No thanks! I would like to know more about CHROMacademy

 Over 3000 E-Learning topics / 5000 Articles & Applications
 

The CHROMacademy Essential Guide to
HPLC Troubleshooting - Autosampler, Column & Detector Issues

The Essential Guide from LCGC’s CHROMacademy presents the second in our series of webcasts on Practical HPLC Troubleshooting.  In this session, Dr. Kevin Schug (Associate Professor, University of Texas, at Arlington) and Scott Fletcher (Technical Manager, Crawford Scientific), present practical troubleshooting and maintenance information associated with HPLC autosamplers, columns and detectors.  This session examines different autosampler designs, injection valve operations and common problems associated with modern HPLC autosamplers.  Sample solvent and injection volume effects on chromatography will also be discussed in detail.  We will consider column hardware design, materials of construction and associated issues – including minimizing dead volume and avoiding / troubleshooting column blockage issues, especially with the popular Sub 2µm column packing materials.  We will also investigate the importance of controlling the temperature of the column during analysis and the potential chromatographic effects of poor temperature control.  We will conclude by reviewing common HPLC detector hardware problems and the chromatographic issues associated with detector hardware and acquisition settings.   We will build a checklist of suggested maintenance operations as well as outlining common diagnostic tests.
 
   

The Camtasia Studio video content presented here requires JavaScript to be enabled and the latest version of the Adobe Flash Player. If you are using a browser with JavaScript disabled please enable it now. Otherwise, please update your version of the free Adobe Flash Player by downloading here.

 

             
Presenters: Scott Fletcher
Technical Manager
Crawford Scientific
  Kevin Schug
Associate Professor
University of Texas at Arlington

 


Topics includeautosampler

  • Autosampler designs, common problems and avoiding carryover
  • Sample solvent and injection volume effects
  • Column hardware design, construction and mechanisms of column blockage
  • Column temperature control issues
  • Common HPLC Detector problems
  • Chromatographic symptoms and diagnostic tests
  • Common maintenance operations
 

Who Should Attend:

  • Anyone who uses HPLC equipment and who wants to improve their troubleshooting and instrument maintenance skills

Key Learning Objectives:

  • Learn different autosampler designs and injection valve configurations and operations
  • Explore common autosampler hardware problems and associated chromatographic symptoms
  • Appreciate the effect different sample solvents and injection volume  can have on chromatographic performance
  • Examine column hardware design, materials of construction and reasons / solutions for column blockage
  • Appreciate the requirements for column temperature control and chromatographic implications of poor temperature control
  • Examine frequently used HPLC detectors and review common detector hardware problems – especially those associated with flow cells and detector acquisition settings
  • Propose a maintenance schedule and several useful diagnostic tests
   

 

 

The CHROMacademy Essential Guide to HPLC Troubleshooting - Autosampler, Column & Detector Issues

This Essential Guide presents practical troubleshooting and maintenance information associated with HPLC autosamplers, columns and detectors.  This article examines different autosampler designs, injection valve operations and common problems associated with modern HPLC autosamplers.  Sample solvent and injection volume effects on chromatography will also be discussed in detail. 

We will consider column hardware design, materials of construction and associated issues – including minimizing dead volume and avoiding / troubleshooting column blockage issues, especially with the popular Sub 2µm column packing materials. 

We will also investigate the importance of controlling the temperature of the column during analysis and the potential chromatographic effects of poor temperature control. 

We will conclude by reviewing common HPLC detector hardware problems and the chromatographic issues associated with detector hardware and acquisition settings.

 

 

Accucore™ HPLC columns

New 2012-2013 Thermo Scientific Chromatography Columns and Consumables Catalog

Application notes

Troubleshoot an LC issue

Share this page

 
     

Introduction

Autosamplers are widely used in analytical laboratories to increase sample throughput, improve injection precision and enable unattended operation – so reducing the labor costs associated with manual injection.
Most autosamplers use six-port loop injection valves in order to deliver the sample plug to the analytical column. In modern autosamplers the rotor is driven by an electric motor, in older models, compressed air may be used.
All autosamplers have the same basic components which include, the injection valve, a syringe or sampling needle, a loop of either fixed or adjustable volume, a metering pump to aspirate the sample from the vial and an injection port through which the sample is introduced into the loop.

 

Share this page

Whilst manufacturers have developed different autosampler operating principles, all possess four essential components that allow mechanical automations of the injections process;

  1. Samples are contained in standardized vials.  Each sample vial is sealed by a septum, which can be either an integral part of the cap or held in place by the cap, which prevents selective evaporation of the sample solvent causing associated concentration changes.
  2. Sample vials are contained within trays that permit either their serial or random injection.  Such trays can be thermostatically controlled in order to help prevenat the degradation of thermally labile samples.  It is possible to sample from vials, well-plates or a combination of the two.
  3. An injection needle is employed to penetrate the septum and draw the specified sample volume for injection using a highly accurate metering device or small analytical pump head.  Depending on the autosampler operating principle the needle can either be movable or fixed.
  4.  An injection valve that allows the introduction of the sample into the loop prior to injection.  Such automated sample injection valves are operated by either pneumatic or electronic actuators.
 
 

Injection Valves

The valve is the heart of the autosampler system and operates by controlling the flow of eluent through the autosampler device to allow loop filling in one configuration and to sweep the loop contents onto the HPLC in the other configuration. Of course, more sophisticated valves can also be used for applications where perhaps two different sample loops are required or where analyte needs to be directed onto different analytical columns etc. [1, 2]

Valve injection allows the rapid, reproducible and essentially operator independent delivery of a wide range of sample volumes (e.g. from 60nL up to several milliliters), at pressures up to 1200bar (17,500psi) with less than 0.2% error.

Sampling valves can be located within a temperature-controlled oven for use with samples that require analysis at either elevated or lowered temperatures. The most common valves can be obtained in either 4 or 6 port configurations for use in either manual or automated mode.

 
 

 

 

Figure 1.  Autosampler.

 
 

How Injections are Made

Sample injectors for HPLC transfer the sample at atmospheric pressure from a syringe to a Sample Loop. The loop is then connected via a change in valve configuration, to the high-pressure mobile phase stream, which carries the sample onto the column. There are two methods of loading the sample:[1, 2]

  • Complete-filling – where the loop size chosen has the desired injection volume and is totally filled with sample
  • Partial-filling – where the loop chosen is at least twice the required sample volume and is only partially filled

Dual mode injectors allow both complete and partial filling, single mode injectors allow only complete-filling. These two techniques differ in accuracy, precision and the amount of sample required and will be discussed further in a later topic.

Use the Animation 1 to investigate the various steps in a manual injection process.

 

 

 

Animation 1.The injection process.

 
 

Different Autosampler Designs

Autosamplers are widely used in analytical laboratories to increase sample throughput, improve injection precision and enable unattended operation – so reducing the labor costs associated with manual injection.[1, 2]

Most autosamplers use six-port loop injection valves in order to deliver the sample plug to the analytical column. In modern autosamplers the rotor is driven by an electric motor, in older models, compressed air may be used.

 
There are three main operating principles which are used in autosampler design:
  • Pull-to-fill
  • Push-to-fill
  • Integral-loop autosamplers

Each of these variations is implicitly different and has different performance characteristics.

For more information on these autosampler types subscribe to CHROMacademy

 
 

How Injection Valves Work - Anatomy of the Injection Valves

It is necessary to divert the flow of mobile phase away from the sampling system (i.e. the syringe) when aspirating the sample loop prior to injection.  This is achieved using the injection valve containing a Rotor Seal. [1, 2]

The interface between the HPLC capillaries and the stator is known as the Stator Face.

Situated directly beneath the Stator Face is the Stator itself which is used to transport eluent or other liquids from various ports on the valve head to the channels on the Rotor Seal.  Essentially the Rotor Seal ‘joins’ liquid paths via the stator.

As the Rotor Seal turns against the Stator under high pressure it is important that the contact between these components is leak tight, flat and even.  This is achieved by the use internal Springs and Washers, further more Retaining Bolts that hold the injection valve components in place are tightened with equal torque.

Although most modern valves are ‘self-leveling’, older models use ‘set-screws’ which must be adjusted to ensure the stator face is level with the valve body –this is usually checked with the aid of ‘feeler gauges’.

On automated systems the Rotor Seal is moved using an electronic or pneumatic motor.

Figure 2. Anatomy of the injection valve.
 
 

Troubleshooting

The rotor seal is constantly rotating against the ceramic stator under a great deal of torque. Over time the channels on the rotor seal will start to widen. If particulate materials are trapped between the stator face and rotor seal, as the two turn, a scratch may develop between the two channels. The scratch may eventually develop into a ‘cross port scratch’ which ultimately results in sample or eluent leaking out of the device into the waste port during the injection or analysis phase of the injection. The poor mass transfer of the wider sample plug will ultimately result in broad chromatographic peaks and poor peak area (or height) reproducibility.

 

 

 

Figure 3. The rotor seal

There is no remedial action that can be implemented once a ‘cross part scratch’ has been developed and the damaged rotor seal must be replaced.  There are however a number of best practices that can be implemented in order to drastically reduce the likelihood of such a scratch developing;

  • Ensure the correct torque setting of the stator head or alignment in older models – it is essential that the retaining bolts are not OVER tightened
  • Prior to injection samples should be filtered in order to ensure all particulate matter is removed
  • The rotor seal must be compatible with the employed mobile phase and analytes – see later
  • The sample must be miscible in the mobile phase in order to elimate sample preipitaion upon mixing with the mobile phase
 
 

Problems Associated With Injection Valves

The sample first makes contact with the HPLC instrumentation in the Autosampler. Whilst manufactures produce parts from the most inert materials possible (whilst considering cost implications), and to the best standards of design and engineering - the autosampler is still subject to contamination. [1, 2] When analyzing ionic and bio-molecules, stainless steel contact parts are replaced with PEEK and titanium

Contamination is often presented in the form of sample carryover as evidenced when injection of a solvent blank produces a mini-version of the previous sample’s chromatogram. Most carryover occurs in the rotor seal via sample adsorption.

Extra peaks / Ghost peaks that are sharp are often due to sample contamination rather than system contamination. The appearance of broad, less efficient peaks within otherwise reasonable chromatograms may indicate the elution of highly retained species from previous injections.

 

 

 

Figure 4. Contamination and carry over.

 
 

Figure 5. Autosampler contamination.

 
 

Partial and Complete Loop Filling- What are the Rules for Sample Volume in Each Case and why

Complete Loop Filling

As the loop contains mobile phase, the sample solution introduced will necessarily mix and become diluted with this resident mobile phase whilst it is in the process of displacing it. Therefore, in order for the loop to be filled homogeneously, it should be overfilled between 2 and 5 times with the sample solution thereby eliminating any possible dilution effects.[1, 2]

Although the loop effectively controls the volume of sample injected, effectively making the amount of sample introduced from the syringe a noncritical parameter, it is typically recommended practice for maximum precision and linearity to overfill the loop with the same volume of sample solution each time i.e. 3x overfill ± 10%, therefore for a 100μL loop this would equate to a sample volume of 270−330μL per injection. Reproducibility is essential in maintaining precision and accuracy; the absolute volume injected is of less importance.

