Within the first month of work on the project, the field ion source (FI-source) was designed as optional module to replace the electron impact ion source (EI-source) on existing ESD 1000, GAM 2000 and GAM 3000 quadrupol mass spectrometers. Before, an extensive ion optical calculation had the result, that an FI-source with optimal short ion decelleration lens would expand the quadrupol analyzer not in diameter, but in length, just by about 40 mm for ionization and focussing regions.
According to the results of the ion optical calculations, the field ion source was constructed. Since source potential differences exceeding 10.000 Volt need to be maintained in continous operation under de-terioration by sample materials that construction had to envision hard conditions to be expected.
The construction of FI-source did evolve, through three steps for different anode tip holders, since that most important element of ion source by experiments got changes of design and production process for several improvements. The first drawing of FI-source (Drawing No.1) included the first design of anode tip holder, as used in former published FI studies, and has been changed through an intermedi-ate design not practical for industrial production and adjustment (Drawing No.2) to final solution, where the anode tip is more precisely adjusted by a glass capillary in the FI source (Drawing No.3).
According to the result of ion optical calculations and source construction, three identic FI-sources have been manufactured. One of these FI-sources, connected to a quadrupol without pre-Quad, installed on the ion optical bench, is under continous test and shows continous stability in function and high voltage operation in the frequent testing of self-developed anode tips for their function after any changes of their production process.
We knew from former experiments, that to form an optimal Pt-anode tip for the FI-source, in an etching process, a Pt wire had to be periodically dipped into an etching bath with electrolytic activity. For that precise dipping, a motor driven „Dipping Unit“ was in-house designed (Drawing No.4) and produced, and did reliable work through all the experiments to develop an optimal etching process.
The well published original process „electrolytic etching in a salt melt at 800oC“ was reproduced on 100% Pt-wire of 0.1 mm diameter, requiring a well protected oven and careful handling. The rather weak Pt-anode tips did not prove to hold a reliable position on holders shown in drawing No.2 what required a total re-design and production change. We went to a more stiff base material, a Pt/Ir 80/20 wire of 0,25 mm diameter, and to avoid difficulties of „hot etching“-, we developed an „electrolytic etching in a cold salt solution“, which is providing reproducible well formed anode tips, after more experiments and prove to be real „field ion productive“,
- based in a glass capillary segment (drawing No.3) much better adjustable to the correct position
opposite to field cathode in ion source, and much more reliable holding that position,
- utilizing molar polarities of organic molecules to concentrate them by a factor of 107 in ionization
zone of FI source,
- providing over the wide range of 5 to 11 kV of FI source potentials the total ion currents at 3x10-13
to 10-10 A on the Faraday-collector of the quadrupol mass spectrometer (Quad graph No.1) which,
transformed by a Secondary Electron Multiplier (SEM) as used on any Q-MS, would allow for
quantitative analysis of organic mixtures in a 107 dynamic range of concentrations.
To demonstrate the ionizing function of the field ion source first without quadrupol, later with quadrupol, the ion source, connected to a primitive batch inlet system, is still mounted on the ion optical bench opposite to a Faraday collector. That optical bentch is located in a large vacuum tub, pumped by turbomolecular pump and forepump down to 5 x 10-7 mbar. A differential pumping of ion source and external space is not available here.
From reason of limited work capacity and limited time, a clean sample supply to FI-source an opti-mal batch inlet system could not be constructed and produced. The batch inlet system currently in use has been collected from older spare components. In its primitivity, it has several problems: The batch volume is only heatable to maximum 80 oC, inlet line with pressure reduction valve and inlet capillary at room temperature, all components are vacuum pumped by a rotary forepump to 0.05 mbar with oil background. Any inlet of a specific organic compound like benzene or toluene is accompanied with that oil background and with high residiual water background, which, in field ionization, is rather disturbing on base of the waters extraordinary high polarity.
Taking the deficiencies of experimental system into account, the quantitative stability and intensity
of total ion currents produced by the field ion source, demonstrated by Quad graph No.2 at source potentials between 5,5 and 10 kV with a toluene inlet at 1x10-4 mbar total source pressue, indicated
a promising sensitivity and quality of the single tip field ion source.
