Liquid Chromatography Problem Solving and Troubleshooting
Author(s) -
Brian A. Bidlingmeyer
Publication year - 1996
Publication title -
journal of chromatographic science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.362
H-Index - 56
eISSN - 1945-239X
pISSN - 0021-9665
DOI - 10.1093/chromsci/34.10.472
Subject(s) - chemistry , troubleshooting , chromatography , computer science , operating system
Yes, columns change during use; this is why column-to-column reproducibility is of concern to most chromatographers. The instant you place the column in the instrument and start the mobile phase flowing, the column begins the "aging process" Sometimes this aging is slow, and other times it is fast. The degree to which the column changes during use depends on the type of column and the mobile phase that is used. The best way to ensure long life is to follow the manufacturer's operating instructions during use, but there are times when columns must be used in mobile phases that wil l cause "wear and tear" on the bonded phase and/or on the silica. In these situations, you wil l have to accept the change in column performance and should search for a column that wil l minimize the change. For instance, many reversed-phase column manufacturers recommend that columns be used at pH values from 3 to 8. The recommended pH range is intended to protect the column from encountering mobile phases that can degrade the performance. At acidic p H , the silane linkage of the bonded phase can be hydrolyzed. This is particularly troublesome for short-chain alkyl bonded phases like C 4 groups and lower. This behavior wil l vary from manufacturer to manufacturer and wil l depend on how the alkyl silane is attached to the silica. If the silane is attached using a dimethyl alkyl group, the bond is more prone to hydrolysis than if the silane has protecting side groups. The demonstrated stability in Figure 1 is for a cyano column that is often reported to be unstable compared to a C 8 or a C 1 8 . As seen in Figure 1 , the problem is in the lability of the alkyl linkage to the silica and not the cyano phase itself. The change in capacity factor represents a loss of stationary phase; one of the bonded phases hydrolyzes, and the other is stable. An extrapolation of the findings that short chain groups are labile at low pH is that endcapping groups wil l be cleaved from the silica at pH values less than 3. This is a commonly believed phenomenon, and although a completely bonded phase of C 3 wi l l lose phase at low p H , a study of the loss of only the endcapping group in the presence of a longer alkyl chain group has not been done. The stability of phases at low pH is particularly important to those researchers using trifluoroacetic acid (TFA) in their mobile phases. It may be worth investigating several columns to see if the stability might be improved using a different column bonding type. At the basic pH range, the column can change during use, not due to silane hydrolysis, but to the dissolution of the silica. Even at a pH value of 7, the column can change. The change is generally believed to be the result of the creation of additional hydroxyl groups as the silica is dissolved and exposes more silanols. Compounds sensitive to the silanol population wil l exhibit retention time changes and often peak shape changes. An example of this is shown in Figure 2. In this example, two C 8 columns are compared. O n e has a retention time change of 5 5 % over a 125-day exposure to a phosphate buffer mobile phase with a pH of 7, whereas the other column changes only 10%. In this example, there were two main differences between the two columns. O n e difference was the type of silica. The more quickly degraded material was a xerogel type of material, and the other was a solgel type of silica. The second difference was that the more quickly changing column was singly endcapped, whereas the other column was doubly endcapped. Clearly, both columns "aged" during use, as exemplified by the change in the relative capacity factor of the analyte. To use an analogy of steel rusting, steel is painted to keep the "corrosive agent" away from the steel. In Η P L C , the endcapping may be thought of as a coating (like the paint), and a double endcapping may be a more effective coating to protect Figure 1. Stability of alkylcyano bonded phases at low pH. Relative retention of toluene as a function of mobile phase (column) volume pumped across the column. Two different silanes were used to make the bonded phase. Top line: an isopropyl-alkylsilane was used. Bottom line: a dimethyl-alkylsilane was used. Conditions: pH = 2.0 at 50°C.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom