I
I

I

  

IIIII

I

I R I

II

I
I

I“

III)

I
I
I

I
H

II

FACTORS AFFECTING ASYMMETRIC
INDUCTION IN. ADDITIONS T0
2.3- DIMETHYLBUTANAL AND
3,4-DIMETHYL-2-PENTANONE

Thesis for the Degree of PIL D.
MICHIGAN STATE UNIVERSITY
THEODOSIDS C. CHRISTODOULIDIS
1972

 

 

This is to certify that the
thesis entitled
FACTORS AFFECTING ASYMMETRIC INDUCTION

IN ADD IT IONS T0 2 , 3-DIMETHYLBUTANAL
AND 3 , 4-DIMETHYL- 2-PENTANONE

presented by

Theodosios C. Christodoulidis

has been accepted towards fulfillment
of the requirements for

 

 

Ph.D. degreein Chemistry

H [MA/:1—

Major professor

Date August 22 , 197 2

 

0-7639

  
      

 
 

.‘ no av
Wr suns
880K smnm mc.

LIBRAHY BIHDERS

ABSTRACT

FACTORS AFFECTING ASYMME'I'RIC INDUCTION IN ADDITIONS TO

2 , 3—DD’IEI‘HYLHJTANAL AND 3 , lI-DIME'I'HYL-Z—PENTANONE

by

Theodosios C . Christodoulidis

Several models have been developed that predict the course of
asymnetric induction in additions to carborwls directly bonded to asym—
metric centers. The Karabatsos model1 makes not only qualitative pre-
dictions, as the others do, but also quantitative predictions as well.

The model was tested by analysis of the products obtained from addi-

tions to 3,11-dimethy1-2—pentanone of lithium alurrdnum hydride in ether
and tetrahydrof‘uran, and of sodium borohydride in 2-propanol ; additions

to 2,3—dimethy1butanal of methylmagnesium iodide in ether, methylmagnesium
bromide in ether and tetrahydrofuran, methylmagnesium chloride in tetra-
hydrofuran and methyllithium in ether. Besides the nucleophiles , other
variables were also controlled (solvent, temperature, and concentration
of reactants). The ratio A/B of the diastereomeric alcohols, determined
by vpc and nmr, was used to calculate the differences in the free energies

of the diastereomeric transition states.

4‘.
AAGAB—-R'I‘lnA/B

Theodosios C. Christodoulidis
Fran the AAHIIB and AASjCB values, obtained by plotting ln A/B
versus l/I‘, we concluded that most of the reactions were enthalpy con—
trolled. Their stereoselectivities depended on the nature of the at-
tacking nuc leophi les . They were independent of concentrations of the
nucleophiles and the polarities of the solvents . The experimental

results agreed fairly well with those predicted by the Karabatsos model.

 

l. G. J. Karabatsos, J. Am. Chem. Soc., Q2, 1367 (1967).

 

FACIORS AFFECTDVG ASYMMETRIC INDUCTION 1N ADDITIONS TO

2 , 3—DIMETHYLBU'TANAL AND 3 , ll—DIMEITHYL—2-PENTANONE

by

Theodosios c'i‘ Christodoulidis

ATHESIS

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

Department of Chemistry

1972

r. ‘3')

II.
1.

Qt
(M.

gal... j

TOMY PARENTS

ii

ACIWOWIEDGMENTS

The author wishes to express his appreciation to Professor
G. J. Karabatsos for his guidance during the course of this investigation.

Financial assistance from the Michigan State University Chemistry
Department, and from the National Institutes of Health is gratefully

acknowledged .

iii

Tyger! 'I‘yger! burning bright
In the forests of the night,
What immortal hand or eye

Dare frame thy fearful symmetry?

William Blake

iv

TABLE OF CONTENTS
Page

ACMOWLEDGIVIENTS...............iii

LIST OF TABLES . . . . . . . . . . . . . . . Vi
LIST OF FIGURES . . . . . . . . . . . . . . . Vii
INTRODUCTION . . . . . . . . . . . . . . . . 1
RESULTS AND DISCUSSION . . . . . . . . . . . . . 8
A. Calculation othctivation parameters . . . . . . 15
B. ‘Effect of Nucleophile . . . . . . . . . . . 23
C. Solvent Effects . . . . . . . . . . . . . 26
D. Entropy Effects . . . . . . . . . . . . . 27
E. Configurational Studies . . . . . . . . . . 28
F. Conclusions . . . . . . . . . . . . . . 28
EXPERIMENTAL . . . . . . . . . . . . . . . . 30
A. General Method . . . . . . . . . . . . . 30
B. Preparation of 3,4—Dimethyl—2—pentanone . . . . . 31

C. Preparation of 3,fléDimethyl-Z-pentanone—
1,1313% o o o o o o o o o o o o o o 32

D. Additions to 3,A—Dimethyl-2—pentanone—
1,1,1,3-9J4 o o o o o o o o o o o o o o 32

E. Preparation of 2,3—Dimethylbutanal . . . . . . 36
F. Additions to 2,3éDimethylbutanal . . . . . . . U0

G. Preparation of'Threo—3,u-dimethyl-2—
pentanol...............u1

BIBLICX‘zRAPI-IYWLQ

LIST OF TABLES

Table Page

I. The effect of temperature on asyrmetric
inductions of 3,1I-dimethyl-2-pentanone . . . . . . 9

II. The effect of temperature on asymmetric
inductions of 2,3—dimethylbutanal . . . . . . . 11

III. Asymmetric induction of some selected
carbonylsystems............. 111

IV. The effect of nucleophile on asymmetric
inductions of 2 ,3-dimethylbutanal and
3, ll-dimethyl—2—pentanone . . . . . . . . . 16

V. The effect of nucleophile concentration
on asymmetric inductions of 2,3—dimethy1—
butanal and 3,1I-dimethyl-2-pentanone . . . . . . 17

VI. Solvent effects on asymmetric inductions
of 2 ,3—dimethylbutanal and 3, A—dimethyl-
2-pentanone . . . . . . . . . . 18

VII. AAH+andAAS+values. . . . . . . . . . . .522

Figure
l.'1heopenchainmodel. . . . . . . .
2. The rigid and dipolar model
3. Energ relationship in the prediction
of diastereomers A and B . .
ll. Transition states leading to diastereomer A
5. Activation parameter plots for additions of
LiAlHu to 3,ll-dimethyl-2-pentanone . .
6 . Activation parameter plots for additions of
CH3MgBr to 2,3—dimethylbutanal . . . .
7. Activation parameter plots for additions of
CH3MgI and CH3MgCl to 2 ,3—dimethylbutanal .
8. N.M.R. spectra of (a) 3,1l-dimethyl-2—penta-
none and (b) 3,1l—dimethyl-2-pentanone-l,l,l,
3—g1‘ I O O O O O O O O
9. N .M.R. spectrum of 2,3,ll—trimethyl-2-penta-
nol-l,l,l,3—gu
10. N. M. R. spectrum of 3,A—dimethyl—2-pentanol-
1,1, 1,3-d
11. N. M. R. spectrum of 3, A—dimethyl—2—pentanol
12. N.M.R. spectrum of 2,3-dimethylbutana1 .

