Directed resolution of enantiomers via liquid chromatography of diastereomeric derivatives III. A convenient method to determine the absolute configuration of carboxylic acids R1R2HCCOOH

l'etrahedron LettersNo. 16, pp 34.7 - 1420, 1977.

DIRECTED

RESOLUTION

DERIVATIVES

III.

CARBOXYLIC

ACIDS

OF ENANTIOMERS

A CONVENIENT

lnstitut

fur Organische Stuttgart

VIA LIQUID

METHOD

CHROMATOGRAPHY

TO DETERMINE

OF DIASTEREOMERIC

THE ABSOLUTE

CONFIGURATION

OF

RTR’HCCOOH’) Gunter

D-7000

Pergarnon Press. Printedin Great Britain.

Helmchen,

Chemie,

Helene V&her and Willi

Biochemie und lsotopenforschung

Schuhle der Universitllt,

Pfaffenwaldring

55,

80

(Receivedin UK 15 February1977;

accepted

for

publication

7

March 1977)

Some timeago we introduced the term “directed resolution of enantiomers” to denote a method capable of simultaneous (a) separation of enantiomers, absolute configuration solution

of chirol

(b) determination

compounds. In papers I

of corboxyiic acids and omines con generally

2)

of enontiomeric

tion of the diastereomers. group internally group2)

condition

(c) is fulfilied

Condition

by o model which correlates elution

This model was previously

of

restricted

secondary amides

(b) is met by high performance liquid order and configura-

to amides R’R2XCCONHCHA’A2,

hydrogen bonded to the amide proton and thereby determining

X being a

the conformation of the ocyl

.

We hove since extended our investigations (a) definition titative

and (c) determination

be achieved by forming diastereomeric

and separating them by liquid chromatogmphy on silica gel. chromatography (HPLC)‘,

purity

9 of this series it was shown that directed reond It

prediction

tion of enontiomers. R1R2HCCOOH,

method: (c) quon-

stereochemicol analogy model) and (d) prepamtive

We now report our observations

o class containing

aspects of the directed resolution

(b) measurement of precise separation focton by HPLC,

of separation factors (Ltgi-Ruth

macological and stereochemicol

concerning its application

numerous compounds which are frequently

to amides of corboxylic

sepamacids

encountered ond often of phar

importance. concepts 3

Our method is based on the following

(1) Secondary omides odopt essentially (on silica

to include the following

of the scope of opplicobility,

:

the same conformations

in polar solutions

and in the adsorbed state

gel).

(2) In the adsorbed state o parallel alignment of the planar amide group and the surface of the silica gel is preferred

(the latter is supposed to be planar in micro regions’).

(3) Apolor groups (alkyl, oryl) outside the omide plane cause o disturbance of this preferred arrangement in proportion

to their steric bulk in a direction

perpendicular

to the amide plane I)

. Such groups are classified

as large and small by indices L and S, respectively. (4) That member of a diostereomeric

pair in which both faces of the omide plane are more shielded thon the

least shielded face in the other member is eluted first.

la8

No. 16

(5) There is an attractive

intemction

between small polar groups and the silica

hydrogen bond donors not internally

bonded to the amide group. Formally,

gel, particularly,

if they ore

such groups are assigned to the

S (small) class. On the basis of these points,

specific rules can be formulated for various classes of secondary amides if their

conformations

can be defined independently

Fortunately,

the conformation of the carbinylimine

(la) Amides RCONHCHA1A2 tion characterized

of the individual

properties of the groups referred to under (3)-(5).

moiety is always the same:

(as well as corresponding esters) have invariably

by an approximately

This statement is substantiated

antiplanar

been found to adopt a conforma-

arrangement of the fmgment HNCH

by numerous experimental

observations

38)

(see the figures below).

and by quantum chemical calcula-

tions9). A statement of such genemlity variety of the functional constant conformations

is not possible with respect to the conformation

groups to be considered is much larger.

