Applications of Laser Fluorimetry to Microcolumn ... - ACS Publications

than eight orders of magnitude in concentration. ... 0097-6156/ 86/ 0297-0120$06.00/ 0 ... In order to achieve the full potential of microcolumn LC, i...
0 downloads 0 Views 1MB Size
8 Applications of Laser Fluorimetry to Microcolumn Liquid Chromatography V. L. McGuffin and R. N. Zare

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

1

Department of Chemistry, Stanford University, Stanford, CA 94305

Laser-induced fluorescence (LIF) is examined as a sensitive and selective detector for microcolumn liquid chromatography (LC). This detector employs a cw helium-cadmium laser (325 nm, 5-10 mW) as the excitation source together with a simple f i l t e r / photomultiplier optical system. Several flowcells are evaluated: (1) a flowing droplet; (2) an ensheathed effluent stream; and (3) a fused-silica capillary. The latter is judged to be the most promising for general use. The LIF detector is capable of sensing femtogram amounts of model solutes (coumarin dyes), and the response is linear over more than eight orders of magnitude in concentration. The potential of combining the high separation power of microcolumn LC with the high sensitivity of LIF detection is demonstrated through the analysis of multicomponent mixtures of polynuclear aromatic hydrocarbons and derivatized amino acids.

The i d e n t i f i c a t i o n and quantitation of individual components i n complex samples presents a challenging a n a l y t i c a l problem that demands continual improvement of existing methodology and i n s t r u ­ mentation. This demand has been a primary motivation for the development of high-efficiency separation methods, such as gas (GC), s u p e r c r i t i c a l f l u i d (SFC), and l i q u i d (LC) chromatographic techniques. In p a r t i c u l a r , microcolumn l i q u i d chromatography i s rapidly gaining popularity f o r the separation of multicomponent mixtures of nonvolatile compounds. The microcolumns currently under development are of three general types: narrow-bore or microbore packed columns (1-4), semipermeable packed c a p i l l a r i e s (5-7), and open tubular c a p i l l a r i e s (8-11). Although the t h e o r e t i c a l potential and the current state of the art are 1

Current address: Department of Chemistry, Michigan State University, East Lansing, M I 48824 0097-6156/ 86/ 0297-0120$06.00/ 0 © 1986 American Chemical Society

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

8.

McGUFFlN AND ZARE

Applications of Laser Fluorimetry

121

different for each column type, these microcolumns have demonstrated several d i s t i n c t advantages over conventional LC columns. F i r s t , microcolumns are capable of achieving much higher chromatographic e f f i c i e n c y than their conventional counterparts, exceeding one m i l l i o n theoretical plates (4,11). Consequently, microcolumns w i l l improve the separation of very complex samples or hard-to-resolve solutes. A l t e r n a t i v e l y , a given separation e f f i c i e n c y can be attained i n a shorter analysis time using semipermeable packed c a p i l l a r y or open tubular c a p i l l a r y columns because of t h e i r high permeability (12). Another desirable attribute of microcolumns i s their low volumetric flowrates (1-50 jiL/min), which result i n s i g n i f i c a n t l y reduced consumption of both sample and solvent. Although the economic and environmental advantages of reduced solvent consumption are immediately obvious, other benefits have become apparent with the continued development and use of microcolumn LC. For example, separations may be performed using novel mobile and stationary phases that are too expensive, rare, or toxic to be used with conventional columns (13). Furthermore, novel detection techniques can be implemented that are largely incompatible with conventional columns, among which flame- or plasma-based detectors (14-18) and mass spectrometry (19,20) are representative examples. Despite the many advantages of microcolumn LC, there are several limitations that may ultimately r e s t r i c t the p r a c t i c a l application of this technique. Long analysis times, ranging from a few to many hours, are frequently necessary to provide resolutions of very complex samples. This problem i s not c h a r a c t e r i s t i c of microcolumns themselves, but i s inherent i n a l l high-efficiency LC separations because of slow d i f f u s i o n a l processes i n the condensed phase. Another l i m i t a t i o n of microcolumn LC i s the stringent technological requirements imposed on a n c i l l a r y chromatographic equipment such as pumps, sample i n j e c t o r s , connecting hardware, and detectors. Although s i g n i f i c a n t improvements have been achieved for many components of the chromatographic system, there has been a conspicuous lack of sensitive, low-volume detectors. In most current applications of microcolumn LC, the detectors are merely miniaturized versions of those employed with conventional columns, such as UV-absorbance, fluorescence, and electrochemical detectors. Whereas such devices are adequate at the present stage of column development, the technological limitations and the consequences of further miniaturization become immediately apparent. In order to achieve the f u l l potential of microcolumn LC, i t i s necessary to develop new detection systems that are i n compliance with the rigorous requirements of this a n a l y t i c a l technique. Among the many p o s s i b i l i t i e s , laser-based spectroscopic detectors appear to be p a r t i c u l a r l y well suited for this application. The i n t e n s i t y of the l a s e r radiation permits very high s e n s i t i v i t y to be achieved using those spectroscopic techniques i n which the signal i s proportional to source i n t e n s i t y ; for example, fluorescence (21-23), phosphorescence (24-26), l i g h t scattering (27,28), and thermooptic or thermoacoustic measurements (29-31). Moreover, the highly collimated laser radiation can be readily focused into flowcells of nanoliter volume, as required f o r

