In this installment, we finish our discussion of energy and elements used in atomic spectroscopy by examining an important analytical technique: inductively coupled plasma-mass spectroscopy (ICP-MS).
In this issue, we will finish our discussion of energy and elements used in atomic spectroscopy by examining an important analytical technique: inductively coupled plasma-mass spectroscopy (ICP-MS). ICP-MS is an often-crucial technique in the detection of very low levels of potentially toxic elements. To use this technique to its fullest potential, it is important to understand its structure, operation, and functionality in order to optimize the analysis of toxic and trace elements that are important to the cannabis industry.
Analytical science has developed over the centuries as a method to identify and quantitate target compounds, elements, and organisms that impact life in both positive and negative fashions. Many analytical techniques have been streamlined and refined to respond to increasing smaller quantities of these targets as we discover the minute amounts that can be a detriment to life and health. One of the most powerful tools in the inorganic or elemental analysis arsenals is inductively coupled plasma-mass spectrometry (ICP-MS).
Mass spectrometry in its many forms continues to be the driving force of research and trace analysis for many components and elements due to its increasing sensitivity and accuracy. While other elemental techniques can be more selective or sometimes easier to use; ICP-MS continues to grow as the standard for minute trace level analytical instrument for a wide variety of disciplines and markets including cannabis.
The basis for an ICP-MS starts with similar interfaces we have discussed before with other elemental techniques such as ICP-OES with sampling, nebulization, and ionization by the plasma torch. The difference comes when the sample passes into the sample introduction area of the mass spectrometer where the sample is subjected to increasing vacuum pulled by first a rotary vacuum pump and then a turbo-molecular vacuum pump before introduction to the MS optics and hardware (see Figure 1).
The pressure of the initial sample introduction and ionization starts at atmospheric pressure (760 Torr) and is decreased first to about 5 Torr (interface vacuum) by a rotary vacuum pump as it passes through the first sample introduction lens called a sampling cone that is usually composed of nickel and copper. This area is often referred to as the first vacuum stage. The second optic the sample reaches is the beginning of the high vacuum area or second stage (~1 x 104 to ~1 x 106 Torr) generated by a turbo pump. The skimmer is a nickel cone sometimes coated or tipped in platinum which directs the ion flow to the quadrupole. The platinum tipped cones are most commonly used in applications with high organic content or use highly corrosive acids such as HF which can damage the cone or contribute to interference.
In some instruments there may be some optics or mirrors that direct or divert ion flow from the skimmer cone before the flow enters the quadrupole to reduce interferences. The sampling cone and the second skimmer are also often slightly offset to reduce the flow of ions and interference from entering the MS (see Figure 2). There are some ICP-MS configurations that offer additional skimmers and extraction lenses to further reduce the background flow that can cause interference in the system.
The nebulized sample that is ionized by the plasma moves through the sample interface (between the sample cone and skimmer), which allows the sample to cool as it passes through this water-cooled zone and transition to a higher vacuum area of the system before entering the second stage vacuum area.
Once the sample is pulled into the high vacuum area, it is directed and separated by a series of lenses and mirrors before being filtered by the quadrupole or magnetic sector. The quadrupole mass detectors are the most common type and are similar to the single quadrupole mass spectrometer for liquid chromatography we discussed in previous columns. The quadrupole consists of two pairs of parallel rods usually composed of a material such as molybdenum in which direct current (DC) and radiofrequency (RF) are applied to create a charged tunnel through which analyte ions and fragments are filtered by their mass-to-charge ratios.
One pair of the rods have a positively charged DC and RF applied while the second pair of rods have a negatively charged and 180 degree offset RF to the first pair RF setting creating horizontal and vertical fields of current. The ions that exit the plasma are positively charged and move towards the negatively charged rods and away from the positive rods. An alternating current is applied that moves the ions from the negatively charged rod down the tunnel in a spiral motion as the polarity shifts to be separated by their mass to charge ratio (see Figure 3).
Lighter ions are affected more by the AC field and follow the RF fluctuations while heavier ions are less affected by the AC filed and follow the DC field. One m/z ratio travels down the path at a time and this can allow for the separation
The determination of any concentration of a particular element for ICP-MS will depend on the abundance and number of isotopes and isobars for that target element. Isotopes are atoms of the same element that have the same number of protons (same atomic number) but different numbers of neutrons (different atomic mass). Some elements are polyatomic meaning they have multiple isomers with different masses, in natural abundance ratios (see Table I).
For most elements, such as chromium there is one isotope that is the most abundant of the possible isomers, which is most commonly selected to analyze for that particular element. Other elements such as mercury have multiple isotopes with similar abundances that can be used depending on the method. The choice of which isotope to use can often be influenced by the other elements that could be present in the sample and interfere with monitoring of a particular isomer.
