A Study of Some Factors that Influence
Asymmetric Induction in Additions to 3-Pheny1-2-pentanone-1,1,1,3-g4
and 2-Pheny1butana1-2-g:l

Thesis fOr the Degree of M.S.
Michigan State University
Luis Valles
1970

ABSTRACT
A Study of Some Factors that Influence Asymmetric
Induction in Additions to 3-Phenyl-2-pentanone-l,l,l,3-g4
and 2-Phenylbutal'ialoz-g1

by Luis Valles

The difference in the free energy between the diastereomeric
transition states A and B can be qualitatively predicted from the

relative magnitude of M ++ 0 and L ++.o interactionsl. For a system

M (g .
L s
R 5 M

A R B

where L - phenyl, M s ethyl, and s 8 hydrogen (deuterium), the
predicted oasis value is about 500 cal/mole in favor of A. To test

the validity of the model, diastereomeric product ratios were measured
by NMR from additions to 3-phenyl-2-pentanone-l,l,l,3-_d4 and 2-phenyl-
butanal-Z-d . The results obtained were as follows: Additions to the
ketone of methyl lithium, methylmagnesium bromide, and methylmagnesium
chloride gave AAGtB values slightly higher than those predicted by the
model, whereas addition of methylmagnesium iodide gave values that were
slightly lower. Reduction of the aldehyde with either lithium aluminum
deuteride or sodium borodeuteride gives AAGXB values below 500 cal/mole.
Solvent and temperature changes affected these values to some extent,

but not sufficiently to invalidate the model.

The lack of good agreement between the calculated and the
experimental oasis values means that factors other than M ++ 0 and
L ++IO interactions, such as solvent, entropy, and other non-bonded

interactions contribute to the experimental AAGtB values.

 

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

A STUDY OF SOME FACTORS THAT INFLUENCE ASYMMETRIC
INDUCTION IN ADDITIONS TO 3-PHENYL-2-PENTANONE-l,l,l,3-g4
AND 2-PHENYLBUTANAL-2-gq

by

Luis Valles

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
Department of Chemistry
4 i970

CJpZZQZ
(9.3510

A mis padres y hermanos con todo carifio

ACKNOWLEDGEMENT

The author wishes to express his sincere appreciation to
Professor Gerasimos J. Karabatsos for his understanding, encouragement,
and help during the course of this investigation.

He thanks the Ford Foundation for its generous financial support.

Introduction ..........................
Results and Discussion .....................

l.

2
3.
4
5

Experimental. . . . . . ..... . .............

l.

m N 05 01 «h on N
O O O C

10.

ll.

TABLE OF CONTENTS

Effect of nucleophile. Grignard reagents ......
. Hydrides ........................
Solvent .........................
. Temperature .......................
. Nature of the carbonyl compound .............

Origin of reactants and solvents ............
Nuclear Magnetic Resonance Spectra ...........
. Infrared Spectra . . ....... . ..........
Preparation of 3-phenylpentanone ............
Preparation of 2-methyl-3-phenyl-2-pentanol .......
Deuteration of 3-phenylpentanone ............
. Purification of 3-phenyl-2-pentanone-l,l,l,3-_c_i4 .....

. Reaction of methylmagnesium iodide with
3-phenyl-2-pentanone-l,l,l,3-g4. . . ..........

Reaction of methylmagnesium bromide with

3-phenyl-2-pentanone-l,l,l,3-g4 in ether ........
Reaction of methylmagnesium chloride with 3-phenyl. -2-

pentanone-l, l ,l ,3-94 in tetrahydrofuran ..... .
Reaction of methyl lithium with 3-phenyl-2-pentanone-

l,‘,1,3'g400000000000 oooooooooooo

Page
1
8
8

19
2l
22
24
25
25
25
26
26
27
28
29

31

32

TABLE OF CONTENTS (Continued)
Page
ll. NMR study of 2-methyl-3-phenyl-2-pentanol-l,l,l-3-g4. . . 33
l2. Synthesis of 2-phenylbutanal ............... 35
l3. Deuteration of 2-phenylbutanal .............. 37

14. Reaction of lithium aluminum deuteride with
Z-phenylbutanal-Z-QJ in ether . . . . .......... 38

15. Reaction of Sodium Borodeuteride with 2-phenylbutanal-
z-g, in tetrahydrofuran ........... . ..... 40

l6. Reaction of sodium borodeuteride with 2-phenylbutanal-
z-g, in isopropyl alcohol . . . . . . . . . . . . . . . . 40

I7. NMR study of 2-phenyl-l-butanol-l,2-d (Figure 6) . . . . 42

l8. Purification of 2- phenyl -l -butanol -l ,2 -92
by chromatography . . . . . . . . . . . ........ 42

Bibliography. ......................... 44

ii

LIST OF TABLES

Table
I. Additions to 3-phenyl-2-pentanone-l,l,l,3-94. . . . . .

II. Reduction of 2-phenylbutanal-2-gh . . . . . ......
III. Some additions to and reductions of carbonyl
compounds obtained by other investigators . . .....

IV. Experimental results for additions to 3-phenyl-
2-pentanone and 3-phenyl-2-pentanone-l,l,l,3--_d4 . . . .

V. Experimental results for reduction of
2-phenylbutanal-2-gq. . . . . . . ...........

iii

Page

10

ll

30

Figure

II.

III.

IV.

VI.

LIST OF FIGURES

Page

Ground states of two dissymmetric enantiomers and
their diastereomeric transition states ........ 2

Six characteristic conformations of a carbonyl
compound whose a-carbon atom bears three groups
of different size. . . . . . ...... . . . . . . . 4

Variation of A/B ratio of 2-methyl-3-phenyl-2-pentanol-
l,l,l,3-d in about 20% of alcohol in pyridine, from
reaction-3f 3-phenyl- -2- -pentanone- -l ,l ,l ,3 -g4 at 2° with

a) MeMg I in ether (A/B . 6l/39); b) MeMgBr in ether

(A/B - g79/2l); c) MeMgCl in THF (A/B - 86/l4); d)

MeLi in ether (A/B - 89/ll). . . . ....... . . . 17

Variation of A/B ratio for the neat 2- ~phenyl -l- butanol-

l ,2 d from reduction of 2-phenylbutanal-2- -g, with

NaBD in isopropyl alcohol; a) at 82° (A/B . 55/45);

b) aI 25°( (A/B - 60/40); c) at -63° (A/B . 68/32). . . 23

NMR Spectrum of neat 2-methyl- ~3-phenyl -2- -pentanol- -l ,l,3-
d from addition of MeLi to 3-phenyl- -2- -pentanone-
T41 ,l A3 -_4 in ether at 2°. . . .......... 34

NMR spectrum of neat 2- -phenyl- -l -butanol- l A2 -_;m from
n

reduction of L ~phenylbutanal L ~94. with NaBD4
at 2°. . . . . . . . . . . . . . ....... 41

iv

INTRODUCTION

Asymmetric centers arise mostly as the result of a conversion of
trigonal carbon atoms into tetrahedral ones at the site of unsaturated
functions such as carbonyls, enols, enamines, and at isolated or
conjugated olefinic double bondsl. Thus, the addition of methyl-
magnesium iodide to propionaldehyde produces a dissymmetric molecule.
There is no reason for the Grignard reagent to have anything but an
equal probability of approaching from either side of the propion-
aldehyde molecule, thus producing equal amounts of (+)- and
(-)-butanol-2. The transition states leading to (+)- and (-)-
butanol-2 are said to be enantiomeric and since propionaldehyde is
a single non-dissymmetric molecule the free energy difference between
the transition states is zero. It follows that the two reactions
occur at exactly the same rate and the products are formed in exactly
equal amountsz.

When a dissymmetric grouping is already present in the molecule
and a second grouping (e.g. an asymmetric carbon atom) is created,
an exactly lzl mixture of the two possible stereoisomers (which are
now diastereomers) is not expected. This follows from the fact that
diastereomerically related transition states, like diastereomerically
related ground states, differ in free energy and, therefore, in

stability. This difference shows in the rates of formation and in

the rates of reaction of stereoisomers. A reaction in which new
dissymmetric groupings are produced in unequal amounts is called
assymmetric induction or assymmetric synthesisz.

