Among the imaging techniques available, DESI imaging is the most accessible entry point: the sample sits in open air, no matrix is applied, and a tissue section can go from cryostat to acquisition with very little in between. That simplicity is genuine and it is the main reason to choose the technique. It also has costs, and they are worth understanding precisely rather than generally.
Key Takeaways
|
The Ambient Ionization Concept
Ambient ionisation means generating ions from a sample in its native state, at atmospheric pressure, without prior preparation or enclosure. For imaging, that changes the workflow more than it changes the physics. There is no matrix to select, apply, and optimise, no vacuum compatibility requirement, and no crystal size setting a ceiling on resolution. What limits resolution instead is the geometry of the sampling probe.
That distinction matters for how the technique is usually adopted. As a society-published guide to mass spectrometry imaging notes, DESI analyses tissue in the near-native state without extensive preparation, whereas MALDI and SIMS typically operate under vacuum. A laboratory with a suitable mass spectrometer can therefore begin ambient imaging with comparatively modest additional investment, which is why it functions as an entry route into the field.
Working in analytical science?
Register for a FREE Separation Science account to subscribe to the Separation Science Newsletter.
One clarification before going further, because the terminology is often muddled. Post-ionisation approaches, in which a second laser or energy source ionises neutral material released from the surface, are sometimes grouped with ambient methods. They are a different thing: post-ionisation is a sensitivity enhancement applied within a desorption experiment, and it can be implemented under vacuum or at atmospheric pressure. Whether ionisation occurs in open air is a separate question from whether a second ionisation step is used.
How Does DESI Imaging Work?
A pneumatically assisted electrospray directs charged solvent droplets at the sample surface at a shallow angle. Those droplets strike the surface, dissolve analytes into a thin liquid layer, and the impact of subsequent droplets ejects secondary droplets carrying dissolved material toward the mass spectrometer inlet. Ionisation happens largely in that secondary droplet population, by mechanisms comparable to conventional electrospray.
Four parameters govern the result, and they are all geometric or chemical rather than instrumental.
- Solvent composition. Determines what dissolves and therefore what is detected, and as shown below determines whether the section survives.
- Spray angle and distance. Sets the sampled footprint, which is the primary limit on lateral resolution.
- Gas and solvent flow. Affects both extraction efficiency and how far material spreads laterally before collection.
- Stage speed. Since acquisition is usually continuous along a line rather than discrete per pixel, stage speed and scan rate together define effective pixel size.
The important consequence is that a DESI experiment is optimised chemically. In MALDI, the matrix decision dominates, as covered in MALDI Imaging Mass Spectrometry: How It Works. In DESI, the solvent decision does the equivalent work, and it is adjusted per application rather than selected from a shortlist.
Nano-DESI and the Resolution Question
Resolution is where ambient methods have historically been weakest and where most of the development effort has gone. It is also where published figures need reading carefully, because three different numbers circulate and they mean different things.
Approach | Reported Spatial Resolution | What Sets the Limit | Notes |
Conventional DESI | Typical ambient MSI around 100 microns | Spray footprint and lateral solvent spreading | Robust and simple; the usual starting point |
Nano-DESI | Commonly 20 to 200 microns | Size of the liquid bridge formed at the surface | Two capillaries form a continuous flowing solvent bridge |
Nano-DESI with shear force probe | Better than 10 microns | Precise control of probe-to-sample distance | Distance control is the enabling development |
Nano-DESI, proteoforms | Down to 7 microns demonstrated | As above, but signal-limited | Protein signals fall substantially at high resolution |
Nano-DESI, denatured proteins | Around 200 microns typically | Analyte abundance and ionisation efficiency | Same instrument, coarser resolution, because the analyte is harder |
LESA | Sampling spot around 600 microns, pixel size around 1 mm | Diameter of the liquid microjunction contact | Coarse by design; suits discrete spot sampling rather than fine imaging |
Table 1. Reported spatial resolutions across ambient imaging approaches, compiled from the peer-reviewed sources cited in this section. Note that figures are analyte-dependent as well as instrument-dependent.
A tutorial review of nano-DESI mass spectrometry imaging in ACS Measurement Science Au sets out the progression. Routine operation with a three-point plane calibration to correct for sample tilt supports imaging in the 20 to 200 micron range. Introducing a shear force probe to control the distance between sample and probe, which generates a small liquid bridge at the surface, has enabled resolution better than 10 microns. The same review notes that solvent composition is optimised specifically to improve analyte extraction and ionisation efficiency, and that the field has had to establish normalisation methods to measure concentration gradients accurately because matrix effects are significant.
Three Numbers That Are Not the Same Thing First, acquired resolution on abundant analytes. Nano-DESI has reached around 10 microns for abundant lipids and metabolites. Second, acquired resolution on difficult analytes: the same technique typically manages around 200 microns for denatured proteins, and work demonstrating proteoform mapping down to 7 microns notes explicitly that protein signals decrease substantially in high-spatial-resolution experiments. Resolution is a property of the experiment, not the instrument. Third, and most often misreported: 5 micron molecular images have been produced by fusing lower-resolution ion images with high-resolution optical microscopy of a stained section. That is a computational result, a predictive image informed by two measurements, rather than resolution achieved at acquisition. It is a legitimate and useful approach, and it is not the same claim. Any figure quoted for ambient MSI resolution should be checked for which of these three it describes. |
One further capability worth knowing. Nano-DESI generates multiply charged protein ions, which is advantageous for identification by top-down proteomics, and native variants using ammonium acetate-based aqueous solvents have been used to map intact proteins and complexes. That is a genuine point of difference from MALDI, which produces predominantly singly charged ions. The work on native nano-DESI imaging of proteins and complexes sets out both the approach and the resolution constraints that come with it.
