<%flg="B"%>
 

What Do You Do If ?

The following information answers some frequently asked questions about ZORBAX RP-HPLC columns for analytical chemistry. The topics discussed include:

 

Peak Tailing and Basic Compounds

Question: I’m trying to separate some very difficult basic compounds at pH 2.5 using a C18 column and their tailing factors are still too high — around 2.0. Which column should I try next?

Answer: The Bonus-RP column would be a good column to try. This column has an amide group in the alkyl chain. This amide group reduces interactions between basic compounds and the residual silanols on the silica surface by acting as an internal competing base and reducing peak tailing of basic compounds. The Bonus-RP column, while ideal for the mid-pH range where silanol interactions are more likely, will also improve peak shape at low pH. The Bonus-RP column uses the same sterically protecting bonding used for StableBond columns. This improves the lifetime of the Bonus-RP column at low pH and makes it a good choice in situations like this.

 

Question: Should I just select the Bonus-RP column for all my method development with basic compounds?

Answer: Basic compounds are best analyzed following the method development scheme outlined earlier in this brochure. Select an SB-C8 or SB-C18 column for initial development and use a buffered low pH mobile phase. Many times this approach provides a good separation and the StableBond columns will have exceptional lifetime at low pH even at high temperatures.

 

Question: If one peak in a chromatogram is tailing but the others are not, what is the likely cause?

Answer: Since I don't know the chemistry of the sample or details regarding the mobile phase or the column it is difficult to answer this question. Peak tailing can be caused by a variety of reasons and I would prefer to ask you several questions about your sample before I submit a response. Since most of the peaks in your chromatogram are well shaped and only one is tailing, I suspect that the chemistry of the column and sample are such that a secondary interaction is inducing the tailing. Modifying the mobile phase or selecting another HPLC column can reduce these secondary interactions.

Back to Top

Increasing Retention of Basic Compounds

Question: I can’t get enough retention of my basic compounds at low pH, what column should I try next?

Answer: The Eclipse XDB column can be used in the intermediate pH region — from pH 3 - 8 for the longest lifetime. The primary advantage of using the intermediate pH region is a possible increase in retention for basic compounds. Most basic compounds have pKa values of 5 or greater, therefore at pH 6 - 8 some of these compounds may become non-charged and the column will retain them more. In addition, at pH > 5 the residual silanols on the silica surface will become charged. This can lead to stronger interactions with basic compounds and increase retention. The Eclipse XDB column is the ideal column to use in this mid-pH region because the dense-bonding and double endcapping will cover the most active silanols on the surface of the column, and any residual silanols can contribute to increased retention of basic analytes without causing excessive peak tailing. Figure 1 shows a plot of retention vs pH for a group of basic compounds. At pH 6 and higher the retention of all of these compounds increases due to increased interaction with the column, though at pH 6.5 only Triprolidine is non-charged.

If retention has not increased enough, then the next step would be to try the Extend-C18. All of the basic compounds shown in Figure 1, except Triprolidine, have pKa values of 9.0 - 9.2. Therefore they must be analyzed at pH 10 or higher before they are non-charged. The Extend-C18 can effectively be used at this high pH to improve the retention of basic compounds.

FIGURE 1

Effect of pH on Retention — Antihistamines

Column: Eclipse XDB-C8 (Agilent Part No. 993967-902)
Mobile Phase: 75% 25 mM Phosphate buffer
25% Acetonitrile
Temperature: 40°C

 

Back to Top

Choosing Bonded-Phases

Question: What column do you recommend to analyze raw materials?

Answer:There is not enough information provided in this question to give a specific recommendation. Column choice, and appropriate mobile phase, depends on many characteristics of the sample such as polarity, pKa of ionizable functional groups, solubility vs. pH, and molecular weight.


Question: I’m currently using a C18 column and my separation has a couple of peaks that elute early and a couple of peaks that elute late. What can I do to reduce the analysis time and maintain resolution of the early eluting peaks?

