21

Hogen-Esch and coworkers3 have been able to separate "up to pentamers and

beyond" of 2-vinylpyridine oligomers by gradient-elution preparative liquid

chromatography. They were able to identify all the stereoisomers of dimer, trimer

and tetramer by capillary gas chromatography and 13C NMR. An important tool

that enabled them to solve for all tetramer assignments was base-catalyzed

epimerization experiments. Distributions of stereoisomers of completely epimerized

dimer, trimer and tetramer3 indicate the nearly equal thermodynamic stability of











the meso and racemic dyads of oligo(2-VPy) at stereochemical equilibria in DMSO at

25 oC. Under the same conditions, similar results were obtained for oligomers of

styrene.22 And similarly, complete epimerization of 4-vinylpyridine tetramer under

the same conditions showed apparently equal meso/racemic dyad stabilities.23 These

results probably indicate the influence of the pendant group size, rather than any

dipolar contribution of each such moity to conformational energies in the strongly

polar solvent.

Suter24 has found that the poly(t-butyl vinyl ketone) dyads are quite different

in energies, with the meso dyad thermodynamically favored over the racemic dyad

by at least 1 kcal/mol. Suter et al. state,25"To our knowledge, this is the first

monosubstituted vinyl polymer in which the meso dyad is estimated to be more

stable than the racemic one." Thus in order to make stereochemical assignments

based on total epimerization of t-BVK oligomers, it was necessary to compare the

experimentally determined distributions of stereoisomers to values calculated using

Flory's theories of conformational analysis of systems in stereochemical

equilibria22,26 and the dyad rotational energies25,27 computed for model P(t-BVK)

dyads.

The value of partial epimerization experiments performed on isolated,

individual oligomeric stereoisomersof 2-vinylpyridine was based on the finding that,

of the acidic methine positions along the backbone, the outer ones racemize more

rapidly than the inner ones.3 Assuming the rate of epimerization to be the same for

a methine carbon that is part of either a meso or racemic dyad, Huang et al.3

calculated the ratio ko/ki to be greater than 60. Since partial epimerization is such an













important technique for assigning stereochemistry in higher oligomers, its

applicability to t-BVK oligomers also had to be proven.


-R
R
--R
R-

(MMY)


--R

--R

+R


--_R
(mmm)






R_.
R ,

-T-R
(nmr)


--R
R ko (nmn)
+
R (rrr)

(R)
(mjr)


Figure 1-1. Epimerization of tetramer



An important structural feature influencing stereoregulation in anionic

polymerization is the presence or absence of intramolecular coordination to counter-

ion by pendant group functionalities that can act as Lewis bases (most often

heteroatoms).28 Many interrelated factors influence the degree to which

intramolecular coordination occurs: nature of the counter-ion and its tightness of

pairing with the carbanion chain end,29 competition with solvation,temperature and

the proximity of the chain-bound chelating ligands to the counter-ion. The latter

being dictated by the relative energies of accessible chain conformations. Mathis and

Hogen-Esch2 dramatically demonstrated the effect of coordination (or its apparent


(mn) ki
+
(M)











absence) by the penultimate unit in reactions of 'living' 2-vinylpyridine lithiated

dimers.
CH3 CH3
Li\ CH3 Li\



A B

Methylation of carbanion A with CH3I at -78 oC in THF resulted in greater than

98% "meso" dimer; methylation of B under the same conditions was 76%

racemicc". Addition of 2-vinylpyridine to A gave trimer whose first dyad was 64%

"meso" and for B, it was 73% racemicc". It seems that due to coordination of Li by

the penultimate pyridine in A, Li preferentially resides on the pro-"meso" side of

the carbanion. From CPK models, the ring methyl in B can be seen to interact with

the methyl group at the end of the chain when similar coordination is attempted.

The stereoregularity of the anionic polymerization of vinyl ketones is also

influenced by factors affecting intramolecular coordination. In the non-polar

solvents hexane and toluene, poly (t-BVK) prepared anionically was found to be

crystalline and highly isotactic; whereas in THF or ether, an amorphous, atactic

polymer resulted.30-32 Indeed, Tsvetanov et al.19 have reported IR evidence of

penultimate coordination in 'living' oligomers of isopropenyl methyl ketone in THF.

The purpose of this investigation was to prepare, separate, and identify all
stereoisomers of the oligomers dimerr, trimer, and tetramer) of tert-butyl vinyl

ketone. Model initiators were to be used for two reasons: since symmetry reduces

the total number of possible stereoisomers and in order to be able to study the effect

of unit by unit increase on the stereochemistry of oligomerization. With the NMR












assignments of stereochemistry firmly established for the oligomers, it was hoped

that the microstructure of polymers (prepared by various techniques) could be

analyzed for triad tacticity in the main chain and the chain end (using C-13 labelling)

with expectations of fitting known statistical models.

Another goal of this study was to describe as completely as possible the nature

of the enolate species that initiate and propagate the chain forming reaction. This

was to be accomplished both directly (principally with NMR spectroscopy) and

indirectly by trapping the species and analyzing the products. It was hoped that

geometrical isomers of the trapped enolates could be separated and regenerated to

determine the effect of initiation by the (E) versus the (Z) lithium enolate on the

oligomerization stereochemistry.


CH3 t-Bu H t-Bu


H OLi CH3 OLi

















CHAPTER 2

EXPERIMENTAL
Of the several synthetic routes to the monomer, t-butyl vinyl ketone

published,32-37 the procedure judged to be most satisfactory was that by

Overberger and Schiller35 via the Mannich base of pinacolone (t-BMK).


Preparation of the Mannich Base of Pinacolone


HCHO, (CH3)2NH-HCI
(CH3)3 CCOCH3 > (CH3)3CCO(CH2)2N(CH3)2-HC1 (2-1)
HCI, EtOH
(1) (2)

Pinacolone (from Aldrich Chemical Co.) was purified by fractional distillation

through a vacuum-jacketed column packed with glass helices (bp. 105-106 0C). In

a 1000 mL three-necked flask equipped for vigorous stirring, pinacolone (2.0 mol)

and dimethylamine hydrochloride (2.3 mol) were added to 260 mL ethanol.

Concentrated hydrochloric acid (2.0 mL) was added and the "slurry was stirred and

cooled in a ice-water bath" as indicated.35 After filtration, a very impure (mp.

120-145 oC) hygroscopic solid was obtained for which the IR spectrum lacked the

expected strong carbonyl absorption. After reviewing experimental techniques for

the Mannich reaction, it was realized that refluxing the reactants for a certain length

of time would be necessary. This was done and to be safe, the extent of reaction

was monitored by periodically withdrawing several drops of the reaction solution











and testing for aldehyde using Tollen's reagents. After refluxing and stirring for

three days, the reaction was found to be complete. On cooling, beautiful white

crystal flakes formed. These were filtered, washed and recrystallized from ethanol.

The yield was 82%, mp.176-177 oC (lit.35 mp. 130 oC).

IR: 2800-2300 cm-1 (s, b) several peaks, 1700 cm-1 (vs), sharp C = 0;

1H NMR (60 MHz, in CDC13 with TMS): singlet at 1.19 ppm (9H), singlet at

2.38ppm (6H) and multiple at 2.83 ppm (4H).

Analysis: Calculated for C9H20 Cl NO: C, 55.80; H, 10.41; N, 7.23; Cl, 18.30.

Found: C, 55.69; H, 10.46; N, 7.24; Cl, 18.30.




Preparation of t-Butyl Vinyl Ketone


H20
(CH3)3CCO(CH2)2N(CH3)2-HC1 > (CH3)3CCOCH=CH2 (2-2)

(2) A (3)


The Mannich base was dissolved in an equal amount of water and the solution

added to a 300 mL three-neck flask via a dropping funnel so as to maintain volume

to half-full throughout the reaction. Connected to the flask was a heated Vigreux

column and Liebig condenser. An oil bath was used to heat the flask and contents

while stirring. Only when the bath reached 180 oC did product begin to appear.

The condensate was collected directly in a separatory funnel. Considerable water

distilled and was extracted with ether. The product t-BVK (3) was dried with

anhydrous magnesium sulfate.












It was determined that as the reaction proceeded, the t-BVK was contaminated

with greater amounts of pinacolone and unidentified higher molecular weight

components.


Table 2-1. Mannich Base Decomposition Products Collected

Fraction Vol (mL) t-BVKa t-BMKa Xb
1 2 99.2 0.3 0.2
2 33 95.4 4.0 0.3
3 9 82.2 16.2 1.5
a Percentages from GC analyses. b X- a higher Mwt impurity


The pinacolone results from an apparent reversal of the Mannich reaction (2-1).

By the end of the last fraction to distil over, the contents of the reaction flask had

turned dark brown and viscous.

The dried t-BVK was purified by reduced pressure, spinning-band distillation

(68 oC, 93 mmHg). To prevent polymerization of the t-BVK in the distillation pot,

it was necessary to add hydroquinone to the monomer. Attempts to dry the t-BVK

over calcium hydride resulted in polymerization. So the purified t-BVK was

degassed on the vacuum line and dried again with fresh anhydrous MgSO4 before

sealing in ampoules with trace (<0.2%) hydroquinone. The ampoules were stored

in a freezer (-20 oC). The yield was approximately 55%.

IR: 1690 cm-1 (vs), somewhat broad (C = 0); 1610 cm-1 (vs), sharp (C = C);

1400 & 1363 cm-1 (s) (t-butyl group).












1H NMR (60 MHz, in CDC13 with TMS as reference): singlet at 1.17 ppm (3H),

7.2-5.5 ppm vinyl (1H), ABC pattern: (Spin simulated on Varian XL

200 computer)

ABC pattern: 8A= 6.88 8B= 6.40 8C= 5.71 ppm,

JAB = 16.9 JAC = 10.2 JBC = 2.2 Hz

13C NMR (25 MHz, 20% in CDCI3), in ppm from TMS: 26.0, t-butyl methyls;

42.9, t-butyl quaternary; 128.2, vinyl methylene; 130.8, vinyl methine;

204.0, carbonyl; (assignments based on APT spectrum).


Preparation of t-Butyl Ethyl Ketone



1.t-BuMgCl, THF,-78 oC
CH3CH2CO-O-COCH2CH3 -! t-BuCOCH2CH3 (2-3)
2. H+, H20 (4)


2,2-Dimethyl-3-pentanone (t-BEK) 4 was directly prepared by the method of

Ansell et al.38 from the Grignard reagent. A 1000 mL, three neck flask equipped

with dropping funnel (with pressure equalizing side-arm) and magnetic stirrer was

flame dried under high vacuum; 200 mL of dry THF was transferred under vacuum

into the flask. After the contents of a sealed bottle (Aldrich gold label) of propionic

anhydride (50 g, 0.38 mol) were added under Ar, 200 mL of a fresh solution of

t-butylmagnesium chloride, 2.0 M in THF, was transferred via cannula under Ar

into the dropping funnel. The anhydride solution at -78 oC (dry ice / isopropanol

bath) was vigorously stirred while the Grignard solution was added dropwise over a











period of 2 h. After warming saturated aqueous ammonium chloride was added and

the solution was partitioned between pentane and 1 M aq. NaOH. The solution of
pentane extract was dried over 5A molecular sieve. The t-BEK was purified by
fractional distillation (bp. 125 oC). Complete drying of t-BEK was accomplished
by stirring with calcium hydride overnite. After being degassed on the vacuum line,

the dry t-BEK was distilled onto fresh CaH2 and then transferred into ampoules

which were flame sealed. The yield was 58%.
t-Butyl Ethyl Ketone Analysis:
IR 1702 cm"1 ,vs (C=O); 1363 and 1390 cm-1 s (t-Bu group)

1H NMR (200 MHz, in CDCI3 with TMS) triplet (7.3 Hz) at 0.96 ppm (3H),

singlet at 1.08 ppm (9H), quartet (7.3 Hz) at 2.46 ppm (2H)

13C NMR (25 MHz) 20% in CDC13, TMS as reference 8.2, 26.6, 29.6, 44.0, and

216.3 ppm
MS molecular ion m /e 114, base m/e 57.


