- Email: [email protected]
3D reconstruction of a micro pipette tip
Microelectronic Engineering 86 (2009) 868–870
Contents lists available at ScienceDirect
Microelectronic Engineering journal homepage: www.elsevier.com/locate/mee
3D reconstruction of a micro pipette tip H. Ostadi, M. Malboubi, P.D. Prewett, K. Jiang * Centre for Biomedical and Nanotechnology, Department of Mechanical Engineering, University of Birmingham, Birmingham, B15 2TT, UK
a r t i c l e
i n f o
Article history: Received 26 September 2008 Received in revised form 14 November 2008 Accepted 14 November 2008 Available online 6 December 2008 Keywords: Focused ion beam Patch clamping 3D reconstruction Micro pipette
a b s t r a c t Glass micro pipettes have been widely used for patch clamping. Although many researchers agree that the shape and roundness of pipette tip are important for cell grabbing and gigaseal formation, no numerical values of the geometry have been reported so far. In this research, a FIB/SEM system was used as a nano tomography tool to obtain the 3D shape of a pipette tip. The process involves a cycle of cutting a slice of the pipette using focused ion beam (FIB), taking an SEM image of the new surface, and then milling and imaging again to produce a stack of SEM images. A 3D reconstruction of the pipette was made based on the SEM images. The roundness of the pipette tip was measured and is presented in this paper for the first time. Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction Patch clamp recording is an electrophysiology technique allowing study of ion channels of cells. It is a refinement of the voltage clamp method [1]. It has important applications in cell study and is an important tool to pharmaceutical research. In patch clamping, a glass pipette with a tip diameter of a few microns is pressed against a cell membrane to form a high resistance seal between the inside and outside of the pipette nozzle, known as gigaseal. It is found that achieving a gigaseal is relevant to the tip geometry of the internal wall of the pipette. The physical and chemical mechanisms behind the gigaseal formation are not fully elucidated. It is not only determined by the surface properties of the pipette tip and the cell itself, such as cleanliness, smoothness or roundness but also to a large extend by the skill and patience of the operator. The patch clamp pipette is not the only, but one of the most important factors in gigaseal formation. The tip is the place where the pipette and cell membrane interact with each other. Exact study of this area will help us to better understand the phenomenon of gigaseal formation and may lead us to optimize the shape of the patch clamp pipette [2]. In the 3D reconstruction of a pipette tip, an FEI dual beam Strata 235 focused ion beam (FIB) system was used as a nano tomography tool. The process involves a cycle of milling a slice of the pipette using FIB, taking an SEM image on the new surface, and then milling and imaging again to produce a stack of SEM images. A 3D reconstruction of the pipette was made based on the SEM images.
* Corresponding author. E-mail address: [email protected] (K. Jiang). 0167-9317/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.mee.2008.11.041
Similar approached can be found to be used in many other areas [3–6].
2. Experiment The first step in 3D reconstruction of a micro pipette is obtaining the volumetric information. The dual beam FIB makes this process possible by providing high resolution volumetric information. A micro pipette with a tip diameter of 1.3 lm was placed facing the electron beam. Fig. 1a shows the schematic of pipette, electron beam and ion beam configuration. The angle between I-beam and E-beam was 52°. Therefore, the angle between the imaging plane and the sample was 38°, referring to Fig. 1b. This information was used later for reconstruction. Each slice of the sample was milled off using Ga+ ion beam at 30 Kv and 100 pA for 90 s and dwell time of 1 ls with overlap parameters of 50%. 60 slices with a total thickness of 3 lm were removed and SEM images of the slices taken. The pixel size of the SEM images was 4.5 nm. Fig. 2 shows an image of the 20th slice after milling and its internal edge. (1) Image alignment: We used feature based alignment method [4]. A fixed feature which has not been milled during slicing and not affected by the ion beam is the right side of the pipette in Fig. 1a corresponds to the south part of the pipette in Fig. 2a. We used this part of the pipette to align all of the images. (2) Edge detection: The edge of the internal circle of pipette was detected using Canny algorithm [7]. The basic idea of this algorithm is to detect at the zero-crossing of the second derivative of the smoothed images.
