TIME FOCUSING IN A FIELD PULSED ATOM-PROBE WITH A REFLECTRON

HAL Id: jpa-00229992

https://hal.archives-ouvertes.fr/jpa-00229992

Submitted on 1 Jan 1989

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire

destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

TIME FOCUSING IN A FIELD PULSED ATOM-PROBE WITH A REFLECTRON

W. Drachsel, L.V. Alvensleben, A. Melmed

To cite this version:

W. Drachsel, L.V. Alvensleben, A. Melmed. TIME FOCUSING IN A FIELD PULSED ATOM- PROBE WITH A REFLECTRON. Journal de Physique Colloques, 1989, 50 (C8), pp.C8-541-C8-545.

�10.1051/jphyscol:1989893�. �jpa-00229992�

COLLOQUE DE PHYSIQUE

Colloque C8, Suppl6ment a u n o l l , Tome 50, novembre 1989

TIME FOCUSING I N A FIELD PULSED ATOM-PROBE WITH A REFLECTRON

W. DRACHSEL, L.v. ALVENSLEBEN* and A.J. M E L M E D * * '

"'

Fritz-Haber-Institut der ax-Planck-Gesellschaft, Faradayweg 4-6,

?-zoo0 Berlin 33, F.R.G.

Institut fiir Metallphysik, Universitst Gdttingen, Hospitalstrasse 3-5, 0-3400 Gdttingen, F.R.G.

* * ~ a t i o n a l Institutes of Standards and T e c b o l o g y , Gaithersburg, M D

Abstract - The poor termination of a tip in the discharge line of an atom-probe and the field ionization of the species a t different times during the voltage pulse cause a broadening in the kinetic energy of the ions, thus deteriorating the ToF-spectra. To improve the resolution of the atom-probe, the Poschenrieder-filter was introduced quite early. Despite the elaborate designs so far in use, the simpler design of a reflectron, a s is commonly in use with other ToF- techniques, was never tried in this context.

To test this idea, a reflectron with an acceptance angle of 1.5" for a 0.8 m ToF-length was constructed. The time focusing is of first order ( a s with a Poschenrieder filter), which gives a n improvement a1 an assumed AE/E=10 % from Atlt=0.05 to Afft=0.001 with compensation.

The instrument was tested with a field pulsed atom-probe and a n improvement of m/Am from -30 to > 400 was achieved.

1 - INTRODUCTION

The application of field pulsed atom probes in material science is well established, despite their limited mass-resolution. The reason for the poor mass resolution is a broadening in the kinetic energy of the pulsed field desorbed ions. Two effects determine the ion energy: (a) the moment of field ionization during the field pulse and (b) the "ringing" of the field pulse, due to imperfect termination of the pulse cable. This can lead, under unfavorable conditions, to a AVN-0.3 (q.V=kinetic energy), which i s then the same for the relative mass resolution. This problem was recognized very early Ill, and energy compensation by a Poschenrieder deflection lens I21 a s a flight time focusing device was used by several authors 131. This device has first order compensation and focusing properties. A mass resolution of Am/m (10 % valley) of 850 can be achieved with a commercially available instrument/4/. The electrode design i s very elaborate and relatively bulky, if a big acceptance angle is chosen. Since time focusing i n other ToF-mass spectrometers is normally done by the simpler reflectron 151, we wanted to improve an existing atom-probe by attaching this device.

2 - PRINCIPLE OF OPERATION

The reflectron a s a time focusing device for ToF was introduced i n 1972 161 a s a first order compensation device a n d later for second order compensation too 171. As the AVN broadening for a n atom probe is below 10 %, a first order correction was thought to be sufficient for this case.

(''present address : Custom Probes Unlimited, Box 3938. Gaithersburg. NLI 20878, U . S . A .

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1989893

kur~i~c ^energy

I/

TOF-detector / (chevron channel plare.

phosphor screen I

U r e f

Fig. 1 - Principle scheme of the reflectron.

The field ions in an atom probe have a flight time t, which is correlated with the effective voltage V, and the field free length L a s

t Z C . L

2q-v ⁽¹⁾

The relative shift in time depends on AV by

The ions t h a t enter the reflectron are continuously decelerated (fig. 1) due to the constant retarding field until they stop a t a depth 1 and are then driven backwards a t a slightly tilted angle. In the entrance plane, the ions hit the ToF-detector. Considering deviations from a perpendicular entrance, the ions move on a parabolic curve. The travel time in the reflectron for the case of perpendicular entrance is simply given as (a = acceleration)

where I, is the length of the reflectron and VK the applied voltage. The total time of flight is

and by differentiating with respect to V, the first order shift in t can be calculated a s

The condition for first order time focussing is that At+O, realized by tl,=tl, from equation (3). This means t h a t for an average ion the condition for the pathlengths 1 = 114 L according to (4) and (5) has to be fullfilled. If a AVN of 10 O/n is expected, the total length of reflectron 111 should be 10 %greater than 1, and V I ~ should be set to 1.1 V. That means an atom probe of one meter length requires a reflectron length of 0.25 m

+

be simply understood a s the result of the higher speed ions penetrating deeper and thus needing more time in their flight.