LIST OF FIGURES

Vii

Page

19

2O

21

33

35

37
38
39

INTRODUCTION

The problem of asyrmetric induction started at the end of the last
century with the classic work of Emil Fischer on the synthesis of D—glu—
cose.1 As early as 1901:, w. Markwald gave the following definition of
asymmetric synthesis.2 "Asymmetrische Synthesen sind solche, welche aus
symmetrisch constituirten Verbindungen unter intermediarer Benutzung
optisch-activen Stoffe, aber unter Vermeidung Jedes analytischen Vor-
ganges , Optisch-active Substanzen erzeugen."

A more inclusive definition than that of Markwald's is: Asym-
metric synthesis is a reaction in which an achiral unit in an ensemble
of substrate molecules is converted by a reactant into a chiral unit in
such a manner that the stereoisomeric products are produced in unequal
amounts. That is to say, an asymmetric synthesis is a process which
converts a prochiral unit into a chiral unit so that unequal amounts of
stereoisomeric products result . 3

Recently Y. Izumi proposed the division of asymmetric synthesis
into enantioselective syntheses and diastereoselective syntheses .“

From the pioneering work of McKenzie,S Erlenmeyer and P. Ritchie,
we care to the work of Prelog‘5 who reinvestigated McKenzie's work and
tried to rationalize the asymmetric induction by assuming steric inter-
actions among the possible reacting conformers .

In the early 1950's Cram and Abd Elhafez published a paper in
which they formulated a rule of steric control of asymmetric induction.7

"In non-catalytic reactions of the type sham (formulas), that diastereomer

0
Rs m
R'—._—-_==> é————- R'
“E

I

Figure l. The open chain model. Steric bulk is RS<RM<RL

will predominate which would be formed by the approach of the entering
group from the least hindered side of the double bond when the rota-
tional conformation of the 0-0 bond is such that the double bond is
flanked by the two least bullqr groups attached to the adjacent asym-

metric center. "7

This empirical model was fairly successful but very
soon, after its publication, exceptions to it were reported.

'IWo more models were developed in 1959, dealing with the dis-
crepancies of the open chain model. These were Cram's rigid model I13,

and Cornforth's dipolar model 9 III, depicted in Figure 2.

zg'w
O O (N) 0
RL Rs
(-—-——- R!
% Rs
R R Cl
II III

Figure 2. The rigid model II, and the dipolar model III.

For the kinetically controlled asymmetric reactions AAG°= 0 but
MG+ 7‘ O, as illustrated in Figure 3. These reactions are stereoselec-
tive and the degree of stereoselectivity depends on the energy differ-

ence between the two diasterecmeric transition states.

5+
_ .1} _____
+ AAGI 7‘ O
_ _ -L __ __
AAG°= O
R,S

Figure 3. Eherg relationship in the production of diastereomers
A and B. Transition states are indicated as A4: and B+.

Quantitatively the per cent stereoselectivity e is defined as
e = 100 (A—B)/(A+B), where A and B are the diastereomers formed. Thus,
the greater the AAG+ is, the greater the e ; and if AAGI is zero, then
0 is zero.

In 1967 Karabatsoslo suggested a semiempirical model on the basis
of which approximate quantitative predictions could be made. In devising

the model he assumed the following: (1) "Little bond breaking and making

has occurred at the transition states.

Consequently, the arrangement of

the groups of the asymmetric carbon atom with respect to the carbonyl

group is similar to that about sp2 - sp3 carbon - carbon bond." (ii)

"'Ihe diastereomeric transition states that control product stereospecifi—

city have the smallest group closest to the incoming bulky group R' ."

 

 

VI

Figure A. Transition states hading to the diastereoner A.

Considering the three transition states depicted in Figure ’4, Karabatsos

arrived at the conclusion that transition state IV should be the most stable.

This conclusion was based on the evaluation of the following interactions

in each conformer.

 

 

Conformer Interacti ons
o B y 6
IV R '«RS R“RS R0131L PM" 0
V R 'HRL RWRL R9}?M RS” 0
VI RH—vRM RflRM R«RS 13L“ 0

 

Interactions “IV and BIV in IV are energetically more favorable
than “V and IV in V. YIV g 8V and cancel each other out. From nmr and
microwave data it is knom that interaction 6 IV favors conformer IV over
V.ll 'Ihus, IV ouglt to be more stable than V.

From a comparison of IV and VI it is evident that interaction
YIV is energetically less favored than BVI' However, “IV and 6IV are
favored over the corresponding “VI and YVI’ Hence, IV ought to be more
stable than VI. Repeating the same process for the analogous three tran-
sition states leading to diastereorer B,
we find VII to be the most stable tran—
sition state. No», we have to compare
the energetically favored diastereorer'ic

transition states IV and VII. and

0‘IV

5IV are the interactims favoring IV

 

over VII, and YVII is the interaction
VII favoring VII. From these interactions
only the (RHO), or 6, is evaluable
spectroscopically.11 For example, (Me+~O) - (iPraO) is expected to be
about 4400 cal/mole. The other gauche interactions, o -y, are unpredictable.
Karabatsos used the values he obtained for the 6 interactions from nmr data
to predict the free energy difference of the two diastereoreric transition
states and from these the product ratio A/B. 'Ihe eXperimental AAGIAB were

obtained by using the Curtin-Hammett principle.”

AAG+AB = —R‘I‘ln A/B (l)

'Ihese predictions were made with the understanding that the ex—
perimental AAGIAB values, for a given chiral center, will vary as long
as the groups R and R' will vary in size. The model does not take into
account the extent of bond breaking and making in the transition states,
any differences in solvation, or any differences in the entropies of the
two transition states. Recent investigations in these laboratories ,1“ on
3-methoxy-2-butanone and 3-phemrl-2-butanone , revealed that the above
factors are important.

Noting that both previous compounds studied in these laboratories
contained polar groups, we decided to study 2,3-dimethylbutanal and 3,14-
dimethyl-2—pentanone . The fact that the chiral centers of these two com-
pounds do not contain any polar groups makes them excellent substrates to
test the Karabatsos model.

Before discussing the results obtained in this investigatiol a few
words about the "stereochemical analogy model" introduced by E. Ruch and
I. Ugil3 are in order. 'lhis mathematical model, based on group theory,

allows one to make quantitative predictions as outlined be low. The authors

L R x R

 

1 Q R 3 x
\uC-——V-—-—C/ X‘Y - °/ + —c‘\/
L O“ \\ r C\
3 u 11—37 \
L2 u—y
A B

assumed that all ligands are achiral, or that their chirality does not in—
terfere with the stereochemistry of the reaction. The product ratio A/B
is given by the concentration ratio of the transition states , since the

reaction is kinetically controlled. The stereoselectivity of the reaction

is defined as in eq. 2, which represents a linear free energy relation-
ship. 'lhe polynomial is a product of the ligand parameters A1,A2,l3.

