Turning up the heat
4 Oct 2012 by Evoluted New Media
David Collins explores why High Temperature Liquid Chromatography (HTLC) has become such a hot topic, the benefits, precautions and risks that must be considered when applying elevated temperatures to a separation
Historically, the use of elevated temperatures in liquid chromatography has not generated much interest and certainly it was not widely used, even though it was known to be an important separation parameter. Two exceptions to this were its application in size exclusion and ion-exchange chromatography1. For the most part, the role of temperature in the chromatographic process was often one of simple (or convenient) thermostatting rather than being employed as a useful chromatographic tool. Part of the reason for this was due to the lack of commercially available column ovens capable of high temperatures and more significantly, it was because reliable and thermally stable stationary phases simply didn’t exist. Plus, during the latter half of the last century pumping technology was developing quickly and improvements in pumping control and mixing was enabling the use of reproducible, compositional mobile phase gradients with the result that high temperature separations attracted very little attention. As recently as 1996, a publication reported that half of all HPLC systems being sold did not come with a column oven2.
In the past few decades, column technology has moved on considerably and in recent years there has been renewed interest in HTLC, particularly given the growing popularity of microbore and capillary scale LC. The low thermal mass of these small columns allows for almost instantaneous thermal transfer and today the use of higher than ambient temperatures is recognised as a valuable tool in LC, particularly in reversed-phase separations. There have been some excellent reviews and book chapters published on the subject of high temperature LC1, 3-12 but why has it become such a hot topic, what are the benefits?
The most obvious advantage of running a separation at higher than ambient temperatures means that there is reduced backpressure due to a decrease in mobile phase viscosity. This drop in pressure means that the user can run at higher flow rates, therefore increasing the speed of the separation. Alternatively this drop in pressure might allow the user to use longer columns with higher plate numbers. In addition to this, there is an inverse relationship between temperature and analyte retention (there are a few exceptions), particularly evident in reversed-phase LC, and so by using higher temperatures a user may elute analytes faster. Faster flow rates and lower retention factors mean that run times can be considerably shorter. Figure 1 shows a comparison of the retentions of some alkylbenzenes at different temperatures and flow rates.
[caption id="attachment_30435" align="alignright" width="200" caption="Figure 1. Separation of five alkylbenzenes (toluene, ethylbenzene, propylbenzene, butylbenzene, and pentylbenzene) at varying temperatures and flow rates: (red) temperature 85 °C, flow rate 4 ?L min-1, (orange) temperature gradient 25 – 85 °C from 3.5 min to 6.5 min, flow gradient 1 – 4 ?L min-1 from 4.3 min to 8.0 min, (green) temperature 25 °C, flow gradient 1 – 4 ?L min-1 from 5.3 min to 6.3 min, (blue) temperature 85 °C, flow rate 1 ?L min-1, (purple) temperature gradient 25 – 85 °C from 3.5 min to 6.5 min, flow rate 1 ?L min-1, (black) temperature 25 °C, flow rate 1 ?L min-1. LMA-EDMA monolithic column, 150 mm x 100 µm I.D.; mobile phase 50:50 ACN/H2O"]
[/caption]
Apart from decreasing analysis times, the use of high temperature has many other significant effects. Analyte diffusivity increases at higher temperatures which improves column efficiency due to improved mass transfer. Resolution too can be much improved, especially in the case of large molecules, and in many instances the analyte signal-to-noise ratio can also be increased. Selectivity can be altered and so peaks that co-elute at room temperature can often be separated at higher temperatures or through the application of thermal gradients (also known as temperature programming). Isobaric (constant pressure) separations are also possible using temperature programming and this offers an excellent alternative to solvent gradient elution where the required range of the elution strength is not very wide. Higher temperatures also reduce the dielectric constant of water, so hotter water begins to take on the characteristics of an organic solvent. This means that as higher temperatures are used, less organic solvent is required in the separation. Indeed, a growing area of HTLC is ‘green’ chemistry, where only pure water is used as the mobile phase for reversed-phase separations8-9. Several comparisons13-15 of solvent composition against temperature have shown that a 1% increase in MeOH concentration was equivalent to an increase in temperature of around 4ºC. Other comparisons have been made with acetonitrile-water mobile phases with similar results. Of course, there is nothing stopping us from using both solvent and temperature gradients together, and indeed the simultaneous use of the two is considered to be a highly effective way to control peak resolution and selectivity. In general, it is the entropy dominated separations that will benefit most from temperature programming over solvent gradient programming.
In addition to the many advantages of applying high temperature to the separation itself, there are also some other positive aspects which are not so obvious, particularly concerning the detector. Often, in the course of HTLC, post column cooling is required prior to UV or fluorescence detection; however there have been many studies done which show that high mobile phase temperatures improved detector response, particularly for evaporative light scattering detectors. High eluent temperatures are also of benefit when using LC-MS and studies have shown many advantages including reduced run times and improved sensitivity and detection16-17.
