Production of negative ion beams through charge transfer between negative hydrogen ion beams and non-metallic gases

Negative ion beam applications in tandem accelerators are used for nuclear research, environmental studies, materials analysis, medical treatments, and ion implantation in semiconductor devices. Conventional methods for generating negative ions for tandem accelerators rely on metallic vapors (typically alkali) for charge exchange, which pose challenges like contamination, electrical shorting and breakdowns, and maintenance issues. To address these drawbacks, this work explores an alternative approach to produce negative ions using a non-metallic charge exchange process. It involves directing negative hydrogen ions into neutral gases within a specially designed charge exchange cell equipped with an electrostatic accelerator. The method is applied to various gas targets, including He, H2 and O2, to accelerate and measure resulting negative ions. This innovative approach aims to mitigate contamination concerns associated with metallic vapor double-charge exchange methods and explore novel avenues for negative ion production through charge transfer. Any newly formed negative ion beam current conversion ratios from the incident H− beam will be reported as progress in this research.


Introduction
Tandem accelerators double the ion beam energy by stripping two electrons from negative ions using a gas cell or foils within a positive electrode.One area where tandem accelerators are leveraged is the ion implantation for semiconductor devices, a multi-billion-dollar industry [1].Light ions like hydrogen and helium can be used to modify the properties of power electronic devices through ion implantation [2] [3].Negative ions like helium are traditionally formed through charge exchange with metallic vapour but these vapours present contamination and maintenance challenges.To address this, we explore an innovative approach using D-Pace's TRIUMF licensed H − ion source [4] to impinge H − beams onto non-metallic gases with the goal of transferring an electron from the H − beam to the neutral gas target.A purpose-built in-vacuo electrostatic accelerator accelerates resulting negative ions, followed by separation using a mass spectrometer system.Initially, helium is chosen as a target due to its industrial applications.The production of negative ion formation using H 2 and O 2 as targets is also explored.The general expected charge transfer reaction is H − + X → H + X − , with X representing the gas targets.This work builds upon prior research by Doupé and Litherland, demonstrating nearresonant charge transfer between a Cu − beam and ICl gas [5].Near-resonance occurs when the difference in electron affinity between the projectile and target is within 1 eV, yielding significant charge transfer cross-sections [6].This paper discusses the experimental setup and reports the beam current intensity of any newly created negative ion beams, offering a novel approach to negative ion production.

Experimental setup
D-Pace's TRIUMF licensed H − ion source generates the incident beam, which is directed into a vacuum box containing the neutral gas target and an electrostatic accelerator specifically designed and constructed for these experiments.Newly formed negative ions (X − ) will be accelerated by the electrostatic accelerator along with the incident H − beam.The ion beams exiting the accelerator unit will have a different mass/energy, allowing them to be separated by the mass spectrometer system.A fixed Faraday cup downstream of the mass spectrometer dipole magnet is then capable of measuring the beam current of X − through a Keithly 6485 picoammeter with a resolution of 0.1 nA.A diagram of the test stand for the experiment including all components listed above and example beam energies is shown in figure 1.
Figure 1.A test stand diagram of the experimental setup which shows the beam path of the H − driver beam and the newly created X − ion beam.Example beam energies are used to display the test scenario but it should be noted that these energies are adjustable within the ranges described in table 1.
The adjustable experimental parameters are summarized in table 1.It should be noted that the parameters of the incident H − beam are limited by the amount of beam power that the electrostatic accelerator components can handle and not due to the performance of the source.A labeled diagram of design for the electrostatic accelerator as well as an image of the constructed accelerator unit are displayed in figures 2 and 3, respectively.The incident H − beam at 30 keV (mass M − H , energy E − H ) is detected at the Faraday cup when 8.3 A of current is applied to the mass spectrometer's dipole magnet (0.13 T magnetic field strength).Mass spectrometer current for detecting negative ions with different mass and energy is determined by the centripetal and Lorentz forces, following equation 1.
Thus, ions with a smaller mass and energy are detected on the Faraday cup at a lower mass spectrometer current.Ions created further downstream within the accelerator unit would be detected at lower magnet currents due to reduced exposure to the accelerating potential.