61nC

SR
r[here is no stereoselectivity when two or more of the ligands, Ln’ become
equal. The ratio CRR/CSR is the concentration ratio of the diastereoreric
reaction products. A norenclature factor 6 , being equal to +1 for the R-
sequence and to -l for the S-sequence, was introduced, to use the Cahn,
Ingold and Prelog R ,S ncmenclature.15 From experimental data,17 they cal-
culated the A values of nine ligands and the p values of two reactions.“5
This method is of limited value thougu, since the A and p values vary with
different reactions. Also, it cannot be used in cases of adjacent chiral
centers, thus excluding the large number of 1,2-asymmetric inductions,

where the Karabatsos model can be used to make quantitative predictions.

 

-__—

RESULTS AND DISCUSSION

The results of the additions of lithium aluminumlhydride and sod-
ium borohydride to 3,4-dimethyl-2—pentanone are summarized in Table I.
Those of the additions of methylmagmesium iodide, methylrnag’iesium bromide,
methylmagnesium chloride and.methyllithium to 2,3—dimethylbutanal, are
presented in Table II.

The above mentioned reactions were carried out at various substrate
concentrations, solvents, and temperatures. The transition states leading

to the diastereoreric product ratio A/B are depicted below (VIII, IX).

IO iPr

-_R' '______|

VIII IX

When R=H and R'=- -CH3, the predominant diastereorer is the erythro—3,A—
dimethyl-Z—pentanol (R—R and 8-8 mixture).

 

 

 

 

 

 

em a new: 0.0: " 0.0m mm NH
em a gem- o.He H o.mm om H.o see sesame as
em a mem- o.mm H o.wo mm- OH
em a omm- o.mm " o.mo m.H a
as a mam- o.em H o.mo mm m
me a mem. m.mm ” m.Ho mm mo.o steam emaaas e
em a 0mm- o.mm " o.me me. o
mm H gml m.mm n 56m mm... m
m: n mmml m.mm " mém m a
me a mem. m.mm " s.ao mm m
as a mmm- e.mm " m.Ho mm H.o bosom senans N
am a sea- e.me " o.om mm s.o bosom emaans H
Home em 0 < .ewme o . osoo outflow 238282 Se
b. s. osossosooumuafiosaous . m
mo 953035 oEumEEnmm co gufioflfip mo poommo one .H mama

lO

.ofipmp occuox\oafieoooaosc H "

amm.H no use oofiphmo mm3_cofipomom

m

.oaosxasoo

 

.mHOQMpcoolmlamepoeaona.m OfigmeomgopmMHo mo oapmm omnpzpm\omae9o

.oapmp econox\oafinoooaosc H "

.poo>H0m mo H\oaflgoooaosc mo mofiozo

mo.m um poo oofighmo mums mCOHpowom

n

.emrm.H.H.Huosossosooumufiseoosaoue.m one eons o3 otfi mess eHs

 

 

 

o: a cam- o.mm " o.Ho me omm.o momma emmsz AH
am a cam- o.mm " e.oo mm- on
mm H emml m.mm ” m.mm 0H! ma
mm H mam! m.:m " m.mm m :H
on n cam- o.em " o.mo mm H.o one emaaaq ma
mwoea so u a .omms o.oeoo bsosaom ofifieoooaosz ssm

 

 

Aoossaosoov H money

ll

 

 

 

:NH H mmml O.mH u O.Nm m.H mm
NMH H 0mm! N.NH H w.Hm mm mm
NMH H mam! w.wH u :.H@ mm 5m
9:. ... so. mas. u now so do see newsman on
mOH H 0H5! m.mH " m.om mml mm
mOH M NONI m.HN u m.ww m.H 2N
OHH H mmwl N.mm " w.>n Hm mm
m: a em? eon . :5. mm mos bases 3sz mm
mm H szl N.HN n m.m> NNI Hm

hm H wzml :.mm H m.m> m.H ON

50 H wzwl o.mm " o.m> mm mH

mm H 03ml 0.0N u 0.35 mm H.o Hmnpm HwZMmO ®H

omnfiog om " < . 9W9 n. 080 pcofiuom 33903032 com

 

 

sfisqsosogfisaonma do nsoflosofi 3585mm so assessments. oo boots o5. .HH Ema.

l2

 

 

 

own A mow- m.mH " N.Hm m as
Add A cam- 0.0m " 0.0m am on
:ma n Rm: how. A TE em mm
as e. as- 9mm " at so do so. Seams am
sea a mag. o.ea " o.Hm Hm- em
gas a ewe- m.eH " m.om m.H om
mmH a afim- 0.0m " 0.0m mm mm
was A one. an " new mm mes meson seams em
sea a New- 0.0m " o.ow Hm: mm
HHH A see. m.om " m.me m.H mm
mas a meet o.om " e.me mm Hm
ema n saw- o.Hm " o.me mm H.o bosom smegmmo om

ms 00

$.02 om " a. .989 bosoo boot/How oHEoooHosz Se

 

 

Aoossaoeoov HH mamas

l3

.mHoEKHmoo

 

.nHossosooumangoosHoue.m onoeoogoonoHo oo oHosb oogeexogsugumo

.pco>Hom mo H\oHH£oooHozc mo moHoz

o

.oHpmp moznonm\oHHnoooHosc H " mew.H no use oothwo one: mCOHuomomw

 

 

 

em a mew: m.mm ” m.me H w:
om H mmml m.mm A m.He m.Hm e:
a a. am. one. " one mm moo hoses ammo es
mm A mam: o.mm " o.me mm: m:
em a 2mm: m.mm u m.:e H 3:
mm a gem- o.mm " o.ms m.Hm me
Hm H mom: 0.5m u o.me mm H.o Locum qumo me

me 00

Hy+o<< om " < .osoe n.o:oo pcm>Hom oHfinoooHosz 25m

 

 

Aoossaosoov HH mumae

lll

 

Homsz

 

 

 

 

mm OOH- OOH- O mosmH

mm OOH- OO- mm bosom OoHeHH omogmmmovoOmmO
mm Oom- OHH- O sebum Hoe:

mm OOm- OOm- mm steam Hmzoz omOmOAHHmOOmeO
Hm OON- OHs- OO one Homes:

Hm OOm- OOm- mm noses smwzmz

Hm OON- OOH- mm bosom szoz moOOOHmmOovoogmH
Hm OON- Ome- Om mogmH Homsz

Hm OOm- OOm- OO age eomsz

Hm OON- Omm- mm bosom OOHOHS omogmmOOOQOboH

ma Os
.oom .OHsO .ogo _+O<O .osoe peo>Hom oHHeoooHosz oosbpnbsm

 

 

.mEopmgHm Hagoono oopooHom 95m no 83965 oupéme. .HHH mama.

15

When R= -CH and R'=H, the threo alcohol (R—S and 8-H mixture) predomi—

3
nates. Verification of the above assignments will be given later in
Section E. Table III contains asymmetric induction results of sore sel—

ected carbonyl systems .