Preparation of the Silyl Enol Ethers of t-BEK and t-BMK


LDA Me3SiC1
(CH3)3CCOCH2R > (CH3)3CC(OLi) = CHR -
THF, OoC Et3N
(1) R = H (5) R =H (2-4)
(4) R = CH3 (6) R =CH3

(CH3)3CC(OSiMe3) = CHR

(7) R=H
(8) R = CH3












The procedure was adapted with only minor changes from that of House et

al.39 for the preparation of trimethylsilyl enol ethers under kinetic control. Into a

specially constructed, high-vacuum dropping funnel (see Fig. 2-1), the ketone

(t-BEK or t-BMK) was transferred through the vacuum line from fresh CaHll2.

Diisopropylamine (Aldrich) and triethylamine (Eastman Kodak) were distilled from

CaHll2 and sealed under vacuum in volumetric ampoules; these were then attached to

the 500 mL flask by glassblowing. The flask was flame dried under high-vacuum

after which, under Ar, the dropping funnel with ketone was connected in place.

Through a side-arm, a pinch (several milligrams) of triphenylmethane was added as

an indicator. The required molar equivalent of recently titrated n-butyllithium in

hexane (Aldrich) was transferred via cannula from a graduated cylinder under Ar

pressure into the flask through the side-arm, which was sealed afterward under

vacuum.

One run with an ampoule containing 43 mL of diisopropylamine began with the

transfer of 118 mL of 2.6 M n-butyllithium in hexane. The hexane was evaporated

and dry THF distilled in through the line whereupon the solution turned the bright

red of the triphenylmethide indicator. The diisopropylamine was introduced into the

solution kept at -780C by rupturing the ampoule's breakseal and the solution was

allowed to warm to ambient temperature while stirring and degassing (butane).

Then once cooled to 0 oC (ice / water bath) the ketone was added dropwise with

stirring until the color nearly disappeared (approx. 15 min). Stirring at 0 C

continued for another 5 min before cooling to -78 oC and distilling in 61 mL of

trimethylsilyl chloride (dried over CaHll2) with stirring. Fifteen mL of triethylamine

























iPr NH




Et N


Figure 2-1. Apparatus used to prepare the silyl enol ether compounds











was added and the solution was warmed to room temperature while stirring. A

white solid (LiCl ) precipitated. The solution was partitioned between pentane and

saturated aqueous sodium bicarbonate (thrice), followed by three washings with

saturated aqueous ammonium chloride and finally deionized water. After drying

with anhydrous MgSO4 and filtering, the pentane was removed using a rotary

evaporator. Fractional distillation through a short Vigreux column yielded major

fractions of the trimethyl silyl enol ether in 98% purity (by GC); t-butyl ethyl ketone

gave Si-tBEK (bp. 90 oC, 54 mmHg; 61% yield) and t-butyl methyl ketone

Si-tBMK (bp. 76 oC, 90 mmHg; 63% yield). Impure fractions with substantial

product were cleaned-up by preparative LC (SiO2 column) with constant elution

(14% diethyl ether in hexane) and later fractional distillation.

Further drying (over CaH2) and degassing of the redistilled silyl ethers was

done on the vacuum line before sealing in volumetric ampoules. The density of the

silyl enol ether of t-butyl ethyl ketone was measured to be 0.818 0.005 g/mL at 24

oC.

Trimethylsilyl Enol Ether of t-Butyl Ethyl Ketone (8)

IR: 1668 cm-1 vs, sharp (C=C)1395 & 1360 cm-1 moderately, sharp (t-butyl)

1320 & 1258 cm-1 vs, (split) (Si(CH3)3).

1H NMR (300 MHz, 50% in CDCl3, reference TMS): singlet at 0.15 ppm (9H),

singlet at 0.97 ppm (9H), doublet at 1.43 ppm, J = 6.7 Hz (3H), quartet

at 4.51 ppm, J = 6.7 Hz (1H).












13C NMR (75.5 MHz, 65% in CDC13, external reference C6D6 central peak =

128.0 ppm): 1.4 ppm (q, 1JcH = 118.4 Hz) silyl methyls, 12.0 ppm (q,

1JCH = 125.8 Hz) allylic methyl, 29.0 ppm (q, lJCH = 125.7 Hz) t-butyl

methyls, 36.6 ppm (s) t-butyl quaternary, 97.7 ppm (d of q, JCH =

154.2 Hz, 2JCH = 6.6 Hz) vinyl methine, 159.6 ppm (s) vinyl ether.

29Si NMR (59.6 MHz, 65% in CDC13, reference TMS): 13.94 ppm (m, 2JSiH =

6.6 Hz).
Trimethylsilyl Enol Ether of t-Butyl Methyl Ketone (7)
IR 1621 cm-1 strong (C=C) with shoulder at 1660 cm-1, 1360 & 1387 cm-1 med

(t-butyl), 1257 & 1300 cm-1 strong (Si(CH3)3).

1H NMR (300 MHz, 50% in CDC13, reference TMS): singlet at 0.20 ppm (9H),

singlet at 1.05 ppm (9H), doublet (1.4 Hz) at 3.91 ppm (1H), doublet
(1.4 Hz) at 4.07 ppm (1H).

13C NMR (75.5 MHz, 50% in CDC13, reference CDC13 77.0 ppm): 0.1 ppm (q,

1JCH = 115.8 Hz ) silyl methyls, 28.1 ppm (q, 1JCH = 125.9 Hz)
t-butyl methyls, 36.4 ppm (s) t-butyl quaternary, 85.8 ppm (d of d 153.8
Hz, 159.1 Hz) vinyl methylenes, 167.0 ppm (s) (=C-O).












Drying and Dividing into Volumetric Ampoules
The technique for sealing an exact volume of liquid in an ampoule under

vacuum was taught to me by Dr. Mikio Takaki.

The apparatus used is shown in Fig. 2-2. Each ampoule is constructed in such

a way that the point at which the 5 mm OD tube joins the 12 mm OD tube of the

ampoule is slightly constricted (ID 2mm). Before glassblowing them onto the

apparatus each ampoule is weighed empty, then filled with water to the constriction

and reweighed. Filling it with water required use of a capillary PE (polyethylene)

tube adaptor for the wash bottle. (This was made by heating and pulling PE tubing.)

The ampoule was dried before attaching it, since rapidly evaporating water would

freeze and cause the breakseal to rupture.

After stirring the liquid to be dried overnight with CaH2 in flask A, it is

degassed and transferred onto fresh CaHll2 in B. Afterwards the liquid is transferred

into flask C, degassed and flame sealed at points a and b under vacuum. The liquid

is then poured into ampoules 1,2 and 3 filling them consecutively. The apparatus is

then inverted so that the liquid drains from the 5 mm ampoule stems (surface tension

at the constriction prevents emptying the ampoule) into the overflow ampoule 0.

Ampoule 0 is the first to be sealed from the apparatus. When flame sealing the

ampoules, the flask is touched just before hand with a cold daubber (dry ice bath or

liquid N2) only momentarily and the seal made at least 3 cm above the level of the

liquid.


















vacuum
line

f


a 2







AO fresh CaH2

liquid + CaHl2


Apparatus for drying and dividing into volumetric ampoules


Figure 2-2.












Preparation of 13C Labeled t-BEK

1. n-BuLi, THF
LCH3)3CC[OSi(CH3)3]=CH2 (CH3)3CC(O)CH213CH3 (2-5)
(7) 2. 13CH3I (9)


In order to label the initial end of the oligomer chains for 13C NMR study, the
13C enriched t-butyl ethyl ketone (9) was synthesized as follows. 13C Enriched

(99%) methyl iodide (Cambridge Isotopes) was degassed and dried twice over

CaH2 before sealing in an ampoule with break seal. This and an ampoule of

unlabeled CH3I were glassblown onto the reaction vessel. An ampoule containing

a 10-20% excess molar equivalent of dried Si-tBMK (7) was also attached by

glassblowing. Once the apparatus was thoroughly dry, an equivalent amount of

recently titrated n-BuLi in hexane was transferred into the flask under Ar by

syringe through a side arm which was afterwards sealed by flaming. Dry THF from

the line was transferred in under vacuum. Then the seal to the Si-tBMK (7) was

broken and the reagents allowed to react with stirring at room temperature for at

least 30 min. The THF solution of the lithium enolate was cooled to -78 oC (dry

ice/iso-PrOH bath) and the 13CH3I added. A white precipitate (Lil) was noticed.

The mixture was kept at -78 oC for at least 24 h before adding the excess CH3I. It

sat another day at -78 oC and then let warm to room temperature for 1 h before

working-up.

Upon opening the vessel, pentane was added and the solution washed twice

with saturated aqueous NH4Cl and twice with saturated aqueous NaHCO3. It was












dried over fresh 5A molecular sieve and pentane and THF removed by fractional

distillation. GC analysis revealed the presence of t-BMK, unreacted Si-tBMK (7)

and the dimethylation product t-BiPK (10). The ketones were readily separated by

preparative LC using 5% ether in pentane over SiO2, but the Si-tBMK(7) was

eluted with t-BEK*(9) even with pentane as the eluent. Therefor the mixture was

treated with tetrabutylammonium fluoride and water to convert the Si-tBMK(7) to

the ketone t-BMK. This reaction was monitored by GC analyses. Once the reaction

was completed the solution was dried and the product cleaned-up by LC and

purified by fractional distillation. The yield was an abysmal 23% so no attempt

was made to remove the final traces of THF or hexane. It was dried over CaH2,

quantitated, degassed and sealed in an ampoule under vacuum.

Proton decoupled 13C NMR showed the enhanced intensity of the signal at

8.2 ppm and 2 Hz splitting of the carbonyl signal at 216.3 ppm due to 2Jc-C-

Titration of Alkyllithium Solutions

The methyllithium in diethyl ether (Aldrich) was found to lose potency even

though it was stored under Ar in the freezer and required titration before every use.

n-Butyllithium in hexane (Aldrich ), kept sealed and under Ar on the shelf, was

quite stable and did not require titration more often than once every couple of

months. For both the method of Ronald employing 2,5- dimethoxybenzyl alcohol

DMBA (Aldrich) as both titer and indicator was used. The first equivalent of alkyl-

lithium deprotonates the alcohol functionality and the benzoxide salt is colorless.

The dianion is dark red in THF and its presence indicates the end-point.

A dry flask was weighed empty and with six drops of DMBA (ca. 100 mg).











This was degassed on the vacuum line and dry THF distilled in. A syringe (2 mL)
with a teflon plunger was flushed with the alkyllithium solution and refilled. With
the DMBA in THF vigorously stirring under Ar and at room temperature the alkyl
lithium solution is added dropwise after passing the needle of the syringe through a
septum. The persistence of the red coloration for longer than 15 s. marked the end
point. This procedure was repeated two more times for each determination and an
average molarity calculated.




Oligomerization of t-Butyl Vinyl Ketone
Li-tBEK Initiated
n-BuLi
(or MeLi)
(CH3)3CC(OSiMe3)=CHCH3 (CH3)3CC(OLi)=CHCH3 (2-6)
(8) THF (6)

CH3 -CH2)
CH2=CHC(O)C(CH3)3 '(-H CHCH=--C(OLi)C(CH3)3
THF, -780C I
(CH3)3C
(II Li) (V Li) (2-7)


CHI3 CH2 CH3
CH31 (CH 'n -CHi
THF, -78C > C=O I


(II) (V)


(2-8)












or H+ \i "., 2
or I> 1 I

(CH3)3 (CH3)3C ==
(IIp) (Vp) (2-9)


CH3 /CH2.
or Me3SiC1 ( 'CH I nCH-C(OSiMe3) C(CH3)3
Et3N I O
(CH3)3c
(II Si) (V Si) (2-10)
for all II, n=1; III, n=2; IV, n=3; V, n=4;


A sketch of the apparatus used for the oligomerizations is shown in Fig. 2-3.
With all reagent ampoules sealed on by glassblowing, the vessel was kept under a
vacuum of 10-6 mmHg overnight before beginning. From the known volume of
silylated ketone (typically 15 mmol) and exact molar equivalent of freshly titrated

alkyllithium solution (n-BuLi in hexane or MeLi in Et20O) was injected into the flask

under an Ar atmosphere. After that side arm was sealed by torching, dry THF (250
mL) was vacuum distilled in from the line and cooled to -78 oC by a dry
ice/iso-propanol slush. The Si-tBEK (8) was added by rupturing the breakseal with
the glass-enclosed bar magnet manipulated by a horseshoe magnet. While stirring,
the solution was warmed to room temperature and kept there for at least 30 min.
The resulting colorless enolate initiator (6) was typically at 0.06 M concentration.
Meanwhile, with the teflon "Rotaflo" (Fisher) stopcocks closed, the
ampoule(s) of monomer plus hydroquinone inhibitor was opened and their contents






















high vacuum
manifold
RU in


Pyrex Wool Plug


Figure 2-3. Oligomerization apparatus


homnw.












distilled onto the fresh CaH2. The purpose of the CaH2 was two-fold: to remove

the acidic hydroquinone and to better dry the t-butyl vinyl ketone. The monomer

was then distilled directly into the graduated cylinder which was immersed in a dry

ice bath and the stopcock to CaH2 flask was closed. The monomer was degassed at

-78 oC and then warmed to 0 oC (ice/water).

