The Supercritical Fluid Chromatography Engineering Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This paper shows some of the different supercritical fluids available for use in SFC, and discusses the advantages that SFC has over HPLC and GC and how these can provide benefits to the industries which adopt this technique. Nevertheless, HPLC can't be fully replaced as it still needed for samples that aren't compatible with SFC, but together they can be used to accomplish experiments. The different setups available for SFC and how they compare along with the mobile phases and solid phases which are currently used are highlighted within the paper.

Supercritical fluid chromatography (SFC) is a normal phase chromatographic technique which was overlooked when first introduced, as reversed phase high performance liquid chromatography (HPLC) was preferred. However, the use of supercritical fluid chromatography has recently been adopted as a main chromatographic technique, as from about 1984 packed and capillary SF columns became more readily available [1, 2]. As a result there is still a lot of research being undertaken in the areas of SFC method development and in hardware development [3]. The key difference between SFC and gas chromatography (GC) and HPLC is that a supercritical fluid is used as the mobile phase [1].

A supercritical fluid is an element or compound above its critical temperature and pressure and therefore is an excellent mobile phase as it is both gaseous and solvating (Figure 1).

Figure 1 - Phase diagram showing the critical temperature and pressure values of pure carbon dioxide [4].

Supercritical fluids are obtained either by heating a gas above its critical temperature or by compressing a liquid at a higher pressure than its critical pressure. Once in a supercritical state a variance in either temperature or pressure will result in a change in the density of the mobile phase.

There is a range of different supercritical fluids available, some examples and their critical temperatures, pressures and densities are shown in Table 1[4].

Table 1 A range of different supercritical fluids and their critical temperatures, pressures and densities


Critical temperature (oC)

Critical pressure (bar)

Critical density (g mL-1)





























The critical conditions of carbon dioxide make it the choice of mobile phase in SFC for the analysis of the thermally labile compounds.

Advantages and disadvantages of SFC

Even though HPLC is a widely used technique for the separation of analytes in many classes, SFC has clear advantages over it. In HPLC a substantial amount of organic solvent is generated with each separation, which then needs to be disposed. The disposal of the organic solvents is expensive at $5 - $10 per gallon, whereas SFC uses considerably less, or no organic solvent, which leads to a decrease in analysis costs [5]. In the replacement of organic solvents an inert environmentally friendly mobile phase is used, often carbon dioxide which can be collected from the atmosphere [6]. Also, without the use of organic solvents the product is more concentrated compared to HPLC where the solvent must be evaporated. Without the need to evaporate any solvent there is a reduction in energy and labour costs [6].

SFC is similar to GC in that both gases and supercritical fluids have lower viscosity and higher diffusion coefficient than the liquids used in HPLC which allows for quicker, more efficient separations. The separation time can be cut down from hours or days to a few tens of minutes [7]. As seen in Table 2, supercritical fluids lie between liquids and gases, allowing for SFC to use the features of both HPLC and GC.



Supercritical Fluid


Density g/cm3

(0.6-2) x 10-3



Diffusion Coefficient cm2/s

(1-4) x 10-1


(0.2-2) x 10-5

Viscosity g cm-1 s-1

(1-3) x 10-4

(1-3) x 10-4

(0.2-3) x 10-2Table 2: Comparison of Properties of Supercritical Fluids, Liquids and Gases [5]

Due to supercritical fluids having gas like and liquid like density they have a greater solvating power than either gases or liquids. Therefore, SFC has a larger molecular range which includes non-volatile molecules which methods like GC do not include [5, 8]. Also, unlike GC which does not analyse thermally unstable compounds, SFC is able to, due to the low critical temperatures of supercritical fluids such as carbon dioxide (31oC)[5] . An advantage of supercritical fluid carbon dioxide is that it has a varied solvating strength that allows for selective separations [9]. Along with this by altering the temperature and/or pressure it is possible to achieve higher selectivity.

The range of detectors is also wider for SFC compared to GC or HPLC because in SFC the mobile phase can be liquid or gas like, so both GC and HPLC detectors can be used [5]. For example SFC with flame ionization detection (FID) can provide quantification of resolved materials with a sensitivity of 0.1 ng [8]. Due to the range of detectors available for SFC and the low critical temperature of the carbon dioxide mobile phase, the detection and analysis of thermally labile compounds has been successful[7, 9].

Pinkston et al. compared HPLC/MS and SFC/MS in the elution and detection of pharmaceutical drug compounds. From the library 75% were detected by SFC/MS and 79.4% by HPLC/MS. They found that compounds containing phosphate, a phosphonate, or a bisphosphonate were not observed using SFC/MS [10].

