- Email: [email protected]
Cleaning AFM colloidal probes by mechanically scrubbing with supersharp “brushes”
ARTICLE IN PRESS Ultramicroscopy 109 (2009) 1061–1065
Contents lists available at ScienceDirect
Ultramicroscopy journal homepage: www.elsevier.com/locate/ultramic
Cleaning AFM colloidal probes by mechanically scrubbing with supersharp ‘‘brushes’’ Yang Gan a,, George V. Franks b a b
School of Chemical Engineering and Technology, Harbin Institute of Technology, Harbin, Heilongjiang 150001, China Department of Chemical and Biomolecular Engineering, University of Melbourne, VIC 3010, Australia
a r t i c l e in fo
Keywords: AFM Grating Contaminant Colloidal probe Clean
abstract Contamination control of atomic force microscope (AFM) tips (including standard but supersharp imaging tips and particle/colloidal probes) is very important for reliable AFM imaging and surface/ interface force measurements. Traditional cleaning methods such as plasma, UV–ozone and solvent treatments have their shortcomings. Here, we demonstrate that calibration gratings with supersharp spikes can be employed to scrub away contaminants accumulated on a colloidal sphere probe by scanning the probe against the spikes at high load at constant-force mode. The present method is superior to traditional cleaning methods in several aspects. First, accumulated lump-like organic/ inorganic material can be removed; second, removal is non-destructive and highly efficient based on a ‘‘targeted removal’’ strategy; third, removal and probe shape/morphology study can be completed in a single step (we report, to our best knowledge, the first evidence of the wear of the colloidal sphere during force measurements); and fourth, both colloidal/particle probes and standard but supersharp AFM imaging tips can be treated. & 2009 Elsevier B.V. All rights reserved.
1. Introduction Nowadays, atomic force microscope (AFM) is a key technique for surface morphology characterization and surface/interfacial force studies. Contamination control of the AFM tips is important for reliable AFM imaging and force measurements, but has often been overlooked by many. Frequently one finds that tips can become contaminated during storage at ambient conditions with inorganic/organic matters [1], or when exposed in ambient conditions for a short period, or when in use by hard contact with a sample during AFM imaging or force measurements [2]. Contaminants, even only a few monolayers thick, can significantly alter the surface chemistry of the tip [3]. Moreover, accumulated contaminants change the tip geometry, resulting in distorted AFM topographs [4]. (Various methods for determining the profile of AFM tips have been reviewed recently in an excellent article by Yacoot and Koenders [5].) Several cleaning methods, such as using plasma [3,6–9], UV–ozone [10,11] and various solvents [1,12,13] treatments, have been proposed and applied to remove contaminants from the AFM tip. However, these methods have their shortcomings. First, generally a thin layer of organic contaminants instead of lump materials can be removed; second, reflective metallic materials such as Au and Al may be damaged or removed
after harsh or prolonged treatment [1]; third, all the above methods aim to clean the whole probe instead of the tip or the desired area on the tip; and last but the most important, after cleaning, the conventional cleaning procedures require ex-situ techniques, such as electron microscope, to study the surface morphology to check whether the tip is indeed free of contaminants. Several groups [14–23] demonstrated beautifully that grating with supersharp spikes [23] is a powerful tool to characterize both routine AFM tips and particle/colloidal tips: both the tip’s radius and surface roughness can be determined by scanning the probe over the grating to obtain a reversed image because the tips of spikes are sharper than the routine AFM tips and spheres. Here, we extend the application of the grating with supersharp spikes to remove contaminants from AFM tips. We demonstrate that grating can be used as a ‘‘brush’’ to scrub away accumulated contaminants on the tip that are not easily removed by traditional methods. We also discuss the other advantages of our method and compared with the method developed by Nie and McIntyre [24].
2. Experimental 2.1. AFM and probes
Corresponding author.
E-mail address: [email protected] (Y. Gan). 0304-3991/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ultramic.2009.03.019
A Nanoscope III, MultiMode AFM having a liquid cell (Veeco Digital Instruments Inc., Santa Babara, CA, USA) was used.
ARTICLE IN PRESS 1062
Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
Standard triangular-shaped silicon nitride cantilevers with pyramidal tip were used. The long wide cantilever of a nominal spring constant of 0.12 N/m was used for force measurements. All measurements were performed at ambient conditions. A silica sphere (Bangs Laboratories, Inc., Fishers, IN, USA) of 5 mm in diameter was glued to the tip of the cantilever with a twocomponent epoxy resin (Selleys, Australia) following the cantilever-moving method described in Ref. [25].