With the initiator solution at -78 oC and stirring vigorously (but not splashing)

the monomer vapors are slowly (> 2 h) distilled in. It was learned from Dr. Jan

Lovy that a low temperature heat gun directed at the monomer vapor inlet tube

prevented condensation in the -78 oC environment. Dropwise addition of monomer

invariably resulted in mostly polymer. Also, if stirring of the enolate solution were

too laminar or monomer distillation too rapid, a polymer film would be formed on

the surface. With the monomer at 0 oC there was a minimum of bumping during

distillation, nonetheless a tiny stir bar was placed in the graduated cylinder to assure

smooth transfer. Stirring of the enolate solution at -78 oC was continued for 30

min. after the addition of monomer was completed.

At this point, the THF solution of 'living' oligomer was either terminated

directly by methylation, protonation or silylation, or divided and the various

portions terminated differently. When dividing, the apparatus was constructed with

an additional side arm of heavy-walled tubing leading to a 200 mL round-bottom

flask that either had one ampoule containing a terminating agent or had an array of

ampoules for manifold division of the living oligomer solution. The entire apparatus

was sealed from the line for dividing.

Termination by methylation was done by reaction with methyl iodide at -78 oC.












At least a three-fold excess of CH3I was added from the attached ampoule while

stirring and the mixture was kept closed at -78 oC for more than 20 h. Then it was

warmed to room temperature while stirring for an hour before opening and

working-up the contents. Protonation was accomplished by the addition of excess

10% acetic acid in methanol to the solution at -78 oC, stirring at room temperature

for 15 min and working-up. For silylation, twice the necessary molar equivalency

of trimethylsilyl chloride was transferred into the solution at -78 oC from the line

and half an equivalent of triethylamine added from an ampoule, all with stirring.

The mixture was allowed to warm to room temperature for 30 min while stirring and

then worked-up.

The work-up of the terminated oligomer solution was the same in all three

cases. The THF solution was added to an equal volume of pentane and washed

three times each with half volumes of saturated aqueous NH4Cl, followed by

saturated aqueous NaHCO3 and finally water. The pentane solution was dried over

5A molecular sieve and the pentane was then removed with a rotary evaporator.

Whenever the formation of polymer was noted during an oligomerization, the

concentrated worked-up solution was poured into methanol and the precipitate

removed by centrifugation.

Oligomer solutions were analyzed by capillary GC before and after work-up.










Li-tBMK Initiated Oligomerization


(CH3)3CC(OSiMe3)=C
(7)


CH2=CHC(O)C(CH3)3
THF, -78C


CH3I
THF, -78C


n-BuLi
(or MeLi)
(r (CH3)3CC(OLi)=CH2
THF (5)


H (CH nCH2)CH=--C(OLi)C(CH3)3
c=0
(CH3)3CA
(pII Li) (pV Li)


H 1- CH2~) /CH3
(-CH rntCH
I ICH3 =
(CH3)3Ce- (CH3)3KC/


(DII) (DV)


(2-13)


H+
or >


(2-14)


The procedure followed for the oligomerization initiated with the lithium
enolate of pinacolone (Li-tBMK, 5) was identical to that just described in all respects
but one. It had been noticed that the lithium enolate solution, once formed from


(2-11)


(2-12)


:H2


(pIIp) (pVp)












treatment of the silyl enol ether with alkyllithium (Eq. 2-11), became turbid on

cooling to -78 oC. The white suspension remained after the addition of the first

equivalent of monomer so the mixture was allowed to warm to room temperature,

stirred for 15 min and recooled to -78 oC before adding any more monomer. The

solution cleared on warming and remained clear on recooling. Continued

oligomerization, termination and work-up proceeded as before.


Polymerization of t-BVK
Besides the inadvertent occurrence of polymer as a side-product with some of

the anionic polymerizations in THF and the spontaneous polymerization of

uninhibited and purified monomer, several methods were used to directly prepare

poly t-BVK.

Group Transfer Polymerization

The silyl enol ether of t-BEK (8) directly initiated the polymerization of t-BVK

(3) in the presence of a bifluoride catalyst. The catalyst was prepared by heating

tetra(n-butyl)ammonium fluoride trihydrate40 (Aldrich) at 110 oC overnight under

vacuum. Dry THF was distilled onto the resulting pale yellow-brown 'glass'.

(Under argon and with a serum cap, it turned blue when shaken to dissolve, then

after a few minutes it was yellow-brown again.) One drop of the catalyst solution

(0.16 M) into an equimolar mix of Si-tBEK (8) and t-BVK (3) in THF (0.4 M)

under Ar and at room temperature yielded polymer that was soluble in THF, acetone

and chloroform, and precipitated in methanol and DMSO.

Free Radical Polymerization

To keep the degree of polymerization, DP, low (ca. 100), a suitable solvent











was chosen to act as a chain transfer agent in accordance with the equation41


1/DP = Cs [S]/[M]

where C. is the chain transfer constant to solvent S; M is monomer.

From the polymer handbook42 Cs for CC14 is ca. 10-4; so with [M] = 10-2 [S] the

polymerization was done. The CC14 was degassed by three freezing/thawing cycles

and AIBN (2% of [M]) was used to initiate the chain polymerization. The stirring

solution was kept at 60 oC under Ar for 20 h. A white polymer with a waxy texture

was recovered from precipitation in methanol.

Anionic Polymerization in Hexane

An attempted anionic oligomerization in hexane led to more than 90% polymer
yield. The n-BuLi and Si-tBEK (8) mixture in hexane showed little evidence of

having reacted even after stirring for 20 h at room temperature. Thus the

polymerization was essentially initiated by n-BuLi. Before the complete addition of

the first equivalent of monomer, a precipitate clouded the swirling solution. After

the total 1.6 mol equivalents of t-BVK (3) were added, solid living' polymer was

visibly coagulated on the sides of the flask. Tetrahydrofuran was added immediately

before the mixture was divided and terminated with CH3I and acidified MeOH. The

polymer precipitate in MeOH was centrifuged and washed with MeOH; two solid

layers were apparent. The upper layer was a semi-translucent, waxy substance. It

dissolved in chloroform. No solvent could be found to dissolve the lower layer of

polymer. Oligomers were isolated from the clear supernatant.











Butadiene-t-BVK Block Copolymer

It was hoped that a long polybutadiene chain would help maintain the solubility

of an attached growing poly t-BVK chain in non-polar solvent hexane.

With 2 mmol of n-BuLi in 250 mL of hexane (distilled from the liquid alloy,

Na/K) stirring at 0 oC, 100 mmol of butadiene (dried over CaH2) was transferred in

through the line (bp. -4 oC). This was left stirring at room temperature overnight.

Upon cooling to -78 oC, t-BVK was distilled in very slowly. The first traces caused

the solution to yellow, which disappeared after some 10 s. After only 4 mmol of

t-BVK was added, the solution became turbid white. By the time the total 20 mmol

of t-BVK had been added the white suspension had the consistency of thick soup. It

was terminated by addition of 13CH3I in THF.

The polymer was precipitated in excess methanol and filtered. It was then

dissolved in chloroform, reprecipitated in methanol, collected and finally washed

with acetone. From the vinyl region of the 13C NMR, the polybutadiene block of

the copolymer appears to be > 90% the 1,4-addition product with a roughly 50/50

random distribution of cis and trans units, much as expected.43 The peak shape in

the carbonyl region of the 13C NMR appears unusually sharp for poly(t-BVK),

probably indicating a highly stereoregular structure.32 From a comparison of the

integrated 1H NMR of the t-butyl, methyl and allylic methylene peak areas, the ratio

of BD to t-BVK units in the copolymer was determined to be 3.5 quite comparable

to the mol ratio of 3.3 for monomers added.

Epimerizations
Potassium t-butoxide was used in all epimerization experiments as the base for

deprotonating the acidic backbone methine positions alpha to the carbonyls of the












oligomers. It was surmised to be bulky enough to hinder attacking the carbonyl

directly. Other workers reported successes using KOt-Bu to epimerize vinyl

oligomers.3,24 KOt-Bu was prepared by refluxing t-butanol (dried with CaHll2)

with an excess of filtered potassium metal in THF under Ar. It was found that the

trimer in 0.3 M KOt-Bu in THF at 24 oC under Ar totally degraded within 15 min.

The addition of t-butanol to the KOt-Bu solution reduced side-reactions and made

kinetic control of epimerization possible.

Partial Epimerization

The mixture of oligomer (0.05 M), KOt-Bu (0.04 M) and t-butanol (1.0 M) in

THF was stirred under Ar. At regular intervals aliquots were withdrawn by syringe

through a septum and squirted directly into a test tube containing 0.5 mL CHC13 and

1.0 mL of saturated aqueous NH4Cl. After mixing and settling the CH3CI layer

was analysed by capillary GC.

For the partial epimerization of the isotactic trimer of vinyl pyridine using

KOt-Bu in DMSO,44 it was found that the heterotactic isomer is formed much more

rapidly than the syndiotactic. This was interpreted as meaning that the outer methine

positions are more accessible to deprotonation and racemize before the inner methine

carbon under the given conditions.

By shaving peaks and recycling fractions in the preparative LC, isotactic

t-BVK trimer of 81% purity was the best achieved. It was subjected to the partial

epimerization conditions and the results are listed in Table 2-2. Especially in the

early stages it is clear that loss of the mm isomer corresponded to gain in mr/rm.

This experimental result agrees with that for the vinyl pyridines that were epimerized












under more severe conditions. Thus it was deemed safe to use partial epimerization

of isolated stereoisomers of tetramer (IV) as an aid in their identification.




Table 2-2. Partial Epimerization of Isotactic Trimera

t (min) mm mr/rm rr

0 81.4 18.6 -

17 79.7 20.3 -

68 74.0 24.3 1.7

177 66.2 30.0 3.8

374 59.5 35.0 5.5
a Trimer (0.024 M), KOt-Bu (0.06 M), t-BuOH (1.0 M) in
THF at room temperature under Ar.




Total Epimerization

Some 30 40 mg of dry oligomer in a tube was degassed on the vacuum line.

Two mL of 1.0 M KOt-Bu in dry t-butanol was added to the tube under Ar, which

was subsequently degassed and sealed under vacuum. The tube and content were

kept at 50 oC for one week with agitation. After one week the tube was opened, the

base neutralized with saturated aqueous NH4C1 and the epimerized oligomer

partitioned into pentane. The distribution of stereoisomers was determined by

capillary GC and compared to that calculated from conformational analysis.












Instrumental Analyses
Gas Chromatography

Routine analyses of mixtures of oligomers were done on a Hewlett-Packard

5880A gas chromatograph equipped with a capillary column in a temperature

programmable oven as well as a flame ionization detector and a microprocessor.

The column (HP# 19091-60750) was fused silica capillary (0.2 mm ID) coated with

0.11 pm film of silicone gum (Gen.Elec. Co. SE-54, which was methyl, 5%

phenyl, 1% vinyl cross-linked polysiloxane). The carrier gas was helium (Airco)

which was scrubbed with pre-column molecular sieves. The flow rate was set so

that the optimum number of theoretical plates for the column was achieved; i.e., the

minimun value from the Van Deemter plot published in the HP 5880 literature was

used to calculate optimun flow for that column.

Various step programs were used to increase the oven temperature depending

on desired speed of analysis vs resolution They were done so that oligomers

eluted on the plateaus of the steps. All stereoisomers of tetramer, Mw = 464, could

be separated. With the oven temperature at the limit for the column coating

(325 oC), octamer (Mw = 912) was eluted, though poorly resolved.

