|
Home > Magazine > Other Techniques
Reviving Thin Layer Chromatography
TLC lends a helping hand to HPLC workflowsFredric Rabel
Today, chromatographers know about thin layer chromatography (TLC), but most have not used it since their undergraduate days. Although devices are available to automate some TLC tasks, the majority of the steps used in TLC are mainly manual, thus accounting for its minimal use in the analytical laboratory where automation is so important to productivity and record keeping.
 Figure 1: TLC plates spotted with 40 + samples, using opposite ends of the plate. | Still, the sales of TLC products are as strong as ever, meaning it is being used in many laboratories, albeit in a minor capacity. TLC users are just not discussing their work or publishing new applications. Although many people might think a TLC method cannot be validated, such is not the case. As with GC or HPLC, TLC also demands attention to detail, the correct equipment and well-trained personnel to generate good, reproducible results.
It's likely that researchers will not be making a mass move to TLC. Regardless, this article will discuss some of the possibilities for using TLC in the HPLC laboratory. The various TLC uses discussed will be to:
1. aid in sample preparation
2. aid in choosing a HPLC mobile phase, and
3. aid in selecting regeneration or cleaning solvent(s) for an HPLC column.
Every reader will have to decide for himself whether to incorporate one or another of these techniques. Each possibility has merit for its use and should at least be considered in helping develop a more robust HPLC separation.
As an aid in sample preparationMost HPLC methods fail because extraneous materials get onto the top of the analytical column. The results are decreased resolution, band broadening and an increase in back pressure. Considerable time is spent on sample cleanup in the initial stages of the method development. Unfortunately, there is no way to know just how good this has been without having a few hundred samples injected onto the column of choice with the optimized mobile phase. Doing a study of column destruction or failure with so many samples is seldom done. Only after the protocol has been put into actual use is such data available.
To simplify the discussion, this article will assume that the sample is a drug being extracted from plasma. Thus, it has to be deproteinized and then placed on various C8 or C18 solid phase extraction (SPE) tubes from a number of manufacturers. Each of these SPE bonding chemistries is as different as the bonding chemistries are for HPLC columns, so they will often give very different extraction efficiencies. The second or third choice may be the one that wins out after the extraction has been optimized for the components of interest. Also assumed is that the developing solvent for the TLC separation is known, or can be found in the literature (see References at the end of this article). One last requirement is that a visualization reagent is known for the compounds needed to be seen. A Web source for visualization information online is also provided below.
The extracts can be easily be tested on a TLC plate to see what elutes after the capture, wash and release cycles. A large matrix of solvent combinations, buffer strength and pH changes can be made and tried sequentially. Collected samples after these steps can be spotted against standards to see approximate amounts and placement of drugs and metabolites. Analyzing all of these sample (there could be hundreds) generated from such a matrix is more rapid using TLC (which can hold up to 40+ samples per plate, as in Figure 1). Placing them one after the other on an HPLC column would take considerable time, even if the analysis time allowed is 10 minutes. Using both ends of a TLC plate also cuts the cost per analysis or per plate in half. Development is done sequentially from one end, then the other. Any notes/sample numbers can be written on the back of the support with an appropriate marker.
 width="200" height="108"> Figure 2: Chromatographic differences between TLC and HPLC | At this point in the method, where little is known about the impurities that might be present, using any HPLC column would eventually result in high backpressure. The backpressure problem might not arise from the particulates since the samples are always filtered before injection, but to unknown components that would be sorbed, often irreversibly, onto the guard or analytical column.
With the use of smaller (3 and sub-2 µm) particle size columns, such clogging and backpressure can be an even greater problem. It might be noted that monolithic silica columns with their greater porosity and low backpressure, suffer much less from these problems, often lasting three to four times longer.
If applying these same impure extracts to a TLC plate, no damage is incurred since the plate is only used once. TLC also has the advantage of not hiding anything on the plate from the origin to the solvent front. By using suitable visualization techniques (specific reagents fluorescence or fluorescence quenching), any components of the sample can be seen.This is unlike a stainless steel or polymer column into which the HPLC packing is placed. Nothing can be seen, only deduced from comparison of an initial pressure reading compared to later readings after a number of samples have been injected.