2.2. Grating with supersharp spikes A silicon calibration grating with supersharp spikes was used to reversely image the silica sphere and employed as a brush to remove adhered contaminants from the sphere. The nominal tip radius of spikes is less than 10 nm and the tip angle is less than 201 (model TGT1, ND-MDT Co., Moscow, Russia). The spikes form a 2.12 mm pitch with the diagonal distance of 3 mm. The tip height ranges from 600 to 800 nm. The grating occupies a 4 mm2 surface area. There are thus about 400,000 spikes on the grating. A SEM micrograph of the grating is shown in Fig. 1 [23]. The surface morphology of the freshly prepared colloidal probe was checked by scanning over the grating as in Ref. [14] to obtain a reversed image of the sphere to ensure that the surface was free of any accumulated material.
2.3. Substrate solution and force measurements Oxidized silicon wafer samples were put into a UV/ozone cleaner for 10 min followed by 10% NaOH etching for 20 min, then rinsed with copious Milli-Q water. Root mean square roughness of substrate was 0.1 nm over a 1 mm2 area. The concentrated KCl solution was prepared with ACS-grade reagents. Deionized water filtered with a Milli-Q water system was used throughout. A clean glass syringe was used to inject solution into the AFM liquid cell; a 0.1 mm pore-sized nylon syringe filter (GE Osmonics, Inc., Chicago, IL, USA) was placed between the syringe and the inlet of the AFM liquid cell to filter out larger airborne particulate matters. The salt solution of 10 ml was circulated through the space formed by the liquid cell, substrate and O-ring. Allowing 10–15 min for the system to become stable, the colloidal probe was engaged onto the substrate (oxidized silicon) to capture a number of force
Fig. 1. SEM micrograph of Si grating with supersharp spikes. Reproduced from Ref. [23].
curves. Force measurements were repeated on various locations on the substrate. 2.4. Contaminant removal Contamination of the probe can be determined to have occurred when the force curves measured change from well behaved and reproducible to erratic with features typical of contamination. Typical indicators of contamination include compressible regions upon contact during approach and longrange adhesion on retraction. When force curves were found to show these abrupt and persisting changes that were characteristic of contamination, the force measurements were stopped. The suspected contaminated probe was taken out from the liquid cell, rinsed with Milli-Q water, blow dried with high-purity nitrogen gas. (We found that rinsing and drying did not remove contaminants.) Then the probe was placed again into the AFM. Note that contaminants removal is completed in air. This time, the probe was scanned in the constant-force mode against the grating to first obtain the pristine surface topography of the sphere at low force (usually a few nano-newtons in order not to destruct and remove the contaminants). As soon as any accumulated material was located on the tip, the scanning was continued but at a higher force (usually tens of nano-newtons) to scrub the sphere harder against the spikes until the contaminating material disappeared. In order to make sure that contaminants were indeed removed, we captured another image by varying the slow scan direction once from 01 to 901. Furthermore, the cleaning process was repeated with a second spike that had not touched the sphere before in order to minimize the risk that the removed contaminants re-accumulated from the same spike after repeated scanning. A UV cleaner (BioForce Nanosciences, Inc., IA, USA) was also used to clean AFM probes in this study. The peak intensities of the UV lamp were at wavelengths of 185 and 254 nm. The UV light intensity is slightly below 20 mW/cm2 at the 254 nm wavelength. The distance between the lamp and the probe was less than 5 mm. The exposure time was 10 min.
3. Results Fig. 2a shows a typical force curve captured when the colloidal sphere on a colloidal probe has been contaminated after the sphere has repeatedly been in contact with the substrate in concentrated KCl solution. Note that on the approaching curve, at a close separation of 5 nm, a compressible barrier – instead of a sharp jump-in point, which should be observed in a good force curve – appeared before the probe, which was in hard contact with the substrate. This barrier persisted on force curves obtained at several different locations on the same substrate, indicating that the contamination was attached to the probe. The contaminated sphere was first checked by scanning at low force against the grating with sharp spikes to obtain the surface morphology (reversed imaging) of the silica sphere. (Note that the topography obtained is only part of the sphere that faces the substrate during force measurements.) As expected, a lump of material of about 100–150 nm in size was found to be lying just off-centre from the contact area (Fig. 3a). It is worth noting that an area at the centre of the image appeared rather smooth. This smooth area was caused by wearing of the sphere against the substrate during capturing force curves because relative sliding occurred when the sphere and substrate were in hard contact on both the approaching and retracting runs. We believe that this was the first report of wearing of a colloidal sphere during force measurements. The change in morphology of the surface of a colloidal sphere will affect the contact area and
ARTICLE IN PRESS Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
1063
may significantly influence the short-range interfacial forces, particularly the accuracy of adhesion measurements. Further studies of the wearing process and the influence of sphere wearing on the adhesion properties are underway. The lumpy contamination was successfully removed by repeatedly scanning the probe against the grating at higher force. Fig. 3b shows that the lump material disappeared whereas the surface morphology of the sphere, including the smooth worn part of the sphere, remained unchanged after repeated scanning. Then the probe was reused to capture force curves against a new substrate in the concentrated salt solution. Fig. 2b clearly shows that a sharp jump-in can be observed at a separation of 8 nm on the approaching curve (a plateau in jump-off point is only specific for this case).
