talk A Cazenave Gassiot Cambridge 2006gjl/docs/2006/Talk_A_Cazenave... · 2018. 6. 1. · SFC User Meeting -Cambridge -2006 14 0 0 t t t k R − = More than the retention time, the

Screening of a SulfonamidesLibrary

by

Supercritical Fluid Chromatography

Coupled to Mass Spectrometry (SFC-MS).

Preliminary properties-retention study.

SFC User Meeting

Cambridge

31stMay –2ndJune 2006

Amaury Cazenave Gassiot

Dr. G. John Langley

SFC User Meeting -Cambridge -2006

2

OUTLINE

�Introduction

�SOTLIB Library

�Isocraticstudy

�Retentionmodelsin chromatography

�Polycraticstudy

-Looking for linear re

lationships

-Correlating d

ata

with p

hysico-c

hem

ical pro

perties

�Conclusion and future work


3

INTRODUCTION

�SFC-MS appears more and more as a technique

complementary to HPLC for high throughput analysis

�Importance of knowing which technique is more suitable

for a specific type of analytes

�Possibility of outlining a set of properties-based rules

allowing prediction of the retention of a given compound

by SFC ?

�To investigatethosequestions: design of a smalllibrary

of sulfonamidescompounds to bescreenedby SFC


4

SOTLIB LIBRARY: NEUTRAL SET

�Coupling of 4 sulfonylchlorides with 3 amines

�All compounds synthesized and analysed by SFC

SO O

N1

SO O

N

F

F

F4

SO O

N7

SO O

NN

H2

10

SO O

N2

SO O

N

F F

F5

SO O

N

8

SO O

NN

H2

11

SO O

NN

3

SO O

NN

F

F

F

6

SO O

NN

9

SO O

NN

NH2

12


5

SOTLIB LIBRARY: BASIC SET

�Coupling of 4 sulfonylchlorides with 5 amines

�All compounds synthesized and analysed by SFC

SO O

NN

13

NN

SO O

14

NN

SO O

15

SO O

N

NEt 2

NEt 2

16

S OO

N

N

17 S OO

N

N

F F

F

22

SO O

N

NEt 2

NEt 2

F F

F

21

NN

SO O

F F

F

20

NN

SO O

F

F

F

19

SO O

NN

F

F

F

18

SO O

NN

23

NN

SO O 24

NN

SO O

25

SO O

N

NEt 2

NEt 2

26

S OO

N

N

27 S OO

N

N

NH2

32

SO O

NN

H2

NEt 2

NEt 2

31

NN

SO O

NH2

30

NN

SO O

NH2

29

SO O

NN

NH2

28


6

ISOCRATIC STUDY: NEUTRAL SET (I)

�Screening of the library using isocratic elution

�Neutral set studied on:

-2-ethyl-pyridylcolum

n 4

.6x250m

m-cyano

colum

n 4

.6x250m

m-diolcolum

n 4

.6x250m

m-bare silica

colum

n 4

.6x250m

m

�Chromatographic Conditions:

-mobile phase: 20%

v/v

MeOH in C

O2

-flow: 4m

L.m

in-1

-temperature

: 35°C

-outlet pressure

: 100 b

ars

-injection volume: 4µL (~0.5

mg.m

L-1)


7

SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2

ISOCRATIC STUDY: NEUTRAL SET (II)

�Retention time related to structure of compounds

�2-ethyl-pyridyl (2-EP) and cyanocolumns:

t R� ��

t R� ��

�On 2-EP phase: co-elution of 2 compounds, despite

distinct physico-chemical properties.


8

SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

�Different trend observed on dioland silica phases:

benzimidazolederivatives eluted the latest

�On bare silica column: compounds 3 and 9 co-eluting

t R� ��

t R� ��

ISOCRATIC STUDY: NEUTRAL SET (III)


9

COLUMN COMPARISON

�Compounds not more

retained by a particular

column

�Same interactions

between compounds

and all four columns ?

�Retention mechanism

mainly due to free silanol

groups ?