Answer: First, you could try a gradient elution method on the C18 column. But many people do not like to use gradients, so choosing a different bonded-phase may help. Short chain polar bonded-phases such as the SB-CN and SB-Phenyl are ideal for separations like this. The increased polarity of these phases reduces retention of the later eluting hydrophobic compounds while often maintaining the retention of earlier eluting hydrophilic compounds. Figure 2 shows this clearly. Using the same mobile phase conditions these three compounds are well resolved on the SB-CN in about 5 minutes. The same separation on the SB-C8 column takes nearly twice as long and provides incomplete resolution.

FIGURE 2

      SB-C8 (Agilent Part No. 866953-906)                 SB-CN (Agilent Part No. 866953-905)

Separation of Acid/Base Sample on StableBond

Columns: 4.6 x 75 mm, 3.5 µm
Mobile Phase: 80% 25 mM NaH2PO4, pH 3.0 20% MeOH
Flow Rate: 1.0 mL/min
Temperature: 35°C
Detection: UV 254 nm

Sample:
1. Barbital
2. Sulfamethoxazole
3. Caffeine

 

Question: When should I select Phenyl bonded-phases?

Answer: Both SB-Phenyl and Eclipse XDB-Phenyl bonded phases provide unique selectivity and are often a good choice for changing the selectivity for two closely eluting analytes. Phenyl bonded phases are less retentive than C8 bonded-phases so they offer a second option for reducing retention of late eluting hydrophobic compounds and minimizing analysis time (Figure 3).

FIGURE 3              SB-C18 (Agilent Part No. 863953-902)

                           SB-Phenyl (Agilent Part No. 863953-912)

Distinct Selectivity of ZORBAX SB-Phenyl and SB-C18 Bonded Phases

Columns: 4.6 x 150 mm, 3.5 µm
Mobile Phase: 84% MeOH 16% 0.1% TFA H2O, pH 2.0
Flow Rate: 1.0 mL/min
Temperature: 30°C
Detection: UV 310 nm

Sample:
1. Oxybenzone
2. Padimate-O
3. Ethylhexyl-p-methoxycinnamate
4. Ethylhexyl-salicylate

 

Back to Top

Temperature Control

Question: What temperature should I use for my separation?

Answer: Temperature control for separations is important for long-term retention reproducibility, one factor of method ruggedness. Controlling temperature at 35 - 40°C is normally sufficient for good method reproducibility and ruggedness. In addition, the use of elevated temperature can have other benefits. First, it reduces the system operating pressure by reducing the viscosity of the mobile phase.

Second, it will reduce analysis time, which can substantially increase productivity (Figure 4).

Third, temperature may change the selectivity of a separation. Not all compounds have the same response to temperature so the selectivity of a separation can change dramatically when temperature is increased or decreased. StableBond columns have high temperature limits — SB-C18 can be taken up to 90°C, at low pH, and SB-C8 , SB-Phenyl , SB-CN and SB-C3 can be taken up to 80°C, making it possible to optimize your separation without changing columns, particularly if you are analyzing ionizable compounds.

FIGURE 4

High Temperature Can be Used with StableBond Columns for High Throughput Analysis

Columns: 4.6 x 75 mm, 3.5 µm
ZORBAX Rapid Resolution SB-C18 (Agilent Part No. 866953-902)
Mobile Phase: 26% MeOH
74% H2O + 10 mM Hexane Sulfonate and 0.1% phospheric acid

Sample: Water Soluble Vitamins

Back to Top

Custom Columns

Question: There are a lot of different column configurations (dimensions) available, but I don’t see the one I’m looking for. Can I get a column made in the configuration I want?

Answer: Most likely. Agilent is continually adding to its HPLC column offerings, so check with your Agilent column distributor or in the US call Agilent Customer Assist at 800-227-9700 and ask for HPLC column support to find out if the column configuration you need is currently available. For locations outside the U.S., your authorized Agilent column distributor, listed on the Agilent website, can also help you find out the cost and delivery time. If the column is not available, a column can be packed in the configuration of your choice with available bonded-phases. Select an internal diameter and specify a column length, bonded-phase, and particle size using product number 899999-999. A special products quote (SPQ) will then be issued to track your order.

