Ionization mechanism

initial findings

SESI and ESI (Electro-Spary Ionization) have been linked since the initial development of ESI. Fenn and colleagues already noted that gas-phase species in the counterflow gas were also ionized by the spray. Still today, the routine calibration of mass spectral data relies on ubiquitous plasticizes and other contaminants (i.e. polydimethylcyclosiloxanes, phthalates) that are ionized in regular ESI via Secondary Electrospray Ionization mechanisms. Indeed, most laboratories and MS users have databases of identified contaminants, which appear in the mass spectra regardless of the liquid sample. The persistence of the signals produced by these commonly known contaminants shows that the ionization mechanism of SESI is very efficient.

In an early stage, several authors started to demonstrate the potential of SESI-MS in different applications using home-made non-optimized SESI sources.


Droplet vs ion pathway

Initial discussions on the ionization mechanism considered two possible ionization mechanisms:

  • Droplet pathway: gas phase molecules are first adsorbed by the electrospray droplets, and then as the droplet collapses, they are ionized via regular ESI ionization mechanisms.

  • Ion pathway: the ions or ion clusters produced by the electrospray transfer their charge to the gas phase molecules when they collide.

SUPER SESI is dominated by the ion-pathway. This discussion is still open when large droplets and ions coexist. However, SUPER SESI uses a nano-electrospray very close to the boiling point of the liquid. This results in very small nano-droplets that evaporate very quickly. The region containing droplets is extraordinarily small, and the ionization region is mostly populated by ions (and clusters) in the lowest energy level.

Having one single ionization mechanism is important for two reasons:

  • This eases studying and understanding the ionization mechanism

  • This simplification allows us to develop specific numerical models, which are at the heart of the optimization


Aerosols and vapors

Many of the species detected by SUPER SESI in breath have remarkably low vapor pressures. Do these molecules stay isolated in the gas form, or do they condense to form aerosols? We still don't have a clear answer to this. The lungs produce aerosols with a log-normal distribution centered about 0.3 micrometers. In SUPER SESI, these aerosols are evaporated in the sample line, which operates at a high temperature, and the resulting vapors are ionized.

The effect of humidity

Rising the humidity level of the gas introduced through the sample line accelerates the charge transfer reaction rate in SESI. It is hypothesized that larger clusters transfer their charge more efficiently than smaller clusters or even bare ions. A study introducing regular and deuterated ethanol and water in the sample gas, showed that the gas-phase proton-affinity of the molecules clustering around the charging ion, and the molecules of interest is a key parameter.

The humidity has to be carefully considered, particularly when developing a new application or experimental set-up. FIT's solutions address this problem to ensure that no variations in the humidity level will affect your measurements.

  • Exhaled air from the alveolar region of the lungs has a very steady humidity level, with a dew point regulated by the body temperature (36-37ºC). For breath research and analysis, the humidity is self regulated.

  • Headspace analysis, (or for the analysis of other humid samples), controlling the temperature and the humidity of the sample is crucial to ensure a constant humidity level. Adding extra humidity helps to boost some signals and to eliminate possible variations.


dynamics of the expanding plume, first models

Even when the droplets undergo complete evaporation and the cluster size distribution reaches its equilibrium, the ionization environment in a SESI is not uniform because ions are constantly expanding due to coulombic repulsion. This is one of the main difficulties when it comes to accurately model, quantitatively understand, and optimize a SESI design.

The first attempts to model an expanding Secondary Electro-Spray Ionizer used analytical solutions to the equations governing the SESI plume, which could be greatly simplified because they tended to infinity at the tip of the electrospray. Interestingly, these models also allowed the effect of charge competition to be quantitatively estimated.

