Frequently Asked Questions (FAQs): HPLC and UHPLC
- How much sample can I inject on my LC column and still avoid band broadening?
- How much should I change my injection volume if I change the size of my column?
- What happens if my sample solvent is stronger than my mobile phase?
- Can I get a sharper peak by injecting my sample in a weaker injection solvent (such as 100% water for reverse phase)?
- How do I address an increase in backpressure and other signs of blockage or contamination?
- How do I determine total column volume or void volume for LC?
- Why am I seeing bleed from my Biphenyl column on my UV but not on my mass spec?
- What should I use to analyze explosives (as per EPA Method 8330B) by HPLC?
Raptor LC Columns
- How does the new Raptor Biphenyl column differ from other Biphenyl columns from Restek?
- How exactly does an SPP silica-based column produce higher efficiency and resolution?
- Is special conditioning needed for the Raptor Biphenyl column prior to its first use, or if it has been sitting idle?
- How much equilibration time is required in between gradient runs on a Raptor Biphenyl column?
- What mobile phase solvents are compatible with SPP or Raptor columns?
- Can I pump solvent through the Raptor Biphenyl column backwards to clean it?
- How much can I inject onto a Raptor column?
- How do Raptor ARC-18 columns differ from ordinary C18s?
- How well does the Raptor ARC-18 column work for acids and bases?
- Can the Raptor ARC-18 column be used with 100% aqueous mobile phases?
- How do we know that the Raptor columns are rugged?
- What type of LC guard column system do I need?
- Is there a Restek guard cartridge to go with my UHPLC column?
- Which guard cartridge should I use?
- Which cap frit should I use with my Trident guard system?
- Can I use a Restek guard cartridge with my column from another vendor?
- What size threads are on the end fittings of my HPLC column?
- What does “10-32” mean?
- Are there any other size threads used for LC, besides the 10-32?
- How are stainless-steel fittings for HPLC different from polymer-based fittings?
- Which fittings can be used for UHPLC?
- How do I tighten my fittings?
- Which fittings do I use for which tubing?
- How do I know what the internal diameter of my LC tubing is?
- How do I avoid problems with HILIC methods?
- Is my Raptor HILIC-Si column ready to use right out of the box?
- How long should I equilibrate my Raptor HILIC-Si column between injections?
- What injection solvent should I use for HILIC separations?
- What kind of pH effects do I have to be aware of with HILIC separations?
- Can I use buffers for HILIC separations? What kind and what concentration?
What are PFAS Delay Columns?
- What are PFAS?
- What are GenX and PFBS?
- What kind of labs do PFAS testing?
- What are typical PFAS analysis levels?
- What’s different about PFAS analysis compared to other LC-MS/MS analyses?
- How do I know if my LC system has PFAS contamination?
- Why are system-related PFAS important to isolate from my samples?
- How does the PFAS delay column help?
If your question doesn’t appear on the list, please contact Restek’s expert chemists, for help troubleshooting HPLC and UHPLC systems and analyses. (Remember to include your company name and complete mailing address.)
Injection volumes, as well as optimal flow rates, are limited by the size of the column. Ideally, if your sample is in the same solvent as your mobile phase, approximate volumes should range as follows:
|Column ID||Volume (µL)|
|2.1 mm (30 -100 mm length)||1-3|
|3.0-3.2 mm (50-150 mm length)||2-12|
|4.6 mm (50-250 mm length)||8-40|
Optimal injection volumes are directly related to the cylinder volume of your column and are, therefore, dependent on the cross sectional area (A=π r2) and length (L) of your column. Since that is the case, you can estimate any adjustment from an existing method for injection volume.
If you are converting to a different size ID (with packing material and length remaining the same), just multiply your current volume by the ratio of the radii squared to determine the correct volume for your new method. For example, if you are currently injecting 20 µL on a 150 x 4.6 mm column and then switch to a 150 x 3.0 mm column, you could estimate the adjusted volume by multiplying 20 x (1.52)/(2.32). Your new volume should be about 8.5 µL.
We do not recommend injecting in a stronger solvent because it usually results in peak distortion, broadening, poor sensitivity, and shortening of retention times. This happens because some analytes will tend to travel too quickly through the column, instead of eluting in a symmetrical band. If you absolutely must do this, keep the volume as small as possible and make sure the solvents are miscible.