However,

of the acyl group because the

for seveml types of acyl groups

are also found:

(lb) If there are no hydrogen bonds between R’ or R2 and the amide group the average conformation R’R’HCCONHR

is characterized

by an antiplanar

disposition

of the fragment HCC=O

of amides

(see the figures below).

la) and has since been corrobomted by extensive This conformational assignment has been discussed before NM

investigations

Combination

11)

, x-my crystallogmphy

of the concepts (l)-(S),

the foundation

to the rule illustmted

fore the enantiomer of B). This rule,

12) and quantum chemical calculations 9)

in conjunction

.

with the observations presented in (la) and (lb),

provides

by the figure: A is eluted before B (and the enantiomer of A is eluted beof course, also applies to diastereomen

derived from enantiomeric

amines:

A is eluted before the enantiomer of B (and the enantiomer of A is eluted before 8).

RESULTS.

B

1st eluted

A

The validity

configuration.

2nd

eluted

of this rule has been checked with a large number of diastereomeric

A subset is given in the table.

of known 13) configuration

were determined by synthesis

or by our recently developed PNMR

also been extended to corboxylic and by elemental onalysis.

Configurations

acids

method (PNMR)”

pairs of known (S) from components

which in the meantime hos

11). Each compound was chomcterized individually

by spectml doto

[f]

_

‘R(2) = retention

CH,

C,H, _

‘gH5

‘gH5

‘gH5

‘gH5

‘gH5

‘gH5

-

P-naphthyl

‘gH5 I-naphthyl

42 153 347

1.30 1.80

2.90

[f]

626

424

242

1.51

1.07 [d]

[e]

349

1.81

2.06

308

233

242

253

23.4

23.0

23.0

23.0

23.0

23.0

23.0

23.0

23.5

23.0

23.0

23.0

23.0

0 110

23.0

23.0

23.0

22.9

23.0

temp. [“Cl

173

1.69

1.48

1.51

1.54

1.21

1.0

1.34

349

312

1.70 1.81

313

(AG) [ca 1 [cl

1.70

-A 139

Ccl

1.27

a

l+l.

are rounded off sepamteiy.

[d] Pressure: 50 bar.

0.01

and 4 Cal.,

[e] Pressure: 200 bar.

for a and A (AG)

pressure: 100 bar; appamtw: as described in 3). [b] See text (Discussion). time of 1st and 2nd eluted component, respectively; to = ret. time of a non-ads. subst. (pentane);

8t2 if not otherwise stated;

-

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

CH3

‘gH5

CH2C6H5

COOCH3/CH3

‘gH5 Fermcenyl

“C4H9

!?C6H 13

C2H5

AL

amides A and B.

measurements were performed with each sample; maximal standard deviation

‘R(l)’

acetate

a

respectively; the values of a and A(AG) Eluens: I.b. petroleum ether-ethyl acetate

-RTln a; 5-10

a = tR(2)-to/‘R(l)-to’

[‘I

S/PNM

S/PNMR

PNMR

S/PNMR

PNMR

S/PNMR

S/PNMR

S/PN MR

S/PNMR

S/PNMR

Eluens: hexane-ethyl

A(AG)=

_

CH3

CH3

CH3

CH3

CH3

A

of diastereomeric

COOCH3

C,H,

RL

[a]

PNMR

CH,CH,OH

RS

sepamtion

COOCH3

[b]

liquid chromatogmphic

S/PNMR

PNMR

S/PNMR

S/PNtv’R

PNMR

S/PNMR

Config.

High performance

[a]

No.

Table.

1420

No. 16

DtSCUSSlON.

The results presented in the table are in excellent agreement with the concepk we have deline-

ated. A number of poink deserve special comment: (a) For the sake of convenience we have used hexane-ethyl acetate as the eluent throughout this investigation, tion.