 Figure 3: Relationship between TLC Rf values and HPLC capacity factor | The TLC sample can be put into any solvent for spotting since it is evaporated off, allowing a wide range of solvents to be used to obtain a clear solution. Remember that a cloudy solution means something is insoluble or immiscible and not a good sample for either TLC or HPLC. Most likely, any other solvent combination used for the separation will also not dissolve many (or any) of these insoluble or immiscible components.
Any components that do not dissolve in the mobile phase will be found at the sample spot origin. Since this is the sample cleanup stage, you must ask yourself: are the spots that are moving in the TLC layer the only components needed for separation and/or quantitation? Are there other important compounds remaining behind at the origin? Only standards, intermediates and decomposition products placed on the plate will tell you whether or not you have completed the optimization of the sample preparation.
The cleanup attempts should yield the least amount of impurities so that they do not hinder the finally analysis. Any information observed on the plate during this sample cleanup phase is also important to the TLC help in cleaning or regenerating the HPLC column, to be discussed below.
As an aid in choosing an HPLC mobile phaseThe information gathered from TLC separations, along with the solvents used to make the separation possible, can also be used to lead to the final optimized mobile phase for the separation. There are, however, many differences between the separation that occurs on a TLC plate and in an HPLC column, as illustrated in Figure 2. Perhaps the most important is the difference between the dry plate and any vapor effects possible with it versus the pre-wetted HPLC column packing.
 Figure 4: Illustration of tailing and bearding spots on a TLC plate | The method to develop a suitable mobile phase for HPLC from TLC work is to spot the sample on a plate and use the most polar combination possible. For a reversed phase separation on a bonded RP18 plate, this would be to use a high proportion of the organic solvent. Assume methanol/water combinations were to be tested first.
Attempt the separation with pure methanol after spotting the sample. All the components will move with the solvent front, meaning the mobile phase is too rich in methanol (the stronger eluent of the two), and no separation is possible. This initial run with methanol is made to ensure little is left at the origin, thus preserving the integrity and usefulness of the guard and analytical columns. It also ensures that there will be few, if any, late-eluting compounds. These can cause problems when they elute later during the second or third injection if not enough time has passed when using the weaker eluting mobile phase. This initial attempt at optimizing the mobile phase is followed sequentially on another plate (use the 1 × 3” plates for this beginning optimization). Then, develop with methanol/water (95:5), continuing this increase in water content until a good separation is seen. Generally, this means all the spots are spread over most of the TLC plate, from the origin to the solvent front.
However, because of the difference in the mechanism of TLC and HPLC, the mobile phase should be adjusted so that most of the compounds (or the ones most critical to a successful separation) are in the lower half of the TLC plate. This will keep the compounds longer on the HPLC column for a better separation. If the compounds elute too fast, they will be found with less separation between them and closer to the solvent front.
Rf values (obtained from a TLC separation) and k’ (capacity factor, as measured from an HPLC separation) are reciprocals of one another. The higher the Rf, the shorter the k’, and the less likely it is that a separation will occur. Lower Rf values will give moderate k’ values, allowing the compounds more time for more sorption/desorption steps to occur to yield a better separation.
 Figure 5: Nomogram for conversion of binary solvent systems in reversed phase separations | Most chromatographers have the standard TLC plate in their laboratories, made up of silica/bonded silica, with an average particle size of 15 µm. On this layer, the spots will not be as compact and separations not as fast, compared to the HPTLC plate, since you develop to 10 cm from the origin to obtain sufficient resolution. The advantage of the HPTLC plate is this compactness of the spot, and the shorter 5 cm developing distances needed. If using a HPTLC plate made with an average particle size of 5 µm silica gel, it will not have the efficiency of a 5-µm HPLC column. Such layers are spread onto a plate with gravity being the only force to compact the particle bed. HPLC columns are packed at very high pressures, giving much denser chromatographic beds. This results in less band broadening during the separation process to generate higher efficiency. Thus, you might see spots overlapping on a TLC plate, but the equivalent HPLC column will give much better resolution, providing baseline separation.
Often, tailing (spot with a comet-like tail) or bearding (with the ‘tail’ leading the spot) is seen in a TLC separation. This is due to some ionization of the acidic or basic groups in the chemical structure of one or more of the components in the sample. Adding 1 to 2% of an acid (usually phosphoric or acetic acid) or a base (ammonium hydroxide or trimethylamine) to the mobile phase suppresses the ionization to make the spots circular.
Once a suitable TLC solvent combination of methanol/water has been found that should also work on a HPLC column, one additional step should be tried: substitute one of the other organic solvents recommended for reversed phase; either acetonitrile (AcCN) or tetrahydrofuran (THF) can be used. The nomogram shown in Figure 5 can be used to give approximately the same solvent strength and location on the TLC plate or through an HPLC column. Just move a straight edge to the solvent combination that has worked best for a separation, and try the other solvent combinations to see if they yield better results.
Then, of course, the final optimization is done on the HPLC column or columns being tested for the final method. At this point some other combinations of the three solvents mentioned above for the reversed phase mode can be tried. The addition of an acid, base or buffer is also critical at this point to suppress ionization to get narrower and more symmetrical peak, since the separation mechanisms and how they are affected by ionization are equivalent in the two techniques.
As an aid in selecting regeneration or cleaning solvent(s) for an HPLC column
 Figure 6: Using TLC to find a solvent combination for HPLC column cleaning | Recalling some of the statements made above when optimizing sample preparation, it is desirable to move all of the components in a sample away from the origin on the plate. This is particularly important after the optimization of the sample preparation and the sample is dissolved and developed in the mobile phase. This should also have yielded no remaining components at the origin with a final test on a TLC plate.
For the longest use of any HPLC column, any components caught on the analytical column (setting aside the guard column for this part of the discussion), will compromise the inlet bed of the column. These might be particulates from the sample or components that become insoluble in the mobile phase. Either will add to back pressure, peak broadening and/or decreased efficiency.
As mentioned, most HPLC columns are constructed of stainless steel or a polymer. Thus, inspection of the inlet bed and anything it has collected is impossible, except when the inlet fitting and frit are removed. However, from previous work with TLC, any remaining components would have been seen at the origin with the mobile phase used for the HPLC separation.
It is acceptable to have insoluble components collect at the inlet of a column if they are not necessary for quantitation, and if they can be easily solublized and washed from the HPLC column. This is possible with many compounds, with the exception of proteinaceous materials. Using the TLC method as above, continue to add samples to the mini-plates and try different solvent combinations to see what might elute the insoluble components. In the example shown in Figure 6, the original mobile phase for the reversed phase separation is methanol/water (60:40). The separation is seen in the far left chromatogram. The middle chromatogram shows that methanol alone still does not remove the materials at the origin. The right chromatogram shows that this material (one or more components) can be removed with methanol and some acetic acid added to it.
If no sample combination can remove this material, then it is imperative to use a guard column. This protector column ahead of the analytical column will collect this material. After the back pressure has reached a few hundred psi greater than at the beginning of its use, then the guard column is replaced with a new guard column. Never attempt to clean a guard column since it is not possible and is a waste of time and solvent.
TLC can be used as an aid in the HPLC laboratory. It can speed up the analysis of samples when optimizing the sample preparation of a complex mixture or biologically derived sample. TLC can also help determine the solvent combinations that may possibly work for the HPLC separation. Lastly, it can show what is not moving or eluting from the HPLC column, since everything is shown on the TLC plate, from the origin to the sample front.
ReferencesApplied Thin-Layer Chromatography, Best Practice and Avoidance of Mistakes, 2nd Edition, E. Hahn-Deinstrop, Wiley, NY (2007) (ISBN: 978-3-527-31553-6).
Thin-Layer Chromatography — A Modern Practical Approach, Peter E. Wall, Royal Society, London (2005).
3. Handbook of Thin-Layer Chromatography, 3rd Edition, J. Sherma & B. Fried, Dekker, NY (2003) (ISBN: 0-8247-0895-4).
4. Milestones in TLC, J. Sherma, EMD Chemicals TLC Catalog, 2005, pp 2-9 (EMD Chemicals, Inc., Gibbstown, NJ).
5. TLC Visualization Reagent PDF:
http://www.emdchemicals.com/analytics/literature/TLC_Visualization_Reagents.pdf
Laboratory Equipment
Rockaway NJ 07866
Company’s Other Products
|
|
|
|
|