4. Can I get a sharper peak by injecting my sample in a weaker injection solvent (such as 100% water for reverse phase)?
In this scenario, the sample is initially concentrated onto the head of the column and moves through the column in a tight band. This technique, referred to as “on-column compression” or “point-injected”, is sometimes used to minimize band broadening for a larger volume sample injection (see FAQ #8 above). Keep in mind that your sample components must be soluble in mobile phase as well, in order for this to work.
Please follow the recommendations in our blog "Building up Pressure on HPLC?" This will help you identify or confirm the source of pressure. If the column itself is found to be the source of difficulty, please follow the column regeneration steps given on our LC Column Cleaning Recommendations page. To help avoid future occurrences, Restek highly recommends the use of guard columns to protect your reversed phase and normal phase analytical LC columns from both frit blockage and column contamination.
The term "column volume" usually refers to the void volume, which represents the volume of mobile phase that is between the silica particles. This area is referred to as the interstitial space. You can estimate void volume by multiplying the total column volume (pi x radius2 x length) by a factor that estimates the typical packing efficiency for a particular column type. For fully porous columns, the equation for void volume (in mL) is V = (0.68) pi r2 L, where V = column volume in mL, r = column radius in cm, and L = column length in cm. For superficially porous columns, such as our Raptor columns, the factor is different and the equation is V = (0.50) pi r2 L.
Void volume is more commonly estimated experimentally by injecting a standard containing an analyte that is known to have no, or negligible, retention on that particular column phase. A good example of this for reversed-phase HPLC is uracil. One should be aware that this estimation is also subject to extra column dead volume for the specific instrument that is being used, so it may vary slightly.
A small amount of phase bleed is inherent for all phases, including phenyl phases, and is somewhat dependent on the size and dimensions of the column. This bleed is usually negligible and does not affect retention times, but may be visible, particularly by UV detection. It can often be reduced after conditioning. Bleed may also be minimized by using an isocratic elution, a shallower gradient, and/or incorporating a gradient flush between runs.
While no one LC column can provide baseline separation for all of these analytes combined, the Raptor Biphenyl and Raptor ARC-18 columns from Restek are an outstanding choice for primary and confirmation analysis. Fully porous HPLC particles, namely the Ultra C8 and Ultra Aromax columns, are also an option. Keep in mind that a variety of column phases may provide partial solutions for this method, but Restek has found these pairs to give optimal results.
The Raptor Biphenyl is made with superficially porous particles of silica, often abbreviated as SPP. The SPP silica has a solid core with a porous outer layer and is made with a diameter that is typical of HPLC particles, in this case, 2.7 µm. SPP columns exhibit higher efficiencies and resolution, often approaching that of smaller (such as 1.9 µm) UHPLC particles that are based on fully porous silica. There are also a couple of additional benefits to using solid core particles. SPP silica particles can be operated at high flow rates without significantly impacting column resolution, and they do not require higher pressure system components like UHPLC columns do. (It’s nice to achieve a separation that looks like UHPLC on a regular HPLC system!)
Greater efficiency is accomplished primarily by shorter transfer time for solutes traveling into and out of the particles and mobile phase, but is also enhanced by more uniform particle size, which results in more consistent packing. From a more theoretical viewpoint, the Van Deemter equation is the best approach to explain this phenomenon. Better efficiency is represented here as a shorter plate height (H).
The Van Deemter equation describes the relationship between column flow rate and peak efficiency, referred to as band broadening.
The solid core of SPP silica results in lower values for the A and B terms (eddy diffusion and longitudinal diffusion) and it results in better mass transfer of solutes, translating to lower values for the mass transfer component, the C term. Therefore, a shorter theoretical plate height H and higher efficiency can be achieved at high flow rates. This is also the case for UHPLC particles, but with SPP particles of normal analytical size, the column pressure is not dramatically higher, as is the case for UHPLC. SPP particles are usually also more uniform in shape and result in more consistent packing, which improves overall efficiency as well. All of these factors produce the significant reduction in band broadening that we commonly associate with SPP silica columns.
11. Is special conditioning needed for the Raptor Biphenyl column prior to its first use, or if it has been sitting idle?