H. Ostadi et al. / Microelectronic Engineering 86 (2009) 868–870
869
It seeks out the zero-crossings of:
o2 ðM IÞ=on2 ¼ oð½oM=on IÞ=on Where M and I are image matrix and unit matrix respectively and n is the direction of the gradient of the smoothed image. This has been done using MATLAB for all of the slices. (3) Back projection: As far as the image plane has a 38° with respect to the sample slices (Fig. 1b) one can calculate the position of each point of the sample slice. Assuming that x (horizontal) and y (vertical) axis are in the image plane and z is the norm of the surface, then Fig. 1. (a) A schematic of pipette, E-beam and I-Beam configuration, (b) a schematic of projected plane and the sample slices. The dark grey shows the projected plane.
Fig. 2. (a) An SEM image of a pipette after milling, (b) the edge of the internal circle of the pipette detected using Canny algorithm.
Fig. 3. (a) A 3D structure of the pipette tip after reconstruction, (b) an isometric view of the pipette and (c) the reconstructed sample in the Y–Z plane.
Fig. 4. (a) A projected image of the 3D pipette tip on the X–Y plane, (b) an image of the first slice and fitted circle. The fitted circle is shown in dashed line.
870
H. Ostadi et al. / Microelectronic Engineering 86 (2009) 868–870
xssp ¼ x yssp ¼ y=cos 38
where indices ‘‘ssp” is for the sample slice position. For zssp, the thickness of the slices are 50 nm so the relative distance between the slices remains 50 nm (or 11 pixels) and the initial angle of the image plane and projected plane is 38°. Fig. 3a shows the 3D structure of the pipette tip reconstructed using MATLAB. The units of X,Y,Z axes are in pixel and each pixel is 4.5 nm. Fig. 3b is the isometric view of the pipette and Fig. 3c shows the Y–Z plane view of the reconstructed image. In order to examine the shape of pipette tip, a perfect circle was fitted to each slice based on the least squared fitted circle method[8] and maximum deviation of the pipette shape from the circle was obtained. Fig. 4a shows the projected image of 3D pipette tip on the X–Y plane for roundness examination. Fig. 4b shows the first slice image and fitted circle. The maximum deviation from fitted circle in Fig. 3b was 43 nm. The average of maximum deviations of all slices was found as 67 nm. This is the first direct roundness measurement of a pipette known to the authors. 3. Conclusions There is a lack of knowledge in measurement of micro pipette tip to understand the gigaseal formation. Using a focused ion beam and SEM system, one can obtain the key parameters of patch clamping such as shape and roundness. Three micrometer long
section of a pipette with the tip diameter of 1.3 lm has been examined and 60 images with 4.5 nm pixel size were taken. The internal walls of the pipette tip, where the contact between the pipette and a cell occurs, were detected using Canny algorithm. After 3D reconstruction of the micro pipette tip in MATLAB, we examined the roundness of each slice with least square fitted circle. The average of maximum deviation of all slices were found as 67 nm, or 10% in roundness error.
References [1] E.R. Kandel, J.H. Schwartz, T.M. Jessell, Principles of Neural Science, fourth ed., McGraw-Hill, New York, 2000. [2] Wee-Liat Ongi, Levent Yobas, Wei-Yi Ong, A missing factor in chip-based patch clamp assay: gigaseal, Journal of Physics 34 (2006) 187–191. [3] R.M. Langford, G. Dale, P.J. Hopkins, P.J.S. Ewen, A.K. Petford-Long, Focused ion beam micromachining of three-dimensional structures and three-dimensional reconstruction to assess their shape, Journal of Micromechanics and Micro Engineering 12 (2002) 111–114. [4] A.J. Kubis, G.J. Shiflet, D.N. Dunn, R. Hull, Focused-ion beam tomography, Metallurgical and Materials Transactions A 35 (7) (2004) 1935– 1943. [5] Nan Yao, Focused Ion Beam Systems Basics and Applications, Cambridge Press, Chapters 5–6, 2007. [6] J.R. Wilson, W. Kobsiriphat, R. Mendoza, Hsun-Yi Chen, J.M. Hiller, D.J. Miller, K. Thornton, P.W. Voorhees, S.B. Adler, S.A. Barnett, Three-dimensional reconstruction of a solid-oxide fuel-cell anode, Nature Materials 5 (7) (2006) pp. 541–544. [7] J. Canny, A computational approach to edge detection, IEEE Transaction Pattern Analysis and Machine Intelligence 8 (1986) 679–714. [8] W. Gander, G.H. Golub, R. Strebel, Least-squares fitting of circles and ellipses, BIT Numerical Mathematics 34 (1994) 558–578.