The microprocessor reported peak retention time (minutes), integrated areas,

type and percent of total area. Retention times were highly reproducible ( 0.1%

for consecutive injections); nonetheless standards were kept and used to eliminate

ambiguity. Peaks of the type not resolved to baseline were integrated in such a way

that the area perpendicularly beneath the peak to the bottom of the valley where it












joins another is included. This tended to inflate the smaller peaks and diminish the

larger ones not completely resolved, but the effect on analyses was considered

insignificant.

The reproducibility of integrated areas depended on the sample size injected

(generally, 1 tL of a 10% solution) and since only 0.5% of that actually goes into

the column, the possible variations were generally large. However, the percent of

total area values used to calculate the distribution of stereoisomers were, all in all,

highly reproducible with nominal variations of 0.2%.


Preparative Liquid Chromatography

All mixtures of oligomers were separated by passing their hexane/ether

solutions over silica gel. The high performance liquid chromatograph used was an

Altex Model 332 system (now Beckman Co.) with programmable gradient elution.

The two solvent pumps were fitted with preparative heads. An analytical cell with

longer pathlength was used in the constant wavelength (254 nm) UV detector Model

153 for sensitive detection of these low absorbance polyketones.

The preparative SiO2 column used was Merck's Lobar B (310x25(ID)mm)

packed with 40-63 pm silica gel (LiChroprep). It was a glass column with a

pressure limit of 90 psi; the system was fitted with a pressure release valve in-line

before the injection port. Since the column efficiency diminished with usage, it was

regenerated by flushing it first with THF, then dry MeOH. After that it was

connected to an Ar tank and purged of all solvent, then wrapped with heating ribbon

and gradually heated to 250 oC under Ar flow. After cooling it was connected to an












02 tank and reheated to 250 oC (behind a shield) with 02 flow. It was finally

purged with Ar before reconnecting to the HPLC.

Gradient elution of these essentially non-polar oligomers was achieved using

varying proportions of hexane (HPLC grade Fisher) with anhydrous diethyl ether,

always freshly prepared before a separation. In general a linear increase from 2.6 to

13.0% ether in hexane over a period of 200 min with a constant flow rate of 5.6

mL/min produced an adequate separation of all oligomers through hexamer. Later

refinements improved this somewhat. A convex gradient (steeper in the beginning,

more gradual toward the end) speeded up the operation. Better reproducibility was

achieved when the eluent composition was altered by a third component, constant

0.05% iso-propanol throughout. Still one of the most successful separations of only

dimer, trimer and tetramer involved simply a constant 7.8% ether in hexane at a

constant 8 mL/min.

Most LC fractions were also analyzed by capillary GC.

NMR Spectroscopy

The availability of several NMR spectrometers here in the Chemistry

Department made this aspect of the investigation quite pleasant.

1H NMR spectra were obtained routinely on the 60 MHz continuous wave

Varian 360 with its permanent magnet or the 100 MHz Fourier transform JEOL FX-

100 with its electromagnet. As an identification aid, some homonuclear decoupled

1H NMR experiments were done with the FX-100. When the added resolution that

sometimes accompanies increased field strength was desired, samples were

submitted to Dr. Brey to be run on the 300 MHz superconducting Nicolet NT-300

(financed by the Instrument Program of the NSF Chemistry Division).











For 13C NMR spectra, the JEOL FX-100 instrument was the workhorse until

the arrival of a Varian 200XL superconducting NMR spectrometer. All T,

measurements, APT experiments and computer-simulated spectra were done on the

Varian XL-200. Several key 13C NMR studies were performed on the Nicolet

NT-300 including selective 1H decoupling of the silylated enolates. 29Si NMR

measurements were done on the Nicolet NT-300.
Infrared Spectroscopy

Those few IR spectra that are mentioned here were done neat on NaCI plates
using a Perkin-Elmer 281 IR spectrophotometer with data station.

Gas Chromatography / Mass Spectrometry

Samples rich in particular stereoisomers of dimer(II), trimer (III) and tetramer
(IV) were run on a GC/MS system by Dr. Roy King. The mass spectrometer was

an AEI MS30 with a Kratos data system. The gas chromatography was a Pye Series

104 with a polysiloxane coated particle packed 4'xl/4" glass column.

X-Ray Diffraction Study of Crystalline Heterotactic Trimer

The crystallographic study of the mr / rm trimer (III) was done by Drs. G.J.

Palenik and Anna E. Koziol. The crystal was grown by slow evaporation of solvent

from hexane/ethyl acetate solution in an uncapped NMR tube.

Crystal data: monoclinic, la, a=12.141(3) A, b=14.251(7) A, c=13.424(4) A,

13=94.03(2) o, V=2317(1) A3, Z=4

Intensity data: Nicolet R3m diffractometer; Mo Kac radiation, graphite mono-

chromator; (0-20 scan to 20=46.00; 2572 unique observed reflections;







34



Structure solution and refinement: SHELXTL programs; direct methods and

Fourier synthesis; least squares refinement; R(usual) = 0.0734,

R(weighted) = 0.0483; goodness-of-fit = 3.36















CHAPTER 3
IDENTIFICATION OF OLIGOMER STEREOISOMERS

Dimer
Dimer was almost always the major oligomer formed in the oligomerizations in
THF regardless of monomer to initiator ratio ([M]/[I] varied from 1.1 to 3.9), and

was readily separated from the other oligomers by preparative solid/liquid

chromatography (SiO2). The methylated dimer (II) exists as two possible

diastereoisomers: meso (m) and racemic (r).





0 40 0 0

(m) (r)



When the LC eluent polarity was sufficiently low (constant 5% ether in hexane

over SiO2), the two diastereoisomers separated with the (r) being the first to elute.

These isomers were quantitatively analyzed by GC using wall coated (SE-54)
capillary columns.
The stereoisomers were identified by 1H NMR on the basis of their methylene

regions45 (see Fig. 3-1). In the meso isomer these protons (Ha + Hb) are

diastereotopic and exhibit a chemical shift difference of 0.6 ppm. For the racemic










dimer the methylene region is seen as a doublet of doublets, not the triplet which
might have been expected. The A2B2 pattern for these equivalent, enantiotopic
protons exhibits different vicinal couplings due to conformational effects, as
explained by Bovey45 for racemic 2,4-diphenylpentane.
The mass spectra of these methylated dimers from GC/MS show that their
principal fragmentation involved homolytic cleavage of bonds on both sides of the
carbonyl with t-butyl+ (C4H9+, m/e = 57) the most intense (or base) peak and its
loss M-57 (m/e = 183) leading to the next most intense. Likewise the pivaloyl cation
(t-BuC=O+, m/e = 85) or that resulting from the loss of its radical M-85 (m/e = 155)
are seen as important. The peak at 69 is undoubtably due to the resonance stabilized
cation (C4H50+),



C 0@ C
I I II

CH2 CH3 CH2 CH3 CH2 >H


which may in turn be a result of rearrangement of the largest (m/e = 183) cation
fragment.
H0C


CH>K30


CH3










MESO


H2


Ha- -Hb
H --


ftlBCf


3.0 2.5


RACEMIC
H3
0
Ha- --Hal

0

H3





3.1 PPM


1.5 1.2 1.0

Figure 3-1. 100 MHz 1H NMR Spectra of the diastereomers of methylated dimer
15% in CDCI3 at ambient temperature










The peak at 114 can only be due to this same type rearrangement of the
molecular ion, called the McLafferty rearrangement.46

0


4- + &4-
CHH
CHm/e= 114


It was thought that the intensity of appearance of this fragment might be an MS
handle on stereochemistry, since the relative orientations of the groups about chiral
centers C(4) and C(6) depend on which diastereomer is being examined. The
intensity of the 114 peak (relative to the base t-Bu+ peak ) is 10.1 for the racemic
dimer and 12.9 for the meso. Only the chair cyclohexane-like transition state for this
C-C bond scission leads to the more stable (Z) enol. Both isomeric transition states
exhibit 1,3-diaxial interactions and unless the methyl/carbonyl interaction is favored
due to H-bonding association (which is not at all likely considering the low acidity


H 3 CH CHH

H "CH, H


(m)


(r)












of the methyl hydrogens), it is difficult to explain the different intensities. Put in

perspective, the difference is probably not significant, since the tremendous energy

imparted to the molecule to fragment bonds overshadows the relatively small

differences in energy discussed.

The 13C NMR spectra of these methylated dimers reveal their symmetry. Each

stereoisomer has only one 1H decoupled 13C resonance each for the end methyl,

methine, carbonyl, quaternary and t-butyl carbons. Among these the end methyl and

carbonyl signals show the best stereostructural differentiation as had been seen

earlier in these labs with vinyl pyridine oligomers.3 The chemical shifts are listed in

Table 3-1 along with those for the unsymmetrical protonated and n-BuLi initiated

(CH3I terminated) dimers. The t-BVK dimer terminated by protonation (usually

10% HOAc in MeOH) has only one chiral center and thus has no diastereoisomers.

Its two carbonyl signals are well separated.

The effect of the n-pentyl group at the initial end of these short two-unit chains

is to shift all carbonyl signals upfield; the end methyls for the two diastereomers

move closer together. End group effects on the 13C chemical shifts of styrene

oligomers have been studied by Sato and Tanaka7 and the oligomers with the longer

n-alkyl ends proved better models for polymer stereochemistry. So it might be

anticipated that the changes in chemical shifts seen here portend trends for higher

oligomers.

















Table 3-1. Dimer 13C NMR Chemical Shifts


R CH3

0 0





R Stereoisomer 5t-Bu 8C=O 8 End CH3

CH3 meso 26.3 219.2 18.5

CH3 racemic 26.3 218.6 17.0

H --- 26.1 215.4 18.1

26.5 219.4

n-C5H11 "meso" 26.2 218.0 18.4

26.6 218.6

n-C5H11 racemicc" 26.4 217.8 17.5

26.7 218.1
Approximately 15% in CDC13 at ambient temperature; in ppm from TMS.











Trimer
Of the two symmetrical methylated t-BVK trimers (mm and rr), only the

isotactic (mm) isomer has its methylene protons in clearly different chemical

environments.


(mm) (mr*) (rm *) (rr)
CH3 CH3 CH3 CH3
--R R --R R
Ha -Hb H Hb H- -H Ha -H
R R R-- R--
H0 Hb H- H6 Hb H0 H0 H6
Ha--Hb H--Ha Hb -Ha Ha--Ha
R R-- R --R
CH3 CH3 *CH3 CH3


Isotactic Heterotactic Syndiotactic


Preparative LC was used to isolate a fraction rich in the isotactic trimer (71%)

by shaving peaks and recycling appropriate portions. Its 13C NMR (Fig. 3-2)

clearly shows it to be symmetrical: only one end methyl peak and two carbonyl (and

t-butyl) peaks of unequal intensity (two outer and one inner). The 1H NMR (Fig.

3-2) shows that its methylene signals are separated by 0.7 ppm, but even a high field

instrument was unable to differentiate the upfield CH2 signals from the t-butyl

absorption. The integration of the downfield methylene signal was compared to the

total methine integrated signal; the expected ratio of 2/3 was obtained, confirming the

identity of this stereoisomer.

When the LC fraction of total trimer was left to slowly evaporate, crystalline

heterotactic trimer was formed. Once isolated and washed, the capillary GC showed























3.5


2'.5


1.5


aJL


1.0


CARBONYL REGION


220


Figure 3-2.


4'0


eb


0:5


Isotactic trimer, 71% pure as determined by GC (the impurity is
heterotactic trimer) (a) 300 MHz 1H NMR (b) 25 MHz 13C-{ 1H}
15% in CDC13 at 300C


~






43


it to be pure. Mass spectrometry confirmed its molecular weight and the key regions
of the 13C { 1H} NMR spectrum (Fig. 3-3) left no doubt as to its identity. Also
Drs. G. Palenik and A. Koziol had done an X-ray diffraction analysis of these
crystals regrown from hexane/ethyl acetate. Its structure is depicted in Fig. 3-4 and
selected bond lengths and dihedral angles are given in Table 3-2





CH3


-I
D-,- -
+-C-

^-


_c-p




-H3


Carbonyl Region:


End Methyl


219.3 218.5


217.7 ppm


17.4 17.0 ppm


Figure 3-3. 25 MHz 13C-{lH} NMR of heterotactic trimer


Region:



















iC(16)


C(9)


C(10)


iC(171


C(3)


10(3)


C(21) (


C(22)


Figure 3-4. Molecular structure of heterotactic trimer determined from X-ray
analysis of the crystal.