A. Calculation of Activation Parameters.

 

The difference in the free energies of activation of the two
diastereoreric transition states was calculated by using the Curtin—

Hammett principle, 12

+ _
AAGAB — -RTln A/B

where A/B is the ratio of the diastereoreric products A and B, R is the
ideal gas cmstant, and T is the absolute temperature (Tables I - VI).
The errors in AAGtB values were calculated as follows: The maximum de-
viation in the temperature of i2° was taken into account together with
the maximum possible fractional error in the ratio A/B which ranged from
2-8%. This error was obtained by integrating eight times the areas under
the appropriate signals in the nmr and vpc tracings.

The equation given above was combined with

+ _ + +
AAG - AAHAB -TAASAB

AB
to give:

1n A/B = -AAHiB/RI‘ + AAStB/R

By plotting ln A/B gs. l/T (Figures 5-7) we obtained the various MB:B
and MBA-B values. The values and the errors reported in Table VII were

calculated by using the ICINFIT corputer program of Dye and Nicely. 18

16

.onm pm mmB CH 2 H.o .omm pm Hmnpm CH 2 H.oo

o

.om: pm mOLmH CH 2 mmo.o .omm pm Hmzpm CH mo.ow

o

 

 

HO n OOO- Oh “ OT eH-Hmmo we
ONH n RO- EON u mag OHowzmmO Om
HOH a emO- O.Hm " O.Oe osmwzmmo Om Omomwmmo
mAmmOVmO
mm o OOO- 0.0m H O.ee OHmzmmO OH
O: a OAN- O.Om " O.HO bemmsz sH
A: a mHm- O.em " O.mO semHOHH O mmooomwmmo
mgmmovmo
mwoaa m n a oHHsoooHesz a use oosgonosm

 

 

.oeossosoo-m-HzepoEHs-e.m one
HessostsgsosHo-m.m oo nsoHoososH OHboosegns so oHHgoooHoss go oooooo bee .OH mqmae

omm pm pogo CHQ 0mm um «Honor :Hm

 

l7

 

 

om H mmml m.mm ” m.Hn mo.o 5:
HO a. sm- OON " One H.O names me
mmH m OHO- OOH ” O.OO mOO mm
OHH ... NOT OON H lie. HO somzmeo Hm
OHH a mms- Nam " OE mOO mm 98ng
mimosa
a a es- O.mm ” One H.O sHmzmeO OH
m: H wmml m.wm " m.Hm mo.o b
we a NOT gm " THO HO N
am A CT lime “ Oom 5O mafia-B H mmooomommo
mHmmOEHw
Mesa m H a .88 oHEQOOHosz a com oosbonosm

 

 

decompose-THREE: . m one HsfiosoHEosHHO-m . O
CO mCOHposoCH oEumfiEHnm Cc COHpmppCoocoo oHHCQooHoHE mo vocab one .> as

18

.mHOE\Hmon

.omm be oboe/Hon fi 2 H.Oe

 

 

 

NMH H 0mm! N.NH ” w.Hw mm& mm
OHH ... HE- OON ” sag bosom magmas Hm 2%me
mgmmoomo
m: H OHM! o.>m u o.mw mmB ma
m: H mmml m.wm ” >.Hm hmnum 23H¢HA m mmOOO mmo
mhmmovmo
WHO: m n < usosHom soHBoooHosz a 5m Bentham

 

 

.osoasosoo-m-HBOQHHO-s . M Home

.HsssesoHEomfio-m.m .8 838:9: OBOE—Ens so Boots oeosHom .9 Smog.

 

19

 

o
\
o
A
A
O 0.1M LiAlHu
‘6 0.05M LiAlHu
I l l
300 305 ”.0
lOOO/T °K"l
Figure 5. Activation Parameter Plots of the Reaction of 3,14—Di-

methyl-2—pentanone with LiAlHu in ether.

lnA/B

--l.5

 

  

C) 0.3“ CHngBr/Ether
A O . 05M CH3MgBI'/Ether
I3 0.1M CHngBr/THF

2O

 

J l L
3.0 3.5 11.0
lOOO/T °K‘1
Figure 6. Activation Parameter Plots of the reactions of 2,3—

Dimethylbutanal with CH3MgBr in ether and THF.

21

    

0.1M CH3MgI
O 0.05M CH3MgI

A 0.]M CH3MgCl

 

l 1 1
3.0 3.5 13.0

lOOO/T

Figure 7. Activation Parameter Plots of the Reactions of 2,3—
Dimethylbutanal with CH3MgI in Ether and CH3MgCl in
THF.

22

.56 CH0

 

 

 

.OHoesto OOO- oo oo Hooos_nonososmoe one He ooOOHooso nH mwmeeb
.oHoeHsOs
mm. a OH.H- OO a OOH- H.O
OO. H me.H- ON a Oee- mO.O bosom
em. a mm.H- mOH a OOO- H.O age emHaHH mmOOOmOmmO
mflmmovmm
mo. H HO.O HHH a mOm- H.O bosom Hommo
mm. H OO.O mm H HOO- H.O ems HOOZMmO
NH. a OO.O mm a omo- H.O
Om. a Ono HO ... OHm- mOO bosom Hmzmmo
OH. H HH.N we a emm- H.O ems
Hm. h mm.m OO a OOH- mO.O
OH. H sm.m mm a OOH- H.O noose smegmmo omomommo
dimmers.
MOS efimfi .osoo ososHom oHEoooHosz openness

 

 

.HH Ode H npooe sH ooosooog

m2. m2.

mCOHpommh on» How mdeg +m<< 98 +33 .HH> mam-SH.

23

B. Effect of Nucleophile.

Grigpard Reagents.
The reactions of 2,3—dimethylbutanal with methyl Grignard rea—
gents revealed stereoselectivity in the following order (runs 18, 30,

39, Table IV).

CH3MgC1 3 CH3MgBr > CH3MgI
This order is probably due to the combination of two effects.19 First,
the increase of the steric effect in going from methylmagnesium chloride
to methylmagnesium iodide; and, secondly, the corresponding decrease of
the inductive effect of the halogens. Why the effect should be as found,
however, is not understood, as the mechanism of the Grignard addition to
aldehydes and ketones is not corpletely known. Smith,“ in spectros-
copic and kinetic studies of the reaction of methylmagnesium bromide with
2,ll—dimethyl-Ll'-methylmercaptobenzophenone in diethyl ether, obtaired
data below 9a. 0.3M methylmaglesium bromide that were consistent with
complex formation followed by first-order cmversion to product. The

proposed mechanism was:
ketone + Grignard—éK—J; complex __k_. product (A)

At concentrations above _c_a_. 0.3M the mechanism was found to be faster
than predicted on this basis. Billet and sham-25 studied the reaction
of A—methylmercaptoacetophenone in ether with methylmagnesium bromide

at 25° . The pseudo-first-order rate constant for the reaction was found
to increase from 0.3 to 1.3 sec":L with an increase in the Grignard con-

centration fron 0.05 to 0.6M. The following scheme involving dimerization

214

of the methylmagnesium bromide was prOposed for the reaction of ll—

methylmercaptoacetophenone with methylmagnesium bromide.

k
K + Gltxl: Cl—l—QP

K
2G——-2—-—-»IG

l 2

K k
K+ngi202—3—>P

"It is of interest to note that, as with the previous example,”b to the
extent that this scheme represents the actual chemical system, the com-
plex containing the dimeric Grignard yield product pa. ten times faster
than the complex between ketone and monomeric Grignard. Thus, the for—
mation of a more reactive corplex between the ketone and the dimeric
Grignard can account for the observed pseudo-first-order rate constant
being larger than predicted on the basis of the scheme outlined in eq.
A."25

The corposition of Grignard reagents has been extensively studied.
Recently Parris and Ashby20 came out in support of the Schlenk type equi-

librium of the Grignard corpounds.