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

122

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

microcolumn LC, without concomitant loss of radiant power. Other properties of laser sources, such as narrow s p e c t r a l bandwidth, p o l a r i z a t i o n , and temporal characteristics (pulsed or cw), have been favorably exploited f o r l i q u i d chromatographic detection as well (22). Laser-induced fluorescence (LIF) i s one of the simplest and most promising applications of laser-based detection i n l i q u i d chromatography. Previous investigations with conventional LC columns have c l e a r l y indicated the high s e n s i t i v i t y and s e l e c t i v i t y that can be attained (33-40). Based on these preliminary and highly promising results, several laboratories have undertaken concurrently the combination of microcolumn l i q u i d chromatography and laser fluorimetric detection (41-44). In the present study, we have developed a s e n s i t i v e low-volume LIF detector that i s compatible with a l l of the microcolumns described previously, yet i s simple and r e l i a b l e to operate. Several flowcells were constructed and evaluated, including a f u s e d - s i l i c a c a p i l l a r y (45), a miniaturized flowing droplet (33,46) and an ensheathed effluent stream (21,22,35). The LIF detector was optimized and characterized with respect to s e n s i t i v i t y , linear dynamic range, and dead volume. F i n a l l y , the outstanding performance of this detection system was demonstrated through the analysis of polynuclear aromatic hydrocarbon and derivatized amino acid samples. Experimental

Section

A schematic diagram of the l i q u i d chromatography system and laser fluorescence detector i s shown i n Figure 1. The key components of this a n a l y t i c a l system are described sequentially i n what follows: Liquid Chromatography System. The solvent delivery system was constructed of two 10-mL s t a i n l e s s - s t e e l syringe pumps (MPLC Micropump, Brownlee Labs, Santa Clara, CA). By s p l i t t i n g the pump effluent between the microcolumn and a r e s t r i c t i n g c a p i l l a r y (1:20-1:2000), i s o c r a t i c separations were achieved reproducibly at column flowrates as low as 0.005 yL/min, and gradient separations as low as 0.1 yL/min. Samples of 0.5 to 50 nL volume were introduced by the s p l i t i n j e c t i o n technique with a 1- L valve i n j e c t o r (Model ECI4W1., Valco Instruments Co., Inc., Houston, TX). The i n j e c t i o n valve, s p l i t t i n g tee, and microcolumn were maintained at constant temperature (±0.3°C) i n a thermostatted water bath. Three d i f f e r e n t types of microcolumns were prepared. Packed microcolumns were fabricated from f u s e d - s i l i c a tubing (HewlettPackard, Avondale, PA) of 0.20 to 0.32 mm inner diameter and 1 to 2 m length. This tubing was packed under moderate pressure (400 atm) with a s l u r r y of the chromatographic material [Aquapore RP-300 (10 ym), Brownlee Labs; Micro-Pak SP-18 (3 urn), Varian Instrument Group, Walnut Creek, CA] i n an appropriate solvent (3)· In this manner, narrow-bore packed microcolumns reproducibly yielded 150,000 or more theoretical plates. Semipermeable packed c a p i l l a r y columns were prepared by loosely packing a Pyrex glass tube with an i r r e g u l a r s i l i c a adsorbent [LiChrosorb Si-60 (30 μη),

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

8.

123

Applications of Laser Fluorimetry

McGUFFlN AND ZARE

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

THERMOSTATTED WATER BATH

He-Cd LASER 325 nm

UVABSORBANCE DETECTOR

€:ο:ΗίΓ:1] (a)

(b)

FUSED-SILICA FUSED-SILICA CAPILLARY FLOWCELL CAPILLARY FLOWCELL (Transverse Excitation) (Longitudinal Excitation)

PICOAMMETER

(O FLOWING DROPLET CELL

CHART RECORDER OR COMPUTER

(d) ENSHEATHED LAMINAR FLOWCELL

COLUMN PACKING

Figure 1. Schematic diagram of the l i q u i d chromatography system and laser fluorescence detector with d i f f e r e n t f l o w c e l l s : (a)-(d). I = i n j e c t i o n valve, Τ » s p l i t t i n g tee, M = metering valve or r e s t r i c t i n g c a p i l l a r y , L = lens, F = f i l t e r , A = aperture, PMT * photomultiplier tube.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

