Chapter 1: An Introduction to Ambient Ionization Mass Spectrometry

The introduction of desorption electrospray ionization mass spectrometry (DESI MS) by Cooks and coworkers in 2004 brought, for the first time, widespread attention to the concept of open-air surface analysis under ambient conditions.¹  Contemporary with the disclosure of DESI, work carried in parallel by other research teams explored a similar philosophy in chemical analysis. Examples include the patent on the ion source named direct analysis in real time (DART) filed in December 2003,²  Shiea's work on open-air laser-based ion sources,³  and work by the Van Berkel group at Oak Ridge National Laboratory on surface sampling probes (SSPs) for direct sampling of thin-layer chromatography plates first published in 2002.⁴  DESI, DART, and other ambient MS techniques enabled an exciting new perspective on ways to perform both qualitative and quantitative chemical investigations on samples not typically amenable to direct MS analysis. As a bonus, direct analysis on native surfaces could be done, in most cases, without sample preparation.

Considering the pressures on modern analytical laboratories in terms of workload, turnaround time, and cost per sample, it is not surprising, in perspective, that ambient MS would rise so quickly to the forefront of analytical science. Our particular interest, as a research group, stemmed from our involvement in large surveys to study the quality of anti-infective medicines used to treat malaria. These surveys required that we rapidly screen large numbers (thousands) of drug tablets for the presence of falsified and other poor-quality medicines. Being able to simply hold a tablet in front of the atmospheric-pressure inlet of a high-resolution mass spectrometer while exposing it to an ionizing plasma and obtaining a pass/fail result in seconds almost seemed like magic. It goes without saying that initial experiments that our group did with DART before the technique was officially released, got us interested almost instantly. Typical chromatographic analysis for testing drug quality requires hours of sample preparation and at least tens of minutes for chromatographic analysis. The time savings with ambient MS instantly made this type of techniques a central step in our multitiered approach to detect and source falsified and other poor-quality pharmaceuticals. After more than 8 years using various ambient MS methods for falsified drug analysis, we can confidently say that this family of desorption/ionization techniques have definitely enabled unique analytical workflows that were not possible before 2004.

As skeptical scientists, we should still ask ourselves what is truly new with respect to ambient MS approaches. Will the excitement about ambient MS withstand the challenge of time? Will we see ambient MS approaches being routinely used in laboratories worldwide 20 years from now? The answers to these questions lie in the true usefulness of ambient MS. For analytical technologies to become widely adopted they have to offer capabilities that are sufficiently different from existing approaches. The key advantages of ambient MS approaches are in the format in which well-established ionization mechanisms are implemented to enable surface analysis. DART, for example, makes use of ionization mechanisms that predominantly follow atmospheric-pressure chemical ionization (APCI) pathways, but in an open-air format. This technique, however, has enabled experiments that are not easily performed by APCI. APCI, in its most common format, requires a liquid sample. This is not the case with DART, by which one can also readily examine solids and gases directly. Direct infusion APCI, which overcomes the chromatographic step, still requires sample dissolution and pre- and postanalysis rinsing of the tubing connecting the pump propeling the liquid into the ion source. The lack of any plumbing makes DART much more impervious to memory effects that arise from rapidly injecting samples with analytes in a widely varying concentration range. Therefore, the advantages that make DART MS attractive compared to direct infusion APCI MS are related to its minimum carryover, as all parts in contact with the sample are disposable, and its shorter analysis time, as there is no need for cleaning parts. More importantly, DART has also been shown to have advantages over APCI when compared side-by-side as ion sources in LC MS. For example, LC DART MS has shown less that 11% ionization suppression in the analysis of parabens in sewage-plant effluents in comparison to APCI, which showed ionization suppression that ranged between 20 and >90%.⁵  DART MS has enabled applications such as rapid forensic screening,^6–8  rapid metabolomic profiling,^9–14  rapid bacterial typing,¹⁵  chemical profiling of live animals to study pheromone-mediated behavior,¹⁶  fast screening of counterfeit drugs,^17–19  low cost authentication of food products,²⁰  and rapid detection of warfare agents,^21,22  among others.