R2Mg+MgX2<¢ K >2RMgX

 

The equilibrium constant K, determined by nmr temperature studies, was
found to change with R, halogen, and solvent. In diethyl ether bromides
and iodides were monoreric at concentrations between 0.05 and 0.1M. At
concentrations between 0.3 and JM increasing association to diners oc-
curred. In tetrahydrofuran, regardless of the nature of the halogen in-

volved, only monomers were found.

25

To make sure that monoreric Grignards were the attacking species,
we used concentrations in the range of 0.05 to 0.1M (Tables II and IV).
The stereoselectivity of methylmagnesium iodide (run 19 y_s_. 23) and
methylmagnesium bromide (run 31 _v_s_. 35), Table V, remained the same

despite the change in concentration from 0.05 to 0.1M.

Methyllithium. The mechanism of the addition of organolithium reagents
to carbonyl compounds was found to be first-order in each reactant, se-

cond-order overall . 2 5
ketone + RLi Jai't—r; complex —S—lfl+ products

The mechanism is believed to involve the formation of 1:1 coordination
complex between the carbonyl corpound and the organolithium in a fast
and reversible step, folloved by a rate determining step to give product.

The observation of 13C -7Li spin—spin cooling in ether and te-
trahydrofuran, defines the structure of the methyllithium tetramer in
these solutions.28

The additions of methyllithium to 2,3—dimethylbutana1 gave
smaller MCI-1:3 values than those of the Grignards (run 142 YE- 18, 30, 39,
Table IV). Furthermore, as was the case with the Grignards, no change
in stereoselectivity was observed with increase in methyllithium concen—

tration from 0.05 to 0.1M (run A7 y_s_. 143, Table V).

Metal Hydrides. The formation of alkoxyaluminum hydrides in the reduc-
tion of carbonyl compounds, and their interference with the mechanism
of the reduction, has been extensively studied. Eliel29 prOposed that

alkoxyaluminum hydrides were not involved in the reduction of 3,3,5-

26

trimethylcyclohexanone with lithium aluminum hydride, but that, as soon
as they formed, they disproportionated to lithium aluminum hydride and
lithium aluminum tetraalkoxide. Thus, the lithium alwmunwn hydride was
the only effective reducing agent througiout the reduction. Further-
more, it is kno-m that alkoxyalumrinum hydrides are less effective re-
ducing agents than AlHLI'.

Sodium borohydride reductions of aldehydes and ketones in iso—
propyl alcohol exhibit simple second-order kinetics, first-order in
borohydride and first-order in the carbonyl derivative.30

Lithium aluminum hydride reduction of 3,4—dimethy1-2—pentanone
revealed no change in stereoselectivity in the range 0.05 to 0.1M (run
7 1s. 2, Table V), and only a slight decrease in the range 0.1 to 0.7M
(run 2 y_s_. 1). Sodium borohydride was less stereoselective than lithium
aluminum hydride (run 17 _v;s_. 8, Table IV).

C. Solvent Effects .

Solvent can play an important role in the stereoselectivity of
reactions. Solvation of the transition states usually takes place
through hydrogen bonding, ion-1m, ion-dipole , or dipole-dipole inter-
actions. The more polar solvents stabilize the more polar transition
states.”a

Changing the solvent from ether to tetrahydrofuran, in the re-
actions of 3,1-l-dimethyl-2—pentanone with lithium aluminum hydride and
2,3—dimethylbutanal with methylmagnesium bromide (runs 3 _v§_. l3 and 31
_v_s_. 28, Table VI), did not alter the stereoselectivity of the reactions.
Thus, the greater stereoselectivity in tetrahydrofuran than in ether,

observed in the reactions of 3—methoxy-2—butanone,1"a was not observed

27

here. The suggested greater solvation of the hydride species in tetra—
hydrofuran than in ether ,29 and the effect of increased solvent polar—
ity,”a were of minor importance. Since both 2, 3—dimethylbutanal, and 3,
ll-dimethyl—2—pentanone contain no polar groups, besides the carbonyl
group, it is not surprising to find that the population of the two tran—
sition states VIII and D{ remained unaffected with a change in solvent.
Solutions of methylmagnesium chloride in ether, and methylmagnesium io-
dide in tetrahydrofuran were not studied, because the former dispropor-
tionates in ether at room temperature, and the latter is unstable in
tetrahydrofuran.20

D. Entropy Effects.

In the reductions of 3,14-dimethyl-2-pentanone with lithium alum-
inum hydride, both AAHIIIB and AASigB had the same, negative sign (Table VII).
The AAHfB and AASIIB had opposite signs in the reaction of 2,3—dimethylbu—
tanal with methyl Grignards and methyllithium. The methylmagnesium iodide,
metrwlmagnesium chloride and methyllithium additions were enthalpy con-
trolled. The entropy was positive and small, ranging from 0.014 to 0.76
e.u. Only the Reaction of methylmagnesium bromide in ether and tetrahy-
drofuran is an exception, as it is entropy controlled. These findings
are in concert with the restrictions placed on the Karabatsos model,
namely that entropy differences between the two diastereoreric transition
states must be negligibly small. Thus, our findings differ from those
obtained in the additions of methyllithium, lithium aluminum hydride,
and methylmagnesium iodide to 3—phenyl-2—butanone-l,l,l,3—d_u,1"b and
methyl Grignard and metal hydride reductions of 3—methoxy—2—butanone-l,1,

l,3-d_u,1“a which were entropy controlled. They also differ from those

28

obtained in the reductions of 3—methyl-2—pentanone and 3,4,A-trimethyl-
2-pentanone with metal hydrides , in which cases entropy differences

caused an inversion of the diastereoreric product ratio A/B.23

E. Configurational Studies .

In order to prove that the stereochemistry of the 3,4—dimethyl—2-
pentanols is as predicted by the Karabatsos model, we independently syn-
thesized one of the two alcohols. Pure t_hr_eg-3,ll—dimethyl-2-pentanol
was prepared by hydroboration of t_ra_n_s:3,ll—dimetm1—2-pentene,27 according

to the equation:

0 0H
_ /H 1. B2H6 CH3 H H CH3
_ H 0 0H” +
\CH3 2. 2 2,/ CH H
3 H CH3
iPr iPr

S—R R-S
1 MO J

We were unable to detect amt erythro-3JI-dimethyl-2-pentanol, or 2,3-

 

L*’\/5§’

dimethyl—B—pentanol in the reaction product. The pure t_hr_eg-3 ,ll-dimethyl-
2-pentanol matched the major diastereorer-1c alcohol A produced from the
reaction of 3,11-dimethyl-2-pentano'ie with metal rwdrides , and the minor
diastereoneric alcohol B obtained from the methyl Grignard and methyl-
lithium additions, to 2,3—dimethylbutanal. Thus, the stereochemistry was
found to be as predicted by the Karabatsos model.