124

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

E . Merck Reagents, Darmstadt, F . R . G . ] , and subsequently extruding the packed c a p i l l a r y with a glass-drawing apparatus to 70 μη i . d . and 25 m length (7). Open tubular c a p i l l a r y columns were fabricated from f u s e d - s i l i c a tubing of 10 to 100 μη i . d . and 5 m length ( S c i e n t i f i c Glass Engineering, I n c . , Austin, TX), and were used without further surface modification. The l a t t e r microcolumns were employed s o l e l y to ascertain compatibility with the laser fluorescence detector and to optimize flowcell design. A variable-wavelength UV-absorbance detector (Model Uvidec 100-V, Jasco I n c . , Tokyo, Japan) was modified to permit "on-column" detection with packed and open tubular f u s e d - s i l i c a microcolumns, as described previously (3,45,47). The UV-absorbance detector was placed i n series before the laser fluorescence detector (see Figure 1), and was used for comparisons of s e n s i t i v i t y , s e l e c t i v i t y , and dead volume (44). Laser Fluorescence Detector. A helium-cadmium laser (Model 4240B, L i c o n i x , Sunnyvale, CA) was chosen as the excitation source because of i t s s t a b i l i t y and convenient wavelengths (325 and 442 nm). The UV laser radiation (325 nm, 5-10 mW cw) was isolated with a d i e l e c t r i c mirror and was focused on the miniaturized flowcell with a quartz lens. Sample fluorescence, collected perpendicular to and coplanar with the excitation beam, was spectrally isolated by appropriate interference f i l t e r s and then focused on a photomultiplier tube (Centronic Model Q 4249 B, Bailey Instruments C o . , I n c . , Saddle Brook, N J ) . The resulting photocurrent was amplified with a plcoammeter (Model 480, Keithley Instruments, I n c . , Cleveland, OH), and f i n a l l y was displayed on a stripchart recorder (Model 585, Linear Instruments Corp., Reno, NV). Several miniaturized flowcells were evaluated, including a f u s e d - s i l i c a c a p i l l a r y , a suspended flowing droplet, and an ensheathed effluent stream (Figures l a - d ) . The f u s e d - s i l i c a c a p i l l a r y flowcell was formed by removing the protective polyimide layer from a short section of f u s e d - s i l i c a tubing. When the LIF detector was used alone, the flowcell was simply an extension of the column i t s e l f (0.20 to 0.33 mm i . d . ) , thereby eliminating dead volume from connecting tubes and unions. However, when the UV-absorbance and LIF detectors were employed i n s e r i e s , i t was necessary to use a c a p i l l a r y of smaller diameter i n order to reduce laminar dispersion between the detectors. In such cases, a fused—silica c a p i l l a r y of 0.035—0.100 mm i . d . and 1 m length was attached to the microcolumn outlet with PTFE (Teflon) tubing, and formed the flowcells for both UV-absorbance and LIF detectors. By varying the inner diameter, flowcells were constructed with illuminated volumes from 1 to 100 nL, and corresponding o p t i c a l pathlengths of 0.035 to 0.33 mm. Illumination was achieved i n either the transverse (Figure la) or longitudinal (Figure lb) d i r e c t i o n with respect to the flowcell a x i s . The l a t t e r method, which employed a UV-transmitting o p t i c a l waveguide inserted d i r e c t l y into the c a p i l l a r y f l o w c e l l , was more e a s i l y aligned and permitted control of the illuminated volume without requiring a change i n c a p i l l a r y diameter. The flowing droplet c e l l , i l l u s t r a t e d i n Figure l c , was a straightforward miniaturization of the windowless fluorescence flowcell described for conventional

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

8.

McGUFFlN AND ZARE

125

Applications of Laser Fluorimetry

l i q u i d chromatography by Diebold and Zare (33). This c e l l was formed by suspending a flowing droplet of the microcolumn effluent between a f u s e d - s i l i c a c a p i l l a r y and a quartz rod of similar outer diameter. By varying the diameter and spacing of the c a p i l l a r y and rod, the droplet volume could be adjusted from approximately 50 to 200 nL, with corresponding o p t i c a l pathlengths of 0.15 to 0.5 mm. The ensheathed effluent f l o w c e l l , shown schematically i n Figure Id, was both s t r u c t u r a l l y and functionally s i m i l a r to those described by Hershberger et a l . (35) and Dovichi et_ a l . (21,22). This f l o w c e l l was obtained from a commercial flow cytometer (System 50, Ortho Diagnostic Systems, Inc., Westwood, MA), and consisted of a quartz tube of square cross section having a 0.25 mm square central bore. The effluent was introduced d i r e c t l y from the f u s e d - s i l i c a microcolumn, which extended just beyond a conical s t a i n l e s s - s t e e l j e t at the bottom of the quartz tube. The sheath solvent, which was of similar composition to the column effluent, was supplied by a gas-pressurized l i q u i d reservoir through two s t a i n l e s s - s t e e l tubes perpendicular to and s l i g h t l y below the column i n l e t . Turbulence and mixing of the effluent and sheath streams were not observed when pulseless, laminar flow conditions were maintained. The o p t i c a l pathlength and e f f e c t i v e volume of the effluent stream were controlled by hydrodynamic focusing, by varying the r e l a t i v e flowrates of the column effluent and the sheath solvent. In this manner, the pathlength was readily varied between 5 and 20 μη, with corresponding illuminated volumes of 20 to 300 pL (calculated values). The large volume of sheath solvent employed, t y p i c a l l y 0.5 to 5.0 mL/min, eliminated the advantage gained by microcolumns i n reducing solvent consumption. A n a l y t i c a l Methodology. An N.B.S. Standard Reference Material (SRM 1647) containing sixteen polynuclear aromatic hydrocarbons was diluted twenty-fold with methanol/methylene chloride (1:1) p r i o r to analysis. Amino acids were derivatized with 1-dimethylaminonaphthalene5-sulfonyl chloride (dansyl chloride) according to the procedure of Tapuhi and coworkers (48-50). A 10"" M stock solution of twenty common L-amino acids (Sigma Chemical Co., St. Louis, MO) was prepared by dissolving the c a r e f u l l y weighed standards i n 0.1 M aqueous hydrochloric acid. Aliquots of this solution were transferred to conical v i a l s , evaporated to dryness, and redissolved i n 500 \JL aqueous buffer (0.04 M lithium carbonate, pH 9.5). A 500 pL volume of dansyl chloride solution (5 x 10"" M i n a c e t o n i t r i l e ) was added, and the derivatization was allowed to proceed i n the dark at 35°C f o r one hour. The reaction was terminated by the addition of 2% methylamine hydrochloride, and the derivatized sample was analyzed immediately by microcolumn l i q u i d chromatography with UV-absorbance and LIF detection. Organic solvents employed i n t h i s investigation were high-purity, d i s t i l l e d - i n - g l a s s grade (Burdick & Jackson Laboratories, Inc., Muskegon, MI); water was deionized and doubly d i s t i l l e d i n glass (Mega-Pure System, Corning Glass Works, Corning, NY). 3