In order to calculate the mass resolution we have to know the minimum expected deviation in arrival times. In fig. 2, the ideal resolution of an atom probe with a reflectron is calculated from the relative energy width AVN, neglecting other terms.

This shows that an atom probe with a mass resolution of 10 (if this value is limited mainly by energy broadening) should ideally obtain a mass resolution of 360 by addition of a reflectron.

Fig. 2 - Ideal resolution of ToF-atom probe.

The tilt angle remains to be discussed. The detector is not quite perpendicular to the reflected beam.

With the actual layout, the center beam hits the detector a t an angle of 8 4 3 , and a t the edges a t 85.2"

and 83.8". From this geometry, a AVt of 1.2.10-3, results. As a further improvement, the detector could be tilted by this value. In practice, these deviations can be tolerated for a Afft -resolution of a s the ion beam is focused by an external einzellens on the center part of the channel plate detector anyway.

3

-

A reflectron was built with 1 ~ = 0.22 m for an atom probe of 0.8 m length. The acceptance angle was 1.5", the center beam was tilted 5.5" against the figure axis. A fine mesh of 20 mm diameter and 90 % transparency was placed a t the entrance to avoid fringing effects. The detector consisted of a chevron channel plate of 40 mm effective diameter and a fluorescence screen with an electronic pickup for the ToF-signal. The constant retarding field was established by 40 electrode rings connected by a chain of resistors. The reflectron voltage of V~=V13+a.V-p was connected a t the endplate. The pulse fraction factor, a, had to be determined from other experiments. With the einzellens of the atom probe, the ion beam could be nicely focussed on the detector, thus increasing the acceptance angle from 1.5" to 8" and keeping down the AVt-value due to a slightly tilted detector, a s the focussed spot was of the order 5 mm in diameter.

An adaptation of the ToF-electronics and the mass evaluation program to the addition of the reflectron was not made in this work. For this reason, the improvement in the He-spectrum was apparently quite small (three times in mass resolution). The time resolution of the electronics was not sufficient to detect a real difference with or without the reflectron a t these low flight times. This problem is overcome however a t higher masses. The field evaporation of tungsten in helium gives rise to seven mass lines as W" and W He%hich could not be resolved with the atom probe alone (fig. 3). The reason for this low mass resolution of -30 in this case is caused by the poor coupling of the HV-pulse from the transmission line to the tip. With introduction of the reflectron, the mass lines can be clearly separated with a resolution of Amlm=400 (10 % valley), as is demonstrated in fig. 4. The applied tip voltage a s well the pulse voltage in both cases were similar. The instability of the reflection voltage supply was lower than 10

".

factors, so t h a t the mass resolution could be improved further. The reflectron is a first order compensation instrument, a s is the Poschenrieder filter, and i t should achieve same mass resolutior as he latter one, in principle.

m/n (amul

Fig. 3 - Mass spectrum of tungsten and tungsten helides with an atom probe.

m/n (amu)

Fig. 4 - Same spectrum, with reflectron attached.

4 - CONCLUSION

The application of the reflectron for flight time focussing yields an exceptional improvement of mass resolution for the straight atom probe used i n this experiment. This is of great importance for separating metal hydrides or studying multi-element alloys with ingredients of similar mass. Despite t h e achieved improvement, there is still a series of open questions. These include the influence of the grid, the influence of the einzellens on the performance and transmission characteristics, and the influence of the pulse fraction. These problems will be addressed by further investigations. In summary, these preliminary results show t h a t the reflectronis a simple alternative to the wide-spread Poschenrieder device in flight time focussing.

REFERENCES

111 MULLER, E.W., PANITZ, J.A., AND McLANE, S.B., Rev. Sci. Instrum. 39 (1968) 83 121 POSCHENRIEDER, W.P., Int. J. Mass Spectrom. Ion Phys. 9 (1972) 357-

I31 MULLER, E.W., AND KRISHNASWAMY, Rev. Sci. Instrum.

2

151 NEUSSER, H ~ J . , BOESL, U., WEINKAUF, R., AND SCHLAG, E.W., Int. J. Mass Spectrom. Ion Phys. a ( 1 9 8 4 ) 147

161 MAMYRIN, B.A., KARATAEV, V.I., SHMIKK, D.V., AND ZAGULIN, V.A., Sov. Phys. J E T P

3

(1973) 45

I71 MAMYRIN, B.A., AND SHMIKK, D.V., Sov. Phys. J E T P

49