F. Conclusions.
This study of the additions to 2,3—dimethylbutanal and to 3,4-

dimethyl-2-pentanone showed that the nucleophi 1e concentration and the

29

solvent polarity do not affect the stereoselectivity of the reactions.
The experimental results closely paralleled those predicted by the Kar-
abatsos model. Any discrepancies, from the predictions, were not very
significant. Small discrepancies are to be expected, however, since the
model's predictions are based on the measurable RHO interactions. The
other, gauche, interactions in the transition state, being unmeasurable,
are disregarded.

It would appear that the extension of the model to other chemical

systems, e.g. imrlnes, seems very promising.

A. General

In the asymmetric induction studies, the apparatus consisted
of a three-necked flask equipped with a thermoreter, a magnetic stirrer,
condenser, and a septum cap. The system was closed by placing a bal-
loon at the top of the condenser and was flushed with nitrogen.

All solvents used were distilled from lithium aluminum twdride,
placed in flame dried flasks, closed with septum caps, and stored under
nitrogen. The necessary amounts of solvent and reagents were introduced
into the reaction flask with syringes. The temperatures of the reaction
mixtures were controlled with carbon tetrachloride-dry ice, Egg—amyl
alcohol-dry ice, water-ice, or water baths (for —22°, -l2°, 1°, 22°
respectively). The higher temperatures, 35° in ether and 60° in tetra-
hydrofuran, were controlled by refluxing the solvent. This method in-
troduced a maximum deviation in temperature of i 2° .

Re nts: Solutions of methylmagnesium bromide (3.0M) in ether
and methylmagnesium chloride (3.16M) in tetrahydrofuran (THF) were ob—
tained from Alfa Inorganics Inc. Solutions of methyllithium (2.11M) in
ether, lithium aluminum hydride (2 .llIM) in THF , and lithium aluminum hy-
dride (llJ-IM) in ether were obtained from Foot Mineral Corpany. A sol-
ution of methylmagnesium bromide (3.0M) in THF was prepared from the
ether solution by removing the ether under vacuum and adding anhydrous
THF.

30

31

B. Preparation of 3,lI-Dimethyl-2-pentanone

 

To a cooled suspension (-5 to 0°) of 26g (0.137 mole) of copper
(I) iodide, prepared according to Kauffman and Pinnell,31 in 100ml of
ether, was added 100ml of methyllithium, 2.14M in ether (Alfa) . A solu-
tion of 7.9g (0.080 mole) of 3—methy1—2—penten—2—one (Aldrich) in 50ml
of ether was added over a period of 15 min. to the lithium dimethyl cu-
prate suspension, by using the procedure of House and Fischer, Jr.32
After 15 min. of stirring, the mixture was quenched with an aqueous sol-
ution (pH 8—9) of ammonium chloride—ammonia. The other layer was washed
three times with aqueous ammonium chloride solution and was dried over
anhydrous magnesium sulfate. Upon removal of the solvent and distilla-
tion, a fraction of 5.3g (58% yield) of 3,11-dimethy1-2-pentanone was
collected: ‘bp 13o-l3u0. Literature value:33 bp 128-1330 at 719mm.
Vapor phase chromatograms were taken on an Aerograph A90-P3, by using a
20% Carbowax DMCS, Chrorosorb W, 20 ft. column, He pressure 21 psi, and
column temperature 130°. The ketone was 96% pure, and had a retention
time of 15 min. Nmrr spectra (140mg of ketone in 0.250m1 of carbon tetra-
chloride with 12mg of Eu(FOD)3) were obtained with a Varian A 56/60D
spectrometer. The spectrum consisted of a quartet centered at 6 1.07,
a doublet centered at 6 1.23, a mmltiplet centered at 6 1.80, a multi-
plet centered at 6 2.34 and a singlet at 6 2.311 (ratios, 6.0 : 3.0 : 1.0 :
1.0 : 3.0, respectively).

Mass spectrum showed most abundant peaks at:

mve R.A.3“ mve R.A. m/e R.A.
39 10.5 55 12.u 72 38.8
A1 17.5 57 8.0 99 7.3

143 100.0 71 7.9 1111 6.7

32

C. Preparation of 3,4—Dimethy1—2—pentanone—1,1,l,3-<_iu

 

In a 50ml round-bottomed flask equipped with a magnetic stirrer,
a condenser, and closed at the top with a calcium chloride drying tube,
was placed four grams (0 . 035 mole) of 3 , ll-dimethyl—2—pentanone together
with 15ml of 90% deuterium oxide. The mixture was refluxed at 100° for
one day. The pH of the deuterium oxide was maintained at 10 by using a
few drops of lithium deuteroxide . The same treatment of the ketone was
repeated two more times with 99.5% deuterium oxide, and refluxed for two
days . After extraction with ether, the ether layer was dried over an-
hydrous maglesium sulfate and distilled. The fraction boiling at 131-
13ll° was collected to give 3.2g (80% yield). The nmr spectrum of a 20%
solution of this ketone in benzene (Figure 8—B) , exhibited a quartet
centered at 6 0.73, a singlet at 6 0.87, and a multiplet centered at
6 1.77 (ratios, 6.0 : 3.0 : 1.0, respectively).

D. Additions to 3,Ll-Dimethy1-2-pentanone-l,1,l,3111,

 

1. Addition of Methylmagesium Bromide.

 

In the apparatus described previously (Section A) were placed
20ml of anhydrous ether and 11ml of methylmagnesium bromide (3.0M solution
in ether). While maintaining a temperature of 35° , a solution of 0.614g
(5.142 moles) of 3,4—dimethyl—2-butanone—l,l,l,3-<_iu in 1.8ml of ether was
added by means of a syringe. After stirring for three hours the mixture
was quenched with a 10% ammonium chloride-ammonia solution (pH 9). The
ether layer was separated, washed twice with 15ml of water and dried
over anhydrous magnesium sulfate. The ether was removed by distillation

and the crude 2,3,h-trimethy1—2-pentanol—1,l,1,3-d

4‘ was further purified

33

 

682% 5 .JAJJJucEfiocooumufiEomEouim as
com macroecooumlfiafiaflonim 2: no 8338 RON no 88on .m.s.z .m mamas

01m o.m o.m m 0.0 0A o.m o.m

h -

 

' - u d u d

 

 

 

 

A5 A5

 

314

by vpc (6 ft. x 1/4 inch, column, Chromosorb W). The nmr spectrum of
a 14% solution of the alcohol in pyridine (Figure 9) showed a multiplet
centered at 6 0.97, a singlet of unresolved diastereotopic methyl protons

at 6 1.32, a multiplet centered at 6 2.25, and a singlet at 6 11.85.