 

The Case: You are using complete loop filling (20μL loop), and no visible leaks can be found. How to make sure that no injection issues are taking place?

The Solution: inject 40, 60, 80 and 100 μL. Your loop has been filled twice, three, four and five times (the total loop volume is in this case is 20μL). If everything is OK, the peak area should be constant; if the peak increases or decreases, the injector should be investigated for leaks or mixing (dilution) of sample and eluent in the loop (increase sample amount injected).

 
 

Partial Loop Filling

The partial loop filling technique is typically utilized when it is important to preserve valuable or limited sample. To obtain the best possible precision the volume of sample solution injected should be no greater than 50% of the injector loop capacity. During loading of the loop the sample solution introduced “pushes” the mobile phase ahead of it out, and during this exchange process the front of the sample effectively becomes diluted.

This dilution effect is caused by the parabolic flow profile of a liquid moving through a piece of tubing,  -  the centre of the fluid stream travels faster than the sample liquid lying closer to the tubing wall (laminar flow).

Figure 6. The front of the sample band starts to mix with the trapped mobile phase resulting in sample dilution.

 

One practical way to prevent such sample injection discrimination is to inject an air bubble immediately in front of the sample volume to be injected. In this way any possible mixing of the sample solution with the mobile phase is prevented. [1, 2]

Figure 7. Air injection.

 

The consequence of this phenomenon is that the diluted sample effectively occupies approximately 2μL of loop volume for every 1μL of sample loaded from the syringe. Therefore, ensuring that <50% of the effective loop volume is loaded prevents the diluted front of the sample band from leaving the loop causing associated precision and irreproducibility errors.

 
 

Troubleshooting the Autosampler

In general terms, the source of carryover is any dead volume within the autosampler system capable of acting as a sample reservoir (localised sample retention).[1, 2]

The consecutive injection of blanks will help to identify whether the source of your problem is carryover or contamination.  If peak size reduces with subsequent sample injection then autosampler carryover should be suspected.  Otherwise, if peak intensity remains constant, then contamination can be suspected as the probable cause.

If contamination seems to be the source of your problem, then begin by injecting blanks prepared by using fresh and high quality chemicals and solvents.  Make sure all glassware is clean.  Perform injections with different amounts of sample to confirm whether or not the peak size responds proportionally to the amount of sample injected.

 
 

The Sample Needle

Sample needle blockages are one of the most frequently occurring autosampler problems. Judicious sample preparation will help minimise the number of problems created by particulate material, with either filtering or centrifugation of the sample solution prior to injection helping to minimise the problem.

A recommended cleaning regime for a blocked needle is as follows:

  1. Soak the needle in an appropriate solvent i.e. isopropanol. Do not sonicate the needle if the component has been welded, as the possibility exists that it may fracture or break at the joint
  2. Attempt to dislodge the obstruction with a reaming wire
  3. Connect the needle to the pumphead with a length of capillary and flush with progressively higher flow rates to “blow out” the blockage
 
 

The High Pressure Needle Seal and Capillary

On some autosampler designs sample and mobile phase first start to mix in the capillary just after the High pressure Needle seat. This capillary is prone to blockage by precipitated sample or buffer species. Pressure spikes on injection are a good indicator of such a blockage.  In some designs the needle seal may leak after repeated use, causing a mobile phase leakage and system pressure reduction during the injection cycle.

 
 

The Rotor Seal

Sample carryover, and/or poor peak area (height) precision, can indicate a worn rotor seal. One might also observe a leak from the injector waste port during the injection cycle.

The rotor seal should be evaluated for signs of wear, especially a deepening at the ends of the etched crossport grooves which can become sites of localised sample retention or for the afore mention ‘cross port scratches’. These are then washed clean on the subsequent blank injection producing a mini-chromatogram or “ghosting” peaks.[2]

Some rotor seal material may adsorb certain analyte types causing poor peak area (height) reproducibility.  One should investigate different materials for the rotor seal -  Vespel, Tefzel and PEEK varieties are available.

Figure 8. Rotor seal construction.
 

Many polymeric composites have been developed, since no single rotor seal material will perform satisfactorily in every situation.  All rotor seal materials are developed to cope with extreme conditions of temperature (100-350oC) and pressure (300 – 1000 psi) and exhibit great inertness towards mobile phase and sample matrix components.  However, rotor seals have their limitations:

  • polyimide/PTFE/carbon composites are well known for being attacked by ammonia, hydrazines, amines and solutions having a pH of 10 or more
  • polyaryletherketone/PTFE composites are softer and are attacked by high concentrations of sulphuric and/or nitric acid.  Long exposure to DMSO, THF or liquid methylene chloride should be avoided.  They are however stable over the entire pH range and are the rotor seal of choice when analysing ionic or bio-molecules

Thorough and regular system flushing can help minimise rotor seal contamination.

 
 

Peak Distortion Effects

Peak distortion can occur due to a mismatch between injection solvent and mobile phase strength.  In particular, when large amounts of sample are injected in a solvent that is much ‘stronger ‘than the mobile phase.  The diagram below shows typical peak distortion problems.

 

 

 

 

Figure 9. Peak shape abnormalities currently found when solvent strength is not considered when injecting.

 

 
 

In real life, peak distortion due to solvent strength mismatch is typically most severe in early eluting peaks.  As a rule of thumb, when stronger injection solvents are used, retention times are usually reduced.

The reason for this retention time reduction and peak shape distortion lies in the different “solvent environments” that peaks experience during separation.  After injection, the plug of strong injection solvent travels through the column and becomes diluted.  Analytes within the plug of strong solvent will travel faster across the column than those which have mixed with eluent.  As a consequence, sample components show differential migration rates within the column until the plug of strong solvent is fully diluted.  In order to avoid/correct this unwanted situation, sample should be injected in either:

  • a weaker solvent
  • reduced injection volume, so equilibration is almost instantaneously achieved through mixing in the tubing between the autosampler and column
 

Peak distortion problems, due to extra-column band broadening, can also be reduced by focusing the sample at the head of the column.  Different strategies have been proposed to achieve this.

 
 

Matching Injection Volume with Sample Solvent Strength

Under ideal conditions the strength of the sample solvent used would make no difference to the chromatographic separation, because the solvent would be diluted instantaneously to the same composition as the mobile phase. However, in practice it takes a finite period of time for the injected sample solvent to become diluted with the mobile phase. [2]

The following table provides guidelines for sample injection volumes based on the sample solvent strength with respect to that of mobile phase.

 
Sample Solvent Strength Maximum Injection Volume
100% Strong Solvent ≤ 10μL
Stronger than Mobile Phase ≤ 25μL
Mobile Phase ≤ 15% of Peak Volume
Weaker than Mobile Phase Large

Table 1.  Guidelines for sample injection volume based on the sample solvent strength.

 

Solvent Strength in HPLC Reverse Phase

Reversed phase mobile phases usually consist of water and an organic solvent often called a ‘modifier’. When ionisable compounds are analysed, buffers and other additives may be present in the aqueous phase to control retention and peak shape.

Chromatographically, in reversed phase HPLC water is the ‘weakest’ solvent. As water is most polar, it repels the hydrophobic analytes into the stationary phase more than any other solvent, and hence retention times are long – this makes it chromatographically ‘weak’. The organic modifier is added (usually only one modifier type at a time for modern chromatography), and as these are less polar, the (hydrophobic) analyte is no longer as strongly repelled into the stationary phase, will spend less time in the stationary phase, and therefore elute earlier. This makes the modifier chromatographically ‘strong’ as it speeds up elution.

 

 

 

 

Figure 10. Solvent strength in HPLC reversed phase (the numbers in brackets refers to the solvent polaraity, the inverse of solvent strength in reversed phase HPLC).

 

 

100% Strong Solvent

In this demanding situation it is necessary to keep the injection volume as small as possible, thereby minimising peak distortion. The recommendation is for no more than 10μL.  However,  lower concentrations of strong solvent (lower than 100%), will permit larger the injection volumes (larger than 10μL).

Stronger than Mobile Phase

When the sample solvent is no more than approximately 25% stronger than the mobile phase, then injection volumes as high as 25μL should be possible. However, always examine the chromatogram for possible peak distortion, given the inter-relationship of this and the preceding rule.

Generally, analyte solubility issues are the primary reason for increasing the sample solvent strength. To minimise such solvent strength variations on injection, dissolve the analyte at a high concentration in the strong solvent, then dilute with the weaker solvent to progressively correct for the difference in eluotropic strength. In a worst-case scenario dissolve in 100% strong solvent and inject as small a volume as possible.

Mobile Phase

The best injection scenario is when the sample solvent and the mobile phase are matched, both in terms of eluotropic strength and pH. In this situation, given the solvent homogeneity, you don’t have to be concerned with dilution effects.

As a rule of thumb, when injecting in mobile phase (reversed phase), injection volumes up to 15% of the volume of the first peak of interest can be made.  This is known as the 15% rule.

 
The 15% Rule:  if injecting in mobile phase (reverse phase conditions), then it is possible to inject approximately 15% of the volume of the first peak of interest.

Example: The first eluting peak in your chromatogram is 12 seconds wide at its base.  The eluent flow rate was 0.8 mL/min.  Calculate the maximum injection volume in mobile phase if the 15% rule holds.

Therefore:

Important: If stronger injection solvents are used, smaller volumes must be used and vice-versa, weaker injection solvents permit larger injection volumes being used.
 
 

Weaker than Mobile Phase

To assess the effect of the mobile phase strength consider the “Rule of Three”, which states that a 10% change in the strong solvent will result in an approximate threefold change in analyte retention time. Therefore, if a chromatographic peak had a retention time of 5min in a
mobile phase of 40:60 water:acetonitrile, then in a mobile phase of 60:40 water:acetonitrile its retention time would increase to approximately 45min (3 × 3 × 5min = 45min).

Consequently, if the sample were to be injected in 60:40 water:acetonitrile instead, then the analyte molecules would travel much more slowly through the column, until the sample solvent was fully diluted with the mobile phase.

Chromatographic bands travelling more slowly than the mobile phase tend to compress, because as mobile phase reaches the analyte molecules at the peak “tail” they will begin to move faster down the column, catching those analyte molecules further down the column that are still effectively residing in a weaker mobile phase strength.

As the difference between the solvent strength of the sample solvent and the mobile phase increases, it is possible to inject progressively larger and larger volumes. This effectively allows analyte on-column sample “focussing”, or concentration, and is often used, for example, in environmental analysis to assist in the detection of trace quantities of pesticides.

Exceptions to the above are whenever the mobile phase contains trace additives, and then it is always advisable to inject samples dissolved in the mobile phase. e.g. ion-pair separations rely on the equilibrium between the ion pair reagent in the mobile phase and the stationary phase. Injecting a sample in a solvent that doesn’t contain the ion pair shifts this equilibrium to the mobile phase, with resulting peak distortion and reproducibility issues.