3

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

126

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

Results and Discussion Evaluation of Flowcell Performance. There are many important factors to be considered i n the evaluation of f l o w c e l l performance for microcolumn l i q u i d chromatography. Foremost among such c r i t e r i a i s s e n s i t i v i t y ; i . e . , the signal-to-noise r a t i o that can be obtained for a standard sample i n j e c t i o n . I f we assume that the flowcells are uniformly illuminated i n the laser beam and that the fluorescent emission i s collected from each c e l l with the same e f f i c i e n c y , then the s i g n a l i n t e n s i t y should be primarily a function of the o p t i c a l pathlength of each f l o w c e l l . Accordingly, the flowing droplet c e l l , which had the longest pathlength (0.15-0.5 mm), provided the largest photocurrent, while the f u s e d - s i l i c a c a p i l l a r y (0.05-0.32 mm) and the ensheathed effluent (0.005-0.02 mm) flowcells yielded proportionately smaller responses. However, the background noise l e v e l also appeared to be nearly proportional to pathlength, indicating that the predominant source of noise was fluorescent or scattered (Rayleigh and Raman) l i g h t from the effluent i t s e l f , rather than from c e l l walls. S p e c i f i c a l l y , Raman scattering from the solvent seemed to be the major component of background interference, p a r t i c u l a r l y when the detected emission wavelength was near the e x c i t a t i o n wavelength. Thus, while the flowing droplet and ensheathed effluent flowcells were able to eliminate or discriminate against l i g h t o r i g i n a t i n g from c e l l walls through s p a t i a l f i l t e r i n g , this did not translate d i r e c t l y to an improvement i n signal-to-noise r a t i o . Indeed, t h i s figure of merit was approximately the same for each of the flowcells investigated when appropriate spectral f i l t e r i n g was employed. Consequently, other performance c r i t e r i a , such as dead volume, s i m p l i c i t y of operation, and other factors dictated the preference i n f l o w c e l l design. It i s imperative that the volume of the f l o w c e l l and associated connections does not contribute excessively to the dispersion of the chromatographic peaks. Moreover, i t i s desirable to have a single flowcell design that w i l l accommodate microcolumns of d i f f e r e n t types and s i z e s , and w i l l adapt to t h e i r different volumetric requirements. The flowing droplet c e l l , which could be varied from 50 to 200 nL, was found to be suitable f o r packed microcolumns of 0.5 mm or greater inner diameter, but to suffer from reduced s e n s i t i v i t y and s t a b i l i t y f o r columns of smaller bore. On the other hand, the very small volume of the ensheathed effluent f l o w c e l l (20-300 pL) made i t suitable f o r open tubular and semipermeable packed c a p i l l a r y columns, but somewhat impractical for columns of larger bore because of reduced s e n s i t i v i t y and increased consumption of ensheathing solvent. In contrast, f u s e d - s i l i c a c a p i l l a r y c e l l s of varying dimensions could be r e a d i l y interchanged to provide a suitable compromise between s e n s i t i v i t y and dead volume f o r a l l of the microcolumns under study. Another important consideration i s the compatibility of the f l o w c e l l with the normal range of operating conditions i n l i q u i d chromatography. F i r s t , the flowcell should be compatible with a wide variety of solvents f o r both normal- and reversed-phase separations. Second, i t should readily accommodate variations i n solvent flowrate to permit both h i g h - e f f i c i e n c y and high-speed

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

8.

