Anal. Chem. 1994,66, 3639-3643
Universal Spectropolarimeter Based on Overtone Circular Dichroism Measurements in the Near-Infrared Region Chieu D. Tran’ and Victor I. Grishko Department of Chemistry, Marquefte University, Milwaukee, Wisconsin 53233
A novel and highly sensitive circular dichroism spectropolarimeter for the near-infrared region has been developed. In this instrument, a solid state titanium-sapphire laser that can be spectrally tuned from 670 to 1030 nm was used as the light source. The laser beam was converted into linearly polarized light by a polarizer and into left circularly polarized light and right circularly polarized light at 42 kHz by a phstoelastic modulator (PEM). A limit of detection of 1.1 X AU was achieved by this instrument for (+)-Co(en)s3+ at 765 nm. Further improvement including employing double modulation (at 42 kHz by the PEM and at 85 Hz by a chopper just before the laser beam was converted to CPL), demodulating and amplifyingthe signal with high-performancelock-in amplifiers, was made to enable the instrument to have the required sensitivity for the measurements of the circular dichroism of overtones and combination transitions of saturated chiral compounds, e.g., (R)-and (@-camphor,(R)-and (@-2-octanol, and ( R ) -and (S)-2-amino-l-octanol. Because the measured CD spectra originate from the overtones and combination transitionsof the C-H and 0 - H groups, the spectropolarimeter can be used to detect virtually any compounds that have 0-H and/or C-H groups.
Laser, with its unique properties, namely, its high output power, better beam collimating and focusing, higher degree of spectral and polarization purity, enables the development of sensitive spectropolarimeters which would not be feasible with other light sources. In fact, M I A values as low as 1od have been measured by use of these newly developed laser-based ~pectropolarimeters.~”-~3 The reported laser-based spectropolarimetershave proven to be effective but only for color chiral compounds because these instruments are based on measurements of the CD in the visible region. Many attempts have been made in the last few years to extend the developed instruments to the shortwavelength UV region. Unfortunately, to date, it has proven to be very difficult, if not impossible, to achieve this goal. A variety of reasons are responsible for this shortcoming including the high degree of instability of the laser in the shortwavelength UV region (as compared to the visible region) and the fact that the transparency and anisotropic properties of optical components in the visible are much different from those in the UV region. Moreover, it is important to point out that a laser-based UV spectropolarimeter, if it were developed, would not be a universal instrument because certain types of chiral compounds, including saturated chiral compounds, do not absorb in the UV region. An instrument based on the circular dichroism measurement in the infrared and/or near-infrared regions can, in principle, be an universal spectropolarimeter because these regions cover the fundamentals, overtones, and combination absorption of C-H, 0-H, and N-H groups, which are present in all types of organic molecules. A vibrational CD spectropolarimeter has, in fact, been developed but its application to analytical chemistry is relatively limited because of the difficulties associated with the transparency of optical components and sample and also because of the requirement for sample Overtones circular dichroism does not suffer from these limitations and, hence, should be a universal chiral analysis technique. The overtone CD has,
