Ultrahigh Resolution Chromatography - American Chemical Society

11 Carbon Dioxide Based Supercritical Fluid Chromatography Column Efficiencies and Mobile Phase Solvent Power L. G. RANDALL Downloaded by UNIV OF PITTSBURGH on May 3, 2015 | http://pubs.acs.org Publication Date: April 26, 1984 | doi: 10.1021/bk-1984-0250.ch011

Hewlett-Packard, Avondale, PA 19311

Efficiencies attainable in supercritical fluid chromatography (SFC) are compared to those encountered in packed column high performance liquid chromatog raphy (HPLC) and capillary column gas chromatography (GC) systems. Since SFC efficiencies are quite accept able for high resolution chromatography, exploration of of the available solvent power range of supercritical fluid mobile phases is warranted. In particular, the application of the Snyder solvent classification scheme recently used for the optimization of LC separations is proposed as a framework to aid the supercritical fluid chromatographer in the selection of appropriate modifi ers to alter the solvent power of carbon dioxide. Pre liminary experiments to test the feasibility of this approach have shown that the chromatographic capacity factors and separation ratios can be greatly affected by not only the modifier identity but also the modifier con centration in the modifier/carbon dioxide solvent mixture. One o f the most compelling reasons to use s u p e r c r i t i c a l f l u i d s as solvents i n e x t r a c t i o n s and chromatographic systems i s that the solvent powers o f these f l u i d s approach those o f t y p i c a l l i q u i d s while the v i s c o s i t i e s are g a s - l i k e and the d i f f u s i v i t i e s can approach values intermediate between t y p i c a l gases and l i q u i d s . Moreover, the s u p e r c r i t i c a l f l u i d solvent power i s v a r i a b l e , being a n e a r l y l i n e a r f u n c t i o n o f the f l u i d d e n s i t y , so a wide range o f solvent powers i s a v a i l a b l e f o r each s u p e r c r i t i c a l f l u i d s o l v e n t . This means that a p h y s i c a l parameter, the density (or p r e s s u r e ) , can be v a r i e d to change the solvent power. For those who work p r i m a r i l y with gas chromatography (GC) where the mobile phase i s used only f o r zone movement, and temperature, not the mobile phase composition, i s the parameter used t o c o n t r o l analyte concentration i n the mobile phase, the importance o f a v a r i a b l e solvent power coupled with t r a n s p o r t p r o p e r t i e s somewhat l e s s than favorable f o r GC systems may not be r e a d i l y apparent. However, i t has been noted

0097-6156/84/0250-Ό135S09.50/0 © 1984 American Chemical Society

In Ultrahigh Resolution Chromatography; Ahuja, S.; ACS Symposium Series; American Chemical Society: Washington, DC, 1984.

Downloaded by UNIV OF PITTSBURGH on May 3, 2015 | http://pubs.acs.org Publication Date: April 26, 1984 | doi: 10.1021/bk-1984-0250.ch011

136

ULTRAHIGH RESOLUTION CHROMATOGRAPHY

that o n l y about 10% o f the two m i l l i o n known compounds are amenable to a n a l y s i s by GC ( 1 ) . High performance l i q u i d chromatography (HPLC) i s used to separate many mixtures that cannot be analyzed by GC be cause the mobile phase serves two purposes: s o l v a t i o n and zone movement. Because o f the mobile phase i n t e r a c t i o n with the s o l u t e and the mild o p e r a t i n g temperatures of HPLC, HPLC i s s u i t a b l e f o r a n a l y s i s o f i n v o l a t i l e and/or thermally l a b i l e compounds (the "other 90%"). With a s u p e r c r i t i c a l f l u i d being a hybrid o f gases and l i q u i d s as we normally encounter them, i t follows that s u p e r c r i t i c a l f l u i d chromatography (SFC) i s a chromatographic technique that i s a combination of and complementary to GC and HPLC. When making the d e c i s i o n as to which o f these three chromatographic techniques to use f o r a s e p a r a t i o n , the i n v e s t i g a t o r must consider analyte v o l a t i l i t y and t h e r m o l a b i l i t y , mobile phase solvent power ( i f any), the required chromatographic e f f i c i e n c y , and f i n a l l y the speed o f the a n a l y s i s . As the t i t l e i m p l i e s , the main purpose o f t h i s work was to study mobile phase solvent power. This r e s u l t s from the f a c t t h a t , even though a s u p e r c r i t i c a l f l u i d has a v a r i a b l e solvent power, there i s a maximum value o f the solvent p o w e r — e s s e n t i a l l y that o f the substance as a l i q u i d . I f that maximum solvent power i s not high enough, a higher value can be achieved by u s i n g a mixture o f the chosen s u p e r c r i t i c a l f l u i d with a second, h i g h e r s o l v e n t - s t r e n g t h component. The second component i s o f t e n r e f e r r e d to as a " m o d i f i e r . " Once the necessary mobile phase solvent power has been e s t a b l i s h e d (implying p r i o r e l i m i n a t i o n o f the GC c h o i c e ) , the chroma tographic e f f i c i e n c y and speed o f a n a l y s i s should be considered i n making the choice between c a p i l l a r y column SFC, packed column SFC, and conventional packed column HPLC. Before e x p l o r i n g the s p e c i f i c t o p i c of mobile phase composition i n carbon dioxide based chromato graphy, the question o f whether the realm o f u l t r a h i g h r e s o l u t i o n chromatography i s a c c e s i b l e to SFC should be considered. EFFICIENCIES IN SFC I t i s well-known that extremely high e f f i c i e n c i e s are p o s s i b l e i n c a p i l l a r y column GC: e.g., 250,000 t o t a l p l a t e s f o r a 50 m column y i e l d i n g 5,000 plates/meter (2,3). By comparison, i n HPLC, c l a s s i f i c a t i o n o f columns as to high r e s o l u t i o n or u l t r a h i g h r e s o l u t i o n appears to be l e s s common. A d i s c u s s i o n about what parameters are important i n c h a r a c t e r i z i n g column performance has been presented by Snyder and K i r k l a n d (*•). In p a r t i c u l a r , they o u t l i n e the r e duced parameter approach of Giddings ( 5 ) , where the use o f the reduced p l a t e height (h = HETP/dp) and the reduced mobile phase v e l o c i t y ( ν = ûdp/Pi2 ) y i e l d s one general r e l a t i o n s h i p . (The terms are defined as f o l l o w s : h — r e d u c e d p l a t e h e i g h t , HETP— height equivalent to a t h e o r e t i c a l p l a t e , d — p a r t i c l e diameter, ν — r e d u c e d mobile phase l i n e a r v e l o c i t y , û — m o b i l e phase average l i n e a r v e l o c i t y and P12 — b i n a r y d i f f u s i v i t y . ) A very important point i s that the reduced p l a t e height i s a f u n c t i o n o f only the reduced p



11.

RANDALL

Carbon Dioxide Based Supercritical Fluid Chromatography

137

v e l o c i t y ( v ) and not o f packing p a r t i c l e diameter. Furthermore, the p l a t e height c o e f f i c i e n t s f o r the c o n t r i b u t i o n s o f the d i f f e r e n t band broadening processes to the column p l a t e height have constant, minimum values that are a l s o independent of p a r t i c l e diameter from column to column f o r "we11-packed" columns. A g r a p h i c a l p r e s e n t a t i o n y i e l d s an optimum reduced v e l o c i t y o f about 3 and a minimum reduced p l a t e height o f about 2 f o r a l l columns. I f i t can be a s sumed that p a r t i c l e diameters of 3 ym are p r e s e n t l y s t a t e - o f - t h e art f o r LC systems, then the l i m i t i n g minimum p l a t e height (or height equivalent to a t h e o r e t i c a l p l a t e ) i s 6ym and a 25 cm packed HPLC column would have 42,000 p l a t e s , corresponding to 168,000 plates/meter. This might be designated as an u l t r a h i g h r e s o l u t i o n LC system. A "good column" ( o r , perhaps, a high r e s o l u t i o n column) g e n e r a l l y has reduced p l a t e heights from 2 to 3.5 near the optimum l i n e a r v e l o c i t y ( 4 ) . This would correspond to a lower value f o r the high r e s o l u t i o n range (10 ym p a r t i c l e s , h = 3.5, 25 cm long column) of 7100 p l a t e s per column or 29,000 plates/meter. In SFC e i t h e r c a p i l l a r y or packed columns may be used. Con s i d e r f i r s t a comparison between packed column LC and SFC systems. Experimentally obtained (5) p l o t s of the HETP as a f u n c t i o n o f mobile phase l i n e a r v e l o c i t y f o r a packed column are shown i n Figure 1a. In the LC operating mode a mixture o f a c e t o n i t r i l e and water i s the mobile phase and i n the SFC mode carbon d i o x i d e i s the mobile phase. The solvent powers were adjusted i n each operating mode so that the c a p a c i t y f a c t o r o f pyrene was about the same. In one case the adjustment involved the a c e t o n i t r i l e / water r a t i o and i n the other, the carbon dioxide d e n s i t y . The f i r s t notable point i s that the minimum p l a t e heights are the same. This i s reasonable and p r e d i c t a b l e since the minimum p l a t e height (HETP,^) i n a van Deempter curve f o r a packed column i s a f u n c t i o n * of the packing p a r t i c l e diameter (dp) and the analyte c a p a c i t y f a c t o r ( k ) : f

HETP

min

= f ( d , k») p

•Assuming the simplest case o f no eddy d i f f u s i o n c o u p l i n g (the c o u p l i n g approach (6) d e s c r i b e s v e l o c i t y i n e q u a l i t i e s from l a t e r a l d i f f u s i o n and c l a s s i c a l eddy d i f f u s i o n ) so that the "A" term i s independent of the l i n e a r v e l o c i t y (u) and a l s o assuming no r e s i s tance to mass t r a n s f e r i n the t h i n l i q u i d f i l m (2) o f the s t a t i o n ary phase, the van Deempter r e l a t i o n s h i p f o r a packed column i s HETP = A + Β/û + CD

=

2( λ ) d

2

p

+ 2 Dj^ /û + dp ψ û / D

1 2

f

where HETP, dp, D^ , and k are defined i n the t e x t , λ i s the eddy d i f f u s i o n c o e f f i c i e n t , and ψ = (1 + 6k + 1 1 k ) / ( 2 4 ) ( 1 + k ' ) . f

ûopt = 7z

Then,

D^/d

p

and

f2

2

HETPmin = 2 d λ + p

2/2d *H p

where ïïop i s the optimum l i n e a r v e l o c i t y corresponding to HETPmin. t



138


0

0.2

0.4

0.6

0.8

1.0

1.2

0.010


11.

RANDALL

< Figure

1.