McGUFFIN AND ZARE

Applications of Laser Fluorimetry

127

applications. Furthermore, for optimal analysis of complex samples, the f l o w c e l l should allow gradients i n either solvent composition or flowrate to be employed. F i n a l l y , i t i s desirable for the detector c e l l to be i n s e n s i t i v e to temperature and pressure fluctuations i n order to minimize background noise. Although our investigations encompassed a wide range of operating conditions, the results can be b r i e f l y summarized i n the following manner: For a fixed set of operating conditions (solvent composition, flowrate, temperature, pressure), the flowing droplet and ensheathed effluent flowcells could be r e a d i l y optimized. However, because these c e l l s have f l e x i b l e boundaries i n the o p t i c a l region, they were strongly dependent upon the physical properties of the effluent and could not readily tolerate changes i n operating conditions without reoptimization. Hence, these c e l l s were not compatible with gradients i n either solvent composition or flowrate. Moreover, both the flowing droplet and ensheathed effluent flowcells were sensitive to pressure and flow fluctuations from the pump, mechanical v i b r a t i o n , and bubble formation. Furthermore, these c e l l s were e s s e n t i a l l y destructive; i . e . , they caused d i l u t i o n or mixing of the column effluent such that another detector could not r e a d i l y be employed i n series. In contrast, the f u s e d - s i l i c a c a p i l l a r y c e l l exhibited only nominal dependence upon the chromatographic conditions. Because of i t s operational s i m p l i c i t y , compatibility with many solvents as well as gradient elution, compatibility with a wide range of microcolumn types and s i z e s , and nondestructive nature, the f u s e d - s i l i c a c a p i l l a r y flowcell was judged to be the most v e r s a t i l e and useful among those investigated. Characterization of LIF Detector. In a previous study (44), the LIF detector was characterized with the f u s e d - s i l i c a f l o w c e l l using transverse e x c i t a t i o n . The l i m i t of detection, measured at a signal-to-noise r a t i o of seven (99.9% confidence l e v e l ) for three r e p l i c a t e measurements, was determined to be 2.3 x 1 0 " ^ g of coumarin 440 (1.3 x 10"*^ moles) injected i n a 25 nL sample. Whereas detection l i m i t s as much as an order of magnitude lower have been reported for laser fluorimetry (22), such results were obtained under rather rigorous conditions. In contrast, the detection l i m i t demonstrated i n the previous study (44) was achieved with a r e l a t i v e l y low-power laser and simple o p t i c a l system using a representative solute analyzed under normal chromatographic conditions. Hence, t h i s high s e n s i t i v i t y can be routinely achieved under p r a c t i c a l operating conditions. The fluorescence s i g n a l was found to be l i n e a r l y related to solute concentration over at least eight orders of magnitude, extending from the detection l i m i t (5 x 1 0 " M) to the s o l u b i l i t y l i m i t (2 x 10~ M) of coumarin 440 i n methanol. I t was suggested that the small diameter of the c a p i l l a r y f l o w c e l l e f f e c t i v e l y reduced nonlinear behavior i n both the absorption and emission processes, thereby reducing both inner and outer f i l t e r effects and extending the linear response into very high concentrations (44). This extraordinary l i n e a r range f a c i l i t a t e s the simultaneous quantitation of both major and minor components i n complex sample matrices. 10

2

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

128

There are several important sources of band broadening i n the LIF detector, including both volumetric and temporal contributions. These contributions were assessed independently i n order to e s t a b l i s h the major sources of dispersion and to minimize t h e i r effects (44). The volumetric variance, given by the second s t a t i s t i c a l moment of the chromatographic peak, was determined to be from 0.06 nL to 0.06 pL for c a p i l l a r y flowcells of 0.035 to 0.33 mm diameter, respectively. This variance was predominantly due to laminar dispersion occurring between the microcolumn f r i t and the illuminated region (see Figure l a ) . The temporal dispersion, given by an exponential time constant of 0.7 ms, was equivalent to 0.0001 n L i n variance u n i t s . Thus, both volumetric and temporal variances were s u f f i c i e n t l y small for almost any application i n microcolumn l i q u i d chromatography, including both high-speed and high-efficiency separations.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

2

2

2

Applications. Many molecules of environmental and biochemical significance are inherently fluorescent, among which the polynuclear aromatic hydrocarbons (PAH) are a representative example. Although such compounds may occur i n very complex sample matrices, they can be e f f i c i e n t l y separated by microcolumn LC and s e n s i t i v e l y detected by laser fluorimetry. An exemplary chromatogram i s shown i n Figure 2, wherein a Standard Reference Material (SRM 1647) containing sixteen polynuclear aromatic hydrocarbons has been analyzed. This separation was achieved on a f u s e d - s i l i c a microcolumn (0.2 mm i . d . , 1.3 m length) containing a 3-pm o c t a d e c y l s i l i c a packing material (N=150,000) using an i s o c r a t i c mobile phase of 92.5% aqueous a c e t o n i t r i l e . For comparison of s e n s i t i v i t y and s e l e c t i v i t y , the chromatograms obtained with UV-absorbance and laser-induced fluorescence detection are i l l u s t r a t e d i n Figure 2. Several important features of LIF detection become apparent: Whenever fluorescence detection i s applicable, the s e n s i t i v i t y i s generally far superior to UV-absorbance and other common detection methods, frequently by several orders of magnitude. However, not a l l solutes of interest are n a t u r a l l y fluorescent, and very s i m i l a r molecules may have widely d i f f e r i n g absorption c o e f f i c i e n t s and quantum y i e l d s . This becomes evident when comparing the r e l a t i v e response of benzo(g,h,i)perylene and indeno(l,2,3-cd)pyrene, two s i x - r i n g polynuclear aromatic compounds present i n equimolar amounts i n the Standard Reference Material. In this case, fluorescence of the alternate PAH benzoperylene i s s i g n i f i c a n t l y lower than that of the non-alternate indenopyrene due to differences i n the absorption c o e f f i c i e n t (51) and to quenching by the solvent a c e t o n i t r i l e . This high degree of s e l e c t i v i t y can be advantageous for the s i m p l i f i c a t i o n of complex chromatograms and f o r the reduction of background interferences. Because most common UV lasers operate at fixed wavelengths, the range of application of laser-induced fluorescence may be somewhat limited. The He-Cd laser provides additional v e r s a t i l i t y because two excitation wavelengths are available, 325 nm and 442 nm. This feature i s i l l u s t r a t e d i n Figure 3, taken from the recent work of Guthrie, Jorgenson, and Dluzneski (41), where a solvent-refined coal f l u i d has been analyzed using an open tubular

