Flow Cell Sfcftir

The flow cell approach involves connecting a high pressure flow-through cell at the end of the separation column and positioning the flow cell in the FTIR beam so that the analytes can be monitored in real time as they elute from the column and flow through

Table 1 Key advances in the development and application of flow cell and mobile elimination interfaces for SFC/FTIR

Year

Development

Flow cell approach 1983

1985, 1986

1986

1987

1988

1989

1990

1991 -94 1994

Mobile phase elimination approach 1986

1986

1988

1989

1992 1995

First demonstration of flow cell SFC/FTIR using carbon dioxide as the mobilephase (Shafer and Griffiths)

Demonstration of packed column SFC/FTIR with 10 |L flow cell

Evaluation of carbon dioxide and Freon-23

First use of xenon as a mobile phase for capillary SFC/FTIR

Development of Gram-Schmidt orthogonalization for removing baseline drift from reconstructed chromatograms as a result of pressure programming the mobile phase

Introduction of commercial thermostated 1.4 |L flow cell for capillary SFC/FTIR Application to thermally labile compounds such as pesticides and vitamins demonstrated

Detection limit for caffeine determined to be &2 ng

Development of 800 nL flow cell for use with 50 |im i.d. capillary columns

Use of make-up gas to minimize band broadening in flow cell studies

Stopped flow SFC/FTIR used to improve signal-to-noise ratio of flow cell measurements

Comparison of xenon and carbon dioxide as mobile phases for SFC/FTIR Investigation of 500 nL flow cell for use with 50 |im i.d. capillary columns Practicability of multihyphenated SFC-UV-FTIR system demonstrated

First demonstration of packed column SFC with deposition on moving KCl plate. Offline diffuse reflectance IR spectra measured from analytes (Shafer and Griffiths; Pentoney)

Use of FTIR microscope to measure IR spectra from 150 |im diameter spot deposited from capillary SFC and restrictoron to a ZnSe plate (Pentoney etal.) Demonstration of offline mobile phase elimination approach to polymer additives, phenolic carboxylic acids, steroids, polycylic aromatic hydrocarbons Comparison of IR sampling techniques evaluated for SFC/FTIR Conventional transmission sampling confirmed to give best sensitivity, repro-ducibility and adherence to Beer's law

Adaption of first commercial online direct deposit GC/FTIR interface for SFC/FTIR

Performance of real-time direct deposition SFC/FTIR interface characterized Identifiable spectra from 600 pg caffeine demonstrated (Norton and Griffiths) Application to sulfanilamides, silicone oligomers, polyethylene glycol and nonyl phenol oligomers, herbicides the cell. The IR absorbance is directly proportional to the concentration of the sample in the chromatographic peak that is flowing through the cell at any given time, according to the Beer-Lambert law. A flow restrictor is usually placed after the cell to maintain high pressure in the flow cell, but since FTIR is nondestructive, additional detectors such as a flame ionization detector may be connected in series after the flow cell. A schematic diagram of a SFC/FTIR with a flow cell interface is shown in Figure 1. One of the main advantages of online FTIR detection is that chromatograms based on total or selected IR absorb-ance of the organic molecules can be reconstructed using the Gram-Schmidt algorithm. The retention time data obtained from the Gram-Schmidt reconstructed (GSR) chromatogram is complemented by structural information provided by interpreting IR spectra of the separated components, or by compar ing them with a spectral library. The major disadvantage of this interface is the low detection sensitivity obtained, due to the fact that in a dynamic system only a limited number of spectra can be measured as an analyte peak passes through the flow cell.

Flow Cell Compatible Supercritical Mobile Phases

Traditionally, analytes that are incompatible with GC have been separated by liquid chromatography (LC). However, most LC solvents absorb so strongly in the IR region of the spectrum that flow cells with very short pathlengths (typically 100-500 |im) must be used to prevent the solvent bands from completely obstructing wide regions of the spectrum. Detection limits are therefore very high ( > 20 times higher than in GC/FTIR). One of the advantages of supercritical carbon dioxide, the most commonly used SFC mobile phase, is that it has extensive regions of transparency

Ftir Schematic Diagram

Figure 1 Schematic diagram of SFC/FTIR system with a flow cell interface.

obscures the 3504-3822 cm 1 region. Consequently, vibrational modes of functional groups such as the O-H stretch (3500-3800 cm"1), C=N stretch (2100-2200 cm"1), C-Cl stretch (600-800 cm"1) and aromatic C-H out-of-plane bending modes that absorb below 750 cm"1 cannot be detected in CO2. It should also be pointed out that the position of absorption bands in the spectra of compounds measured in carbon dioxide vary with the pressure and temperature of the mobile phase. For example, variations up to 10 cm" 1 have been observed in spectra measured in liquid versus supercritical CO2. Modifications of intensity distribution and band width variations were also reported. As a result, the SFC/FTIR spectra obtained using a flow cell are very similar but may not be identical to vapour-phase spectra. However, SFC is generally applied to involatile materials for which

Figure 1 Schematic diagram of SFC/FTIR system with a flow cell interface.

in the IR region of the spectrum, making it more compatible for flow cell SFC/FTIR. However, not all functional groups can be detected and in these cases xenon or other fluids have been used.

