Initial Baseline noise HPLC - First few minutes

[Mouleyre]

When doing analysis by HPLC, often in the first minute there is a lot of noise on the baseline. What causes it? Something to do with the injection?

[Multidimensional]

Except for the t0 peak (observed when the injector pulse and/or unretained sample actually reaches the detector), no "noise" of any kind should appear on the baseline at all. The system should be fully equilibrated before the analysis starts. You provided almost no information so it sounds like you may be using HPLC for the first time. Please get some local help from an experienced user and review these initial steps.

Basic troubleshooting steps:
Q1: At what time(s) does the noise appear? Is it during the t0 time?
Q2: If you inject a true blank (a blank is just the mobile phase alone, at the same volume w/o the sample?
Q3: If you inject NOTHING, and only move the injector from LOAD to Inject (manually or via an A/I), do you still see the "noise" appear?

For assistance, please provide basic HPLC method details to be your question in context (Basic info equals: flow rate, column dimensions and particle size, measured column void volume, exact mobile phase composition, gradient or isocratic, sample name, injection volume, injection solution used, detection mode with ALL parameters (e.g wavelength & bandwidth etc). Include a chromatogram, with scales shown, illustrating what you describe as "noise".

[Mouleyre]

Except for the t0 peak (observed when the injector pulse and/or unretained sample actually reaches the detector), no "noise" of any kind should appear on the baseline at all. The system should be fully equilibrated before the analysis starts. You provided almost no information so it sounds like you may be using HPLC for the first time. Please get some local help from an experienced user and review these initial steps.

Basic troubleshooting steps:
Q1: At what time(s) does the noise appear? Is it during the t0 time?
Q2: If you inject a true blank (a blank is just the mobile phase alone, at the same volume w/o the sample?
Q3: If you inject NOTHING, and only move the injector from LOAD to Inject (manually or via an A/I), do you still see the "noise" appear?

For assistance, please provide basic HPLC method details to be your question in context (Basic info equals: flow rate, column dimensions and particle size, measured column void volume, exact mobile phase composition, gradient or isocratic, sample name, injection volume, injection solution used, detection mode with ALL parameters (e.g wavelength & bandwidth etc). Include a chromatogram, with scales shown, illustrating what you describe as "noise".

I am talking about the t0 peak. Can you give more details on what is an injector pulse please.

[Multidimensional]

The injector "pulse" results from the system pressure change which occurs between the load and inject positions. This "pulse" can be observed as a pressure spike, going UP and Down in amplitude. Depending on the detector used, the pulse can often be observed on the baseline if you reduce the scale. Measuring the column void volume and recording Tzero (start by calculating it) the column void volume (when multiplied by time, provides Tzero) are the most basic fundamentals of HPLC analysis. We teach them in the very first week of training and a practical understanding is required BEFORE running any HPLC analysis or method. You can learn these basic fundamentals before using HPLC by reading one of the many practical classical books on the topic (i.e. "Introduction to Modern Liquid Chromatography") as well as our own professional training website; https://hplctips.blogspot.com/ . Many years of training are needed to learn what is noise, what are sample related peaks, which settings are correct, etc (with an experienced professional to train you, as you can not learn this technique just from "doing"). However, the basics can be learned by anyone from these resources, but practical bench time is needed to actually start learning the technique. Perhaps you can find someone at your school or business who has industrial training in HPLC that is willing to help you learn these basics?

[Multidimensional]

"Determination of HPLC Column Void Volume / Dead Volume, Dead Time (T zero)"; https://hplctips.blogspot.com/2011/05/d ... -time.html

[vmu]

Pressure changes are transmitted through the mobile phase with the speed of sound in the mobile phase, this speed being much higher than L/t0.

Any peak on a chromatogram is a result of the change of some detectable property caused by the (small or large) change of the composition of the eluate. The peak(s) near t0 is(are) due to the changes in concentrations of the components that move along the column practically unretained (these components are some constituents of the eluent or the unretained components of the injected sample).

