HPLC to UHPLC Conversion Notes (Column Dimensions, Flow Rate, Injection Volume & System Dispersion)

The use of ultra-high performance liquid chromatography (UHPLC) columns to reduce analysis times and sometimes improve detection limits is a hot topic. UHPLC presents a number of new issues. The incorporation of smaller 1.9 to 3.0 micron particles and smaller frits will raise backpressures and increase system wear and tear. Smaller diameter lines are often used (I.D. of 0.12mm or less) which can increase blockages and clogs if you do not filter your mobile phase and samples through a 0.45 or 0.2 micron filters. Piston seals and valve rotors can wear out early due to the very high pressures, heating and stress imposed on them. You should monitor your HPLC system carefully over time and consider increasing the frequency of preventative maintenance and inspection services as well. However, the smaller particle sizes can provide better resolution in some applications so they are well worth evaluating.

I must answer twenty or so questions each week in the area of UHPLC. The most common questions deal with selection of an UHPLC column and making adjustments to a method for the changes which effect: (1) Column Dimensions; (2) Flow Rate; (3) Injection Volume and (4) System Dispersion. The good news is that some of these questions can be answered with some basic math while others just require a basic understanding of how the system works.

(1) COLUMN DIMENSIONS: Let's start by making things as simple and brief as possible (this is supposed to be a "hint & tip", not a thirty page article). When initially converting from a convention HPLC column (e.g. with 5 micron particles) to an UHPLC column (e.g. with 1.9 to 3 micron particles), initially select a column with the same I.D. and length for the calculation. This way only the particle size changes. *I like to change one variable at a time. If you would like to change the column length to take advantage of some of the increased efficiency (and decrease the pressure!) which results from smaller particles, then please refer to the following equation.

     EQUATION A:  'Lc2' = ('Lc1' * 'p2') / 'p1'

[ 'Lc1' = Length of Column #1 in mm; 'Lc2' = Length of Column #2 in mm; 'p1' = particle size of Column #1 in microns; 'p2' = particle size of Column #2 in microns].
                    
   Example: Column # 1 is a standard HPLC column;  4.6 mm x 250 mm (5u) Column. You want to find out the length of an equivalent column which uses 1.9 micron particles instead of the 5 micron particles.

   'Lc2' = (250 * 1.9) / 5 ; Answer is: 'Lc2' = 95 mm. *So a 10 cm long column would be a good choice here.


(2) FLOW RATE: Flow rate is directly proportional to column diameter and as we saw above in Equation A, the particle size can also affect it too. If you keep the column length and internal diameter the same, then the linear flow will be unchanged with the same particle size. A change to the particle size alone will change the flow rate as follows: 'Fc2' = 'Fc1' x ('p1'/'p2').

A change to a smaller diameter column to compensate for the improved efficiency will require a change to the original flow rate to preserve the linear velocity. Please refer to the following equation.

     EQUATION B:   'Fc2' = ('d2' / 'd1')^2 * 'Fc1'

['Fc1' = Flow Rate of Column #1 in ml/min; 'Fc2' = Flow Rate of Column #2 in ml/min; 'd1' = Column #1 Diameter in mm; 'd2' = Column #2 Diameter in mm].
                     
   Example: Column # 1 is a standard 4.6 mm ID Column. You want to find out what the linear flow rate should be if you use a smaller diameter column (2.1mm in this example).

   'Fc2' = (2.1/4.6)^2 * 1.000 ; Answer is: 'Fc2' = 0.208 ml/min. *A flow rate of 200 ul/min would be fine. 

However, one other factor should be considered. The optimum flow rate for sub 2.5u particles are often about double that of the "normal" linear flow rate used with conventional particles (>2.5u). Evidence for this has been shown through analysis of the van Deemter curve with the tiniest particles showing much flatter curves. Retention (K prime) can often be maintained by combining twice the normal flow rate and speeding up the gradient time by a factor of 2. So a method utilizing std sized particles with a linear flow rate of 0.200 ml/min might benefit from a faster flow rate of 0.400 ml/min and a twice as fast gradient composition change.