4. Discussion In this section, we will discuss the removal process, and the advantages of the present method compared with traditional methods. We note that Nie and McIntyre developed recently a novel non-destructive cleaning method by using a porous polymer film (a biaxially oriented polypropylene (BOPP) film) to scrub away contaminants [24]. The merits of two methods are discussed. 4.1. The removal process Removal of contaminants using the supersharp grating is accomplished via a mechanical scrubbing action. Fig. 4 illustrates how the attached contamination on the sphere is scrubbed away mechanically by scanning against the spikes at high force. With Nie and McIntyre’s method [24], the contaminated tip is cleaned by pushing the tip into the BOPP film for several times. 4.2. The ability to remove lump contamination
Fig. 2. Typical force curves obtained in concentrated 4 M KCl solution using a contaminated (upper graph) and cleaned probe (lower graph). The approaching and retracting force curves are indicated by arrows in two graphs.
UV/ozone treatment works by oxidizing organic matters to volatilize only small molecules. We also tried but failed to remove lumpy contaminants as shown in Fig. 3a by UV/ozone treatments. UV/ozone treatment can be very efficient for removing thin (a few molecular layers) organic materials, but performance is poor with
Fig. 3. The AFM topograph (reversed image) of the sphere before (left) and after (right) being cleaned by scrubbing against the grating. The lumpy materials are attached contaminants.
ARTICLE IN PRESS 1064
Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
Fig. 4. The mechanical removal of the attached material on the sphere by scanning against the spikes at high force: (a) the contaminants are to be removed and (b) the contaminants are scrubbed down by the spikes.
bulk contaminants like those lumpy ones shown in Fig. 3a. It is probable that some organic contaminants remained on the tip even after being mechanically scrubbed for two reasons. First, organics might present as a molecular thin film that was too thin to be removed mechanically. Second, a spike was not infinitely sharp so that some areas on the surface could not be accessed by the spike. However, we did not find significant changes in both morphology and force curves after further UV/ozone treatment. This means that organic contaminants were minor in amount. We did not analyze the chemical composition of the contaminants – they might be organic or inorganic matters or even a mixture. UV/ozone, plasma, oxidant solvent treatments can only oxidize organic matters. The present scrubbing method appears especially appealing as both inorganic and organic materials can be removed. 4.3. Non-destructive, ‘‘targeted removal’’ method Other cleaning methods, as discussed in Section 1, may alter the structure or even damage the probe since the whole probe is exposed to a strong oxidizing environment. The reflective metallic layer on the back of the cantilever might be damaged [1]. Particularly for colloidal probes, the resin (which was used to glue the sphere to the cantilever) might be etched, resulting in weakened bonding between the sphere and the cantilever. As distinct from the ‘‘overall removal’’ cleaning route of the traditional method, the present method acts as a ‘‘targeted removal’’ strategy. Only the contaminants on the contact surface area of the sphere, instead of the whole sphere or the whole probe, are cleaned by applying mechanical force. The cleaning efficiency is thus very high and the cleaning process is completely nondestructive as evidenced from Fig. 3b. With Nie and McIntyre’s method [24], the cleaning process is also non-destructive. However, the present method is unique because scanning with supersharp spikes can identify the contaminants unambiguously and clearly. 4.4. Cleaning and characterization completed in a single step With traditional cleaning methods, the probe has to be checked with other techniques such as electron microscope [26] or contact angle measurements [1,9] to check whether the probe is free of contaminants. On the contrary, with the present method, the grating is used as both a contamination removal tool and a characterization tool to study the cleaning results. Therefore, the
present method is more efficient than traditional methods. Furthermore, because there are 400,000 spikes on a grating, the grating can be used for many times as it will be very rare for a new probe to touch the used spikes. The present method is thus very cost effective. With Nie and McIntyre’s method [24], cleaning and characterization can also be completed in a single step – cleaning by pushing the tip into the film and characterization by scanning over the film to monitor the morphology change.