Another advantage SFC has over HPLC is in the separation of chiral compounds. In HPLC the process is very time consuming, in SFC however, due to the lower viscosity of the supercritical fluids, the chiral separation can be run at a flow rate of up to five times faster than that of the HPLC while avoiding pressure build up. The higher flow rate of SFC consequently means that the productivity is higher than HPLC methods [6]. A comparison between HPLC and SFC has been conducted for trans-Stilbene oxide (Figure 2 and 3, Table 3).

Figure 2 [11]

Figure 3 - Comparison of HPLC and SFC elution times for trans-Stilbene Oxide [12]

Table 3: Comparison of chromatographic conditions [12]



Column (4.6 mm ID x 250 mm L)



Column temperature (oC)



Mobile phase

Hexane/IPA (60:40)


Mobile phase flow rate (mL/min)






Modifier flow rate (mL/min)



Pressure (MPa)



Wavelength (nm)



When used in large scale separations, fluid carbon dioxide can be recycled and then reused this minimises the amount of waste generated [7].

Due to the fact that SFC has features of both GC and HPLC, SFC offers diversity in the columns that can be used which are either open tubular (GC) or packed (HPLC). In packed column SFC by choosing suitable column dimensions and particle size [13], this can cause an increase in the number of theoretical plates (over 100,000) [6, 13]. Along with the column and particle size the choice of stationary phase is important, (see the section 'Mobile phases and stationary phases used in SFC').

Further advantage of SFC is that it is very clean; any contaminants of the mobile phase are usually gases. Due to the mobile phase being free from dissolved oxygen the mobile phase is easily removed [6].

However, there are some disadvantages of SFC. One is when carbon dioxide is used as the mobile phase is it does not elute very polar or ionic compounds; this can be easily overcome by using an organic modifier (see 'Mobile phases and stationary phases used in SFC'). Another disadvantage is that the availability of SFC is limited [6].

Instrumentation used in SFC

Supercritical fluid separation was first conducted by Kesplar et al. in 1962; the first groups to use carbon dioxide as the mobile phase were Sie et al. and Giddings et al. In the beginning SFC systems used HPLC instruments which were modified for SFC use. The newer instruments now often include a pumping system, modifier module, post-column nozzle and a separator detector [6] Figure 4. Lee et al. were one of the first groups to use capillary columns for SFC [2].

Figure 4 - Standard SFC schematic [14]

The mobile phase in SFC is pumped as a liquid and then heated up past supercritical temperature until it reaches the supercritical region. It passes through the injection valve before the sample is introduced, which carries the sample into the analytical column. To ensure the mobile phase stays supercritical, pressure restrictors are placed after the detector or at the end of the column. The pressure restrictors are heated as to avoid clogging [1].

As SFC uses a supercritical fluid as mobile phase, there are two possible types of column setups; typical dimensions for each type can be seen in Table 4.

Table 4: Typical Dimensions of packed and capillary columns [1, 4]:

Internal Diameter (mm)

Length (m)


2.0 - 4.6

0.03 - 0.25


0.025 - 0.1

1 - 35

One setup is HPLC like which consists of two reciprocating pumps, a packed column and an optical detector. The pumps mix the mobile phase as well as introduce the modifier, and the column is placed in an oven, in this setup the pressure and flow rates can be controlled separately [1]. Packed column SFC has recently become popular again over the past decade due to drug discovery and the pharmaceutical industry, as it offers the use of an environmentally friendly mobile phase, carbon dioxide, and decrease in waste generation and provides purified materials even on a large scale. When used for drug discovery packed SFC is usually coupled with a mass spectrometer detector [6, 15]. In SFC there is lower eluent viscosity and higher diffusion coefficient which, as a result, leads to an increase in efficiency and a shorter separation time, the low viscosity causes only slight pressure drops which in turn allows for the flow rate to be quicker (3-5 mL min-1) compared to that of HPLC (typically ~1 mL min-1)[5, 15].

The other column setup is capillary SFC which is an extension of GC that includes a syringe pump and a capillary column inside a GC oven with a restrictor with a flame ionisation detector (FID). However, in capillary SFC the flow rate of the pump controls the pressure of the system [1, 13]. Other detection methods are also used for capillary SFC. One method is Fourier transform infrared (FT-IR) spectrometry where the eluent used is xenon as it has no inherent background spectrum [2, 16]. Capillary SFC is used for high separation power and is more suited for fluids with low density. However, capillary columns have some limitations these include sample loading capacity, detection limits and quantitation [1, 13].