Assymmetric syntheses are particularly effective in biological
systems. The reactions take place on enzyme surfaces, and generally
the stereoselectivity is close to lDOZ. By contrast, most non-
biological processes fall well below this mark. Dne kind of
assymmetric synthesis is that in which the starting material is
dissymmetric and two enantiomers exist, each capable of leading

to two diastereomeric transition states (Figure 1). Since they are

l ---ss

 
 

Figure l

Ground states are indicated by black lines. Transition states by

dashed lines, and mirror images relationship by vertical lines.
enantiomers, they have the same ground state energy; but each enantiomer
reacts at the same rate gig_two diastereomeric transition states; in
other words, each isomer gives rise to two diastereomeric transition
states, the populations of these two transition states being exactly

the same starting from either A or B3.

Mislow's analysis is of further importance in that it identifies
the concept of asymmetric synthesis within the framework of the general
phenomenon of stereoselectivityz. Of these reactions, additions to
carbonyls are the most widely studied. A carbonyl group is converted
to an asymmetric center by (a) reduction with an hydride or (b)
treatment with an organo-metallic reagent“.

The direction of the asymmetric synthesis appears to be governed
by differences in the free energies of the diastereomeric transition
states which arise from differences in non-bonded interactions between
the groups attached to the asymmetric carbon atom and the carbonyl
groupz.

The importance of the initial confbrmation of the molecules is
illustrated by the semi-empirical rule of Cram and Elhafez, according
to which the preponderant isomer is the result of attack from the
least hindered side (I and II) when the carbonyl group is already
situated between the two smallest substituents5, provided that the

 

I II

Open-chain model: relation between direction of
attack and size of substituents (s: smallest; M:
medium sized, and L: largest substituentL

I - Transition state of lowest energy.

II - Predominant diastereomer.

reaction is non-catalytic and that the products are formed in a
kinetically controlled process and not in a subsequent equilibration.
However, it must be noted, with respect to the six characteristic
conformations of a carbonyl compound whose a-carbon atom bears
three groups of different size (Figure 2), that the attack on
conformation b and c from the least hindered side would yield the

same isomer as that resulting from attack on a, whereas the other

S?

A

a § §I% d

@sz
b (\® C) e
0\ 3% §

c ‘I C2 f

0’\

Figure 2

epimer would result from an attack, under the same conditions, on

a conformation such as d, e, or f1.

A second model was also considered for systems in which the
asymmetric center in the starting material carries a group such as
0H which is capable of complexing with organo-metallic reagents
(III). It involves a relatively rigid, five-membered ring which

fixes the conformation of the reacting speciesS.

x\ ’R'

Z
-0» \0 0 ‘92
\ 2 —-> \ _R.
,I C..— C L uni C -—-C
L I \R 1 \
5 R
III 5 IV
Transition state Predominant diastereomer
of lowest energy Rigid Model

A third model, dipolar model (V)’has been used to account for the
stereochemistry of additions of Grignard reagents and lithium alkyls

to a-chloro-carbonyl compoundsZ9.
Cl

\ ;'
2“?_“\>:\

Karabatsos has developed the Cram model to a point where semi-
quantitative predictions of product stereospecificity are possible.
The model is based on the fbllowing assumptions: (a) the transition
states resemble the reactants, i.e., little bond breaking and making
has occurred in the transition state; (b) the two transition states
of lowest energy have the smallest group of the asymmetric carbon

atom closest to the incoming bulky group; (c) the diastereoisomeric

product ratio A/B reflects the relative magnitude of the carbonyl-

eclipsed group interaction M ++ O in (VI) and L ++ O in (VII)7.

   

VII

The conformations of the preferred transition states are similar
to those of the correSponding aldehydes with either M or L groups
eclipsing the carbonyl oxygen.

The ratio A/B can be qualitatively predicted from the relative
stabilities of VI and VII. When the groups 5, M, and L are alkyls
or aryls, the evaluation must be primarily based on the relative
importance of the interactions:

R' ++ M, R ++ L, and M ++ 0, in VI and
R'HL, RHM, and L+—>0, inVII.

The first two interactions in both transition states are unpredictable
and variant; only the third interaction, M ++ D vs. L ++ 0, is
available. For example, when M is ethyl and L is phenyl, M ++ 0

is favored over L ++-0 by about 500 cal/mole. On this basis, the
greater stability of VI over VII would lead to A/B ratios greater

than unity. Karabatsos has used these latter interactions to predict

the experimental AAGia values, where:

* = * - i = - 9
AAGAB AGA AGB RTlnA/B

He has pointed out7 the fbllowing limitations in predicting such
values:

The stereoselectivity of reactions of this type would be susceptible
to solvent and perhaps reagent. It was estimated by Karabatsos7
that solvation of the two diastereomeric transition states might
contribute as much as 50-100 cal/mole to the AAGxB values. The
unknown [(R ta: L) + (R' ++: M)] - [(R ++' M) + (R' ++~ L)] value
will depend on R and R'. This value would tend to make AAGAB more
negative as R' increases in size, although this value will probably
be small. The variation of bond breaking and making at the transition
states will also contribute to fluctuations in AAGfiB, as will
differences in the entropy of the two diastereomeric transition

states. Such entropy contributions to AAGXB will limit the usefulness

of the model, as the model applies to cases where AAHT controls AAGAB7°
The present work was undertaken to study some of the above

mentioned factors in systems represented by
sML-C - CO-R

where s, M, and L are hydrogen (deuterium), ethyl, and phenyl
respectively, and R is hydrogen, or methyl (methyl-93). The reactions
chosen involve the reduction of the aldehyde with lithium aluminum
deuteride and sodium borodeuteride in ether, tetrahydrofuran, or
2-propanol, and the addition to the ketone of methylmagnesium
halides and methyl lithium in ether or tetrahydrofuran.

RESULTS AND DISCUSSION

The experimental results from additions to 3-phenyl-2-pentanone-l,
l,l,3-g4 are summarized in Table I; and those from reductions of 2-
pehnylbutanal--2-g_1 in Table II. Table III summarizes some results,
mainly those in which the open-chain model applies, obtained by other
investigators.

The error fOr the A/B ratios of 2-methyl-3-phenyl-2-pentanol-l,
l,l,3-g4 is estimated at 2%, and for 2-phenylbutanol-l,2-gz, at 3%.

The reported errors fOr AAGtB values were calculated on this basis.
Factors other than M ++ D and L ++ 0 interactions, such as other non-
bonded interactions, the entropy, the solvent, are surely playing an
important role on the total AAGtB. Nevertheless, the Karabatsos model
can explain the results quantitatively, even in those cases where they
do not agree too well with those results predicted by the model. In
the ensuing discussion some trends in the diastereomeric product ratios

wfill be pointed out and some explanations of these trends will be

offered.

0 ‘0. ' 0,.l -°'.).

Generally the results from Grignard addition to a carbonyl
compound varies with the nature of the halogen (entries: 2-9; 43-45;
29-54; 52-55-59; 56-61; 79-82; Bl-84; 89-91-93; and 90-92-94), where

the order of decreasing stereoselectivity is: R'MgCl > R'MgBr >

.mmwcc com opoE\Pou com-

mo mapn> m muuwcmea Pecos momumnmemx mg» .cmmogna; new .ngpm .chmga meomn scum congeu uwgumsszmm men :mnzu
.oooxom< :N m— xgucm new .sz cop m up .op zuucm com ugmuxm .Pmcos oou< cm cw :mxmp mew: mguuonm one