DESI, on the other hand, is in many ways complementary to DART (Table 1.1) and makes use of desorption mechanisms that involve a continuous solid–liquid extraction process, while capitalizing on the known ionization mechanisms of ESI. As with DART, DESI has enabled applications that are not possible by ESI. Examples include imaging of tissues in reactive mode,²³  in vivo imaging of secondary metabolites,²⁴  intraoperative lipid profiling of brain tumor tissue sections,²⁵  direct detection of chemical warfare agents,²⁶  imaging of counterfeit pharmaceuticals from developing-world countries,²⁷  clustering based on sample composition,²⁸  and imaging products of heterogeneous model prebiotic reactions on the surface of minerals,²⁹  to name a few examples. We have chosen the following two case studies to showcase in more detail unique advantages brought by DART and DESI, the two flagships of ambient MS approaches:

• (i) The work by Musah et al.³⁰  illustrates the effectiveness of DART MS in forensic drug chemistry, demonstrated by the detection of synthetic cannabinoids in herbal blends. Detection and identification of these compounds is challenging given the wide range of active ingredients, and the variety of botanical matrices in which they are found. In addition, these substances are not part of routine drug screens, and metabolites in urine would not show positive for marijuana use. To make this application even more challenging, none of the synthetic cannabinoids trigger a positive drug test using standard immunological screening procedures, and they are particularly problematic for screening methods that rely on a library search for identification because these substances are rarely included in standard databases. Additionally, the matrix in which synthetic cannabinoids are found can be comprised of several types of plants, making their detection even more arduous. This work shows that DART MS is capable of overcoming these difficulties and identifying this type of compound with a high degree of confidence without using a database. Plant leaves doped with the AM-251 and JWH-015 synthetic cannabinoids can be analyzed by simply holding the leaves with tweezers between the ionization source and the mass-spectrometer inlet. Despite the complex mass-spectral profiles given by the plant matrices, DART MS was capable of identifying the target compounds without the need for extraction or other sample preparation steps. Ionization suppression did not significantly affect analysis since 300 µg of cannabinoid were easily detected within an excess of background matrix. Unique advantages of DART MS illustrated in this case study, and that can be extended to other applications include: (i) no need for solvents, extractions, sample processing, or preparation steps; (ii) resulting spectrum produced in seconds; (iii) high-throughput analysis with no carryover between samples, and (iv) no plant matrices interference in the detection of target compounds. In contrast, comparative analysis performed by gas chromatography/mass spectrometry (GC/MS) required approximately 3 days to be completed. Analyte solvent extraction is usually undesirable given the drawbacks associated with additional sample preparation steps such as higher blanks, low recovery, and decreased sample throughput. Given that illicit drug manufacturers have demonstrated the capability to rapidly modify the components and formulations that they market, instrumentation and methodologies that can readily identify the presence of illicit substances are highly desirable.

• (ii) The study by Wu et al.³¹  is another example highlighting how ambient techniques can tackle challenging applications. This work illustrates the capability of reactive DESI to detect cholesterol in human serum and in rat brain tissues with high sensitivity and selectivity by incorporating betaine aldehyde into the spray solvent. The experiment combines a chemical derivatization in situ that takes place in the short timescale of the solvent extraction–ionization–detection process to efficiently detect a nonpolar compound. A rapid and selective nucleophilic addition occurs at the spot being sampled, generating a positively charged hemiacetal, which allows the detection of a low proton affinity analyte that is hardly ionized by common soft ionization techniques. The capability of quantitative analysis of free cholesterol in human serum was demonstrated using the standard addition method. Serum calibration solutions, spiked with isotopically labeled cholesterol-d₇ as internal standard, were manually spotted on a glass slide and analyzed with reactive DESI MS. Matrix interferences were mitigated due to the high selectivity of the nucleophilic addition reaction. The performance of this quantitation method was comparable to GC/MS and ESI MS, but accomplished in a shorter timescale. A detection limit of 1 ng was achieved with reactive DESI when a solution of 1 µg mL⁻¹ was spotted onto the surface. In addition, using 65 ppm betaine aldehyde in acetonitrile:water:dimethylformamide (8:3:1), cholesterol in rat brain tissue was imaged under ambient conditions giving a full 2D image at a 200 µm pixel size resolution within an hour. This ambient MS technique provides unique advantages for cholesterol detection in comparison to colorimetric extraction assays, and traditional hyphenated techniques in terms of high-throughput analysis, as there is no need for sample pretreatment such as derivatization, extraction or other time-consuming steps. If the sample amount is not a critical limitation, this approach can enable successful quantitation of cholesterol in biological fluids. Similarly, when high lateral resolution is not needed, reactive DESI provides imaging capabilities for mapping low-polar compounds in biological tissues under atmospheric pressure and using a soft ionization technique with no need for matrix addition.