So, why isn’t HTLC more popular, particularly in industry? There continues to be a widespread reluctance to use high temperature mainly due to the popularity of silica based stationary phases and their intolerance of elevated temperatures. Silica based phases have a reputation as being notoriously unstable when used at elevated temperatures, and in some cases even short term exposure to high temperatures will permanently degrade the column. High temperature often increases the solvating strength of the mobile phase, and non-bonded coatings may be easily removed. The underlying supports of the stationary phase can also be attacked causing the structure to either collapse or simply wash off the column (column bleed). Occasionally a column may be conditioned by cycling the column temperature under flow or by flushing the column with a strong mobile phase at high temperature. However, depending on the type of column there may be a loss of stationary phase, reducing efficiency and ultimately resulting in the destruction of the column. Having said this, there are many columns now commercially available that have excellent thermal stability and which demonstrate excellent reproducibility when used at elevated temperatures. In addition, even separations done on silica columns which are prone to degradation at elevated temperatures can often benefit from temperature programming provided the upper temperature limits are not excessive. Over 100 ºC most silica based C18 phases will rapidly degrade; however, once the temperature is kept below 70 ºC most reversed-phase columns will experience no problems whatsoever. On the other hand, phases such as cross-linked polystyrene, graphitic carbon, and zirconia (to name a few) have demonstrated excellent stability at temperatures exceeding 240 ºC.
Analyte stability may also be affected by the application of high temperatures particularly if the analyte in question is a thermally labile species. Many compounds (especially biomolecules and bio-active compounds) will undergo rapid degradation at elevated temperatures and this is probably the main reason that HTLC has yet to be accepted by the wider pharmaceutical industry. However, the occurrence of on-column degradation depends not only on the reaction rate of the analyte at the given conditions, but also the length of time which the analyte is exposed to these conditions. Theoretically, any on-column reaction might be rendered insignificant if the increase in the analytes reaction rate at high temperatures was offset by the reduction in the time that the analyte spends in the column18. In other words, although the compound might degrade at the applied temperature, the combined effects of reduced retention and higher flow rate might well result in the analyte eluting before it has time to degrade. A recent work illustrated this particularly well by separating three proteins at 120 ºC on a silica based column using a mobile phase consisting of 0.1% TFA19.
At this stage you might be asking yourself whether you too can implement HTLC in your separations and indeed you probably can however, there are a number of considerations that must be taken into account besides the stability of the column and analytes in question. In order to minimise band broadening the mobile phase should be pre-heated to within at least 6 ºC of the column temperature and this is critical for columns larger than 2 mm in diameter. Likewise, at the column exit, the eluent should be temperature controlled depending on the mode of detection being employed. For most standard optical detectors the mobile phase may need to be cooled prior to entering the detector. At higher temperatures a pressure restrictor (usually just a coil or section of narrow bore tubing or capillary) will often be required between the column and detector, the purpose of which is to prevent the mobile phase from boiling before it has a chance to cool sufficiently. If the column ID is too large (typically >2 mm) then there will be considerable thermal lag, so rapid temperature programming will not work effectively and will probably cause band broadening (this can occur due to axial and/or radial temperature gradients within the column). Smaller bore columns have a much lower thermal mass and so will thermally equilibrate quicker allowing the use of much faster temperature gradients. These smaller columns also lend themselves well to sample focusing at the head of the column – this is achieved by holding the column inlet at a low temperature, typically less than 5 ºC. This effectively allows solute enrichment at the head of the column and previous works have shown sample volumes to be increased by a factor of 103 simply by employing this method20-23.
Nowadays most commercially available LC systems come with some degree of column heating, usually in the form of a column compartment, however some instruments also offer extended compartments which house not only the column, but also the injector and inlet tubing. The vast majority of these column ovens are of the air bath type, blowing heated air into the column compartment. Other types of column oven include block heaters and water jackets, however air bath ovens are by far the most popular, although in practice they are the slowest and most inefficient. Direct contact heating, either through heating blocks or water jackets are by far the most efficient due to their superior heat capacity, however they are usually limited in their temperature range and a prohibitively slow rate of heating and cooling when used in temperature programming applications. Rates of heating typically range from 5 – 10 ºC min-1 for air bath ovens to around 30 ºC min-1 for direct contact heaters. Given the fact that capillary and microbore scale HPLC lends itself well to rapid separations it makes sense that column heating equipment should at least be able to keep up with the separation in terms of performance, however, the slow performance of most column heaters makes them unsuitable for use with rapid temperature programming. In addition to their slow response, the bulk of these column ovens are further limited by their temperature range, which is usually 5 or 10 ºC below ambient to 80 or 90 ºC.
[caption id="attachment_30436" align="alignleft" width="200" caption="Figure 2. TEC capillary/microbore scale column heater developed by the Irish Separation Science Cluster"]
[/caption]
Another breed of column heater (see Figure 2) has recently been developed within the Irish Separation Science Cluster, ideally suited to capillary and microbore scale columns24. This type of column heater is based on an array of thermoelectric (TEC) modules, essentially small microchips made up of hundreds of P-N junctions, allowing the column to be heated or cooled at rates exceeding 400 ºC/min making it ideal for use with rapid separations. Furthermore, the temperature range of this heater is very wide, from -15 ºC to 200 ºC and it can accommodate columns up to 5 mm in diameter and 225 mm in length. Direct contact heating/cooling blocks are coated with a thick layer of flexible thermally conductive silicon and so create excellent thermal contact with whatever type or size of column (or indeed fittings) used.