Table 3-2. Heterotactic Trimer Crystal

Bond Lengths (A)

Backbone


C(1)-C(2)

C(2) C(3)

C(3) C(4)


1.53

1.55

1.53


C(4)- C(5)

C(5) C(6)

C(6) C(7)


Carbonyls

0(1) C 1.20 0(2) C 1.22 0(3) C 1.18

t-Butyls


C(9) CH3 1.51

C(14) CH3 1.56

C(19) CH3 1.48

Dihedral Angles


1.47

1.52

1.52


1.50

1.52

1.52


C(1)C(2) C(3)C(4)

C(2)C(3) C(4)C(5)

C(3)C(4) C(5)C(6)

C(4)C(5) C(6)C(7)


70.50

-176.10

65.50

58.90


Conformation

gauche

trans

gauche

gauche


Carbonyls


0(1)C(8) C(2)C(1)

0(2)C(13) C(4)C(3)

0(3)C(18) C(6)C(7)


1.54

1.55

1.55


Backbone


Dyad

m

m

r

r


54.70

40.80

52.30











From the dihedral angles of the backbone of this crystalline heterotactic trimer
the conformation is determined to be gtgg (mr) which was calculated to be the most
stable conformer in solution (see conformational analysis section). The carbonyls
are seen to be nearly bisecting the CCC angles at the methine positions in the chain
backbone. This agrees well with Suter's calculated orientation47 for the model
compound, 2,2,4-trimethyl-3-pentanone as shown.

0
R




H


One methyl of the t-butyl group eclipses the carbonyl in Suter's favored
conformation of the t-butyl isopropyl ketone. Similarly the slightly shorter C,C
bonds of the t-butyls in the two outer pivaloyl groups of the crystalline heterotactic
trimer correspond to methyls eclipsing carbonyls. The t-butyl of the inner pivaloyl
is slightly askew.
The 13C NMR spectrum (Fig. 3-3) of this heterotactic trimer shows two peaks
for the end methyl groups and three for the carbonyls as expected for this
asymmetric molecule. In order to assign these to the meso or racemic dyad (and
inner vs outer for the carbonyls), the following experiment was undertaken. A
solution of 'living' oligomer (in THF at -78 oC) was divided and the halves were
terminated in different fashions: one was protonated, and the other, alkylated with
13C labeled methyl iodide. With the 13C label the two ends of the asymmetrical







47



oligomers can be differentiated (provided that they don't result in equal amounts as
happened in one experiment). Protonation traps the stereochemical information
present in the 'living' oligomers as the result of monomer addition, since no
new chiral center is formed in the chain. This is represented for the meso 'living'
trimer:


L0


CH3 rH

'LIVING'
M- TRIMER


13CH31





4 0


CH3
MM*


13C
0
to)


CH3
MR*


(M-) = (MM*) + (MR*)


The sum of the mol fractions of isotactic and the (mr*) heterotactic trimer must
be equal to the mol fraction of living (m') trimer present at the time of methylation
(determined from the protonated portion), i.e.,


(m) = (mm*) + (mr*)


(3-1)











Likewise for the other isomer of living trimer.


(r) = (rm*) + (rr*)


(3-2)


Since the identities of the isomers of protonated trimer had yet to be established, they
were also determined in this same experiment. Capillary GC of the solutions of both
sets of oligomers separated all diastereomers of trimer allowing analysis. The
following GC trace of the methylated trimer shows the lack of complete resolution of





ISOTACTIC
HETEROTACT IC
r SYNDIOTACTIC







isotactic and heterotactic isomers; nonetheless, the integrated areas gave reliable
quantitation (as verified by 13C NMR). For the two peaks of the protonated lot,


(m-) + (r-) = 1


(3-3)


The question was only which was which. The 13C NMR of the end methyl region
for a fraction of trimer (see Fig. 3-5), separated from the other oligomers by












CH3 CH3


o0 0o 0



I
H







18.5 18.1 17.4 17.0 ppm


mr,

MM*



rr*




18.5 18.1 17.4 17.0 ppm


Figure 3-5. 25 MHz l3C-{ H} NMR of the methyl end groups of trimer
(a) normal methylation, (b) terminated with rCH3I.












preparative LC, allowed quantitation of the heterotactic peaks labeled on the one end

or the other; the question again to be answered was which was which.


(mr*) + (rm*) = (mr/rm) (3-4)


The four unknowns were solved directly from these four simultaneous equations.

As noted in the spectrum of 13C labeled end methyls, the racemic ends of the

heterotactic trimer show greater enrichment. This fact was used to advantage when

examining the carbonyl region of the 1H decoupled spectrum (Fig. 3-6). Splitting

on several downfield peaks appears like triplets, but on closer examination the

central peak of each varies in intensity. The doublet is due to the two bond C,C

coupling [J2 (13C-C) = 2 Hz] normally not seen because of the very low natural

abundance of the 13C isotope. The central peak is the unsplit signal of the carbonyl

located at the unlabeled end. Of the three heterotactic peaks, the upfield absorption

showing no splitting is clearly that of the inner carbonyl carbon and it is concluded

that the resonance with the highest intensities for the split peaks and the lowest for

the unsplit must be the outer carbonyl on the racemic side (mr*).

The isolation of a fraction of trimer rich in syndiotactic isomer completed the
13C NMR picture for all stereoisomers of trimer carbonyls and can be seen in Fig.

3-6. Of these, the inner resonance lines are of key importance as models for

interpreting polymer tryad distributions (relative amounts of isotactic, heterotactic or

syndiotactic three unit segments in the chain). But, unfortunately, this system

defies routine analysis since the heterotactic peak is upfield of the isotactic and

syndiotactic signals. Similar irregularities were noted for the 13C NMR spectrum of















INNER
H




OUTER
e OUTER
OUTER INNER
I I







219.0 21 .0 PPM



S
outer


) ^inner

H S H







219.0 218.0 PPM


Figure 3-6. Carbonyl region of trimer spectra (a) 75 MHz 13C { H} NMR of
mix of end "CH3- enriched trimers. (b) 25 MHz 13C { H} NMR
of a mix of isomeric trimers rich in syndiotactic.












the terminal methyl groups. This anomalous ordering may be due to end group

effects7 which cause these short chains to be in different conformations than would

be found for the same tryad located in a long polymer chain.

A great deal of caution must be exercised in obtaining a 13C {1H} spectrum

for quantitation.48 Besides such considerations as sufficient power for the 900

pulse, maximum digital resolution, and use of gated-decoupling to avoid NOE

(Nuclear Overhauser Enhancement), it is necessary that the delay between each
pulse/acquisition be long enough (~ 5 Tl) to obtain complete relaxation of the 13C

nuclei measured. For that reason, the spin-lattice relaxation times, T1, for all
carbons of the heterotactic trimer were measured and are collected in Table 3-3.












Table 3-3. T1 Values for Heterotactic Trimer


Methyls


ends


t-butyls




Methylenes



Methines




Quaternary


Carbonyl


central

r-side

m-side


13C 8 (ppm)

17.06

17.38

26.09

26.29

26.81

34.70

36.18

36.62

37.11

40.51

44.42

44.60

44.64

217.43

218.15

218.81


Tl(s)

1.44 0.26

1.56 0.23

1.42 0.13

1.46 0.13

1.37 0.14

0.57 0.06

0.67 0.07

1.22 0.08

1.38 0.07

1.37 0.08

24.80 0.98

22.12 0.57

25.31 0.73

17.07 0.47

16.74 0.27

19.34 1.23












Tetramer

From the following capillary gas chromatograph of the total tetramer product

resulting from methylation of the oligomerization mixture in THF, two

stereoisomers are seen to predominate. Six diastereomers of methylated tetramer IV

are possible and all are visible in the GC.


Of these, four isomers are symmetrical: mmm, mrm, rmr, and rrr; and two

unsymmetrical: mmr/rmm and mrr/rrm. The two major components of tetramer

were each isolated by preparative LC. From the 13C NMR spectrum (Fig. 3-7), it's

clear that this first isomer is symmetrical. And, even though the sample was only

75% pure, the 1H NMR (Fig. 3-7) shows the chemical shift separation of the

methylene protons as we'd seen before for the meso dimer and the symmetrical mm

trimer. Evidently this symmetrical tetramer is the isotactic stereoisomer, mmm.









CARBONYL REGION



(a)




218.5 218.0


(b)


UPFIELD REGION







I I I
40 35 30 25 20


3.0 2.5 2.O I5


I1 0X5 6


Figure 3-7. NMR Spectra of isotactic tetramer (75% pure, from GC)
(a) 25 MHz 13C { 1H. (b) 100 MHz 1H in CDC13











The predominant tetrameric isomer formed in the oligomerization terminated by

methylation (eqns. 2-7, 2-8) is an unsymmetrical compound. This is apparent in

the two regions of 13C NMR shown in Fig. 3-8: four carbonyl peaks and the

methyls at each end of this four-unit chain have different chemical shifts. This same

stereoisomer was labeled by terminating the oligomerization mixture with 13CH31

and isolated as before. Suprisingly only one end of this unsymmetrical tetramer was

labeled (Fig. 3-8b). The other end methyl 13C NMR peak at 17.4 ppm was so

diminished as to be insignificant. The splitting of the carbonyl peak at 218.3 ppm

due to the two bond 13C,C coupling can be seen to give a clean doublet Fig.3-8. It

was a very different result from that seen for the 13C end-labeled unsymmetrical

trimer. Why? Either the methylation of the carbanion end of the tetramer was highly

stereoselective (whereas it wasn't for dimer or trimer ) or the two major

stereoisomers of tetramer resulted from the methylation of the same carbanion.

The answer to this question lay in the results of the oligomerization experiment,

in which two parts of the solution were terminated differently. The protonated part

showed the expected four isomers of tetramer: mm-, mr-, rm-, and rr- with the first

GC peak to elute being ca. 75% of total protonated tetramer IVp (see Fig. 3-11).

This major isomer of IV p had been characterized by 13C NMR: four carbonyl

peaks at 214.5, 217.9, 218.2 and 218.7 ppm; and the end methyl group at

18.25 ppm (from TMS).

The two predominant isomers of methylated tetramer IV amounted to just

slightly more than 75% of the total (see Table 3-4). It seemed most probable then














END METHYL REGION


CARBONYL REGION


(a)


2i5 218


(b) 2 Hz




219 218


17.4 17.1 ppm


18.3


17.1 ppm


Figure 3-8. 25MHz 13C { 1H} NMR of the unsymmetrical methylated tetramer
in CDC13 at 300C (a) unlabeled, (b) terminated with 13CH3I












that since the first of these was the mmm isomer, the other unsymmetrical isomer

must be the mmr and the principal protonated tetramer, the mm- isomer.


13 R13
13 CH3 I I iI | CH3 + CH3 I II III CH3
CH3 R R R R R RR

CH3 i', R MMM* MMR*
R R R
HCH3 '
MM- C j--ill!!!! R

R R R

MM-



Partial epimerization of the isotactic tetramer provided supportive evidence for

the assignments. As mentioned in the experimental section for trimer and found for

oligovinylpyridines,44 the external methine positions (a to the carbonyls) are

racemized much faster than the internal methines when treated with potassium

t-butoxide/t-butanol. With the reaction monitored by GC as a function of time, the

first isomer to appear at the expense of the mmm tetramer was expected to be the

mmr/rmm stereoisomer (see Fig. 3-9). It had the same retention time as the principal

peak for the methylated tetramers, thus corroborating the mmr assignment made

earlier.

This unsymmetrical tetramer was also subjected to partial epimerization by

treatment with KOt-Bu. Racemization of the chiral methine carbon at the racemic

end of the molecule leads to the mmm tetramer, inversion at the meso end gives rmr.