Very often, only one form of enantiomers is pharmaceutical active and has therapeutic The other or others can reverse or otherwise limit the effect of the desired enanti~ m e r . ’ -It~ is thus hardly surprising that chirooptical analysis has increasingly become important in recent years.68 Circular dichroism (CD) is generally preferred to the optical rotatory dispersion (ORD) because the former provides the direct chirooptical information on the chromophore. Traditionally, the CD measurements are considered to be difficult. This is because the signal is not only very small (only about of an absorbance unit9 ) but is also difficult to measure (small ac signal riding on top a large dc signal). In fact, the minimum detectable value as determined by commercially available spectropolarimeters is about IO4 of an absorbance unit.lq2 Efforts in the last few years have been centered on the (10) Synovec, R. E.; Yeung, E. S.Anal. Chem. 1985,57, 2606-2610. development of novel and ultrasensitive circular dichroism (11) Synovec, R. E.; Yeung, E. S.J. Chromatogr. 1986, 368, 85-93. (12) Chan, K. C.; Yeung, E. S . J. Chromatogr. 1989, 457, 421-426. spectropolarimeters. The most notable advance is perhaps (13) Rosenzweig, Z.; Yeung, E. S. Appl. Specfrosc. 1993, 47, 2017-2021. the development of laser-based spectr~polarimeters.~~~~(14) Nafie, L. A.; Keiderling, T. A.; Stephens, P. J. J . Am. Chem. Soc. 1976,98, ~~
(1) Chem. Eng. News 1990, 68 (19), 38-44; 1992, 70 (28), 4 6 7 9 . (2) Tran, C. D.; Dotlich, M. J . Chem. Educ., in press. (3) Armstrong, D. W.; Han, S.M. CRC Crit. Rev. Anal. Chem. 1988, 19, 175224. (4) Hinze, W. L. Sep. Pur$ Methods 1981, 10, 159-237. (5) Armstrong, D. W. Anal. Chem. 1987, 59, 84A-91A. (6) Purdie, N. Prog. Anal. Spectrosc. 1987, 10, 345-358. (7) Purdie, N.; Swallows, K. A. Anal. Chem. 1989, 61, 77A-85A. (8) Bobbitt, D. R. In Analytical Applications of Circular Dichroism; Purdie, N., Brittain, H. G., Eds.; Elsevier: New York, 1994; Chapter 1, pp 15-52. (9) Rau, H. Chem. Rev. 1983, 83, 535-547. 0003-2700/94/0366-3639$04.50/0 0 1994 American Chemical Society
2715-2723. (15) Diem, M.; Roberts, G. M.; Lee, 0.; Barlow, A. Appl. Spectrosc. 1988, 4.7, 20-27. (16) Osborne, G. A.; Cheng, J. C.; Stephens, P. J . Rev. Sci. Instrum. 1973, 44, 1C-15. (17) Freedman, T. B.; Nafie, L. A. In Merhods in Enzymology; Riordan, J. F., Vallee, B. L., Eds.; Academic: New York, 1993; Vol. 226, Chapter 13, pp 470-482. (1 8) Polavarapu, P. L. In VibrationalSpecfraandStrucfure;Durig, J. R., Sullivan, J. F., Eds.; Elsevier: Amsterdam, 1989; Vol. 17B, pp 319-342. (19) Diem, M. In VibrationalSpectraandStructure; Durig, J. R., Sullivan, J. F., Eds.; Elsevier: Amsterdam, 1990; Vol. 19, pp 1-54.
Analytical Chemistty, Vol. 66, No. 21, November 1, 1994 3639
in fact, been theoretically treatedz2J3 and experimentally ~ e r i f i e d . ~However, ~J~ the sensitivity of the instrument used in this single studyz4 was so poor that application of this technique to the chirooptical analysis of all types of chiral compounds has not been realized. The limitation on the sensitivity can be ameliorated if a near-IR laser such as a titanium-sapphire laser is used instead of the conventional light sources. The Ti-sapphire laser is a solid state laser that can be spectrally tuned from 650 to 1100 nm. Substantial instrumental improvement can be achieved when the advantages of this laser are fully exploited. Specifically, the wellcollimated, small-diameter, highly polarized monochromatic laser beam facilitates the complete conversion of the linearly polarized laser light to circularly polarized light. A spectropolarimeter based on this laser is particularly suited for samples that have small volume and low concentration. Additional improvement can also be achieved by use of the recently available high-performance and digital signal processing lock-in amplifiers, the low-noise preamplifier, and the photoelastic modulator, which has a larger acceptance angle and operates at high frequency. The information presented clearly indicates that it is possible to fully and synergistically exploit recent advances in laser technology and electronics to develop a novel and sensitivenear-IR circular dichroism spectropolarimeter. Such considerationsprompted us to initiate this study, and in this paper, we report the instrumentation development of the first near-IR spectropolarimeter based on the Ti-sapphire laser. Calibration of the instrument using nickel ((+)-tartrate)z and (+)- and (-)-C0(en)3~+ complexes, and its preliminary applications, include measurements of overtones and combination circular dichroism of saturated compounds (i.e., camphor, 2-octanol, and 2-amino-1-butanol) will also be reported.
