need to check one thing at a time, because you don't what cause the issue. First, use the best solvent you can find. Then change one thing at a time using your problematic method to start with. For instance, start with water. Change to the best quality with everything in place. If it is still bad, it is not water. Next check the salt (source and grade). Do this one thing at a time and then you will find the source. If you change many things at one time, you would not know what is the source of the problem.
I can't comment on the in-line mobile phase cleaning columns because I haven't used them.
I assume you are running reverse phase HPLC? Ghost peaks are usually impurities on solvents, commonly the water although there are sometimes polar impurities in the organic solvent that accumulate. Also check any modifiers that you might use to adjust the pH of the mobile phase. You may need to use a higher quality solvent, also make sure people aren't putting unused solvent back in the container.
there may be many reasons for this situation. Agree with previous speakers, much depends on the purity of solvents but also bidistilled water and its quality. This problem is also encountered when working on the buffers, it is necessary to long and thorough washing of the column, preferably warm water, redistilled at low flows. pH changes using eg phosphoric acid causes this just catch an.
need to check one thing at a time, because you don't what cause the issue. First, use the best solvent you can find. Then change one thing at a time using your problematic method to start with. For instance, start with water. Change to the best quality with everything in place. If it is still bad, it is not water. Next check the salt (source and grade). Do this one thing at a time and then you will find the source. If you change many things at one time, you would not know what is the source of the problem.
The Ghost-Guard-LC and similar products are very effective to clean your mobile phase. Trace impurities in the mobile phase are offen causes ghost peaks! We use the Ghost-Guard-LC in all our gradient methods with good success. You can use the Ghost-Guard-LC in "On-Line" and in "In-Line" mode.
I used the Shimadzu product and it worked. The biggest challenge was convincing everyone else at my company that I had ruled out the "simple" causes. Ghost peaks are a huge issue when a steep gradient is needed for trace impurity analyses. What most people don't get is that this is a situation where robust systematic trouble shooting fails to solve the problem. Ghost peaks are ppt level contaminants that interfere with ppm analytes. Ppt contamination is so hard to troubleshoot. Someone could sneeze in the next lab over and you have a new ghost peak. Labs simply are not capable at preventing this level of contamination.