4.5. Suitable for both colloidal probes and routine AFM probes Though we applied the present method only to colloidal probes, there is no reason why this method cannot also be used to treat routine AFM probes because the tip radius of spikes on the grating is less than 10 nm – sharper than routine AFM probes (usually with a tip radius of 20–50 nm). Due to the small pore size of film used by Nie and McIntyre [24], it will be difficult to treat colloidal probes with micron-sized spheres with their method.
5. Conclusions We demonstrated that calibration gratings with supersharp spikes can be employed to remove contaminants attached to a colloidal probe. The contaminants are scrubbed away mechanically by scanning the probe against the spikes at high load in constant-force mode with the AFM. The present method is superior to traditional contaminants removal methods in several aspects: (1) Lumpy organic/inorganic material can be removed. (2) Removal is non-destructive and highly efficient based on a ‘‘targeted removal’’ strategy. (3) Removal and morphology study can be completed in a single step. (4) Both colloidal/particle probes and standard AFM imaging can be treated.
Acknowledgements Y.G. is supported by Overseas Talent Attraction Program of HIT (GFCQ18600003) and sponsored by SRF for ROCS. The authors thank Erica J. Wanless of the University of Newcastle for support.
ARTICLE IN PRESS Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
References [1] Y.S. Lo, N.D. Huefner, W.S. Chan, P. Dryden, B. Hagenhoff, J. Beebe Jr., Langmuir 15 (1999) 6522. [2] M. Enachescu, R.W. Carpick, D.F. Ogletree, M. Salmeron, Journal of Applied Physics 95 (2004) 7694. [3] L. Sirghi, O. Kylian, D. Gilliland, G. Ceccone, F. Rossi, Journal of Physical Chemistry B 110 (2006) 25975. [4] H.Y. Nie, N.S. McIntyre, Review of Scientific Instruments 78 (2007) 023701. [5] A. Yacoot, L. Koenders, Journal of Physics D: Applied Physics 41 (2008) 103001. [6] G.U. Lee, D.A. Kidwell, R.J. Colton, Langmuir 10 (1994) 354. [7] T.J. Senden, C.J. Drummond, Colloids and Surfaces A 94 (1995) 29. [8] A. Feiler, I. Larson, P. Jenkins, P. Attard, Langmuir 16 (2000) 10269. [9] E. Bonaccurso, G. Gillies, Langmuir 20 (2004) 11824. [10] T. Arai, M. Tomitori, Applied Physics A 66 (1998) 319. [11] M. Fujihira, Y. Okabe, Y. Tani, M. Furugori, U. Akiba, Ultramicroscopy 82 (2000) 181. [12] P. Hinterdorfer, W. Baumgartner, H.J. Gruber, K. Schilcher, H. Schindler, Proceedings of the National Academy of Sciences of USA 93 (1996) 3477. [13] M. Tomitori, T. Arai, Applied Surface Science 140 (1999) 432.
[14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26]
1065
C. Neto, V.S.J. Craig, Langmuir 17 (2001) 2097. D.L. Malotky, M.K. Chaudhury, Langmuir 17 (2001) 7823. V.N. Konopsky, Optics Communications 185 (2000) 83. A. Drechsler, K. Grundke, Colloids and Surfaces A 264 (2005) 157. M.J. Bunker, M.C. Davies, X. Chen, M.B. James, C.J. Roberts, European Journal of Pharmaceutical Sciences 29 (2006) 405. J.C. Hooton, C.S. German, S. Allen, M.C. Davies, C.J. Roberts, S.J.B. Tendler, P.M. Williams, Pharmaceutical Research 20 (2002) 508. M.J. Bunker, C.J. Roberts, M.C. Davies, M.B. James, International Journal of Pharmaceutics 325 (2006) 163. S. Robert, B. Bharat, D.H. Bryan, J.K. Andrzej, G. Gerard, Journal of Applied Physics 99 (2006) 014310. T. Svaldo Lanero, O. Cavalleri, S. Krol, R. Rolandi, A. Gliozzi, Journal of Biotechnology 124 (2006) 723. V. Bykov, A. Gologanov, V. Shevyakov, Applied Physics A 66 (1998) 499. H.-Y. Nie, N.S. McIntyre, Langmuir 17 (2001) 432. Y. Gan, Review of Scientific Instruments 78 (2007) 081101. B. Ska˚rman, L.R. Wallenberg, S.N. Jacobsen, U. Helmersson, C. Thelander, Langmuir 16 (2000) 6267.
