We finish looking at liquid chromatography with a deeper examination of the mechanics, physics, and chemistry of different spectrometers that are used in analytical laboratories.
Over the last two columns we have looked at the basics of chromatography and examined method development. In this column, we will finish looking at liquid chromatography with a deeper examination of the mechanics, physics, and chemistry of different spectrometers that are used in analytical laboratories. We will look deeper at mass spectroscopy and its relationship with liquid chromatography and thereby add to our working knowledge base of liquid chromatography processes. It is hoped that these tools allow the chromatographer or spectroscopist to maximize their efficiency and develop sound methods for analysis of their targets by giving an in-depth view of how the different chemistries, technologies, and instruments function to produce results.
One of the first instruments a chromatographer encounters early in their career is often a liquid chromatography system with some form of an energy-based detector. It is a principal starting point of liquid chromatographic separations that can be used for the analysis of a variety of compounds and often is the conjoined partner of a spectrometry-based liquid chromatography–mass spectrometry (LC–MS) system. At the heart of most high performance liquid chromatography (HPLC) systems is a spectrometry or spectrophotometry configuration in which energy (or light) plays an important role in analysis.
There is often a lot of confusion between the terms chromatography, spectroscopy, spectrometry, and spectrophotometry. These concepts are intrinsically linked, but are still different approaches to looking at (or measuring) matter with light (or energy). The terms are all based around the Greek word origins for color (chroma-) and light (photo-) combined with the terms: ‘to write’ (graphein), ‘to measure’ (metria), ‘to see’ (skopia), or the Latin ‘to look at’ (specere). The different disciplines can be understood as a general sphere of study with levels of increasing specialty or specificity.
On the highest level, especially for analytical scientists, there is spectrometry which is the measurement of the interaction of matter with energy (in particular electromagnetic radiation). There are several types of spectrometry defined by their target and method of measurement: neutron triple-axis, ion-mobility, Rutherford backscattering, mass spectrometry, and spectroscopy (Figure 1).
The most elite (and least available) form of spectroscopy is neutron triple-axis spectrometry. Neutron triple-axis spectrometry or triple-axis spectrometry (TAS) (Figure 2a) is the measurement of inelastic neutron scattering at points of physical space or energy monitored within the range of the spectrometer. Inelastic neutron scattering measures the change in kinetic energy during a collision of neutrons and a sample. Inelastic neutron scattering is an experimental technique commonly used in condensed matter research to study atomic and molecular motion as well as magnetic and crystal field excitations (2–4). It distinguishes itself from other neutron scattering techniques by resolving the change in kinetic energy that occurs when the collision between neutrons and the sample is inelastic. Neutron scattering occurs between a neutron and a target nucleus, in elastic scattering a single neutron is emitted from the sample nucleus and no energy is transferred into excitation of the target nucleus. Inelastic neutron scattering occurs when some of the energy of the incident particle is either absorbed, increased, or lost with the recoiling of the sample nucleus leading to an excited state. In this situation, momentum is conserved but the kinetic energy of the system is not. These instruments are used in research capacities at high level research institutes and are not commonly encountered by most spectroscopists.
If the most difficult and least accessible forms of spectroscopy is TAS then one of the most accessible forms of spectroscopy is ion-mobility. Ion-mobility spectrometry (IMS) (Figure 2b) targets, separates, and identifies ionized gas-phase molecules by their ion mobility in a carrier gas. The simplest model of IMS is the drift tube (and its varied configurations) which allow sample molecules carried by the carrier gas to flow through a tube at controlled intervals to interact with the applied electric field and the neutral drift molecules which allow for the separation of ions. The time the ions take to traverse the tube is measured from fastest to slowest. Often these spectrometers are coupled with other systems and detectors such as gas chromatography (GC)–MS or LC–MS. The most common use of these detectors is for high-speed detection of environmental or in situ targets. They are most often seen in portable or field instruments used to detect drugs, explosives, or chemical weapons. Their speed and compact forms make them common spectroscopy tools in use by a wide variety of skill levels including non-spectroscopists.