This proceeded cleanly as seen in the GC trace shown in Fig. 3-10. Not only did













CH3
--R
R KOtBu
--R
H+
--R
CH3

MMM


CAPILLARY GAS CHROMATOGRAMS


MMM


Figure 3-9. Partial epimerization of isotactic tetramer


R
CH3 '2 I | I | CH3
SRI KOtBu R R R
CH3 I I I I CH3
R R R H+ R


CH3-t I- I t CH3
R R R


RMM


MMM



RH+ R
) CH3 I I Il -CH3
R R
RMR


CAPILLARY GAS CHROMATOGRAMS


RMM/MMR


RMM/MMR


Figure 3-10. Partial epimerization of unsymmetrical tetramer mmr/rmm


CH3
- R
-R
--R
CH-R
CH3


CH3

H+ R
-R

CH3

RMM


MMM
MMR/RMM












this further support the assignment of mmr but it provided evidence to newly assign

a minor oligomerization component of tetramer, the symmetrical rmr isomer.

Still most of the stereoisomers of tetramer (protonated and methylated)

remained unidentified. Partial epimerization of isotactic protonated tetramer (mm-)

was undertaken with hopes of alleviating this situation. As usual the course of the

reaction was followed by GC, but resolution of these stereoisomers pushed the

technique to the limit. Even with a 100 m long capillary column whose wall was

coated with non-polar, cross-linked polysiloxane, retention times in excess of 2 h at

constant temperature were required to separate the last two components. Two peaks

were seen to grow fastest as the isotactic diminished (Fig. 3-11). In GC elution

order, the fourth peak increased more than the second.

Interpretation of the results, however, was confounded by the difference

between the possible enolate carbanions formed at the two ends:

tertiary-vs-secondary. The secondary carbanion is more stable and the base is

expected to encounter less hindrance in approaching these methylene protons;49

however it is achiral. Deprotonation and reprotonation of this acidic end position is

inconsequential to the molecular stereochemistry, unless the carbanion is involved in

kinetically significant secondary reactions.

One such reaction may be back-biting, or intramolecular self-epimerization.

Back-biting by the more stable secondary enolate carbanion of the mm- tetramer

would lead to formation of the rr- stereoisomer directly as illustrated in Figure 3-12.









61









\. 0 i-4 Ux f- C,'
eoo are show in


























LU




S aW
I--

U-









Figure 3-11. Partial epimerization of isotactic protonated tetramer (IVp)
by treatment with equimolar KOt-Bu in t-BuOH at ambient
temperature. GC traces of aliquots are shown.












--R --R --R

R e R H+ R

-R -R R--


R R R

mm- xx- rr-

Figure 3-12. Intramolecular self-epimerization




Total epimerization of the methylated tetramers yielded more stereochemical

information about these oligomers. The mix of tetramers was epimerized at 50 oC in

1.0 M KOt-Bu in t-butanol in a sealed tube under reduced Ar pressure for one week.

The tube was opened, the base was neutralized with saturated aqueous NH4C1 and

the contents were analyzed by capillary GC. The distributions of stereoisomers of

tetramer are listed before and after this treatment in Table 3-4 in order of elution of

GC. Also the distribution expected under conditions of stereochemical equilibrium

was calculated assuming Flory's rotational isomeric state model50 and using Suter's

computed values for the relative energies of meso and racemic dyads in different

conformations25,27 (see chapter on conformational analysis of oligomers for

details). These calculated values are also listed below along with the indicated

tetramer stereochemistry. The good agreement of experimental and calculated values

tends to support the assignments. The assignment of the mrm and the mrr/rrm pair

was based on the observation with the vinylpyridine tetramers that GC elution order











Table 3-4. Total Epimerization of Tetramer
Distribution of Stereoisomers (%)
GC
Elution After Assignment
order Before 1 week Calculated Assignment Basis

1 30 25 27 mmm NMR, Calc, Epn

2 5 14 17 mrm GC, Calc

3 5 19 18 mrr / rrm GC, NMR

4 58 28 27 mmr/rmm NMR, GC, Epn,
Calc

5 1 7 5 rrr GC, Calc
6 1 7 7 rmr Epn, Calc






was determined by the external dyads,3 where Huang et al. found the order to be 1)

r...r, 2) m...r/r...m, 3) m...m. So the tBVK tetramer stereoisomers were grouped:
1) m...m, 2) m...r/r...m, 3) r...r since one from each group had been identified

earlier and the nrrr isomer was calculated to be present in the least amount.
At this point the following equations could be applied describing the

protonation and methylation products.
(mm-) = (mmm) + (mmr) (3-5)
(mr-) = (mrm) + (mrr) (3-6)
(rm-) = (rmm) + (rmr) (3-7)
(rr-) = (rrm) + (rrr) (3-8)











(mmr) + (rmm) = (mmr / rmm)
(mrr) + (Tm) = (mrr / rrm)


(3-9)
(3-10)


Several such "divided oligomerizations" were done. The results of one are
given in Table 3-5 as an illustration. It is believed that all possible combinations of
doubtful assignments were tried in order to achieve this solution.


Table 3-5. Tetramer Calculation the Best Fit
Methylated Tetramera
(mmm) = 19.5 (mmr) + (rmm) = 52.5
? (mrm) = 8.1 (rrr) = 3.0
? (mrr) + (rrm) = 12.7 (rmr) = 4.1


Protonated Tetramera
(mm-) = 67.9 =

? (mr-) = 17.8 =

? (rr-) = 6.1 =
? (rm-) = 8.2 =
a Percentages from GC.


Identities
19.5 (mmm) + 48.4 (mmr)

8.1 (mrm) + 9.7 (mrr)
3.0 (rrm) + 3.0 (rrr)
4.1 (rmr) + 4.1 (rmm)

? Indicates assignments in doubt before calculation.











End Methyl Group 13C NMR Assignments

Once these GC assignments were firm, a variety of experiments were

undertaken in order to complete the 13C NMR assignments of the tetramer end

methyl group signals. One of these involved initiating the oligomerization of t-BVK

with the lithium enolate of the 13C labeled t-butyl ethyl ketone prepared as illustrated


II 1. LDA, THF, 0C 0,L
S13C 13,-,
3. Ph3CL, THF


After addition of monomer the solution of living 13C labeled oligomers was

divided into two parts that were protonated and methylated respectively, and

worked-up by preparative LC in the usual manner. The 13C NMR spectra of

numerous LC fractions were compared to their GC analyses. The predominant

unsymmetrical methylated tetramer (IV) (mmr/rmm) now showed an almost

exclusive resonance at 17.4 ppm (*mmr) for the end methyl group with a negligible

signal at 17.1 ppm (*rmm). This is just the opposite of what was observed when

the terminal end methyl group was enriched and was expected (based on the results

already discussed).

Another experiment was designed to enhance the yields of non-isotactic

stereoisomers of protonated tetramer (IVp). By using the lithium enolate of

pinacolone (derived from its silyl enol ether) as initiator and terminating the

oligomeric chains with 13CH31, a 'reverse' protonated tetramer (pIV) was obtained

for which the amounts of the four stereoisomers were roughly equal.












Me3SiO\ 1. nBuLi, THF
tBC=CH2 2.n tBVK -78

3. 13CH31I

(7)


H CH2) C H3

(VC=0 n=0
tBu tBu


(pIV, n=3)


With more of these normally insufficient isomers available to work with,

separation and characterization went smoothly. The various LC cuts of tetramer

pIV were analyzed by GC, so the 13C NMR shifts of the methyl end groups could

all be directly assigned. These chemical shifts are listed with those for the methyl

end groups of the other oligomers in Table 3-6. It seems logical to expect to be able

to assign the protonated tetramer IVp (mm-, mr-, rm-, rr-) methyl end group

chemical shifts on the basis of comparison to those for trimer III (mm, mr, rm, rr)

whose assignments had been worked out earlier. But as can be seen from the values

listed in Table 3-6, this would have resulted in erroneous assignments of (mr-) and

(rr-).

The tetramer region of a gas chromatogram for one particular LC fraction from

a 13CH3I terminated oligomerization (shown below) may be compared to the

13C NMR end methyl signals (Fig. 3-13).




MMM
MRM
MRR/RRM
RMR MMR/RMM
RMRRR
> ---- = --- RMR











This sample also had some trimer present. The GC peak areas were normalized so

that all values represented the mol percent of total oligomer present. The 13C NMR

of its enriched end methyl region (Fig. 3-13) was taken with maximum digital

resolution, using a 900 flip angle for the rf pulse, gated proton decoupling to avoid

N.O.E. and with the total pulse delay plus acquisition time of more than 8 sec. The

integrated signal intensities are compared to the normalized GC areas below.


Table 3-6 GC and 13 1~-H) NMR Integrated Areas for one LC fraction
containing 1-C End-labelled Trimer (III) and Tetramer (IV)

13C NMR
GC (mol %) End Methyl (%)
II (mm) = 11.5 (mm) + (rrm*) = 17.7
(mr) + (rm) = 9.7 (rmr) + (mmm) = 31.4
(mrr*) = 13.4
IV (mmm) = 25.2 (mrm) = 7.3
(mrm) = 7.6 (rmm*) = 2.4
(mmr) + (rmm) =18.0 (rm*) = 4.8
(mrr) + (rrm) = 19.5 (mmr*) = 15.3
(rrr) = 2.4 (mr*) + (rrr) = 7.7
(rmr) = 6.1

Many such comparisons were done before making the assignments listed in

Table 3-7. It should be noted that the end methyl group chemical shifts for certain

stereoisomers exhibit a strong temperature dependence. Foremost among those are

the (*mr) and (*rm) trimer II and the (mrm), (*mmr) and (*rmm) tetramer IV

signals. In fact the (*mmr) and (*rmm) peaks have moved 0.2 ppm downfield with

increasing the temperature from ambient to 40 oC. The overall tendency is for the
end methyl peaks to bunch together more with increasing temperature. Also

different solvents (and concentrations) affect the chemical shifts of some
stereoisomers more than others.
























Assignments


IV rrm*
I l mm
IV rmr
IV mmm
IV mrr*
IV mrm


18.5


6

10
8



5 p S W I a 5 5 I


18.0


17.5


Figure 3-13. 25 MHz 13C NMR with gated 1H decoupling of 13CH3 end groups
of mix of tetramers (IV) plus trimers (III) in CDC13 at 40 C


IV rmm*
III rm*
IV mmr*
III mr*
IV rrr


17.0


I













Table 3-7. End Methyl Group 13C NMR Chemical Shifts

Methylated Oligomersa
Dimer H (m) 18.46 (r) 17.05


Trimer III


(mm)

(*mr)


18.55

17.44


(rr)
(*rm)


18.20
17.08


Tetramer IV


(mmm) 18.43 (rrr)
(*mmr) 17.57 (*rrm)

(mrm) 17.82 (rmr)

(*mrr) 18.58 (*rmm)

Protonated Oligomersa


Dimer IIp


18.09


Trimer IIIp

Tetramer IVp


(m-) 18.34

(mm-) 18.25

(mr-) 17.73


(r-) 17.21

(rr-) 17.66

(rm-) 17.20


a In ppm from TMS for concentrations 10-15% in CDCI3 at
40 oC for the methylated and at ambient temperatures for
the protonated oligomers.


17.11
18.02

18.45

17.33


















CHAPTER 4

OLIGOMERIZATION STEREOCHEMISTRY

With the identities of all the stereoisomers of methylated and protonated dimer

(II & IIp), trimer (III & IIIp) and tetramer (IV & IVp) established, we could

now direct our attention to analyzing the stereochemistry of reactions involved in

oligomerization, namely vinyl addition and methylation. The stereochemistry of the

oligomers terminated by silylation with Me3SiC1 is discussed in the chapter titled

"Structure of Enolates".