As mentioned FID is mostly used for capillary SFC, although in certain cases FID can be used with packed column SFC when a non-flammable mobile phase is used. The mobile phase that is used is usually carbon dioxide which requires an organic modifier (esters or lower alcohols [17]) to deactivate any unbounded silanol groups in the stationary phase, [18] thus causing the mobile phase to become flammable. The use of modifiers causes a high background signal and a loss of sensitivity [17]. To avoid the use of modifiers, open-tubular capillary columns can be used, since silanol groups are not present in the stationary phase [18].

Compared to capillary columns, packed columns display higher efficiency per unit time; also separations can be transported directly from analytical or preparative liquid chromatography (LC) to SFC. Moreover, a standard liquid chromatograph can easily be converted into a supercritical fluid chromatograph [19].

'It has been found that certain separations that were developed on a 50 μm i.d. capillary column can be repeated with the same or better performance on a 1 mm i.d. ("microbore") packed column. The packed column system has the additional advantage of yielding excellent peak area precision' [13]. It is also shown that the combination of water and formic acid is an effective modifier for CO2 which can be used with FID [13].

A study using the water and formic acid modifier was conducted by H. E. Schwartz et al. Formic acid is used as it has low background noise and therefore is more favourable than other modifiers. However, another problem arises when using formic acid where large gradient 'humps' appeared during the run, these were most probably because of organic impurities in formic acid. A way round this problem involves the addition of water to the carbon dioxide via the use of an 'aquafier' system; the 'aquafier' system used by H. E. Schwartz et al. was a "15 cm x 4.6 mm i.d. silica column (100-200 mesh) that was saturated with ca. 40% w/w water". The column was placed between the pump outlet and injection valve. A test mixture of the formic acid and water modifier was performed by H. E. Schwartz et al. and prodcued the chromatogram as seen in Figure 5 [13].

Figure 5 - Chromatogram of a test mixture of formic acid/water/CO2 mobile phase.

Peak identification (from left to right): n-eicosane, anthraquinone, n-triacontane, tocopherol acetate, cholesterol [13].

In Figure 5 the baseline rises this was due to the pressure program used, however due to the addition of water to the mobile phase which prevented the accumulation of formic acid on the head of the column no 'hump' is visible. In Figure 1 it can also be seen that all the peaks have good shape and resolution even for the more polar compounds like anthraquinone, tocopherol acetate and cholesterol [13].

Mobile phases and stationary phases used in SFC

In SFC the density of the mobile phase is about 200-500 times greater than that in gas chromatography. Compounds with high molecular weights are not usually detectable in gas chromatography, however with the density of the mobile phase being greater they can therefore be chromatographed [20]. A wide range of compounds has been tested for use as SFC mobile phases (Table 1). However, a variety of these required special conditions, and would therefore not be suitable. This resulted in carbon dioxide (CO2) being used as it is easily obtainable, low cost and safe[2], along with the critical temperature being 31oC and critical pressure being 73.8 atm[16]. A problem with CO2 as a mobile phase in a packed column is that if CO2 mobilizes a species then there is a possibility that the compound will be irreversibly adsorbed onto the column. This is because of the high activity of most sorbents; the problem can be avoided by the use of modifiers. This however, does not happen in capillary SFC as inert fused silica open-tubular columns are used [16].

There are two main reasons why modifiers are added to the mobile phase. First is that only a small amount of modifier is added in order to deactivate the sorbent active sites. Second is when the modifier is added in higher concentrations (level of modifier needed is ≥1%) it improves the solubility of the analyte in the mobile phase [16].

One problem with using modifiers is they have a high response when a FID is used; an alternative to using FID which helps relieve the baseline problem is the use of a ultra-violet (UV) absorption detector, as SFC mobile phases are UV-transparent, although it is not as applicable to organic compounds compared to FID [16]. This is only true for packed SFC, as when capillary SFC is used most separations are done using only CO2, which is compatible with FID. Having only CO2 as the mobile phase can cause slight defects on the chromatograms such as very broad peaks which are not well resolved, as well as longer retention times, this is solved by adding a small amount of water to the mobile phase, hence improving the peaks and decreasing the retention time [20].

Modifiers which can be used with the mobile phase include methanol, acetonitrile, chloroform and formic acid. Methanol is the most popular modifier being used in both packed and capillary SFC, even though the addition of water speeds up elution of polar compounds in capillary SFC [20]; methanol has a greater effect when used with silica-packed columns [16]. The solubility of methanol, acetonitrile and chloroform in CO2 was studied by K. L. Maguire and R. B. Denyszyn, they found out that when the pressure is below 90 for methanol/CO2 there was little effect on solubility, but when raised above 90 there was a substantial increase. Acetonitrile/CO2 had very little pressure dependence but small temperature dependence. Finally, with chloroform/CO2 both pressure and temperature had a small effect on solubility, when either was raised the solubility of chloroform increased [16].