With so many new ionization techniques being reported since 2004, distinguishing ambient ionization techniques from more conventional atmospheric-pressure ionization techniques can help delineate the different applications that are best paired with each approach. To this purpose, we propose a set of basic characteristics that should be present in techniques to be part of the “ambient ionization/sampling” MS field. First, ambient MS techniques should be able to carry ionization in the open air. This is a critical feature when examining objects of unusual shape or size such as plants, solid phase extraction fibers/bars, tablets, fabrics, etc. in direct analysis applications. Direct surface analysis capability is another key attribute of ambient MS techniques. This is particularly useful for surface analysis of solids, avoiding many, if not all sample preparation steps typically required in MS-based chemical analysis. Ambient MS ion sources are easily swappable in most types of mass spectrometers fitted with atmospheric-pressure interfaces. No modification to the ion transfer optics or the vacuum interface are generally needed for ambient MS operation, with the exception, in some cases, of suction interfaces to reduce the gas load and prevent damage to the vacuum system. It goes without saying that ambient MS techniques should generate ions without significant fragmentations, as is the case with their atmospheric-pressure counterparts.

As in ambient MS, many newly reported ionization approaches also strive to incorporate sample-preparation steps into the ionization process or analyze samples in its native form. Examples include paperspray ionization,^32,33  extractive electrospray ionization (EESI),^34,35  and fused droplet ESI (FD-ESI).³⁶  Sometimes these approaches are bundled into the ambient MS field, but not being surface-analysis techniques, we argue that this may not be correct. Paperspray ionization incorporates simple chromatographic separation and/or solid-phase extraction processes so they occur simultaneously with the ionization process, allowing direct analysis of dried biofluid samples.³³  EESI and FD-ESI incorporate a continuous liquid–liquid extraction step into the ionization process, leading to increased salt tolerance than when compared with ESI. This feature is useful for the simplified extraction of trace analytes, such as melamine, from complex samples such as milk.³⁵  Extensions of the paperspray concept can be found in the recently described “tissue-spray”,³⁷  and leaf-spray techniques.³⁸  In these cases it is possible to perform electrospray ionization of tissue components directly by wetting the sample with solvent. The sample is usually cut to have a sharp end from which to initiate the electrospray process. Paperspray, EESI/FD-ESI, tissue-spray, leaf-spray and similar techniques are best classified as direct ionization techniques more closely related to ESI than to DESI and DART.

A number of review articles and tutorials are available on the topic of ambient ionization and ambient imaging.^6,7,39–54  The classification of the various ambient MS techniques in subclasses varies in these reviews, with a certain degree of overlap. Our group^39,42,51  and others^49,50  have classified ambient MS techniques based on their intrinsic desorption/ionization mechanisms, but these divisions are sometimes debatable. This is the case when several concurrent desorption/ionization mechanisms occur. The subdivisions that we propose are as follows:

• One-step techniques where desorption occurs by solid–liquid extraction followed by ESI, APPI, sonic spray, or CI ion production mechanisms.

• One-step plasma-based techniques involving thermal or chemical sputtering neutral desorption followed by gas-phase chemical ionization.