Carbon Dioxide Based Supercritical Fluid Chromatography 139

Van Deempter curves f o r packed column HPLC (#1), packed column SFC (#2), c a p i l l a r y column SFC (#3, #6, #7)· and c a p i l l a r y column GC (#4, #5). Curves 1 and 2, experimen t a l data ( 5 ) ; Curves 3 - 7# c a l c u l a t e d (see t e x t ) . #1 HPLC — pyrene; H y p e r s i l ODS, 5 ym, 10 cm χ 4.6 mm k = 2.85 CH CN/H 0 (70/30), 40°C; take-up s o l v e n t , mobile phase #2 SFC — pyrene; H y p e r s i l ODS, 5 ym, 10 cm χ 4.6 mm k» = 2.30 CO2, 0.8 g/mL, 40°C; take-up s o l v e n t , pentane #3 SFC — c a l c u l a t e d f o r pyrene; d = 50 ym k = 2.30 CO2, 0.8 g/mL, 40°C Dx = 0.00008 cm /s, obtained from Curve 2 #4 GC — c a l c u l a t e d ; d = 50 ym k» = 2.30 Di2 t assumed 0.2 cm /s (0.01 - 1 cm /s (2)) #5 GC — c a l c u l a t e d ; d = 250 ym k r 2.30 D12 , assumed 0.2 cm /s #6 SFC — c a l c u l a t e d f o r naphthalene; d = 50 ym k' = 0.8 C0 , 0.3 g/mL, 40 C = 0.000245 cm /s (13) #7 SFC — c a l c u l a t e d f o r pyrene; d = 50 ym k = 0.32 (14) n - C H , 0.08 g/mL, 210°C D12 = 0.0036 cm /s (14) f

3

2

c


f

2

2

c

2

2

c

1

2

c

e

2

2

c

f

5

12

2

Therefore, any column that i s packed well f o r u l t r a h i g h r e s o l u t i o n LC separations does not lose i t s maximum e f f i c i e n c y i n SFC separa t i o n s . However, the l i n e a r v e l o c i t y (û) must be s i g n i f i c a n t l y higher, by a f a c t o r o f at l e a s t three, t o obtain the most e f f i c i e n t operation. Again, t h i s i s expected since the optimum l i n e a r v e l o c i t y i s i n v e r s e l y p r o p o r t i o n a l t o the p a r t i c l e diameter and d i r e c t l y p r o p o r t i o n a l t o the mobile phase/solute binary d i f f u s i v i t y (D12) and the s o l u t e c a p a c i t y f a c t o r ( k ) : f

S

o p

t

=

«Vd

p

f

D , k»> 12

I f the e f f i c i e n c y per u n i t time, or the r e s o l u t i o n per u n i t time, i s considered instead o f the o v e r a l l e f f i c i e n c y , then i t can be s a i d that the SFC operating mode i s p r e f e r a b l e to the LC mode. The experimental curves i n Figure 1a can be used t o c a l c u l a t e D f o r pyrene i n carbon dioxide at the c i t e d temperature and mobile phase d e n s i t y : D i = 0.00008 cm /s, a value that i s q u i t e 1 2

2

2



140

reasonable ( 7 ) . This value and the Golay equation* (HETP = B/û + Cû) f o r an open tubular column can be used t o generate a t y p i c a l van Deempter curve f o r c a p i l l a r y SFC, as shown i n Figure 1b f o r a column inner r a d i u s , r , o f 25 ym . The experimental data from Figure 1a have been p l o t t e d again i n Figure 1b f o r d i r e c t comparison o f the curve shapes, the minimum values o f HETP, and the o p t i mum l i n e a r v e l o c i t i e s . Of course f o r c a p i l l a r y columns, c

H E T P

k

ûopt = f ( 1 / r t D12 t ' ) k

and


f

m i n = « V '> C

T y p i c a l c a p i l l a r y column GC curves can be s i m i l a r l y generated, as shown i n Figure 1c. In t h i s case an intermediate value f o r the binary d i f f u s i v i t y , 0.2 cm /s, was used along with two column inner diameters,** 250 ym and 50 ym. The 50 ym diameter i s unusually small f o r u l t r a h i g h r e s o l u t i o n c a p i l l a r y GC work while the 250 ym diameter i s i n the middle o f the diameter range commonly used i n t h i s area. As seen i n Figure 1c, a comparison o f the 50 ym diameter SFC curve t o the 50 ym GC curve (Curves 3 and 4) a t the same k emphasizes that the minimum HETP p o s s i b l e i n each case i s the same ( j u s t as f o r the packed column LC/SFC comparison). Thus, the e f f i c i e n c i e s as measured by t o t a l p l a t e s or plates/meter are e q u i v a l e n t . However, higher optimum mobile phase l i n e a r v e l o c i t i e s are p o s s i b l e f o r c a p i l l a r y GC (because o f the higher d i f f u s i v i t y ) so that c a p i l l a r y GC i s s u p e r i o r t o c a p i l l a r y SFC i n terms o f e f f i c i e n c y per u n i t time. In order t o present an order-of-magnitude comparison o f a l l the p o s s i b l e techniques under c o n s i d e r a t i o n , a l o g / l o g coordinate system was used f o r Figure 1c. Hence, the curve shapes appear more shallow compared t o conventional l i n e a r p l o t s . The statement that s u p e r c r i t i c a l f l u i d s are a hybrid o f gases and l i q u i d s as we normally encounter them i s g r a p h i c a l l y represented by the s e r i e s o f t h e o r e t i c a l c a p i l l a r y column SFC 2

f

*For a c a p i l l a r y column, the "A" term i s zero and the column inner r a d i u s , r , r e p l a c e s the p a r t i c l e diameter i n the van Deempter equat i o n , r e s u l t i n g i n what i s known as the Golay equation, so that c

û

t

D12 / r Ψ c

and

HETP^ =

2& v

c

&

**These c a l c u l a t i o n s ignore the f a c t that f o r small column diame t e r s the r e s u l t a n t s i g n i f i c a n t pressure gradients r e q u i r e a more fundamental form o f the Golay equation (8,9). Because the optimum l i n e a r v e l o c i t y becomes i n c r e a s i n g l y more independent o f the diam e t e r as the number o f p l a t e s increases f o r diameters o f t h i s s i z e (8), there i s a t r a d e - o f f between o p e r a t i n g with l a r g e numbers o f p l a t e s but a t l i n e a r v e l o c i t i e s comparable t o l a r g e r diameter c o l umns and fewer numbers o f p l a t e s but a t increased l i n e a r v e l o c i t i e s .


11.


RANDALL

141

curves i n Figure 1d. Again, the simplest form o f the Golay equa t i o n was used to generate these curves. Packed column SFC van Deempter curves would l i e at higher optimum l i n e a r v e l o c i t i e s j u s t as the r e l a t i v e p o s i t i o n s o f Curves 2 and 3 i n Figure 1c j u s t by v i r t u e of the d i f f e r e n c e between the packing p a r t i c l e diameter and the c a p i l l a r y inner radius (5 ym vs 25 ym). The purpose of Figure 1d i s to emphasize the wide range o p t i mum l i n e a r v e l o c i t i e s p o s s i b l e with s u p e r c r i t i c a l f l u i d s as the mobile phase. I t i s important to bear i n mind that the optimum l i n e a r v e l o c i t y i s a f u n c t i o n o f k and D j · Furthermore, k i s a f u n c t i o n o f the mobile phase solvent power ( i t s e l f a f u n c t i o n o f density and chemical composition (χ ) with s u p e r c r i t i c a l f l u i d s ) and the temperature, and the binary d i f f u s i v i t y i s a f u n c t i o n o f d e n s i t y and temperature: 1

f


2

k» = f ( p ,

χ

, T)

and Dl2

= f ( ρ , Τ)

It i s i n s t r u c t i v e to explore these parameters f o r each of the curves i n Figure 1d. Curve 3. In t h i s case the density of the solvent carbon dioxide i s 0.8 g/mL, the temperature i s 40°C, D i s 0.00008 cm /s, and k i s 2 . 3 — a l l parameters chosen to make d i r e c t compar i s o n between packed (experimental data) and c a p i l l a r y columns. The d e n s i t y of 0.8 g/mL, while at the high end o f the 0.1 to 1.0 g/mL range commonly a s s o c i a t e d with s u p e r c r i t i c a l carbon dioxide as a solvent (10), i s a t y p i c a l density f o r d i s s o l v i n g l a r g e r aromatic compounds (11, 12). (For example, Bowman (11) has shown that the threshold d e n s i t y * of carbon dioxide i s 0.1 g/mL f o r d i s s o l v i n g naphthalene, 0.25 g/raL f o r d i s s o l v i n g anthracene, 0.5 g/mL f o r d i s s o l v i n g t e t r a c e n e , and 0.9 g/mL f o r d i s s o l v i n g pentacene at 40°C. Gere (12) has presented chromatographic data showing that at 35°C the c a p a c i t y f a c t o r f o r coronene on a packed reversed phase column ranges from k = 100 at a carbon dioxide d e n s i t y o f 0.8 g/mL down to k = 35 f o r carbon dioxide at 1.0 g/raL.) 12

2

f

f

f

Curve 6. The values used to c a l c u l a t e t h i s curve were ρ = 0.3 g/mL, D = 0.000245 cm /s, and k = 0.8, a l l f o r carbon dioxide at 40°C. The binary d i f f u s i v i t y i s from a range of experimental values measured by F e i s t and Schneider (13) f o r 2

f

12

*The threshold density i s that density of solvent gas at which a s o l u t e begins to d i s s o l v e i n a s u p e r c r i t i c a l f l u i d (or dense gas) at l e v e l s that are d e t e c t a b l e (11). In Bowman's work i t was UV d e t e c t i o n using a mercury lamp; f o r aromatic compounds the s o l u t e s o l u b i l i t i e s corresponding to threshold d e n s i t i e s of carbon d i o x ide were on the order of micrograras-to-nanograms per m i l l i l i t e r of carbon d i o x i d e .