(3) INJECTION VOLUME: A change in the column dimension may require a change to the injection volume (note: "volume" and concentration are two different things. If the solution concentration remains the same and you inject less, the on-column sample concentration will also be less). The smaller the internal volume of the column, the smaller the injection volume. To calculate the linear change in volume, please refer to the following equation.

     EQUATION C:   'V2' = 'V1' * {('d2'^2 * 'L2') / ('d1'^2 * 'L1')}

['V1' = Injection Volume #1 in ul; 'V2' = Injection Volume #2 in ul; 'L1' = Column #1 Length in mm; 'L2' = Column #2 Length in mm; 'd1' = Column #1 Diameter in mm; 'd2' = Column #2 Diameter in mm].
                     
   Example: Current injection volume is 10 ul. Column # 1 is a standard 4.6 mm ID x 250 mm Column. You want to find out what the equivalent injection volume should be for a 2.1 mm ID x 150 mm column.

   'V2' = 10 * (2.1^2 * 150) / (4.6^2 * 250) ; Answer is: 'V2' = 1.25 ul.



(4) SYSTEM DISPERSION: When converting HPLC methods to "UHPLC" methods, few parameters effects the results obtained more than the HPLC system's System Dispersion. The volume of liquid that is contained between the injector needle and flow cell (with the column removed or by-passed) is know as the system dispersion volume. This volume is determined by how the specific HPLC is designed and plumbed. On most HPLC systems, it can be easily changed and optimized to fit the specific application desired and only requires that you have a solid understanding of how the HPLC system works. The choice of connection tubing ID and length, how the autoinjector is programmed, its loop size and the detector's flow cell volume all contribute to the system dispersion volume. In the same way that changes to the total column volume can effect the peak shape and resolution, the internal system dispersion volume also contributes to the results. 

With standard sized analytical columns (i.e. 4.6 x 250 mm), the typical HPLC's system volume is so small relative to the volume of the column (e.g. 100 ul system dispersion vs 2900 ul column volume, or 3.5%) that it does not negatively impact the chromatography. However, anytime we utilize a tiny HPLC column whose column volume is a fraction of that found in a standard column (i.e. 100 ul system dispersion vs 2.1 x 50 mm column with 120 ul volume), diffusion and band spreading can quickly become so significant that effective plate numbers are quickly reduced below values found on a standard sized column. As column volume decreases (and approaches the system volume) the total system dispersion volume must also decrease. In general, try and keep the system dispersion volume at or below 10% of the column dead volume. This is most easily accomplished by reducing the number of connections and fittings used, reducing the lengths of all tubing used, using much narrower ID tubing (e.g. 0.12 mm vs 0.17 mm ID), reducing flow cell volume and reducing the flow-through volume in the autoinjector (i.e. loop size, needle seat, etc). The injector is often the largest contributor to the system dispersion so concentrate efforts here (e.g. after injection, switch the loop out of the flow path).

Using Smaller Diameter HPLC Columns (Calculate Linear Velocity)

Lots of 2.1mm ID chromatography columns are appearing on the market right now. Since most of us are using 4.6 mm ID columns to develop HPLC and UHPLC methods, use of these smaller ID columns requires a few adjustments be made to the method and often, the HPLC system. If gradient elution is used, then the gradient profile must be changed to compensate for changes in void volume of the column and the dwell volume of the system. Injection volume must also be adjusted in a linear fashion too. Additionally, to maintain the same initial mobile phase linear velocity through the column as we had before (to obtain the same approximate retention times), the flow rate must also be adjusted. *We will discuss how to calculate the change in flow rate in this installment.

In order to reproduce your original method, we must first adjust the flow rate for the new, narrower bore column. The formula to do this is very simple. We decrease the flow rate by using the square of the ratios of the column diameters times the flow rate.

Linear Velocity Change Formula:

( C₁  / C₂ )² x original flow rate (ml/min) = new flow rate (ml/min).

Where:  C₁  =  Diameter (mm) of new (smaller) column;
              C₂ =   Diameter (mm) of original column.
------------------------------------------------------------------------------------------------------------------------------

Example #1: Find the new linear flow rate if we use a 2.1 mm ID column in place of a 4.6mm column with an initial flow rate of 1.000 ml/min.