 

Back to Top

Rapid Resolution Columns

Question: I typically select 4.6 x 250 mm, 5 µm columns for my analytical work because I have complex samples, but I need to reduce my analysis time and increase my sample throughput. What can I do?

Answer: Rapid Resolution columns are ideal for your needs. A Rapid Resolution 4.6 x 150 mm, 3.5 µm column will reduce your analysis time by 40% while maintaining your resolution. Your gains are in reduced analysis time, whether you are doing isocratic or gradient separations, and substantial solvent savings (Figure 5). You may be able to choose even shorter Rapid Resolution column lengths and maintain the desired resolution (Rs) because Rs a N1/2. This means that decreasing column length, and therefore efficiency, will not decrease Rs by the same amount. It is likely that resolution will be maintained on even shorter column lengths and choosing 75 mm or even 50 mm column lengths can reduce analysis time even more.

FIGURE 5
    
    SB-C8 (Agilent Part No. 883975-906)            SB-C8 (Agilent Part No. 866953-906)

Rapid Resolution Chromatography

Column: 4.6 x 150 mm, 5 µm
ZORBAX StableBond SB-C8 (Agilent Part No. 883975-906)
Gradient Time: 30 min.
Flow Rate: 1.0 mL/min
Analysis Time: 24 min.
Column: 4.6 x 150 mm, 5 µm

Column: 4.6 x 75 mm, 5 µm
ZORBAX StableBond SB-C8 (Agilent Part No. 866953-906)
Gradient Time: 15 min.
Flow Rate: 1.0 mL/min
Analysis Time: 12 min.

Sample:
1. Tebuthiuron
2. Prometon
3. Prometryne
4. Atrazine
5. Bentazon
6. Propazine
7. Propanil
8. Metochlor

 

Question: But are these columns more likely to fail than 5 µm columns?

Answer: No, they are not. The Rapid Resolution columns are as rugged as the 5 µm, 250 mm columns. There are two types of accelerated column failure attributed to using smaller particles. First, smaller particle columns are thought to plug faster. This is not true when a standard 2 µm frit is used at the top of the column. Because of careful particle size control and the use of 3.5 µm particles, ZORBAX Rapid Resolution columns will not contain any particles as small as 2 µm. This means a standard frit can be used on the column and the Rapid Resolution columns will be no more prone to plug than a 5 µm particle size column.

Second, columns with smaller particles are thought to have shorter lifetimes because the column beds compress, leaving voids that cause peak broadening and tailing. It is true that 3.5 µm particle size columns will operate at slightly higher pressures than 5 µm columns, but ZORBAX particles can easily withstand these increases in pressure. ZORBAX particles are packed at 8000 psi and can easily withstand pressures up to 5000 psi in routine use. A Rapid Resolution 4.6 x 150, 3.5 µm column will typically be operated below 3000 psi, so the column bed will not compress when using ZORBAX Rapid Resolution columns. So both the rugged ZORBAX particles and the standard 2 µm column frit assure you long column lifetime when using Rapid Resolution columns.

FIGURE 6

 

 

 

Run Time: 1.8 min.
Equilibration Time: 1 min.
Total Analysis Time: 2.8 min.

 

 


Fast Gradient Analysis of Heart Drugs on a Very Short Column

Column: 4.6 x 50 mm, 3.5 µm
Rapid Resolution ZORBAX Eclipse XDB-C8 (Agilent Part No. 935967-906)
Mobile Phase: A: 25 mM Na2HPO4, pH 3.0
Gradient: 42 - 74% B in 1.8 min.
Flow Rate: 3.0 mL/min
Temperature: 35°C

Sample: Cardiac Drugs
1. Diltiazem
2. Dipyridamole
3. Nifedipine
4. Lidoflazine
5. Flunarizine

The very short Rapid Resolution columns can also be used at high flow rates to further reduce analysis time. Figure 6 shows a very rapid gradient analysis at 3 mL/min on a 4.6 x 50 mm Rapid Resolution Eclipse XDB-C8 column.