• Two-step techniques involving thermal desorption or mechanical ablation in the first step followed by a second, separate step where secondary ionization occurs.

• Two-step techniques involving laser desorption/ablation followed by an independent secondary ionization step.

• Two-step methods involving acoustic desorption approaches.

• Multimode techniques combining two or more ambient MS techniques.

• One-of-a-kind techniques that make use of other principles for desorption or ionization that do not belong to any of the previous categories.

Table 1.2 describes the techniques that fall into the aforementioned division. It is clear from this table that not all reported techniques are truly different from each other, sometimes the differences simply being a specific set of experimental conditions (such as type of laser used, flow regime, etc.). Distinguishing between true innovation and the simple rebranding of already-reported techniques continues to be a challenge. For these reasons, authors are strongly discouraged to give existing techniques new names. In order to provide a general overview of the most common ambient MS approaches, a brief description of them is given here.

Two groups of techniques, and desorption atmospheric-pressure photoionization (DAPPI) fall in this category. The first group includes DESI and its variants such as reactive DESI, transmission mode-DESI (TM-DESI), desorption ionization by charge exchange (DICE). It can be tempting to include easy ambient sonic-spray ionization (EASI) in this group, but it must be taken into account that the ionization mechanisms in DESI and EASI are different. The second group is based on the formation of liquid microjunctions (LMJs), and comprises LMJ-surface sampling probe (LMJ-SSP), liquid-extraction surface analysis (LESA), and nanospray desorption electrospray ionization (nano-DESI).

DESI is a one-step technique where desorption occurs by solid–liquid extraction followed by ESI-like ion-production mechanisms. Solid-phase analytes are extracted into a thin liquid film that is formed on the sample surface. Small, charged solvent microdroplets continuously impact this thin film, producing secondary droplets during the collision, which are driven upwards while transporting the extracted analyte. These droplets are suctioned into an extended atmospheric-pressure ion-transfer capillary that transports ions and droplets towards the vacuum regions. A pneumatically assisted ESI probe⁵⁵  is attached to an XYZ adjustable mount to aim the spray nozzle at a surface/sample,¹  and the coaxial high flow nebulizing gas is supplied from a pressurized cylinder (100–200 psi) to induce the formation of small, charged solvent microdroplets. The five main geometrical variables to consider when fine tuning a DESI source are the incident angle, the collection angle, the sample spot-to-MS inlet distance, the tip-to-surface height, and the MS orifice-to-surface height. As in ESI, a high-voltage connection is required to produce the primary electrically charged droplets that are emitted from the spray nozzle. When sampling is performed through a transmissive/porous material the operation mode is known as TM-DESI,⁵⁶  and if specific reagents are incorporated in the solvent system to enhance the selectivity and sensitivity for certain analytes, the technique is named reactive DESI.⁵⁷  A similar technique to reactive DESI has been named DICE,⁵⁸  which targets low-polarity analytes by utilizing nonpolar spray solvents and electrochemical oxidation at the metal spray needle/solvent interface.⁵⁹  The process proceeds via charge exchange between the charged spray solvent (commonly toluene) and the low polarity (neutrally desorbed) analyte that are not efficiently ionized with conventional DESI spray solvents. When the same experimental setup as DESI (the nebulizing gas flow and polar solvent system) is utilized, but no high voltage or heat is applied, the technique is known as EASI. This technique was initially referred to as desorption sonic spray ionization (DeSSI)⁶⁰  given the prevailing ionization mechanisms involving bipolar-charged droplets formed at atmospheric pressure that lead to an ion current dependent on the supersonic gas velocity.⁶¹ 