              ( 2.1 / 4.6 ) ² x 1.000 = 0.208 (208 ul/min)


Example #2: Find the new linear flow rate if we use a 2.1 mm ID column in place of a 4.6mm column with an initial flow rate of 2.000 ml/min.

              ( 2.1 / 4.6 ) ² x 2.000 = 0.416 (416 ul/min)



Example #3: Find the new linear flow rate if we use a 1.0 mm ID column in place of a 4.6mm column with an initial flow rate of 1.500 ml/min.

              ( 1.0 / 4.6 ) ² x 1.500 = 0.071  (71 ul/min)



If we assume that the original flow rate is 1.000 ml/min then we can also use this table to get an idea of how the flow rate changes with decreasing column diameter (same particle size and support).

Column I.D. (mm)                 Flow Rate (ul/min)
            4.6                                          1,000
            2.1                                             208
            1.0                                               47
            0.3                                                 4
            0.15                                               1



Summary: Scaling down a method which was originally developed on a 4.6 mm ID column for use on a 2.1 mm ID column (with the same particle size) requires that the flow path of the HPLC system be optimized (reduced) to minimize diffusion and the flow rate reduced five time to achieve the same linear velocity. If the particle size is also going to be reduced from 5u to 2.5u or smaller, then increases in the flow rate may be considered to take advantage of the optimized plate counts using optimized linear velocities (which are much higher for smaller particles).

Tips and Advice for Priming your HPLC PUMP (or similar pumps, FPLC or UHPLC Pump)

The single most important component of any HPLC system is the Pump module. We often refer to it as "the heart of the HPLC system". 

• You may have the most sensitive HPLC detector, the best column, a perfect method of analysis, but none of this will matter unless the HPLC pump(s) that provide mobile phase to the system operate perfectly, all of the time. If you have a poor quality (or poorly maintained) system, then you will spend much of your time trying to establish reliable flow through the HPLC system, not running samples. 
• Before using an HPLC system, you should prime all of the lines in your HPLC pump. This is needed to purge any air from the tubing, introduce fresh mobile phase to each line and then to VERIFY that each channel delivers the reported amount of fluid to the column (measure it).
  

• NOTE: This is a LONG, detailed article with lots of information, Hints and Tips. It is available in PDF format for download, here.

The HPLC pump's ability (stability) to provide reliable operation depends on: 

(1) The Chemical, Physical and Miscibility properties of the Liquid(s) being pumped;

(2) The Amount of dissolved gas inside the liquid (must be minimized);

(3) The Temperature of the room (or HPLC) must be stable;

(4) The Position of the mobile phase bottles (relative to the pump, above or below);

(5) The Solvent Pickup Filters used (are clean and appropriate in material & porosity);

(6) The Fittings used are correctly installed & tightened;

(7) The types of Tubing used are chemically, temperature and pressure compatible (esp. the Inside Diameter of the tubing);

(8) The Selected Flow Rate(s) and Back-pressure are within the optimal range of the pump;

(9) All mobile phase solutions are Filtered, Freshly prepared and Degassed;

(10) How often the Pump is properly Inspected, Cleaned & Serviced.

 The HPLC pump is the most important part of your HPLC system. Take care of it. Neglect or abuse it, and you may lose time and money. Almost every problem you experience using an HPLC will be related in some way to the pump. Make sure you understand the flow path of the system in detail, and have the training to setup and use it properly. Take a hands-on training class (not a video or web based tutorial) to learn how to use the pump on your specific HPLC system. Learn how to run simple verification tests to check the flow rate (best done with a graduated cylinder). Never rely on the software values, check and verify everything yourself. Priming and flushing are needed any time air bubbles are present, mobile phase solutions are changed or the system has sat unused (this includes overnight). Always flush multi-channel pumps and valves (i.e. Binary, Ternary, Quaternary...) using a setting of 100% channel composition. Run one channel at a time at 100%, not 25% or 50% to flush channels (a common novice mistake). Flush ALL channels on a regular basis.