142

f

naphthalene i n carbon dioxide at 40° C. The k f o r naphthalene i s a value determined experimentally on the same type o f reversed phase packed column as Curves 1 and 2 a t the c i t e d average column d e n s i t y , so i t s use here i s analogous to using k = 2.3 f o r pyrene from Curve 2 t o Curve 3· When Curves 3 and 6 are compared i n Figure 1d, i t i s apparent that the minimum HETP l i e s lower, as expected f o r a lower k . The optimum l i n e a r v e l o c i t y i s higher i n Curve 6 than i n Curve 3 not only because the d i f f u s i v i t y i s higher but a l s o , to a l e s s e r extent, because k has decreased. The decrease i n k' i s probably due to two reasons: 1. naphthalene i s more s o l u b l e i n carbon d i o x i d e than pyrene (as evidenced by threshold d e n s i t i e s ) and 2. at 40°C naphthalene i s f a i r l y v o l a t i l e , having a vapor pressure o f almost one Torr at atmospheric c o n d i t i o n s . f

f


1

Curve 7. The values t o c a l c u l a t e t h i s curve were taken from c a p i l l a r y SFC work by Peaden et a l (14) who presented van Deempter curves obtained with pyrene i n n-pentane at 210°C and 29 atm f o r s e v e r a l c a p i l l a r y column diameters. A value f o r D12 f o r pyrene i n n-pentane was c a l c u l a t e d from one o f those curves ( f o r a 300 μπι diameter column and a k of 0.32) i n j u s t the same manner as de s c r i b e d above f o r Curve 2 y i e l d i n g D = 0.0036 cm /s.» The d e n s i t y o f n-pentane, estimated from experimental and reduced** pressures and temperatures, i s about 0.08 g/mL. (From c r i t i c a l values o f 33.3 atm and 469.8 Κ (15) the reduced pressure i s 0.87 while the reduced temperature i s 1.028, corresponding t o a reduced d e n s i t y o f about 0.35 by corresponding s t a t e s . Hence, t h i s mobile phase i s p o s s i b l y b e t t e r r e f e r r e d to as a dense gas: i . e . , f o r reduced p r e s sures > 0.7, gas d e n s i t i e s r a p i d l y approach l i q u i d d e n s i t i e s (16).) For Curve 7 the minimum value o f the HETP i s lower than f o r Curves 3 and 6 (lowest k ) , and the optimum l i n e a r v e l o c i t y occurs at even higher values (lowest k and highest D , with the e f f e c t o f the increased d i f f u s i v i t y being f a r g r e a t e r ) . In a l l three curves the reduced temperature i s about the same; f o r Curves 6 and 7 the s o l u t e v o l a t i l i t i e s at the r e s p e c t i v e temperatures are probably about the same. Since the maximum solvent powers o f carbon d i o x i d e and pentane are s i m i l a r (17, 18), the d e n s i t y - r e l a t e d v o l a t i l i z a t i o n o f pyrene i n t o the pentane mobile phase a t such a low density i s f a r l e s s s i g n i f i c a n t than the high o p e r a t i n g temperature. This i s substantiated by the f a c t that pyrene i s f

2

12

1

f

12

* This value appears reasonable when compared t o a s e l f d i f f u s i v i t y value very roughly estimated using a g r a p h i c a l p r e s e n t a t i o n o f (Ρ·ϋ)/(Ρ·ϋ)° as a f u n c t i o n o f reduced pressure at v a r i o u s reduced temperatures i n the manner described by B i r d , Stewart, and L i g h t f o o t (15). Ρ i s the pressure, D i s the s e l f d i f f u s i v i t y , (Ρ·ϋ) i s t h e i r product at some high pressure and (P«D)° i s t h e i r product at some low reference pressure. ••Reduced parameters:

Actual value d i v i d e d by c r i t i c a l point value.



11.

RANDALL


143

e a s i l y analyzed by GC at temperatures between 200-250°C. Another example o f the e f f e c t o f temperature i s the coronene data o f Gere (12): with a constant d e n s i t y o f 0.8 g/mL (carbon d i o x i d e ) an increase i n operating temperature from 35°C to 100°C causes a decrease i n k' from 100 t o 7.9. (Temperatures i n excess o f 250°C are needed t o analyze coronene by c a p i l l a r y GC.) Thus, with these two examples i t can be seen that operating temperature i s very important. For thermally s t a b l e coumpounds, analyses can be done f a s t e r , with some improvement i n e f f i c i e n c y , at higher operating temperatures. For such a n a l y t e s , the chromatographer may even wish t o use mobile phases with higher c r i t i c a l temperatures so that system pressures are lowered. Other d i s c u s s i o n s about the simultaneous e f f e c t s o f temperature and density may be found i n the l i t e r a t u r e (19· 20, 21). What Figure 1d demonstrates i s that with a s u p e r c r i t i c a l f l u i d mobile phase, the chromatographic c o n d i t i o n s w i l l range from those o f l i q u i d chromatography t o those o f high pressure gas chromatography. Therefore, the s e l e c t i o n o f the column and chromatographic c o n d i t i o n s i n SFC i n v o l v e s the o r i g i n a l set o f c o n s i d e r a t i o n s : r e s o l u t i o n , r e s o l u t i o n per u n i t time, sample c a p a c i t y , mobile phase solvent power, and analyte v o l a t i l i t y and thermolability. Various approaches t o the e v a l u a t i o n o f e f f i c i e n c y ( i . e . , r e s o l u t i o n ) f o r SFC as w e l l as LC and GC are presented i n Table I where i t can be seen that SFC, with packed or c a p i l l a r y columns, compares f a v o r a b l y with the a l t e r n a t i v e s o f packed column HPLC and c a p i l l a r y column GC techniques. In l i g h t o f that c o n c l u s i o n , i t i s reasonable t o explore the parameter o f solvent strength i n SFC. In those cases where analytes are i n v o l a t i l e and/or thermally l a b i l e , a range o f r e l a t i v e l y high mobile phase d e n s i t i e s are required a t mild temperatures. When the maximum solvent power o f a chosen s u p e r c r i t i c a l f l u i d i s not enough, then that parameter must be changed by changing the chemical i d e n t i t y , or f u n c t i o n a l i t y , o f the s o l v e n t . The remainder o f t h i s paper w i l l deal with p r e l i m i n a r y experiments that explore the solvent power o f carbon d i o x i d e / m o d i f i e r mixtures. While the work that w i l l be described was pursued f o r packed column SFC, i t i s c e r t a i n l y a p p l i c a b l e t o c a p i l l a r y SFC as w e l l . PROPOSED FRAMEWORK FOR EXPLORING MOBILE PHASE SOLVENT POWER AND SELECTIVITY Comparison o f S u p e r c r i t i c a l F l u i d Solvents to Conventional L i q u i d Solvents As stated i n the i n t r o d u c t i o n , the v a r i a b l e solvent power o f a s u p e r c r i t i c a l f l u i d i s a n e a r l y l i n e a r f u n c t i o n o f the d e n s i t y o f the f l u i d . I t i s i n s t r u c t i v e t o compare t h i s way o f s e l e c t i n g a solvent power t o the more f a m i l i a r operation used with o r d i n a r y l i q u i d s o l v e n t s . Figure 2 i s a schematic o f s o l u t e s o l u b i l i t y as a f u n c t i o n o f solvent power where the solvent power ranges from that o f non-polar hexane t o polar water; a multitude o f s o l v e n t s l i e s i n between, only a few o f which are shown. Consider a hypo-



2.3

0.000024

3

2.3

3

2.85

2.3

0.00008

0.00008

5

0.000024

2.3

50

5

2.3

0.00008

10 cm

10 cm

10 cm

10 cm

50 m

50 m

50 m

50 m

2.3

2

2.3

Column Length

0.2

k»

0.2

a

- L -

12

(cm /s)

D

0.00008

f

S

250

50

250

(μπι)

de min

2 >

0.71 0.125 0.213

0.012 0.006

2

0.41

0.012 0.006^

9

0.086

0.017

214

42.8

(cm/s)

opt

0.037

0.187

0.037

0.187

(mm)

H E T P

c

d

0.78

1.33

0.23

0.41

970

4900

0.39

1.95

(min)

to

p

c

2.58

81,000

2

52

45,700 4,570 8,333

5.22

8,100

15

81,000 8,100

16,667

50 175

40,400 4,040

1.34

8,333

2

3.4

16,667

0.13 2600 13.140

130,000 657,000

267,000 1,351,000

340 8400

2600 13,140

130,000 657,000

267,000

- N/L -

- Ν -

1,351,000

- η -

e

Effective Effective Effective Plates/ Plates/ Plates/ Meter Second Column

0.77

3200

16.170

1.3

6.4

(min)

Total Plates/ Column

Order-of-Magnitude Comparison of C a l c u l a t i o n s f o r C a p i l l a r y Column GC, C a p i l l a r y Column SFC, Packed Column SFC, and Packed Column LC f o r T y p i c a l High R e s o l u t i o n Dimensions ( d = 250 ym, dp = 5 ym) and Best-Case Dimensions ( d = 50 ym, d = 3 ym). (Experimental values i n i t a l i c s . )

«

2

/2

»

(η) (k'/(k*

op

» L/U t

R

t

2

+1))

ψ*

»

+ 2 /I p

d

tQ(l + k*)

see text for further

o p t

details, details.

factor.

ψ* ; see text for further

where k' is the capacity

or /2 ^/dp ψ*

p

or 2\d

m

n

s

Q

8

From reduced

plate

height p

(h) where h = HETP/d

so h

n i n

- HETP^n/dp

~ 2.

The numbers here are unrealistically optimistic since the calculations ignored the large pressure drop in a very small capillary column and its effect on the diffusivity. Rigorous calculations would have given a slightly higher HETP ^ and a lower Q pt ( *^). For columns with plate numbers greater than 100,000, û becomes increasingly independent of the column inner diameter. For columns of 30-50 \m inner diameter and up to 9 m long, over 150,000 plates are possible (23).

Ν

Ù0

c

and

r

' opt « & h2^c

5

' HETP^n

w

For calculated GC values, the intermediate value of 0.2 cm /sec for D^2 *s used. (The common range for Dj2 * 0.01-1 cm /sec (2). The value of 0.2 cm^/s seems reasonable in light of values of P12 calculated from van Deempter curves in Jtef. 2, Figure 1-4, where heptadecane was the compound of interest. Also approximations of Όχ2 ^or pyrene using data for benzene/helium (22)—i.e., assuming the size of pyrene is 4 times that of benzene—indicate this is not unreasonably low, particularly since the LC and SFC curves are for a heavier solute, pyrene.) For SFC and LC, values from experimental van Deempter curves in Figure 1 were used for all D^ '

LC

SFC

GC

Table I.


> χ

70

Ο

ι

Ο

70

m Ο r c H δ z η x

70

> X Ο X

70

C

11.

RANDALL


145

t h e t i c a l mixture o f A, B and C where only A i s s i g n i f i c a n t l y s o l u b l e i n hexane, A and Β are s o l u b l e i n a 50/50 mixture o f hexane and ether, and a l l three components are s o l u b l e i n e t h e r . These mixture components can be separated from each other by v a r y i n g the solvent power from that o f hexane t o that o f e t h e r — i . e . , the chemical i d e n t i t y o f the solvent must be changed.


f

Figure 2.

Generalized behavior o f s o l u t e s o l u b i l i t y as a f u n c t i o n of solvent power. Solvent placement i s approximately according t o the Snyder p o l a r i t y ranking and the curves of o i l and sugar are based upon various published data.

Another way t o e f f e c t the separation i s t o use s u p e r c r i t i c a l carbon d i o x i d e . As i n d i c a t e d below the x-axis (Figure 2), the solvent power o f carbon d i o x i d e ranges from low values a t low f l u i d d e n s i t i e s t o a maximum value, about that o f l i q u i d carbon d i o x i d e , at higher d e n s i t i e s . Here the chemical i d e n t i t y o f the f l u i d remains the same; only the system pressure i s changed. I f the maximum solvent power i s not s u f f i c i e n t f o r a p a r t i c u l a r a p p l i c a t i o n , there are two c h o i c e s : 1. change the i d e n t i t y o f the s u p e r c r i t i c a l f l u i d (e.g., t o s u p e r c r i t i c a l acetone, methanol, ammonia—often with a concomittant l a r g e change i n operating temperature because o f higher c r i t i c a l temperatures) or 2. use



146


mixtures of carbon dioxide and other more polar s o l v e n t s . The most e f f e c t i v e way to make t h i s choice r o u t i n e l y with minimal t r i a l - a n d - e r r o r experimentation i s to evaluate the solvent power of carbon dioxide with reference to other s o l v e n t s . This allows the use of e x i s t i n g solute s o l u b i l i t y information a v a i l a b l e f o r solvents determined to be s i m i l a r to carbon d i o x i d e . A semie m p i r i c a l approach provided j u s t an estimate i n 1968 (24). More r e c e n t l y , two other approaches have been undertaken (17» 18). They were the c a l c u l a t i o n of the Hildebrand s o l u b i l i t y parameter as a f u n c t i o n of d e n s i t y using tabulated thermodynamic data f o r carbon d i o x i d e and Raman spectroscopy of t e s t s o l u t e s d i s s o l v e d i n s u p e r c r i t i c a l carbon d i o x i d e compared to l i q u i d solvents to evaluate s o l v e n t - s o l u t e i n t e r a c t i o n s . The r e s u l t s of these recent approaches i n d i c a t e d that while the maximum solvent power o f c a r bon d i o x i d e i s s i m i l a r to that of hexane, probably somewhat higher, there i s some s o l v e n t - s o l u t e i n t e r a c t i o n not found with hexane as the s o l v e n t . The l i m i t i n g solvent power of carbon dioxide i s r e solved by choosing the a l t e r n a t i v e of a s u p e r c r i t i c a l f l u i d mixture as the mobile phase. The component added to the s u p e r c r i t i c a l f l u i d to increase i t s solvent power and/or to a l t e r the chromatograph column i s r e f e r r e d to as the "modifier." As i n HPLC, the problem i s now what to use as the second component i n the mobile phase solvent mixture. There are many organic solvents that are s o l u b l e i n l i q u i d (25, 26, 27) and s u p e r c r i t i c a l (27, 28) carbon dioxide and so the chroraatographer i s presented with numerous p o s s i b i l i t i e s from which to choose. In a way, i n SFC there i s an even greater range of modifier choice than a v a i l a b l e i n conventional HPLC because the SFC instrumentat i o n allows pumping of h i g h l y compressible l i q u i d s with low b o i l i n g p o i n t s , with carbon d i o x i d e being an extreme example. Of course, d e t e c t o r - c o m p a t i b i l i t y , t o x i c i t y , expense, and p u r i t y are modifier c o n s i d e r a t i o n s that s t i l l apply i n SFC as well as i n HPLC. The Snyder solvent c l a s s i f i c a t i o n scheme (29, 30) that has been s u c c e s s f u l l y a p p l i e d to LC systems (31-37) f o r LC separation o p t i m i z a t i o n appears to be a general framework that can be used f o r guidance i n the s e l e c t i o n of the m o d i f i e r . Some p r e l i m i n a r y experiments have been done to determine the a p p l i c a b i l i t y of t h i s approach to SFC systems as a prelude to a more in-depth, s t a t i s t i c a l l y - d e s i g n e d modifier s e l e c t i o n study. These f i r s t experiments are the t o p i c o f t h i s paper; the b a s i s of the proposed modifier s e l e c t i o n scheme w i l l now be described i n d e t a i l . Proposed Framework Snyder has discussed l i q u i d solvent c h a r a c t e r i z a t i o n on s e v e r a l occasions (29,30,38,39). One of the very i n t e r e s t i n g points i s that when s e v e r a l solvents have e s s e n t i a l l y the same solvent power as described by the Hildebrand s o l u b i l i t y parameter (40), those s o l v e n t s o f t e n do not d i s s o l v e a p a r t i c u l a r s o l u t e to the same extent. This i s a t t r i b u t e d to the f a c t that the Hildebrand s o l u b i l i t y parameter does not, and cannot by regular s o l u t i o n theory,


11.

RANDALL


r e f l e c t s o l u t e - s o l v e n t i n t e r a c t i o n s stronger than d i s p e r s i v e type f o r c e s . The Hildebrand s o l u b i l i t y parameter f o r a s o l v e n t , the square root o f the molecular cohesive energy per u n i t volume, i s obtained from thermodynamic p r o p e r t i e s o f only the s o l v e n t — i . e . , δ

=

[(ΔΗ*

- RT)/V]** ν

where θ i s the Hildebrand s o l u b i l i t y parameter, Δ Η i s the heat o f v a p o r i z a t i o n , R i s the gas constant, Τ i s the tempera t u r e , and V i s the molar volume. Snyder has derived (29) and r e f i n e d (30) a solvent c l a s s i f i c a t i o n scheme, using data o f Rohrschneider (41), i n which each solvent i s assigned an o v e r a l l " p o l a r i t y , " P . (The p o l a r i t y i s a l s o r e f e r r e d t o as the p o l a r i t y index or the chromatographic strength.) The p o l a r i t y i s an e m p i r i c a l ranking o f solvents and ranges from a value o f 0.1 f o r hexane t o 10.2 f o r water. This measure o f solvent power, s i m i l a r t o the Hildebrand s o l u b i l i t y parameter, i s derived from the i n t e r a c t i o n o f each solvent with a set o f t e s t s o l u t e s (ethanol, dioxane, nitromethane) that probe three p o s s i b l e types o f solvent i n t e r a c t i o n s with the s o l u t e : proton acceptor, proton donor, and strong d i p o l e . Hence, the o v e r a l l p o l a r i t y , P , i s the sum o f that part o f the solvent p o l a r i t y due to each type o f s o l v e n t - s o l u t e i n t e r a c t i o n :


f

f

Ρ*

=

p

' p r o t o n acceptor

+

p f

p r o t o n donor

+

p t

strong

dipole

Solvents can be compared i n terms o f t h e i r s e l e c t i v i t i e s by c o n s t r u c t i n g t r i c o o r d i n a t e p l o t s o f the three s e l e c t i v i t y f r a c t i o n s . The s e l e c t i v i t y f r a c t i o n o f each o f the three i n t e r a c t i o n s can be defined as that amount o f the solvent p o l a r i t y due t o the p a r t i c u l a r i n t e r a c t i o n d i v i d e d by the t o t a l p o l a r i t y . Then each s e l e c t i v i t y f r a c t i o n ranges from 0 t o 1 — e.g., from no proton acceptor character to pure proton acceptor c h a r a c t e r . Snyder presents s e l e c t i v i t y f r a c t i o n s f o r 80 solvents (30); the placement of a few o f those solvents within the s e l e c t i v i t y t r i a n g l e i s shown i n Figure 3· Quite i n t e r e s t i n g l y the 80 s o l v e n t s under consideration f e l l into eight d i s t i n c t s e l e c t i v i t y groups—e.g., a l c o h o l s grouped together; ethers were nearby but d i s t i n c t ; and " u n i v e r s a l " solvents l i k e 2-methoxyethanol, d i r a e t h y l s u l f o x i d e , and dioxane were centered within the t r i a n g l e . I t should be emphasized that the s e l e c t i v i t y mapping shown i n Figure 3 says nothing about the o v e r a l l p o l a r i t y (solvent strength) of each s o l v e n t . In the o v e r a l l p o l a r i t y ranking, toluene with a P o f 2.4 l i e s s i g n i f i c a n t l y lower than diraethylsulfoxide with a P of 7.2. The t o t a l c l a s s i f i c a t i o n scheme i s summarized i n Figure 4, where the o v e r a l l p o l a r i t y P increases along a l i n e a r v e r t i c a l axis, piercing selectivity triangles. When i t i s d e s i r e d t o study a wide range o f solvent s e l e c t i v i t i e s , only three s o l v e n t s — e a c h from a s e l e c t i v i t y t r i a n g l e v e r t e x — a r e needed, not the huge numbers that are p o s s i b l e . Any solvent whose s e l e c t ! v i t y ^ | ^ g p | ^ g ^ h ^ ^ ^ j ^ p e n s e l e c t i v i t y 1

f

f

Society Library 1155

16th St. N. W. Washington, 0 . C. 20030 In Ultrahigh Resolution Chromatography; Ahuja, S.; ACS Symposium Series; American Chemical Society: Washington, DC, 1984.



148

Figure 3.

S e l e c t i v i t y mapping f o r t y p i c a l l i q u i d solvents using s e l e c t i v i t y f r a c t i o n values tabulated by Snyder (30). Ρ ' (Snyder Polarity Index)

water ( 1 0 . 2 ) ^ © , f ormamide (9.6) —

dimethyl sulfoxide (7.2) 2-methoxyethanol (5.5) methanol (5.1) 2-propanol (3.9) chloroform (4.1)

Dispersive Interaction

Figure 4.

—

methylene chloride (3.1) toluene (2.4) carbon tetrachloride (1.6) carbon disulfide (0.2) hexane (0.1) leo-oc tart* (0.1)

C l a s s i f i c a t i o n o f common l i q u i d s according to o v e r a l l p o l a r i t y (Snyder P o l a r i t y Index) and solvent s e l e c t i v i t y . O v e r a l l P values and solvent s e l e c t i v i t y f r a c t i o n s are those tabulated by Snyder (30). f

t r i a n g l e can be approximated by the proper combination o f the three vertex solvents i n a d i l u e n t — i . e . , the d e s i r e d s e l e c t i v i t y i s achieved by the proper ternary combination and the d e s i r e d o v e r a l l p o l a r i t y , by the proper quaternary combination (the s e l e c t i v i t y ternary mixture i n a d i l u e n t f o u r t h component).



11.