Of particular

acetate,

tetmhydrofumne).

cessary for the determination

amine.

in many cases other solvents give better separa-

value are mixtures of hydrocarbons (alkones,

chloroform,

and optimization

however,

purity.

reagent as it is crystalline,

pure state.

tionally

Because of their ready availibi lity

advantageous and possible on a micro scale.

The development of enantiomer

investigation

order

phenylglycine

the determination

of ab-

and the order of TLC spots has to be es(oxalylic

chloride) or imidazoli-

(e) The pairs -150, b and -16a, b show excepamides of type ‘6 are easily hydrolized

separation reagents on the basis of these findings

by us 14). (f) I n order to examine the efficiency

the pairs of the table with MERCK-LOBAR

of elution

I-Q-naphthyl)-ethyl-

I-phenylethylamine,

(d) Experimentally,

active components, respectively,

high sepamtion effects (see concept (5)). Furthermore,

lute acids.

is only ne-

has a high UV absorption and is easily pre-

Our experience shows that prepamtion of amides via acid chlorides

des is particularly

(ethyl

is extremely simple: A mixture of diastereomeres and a pure or enriched sample has to be

prepared from mcemic and optically tablished.

(b) HPLC

TLC is adequate for the determination

methyl ester and other amino acid esters are also useful reagents. solute configumtion

effect.

(c) Highest separation factors have been obtained with

This compound is an excellent

pared in an enantiomerically

Rotic solvents seem to have a detrimental

of enantiomeric

of sepamtions.

benzene) with polar aprotic solvenk

of our method for enantiomer

a > 1.05 were separated on a preparative scale (l-20

columns or a similar prepamtive

by di-

is under active separations

15) all

g) by chromatogmphy with

low pressure technique developed in our laboratory

14 .

1) Support of this research by the Deukche Forschungsgemeinschaft is gratefully acknowledged. This work has been presented in part in lectures (Regensburg, 22.4.75 and 31.3.76 (Chemiedozenten-Tagung); Aachen, 20.1.77). 2) G. Helmchen, R. Ott, K. Sauber, Tetmhedron Lett. 1972, 3873. 3) G. Helmchen, W. Strubert, Chromatogmphia 7, 713 m. 4) 3) and footnote 13 of 2); for applications of thE method see: 4a) R. Eberhordt, C. G lotzmann, H. Lehner, K. Schlgl, Tetrahedron Lett. 1974, 4365, 4b) C.G. Scott, M.J. Petrin, T. McCorkle, J. Chromatogmphy 125, 157 (1976). 5) It is tobe emphasized that these concepts are based only on experience with silica gel. 6) L.R. Snyder, “Principles of Adsorption Chromatography, Marcel Dekker, New York, 1968, p. 159. 7) We assume that this arrangement is due to double hydrogen bonding of the type: Si-OH ***O=C-N-H*** O(H)-Si . This assumption is supported by the observation that sepamtion factors of diastereomeric esters are generally smaller than those of corresponding amides. 8) W.H. Pirkle, J.R. Hauske, J. Org. Chem. 41, 801 (1976). 9) L. Pavlidou, G. Helmchen, to be published FCILO method). 10) G. Helmchen, Tetmhedron Lett. 1974, 1527. 11) G. Helmchen, to be published. 12) A.T. Hagler, L. Leiserowitz, M. Tuval, J. Am. Chem. Sot. 98, 4600 (1976). 13 Contrary to a statement in 4a), the absolute configumtion of lql-naphthyl)-ethylamine is securely known: M.G.B. Drew, Acta Cryst. B 25, 1320 (1969). 14) G. Nill, Diplomarbeit, Stutts 1976. 15) To date, the cleavage of amides has been proven to be mther difficult. The best method for obtaining the acid appears to be the ni trosamide rearrangement: G . Helmchen, V. Relog, Helv. Chim. Acta 2, 2599 (1972). The amine can only be genemted easily from amides of type 16. 16) B. G latz, Dissertation, Stuttgart 1976. 5000-10000 theoretical plateTm (naphthaline) are achieved routinely with this technique.