The second family of one-step solid–liquid extraction-based techniques is based on the formation of a liquid junction.⁶²  When surface sampling is performed by a semistatic liquid junction,⁶³  the technique is known as LMJ-SSP,⁴  now commercially available under the name of flowprobe™. The LMJ probe assembly consists of two concentric tubes, with the outer (larger internal diameter) tube supplying the fresh extraction solvent to the surface, and the inner tube applying suction to pull the solution to an ESI probe for direct analysis, or for offline collection prior to a chemical separation. The variables that influence performance are the liquid junction height and the inner capillary retraction. Similarly, when the microliquid extraction from a solid surface is integrated with nanoelectrospray mass spectrometry, the technique is known as LESA, which is an adaptation of the Nanomate^® robotic pipette chip-based infusion nano-ESI system.^64,65  LESA is fully automated by means of a robotic arm and uses disposable pipette tips creating LMJs with spatial resolutions of ∼1 mm, and single-use nano-ESI nozzles eliminating spot-to-spot sample carryover. Nanoelectrospray is initiated by applying the appropriate high voltage to the pipette tip and gas pressure on the liquid. The smallest scale LMJ approach to ambient surface analysis has been called nano-DESI.^66,67  This approach uses two fused-silica capillaries connected by a solvent bridge formed on the sample surface. It involves a solid–liquid extraction mechanism as part of the desorption process, and does not employ a nebulizing gas.

The remaining technique belonging to this classification group is desorption atmospheric-pressure photoionization (DAPPI),⁶⁸  which is based on a combination of thermo/chemical desorption processes and atmospheric-pressure photoionization mechanisms.⁶⁹  A heated mix of atomized gas and solvent vapor produced by a microchip nebulizer is aimed at the sample. In addition, the desorption region is exposed to ultraviolet (UV) radiation. In positive mode, ions are produced by photoionization of desorbed neutral analytes, charge-transfer reactions with solvent species or dopant molecules, and/or ion–molecule reactions involving protonated solvent/dopants.⁷⁰  In negative-ion mode similar pathways lead to ion production, with the additional electron-transfer mechanisms.

The techniques that fall into this category involve metastables and reactive ions. These species react with the analyte directly or indirectly through proton- and charge-transfer reactions. Optional heating can be used to enhance desorption, and samples can be placed in a glancing or transmission geometry. Plasma-based ambient MS techniques can be grouped based on the following: (a) design features for removal of plasma species from the flowing gas stream (nitrogen, helium, argon etc.) previous to interaction with the sample, (b) presence or absence of heating used for enhancing desorption, and type of heating employed, (c) discharge operation mode (DC, AC pulsed mode), (d) current–voltage regime chosen for operation to distinguish between glow, corona, spark, etc., (e) discharge configuration (annular, point to plane), and (f) applicability to miniaturized instrumentation based on the discharge gas consumption (mL min⁻¹–L min⁻¹). Based on the number of publications, DART, flowing atmospheric-pressure afterglow (FAPA),^71,72  and low-temperature plasma (LTP)⁷³  are the main techniques, followed by others such as dielectric barrier discharge ionization (DBDI),⁷⁴  desorption atmospheric-pressure chemical ionization (DAPCI),⁷⁵  and desorption corona beam ionization (DCBI).⁷⁶  Other plasma-based ambient techniques with a smaller number of examples include PADI,⁷⁷  a technique that employs a radiofrequency-driven glow discharge plasma in direct contact with the sample;⁷⁸  atmospheric-pressure thermal desorption/ionization⁷⁹  (APTDI), a technique suitable for the analysis of organic salts; helium atmospheric-pressure glow-discharge ionization⁸⁰  (HAPGDI), an early name given to FAPA; and microhollow cathode discharge (MHCD) microplasmas.⁸¹  The “atmospheric solids analysis probe” or ASAP ion source,⁸²  which is commercially available, is based on the introduction of a probe directly into the plasma of a modified APCI ion source, and probably shares ionization mechanisms with standard APCI ion sources, with desorption occurring by a stream of heated gas as in DART and FAPA.