For the most part, the Raptor Biphenyl column behaves just like any other reversed-phase column. However, in certain circumstances, longer equilibration times may be needed. Switching between organic solvents, such as acetonitrile and methanol, may require a 15-20 minute flush in high organic mobile phase.
Whether you are using fully porous silica or SPP silica, some equilibration time is needed between runs if you are using a gradient and the amount of time is similar for both types of columns. Usually, the equivalent of 7 column (void) volumes is sufficient unless you are using an ion-pairing technique.
Any solvent that is commonly used for reversed-phase LC will work fine, including but not limited to water, methanol, and acetonitrile.
Similar to UHPLC columns, it is not recommended to reverse the flow for these columns. However, you can still pump through a series of solvents, as long as they are miscible.
Injection volume depends on a number of factors including column dimensions, sample solvents, and analysis requirements. As is always a good practice with chromatography, try to inject as little as possible and in the same or weaker solvent than your mobile phase.
The significant difference is the ruggedness of the bonded phase. With the ARC-18, any residual silanol groups are shielded and made inert through steric protection. The result is a wider operating pH range of 1.0–8.0. The ARC-18 is particularly useful between a pH of 1.0 and 3.0, where other C18 column phases may begin to degrade under these harsh conditions. Like the Raptor Biphenyl column, the stationary phase is bonded to superficially porous silica particles (SPP). Please see our earlier FAQ regarding the advantages of SPP.
The ARC-18 provides added retention for charged bases and, in many cases, is preferred over a conventional end-capped C18. For neutral acids, it works well and is preferred over end-capped C18 phases, particularly at pH < 3. The ARC-18 also works for neutral bases and charged acids, but provides more advantages and performs best at the lower pH ranges.
No. We recommend using the Raptor ARC-18 column with at least 5% organic in the mobile phase. For applications requiring higher aqueous content, we suggest the Ultra Aqueous C18 or Pinnacle DB Aqueous C18 columns.
We use frits that are less prone to clogging from sample matrices, and the column packing is less likely to be damaged by higher pressures that might develop. Added protection of a guard column is also available and recommended to extend the life of the column further. Visit www.restek.com/raptor to see these columns in use and put to the test against the competition, but as the ultimate proof, we encourage you to try them for yourself.
Guard column systems consist of holders and cartridges, and the correct choice is determined by the technique and analytical column that you are using. Restek offers three guard column systems—EXP, Roc, and Trident (direct and in-line)—that are each designed to match one of our LC column families and are intended for a specific technique.
- Holder is compatible with all HPLC through UHPLC instruments and cartridges (rated to 20,000 psi [1400+ bar]).
- Designed for use with Raptor and Force LC columns.
- Compatible with traditional HPLC instrumentation.
- Designed for use with Roc HPLC columns.
- Compatible with traditional HPLC instrumentation.
- Designed for use with legacy HPLC columns (Ultra, Pinnacle II, Pinnacle DB, Allure, Viva).
- Available in three levels of protection.
- Compatible with traditional HPLC instrumentation.
- Designed for use with legacy HPLC columns (Ultra, Pinnacle II, Pinnacle DB, Allure, Viva).
- Requires a coupler to attach to the analytical column.
For more details, see our LC guard column selection guide.
We have guard cartridges and holders available for Raptor 1.8 µm UHPLC columns. As an alternative for UHPLC, you may find an in-line filter to be beneficial.
UltraShield UHPLC Precolumn Filter
Similar to our Trident direct guard system, the UltraShield filter screws directly into the column end fitting at the inlet. The UltraLine filter requires a tubing connection on either side; if it is installed between the injector and the column, the EXP coupler is recommended.
The other difference is in the frits, or filters as you may call them. The UltraShield unit has a titanium filter that is not replaceable. Therefore, the whole filter assembly is used as a disposable item. On the other hand, the UltraLine filter has a replaceable stainless steel filter, and more of them can be ordered as needed.
Filters in both units have a porosity of 0.5 µm and both holder assemblies are leak tight up to a pressure of 15,000 psi. Therefore, they are functionally equivalent.