 

Question: What do you think about neutralizing extra-column volume by packing short columns with wide-bores (7mm)?

Answer: As long as you are willing to accept operating at a higher flow rate. If you operate at 1.0 mL/min on a 4.6 mm i.d. column, then operating at 2.3 mL/min would be the equivalent linear velocity when using a 7.0 mm i.d. column. Solvent waste may be an issue.

Back to Top

High-Throughput HPLC/Combinatorial Chemistry

Question: My work requires very high sample throughput. How fast can I do a gradient separation?

Answer: More and more people, especially those who analyze combinatorial chemistry samples, need to analyze a lot of unknown samples quickly. The best way to do this is on short, Rapid Resolution columns with rapid gradient times. The best gradient time is the one that resolves all of your analytes in the least amount of time. On very short columns — 30 and 50 mm lengths — a good starting point would be a 2 - 5 minute gradient. From there optimize your separation for organic range and gradient time. On these short columns you can easily increase the flow rate to further decrease analysis time to 30 seconds without exceeding the pressure limits of the columns. Column re-equilibration times are typically as short as the analysis times — 3 - 5 minutes for the 50 mm columns.


Question: Can I really effectively use these very short columns on my HPLC instrument?

Answer: Yes, you can. For columns of 3.0 mm i.d. and above, no instrument adjustments are necessary. For gradient separations with 2.1 mm and 1.0 mm i.d. columns the ideal HPLC is a high-pressure mixing instrument — like the Agilent 1100 HPLC with the binary pump, because it minimizes the gradient delay volume. Using a low volume mixer and the injector by-pass (or micro-injector) further minimize gradient delay volume. Narrow i.d. tubing and a low volume detector flow cell are preferred but not necessary. These changes are easy to make and allow you to effectively use columns as small as the 2.1 x 30 mm columns or even the 2.1 x 15 mm columns (Figure 7).

FIGURE 7

        SB-C18 (Agilent Part No. 873700-902)         SB-C18 (Agilent Part No. 875700-902)

Gradient High-Throughput Narrow-Bore HPLC . . in Less Than 60 Seconds

Column: 2.1 x 30 mm, 3.5 µm
Rapid Resolution StableBond SB-C18 (Agilent Part No. 873700-902)
Gradient: 5 - 25% B in 1.0 min.
Flow Rate: 1.0 mL/min
Analysis Time : 100 sec.

Column: 2.1 x 15 mm, 3.5 µm
Rapid Resolution StableBond SB-C18 (Agilent Part No. 875700-902)
Gradient: 10 - 60% B in 30 sec.
Flow Rate: 2.0 mL/min
Analysis Time : 47 sec.
Mobile Phase: A: Water with 0.2% formic acid
B: Methanol with 0.2% formic acid
Temperature: 35°C

Sample: Organic Acids
1. Gallic
2. Protocatechuric
3. Hydrocaffeic
4. Gentisic
5. Vanillic
6. Syringic
7. Sinapinic
8. Salicyclic

 

Back to Top

LC/MS

Question: Very short narrow-bore columns seem ideal for LC/MS. What bonded phase should I use?

Answer: LC/MS requires volatile mobile phases and the most suitable buffers are acetate and formate. The buffer range for acetate is from pH 3.8 - 5.8 so the best bonded-phases to start with would be the Eclipse XDB C18 and C8. Formate is also widely used and has a buffer range from pH 2.8 - 4.8. When using this buffer you can select StableBond columns then Eclipse XDB columns and follow the same method development scheme as discussed in the method development section. For high pH, using ammonium hydroxide at pH 10.5, select Extend-C18.