Contents lists available at ScienceDirect
Microelectronic Engineering journal homepage: www.elsevier.com/locate/mee
3D reconstruction of a micro pipette tip H. Ostadi, M. Malboubi, P.D. Prewett, K. Jiang * Centre for Biomedical and Nanotechnology, Department of Mechanical Engineering, University of Birmingham, Birmingham, B15 2TT, UK
a r t i c l e
i n f o
Article history: Received 26 September 2008 Received in revised form 14 November 2008 Accepted 14 November 2008 Available online 6 December 2008 Keywords: Focused ion beam Patch clamping 3D reconstruction Micro pipette
a b s t r a c t Glass micro pipettes have been widely used for patch clamping. Although many researchers agree that the shape and roundness of pipette tip are important for cell grabbing and gigaseal formation, no numerical values of the geometry have been reported so far. In this research, a FIB/SEM system was used as a nano tomography tool to obtain the 3D shape of a pipette tip. The process involves a cycle of cutting a slice of the pipette using focused ion beam (FIB), taking an SEM image of the new surface, and then milling and imaging again to produce a stack of SEM images. A 3D reconstruction of the pipette was made based on the SEM images. The roundness of the pipette tip was measured and is presented in this paper for the first time. Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction Patch clamp recording is an electrophysiology technique allowing study of ion channels of cells. It is a refinement of the voltage clamp method [1]. It has important applications in cell study and is an important tool to pharmaceutical research. In patch clamping, a glass pipette with a tip diameter of a few microns is pressed against a cell membrane to form a high resistance seal between the inside and outside of the pipette nozzle, known as gigaseal. It is found that achieving a gigaseal is relevant to the tip geometry of the internal wall of the pipette. The physical and chemical mechanisms behind the gigaseal formation are not fully elucidated. It is not only determined by the surface properties of the pipette tip and the cell itself, such as cleanliness, smoothness or roundness but also to a large extend by the skill and patience of the operator. The patch clamp pipette is not the only, but one of the most important factors in gigaseal formation. The tip is the place where the pipette and cell membrane interact with each other. Exact study of this area will help us to better understand the phenomenon of gigaseal formation and may lead us to optimize the shape of the patch clamp pipette [2]. In the 3D reconstruction of a pipette tip, an FEI dual beam Strata 235 focused ion beam (FIB) system was used as a nano tomography tool. The process involves a cycle of milling a slice of the pipette using FIB, taking an SEM image on the new surface, and then milling and imaging again to produce a stack of SEM images. A 3D reconstruction of the pipette was made based on the SEM images.
* Corresponding author. E-mail address: [email protected] (K. Jiang). 0167-9317/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.mee.2008.11.041
Similar approached can be found to be used in many other areas [3–6].
2. Experiment The first step in 3D reconstruction of a micro pipette is obtaining the volumetric information. The dual beam FIB makes this process possible by providing high resolution volumetric information. A micro pipette with a tip diameter of 1.3 lm was placed facing the electron beam. Fig. 1a shows the schematic of pipette, electron beam and ion beam configuration. The angle between I-beam and E-beam was 52°. Therefore, the angle between the imaging plane and the sample was 38°, referring to Fig. 1b. This information was used later for reconstruction. Each slice of the sample was milled off using Ga+ ion beam at 30 Kv and 100 pA for 90 s and dwell time of 1 ls with overlap parameters of 50%. 60 slices with a total thickness of 3 lm were removed and SEM images of the slices taken. The pixel size of the SEM images was 4.5 nm. Fig. 2 shows an image of the 20th slice after milling and its internal edge. (1) Image alignment: We used feature based alignment method [4]. A fixed feature which has not been milled during slicing and not affected by the ion beam is the right side of the pipette in Fig. 1a corresponds to the south part of the pipette in Fig. 2a. We used this part of the pipette to align all of the images. (2) Edge detection: The edge of the internal circle of pipette was detected using Canny algorithm [7]. The basic idea of this algorithm is to detect at the zero-crossing of the second derivative of the smoothed images.