General SESI eqs.png

In these equations, b is more concentrated and has a higher proton affinity than s.

c produced via ESI mechanism
s,b ionized via SESI mechanisms
nc,s,b ion concentration
Ns,b neutral molecule concentration
E Electric field
Vf Flow velocity field
Zc,s,b Ionic mobilities
Kcs,cb,sb reaction rate constants
e elementary electric charge
ε0 electric permitivity of the vacuum
τ time

Some interesting conclusions of these mathematical models:

  • The concentration of ionized molecules is uniform across the electro-spray plume.
    The coulombic expansion of the spray plume dilutes the charging ions and the ionized molecules. For this reason, the concentration of charging ions is not uniform. However, when no charge competition effects are present, the molecules in the gas phase are constantly replenished by charge transfer reactions. The mathematical model shows that these two effects cancel out, resulting in a uniform concentration of ions.

Concentration Sample eq.png
Concentration Competitor eq.png

This geometrical uniformity is highly beneficial because signal intensities do not depend much on tight mechanical tolerances.

  • Charge competition effects are produced when a concentrated vapor hinders ionization of other species with lower proton affinity. For these effect to be noticeable, the concentration of charged vapors (nb* )and charging ions have to be of the same order.
  • These mathematical models established the basis upon which the first optimized SESI configuration was engineered.


charge competition effects

Charge competition effects can be observed if the gas supply of the mass spectrometer is contaminated, or if the sample gas is strongly dominated by one species. This can be troublesome if the gas supply of the mass spectrometer is not carefully controlled. In order to prevent this, SUPER SESI incorporates a filter that eliminates those contaminants.


dynamics of the expanding plume,
Numerical modelling & Ionizer design

a) First SESI numerical method, and optimized LF-SESI as described in  Sensors and Actuators 2015 .

a) First SESI numerical method, and optimized LF-SESI as described in Sensors and Actuators 2015.

Having an accurate numerical model to simulate the electrospray plume is essential to fine-tune and optimmize any engineering design. SESI is not an exception to this rule. However, simulating a SESI is a difficult task, even if the details of the chemistry are all wrapped in a reaction rate constant. There are three main challenges to properly simulate a SESI:

  • It involves many physical processes (fluid dynamics in the gas phase, fluid dynamics in the liquid phase, electrostatics with moving ions, chemical reactions, diffusion, and even turbulence)

  • All these processes are highly coupled

  • It involves very different scales, with the electrospray singularity producing numerically unstable solutions that tend to infinity

b) FIT's first SESI numerical method, and first electrode-less SESI configuration.

b) FIT's first SESI numerical method, and first electrode-less SESI configuration.

The first optimized SESI source was designed by FIT's founder to be integrated in an explosives detector, coupled with an Ion Mobility Filter. At the time, no numerical model was available, and fine tuning was not an option. For this reason, knowing that the exact configuration within the SESI was not known, the design criteria was to 'withstand' turbulence or any other unknown, rather than to be streamlined. The resulting configuration, termed Low Flow SESI used a set of plate electrodes to keep the different gases separated and to create strong electric fields that guide the ions no matter what.

A first numerical method that solved the problem associated with the singularity of the electrospray tip and the expansion of the plume was developed in 2015, and it was used to optimize the diameters and distances of a Low Flow SESI to be coupled with a Thermo instrument. This source improved ionization efficiency, but at the cost of increasing the internal surface area, and thus the background levels.

c) FIT's new numerical method, featuring the effect of the electrostatic forces on the gas velocity field, and the toroidal vortex.

c) FIT's new numerical method, featuring the effect of the electrostatic forces on the gas velocity field, and the toroidal vortex.

Removing the electrodes was the next step to improve the background levels. The first Electrode-less SESI was designed in 2016. This design used a similar numerical method. For the first time, this method allowed the internal geometry of the SESI to be fine-tuned so that the different gassed were not mixed and ions were transferred to the mass spectrometer. This source is not immune to turbulence, it simply is laminar.

FIT's newest numerical method retrofits the flow fields with the electrostatic forces produced by the expanding plume. The electrostatic forces generate a toroidal vortex right in the ionization region. In the new SUPER SESI, this vortex is stabilized, locked in a fixed position, and the flow induced by it is used to draw gas phase molecules into the ionization region. This results in an improved ionization efficiency.