Back to Top

Column Cleaning

Question: What solvents do you recommend to clean the columns? I use THF

Answer: I generally recommend using the organic solvent acetonitrile to clean reversed-phase, C8 or C18, silica-based HPLC columns. THF is a strong reversed-phase organic solvent and that may be difficult to remove from hydrophobic bonded-phases, causing chromatographic variability. For most situations, acetonitrile is a strong enough solvent to remove sample and mobile phase components that have accumulated on your column during routine isocratic use.

Before cleaning the column with organic solvent however, be sure that you have removed any buffer salts that may be in your mobile phase and column to avoid precipitation of these salts in the column. To do that, prepare your mobile phase without buffer salts and flush your column with 15-20 column volumes of this mixture. That is, if your mobile phase is fifty-percent acetonitrile and fifty-percent 50 mM phosphate buffer, then the appropriate first step would be to remove the buffer salts with 50/50 mixture of acetonitrile and water. Then I would suggest flushing the column with pure acetonitrile thoroughly for 30-40 column volumes. If you believe there may particulates at the top of the column, e.g., pressure may be higher than normal, you should consider disconnecting the column from the detector and reversing the direction of flow before the cleaning step. Do no reconnect the column to the detector, but allow the cleaning solvent, in this case, acetonitrile, to collect in a beaker.

Since acetonitrile is commonly recommended for long-term column storage, the column can be easily stored after cleaning using the procedure described above. When ready to re-use this stored solvent, remember to flush the column with your mobile phase composition without buffer salts before introducing the buffered mobile phase, again to reduce the possibility of salt precipitation.

Question: What about the use of detergents for cleaning?

Answer: Detergents are not recommended for cleaning reversed-phase (RP) HPLC columns. Ionic detergents are typically long-chain carbon compounds having an ionizable group at one end and have been used as ion-pairing agents. Because these long carbon chains partition well into the bonded phase of reversed-phase columns (i.e. the bonded-phase of the C18 reversed-phase column strongly retains the long carbon chain of a detergent), the removal of detergents can be difficult, if not impossible.


Question: Some manufacturers’ indicate (newer) columns can be reversed to remove blockage from the front of the column. Is this generally recommended or does it depend on the column?

Answer: HPLC columns are more efficient and packed better than years ago, therefore this is generally recommended for most reversed-phase and normal phase silica based columns. This procedure is designed to remove particles from the column frit when high pressure occurs at the column inlet, but it will not work all the time. Because it does not require opening the column it is worth trying. Make sure when you do this that you disconnect the column from the detector and make sure the particles that plugged the column are not coming from the HPLC system or you may just plug the frit at the back end of the column and this may not be replaceable. Columns can also be turned in the reverse direction for washing/cleaning with stronger solvents to remove adsorbed material. This has the benefit of not exposing the rest of the column to the contaminants. When this is done the column should also not be attached to the detector.


Question: What's the best procedure for cleaning out ion pair reagent materials from a column?

Answer: Some detergents are used as ion-pair (IP) reagents, e.g. SDS with a carbon chain length of twelve. However, most commonly used ion-pair reagents have shorter carbon chains, e.g. hexane sulfonate has carbon chain length of six. IP reagents with longer carbon chains are more difficult to remove from RP-HPLC columns. Studies have shown that ion-pair reagents and some detergents are best removed with long washes with a 50/50: v/v, methanol/water mobile phase system. If you can regain the retention, selectivity and efficiency (resolution) of a known separation (e.g., QC sample) after washing, then you might have successfully removed the ion-pair reagent. I would not recommend using this column for developing a new method as exposure to ion-pair reagents may change the retention characteristics of a column permanently. If scouting runs for a new method are conducted on a column exposed to ion-pair reagents, a new column should be purchased as soon as possible to verify that the selectivity is not different. In my opinion, columns exposed to IP reagents should be dedicated to the ion-pair method.


Question: How do you clean C4 column?

Answer: Much as you would any RP-phase column and I've included some general instructions provided that you are using standard mobile phase conditions to separate small molecules. If you are separating peptides and proteins or if your sample is dissolved in plasma, then the guidance is different and I recommend that you call 800-227-9770, press option 1 and ask for HPLC column technical support for more information (within the US, other areas contact your local Agilent sales office.


