Injection Solvent Effects

The Problem
Chromatograms showing early eluting peaks which are distorted or split, plus later eluting peaks which are symmetrical and sharp suggest a specific cause -- the injection solvent might be stronger than the mobile phase. This strong solvent effect is demonstrated in Figure 10-1A. Here the injection solvent is 100% acetonitrile (100% strong solvent), and the mobile phase is considerably weaker, 18% acetonitrile and 72% water. The first peak is split and is noticeably broader than the second peak. When the injection solvent is changed to mobile phase both peaks are well-shaped and sharp, see Figure 10-1B.

FIGURE 10 Injection Solvent Effects A. Injection Solvent: 100% Acetonitrile B. Injection Solvent: Mobile Phase Column: ZORBAX® SB-C8, 4.6 x 150 mm Isocratic Conditions: 72% H20, 18% ACN Injection Volume: 30 µL Sample: 1. Caffeine  2. Salicylamide

The Explanation
When an injection is made, peak broadening will occur. The optimum composition and volume of injection solvent will keep this to 10% or less. In this example, when the injection is made, the local concentration of the injection solvent is not in equilibrium with the mobile phase. In other words, when a sample is injected the solvent is not immediately diluted with the mobile phase. Therefore some of the analyte molecules migrate down the column with stronger solvent, and other analyte molecules are diluted and migrate with the mobile phase, resulting in the split peak.

If injection size and solvent strength differences are small, solvent injection effects can be more subtle. The first peak may be broader than the second. If this broadening causes loss in needed resolution (Figure 10-2), this situation still demands attention. In Figure 10-2A, a short column is used, and a 5 µL injection is made. This is close to the optimum injection size of 4 µL. Using the same 5 µL injection size with a slightly stronger solvent causes extra band broadening that is visible as a loss in resolution, from Rs = 2.1 down to Rs = 1.5 (Figure 10-2B). A Rs of 2 or more is recommended for a rugged method with resolution between critical peaks that will remain even as the column ages, or with day-to-day method variation. A Rs of 1.5 is just baseline resolution, and any method variation and column aging could compromise the results. Doubling the injection volume (10 µL) causes an additional loss in resolution (Figure 10-2C) and would make the method unacceptable.

The Solution
The simplest solution to the problem is to dilute or dissolve your sample in the mobile phase before making an injection. This will allow you to make an injection that is 15 - 20% of the volume of the first peak. When using a 4.6 x 250 mm column for a peak with a k = l, the injection volume of sample dissolved in mobile phase will be < 30 µL. This will keep the band broadening due to injection to 10% or less.

The sample can be dissolved in a solvent weaker than the mobile phase as well. If the solvent is much weaker than the mobile phase, then the solutes are concentrated at the top of the column, and large injection volumes can be used. It is important to keep the analyte concentration low enough not to overload the column when a large injection volume is used.

If your sample must be dissolved and injected in a solvent that is stronger than the mobile phase, it is important to keep this volume low. Using a 4.6 x 250 mm, the injection volume should be no more than 20 - 25 µL (k of first peak = 1 - 3) if the injection solvent is somewhat stronger than the mobile phase. If the injection solvent is 100% strong solvent, then the injection size should be 5 - 10 µL.

To minimize peak splitting and band broadening effects extra sample preparation steps may be required. These may be as simple as diluting the sample with the injection solvent and injecting more of it; the result will be improved chromatographic efficiency, reproducibility, and accuracy.

The best recommendation is to completely dissolve your final sample in the mobile phase or a weaker solvent. If not, peak shape problems will persist.

Previous Section  |  Table of Contents  |  Next Section

MAC-MOD Analytical, Inc. -- info@mac-mod.com -- 1-800-441-7508