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Figure 2. Chromatogram of polynuclear aromatic hydrocarbon standards. Column: f u s e d - s i l i c a c a p i l l a r y (0.2 mm i . d . , 1.3 m length) packed with Micro-Pak SP-18 (3pm); Mobile phase: 92.5% aqueous a c e t o n i t r i l e at 1.2 pL/min; Sample i n j e c t i o n : 50 nL, 0.2-2.5 ng per component. Reproduced with permission from Ref. 43. Copyright 1984, American Assoc. for the Advancement of Science.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

2

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

130

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

LASER

FLUORESCENCE

λ = 325 nm β χ

λ e m >380 nm

λ = 4 4 2 nm ΘΧ

λ e m >470 nm

10

20

15

25

mm. Figure 3. Characterization of a solvent-refined coal sample by laser fluorimetric detection with dual excitation wavelengths. Column: open tubular glass c a p i l l a r y (16 pm i . d . , 2.3 m length) with octadecylsilane bonded phase; Mobile

phase:

permission

50% a q u e o u s

from

Ref. 41.

acetonitrile Copyright

,

425 p s i . R e p r o d u c e d

1984,

Preston

with

Publications.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

8.

McGUFFlN AND ZARE

Applications of Laser Fluorimetry

131

c a p i l l a r y column with a chemically bonded octadecylsilane stationary phase. I t i s evident that UV e x c i t a t i o n at 325 nm (0.30 mW) i s appropriate for the detection of small polynuclear aromatic hydrocarbons, while v i s i b l e e x c i t a t i o n at 442 nm (4.1 mW) permits greater s e n s i t i v i t y for PAH compounds of higher molecular weight. The complementary information contained i n these chromatograms provides a powerful a n a l y t i c a l tool for the characterization of complex coal-derived samples. Molecules that are not amenable to LIF detection because of their low absorption cross sections or low fluorescence quantum yields may be analyzed through a variety of indirect fluorescence techniques. For example, non-fluorescent solutes may be detected by the extent to which they enhance (52) or quench (53) the emission of a fluorescent dye added to the mobile phase. A l t e r n a t i v e l y , simple displacement of the dye molecule i n the column effluent by a non-fluorescent solute w i l l result i n a reduction of the fluorescence i n t e n s i t y , which can provide nearly universal detection c a p a b i l i t y with adequate s e n s i t i v i t y for many applications (54). F i n a l l y , non-fluorescent molecules may be rendered detectable through the incorporation of a fluorescent l a b e l that i s optimized for the laser excitation wavelength. Derivatization methods are p a r t i c u l a r l y a t t r a c t i v e because they introduce an additional dimension of chemical and spectroscopic s e l e c t i v i t y that can simplify analyses i n a predictable and reproducible manner. A wide variety of fluorescent molecular probes have been demonstrated to be suitable for excitation by the He-Cd laser: 4-bromomethyl-7-methoxycoumarin has been employed for the detection of carboxylic and phosphoric acids (44,55), 7-chlorocarbonylmethoxy-4-methylcoumarin f o r hydroxyl compounds (42), 7-isothiocyanato-4-methylcoumarin for amines and amino acids (55), 7-diazo-4methyl-coumarin f o r a variety of aromatic compounds (55), and terbium chelate molecules with long fluorescence lifetimes (~ 1 ms) for protein analysis (56). In t h i s study, we examined the u t i l i t y of l-dimethylaminonaphthalene-5-sulfonyl chloride (dansyl chloride) as a s e n s i t i v e and s e l e c t i v e reagent f o r the determination of biogenic amines and amino acids. The use of dansyl chloride as a d e r i v a t i z a t i o n reagent f o r amino acids and peptides i s a well established and well accepted methodology. Nevertheless, improvements i n the separation and detection of these derivatives may have a s i g n i f i c a n t impact on c l i n i c a l and biomedical research. The fluorescence e x c i t a t i o n and emission spectra of the dansyl amino acids are shown i n Figure 4, whereupon the l a s e r excitation wavelength (325 nm) and f i l t e r emission wavelength (546 nm, 10 nm FWHM) have been indicated. I t i s evident that t h i s fluorescent l a b e l i s not excited optimally by the He-Cd laser, yet there appears to be s u f f i c i e n t overlap to permit sensitive detection. The analysis of twenty common amino acids derivatives (50-100 pmol each) i s demonstrated i n Figure 5. This separation was achieved using a f u s e d - s i l i c a microcolumn containing Aquapore RP-300, a wide-pore o c t y l s i l i c a packing material of 10-pm nominal diameter. The mobile phase consisted of an optimized gradient from 5% to 32% 2-propanol i n a formic and a c e t i c acid buffer (pH 3.2). The chromatographic separation shown

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

I

ι

200

I

300

ι

I

ι

400 WAVELENGTH

I

500

ι

1

1

600

L

700

(nm)

Figure 4. Fluorescence excitation and emission spectra of dansyl amino acids.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