Question: What is the purpose of the H3PO4 wash of the column?

Answer: A phosphoric acid wash has been shown to be effective at reducing tailing caused by the sample complexing with metals in the HPLC system. Typically a 1% phosphoric acid wash of the system and column is suggested to eliminate this tailing and it works. It is perfectly reasonable to use these recommended wash conditions with Agilent ZORBAX StableBond reversed-phase HPLC products. The StableBond HPLC column is particularly stable at low pH - the SB-C18 column is stable at a pH of 0.8 and 90°C. How can you tell if your peak tailing is caused by metal complexation? Look to see if a lone pair of electrons on either a N or O atom can chelate with the metal to form a 5 or 6-membered ring. Metal complexation is a commonly overlooked cause of peak tailing and metals are presents in every HPLC system.


Question: I'm working with environmental samples that may contain sulfur. I suspect that a few samples insufficiently cleaned up. Column performance has declined significantly. Is there any way to restore performance?

Answer: I assume you are using a reversed-phase column. Please review the recommended HPLC column cleaning procedure . I do suggest that if column performance declined quickly that you consider using a guard column and possibly consider including a preliminary clean-up step before injection.


Question: How can one properly wash GPC columns (safely)?

Answer: Wash the column in the reverse direction, not attached to the detector and at half the recommended flow rate (keep the pressure below the recommended maximum). First choose a solvent that will dissolve what you believe has contaminated the column. Most GPC columns are PS-DVB and you need to check the solvent compatibility before using a solvent. Many wash solvents are a higher viscosity than the typical eluting solvents so a lower flow rate with careful attention to pressure is needed.

Anionic samples can adsorb onto PS-DVB and if these have contaminated your GPC column a wash solution with a salt is recommended. Check to see what types of salts are recommended for the column. In some cases the polarity of the material adsorbed may require washing with organic solvents modified with acid (formic or acetic) or base (triethanolamine) (check the pH range) or some water may be compatible with an appropriate organic solvent. If more hydrophobic material were retained, elevated temperature along with an appropriate organic solvent would be recommended. Once again you need to check the maximum temperature range allowed for your column.

If you wash carefully the column should not degrade from the solvent switching.

Back to Top

 

Peak Fronting

Question: 3. When do you see peak fronting?

Answer: Peak fronting can occur under a variety of conditions. The most well known is peak fronting that occurs due to column overload. In this case you will typically see the peak retention time shift slightly shorter. This is also the simplest to evaluate, because you can just inject less sample and evaluate the peak shape.

But there are at least four other causes of peak fronting. These are column channeling, ionic interactions between the analytes and the silica, poor solubility of the analyte in the mobile phase, and wettability problems between the mobile phase and bonded-phase. A problem with column channeling should result in peak fronting on all peaks or at least the largest peak in the chromatogram if the others are very small. If this occurs you need to replace the column. To evaluate this problem try a new column.

Increasing the buffer ionic strength or changing the pH of the mobile phase can often improve ionic interactions, which cause fronting. Increasing the ionic strength can reduce interactions between the silica and ionic analytes and changing the pH can have the same effect.

Solubility needs to be assessed by trying to improve the solubility of the analyte and evaluating the resulting chromatogram. For instance, you can increase the time you sonicate your sample and re-inject or dissolve it in a solvent where the sample has good solubility and then dilute and inject in the mobile phase. Also any sample you suspect has a solubility problem should be filtered. You can also mix your sample and mobile phase off-line to see if it is visibly soluble in the mobile phase.

Wettability refers to the ability of the mobile phase to fully penetrate the bonded-phase so that analytes interact with all of the bonded-phase. In cases where there is a highly aqueous mobile phase with a C18 column complete wettability may not be achieved. The bonded-phase can fold over on itself. The results may be loss of retention and distortions in peak shape. If you can, increase the amount of organic in the mobile phase and re-evaluate peak shape. If not you may need to consider selecting a column designed for use in very high aqueous mobile phases.

