Please take the time to first read up on and learn about what HPLC is and what the key fundamental concepts are when running a size exclusion chromatography (SEC) method. Knowledge of these basic concepts is critical to answering your question and understanding how the analysis is designed to work.

Water should NOT be retained on your column (this is normal). It should elute out early, at or near the Tzero time. Injecting the mobile phase should show the same result, though the signal should be small and account mostly for the backpressure change within the HPLC system from the injector valve switching from the LOAD to INJECT positions. To perform chromatographic analysis, try injecting some standards of appropriate Mw and structure which apply to your proposed sample analysis. Select stds which are retained within the pore range of the column(s) selected. This is needed for analysis and provides a great learning tool to "see" the chromatographic process in real-time.

Suggestion: Your mobile phase is going to quickly grow cultures of bacterial, algae and maybe even mold in just about one day. Be sure to prepare FRESH mobile phase each day, then filter it through a 0.45 or 0.2 filter before use. Always flush out all buffers from the HPLC system at the end of the day.

Contamination of the column and flow path of the HPLC may result in corrosion, damage and/or lost time. Be sure to get some help from a local, trained chromatographer.

William gave a good answer. Think about what happens if you displace a mobile phase with buffer with something with no buffer. What happens to the refractive index?

The RI is very sensitive - very small changes in RI from trace amounts of additives, salts, etc. will be seen. the eluent and the sample solvent may be the “same” but they aren’t exactly the “same”!

William Letter Thanks for your suggestion. I have used standard, Polyacrylic acid 350k, 695k, and 1.3m, which the range recommended for the molecular weight target of my sample (below and higher range than molecular weight target). But the standards given negative peak also, which when I overlay them with the water, their peaks at identical peak. So I ask this question, may when I injected water, and then resulted the negative peak with significant mV, is there any problem in my column or etc?

By the way, the column unused for 2 years. Then I use it nowadays. May this causing this problem? As you mentioned above, we have to always flush out all buffers from the HPLC system at the end of the day.

Jack Silver Thanks for your answer. Yes logically, it will give a peak. But, is it possible that the peak shown in significant mV value? Here I attached the GPC Chromatogram of injected water and the polyacrylic acid 350k standard (diluted in water with concentration 0.1%)

James E Hanson Thanks for your answer. Okay maybe if I injected water only, there is a peak because the mobile phase contains salt. But how if a peak still shown when I injected the mobile phase also? Is there any reasons which may cause this one? Or maybe there is trouble in the column? Thanks by the way.

Troubleshooting starts by simplifying things and following good chromatography fundamentals. Configure the RI for normal, positive signal output (The system can not integrate negative peaks properly. Please run in pos voltage offset mode). Please get rid of the sodium azide and prepare freshly filtered and pH checked mobile phase daily ; In your example: 10 mM NaH2PO4.2H2O with the flow rate set to 1 mL/min . Equilibrate the solution and monitor the backpressure, which should be stable. Once it is stable, flush the RI's reference cell and establish analysis conditions.

• Now, run an analysis and inject a sample of mobile phase. You should only see a tiny peak, relative to the scale. Could we see the entire chromatogram (start to end) showing the baseline with injection of mobile phase?
• If this shows a normal result, then dissolve your three standards (same approx conc. each) in the MOBILE PHASE and run an analysis using that mixture. Please share the resulting chromatogram.

Michael Yonathan Suryana Yes, you may have a significant signal from water without the salt. William gave some good suggestions to try.

Jack Silver Thanks for your help.

William Letter Sorry for my late reply. I have injected as per your suggestion. Here I attached the mobile phase, PAA 350k, PAA 695k, and PAA 1.3m chromatogram. I attached also the overlay chromatogram of those standards. Looks that buffer phase doesn't give significant peak. But the standards look have similar retention time. How is your opinion about this?

Please provide the details of the column you used for the above analysis. We need the exact FULL NAME of the column, the dimensions (L x ID, particle size).