The remaining three types of spectroscopy or spectrometers are where most spectroscopist have the most experience depending upon the type of targets and samples that are being examined. Rutherford backscattering spectrometry (RBS)
(Figure 2c) is a technique where the structure and composition of a material is determined using a high-energy ion scattering to measure the backscattering from a beam of high energy ions (protons or alpha particles) onto a sample. This type of spectrometry is found most commonly in the material sciences.
The last two types of spectrometry are of particular importance to most analytical laboratories and include mass spectrometry and optical spectroscopy.
Mass spectrometry (MS) measures the mass-to-charge ratio of charged particles which results in a mass spectrum (intensity versus mass-to-charge plot). The main parts of a mass spectrometer are a sample inlet, an ion source, a mass analyzer, and a detector. Sample inlets allow for the controlled introduction of a gaseous or vaporized liquid sample (or solid via a heated probe) through an aperture where the sample passes to an ionization source which generates ions.
Samples are ionized by several different mechanisms depending on the type of sample and dissociated into fragments that are characteristic for the molecules or elements in the sample. The fragments of smaller mass sometimes carry a charge and can be measured with relative intensity or abundance that provide structural or isotopic information needed for identification. Most ionization techniques fall into either ‘hard’ ionization or ‘soft’ ionization depending on the ionization energy involved and the degree of fragmentation which results.
Hard ionization uses high quantities of energy in fragmenting the target molecules and result in a large number of fragments from the rending of bonds in the original molecule. The fragments tend to have lower mass-to-charge ratios (m/z) than the parent molecule. The most common hard ionization technique for organic molecules is electron impact ionization (EI) that employs a high-energy electron beam (~70 eV) to form radical cations which then decompose to smaller fragments. These fragments are the basis for the mass spectrum (sometimes referred to as the molecular fingerprint) and its use in identifying compounds (Figure 3). EI requires a system that can be kept under high vacuum such as a gas chromatography system (GC–MS) and is not well suited for LC systems that are mostly operated at atmospheric pressure. Many libraries exist for typical mass spectrum using GC–MS with an EI source to ascertain identities of unknown molecules or species detected in samples.
Soft ionization uses small amounts of energy to ionize molecules and result in only a small number of fragments. The most commonly used soft ionization techniques in an analytical laboratory include chemical ionization (CI), atmospheric-pressure chemical ionization (APCI), electrospray ionization (ESI), atmospheric pressure photoionization (APPI), and matrix-assisted laser desorption (MALDI).
In chemical ionization (CI) techniques, ion fragments are produced by the collision between sample molecules and a collision gas. This type of ionization requires lower energy than other types of ionization depending on the type of sample and collision gas. CI often provides a simpler spectrum with little to no fragmentation. In CI and other soft ionization techniques, the molecular ion peak [M+1]+ is present and is helpful to determine molecular mass. This simpler spectrum can limit the amount of structural information for a particular sample or element but can be useful when other stronger ionization techniques such as EI make molecular ion peaks undetectable. CI techniques, similar to EI techniques, tend to be used in conjunction with GC–MS systems under high vacuum.
Atmospheric pressure ionization sources such as ESI, APCI, APPI, and MALDI are characterized with ionization occurring at atmospheric pressure before the resulting ions are introduced into the mass detector under vacuum. These techniques (except for MALDI) are associated with liquid chromatography systems (LC–MS). MALDI, which stands for matrix-assisted laser desorption/ionization, fits neither a pure gas chromatography nor liquid chromatography system. MALDI is an ionization process where a solid sample or a target sample mixed with a solid matrix is applied to a sampling support and a laser ablates and desorbs the sample. The sample molecules are then ionized by protonation or deprotonation by ablated gases. This technique is most often seen for use with large biological molecules and tissues. Often the choice between GC–MS or LC–MS sources and MALDI are dependent upon the sample form, target analyte size, and molecule polarity (Figure 4).
A pulsed laser irradiates the sample, triggering ablation and desorption of the sample and matrix material. Finally, the analyte molecules are ionized by being protonated or deprotonated in the hot plume of ablated gases, and then they can be accelerated into whichever mass spectrometer is used to analyze them.