Measurement of the accurate (physically meaningful) t0 in many cases is not so simple as it may seem.
https://www.chromforum.org/viewtopic.php?t=122416
https://www.chromforum.org/viewtopic.php?t=122417

[Multidimensional]

Actually, measuring the Tzero as a practical matter IS rather simple to do, as noted in the previous article I referenced. No need to make it complicated at all, esp for new users. It can and should be estimated first, then measured as previously described. Besides injecting samples which are not retained on the specific column support used, the observation of the "pressure" blip (or spike etc) can be easily observed on most detectors commonly used (i.e. UV) and this peak is an accurate measurement of Tzero for HPLC analysis and use in calculations. No speed of sound issues need apply ! LOL. The change in pressure from the valve position switch (LOAD to INJECT) results in a real pressure change that is measured by the pressure transducer in the flow path between the pump and injector valve. This makes it really easy and simple. This is part of basic HPLC training (IOW: the training no one ever seems to get when they need it) and should be shared with others.

[vmu]

I think that it is good for any chromatographer to:

- know that training (studying, learning) is better than ignorance;

- know about different sources of knowledge (training courses, blogs, forums, experienced colleagues, literature from equipment vendors, textbooks, monographs, peer-reviewed scientific journals, universities);

- compare and critically analyse information from different sources;

- know the relationships between volumes and times in LC (Vr = F*tr, V0 = F*t0);

- have basic understanding of the internal geometry of a chromatographic column;

- know the difference between the hold-up volume (or void volume; the total mobile phase volume in a column) and the column interparticle volume;

- know about the size exclusion and ion exclusion effects; know that these effects can be easily encountered even in RPLC;

- know that anions (nitrate and nitrite mentioned in the blog referred to above in this thread) are excluded significantly from the pores of the usual reversed-phase sorbents when the eluents with low (less than 100 mM) ionic strength are used (e.g. https://doi.org/10.1016/j.chroma.2009.12.010); this usually leads to underestimated values (as low as the column interparticle volume) of V0 when anions are used as V0(t0)-markers in RPLC;
 
- know that determination of the column interparticle volume is a separate important task (in science);

- be aware of the concept of zone retention factor k" used in science in addition to the commonly known concept of phase retention factor k' (e.g. https://doi.org/10.1016/j.chroma.2008.02.018);

- know that using as a t0-marker the first peak (of odd shape and of origin not known exactly) the analyst comes across on a chromatogram is a bad idea; depending on conditions, the retention volume of this peak can vary between the column interparticle volume and the values that indicate some adsorption;

- know that the use of a dedicated t0-marker (such as uracil, thiourea, etc) is better than using the first strange peak; however, it does not guarantee the absolute absence of retention; with such markers, one can often obtain reasonable and acceptable for some purposes t0 (V0) values, but these values are still estimates, the accuracy of which depends on the experimental conditions;

- always compare his/her experimental V0 values (Vexper) with the column volume (Vcol = L*pi*(d^2)/4);
* the ratio Vexper/Vcol close to 0.36-0.42 indicates nearly total exclusion from the sorbent pores (if the sorbent particles are porous) for most LC columns filled with rigid spherical particles;
* the ratio Vexper/Vcol higher than 0.7-0.9 means the chosen t0-marker is actually retained by the stationary phase.

[vmu]

The change in pressure from the valve position switch (LOAD to INJECT) results in a real pressure change that is measured by the pressure transducer in the flow path between the pump and injector valve.

The pressure change occurs exactly at the start of the analysis (t = 0 min). This change can be observed on the pressure trace (if recorded) at t = 0 min. However, the first baseline disturbance registered by the LC detector is at t near t0 (which is undoubtedly higher than zero). The pressure change caused by the valve switching might cause some disturbance in the composition of the eluent, and it is the zone of changed composition that moves along the column with the speed near L/t0, resulting in peak(s) near t0. However, in general one cannot say how exactly the eluent composition is altered in this case. Depending on the eluent composition, this concentration disturbance may appear very close to the real t0 or before t0 or after t0 on a chromatogram.

[JI2002]

Pressure change during injection causing the t0 signal is completely false.