OK, so what can you do to make sure your HPLC pump is properly primed with fluid and operating to the best of its ability?

Start, by reading the operator's manual for your pump. Review the procedures for connecting it to the system, become familiar with the flow path and understand the procedures to prime the pump heads. Practice these procedures.

If an inline vacuum degasser is used, become familiar with the specifications, chemical compatibility (some are not compatible with solvents such as strong acids, strong bases, THF, chloroform, fluorinated additives and so on) and internal channel volume of each chamber used. It is useful to know what the degasser chamber volume is to figure out what the total channel priming volume is. This may be different for similar systems. Check, measure, verify, do not assume.

Priming Volume: The total volume contained in each channel's low-pressure line from the mobile phase bottle to the degasser + the degasser chamber channel volume + the total volume in the line from the degasser to the pump head (or multichannel valve) = the total minimum volume you must flush out before using the system. Because flushing just the minimum of volume (1x) of fluid through the channel is unreliable, flush 2x, 3x or more times this total volume, per channel (or as much fluid as it takes), to prime each channel. *If no degasser is present, then just calculate the volume contained in the low pressure tubing from the bottle to the pump head/valve. Set the pump to direct the flow to waste and use a high initial flow rate to speed up the priming process.

Use fresh mobile phase (prepared daily and filtered). Make sure the solvent pickup filters are clean. If possible, have the bottles placed higher than the pump's inlet (once flow has been established, this will allow natural siphoning to push liquid towards the pump head). Prime all of the lines used. The pumps run on liquid, not air so try and fill any of the lines with pure mobile phase before you connect them to the pump and/or degasser (If all of the lines are prefilled with fresh liquid, you can skip this part).  

There are two ways to PRIME EACH line (Flushing the Channels).

• *First, open any Prime/Purge or Waste Valve so the mobile phase is directed to waste, not the injector, column or detector. Our goal is to initially fill the lines with liquid, quickly, and we do not want these fluids to go through the entire HPLC system (i.e. column), just the HPLC pump.

(1) Wet Priming use a syringe fitted with a Luer-to-threaded fitting adapter (usually 1/4-28) to draw liquid through the tubing in the mobile phase bottle and into the pump's degasser and/or pump head's inlet. Be sure to have this type of syringe available (very useful). Never push fluid, only draw fluid through the tubing, just like the pump does. Connect the syringe to the mobile phase bottle lines, degasser ports and/or pump head multichannel valve or pump head inlet, as needed, to draw liquid through until all lines are filled.

(2) Dry Priming using the HPLC pump to draw the mobile phase out of the bottles, through the lines, degasser channels and to the pump head or multichannel valve. Note: "Dry" because the lines (low pressure tubing) are initially dry when we start. Always do this one channel at a time (e.g. A = 100%). This insures no miscibility or mixing problems and is standard procedure. Start with a modest flow rate to get the fluid moving through the lines, then increase the flow rate to speed up the process. The low pressure Teflon tubing is transparent so you can watch this process. Repeat with each channel. Note: Some HPLC pumps will struggle to perform this type of dry priming, as they will be unable to draw the liquid up from the bottle and/or pump the air out of the system. Pre-priming the lines using a syringe (as in #1 above) will help solve this. Running the pump with just air inside the lines may result in increased wear on the system (esp the piston seals) so if the system struggles to fill with liquid after one minute, discontinue and manually wet prime each line.

NOTES: 

• The back-pressure shown on the system readout should be very low during this initial  priming process (e.g. < 15 bars) as the HPLC system should not be plumbed with the column or detector inline, during the priming process (it should be by-passing those parts). Only the viscosity of the solution, the selected flow rate and the internal diameter of the tubing going into and out of the pump will contribute to the observed back-pressure, and this should be very low value.

• Once you have verified that liquid is exiting through the pump head waste port, you can increase the flow rate to speed up the priming process, but pay attention to the back-pressure. It should increase as the flow rate increases and drop as the flow rate drops. Continue to prime each channel in this way, one-at-a-time, until all channels are primed and flushed with liquid. You can gradually slow the flow rate down as you stop, to transition from one channel to another.
  