The UltraShield filter is compatible with any family and phase of Restek UHPLC columns, whereas, the Raptor UHPLC guards are compatible with matching phases of 1.8 µm Force and 1.9 µm Pinnacle DB UHPLC columns.
For more details, see our LC guard column selection guide.
Once you know which guard column system you need, the next step is choosing an LC guard cartridge with the packing and dimensions that best match your analytical column. In order to obtain maximum protection without compromising selectivity or efficiency, you should select an LC guard cartridge that contains the same packing material—both stationary phase and silica—as your analytical column. For example, the best LC guard column to use with a Raptor Biphenyl 2.7 µm analytical column is a Raptor Biphenyl 2.7 µm guard cartridge.
In addition to choosing a guard cartridge packing that matches your analytical column, you need to select an LC guard cartridge with the correct inner diameter (ID). A good general rule is that the LC guard column cartridge ID should be the same as, or one size smaller than, the ID of the analytical column. So, a 4.0 mm ID guard cartridge should be used with a 4.6 or 4.0 mm ID analytical column, while a 2.1 mm ID guard cartridge is recommended for 3.2, 3.0, and 2.1 mm ID analytical columns.
See our LC guard column selection guide for details on which guards are compatible with which specific analytical columns.
The cap frit on the Level 1 and Level 3 Trident filter end fitting will catch larger particulates and maximize the life of the guard cartridge. Porosity is chosen based on the nature of the sample matrix and, in some cases, the nature of the mobile phase mixture. A frit with a porosity of 2 µm can be used in most cases and is the standard size when ordered. A 0.5 µm frit offers more protection for samples that are not prefiltered or that lack cleanup during sample preparation. Smaller porosity may also be more beneficial when using a 3 µm particle size analytical column or when using a mobile phase that contains higher levels of buffer salts.
You can use Ultra or Roc 5 µm guard cartridges for fully porous particle (FPP) columns with a particle size of 3 µm or greater, provided the packing material is very similar. For “core-shell” or superficially porous particles (SPP), a Raptor 2.7 µm or 5 µm guard can be used for similar phases and particle sizes. For FPP or SPP particles less than 2 µm, a Raptor UHPLC guard column can be used for similar phases. Make sure to use the appropriate Roc, Trident, or Raptor guard holder. Some of the column stationary phases/packing materials that are similar to ours are listed in this chart for easy reference.
In some cases, the stationary phase may be so unique that you might need to purchase guard cartridge supplies from the same vendor in order to obtain the best results. Feel free to call us if you have questions about this.
We recommend using only Restek guard cartridge holders with Restek guard cartridges in order to ensure a proper fit.
Fittings on all HPLC and UHPLC columns have 10-32 threads. However, you will find that fittings and columns from Waters, Rheodyne, SSI, Gasukura have different seating depths, as shown below, even though the threads are still 10-32. Restek’s universal 10-32 PEEK column connectors and new EXP fittings will accommodate any of these configurations.
|Fitting styles differ among various manufacturers, but all HPLC and UHPLC column fittings have 10- 32 threads.|
This is a naming convention adopted by the American National Standards Institute (ANSI). For HPLC fittings, the first number indicates the diameter of the threaded portion of the nut (not the tubing that goes inside). When the diameter is less than 1/4”, as it is in this case, a gauge value is used (gauge 10). The number following the hyphen refers to the number of threads per inch, thus indicating the “pitch” of the threads. An external screw thread size and tolerances chart is available here: http://www.engineersedge.com/screw_threads_chart.htm
10-gauge fittings are approximately 3/16” in diameter.
Size 10-32 fittings have a thread count of 32 threads per inch. Note that the first thread is used to align the fitting at zero and is not included in the thread count.
The 10-32 threads are probably the only ones you will use with HPLC. However, you may occasionally see ¼ x 28 or M6 (abbreviated form for M6 x 1) fittings used in laboratory settings or for lower pressure connections. The ¼ x 28 size is based on English measurement units like the 10-32. ¼ x 28 threads are further apart and slightly larger in diameter, and usually they do not withstand higher pressures as well. The M6 fittings are metric threads that are actually 6 mm in diameter and 1 mm apart.