Methylation Kinetics and Stereochemistry

In the early days of this investigation, incomplete methylations due to

insufficient reaction times resulted in intractable mixtures that were nearly

impossible to characterize. The work-up technique employed contributed to the

problem since the solution of oligomers was allowed to warm to room temperature

and was concentrated by evaporating solvent before removal from the vacuum line

for the extraction. Since these oligomers were partly living and not completely

terminated, numerous side-reactions had undoubtedly occurred, including

condensations,15 epimerizations and fragmenting depolymerizations.51

This problem was avoided once it was realized how much time was required to

completely methylate these oligomers in THF using CH3I. This was determined by

studying the rate of methylation of the initiator (Li-tBEK, 6) by CH3I in THF at












-780C. The course of the reaction was followed by removing aliquots under Ar,

quenching them in acidified methanol and measuring by GC the relative amounts of

the protonated product, t-BEK (4) and t-BiPK (10) the methylated product. The

time-conversion curve for the reaction of a more than 5-fold excess of CH3I added

to 0.04 M lithium enolate of t-BEK (4) (prepared from the silyl enol ether) in THF

kept at -78 oC is shown in Figure 4-1. The results show that at least 20 h was

required to assure complete methylation under these conditions. This was

surprisingly long, but then the lithium enolate present in THF was clean, the only

side-product being the inert TMS. No amines were present, such as results when

LDA is used to form the enolate.15 Amines would reduce methylation time by better

solvating the Li+ counter-ion.

House et el.52 has suggested that the decreased reactivity of some alkali metal

enolates in ethers toward alkylation may be attributed to aggregation. Plots of In

[t-BEK] and [t-BEK]-1 vs time of reaction are included in Fig. 4-1. The excellent

linear correlation (0.998) for the [t-BEK]-1 vs time up to 60% completion indicates

that this reaction with excess CH3I is second order with respect to the lithium

enolate of t-BEK, i.e.


d[Li tBEK1 = k [Li tBEK] 2
dt


Methylation with methyl iodide involves a bimolecular nucleophilic

displacement of an iodide ion which associates with a lithium ion to form the

co-product LiI. The source of the lithium ion that actually associates with the

displaced iodide ion is open to question. The fact that the kinetics of methylation of







72







3.0-

-1.0


2.6- -

\-0.8


2.2- -

X o -0.6
tBEK
1.8- -


-0.4


1.4-

-0.2


1.0-



20 60 100 140 180

TIME (min)



Figure 4-1. Kinetics of the methylation of Li-tBEK in THF; X is the mol
fraction of tBEK (from protonating unreacted Li-tBEK)












the initiating lithium enolate (Li-tBEK, 6) were second order with respect to

Li-tBEK for the first half-life may indicated the necessity of the presence of another

lithium enolate 'ion-pair' to lend electrophilic assistance. The possibility that the

reacting enolate species is dimeric in THF will be discussed in more detail in the

chapter on the structure of enolates.

The stereochemistry of methylation of living' dimer (II Li) with methyl iodide

in THF at -78 C is non-selective. The results of GC analysis of methylated dimer

for several oligomerizations are listed in the Table below


Table 4-1. Percentage of Dimer and Stereochemistry of its Formation


Experiment
designation [M]/[I]

olig *12-3 2.1

mem olig 3.7

olig 7-7 3.9

olig 5-23 1.9

olig 3-26 2.2
a % of total oligomers II thru V.

reactions done in THF at -78 C


Dimer a Meso IIb

41 45

50 59

42 50

44 58

34 60
b % of dimer (m)/[(m) + (r)]


In order to determine the methylation stereochemistry for higher oligomer, it is

necessary to work with the data derived from divided oligomerizations to compare

the protonated and methylated parts. The data for two divided oligomerizations is

given in GC elution order in the Tables 4-2 a, b.













Table 4-2a. Stereoisomer Distribution of the Protonated Portion
from Oligomerizations Divided before Termination


Trimera

(m-) (r-,


olig *12-3 62 38

mem olig 67 33
a Percentages as determined by GC.


mm-)

68

74


Tetramera

(mr-) (rr-)

18 6

18 2


Table 4-2b. Stereoisomer Distributions of the Methylated Portion
from those Oligomerizations

Trimer


olig* 12-3

mem olig


(mm)
23

31


(mr /rm)
64 (39,24)

60 (36,24)


Tetramer

(mmm) (mrm) (mrr,rrm) (mmr,rmm)

olig* 12-3 20 8 13 (10,3) 52 (48,4)

mem olig 23 9 10(9,1) 55 (51,3)

Percentages as determined by GC and (calculated values)


(rrr) (rmr)
3 4

1 3


(rm-)
8

6


(











In order to calculate the overall percentage of meso-dyad ended chains, the
following relationships are employed:
For trimer, (m-) = (mm) + (mr) (4-1)
trimer, (r-) = (rr) + (rm) (4-2)
meso ends = (mm) + (rm) (4-3)
Values can be checked: (mr) + (rm) = (mr / rm) (4-4)
and for tetramer, (mm-) = (mmm) + (mmr) (4-5)
(mr-) = (mrm) + (mrr) (4-6)
(rr-) = (rrm) + (rrr) (4-7)
(rm-) = (rmm) + (rmr) (4-8)
meso ends = (mmm) + (mrm) + (rrm) + (rmm) (4-9)
Checks: (mmr) + (rmm) = (mmr / rmm) (4-10)
and: (mrr) + (rrm) = (mrr / rrm) (4-11)


The arithmetic was done and the results for trimer show an overall 47% and

55% meso stereochemistry of CH3I methylation (eqn. 4-3) for the two

oligomerizations designated olig* 12-3 and mem olig, respectively. This result
does not seem to be very different from that of the dimer. But the meso living
trimer (m-, III Li) shows a slight tendency to undergo methylation with
formation of a racemic dyad (ratios (mr) : (mm) = 1.7 and 1.2, respectively); and
methylation of the racemic living trimer favors even more the formation of the
meso end dyad from methylation (ratios (rm) : (rr) = 1.7 and 2.7, respectively).
And for the so-called "reverse" oligomerization that was initiated with Li-tBMK

(5) and terminated with CH31, the stereoisomers of trimer (pllI) were present in












the ratio (-m)/(-r) equal to 44/56. So even for living trimer that has no

diastereomers (pll Li), racemic methylation stereochemistry is slightly favored.

For tetramer (IV) the overall stereochemistry of methylation was 65% (and

64% for the mem oligomerization) racemic. This stereoselection is dominated by

the preference (70%) for racemic methylation of the isotactic (mm-) living

tetramer. All other stereoisomers of living tetramer for both runs show

non-selective methylation stereochemistry with meso to racemic and ratios equal to

1.0. The isotactic living tetramer is the predominant living tetramer in the

oligomerization solutions. The ratio of its methylated products, (mmr) : (mmm)

was found to be 2.4 and 2.2 (70% racemic) for the two oligomerizations listed in

Tables 4-2 a and b, respectively.
Stereochemistry of Vinyl Addition in THF

The most direct view of the stereochemical preferences for monomer addition

to living oligomers comes from examining the distribution of stereoisomers of

the protonate oligomers. These are listed for several oligomerizations in

Table 4-3. The predominance of meso protonated trimer (m-, IIIp) in all cases

indicates the more facile attack of monomer on the pro-meso side of the lithium

enolate functionality of living dimer (II Li). As pictured at the top of page 78, it

means that monomer located above the plane of the paper presents a better

bonding situation than below it. Since the four atoms: 0 (enolate), C(l), C(2)

and C(3) all lie in a plane, the Newman projection clearly depicts that, with the

carbonyl oxygen coordinating with the Li atom, approach from below is

somewhat hindered.













Table 4-3. Protonated Oligomerization Products Prepared in THF


Stereoisomers
IIIp m-, r-
IVp mm-, mr-, rr-, rm-


Olig* 12-3


Mem olig


Mani 2-17


Mani 3-8


Quench 3-29a


Pina 7-28b,c

Pina 6-4b


38-26-24-9


21 19 32 21


25- 21 33- 18


28 20 29 16


58-13-20-8


10-18-15-36

26 21 32 14


IIIp
IVp

IIp
IVp

I'p
IVp

I'p
IVp

I'p
IVp


62, 38
68, 18, 6, 8

67, 33
74, 18, 2, 6

70, 30
74, 20, 2, 5


71, 29
76, 20,


89, 11
95, 2, -, 3


pIVp 56 44

plVy 55 45


a Li-tBEK prepared from Ph3CLi + t-BEK. b
Li-tBMK prepared from Ph3CLi + t-BMK.


Li-tBMK initiated.


Oligomers
[M]/[I] IIp-IIIp-Vp-Vp














5 3H H
4 2 H viewed

O i H down
0 Li
0 2 C(3) C(4)







The sum of the fractions of the two tetramer (IVp) stereoisomers [(mm-) +

(mr-) = 86 to 96%] formed by monomer addition to the meso living trimer (m-, III

Li) is substantially greater than the amount of living meso trimer trapped by

protonation (m-, IIIp). This probably indicates the greater reactivity of the meso vs

the racemic living trimer (III Li) with respect to vinyl addition. But it must be

borne in mind that these are intermediates in a chain reaction; i.e., they are formed

from lower oligomers at different rates and are consumed at different rates to give

higher oligomers. The data could mean that both the rr- and rm- living tetramers

(IV Li) are much more reactive than the mm- and mr- and, by depletion, shift the

distribution.

Examination of the data for Quench 3-29 in Table 4-3 helps clarify this

situation. The oligomerization denoted Quench 3-29 went badly, but has proved
informative. Monomer with the inhibitor hydroquinone had not been distilled from

CaH2 before adding to the initiator solution. And in spite of the fact that excess












Me3SiCl/Et3N had been added to terminate the oligomerization by silylation, all

oligomers were found to be protonated. Apparently hydroquinone had distilled with

the monomer resulting in protonation. From the percentages of dimer (lip) through

pentamer (Vp) Table 4-3, it can be seen that the degree of oligomerization for

Quench 3-29 is lower than the rest (as expected from monomer to initiator ratio).

The stereoisomer distribution shows more isotactic isomers present in the trapped

products at the earlier period of oligomerization. This supports the contention that it

was the greater reactivity of the meso (vs racemic) living trimer that accounts for the

observation mentioned above, which was:

(mm-) IVp + (mr-) IVp > (m-) III

From the distribution of stereoisomers of IVp for the oligomerizations listed in

Table 4-3, it is evident that vinyl addition to living trimer (III Li) is meso

stereoselective, regardless of whether iI Li is m- or r-.

The fraction of end dyads of tetramer (IVp) that are meso are calculated as

follows:


from III Li (m-), (mm-, IVp) / [(mm-, IVp) + (mr-, IVp)]


from III Li (r-), (rm-, IVp) / [(rm-, IVp) + (rr-, IVp)]


and the values are listed in Table 4-4 on the next page.











Table 4-4. Percentages of Meso Dyads at Chain Ends Formed
from II Li, III Li, m- and III Li, r-


IIIp IV
from, II Li III Li, m- III Li, r-

Olig* 12-3 62 79 57

Mem olig 67 80 75

Mani 2-17 70 79 71

Mani 3-18 71 79 80

Quench 3-29 89 98 >90

Percentages calculated from GC data.






The results of oligomerizations designated Pina (initiated by the lithium enolate

of pinacolone (5) ) in Table 4-3 show much less meso tetramer plVp (55%) than

expected from the results just discussed for the other oligomerizations. Only two

diastereomers of tetramers pIVp are possible and their stereochemistry is

determined by vinyl addition to the living trimer pIII Li. Either meso

stereoselection for vinyl addition is much lower for this living trimer (HIII Li vs III

Li) or the resulting meso living tetramer pIV Li is much more reactive toward vinyl

addition than its racemic counter part (relative to the other living tetramers IV Li).

Unfortunately diastereomers of pentamer pVp proved impossible to separate by our

analytical techniques, so that the problem remains unresolved.












It should be noted that the stereochemistry of methylation of "meso" living

trimer (predominantly racemic) and the apparent stereochemistry of vinyl addition to

living trimer (predominantly meso) are contrary (Table 4-3). The chemistries of the

two reactions are quite distinct but there does not appear to be any good explanation

of the difference.

Though the identification of the protonated pentamers had not been established,

two isomers can be seen to predominate. They elute first and second among

protonated pentamers in the GC. (For protonated tetramers, mm- and mr- elude

first and second respectively and evidence was cited earlier to indicate that the

oligomer chain ends apparently determine GC elution order.) The 13C NMR

chemical shifts of the 13C labeled methyl group at the initial chain end of these

protonated pentamers are 18.13 and 17.66 ppm for the first and second isomers

(Vp) respectively. (Recall that *mm- (IVp) and *mr- (IVp) methyl end group

chemical shifts are 18.25 and 17.73 ppm respectively). Therefore the likely identity

of these two pentamers (Vp) can be assumed to be mmm- and mmr-. The relative

amounts of the two isomers in the two runs under scrutiny were 19 and 45% for

olig* 12-3 and 24 and 61% for mem olig; that meant ratios of 2.4 and 2.5 for the

presumable (mmr-) : (mmm-).