Separation is shown, but the resolution is poor between peaks because the Mw shown for each is just an average value, a range of Mw each. If the correct column is used, it may be possible to sharpen up the peaks a bit more and improve the separation, but overlapping may still exist. To find out what is possible, we need to determine if the column you have selected is appropriate for this analysis so please respond with the details requested.

The analysis with mobile phase only does show drift which is larger in amplitude than expected over a short time period. This means you do not have control of the analysis conditions. The column is not equilibrated, the temperature of the column is changing OR the refractive index temperature is changing (of the flow cell). DRIFT is normal with Refractive Index detectors, but it must be minimized/reduced for the analysis to be reproducible and informative.

• "HPLC Baseline Stabilization Tips for Refractive Index Detectors (RI or RID)"; https://hplctips.blogspot.com/2018/02/hplc-baseline-stabilization-tips-for.html

William Letter Ultrahydrogel Linear (7.8x300 mm) and 500 (7.8x300 mm).

I have equilibrated it for long time, also purging and flushing it, but the drifting is still appears. By the way, GPC instrument in company I work now, is stopped for around 3 years previously, from 2017-2019. Then I started to reoperate and evaluate this instrument. May this cause the drifting and the bad peak resolution, Sir?

By the way thanks for your cited reference

OK, that sounds like Waters Ultrahyrogel. *Those columns provide very little surface area to resolve such broad Mw Peaks apart all by themselves. The Linear, which is just a non-specific, general purpose column and the 500A column which has a spec of 5K to 400K.

To resolve apart such large and "wide" molecules as you propose, 350K, 695K and 1.3 M, you may need 3, 4 or even 5 properly sized SEC columns in-series. This is common for SEC work as the average Mw's of each sample are so broad that you will need a LOT of surface area to retain and resolve them apart. We place multiple columns in series, selected for the correct Ang size each, to increase the available surface area.

Additionally, please make sure you have been fully trained in how to use the HPLC and RID before use. Success is based on training and understanding.

Regarding the issues with stabilizing the HPLC and RID. This is addressed in the article that I provided. - Using degassed mobile phase, insulating the tubing and connection lines, correctly flushing the reference cell, insuring the the flow cell temperature and column temperature are the same (AND CONTROLLED). Room temperature must also be stable too for the detector to be stable. If the HPLC system has any ongoing problems with the pump, then you may never achieve equilibration so make sure it has been professionally checked and serviced before use.

William Letter Yes Waters Ultrahydrogel. Oh I see. Waters application expert also mentioned that statement also, but he is not yet recommended me about using column in-series. Thanks for your suggestion.

This is the problem here in Indonesia, because GPC is a rare instrument, and cause the trainer also have some limitations. Luckily, I can discuss with you, Sir. I really appreciate it.

Yes I have read your cited article and I try to implement it. By the way, is flow cell temperature same with "Internal Temperature" term? And, is column temperature same with "Ext. Heater 1"? (I attached the screenshot of overall user interface and also the detector display)

• If your room temperature is stable and at or below 22C, the column heater and RID flow cell temperature of 37C should be fine.
• What is the ROOM temperature? Is is stable?
• The baseline noise shown in the chromatogram, showing only ~ 2.5 mV of vertical oscillation may not present any problems at all. It is the S/N ratio of the baseline noise to the analysis peak that really determines if any integration or RSD problems may cause problems. *There appears to be no problem of any kind with the HPLC system.
• Your original question asked why "water" gave a sharp peak, but the very high Mw samples resulted in a broad peak. VERY large molecules (e.g. protein or large crosslinked polymers) will be observed as large and broad peaks because the molecule spends a lot of time interacting with the column (= late retention for large Mw, short Rt for low Mw). Salts, very low Mw compounds and solvents will elute quickly as sharper peaks. This is normal and the basis of SEC. Adding multiple columns in-series is one way to increase the available support. If done correctly, with the needed changes to the method (salts/buffer/pH), then it may be possible to resolve them apart better. Not everything can be baseline resolved though, so you should take the time to review similar examples using different columns and methods to determine if investing in more columns is worth it.