Contents lists available at ScienceDirect
Ultramicroscopy journal homepage: www.elsevier.com/locate/ultramic
Cleaning AFM colloidal probes by mechanically scrubbing with supersharp ‘‘brushes’’ Yang Gan a,, George V. Franks b a b
School of Chemical Engineering and Technology, Harbin Institute of Technology, Harbin, Heilongjiang 150001, China Department of Chemical and Biomolecular Engineering, University of Melbourne, VIC 3010, Australia
a r t i c l e in fo
Keywords: AFM Grating Contaminant Colloidal probe Clean
abstract Contamination control of atomic force microscope (AFM) tips (including standard but supersharp imaging tips and particle/colloidal probes) is very important for reliable AFM imaging and surface/ interface force measurements. Traditional cleaning methods such as plasma, UV–ozone and solvent treatments have their shortcomings. Here, we demonstrate that calibration gratings with supersharp spikes can be employed to scrub away contaminants accumulated on a colloidal sphere probe by scanning the probe against the spikes at high load at constant-force mode. The present method is superior to traditional cleaning methods in several aspects. First, accumulated lump-like organic/ inorganic material can be removed; second, removal is non-destructive and highly efficient based on a ‘‘targeted removal’’ strategy; third, removal and probe shape/morphology study can be completed in a single step (we report, to our best knowledge, the first evidence of the wear of the colloidal sphere during force measurements); and fourth, both colloidal/particle probes and standard but supersharp AFM imaging tips can be treated. & 2009 Elsevier B.V. All rights reserved.
1. Introduction Nowadays, atomic force microscope (AFM) is a key technique for surface morphology characterization and surface/interfacial force studies. Contamination control of the AFM tips is important for reliable AFM imaging and force measurements, but has often been overlooked by many. Frequently one finds that tips can become contaminated during storage at ambient conditions with inorganic/organic matters [1], or when exposed in ambient conditions for a short period, or when in use by hard contact with a sample during AFM imaging or force measurements [2]. Contaminants, even only a few monolayers thick, can significantly alter the surface chemistry of the tip [3]. Moreover, accumulated contaminants change the tip geometry, resulting in distorted AFM topographs [4]. (Various methods for determining the profile of AFM tips have been reviewed recently in an excellent article by Yacoot and Koenders [5].) Several cleaning methods, such as using plasma [3,6–9], UV–ozone [10,11] and various solvents [1,12,13] treatments, have been proposed and applied to remove contaminants from the AFM tip. However, these methods have their shortcomings. First, generally a thin layer of organic contaminants instead of lump materials can be removed; second, reflective metallic materials such as Au and Al may be damaged or removed
after harsh or prolonged treatment [1]; third, all the above methods aim to clean the whole probe instead of the tip or the desired area on the tip; and last but the most important, after cleaning, the conventional cleaning procedures require ex-situ techniques, such as electron microscope, to study the surface morphology to check whether the tip is indeed free of contaminants. Several groups [14–23] demonstrated beautifully that grating with supersharp spikes [23] is a powerful tool to characterize both routine AFM tips and particle/colloidal tips: both the tip’s radius and surface roughness can be determined by scanning the probe over the grating to obtain a reversed image because the tips of spikes are sharper than the routine AFM tips and spheres. Here, we extend the application of the grating with supersharp spikes to remove contaminants from AFM tips. We demonstrate that grating can be used as a ‘‘brush’’ to scrub away accumulated contaminants on the tip that are not easily removed by traditional methods. We also discuss the other advantages of our method and compared with the method developed by Nie and McIntyre [24].
2. Experimental 2.1. AFM and probes
Corresponding author.
E-mail address: [email protected] (Y. Gan). 0304-3991/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ultramic.2009.03.019
A Nanoscope III, MultiMode AFM having a liquid cell (Veeco Digital Instruments Inc., Santa Babara, CA, USA) was used.
ARTICLE IN PRESS 1062
Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
Standard triangular-shaped silicon nitride cantilevers with pyramidal tip were used. The long wide cantilever of a nominal spring constant of 0.12 N/m was used for force measurements. All measurements were performed at ambient conditions. A silica sphere (Bangs Laboratories, Inc., Fishers, IN, USA) of 5 mm in diameter was glued to the tip of the cantilever with a twocomponent epoxy resin (Selleys, Australia) following the cantilever-moving method described in Ref. [25].