The mostly widely used source for LC–MS is an ESI source. Electrospray ionization (ESI) produces ions using a nebulizing needle or probe in addition to drying gases and an applied energy voltage to create an aerosol. ESI is considered to be a soft ionization technique that can create multiply charged ions which can be several orders of magnitude larger than the parent compound. This multiply charged ion species can be seen in molecules such as proteins and peptides.
During ESI, a nebulizer needle with a nebulizing gas, such as nitrogen, flows through a fused silica capillary. There is a charge applied which forces the oppositely charged ions to lag in the capillary and forces the similarly charged ions to gather at the end of the capillary tip. The flow from the capillary and the nebulizing gas force a discharge from the tip and form a Taylor cone of clustered similar charged ions, which then converge into droplets with multiple charges (electrospray). The droplets make up the aerosol plume composed of solvent, gas, and charged ions (Figure 5a).
As the ion droplets progress, repulsion of like charges and surface tension cause the Rayleigh limit to be reached and the expulsion of single charged ions and the formation of smaller droplets which then continue to desolvate as the droplets are evaporated by heated drying gas until only ions remain to enter the capillary inlet to the mass analyzer (Figure 5b).
The initial formation of droplets can be affected by multiple variables including choice of mobile phase, mobile phase additives, applied voltage, and nebulizer gas flow and temperature. In past columns, we examined the importance of the chemistry of solvents and mobile phase to aid in ionization prior to reaching the ionization source by creating charged species. In the ionization source, mobile phase composition and additives can play an important role in droplet size and surface tension.
Solvents with lower surface tension can facilitate the release of single charges from the multiply charged droplets. Water has the highest surface tension; therefore, methods with high aqueous mobile phase will need changes to the other parameters or addition of other solvents to aid in surface tension (Table I). The composition of the mobile phase can also affect the rate of desolvation with more volatile solvents evaporating quickly than less volatile solvents.
The first step in creating ions in ESI is producing a robust and reproducible electrospray by altering the applied voltage (capillary voltage), drying gas flow, and temperature. First, we will examine the effect of changes in voltage to the spray. The average applied voltage is about 3000–3500 V. The voltage helps to form a Taylor cone and pull the charged ions towards the capillary. A Taylor cone is a cone of material formed during electrospray in which a jet of charged particles is formed above a threshold voltage. This cone formation allows then for a jet of charged particles to travel and spread in the desolvation zone to the capillary aperture.
Most LC–MS systems accommodate between 2000–6000 V, which can be optimized for individual targets using flow injection analysis (FIA) studies where one sets up a series of injections of the target analyte with stepwise changes to different MS parameters. Modern LC–MS manufacturers often include instructions on the correct set up of an FIA experiment.
At lower voltages, a Taylor cone may not form, and the droplets fall out of the stream to waste, resulting in a lack of signal. As the voltage increases toward optimum, a cone and jet are formed (axial spray), usually in the range of about 3000–3500 V. Axial spray mode is the optimal condition for reproducible signal. As voltage increases, the cone and flow can start to destabilize (rim emission) which can still result in signal, but that signal may be unstable. Finally, signal can be lost by coronal discharge if the voltage exceeds a certain potential where all the droplets and solvent are completely ionized and evaporated before reaching the capillary.
The flow and temperature of the drying gas and nebulizer pressure can change the number of ions reaching the mass detector. These parameters are interconnected and are dependent upon several factors including type of ionization source and HPLC flow rate (Table II). Table II represents the most common range of settings for the different types of ionization sources at different flow rates. Best practices always dictate to start at the lowest value and increase slowly to obtain the best results with the minimal output.
There are several drawbacks to using ESI, starting with the types of target analytes ionized by ESI. Successful ESI targets are moderately polar and easy to ionize. ESI is most efficient with less aqueous mobile phase mixes and more organic mobile phases with low surface tension. In addition, ESI requires somewhat lower flow rates and often will need to have flow partially diverted prior to entering the source so as not to overwhelm the source’s ability to desolvate the droplets.