Other Ambient Techniques
DESI and nano-DESI dominate, but the ambient family is broader and the alternatives suit different sampling problems.
- LESA, liquid extraction surface analysis. A solvent droplet is held between a pipette tip and the surface, forming a liquid microjunction that extracts analytes from a discrete spot. Reported sampling diameters of around 600 microns and pixel sizes near a millimetre make it coarse for imaging, but it is well suited to sampling defined regions and to native protein work.
- Laser ablation with electrospray post-ionisation. Material is ablated from the surface by a laser and ionised in an electrospray plume, combining laser sampling precision with electrospray ionisation characteristics.
- Atmospheric-pressure MALDI. MALDI performed without vacuum, which retains matrix chemistry while gaining the sample handling advantages of open-air operation. Useful when volatile or vacuum-sensitive analytes are of interest.
This is also the right place to resolve the terminology point raised earlier. Post-ionisation, sometimes labelled MALDI-2, applies a second ionisation step to neutral material released from the surface and can substantially improve sensitivity. It is a valuable development and it belongs to a different classification: it concerns how many ionisation events occur, not whether the experiment runs at atmospheric pressure. A reader looking for post-ionisation will find it discussed alongside MALDI instrumentation rather than among ambient methods, in Mass Spectrometry Imaging: Principles, Techniques, and Applications.
Why Does Minimal Preparation Matter?
Beyond convenience, and this is the strongest practical argument for ambient imaging. Because no matrix is applied and the section is analysed in near-native state, the same tissue section can often be stained and examined histologically after imaging. That turns an ion image and a pathology image into two measurements of the same object rather than two adjacent sections.
The published position is more specific than the general claim, and considerably more useful. Work on non-destructive, histologically compatible tissue imaging by DESI examined exactly this and found that compatibility depends on the solvent system.
Solvent System | Effect on the Section | Consequence |
Dimethylformamide and ethanol, 1:1 | No effect on tissue morphology observed relative to controls; no spatial delocalisation of cellular protein content during subsequent staining | The same section can be imaged, then H&E stained or subjected to immunohistochemistry |
Acetonitrile and water, 1:1, conventional | Tissue destruction occurs | The section is not available for subsequent histology |
Table 2. Solvent choice and histological compatibility in DESI imaging, from the cited study. Findings are for the tissue types and protocols examined there; validate on your own material before relying on it.
That study went further than morphology, confirming by pathological evaluation that sections first imaged with the compatible solvent showed no delocalisation of stained protein content during either H&E or p63 immunohistochemistry protocols, and demonstrating direct overlay of DESI ion images with the H&E stain of the same section. In one bladder cancer example, a region called negative by the ion distribution was confirmed as normal tissue by pathological evaluation of the overlaid image.
The practical implication is a decision rather than a feature. If same-section histology matters to your study, the solvent has to be chosen for it at the outset, accepting whatever that costs in extraction efficiency for your analytes. If it does not, a conventional solvent system may detect more. Describing DESI as simply non-destructive skips over the choice entirely.
DESI or MALDI?
They answer different practical questions, and the decision is usually made on preparation and section fate rather than on performance.
Consideration | Favours DESI | Favours MALDI |
Sample preparation | No matrix; minimal handling; fastest route to a first image | Matrix application required, and it is the critical variable |
Best achievable resolution | Better than 10 microns with nano-DESI and distance control | Single-digit microns routinely, and sub-micron crystals demonstrated |
Analyte range | Metabolites, lipids, and proteins including native and intact species | Widest overall: metabolites through to intact proteins and glycans |
Charge states | Multiply charged proteins, which aids top-down identification | Predominantly singly charged, giving simpler spectra |
Low mass region | No matrix ions, so the low mass range is clearer | Matrix-derived ions crowd the low mass range |
Section after analysis | Can be preserved for histology with a compatible solvent | Consumed at ablated positions |
Throughput on large areas | Continuous line scanning is efficient at moderate resolution | Fast at coarse rasters on TOF platforms |
Table 3. Choosing between the two dominant approaches. Most laboratories use whichever their instrument supports, so the useful reading is which constraints you are accepting.
The row most often overlooked is the low mass region. Because DESI applies no matrix, the low mass range is not congested by matrix clusters and fragments, which is a genuine advantage for small-molecule work and one that partly offsets the resolution gap. Set against that, the section fate row is the one that decides studies where tissue is scarce or where pathology review is part of the design.
For where both approaches sit relative to the antibody and sequencing-based spatial techniques, see Spatial Analysis in Analytical Science: Mass Spectrometry Imaging and Spatial Omics.
This article was produced under Separation Science's AI Editorial Guidelines.