Figure 5. Chromâtogram of dansyl amino acid standards. Column: f u s e d - s i l i c a c a p i l l a r y (0.32 mm i . d . , 1.3 m length) packed with Aquapore RP-300 (10 pm); Mobile phase: l i n e a r gradient from 5% to 32% 2-propanol i n 0.05 M formic acid/0.06 M acetic acid buffer (pH 3.2) i n 200 min, 2.0 pL/min, 55°C; Sample i n j e c t i o n : 130 nL, 50-100 pmol per component.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

134

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

here required four hours for completion when run at optimum flowrate (2.0 pL/min), yet i t c l e a r l y reveals the high resolving power of microcolumn LC. I t i s possible and frequently desirable to compromise separation e f f i c i e n c y to achieve faster analysis time; a s l i g h t l y i n f e r i o r chromâtogram, s t i l l with baseline resolution, can be obtained i n less than two hours. For purposes of comparison, the separation was monitored using both UV-absorbance and laser fluorescence detection. The UV-absorbance chromatogram (Figure 5, upper trace) shows r e l a t i v e l y poor s e n s i t i v i t y f o r the dansyl amino acids, even at the optimum absorbance wavelength. Because this detector i s highly sensitive to changes i n absorbance and r e f r a c t i v e index of the mobile phase during gradient elution, the small peaks of interest are superimposed upon a steeply sloping baseline. In contrast, the fluorescence chromatogram (lower trace) demonstrates a superior signal-to-noise r a t i o with the peaks of interest superimposed upon an absolutely f l a t baseline. The high s e n s i t i v i t y of the LIF detector permits the determination of a l l dansyl amino acids, as well as some impurities therein, at the subpicomole l e v e l . Summary The development of suitable detectors i s currently one of the l i m i t i n g factors i n the routine p r a c t i c a l application of microcolumn l i q u i d chromatography. The laser-induced fluorescence detector described herein appears to f u l f i l l the rigorous requirements because of i t s high s e n s i t i v i t y (a few femtograms of analyte detected), remarkable l i n e a r i t y (greater than 10° dynamic range), and small detection volume (0.06 nL to 0.06 pL for flowcells of 35 to 330 μπι i . d . , respectively). Although many improvements are possible, such refinements would involve a s a c r i f i c e i n the inherent s i m p l i c i t y , r e l i a b i l i t y , and a f f o r d a b i l i t y of the present detection system. This system can already be applied to a wide variety of interesting a n a l y t i c a l problems, of which the separations of polynuclear aromatic hydrocarbons and dansyl amino acids are only representative examples. I t would appear that the combination of laser-induced fluorescence detection and microcolumn l i q u i d chromatography shows exceptional promise for the analysis of complex samples of biochemical and environmental o r i g i n . 2

2

Acknowledgment s We are g r a t e f u l to Brownlee Labs, Santa Clara, CA, f o r the loan of a prototype MPLC Micropump solvent delivery system. We also thank Raymond Dandeneau of Hewlett-Packard and Ernest Dawes of S c i e n t i f i c Glass Engineering for providing f u s e d - s i l i c a c a p i l l a r y tubing of non-standard dimensions, and Bojan Petek of the San Francisco Laser Center for advice on the use of o p t i c a l f i b e r s . This research was supported by the National Institutes of Health under grant number 9R01 GM29276-06. R. Ν. Z. g r a t e f u l l y acknowledges support through the Shell Distinguished Chairs Program, funded by the Shell Companies Foundation, Inc.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

8.

McGUFFIN AND ZARE

Applications

of Laser Fluorimetry

135

Literature Cited 1. 2. 3. 4.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27. 28. 29. 30. 31. 32. 33. 34.