One type of interference for a target isotope is the occurrence of an isobar for the monitored mass. Isobars are atoms with the same mass number but different atomic numbers such as calcium, argon, and potassium (see Table II).
In the example of Table II, both the argon and calcium have their major abundance at mass 40 which means that there is a built-in overlapping interference between 40Ca and 40Ar, which can be problematic in ICP-MS systems that rely on argon gas.
Interference for ICP-MS fall into similar categories as were found in the ICP-OES analyses in our previous column: physical, chemical, and spectral.
Physical interferences are the differences between the samples being tested on any standards, calibration curves, or blanks that can alter the sample introduction and nebulization. Most of the difference stems from the makeup of the matrices of the sample and standards.
Chemical interferences are differences in the way in which a sample, standard, or element reacts in the plasma. These interferences can result in ionization effects, formation of molecules, or changes to the plasma.
Spectral interferences are changes that can mask the detection of the target element through the detection of isobars or oxidation of an element in the plasma, or double charging of an ion. The conjunction of multiple ions in the plasma can create polyatomic interference complex that can alter the target mass detection (see Table III).
Many common elements can contribute to polyatomic interferences, which can alter response at the most abundant masses. Gases such as oxygen and argon are large contributors along with common contaminants such as salts, and instrument components like the platinum from tipped cones. Most instrument manufacturers either have hardware filters or software corrections to reduce or eliminate polyatomic interferences such as the addition of collision gases, application of a collision cell, or multiple mass filters.
The physical filters such as collision or reaction cells or additional quadrupoles take advantage of the size and energy differentials between a single isotope atom versus a polyatomic complex. The size and the chemistry can make the polyatomic complex subject to either physical, chemical, or energy-based separation from the target analyte.
The benefit of ICP-MS is that the technique can be applied to a broad range of targets and samples with a high degree of accuracy and sensitivity, but this can also be its detractor when it lacks some degree of selectivity especially in the cases of interferences. Despite the challenges in cost, operation, and methodology, ICP-MS is the clear tool for trace elemental analysis and high-throughput laboratories. ICP-MS can monitor the greatest number of elements to the lowest concentrations while performing quantitative and isotopic analysis. The question is not whether a cannabis or hemp laboratory should be using ICP-MS; it is which type should be selected.
Cannabis testing laboratories have become the latest battleground for analytical instrumentation vendors. The entrepreneurial spirit in the cannabis industry has seen these laboratories springing up all over the US, trying to capture a slice of the lucrative testing market as new states begin the process of legalizing the use of medicinal and adult recreational cannabis consumer products. An investment of $1-2 million can get you all the required analytical instrumentation to start competing for business in one of the 38 states that currently allow for the use of these products. However, because of the competitive nature of this industry, the lowest bid typically gets the business. As a result, instrumentation vendors that have all the analytical instrumentation in their toolkit would normally beat out other manufacturers that offered only some of the equipment, which invariably meant that for any single technique like ICP-MS, the best instrument wasn’t always the one that was purchased.
Imagine you have just started a new job as a chemist at a start-up cannabis testing laboratory and you have been given the responsibility for buying a new ICP-MS instrument. You have previously used the technique in your last position at an environmental contract laboratory and knew the requirements for potable and wastewater samples, but you did not have a good understanding of the analytical demands of a cannabis testing laboratory. With that as background information, let’s first remind ourselves of the many benefits of ICP-MS for trace element analysis and specifically what’s important for carrying out the measurement of heavy metals in cannabis consumer products. There are several excellent commercial ICP-MS systems on the market—all with very similar specifications—so how do you choose the one that best fits your cannabis testing needs? How do you go about comparing the different designs, hardware components, and performance features, all of which are of critical importance in the decision-making process?
First, it’s very important to decide what your objectives are, particularly if you are part of your laboratory’s evaluation committee. You can have more than one objective, but they must be clearly defined. Every cannabis testing laboratory’s application demands are unique, even if they share some similarities, so it is important to prioritize them before you begin the evaluation process. Analytical capability, usability, and reliability are typically the areas that are considered the most important, so let’s take a closer look at them.
The major reason that the trace element community was attracted to ICP-MS more than 30 years ago was its extremely low multielement detection limits. Other multielement techniques, such as inductively coupled plasma-optical emission spectrometry (ICP-OES), offered very high throughput but could not achieve ultra-trace levels. Graphite furnace atomic absorption (GFAA) spectrometry offered much better detection capability than ICP-OES, it did not offer the sample throughput. In addition, GFAA was predominantly a single-element technique and was therefore impractical for carrying out rapid multielement analysis (see Table IV).
These limitations quickly led to the commercialization of ICP-MS as a tool for rapid ultra-trace element analysis. There are a number of performance metrics that are commonly used to measure the capability of an ICP-MS, including:
So, let’s take a closer look at each of these capabilities related to the demands of the cannabis industry, particularly how they might impact the evaluation process. It is not meant to be an exhaustive look but an overview of the most important criteria.
Most states that have approved the use of medical or recreational cannabis regulate four heavy metals: Pb, Cd, As and Hg, known as the “big 4.” Take for example the action limits for these four metals for orally consumed cannabis products for the state of California, which are shown in Table V. Even when the concentrations in solution are calculated after sample preparation, it can be seen that ICP-MS detection limits comfortably meet these limits. Even if the inhalation action limits were used, which are typically 10x lower than the oral regulations, the technique would have no problem in meeting these limits as can be seen in the ICP-MS factor improvement.
So, it’s important to emphasize that all instruments on the market should offer similar detection capability. By all means, compare detection limits and real-world limits of quantitation of different instruments, but with cannabis-related samples it is unlikely you will be pushing the capabilities of the technique.
Accuracy is a very difficult aspect of instrument performance to evaluate because it often reflects the skill of the person developing the method and analyzing the samples, instead of the capabilities of the instrument itself. If evaluated correctly, it is a very useful exercise to go through, particularly if you can get hold of a cannabis reference material whose values are well defined. However, when attempting to compare the accuracy of different instruments, it is essential that you prepare every sample yourself, including the calibration standards, blanks, unknown samples, quality control (QC) standards, or certified reference material (CRM), if available. I suggest that you make up enough of each solution to give to each vendor for analysis. By doing this, you eliminate the uncertainty and errors associated with different people making up different solutions. It then becomes more of an assessment of the capability of the instrument, including its sample introduction system, interface region, ion optics, mass analyzer, detector, and measurement circuitry, to handle the unknown samples, minimize interferences, and get the correct results.
It is recognized that the major source of drift and imprecision in ICP-MS, particularly with real-world samples, is associated with either the sample introduction area, the interface cones, or the ion optics system. Some of the most common problems encountered with cannabis plant materials, and cannabinoid products include:
These can all be somewhat problematic depending on the types of cannabis samples being analyzed. However, the most common and potentially serious problem with real-world cannabis matrices is the deposition of sample material on the interface cones and the ion optics over time. It probably won’t significantly impact short-term precision because careful selection of internal standards matched to the analyte masses can compensate for slight instability problems. However, sample material, particularly matrix components found in cannabis matrices, can have a dramatic effect on long-term stability. The problem is exaggerated even more in a high throughput testing laboratory, because poor stability will necessitate more regular recalibration and might even require some samples to be rerun if QC standards fall outside certain limits. There is no question that if an instrument has poor drift characteristics, it will take much longer to run an autosampler tray full of samples, and in the long term, this will result in much higher argon consumption and more regular routine maintenance.
When ICP-MS was first commercialized, it was primarily used to determine very low analyte concentrations. As a result, detection systems were only asked to measure concentration levels up to approximately five orders of magnitude. However, as the demand for greater flexibility grew, such systems were being called upon to extend their dynamic range to determine higher and higher concentrations. Today, the majority of commercial systems come standard with detectors that can measure signals up to 10 orders of magnitude. This means they can measure concentrations from low parts per trillion levels up to hundreds of parts per million. This capability has definitely enhanced the flexibility and capability of the technique, but if you are going to be using the instrument primarily to measure low levels of heavy metals in cannabis related samples, it is unlikely you will need to evaluate the instrument’s extended dynamic range feature.
There are two major types of interferences that have to be compensated for in ICP-MS: spectral and matrix interferences Although most instruments approach the principles of interference reduction in a similar way, the practical aspect of compensating for them will be different, depending on the differences in hardware components and instrument design. If the sample is digested correctly and the cannabis material is completely dissolved, matrix interferences should not pose a significant problem, as long as the acid concentration is similar in blanks, standards, and samples. The only potential problem in a high throughput laboratory is that the interface cones should be cleaned on a regular basis, so the organic matrix components do not build up and block the cone orifice.
