Why Does My HPLC Baseline Drift?

A drifting baseline is one of the most common frustrations in HPLC. Instead of holding steady, the detector signal creeps steadily upward or downward across the run, making peaks harder to integrate and sometimes burying small ones entirely. The good news is that baseline drift almost always has a traceable cause. This guide walks through the usual suspects and how to fix them systematically.

If you are new to interpreting the detector trace, it helps to first understand what a healthy signal looks like — see our guide on how to read an HPLC chromatogram.

What baseline drift actually is

Baseline drift is a gradual, unintended change in the detector signal over the course of a run — distinct from sharp, jittery noise. An ideal baseline is flat, low in noise and free of artefacts such as ghost peaks; the closer the real baseline comes to that ideal, the more confidently you can quantify at low concentrations [1]. Drift can run positive (upward), negative (downward), or follow a slow cyclic pattern. The direction and shape of the drift are your first diagnostic clues, so always note them before you start changing things — separating “drift” from “noise” is the first step in narrowing the list of causes [1].

Cause 1: Temperature fluctuations

Small changes in column or mobile phase temperature are a classic source of slow, often cyclic drift. Many detectors — particularly refractive index detectors and UV detectors at high sensitivity — are sensitive to temperature [2].

The fix: House the column in a temperature-controlled column oven and keep it away from drafts, direct sunlight or air-conditioning vents. Stable temperature is one of the simplest ways to steady a baseline. This is also why retention times shift when temperature is uncontrolled — the two problems often appear together.

Cause 2: Mobile phase problems

A non-homogeneous or contaminated mobile phase is one of the most frequent culprits, usually producing an upward drift in absorbance. Solvents and additives can contain impurities that are strongly retained under the conditions of the run, accumulate on the column and then bleed off later — a documented cause of baseline anomalies traced back to the reagents themselves [1].

Certain additives are notorious here. Trifluoroacetic acid absorbs UV strongly, and detecting at low wavelengths while using a high concentration of an absorbing additive such as formic acid can pull large amounts of light out of the beam [1].

The fix: Prepare fresh mobile phase with high-purity (HPLC-grade) solvents and water. Mix thoroughly, and replace old or degraded solvents. For more on preparing solvents and buffers correctly, see our mobile phase basics guide.

Cause 3: Gradient drift from differential solvent absorbance

In gradient runs there is a cause that has nothing to do with contamination: the two solvents simply absorb UV light to different degrees. As the composition changes during the gradient, the baseline follows it. Above roughly 250 nm the absorbance of mobile-phase components is usually minimal and drift is seldom a concern; below 220 nm, differential absorbance can be severe enough to make certain solvents impractical. One remedy is to add a UV-absorbing component to the other solvent so that the two are balanced [3].

The fix: Where the method allows, monitor at a higher wavelength, or match the additive concentration in both solvents so the absorbance difference disappears.

Cause 4: Insufficient equilibration

If you start injecting before the column has fully equilibrated with the mobile phase, the baseline will drift as conditions continue to settle. This is especially common in gradient methods.

The fix: Allow adequate equilibration time before your first injection, and run a blank gradient between analyses. A blank run also lets you record and later subtract any inherent baseline movement during data processing. In HILIC the equilibration requirement is far longer than most reversed-phase habits suggest — see our article on how HILIC retention works.

Cause 5: Air bubbles and degassing

Dissolved air coming out of solution forms bubbles that disrupt flow and detector response, often causing an upward-drifting or unstable baseline.

The fix: Degas your mobile phase properly (in-line vacuum degassers are standard on modern systems) and make sure lines are primed and bubble-free.

Cause 6: Detector and flow cell

With UV detection, a weak or ageing lamp, an obstructed light path or an improperly installed flow cell all disturb the signal. Detector problems typically show up as baseline noise, drift or a sudden loss of sensitivity, and they affect the response rather than the retention times [1][4]. Most modern systems include a lamp-intensity diagnostic, which is a sensible first test when the baseline looks fuzzy [1].

Also check that the reference wavelength is set and appropriate — on gradient runs this alone can account for drift that otherwise looks mysterious.

A systematic approach: isolate the source

When the cause is not obvious, isolate whether the problem lies in the fluid path or the electronics with one simple test:

• 1. Turn off the pump so flow is zero.
• 2. Watch the baseline for 5–10 minutes. If it improves with no flow, the problem is in the fluid path (mobile phase, bubbles, column). If it does not improve, the problem is more likely electrical or detector-related.

A second useful rule when peaks are also affected: does the problem hit all peaks or only one or two? All peaks similarly affected points to a physical or system-wide cause; one or two distorted peaks point to a chemical or column–analyte interaction [4].

Quick reference checklist

• Temperature — use a column oven, avoid drafts.
• Mobile phase — fresh, high-purity, well-mixed solvents.
• Wavelength — check for differential solvent absorbance below 220 nm.
• Equilibration — allow time; run blank gradients.
• Degassing — remove dissolved air and bubbles.
• Detector — lamp diagnostic, reference wavelength, flow cell.
• Isolate — pump-off test to separate fluidic from electrical causes.

Key takeaways

Baseline drift is rarely random — it is a signal that something in the system needs attention. Note the direction and shape of the drift first, then work methodically through temperature, mobile phase, wavelength, equilibration, degassing and detector settings. In most cases, a stable baseline is only a few systematic checks away.

References

1. Stoll DR. Essentials of LC Troubleshooting, Part 4: What Is Going On with the Baseline? LCGC International — drift versus noise, mobile-phase impurities accumulating on-column, UV lamp intensity and light-path problems. chromatographyonline.com
2. Optimizing the Refractive Index Detector — solvent composition changes over time and temperature effects as sources of baseline drift. Agilent 1200 Series RID User Manual. community.agilent.com
3. Dolan JW. Gradient Elution, Part V: Baseline Drift Problems. LCGC International — differential UV absorbance of gradient solvents, wavelength dependence and compensation. chromatographyonline.com
4. LC Troubleshooting Essentials: A Guide to Common Problems and Solutions for Peak Tailing, Ghost Peaks and Pressure Spikes. LCGC International, 2026 — detector faults as a source of drift and noise, and the “all peaks or only some?” rule. chromatographyonline.com