DART uses a negatively biased point-to-plane atmospheric-pressure glow discharge physically separated from the ionization region by one or several electrodes. The metastable species formed within the discharge supporting gas, typically He or N₂, generate protonated water clusters by Penning ionization of atmospheric water molecules from naturally present moisture.⁸³  These hydronium ion–water clusters are the reactive reagents involved in proton-transfer reactions with analyte molecules thermally desorbed by the heated gas stream.^39,83  The thermal conductivity of the gas used for desorption is a critical parameter affecting the sensitivity due to the thermal desorption step, with He being approximately one order of magnitude more conducting than N₂. In the majority of applications, the utilization of DART has mainly been focused on analytes with masses below 1 kDa. In contrast to DART, in FAPA there is no filtering of plasma species by any electrodes before interaction with the sample, leading to a higher reactivity, and is operated in the current-controlled glow-to-arc regime (∼25 mA), whereas DART typically operates at lower currents. A third difference with DART is that FAPA achieves heating of the gas stream through Joule heating within the electrical discharge and not by an external heater. DAPCI⁸⁴  and DCBI⁷⁶  also use APCI-like ionization mechanisms, but the desorption process is based on chemical sputtering by charged solvent gaseous species formed by the plasma. In DAPCI, gas-phase solvent vapors are ionized by corona discharge ionization. DCBI uses a helium plasma sustained in the corona regime (10–40 μA discharge current under a 3 kV potential difference) and can be operated in modes that seem equivalent to temperature-ramped DART,⁸⁵  low-current FAPA and DAPCI. Thermal mechanisms and chemical sputtering mechanisms are most likely responsible for desorption in DCBI. Solvent vapors can be selectively added to the DCBI probe, achieving DAPCI-like desorption and ionization.

Neutral desorption extractive electrospray ionization (ND-EESI),⁸⁶  beta electron-assisted direct chemical ionization (BADCI),⁸⁷  atmospheric-pressure thermal desorption–secondary ionization (AP-TD/SI),⁸⁸  and probe electrospray ionization (PESI)⁸⁹  belong to this family of two-step desorption–ionization techniques. In ND-EESI⁹⁰  a plume of neutrals is generated from the surface of a sample by the impact of a gas stream. This plume is subsequently ionized in an ion cloud created by ESI of solvents.³⁴  Uncharged analytes in the sample spray beam are ionized through charge-transfer reactions taking place during collisions between neutral aerosols and ESI ions in the gas phase. In this way desorption and ionization events are separated in time and space, enhancing ionization efficiency in complex samples. Probe electrospray ionization (PESI) is a two-step technique that uses disposable acupuncture needles as solid-sampling electrospray probes.⁸⁹  When the probe picks material up from a biological sample, a very small amount of water (∼pL)⁹¹  carried on the needle surface is sufficient to induce an electrospray when a high voltage is applied to the probe. Automation can be achieved if the needle is attached to a motorized controller to control the depth and rate of sampling. Samples with complex matrices such as biological tissues can be analyzed intact with a single needle with no clogging, and no carryover if the needle surface is clean after analysis in repeated measurements.

In this group of techniques the analyte is desorbed or ablated from a surface by an IR or UV laser with or without a matrix. The generated sample plume is subsequently merged with an electrospray droplet cloud or a plasma stream, depending on the ionization source utilized for the second step. Analytes are ionized through ESI mechanisms, or charge/proton-transfer reactions. As desorption and ionization processes are separated in space and time, samples are not directly in contact with the ionizing plume, and desorption and ionization can be optimized independently.

The most used name for IR laser-sampling/ESI ionization hybrid techniques has been laser-ablation electrospray ionization (LAESI),⁹²  which is commercially available. However, the first technique that coupled laser sampling to an ESI source was electrospray-assisted laser desorption ionization (ELDI),³  which used a 337 nm nanosecond pulsed nitrogen laser. In contrast, when the IR-ablated sample is picked up in the open air by a plasma stream operated in the glow regime, the technique has been named infrared laser ablation metastable-induced chemical ionization (IR-LAMICI).⁹³  Additional techniques that belong to this family and share some instrumentation aspects with LAESI are: infrared laser-assisted desorption electrospray ionization (IR LADESI),⁹⁴  laser desorption electrospray ionization (LDESI),⁹⁵  laser ablation mass spectrometry (LAMS),⁹⁶  laser desorption spray postionization (LDSPI),⁹⁷  and laser electrospray mass spectrometry (LEMS).⁹⁸  The latter uses a high-intensity nonresonant femtosecond laser (laser pulse ∼ 10¹³ W cm⁻²) allowing analysis of anhydrous samples. An interesting LEMS capability is the preservation, at atmospheric pressure, of condensed-phase protein conformation upon transfer into the gas phase for capture and ionization in the electrospray plume.⁹⁹ 