 
 

General Autosampler Considerations

Mechanical Problems

Reliable autosampler operation requires that it be kept clean and regularly checked for proper adjustment. Ensuring mobile phase buffers and other reagent spills are removed quickly will help prevent any salt deposits building up, which could corrode or restrict the autosampler movement. Sample tray alignment should be checked periodically, many manufacturers now incorporate a software error message to inform of a missing or incorrectly aligned autosampler tray. [2]

 
 

Tray Position and Sample Identity

Incorrectly identified samples can have potentially serious consequences on the reported results. Whilst injection inaccuracies can be corrected by the use of an appropriate internal standard, and a blockage or mechanical failure evidenced by no injection being made, positioning of the autosampler is crucial for correct injection.

A quick visual check of the chromatograms, associated sample data and uniquely labelled vials to see if the standards correlate properly, before removing any vials from the autosampler tray, will enable the identification of any samples requiring re-analysis.

Sample Carryover

Sample carryover is evidenced when injection of a solvent blank produces a mini-version of the previous sample’s chromatogram. Most carryover occurs in the injection switching valve via sample adsorption.

Sample carryover between injections can be eliminated in a number of ways:

 
 

Implement a Needle-Rinsing Regime between Injections

Integral-loop autosamplers are configured to maintain the injection orientation on sample introduction, effectively allowing the continual washing of the injection system and switching valve over the duration of that chromatographic run, thereby reducing or eliminating sample carryover.

Other autosampler designs provide for an injection needle and/or syringe rinse between samples. This rinse solution may be from a separately located and independently filled source, or located directly in the autosampler tray, in a position defined by the user.

Minimise the Impact of Carryover.

Sample dilution, in addition to improving autosampler precision as discussed earlier, will reduce carryover because autosamplers have a finite hold-up volume i.e. 0.1μL which represents residual sample volume. Therefore, with a 10μL sample injection this residual volume gives a 1% carryover, but would only give 0.1% carryover for a corresponding 100μL injection.

As an alternative to sample dilution, arrange samples from low to high concentration, as a fixed-volume carryover will produce fewer problems when a low concentration sample is followed by a high concentration sample rather than vice versa. [3]

For further information relating to carryover, the interested reader is directed to following excellent references: [4, 5]

 
 

Leaks

Air can potentially be introduced through the permeable walls of any PTFE tubing being used in the autosampler. This would not represent a problem if stainless steel or PEEK tubing were used instead in the sampling portion of the unit.

A poorly connected or partially tightened nut connecting the sample loop to the injection needle may also allow air ingress or sample leakage when sample solution is withdrawn from the vial. An incorrectly assembled switching valve body, or a compromised component lying within, may produce a leak in the valve body. Strip to evaluate, and reassemble according to the manufacturer's instruction to rectify.

 
 
     

Overview

The column is the only device in the HPLC system which actually separates an injected mixture. Column packing materials are responsible for the separations taking place, and therefore, they are of primary importance for successful separations.

The performance of the separation of HPLC columns is influenced by the packing material (particle size, shape and distribution).[6, 7, 8]

A wide range of column construction materials are currently available, such as: stainless steel, titanium based alloys, PEEK and glass.  The table below illustrates some commercially available HPLC columns.

 

 

 

Share this page

 
Column Properties Diagram
Stainless Steel Perhaps the most widely used material in column manufacturing.  Stainless steel columns are highly inert and will stand extreme conditions of pressure (usually above 400 bar.  Some high pressure applications could exceed 1300 bar).
Titanium There are situations where stainless steel is not inert enough, for example, when dealing with very aggressive or corrosive solvents or ionic compounds.  Titanium-Zirconium allows are well known for their superior mechanical strength and high inertness.
PEEK These metal free columns can in excess of 90 bar and are excellently suited for most HPLC solvents and additives and in particular for the analysis of ionic and bio-molecules.  However, concentrated nitric and sulfuric acids, tetrahydrofuran and dimethylsulfoxide should be avoided.
Glass These metal free columns are low cost and exhibit inertness, chemical resistance and high efficiency.  They are compatible with most HPLC solvents and additives.  However, hydrofluoric acid should be avoided.  Glass columns are designed to cope with moderately low pressures (10 – 20bar); however, modern glass columns can stand pressures up to about 90 bar.

Table 2.  Selected column construction materials.

 

The physico-chemical properties of the support material have been described at length in CHROMacademy, please follow the links below:

HPLC columns can also be classified by considering their capacity / loadability.

Figure 11. HPLC column classification.
 
 

Column Geometry and Installation

In order to achieve optimum results, low dead volume fittings and accessories should always be used, avoiding band broadening and peak shape problems will be avoided. [1, 2, 3, 9, 10]

It is a good practice to utilize low volume in-line filters to protect your HPLC column from particulate matter.  By so doing, the life time of your valuable column will be enhanced while preventing the frit from being clogged.

The figure below provides a deep inside of a commercial HPLC column.

Figure 12. Commercial HPLC column.
 

Frit: A major cause of column deterioration and damage is the buildup of particulate and chemical contamination at the head of the column. This can lead to increased back pressure and anomalous chromatographic results.  HPLC Columns normally contain stainless steel inlet and outlet frits (acting as filters) which retain the column packing and allow the passage of the mobile phase.  The pore size of the frit must be smaller than the particle diameter of the packing, e.g., a 0.5 μm frit for 1.8 μm packing.  Sample depositing on the column inlet frit can lead to peak shouldering as demonstrated below. 

 
Figure 13. Column inlet frits.
 

Removal of the blockages on the column inlet frit is time consuming and not with out risk, The column is required to be reversed and a low mobile phase flow, usually ranging from low to high strength and back down to low is employed.  This will by no means guarantee to restore the column to its full pre-blockage working order.

A better solution is to employ an in-line filter that is situation prior to the analytical column.

 
Figure 14. In-line filter.
 

For more information relating to column care please refer to follow document: [12]

 
 

Tubing and Accessories

Stainless steel tubing is required for very high pressure applications where other types of tubing such as PEEK aren’t feasible.  Stainless steel tubing is reasonably flexible, but is best when bends are kept to a minimum.

The tubing length and ID in particular must be minimized before the column so that analyte band broadening and gradient dwell time is as short as possible. [13]

Fittings are used to connect the various components of an HPLC system together while maintaining a leak free flow-path.  Using the correct zero dead volume fitting eliminates dead volume that can contribute to peak broadening, decreased resolution, split peaks, sample carry-over and gas collection. No matter how sophisticated the instrumentation, it cannot compensate for poor fittings.

Fittings and column connections have specialized assembly procedures that must be followed. Not doing so risks excessive band broadening, peak area reproducibility and loss of accuracy.  Always follow your manufacturer’s instruction to attain a ‘zero dead volume’ connection.

Band broadening can occur not only in the analytical and guard columns, but also because of the dead volume in the injector, detector, or plumbing connecting the various components of the HPLC system (extra-column band broadening). This effect, called extra-column dispersion must be minimized for high efficiency. The proper choice and use of tubing and fittings is critical to achieve excellent results.
 

 

 

 

Figure 15. Schematic showing the effect of incorrectly matching fittings.

 

 

Extra column volume can be effectively reduced by ensuring that all connections are correctly made and are zero dead volume.  Due to traditional end-fitting materials such as PEEK (polyetheretherketone) materials not being able to seal under the increased pressures generated in UHPLC, stainless steel fittings are commonly used.  As the stainless steel fitting cannot deform to seal into the fittings in which they are installed it is essential that the correct type of fittings are used. 

The use of optimized lengths and internal diameter of capillary tubing is also important to retain the highest efficiency in UHPLC systems.  Tubing length is important, but the internal diameter is of much greater importance and this is demonstrated by the Aris – Taylor equation shown below:

Where; F is the flow through capillary tubing, L is the length of the capillary tubing, DM is the analyte diffusion coefficient and id is the capillary tubing internal diameter.

The overall dispersion of the analyte band is proportional to the length of the capillary tubing but is proportional to the capillary tubing internal diameter raised to the 4th power, thus producing a profound effect upon analyte peak dispersion.  This same principle may be applied to flow cell volumes, which are minimized in UHPLC systems to avoid similar dispersion effects.

 

Important

When dealing with tubing and accessories, please bear in mind the following pointers:

  • Nuts and ferrules will change from vendor to vendor; do not mix products from different manufacturers
  • Do not over tighten nuts and/or ferrules
  • Do not reuse an installed nut or ferrule for any other connections
  • Always use the appropriate tools
  • Neither tubing, nor accessories should react, interact, deteriorate or compromise the integrity of your sample, i.e. they should be ‘inert’ towards your sample
  • Low dead volume accessories should always be used
 
 

Column Nuts and End Fittings

Nuts are manufactured to fit one specific column end fitting type and the nut and the user should check the compatibility between the column and end fittings used.   The figure below demonstrates what happens when three commonly utilized styles of end fittings are not correctly matched. [14]

Figure 16. Fittings from selected manufacturers.

 

Figure 16b. End fitting geometries (dead volume highlighted in red colour) from selected manufacturers.

 
 

Column Care

Filtering

Filtering is one of the crucial aspects to consider to maximize column lifetime.  It has long been recognized that unfiltered mobile phase and/or samples will lead to clogs and over-pressurization of the system as discussed earlier.

Animation 2.  Sample particulate material can block HPLC columns leading to high system back pressure and shortened column lifetime. [15]

 

Filtration must be implemented whenever buffers are used (no matter how pure they are) to remove undissolved particulates.

The following table will help to plan the correct filtering strategy for a mobile phase.

Situation Filtration Strategy
Traditional HPLC columns 0.45 micron filter
Sub-two-micron columns (frit porosity in the order of 0.5 micron) 0.2 micron filter

Table 3.  Mobile phase filtering.

 

Filtration of HPLC samples should always be considered as no matter how the sample was prepared, there is always a chance that small particles remain in the matrix.  This can conveniently be achieved by using syringe filters as depicted below.

Animation 3.  Ultrafiltration.

Make sure that your selected filter is compatible with the sample matrix and analytes and should have neither affinity nor reactivity towards any of the sample components.

 

Figure 17. Syringe filters.

 
 

Guard Columns

Guard columns are shorter columns packed with the same stationary phase as found in the analytical column (but not of the same high quality and hence much cheaper).  These devices are placed in front of the analytical column to prevent impurities reaching the column – acting as a chemical rather than a physical filter.

Chemical damage principally occurs at the column inlet, where sample adsorption and stationary phase damage (dissolution of silica, channeling, etc.) are not uncommon.  The guard column will assume the role of the of the top portion of the HPLC column, so any damage to the analytical column will be mininized, increasing the overall column life.  Regular replacement of the guard column is advisable.

A guard column is inserted between the injector and analytical column to protect the latter from chemical damage (silica dissolution / bonded phase hydrolysis) or loss of efficiency due to the presence of particulate matter or strongly adsorbed impurities from analytical samples.  For maximum protection against contaminants and particulate matter, the guard column can be placed between a set of frits (that act as filters).  However, the primary function of the guard column is to act as sacrificial stationary phase in order to preserve the much more expensive stationary phase employed in the analytical column from aggressive solvents or solutes that would irreversibly damage or bind.

 
 

Figure 18. Guard column installation.  Courtesy of Optimize Technologies, Oregon, OR.