�Method: isocratic 20% MeOH, 4mL/min, 35°C, 100 bars

Retention time vs compound number

Compound Number

02

46

810

12

Retention Time (min)

01234

2-EP 60A 6u 4.6x250mm

Cyano 60A 6u 4.6x250mm

DIol 60A 6u 4.6x250mm

Silica 60A 6u 4.6x250mm


10

�Screening of the library using isocratic elution

�Basic set studied on:

-2-E

P c

olum

n 4

.6x250m

m

-cyano

colum

n 4

.6x250m

m-diolcolum

n 4

.6x250m

m

�Chromatographic Conditions:

-modifier: M

eOH +

0.1

% v

/vDEA

-mobile phase: 20%

v/v

modifier in C

O2

-flow: 4m

L.m

in-1

-temperature

: 35°C

-outlet pressure

: 100 b

ars

-injection volume: 4µL (~0.5

mg.m

L-1)

ISOCRATIC STUDY: BASIC SET (I)


11

�Retention time related to structure of compounds

�BUT: only with regards to sulfonylpart

�Sametrend as observedfor neutrals:

t R� ��

ISOCRATIC STUDY: BASIC SET (II)

F F

F

SO O

N

R2R1

SO O

N

R2R1

SO O

N

R2R1

SO O

N

R2R1

NH2

� ��Diol columns: 4 compounds not eluted:

SO O

NN

H2

NEt 2

NEt 2

31

SO O

N

NEt 2

NEt 216

SO O

N

NEt 2

NEt 2

26

SO O

N

N

NH2

32


12

COLUMN COMPARISON

�More differences than

observed with neutral set

�However retention close

on all columns for most

of the compounds

�Retention mechanism

mainly due to free silanol

groups ?

�Isocratic elution, 20% MeOH+0.1%DEA, 4mL/min, 35°C,

100 bars

Retention time versus basic compounds number

Compound Number

12131415161718192021222324252627282930313233

tR (min)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

2-ethyl-pyridyl 60A 6u 4.6x250mm

cyano 60A 6u 4.6x250mm

diol 60A 6u 4.6x250mm


13

LIMITS OF THE ISOCRATIC APPROACH

�No simple correlation highlighted between physico-chemical

properties and tR

�Is the isocratic approach right?

�Indeed: methanol composition (20%) chosen arbitrarily

�So, why 20% and not 30% ?

�Retention time: the most obvious parameter, but is it the

most appropriate ?


14

0

0

t

tt

kR−

=

�More than the retention time, the capacity ratio kis a

convenient way to normalize the retention of an analyte:

�Most successful models of retention developed by Snyder1

and Soczewinski2, in reverse phase systems like ODS/H2O+MeOH:

ϕn

log

log

−=

wk

k

�In normal phase system, strong deviation observed, but linear

correlation can be found in log-log coordinate system:

ϕmlog

const

log

−=

k

1.

Snyder, L

. R., Principles of Adsorption chromatography. First ed.; M

. Dekker: N

ew Y

ork

, 1968

2.

Soczewinski, E

., M

echanistic m

olecular m

odel of liquid-s

olid chro

mato

gra

phy -

Rete

ntion-e

luent com

position

relationships. Journal of Chromatography A 2002,965, (1

-2), 1

09-1

16

RETENTION MODEL IN CHROMATOGRAPHY (I)


15

% Modifier vs. log k

% Modifier

05

10

15

20

25

log k

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

log kw

�Regression analysis: calculation of intercept log kwand slope

�log kw: standardised parameter more reliable than arbitrary

isocratic log k

�Start of “polycratic”study, aim:calculation of intercept

“log k0”and slope S

RETENTION MODEL IN CHROMATOGRAPHY (II)

log k= S x ϕ ϕϕϕ+ log kw


16

�Requirement: keep log kwithin a range of “practical

interest”as defined by Schoenmakeret al.1:

0 < log k< 1

�Getting the retention factor within the range:

-slow eluting compounds: no d

ifficulties since incre

asin

g

ϕallows to e

lute

them

soon e

nough

-fast eluting compounds: decre

asin

g ϕ

doesn’t

syste

matically a

llow sufficient in

cre

ase in rete

ntion

to com

ply w

ith requirem

ent

�Consequence: some compounds to be removed from study

�Analyses carried out three times, plotting of mean values

«POLYCRATIC»STUDY

1.

Schoenm

akeret al.Journal of Chromatography1979, 185, 179-1

95


17

2-EP COLUMN

�10 compounds within krange

�plot ϕ ϕϕϕvs.log k: R2> 0.98

�log k0and Scalculated for the 10 compounds

SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2


18




SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2

SILICA COLUMN


19


�plot ϕ ϕϕϕvs.log k: R2> 0.99 (except compound 6 at R2=0.92)


SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2

CYANO COLUMN


20




SO O

N

F

F

F

SO O

N

SO O

N

SO O

NN

H2

SO O

NN

F

F

F

SO O

NN

SO O

NN

SO O

NN

NH2

SO O

N

F F

F

SO O

N

SO O

N

SO O

NN

H2

DIOL COLUMN


21

CORRELATING DATA WITH COMPOUNDS

PROPERTIES (I)

�On none of the columns, log k0and Scould be correlated

with lipophilicityconstants of the analytes (log P[ACD and

Spartan], log Dat pH 7.4) nor with pKavalues.