Combining that field ion source with a quadrupol on the ion optical bench, the quadrupol did trans-mit that TI (total ion current) at full intensity to a Faraday detector. However, some scans of the quadrupol through mass range 50 to 150, on base of that strong total ion current for one organic product, toluene C7H8 with mass 92, continous introduced to FI-source at a partial pressure of 1 x 10-4 mbar, had the result that the molecular ion signal at mass 92 is shown at the quadrupol analyser on top of a base of a high intensity multi ion continuum (Quad graph No.3).
One member of the projectteam, has used a single tip FI-source with a magnetic sector field mass spectrometer many years ago, differentially pumped to 10-8 mbar at ion source and analyser, and thermostated to 150oC for ion source, inlet line and and batch inlet system. In that early work, he never had problems by an ion continuum to superimpose the mass spectra.
In our current experiments, the disturbing ion continuum may contain three different types of non-specific ions, caused and produced by problems of the currently used experimental set-up :
1. Ion source, batch inlet system and inlet line are not held at a uniform high temperature to keep
the background of system low, especially the water background.
2. The inlet system and inlet line are vacuum pumped, only to a 5x10-2 mbar vacuum, by an oil filled
forepump, exposing that part of system to oil background.
3. In FI, the primary ions start with high energy from ionization zone at anode of FI-source. Many of
these ions collide with edges of sources cathode, and molecules carried with the ion beam are
getting excited, enter the quadrupole and loose under action of quadrupoles R/F field an electron
for late ionization. Their initial ion energy non specific, they form a portion of the ion continuum.
Experiments still must stay in action to minimize that continuum, to avoid continuum ions and to get analyte ion signals separated and better indicated. Our first positive results indicate :
- A grid in front of the ion detector can get the continuum to a constant smaller intensity, however will also reduce the analyte ion signal.
- A ferrite insert at the ion exit of the quadrupol can reduce the continuum by 90 % and will have a
minor negative impact to the signals of analyte ions.
- To separate the continuum from analyte ions, inclining the axis of the FI-source to that of the analyzer with a pair of deflection plates in between, as shown by drawing No. 5, should bring some result.
- A repeller electrode installed near the quadrupoles exit, to bent the ion path to Faraday collector
and SEM by 90o, will improve that separation.
- Finally, the GAM standard pre-quadrupol with ferrites at his ion entrance and/or at his ion exit will
minimize ion continuum even more and will transfer analyte ions to analyzer well prepared.
Current results demonstrate, the single tip field ion source produces safe and efficient molecular ion currrents with good intensity and sensitivity, which can satisfy the requirements of quick quanti-tative organic mixtures analysis. We confidently continue our efforts to improve the experimental set-up to a state, that FI-mass spectra arisea on a IPI mass spectrometric analyzer at a quality, as published in „Analytical Applications of Field Ion Mass Spectrometry“, Fresenius Z. Anal. Chemie 1963, 197 by
H.D.Beckey and G. Wagner, and shown here by two spectral displays of that publication.
With start of project, we intended to combine that ion source to the new IPI-QMS1-quadrupol analyzer (Drawing No. 6), well housed and protected in his recipient and best suited by standard pre-filter quadrupol of 50 mm length, to transfer field ions coming from ion source with uniform initial energy, to the analyzer quadrupol well separated from continuum ions ariving at non specific energies.
Actually, first serial IPI-QMS1-units have been ordered in time, but were delayed in production by several improvements, caused by tests of prototypes and now come along with a pre-quadrupol on their standard analyzer. In a first step, we will improve our set-up for tests on the ion optical bench by an IPI-QMS1 dual quad analyzer, to improve the analyte ion transfer a n d the separation from continuum background. With satisfactory function in these tests, we can transform one of the then existing IPI-GAM units to the FI-Q-MS by installation of FI-source to the analyzer into a special recipient extension, prepared for that bent analyzer configuration and for differential pumping, and with space in the frame for the final batch inlet system and high voltage electronics.
This IPI-GAM unit with analyzer will be available at IPI, however, its extension of space for bent source added to analyze geometry and with additional turbo-molecular vacuum pump for differential pumping, batch inlet system and high voltage electronics will add their costs to projects expenses. That configuration of a FI-Q-MS will constitute the feasibility study final Prototype with embedded computer, running under control of IPI-Q-MS1 software.