 

 

Please bear in mind that the use of guard columns can compromise the efficiency of very low volume columns and or columns packed with very small particle sizes (which require very low dead volume).  In such cases, low volume inline filters should be used instead.[16, 17, 18, 19]

When to use a guard column.

Situations where guard columns should be used include:

  • Contaminated samples
  • High sample throughput required
  • High temperature required
  • High back pressure developed after modest column usage
  • Variable retention time after a few injections
 

For additional information the interested reader is directed to following articles:[17, 18]

 
 

Voids in the Stationary Phase

Working outside the ideal pH range of your column could compromise the integrity of the stationary phase (silica dissolution), so voids are likely to develop at the head of the column.

Silica is soluble at high pH conditions; therefore, silanol accessible functional groups (which are present in all silica based columns) can be attacked by strong bases while developing voids in the packing material, with the consequent loss of efficiency.

HPLC columns are designed to operate under high pressure conditions; however, columns can be damaged by sudden variations in pressure.  Reasons for pressure variation include:

  • Slow rotation of the sample injection valve
  • Fast startup of the pump
  • Column switching operations

Voids are also formed when silica dissolution occurs and particles collapse causing bed compression when the column is pressurized.

Peak shouldering and broadening is one of the most common problems due to voids in the stationary phase.[20]

Figure 19. Peak splitting, shouldering and broadening due to column voids.

Where as in older and in cartridge type columns the inlet frit could be removed and loose stationary phase added as temporary fix, newer columns do not allow this with end fittings being permanently fixed in place.

Reasons for peak shouldering and splitting include:

  • Column void
  • Column contamination
  • Contaminated or partially blocked frit
  • Injection solvent stronger than mobile phase

Note that if peak shouldering affects only one peak within the chromatogram, then rather than stationary phase voids, co-elution should be suspected.

 
 

Column Overload

This condition will cause different problems: poor peak shapes (broadening, tailing, fronting and asymmetry), variable retention times, selectivity issues, etc.  Column capacity depends upon many factors; however, the table below provides the means to quickly identify situations where a real possibility of column overload exists.

Column Type Typical Dimensions Capacity (mg)
Analytical 25 cm × 4.6 mm < 0.5
Semi-Preparative 25 cm × 10 mm < 100
Preparative 25 cm × 21 mm < 500

Table 4.  HPLC column capacity (oversimplified).

* Values per total amount of analyte

Please note that the previous table is intended to create awareness of the possibility of overloading your column.

There are two forms of column overloading: concentration and volume overloading. Both of them lead to a decrease in chromatographic resolution.[21]

 
 

Concentration Overloading

When the sample of interest has good solubility in the mobile phase, then highly concentrated sample can be injected.  When the column capacity is exceeded, the situation is known as concentration overload.

When performing concentration overloading, the retention factor (retention time) is reduced and peak shapes tend to be triangular rather Gaussian.

  Figure 20. Concentration overloading.
 
 

Figure 21. The effect of increasing sample concentration on a 50 x 4.6mm id column packed with C18 material and eluted at 1mL/min with acetonitrile:water (60:40).

Please note the when only marginally concentration overloading the column then peak fronting will be observed.  As the analyte concentration is successively increased then the peak shape deformation will change from fronting to broadening and finally to the tailing peak as depicted above.

 
 

Volume Overloading

Column overloading can occur when injecting large volumes of sample.  In this case, peaks tend to be broad rather than the expected Gaussian peaks and  retention time will increase.   There is also a point after which further increasing the volume of injected sample, does not lead to an increase in peak height – this is due to overloading the detetcor.

 

Figure 22. Volume overloading.

 
 

Other Column Care Considerations

In general terms, HPLC columns are stable within a pH range of 2 to 8.  However, modern HPLC columns can endure extreme conditions of pH.  Please always refer to the column manufacturers operating guidance when operating outside of these traditional pH limits. 

Aggressive Solvent:  There are certain eluent systems and components that are aggressive towards certain columns (or components).  For example, hydrofluoric acid will attack glass columns.  Solutions of strong acids or bases will promote corrosion of stainless steel columns.

Never allow your column or HPLC system to stand with water or buffer in it for long periods of time as this will promote bacterial growth or component corrosion respectively.

Solvent Compatibility:  Make sure the new mobile phase is miscible with that in the column and will not cause a precipitate to form.  If the new mobile phase is not compatible in the aforementioned way, then use an intermediate solvent and flush for at least 20 column volumes.  Please follow the link below for additional information:[22]

Column Equilibration:  The equilibration time of a column is determined by the column dimensions.  As a rule of thumb, columns are equilibrated after being flushed with 10 column volumes.  A column volume can be calculated using the below equation:

where:

DC - column id,
L - column length

W - column porosity (~68% or 0.68)

 

The table below reports equilibration time (approximated) for selected HPLC columns.

 
Column Dimensions Column Volume Flow Rate (mL/min) Equilibration Time (min)
250 mm × 4.6 mm 2.83 1.0 28.3
150 mm × 4.6 mm 1.70 1.0 17
100 mm × 4.6 mm 1.13 1.0 11.3
50 mm × 4.6 mm 0.57 1.0 5.7
250 mm × 4.0 mm 2.14 1.0 21.4
125 mm × 4.0 mm 1.07 1.0 10.7
150 mm × 2.1 mm 0.35 1.0 3.53
100 mm × 2.1 mm 0.24 1.0 2.36
50 mm × 2.1 mm 0.12 1.0 1.18

Table 5.  Equilibration time for selected HPLC columns.

Note that if you increase the flow rate, then shorter equilibration times will be expected.

 
 

Column flush

Figure 23:  Flushing the HPLC column.

For more information in column flushing, cleaning and regeneration, please refer to the following documents: [23, 24]

 
 

Column Storage:  For short term storage (i.e. overnight or even a couple of days) columns can be stored in mobile phase (given that it inhibits microbial growth); however, consult your column manufacturer for more details.  Long term storage should be performed according to the guidelines given by your column manufacturer.

Column Cleaning and Regeneration:

Irreversible adsorption of impurities on the column will cause a number of problems such as peak shape (peak broadening, splitting, etc.), selectivity and retention time issues.  There are many protocols for column regeneration; however, the best approach is to consult your column manufacturer.

The following are column regeneration protocols for generic columns.

Regeneration of reversed-phase columns (C18, C8, C4, C1, C30, CN or phenyl):

  1. Flush the column with 20 column volumes of water
  2. Flush the column with 20 column volumes of acetonitrile
  3. Flush the column with 5 column volumes of isopropanol
  4. Flush the column with 20 column volumes of heptane
  5. Flush the column with 5 column volumes of isopropanol
  6. Flush the column with 20 column volumes of acetonitrile

 

Regeneration of normal-phase columns:

  1. Flush the column with 20 column volumes of heptane
  2. Flush the column with 5 column volumes of isopropanol
  3. Flush the column with 20 column volumes of acetonitrile
  4. Flush the column with 20 column volumes of water
  5. Flush the column with 20 column volumes of acetonitrile
  6. Flush the column with 5 column volumes of isopropanol
  7. Flush the column with 20 column volumes of heptane
 
 

Channeling:  Channels occur when voids are created in the packing material of an HPLC column.  The presence of channels will cause peak fronting, as solvent (and solutes) will travel faster through them than in the rest of the column.

 

Animation 4.  The presence of channels will generate speed migration differences between different components of mobile phase, thus yielding peak shape problems. [25]

 

Channels are produced for a number of reasons; the most important, however, are related to the packing process itself (when manufacturing the column), the erosive action of mobile phase and components (solids, bubbles, etc.) and mechanically stressing or shocking the column such as dropping it.  There is no remedial action for a channeled column and it must be replaced.

 
 
     

Temperature plays an important role in HPLC; this is because both the kinetics and thermodynamics of the chromatographic process are temperature dependent.

Column temperature control is essential to ensure reproducibility of retention times and the stability of the detector baseline.  A small temperature variation could lead to errors in quantification or peak identification.

When the column is under well-thermostatted conditions, a radial temperature profile will develop.

 

 

 

 

Share this page

 

Figure 24. Oversimplified temperature profile in a HPLC column with temperature control.

 
 

Frictional Heating

In nearly all reversed phase separations, an increase in temperature will reduce analyte retention. Additionally solvent viscosity is reduced at elevated temperature which in turn means lower backpressure.

While flowing through the column, the eluent system experiences frictional forces which increase its temperature (frictional heating).  These frictional forces are related to the speed of the eluent system, its chemical nature and the contact area with the packing material (which in turn depends upon the particle size).

Frictional heating of the mobile phase can cause a non-uniform increase in temperature inside the column that can adversely affect the separation (peak shape, broadening problems and retention factor).  A typical frictional heating profile within a column is shown below.[26, 27, 28]

Figure 25. Oversimplified temperature profile in a HPLC column with no temperature control.
The temperature increases in the flow direction (left to right); each temperature zone is identified with a different colour.

In order to avoid frictional heating use a pre-column heat exchanger and a thermally controlled column compartment.

Certain detectors are affected by the temperature of the column’s effluent (for example refractive index detectors).  Post-column temperature control should be implemented in such cases.

 

Column Thermal Deterioration

Temperature control is of overriding importance to HPLC, not only to ensure analysis reproducibility but to keep your column in good working conditions.  Silica support materials are prone to attack by aggressive mobile phases (in terms of pH) and this process is increased at temperatures above 60oC.  The combination of high temperature plus extreme pH (above 7) will lead to faster column deterioration.

 
 
     

Introduction

The most common HPLC detectors are based on the absorbance of UV (or Visible) light by the analyte molecule. These detectors are popular because of their low cost, robustness, reasonably low detection limits, and ease of use. There are several common types of ‘UV’ detectors including; single wavelength, multiple wavelength and (photo)diode array configurations. The diode array detector also allows some limited qualitative functionality via the ability to dynamically collect UV spectra. Several other detector types are available which use other physico-chemical properties of the analyte molecule and these include:

  • Flouresence
  • Evaporative Light Scattering
  • Refractive Index
  • Electrochemical
  • Electrical Conductivity

These detectors tend to be employed when high sensitivity is required and/or the analyte molecule does not absorb UV radiation. Mass Spectrometric detectors have become increasingly important and these are now widely used in many areas of analytical chemistry as they can be used for the detection of low analyte concentrations, for qualitative analyses or in situations where separation of analyte components is difficult. Refractive index detection is useful when analyte molecules do not respond to any other detector type or if the analyte shows large differences in refractive index from the mobile phase used.

 

 

Share this page

 

Figure 26. HPLC Detection.

UV-Vis detector has become the standard in HPLC, not only for its flexibility and power but for economic reasons also; therefore this part of the essential guide is dedicated to them.

 
 

General Terms and Concepts

Selectivity

Non-selective detectors react to the bulk property of the solution passing into the detector. When a compound elutes from the column, this bulk property changes and the change is measured and recorded (i.e. refractive index).

Selective detectors do not react to the bulk solution passing through but measure a response due to a specific property of the solute molecule (i.e. UV absorbance).