These are the typical causes of peak fronting and some solutions. Please note that some of these problems may also result in peak tailing, though the solutions would still be the same.

Back to Top

 

Wavy Baseline

Question: What might cause a rhythmic,wavy baseline?

Answer:This is rarely a column problem. The most common causes of this problem are related to the pumping system.

a) If you have an isocratic pump, your pump seals may be worn. Typically your pump will have two pistons and seals. One may be more worn than the other causing flow and pressure variations in a very rhythmic pattern, which will also be seen by the detector. You can confirm a pump seal problem by varying the flow rate. The frequency (in time) of the baseline fluctuations will increase/decrease proportionally to an increase/decrease in the flow rate if the pump seals are worn. The solution is to replace the pump seals.

b) If you have a multiple pump system (binary, ternary, or quaternary), the problem may also be pump seals but could also be a function of insufficient mixing or of a malfunctioning proportioning valve. The former can be solved by adding a mixing column (ask your manufacturer for a recommendation) and the latter by a service call.

Another possibility is an aging lamp although these signal fluctuations are not as perfectly periodic as a pump problem. An aging lamp can be distinguished from a pump problem with the flow rate experiment described above. If the frequency of signal fluctuations (in time) don't change when the flow rate is varied, the problem may be the lamp.

Back to Top

Use of Guard Column

Question: What does the filter do that the guard-column does not?

Answer: Nothing really - both a filter and a guard can capture small particulates that are either coming from your sample, mobile phase or are from wear-and-tear of instrument seals and gaskets. A guard, on the other hand, traps sample material that is strongly retained on your column, material that would not elute from the column under the experimental conditions defined by the method. These strongly retained materials can build over time at the head of the column and generally lead to high pressure and/or poorly shaped peaks. In some cases, these retained materials cannot be removed even after extensive washings with strong solvent. By using a guard, your column is protected from material that is permanently retained.

Back to Top

Measurement of Column Performance

Question: How do you evaluate for column voids?

Answer: First check your method and find out if you have operated at a pH higher than is recommended for the column. This is probably the major cause of column voids because silica can dissolve causing the column void. The sample injection solvent as well as the mobile phase needs to be considered here. If there is a column void, you will see a change in peak shape - tailing, broadening, or split peaks - on every peak in the chromatogram. A void will not usually cause a change in only one peak in the chromatogram. It also does not typically cause a change in analyte retention. You can also turn the column around and if there is a column void then the peak shape should be poor in the reverse direction as well. The only definitive way to check for a column void is to open the column and this should be done only as a last resort in identifying the problem with the column.


Question: What is meant by peak plate?

Answer: Plates" allow you to compare the efficiency of different columns and its measure is determined by the width of a peak in a chromatogram relative to its retention. In general, for the same retention , the more narrow a peak, the more efficient the column and the more efficient the column has more "plates". HPLC instruments often determine column plates or efficiency by measuring the peak width , typically at half-height, using the following equation:

N = efficiency = plates = 5.54(tR/w1/2)2

w1/2 = peak width at half-height in min.

tR = retention time in min.

The efficiency of a column is reported by the manufacturer of your column and is generally provided in a column report. For a more detailed discussion on how plates and column efficiency relate to resolution and your analyses refer to the text Practical HPLC Method Development written by Lloyd R. Snyder, Joseph J. Kirkland and Joseph L. Glajch, Wiley-Interscience , Second Edition, 1997.

Question: Do you think that using toluene as a standard for measuring GPC column efficiency is appropriate?

Answer: Yes, I do. Conditions for this evaluation are chosen so that the small molecule toluene does not interact with the packing, but diffuses through all the pores of the GPC column packing. By conducting the experiment in this way, the width of the eluting peak is a function of the particle size of the packing and how well the column bed is packed. That's what you want to know when you are measuring efficiency. This sample is commonly used and recommended by most manufacturers to test efficiency of these columns. Alternate standards may be used to determine other characteristic of a GPC column, e.g., exclusion volume.


