Select Page
Welcome to our Knowledge Base

Decrease in detector response

You are here:
< All Topics

Applies to

Electrochemical detector

Summary

For the same standard solution at the same detector and LC settings, a decrease in detector response is noticed over a period of hours or days.

Cause

Decrease in detector response can be caused by:

  1. Contamination of the working electrode. This is often caused by either a sample or mobile phase constituent. Sometimes it is something that slowly desorbs from the column or other parts such as tubing or pump heads and adsorbs to the working electrode. In case of metal electrodes the formation of metal oxide may change the electrode response.
  2. Also injection of high concentration of standards (>10 µmol/L) may result in electrode contamination.
  3. Incorrect potential settings. If the cell potential is not optimized, small changes in the potential may result in considerable decrease in peak height. This may occur with poorly defined or buffered mobile phase. The reference electrode and thus the cell potential may change with the pH or with the chloride concentration.
  4. Malfunctioning reference electrode. In case of a salt bridge REF this may be caused by a gas bubble in the reference compartment, clogging of the cotton wool frit or dilution of the saturated KCl in the REF. In situ reference electrodes (ISAAC and HyREF) depend highly on the mobile phase composition. If no chloride is added (ISAAC) or if the mobile phase is poorly buffered (HyREF) these reference electrodes will not function properly.
  5. Electronic malfunction. A poor contact in the cell or cell cable.
  6. Data acquisition not capable to measure below zero (or -100 mV). In combination with drop in background current (not always notable!) it will result in a part of the peak become below the detection window. Only tops of peaks are measured.
  7. Chemical instability of sample.
  8. Syringe speed autosampler too high for micro LC or viscous solution.
  9. Another cause not related to the detector, such as a problem with injection.
  10. Sencell: wrong spacing setting (set spacing to 1 or 2 to check, not 0).

Diagnosis

Contamination of the working electrode: is recognized by a continuous loss of signal. Each run results in a decreased peak height and area. After servicing the working electrode the signal is restored. Incorrect potential settings: the signal improves considerably by increasing the working potential with steps of 100 – 200 mV. Chemical instability: the signal improves with a new, freshly prepared standard. Data acquisition: most software packages measure between +/- 1 or 10 Volt and don’t have any problems with a decrease in baseline. However, exceptional situations are limited for example 0 – 1 Volt. Check specifications of the AD convertor of your data acquisition package.

Remedy

Contamination of the working electrode: if mobile phase contains contaminant switch to highest purity solvents. If sample contains contaminant look for a sample pretreatment. Dilute sample if possible, or use larger spacer thickness. If not possible, more frequent polishing the cell is an alternative. Flow cell or REF related problem: replace malfunctioning parts. Incorrect potential settings: the potential settings must be optimized by constructing an I/E curve. Chemical instability: additives to sample buffer, or adjusting the pH can improve chemical stability. Search literature (handbook) for details. Data acquisition below zero: program an autozero event at start of a run, also using a +10% will help. Syringe speed: should be set to low when working with microbore needle or viscous solution. Avoid pressure drop or bubble formation in the injection flow path. See also next point. Other cause: try systematic elimination of all possible causes, for example replace the autosampler for a manual valve to see if the problem is related to the autosampler.

Was this article helpful?
0 out Of 5 Stars
5 Stars 0%
4 Stars 0%
3 Stars 0%
2 Stars 0%
1 Stars 0%
How can we improve this article?
Please submit the reason for your vote so that we can improve the article.
Previous After installing a (new) column the sensitivity is lost
Next Excessive peak broadening with partial loop fill
Table of Contents