.Apogoupa Lo umm ou mpv «cwvmgxa we: now: acm>Pom ugh .LmNNEocuomam «22 mg» gap: «was mu: m
xgpcm so; cowumgmmucu .memsmcmpn a saw: mums ma: mxmwn owcmsomemummwu asp ecu mo omen mg» mo comumgmmucum

a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN . mNN- N.¢N o.NN N . NPNNN NM- = = , . NF
¢m « mmml _..¢N NrON N « ¢_.\©® m = .. . .. NP
em . ONON- «.mN «.mN N . NNNNN we LIN , .umzmxu = F_
Nm « Npml NNF N a quw mm! .. .. = 0—.
mm « NNNI D.MN m.mm N « _.N\mh N = = = G
No . NON- N.¢N N.NN N . «NNNN mm . LNmzmzu = N
omN . oeNP- m.eN . m.eN N . m \mm Nm- = g . N
NF. . oep_- P.¢N _.oN N . .PNNN N = = = o
om . cam- N.¢N F.0N N . N_\PN mm . LNNIQ . m
Ne . NeN- N.mN m.NN N . NNNNG ON- . = g e
oe . NNN- ¢.NN m.mN N . NNNNN N_. g = g N
me . eeN- N.NN m.mN N . NNNPN N . g ; g N
mm . Nee- m.mN m.NN N . NNNNN mN Laepu Nazmzu mauouamzeuvoumeu _
«NoeNNau N <

m< u N=a>_om

Nae mze LOL “No NN< . . apweaoapuaz «autumnam Ngueu
u e p.323 $3.55 a 9:8. .5538.
T

e-N.F..._-aeocaucma-N-Nzeo;N-N 0» accepeuu< ._ NNN<N

10

e
_ Lo; wpoe\~mo com- Lo mzpm> m meowumga Faves momumnocmg mg» .cmmoecx; use vague .chmca sup: swamAN m Log

2

.N22 cap as nonwmuao mguuoam
New: on» no emumswcmpa a gap: mums mm; mxmma UPLNEomLmummwn oz“ we» mo amen on» No copumgmmpcp «new

 

 

 

 

 

 

 

 

 

eN.. NSF- . meNNm NN = g g NN
mm . NPN- ¢.¢mm e._mm . NNNNe me- = = . ON
eN . oeN- N.Nmm m.NeN . oe\oe mN Foeaaoea-N . g N.
om . mmp- N.omN N.N¢N . NNNNN N = = . NP
_N . NN_- o._mN N.N¢N . ¢¢\em me LIN Name: = N.
mm . mNN- N.Nmm N.omm . NN\NN NN- . = = e.
we . NeN- . NNNFN ,N .. = .. mp
NN . Nap- . Ne\Nm mm LaguN, «NNNPA ozufimzeuvaumzNu ¢_
“mmuMwuu .uwmmmsmwumflm;o mm\< .aflMh wmwnwwwm upwgqoopuaz mumgumnsm xuucu
.m-N-_~=oN=o~Nca;N-N Lo coeouzemm .NN NANNN

.wFOE\Pmu 0001 u me<<

com mapm> O Ouuwcmca pmvos Oomamnmgmg as» .masogm cmmogux; use .ngpme .Facmcn saw; Seaman O Noun

 

 

 

 

 

 

 

 

 

 

ll

 

 

 

 

 

 

 

 

 

 

 

NO OON- ONNON OO = LOOOOE =- NO-
NO OOO- ONNNO OO- -: = = OO
-NP OOO- ON\ON, ON tr- NOzaz, OOOOONOIOOOOOOIO Om-
mN OON- OFNOO ON- -F .Lr -Lr Om-
NN OOO- -O.NN\O»NN O tr = -Lr NO
O OOO- O.ON\OOON ON t-- t- tr, Om
ON OOO- O.NN\O.NO Om .p OIP<OO OOOOOOOIOOOIONIO ON
ON, OONO- O.FO\O.OO ON .ipw OpOzNOOO e, mm
O ONO.v ONNOO. ON -e- LOOOOOOO IF NO-
ON -OOON OONON O = NO 29% L- Ow
O OFOO OONOO ON LOOOO NOzoz OIOAOIOOOIOOOO Om»
OF OOO- OO\ON -OO- -= = : NN
b ONO- 58 ON :95. r L- Li ON
h _- bm _.1 WNNFM c #1 (L.- F mm
NF OOP- OONOm- Om. NOON OOOOz -LP ON
NO ONO- -OONOO NO- -F -: en Ow
-N_ mev-OO -NO\NO, ON LOOOO OOOOOO OOOOOOOOOOOOIO NN
a _
O NOO>FOO
oucmgmmmm omMOOO m\< .asmh coONUOmm op_;aoopu=z «pneumazm zgucu

 

mvcaogeou Pheongmu mo meowuuavmm can on mcovumuv< meow .HHH m4m<h

12

NP augmeomm; Eoguo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DDn Fl M \Nm DWI mewm = = F0
82- O \OO W NE L. F 8
DOS. O \NO ON .293 __ tr NO
ONE- OZOO OO O: NOOzNI-OO LT NO
El m \Pm le .. F! LEN Om
ONON» ONNOO mm = LOOzszO Lr- Om
DFNI DP\g mml LW = LT mm
OON- DNNON m .. b LT Nm
OON- .mNNNN OO = NO: OOO ONOOOOOOIOOOOONIO NO-
ON -ONO- -ONNON ON -Lr OONONO ONOOOONOOOvapmmplui-Nuw.
OOOT O NO mm- ..l F F OO
ONO- NNNON Om LOONO -e -:Lr, my
ONN- -ONNON Om OOONOON ONO: = NO
OOOT O (NO m NE F T OO
bNDPI b P5?” mm (.0me .. if M?
ONON- NNNON OO NE 893: IF) EN
1 0mm! 0 E Pm mm L.’ =[ INF NV
ONO- NNNNO O. LOOOO NOOzOz OOOOONOOOONOOOIO NO
ONOENNOO O. NOONNOO
«mucogmmmm mmc<< m\< .nso» cowaummm oppgaompuzz mumgumnam xgucm
N.O.NOOOV NON OOO<N

l3

Ouuwumga vaoe OoOanagox Ogu .cmmoev»;

.mpos\—mu com- u

mwu<<

use chmzn .NOOoNOoOO OOONOOO Osoucmu ONNNOEEOOO NOON

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.NN OON- NONOO N -: LO = ON
b 9:- OONO-O ON SOON LEO: OOOOONOOOOOOOLN-O ON
ON, om + OONNO Om OOOOOOO- NOON-O = NN
NLOOZO
ON OON- O.OONO.OO ON = + LOOzea-O -Lr. ON
ON OOO- O.OO\O OO -ON LOOOO LOOZNN-N OzONOzOOOIOLN-N ON-
-NN OOO- ONNOO OO- F-- O-- tr ON,
NO OOO- NONOO- ON Iota-O e If ON
-NN, OON- OONNO OO LIN, OOOOz -: ON
NO, ONO- ONNNN m. tr FY tr OO
-NN ONO- OONON ON, -LO OONONO OOONOOOOOOONN-N .mO
N SON- NONOOO ON- L. Li L) NON
ON OOON- N.NNO ON -Lr- = OO-O-OONOOOOOIOOOO Om
ON OOO- NNNOO ON F- -LO LN-O-OOOOIOOOONOOO OO
O OOO-v ONNOOA OO -F tr OIOOOONOIOOOIOMNN NO
-O ONO- OONNO ON LOOOO OIN<OO NOONIOOONOOOOOOONNO NO
m~o5\Pmu uo pcm>~om
oucmgmmum Omwwcq m\< .asmh copuuumm mpvcaompuaz muugumnam zuucu

 

~.u.ucouv Hum m4m<h

l4

.ONOsNNOO OON- - N

Ko<<

ONOOOONO Pecos OoONOONNOx mg» .comogu»; Ocm .Oxgums .Nzxwzopuau OOONNOO Ogmucmu OONOOEEOON NOON

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN- ONN+ O.OO\O.ON ON NOON-N NNONN-NOOO .. NON
NN OON- NONNO ON NONON ONNOON -p OON
.NN, ONN- NNNNO‘ ON, ONO-NOON ONNOz NNOOONNNNOOONONNO OO
.NN ONO- OONOO O NOOON . = NO
NN ONN- OONOO O OOOOOON‘ NOON = NO
NN ONN- -ON\OO ON LO NONON ONO-OWWNMOONONNO OO
NON OONN- ONNOO ON LOT. ONNOON LN-OOOOONOOONOLN-O OO
NN OOO- NNNOO Nw -Fw tr -Lr OO
-NN ONN- -ONNOO -ON L. NOONNNOO tr- NO
NO ONO- NNNNO N F e F NO
NO ONN- ONNNO ON LO NNONNNOO -Lr NO
NN OOO- ONNON‘ N O- -Lr --O OO
LOO OON- ONNOO ON = NOzNNOO ONOOOONONOOONOLN-N ON
-N_ OOON- O \NO NO- LO, Lr- tr . NN
-NN OOON- NONNN N NOOON -F .. \NN
NN- ONO- NNNNO ONw OOOOOON ONO: -e, ON
Nr OOO- NNxNN N NE _. LT OP
-NN ONN- ONNON ON NOOON .N% F ON
N— OOOO NNNNN OO NON .NO NON -e- -NN
NO Os- NNNNO N 2.23 NONE: NOOEONOOOOOLO-O NO
ONOONNOO O. OOOONOO
mesmeummm ONKOOO N\< .OEON cowuumom OPNOOOOFOOz museumnam Ogucm

 

A.v.ucouv Hum u4m<h

15

.m

NOENNNO OON- - NM

w<< 50%

ma—O> cm ONOOOONO Owcos OoOuNOONNx asp .Oaaogm :mmoeu»: ONO .FNOOOE .chum ONO; Ogmucmu uNNumEEOON NOON

 

 

 

 

 

 

 

 

 

 

 

 

 

NN OOO- ONNON ON = ONNOOO ONNOOONONOOONOONNO N.—
NN OOO- ONNON ON = LNONONNO ONONONOOONOONNO NNN
O OOO- ONNON ON = ONNNOO NNOOONONOOVNOONNO NNN
O OOO- NNNNO ON . NONNNO ONONONOOONOONNO ON.
ON OON- NNNNO ON . . ON-OOONNNNOOONONNO OON
ON OON- O.O_\O.ON ON = : LN-NOONNNNOOONONNO NO.
ON NNO- NNNNO ON = = ONNOOONNNNOOONONNO NO.
ON NNN- O.NNNO.NO ON = = NNOOONNNNOOONONNO OON
NN ON - OONNO ON- = . .. . OO—
NN NN + O._ONO.NO O = = g OO.
NN ON + O.NONO.NO ON . ONNNON NNOOONNNOONOONNO NON
NN ONN- OONOO O NONON NNOzOz ONO-ANNOONOONNO NON

mucmcmwmm ammmflwou N\< .ammp wuwwwmwm upwgaoopuaz mumgumasm Opus“

N N

 

A.u.ucouv HHH w4m<h

l6

R'MgI (Figure 3). This order might be accounted for by two fundamental
factors: the greater inductive effect in going from iodide to bromide
to chloride; and De increased steric effect in going from chloride to
bromide to iodidelo.

Since an increase in the electronegativity of 211 decreases the

Z
N ” ,/ z
*1‘N‘TJL\N
a' H I H
R I
VIII H IX

ratio IX/VIII, we can apply this same consideration in order to
explain the above results: an increasing positive charge is developed
on the magnesium atom as the electronegativity of the halogen on the
carbonyl complex (X) increases from iodide to bromide to chloride.
Then, as magnesium becomes more positive, the stereoselectivity of the

reaction increases,and AAGAB becomes more negative. However, detectable

6+
;--------g-----6-

X

differences do not always occur (entries: 89-91-93; and 90-92-94).
The dependence of stereoselectivity on the halogen might also

be the result of differences in the structure of the Grignard reagent.

Since Grignard reagent composition is no longer represented by the

unsymmetrical dimer R'zMg'MgXZ, the mechanism of Grignard addition

to ketonesl°, represented by attack of R'zMg-ng2 on the carbonyl

group, has been seriously questionedlz. A new mechanism seems to be

more probable:

l7

.N ONOON ONO .NNNNON - NNNO LOOOO ON NOON MO
ON NNON - NNOO NON ON NOOzO: NO O:NNON u NNO LONOO ON LNOzOz O
N ONNNO - NNNO NOOOO ON NOzOz NO ONO: oN ON O-N.N...N-OOOOOOOOO
iN pxcmga-m mo copuuaog seem .oOOOONNO :N pogoupu mo NON uaonm c.
.Om-N._.N.N-NOONOOON-N-NNOONO-N-NNOOOe-N NO OOOON NNN NO OONOONLOO

0

Am :

 

 

 

 

 

 

 

.m «LOOOO

 

 

 

G + K, 1:;:£: c
C + G v— P'
P' 1E=5= P + G

where G - Grignard, K s ketone, C t complex, P a product, p' O
product complexed to Grignard reagent.

R-?-0 + R‘ng ':5==E: RN_?..0...M9__R- O
R

 

R
I— ." R
R“ \Ofi‘Mg—R' | |
R/Sj; f) -—9 12—1—0ng + ng2
lg .
__ x _ 1r

 

In diethyl ether, Grignard bromides and iodides are monomeric
at low concentration (0.05 M or less), and associated only at high
concentration, whereas the Grignard chlorides are associated even at
low concentration‘“. At the concentration we used (Tables IV and V),
even the bromide and iodide Grignard might be dimeric. In tetra-
hydrofuran all of the Grignard reagents are monomeric over a wide
concentration range‘s. The difference between diethyl ether and

tetrahydrofuran is one of degree rather than of kind when you take

l9

into consideration their similar structure, and the difference in
basicity of the two solventsls.

aaGfiB values become more negative as R', on the nucleophile,
increases in size. Entries 31-32 are the only examples where this
effect can be observed directly, but even in those cases where
another factor changes, the direction in which these changes occur
agrees with this statement (entries 40-53; 41-54; and 9-55). The
Karabatsos model explains these results, since the difference between
R' ++ M (VI) and R' ++~L (VII) would becomes larger, favoring VI
over VII as R' increases in size17. when one of the three groups on
the asymmetric carbon atom is isopropyl, AAGXB values decrease sharply
(entries 79-90; and 81-91).

Methyl lithium exhibits higher stereoselectivity than the bromide
or iodide Grignard reagents and similar stereoselectivity to that of
the chlorides (entries 2-6-9; 7-l0; 79-82-87; and 43-45-48). If one
of the three groups is a phenyl, the stereoselectivity is much higher
in ether than in pentane (entries 5-6-7-48-49-87-88; and 47-86). But
when the three groups are hydrogen, methyl, and cyclohexyl, the reverse
effect is observed (entries 97-98). These results are one more evidence

on the solvent dependence of the pheny1 group‘s.

WW

Reductions of carbonyl compounds with lithium aluminum deuteride
in ether are too sensitive to variation in structure. The highest
stereoselectivity is observed when L a phenyl; M . methyl, ethyl, iso-
propyl, or 5-buty1; s - hydrogen; and R (group attached to carbonyl

carbon atom) is t-butyl (entries 66-67). As R increases in size the

20

stereoselectivity increases too (entries 34-63-64-65-66; and 14-111).
when there is no phenyl group on the asymmetric carbon atom, the
stereoselectivity decreases abruptly (entries 105-106-107-109) and
inversion is observed in some cases (entries 103-104), except when R
is iSOpropyl (entry 108).

The stereospecificity of these reductions is highly dependent on
the temperature (entries 68-69-70; 14-15-16; and 34-37-38). No such
dependence is observed when there is no phenyl group on the asymmetric
carbon atom (entries 104-105).

No variation in stereospecificity is oberved as the temperature
changes in the reduction of aldehydes with sodium borohydride or
deuteride in tetrahydrofuran (entries 17-18; 24-25; and 71-72), but
in isopropyl alcohol the stereospecificity is higher and increases as
the temperature decreases (entries 19-20-21; 26-27; and 73-74).
Reduction of cyclohexanones with lithium aluminum hydride leads to a
higher stereospecificity than with sodium borohydride2°. This is also
true for reduction of acyclic aldehydes (entries 15-18; and 69-72),
although there is not too much difference in the solvent properties of
diethyl ether and tetrahydrofuran. Since the mechanism of these
reactions is not well known19, and since they were carried out (Tables
I, II, III) in different solvents, it is not easy to make consistent
conclusions, and more experiments are needed.

For the reactions where AAGAB values agree with those predicted by
the model, one may conclude that the transition states are reactant-like.
0n the other hand, when poor agreement is observed, the conclusion

that extensive bond making and breaking many have occurred, in these

2l

transition states“, may or may not be correct. More bond breaking
and making might have occurred in the transition states for reductions
of aldehydes with lithium aluminum deuteride and sodium borodeuteride,
where M is methyl or ethyl, and L is pheny1, as attested by the small
AAGAB values. However, when the larger isopropyl group has replaced
the smaller methyl or ethyl group on the asymmetric carbon atom, the
AAG B values are more negative than those predicted by the model

(entries 68-69-70; and 71-72-73-74).