CO2 has two strong infrared-active fundamental vibrational modes, the antisymmetric stretching band that is centred at approximately 2350cm"1 (v3) and the bending mode near 668 cm"1 (v2). The symmetry-forbidden symmetric stretch (v1) is centred at about 1390 cm"1. These bands obscure the IR spectrum from 2137 to 2551 cm"1 and below 800 cm"1, as shown in Figure 2A. The intensity of v1 is enhanced by Fermi resonance with 2v2, so that two bands at 1390 and 1285 cm"1 are observed in the spectrum. In SFC, the mobile phase is typically programmed from a low density to a higher density to increase the solvent power of the fluid and elute progressively more involatile analytes. As the density of the CO2 is programmed upwards, these Fermi resonance bands also increase in intensity, as shown in Figure 2B, but are generally not a problem to deal with, as they may be subtracted from the spectra of eluting components if short cell pathlengths (<10 mm) are used. These two Fermi modes also interact with v3 to generate a very intense doublet, (v1 + v3) and (v3 + 2v2), that

Figure 2 (A) IR spectrum of carbon dioxide at 300 atm and 25°C using an optical pathlength of 10 mm. (B) Increase of absorption of the Fermi resonance bands at 1390 and 1285 cm-1 with increasing CO2 pressure.

vapour-phase spectra do not currently exist and hence, in the future, libraries of spectra recorded in CO2 will probably need to be generated.

In cases where detection of the above functional groups is necessary, xenon may be used as an alternative mobile phase. Although xenon is much more expensive than carbon dioxide, it has the advantage of being completely transparent in the IR and background subtraction is less of a problem. However, significant purging with xenon is required to remove the carbon dioxide from the pump and chromatograph completely when changing mobile phases. Spectra measured in xenon also more closely match available vapour-phase library spectra than spectra obtained in carbon dioxide. Xenon is nontoxic, has critical parameters (Tc = 16.6°C, Pc = 57.6 atm) that are similar to carbon dioxide and is compatible with other chromatographic detectors such as the flame ionization detector. Even though only a limited amount of research has been undertaken with xenon as a mobile phase, most reports indicate that it has similar chromatographic properties to carbon dioxide for nonpolar analytes, but is inferior to carbon dioxide for polar analytes.

As far as other fluids are concerned, supercritical nitrous oxide has been shown to be unsuitable for flow cell SFC/FTIR, whereas supercritical sulfur hexafluoride affords transparency solely in the 3000 cm-1 region (i.e. for detection of C-H stretching bands). It has been shown that several chloro-fluorocarbons that have moderate critical temperatures and pressures, such as Freon-23 (trifluoro-chloromethane), have significant regions of transparency and complement carbon dioxide. For example, the O-H stretching region at 3600 cm"1 cannot be monitored in carbon dioxide, but can be monitored in Freon-23. Finally, the addition of polar modifiers is often required to increase the solvent strength of carbon dioxide or to deactivate active sites on stationary phases, especially in packed column SFC. Unfortunately, the addition of a polar modifier such as an alcohol to carbon dioxide reduces the applicability of the online method. Addition of 1% hexanol to CO2, for example, blocks large regions of the mid-IR spectrum, while addition of 0.2% methanol has been shown to reduce the IR-accessible windows to 34002900 cm-1, 2800-2600 cm"1, 2100-1500 cm"1 and 1200-1100 cm"1.

Flow Cell Design Criteria

A number of factors must by considered when using a flow cell interface: choice of chromatographic column (i.e. packed or open tubular), flow cell dimensions, materials of construction of the cell body, windows and seals, the mobile phase and the detection conditions. A schematic diagram of a high pressure flow cell is shown in Figure 3. The major chromato-graphic requirements of the cell are that it must withstand the high pressures used in SFC (up to 500 atm) and that it has a small volume in relation to the volume of a typical peak eluting from the column to prevent band broadening. To withstand high pressure, most flow cells have been constructed from stainless steel, and ZnS, CaF2 or ZnSe window materials, have been used with polytetrafluoroethylene (PTFE) or Kel-F seals. To prevent band broadening, the cells must be designed to minimize dead volume in the connections and flow path and the cell volume should not exceed 0.3 SD or about 25% of the volume of a Gaussian peak. However, the peak volume is dependent on the column type. For example, peak volumes in typical 4.6 mm i.d. packed columns are in excess of 40 |L and cell volumes of 10 |L are therefore optimal. For capillary SFC, open tubular columns are typically 50 | m in internal diameter and a typical peak has a width at half height of about 15 s at 150 atm when operated at a flow rate of 1.5 |L min"1. This corresponds to a peak volume of 400 nL and hence the optimal detector volume is in the region of 100 nL. Unfortunately, cells with such small volumes are very difficult to make and are also in conflict with the requirements of the FTIR. Inevitably, a compromise between chromatographic and spectroscopic requirements must be reached.