Sensitivity

The smallest detectable signal is usually estimated to be equivalent to three times the height of the average baseline noise at or around the peak of interest – this would give a signal to noise ratio of 3:1 for the ‘Limit of Detection’ of the detector . If the amount of analyte injected is less than this, then the signal ceases to be distinguishable from noise.

For quantitative analysis a signal to noise (S/N) ratio of 10:1 is recommended for the ‘Limit of Quantitation’.

 

Figure 27. Sensitivity and signal to noise ratio.

The sensitivity and relative selectivity of some common HPLC detectors are compared in the table below.

Detector Selectivity Sensitivity Min Mass Detected
Refractive Index Low 1 – 5 μg
Conductivity Low 10 – 50 ng
UV/Visible Medium 0.5 – 1.0 ng
Electrochemical High 50 – 500 pg
Fluorescence High 10 – 100 pg
Mass spectrometer High 10 –100 fg
Evaporative Light Scattering Low 0.1 – 1.0 ng

\

 
 

The Flow Cell

The most commonly used detector for HPLC analysis is the UV – Visible detector. The detector fulfills many of the criteria for an ‘ideal detector’ including:

  • Sensitive
  • Wide Linear Range
  • (Relatively) Unaffected by Temperature Fluctuation
  • Selective and suitable for gradient elution
The schematic animation below shows the basic flow cell of a UV detection system. As the mobile phase leaves the analytical column it enters, and fills the flow cell. Light from the UV (Deuterium) or Visible (Tungsten) lamp shines through the flow cell and its content.
 

Animation 5.  The UV Flow Cell

 

Light intensity of the emergent light from the flow cell is measured using photo diodes, which produce an electrical signal when exposed to light. The greater the intensity the greater the absorbance and the larger the resultant signal. Analytes which contain a UV chromophore(s) will cause a large absorbance difference when they elute into the flow cell and the light exiting the flow cell will reduce markedly – generating a large response.

By constantly measuring this electrical signal it is possible to produce a plot of Absorbance against Time – the Chromatogram.

 
 

Multi-Wavelength Detectors

Fixed wavelength detectors are known for being more sensitive than their multi-wavelength counterparts by an order of magnitude.  This advantage, however, is outweighed by the flexibility of multi-wavelength instruments.

There are two basic types of multi-wavelength detectors currently used with HPLC instrumentation, they are:

  • Variable wavelength detector (VWD)
  • Diode Array detector (DAD) –also known as Photo Diode Array detector (PDA)

The following links provide a good introduction to optical detector types:[41, 42]

Variable Wavelength Detector (VWD)

In this case, the wavelength for detection can be varied from a single to multiple wavelengths with grating adjustments.

Figure 28 shows the optical path of a conventional variable wavelength detector. Polychromatic light from a deuterium and/or tungsten lamp is focused onto the entrance slit of a monochromator using spherical and/or planar mirrors.

 

Figure 28. Variable Wavelength Detector (VWD).

 

The monochromator selectively transmits a narrow band of light to the exit slit.

The measurement wavelength is selected via the data system – this is achieved using a grating mounted on a turntable. The grating is positioned to allow light of the chosen wavelength to pass to the send mirror and so on through the flow cell or the reference diode.

The magnitude of analyte absorbance is determined by measuring the intensity of the light reaching the photodiode without the sample (reference photodiode) and comparing it with the intensity of light reaching the photodiode after passing through the sample (sample photodiode).

Many variable wavelength detectors are time programmable and can be programmed to switch wavelength during an analysis in order to select a suitable wavelength for each analyte. It is not easily possible to dynamically record sample spectra with this type of detector – although it is possible using stopped flow techniques.

 

Figure 29. UV Wavelength.

 
 

Diode Array Detectors (DAD)

The diode array detector can provide detection at a single wavelength or multiple wavelengths. This detector also has the ability to dynamically acquire and store spectra for peak purity analysis, library searching and/or the creation of extracted signals.

 
Figure 30. Diode Array Detector (DAD).
 

The combined tungsten and deuterium lamps emit radiation from 190-850 nm. The radiation is collimated through the flow cell, then a mechanically controlled slit. The radiation is dispersed at the holographic grating into individual wavelengths of light. Each photodiode receives a different narrow wavelength band. A complete spectrum is taken approximately every 12 ms and spectra and signals are created and stored.

The array consists of over 1000 diodes, each of which measures a different narrow-band wavelength range. Measuring the variation in light intensity over the entire wavelength range yields an absorption spectrum. The entrance slit can be programmed so if high sensitivity is required, the slit is opened to wide and if maximum spectral resolution is desired, then the slit is narrowed. Spectral resolution and detector sensitivity are inversely proportional.
 
Figure 31. Uses of diode array detectors.
 
 

UV absorbance

In UV-Visible detection, the useful detection wavelength range is between 190 nm and 850 nm (with a tungsten lamp). Deuterium lamps are used for excitation of analyte molecules in the UV region (~180 – 380nm), Tungsten lamps are used for measurements in the visible region (380 – 800nm). Below about 210nm it is possible for the solvent used in the mobile phase to interfere with the analyte absorbance measurement and ultra pure solvents must be employed.

Electrons tightly bound in single carbon/carbon or carbon hydrogen bonds absorb electromagnetic energies corresponding to wavelengths less than 180 nm, below the useful operating range for a typical UV-Vis detector. Analyte molecules containing only C-C or C-H bonds do not show high sensitivity in UV-Vis detectors. Unshared electrons in the outer orbital’s of constituent atoms may exhibit larger absorbance’s in the useful range. This would include the unshared electrons of sulfur, bromine, and iodine.

Electrons within unsaturated system such as double or triple bonds within organic molecules are relatively easily excited by UV radiation, and generally show absorbance within the useful UV-Visible region of the spectrum. Therefore, compounds with unsaturated and aromatic characteristics generally exhibit useful absorbance spectra. Approximate absorbance wavelengths for some common functional groups are shown below.

 
Chromophore Structure λmax (nm)
Amine -NH2 195
Ethylene -C=C- 190
Ketone 195
Ester -COOR 205
Aldehyde -CHO 210
Carboxyl -COOH 200-210
Nitro -NO2 310
Phenyl 202, 255
Naphthyl 220, 275

Table 7.  Approximate absorbance wavelengths for some common functional groups.

 
 

The Electromagnetic Spectrum

Electromagnetic radiation varies in both frequency and wavelength.  The range of all possible frequencies of electromagnetic radiation is known as the “Electromagnetic Spectrum”.

Figure 32. The Electromagnetic Spectrum.

 

UV detectors can be used with a gradient providing that the solvents do not absorb significantly over the wavelength range that is being used for analyte detection.

UV cutoff can be regarded as the wavelength at which the solvent absorbance in a 1 cm path length cell is equal to 1 AU (absorbance unit) using water in the reference cell.

Figure 33.  UV Cutoff interpretation.

 
 

Broadly speaking, reversed phase solvents are transparent to UV light and are suitable to UV detection.  In contrast, normal phase solvents are more likely to absorb UV light and care must be taken when dealing with them.  Table 8 helps to select HPLC solvents for UV detection.

 
Solvent* UV Cutoff (nm)   Solvent* UV Cutoff (nm)
Ethyl Acetate 256 Ethyl sulfide 290
DMSO 269 Isopropyl ether 220
Acetonitrile 190 Isopropyl chloride 225
Amyl alcohol 210 Isopropanol 205
Acetone 330 Methanol 205
Amyl chloride 225 Methylene chloride 233
Benzene 280 Methyl-isobutyl ketone 334
Cyclohexane 200 Methyl ethyl ketone 330
Cyclopentane 200 1-Nitropropane 380
Carbon disulfide 380 Nitromethane 380
Carbon tetracholoride 265 Octane 215
Chloroform 245 n-Pentane 190
Diethyl amine 275 Petroleum ether 210
Ethanol 210 n-Propanol 210
Ethylene glycol 210 n-Propyl chloride 225
2-Butoxyethanol 220 Pyridine 330
p-Dioxane 215 Toluene 385
Ethyl acetate 256 Tetrahydrofuran 230
Ethylene dichloride 230 Xylene 290
Ethyl ether 220    

* Solvents with high UV cutoff values such as acetone (UV cutoff 330 nm) are unsuitable for analyses at low wavelengths such as 205nm.

Table 8.  UV transparency of selected HPLC solvents.

 
 
Buffer pKa Concentration pH Range UV Cutoff (A > 0.5)
Trifluoroacetic acid <<2 (0.5) 0.1% 1.5 - 2.5 210 nm
KH2PO4/phosphoric acid 2.12 0.1%

1.1 - 3.1

<200 nm
Tri-K-Citrate/hydrochloric acid 1 3.06 10mM 2.1 - 4.1 230 nm
Potassium formate/formic acid 3.8 10mM 2.8 - 4.8 210 nm
Tri-K-Citrate/hydrochloric acid 2 4.7 10mM 3.7 - 5.7 230 nm
Potassium acetate/acetic acid 4.8 10mM 3.8 - 5.8 210 nm
Tri-K-Citrate/hydrochloric acid 3 5.4 10mM 4.4 - 6.4 230 nm
Ammonium formate 3.8 50mM 2.8 - 4.8 N.A.
9.2 50mM 8.2 - 10.2
Bis-tris propane HCI/Bis-tris propane 6.8 10mM 5.8 - 7.8 215 nm
Ammonium acetate 4.8 50mM 3.8 - 5.8 N.A.
9.2 50mM 8.2 - 10.2
KH2PO4/ K2PO4/ 7.21 0.1% 6.2 - 8.2 <200 nm
Tris HCI/Tris 8.3 10mM 7.3 - 9.3 205 nm
Bis-tris propane HCI/Bis-tris propane 9.0 10mM 8.0 - 10.0 225 nm
Ammonium hydroxide./ammonia 9.2 10mM 8.2 - 10.2 200 nm
Borate (H3BO3/Na2B4O7 10 H2O) 9.24 N.A. 8.2 - 10.2 N.A.
Glycine  HCI/glycine 9.8 N.A. 8.8 - 10.8 N.A.
1-methylpiperidine HCl/1-methylpiperidine 10.1 10mM 9.1 - 11.1 215 nm
Diethylamine  HCI/diethylamine 10.5 N.A. 9.5 - 11.5 N.A.
Triethylamine  HCI/triethylamine 11.0 10mM 10.0 - 12.0 <200 nm
Pyrollidine  HCI/pyrollidine 11.3 N.A. 10.3 - 12.3 N.A.

Table 9.  UV transparency of selected HPLC buffers and additives.

 
 
Mobile Phase UV Cutoff (nm)
1% Acetic acid 230
10 mM Ammonium bicarbonate 190
0.1% BRIJ 35 190
0.1% CHAPS 215
50 mM Diammonium phosphate 230
10 mM Ammonium acetate 205
1 mM Na2-EDTA 190
0.1% HCl 190
10 mM HEPES (pH 7.6) 225
10 mM MES (pH 6.0) 225
10 mM Potassium phosphate-monobasic 190
10 mM Potassium phosphate-dibasic 230
PIC Reagent B-6 (1 vial/liter) 225
PIC Reagent D-4 (1 vial/liter) 190
PIC Reagent A (1 vial/liter) <200
1 M NaCl 208
10 mM Sodium citrate 225
10 mM Sodium acetate 205
0.1% SDS 190
10 mM Sodium formate 200
0.1% Trifluoroacetic acid (TFA) 205
0.1% Triethylamine 235
0.15 Tween 20 190
0.1% Triton-X 100 240
20 mM Tris-HCl (pH 7.0) 204
20 mM Tris-HCl (pH 8.0) 212

Table 10.  UV transparency of Selected HPLC Mobile Phases.