RANDALL


149

Within the l a s t s e v e r a l years HPLC separations have been optimized i n terms o f the most appropriate mobile phase composi t i o n f o r a p a r t i c u l a r s e t o f s o l u t e s by e x p l o r i n g the whole plane o f solvent s e l e c t i v i t i e s using t h i s solvent c l a s s i f i c a t i o n scheme with a minimal number o f measurements i n s t a t i s t i c a l l y designed experiments. For reversed phase HPLC systems, the s e l e c t i v i t y t r i a n g l e i s o f t e n defined by methanol, a c e t o n i t r i l e , and tetrahydrofuran with water as the d i l u e n t ( 3 7 ) . However, i n normal phase adsorption systems (or l i q u i d - s o l i d chromatography) the i n t e r a c t i o n o f the mobile phase solvent with the s o l u t e i s o f t e n l e s s important than the competing i n t e r a c t i o n s o f the mobile phase solvent and the s o l u t e with the s t a t i o n a r y phase adsorption s i t e s . Solute r e t e n t i o n i s based upon a d i s p l a c e ment mechanism. Multicomponent mobile phases and t h e i r combination to optimize separations i n l i q u i d - s o l i d chromatography have been studied i n d e t a i l (31-35). Here, solvents are c l a s s i f i e d as t o t h e i r i n t e r a c t i o n with the adsorption surface (Reference 32, in particular): 1. n o n - l o c a l i z i n g solvents that are non-polar t o intermediate p o l a r i t y . The i n t e r a c t i o n with the surface ranges from d i s p e r s i v e t o polar but any part o f the surface i s used f o r i n t e r a c t i o n ; the solvent i s not l o c a l i z e d on an a c t i v e s i t e . Solute r e t e n t i o n i s a f u n c t i o n o f solvent strength ( ε θ ) , simply r e l a t e d t o the mole f r a c t i o n o f each component i n the mobile phase. 2. strong d i p o l a r , l o c a l i z i n g s o l v e n t s . These mobile phase s o l v e n t s i n t e r a c t s t r o n g l y with adsorption s i t e s ; t h e i r solvent strengths depend upon the mole f r a c t i o n o f the solvent on the s t a t i o n a r y phase. When the a c t i v e s i t e s o f the adsorbent surface are covered by the l o c a l i z i n g solvent molecules, f u r t h e r i n t e r a c t i o n o f the solvent molecules with the surface i s n o n - l o c a l i z i n g . When l o c a l i z i n g s o l u t e s and solvents are present, an i n t e r a c t i o n between the two e f f e c t s r e s u l t s i n a cross term between the r e l a t i v e l o c a l i z a t i o n o f the s o l u t e and the solvent l o c a l i z a t i o n and i t s surface coverage. Strong d i p o l a r , l o c a l i z i n g solvents compared a t about the same solvent strength and the same solvent param e t e r (32) e x h i b i t no s o l v e n t - s p e c i f i c l o c a l i z a t i o n — i . e . , the s o l u t e c a p a c i t y f a c t o r (k») and separation r a t i o ( α ) are f u n c t i o n s only o f the mobile phase solvent parameter, 3. b a s i c , l o c a l i z i n g s o l v e n t s . A s o l v e n t - s p e c i f i c l o c a l i z a t i o n i n t e r a c t i o n e x i s t s f o r these s o l v e n t s ; s o l u t e c a p a c i t y f a c t o r s and separation r a t i o s are functions o f the solvent i d e n t i t y as well as the mobile phase solvent parameter. Therefore, i n l i q u i d - s o l i d chromatographic systems s e l e c t i v i t y i n a separation i s determined by the mobile phase solvent strength and i n t e r a c t i o n with the adsorbent. The general s e l e c t i v i t y t r i a n g l e




150

for such systems i s shown i n Figure 5a; i n Figure 5b, examples o f n o n - l o c a l i z i n g , n o n - s o l v e n t - s p e c i f i c l o c a l i z i n g , and s o l v e n t s p e c i f i c l o c a l i z i n g solvents are shown as well as the a c t u a l l i q u i d - s o l i d s e l e c t i v i t y t r i a n g l e solvents heretofore used. ( 3 D While remembering t o keep the d i f f e r e n t separation mechanisms and t h e i r d i f f e r e n t solvent c l a s s i f i c a t i o n schemes i n mind, the chromatographer i n SFC systems should be able t o use such s e l e c t i v i t y t r i a n g l e s for appropriate choices o f m o d i f i e r . For SFC experiments with bonded phase columns, methylene c h l o r i d e as the l a r g e d i p o l e moment, chloroform as the proton donor, and 2propanol as the proton acceptor were chosen as the s e l e c t i v i t y t e r n a r y s o l v e n t s . This s e l e c t i v i t y t r i a n g l e i s compared t o that commonly used (37) t o optimize reversed phase systems i n Figure 5c. The m i s c i b i l i t y o f carbon dioxide with a l a r g e number o f s o l v e n t s allows a l a r g e s e l e c t i v i t y t r i a n g l e t o be chosen. A c e n t e r - p o i n t s e l e c t i v i t y can be achieved by an equal volume mixture o f each o f the three components ( i . e . , 1/3 chloroform + 1/3 methylene c h l o r i d e + 1/3 2-propanol). Since the s e l e c t i v i t y f r a c t i o n s o f 2-methoxyethanol are very n e a r l y the a r i t h m e t i c averages o f each o f the s e l e c t i v i t y f r a c t i o n s o f the three SFC vertex m o d i f i e r s and since 2-methoxyethanol has a high o v e r a l l p o l a r i t y ( P = 5.5)· that solvent was a l s o chosen as a p o s s i b l e m o d i f i e r t o be s t u d i e d . f

• Figure 5. Various s e l e c t i v i t y t r i a n g l e s . a. General s e l e c t i v i t y t r i a n g l e for l i q u i d chromatography

adsorption

b. S e l e c t i v i t y t r i a n g l e used f o r o p t i m i z a t i o n o f l i q u i d adsorption chromatographic separations along with examples o f l o c a l i z e d b a s i c , l o c a l i z e d d i p o l a r , and non-localized solvents. c. Comparison o f the s e l e c t i v i t y t r i a n g l e used f o r the o p t i m i z a t i o n o f reversed phase l i q u i d chromatographic separations t o that chosen f o r the s u p e r c r i t i c a l f l u i d modifier s e l e c t i v i t y survey. d. The two adsorption chromatography s e l e c t i v i t y t r i a n g l e s a v a i l a b l e with chosen SFC m o d i f i e r s compared t o that used i n LC systems. e. Comparison o f s e l e c t i v i t y t r i a n g l e s used i n LC and SFC.

. . . .

L i q u i d adsorption chromatography L i q u i d reversed phase p a r t i t i o n chromatography Adsorption and p a r t i t i o n chromatography with s u p e r c r i t i c a l f l u i d mobile phases


11. RANDALL

Carbon Dioxide Based Supercritical Fluid Chromatography Proton Acceptor (PA) 0.2 > Solvent-specific Localizing (Localized Base)


Non-Localizing

0.2 Proton Donor (PD)

Strong Dipole (DIP)

®

•Localized Base 13— 2-propanol 12— pyridine 11— ethyl ether 10— tetrahydrofuran 9— triethylamine 8— dimethylsulfoxide •Localized Dipole 7— nitromethane 6— ethyl acetate 5— acetone 4— acetonitrile •Non-Localized 3— chloroform 2— benzene 1 — methylene chloride

1— 2— 3— •SFC 4— 5—

acetonitrile tetrahydrofuran £ methanol

chloroform methylene chloride 6— 2-propanol 7— 2-methoxy- , ethanol / 4

/PD 0.2

©

0.2 'PA\ y

r 0.6 A Localized Base •Localized Dipole # Non-Localized

0.4/

0.6,

'PD 0.2

0.4

-y

/ 0.6


/-

151



152

For adsorption chromatography on s i l i c a * with a s u p e r c r i t i c a l mobile phase, the s e l e c t i v i t y t r i a n g l e vertex solvents would be methylene c h l o r i d e or chloroform as the intermediate p o l a r i t y nonl o c a l i z i n g s o l v e n t , 2-methoxyethanol as the n o n - s o l v e n t - s p e c i f i c l o c a l i z i n g s o l v e n t , * * and 2-propanol as the s o l v e n t - s p e c i f i c l o c a l i z i n g s o l v e n t . Therefore, the SFC modifier s e l e c t i v i t y t r i a n g l e chosen for p a r t i t i o n i n g s t a t i o n a r y phases encompasses two s e l e c t i v i t y t r i a n g l e s f o r SFC adsorption chromatography (Figure 5d). F i n a l l y , a l l s e l e c t i v i t y t r i a n g l e s discussed here are compared i n Figure 5e. In these SFC s t u d i e s carbon dioxide i s the d i l u e n t , analogous to the use o f hexane i n LC s t u d i e s and reasonable i n l i g h t o f the r e s u l t s presented i n References 17 and 18 and o f exploratory chromatographic e x p e r i m e n t s — a l l showing that while carbon doxide may have s l i g h t s e l e c t i v i t y (towards the strong d i p o l e vertex, perhaps due t o i t s quadrapole moment), i t has an o v e r a l l maximum p o l a r i t y o f about that o f hexane. RESULTS AND DISCUSSION The purpose o f these f i r s t experiments i n the modifier survey work was t o determine the a p p l i c a b i l i t y o f the Snyder solvent c l a s s i f i c a t i o n scheme t o the choice o f modifier t o see i f an in-depth study would be warranted. The experimental c o n d i t i o n s are described i n the f i n a l s e c t i o n o f t h i s paper. The i n i t i a l experiments at the v e r t i c e s o f the s e l e c t i v i t y t r i a n g l e s d i d show that l a r g e d i f f e r e n c e s i n separation could be induced by the modifier c h o i c e . I t was seen that the c a p a c i t y f a c t o r s and separation r a t i o s were not only f u n c t i o n s o f the modif i e r i d e n t i t y but a l s o f u n c t i o n s o f the modifier concentration i n the modifier/carbon d i o x i d e mobile phase mixtures. Furthermore, adsorption chromatography with a s u p e r c r i t i c a l f l u i d mobile phase does e x h i b i t f a s t column r e - e q u i l i b r a t i o n times upon a change i n mobile phase composition. However, methylene c h l o r i d e and c h l o r o form were not acceptable m o d i f i e r s on the f u l l y a c t i v e s i l i c a adsorbent, although chloroform, i n p a r t i c u l a r , d i d demonstrate a s e l e c t i v i t y d i f f e r e n t from 2-propanol and 2-methoxyethanol as m o d i f i e r s . F i n a l l y , chromatographic peak d i s t o r t i o n , the onset o f which appeared to be s y s t e m - s p e c i f i c , was often but not always observed. Each o f these p o i n t s w i l l now be discussed i n more d e t a i l followed by a concluding o u t l i n e o f p o s s i b l e experiments suggested by these r e s u l t s .

•Snyder and G l a j c h (32) note that a l c o h o l s can be used as s o l v e n t s p e c i f i c l o c a l i z i n g solvents f o r adsorption on s i l i c a , with behavior p r e d i c t a b l e by t h e i r model; t h i s i s not true f o r alumina systems (34). **This solvent appears t o l i e a t the d i v i d i n g l i n e between nons o l v e n t - s p e c i f i c and s o l v e n t - s p e c i f i c l o c a l i z i n g s o l v e n t s ; i t s a c t u a l c l a s s i f i c a t i o n has not been demonstrated experimentally.


11.