W

From the data on Tables I, II, and III it is not difficult to
observe that when stereospecificity decreases due to a change in solvent,
the order is: 1, methyl lithium: its stereospecificity is higher in
ether than in pentane (entries 47-48-49; and 86-87-88). 2, sodium
borohydride or deuteride: it is higher in isopropyl alcohol than in
tetrahydrofuran (entries 17-18-19-20-21; 24-25-26-27; and 71-72-73-74).
3, Grignard reagent chlorides in tetrahydrofuran and in diethyl ether:
here stereospecificity is almost undetectable when changing solvent
(entries 44-45-46; 57-58-59-60-61; and 83-84-85).

The dielectric constant of a solvent increases as the temperature
decreases, and this in turn increases the ratio A/B (entries 19-20-21).
This is general behavior in these kind of reactions. The nature of the
solvent can have even more importance than the nucleophilefi'7 on the
stereospecificity of the reaction21. Solvation through hydrogen bonding,
ion-dipole, or dipole-dipole interactions occurs by means of the close
association of suitable functional groups, and so has many of the

characteristics of a chemical reactionzz.

22

W

The dependence of AAGAB values on the temperature is not constant
in the system under study, C2H5CD(C6H5)C0R (for R a CD3 or H). Thus,
while addition to the ketone of methylmagnesium iodide in ether (entries
2-3-4) and methylmagnesium chloride in tetrahydrofuran (entries 11-12-13)
and reduction of the aldehyde with sodium borodeuteride in tetrahydrofuran
(entries 17-18) reveal an decrease in the negative AAGtB values as the
temperature decreases, addition of methyllithium and methylmagnesium
bromide,both in ether, to the ketone (entries 5-6-7; and 8-9-10), and
reduction of the aldehyde with lithium aluminum deuteride in ether
(entries 14-15-16), show opposite behavior. Similar results are obtained
for the system where the three groups are hydrogen or deuterium, methyl
and phenyl (entries 22-23; 24-25; 26-27; 29-30; 34-38; 39-40; 41-42;
44-46; 48-49; 51-53; 54-56; 57-60; 58-61; 66-67). But when isopropyl is
substituted for methyl, the decrease in AAGKB values with a decrease in
temperature is constant (entries 73-74; 78-80; 81-82; 83-85; 87-88;
89-90; 91-92; and 93-94). Actually, only in the addition of methyl-
magnesium bromide in ether (entries 8-9-10), where there is a difference
of 0.204 kcal/mole in a range of 88°; in the addition of methylmagnesium
chloride in tetrahydrofuran (entries 11-12-13), where the difference is
0.295 kcal/mole for a range of 104°; and in the reduction with sodium
borodeuteride in isopropyl alcohol (entries 19-20-21), with a difference
of 0.172 kcal/mole for 145‘; is the change of AAGtB with temperature
outside experimental error. So, in general, one may say that where the
Karabatsos model applies the entropy effect on AAGta is small, with

enthalpy controlling the difference in free energy.

23

.NNNNNO - NNNV

.NO- NO NO m8in - NNNV .NN NO NO mNOONOO - NNOO oNN NO NO
Opozomhm ONOONOOON ON omoz ONO: iN-pmcmuanpacmsn-N No cowuuanmc
NOON O-N.N-NOOOOOO-N-NNOOOO-N NOON ONO LON OOOOL ONO NO OOOOONNOO

o.
\ .—

 

.O «NOOON

 

24

mm

In additions of Grignard reagents to ketones, where the largest
group is phenyl (entries 1-39-78; 8-43-81; 9-42-82; and 11-44-83),
the ratio A/B decreases as M (medium group in size) increases. Those
-resu1ts are expected since the A/B ratio does not depend on the
R' ++’H and R' ++ M interactions (R' is the incoming group),but on
M ++~0 and L ++ 0 interactions’. No changes are observed when the

alkylating reagent is methyl lithium (entries 7-49-88).

EXPERIMENTAL

WWW-mm- -
a-Phenyl-n-butric acid: Eastman Organic Chemicals.
2-Propanol: Fisher Scientific Company.
Sodium borohydride: Metal Hydrides, Inc.
Lithium aluminum deuteride: Isotopes, Inc.
Sodium aluminum hydride, sodium borodeuteride, methylmagnesium
bromide, methylmagnesium chloride, and methyllithium: Alfa
Inorganics, Inc.
Magnesium: Mallinckrodt.
Methyl iodide: Aldrich Chemical Company, Inc.

Tetrahydrofuran: Baker Analyzed Reagent.

Diethyl ether: Allied Chemical, Specialty Chemicals Division.

WWW~ - Nmr spectra were determined at

60-Mc on a Varian A-60, or 56/600 spectrometer, or at 100 MHzon a Varian
spectrometer, as indicated in Tables I and II. TMS was used as internal
reference in all cases. The diastereomeric product ratio A/B was
determined by at least three integrations at a sweep width of 50 c/s and
a sweep time of 250 sec. When no good resolution was obtained, the
integration was carried out with a planimeter. The concentrations of

the solutions are expressed as percent by volume.

25

26

LOISOCEQ~§B£8$€E° - The spectrometer used to obtain the infrared
spectra was a Perkin-Elmer model 2378. These spectra, taken neat in
sodium chloride cells, were used to check the completeness of the
reduction, and the conversion of OH to OD.

The carbonyl group of 3-phenylpentanone absorbs at 1700 cm"1
(C - O stretching). The disappearance of this band when treated with
the Grignard reagents or MeLi, and the appearance of a strong absorption
at 3425 cm'], associated with the -0H stretching, was used to assess the

completeness of the reaction.

Wat-imm- - A mixture of 49.269 (0-3 Wes)

of a-phenylbutyric acid and 1439 (1.2 moles) of thionyl chloride was
heated in a steam bath under reflux for 31nn The excess thionyl
chloride was removed with a rotary evaporator. A yield of 93% acid
chloride was collected at 114-117° under 15 mm pressure.

In a 1000 m1 3-necked round-bottomed flask, equipped with a stirrer,
a condenser'and a special apparatus for the addition of solids, was
placed 0.375 moles of methylmagnesium bromide in 450 m1 of ether. To
this was added 409 (0.217 moles) of cadmium chloride. The reaction
mixture was maintained for 15 minutes in the ice-bath, for 20 minutes
at room temperature, and fbr 40 minutes at reflux in a steam-bath.
The ether was distilled and benzene was added to the remaining solid.
After distillation of 50 ml, more benzene was added. A total of 400
m1 of benzene was used. After the solution was heated in a steam-bath
for ten minutes, a small amount of the acid chloride mixed with benzene
was added to it until the reaction began. The flask was then cooled
with an ice-bath and the rest of the acid chloride (0.282 moles) was

27

added over a period of 15 mflnutes. The reaction was completed by gently
warming at moderate reflux for an hour. The mixture was poured upon ice
and a sufficient amount of sulfuric acid (10%) was added to dissolve the
white precipitate. The benzene layer, after beflg‘washed successively
with water, dilute sodium hydroxde, and water, was dried over anhydrous
magnesium sulfate. Six fractions (58% yield) were collected.

Identification: The melting point of its semicarbazone is 191-192°
(191 in the literature). It distilled under 8mm of Hg at 103-110°.

NMR spectrum (neat): area ratio
a singlet at r 2.82 5
a triplet at r 6.46 1
a multiplet centered at T 8.00
2
a singlet at r 8.02
a triplet at r 9.22 3
'.-.. . - . 'u‘ - -- - - -5‘ ,. - . - A solution of methyl-

magnesium iodide was prepared under nitrogen from 3.39 (0.022 moles)

of methyl iodide and 0.559 (0.023 moles) of magnesium in 10 m1 of

ether. To this solution was added 2.59 (0.0154 moles) of 3-phenyl-2-
pentanone dissolved in 10 ml of ether, at room temperature, in a

period of 10 minutes. The reaction mixture was then heated to reflux
fer an hour, cooled, and hydrolyzed with 10 m1 of ammonium chloride
(10%). The mixture was transferred to a separatory funnel and shaken,
and the phases separated. The aqueous phase was washed with 30 ml
portions of ether. The combined organic phases were washed with 40 m1
portions of saturated sodium chloride and then were dried with anhydrous

sodium sulfate. The ether was removed in a rotating vacuum evaporator.