The next decade promises to be very exciting with regard to high temperature LC, particularly as the boundaries of what is possible are pushed further and further. Although the application of elevated temperatures may not become common place across all modes, it is most likely, given that it is such a useful tool, that temperature programming will become the norm. Having said this, there is a worldwide push towards reducing solvent consumption and separations carried out at high temperature is the logical solution. With the market moving towards greener and faster separations we will probably see the sizes of analytical columns becoming smaller and smaller, more suited to rapid temperature programming and high temperatures. In addition, with mass spectrometers becoming more affordable, we can expect an increase in the use of LC-MS in analytical labs, a technique which benefits from the application of elevated temperatures.
References
1. Greibrokk, T., Andersen, T., Journal of Chromatography A, (2003), 1000, 743 2. Zhu, P.L., Dolan, J.W., Snyder, L.R., Djordjevic, N.M., Hill, D.W., Lin, J.T., Sander, L.C., Van Heukelem, L., Journal of Chromatography A, (1996), 756, 63 3. Jones, B., Journal of Liquid Chromatography & Related Techniques, (2007), 27, 1331 4. Tan, I., Roohi, F., Titirici M.M., Analytical Methods, (2012), 4, 34 5. McNeff, C., Yan, B., Stoll, D., Henry, R., Journal of Separation Science, (2007), 30, 1672 6. Vanhoenacker, G., Sandra, P., Journal of Separation Science, (2006), 29, 1822 7. Guillarme, D., Heinisch, S., Separation & Purification reviews, (2005), 34, 181 8. Coym, J., Dorsey, J., Analytical Letters, (2005), 37:5, 1013 9. Smith, R., Journal of Chromatography A, (2008), 1184, 441 10. Teutenberg, T., Analytica Chimica Acta, (2009), 543, 1 11. Heinisch, S., Rocca, J.L., Journal of Chromatography A, (2009), 1216, 642 12. Teutenberg, T., High Temperature Liquid Chromatography – A User’s Guide for Method Development, RSC Publishing, 2010 13. Chen, M.H., Horvath C., Journal of Chromatography A, (1997), 788, 51 14. Bowermaster, J., McNair, H., Journal of Separation Science, (1984), 22, 165 15. Tran, J.V., Molander, P., Greibrokk, T., Lundanes, E., Journal of Separation Science, (2001), 24, 930 16. L. Pereira, S. Aspey, H. Ritchie, J. Sep. Sci. 30 (2007) 1115. 17. M. Albert, G. Cretier, D. Guillarme, S. Heinisch, J.L. Rocca, J. Sep. Sci. 28 (2005), 1803 18. Antia, F.D., Horvath, C., Journal of Chromatography, (1988), 435, 1-15 19. Yang, X., Ma, L., Carr, P.W., Journal of Chromatography A, (2005), 1079, 213 20. P. Molander, K. Haugland, D.R. Hegna, E. Ommundsen, E. Lundanes, T. Greibrokk, J. Chromatogr. A 864 (1999) 103. 21. P. Molander, A. Holm, E. Lundanes, E. Ommundsen, T. Greibrokk, J. High Resolut. Chromatogr. 23 (2000) 653. 22. P. Molander, A. Thomassen, E. Lundanes, G. Fladseth, S. Thorud, Y. Thomassen, T. Greibrokk, J. Sep. Sci. 24 (2001) 947. 23. B.A. Ingelse, H.-G. Janssen, C.A. Cramers, J. High Resolut. Chromatogr. 21 (1998) 613. 24. Collins, D., Nesterenko, E., Connolly, D., Vasquez, M., Macka, M., Paull, B., Analytical Chemistry, (2011), 83, 11, 4307
The author: David Collins David works as part of the Irish Separation Science Cluster (ISSC) based at Dublin City University. At LAB INNOVATIONS he will be delivering a talk on ¬emerging technology from within the group, specifically in instrumentation development.
- easyFairs LAB INNOVATIONS 2012 is being held on the 7th and 8th November at the NEC, Birmingham and is the UK’s only event dedicated to showcasing the latest laboratory technology and consumables, analytical and biotech equipment.
Not to be missed features at the show include: • A top flight conference programme designed by the Royal Society of Chemistry and delivered by experts The conference will focus on analytical chemistry – including the latest advances in measurement, life sciences and environmental research – health and safety, plus continual professional development. • A free-to-attend seminar programme devised by Campden BRI on micro and analytical topics within food and beverage. • A major exhibition with 12 worldwide product launches or UK debuts already confirmed. • A demo area for the latest technologies. • Lions’ Lair where companies will pitch their most innovative products to a panel of experts from the UK’s leading trade bodies.
To find out more about visiting the show go to www.easyFairs.com/LABINNOVATIONS