The most noticeable difference is that stainless-steel fittings are “swaged” onto the tubing. They are tightened with a wrench and once the nut is firmly secured, the ferrule is permanently attached to the tubing. Although this produces a good seal that withstands high pressures, the ferrules cannot be removed and reused. In order to replace a ferrule or reinstall on new tubing, the entire fitting and small section of tubing must be cut, sacrificing the ferrule, as well as a small length of the tubing. (Typically, the nut can still be reused.)
Preseated stainless-steel fittings installed on a Restek mixing capillary (cat.# 26533)
The good news is that polymer-based fittings, such as Restek’s Universal 10-32 connectors and hybrid-ferrule EXP fittings, do not require permanent swaging of the tubing; thus, all of the parts can be reused. The EXP fittings are especially useful because they can withstand higher pressures, while the ferrules are still reusable to some degree. The number of times an EXP fitting can be reused is directly related to the amount of torque that is applied. The ferrule can be reused many times when the fitting is finger-tightened, while it can be reused just a few times if the fitting is wrench tightened for UHPLC. The limitation is due to a certain amount of deformity for the ferrule, not due to swaging.
PEEK finger-tight fittings (cat.# 27710)
You can either use the stainless steel fittings that are like the ones that come with your UHPLC system or you can use EXP fittings. The EXP fittings can be used up to 20,000 psi when tightened with a wrench. In any case, always make sure you are using a fitting with zero dead volume so that the high efficiency provided by UHPLC is not compromised by extra dead volume.
Our polymer-based universal connectors and PEEK union connectors only need to be hand-tightened, whereas any of the stainless steel fittings need to be wrench-tightened. The EXP fittings can be used either way. They are hand-tightened for use up to 8,700 psi or wrench-tightened for use up to 20,000 psi. Note that over-tightening causes galling and will destroy the threads. Fittings that need to be wrench-tightened generally require ¼-turn past hand tight to achieve a leak-tight seal. Unfortunately, there is no universal torque setting.
Convention tells us to match fitting material to tubing material: use PEEK fittings with PEEK tubing and use stainless steel fittings with stainless steel tubing. The main reason for that is simply system pressure. However, some stainless steel fittings can be used with PEEK tubing. For example, EXP fittings utilize a titanium/PEEK hybrid construction and can be used with either PEEK or stainless tubing. However, if inertness is a consideration, stainless steel fittings should be avoided and PEEK fittings are a better choice. Please note that PEEK is not recommended for use with some strong acid solutions, such as aqua regia, and it can be affected by halogenated solvents, as well as THF under certain conditions. Stainless steel is safe for most mobile phase reagents and solvents, but it is less inert, particularly toward mineral acids, bases, and oxidizing agents.
EXP fittings utilize a titanium/PEEK hybrid construction and can be used with either PEEK or stainless tubing.
Other types of tubing, such as PTFE and Tygon tubing, work well with PEEK or polyethylene (PE) fittings, unions, and adaptors. Many of these fittings are not screw-threaded but rather slip-on, because of their use in low pressure applications. A great example of this is mobile phase delivery in HPLC from bottle/reservoir to pump. This is a low-pressure delivery (HPLC and UHPLC systems hold high pressures from pumps to the end of the column), and PTFE tubing and PTFE or PE fittings are commonly used.
PEEK union connector (cat.# 27715)
Bluestem glass solvent filter
UHPLC systems are plumbed with extremely narrow-bore tubing that is indiscernible to the naked eye. For this reason, tubing lengths are color-coded based on internal diameter. Restek offers stainless steel capillary tubing with colored bands: red = 0.005”, yellow = 0.007”, blue = 0.010” and orange = 0.020”. Color coding varies by manufacturer; it is not universal. It’s also important to note that stainless steel capillary tubing should be cut to length by the manufacturer, not by the customer. It is impossible to achieve a clean, square cut without professional machining cutters. Stainless steel fittings with pre-swaged ferrules or EXP fittings with hybrid ferrules are used with stainless steel tubing to achieve leak-tight seals for UHPLC at up to 20,000 psi.
The internal diameter of LC grade stainless steel tubing is identified by the colored band on the pre-cut lengths.
HILIC methodology is a powerful tool, especially for analyzing polar compounds, but it can be challenging to implement and there are several important considerations that you need to be aware of in order to avoid common pitfalls. Our technical article introduces the HILIC technique, discusses frequent problem areas, and helps you successfully incorporate HILIC methods into your laboratory’s repertoire.
Prior to use, your Raptor HILIC-Si column must be properly conditioned with the mobile phase that will be used during analysis. For isocratic HILIC methods, at least 50 column volumes should be used and, for gradient HILIC methods, at least 10 blank injections running the complete time program should be performed. Conditioning is also required when you change the mobile phase composition or the concentration of any additives. Use the same number of column volumes or blank injections as you did when conditioning the column for the first time. Table I below gives the column volumes for Raptor HILIC-Si columns based on their dimensions.
Table I: Column Volumes (mL) Based Upon Raptor HILIC-Si Column Dimensions
|Column ID||Column Length|
In addition to the initial conditioning of the column with the mobile phase, it is critical that the column be re-equilibrated between injections. Because the separation mechanism in HILIC methods involves the adsorption of a water layer on the particle surface, it is very important to completely reestablish or "reset" this water layer in between injections to ensure retention time reproducibility. We recommend equilibration with a minimum of 10 column volumes starting when the gradient program returns to initial conditions, or in isocratic separations, at least 10 column volumes starting from the retention time of the last peak. The number of column volumes required for complete re-equilibration can be very analyte dependent, especially for isocratic separations, so make sure you investigate retention time reproducibility during HILIC method development.
The injection solvent should be as close of a match as possible to the initial mobile phase conditions, which are high in organic content for HILIC separations. By matching the injection solvent to the initial mobile phase conditions, you get better peak shape, increased retention, and higher sensitivity.
The effect of pH on analyte charge state varies based on each compound’s pKa, so pH effects must be evaluated carefully during method development. With HILIC methods, the high concentration of organic solvent in the mobile phase raises the pH, and the actual eluent pH can be 1–1.5 units higher than in the aqueous portion alone. The charge state of the column itself can also be affected. For example, in a Raptor HILIC-Si column, the bare silica has a pKa between 3.8 and 4.5, so the mobile phase pH changes the charge of the silica surface, making it neutral in very acidic conditions and ionized (negatively charged) as the pH begins to approach 3.8 and above. For this reason, if your analyte has one or more protonated amine or quaternary amine groups, it’s a good candidate for analysis on a Raptor HILIC-Si column.
Many HILIC separations use a mass spectrometer for the detector, so volatile buffers like ammonium formate and ammonium acetate are very common. However, the high organic content of the mobile phase can cause buffer salts to precipitate, which can lead to downtime for instrument maintenance. In addition, high buffer concentrations can impact chromatography by reducing analyte retention. To avoid these effects, method optimization is required and 10 mM is a good starting point for buffer concentration. Both the A and B mobile phases should be buffered equally in order to keep the ionic strength constant during a gradient for the most consistent MS detector response. Check with your MS vendor for the maximum buffer concentration they recommend for your ESI source.
PFAS are per- or polyfluorinated alkyl substances that are used as surface-active agents in the manufacture of a variety of products, such as firefighting foams, coating additives (e.g., nonstick pots and pans), textiles (waterproof clothing, stain resistant carpeting), and cleaning products. They are extremely resistant to breakdown in the environment and are found throughout the world in in soil, air, groundwater, municipal refuse, and landfill leachates. Two of the most common PFAS are perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS).
These are two compounds that have been made within the last 10 years to replace PFOA and PFOS. There are still health and environmental concerns about the safety of GenX (ammonium salt of hexafluoropropylene oxide dimer acid, or HFPO-DA, introduced in 2009) and PFBS (perfluorobutanesulfonic acid, released in 2003) and many environmental labs are beginning to test for them along with PFOA and PFOS.
PFAS analysis is done by contract environmental labs, state and local government labs, and municipal water treatment labs. Other labs that may do PFAS testing include universities and food and beverage labs testing their incoming water supply. PFAS testing has also recently expanded into air testing, so air quality labs will likely add PFAS to their analyte lists.
PFAS detection levels vary by region and across matrices such as drinking water, wastewater, seawater, soil, etc. Drinking water is the matrix of most concern in relation to public health and many environmental agencies are trying to set safe consumption levels. Most are in the form of health-based recommended levels with no regulations in place as of Feb 2019. Currently, in most countries, designated levels of concern are in the low ppt (ng/L) range for most PFAS compounds in drinking water. Some examples are given below.
- United States: health advisory level of 70 ppt in drinking water for PFOS and PFOA combined
- Sweden: action level of 90 ppt in drinking water for 11 PFAS compounds combined
- Italy (Veneto): threshold values in drinking water, PFOS ≤ 0.03 ppt, PFOA ≤ 0.5 ppt, other PFAS ≤ 0.5 ppt
- Germany: health-based precautionary value of 100 ppt in drinking water for PFOA, PFOS, and many other PFAS
LC-MS/MS analysis is a very sensitive, selective technique that is used in many different applications and is ideally suited for multi-analyte analysis. Due to their chemical/physical properties and long-lasting inertness, plastic fluoropolymers, such as polytetrafluoroethylene (PTFE), are used to make many LC-MS/MS components. You can find PTFE in LC pump seals, mobile phase transfer lines, the plastic linings used in degassers, etc. Unfortunately, these plastic parts can leach PFAS into the mobile phase, which creates system-related background PFAS contamination that interferes with trace-level PFAS analysis. Because PFAS detection levels in drinking water are in ppt range, even slight interference can bias quantitative analysis.
PFAS interference can also come from the solvents (even new bottles) used to make mobile phase due to the ubiquitous use of PFAS compounds. PFAS are found in air as well, so you can assume PFAS contaminants are literally everywhere at low concentration. Basically, system-related PFAS come from components in all stages of the LC-MS/MS workflow including the sample collection bottle, mobile phase cap/lining (made of PTFE), solvent inlet tubing (often FEP or PFA), degasser, LC pump parts, even the autosampler vial septum (many are PTFE-lined silicone).
The degree of interference that is observed will vary across LC instrument manufacturers and also depends on method parameters (e.g., equilibration time, solvent choice, target analytes, etc.) Note that the longer equilibration times produce more interference because PFAS have more time to leach out of the plastic parts.
The easiest way to identify PFAS contamination in your LC is to pump mobile phase through the column for 30 minutes to give system-related PFAS a chance to build up at the head of the column. Then, inject a solvent blank and when the gradient runs through the column any PFAS that built up on the analytical column will elute and appear at the retention time matching where PFAS in your sample would elute.
Next, inject another solvent blank immediately after the solvent blank injection with the long equilibration time and look at the retention times for PFAS compounds. The purpose of this second injection is to not give enough time for the system-related PFAS to build up in your analytical LC column.
Compare the results of the immediate injection to those of the long equilibration time injection. If there are PFAS peaks in the long equilibration time injection, but not in the injection that immediately followed it, your LC contains system-generated PFAS contaminants that could interfere with trace analysis. If there are not any peaks at the retention times of the PFAS you are looking for in both injections, your instrument may not have a significant amount of system-related PFAS interference. This can occur if you are using an old LC unit with a new mass spectrometry system because most of the leachable system-related PFAS have already leached out. However, this is not a permanent state; if you replace plastic components in any part of the instrument before the injector, it is highly likely that system-related PFAS interference will leach out unless the new parts are PFAS-free.
If you are analyzing PFAS at very low levels, like parts per trillion (ppt), for accurate identification and quantitation, you need to make sure that the PFAS detected belong only to your sample. If background PFAS contamination from your LC coelutes with sample PFAS, they will interfere and cause inaccurate results.
A PFAS delay column is installed before the injector so it traps system-related PFAS and delays their elution. This prevents them from coeluting and interfering with the PFAS in the sample. Functionally, as the gradient starts, the trapped PFAS are eluted from the delay column and then travel to the analytical column. They arrive at the analytical column after the PFAS from the injected samples, so they do not coelute with the sample PFAS.
PFAS in the sample will elute as a normal-shaped, symmetrical peak but the delayed system-related PFAS have been continuously moving throughout the system with the gradient. They were never focused on the analytical column, so when they elute the “peak” they generate is just an elevated baseline. You know it belongs to one or more PFAS because the signal matches the MS/MS transition of the compounds you are monitoring. However, because it is after the retention time window of the same PFAS in your sample, it’s not detected and quantitated as part of the sample.