You can purchase the Measure Your Gradient test mixture here (enough to make 1.5 mL). They are offered at cost, for $40 with domestic ground shipping or for $130 with international shipping.

The Measure Your Gradient test mixture comes with each of the standards dissolved in 75 µL of acetonitrile. Before you use the test mixture, add 1425 µL of water to bring the total volume up to 1.5 mL. After adding the water, the concentration of each standard will be as follows:

100 µM N-methylacetamide,
100 µM N-ethylacetamide
100 µM N,N-dimethylpropionamide
100 µM benzamide
25 µM methylbenzamide
25 µM ethylbenzamide
25 µM n-propylbenzamide
25 µM n-butylbenzamide
25 µM n-pentylbenzamide
25 µM n-hexylbenzamide
25 µM n-heptylbenzamide
25 µM n-octylbenzamide
25 µM n-nonylbenzamide
25 µM n-decylbenzamide
25 µM n-undecylbenzamide
25 µM n-dodecylbenzamide
25 µM n-tridecylbenzamide
25 µM n-tetradecylbenzamide
25 µM n-pentadecylbenzamide
25 µM n-hexadecylbenzamide

Once the water is added, the test mixture becomes cloudy because the larger benzamides are only sparingly soluble in water. That isn't a problem - the larger benzamides quickly and fully dissolve once the mobile phase composition goes to a high level of acetonitrile (like at the end of a gradient). However, the benzamides don't stay suspended in the test mixture forever. After several months, the benzamides will fall out of the suspension, leaving very little left in solution. This is greatly accelerated when the test mixture is cooled, so do not refrigerate the test mixture or leave it in a cooled autosampler. You can tell if the test mixture is old and needs to be replaced when it becomes clear.

The column you use may be of any inner diameter or length, but it must contain the Eclipse Plus C18 stationary phase with 3.5 µm diameter particles. At the moment, this is the only stationary phase for which we have measured the isocratic retention of each of the standards. We intend to add more in the future.

Solvent A must be 0.1% (by volume) formic acid in water and solvent B must be 0.1% (by volume) formic acid in acetonitrile. Prepare them carefully, but we found that +/- 10% error in the formic acid concentration is tolerable.

The "instrument dead time" is the time it takes an unretained solute to reach the detector when the HPLC column is removed. It is an important value needed to accurately back-calculate your gradients. You will only need to measure this once unless you reconfigure the tubing (inner diameter/length) between the point of your sample injection and the detector.

To measure your instrument dead time, replace your HPLC column with a union. Then run the Measure Your Gradient test mix in a 1 min long isocratic method (50% solvent A/50% solvent B is fine) at a relatively low flow rate (200 µL/min is fine). If you are using a mass spectrometer as the detector, scan for the 150 m/z ion (ethylbenzamide). If you are using a UV absorbance detector, set it to 210 nm. Use a sampling rate of at least 5 Hz if possible. The retention time you measure is your instrument dead time at that flow rate.

Important: Remember that your instrument dead time is a function of flow rate. For example, if you change the flow rate to one twice as fast as the one in which you measured your instrument dead time, divide your instrument dead time by 2.

HPLC method
Virtually any gradient program and flow rate may be used in your method but these four rules must be followed:

1. The column temperature must be set to 35 °C.
2. The column must contain the Eclipse Plus C18 stationary phase (3.5 µm particles), though it can be of any length and inner diameter.
3. The initial solvent composition must be ≥5% B and the final solvent composition must be ≤95% B.
4. The needle wash solvent should generally be ≤5% B. The needle wash can affect the retention times of the early-eluting standards if it is too strong.

Note: When measuring your gradient delay volume, you will get more precise results if you use a fast, steep gradient. A 5 min gradient from 5% to 95% B works well.