When the IR or UV laser excites an exogenous matrix that cocrystallizes with the analyte, and a voltage (∼500 V) is applied to the stainless steel target plate, the technique is called matrix-assisted laser desorption electrospray ionization (MALDESI).¹⁰⁰  The parameter settings that influence performance in this group of techniques comprise the source geometry, i.e. the laser incidence angle (90° or 45°) and the ionization source; the laser wavelength; the duration of the laser pulse; the pulse energy; the repetition frequency, and the use (or not) of a matrix (endogenous or exogenous). IR lasers are usually tuned for 2940 nm with pulses of 5 ns duration at 2–20 Hz and pulse energy between 100 µJ and 2.5 mJ. As the IR laser resonantly couples with the O–H water stretch, endogenous water inherently present in biological samples can act as ionization matrix and facilitate desorption.^101–103  Infrared laser ablation can also be combined with photoionization, improving the analysis of compounds with different polarities in a technique named laser ablation atmospheric-pressure photoionization (LAAPPI).¹⁰⁴ 

An additional group of two-step techniques are those that involve laser desorption/ablation in a transmission or reflection geometry to produce a sample plume that is subsequently transferred to a liquid phase, such as a droplet, for analysis.¹⁰⁵  The droplet-capture approach can also be replaced by a continuous-flow LMJ-SSP, operated in a noncontact, surface sampling mode, providing additional means for mass-spectrometry imaging (MSI).¹⁰⁶  The laser-ablated analyte can also be captured on solvent droplets for transferring to a MALDI target,¹⁰⁷  or can be directly captured on a slide to be used as the target for vacuum MALDI MSI, allowing additional imaging capabilities such as high spatial resolution.¹⁰⁸ 

In this group of techniques the analyte is desorbed through a laser-induced acoustic wave or by a piezoelement generating an aerosol plume. The neutral plume is subsequently entrained by reactive ion species or charged solvent droplets from an external ionization source (ESI or APCI).^109,110  As desorption and ionization are two separate steps, the original sample is not in contact with the ionization plume directly. Laser-induced acoustic desorption (LIAD) is a nonresonant laser-based matrix-free desorption approach in which the sample is deposited onto a thin (∼10–15 µm thickness) metal foil (e.g. titanium or aluminum), which is irradiated from the backside with a series of high-energy laser pulses. The acoustic waves created by the laser propagate through the metal foil, causing desorption of nonvolatile, and thermally labile compounds on the other side of the foil. In particular, photosensitive compounds can be analyzed because there is no direct sample exposure to the laser. Metal foils used as substrates should have low reflectivity at the laser wavelength used, low thermal conductivity, and high thermal expansion coefficient to minimize the amount of thermal energy that reaches the analyte.¹¹¹ 

Surface acoustic wave nebulization (SAWN) is also a matrix-free method, which generates low internal energy ions from a planar piezoelectric surface using acoustic waves of 400 µm wavelength. This method produces an aerosol from the liquid sample drop that is deposited on a chip,¹¹²  by applying a radiofrequency signal (∼9.56 MHz) in a pulsed or continuous mode to an interdigitated transducer patterned onto a piezoelectric LiNbO₃ wafer. The liquid surface tension in the droplet is disrupted, atomizing the sample, with additional desolvation of the generated aerosols occurring in the mass-spectrometer inlet.

Experiments combining DESI and DART-type ionization processes can be performed with desorption electrospray/metastable-induced ionization (DEMI).¹¹³  This ion source can be operated in three modes (plasma mode, spray mode or combined mode) affording a versatile platform for ambient MS.