• If liquid has been drawn to the pump head, but the pump head still is not pumping liquid through it, it may be experiencing cavitation (air locked). If there is an outlet port on top of the pump head, the outlet fitting above the pump head can sometimes be briefly loosened with a wrench, allowing the system to push the air out (open it slightly with a wrench, then quickly close it after liquid comes out). Have a towel ready to soak up any fluid that comes out. Keep the area clean and dry. Alternatively, try drawing liquid through this port, while it is running, to gently fill the pump head chamber and remove the air.

• In some case, the inlet or especially the outlet check valve(s) can also become "stuck" open. When buffers are left in the system (they should be flushed out with water), crystals and particulate matter may deposit on the valve resulting in poor sealing, leaks or air being drawn through. Drawing liquid out of the pump head's outlet port with a syringe (or gently pushing it through the pump head) may remove the air bubble, debris and prime the valve, restoring function. Note: If needed, shut down the pump and clean/replace any contaminated or worn check valves before proceeding.
  

• In more extreme cases, you can change the mobile phase going into the pump head to a more viscous intermediate solvent to get things moving (an alcohol such as IPA might work well. If buffers have been used, then always first flush with pure water). 

• Degas all eluents / mobile phase solutions used. All of them. Degassing will help reduce the formation of bubbles inside the pump head. Failure to properly degas the solutions used may result in loss of prime, baseline and/or pressure instability. Make sure your degasser is operating properly (electronic vacuum degassers only last ~ 5 years at most. Be sure to have them professionally serviced). Sonicating fluids at the bench or using vacuum filtration to initially remove gas from the solution will only degas the solution for a short time (minutes). Gas will slowly diffuse back into the solution resulting in baseline noise, drift and pump problems (for HPLC, only use inline degassing or Helium sparging).
  

• Verify the flow rate. It may be unwise to rely on the indicated flow rate shown on the instrument screen or display. It is wise to measure the flow rate of each channel, separately, using a graduated cylinder and a timer. This is the most reliable way to determine what the actual flow rate is through the system (and is also the method we use during performance verification or qualification testing too). To check the flow, make sure the system has been primed and flushed. Install a flow restriction capillary in place of the column (to provide the required back pressure). Set the flow rate to a value which is appropriate for the pump and measure/record the volume delivered vs. time. Example: Using a flow rate of 1.000 mL/min obtain a 10 mL volume, glass laboratory grade graduated cylinder. At time zero, direct the flow from the restrictor's outlet into the graduated cylinder. Measure the volume of fluid collected in 8 minutes. *It should be 8.00 mLs.
  

If you continue to have priming problems and/or air bubbles disrupting the flow there are four more things you can check. 

1. Make sure the solvent pickup filters/frits are clean and unobstructed (these are maintenance items). If the filters are obstructed, then a vacuum may form on the line resulting in pump cavitation and loss of prime. One quick way to check if this might be a problem is to remove the suspect solvent pickup filter from the tubing, then try again. If flow is restored w/o the filter in place, then the filter may have been clogged. Install a new solvent filter as soon as possible. *Never run the HPLC without solvent filters installed. Those filters perform a very important job and protect the flow path of the system.  
2. Service the Pump Heads. Regular cleaning, inspection and replacement of worn parts must be done to maintain the function of the pump. Worn parts will result in failures, instability, lost time, plus invalid data. The pump has many mechanical parts which wear out and require replacement. Most pumps should be inspected/serviced every 6 months. Keep the pumps clean and fully serviced (replace: piston seals, pistons, frits, check valves as needed). Depending on the brand, model and applications, the types of parts needed and the frequency of repairs varies widely. *This is discussed in another article. 
3. If your HPLC system has an inline vacuum degasser (either a standalone or integrated module), it may be damaged, contaminated or broken. The typical service life of an electronic inline vacuum degasser is only five years (some models have even shorter lifespans). Degasser's with internal damage may result in contamination of the mobile phase. A failing or damaged HPLC vacuum degasser may directly contribute to analysis problems (ghost peaks, pressure instability, poor baseline stability...). Have your degasser professionally diagnostically tested and serviced often.  
4. Clean and/or replace any worn or damaged inlet or outlet pump head valves. Not flushing buffers out of the HPLC system on a regular basis or remain in contact with the solution for long periods of time can damage the valves. In some cases, cleaning is all that is needed, but in others, replacement is required to restore function. Be sure to have the system professionally serviced on a regular basis.
   