ESI produces multiply charged species such as dimers and trimers; and can form adducts with a variety of different system contaminants and mobile phase components. Common adducts include sodium (M+23) from glassware; acetonitrile adducts (M+43 ACN+H) and dimers (2M+H). Adduct tables are available from many sources including LC–MS manufacturers. My personal choice in my LC–MS work is an excel adduct calculator available at: www.lc-ms.nl. Some common LC–MS contaminants can become adducts or appear as target fragments in the mass spectrum. These compounds can be from the mobile phase, plastic tubing, modifiers, or glassware (Table III).
The other two types of LC–MS ionization sources are atmospheric pressure chemical ionization (APCI) and atmospheric pressure photoionization (APPI). These two sources are used less frequently than ESI but are powerful tools when the target analytes of interest are not ionizable using ESI due to their polarity or size (Figure 4).
The APCI and APPI sources have a lot of configurational elements in common to the ESI source. All the atmospheric pressure sources have a nebulizer needle to create an aerosol plume. The APCI source contains an additional coronal discharge probe that forces the jet more quickly into the coronal discharge mode, which in ESI with its highly polar targets is unfavorable but for APCI for less polar targets, the added energy allows for a stable creation of ions. The desolvation and ionization of the molecules takes place because of gas phase chemistry (where ESI ionization takes place in the liquid phase).
APCI can reduce thermal decomposition of targets since the droplets are quickly vaporized in a hot source (250–400 °C) which leads to fewer fragments and adducts. APCI has rapid desolvation power with its fast transition to gas phase and higher energy and allows for higher flow rates up to about 2 mL/min without splitting flow to the ionization source. The mobile phase composition can be composed of higher organic solvents with a less polar chemistry due to the gas phase chemistry. It is also due to the gas phase chemistry that factors such as polarity and pH of solvents are less of a concern for ionization.
The final type of ionization source we will examine is APPI. APPI uses the power of photos to ionize molecules in the gas phase which require additional energy to ionize such as nonpolar analytes. To understand how much energy is needed for molecules, we have to take a closer look at proton affinity (PA) and ionization energy (IE) (or ionization potential [IP]). PA is related to the acidity of a molecule we discussed in prior columns in association with pKa. IP is the total amount of energy required to free an outermost shell electron of an atom. Many APPI use a krypton lamp as a source of photos to ionize analytes. The krypton lamp emits about 10 eV of energy which can then directly ionize any analytes with an IE less than 10 eV. In addition, APPI sources can be aided by the addition of easily ionizable solvent molecules called dopants which help to ionize target analytes (whose IP >10) by several methods including: initiating gas phase ion or molecule reactions that subsequently form analyte ions; by undergoing charge exchange or proton transfer. Common dopants for LC–MS include solvents such as acetone (IP = 9.7) and toluene (IP =8.8).
There are many advantages and disadvantages to APPI. Advantages include the ability to analyze a wide range of polarities and molecular sizes in one analysis. APPI is the least subject to adducts, ion suppression, and matrix effects. The APPI source, like the APCI source, is well suited to higher flow rates unlike the ESI source. Some of the disadvantages are that often a dopant is needed to overcome ionization potential and to optimize sensitivity. In general, more response is found with ESI and APCI sources. In the end, the choice is based on the type of target analytes being examined (Table IV).
Once ions are created in the LC–MS, the analyte ions are carried to a detector where they can be fragmented to various degrees with application of charge. For LC–MS, the detectors are often based on one of three configurations: quadrupoles, traps, or time-of-flight. New detectors (tandem MS/MS) are often multiples or combinations of these basic three types (that is, ‘triple quads’ [the combination of multiple quadrupoles], quadrupole and time of flight [Q-ToF], and ion trap ToF (ion trap with time of flight). Orbitrap is a form of an ion trap where the trap acts as the analyzer and detector. The orbitrap has a pouched tube-shaped reservoir with outer electrodes and a central centrode. Ions enter the trap and are trapped. The ions oscillate around and between the electrodes at different frequencies, which results in mass separation. Orbitrap is a form of a the highly costly research Fourier transform mass analyzers and tend to be highly specialized and often cost prohibitive for many analytical laboratories. For the purposes of this column, we will not go into great detail about this specialized type of mass analyzer.