28

The alcohol was collected under 6 mm of H9 at 108-110°.

NMR Spectrum (neat): area ratio
a singlet at r 2.80 5
a singlet at T 7.08 l
a multiplet centered at r 7.48 1
a multiplet centered at r 8.18 2
a singlet at r 8.82 6
a triplet at r 9.25 3

The two diastereotopic methyl groups which appeared as a six proton
singlet in the neat sample were resolved into two singlets of equal

intensity in 20% pyridine at 8.69 and 8.73.

gggtggatigp~gt~gaghgnxlggntgnggg. - In a 250 ml round-bottomed flask

equipped with a magnetic stirrer and a condenser with a calcium chloride
drying tube was placed 219 of 3-phenylpentanone. The ketone was treated
successively at 105° with deuterium oxide:

first exchange........... 50 ml of 90% 020 for 48 hr.
second exchange.......... 50 ml of +99% 020 for 48 hr.
third exchange........... 50 m1 of +99% 020 for 48 hr.
fourth exchange.......... 50 m1 of +99% 020 for 4 days.

The pH was maintained at 9-10 with lithium hydroxide-g, After each
exchange the mixture was transferred to a separatory funnel, and the
water layer washed twice wch ether, which was then removed under
vacuum. The NMR spectrum‘was consistent with complete deuteration of
the a-hydrogens. The ketone, in ether, was dried with anhydrous sodium
sulfate, and after the ether was removed, distilled under 10 mm of H9
at 106-110°. The yield was 95%.

29

NMR Spectrum (neat): area ratio
a singlet at T 2.84 5
a multiplet centered at r 8.10 2
a triplet at r 9.22
Purification of 3-pheny1-2-pentanone-1,l,1,3-94. - The NMR spectrum

flAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNN----
of this ketone showed some impurities: three extra peaks at r 8.82,
7.48, and at 7.08. The IR spectrum showed a strong, broad band at
about 3500 cm'1 for a hydroxyl group. It was not possible to separate
the impurities by distillation. The hydroxyl impurity might be the
alcohol fermed from alkylation with excess methyl cadmium reagent

used for the synthesis of the ketone:

The peak at 8.82 is due to -(CH3)2; that at 7.48 a (triplet) to the
proton on C3, and the peak at 7.08 to the 0-H proton. This was confirmed
with the NMR spectrum of the impurity taken after separation by
chromatography. The ketone and the alcohol have very close boiling
points, at about 250°, and because of this, their separation by
distillation was not possible.

This purification of the ketone was carried out by Gas Chromatography,
in an F and M776 model.

Column: carbowax: 20m; 10-60 w 776
Temperatures: Injection port 250°
Oven: 184°
Detector: 222°
Collector Man: 244°
The carrier gas was nitrogen.

cmgamoguagoeuop NON OON mN .NN OmNgucm .Nmnum "NN o» N Omngcm Nuco>NomO

Om-N.N.N.N-mcocON=mO-N-Nxcm;a-N "ON O» N OmNNNON .mcocONOOO-N-NNOOOO-m ON aspen NONNNNOOOON

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NO ON -NN- --. .Fu1, ON
OO ONN-NNN- ,NN- N --. L- ON-
ON‘ . ONN-ONN- O OO NO.O- NOONmOO NN.

WOO NN N9. 9O Nr- NO
ON -. N N0.0 L-- ON-
ON ON ON NO.O NNONNNO O-
OO ONN-ONN NN ON. NO- i0.0 L- N
OO O N ONO --N -N
NO NNN-NNO NF Or ON OOO PENN M
NN -ONN-ONN ON NN ON- O.O -L- O

O.Om NON-OON O ON Or- ON.N L.- O
OO OON-OON N O ON N.N = N
OO- ONNNOON N O LNON .ON.O NOzNOO N
.I. ONN-NON O .- oON O.O NOzNNO N
u go as Aggy

ONONN .OEON Esauu> «EN» .OemN “MN“Muwmwnn OONNsOomNuaz Nuucu

OONNNNNNOONO OONOOOON ONO.OOOO

 

Om-m.N.N.N-acocoucmO-N-Nxcm=N-m
ONO OOOONOOOO-N-NNOOON-N ON NOONNNOOO NON ONNOOON NOOONONLOONN .>N NONON

31

ReactiOn of methylmagnesium iodide with 3-phenyl-2-pentanone-l,1,1,3-94.

 

l) MeMgI

 

C2H5C0(C6H5)COCD3 :> CZHSCD(C6H5)COD(CH3)(CD3)

ether
2) NH4C1/020

To a three-necked 100 m1 round-bottomed flask with 1.19 (0.045 moles)

of magnesium, a small crystal of iodine and 5 m1 of ether, and equipped
with a stirrer, condenser with drying tube, and addition funnel with a
rubber septum and inlet nitrogen tube, a solution of 6.4 g of methyl
iodide (0.045 moles) in 30 m0 of ether was added slowly at the required
temperature (24°, 2°, -18°, or -20°) over a period of 30 minutes. Once
this reaction was completed, 5.09 (0.03 moles) of the ketone dissolved
in 10 ml of ether was added over a period of 20 minutes. After stirring
for about 5 hours at the reaction temperature, 6 m1 of saturated
ammonium chloride in deuterium oxide was added , and then the required
enough anhydrous sodium sulfate. The solution was decanted, and the
salts washed with two portions of 10 ml of ether. The combined ether
solution was dried, the ether evaporated and the alcohol distilled under

8 mm of H9 at 106-110°. The IR spectrum was run from 4000 to 625 em“:

No absorption fer C-O at 1710 cm"1

was detected. A strong, broad band
at 1145 cm'], characteristic of the tertiary alcohols, as well as a
strong, broad band at 3425 cm"1 (D-H stretching) were present. The 0-D

l

stretch at 2540 cm' was also present.

Reaction of methylmagnesium bromide with 3-phenyl-2-pentanone-1J,l,3-g4

W-

 

1) MeMgBr
C2H5c°(ca"5)coc°3 ether > C2H5C0(C6H5)COH(CH3)(C03)
2) NH4Cl/H20

 

32

A solution of 1.009 (6 millimoles) of the ketone in 4 m1 of ether was
added over a period of 15 minutes to a stirred mixture of 10 millimoles
of methylmagnesium bromide (from a bottle 3M MeMgBr in ether) in 15 m1
of ether, maintained at the desired temperature (35°, 2°, or -53°).
After stirring another ten to fifteen hours at the above temperature,

3 ml of 10% ammonium chloride, and then 10 ml of water were added.

The aqueous layer was separated and the ether layer washed with 10 ml
of water and 10 m1 of saturated sodium chloride before drying with
anhydrous sodium sulfate. The ether was removed and the residual oil

distilled.

'»'- ' ' ‘L' 14.11 . ' " J " --
l,l,l,3-_d4 in tetrahydrofuran. - In a 100 ml three-necked round-
Wcondenser and a rubber septum, 3.5 ml

(10 millimoles) of methylmagnesium chloride (from 3M in THF) and 12

ml of tetrahydrofuran were maintained at the desired temperature

(66°, 2°, or -38°) under a nitrogen atmosphere. To this flask was

added by syringe 1.09 (6 millimoles) of the ketone in 6 m1 of tetrahydro-
furan, over a period of 15 mflnutes, and the stirring was continued for
five to ten hours after the addition. Then, 3 ml of 10% ammonium
chloride, 15 ml of water and 40 m1 of ether were successively added.

The organiclayer was washed with a 30 m1 portion of water and saturated
sodium chloride befbre drying over anhydrous sodium sulfate. The
solvents (tetrahydrofuran and ether) were removed in a vacuum evaporator,
and the oil distilled at 116-119°.