Scott, R. P. W.; Kucera, P. J . Chromatogr. 1979, 169, 51-72. Yang, F. J . J . Chromatogr. 1982, 236, 265-77. Gluckman, J . C.; Hirose, Α.; McGuffin, V. L.; Novotny, M. Chromatographia 1983, 17, 303-9. Menet, H. G.; G a r e i l , P. C.; Rosset, R. H. Anal. Chem. 1984, 56, 1770-3. Tsuda, T.; Novotny, M. Anal. Chem. 1978, 50, 271-5. Hirata, Y.; Novotny, M.; Tsuda, T.; I s h i i , D. Anal. Chem. 1979, 51, 1807-9. McGuffin, V. L.; Novotny, M. J . Chromatogr. 1983, 255, 381-93. Tsuda, T.; H i b i , K.; Nakanishi, T.; Takeuchi, T.; I s h i i , D. J. Chromatogr. 1978, 158, 227-32. Tsuda, T.; Nakagawa, G. J . Chromatogr. 1983, 268, 369-74. Jorgenson, J . W.; Guthrie, E. J . J . Chromatogr. 1983, 255, 335-48. Kucera, P.; Guiochon, G. J . Chromatogr. 1984, 283, 1-20. Knox, J . H. J . Chromatogr. S c i . 1980, 18, 453-61. McGuffin, V. L. L i q . Chromatogr. Mag. 1984, 2, 282-8. K r e j c i , M.; Tesarik, K.; Rusek, M.; Pajurek, J . J . Chromatogr. 1981, 218, 167-78. McGuffin, V. L.; Novotny, M. Anal. Chem. 1981, 53, 946-51. McGuffin, V. L.; Novotny, M. Anal. Chem. 1983, 55, 2296-302. Jinno, K.; Tsuchida, H. Anal. L e t t . 1982, 15(A5), 427-37. Jinno, K.; Nakanishi, T. J . High Resoln. Chromatogr. & Chromatogr. Commun. 1983, 6, 210-1. Henion, J . D. J . Chromatogr. S c i . 1981, 19, 57-64. Schafer, K. H.; Levson, K. J . Chromatogr. 1981, 206, 245-52. Dovichi, N. J . ; Martin, J . C.; Jett, J . H.; Trkula, M.; K e l l e r , R. A. Anal. Chem. 1984, 56, 348-54. Dovichi, N. J . ; Martin, J . C.; Jett, J . H.; K e l l e r , R. A. Science 1983, 219, 845-7. Richardson, J . W.; Ando, M. E. Anal. Chem. 1977, 49, 955-9. Cline Love, L. J . ; S k r i l e c , M.; Habarta, J . G. Anal. Chem. 1980, 52, 754-9. Donkerbroek, J . J . ; van Eikema Hommes, Ν. J . R.; Gooijer, C.; Velthorst, N. H.; F r e i , R. W. J . Chromatogr. 1983, 255, 581-90. Yamada, S.; Miyoshi, F.; Kano, K.; Ogawa, T. Anal. Chim. Acta 1981, 127, 195-8. Jorgenson, J . W.; Smith, S. L.; Novotny, M. J . Chromatogr. 1977, 142, 233-40. Rogers, L. B.; Stuart, J . D.; Goss, L. P.; Malloy, T. B.; C a r r e i r a , L. A. Anal. Chem. 1977, 49, 959-62. Oda, S.; Sawada, T. Anal. Chem. 1981, 53, 471-4. Leach, R. Α.; Harris, J . M. J . Chromatogr. 1981, 218, 15-9. P e l l e t i e r , M. J . ; Thorsheim, H. R.; Harris, J . M. Anal. Chem. 1982, 54, 239-42. Yeung, E. S. Adv. Chromatogr. 1984, 23, 1-63. Diebold, G. J . ; Zare, R. N. Science 1977, 196, 1439-41. Folestad, S.; Johnson, L.; Joseffson, B.; Galle, B. Anal. Chem. 1982, 54, 925-9.

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

C H R O M A T O G R A P H Y A N D SEPARATION CHEMISTRY

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on December 24, 2015 | http://pubs.acs.org Publication Date: January 27, 1986 | doi: 10.1021/bk-1986-0297.ch008

136

35. Hershberger, L. W.; Callis, J. B.; Christian, G. D. Anal. Chem. 1979, 51, 1444-6. 36. Furuta, N.; Otsuki, A. Anal. Chem. 1983, 55, 2407-13. 37. Sepaniak, M. J.; Yeung, E. S. Anal. Chem. 1977, 49, 1554-6. 38. Huff, P. B.; Tromberg, B. J.; Sepaniak, M. J. Anal. Chem. 1982, 54, 946-50. 39. Yeung, E. S.; Sepaniak, M. J. Anal. Chem. 1980, 52, 1465A-81A. 40. Green, R. B. Anal. Chem. 1983, 55, 20A-32A. 41. Guthrie, E. J.; Jorgenson, J. W.; Dluzneski, P. R. J. Chromatogr. Sci. 1984, 22, 171-176. 42. Gluckman, J.; Shelly, D.; Novotny, M. J. Chromatogr. in press. 43. Zare, R. N. Science 1984, 226, 298-303. 44. McGuffin, V. L.; Zare, R. N. Appl. Spectrosc., in press. 45. Yang, F. J. J. High Resoln. Chromatogr. & Chromatogr. Commun. 1981, 4, 83-5. 46. Voigtman, E.; Jurgensen, Α.; Winefordner, J. D. Anal. Chem. 1981, 53, 1921-3. 47. McGuffin, V. L.; Novotny, M. Anal. Chem. 1983, 55, 580-3. 48. Tapuhi, Y.; Miller, N.; Karger, B. L. J. Chromatogr. 1981, 205, 325-37. 49. Tapuhi, Y.; Schmidt, D. E.; Lindner, W.; Karger, B. L. Anal. Biochem. 1981, 115, 123-9. 50. De Jong, C.; Hughes, G. J.; van Wieringen, E.; Wilson, K. J. J. Chromatogr. 1982, 241, 345-59. 51. Peaden, P. Α.; Lee, M. L.; Hirata, Y.; Novotny, M. Anal. Chem. 1980, 52, 2268-71. 52. Asmus, P. Α.; Jorgenson, J. W.; Novotny, M. J. Chromatogr. 1976, 126, 317-25. 53. Haugen, G. R.; Richardson, J. H.; Clarkson, J. E.; Hieftje, G. M. Proceedings of New Concepts Symposium and Workshop on Detection and Identification of Explosives, Reston, VA, 1978. 54. Mho, S.; Yeung, E. S. Iowa State University, Ames, IA, private communication. 55. Gassmann, E.; Kuo, J. E.; McGuffin, V. L.; Zare, R. N. Stanford University, Stanford, CA, unpublished research. 56. Kuo, J. E.; Milby, K. H.; Hinsberg, W. D.; Poole, P. R.; McGuffin, V. L.; Zare, R. N. Clin. Chem. 1985, 31, 50-3. RECEIVED October 4, 1985

In Chromatography and Separation Chemistry; Ahuja, Satinder; ACS Symposium Series; American Chemical Society: Washington, DC, 1986.