2.2. Grating with supersharp spikes A silicon calibration grating with supersharp spikes was used to reversely image the silica sphere and employed as a brush to remove adhered contaminants from the sphere. The nominal tip radius of spikes is less than 10 nm and the tip angle is less than 201 (model TGT1, ND-MDT Co., Moscow, Russia). The spikes form a 2.12 mm pitch with the diagonal distance of 3 mm. The tip height ranges from 600 to 800 nm. The grating occupies a 4 mm2 surface area. There are thus about 400,000 spikes on the grating. A SEM micrograph of the grating is shown in Fig. 1 [23]. The surface morphology of the freshly prepared colloidal probe was checked by scanning over the grating as in Ref. [14] to obtain a reversed image of the sphere to ensure that the surface was free of any accumulated material.
2.3. Substrate solution and force measurements Oxidized silicon wafer samples were put into a UV/ozone cleaner for 10 min followed by 10% NaOH etching for 20 min, then rinsed with copious Milli-Q water. Root mean square roughness of substrate was 0.1 nm over a 1 mm2 area. The concentrated KCl solution was prepared with ACS-grade reagents. Deionized water filtered with a Milli-Q water system was used throughout. A clean glass syringe was used to inject solution into the AFM liquid cell; a 0.1 mm pore-sized nylon syringe filter (GE Osmonics, Inc., Chicago, IL, USA) was placed between the syringe and the inlet of the AFM liquid cell to filter out larger airborne particulate matters. The salt solution of 10 ml was circulated through the space formed by the liquid cell, substrate and O-ring. Allowing 10–15 min for the system to become stable, the colloidal probe was engaged onto the substrate (oxidized silicon) to capture a number of force
Fig. 1. SEM micrograph of Si grating with supersharp spikes. Reproduced from Ref. [23].
curves. Force measurements were repeated on various locations on the substrate. 2.4. Contaminant removal Contamination of the probe can be determined to have occurred when the force curves measured change from well behaved and reproducible to erratic with features typical of contamination. Typical indicators of contamination include compressible regions upon contact during approach and longrange adhesion on retraction. When force curves were found to show these abrupt and persisting changes that were characteristic of contamination, the force measurements were stopped. The suspected contaminated probe was taken out from the liquid cell, rinsed with Milli-Q water, blow dried with high-purity nitrogen gas. (We found that rinsing and drying did not remove contaminants.) Then the probe was placed again into the AFM. Note that contaminants removal is completed in air. This time, the probe was scanned in the constant-force mode against the grating to first obtain the pristine surface topography of the sphere at low force (usually a few nano-newtons in order not to destruct and remove the contaminants). As soon as any accumulated material was located on the tip, the scanning was continued but at a higher force (usually tens of nano-newtons) to scrub the sphere harder against the spikes until the contaminating material disappeared. In order to make sure that contaminants were indeed removed, we captured another image by varying the slow scan direction once from 01 to 901. Furthermore, the cleaning process was repeated with a second spike that had not touched the sphere before in order to minimize the risk that the removed contaminants re-accumulated from the same spike after repeated scanning. A UV cleaner (BioForce Nanosciences, Inc., IA, USA) was also used to clean AFM probes in this study. The peak intensities of the UV lamp were at wavelengths of 185 and 254 nm. The UV light intensity is slightly below 20 mW/cm2 at the 254 nm wavelength. The distance between the lamp and the probe was less than 5 mm. The exposure time was 10 min.
3. Results Fig. 2a shows a typical force curve captured when the colloidal sphere on a colloidal probe has been contaminated after the sphere has repeatedly been in contact with the substrate in concentrated KCl solution. Note that on the approaching curve, at a close separation of 5 nm, a compressible barrier – instead of a sharp jump-in point, which should be observed in a good force curve – appeared before the probe, which was in hard contact with the substrate. This barrier persisted on force curves obtained at several different locations on the same substrate, indicating that the contamination was attached to the probe. The contaminated sphere was first checked by scanning at low force against the grating with sharp spikes to obtain the surface morphology (reversed imaging) of the silica sphere. (Note that the topography obtained is only part of the sphere that faces the substrate during force measurements.) As expected, a lump of material of about 100–150 nm in size was found to be lying just off-centre from the contact area (Fig. 3a). It is worth noting that an area at the centre of the image appeared rather smooth. This smooth area was caused by wearing of the sphere against the substrate during capturing force curves because relative sliding occurred when the sphere and substrate were in hard contact on both the approaching and retracting runs. We believe that this was the first report of wearing of a colloidal sphere during force measurements. The change in morphology of the surface of a colloidal sphere will affect the contact area and
ARTICLE IN PRESS Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
1063
may significantly influence the short-range interfacial forces, particularly the accuracy of adhesion measurements. Further studies of the wearing process and the influence of sphere wearing on the adhesion properties are underway. The lumpy contamination was successfully removed by repeatedly scanning the probe against the grating at higher force. Fig. 3b shows that the lump material disappeared whereas the surface morphology of the sphere, including the smooth worn part of the sphere, remained unchanged after repeated scanning. Then the probe was reused to capture force curves against a new substrate in the concentrated salt solution. Fig. 2b clearly shows that a sharp jump-in can be observed at a separation of 8 nm on the approaching curve (a plateau in jump-off point is only specific for this case).
