Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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)
SFC User Meeting -Cambridge -2006
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.
SFC User Meeting -Cambridge -2006
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)
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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)
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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 ?
SFC User Meeting -Cambridge -2006
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)
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
18
�7 compounds within krange
�plot ϕ ϕϕϕvs.log k: R2> 0.99
�log k0and Scalculated for the 7 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
SILICA COLUMN
SFC User Meeting -Cambridge -2006
19
�11 compounds within krange
�plot ϕ ϕϕϕvs.log k: R2> 0.99 (except compound 6 at R2=0.92)
�log k0and Scalculated for the 11 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
CYANO COLUMN
SFC User Meeting -Cambridge -2006
20
�8 compounds within krange
�plot ϕ ϕϕϕvs.log k: R2> 0.97
�log k0and Scalculated for the 8 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
DIOL COLUMN
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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
CORRELATING DATA WITH COMPOUNDS
PROPERTIES (II)
SFC User Meeting -Cambridge -2006
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:
log k0= 0.063 µ µµµ+ 0.001 A –1.579δ δδδmin–1.25 R=0.94
•when regre
ssin
g S vs.µ, A
and δmin:
S = -0.011 µ µµµ-2.2x10-5A + 0.071δ δδδmin–0.040 R=0.84
exp log k0vspredicted log k0
R=0.93
exp Svspredicted S
R=0.80
CORRELATING DATA WITH COMPOUNDS
PROPERTIES (III)
SFC User Meeting -Cambridge -2006
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:
log k0= 0.124 µ µµµ+ 0.0002 A –1.409δ δδδmin–1.173 R=0.97
•when regre
ssin
g S vs.µ, A
and δmin:
S = -0.0004 µ µµµ-1.1x10-5A + 0.030δ δδδmin–0.008 R=0.95
exp log k0vspredicted log k0
R=0.97
exp Svspredicted S
R=0.95
CORRELATING DATA WITH COMPOUNDS
PROPERTIES (IV)
SFC User Meeting -Cambridge -2006
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:
S = -0.0008 µ µµµ-7.7x10-5A + 0.040δ δδδmin–0.012 R=0.84
exp log k0vspredicted log k0
R=0.99
exp Svspredicted S
R=0.84
CORRELATING DATA WITH COMPOUNDS
PROPERTIES (V)
SFC User Meeting -Cambridge -2006
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
SFC User Meeting -Cambridge -2006
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.
SFC User Meeting -Cambridge -2006
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)
SFC User Meeting -Cambridge -2006
29
ACKNOWLEDGEMENTS