1 MeM c1
C2H5C0(C6H5)COCD3 firH———> C2H500(06H5)C0H(CH3)(CD3)

2) NH4Cl/H20

33

Reaction of methyl lithium*with 3-pheny1-2-pentanone-l,1,1,3-da. -

l) MeLi
C2H5C0(C6H5)COCD3 ether E) CZHSCD(C6H5)COD(CH3)(CD3)

2) NH4Cl/DZO

 

In a completely dried flask with condenser and drying tube, stirrer and
rubber septum, and under an atmosphere of pure, dried nitrogen, a
solution of 2.09 (0.012 moles) of the ketone in 8 m1 of ether was added
dropwise from a syringe to a well stirred solution of 0.018 moles of
methyl lithium (from a bottle 2.1M MeLi in ether) and 15 ml of ether.
The addition was carried out at three different temperatures (35°, 2°,
or -57°). The reaction mixture was stirred at the reaction temperature
for at least two hours after the addition was completed. Then it was
added 8 m1 of saturated ammonium chloride in deuterium oxide. The ether
layer was washed with 15 ml of saturated sodium bicarbonate, 15 m1 of
saturated sodium chloride and 15 m1 of water. The solution was dried
over anhydrous sodium sulfate. The ether was removed in a vacuum
evaporator and the alcohol distilled under 11 mm of H9 at 116-120°.

NMR study of 2-methy1-3-pheny1-2-pentanol-1,1,1-3-g4. -
WWW/MAW

Neat: area ratio
a singlet at r 2.84 5
a singlet at T 7.05 1
a multiplet centered at T 8.10 2
a singlet at t 8.84 3
'a triplet at t 9.26 3

The diastereomeric peak at T 8.84 separates into a doublet in a
solution of pyridine. For the main peak:

34

..N NO LONOO ON Om-N.N.N.N-aneONeea-N-NNOONN-N ON NOON NO
OONNNOOO 5: .OO-N. N. N.N-NOeSOaO-N-NNOONO-N-NNONE-N :2 NO 583m $2 .O 95...:
N o:
A
O.ON ON.O ON.N ON.N OO..N
. . .

ON.m -

 

 

 

 

 

35

T 8.75 in 40% pyridine
t 8.74 “ 25% "
r 8.73 ” 20% “
t 8.72 " 15% "

The resolution of the two diastereotopic methyl peaks in the neat sample
was not possible either in fermic acid, or dimethyl sulfoxide. At the
concentration used in pyridine the diastereotopic peaks move close to
the methylene multiplet, and this can explain the impossibility to
obtain a good separation. The ratios A/B fer different nucleophiles,

solvents and temperatures can be feund in Table I.

Sxntng§1§~pj~2apkgpxlpptapal. - In a 1000 m1 three-necked round-bottomed

flask with over-head stirrer, condenser and an additional funnel with a
rubber septum were mixed. 86.89 (0.4 moles) of 2-benzyl-4,3,6-trimethyl-
5,6-dihydro-l,3(4H) oxazine and 400 m1 of tetrahydrofuran. The mixture
was cooled to -78°. Then,276 ml (0.44 moles) of n-butyllithium was

added in a period of 1 hour. The stirring was continued for an additional

0 + n-But.Li 3.2.9

hour. Then, a solution of 489 (0.44 moles) of ethyl bromide and 100 m1

of tetrahydrofuran was added over a period of 30 minutes. The reaction

0 THF 5
/l\\’,9 + CZHSBr '78° ‘7[:J::1\\r/*
’ N

36

mixture was allowed to warm slowly to room temperature and was stirred
overnight. The mixture was poured into 500 m1 of cooled water and
acidified (pH 2-3) with 9N hydrochloric acid. This acidic solution
was extracted with three 150 m1 portions of pentane and made basic by
the careful addition of 40% sodium hydroxide to the water phase. Ice
was added to keep the mixture cool during the neutralization. The
resulting oil was extracted with three 150 m1 portions of ether, and
these extracts were dried over anhydrous potassium carbonate.
Following ether removal, 250 m1 of tetrahydrofuran and 250 m1 of
95% ethyl alcohol were mixed with the crude dehydrooxazine, while a
temperature of -35° - 40° was maintained; then, 9N hydrochloric acid
was added until a pH of about 7. To the mixture were added alternatively
1.529 (0.4 moles) of sodium borohydride in a minimum amount of water
(20 m1), and 9N hydrochloric acid, so that a pH of 6-8 was maintained.

Then the cold solution was stirred for an additional hour. The mixture

THF

0 ethanol 5 :7[:J\‘1
"fiJ\\z,/¢ NaDfl47
pH-
2”s

was poured into 300 m1 of water and made basic (pH 9-10) by addition

of 40% sodium hydroxide. The layers were separated and the aqueous
solution was extracted with three 200 ml portions of ether. The combined
organic layers were washed with 300 ml of saturated sodium chloride, and
dried over anhydrous potassium carbonate. The ether was removed by

rotatory evaporation.

37

To a 1000 ml flask with a distillation head and an addition funnel
with nitrogen inlet tube was added 1 mole of hydrated oxalic acid and
400 m1 of water. The solution was boiled and the tetrahydrooxazine

added dropwise over a period of 30 minutes. The steam distillation was

0
NCO-2H0
224 2: H
@51\\£Lr¢ H//M\\t/’¢
H
H 25

2 5

 

continued until the distillate was free of organic material. The
distillate was extracted with three 150 m1 portions of pentane and
dried over anhydrous sodium sulfate. The solvent was removed and the
residue distilled (total yield 54.5%).

Characterization: melting point of the semicarbazone 155-156°
(Beilstein 155°). Distillation under 13 mm of Hg pressure, at 103-105°
(Beilstein under 15 mm of H9, at 104-106°).

NMR spectrum (neat): area ratio
a doublet at T 0.52 1
a multiplet centered at T 2.82 5
a multiplet centered at r 6.67 1
a multiplet centered at r 8.18 2
a triplet at r 9.22 3

Q8Hk8COEIQNNRINZEBQ£OXLRE££nkL¢ - In a 250 ml round-bottomed flask

equipped with magnetic stirrer and a condenser with a calcium chloride
drying tube was placed 16.359 of 2-pheny1butanal. The aldehyde was
treated successively at 105° with deuterium oxide:

38

first exchange: 35 ml of 90% 020 for 48 hrs.
second exchange: 20 ml of +99% 020 for 24 hrs.
third exchange: 15 ml of +99% 020 for 24 hrs.
fourth exchange: 15 m1 of +99% 020 for 48 hrs.

The NMR spectrum indicated complete deuteration. The pH was maintained
at 10-11 by addition of potassium carbonate. After each exchange the
mixture was transferred to a separatory funnel and the water layer
washed twice with fractions of 25 m1 of ether, which was then removed
under vacuum. The deuterated aldehyde was distilled under 12 mm of H9
at 103-107°.

NMR spectrum (neat): area ratio
a singlet at r 0.52 1
a multiplet centered at r 2.85 5
a multiplet centered at T 8.15 2
a triplet r 9.22 3

Reaction of Lithium Aluminum deuteride with 2-phenylbutana1-2-d_1
1p~gthgg. - To a stirred mixture of 0.59 (0.012 moles) of lithium

alumnnum deuteride in 20 ml of dry ether maintained at the desired
temperature (35°, 2°, or -78°) was added dropwise 1.59 (0.010 moles)

of the aldehyde in 5 m1 of ether in a period of 10 minutes. The

resulting mixture was stirred for ten to fifteen hours after the addition,
and the excess deuteride was decomposed with 5 ml of deuterium oxide.