• Additional Troubleshooting Info can be found here:

Diagnosing & Troubleshooting HPLC Pressure Fluctuation Problems (Unstable Baseline)

Modern HPLC Method Development Tips (PART I):

This is the first of two articles which will provide suggestions on how to improve the HPLC (UHPLC) method development process. - PART I

Method Development, BEFORE YOU BEGIN

1. Start with a working system: Before you start to develop a method, make sure your chromatograph is operating in top condition! Passing an OQ test and regular maintenance are not enough. You need to have the proper training and experience which will allow you to identify signs of trouble all of the time. For example: Do you see any salt or buffer crystals on the outside of one or more parts (esp the fittings or the pump head)? If you do, then you have a leak or other problem which needs to be addressed. Do you know what the normal backpressure should be when the system is operating correctly at a fixed flow rate with a single pure solvent (or defined mixture) running through? You should, as this provides a quick check that the pump is functioning properly. These types of quick checks should be done each time you use the HPLC and, if applicable, should be part of a general SOP. Is the flow rate shown on the display correct? A small graduated cylinder and a stopwatch (watch, timer) to measure the mobile phase output from the detector can answer this question in a few minutes. Are the pressure and flow rate stable (ripple <1%)? Verify by recording the pressure readings over time and measuring the S/N of the baseline with your detector to estimate the flow rate stability (noise). If they are not stable, then do not start an analysis. Find and fix the cause of the instability before running samples or standards. Some common causes of flow instability include: a sticking check valve, failure to properly degas the mobile phase, a leak, poor mixing or a miscibility problem.

2. Mobile Phase Preparation: Have you prepared enough fresh mobile phase to use during the day? Mobile phase which is mostly aqueous and contains buffer salts often experiences significant growth of microbes in just one or two days (esp. at RT, so store at 4 C). Mobile phase must be clean, filtered and freshly prepared using an established standard method (this should be described in a SOP so it is made the same each time). This applies to all mobile phase additives used too (choose the highest grade possible. “HPLC Grade” or ultra-high purity and look for an analysis sheet). Water used should be freshly prepared, high resistance (~18 Mega ohm) HPLC grade water which has been filtered through a 0.22 u filter.

3. Degassing. Mobile phase solutions should be continuously degassed before being introduced into the HPLC pump(s). This is sometimes done via low pressure Helium gas sparging, but more commonly today, via in-line, electronic vacuum degassing modules. Placing bottles in an ultrasonic bath or using vacuum filtration systems alone to degas solutions before use does a very poor job of degassing and results in the gas slowly dissolving back into the solution the very moment you stop degassing them. Baseline instability, drift and flow rate variation are the result. These problems may contribute to poor quality integration and high %RSD. To insure that your electronic vacuum degasser provides the maximum benefit, it must be purged before use each day to remove the gas which has re-dissolved back inside them. To properly flush each chamber/channel, you will need to know its volume (provided by the manufacturer). *Individual vacuum chamber volumes vary from ~ 0.50 ml to as much as 50.00 ml EACH! At a flow rate of 1.000 ml/min, it could take from 0.5 to 50 minutes just to flush one vacuum chamber (and most systems have 4 chambers!). Luckily, most modern vacuum degassers have chambers with volumes of between 0.5 and 12 mls each and HPLC pumps which can be run at either 5 or 10 ml/min maximum to make this process go faster. Remember to purge these chambers with fresh solution at the start of each day. 

Best Practice: Each channel of your HPLC system should be flushed (purged) with freshly degassed mobile phase at the start of each day, before you run any samples.

 

Before starting ANY HPLC analysis, the HPLC pump must be running and operating with no problems, achieving a stable baseline, steady flow rate with as little pulsation as possible (~ 1% ripple or less). Accuracy depends on this.

Part II of this article discusses initial method development steps.