EXPERIMENTAL SECTION The schematic diagram for the instrument, drawn as the solid lines, is shown in Figure 1. A Ti-sapphire laser (Coherent Corp. Palo Alto, CA, Model 890) pumped by a Coherent argon ion laser (Model Innova 70-5) was used as the light source. With the use of two sets of high reflector and output coupler (i.e., short-wavelengthand long-wavelength set), this Ti-sapphire laser can be spectrally tuned from 670to 850 nm and from 870 to 1030 nm, respectively. Spectral tuning of the laser was achieved by means of an intracavity birefringent filter that was driven by a computer-controlled steppingmotor. The output laser beam was very well linearly polarized. A Glan prism polarizer was used, however, to ensure that the beam was completely linearly polarized. Thelinearly polarized beam was then converted into left circularly polarized light (LCPL) and right circularly polarized light (RCPL) at 42 kHz by means of a photoelastic modulator (PEM; Hinds ~~
~~
~~
~
~~
~~
(20) Nafie, L. A.; Citra, M.; Rapnathan, N.; Yu, G. S.;Che, D. In Analyrical Applicariom of Circular Dichroism; hrdie, N., Brittain, H. G., Eds.; Elsevier: New York, 1994; Chapter 3, pp 53-89. (21) Tran, C. D.; Grishko, V. I.; Huang, G. Anal. Chem. 1994, 66, 2630-2635. (22) Lalov, I. J.; Svetogorsky, D. A. J . Chem. Phys. 1984, 80, 1083-1088. (23) Lalov, I. J.; Svetogorski, D. A.; Turkedjiev, N. P. J. Chem. Phys. 1986, 84,
3545-3552.
(24) Keiderling, T. A.; Stephens, P. J. Chem. Phys. b r f . 1976, 41, 46-48. (25) Konno, T.; Meguro, H.; Murakami, T.; Hatano, M. Chem. Lcrr. 1981, 7, 953-957.
3640 Analytical Chemistry, Vol. 66,flo. 21, November 1, 1994
argon laser I
u Figure 1. Schematlc diagram of the near-infrared clrcuiar dichroism spectropolarimeter: CH, chopper; P, polarizer; E M , photoelastic modulator; PD, photodkde;LIA, lock-inampiifler;pre-amp, preampllfler.
Instrument, Hillsboro, OR, Model FSA IR). The intensity of the light transmitted from the sample was measured by a PIN photodiode (United Detector Technology, PIN 10- DP). The output of the photodiodewas amplified and then connected to the Stanford Research System Model SR-510 lock-in amplifier (LIA 1). The output of the lock-in in turn was connected to a microcomputer (IBM compatible with a 486 microprocessor, Milwaukee PC, Milwaukee, WI) through the A/Dof the 12-bit DAS 16 board (Metra-Byte, Taunton, MA) for data acquisition and analysis. Nickel ((+)-tartrate)z complexes were prepared using the procedure reported in 1iterat~re.l~ Specifically, a 0.24 M aqueous solution of nickel(I1) sulfate was mixed with 0.36 M potassium sodium (+)-tartrate (Rochelle salt) solution. Similarly, nickel ((-)-tartrate)z complexes were prepared from nickel(I1) sulfate and potassium sodium (-)-tartrate. The latter was prepared from (-)-tartaric acid. Optically active (+)-C0(en)3~+and (-)-C0(en)3~+were prepared and resolved as described previously.26 Other optically active compounds were obtained from Aldrich Chemicals and used without further purification.
RESULTS AND DISCUSSION The C0(en)3~+complexes are generally used as standard compounds to calibrate the CD spectropolarimeter in the visible region. The CD spectra of the optically active pure (+), (-), and ( f ) - C ~ ( e n ) ~complexes, ~+ which we have synthesized for our previous studies,2b28were measured (in a 1-cm path length cell) using this Ti-sapphire laser-based spectropolarimeter, and the results obtained are shown in Figure 2. As illustrated, the (+)C0(en)3~+complexes exhibit a positive CD band with its maximum at about 765 nm. The (-)-C0(en)3~+complexes expectedly exhibit a negative band with magnitude equal to that of the (+) complexes, and the racemic mixture has no signal at all. These observationsand the fact that these spectra are in good agreement with those reported earlier29930 lend credence to the spectropolarimeter. Since the C0(en)~3+complexes do not absorb at wavelengths longer than 860nm, nickei tartrate complexes weresynthesized (26) Tran, C. D.; Xu, M. Rev. Sci. Imrrum. 1989,60, 3207-321 1. (27) Xu, M.; Tran, C. D. Appl. Specrrosc. 1990,44, 962-966. (28) Xu, M.; Tran, C. D. Anal. Chem. 1990,62, 2467-2471. (29) McCaffery, A. J.; Mason, S.F. Mol. Phys. 1%3, 6, 359-371. (30) Norden, B.; Seth, S.Appl. Specttosc. 1985, 39, 647-655.