2. N .M.R. Solvent Studies of 2,3,ll—T‘rimethyl—2—pentanol—1,1,1,3-<_ill

 

Since the nonequivalence of the diastereotopic methyl protons ,
HA and H8, at 6 1.32, was not observed with the pyridine solution (Fig-
ure 9), the following solvents were used in an attempt to resolve the
diastereotopic protons: Carbon tetrachloride, chloroform, dimethyl
sulfoxide, benzene , chlorobenzene , toluene , phenol, nitrobenzene , forma-
mide , dimethyl for'mamide , acetonitrile , hexamethyl phosporoamide , t-
butyl alcohol, acetone—g6, and carbon tetrachloride/Eu(FOD)3. These
efforts, to find a suitable solvent to resolve the diastereotopic H

A
and HE, remained fruitless.

3 . Additions of Lithium Aluminum Hydride and Sodium Borohydride

 

The procedure described previously (Section D—l) was followed
for additions of lithium aluminum hydride to the carbonyl compounds .
A solution of 0.6ml (2.8L! moles) of lithium aluminum hydride (“AND in
ether was diluted with ether to concentration of 0.1M (runs 2-6) or 0.05M
(runs 7—10). A second solution of 0.2m1 (1.26 moles) of 3,h—dimethy1—
2-pentanone—1,1,1,3nd,4 in 0.5m1 of ether was added slowly, with fast
stirring, to the hydride solution at the appropriate temperature. The
same procedure was used for runs 11-16, with only the solvent being
changed from ether to THF. In run 17, a 0.025M solution of sodium boro—
hydride in isopropyl alcohol was used.

35

.ofioflso 5 Jum.H.HJAqucooumAEomfiéu.s.m.m do 8338 e: .8 528.com .m.z.z .m mama

 

 

 

 

 

 

 

 

 

 

36

14. 'N.M.’R. and 'V.P.C. Studies of3,ll-Dimethyl—2-‘pentanol and
3 , ll—Dimethyl-2—pentanol—1 , 1 , 1 , 3‘94

 

A solution of 3,li-dimethy1—2—pentanol—1,1,l,3—<_ij_l in pyridine
showed the nonequivalent diastereotOpic protons HA and HB’ -CI_1(0H)-CD3,
as two singlets at 6 3.814 and 3.97 (Figure 10). Quantitative measurements
were made by integrating the area under these two singlets. In the case
of 3,1l—dimethy1-2-pentanol the ratio A/B (threo/erythro) was determined
by integrating the area under the two doublets, centered at 6 1.25 and
6 1.29, of the diastereotopic methyl protons -CH(0H)_C_I_{_3 (Figure 11).
Ratios A/B, calculated by vpc, were in good agreement with those
obtained from the nmr measurements . Vapor phase chrometograms were taken
on an Aerograph, by using a Carbowax DMCS, Chromosorb w, 20 ft. column,
He pressure 20 psi and column temperature 130°. The eggthro—3fl-dimethyl—
2-pentanol had a' retention time of 27 min. and the t_hr___e_9_—3,ll-dimethy1-

2-pentanol a retention time of 30 min.

E. Preparation of 2,3—Dimethy1butanal

According to the procedure of Freeman et._a_1_,35 a solution of
142.08g (0.5 mole) of 2,3—dimethy1—1-butene (Chemical Samples Co.) in
600ml of freshly distilled dichloromethane (Matheson-Colleman) was placed
in a three-necked 2-1 flask equipped with an addition funnel, a mechan-
ical stirrer and a drying tube. The temperature of the solution was
kept between 144° by using an ice-water bath. To it was added, drop-
wise through the funnel, a solution of 80.6g (0.52 mole) of freshly dis-
tilled chromyl chloride (Alfa) in Moon of dichloromethane. The mixture

was allowed to stand in the cooling bath, with stirring, for one hour.

37

 

 

.Aoopdh. n B 3.5g u 5
.8838 E film.ddTHoqoecooumAEcmfiouea no :28on .m.z.z .2 names

sm.o H.m :w.m sm.m :.m

_ _ _ _ _

 

 

 

 

 

 

38

w

 

goose? u .H. mounflrmlm n ma
.ofioflso 5 HoqoecoosmuasneenoJJ .Ho 8338 e: .8 =28on .m.2.z .2 are

0.0 04 0.0. 0.m 0.: 0.m 0.0

 

- H

 

 

 

 

 

 

 

 

 

mum

 

39

ooaaoaeocaeoe condo an HostesflsremfiouMJ .Ho 8338 eom no 558% .m.z.z

00

j

 

0.H

 

 

 

 

0A H.m

 

 

mod

. NH magmfim

 

140

Zinc dust, 41.8g (0.614 mole), was added and the mixture was
stirred for an additional 30 min. after which 350ml of ice—water was
added. The mixture was steam distilled, and the dichloromethane layer
was separated from the aqueous one, washed with water, dried over an-
hydrous magnesium sulfate and filtered. The solvent was removed by

distillation through a short Vigreaux column, and the residue was trans—

 

ferred to a smaller flask and redistilled. The distillate was further
purified via; vpc, yielding 8.0g (16%) of pure 2,3—dimethylbutanal: bp
nil-116°. Literature value: 110—112°,35 112—llll°.37 Vapor phase chrom—

atograms were taken on an Aerograph A90-P3, with a 20% Carbowax DM-CS,

 

Chromosorb W, 20 ft. column, He pressure 21 psi and column temperature
130°. The retention time was 10' 30". Nmr of 30% solution in carbon

tetrachloride (Figure 12) consisted of a multiplet centered at 6 1.0,

a multiplet centered at 6 1.8, a multiplet centered at 6 2.1, and a

doublet centered at 6 9.66 (ratios, 9.0 : 1.0 : 1.0 : 1.0, respectively).

F. Additions of Methyl Grignards to 2,3—Dimethy1butanal

A typical addition of methyl Grignards to 2,3—dimethylbutanal
was as follows: In an apparatus, described in Section A, was placed
29ml of ether together with lml (3 moles) of metrwhnagrlesimn bromide
(3.0M in ether). A solution of 0.2ml (0.16g, 1.6 mmoles) of 2,3—di-
methylbutanal in 0.5le of ether was injected. The temperature of the
solution was maintained at 35° with an oil bath. After stirring for
two hours, the reaction mixture was quenched with a 10% solution of
ammonium chloride-ammonia solution (pH 9) . The ether layer was separ—

ated, dried over anhydrous maglesium sulfate and removed by distillation.

 

141

The ratio A/B‘(erythro/threo) of the diastereomeric 3,11-dimethy1—

 

2—pentanols was determined by vpc, as described in Section D—A.