The mostly widely used LC–MS detector is the quadrupole detector (either a single quad or triple quad). The quadrupole (Q) is the mass analyzer for the system. After ionization by one of the many sources previously discussed, the ions pass through a charged capillary tube, other ion optics such as a skimmer, octupole, and focusing lens which create a fragmentation zone before entering the analyzer. The initial areas of the system are usually under low vacuum from the rough pump (3–4 Torr) until the area around and after the interior end of the capillary and skimmer which then transition to high vacuum provided by the turbo pump (9-15 x 10-6 Torr) to keep the quadrupole under high vacuum. The quadrupole consists of two pair of parallel rods, which direct current (DC) and radio frequency (RF) are applied to create a charged tunnel through which analyte ions and fragments are filtered by their mass-to-charge ratios (Figure 6a). In a single quadrupole system, there is only one pathway for detection which often creates the parent ion plus either an adduct or a hydrogen (M+1) or a simple fragment. There are some changes to the system energies that can cause collision induced dissociation (CID) of all the ions present leading to some fragmentation. Raising fragmentation voltage can start to break apart parent ions into smaller fragments but without the mass filtering found in other types of mass analyzers to be discussed.
In a triple quadrupole (QQQ) system, the three sets of quadrupoles can each be set to different frequencies or energies to create additional sets of smaller fragments which can be used in identification (MS/MS). The first and the third quadrupoles act as mass filters and the second quadrupole becomes a collision cell to induce CID on isolated masses to create fragments (Figure 6b). The collision cell is the area of CID where fragments are created by interaction with other ions in the cell.
Ion trap mass (IT) systems have a charged reservoir or trap into which the analyte ions are deposited (Figure 7). In most cases direct current is not used, and the spectrum are measured first by grounding the endcap electrodes and applying a low high-frequency (hf) voltage to the ring electrodes. Then ions of the required m/z are allowed into the trap. The hf voltage is increased to the trap at the ring electrodes, which fragments the analyte ions and filters the mass-to-charge ratios of the unstable oscillating ions through the endcaps to reach the detector. The ion trap can apply multiple different ionization energies to the trap and can create multiple fragments similar to passing through multiple triple quads and is often referred to as MSn due to its ability to produce a large number of fragmentations.
The trade off between different types of LC–MS sources tends to be cost, ease of operation, resolving power, mass resolution, fragmentation ability, and mass accuracy. Resolving power, as discussed in the previous column, is the ability to separate and distinguish two adjacent peaks while mass resolution is the ability of an instrument to distinguish between adjacent ions of equal intensity. Fragmentation ability is the instrumentation’s ability to create product ions from parent fragments after the application of an energy source. Instruments such as a single quadrupole (Q) have a few methods for simple fragmentation, but precise fragmentation is more in the role of other systems such as QQQ or IT. Mass accuracy is the ability of the system to discern a highly accurate mass for an ion; often these types of systems (ToF, orbitrap) fall under the category of high-resolution mass spectrometers (HRMS). These systems can produce highly accurate mass assignments but are high in cost and can be difficult to operate and maintain. Table V compares the different attributes for the discussed LC–MS systems.
Time of flight analyzers measure extremely accurate mass for analyte ions determined by a time-of-flight measurement. Ions enter a charged field where masses are filtered before being exposed to an energy field for acceleration. The selected ion range are accelerated through the field with larger ions having slower speed than smaller ions. The ions then drift without additional energy through the flight path or ion drift tube (Figure 8). The time to pass through the flight path is measured by the detector and the resulting measurement produces an extremely accurate mass calculation which can be used as a method of identification. ToF uses transits in time through the drift region to separate masses and works best when ions are accelerated in small groups towards the detector. While some ToF systems incorporate fragmentation, the main advantage of the system is very accurate mass for identification. For example, if a certain molecular formula has a mass error of +/-0.5 Daltons, then that variation could mean several hundred possible identifications but if the mass resolution of a ToF is 0.001 Daltons or between 1–5 ppm; then the possible identifications can be narrowed down sometimes into the single digits.