4
880
905
I*
3
930
955
980
1005
A
2
zx s
1
6 o
0.0
B
-1 -2 -3 -4
700
750 800 Wavelength, nm
850
M aqueous (b) (-)-C~(en)~~+, and (c) ( f ) - C ~ ( e n ) ~ ~ + solutions of (a) (+)-C~(en)~~+, (in a 1-cm pathlength cell). 6 L1
1
I
4
2
$ x
s
o
-2 -4 4
3
880
"
"
"
900
"
'
" " " "
920 940 960 Wavelength, nm
W 930 955 980 1005 Wavelength, nm Figure 4. Absorption (A) and near-IR circular dichroism (B) spectra of 2.00 M solutions of and (RS+camphor In carbontetrachloride (in 20-cm pathlength cell). - 6 " " " " " "
Flgwe 2. Near-IR circular dichroism spectra of 4.00 X
980
' 1
1000
Flgure 3. Near-IR circular dichroism spectra of 1.20 X
M aqueous solutions of (a) nickel (+)-tartrate, (b) nickel (-)-tartrate, and (c) nickel (fktartrate (in a 1-cm pathlength cell).
and used as standard compounds for calibration in the longer wavelength region. The CD spectra of the nickel (-)-, nickel (+)-,and nickel (f)-tartrate complexes in the 880-1000-nm region are shown in Figure 3. As expected, the CD spectrum of the (-) complexes has the same amplitude but opposite sign to that of the (+) complexes. Again it is pleasing to see good agreement between the observed and reported spectra." Calibration curves were constructed for (+)-Co(en)33+ complexes using CD data at 765 nm. A good linear relationship (correlation coefficient of 0.999) was obtained over a concentration range of 3.00 X 10-4 to 0.100 M. The limit of detection (LOD) calculated as twice the standard deviation of the near-IR CD background signal divided by the slope of the calibration line is estimated to be 7.80 X lo4 M . These concentrations correspond to the L 4 values of 4.2 X 10". This LODvalue is comparable with other LODvalues (1.5 X 10" and 1.8 X lod) that were obtained for the same C0(en)3~+complexes using circular dichroism spectropolarimeters constructed from the visible argon ion laser.1°J3 It is evidently clear that the sensitivity of this spectropolarimeter is high. However, it still is not sufficient for the measurements of overtone CD; Le., no measurable CD signal was observed for neat ( R ) -or (S)-2-octanol or for a 2.00 M
880
"
'
'
905
(e, (s,
solution of (R)- or (&camphor. Additional improvement was required to bring its sensitivity to a level that is suitable for overtone CD measurements. A schematic diagram of the improved spectropolarimeter is shown as dashed lines in the box in Figure 1. In this case, the Ti-sapphire laser beam, after being converted to linearly polarized light (by a polarizer, P) and circularly polarized light (by the photoelastic modulator, PEM), passes through the sample. Its intensity is detected by a PIN photodiode. In order to increase the measured signalto-noise ratio, a mechanical chopper (Stanford Research Systems Model SR450) was inserted between the Ti-sapphire laser and the PEM to modulate the near-IR laser beam at 85 Hz. Two lock-in amplifiers were used to acquire data in a manner similar to those used by others.1"16 Specifically, the output of the photodiode, after being amplified by a Princeton Applied Research Model 5 113 low-noise preamplifier operated with a bandpass filter of 30-100 kHz, was connected to a high-performance lock-in amplifier (LIA 1;Princeton Applied Research Model 5302). This lock in amplifier, which operated with a signal-channel bandpass filter tuning automatically to the reference frequency,demodulated and amplified the signal at the frequency of the PEM (Le., 42 kHz). Its output was then connected to a second digital signal processing lock-in amplifier (LIA 2, Stanford Research Systems Model SR810) whose reference signal was set at the frequency of the chopper (Le., 85 Hz). This lock-in amplifier worked in the synchronous filtering mode. Its output was then connected to the microcomputer in a manner similar to that described for Figure 1. It was found that the sensitivity of this configuration was substantially improved to a level that is sufficient for the overtone CD measurements. Accordingly, the CD spectra of 2.00 M solutions of ( R ) - ,( S ) - ,and (RS)camphor in carbon tetrachloride were measured using a 20cm pathlength cell, and the results obtained together with the absorption spectra of these compounds are shown in Figure 4. As illustrated in Figure 4A, these compounds exhibit strong absorption band at about 9 17nm. This band has been assigned to the third overtone transition of the C-H ~~
(31) Tran, C. D.; Grishko, V. I.; Baptista, M. S. Appl. Spectrosc. 1994.48.833842.
Analytical Chemistry, Vol. 66, No. 21, November 1, 1994
3641
880
905
930
955
980
1005
870
895
920
945
970
995
Wavelength, nm
Wavelength, nm
Flgure 5. Absorption (A) and near-IR circular dichroism (6)spectra of neat (S)-, and (RS)-P-octanol (in 20-cm pathlength cell).