4. Discussion In this section, we will discuss the removal process, and the advantages of the present method compared with traditional methods. We note that Nie and McIntyre developed recently a novel non-destructive cleaning method by using a porous polymer film (a biaxially oriented polypropylene (BOPP) film) to scrub away contaminants [24]. The merits of two methods are discussed. 4.1. The removal process Removal of contaminants using the supersharp grating is accomplished via a mechanical scrubbing action. Fig. 4 illustrates how the attached contamination on the sphere is scrubbed away mechanically by scanning against the spikes at high force. With Nie and McIntyre’s method [24], the contaminated tip is cleaned by pushing the tip into the BOPP film for several times. 4.2. The ability to remove lump contamination
Fig. 2. Typical force curves obtained in concentrated 4 M KCl solution using a contaminated (upper graph) and cleaned probe (lower graph). The approaching and retracting force curves are indicated by arrows in two graphs.
UV/ozone treatment works by oxidizing organic matters to volatilize only small molecules. We also tried but failed to remove lumpy contaminants as shown in Fig. 3a by UV/ozone treatments. UV/ozone treatment can be very efficient for removing thin (a few molecular layers) organic materials, but performance is poor with
Fig. 3. The AFM topograph (reversed image) of the sphere before (left) and after (right) being cleaned by scrubbing against the grating. The lumpy materials are attached contaminants.
ARTICLE IN PRESS 1064
Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
Fig. 4. The mechanical removal of the attached material on the sphere by scanning against the spikes at high force: (a) the contaminants are to be removed and (b) the contaminants are scrubbed down by the spikes.
bulk contaminants like those lumpy ones shown in Fig. 3a. It is probable that some organic contaminants remained on the tip even after being mechanically scrubbed for two reasons. First, organics might present as a molecular thin film that was too thin to be removed mechanically. Second, a spike was not infinitely sharp so that some areas on the surface could not be accessed by the spike. However, we did not find significant changes in both morphology and force curves after further UV/ozone treatment. This means that organic contaminants were minor in amount. We did not analyze the chemical composition of the contaminants – they might be organic or inorganic matters or even a mixture. UV/ozone, plasma, oxidant solvent treatments can only oxidize organic matters. The present scrubbing method appears especially appealing as both inorganic and organic materials can be removed. 4.3. Non-destructive, ‘‘targeted removal’’ method Other cleaning methods, as discussed in Section 1, may alter the structure or even damage the probe since the whole probe is exposed to a strong oxidizing environment. The reflective metallic layer on the back of the cantilever might be damaged [1]. Particularly for colloidal probes, the resin (which was used to glue the sphere to the cantilever) might be etched, resulting in weakened bonding between the sphere and the cantilever. As distinct from the ‘‘overall removal’’ cleaning route of the traditional method, the present method acts as a ‘‘targeted removal’’ strategy. Only the contaminants on the contact surface area of the sphere, instead of the whole sphere or the whole probe, are cleaned by applying mechanical force. The cleaning efficiency is thus very high and the cleaning process is completely nondestructive as evidenced from Fig. 3b. With Nie and McIntyre’s method [24], the cleaning process is also non-destructive. However, the present method is unique because scanning with supersharp spikes can identify the contaminants unambiguously and clearly. 4.4. Cleaning and characterization completed in a single step With traditional cleaning methods, the probe has to be checked with other techniques such as electron microscope [26] or contact angle measurements [1,9] to check whether the probe is free of contaminants. On the contrary, with the present method, the grating is used as both a contamination removal tool and a characterization tool to study the cleaning results. Therefore, the
present method is more efficient than traditional methods. Furthermore, because there are 400,000 spikes on a grating, the grating can be used for many times as it will be very rare for a new probe to touch the used spikes. The present method is thus very cost effective. With Nie and McIntyre’s method [24], cleaning and characterization can also be completed in a single step – cleaning by pushing the tip into the film and characterization by scanning over the film to monitor the morphology change.