Research by G. L. Pariente and P. R. Griffiths showed when carboxylic acid groups were present in the analyte the retention time was greatly increased while still using CO2 mobile phase. The cause of this could be because the solubility of these polar molecules is low and the solvation is not great enough to overcome the strong hydrogen bonds. The alternative mobile phase used was chlorofluorocarbon (CCl2F2), in comparison to CO2 which had a capacity factor greater than 20 for isophthalic acid; CCl2F2 had a capacity factor of 3.9. These results suggest that CCl2F2 has sufficient free energy of solvation to overcome the hydrogen bonds [16]. Alternatively, an additive could have been used such as trifluoroacetic acid (TFA). Additives are chosen from the nature of the compound in need of analysis. For example, TFA was used by J. Zheng et al. when they found that it was possible to elute peptides up to 40 mers in length using SFC, the use for TFA in their instance was to prevent the deprotonation of the carboxylic acid groups and help protonate the amino groups [21].

Even though CO2­ is the most extensively used mobile phase it is no more polar than hexane [22], so alternatives including CCl2F2 have been investigated. However, the critical temperatures must not be too high as one of the main advantages of SFC is that elution can take place at mild temperatures. Another example is ammonia (NH3), as it possesses a high dipole moment and relatively low critical temperature, however supercritical NH3 reacts with siloxane linkages and when left for an extended time the siloxane stationary phase for capillary SFC breaks down too[16]. Therefore, a more useful way of eluting polar compounds is CO2 ­and the use of a modifier [22].

For packed SFC more or less all of the stationary phases used in HPLC can be used in SFC, most of these are 'silica-based, chemically bonded or encapsulated, or polymeric' [15]. Evaluation of stationary phases of SFC was originally carried out by Schoenmakers et al. This was however, only done using pure CO2 as the mobile phase, and certain phases did not perform well, if a modifier was used these phases would have performed better [15].

When CO2 and a modifier are used as a mobile phase the stationary phase also becomes modified in that both CO2­ and the modifier adsorb onto the stationary phase. The stationary phase depends upon the level of absorption, for CO2 all phases adsorb the same but more polar phases adsorb more modifier than less polar phases. This causes the stationary phase to become more polar than the mobile phase, which in turn will cause polar solutes to interact more with stationary phase increasing retention time [15].

Other stationary phases that have been studied include, 'octadecylsiloxane-bonded silica (ODS), cyanopropylsiloxane- bonded silica, divinylbenzene-ODS, polydimethylsiloxane and porous graphitic carbon (PGC)' [15]. Some examples of achiral polar stationary phases include silica, cyano-propyl (CN) and 2-ethyl-pyridine (2-EP) which was specifically designed for SFC [4] (Figure 6).

Figure 6 - Examples of achiral polar stationary phases [4, 11]

In capillary SFC a problem arises in that normal GC stationary phases dissolve in the supercritical fluid mobile phase as they have a high solvating power. In order to correct the problem a non-extractable stationary phase is needed, examples of this are bonded phase where the stationary phase is attached to the column to surface groups via covalent bonds and cross linked phase where polymer chains within the stationary phase are attached to each other.

In order to create non-extractable stationary phase, the process of 'coating' must be undertaken, there are two types of coating, dynamic and static. The most favoured for SFC is static, as dynamic can lead to poor column efficiency and a thick stationary phase is not possible. In static coating the stationary phase is first dissolved in supercritical fluid and forced into the column, to avoid the removal of the phase cross link phase is used as it occurs between the polymers and not between polymer and substrate, and therefore can be applied to glass and fused silica columns[23].


In comparison to HPLC and GC, SFC has clear advantages. However, SFC is still not accepted by all industries and has only come into the major market over the past ten years. Mainly the pharmaceutical industry has adopted SFC, but others like the petroleum industry have taken an interest as it is useful for detection and analysis of biofuels and pesticides [4].

Supercritical fluid chromatography has advantages such as a mobile phase which is efficient, non-toxic and able to elute polar compounds without the use of modifiers. However, with the introduction of CO2 mobile phase it is possible to not only make industries benefit but also the environment by removing excess CO2 ­ ­from the air. The equipment used by SFC is not yet specific to this type of chromatography; but the modifications that have been made have resulted in raising the profile of SFC.