G. Preparation of Threo—3,h-Dimethyl-2—pentanol

 

In a 50ml two—necked round—bottomed flask equipped with a mag-
netic stirrer, a septum cap, a condenser closed at the top with a bal—
loon and under nitrogen atmosphere, was placed a mixture of 0.348 (9.0
mmoles) sodium borohydride in 15ml T'HF. m—3,A-dirrethyl-2—pentene,
2.9llg (0.030 mole), was introduced into the reaction flask, through the
septum cap, with a syringe. The mixture was stirred and 1.51ml (1.7g,
0.012 mole) of freshly distilled boron trifluoride-ether solution in
2ml of T'HF was added drop-wise over a period of one hour. The tempera-
ture was maintained at 25°. After the addition was over, the mixture
was kept at 25° for an additional hour. The excess diborane was decan-
posed with water followed by oxidation with 3.311 of 3N sodium hydroxide
and drop—wise addition of 3.3n1 of 30% solution of hydrogen peroxide.

The resulting mixture was saturated with sodium chloride and the T'HF
layer was separated, dried over anhydrous magnesium sulfate and dis-
tilled. The yield of _Ti_lre_:o_—3,ll-dimethyl-2-pentanol was 95% by vpc. The
alcohol had a Vpc retention time of 30 min. under the conditions: 20%
Carbowax DMCS, Chromosorb W, 20 ft x 1/14 inch column, He pressure 20 psi,
and column temperature 130° . The erflhro alcohol, under identical con—
ditions had a retention time of 27 min.

Nmr spectrum of a solution of the pure three alcohol in pyridine
showed the doublet of the diastereotopic methyl protons , -CH(OH)-§H_3,
centered at 6 1.25. The doublet of the emhro diastereomer, centered

at 6 1.20 (Figure 11), was missing.

 

 

 

BIBLIOGRAPHY

1.
2.

3.

10.

11.

12.

13.

1M.

\OODNQU'I

BIBLIOGRAPHY

E. Fischer, Her}, 21, 3231 (1899).
w. Marckwald, ibid., 3X: 1368 (19014).

J. D. Morrison and H. S. Mosher, "Asymmetric Organic Reactions",
Prentice—Hall, Inc., N. J. (1971).

 

Y. Izumi, Angew. Chem. Int. Ed., 10, 871 (1971).

A. McKenzie, J. Chem. Soc., 85, 12U9 (190M).

 

v. Prelog, Helv. Chim. Acta, 38, 308 (1953).

 

 

D. J. Cram and F. A. Abd. Elhafez, J. Am. Chem. Soc., «7,13,: 5828 (1952).
D. J. Cram and K. R. Kopecky, ibid., 81, 27118 (1959).

J. W. Cornforth, R. H. Cornforth, K. K. Mathew, J. Chem. Soc., 112
(1959).

G. J. Karabatsos, J. Am. Chem. Soc., 89, 1367 (1967).

 

 

(a) G. J. Karabatsos and N. Hsi, ibid., 81, 2864 (1965);
(b) G. J. Karabatsos, D. J. Fenoglio, and S. S. Lande, ibid., 91,
3572 (1969).

(a) D. Y. Curtin, Record Chem. Prog., 15, 111 (195“);
(b) E. L. Eliel, "Stereochemistry of Organic Compounds", McGraw—Hill
Co., New York, N. Y., 1962, Chapter 6.

 

(a) E. Ruch and I. Ugi, Theor. Chim. Acta (Berlin), 3, 287 (1966);
(b) E. RuCh and I. Ugi, in “Topics in Stereochemistry" Vol. A,
E. L. Eliel and N. L. Allinger, Eds., Interscience, N. Y.,

 

1969. p. 99.

(a) A. Yeramyan, Ph.D. Thesis, Michigan State University, East
Lansing, 1972;

(b) S. Mylonakis, Ph.D. Thesis, Michigan State University, East
Lansing, 1971.

U2

 

 

 

15.

16.

17.

18.

19.
20.

21.

22.

23.

211.

25.
26.
27.

28.

29.
30.

(a)

(b)

(a)
(b)

(a)

(b)

J.
D.
G.
T.

L
O
E.
H

“3

R. S. Cahn, C. K. Ingold, and V. Prelog, Angew. Chem. Int.
.§§:s 5, 385 (1966);

R. S. Cahn, C. K. Ingold, and V. Prelog, Experientia, 1’21,
81 (1956).

 

I. Ugi, z. NaturforSCh, 20%, MOS (1965);
I. Ugi, Chimia, 19, 91 (1965).

V. Prelog, O. Ceder, and M. Wilhelm, Helv. Chim. Acta, 38
303, (1955);

V. Prelog, E. Philbin, E. Watanabe, and M. Wilhelm, ib_i_d_.,
’32, 1086 (1956).

 

 

 

. Dye and'V. A. Nicely, J. Chem. Ed., 38’ AA3 (1971).

. Cowan and H. Mosher, J. Org. Chem., 21, l (1962).

 

Farris and E. C. Ashby, J. Am. Chem. Soc., 93, 1296 (1971).

 

Althuis, Ph.D. Thesis, Michigan State University, East

Lansing, 1968 .

 

D. J. Cram and F. D. Greene, J. Am. Chem. Soc., L5, 6005 (1953).

R. N. Nipe, Ph.D. Thesis, Michigan State University, East Lansing,

1970

(a) s. G. Smith and G. Su, J. Am. Chem. Soc., 86, 2750 (196A);
(b) ibid., §§. 3995 (1966).

J. Billet and s. G. Smdth, J. Am. Chem. Soc., 90, A108 (1968).

 

 

C. G. Swain and L. Kent, ibid., 72, 518 (1950).

G. Zweifel and H. C. Brown, "Organic Reactions" Vol. 13, Chap. 1,
A. C. Cope, Ed., John Willey 8. Sons, N. Y., 1963.

L. D. McKeever, R. Waack, M. A. Doran, and E. B. Baker, J. Am.
Chem. Soc., 91, 1057 (1969).

H. Haubenstock and E. L. Eliel, ibid., 83, 2363, (1962).

H. C. Brown and K. Ichikawa, ibid., 83, 373 (1962), and references
cited therein.

 

31.
32.
33.
314.
35.

36.
37.

1414

 

. B. Kauffman and R. P. Pinnell, Inorganic Smithesis, 6, 3 (1960).

. 0. House and W. F. Fischer, Jr., J. Org. Chem., 33, 9149 (1968).

 

G
H
R. Willstatter and D. Hatt, Ann_., 51/8,, 1148 (1919).
R. A. : Abundance relative to base peak as a 100.

(a) F. Freeman, P. J. Cameron and R. H. DuBois, J. Org. Chem.,
3Q, 3970, (1968);

(b) F. Freeman, R. H. DuBois and T. G. McLaughlin, Org. 833.,
2%, ’4 (1971).

J. C. Sheehan and M. G. Howell, J. Org. Chem., 28, 2279 (1963).

 

 

R. A. Barnes and W. M. Budde, J. Am. Chem. Soc., 618’, 2339 (19146).

 

VlHUIH

9

U

711

1293 03046 3

H
I'll
T “l
A ill
G I”

3