RANDALL



Modifier S e l e c t i v i t y Xanthines. I t was found that the c a p a c i t y f a c t o r s and the sepa r a t i o n r a t i o s were both f u n c t i o n s o f the i d e n t i t y o f the m o d i f i e r as well as o f the quantity o f the modifier i n the carbon d i o x i d e . In Figure 6, i t can be seen that with 9.5% 2-methoxyethanol i n carbon d i o x i d e * the f i r s t two components o f a xanthine mixture, c a f f e i n e and t h e o p h y l l i n e , are not well-separated but the e l u t i o n order i s c a f f e i n e , t h e o p h y l l i n e , theobromine, and xanthine. At 6.5% 2-methoxyethanol i n carbon d i o x i d e , there i s e s s e n t i a l l y b a s e l i n e r e s o l u t i o n o f the f i r s t three components i n l e s s than one minute; xanthine (not very s o l u b l e i n l i q u i d 2-methoxyethanol and not at the usual 1 mg/mL sample concentration but at some un determined s a t u r a t i o n concentration) at a k' o f over 5 has begun to t a i l and to disappear i n t o the b a s e l i n e . While i t i s obvious that the c a p a c i t y f a c t o r s , k», increase with a decrease i n modi f i e r concentration from 9.5 t o 6.5% 2-methoxyethanol i n C02, l e s s o b v i o u s l y the separation r a t i o s ( f o r example, with respect to c a f -

1. Caffeine

9.5%

6.5%

2.5%

2. Theophylline

CH

I

I

3

3. Theobromine

4

12

2-Methoxyethanol/CG^

4. Xanthine

S&3 Figure 6.

Separation o f xanthines at various concentrations o f 2methoxyethanol i n carbon d i o x i d e .

* In the 1082 LC/SFC flow c o n t r o l system the c o n t r i b u t i o n o f the pump Β solvent stream t o the o v e r a l l solvent flow, computed from the independent volumetric flow r a t e measurements, depends upon the r e l a t i v e c o m p r e s s i b i l i t i e s o f the two f l u i d s . Here i t i s assumed that pure carbon d i o x i d e and the 10% (molar) m o d i f i e r / carbon d i o x i d e mixtures have about the same c o m p r e s s i b i l i t i e s and so the a c t u a l m o d i f i e r concentration i s given simply by the product o f (%B)(0.1).


154


f e i n e ) a l s o a l l increase with the change i n m o d i f i e r c o n c e n t r a t i o n , and not by a constant f a c t o r . A more dramatic change i n the separat i o n r a t i o s i s shown by the chromatogram of the xanthines with 2.5% 2-methoxyethanol i n carbon dioxide where there i s a change i n e l u t i o n order to c a f f e i n e , theobromine, and then t h e o p h y l l i n e . Estimates of e f f i c i e n c i e s and r e s o l u t i o n can be made by c o n s i d e r i n g the chromatogram f o r 6.5% 2-methoxyethanol i n carbon d i o x i d e . The r e s o l u t i o n between c a f f e i n e (k = 0.88) and theophy l l i n e (k r 1.24) i s 2.17 and between t h e o p h y l l i n e (k» = 1.24) and theobromine (k = 1.7) i s 2.00, y i e l d i n g an average e f f e c t i v e number o f p l a t e s (N) of 884 and an e f f i c i e n c y - p e r - u n i t - t i m e parameter of 20 e f f e c t i v e p l a t e s / s ( t h e o p h y l l i n e ) . This i s somewhat higher than expected f o r an LC system (Table I) and lower than the most e f f i c i e n t SFC system. Compared to the values i n Table I, the l i n e a r v e l o c i t y i s higher than optimum, the d i f f u s i v i t y i s d i f f e r e n t , and the analyte c a p a c i t y f a c t o r ( e l u t i o n time) i s lower. The e f f e c t of the m o d i f i e r i d e n t i t y can be seen i n Figure 7 which shows the separation of the xanthines with 9.5% 2-propanol i n c a r b o n d i o x i d e . Compared to 9.5% 2-methoxyethanol, the c a p a c i t y f a c t o r s of the f i r s t three components are s i g n i f i c a n t l y l a r g e r (by a f a c t o r of 1.6 to 2.5); the e l u t i o n order i s a l s o f

f


f

t

1. 2. 3. 4.

9.5%

P ' = 0.61

P' «

0.45

2-Methoxyethanol/C0

2

Figure 7.

P' =

Caffeine Theophylline Theobromine Xanthine

0.46

2-Propanol/C0

2

Separation of xanthines with 2-propanol as the m o d i f i e r compared to 2-methoxyethanol as the m o d i f i e r .


11.


RANDALL

155

d i f f e r e n t , with t h e o p h y l l i n e preceding c a f f e i n e . Since the over a l l p o l a r i t y , P , of 2-raethoxyethanol i s higher than that o f 2propanol (P = 5.5 vs 3.9· Figure 4), i t would more f a i r to compare the mobile phase mixtures at equal o v e r a l l p o l a r i t i e s . With the assumption that the p o l a r i t y o f carbon d i o x i d e i s 0.1, l i k e hexane and pentane, with no s i g n i f i c a n t s e l e c t i v i t y , i t can be c a l c u l a t e d a f t e r the manner of Snyder* that 6.5% 2-methoxy ethanol i n carbon d i o x i d e has about the same o v e r a l l p o l a r i t y as 9.5% 2-propanol i n carbon d i o x i d e . A comparison o f the separations of the mixture using the equal p o l a r i t y s o l v e n t s shows that the e l u t i o n times and order are s t i l l d i s t i n c t l y d i f f e r e n t f o r the two modifiers. Further experiments with 2-propanol as the m o d i f i e r i n the xanthine separation showed another e l u t i o n order change with a m o d i f i e r concentration change as shown i n Figure 8. Another phenomenon, p a r t i c u l a r l y evident i n the chromatograms at the lower 2-propanol m o d i f i e r percentages, i s peak d i s t o r t i o n . The d i s t o r t i o n appears to occur when the mobile phase solvent power becomes marginal f o r competiting with the column f o r the solute or f o r d i s p l a c i n g the s o l u t e from the column a c t i v e s i t e s . I t s onset seems to be r e l a t e d to the p a r t i c u l a r solute-mobile phase compositionchromatograph column combination under c o n s i d e r a t i o n — i . e . , i t i s s p e c i f i c to each d i f f e r e n t combination. f


f

*

9.5%

i

5.5%

ι»

r 1

4.5%

1. Caffeine 2. Theophylline 3. Theobromine

*

2Propanol/C0

2

Figure 8.

Separation o f xanthines at v a r i o u s concentrations of 2propanol i n carbon d i o x i d e .

* Ρ = ΦΑ k + Β Β ^er* nents A and Β and ?\ , P 1

p

φ

ρ ι

Φ

f B

Φ

a

r

e

Α· Β volume f r a c t i o n s o f compo are o v e r a l l p o l a r i t i e s of A and Β (39).



156


Hormones » Another example o f the dependancies o f component e l u t i o n time and order upon the m o d i f i e r i d e n t i t y and content i n m o d i f i e r / carbon d i o x i d e mixtures i s presented i n Figure 9· The chromatogram with chloroform as the m o d i f i e r demonstrates f a i r l y t y p i c a l chromatographic behavior with methylene c h l o r i d e and chloroform as modif i e r s . In general, these two m o d i f i e r s simply d i d not y i e l d acceptable chromatograms on the H y p e r s i l SIL column; the reason f o r t h i s f a i l u r e w i l l be explored l a t e r i n t h i s d i s c u s s o n . Even i f chromatography with these m o d i f i e r s appears t o be useless a t a f i r s t glance, there i s a d i f f e r e n t s e l e c t i v i t y when using them compared t o 2-methoxyethanol and 2-propanol as seen i n Figure 9 and as can a l s o be seen i n Figure 10.

1. Progesterone H çÇocH 3

3

9.5%

2.5%

2. Vitamin D

4.5%

9.5%

2

3. Methyltestosterone

J IL 4. Estrone 2-Methoxyethanol

Figure 9.

2-Propanol

H

Chloroform

Separation o f a hormone mixture using 2-methoxyethanol, 2-propanol, and chloroform as m o d i f i e r s i n carbon d i o x i d e .

S u b s t i t u t e d anthraquinones. In the l a t t e r f i g u r e , 9-5% 2propanol i n carbon d i o x i d e as the mobile phase r e s u l t s i n a separation very s i m i l a r t o that with 5.5% 2-methoxyethanol i n carbon d i o x i d e . In both cases 1,8-dihydroxyanthraquinone e l u t e s with and r i g h t a f t e r anthraquinone so those components are not separated, even a t much lower m o d i f i e r c o n c e n t r a t i o n s ; the r e t e n t i o n times o f anthraquinone and 1,8-dihydroxyanthraquinone increase together with the t a i l i n g o f the 1,8-dihydroxyanthraquinone becoming more and more pronounced as the m o d i f i e r conc e n t r a t i o n i s decreased. However, with chloroform as the modif i e r , those two components are s i g n i f i c a n t l y s p l i t apart, with 1-aminoanthraquinone e l u t i n g i n between them. Another i n t e r e s t ing feature o f Figure 10 i s that 1,2-dihydroxyanthraquinone was never e l u t e d from the H y p e r s i l SIL column, no matter which m o d i f i e r was used. This d i f f e r e n c e i n isomer e l u t i o n between the


11.


RANDALL

1. Anthraquinone

9.5%

5.5%

9.5%

2. 1,8-Dihydroxyanthraquinone

3. 1-Amlnoanthraquinone

4

5

L.

4. 2-Aminoanthraquinone

«Ai


1.2

2-Methoxyethanol 2-Propanol

Chloroform

5. Impurity from #4 6. 1,2-Dihydroxyanthraquinone (never)

Figure 10. Separation o f a mixture o f s u b s t i t u t e d anthraquinones with 2-methoxyethanol, 2-propanol, and chloroform as m o d i f i e r s i n carbon d i o x i d e .