H. Ostadi et al. / Microelectronic Engineering 86 (2009) 868–870
869
It seeks out the zero-crossings of:
o2 ðM IÞ=on2 ¼ oð½oM=on IÞ=on Where M and I are image matrix and unit matrix respectively and n is the direction of the gradient of the smoothed image. This has been done using MATLAB for all of the slices. (3) Back projection: As far as the image plane has a 38° with respect to the sample slices (Fig. 1b) one can calculate the position of each point of the sample slice. Assuming that x (horizontal) and y (vertical) axis are in the image plane and z is the norm of the surface, then Fig. 1. (a) A schematic of pipette, E-beam and I-Beam configuration, (b) a schematic of projected plane and the sample slices. The dark grey shows the projected plane.
Fig. 2. (a) An SEM image of a pipette after milling, (b) the edge of the internal circle of the pipette detected using Canny algorithm.
Fig. 3. (a) A 3D structure of the pipette tip after reconstruction, (b) an isometric view of the pipette and (c) the reconstructed sample in the Y–Z plane.
Fig. 4. (a) A projected image of the 3D pipette tip on the X–Y plane, (b) an image of the first slice and fitted circle. The fitted circle is shown in dashed line.
870
H. Ostadi et al. / Microelectronic Engineering 86 (2009) 868–870
xssp ¼ x yssp ¼ y=cos 38
where indices ‘‘ssp” is for the sample slice position. For zssp, the thickness of the slices are 50 nm so the relative distance between the slices remains 50 nm (or 11 pixels) and the initial angle of the image plane and projected plane is 38°. Fig. 3a shows the 3D structure of the pipette tip reconstructed using MATLAB. The units of X,Y,Z axes are in pixel and each pixel is 4.5 nm. Fig. 3b is the isometric view of the pipette and Fig. 3c shows the Y–Z plane view of the reconstructed image. In order to examine the shape of pipette tip, a perfect circle was fitted to each slice based on the least squared fitted circle method[8] and maximum deviation of the pipette shape from the circle was obtained. Fig. 4a shows the projected image of 3D pipette tip on the X–Y plane for roundness examination. Fig. 4b shows the first slice image and fitted circle. The maximum deviation from fitted circle in Fig. 3b was 43 nm. The average of maximum deviations of all slices was found as 67 nm. This is the first direct roundness measurement of a pipette known to the authors. 3. Conclusions There is a lack of knowledge in measurement of micro pipette tip to understand the gigaseal formation. Using a focused ion beam and SEM system, one can obtain the key parameters of patch clamping such as shape and roundness. Three micrometer long
section of a pipette with the tip diameter of 1.3 lm has been examined and 60 images with 4.5 nm pixel size were taken. The internal walls of the pipette tip, where the contact between the pipette and a cell occurs, were detected using Canny algorithm. After 3D reconstruction of the micro pipette tip in MATLAB, we examined the roundness of each slice with least square fitted circle. The average of maximum deviation of all slices were found as 67 nm, or 10% in roundness error.
References [1] E.R. Kandel, J.H. Schwartz, T.M. Jessell, Principles of Neural Science, fourth ed., McGraw-Hill, New York, 2000. [2] Wee-Liat Ongi, Levent Yobas, Wei-Yi Ong, A missing factor in chip-based patch clamp assay: gigaseal, Journal of Physics 34 (2006) 187–191. [3] R.M. Langford, G. Dale, P.J. Hopkins, P.J.S. Ewen, A.K. Petford-Long, Focused ion beam micromachining of three-dimensional structures and three-dimensional reconstruction to assess their shape, Journal of Micromechanics and Micro Engineering 12 (2002) 111–114. [4] A.J. Kubis, G.J. Shiflet, D.N. Dunn, R. Hull, Focused-ion beam tomography, Metallurgical and Materials Transactions A 35 (7) (2004) 1935– 1943. [5] Nan Yao, Focused Ion Beam Systems Basics and Applications, Cambridge Press, Chapters 5–6, 2007. [6] J.R. Wilson, W. Kobsiriphat, R. Mendoza, Hsun-Yi Chen, J.M. Hiller, D.J. Miller, K. Thornton, P.W. Voorhees, S.B. Adler, S.A. Barnett, Three-dimensional reconstruction of a solid-oxide fuel-cell anode, Nature Materials 5 (7) (2006) pp. 541–544. [7] J. Canny, A computational approach to edge detection, IEEE Transaction Pattern Analysis and Machine Intelligence 8 (1986) 679–714. [8] W. Gander, G.H. Golub, R. Strebel, Least-squares fitting of circles and ellipses, BIT Numerical Mathematics 34 (1994) 558–578.