Figure 6. Absorption (A) and near-IR circular dichroism (6) spectra of neat (S)-,and (RS)-2-amino-l-butanol (in 20-cm pathlength cell).
(e,
As illustrated in Figure 4B, there is a pronounced circular dichroism signal associated with this transition: the ( S ) camphor exhibits a positive CD band with a maximum at the same wavelength as that for the absorption spectrum @e., 9 17 nm). Again, it is pleasing to observe that the (R)-camphor exhibits a negative band with the same magnitude as that of the S enantiomer, and the racemic mixture expectedly has no signal at all for this transition. These results provide clear and unequivocal evidence that the measured CD spectra are in fact due to the overtone CD of the C-H group and that the sensitivity of the present near-IR CD spectropolarimeter is sufficiently high to measure such small signals. As stated previously, the Ti-sapphire laser can be spectrally tuned from 670 to 1030 nm. Because this region covers the overtones and combination transitions of the C-H and 0-H gr0ups,3l-~~ this near-IR CD spectropolarimeter can also be used to measure the overtone and combination CD of aliphatic alcohols. Presented in panels A and B of Figure 5 are, respectively, the absorption and CD spectra observed for neat ( R ) - ,( S ) - ,and (RS)-2-octanol taken using a 20-cm pathlength cell. As expected from their structures, the absorption spectra of these compounds are relatively more complicated than those ofcamphor. That is, they exhibit two broad absorption bands and a shoulder of a band beginning at about 983 nm (the band that corresponds to this shoulder was not observed because of the limitation in the wavelength of the Ti-sapphire laser). As illustrated in the figure, the broad band at 930 nm can be resolved into three Gaussian bands with peaks at 909.4,917.9, and 928.5 nm. Comparing this spectrum with that of octane, and on the basis of the information reported earlier,3*~~~ these three bands and the band which begins at 983 nm are tentatively assigned to the third overtone and the combination of the stretching and bending vibrations of the C-H groups. The band at about 975 nm is probably due mainly to the (2q + v 3 ) combination transitions of the 0-H group, where V I is the symmetric stretch and v 3 is the a n t i ~ y m m e t r i c . Because ~~,~~ CD spectra are expected to correspond to the absorption (32) Kelly, J. J.; Barlow, C. H.; Jinguji, T. B.; Callis, J. B. Anal. Chem. 1989, 61, 31 3-320. (33) Phelan, M. K.;Barlow, C. H.; Kelly, J. J.; Jinguji, T. M.; Callis, J. B. Anal. Chem. 1989,61, 1419-1424.
3642
Analytical Chemistry, Vol, 66, No. 21, November 1, 1994
(e,
spectra, there are several bands present in the CD spectrum for each compound. For instance, as shown in Figure 5B, (R)-2-octanol exhibits several positive and negative bands in theregionfrom910to955 nm. Thesebandscan beattributed to the CD of the three aforementioned third overtone and combination of the C-H groups. There are relatively larger CD bands at about 970 nm and in the region from 983 to 1005 nm. These bands are probably due to the combination transitions of the 0-H and C-H groups, respectively. It is interesting to observe the fact that while the absorption bands (of the C-H groups) at 909.4, 917.9, and 928.5 nm (Le,, the first broad band) are much stronger than the band at 975 nm and the band beginning from 985 nm, the magnitudes of the corresponding CD bands are in reversed order (i.e., CD bands at about 970 nm and at wavelengths longer than 985 nm are muchstronger thanthosebetween910and955nm). Avariety of reasons may be accountable for this observztion, but the most likely one is probably due to the proximity of the group to the chiral center of the molecule. Since this is 2-octanol, the 0-H group is in fact connected to the chiral carbon. As a consequence, its CD signal is expected to be larger. A strong shoulder beginning at 985 nm that has strong CD band is probably due to the C( l)-H3, C(2)-H, and C(3)-Hz groups. Introduction of only a single amino group into a molecule produces substantial differences in the absorption and CD spectra. Presented in Figure 6 are the absorption (A) and near-IR CD (B) spectra of neat (I?)-,(S)-, and (RS)-2-amino1-butanoltaken in a 10-cm path length cell. The absorptions of these compounds are much different from those of 2-octanol (Figure 5A). In 2-octanol, the absorption of the 0-H group produces a relatively well resolved band at 975 nm (Figure SA). Conversely, for 2-amino- 1-butanol, the absorption band of the 0-H group overlap with those of the C-H and N-H groups. As a consequence, only a very broad band is observed for these compounds (Figure 6A). Much larger differences are, however, observed in the CD spectra of two sets of compounds. While the CD spectra of 2-octanol are well resolved into several bands, those of the 2-amino-1-butanol are not: only a single very broad band was observed. This is hardly surprising considering the unresolved absorption