The stereochemistry of methylation of living isotactic tetramer (mm-,IV Li)

and vinyl addition of monomer to this same living oligomer appear very similar. If

the assumed identities for the predominant protonated pentamers (Vp) were correct,

it means that the electrophile, regardless of whether it is methyl iodide or the

monomer, prefers to approach the pro-racemic side of the enolate end of the living












isotactic tetramer. Or better worded, the electrophile is more likely to encounter

bonding situations on the pro-r side than on the pro-m side of the Li enolate of living

isotactic tetramer.

In view of these results, it seems likely that one particular conformation of the

living isotactic tetramer allows intramolecular coordination of the lithium ion

associated with the enolate end such that approach is mainly open to the pro-r side.

Or it may even be that the dominant conformation in solution (without the necessity

of intramolecular coordination) hinders access to the pro-m side of the enolate. This

will be examined in more detail in the chapter on conformational analysis.

Vinyl Addition in Hexane

All but one of the oligomerizations were done in THF. For that one, hexane

was used as a solvent and most of the monomer was converted to polymer which

was insoluble in all common lab solvents tested. However some small quantities of

oligomers were found that reveal some stereochemical details of the vinyl addition

reaction in this media

Before addition of monomer, the silyl enol ether of t-BEK(8) was stirred with

an equimolar amount of n-butyl lithium in hexane at room temperature for 25 h.

Nonetheless, oligomers initiated by n-BuLi represented the major portion of the

product. In view of the evidence that n-butyl lithium is found as hexameric

aggregates in hexane,53 the hydrocarbon sphere of n-butyl chains would be

expected to make the approach of the already sterically hindered enol 0 to the Li

atoms in the core of the hexamer very difficult indeed. Only one oligomer initiated

by the lithium enolate of t-BEK(6) was seen, the trimer. The solution of living

oligomer (and polymer) was divided before terminating by protonation and












methylation. (In order to hasten the methylation with CH3I, THF was added to the

solution beforehand.) 97% of the protonated trimer (HIp) was detected to be meso

(m-); for the methylated trimer (HI), the mol percent of the diastereomers (mm) -

(mr/rm) (rr) was measured as 51.5 46.5 2.0. Clearly a highly stereoselective

Michael reaction of the dimer enolate (HI-Li) to t-butyl vinyl ketone in hexane is

indicated by these results. The absence of any enolate initiated dimer (Hip or II)

may be an indication of the relative stabilities of the lithium enolate ended oligomers,

dimer (II-Li) vs trimer (III-Li) in hexane.

The n-butyl lithium initiated dimer was the only other oligomer found to be

present in appreciable quantities. The methylated dimers (n-Bull) were

characterized by 13C NMR and found to be 88.5% "meso". (If the initial n-butyl

group were replaced by methyl, the stereoisomer would be meso.) This unusually

high meso-like methylation stereoselectivity for the living n-butyl dimer is somewhat

perplexing, especially since the methylation of the living trimer (III-Li) in the same

reaction pot showed little or no stereochemical preference (52.5% meso).

Thermal History of Living Oligomer Solution

Since methylation required more than 24 h at -78 oC to complete, the question

naturally arose as to whether self-epimerization or other side reactions might be

occurring. That is, are these basic living oligomers causing racemization of the

acidic chiral backbone methines of other chains (or itself) under these conditions?

To answer this, experiments were undertaken in which the solution of living

oligomers was divided at -780C via a manifold into several lots. All samples were

protonated with acidified methanol (10% glacial acetic acid) and worked-up in the












standard fashion. The treatments of the various ampoules of solutions of living

oligomers are summarized:

A protonated in vacuo at -78 oC and kept at -78 oC overnight

B protonated immediately after removal from dry ice slush bath; left at

room temperature overnight

C warmed to room temperature (24 oC) and kept for one hour before

protonation

D warmed to room temperature for 4 h, then protonated

E kept at -78 oC for 55 h; then protonated directly from the bath

F diluted to twice its original volume by vacuum distillation of THF

from lot G; left at room temperature for 37 h before protonation

G concentrated to half volume and left for 37 h at room temperature

then protonated






The distribution of stereoisomers of trimer (IIIp) and tetramer (IVp) as

analyzed by capillary GC are given in Table 4-5. From these results it's clear that

when the solution of living oligomers is kept at -78 oC (even for two days), the

distribution of stereoisomers (for trimer and tetramer, at the least) remains unaltered.

This evidence should remove any doubts about possible side reactions occurring

during termination by reaction with methyl iodide at -78 oC.







85



Table 4-5. Thermal Histories Distribution of Stereoisomers


Treatment:

Trimer Hip


A B E


C D F G


70 70 70

30 30 30


73 70 58 60

27 30 42 40


Tetramer IVp


mm-

mr-

rr-

rm-


74 74 74

20 20 20

2 2 2

5 5 5


69 47 33a 28a

21 30 29 24


Percentages determined by GC; a Extra tetramer peaks are seen for F
and G and the GC analyses of their distribution is therefore doubtful.





Table 4-6. Thermal Histories-Distribution of Oligomers


A B E

25 24 26

21 21 21

33 33 32

18 19 16

3 3 4


C D F G

25 25 15 19

24 33 27 30

28 21 26 25

18 13 17 15

5 7 11 9


3 2


3 7 2

7 17 36


(48)


"Ip

Hip

IVp

Vp

VIp

VIIp


1 1












Before any comment is made about the distributions of stereoisomers for the

lots left at room temperature for different lengths of time, let us examine the

distribution of oligomers analyzed by GC as shown in Table 4-6.

From the changes in the relative amounts of the various oligomers present, it is

apparent that reactions other than epimerization are occurring. Besides

deprotonating other species, carbanions can undergo elimination to yield olefins.

These could be essentially depolymerizations for an anionic system above its ceiling

temperature or eliminations of olefinic fragments other than monomer. Carbanion

attacks on the pendant carbonyl groups along the chain would lead to aldol-type

condensation products. Although these nucleophilic attacks are conceivable, they do

not occur due to the steric bulk of the t-butyl group. The appearance of unidentified

peaks in the tetramer grouping of the gas chromatogram for the two lots of living

oligomer left 37 h at room temperature indicates the complex nature of these side

reactions. Thus for one oligomerization initiated by Li-tBMK and terminated with

13CH3I (forming so-called 'reverse' protonated oligomers), identifiable


side-products were found. Among the pIV tetramers were 'normal' IV tetramers,

methylated at both ends. A mistake in termination technique was the cause of their

formation. Less than an equivalent of the expensive reagent 13CH3I was added to

terminate 'living' oligomers. To assure complete consumption of 13CH3I, the

mixture was allowed to warm for a short period following the usual overnight period

at -78 oC, before adding excess unlabeled CH3I to complete methylation.

Evidently, deprotonation of the methylated oligomers by the living oligomers had












occurred. Similar carbanion equilibria for vinyl pyridine oligomers had been studied

by Meverden and Hogen-Esch.51

Polymer Stereochemistry

Only a qualitative overview of the stereochemical differences among poly

t-BVK samples prepared in different ways is possible. A soluble PtBVK sample of

a low degree of polymerization resisted all attempts to improve resolution of its

NMR spectra including the use of a variety of solvents, high temperature, low

temperature, lanthanide shift reagents and chemical alteration (oxime and hydrazine

derivatives, and LiAlH4 reduction). Regardless of these failures, interesting

similarities and differences can be noticed when comparing the carbonyl region of

the 13C NMR spectra for these different polymers (Fig.4-2).

The 13C NMR spectra of PtBVK from the AIBN initiated polymerization in

CC14, polymer that spontaneously formed in an ampoule of purified (uninhibited)

monomer and the PtBVK synthesized by group transfer polymerization (GTP) are

all very similar. The stereochemical composition of PtBVK polymerized under

free-radical conditions has been determined by A. Klaus 20 to be 55 4% meso. It

is not surprising to see that PtBVK "GTP" has similar stereochemistry considering

the observations for PMMA.54

The spectra of the two PtBVK samples anionically polymerized in hexane,

"Lite" and "Copoly", are very much alike and distinctly sharper than the

free-radically initiated PtBVK spectra. "Lite" refers to the less-dense, waxy

polymer centrifugate separated out from the anionically homopolymerized t-BVK,

and "Copoly", the block copolymer of butadiene (DP = 50) and t-BVK (DP = 15).





















































Pill'Hillull'li aunpreignualia
220 218 216


Figure 4-2


II i 1 llrl I i
214 ppmn


13C- {1H) NMR Carbonyl region of different tBVK polymers


m












Suter et al.25 have demonstrated by X-ray diffraction and 1H NMR in

C1F2CCOOH that PtBVK prepared using n-BuLi in hexane at -78 OC is highly

crystalline and isotactic [(m) > 0.90].

And different yet are the spectra called n-BuLi neat and THF. The first being

polymer formed when n-BuLi in hexane was injected directly into neat t-BVK

(solvent free); THF designates the common PtBVK side-product formed in the

lithium enolate initiated oligomerizations in THF. It is startling just how different

the carbonyl 13C NMR spectrum for the PtBVK THF is from all the rest.

Of the inner carbonyl signals for timer, the heterotactic peak lies upfield and

the isotactic downfield (with syndiotactic near the middle ). It was hoped (if not

expected) that the signal for the carbonyl flanked by (m) and (r) dyads would be

found between those for the inner carbonyls of the (mm) and (rr) triads. Because

this was not the case for trimer, it meant that extension of this data to interpret the

higher n-ad(odd) sequences of poly(tBVK) was not feasible. Instead, defensible

models for pentads and heptads were needed.

The evidence of how vinyl addition to the predominant isotactic living tetramer

(mm IV Li) appears to result mainly in mmr pentamer (Vp) was discussed. If the

polymer formed anionically in THF at -78 oC consisted mainly of these mmr tetrad

sequences repeating, the PtBVK chain
... mmrmnnmrrrrnmmrm rrmm mmrmr rmmr ....

would be comprised of heterotactic and isotactic triads in the ratio 2:1. This is close

to the ratio observed for the major carbonyl peaks at 216.2 and 217.5 ppm for

PtBVK from THF. The lack of proper pentad models leaves this discussion merely

speculative.















CHAPTER 5

STRUCTURE OF ENOLATES

Li-tBEK Initiator

The lithium enolate of t-butyl ethyl ketone (2,2-dimethyl-3-pentanone) was

used to initiate most of the oligomers done in this study. It can, in principle, be

formed as two geometrical isomers:


CH3 t-Bu H t-Bu


H OLi CH3 OLi
(E) (Z)

The stereostructure of the enolate was determined by first trapping it as the

trimethylsilyl enol ether from reaction with an excess of trimethylsilyl chloride .55

This product was examined directly for purity by capillary GC. Once purified by

spinning band distillation (bp. 90 oC at 54 mmHg), it was analyzed by 1H NMR

and 13C NMR. An earlier study56 had "confirmed" the single silyl enol ether of

t-BEK to be the (Z) isomer based on its quaternary 13C NMR resonance at 36.7

ppm being 0.6 ppm greater than that for the quaternary carbon resonance of the silyl

enol ether of 3,3-dimethyl-2-butanone (t-BMK). The reasoning expressed in

the paper was that the trans-y-methyl substituent effect of 0.6 ppm is a normal

value. We sought to establish its geometrical identity on a firmer basis. The totally

proton coupled 13C NMR signal for this quaternary carbon of 8 is shown in

Fig 5-1(b). By irradiating only the protons of this t-butyl group, the splitting of

90
















CH OSi(CH3)3

0=c
H/ C(CH3)3
.2.2Hz


(b)


36.8


.36.7


36.5


I I I I I I I I I I I I 36 .3
36.6 36.5 36.4 36.3


Figure 5-1. 75MHz 13C NMR of the quaternary carbon of Si-tBEK (8) and
Si-tBMK (7). (a) selectively proton decoupled (b) totally coupled


. . .




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EXPTHBS6S_K98LTF INGEST_TIME 2012-02-17T16:42:08Z PACKAGE AA00003803_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES