HAL Id: jpa-00229938
https://hal.archives-ouvertes.fr/jpa-00229938
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
HAL, estdestinée 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.
DIRECT TIP STRUCTURES DETERMINATION BY SCANNING TUNNELING MICROSCOPY
R. García Cantú, M. Huerta Garnica
To cite this version:
R. García Cantú, M. Huerta Garnica. DIRECT TIP STRUCTURES DETERMINATION BY SCAN-
NING TUNNELING MICROSCOPY. Journal de Physique Colloques, 1989, 50 (C8), pp.C8-235-C8-
240. �10.1051/jphyscol:1989840�. �jpa-00229938�
COLLOQUE DE PHYSIQUE
Colloque C8, suppl6ment au n o 11, Tome 50, novembre 1989
DIRECT TIP STRUCTURES DETERMINATION BY SCANNING TUNNELING MICROSCOPY
R. G A R C ~ A
CANT^
and M.A. HUERTA GARNICACentro d e Investigaci6n y de Estudios Avanzados del IPN, Departamento d e ~ n g e n i e r i a ~ 1 6 c t r i c a . Secci6n d e ~ e t r o l o g i a , Apartado Postal 14-740, 07000 Mdxico, D.F., ~ 6 x i c o
Abstract
-
An electrochemical etched scanning tunneling microscope tip is studied, using a long scan tunneling microscope and scanning electron microscopy, in order to characterize the neighborhood of the tip apex. Observed microstructure and protrusions are discussed in relation to chemical etching and mechanical resistance.1.- INTRODUCTION
Since Binnig and Rohrer /l/ developed the Scanning Tunneling Microscope (STM) tip structure determination has been one of the most intriguing problems in the understandinq and function of the STM. Atomic resolution achieved lonq time ago brought the idea of a solely atom tunneling at the tip extreme /2/;
The construction of a tip ending in a single atom, was properly shown using field ion microscopy / 3 / . On the other hand, working tips are prepared by different ways going from grindstone and electrochemical etching to ion milling /4/. Typical radius of 0.1-10 Mm have been observed by different techniques /4-8/. Several attempts for determination of the tip structure using STM were made by producing an indentation with the effective tip in a soft metal surface like silver and gold /9,10/.
A direct observation using STM of a working tip has not been reported yet, although STM could bring information not seen by techniques like Scanning Electron Microscopy (SEM) or Transmission Electron Microscopy (TEM). Indeed three dimensional capability and the high depth of field at any amplification of the STM complement the other techniques.
Theoretical studies related with the spatial distribution of the tunneling current have shown a highly lateral localization for tip radius less than 0.1 pm. The observed results permit to assume an effective tip radius of less than 10 A, probably due to protrusions at the end of the tip.
Protrusions obtained with electrochemical etching have a high mechanical resistance. This can be seen if the tip is used as an indenter, as a puncher to make holes or as a cutting tool. An STM microstructure characterization could be useful to understand the obseped mechanical resistance.
The determination of the tip structure by STM involves problems of positioning as well as convolution effects. Positioning of the sample tip within the measuring volume of the STM requires a three dimensional coarse approach of several microns in each axis, the identification of the sample tip also needs long scan in each axis of the STM unit. In this work
,
topographic structure of an electrochemical etching tip at sub-optical scales is presented.2.- EXPERIMENTAL METHODS
The results were obtained at air pressure with a long scan tunneling microscope /11/ which has a measuring volume of up to 40 X 40 X 25 pm3, this is a new version of a long scan prototype already reported /12/. Vertical " Z N
control is provided with coupled coils fixed in mylar membranes. The coils moving in permanent magnets. This system provides as long as
+
12 pmdisplacement and still control the tip position. Since resonance frequency is 260 Hz for the coupled coils, we use 0.04 Hz frequency for the "X" triangular wave function. X and Y motion are provided with 5-inches long piezoelectric tubes. All axis are mechanically decoupled. At this very slow scanning, each line is drawn in 25 seconds and we spend about 40 minutes for a 100 lines image. This long time determines a limit for amplification of about 250,000X
(40 A ) due to thermal drift. At 100,000X amplification thermal drift is not a
mai or problem.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1989840
The sample tip was made of 0.5 mm diameter wire. It was prepared with NaOH 2N solution with 7 volts DC, positive voltage at the tip, for 12 minutes and changing polarity during a few seconds. Probe tip was made of 0.1 mm diameter wire and sharpened with a similar NaOH 2N solution, with alternating voltages RMS value of 12 V until the tip was "sharpI1 as seen with a 400X optical microscope. Probing tip's. quality looks much better than sample tip's at this amplification. Positioning of the sample tip was made by micrometers, since our scan covers an area of up to 40ym X 40pm, it was possible to do a fine approach within this area to image the tip. A set of scanning electron microscope images was obtained to compare with the observed STM structure.
3.- RESULTS AND DISCUSSION
In order to obtain the image shown in figure 1, we made many consecutive observations with different conditions for tip position, scanned area and plotter amplification. In addition to repeatability for consecutive lines, we can recognize structural repeatability in different tip images.
This makes us feel confident that this is the actual tip. Figure l(a) is a low vertical amplification (10, OOOX)
,
with a scanned area of 8.8 X 5.4 pm2. At this scale, tip can be recognized as a spherical pattern. Surrounding it, there is a flat region corresponding to the maximum depth the probing tip can reach, so, there is a servo control saturation. Even at this low vertical amplification, some structural details can be observed as it is indicated by an arrow. A place is reached where lines become closer and shadow is apparent indicating the back of the tip. In figure l(b), we went deeper, vestical amplification is twice and scanned area is almost the same, 8.8 X 5.1 pm. The scales were chosen in such a way that this image almost corresponds to a real one. An arrow indicates the same structural details as in fig. l(a). We can see how the structural surface profiles are now more evident. Shadow is darker and we circled the place where, apparently, is located the tip's highest point, which can be estimated by measuring the height from the base line.In figure l(c), 100,000X vertical amplification is shown with a scanned area of 5.7 x 3.8 pm2. An arrow indicates the same structural detail observed in (a) ,and (b). The circle shows a protrusion corresponding to the tip estimated maximum. In order to recognize it, we observed the tip structure in a 20 kV SEM.
Figure 2(a) shows a SEM image at 35,000X amplification. Tip is blunt with protrusions as found in STM images. The lack of depth profile definition and focussing problems of the SEMI makes difficult a complete identification of the tip. In figure 2(b), a 500X amplification SEM image with grain structure at tip's shank is shown. This corresponds to a weaker electrochemical etching than the tip. Fig. 2(c) is a 3,500X amplification, where we can observe clear zones out of focus in the vicinity of the grain boundaries and over the grains. We can see more active etching at the grain boundaries, weaker in the bodiest grain and even weaker in spots identified as whiskers or protrusions which have a better grain quality, so they are more stable against etching.
In figure 3, we made a slower STM scanning in order to have a better characterization of the tip. Here a general corrugation can be observed on all of it. This has been reported /13/ for a similar etching of a tungsten wire as seen with SEM. Long protrusions can be observed as well as isolated ones in the middle of a crater. Explanation of these protrusions could be related with electrochemical etching and grain structure of tungsten wire. These protrusions could have mechanical resistance to deformation as big as the whiskers have. For an iron whisker this is 50 times bigger than industrial steel. Indeed, long protrusion could be a whisker lying on the tip surface.
The mechanical resistance of working tips for STM is perhaps due to the selective electrochemical etching which produces only crystal protrusions. An x-ray analysis could be useful to establish this.
There is a point that becomes necessary to discuss in relation to tip convolution. For our particular case the convolution could bring an overestimation of the tip apex as well as introduce artifacts in the STM profiles.
In order to estimate the effect of tip convolution in this case we consider a simple model in which each tip has a spherical geometry at the extreme with radius of curvature RI and R2 respectively for probe and sample. We assume according to the observations of optical microscopy that probe tip has a
Figure 1. STM consecutive tip observations at different conditions. (a) Tip at a low vertical amplification (10,00OX), the scanned area is 8.8 x 5.4 pm2.