Question: Do you think that using high molecular weight standards would be more appropriate for measuring GPC column efficiency?

Answer: For the reasons mentioned above, I do not.


Question: Is there any kind of "generic" column test to track the performance of a column? How often should it be completed?

Answer: In my opinion, the best way to evaluate a columns performance is to use the QC test that should be shipped with each HPLC column provided to you. By comparing the efficiency, retention and peak shape of the peaks in the sample and very importantly the pressure under these experimental conditions, you will be able to tell if your column has changed over time. These tests can also tell you how your column has changed, as these are the same tools we use to diagnose a column problem. Significant changes in retention, can suggest loss in bonded phase and significant changes in peak width/efficiency and pressure can suggest column contamination. Dramatic losses in peak width and efficiency suggest a column void.

Most manufacturers use similar solutes and mobile phase test conditions to QC their columns. However, different bonded phase require modification of the amount of organic modifier to achieve reasonable retention.

What's important is that that you QC-test a column on your system before you use it for a particular project. Then you know how that column performs on your system, before you start injecting samples and before you start having problems. If problems develop, use this QC-test to verify that the HPLC system and the column are performing well. Before you retire a column from a project, clean and QC-test the column. If for any reason, you start a project with a used column, at a minimum, QC-test the column before injecting samples.

Back to Top

Column Stability

Question: I'm having a problem with resolution and selectivity for a phenyl column using different lots, how can I resolve this problem?

Answer: First, evaluate the methods used on each column. Are their any differences in the column histories or column equilibration procedures? Differences in either of these may alter the behavior of analytes on a column. Recently someone had a problem with a change in resolution and selectivity on two phenyl columns, which was resolved by flushing the column once again with 100% methanol. This may make a good starting point for you. In the case I just described we had a third column lot to compare it to so that when we flushed the column and retested the sample in its mobile phase we achieved the expected results and knew that this flush should be added to the equilibration procedures. Our typical recommendation would be to flush the column with 100% acetonitrile or methanol, which ever is used as the organic modifier in the mobile phase. In the case I described, only the methanol changed the column, possibly due to interactions with the phenyl rings in the bonded-phase.

You should also make up fresh mobile phase and test it in case the problem occurred with the mobile phase and make sure the instrument is operating as expected. These steps will eliminate some of the potential sources of column-to-column variations.

The next step is to review your method ruggedness. Are any of your analytes sensitive to pH or ionic strength? If slight changes in pH or ionic strength affect retention then you may need to modify your method and select conditions where changes in these parameters do not affect resolution and selectivity as much. Acidic and basic analytes can be very sensitive to pH and ionic strength changes around the pKa of the analyte.

Finally ask the column manufacturer for assistance. They may be able to supply additional columns from older lots that have worked or from newer lots while the method is evaluated. The manufacturer may also be able to make some suggestions based on your method parameters.

Question: What types of column are dissolved above pH 7?

Answer: Many commercially available silica-based HPLC packings dissolve at pH 7, for the simple reason that silica is soluble at pH 7. Therefore, Agilent has developed technology to significantly reduce silica dissolution and extend the usable pH range of silica-based columns. The ZORBAX Extend-C18 can be effectively used above pH 8 and even up to pH 11.5. Bidentate-bonding is key to this enhanced stability while offering the high efficiency that only silica can provide.

Back to Top

Peak Co-elution

Question: What do you recommend doing when a impurity peak starts to co-elute and actually starts to make itself present in your main peak as a hump to the peak? I have tried backflushing and washing the column.

Answer: I would like to see the chromatograms you are comparing and ask you more questions about this method. It is not clear from your question, if you are you comparing chromatograms obtained from the same column or from different columns.

How much of a change in resolution are you seeing? Has resolution decreased over time or has the separation been disappointing right out of the box? Do you have any information telling you what factors may cause these peaks to converge? Are they sensitive to small changes in temperature, pH or organic composition? Is one compound much larger than the other? These factors and more can be causing the two peaks to merge, but it is not possible for me to determine the cause from the question posed.

Back to Top