The choice of the stationary phase for a given separation problem is the most difficult decision in TLC, especially, of course, for beginners. The most important aspect of this is to find a suitable selectivity. That is to say, the chemical composition of the station- ary phase, and in particular that of its surface, must be suitable for the task. To obtain satisfactory separation efficiency, the mean particle size, the particle size distribution and the morphology of the particle must be considered. The support for the stationary phase should be as well suited for the intended application as the additives in the layer.
Some attention should be paid to the cost-effectiveness of the analysis. For example, rather than preparing a large number of chromatograms individually as and when re- quired, it may well be advisable to secure better reproducibility by using proprietary precoated solvent systems from a commercial manufacturer who provides a certificate guaranteeing the quality of his product.
27 2.5 Criteria for the Selection of Stationary Phases in TLC
2.5.1 How Can the Choice of the Stationary Phase Be Made?
The first step when tackling any new problem should be to carry out a literature search. Here, not only can articles on TLC be helpful, but sometimes publications from other areas of chromatography, e.g. HPLC, can provide exactly the information needed to solve the problem. The following example is from my own experience. An HPLC publication on the characterization of some extracts from medicinal plants used in cosmetics yielded the information that some amino acids in Althaeae officinalis (marshmallow) root showed a certain “fingerprint”. I therefore proposed a possible method of identifying this extract, whose identification would otherwise be very diffi- cult. A quantitative determination of alanine from a dry extract of marshmallow root in an instant tea was also possible after optimization of this HPLC system by means of TLC (see also Sections 7.2.1 and 7.2.2.1).
The results of the literature search serve as a basis for preliminary investigations.
Without such information one usually starts with aluminum foil silica gel 60, from which suitable pieces can quickly be cut out. After this, if necessary, other layers can can be tested to optimize the resolution and the hRf values. For example, according to the literature, plates precoated with silica gel 60 were often used for the analysis of sandalwood, but the chromatograms obtained were not very good. Many other sor- bents were therefore tried, but without success. Only the use of HPTLC precoated plates of the type NH2F254with the solvent systemn-butanol + formic acid + water (19 + 2 + 5) gave good results in the identification of sandalwood, a dye additive pro- hibited in Germany [75].
2.5.2 How Can the Recommendations for Stationary Phases Found in Pharmacopoeias Be Applied to Precoated Layers?
In the preparation of monographs, members of Commissions have not wished to men- tion any proprietary products so as not to appear to show any preference, and descrip- tions are mainly of plates prepared in-house from loose sorbent materials. Conse- quently, uncertainties nowadays frequently arise when trying to follow recommenda- tions in pharmacopoeias. For example, in many monographs in the German pharmaco- poeia (DAB) it is stated that the test parameters “identity” and “purity” can be tested for by TLC (Method V.6.20.2) using a layer of silica gel GR, G F254R, H F254Retc.
Method V.6.20.2 prescribes the use of either precoated or homemade plates, but al- though the silica gels mentioned are reagents described in the DAB, there is no infor- mation about the corresponding precoated plates. Plates precoated with the silica gel sorbents GRand G F254Rare not manufactured commercially (at least in Europe), as the prescribed proportion of gypsum (13 %) leads to very soft layers, so that the plates are unsuitable for transportation and present severe handling problems. In the case of coatings of silica gel HRand H F254R(“with no foreign binders”), these problems are even more severe, so that these materials also cannot be obtained as precoated layers.
The question of which is the corresponding precoated layer therefore arises in rou- tine laboratory work. General recommendations on the subject can only be made with 28 2 Precoated Layers
some reservations, as a complete rearrangement of the chromatographic system, in- cluding not only the precoated layer but also the solvent system and other parameters, is not always readily feasible. Many examples from pharmacopoeia monographs are given in 11 Tables in this book (Tables 4, 6, 9–12, 14, 15, and 17–19) in which a com- plete change of all the conditions has sometimes led to optimum results.
Practical Tipsfor the choice of precoated layers based on pharmacopoeias:
In the case of all sorbents for which DAB recommends the silica gels GR, G254R, HRand H F254Rand which should have a mean grain size of ca. 15 m, preliminary investigations can be made using TLC silica gel 60 with or without fluorescence in- dicator (FI).
Chromatographic silica gel R(DAB) is a very fine grade with a grain size in the range 3–10 m. For this, the HPTLC silica gel 60 precoated layers with or without FI are recommended.
If the chromatographic silica gel cyanopropylsilylated R is recommended, the HPTLC plates CN F254scan be used.
For the chromatographic silica gel octadecylsilylated R (3–10 m), the HPTLC plates RP-18, RP-18 W, RP-18 F254s, RP-18 WF254sand RP-18 with concentration zones give good results.
If the chromatographic silica gel octylsilylatedRis recommended, HPTLC preco- ated plates RP-8 F254scan be used.
Silica gel H and H F254, silanizedR, can be replaced by HPTLC precoated plates RP-2 F254s.
Silica gel anion exchangerR is a very fine silica gel (3–10 m), whose surface is chemically changed by modification with quaternary ammonium groups. In this case, good chromatograms can be obtained on HPTLC plates NH2and NH2F254s. In the case of, e.g., propyl-4-hydroxybenzoate (and all other PHB esters according to Ph Eur), a suitable octadecylsilanized silica gel that contains a fluorescence indi- cator with very intensive fluorescence excitation at 254 nm is:
DC-RP-18 F254s: Merck 1.15389 20 × 20 cm Merck 1.15423 10 × 20 cm
29 2.5 Criteria for the Selection of Stationary Phases in TLC