 
 

DAD Spectra

A diode array detector contains over 1000 diodes, each of which measures a different narrow-band wavelength range. Measuring the variation in light intensity over the entire wavelength range yields an absorption spectrum.

Absorbance spectra can therefore be collected at all data points within the chromatogram. In fact each data point within the chromatogram is the result of the cumulative absorbance at all selected wavelengths.  Use the slide rule on the diagram below to see how the absorption spectra vary across a typical chromatogram.

 
Figure 34. DAD spectra at various times.

The ability to extract spectra has lead to many possibilities in qualitative data analysis.

 
 

DAD Bandwidth

The Bandwidth parameter in Diode Array detection is related to the number of diode responses which are averaged in order to obtain a signal at a particular wavelength.

A wide bandwidth has the advantage of reducing noise by averaging over a greater diode range. Because Noise is random, averaging the response over a larger range of diodes will reduce noise. As the bandwidth is increased, the signal intensity (detector sensitivity) increases as some diodes will result in a lower absorbance, compared to a reading using only the single most intense wavelength (λmax).

A wide bandwidth results in a larger range of wavelengths are averaged when producing a spectral data point - resulting in a loss in spectral resolution.

Click on the arrows in the diagram below to increase and decrease the bandwidth. Note how the resultant chromatogram and peak spectra change. As with most detection systems in analytical chemistry – detector sensitivity and resolution are inversely proportional.

 

Figure 35. DAD band width.

 
Bandwidth (nm) Signal (10-4 A.U.) Noise (10-4 A.U.) Relative Signal to Noise Ratio
4 14.70 0.41 1.00
8 14.51 0.34 1.19
16 13.79 0.26 1.49
32 11.75 0.18 1.82
64 8.40 0.14 1.62
128 6.53 0.13 1.40

Table 11.  Bandwidth and signal to noise ratio in DAD detectors.

 

The diagram below further illustrates how increasing the bandwidth up to an application dependant threshold improves the chromatographic signal.  As the bandwidth is increased, the signal intensity is reduced.  This occurs because the response is an average of the intensity from diodes at lower and higher intensity –according to the analyte spectrum.  As the bandwidth is increased, however, the noise (which is random) is reduced and as the noise falls off at a faster rate than the signal, the signal to noise ratio and hence the detector sensitivity increases.  The ideal bandwidth is determined as the range of wavelength at 50% of the spectral feature being used for determination –this is highlighted below.

 

Figure 36. DAD bandwidth effect on chromatographic signal.

 
 

DAD Slit Width

Some diode array UV-Vis detectors have a variable slit at the entrance of the spectrograph. This is an effective tool for adapting the detector to the changing demands of different analytical problems.

A narrow slit (width) provides improved spectral resolution for analytes who give UV spectra with enough fine detail to be useful for qualitative analysis. For example, improved spectral resolution will increase the confidence of library matching search results when attempting to identify unknown peaks within a chromatogram.

A wide slit (width) allows more of the light passing through the flow cell to reach the photodiode array, hence the signal intensity and detector sensitivity increase. Baseline noise is also reduced – again leading to an increase in signal to noise ratio.

However, with a wider slit, the spectrograph’s optical resolution (its ability to distinguish between different wavelengths) diminishes. The wavelength of light falling on each diode becomes less specific as the light becomes more diffuse. Any photodiode receives light within a range of wavelengths determined by the slit width, and so spectral resolution decreases.

Click on the arrows below to increase and decrease the slit width. Note how the resultant baseline noise and the peak spectra change.
 

Figure 37. DAD slit width effect on chromatographic signal.

 
 

Response Time

Response time describes how fast the detector signal follows a sudden change of absorbance in the flow cell. The detector uses digital filters to adapt response time to the width of the peaks in your chromatogram. These filters do not affect peak area or peak symmetry.

Decreasing the Response Time effectively allows the detector to take more measurements per unit time – this results in a larger data file and a better defined chromatogram. Low response times are characterized by increases in peak height (or area) as well as a noisier baseline.

Increasing the response time results in averaging more data points and so reduces the noise by the square root of the number of data points. The drawback to increasing the response time is a slight loss in the peak height or area. Response time again results in a compromise between how well the chromatographic signal and UV spectrum are defined (i.e. the spectral resolution) and the sensitivity (as defined by signal to noise ratio).

Generally, 20 - 25 data points across a chromatographic peak are required for accurate quantification.

Click on the arrows below to increase and decrease the response time. Note how the resultant baseline noise and the peak spectra change.

 

Figure 38. DAD response time effect on chromatographic signal.

 
 

Reference Wavelength

The reference wavelength compensates for fluctuations in lamp intensity as well as changes in the absorbance / refractive index of the background (i.e. the mobile phase) for example during gradient elution.

This figure below illustrates the effect of reference wavelength on detector sensitivity. Noise is reduced as the reference wavelength is moved closer to the sample signal. Without any reference measurement, all noise and variability in lamp intensity is recorded within the signal. When using a reference signal, all lamp intensity variation and background (mobile phase) variability is subtracted out of the signal being measured. The closer the reference wavelength is to the sample wavelength, the more effectively these background deviations are catered for and the better the detector sensitivity. However, one should not select the reference wavelength too close to the analyte wavelength or the signal intensity may be seriously reduced.

Choice of a proper reference wavelength can reduce variability and drift in the chromatographic baseline resulting in better signal-to-noise performance.

Click on the arrows below to change the reference wavelength. Note how the resultant noise level changes.
 
Figure 39. DAD reference wavelength effect on chromatographic signal.
 
 

Choosing Sample and Reference Settings

The detector can compute and store several signals simultaneously and also manipulate the signals together in order to yield a composite or deconvoluted chromatogram. The following signals are usually collected using diode array detectors:

  • sample wavelength, the center of a wavelength band with the width of sample bandwidth (BW)
  • reference wavelength, the center of a wavelength band with the width of reference bandwidth

The signals comprise a series of data points over time, with the average absorbance in the sample wavelength band minus the average absorbance of the reference wavelength band.

The example opposite shows an empirical method for determining typical sample and reference settings. This approach will usually result in flatter baselines during gradient analysis and lead to better limits of detection.

 

Figure 40. Determining sample and reference settings.

 
 

Basic UV Troubleshooting Considerations

Whenever you have a detector problem you must first ensure the problem exists by repeating the analysis and recreating the problem.

 

Contamination

Chemical contamination of the flow cell itself or of its windows may result either from improper system cleaning or from the sample type itself. Dirty flow cell windows will lead to a progressive loss in both incident and exiting light intensity, affecting the absorbance of analytes and their response over time.[2]

Contamination can lead to drifting baselines, ghost peaks and baseline spikes.  In addition, contamination can cause plugging of the flow cell outlet capillary.  When working at high pressure conditions, a plugged outlet can lead to cracks in the cell windows which manifest as, noisy baselines or a permanently off scale detector response.  Similar problems can be encountered when mobile phase is deteriorated or prepared with low quality chemicals and reactants.

Figure 41.  A loss in Sensitivity caused by dirty flow cell windows.

 

Any absorbance contribution made by the flow cell can be checked and measured through a specific intensity test; this typically involves filling the flow cell with H2O as a reference solvent.

The following generic discussion for cleaning the flow cell is intended as a supplement to specific manufacturer’s recommendations for any particular detector; please consult the operator’s manual before using any of these procedures.

The general cleaning procedures that follow are applicable for all optical detectors, and can take the form of one of the following:

  • Backflushing
  • Solvent cleaning
  • Acid cleaning

Back-flushing the flow cell often removes particulate matter that may have blocked the inlet tubing. Connect the pump to the detector outlet and reverse the flow of eluent through the flow cell, directing the solvent flow to a waste container. If high backpressure was observed, then the pressure should fall as the blockage is effectively removed.[2]

Apply caution, as excessive pressure across the flow cell may cause leakage around the window gaskets or even crack the quartz windows themselves.  To avoid over pressurizing the flow cell set the maximum pressure cutoff for the pump below that of the flow cell, gradually increasing the flow rate in increments of 0.1mL/min to avoid pressure shocking the flow cell. The standard 1.0cm path length flow cell is generally pressure rated to 120 Bar.

Cleaning with solvents is useful if the presence of an organic soluble contaminant or droplets of an immiscible solvent within the flow cell is suspected. By drawing a series of solvents with different physical and chemical properties i.e. viscosity, polarity etc. through the flow cell, each of which is miscible with the preceding and following solvents, any contaminants possessing a broad range of solubilities can effectively be removed. Such a solvent series might consist of methanol, followed by tetrahydrofuran, methylene chloride, before returning to methanol.

Acid cleaning is the most aggressive of the three cleaning procedures described. This process requires the drawing of an acid through the detector flow cell to remove contaminants, after which H2O is passed through to flush out the acid. To ensure removal of all residual traces of the acid, and any potentially UV absorbing residues at low wavelengths, H2O is then pumped through the flow cell. Nitric acid is preferred, as halogenated acids, i.e. HCl, aggressively attack the stainless steel components of the flow cell.

 

Warning: if using PEEK then avoid hydrochloric acid solutions and never use nitric acid.

 

The following generic acid cleaning protocol can be used to clean the UV cell:

  1. Disconnect the flow cell from column and wasting line
  2. Place the flow cell’s outlet line in a beaker
  3. Use a syringe, or any other appropriate means, to draw about 10 mL of propanol through the cell (this will remove any residual mobile phase)
  4. Use a syringe, or any other appropriate means, to draw about 10 mL of distilled water
  5. Use a syringe, or any other appropriate means, to draw about 10 mL of HCN (6.0 N)
  6. Use a syringe, or any other appropriate means, to draw about 20 mL of distilled water
  7. Use a syringe, or any other appropriate means, to draw about 100 mL of HPLC water
  8. If reversed phase mobile phases are to be used, flushed the cell with mobile phase
  9. If normal phase mobile phases are to be used, flushed the cell with 10 mL of isopropanol and then flush with mobile phase

Note that flushing the cell in the reverse direction will increase the likeliness of particulate matter trapped in the detector tubing being back-flushed out of the system.

If these cleaning methods do not resolve a contamination or blockage problem, then the flow cell will have to be either rebuilt or replaced.  Rebuilding is possible and usually less expensive than acquiring a completely new flow cell. Consult your operator manual for specific recommendations and advice.

 
 

Degassing

The presence of gases in the mobile phase will compromise the quality of your HPLC separation and lean to very noisy baselines. By removing dissolved air you help prevent possible oxidative degradation of the sample and mobile phase.