(b) Deeper image of the same region, the vertical amplification has been doubled, the circle shows the apparent highest point. (c) Higher amplification
(100,000X) showing szructural surface profiles not evident in (a), the scanned area is 5.7 X 3.8 pm
.
The arrow indicates the same structural detail observed in (a), (b) and (c), vertical calibration is shown in each image.Figure 2. SEM observations. (a) 35,000X amplification SEM image of the tip structure and its surroundings, it represents approximately the same region as in figure l(c). (b) 500X amplification showing grain structure at tip's shank.
(c) Amplification (3,500X) of the central part of figure 2 (b)
.
Figure 3. A very slow STM scanning brings a better characterization of the tip. This STM image shows a high corrugation and long protrusions on all the tip.2Similar structures had been reported using SEM. The scanned area is 5.8 x 4 urn ; both, X and Z, calibration is shown.
radius of curvature of 0.1 jum and that the sample tip is about 10 urn.
According to figure 4 the position of the point of nearest approach will not be the same for the on-axis and off-axis position. For a lateral displacement (x) of the probe tip of 1 /nm the point of nearest approach will be shifted about 100 A lateral (Ax) and 5 A vertical (Az) assuming a separation between1- sample and probe (d) of 30 A.
For the particular study we presented here, tip convolution is smaller than the particular resolution shown in figures 1 and 3; this also explain the reproducibility of small details in different images at this scale. At higher amplifications a careful analysis is needed to consider tip convolution.
4.- CONCLUSIONS
We show it is possible to get long scan STM images in the range of optical and sub-optical scale. We apply this for electrochemical etching tip characterization. Corrugated structure with protrusions all over the tip were observed, any of these could be a working tip. Their mechanical resistance could be explained in terms of selective electrochemical etching for tungsten.
Long scan STM could be a complementary technique to SEM and optical microscopy.
Figure 4. Scheme of a simple model in which probe and sample tip have a spherical geometry with radius of curvature R1 and R2 respectively. The separation between sample and probe (d) is kept constant in the constant current mode hence the position of the point of nearest approach will be shifted from the on-axis (dotted line) to the off-axis position. For a lateral d-isplacement (X) of the probe tip, Ax and Az are the lateral and vertical shift of the point of nearest approach.
ACKNOWLEDGEMENT
This work was partially supported for the CONACYT-Mexico and for the American Organization of States. We gratefully acknowledge to the British Council at Mexico City for the provided support to attend this Symposium.
REFERENCES
/l/ Binnig, G., Rohrer, H., Gerber, Ch. and Weibel, E., Phys. Rev. Lett.
=(l), 57 (1982).
/2/ Baratoff, A., ~innig, G., Fuchs, H., Salvan, F. and Stoll, E., Surf. Sci.
168, 734 (1986).
/3/ Fink, H. W., IBM J. Res. Dev. 30(5), 460 (1986).
/4/ Hansma, P. K. and Tersoff, J., J. Appl. Phys. =(2), X1 (1987).
/5/ Bielgelsen, D. K., Ponce, F. A. and Tramontana, J. C., Appl. Phys. Lett.
50 (11)
,
696 (1987)./6/
Kuk, Y. and Silverman, P., Appl. Phys. Lett. u(23),
1597 (1986)./7/ Coton, R. J., Baker, S. M., Driscoll, R. J., Youngquist, M.
G.,Baldeschwider, J. D. and Kaiser, W. J., J. Vac. Sci. Tech. &(2), 349 (1988)
.
/8/ Tiedje, T., Varon, J., Deckman, H. and Stokes, J., J. Vac. Sci. Tech.
&5(2), 372 (1988).
/9/ van de Walle, G. F. A., van Kempen, H. and Wyder. P., Surf. Sci. 167, L219 (1986).
/10/ Gimzewski, J. K. and Moller, R., Phys. Rev. =(2), 1284 (1987).
/11/ Garcia Cantu, R. and Huerta Garnica, M. A., (STMf89 Conference).
/12/ Garcia Cantu, R. and Huerta Garnica, M. A., Surf. Sci. 181, 216 (1987).
/13/ Swanson, L. W. and Li, J. Z., J. Vac. Sci. Tech. &(4), 1062 (1987).
