Results and Discussion
Ambient Ion Focusing
When appropriate DC potentials are applied to the individual components of the electrode assembly, a strong focusing effect is observed. A similar effect has been previously observed in experiments in which a spray was generated within an ellipsoidal cavity held at kV potentials, positioned near a grounded surface . The observation of this effect is due to the propensity of ions at atmospheric pressure to follow electric field streamlines when no external pneumatic forces are applied. This is illustrated through a comparison of predicted ion trajectories as shown in Fig. 3.1b and a plot of electric field streamlines originating in same region ions were initialized (Fig. 3.1c). As stated previously, the curvature of the ion path greatly reduces the probability of neutral transmission by avoiding line-of-sight from the sprayer to the detection/deposition surface. Such a geometry is difficult to machine by traditional subtractive manufacturing techniques, yet its production is trivial through additive manufacturing methods such as FDM.
A comparison of the experimentally measured ion intensity (reconstructed intensity plots) at the deposition surface and simulated ion distributions under identical conditions are shown in Fig. 3.5. The similarity between simulation results and experimental measurements show the predictive power of SIMION-SDS. The elongation seen in the reconstructed intensity plots as compared to the simulated 2D ion distribution at the electrode exit is likely a result of both the gap between aluminum housing and the detector surface as well as the width of the pixel array (1 mm). Specifically, elongation along the pixel axis may be the result of ion diffusion during transit of the 0.711 mm gap between the housing and detector surface. This is negated slightly by the application of a potential to the housing, but cannot be eliminated in entirety. Additionally, there are undoubtedly artifacts introduced by the width of the pixels (1 mm) which would contribute to the observed elongation along the scan axis in the reconstructed intensity map. As an example, with a scan rate of 0.1 mm/s and integration time of 100 ms, the detector moves approximately 0.1 mm during a detection cycle, which is only 10 % of the pixel width. Thus the intensity at each position along the scan axis includes the entirety of ions exiting the electrode ±500 ^m from the assigned position, Y(t). Moreover, the true distribution of ions near the emitter tip is unknown and cannot be accurately modeled without complex treatment. In order to simplify the simulation ions were initiated with 3D Gaussian distribution (rxyz = 5 mm) in the source region and space-charge effects were not accounted for throughout ion transit. Gas velocity throughout the ion path was assumed to be static (no local gas flow). In spite of the simplified consideration of simulation conditions, ion intensities as simulated at the detection surface are in good agreement with experimental results, which highlights the utility of the SIMION-SDS algorithm in predicting the performance of the 3D printed polymeric electrodes.
Fig. 3.5 Experimental (a-b) and simulated (c-d) tetraalkylammonium ion intensity at deposition surface for different electrode potentials. In (a) and (c) potentials on electrodes Ej, E2, and E3 were 2.90, 2.60, and 1.80 kV, respectively; In (b) and (d) potentials on electrodes Ej, E2, and E3 were 2.95, 2.12, and 1.77 kV, respectively. In each case Esource was set to 3.00 kV and spray potential was set at 4.65 kV