4.5. Suitable for both colloidal probes and routine AFM probes Though we applied the present method only to colloidal probes, there is no reason why this method cannot also be used to treat routine AFM probes because the tip radius of spikes on the grating is less than 10 nm – sharper than routine AFM probes (usually with a tip radius of 20–50 nm). Due to the small pore size of film used by Nie and McIntyre [24], it will be difficult to treat colloidal probes with micron-sized spheres with their method.
5. Conclusions We demonstrated that calibration gratings with supersharp spikes can be employed to remove contaminants attached to a colloidal probe. The contaminants are scrubbed away mechanically by scanning the probe against the spikes at high load in constant-force mode with the AFM. The present method is superior to traditional contaminants removal methods in several aspects: (1) Lumpy organic/inorganic material can be removed. (2) Removal is non-destructive and highly efficient based on a ‘‘targeted removal’’ strategy. (3) Removal and morphology study can be completed in a single step. (4) Both colloidal/particle probes and standard AFM imaging can be treated.
Acknowledgements Y.G. is supported by Overseas Talent Attraction Program of HIT (GFCQ18600003) and sponsored by SRF for ROCS. The authors thank Erica J. Wanless of the University of Newcastle for support.
ARTICLE IN PRESS Y. Gan, G.V. Franks / Ultramicroscopy 109 (2009) 1061–1065
References [1] Y.S. Lo, N.D. Huefner, W.S. Chan, P. Dryden, B. Hagenhoff, J. Beebe Jr., Langmuir 15 (1999) 6522. [2] M. Enachescu, R.W. Carpick, D.F. Ogletree, M. Salmeron, Journal of Applied Physics 95 (2004) 7694. [3] L. Sirghi, O. Kylian, D. Gilliland, G. Ceccone, F. Rossi, Journal of Physical Chemistry B 110 (2006) 25975. [4] H.Y. Nie, N.S. McIntyre, Review of Scientific Instruments 78 (2007) 023701. [5] A. Yacoot, L. Koenders, Journal of Physics D: Applied Physics 41 (2008) 103001. [6] G.U. Lee, D.A. Kidwell, R.J. Colton, Langmuir 10 (1994) 354. [7] T.J. Senden, C.J. Drummond, Colloids and Surfaces A 94 (1995) 29. [8] A. Feiler, I. Larson, P. Jenkins, P. Attard, Langmuir 16 (2000) 10269. [9] E. Bonaccurso, G. Gillies, Langmuir 20 (2004) 11824. [10] T. Arai, M. Tomitori, Applied Physics A 66 (1998) 319. [11] M. Fujihira, Y. Okabe, Y. Tani, M. Furugori, U. Akiba, Ultramicroscopy 82 (2000) 181. [12] P. Hinterdorfer, W. Baumgartner, H.J. Gruber, K. Schilcher, H. Schindler, Proceedings of the National Academy of Sciences of USA 93 (1996) 3477. [13] M. Tomitori, T. Arai, Applied Surface Science 140 (1999) 432.
[14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26]
1065
C. Neto, V.S.J. Craig, Langmuir 17 (2001) 2097. D.L. Malotky, M.K. Chaudhury, Langmuir 17 (2001) 7823. V.N. Konopsky, Optics Communications 185 (2000) 83. A. Drechsler, K. Grundke, Colloids and Surfaces A 264 (2005) 157. M.J. Bunker, M.C. Davies, X. Chen, M.B. James, C.J. Roberts, European Journal of Pharmaceutical Sciences 29 (2006) 405. J.C. Hooton, C.S. German, S. Allen, M.C. Davies, C.J. Roberts, S.J.B. Tendler, P.M. Williams, Pharmaceutical Research 20 (2002) 508. M.J. Bunker, C.J. Roberts, M.C. Davies, M.B. James, International Journal of Pharmaceutics 325 (2006) 163. S. Robert, B. Bharat, D.H. Bryan, J.K. Andrzej, G. Gerard, Journal of Applied Physics 99 (2006) 014310. T. Svaldo Lanero, O. Cavalleri, S. Krol, R. Rolandi, A. Gliozzi, Journal of Biotechnology 124 (2006) 723. V. Bykov, A. Gologanov, V. Shevyakov, Applied Physics A 66 (1998) 499. H.-Y. Nie, N.S. McIntyre, Langmuir 17 (2001) 432. Y. Gan, Review of Scientific Instruments 78 (2007) 081101. B. Ska˚rman, L.R. Wallenberg, S.N. Jacobsen, U. Helmersson, C. Thelander, Langmuir 16 (2000) 6267.