The ether was separated and dried with anhydrous sodium sulfate. The oil
was distilled in a “macro distillator" under 12 mm of Hg and at 140°

(bath temperature).
1) LiAlD4
CZHSCD(C6H5)CHO ether > C2H5C0(C6H5)CHD(OD)
2) 020

 

39

.mezumeanmp noon on» NN.Omngcm mums» Now «NONONOQEON och

.mm can «N

.—N .ow .mN .mN .NN .mp mmmgucm Now NouwNNNuva-OLUNE a cow: was NH "mgsumgmasmu :oNumpprmNoa

 

 

 

 

 

 

 

 

 

 

 

 

.NozmaogO-N NON can
NN .NN OON.OOO .OegaeeeeNeeeOOO .ON-ON OONNOON .LONOO .NN OON NN .ON .ON OONNOON .OOONNOOO
ONN . ON ON NN NN.O . NN
OO ON NO- ION.O . NN
NO OON N ON ON NN.O . NN
OO . ON N NN.O . ON
N OO NN.O OONOz ON
OO OON NN NN NN- NN.O . NN
OO OON NN ON N NN.O . NN
OON NN N ON OO.O OONONO ON
NNN-ONN ON O ON ONN.O ONNONO ON
oa N \OO 05
N .ONOOON .ee NON O. OOMN N O N
-........ ZN.
m o
.m-N-NOOOOOONNOONN-N NO OONOOOOON eOO OONOOON NOOeNeNNOOxN .N NONON

40

Reaction of Sodium Borodeuteride with 2-phenylbutanal-2-g”1 in

£££KEQXQ£8£EE$Q° - In a three-necked flask with a sealer stirrer,

l) NaBD4
CZHSCD(C6H5)CHO THF :>» C2H5C0(C6H5)CHD(OH)

2) HCl/HZO

 

a reflux condenser and addition funnel was distilled directly 30 ml

of tetrahydrofuran (dried over lithium aluminum hydride. To the same
flask was added 0.29 (4.8 millimoles) of sodium borodeuteride, and then
l.Sg (lO millimoles) of the aldehyde in lo ml of tetrahydrofuran over

a period of l5 minutes at the desired reaction temperature (66° or 2°).
The stirring was continued an additional 5 to 10 hours. Then 7 ml of
2M hydrochloric acid was.added, and the stirring continued for an
additional hour at the reaction temperature. The solvent was removed
under reduced pressure. The residual oil was shaken with three portions
of 5 ml of deuterium oxide, and then stirred for l5 hours at 105° with
l0 ml of +99% deuterium oxide. Distillation was carried out in a

"micro distillator“ under 3 mm of Hg at l00° (bath temperature).

Reaction of Sodium Borodeuteride with 2-Phenylbutanal-2-d1 in Isopropyl

Alggggl. - The equipment used was similar to that in the previous
experiment. The isopropyl alcohol was dried over drierite and distilled
directly in the reaction flask. To a mixture of 0.29 (4.8 mfillimoles)
of sodium borodeuteride and 30 ml of 2-propanol was added 1.59 (lo
millimoles) of the aldehyde in 5 ml of 2-propanol at the reaction
temperature (82°. 24°, or -63°), over a period of 15 minutes, and the
resulting mixture was stirred overnight. Then 10 ml of 2M acid chloride

41

o.¢—

¢~.a
.

 

No coNuuaumN soc»

om.m
.

Nm

om NO are :N NOomOz guNz .m-N-NOcOu=an:mza-N
-~.NaNo:Ou=nuNupazmsa-~ page we Eagpumam mzz

K.\\ o:
/

m¢.w mw.m

.m ogamwu

cm.~

 

 

 

42

was added and stirred for 2 hours. The solvent was removed under vacuum
and ether was added and washed twice with saturated sodium bicarbonate.
The two layers were separated and the ether evaporated.‘ Then, the
residual oil was dried with anhydrous sodium sulfate and distilled in

a “micro distillator“.

NMR study of 2-phenyl-l-butanol-l,2-ga (Figure 6). -

Neat: area ratio
a singlet at r 2.94 5
a singlet at T 5.85 l
a singlet at r 6.48 l
a multiplet at r 8.36 2
a triplet at T 9.24 3

The results of the resolution of the diastereomeric product are for the
neat sample. The diastereotopic protons were not resolvable in carbon
disulfide, l-bromonaphthalene, l,4-dibromobutane; benzene, chloroform,
acetonitrile, dichloromethane, dimethylsulfoxide-g6, pyridine; carbon
tetrachloride, or hexane. All spectra were taken with a l00 MHz
spectrometer. The product ratio A/B for different nucleophiles, solvents,

and temperatures can be found in Table II.

Purification of 2-phenyl-l-butanol-l,2-gK by chromatography. - Because

of the kind of distillation used (micro distillator) it was not possible
to obtain a highly purified sample of 2-phenyl-l-butanol-l,2-92 by
distillation. A pure sample was obtained by VPC under the following

conditions:

43

Instrument: Aerograph A-90-P3, Sargent Model SR

Column: carbowax
Column temperature 2l5°
Carrier gas: helium, under a pressure of 20 p.s.i.

Retention time: 8.8 minutes.

1.

m \l 0‘ U1
0 o o o

10.
11.
12.
13.

14.
15.
16.
17.

18.

BIBLIOGRAPHY
L. Velluz, J. Valls, and J. Mathieu, Angew. Chem., int. ed.,
8, 778 (1967).

D. Rd, and M. A. McKervey, Quart. Rev. (London), 88 (2)
95 (1968).

K. Mislow, "Introduction to Stereochemistry“, Benjamin, New York,
1966, pp. 128-139.

J. H. Stocker, P. Sidisunthorn, B. M. Benjamin and C. J. Collins,
J. Am. Chem. Soc., 88, 3913 (1960).

D. J. Cram, and F. A. A. Elhafez, J. Am. Chem. Soc., 88, 5828 (1952).
D. J. Cram, and K. R. Kopecky, J. Am. Chem. Soc., 81, 2748 (1958).
G. J. Karabatsos, J. Am. Chem. Soc., 88, 1367 (1967).

G.

J. Karabatsos, and T. H. Althius, Tetrahedron Letters, 88,
4911 (1967).

E. L. Eliel, “Stereochemistry of Carbon Compounds", McGraw-Hill,
New York, 1962, p. 151.

D. 0. Cowan and H. Mosher, J. Org. Chem., 81, l (1962).

G. J. Karabatsos and K. L. Krumel, Tetrahedron, 88, 1097 (1967).
E. C. Ashby, Organometallic Chem. Rev., 4, 198 (1968).

E. C.

Ashby, R. 8. Duke, and H. M. Neumann, J. Am. Chem. Soc. ,
88,1964 (1967).

E. C. Ashby, and M. B. Smith, J. Am. Chem. Soc., 88, 4363 (1964).
E. C. Ashby, and H. E. Becker, J. Am. Chem. Soc., 88, 118 (1963).
E. C. Ashby, Quart. Rev. 81, 259 (1967).

T. H. Althius, “Exploration of Factors affecting Asymmetric
Induction in additions to carbonyls directly bonded to asymmetric
gaggers" (Thesis), Michigan State University, Chemistry Library,
9

G. J. Karabatsos, and N. Hsi, J. Am. Chem. Soc., 81, 2864 (1965).

44

19.
20.
‘21.
22.

23.
24.

25.
26.

27.
28.

29.

45

BIBLIOGRAPH (Continued)
H. 0. House, "Modern Synthetic Reactions”, w. A. Benjamin Inc.,
New York, 1965, pp. 24-32.

H. G. Dauben, G. J. Fonken, and D. S. Noyce, J. Am. Chem. Soc. ,
Z8, 2579 (1956).

Dielectric constant of some solvents at 25°, Tetrahydrofuran 1.63;
Pentane 1.76; Diethyl ether 4.34; Isopropyl alcohol 18.3.

E. M. Arnett, H. G. Bentrude, J. J. Burke and P. McC. Duggleby,
J. Am. Chem. Soc., 81,1541 (1965).

Y. Gault, and H. Felkin, Bull. Soc. Chim. Fr., 1342 (1960).
R. Perez-Dsorio and L. Vargas, An. Real Soc. Es ana Fis. uim.
Ser. B, 88, 809 (1964); Chem. 5 r., , .

?1 Chgrest, H. Felkin, and N. Prudent, Tet. Letters, 88, 2199
968

D. J. Cram, F. A. A. Elhafez, and LeRoy Nyquist, J. Am. Chem. Soc. ,
Z8, 22 (1954).

D. J. Cram and F. D. Greene, J. Am. Chem. Soc., 18, 6005 (1953).

D. J. Cram, F. A. A. Elhafez, and H. Heingartner, J. Am. Chem. Soc. ,
18, 2293 (1953).

J. H. Conforth, (Mrs. ), R. H. Conforth, and K. K. Mathews, J. Chem.
Soc., 112 (1959).

 

 

1293 025049