Experimental results using a helium target
The first gas target chosen for study was helium due to the industrial importance of He − in the ion implantation industry, where current alkali vapor-based methods can result in defective silicon chips.He − only has one known metastable state (1s2s2p) 4 , which has a lifetime of up to 359 ± 0.7 µs [7].Thus, an He − beam at a few keV is capable of travelling tens of meters in the lifetime of a He − ion, much longer than the 2 meter beamline of the experimental setup.This metastable helium state is known to have an electron affinity of 0.08 eV [8] while hydrogen has an electron affinity of 0.75 eV [9], a difference within 1 eV, which should allow for nearresonant charge transfer.Neutral helium gas was injected into the accelerator and an H − beam was passed through the target region.The goal was to create He − ions through the reaction H − + He → H + He − .Despite the potential for near-resonant charge transfer, no He − beam current was measured at a resolution of 0.1 nA during experiments.Table 1 shows all of the experimental parameters varied for the tests performed.After further literature research, it was learned that the metastable state (1s2s2p) 4 can only be reached through electronic transition from the (1s2s) 3 excited state.The injected helium target likely occupies the (1s 2 ) 1 state which cannot form He − .4. Experimental results using a hydrogen target H 2 was chosen as a gas target as a proof of concept for this novel method of negative ion production and to explore the interaction between H − beams at tens of keV and neutral H 2 gas.The electron affinity of H 2 is not well defined but a measurement taken by McWeeny gives an approximate value of 2 eV [10].This is possibly within the range of a near-resonant charge transfer which could result in the measurement of signals above a 0.1 nA resolution.

The metastability of H −
2 is well known [11], thus, the possible charge transfer reactions are hypothesized as: A few nA of beam current on the fixed Faraday cup was measured for a series of different experimental parameters.It was found that as the accelerator potential increased, the mass spectrometer current at which the peak of these beam current signals was measured decreased.This is inversely proportional to what is expected from newly created negative ions as per equation 1.These experiments were repeated using argon and helium as targets within the same mass spectrometer current range in order to confirm whether or not the detected beam current was a product of the gas target or the incident H − beam.Figures 4 and 5 display the plots of the inverse proportionality of the accelerator potential and the mass spectrometer current for negative ion detection and the results using multiple gas targets respectively.It is suspected that the incident H − beam is being slowed by the first electrode (see example energies in figure 2) and stripped of an electron when colliding with the gas target.A hypothesis is that the energetic neutral beam would gain a slow electron after exiting the accelerator and before the mass spectrometer dipole in order to be detected at the fixed Faraday cup.

Experimental results using an oxygen target
Due to availability, O 2 was chosen as a target.O 2 has an electron affinity of 0.448 ± 0.006 eV [12] which, when compared with the electron affinity of the incident negative hydrogen ion beam (0.75 eV), gives a difference of within 1 eV, allowing for near-resonant charge transfer.Based on the work performed by Bailey and Mahadevan [13], the hypothesized charge exchange processes are Experiments using O 2 as a target showed beam current peaks of up to 4 nA in intensity above the background level of the picoammeter.These beam current peaks measured on the fixed Faraday cup were found to be measured at larger mass spectrometer currents as the accelerator potential is increased, this is expected for newly formed negative ions by this method.Figure 6 displays the results of beam current measurements over a range of mass spectrometer current values for different accelerating potentials applied.The peaks shown in figure 6 are only measurable when the O 2 gas target is injected and a potential is applied to the accelerator.The peaks appearing below the expected mass spectrometer current locations could be due to negative ion formation further downstream within the accelerator.These are initial results of this work and further experimentation needs to be carried out in order to confirm the species of the ion beam being measured and to determine why multiple beam current peaks are measured for higher accelerator potential values.

Conclusion/future work
The utilization of an H − driver ion source paired with the development of and implementation of a gas cell accelerator unit poses a possible new approach to the production of negative ions.The initial work presented shows that He − ions at a resolution above 0.1 nA are not detected using this method in the current configuration.Using H 2 as a target displays an interesting electron stripping and reattachment process but does not display the hypothesised charge transfer.The the initial results of tests with an O 2 target show promising results with a possible conversion rate of 0.0013% from H − to O − or O − 2 .Testing over a larger range of accelerating potentials is required to obtain more information which is not feasible with the current system.

Figure 2 .
Figure 2. The labeled design diagram for the purpose built electrostatic accelerator which shows example energy values for the incident H − driver beam and a resulting X − ion beam.

Figure 3 .
Figure 3.The constructed electrostatic accelerator placed in the vacuum box outlined in figure 1.

Figure 4 .
Figure 4. Mass spectrometer current as a function of accelerator potential for different incident H − beam energies showing points of maximum negative beam current intensity.

Figure 5 .
Figure 5.The beam current measured on the fixed Faraday cup downstream of the mass spectrometer dipole versus the mass spectrometer current.A 20 keV beam of H − at 3.5 mA (185 µA after the collimator) was incident on hydrogen, helium and argon targets at a flow rate of 6 sccm.

Figure 6 .
Figure 6.The beam current measured on the fixed Faraday cup as a function of the mass spectrometer current for a range of accelerator potential values.The experimental parameters included a 30 keV H − beam at 6 mA (311 µA after the collimator) and 6 sccm of O 2 flow into the accelerator unit.The dashed lines indicate the expected beam current peak location for the respective accelerator potential applied.

Table 1 .
Variable experimental parameters and their corresponding ranges based on the incident H − beam energy.