For maximum IR sensitivity, the dimensions of the flow cell must also be taken into account. According to the Beer-Lambert law, the IR absorbance (A) is dependent on the sample molar extinction coefficient (s), the sample concentration (c) and the cell path-length (l). Thus, increasing l should improve the signal-to-noise ratio of the spectra obtained. However, l can only be increased so much before absorption by the mobile phase (except with xenon) becomes too intense to allow an adequately low noise level after spectral subtraction. Furthermore, in order to maintain a constant cell volume, the light pipe diameter must decrease as l increases, and this leads to a drop in optical throughput. As the minimum attainable beam diameter is approximately 1 mm using state-of-the-art beam-condensing optics, a cell light pipe diameter less than this must reduce the available energy transmitted through the cell. Consequently, the minimum practical cell diameter is 0.5 mm. Pathlengths of flow cells used in packed column SFC/FTIR have ranged from 5 to 10 mm, while those used with open tubular columns have typically been 1-5 mm to reduce the cell volume as much as possible. Construction details of some flow cells reported in the literature and the experimental conditions under which they were used are summarized in Table 2.

50 |im i.d. deactivated fused silica tubing

Figure 3 Schematic diagram showing the important design aspects of a high pressure flow cell used for SFC/FTIR.

50 |im i.d. deactivated fused silica tubing

Figure 3 Schematic diagram showing the important design aspects of a high pressure flow cell used for SFC/FTIR.

A number of flow cell SFC/FTIR studies have been performed to determine the minimum identification limit (MIL), defined as the quantity of compound required for identification by spectral interpretation, and the minimum detectable quantity (MDQ), defined as the quantity of material which must be injected to yield an IR response three times the noise level. Although these limits are dependent on the strength of the IR absorption of the compound, an MIL of 10 ng methyl palmitate and an MDQ of 2 ng caffeine have been reported using capillary columns and a flow cell with a 5 mm pathlength.

Application

One of the advantages of the flow cell approach is that chromatograms based on total or selected IR absorbance of the organic molecules can be reconstructed using the Gram-Schmidt algorithm. Further FTIR can also be used as a chemically specific detector, by reconstructing absorbance data in a specific wavenumber region. the IR spectra of the peaks contained in the associated data files can be retrieved and manipulated to remove the spectral features of the mobile phase. Most packed column studies have been performed with CO2 at constant pressure and temperature. In this condition, the IR spectrum of the mobile phase does not change during the course of the chromatographic run, facilitating subtraction of the background mobile-phase spectrum from the spectrum of each of the analytes and generation of a GSR chromatogram. In capillary SFC, however, the mobile phase is typically programmed from a low to a high pressure and Gram-Schmidt orthogonaliz-ation must be used to prevent this baseline rise. This procedure has been developed: vectors containing the information from the rising baseline are added to the basic set and the chromatogram is then recalculated to remove the baseline drift and to enhance the signal intensity. A flow cell SFC-FTIR analysis of a UV-curing coating in Figure 4 exemplifies these features.

Many other thermally labile and nonvolatile samples have been analysed by flow cell SFC/FTIR, including free fatty acids, sesquiterpene hydrocarbons, carbamate pesticides, double-base propellants, steroids, triglycerides, aromatic plasticizers and aromatic isocyanates.

Table 2 Experimental details of some packed and capillary column SFC-FTIR studies employing flow cells

Flow cell parameter/study

Jordan and Taylor (1986)a

Morin et al. (1986)b

Wieboldt et al. (1988)c

Raynor et al. (1989)d

Mobile-phase

CO2, CClF3

Xe, CO2

CO2

CO2

Column dimensions

25 cm x 4.6 mm i.d.

15 cm x 4.6 mm i.d.

10 m x 100 ^m i.d.

10mx50 ^m i.d.

(5 |m packing material)

(5 ^m packing material)

capillary column

capillary column

Cell volume

8

8 |L

1.4 |L

800 nL

Pathlength

10 mm

10 mm

5 mm

4mm

Window material

CaF2

ZnSe

Not documented

ZnSe

Pressure

200 atm

282 atm

400 atm

400 atm

Temperature

Ambient

Ambient

Ambient to 50°C

Ambient

Beam condenser

Yes, 3 x

Yes, 4x

Yes

Yes

Scans

2s-1

3s-1

4s-1

3s-1

Resolution

4 cm-1

8 cm-1

4cm-1

4cm-1

Jordan and Taylor (1986) Journal of Chromatographic Science 24: 82.

bMorin etal. (1986) Chromatographia 21: 523.

cWieboldt etal. (1988) AnalyticalChemistry60: 2422.

dRaynor et al. (1989) Journal ofMicrocolumn Separations 1: 101.

Jordan and Taylor (1986) Journal of Chromatographic Science 24: 82.

bMorin etal. (1986) Chromatographia 21: 523.

cWieboldt etal. (1988) AnalyticalChemistry60: 2422.

dRaynor et al. (1989) Journal ofMicrocolumn Separations 1: 101.

Solar Panel Basics

Solar Panel Basics

Global warming is a huge problem which will significantly affect every country in the world. Many people all over the world are trying to do whatever they can to help combat the effects of global warming. One of the ways that people can fight global warming is to reduce their dependence on non-renewable energy sources like oil and petroleum based products.

Get My Free Ebook


Post a comment