However, it is absolutely valid to evaluate the instrument’s spectral interference reduction techniques, especially if you are in a state that requires an expanded panel of heavy metals. Of the big four metals, Pb (208 amu), Cd (114 amu), and Hg (202 amu) are at a high enough mass that there are no obvious polyatomic spectral interferences to address. The only element that suffers from any serious spectral overlaps is As because of the polyatomic spectral interferences from the 40Ar35Cl polyatomic ion at mass 75 (produced by argon gas ions combining with chloride ions from the HCl used in the microwave digestion procedure), which interferes with the monoisotopic arsenic mass 75 amu. It is relatively straight forward to mitigate the 40Ar35Cl with a collision-reaction cell (CRC), which utilizes helium as a collision gas to reduce the kinetic energy of the interference and allows the 75As mass to be used for quantitation. However, if other elements such as Cr, Ni, Cu, and Zn are in the regulated panel, it could be more difficult to mitigate the spectral interferences that can potentially interfere with these analyte elements. For that reason, it’s essential if you are in one of the states that requires this panel of heavy metals, you could possibly need the capabilities of a CRC that utilizes reaction mechanisms to minimize these types of interferences. Furthermore, it has also been shown in the open literature that triple or multi-quadrupole ICP-MS technology is needed to resolve some of the more challenging polyatomic spectral interferences on the more difficult elemental analytes.
Most cannabis testing laboratories will experience high sample workloads at some point during their operation, even though it might not be on a regular basis. So, depending on your requirements, you have to decide if high sample throughput is a requirement of your evaluation. Therefore, if speed of analysis is important it is worth carrying out a sample throughput test. Choose a suite of elements that represents your analytical challenge. Assuming you are also interested in achieving good detection capability, let the manufacturer set the measurement protocol including integration time, dwell time, settling time, number of sweeps, points and peaks, sample introduction wash-in, stabilization, wash-out times, and so on to get their best detection limits. Then time how long it takes to achieve detection limit (DL) levels in duplicate from the time the sample probe goes into the sample to the time a result comes out on the screen or printer. If you have time, it might also be worth carrying out this test in an autosampler with a small number of your typical samples. It is important that the DL measurement protocol be used because factors such as integration times and wash-out times can be compromised to reduce the analysis time. However, it’s worth emphasizing if high sample throughput is truly important, there are automated productivity enhancement systems on the market that by maximizing sample delivery and optimizing wash-out of the sample they are realizing a two- to three-fold improvement in multielement analysis times.
Analytical performance is clearly a very important consideration when evaluating the capabilities of an ICP-MS. However, it’s important to assess what we call the intangibles including:
The vast majority of instruments in use today are being operated by technician-level chemists, who may have some experience in the use of AA or ICP-OES, but in no way could be considered ICP-MS experts. Therefore, ease of use including method development and routine operation is critically important to evaluate. Furthermore, good instrument reliability is often taken for granted nowadays, but it has not always been the case. When ICP-MS was first commercialized, the early instruments were a little unpredictable, and quite prone to breakdowns. However, as the technique became more mature, the quality of instrument parts—and hence the reliability—improved. You should therefore be aware of the instrumental components that are more problematic than others and might need to be maintained on a more regular basis. It’s also very important that you talk to other ICP-MS users in the cannabis testing field to get their perspective and, in particular, the quality of the technical support they get from theirinstrument manufacturer. Their experience, both positive and negative, can be very useful for you as you are going through the evaluation process.
Patricia Atkins is a Global Product Manager with Spex, an Antylia Scientific company and has been a member of many cannabis advisory committees and working groups for cannabis including NACRW, AOAC, and ASTM.
Robert (Rob) Thomas is the principal of Scientific Solutions, an educational consulting company that serves the needs of the trace element user community. He has worked in the field of atomic and mass spectroscopy for almost 50 years, including 24 years for a manufacturer of atomic spectroscopic instrumentation. He has served on the American Chemical Society (ACS) Committee on Analytical Reagents (CAR) for the past 20 years as leader of the plasma spectrochemistry, heavy metals task force, where he has worked very closely with the United States Pharmacopeia (USP) to align ACS heavy metal testing procedures with pharmaceutical guidelines. Rob has written more than 100 technical publications, including a 15-part tutorial series on ICP-MS. He is also the editor and frequent contributor of the "Atomic Perspectives" column in Spectroscopy magazine, as well as serving on the editorial advisory board of Analytical Cannabis. In addition, Rob has authored five textbooks on the fundamental principles and applications of ICP-MS. His most recent book is a new paperback version of Measuring Heavy Metal Contaminants in Cannabis and Hemp published in December, 2021. Rob has an advanced degree in analytical chemistry from the University of Wales, UK, and is also a Fellow of the Royal Society of Chemistry (FRSC) and a Chartered Chemist (CChem).
Atkins. P., Thomas R., Energy and Elements, Part III: Understanding and Choosing an ICP-MS, Cannabis Science and Technology, 2023, 6(3), 18-21.