Rapid evaporative ionization mass spectrometry (REIMS),¹¹⁴  laser desorption ionization (LDI),¹¹⁵  switched ferroelectric plasma ionizer (SwiFerr),¹¹⁶  and laserspray ionization (LSI),¹¹⁷  are one-of-a-kind techniques that do not belong to any of the previous categories. REIMS and LDI are of particular interest since they exemplify the contribution that ambient MS approaches may have in a surgical scenario. When a high-frequency electric current is applied to surgical blades as in REIMS,¹¹⁴  or when utilizing surgical CO₂ lasers as in LDI,¹¹⁵  ablated tissues produce aerosols, and charged species by the heat dissipated during the process. These species created during surgery can be removed by suction from the surgical site and transported to the mass spectrometer for analysis.

As with any rapidly moving technology field, predicting the future directions where we will see important developments in ambient MS is a risky game. Several research themes that we suggest will become increasingly important are:

• Hybrid multimode ion sources: Only a handful of examples on the combination of two ambient ion sources exist. One such example DEMI, which allows for experiments that combine DESI and DART-type ionization processes.¹¹³  Another example is the combination of desorption ionization by charge exchange (DICE) with DESI (DICE/DESI).¹¹⁸  In this approach, a tee union creates a zone where immiscible solvents can intermix before being directed to the spray needle. These solvents allow different ionization chemistries to take place. In both of these cases, the ability of obtaining additional chemical information in a single experiment from a broader spectrum of chemical species remains the key advantage.

• Robotization and full automation of ambient MS: Interfacing of ambient MS approaches to robotics and digital microfluidics is an exciting, yet unexplored area of research that could potentially enable automation and integration of many of the existing approaches into platforms that can respond to the demands of the modern analytical laboratory. Ion transfer in ambient MS techniques occurs under atmospheric-pressure conditions where there is a complex interaction between fluid dynamic forces, thermal gradients and electric fields. This results in a scenario where small changes in sample position and orientation result in large changes in sensitivity. Although this can be viewed as a disadvantage, it can also be considered an opportunity. By reproducibly placing the sample in different positions in space one could imagine that different ionization conditions could be achieved in a dynamic, continuous fashion.

• Quantitative measurements: Quantitation by ambient MS, although completely feasible from the perspective of the governing basic principles, is generally quite challenging due to the variability in the sample introduction conditions. If proper care is taken in introducing the sample reproducibly, it has been shown in numerous times that quantitation is indeed possible.^119–121  The remaining challenges involve a better basic understanding of how analytes can be differentially enriched during ablation/desorption, the extent of ion suppression that exists due to charge competition, and further insights into controlling the prevailing ionization mechanisms for a more rational development of quantitative applications.

• Nanoscale imaging: Nanoimaging by ambient MS will enable further reduction in the dimensions of the systems that we are able to study, i.e. from tissues to cells to subcellular compartments. Ambient MS experiments where nanoscale resolution is obtained have been recently reported.¹²²  In another example, a multimodal imaging platform that allows coregistering bright-field fluorescence and mass-spectral chemical images has been recently developed by coupling a laser capture microdissection instrument with transmission geometry for laser ablation to a mass spectrometer with an APCI ion source.¹²³ 

• Machine learning and expert systems: Machine learning of ambient MS fingerprints will allow expert-independent decision making based on highly complex MS data. Combination of high-throughput ambient MS approaches with advanced classification and pattern recognition algorithms will be useful in developing automated systems used in diagnostics and food safety applications, among others.

• Inclusion of ambient MS technology in the operating room: Medical questions can be translated to chemical questions that ambient MS approaches can address to help physicians make decisions.¹²⁴  Ions and aerosols created during electrosurgery are removed by suction from the surgical site and transported to the mass spectrometer for analysis by remote sampling,^115,125,126  providing metabolic information and chemical histology. Identifying the margins of pathological tissue during operation is one example that can improve surgical decision making in real time.