LC–MS software will have several choices regarding the mode of analysis. These modes are mostly independent of the type of ionization source and detector. The basic modes for most systems are scan, single ion monitoring (SIM), and multiple reaction monitoring (MRM). The most general basic mode is scan in which the range for detection of the instrument is set to a given mass range such as 50–400 m/z. The software and detector then scan that range over the set scan time.
There is often the ability in the software to set stepwise scan programs which change variables of scan time, mass range, and dwell time to maximize detection of ions. These programs can be created to eliminate solvent detection and focus the detection on a particular mass range or time in the method run. For example, if the target are cannabinoids in the 300 m/z (+/- 5 m/z) range then the scan range can be set narrowly in 250–350 (or smaller) to focus detection only on those compounds. Further, if the retention time is known, then windows for different scan ranges, and scan times can be set only for times which those peaks are present.
There is also the possibility of creating fragmentation from the settings in scan mode by creating a program that isolates a small range of masses then applies increasing fragmentation voltage to produce fragments (even in systems such as single quads, which do not usually create many fragment ions). The resulting data from scan mode can the be expressed in several types of chromatograms including total ion chromatograms, base peak chromatograms, and extracted ion chromatograms. Total ion chromatograms (TIC) often have the highest response and show all the ions collected during a run. This information can show information about ions that may have been missed if only looking for a single target mass; but this excess of information can also bury the signal of target with reduced response. Base peak chromatograms (BPC) pull the strongest or most abundant spectrum for each peak and can reduce some of the noise seen in TIC. Finally, extracted ion chromatograms (EIC) pull the spectrum for only one mass or small range of masses which can be helpful to really see small, less responsive peaks but also can miss any adducts that may form and not show any potentially important contaminants or other compounds that may coelute with the target peak.
SIM mode produces results similar to what is seen in an EIC but contain only data points on the selected ions with higher resolution. In EIC, the response is much lower that the TIC response from which the data was extracted. In SIM, all the focus of data collection is on only the chosen ion or ions. SIM mode is good for experiments where the masses and retention times for all the targets is well defined. The SIM mode will not see contamination or interfering compounds, but it will create higher response for the targets. In SIM and MRM, the time each ion is scanned (dwell time) can be adjusted to optimize results.
Multiple reaction mode (MRM) is primarily used in tandem mass spectroscopy systems where individual masses (precursor ions) can be isolated and then fragmented to create smaller fragments (product ions) at different energy transitions. This type of mode can be very powerful for identification of unknown compounds.
There are several types of detectors for mass analyzers including: electron multipliers, photomultipliers, and microchannel plates. All three employ dynodes that convert ions into electrons which are then amplified to produce a current. Most quadrupoles and ion traps use an electron multiplier which receives a single ion and then multiplies it down the multiplier to increase signal for detection. The second most common LC–MS detector is the microchannel plate which is used most often with ToF spectrometers. These detectors are two-dimensional parallel arrays of continuous dynode electron multipliers which are highly sensitive with a very fast response.
Finally, photomultipliers use sealed vacuum tubes that are responsive to varieties of light energy to create a signal and are the least often used detectors in LC–MS. These detectors are more often found in applications for optical spectrometry and spectrophotometry, which is the final branch of the spectroscopy family that we will explore more in-depth in the next column.
Spectrometry is an integral part of the analytical laboratory whether it be as part of a GC–MS or LC–MS system. As was found in the selection of columns and mobile phase in previous columns, chemistry and physical properties of the target analytes play a significant role in the choice of components in a spectrometry system. For LC–MS there are multiple areas of the process that can all be optimized to give maximum results. In this column, we took a deep look at many of the types of components available to the LC–MS scientist from the various types of ionization sources, which often begin with an ESI source to more specialized APCI, APPI, and MALDI sources. The difference and appropriate uses for the various types of mass analyzers shows that the choice of detector depends upon the way in which the LC–MS is used. Systems like the single quadrupole are lower cost workhorses for routine quality control and studies while more fragmentation or mass resolving power is needed for elucidating compound identities and come with corresponding costs. We examined the different types of modes, settings, and parameters that could be altered to fine tune methods all in pursuit of good and accurate data.
Patricia Atkins is a Senior Applications Scientist 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.
P. Atkins, Cannabis Science and Technology 4(6), 26-38 (2021).