Detector configuration
If you are using a UV absorbance detector, use the 210 nm wavelength. If using a mass spectrometer to detect the standard compounds, make sure you include the following m/z values in your scan or SIM (positive ion mode):

74 m/z (N-methylacetamide)
88 m/z (N-ethylacetamide)
102 m/z (N,N-dimethylpropionamide)
122 m/z (benzamide)
136 m/z (methylbenzamide)
150 m/z (ethylbenzamide)
164 m/z (n-propylbenzamide)
178 m/z (n-butylbenzamide)
192 m/z (n-pentylbenzamide)
206 m/z (n-hexylbenzamide)
220 m/z (n-heptylbenzamide)
234 m/z (n-octylbenzamide)
248 m/z (n-nonylbenzamide)
262 m/z (n-decylbenzamide)
276 m/z (n-undecylbenzamide)
290 m/z (n-dodecylbenzamide)
304 m/z (n-tridecylbenzamide)
318 m/z (n-tetradecylbenzamide)
332 m/z (n-pentadecylbenzamide)
346 m/z (n-hexadecylbenzamide)

If you wish to have the Measure Your Gradient software automatically extract the retention times of the standards from your LC-MS data file, you must first convert it to one of three supported open file formats. These include the mzXML (*.mzXML), mzML (*.mzML), and AIA or netCDF (*.CDF) formats. If you wish to enter the retention times in manually, you can skip this step.

First, download ProteoWizard and install it on a computer running Windows. Proteowizard is free software that makes is easy to convert LC-MS data files into several different open-source formats.

After installing ProteoWizard, you will have a new program group for ProteoWizard in your start menu. Find the program called "MSConvert" and run it. Then follow the instructions below:

Once the application launches, you will see the following window:

In this window, enter the following information:

1) Enter your column inner diameter and column length in millimeters.

2) Enter the flow rate you used along with the gradient program you ran.

3) Enter the instrument dead time/volume. The selection box next to the instrument dead time box lets you select whether you want to enter it as a time or as a volume. If you enter it as a time, be sure it is scaled correctly for the flow rate used in the gradient.

4) Enter the retention times you measured for each of the standards in the Measure Your Gradient test mix. You can manually enter in all the values, or you can click the "Find retention times automatically..." button to automatically extract the retention times from an mzXML, mzML, or netCDF file of the LC-MS run (see previous step).

Try it out! Right click the link below and select "Save link as..." to download an example mzXML file. It was acquired on a 2.1 x 100 mm column with a 0.4 mL/min flow rate in a gradient from 5% to 95% B in 10 min. The instrument dead time was 0.146 min.

10mingradient.mzXML

5) Click on the "Next Step" button. The window should now look something like this:

The top-left graph shows the ideal (programmed) gradient profile. The bottom-left graph shows the expected dead time (as measured by uracil) as a function of solvent composition. The table on the top-right shows the retention times you entered in the previous step (with the instrument dead time subtracted).

6) Click on the "Back-Calculate Profiles" button to begin back-calculating the gradient. First, the gradient delay is back-calculated. The gradient delay volume is the sum of all the volume in the tubing, connectors, and pump that is between the point where the solvent is proportioned and the column inlet. The gradient delay volume causes a delay between the time the HPLC pump is programmed to produce a certain solvent composition and when it actually reaches the column.

The gradient delay volume is split into two theoretical parts: mixing delay volume and non-mixing delay volume. One can think of mixing volume as a thoroughly mixed reservoir.

To understand its effect on the gradient profile, consider a gradient in which the fraction of solvent B increases linearly over the course of the experiment. As eluent with a higher amount of solvent B enters the mixing volume, it mixes with eluent of a lower solvent B fraction left over in the mixer. This causes the gradient profile reaching the column to not only be delayed, but also rounded, or dispersed.

One can think of non-mixing volume as an open tube where a significant volume of eluent resides, but does not mix as it moves through it (of course, under laminar flow conditions it actually does mix to some degree as it moves through an open tube, but for this illustration, we ignore it). The following figure shows the effect of 1 mL mixing volume or 1 mL non-mixing volume on a gradient profile run at a flow rate of 1 mL/min.

7) After back-calculating your gradient delay volume, the software will ask if you would like to continue back-calculating your gradient.

If you choose to continue, the application will back-calculate your gradient profile with greater precision. It will allow the gradient profile to take on any shape (not just a dispersed and/or delayed gradient), enabling it to account for solvent mis-proportioning. It usually takes 3-10 minutes to complete the final gradient optimization.

8) Copy and paste a report into spreadsheet software. Once finished, click on the "Copy profiles to clipboard" button to copy a report to the clipboard. Then you can paste the report into your favorite spreadsheet software.

If you have any questions, comments, problems, or you encounter a bug, please email Paul Boswell at boswell@umn.edu. We appreciate any and all feedback.