1,8- and the 1,2- dihydroxyanthraquinones seems t o f i t o l d open column s i l i c a chromatography data f o r s u b s t i t u t e d anthraquinones which showed that i n t e r n a l hydrogen bonding between the ketone oxygens and the hydroxy groups on adjacent carbons (the 1,4,5 and 8 p o s i t i o n s ) could occur and thus i n t e r f e r e with solute adsorption (42). Compounds with hydroxy groups i n the other p o s i t i o n s that do not allow i n t e r n a l hydrogen bonding are f a r more s t r o n g l y adsorbed. The same behavior i s observed f o r amino groups. F i n a l l y , amino-substituted anthraquinones are l e s s s t r o n g l y adsorbed than hydroxy-substituted anthraquinones so the e l u t i o n o f 2-aminoanthraquinone before the 1,2-dihydroxyanthraquinone ( s p e c i f i c a l l y the s u b s t i t u t i o n on the number 2 p o s i t i o n ) a l s o agrees with e a r l i e r adsorption chromatography experience ( 4 2 ) . Xanthone, flavone and s i m i l a r compounds. A d i f f e r e n c e i n c h l o r o form s p e c i f i c i t y from 2-methoxyethanol i s again demonstrated i n Figure 11. Xanthydrol i s eluted before xanthone and flavone with chloroform as the m o d i f i e r . Perhaps t h i s shows the u s e f u l c o u p l i n g o f a proton donor solvent and proton acceptor s o l u t e (a l a r g e secondary solvent e f f e c t i n an adsorption system (43))· the i n t e r a c t i o n that was hoped f o r with the s e l e c t i o n o f chloroform as one o f the s e l e c t i v i t y t r i a n g l e m o d i f i e r s and apparent here because o f a l e s s strong adsorption o f the xanthydrol on the f u l l y a c t i v e s i l i c a than some o f the other b a s i c s o l u t e s used i n the preliminary studies. The t a i l i n g shown by the Biochanin A chromatographic peak i n Figure 11 with 3.5% 2-methoxyethanol i n carbon d i o x i d e was t y p i -



158

1. Xanthone

2. Flavone 8.5%

9.5%

3.5%

3. Xanthydrol

4. Valmorin


5 4 1,2,3

2,3 1

5. BiochaninA

2Methoxyethanol/C0

2

Figure 11.

Chloroform/C0

2

Separation o f xanthone, flavone and s i m i l a r compounds with 2-methoxyethanol and chloroform as m o d i f i e r s .

c a l l y observed with hydroxy group s u b s t i t u t i o n s on aromatic r i n g s . An increased m o d i f i e r concentration can help t o minimize the t a i l i n g , as shown by the chromatogram f o r 8.5% 2-methoxyethanol i n carbon d i o x i d e . I t i s somewhat s u r p r i s i n g to the author that the a n a l g e s i c drug, Valmorin, does not e x h i b i t t a i l i n g or peak d i s t o r t i o n a t the lower 2-methoxyethanol c o n c e n t r a t i o n s ; perhaps t h i s i s another case o f i n t e r n a l hydrogen bonding p a r t i a l l y i n t e r f e r i n g with s o l u t e adsorption on the s i l i c a . Table I I i s a summary o f the various s o l u t e - m o d i f i e r combina t i o n s that were t r i e d on the H y p e r s i l SIL column. A d d i t i o n a l i n formation i s presented i n Reference 44. Adsorption

Chromatography

Chloroform and methylene c h l o r i d e as m o d i f i e r s . In g e n e r a l , even though they demonstrated s e l e c t i v i t i e s d i f f e r e n t from the other m o d i f i e r s , chloroform and methylene c h l o r i d e were d i s a p p o i n t i n g as m o d i f i e r s f o r adsorption chromatography on the f u l l y a c t i v e H y p e r s i l SIL s i l i c a column. This may a r i s e more from the f a c t that the s i l i c a column was f u l l y a c t i v e than from t h e i r lower solvent strengths ( ε° ) i n the s i l i c a adsorption e l u o t r o p i c series. Snyder has emphasized the need f o r adsorbent d e a c t i v a t i o n where the d e a c t i v a t i n g molecules are s e l e c t i v e l y adsorbed on the strongest s i t e s f o r decreased surface heterogeneity, increased l i n e a r c a p a c i t y , and high chromatographic e f f i c i e n c i e s (43). I t i s p l a u s i b l e that both o f the p o l a r l o c a l i z i n g m o d i f i e r s (2-methoxy ethanol as the n o n - s o l v e n t - s p e c i f i c and 2-propanol as the s o l v e n t s p e c i f i c l o c a l i z i n g s o l v e n t s ) served as d e a c t i v a t o r s to the strongest a c t i v e s i t e s , with e s s e n t i a l l y i r r e v e r s i b l e adsorption u n t i l column r e a c t i v a t i o n at elevated temperatures with pure



1-Aminoanthraquinone 2-Aminoanthraquinone 1,8-Dihydroxyanthraquinone 1,2-Dihydroxyanthraquinone Anthraquinone

Caffeine Theophylline Theobromine Xanthine Adenine

Diphenylphthalate Dimethylterephthalate Di-n-butylphthalate Phthalic a c i d

Xanthone Flavone Xanthydrol Biochanin A Valmorin

Progeeterone Methyltestosterone Estrone Vitamin D

1-Aminoanthraquinone 1-Nitronaphthalene 1-Naphthyl-acetic a c i d Acenaphthenequinone p-Nitrophenylacetonitrile A n i s y l alcohol 2-Phenylethanol

II

III

IV

V

VI

VII

1

1 3 1 5 5 4 4 4 2,4 2 1 5 2 2 4 4

5

1 1 1 3 1 1 1,2 1,2 1 1 1 5 1 1 1

5

5

1 1 1,2 1,3 1,2 1 1,2 1,2 1 1 1 3 1,2 1 1 1

5 1 1,2 1,2 1,2 5 3 1 1 1

Chloroform

1 1 1 2 2 2 2 5 1 5 5 5 5 5 1,2 1,2 1,2

ι 1 1 1 1 1

2-Propanol

1 1 1 1,2 1 1,2 3 5 1 1,2 1,2 1,2 1,3 1,2 1 1 1

2-Methoxyethanol

2

Mobile Phases

2 5 5 5 5 5 5 5 4 2 1 5 4 1 2 5

5

1 1,2 2 4 2 6 6 6 1,2 5 5 5 5 5 2 2 2

f

Methylene Chloride

of Various Test Solutes with M o d i f i e r / C 0

Sharp chromatographic peaks; 2 — Peak distortion at some modifier concentration (100XB to 15XB the whole range was not necessarily studied for each modifier); 3 — Tailing; 4 — Poorly eluted, distinct from the characteristic peak distortion and from tailing; 5 — Never eluted at maximum modifier concentration; 6 — Never tried solute/modifier combination.

2

Anthracene Anthraquinone Xanthone Acenaphthenequinone

Chromatographic Behavior

I

Table I I .


^ vO

1

S 3

CO

δ*

Ο

5-

o,

> o > r r

70


160

carbon d i o x i d e , as w e l l as the d e s i r e d r e v e r s i b l e l o c a l i z i n g agents a t the remaining weaker a c t i v e s i t e s . Furthermore, i t would be expected that the moderately p o l a r n o n - l o c a l i z i n g m o d i f i e r s , chloroform and methylene c h l o r i d e , could not f u n c t i o n as d e a c t i v a t o r s . Thus, the strongest a c t i v e s i t e s were a v a i l a b l e f o r cheraisorption o f the most p o l a r , l o c a l i z i n g s o l u t e s when chloroform and methylene c h l o r i d e were used as m o d i f i e r s .


Fast column r e - e q u i l i b r a t i o n . Some l i m i t e d measurements were made t o assess the time necessary f o r column-mobile phase e q u i l i b r a t i o n when a b a l l i s t i c change i n modifier concentration was made. Thus f a r , these measurements have involved only 2-raethoxy-

2.5-r-

w

F

•

*-

< ο Si.sH LU ο

+—*-

i-

z LU t •—D

Ψ

\**+• •• ^

•

« - « — «

Β «-

G

Α

ι ι ι I I I »I | I I I I | I I I I | I I I I | I I I I I I I I I I I I I I I Μ ι ι ι I I I I I 10 20 30 40 50 60 70 80 90 100 TIME AFTER "FLOW STABLE" (min)

Figure

12. Column r e - e q u i l i b r a t i o n upon a b a l l i s t i c change i n mo b i l e phase composition. Curves A - F, the same H y p e r s i l SIL column used f o r chromatograms i n t h i s paper; Curves G and H, another H y p e r s i l SIL column. A Β C D

-

%B: %B: %B: %B:

75% 65% 55% 45%

==> 65» ==> 55% ==> 45% ==> 35%

Ε F G H

-

%B: %B: «B: %B:

35% 25% 25% 65%


==> ==> ==> ==>

25% 15% 65% 25%


11. RANDALL


ethanol as the modifier and the xanthines as s o l u t e s and were c a r r i e d out as o u t l i n e d i n the experimental s e c t i o n . The r e s u l t s of these experiments are presented i n Figure 12, i n which the r e t e n t i o n time o f c a f f e i n e i s p l o t t e d as a f u n c t i o n o f time elapsed from the "FLOW STABLE" c o n d i t i o n . G e n e r a l l y , i t can be s a i d that r e - e q u i l i b r a t i o n i s a t t a i n e d i n a matter o f 2 t o 5 minutes, not only f o r small changes o f 10%B (1% a c t u a l m o d i f i e r concentration) but f o r l a r g e r changes o f 40%B (4% a c t u a l modifier c o n c e n t r a t i o n ) — i . e . , the r e t e n t i o n times o f c a f f e i n e r a p i d l y reach values bracketed by the v a r i a t i o n i n r e t e n t i o n times a f t e r e q u i l i b r a t i o n . Furthermore, the r e t e n t i o n time s t a b i l i t y a t lower m o d i f i e r c o n c e n t r a t i o n o f t e n appears l e s s than a t higher concent r a t i o n s , probably due t o the onset o f peak d i s t o r t i o n with i t s attendant d i f f i c u l t y i n peak i n t e g r a t i o n . C e r t a i n l y , more e x p e r i ments with d i f f e r e n t m o d i f i e r s , d i f f e r e n t s o l u t e s , and a f a r l a r g e r range i n m o d i f i e r c o n c e n t r a t i o n are necessary f o r a f a i r and general comparison o f the short r e - e q u i l i b r a t i o n times o f SFC to the t y p i c a l l y l a r g e e q u i l i b r a t i o n times o f normal phase LC on the order o f hours (45» 46). Peak D i s t o r t i o n As noted e a r l i e r , peak d i s t o r t i o n was commonly observed a t lower solvent powers—with a p a r t i c u l a r "lower" solvent power being s p e c i f i c a l l y a s s o c i a t e d with the s o l u t e o f i n t e r e s t . For example, as shown i n Figure 6, very sharp, well-shaped chromatographic peaks are obtained f o r c a f f e i n e , t h e o p h y l l i n e , and theobromine with 6.5% 2-methoxyethanol i n carbon d i o x i d e ; however, with the lower 2.5% 2-methoxyethanol i n carbon d i o x i d e concent r a t i o n , the onset o f d i s t o r t i o n i s e v i d e n t . For anthracene on the same H y p e r s i l SIL adsorption column, no d i s t o r t i o n i s observed with pure carbon d i o x i d e f o r column midpoint* d e n s i t i e s above 0.5 g/mL (60°C, p = 0.60 g/mL, P t = °* * S/mL, s e t t i n g o f 5 mL/min); however, even with a constant o u t l e t d e n s i t y , d i s t o r t i o n can be observed when the flow r a t e i s slow enough that the column i n l e t d e n s i t y and thus the d e n s i t y / pressure-drop p r o f i l e across the column does not provide t h i s minimum midpoint d e n s i t y . The o r i g i n o f t h i s d i s t o r t i o n has not yet been completely determined. At f i r s t column o v e r l o a d i n g by s o l u t e or a poorly packed column were suspected; these p o s s i b i l i t i e s were both e l i m i n a t e d . The f i r s t was explored by lowering the i n j e c t e d s o l u t e quantity t o a few nanograms. In that case the d i s t o r t i o n 32