As the mobile phase moves into the flow cell from the relatively small volume of the capillary tubing into the large volume of the flow cell it experiences a large change in pressure. Any air dissolved in the solvent may degas into the flow cell causing baseline noise.

Air bubbles passing through the flow cell will cause spurious peaks, noisy and unstable chromatographic baselines. Inadequate solvent degassing is the usual cause of bubble problems, especially if high-pressure mixing has been employed.  Then the reduced pressure resulting on solvent mixing allows for subsequent out-gassing, as the air becomes less soluble in the solvent mixture than the original pure solvent.

Dirty flow cell windows can trap micro-bubbles as they pass through. Such micro bubbles themselves normally pass through the flow cell unnoticed, but are now able to self-associate to form a much larger air bubble; their effect will be to contribute to a rise in baseline noise.

 

Animation 6.  Air in the flow cell.

 

If excessive air accumulates in the flow cell it may agglomerate into a more significant air bubble. In this case the bubble will move in time with the very small pressure variations caused by the pump. As the air bubble passes in front of the flow cell windows – changes in absorbance will occur – which manifest themselves as an undulating baseline or noise spikes.

A trapped air bubble will need purging from the flow cell, this may require the use of isopropanol, whose higher viscosity compared to acetonitrile or methanol helps dislodge the air bubble carrying it to waste.  Immediate and apparent off-scale absorbance readings may be the result of a cracked flow cell window. Solvent passing through the flow cell can leak onto the outside of the window surface, causing such absorption changes.

Don’t confuse an air bubble in the detector flow cell with a lamp problem. Air bubbles may flow through the detector and cause a sharp spike, or a disturbance that resembles a chromatographic peak, although of appreciably narrower peak width. In other instances the air bubble may stop in the flow-cell and cause a dramatic baseline shift; usually off-scale.

Inadequate solvent degassing is the usual cause of bubble problems, especially if low-pressure mixing has been employed. Then the reduced pressure resulting on solvent mixing allows for subsequent out-gassing, as the air becomes less soluble in the solvent mixture than the original pure solvent.

 
 

Temperature and Flow Control

Temperature fluctuations will give rise to drifting baselines.  The lack of temperature control will affect signal response with the following detector types:

  • Refractive index
  • Conductivity
  • UV detectors (working at high sensitivity or in indirect more)

To avoid temperature problems, not only temperature control but heat exchangers before the detector are recommended.  The main problem with heat exchangers is the fact that they increase the dead volume, and of course peak broadening.

Mobile phase mixing and/or flow rate variations will give rise to drifting baselines.

For more information, please refer to the following references:[43, 44]

 
 

Lamps

The operating life of any detector lamp is necessarily a function of the total number of lamp hours for which that particular lamp can perform useful work. Therefore, a lamp’s useful life will be proportionally shorter if the application is a demanding trace analysis rather than an assay method, as once the chromatographic baseline noise and/or drift exceeds the stated limits of the method then the lamp will require replacing.

 

Figure 42.  Drifting and Noisy baselines indicative of an ageing lamp.

 
 

Deuterium Lamps

In general terms, the deuterium lamp consists of a glass vessel that is made of fused silica or UV glass.  Inside this vessel, a high purity nickel box (that contains an internal electrode structure) is placed and kept under deuterium (usually 99.7%).  This construction is capable of producing and sustaining an arc discharge that strongly emits ultraviolet light.

The high purity nickel box contains the anode that is usually made of molybdenum, and the cathode, which consists of a complex tungsten spiral or filament coated with emissive material (usually a barium, strontium, calcium mixed oxide).
 

Figure 43.  The deuterium lamp.

 

A deuterium lamp produces intense UV radiation over the wavelength range 190−400nm, by means of a plasma discharge within a deuterium (D2 or "heavy hydrogen") atmosphere. 
The output from a deuterium lamp decreases according to an exponential decay function. Therefore, output decreases slowly until quite near the end of the lamp’s lifetime, when it begins to decrease quite rapidly, finally resulting in a failure to ignite. This rapid deterioration in output may provide the user with some warning of an imminent lamp failure. As the lamp output diminishes there will be a corresponding decrease in the measured S/N, and consequently the analyte sensitivity required by a method will not be attained.

A deuterium lamp maybe considered unusable when its output falls to below 50% of its original value at a specified wavelength, or when a system suitability sample fails to meet a defined S/N ratio. There are three major fundamental reasons for decreasing output with time:

  • Evaporation of tungsten, molybdenum from the filament coating, causing a reduced output and ignition problems
  • Reaction of the coating material with the quartz envelope
  • Solarization (when UV lamps made of fused silica are exposed to UV radiation, they undergo changes that reduce their ability to transmit UV light) of the quartz envelope at 200−250nm, causing the quartz to buildup absorption between these wavelengths

The expected lifetime of a Deuterium lamp is typically between 800 and 1000 hours, with long-life deuterium lamps providing between 1800 and 2000 hours. The tungsten source of the visible portion of many UV/Visible detectors has an extremely long life, and usually fails by burning out.

The useful output life of a deuterium lamp is inversely proportional to the number of ignitions it sees. Therefore, the practice of leaving the lamp on continuously will affect an approximate three-fold increase in its useful life, based on an average 8-hour working day. Turning the lamp on and off during breaks in the analysis or overnight, will therefore not only effectively shorten the lamp’s lifetime, but ultimately will be unproductive as the lamp will then require warming up for approximately 30 minutes in order to stabilize its baseline.

 

Figure 44.  The working state of your lamp will have a major effect on your chromatogram. [43]

Many software systems have a built in diagnostic functionality. It is therefore possible to measure lamp intensity and age.

 
 

Tungsten Lamps

Tungsten lamps are often used in conjunction with deuterium lamps, and emit light in the near UV and visible regions of the electromagnetic spectrum, typically over the wavelength range 340−850nm. [2]

In general terms, the tungsten lamp consists of a glass vessel where a tungsten filament is heated either under vacuum or under an active filling gas.  The incandescent tungsten filament will emit visible light by resistance heating.

The bulb material for tungsten halogen lamps is quartz, which allows working temperatures of up to 900° C and operating pressures of up to 20 bar.

As the temperature is increased, the tungsten filament evaporates more rapidly.  In order to counteract this evaporative process, some designs implement active gases, such as bromine.  Unfortunately, this process cannot be completely ruled out and in the long run the lamp will fail (filament burn outs).

Tungsten lamps should only be held by the base.  At high temperatures, any finger marks will react with the glass (it will turn opaque) thus compromising the work condition of the whole lamp.

Do not exert any shear or torque forces on the lamp or its base, as the lamp may broken.  And allow plenty of time (at least 15 minutes) for the lamp to cool down before touching it, as serious burn injuries may occur otherwise.
 

Figure 45.  The tungsten lamp (note that the vacuum lamp design is achieved by avoiding the use of any gas within the bulb).

 
 

Wavelength Accuracy

Wavelength accuracy should be checked, either by employing a known calibration solution with defined absorbance maxima, or by using a characteristic emission band of the deuterium lamp.

Poor wavelength accuracy will compromise quantitative results.  Wavelength accuracy is determined by comparing the wavelength of a standard with its measured value.  There are two methods recommended by the European Pharmacopoeia to test for wavelength accuracy, some of them are shown below.

  • Deuterium Emission Line Method
  • Holmium Perchlorate Method
 

Deuterium Emission Line Method

This method scans the deuterium (D2) emission lines and it doesn’t require additional instrumentation other than the detector by itself.

 
Wavelengths (nm)

0 ± 0.7
486.0 ± 0.2

656.1 ± 0.2
Photometric Mode Transmission
Scan Range

-2 to -2
484 to 488

654 to 658
Spectral Bandwidth (nm) 0.20
Signal Averaging Time (seconds) 0.033
Data Interval (nm) 0.02
Lamp UV
Beam Mode Single

Table 12.  Deuterium Emission Line Method.

 

Holmium Perchlorate Method

This method scans the deuterium wavelengths of peaks; it uses not only a holmium perchlorate solution (4% w/v solution of holmium oxide -95%- in 1.4 M perchloric acid) but a blank.  The solutions used can be purchased as part of a UV wavelength test kit.

 
Wavelengths (nm)

241.15 ± 1.0
287.15 ± 1.0
361.50 ± 1.0

536.30 ± 3.0
Photometric Mode Absorbance
Scan Range

231.15 to 546.30

Spectral Bandwidth (nm) 0.10
Signal Averaging Time (seconds) 0.10
Data Interval (nm) 0.02
Lamp UV-Vis

Table 13.  Holmium Perchlorate Method.

 

There are other certification tests currently accepted to make sure your UV detector is in good working order.  These tests, however, are outside the scope of the present presentation and interested readers are invited to explore relevant instrument certification information.  The following links can provide further information:

//www.edqm.eu/en/Homepage-628.html
//www.usp.org/

And for more information, you can consult your system provider/manufacturer.

 
 

Scan Rate

Detectors used for fast or ultra high performance HPLC, must have a very high scan rate in order to collect the requisite number of measurements across the peak to properly model the peak profile.

Modern fast HPLC detectors present scanning rates of more than four times the speed of a traditional HPLC detector:[44]
 
Detector Frequency (Hz)
Fast HPLC 80 - 100
Traditional HPLC 15 - 20

Table 14.  Typical separation speeds of HPLC detectors.

 
Increasing the scanning speed (or the number of points registered per unit time), more information is gathered and a more accurate response is obtained.  In the example below, low scanning speeds give low resolution chromatograms and in extreme cases, no signals are obtained.
 

Figure 46.  Determination of Gamma-Hydroxybutyrate (GHB) and Gamma-Butyrolactone (GBL) by fast HPLC/UV-VIS (215 nm).  Column: C18, 50mm×4.6mm×5.0μm (packed with partially porous particles).  Mobile Phase: 2.5 mL/min 70% methanol in 10mM buffer KH2PO4 (pH=3.0).