spectra of these compounds. It is rather interesting, however, to note that the magnitude of the CD band of 2-amino-lbutanol is relatively higher than any CD bands of 2-octanol. It has been demonstrated that the latest advances in laser technology, optics, and electronics can be fruitfully and synergistically exploited to develop a novel near-IR CD spectropolarimeter which has sensitivity so high that it can be used successfully for measurements of the overtone and combination transition CD. Compared to the CD in the UV-visible and infrared region, the near-IR CD has such advantages as the following: (1) Universal applicability The near-IR absorption is due to the overtones and combination transitions of the OH, CH, and N H groups. As a consequence,the near-IR CD technique can be used to chirooptically analyze all chiral organic compounds. (2) Ease of measurement There is no need for sample preparations (e.g., thin path length cell as in the vacuum UV and middle IR, Nujol, KBr pellet). It may be possible, therefore, to use this universal near-IR CD spectropolarimeter for the chirooptical analysis of any chiral compounds in a complex background. The use of the multivariate calibration methods (partial least squares, principal component a n a l y ~ i s ~ ~) Jto5 analyze the overtone CD data is probably needed for such cases. The major limitation of the near-IR CD instrument reported here is its low sensitivity. However, its sensitivity can, in principle, be improved by decreasing the noise and/or increasing the signal intensity. Fluctuation in the laser intensity is probably the major source of noise in this case. Several methods are currently available to stabilize the intensity of lasers. However, as we have demonstrated recently, the most effective methods to stabilize the amplitude of a single wavelength or multiwavelength laser beam are (34) Martens, H.; Naes, T. Multivariate Calibration; John Wiley: New York, 1989; Chapter 1 , pp 1-34. (35) Sekulic, S.;Seasholtz, M. B.; Wang, Z.; Kowalski, B. R.; Lee,S. E.; Holt, B. R. Anal. Chem. 1993, 65, 835A-845A. (36) Tran, C. D.; Furlan, R. J. Appl. Spectrosc. 1993, 47, 235-238. (37) Tran, C. D. Anal. Chem. 1992, 64,971A-981A. (38) Tran, C. D.; Furlan, R . J. Appl. Spectrosc. 1992, 46, 1092-1095. (39) Tran, C. D.; Furlan, R. J. Reo. Sci. Instrum. 1994, 65, 309-314.
those based on the use of either an electrooptic modulator (Le,, Pockels cell)36or an acoustooptic tunable filter.37-39In fact, up to 2 orders of magnitude reduction in the noise level of the Ti-sapphire output power can be achieved by use of these method^.^^-^^ Alternatively, the sensitivity of the spectropolarimeter can also be improved by using a new method to measure the CD signals. Specifically, rather than using the conventional transmission method to measure the nearIR CD as in the present work, the thermal lens effect (TL) can be used to measure the near-IR CD signal. The sensitivity of this TL near-IR CD should be enhanced because, as we have demonstrated recently for the visible regi~n,Z~-~O the sensitivity of the TL-CD is proportional to the excitation laser power. As a consequence, it is relatively higher than the transmission CD methods. The sensitivity of the overtone and combination CD can also be improved if the measurements are performed not in the near-near-infrared region as used in this work (Le., in the 700-1000-nm region) but rather in the region longer than 1000 nm. This possibility stems from the fact that, in the 700-1 000-nm regions, the transitions involved are higher overtones and combinations (e.g., third overtone of the C-H groups). The molar absorptivities for these higher transitions are relatively lower than those of the lower transitions, which can readily be measured in the longer wavelength region. As a consequence, the sensitivity of the CD for the near-near-IR would be lower than that for the near-IR. Experiments are now in progress to explore these possibilities.
ACKNOWLEDGMENT The authors are grateful to the National Institutes of Health, National Center for Research Resources, Biomedical Research Technology Program for financial support of this work. Acknowledgment is also made to the NIH BRSG (Grant RR07196) for partial support of this work. Received for review June 3, 1994. Accepted July 29, 1994.'
* Abstract published in Aduance ACS Absrracrs, September 15, 1994.
Analytical Chemistry, Vol. 66, No. 21, November 1, 1994
3643