i

n

F

L

0

W

o u

•Column midpoint density i s that d e n s i t y corresponding t o the midpoint pressure, P ^ ^ = ( P i + P u t ) / 5 m i d d ° necess a r i l y equal the a r i t h m e t i c average o f the i n l e t and o u t l e t densities, ( p ^ + p )/2. For the experimental c o n d i t i o n s c i t e d i n the t e x t g i v i n g values o f Pmid about 0.5 g/mL, Pmid /Pave about 1.05. 2

n

n

p

n

o

t

o u t

o

i

e s

0

f

s



162


p e r s i s t e d . A second H y p e r s i l SIL column, with an acceptable minimum p l a t e h e i g h t , was s u b s t i t u t e d f o r the f i r s t column; however, the peak d i s t o r t i o n behavior a t low solvent powers was s t i l l observed. I t has been suggested to the author that the d i s t o r t i o n i s due t o using a h i g h l y p o l a r take-up solvent ( 2 methoxyethanol i n t h i s work) that i n t e r f e r e s with the a c t u a l chromatographic e l u t i o n by l e s s polar mobile phases f o r some i n i t i a l length o f the column. Another p o s s i b i l i t y that has been mentioned i s that the i n j e c t i o n procedure causes the d i s t o r t i o n . A l i m i t e d number o f experiments were conducted t o evaluate the l a s t two p o s s i b i l i t i e s . F i r s t , the take-up solvent f o r two t e s t s o l u t e s , anthracene and naphthalene, was cyclopentane (with s o l u t i o n c o n c e n t r a t i o n s s t i l l about 1 mg/mL and i n j e c t i o n a l i q u o t s s t i l l about 0.3 - 0.5 y L i n t o a 20 vL loop) instead o f 2-methoxye t h a n o l . The d i s t o r t i o n p e r s i s t e d even when a g e n t l e stream o f a i r was used a f t e r a l i q u o t d e p o s i t i o n i n the sampling loop t o blow o f f the take-up s o l v e n t . Furthermore, the onset o f the d i s t o r t i o n for naphthalene occurred a t a lower carbon dioxide d e n s i t y than for anthracene. When the column was changed t o a p a r t i a l l y c h e m i c a l l y deactivated column, H y p e r s i l SAS, where short a l k y l chains are bonded t o a c t i v e s i t e s , onsets o f peak d i s t o r t i o n f o r anthracene* and f o r naphthalene were s h i f t e d t o lower carbon d i o x i d e d e n s i t i e s (and s t i l l d i f f e r e n t from each other) than f o r the f u l l y a c t i v e H y p e r s i l SIL adsorption column. Only s l i g h t peak d i s t o r t i o n was observed f o r a H y p e r s i l ODS column a t even lower column midpoint d e n s i t i e s (0.33 g/mL with the back pressure a t 1170 p s i g , the FLOW s e t t i n g a t 1.90 mL/min, and 40°C; d i s t o r t i o n was observed f o r only anthracene and not naphthalene). In order t o explore the second proposed o r i g i n , the i n j e c t i o n procedure, the solvent c o n d i t i o n s were chosen e m p i r i c a l l y so that at a p a r t i c u l a r combination o f temperature, back pressure, and flow r a t e — w i t h that combination o f s e t t i n g s d e f i n i n g solvent power, or more c o r r e c t l y , a solvent power d i s t r i b u t i o n across the column, there was no peak d i s t o r t i o n . Raising the temperature by small increments o f 5 t o 10 °C could produce peak d i s t o r t i o n — presumably due t o the decrease i n solvent power by a decrease i n d e n s i t y . * * Increasing the back pressure s l i g h t l y or i n c r e a s i n g the flow rate ( f o r an increased midpoint d e n s i t y ) eliminated the

* Pout ^ °·35 g/mL, 60°C, and FLOW s e t t i n g =2.00 mL/min—giving P m i d ^ ^ 8 / — r e s u l t e d i n d i s t o r t i o n o f anthracene peak on H y p e r s i l SAS vs those values c i t e d f o r d i s t o r t i o n on the H y p e r s i l SIL. 0

0

m L

=

** Pmid °·53 g/mL—corresponding t o 60°C, a back pressure o f 1540 p s i g and a FLOW s e t t i n g o f 5.00 mL/min—gave no d i s t o r t i o n for anthracene; however, p ^ ^ = 0.43 given by 70°C and the same back pressure and FLOW s e t t i n g gave d i s t o r t i o n f o r anthracene on the f u l l y a c t i v e H y p e r s i l SIL.


11.

RANDALL


163


d i s t o r t i o n at the higher temperatures. This described behavior seems to the author to be more r e l a t e d to some fundamental i n t e r a c t i o n of the s o l u t e , the chromatographic column and the mobile phase than to an i n j e c t i o n procedure. The chemical d e s c r i p t i o n o f t h i s i n t e r a c t i o n i s s t i l l to be determined. I t appears that there e x i s t s some threshold solvent power (defined e i t h e r by the pure carbon dioxide d e n s i t y or the modifier i d e n t i t y and concentration i n a modifier/carbon dioxide mixture) at which the solvent can begin to compete s u c c e s s f u l l y with a p a r t i c u l a r s t a t i o n a r y phase f o r a p a r t i c u l a r s o l u t e . Whether t h i s i n v o l v e s a d e a c t i v a t i o n of a c t i v e s i t e s amenable to s p e c i f i c solute adsorption on the s i l i c a surface or a secondary solvent e f f e c t (43) where the mobile phase i n t e r a c t s with the s o l u t e as well as with the adsorption surface i s unknown. Conclusions

and Future

Experiments

SFC i s competitive with other chromatographies i n terms o f chromatographic e f f i c i e n c y and r e s o l u t i o n . The packed column SFC r e s u l t s should be a p p l i c a b l e to c a p i l l a r y column SFC. The r e s u l t s of these p r e l i m i n a r y experiments i n e s t a b l i s h i n g a modifier s e l e c t i o n framework are g r a t i f y i n g i n that dramatic d i f f e r e n c e s i n the chromatographic behavior were seen from vertex to vertex on the various modifier s e l e c t i v i t y t r i a n g l e s . However, there are many s t u d i e s to be undertaken to make the t r a n s i t i o n from p r e l i m i n a r y r e s u l t s to a coherent, u s e f u l framework f o r r o u t i n e modifier s e l e c t i o n : 1. f u r t h e r p r e l i m i n a r y surveys on various columns, 2. s u b s t i t u t i o n of other m o d i f i e r s f o r methylene c h l o r i d e and chloroform i f the p r e l i m i n a r y r e s u l t s f o r other column s t a t i o n a r y phases i n d i c a t e a continued l a c k o f s u i t a b i l i t y , 3. a s t a t i s t i c a l survey using a set o f t e s t s o l u t e s to y i e l d a mathematical r e l a t i o n s h i p that evaluates the importance o f each of the a v a i l a b l e p a r a m e t e r s — d e n s i t y , temperature, mobile phase s e l e c t i v i t y and c o m p o s i t i o n — f o r normal phase adsorption, normal phase p a r t i t i o n , and reversed phase p a r t i t i o n s t a t i o n a r y phases i n packed column s t u d i e s and a l s o for s e l e c t e d s t a t i o n a r y phases i n c a p i l l a r y column SFC, [At t h i s point the d e n s i t y parameter appears to be l e s s important i n adsorption systems than the mobile phase composition; t h i s i s not true i n reversed phase systems where d e n s i t y i s quite important.] 4. continued e v a l u a t i o n o f normal phase SFC compared to normal phase LC, 5. an in-depth study to determine the o v e r a l l p o l a r i t y o f carbon dioxide and i t s s e l e c t i v i t y f r a c t i o n s ,


164


and

ULTRAHIGH RESOLUTION CHROMATOGRAPHY 6. evaluation o f "modifying the m o d i f i e r s " by adding small q u a n t i t i e s o f s t r o n g l y a c i d i c or b a s i c components, b u f f e r s , i o n - p a i r i n g agents, and o p t i c a l l y a c t i v e agents to the major m o d i f i e r component.*

Carbon dioxide based chromatography, with a mobile phase o f carbon dioxide and mixtures o f carbon dioxide plus m o d i f i e r s ranging from s u b c r i t i c a l t o s u p e r c r i t i c a l s t a t e s , i s a young but e x c i t i n g area f o r study. I n v e s t i g a t o r s can choose from a wide range o f p o s s i b l e experiments: from fundamental physicochemical s t u d i e s t o semi-empirical frameworks t o a n a l y t i c a l methodsdevelopment f o r complex mixture s e p a r a t i o n . I t i s hoped that the p r e l i m i n a r y r e s u l t s described i n t h i s paper w i l l be u s e f u l t o others i n t h i s area and w i l l stimulate future work that w i l l r e s u l t i n routine o p t i m i z a t i o n schemes f o r SFC separations l i k e those p r e s e n t l y under development f o r LC separations. EXPERIMENTAL The c o n d i t i o n s f o r the experiments described i n t h i s paper were the f o l l o w i n g : Sample A l i q u o t — 0 . 5 yL o f mixtures o f a p p r o x i mately 1 mg/mL concentrations; Sample Take-Up Solvent—2-methoxy ethanol, cyclopentane; Temperature—60 °C; Back P r e s s u r e — 3 4 6 bar (5000 p s i g ) unless s p e c i f i c a l l y noted; Column—5 ym H y p e r s i l SIL, 10 cm long, 4.6 mm ID, Hewlett-Packard #79916 SI Opt.554 f o r a l l presented chromatograms, (5 ym H y p e r s i l SAS, 10 cm χ 4.6ramID; m a t e r i a l from Shandon Southern Products Limited, Cheshire, UK; packed in-house, and 5 ym H y p e r s i l ODS, 10 cm X 4.6 mm ID, HP #79916 OD Opt. 554 f o r s p e c i a l d i s c u s s i o n s ) ; Flow S e t t i n g — 4 mL/ min; Linear V e l o c i t y — 0 . 5 cm/s; Detector Wavelengths—254 nm ex cept f o r hormone d e t e c t i o n a t 240 nm; Mobile Phase—Pure C 0 , C 0 plus m o d i f i e r s ; M o d i f i e r Concentration-10% (molar) maximum, p r e mixed i n commercial gas c y l i n d e r supplied to pump "B" and

Ultrahigh Resolution Chromatography - American Chemical Society

Recommend Documents