 
Issue Symptom Potential Remedial Action(s)
Signal outside linear dynamic range
  • Peaks with variable or unexpected intensity
  • Rounded peaks
  • Reduce sample volume and/or concentration
  • Prepare samples making sure that absorbance lies in the range 0.1 < A < 0.7
Contaminated sample or eluent system
  • Peaks with variable/unexpected intensity
  • Variable retention times
  • Prepare fresh sample
  • Use only HPLC grade water and chemicals
Absorbing eluent system Peaks with variable/unexpected intensity Select another eluent system
Wrong detection parameters Peaks with variable/unexpected intensity Evaluate analyte response at different detector settings
Contamination in detector cell Peaks with variable/unexpected intensity Clean cell
Data collector not electrically grounded Spikes in the Chromatogram
  • Replace data collector and assess
  • Check and reseat all visible lead connections
  • If necessary arrange for an expert visit
Data collector failure Erratic baseline
  • Replace data collector and assess
  • Look for expert advice
Electric interference Spikes in the Chromatogram
  • Isolate HPLC system, check if source of problem is internal or external
  • Correct as necessary
  • If necessary arrange for an expert visit
Strongly absorbing component in strong eluting solvent Positive baseline drifting
  • Select detection parameters in such a way that solvent effects are minimized
  • Check correct solvent was used
  • Check detector polarity
Strongly absorbing component in weak eluting solvent Negative baseline drifting
  • Select detection parameters in such a way that solvent effects are minimized
  • Check correct solvent was used
  • Check detector polarity
Faulty lamp
  • Peaks with variable/unexpected intensity
  • Short frequency baseline
  • Erratic baseline
Replace lamp
Air on cell
  • Short frequency baseline
  • Erratic baseline
  • Negative peaks
  • Low intensity peaks
  • Degas mobile phase
  • Flush system to remove air from detector cell or pump
Recorder gain set too low Rounded peaks Adjust gain
Detector and/or recorder time constants are set too high Rounded peaks Reduce settings to lower values (or until no further improvements are seen)
Wrong dead volume Peak broadening Detector should be designed to give the smallest possible internal volume
Recorder polarity Negative peaks Check polarity
Leaky flow cell Low intensity peaks Suspect this problem only if the cleaning process has proven to be ineffective.  Replace as appropriate
Dirty cell
  • Additional signals in the chromatogram
  • Low intensity peaks
Clean cell
Detector signal wires are reversed Negative peaks
  • Fix as appropriate
  • If required arrange for an expert visit
Lens contamination Negative peaks Clean as appropriate
Temperature variation Peaks with variable/unexpected intensity Implement temperature control

Table 15.  Troubleshooting the UV-Vis detector.

 
 
 
     
  1. The Theory of HPLC.  “Autosamplers” from CHROMacademy
  2. “Practical HPLC Troubleshooting & Maintenance”  Crawford Scientific 2010
  3. “Crawford -Scientific Fundamental HPLC With HILIC and SPS” Crawford Scientific 2010
  4. John W. Dolan.  “Autosampler Carryover” Pp 1-3. LC/GC Europe - March 2001
  5. John W. Dolan.  “Reader’s Questions.  Carry Over, Mobile-Phase Temperature, and Column Care”  LC/GC Volume 17.  Number 11.  November 1999.
  6. Gerard Rozing.  “Trends in HPLC Column Formats — Microbore, Nanobore and Smaller”  Agilent Technologies GmbH, Waldbronn, Germany.  Recent Developments in LC Column Technology.  June 2003.  Pp 1-7.
  7. L. R. Snyder and J. J. Kirkland.  Introduction to Modern Liquid Chromatography (2nd edition), John Wiley and Sons, New York, 1979.
  8. G. Guiochon, in C. Horváth (Editor), High Performance Liquid Chromatography, Advances and Perspectives.  Vol. 2, Academic Press, New York, 1980
  9. Chen MH, Horváth C (1997) J Chromatogr A 788:51–61
  10. Felix C. Leinweber, Ulrich Tallarek.  “Chromatographic performance of monolithic and particulate stationary phases Hydrodynamics and adsorption capacity”  Journal of Chromatography A, 1006 (2003) 207–228
  11. SGE Analytical Science.  Application note BR-0342-H_RevD.  August 2010
  12. John W. Dolan.  “Column Care and Feeding”.  LC/GC Volume 11.  Number 1.
  13. Band Broadening” from “The Theory of HPLC” of CHROMacademy
  14. CHROMacademy Resolver Issue 5: High Efficiency HPLC separations.  Sub 2µm (UHPLC) fully porous particles and Superficially Porous Silica particles (SPS)
  15. Sample Preparation.  “Primary Sample Preparation Techniques” from CHROMacademy
  16. Agilent ZORBAX Reliance Cartridge Guard-Columns.  //www.agilent.com/chem/supplies
  17. Ronald E. Majors.  “Top 10 HPLC Column Myths”  Pp 1172-1182.  LC/GC North America.  Volume 24.  Number 11.  November 2006.
  18. John W. Dolan.  “Frit Cleaning”  Separation Science.  May 11, 2011
  19. HPLC and LC/MS.  Thermo Electron Corporation.
  20. Edward Kim.  Care, Maintenance, and Troubleshooting of HPLC Columns.  Agilent Technologies. 2008
  21. Udo Huber and Ronald E. Majors. “Principles in preparative HPLC” Agilent Technologies Inc. Publication Number 5989-6639EN. Printed in Germany. April 2007.
  22. Ronald E. Majors.  “Columns for Reversed-Phase LC Separations in Highly Aqueous Mobile Phases”  LC/GC Europe.  December 2002
  23. Ronald E. Majors.  “The Cleaning and Regeneration of Reversed-Phase HPLC Columns” LC/GC Europe.  July 2003
  24. John W. Dolan.  “Column Flushing Demystified”  LC/GC Volume 7.  Number 7.
  25. Sample Preparation.  “Method Development” from CHROMacademy
  26. Luis A. Colón, José M. Cintrón, Jason A. Anspach, Adam M. Fermier and Kelly A. Swinney. “Very high pressure HPLC with 1 mm id columns" Analyst, 2004, 129, 503-504
  27. H. Poppe, J. C. Kraak, J. F. K. Huber and J. H. M. van den Berg.  “Temperature gradients in HPLC columns due to viscous heat dissipation”.  Chromatographia, Volume 14, Number 9 / September, 1981
  28. André de Villiers, Henk Lauer, Roman Szucs, Stuart Goodall, Pat Sandra.  “Influence of frictional heating on temperature gradients in ultra-high-pressure liquid chromatography on 2.1mm I.D. columns”  Journal of Chromatography A, 1113 (2006) 84–91
  29. Luis A. Colón, José M. Cintrón, Jason A. Anspach, Adam M. Fermier and Kelly A. Swinney. “Very high pressure HPLC with 1 mm id columns" Analyst, 2004, 129, 503-504
  30. H. Poppe, J. C. Kraak, J. F. K. Huber and J. H. M. van den Berg.  “Temperature gradients in HPLC columns due to viscous heat dissipation”.  Chromatographia, Volume 14, Number 9 / September, 1981
  31. André de Villiers, Henk Lauer, Roman Szucs, Stuart Goodall, Pat Sandra.  “Influence of frictional heating on temperature gradients in ultra-high-pressure liquid chromatography on 2.1mm I.D. columns”  Journal of Chromatography A, 1113 (2006) 84–91
  32. H. Poppe.  “Some reflections on speed and efficiency of modem chromatographic methods”  Journal of Chromatography A, 778 (1997) 3-21
  33. The Theory of HPLC.  “Fast HPLC” from CHROMacademy
  34. G. Guiochon, in C. Horváth (Editor), High Performance Liquid Chromatography, Advances and Perspectives.  Vol. 2, Academic Press, New York, 1980
  35. “Band Broadening” from the “HPLC Channel”.
  36. A.-M. Siouffi.  “About the C term in the van Deemter’s equation of plate height in monoliths”  Journal of Chromatography A, 1126 (2006) 86–94
  37. Alain Bertho and Alain Foucault.  “Comments on Van Deemter plot in high speed countercurrent chromatography”  Journal of Liquid Chromatography & Related Technologies.  2001; 24(13); Pp 1979 - 1985
  38. Bruce E. Poling, John M. Prausnitz, John P. O’Connell.  “The Properties of Gases and Liquids”  Fifth Edition.  Chapters 4, 5 and 6.  McGraw-Hill 2000.
  39. Instrumentation of HPLC.  “Detectors” from CHROMacademy
  40. John W. Dolan.  “Optical Detectors.  Part I: General Principles”  LC/GC Volume 2.  Number 4.  Pp 290-292
  41. John W. Dolan.  “Optical Detectors.  Part II: Fixed-Wavelength UV Detectors”  LC/GC Volume 2.  Number 5.  Pp 365-367
  42. John W. Dolan.  “Optical Detectors.  Part III: Variable-Wavelength UV Detectors”  LC/GC Volume 2.  Number 6.  Pp 439-442
  43. Michael D. Nelson and John W. Dolan.  “UV Detector Noise”  ”  LC/GC Volume 17.  Number 1.  Pp 13-15
  44. Anja Prüß, Christine Kempter, Jens Gysler, Thomas Jira.  “Extracolumn Band Broadening in Capillary Liquid Chromatography”  Journal of Chromatography A, 1016 (2003) 129–141
 
 
     
As a member of CHROMacademy, you will also get access to LCGC magazine articles from your favourite authors - John Dolan, John Hinshaw, Mike Balough, and Ron Majors; over 300 news items refreshed daily; vendor application notes, webcasts and podcasts; electronic laboratory tools and calculators; and access to the growing Essential Guides archive, recorded tutorials by industry experts.

Enroll online today at: www.chromacademy.com

Subscribe now for $399
and get instant access to all CHROMacademy Essential Guide Webcasts


subscribe now

  join today
 
 
     
 

The following subjects are covered in CHROMacademy.com

The Theory Of HPLC
Introduction (1.5hrs)
Chromatographic Parameters (3hrs)
Band Broadening (3hrs)
Column chemistry (4hrs)
Reverse phase (partition) chromatography (6hrs)
Ion-Pair Chromatography (3hrs)
Normal phase (absorption) chromatography (3hrs)
Gradient HPLC (3hrs)
Quantitative and Qualitative HPLC (3hrs)
FAST HPLC (4.5hrs)
HILIC (3hrs)
SFC (3hrs)
Ion Chromatography(3hrs)

Theory and Instrumentation of GC
Introduction (1.5hrs)
Chromatographic Parameters (3hrs)
Band Broadening (3hrs)
Gas Supply and Pressure Control (2hrs)
Sampling Techniques (4.5hrs)
Sample Introduction (5hrs)
GC Columns (5.5hrs)
GC Temperature Programming (3hrs)
GC Detectors (2.5hrs)
SFC (3hrs)

Instrumentation of HPLC
Mobile Phase Considerations (3.5hrs)
Solvent Pumping Systems (4hrs)
Autosamplers (4.5hrs)
Detectors (4.5hrs)

Solid Phase Extraction
Molecular Properties (4hrs)
SPE Overview (3.5hrs)
SPE Mechanisms (4.5hrs)
Method Development (5.0hrs)
Primary Sample Preparation Techniques (2hrs)
Liquid / Liquid Extraction Techniques (1.5hrs) Approaches to Automation for SPE (1.5hrs)

Fundamental GC-MS
Introduction (1.5hrs)
GC Considerations (4.5hrs)
GC -MS Interfaces (2.5hrs)

Fundamental LC-MS
Introduction (1.5hrs)
Electrospray Ionisation Theory (6hrs)
Electrospray Ionisation Instrumentation (4hrs)
Mass Analyzers (9.5hrs)
Atmospheric Pressure Chemical Ionisation (3.5hrs)
Atmospheric Pressure Photoionisation (3hrs)
Solvents, Buffers and Additives (3.5hrs)
Vacuum Systems (3hrs)
Flow Rates and Flow Splitting (3hrs)
Orbitrap Mass Analyzers (3hrs)

MS Interpretation
General Interpretation Strategies (11hrs)
Intro to MS Proteomics Research (3.5hrs)

loading data
loading data
loading data
loading data
loading data
Home | About UsContact Us | SubscribeTerms and Conditions | Advertise | Privacy Policy |

loading data

loading data

loading data

 

loading data


loading data