�However, other properties of the neutral set compounds

calculated with Spartan, e.g.:

-to

tal dipole m

om

ent µ

-surface a

rea A

-volum

e V

-electronic charg

es o

n sin

gle a

tom

s

�Multiple regression analysis performed to correlate

properties with retention characteristics log k0and S


22

�On 2-EP column:

•when regre

ssin

g log k

0vs.µ, A

and δmin:

log k0= 0.232 µ µµµ+ 0.001 A –1.325δ δδδmin–2.200 R=0.95

•when regre

ssin

g S vs.µ, A

and δmin:

S = -0.003 µ µµµ-1.6x10-5A + 0.27δ δδδmin–0.023 R=0.90

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600 0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

exp log k0vspredicted log k0

R=0.95

-0.040

-0.035

-0.030

-0.025

-0.020

-0.015

-0.010

-0.005

0.000

-0.0400

-0.0350

-0.0300

-0.0250

-0.0200

-0.0150

-0.0100

-0.0050

0.0000

exp Svspredicted S

R=0.90


PROPERTIES (II)


23

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400 0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

-0.080

-0.070

-0.060

-0.050

-0.040

-0.030

-0.020

-0.010

0.000

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

�On cyanocolumn:

•when regre

ssin

g log k

0vs.µ, A

and δmin:


•when regre

ssin

g S vs.µ, A

and δmin:



R=0.93

exp Svspredicted S

R=0.80


PROPERTIES (III)


24

-0.035

-0.030

-0.025

-0.020

-0.015

-0.010

-0.005

0.000

-0.035

-0.030

-0.025

-0.020

-0.015

-0.010

-0.005

0.000

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400 0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

�On diolcolumn:

•when regre

ssin

g log k

0vs.µ, A

and δmin:


•when regre

ssin

g S vs.µ, A

and δmin:



R=0.97

exp Svspredicted S

R=0.95


PROPERTIES (IV)


25

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

00.2

0.4

0.6

0.8

11.2

1.4

1.6

-0.040

-0.035

-0.030

-0.025

-0.020

-0.015

-0.010

-0.005

0.000

-0.04

-0.035

-0.03

-0.025

-0.02

-0.015

-0.01

-0.005

0

�On silica column:

•when regre

ssin

g log k

0vs.µ, A

and δmin:

log k0= 0.119 µ µµµ-0.001 A –1.714 δ δδδmin–1.178 R=0.99

•when regre

ssin

g S vs.µ, A

and δmin:



R=0.99

exp Svspredicted S

R=0.84


PROPERTIES (V)


26

�Isocratic approach: highlighting of trends in retention

�Polycraticstudy:

CONCLUSION (I)

-supposed better characterization

of th

e rete

ntion

-no correlation

of log k

0or S

with lipophilicityconstants or pKa

-study of 12 neutral sulfonamideson 2

-EP, cyano, dioland b

are

silica

sta

tionary

phases

-linearity

found for log k = log k

0+ S x ϕ

R2>0.97


27

CONCLUSION (II)

�Correlating retention characteristics and molecular

descriptors

-Retention characteristics log k

0and S

correlated with

tota

l dipole

mom

ent µ, surface a

rea A

and a

tom

ic charg

e o

n the m

ost negatively

charg

ed a

tom

δmin

-Encouraging results for log k0

which a

re correlate

d w

ith m

olecular

descripto

rs w

ith R

> 0

.94 o

n a

ll four sta

tionary

phases

-Less promising results for S, especially o

n cyano

and silica colu

mns

(R =

0.8

4).

-Results to be considered with care

, since o

bta

ined o

n sm

all set of

structu

rally sim

ilar com

pounds.


28

FUTURE WORK

�Analysis of a set of 20 basic sulfonamidescurrently

under way

�Study to be extended to compounds of different structures

�Recalculation of molecular descriptors by more powerful

algorithms under way

�Regression analysis using other descriptors has to be

undertaken in an attempt to highlight better correlation

(experimental log Pto be measured)


29

ACKNOWLEDGEMENTS

Documents

talk A Cazenave Gassiot Cambridge 2006gjl/docs/2006/Talk_A_Cazenave... · 2018. 6. 1. · SFC User Meeting -Cambridge -2006 14 0 0 t t t k R − = More than the retention time, the