**1. Introduction**

130 Microelectromechanical Systems and Devices

McCool, J. I. (1986). Comparison of Models for the Contact of Rough Surfaces, *Wear* Vol. 107:

Morrow, A. (2003). *Adhesive Rough Surface Contact*, Ph.D. thesis, University of Pittsburgh,

Morrow, C., Lovell, M.R., Ning, X. (2003). A JKR–DMT transition solution for adhesive rough surface contact, *Journal of Physics D Applied Physics* Vol. 36: 534–540. Myshkin, N.K., Ya. A., Grigoriev, S.A., Chizhik, Choi, K.Y., Petrokovets, M.I. (2003). Surface roughness and texture analysis in microscale, *Wear* Vol. 254: 1001–1009. Myshkin, N.K., Ya., Grigoriev, A., Kholodilov, O.V. (1992). Quantitative analysis of surface topography using scanning electron microscopy, *Wear* Vol.153 (No.1): 119–133. Muller, V. M., Yushchenko, V. S., B. V. Derjaguin. (1980). On the influence of molecular

Nayak, P. R. (1971). Random Process Model of Rough Surfaces, *Journal of Lubrication* 

Nayak, P. R. (1973). Random Process Model of Rough Surfaces in Plastic Contact, *Wear* Vol.

Oden, P.I., Majumdar, A., Bhushan, B., Padmanabhan, A., and Graham, J.J. (1992). AFM

Papoulis, A. (1965). *Probability, Random Variables and Stochastic Processes*, McGraw Hill, New

Peng, Y. F. & Guo, Y. B. (2007). An Adhesion Model for Elastic-Plastic Fractal Surfaces,

Peng, Y. F., Guo, Y. B., Hong, Y. Q. (2009). An Adhesion Model for Elastic-contacting Fractal Surfaces in Presence of Meniscus, *ASME Journal of Tribology* Vol. 131: 024504-1-5.

Sayles, R. S. and Thomas, T. R. (1978). Surface Topography as a Nonstationary Random

Shockley, W., Hooper, W.W., Queisser, H.J. and Schroen, W. (1964). Mobile electric charges

Tabor, D. (1977). Surface forces and surface interactions, *Journal of Colloid Science on Science*

Wang, S. & Komvopoulos, K. (1994). A fractal theory of the interfacial temperature

Williams, E. D. & Bartlet, N. C. (1991). Thermodynamics of Surface Morphology, *Science* Vol.

Xue, X. & Polycarpou A. (2007). An improved meniscus surface model for contacting rough

Yamachika, R. et al. (2004). Controlled Atomic Doping of a Single C60 Molecule, *Science* Vol.

Yan, W. & Komvopoulos, K. (1998). Contact analysis of elastic-plastic fractal surfaces, *Journal* 

surfaces, *Journal of Colloid and Interface Science* Vol. 311: 203–211.

on insulating oxides with application to oxide covered silicon p-n junctions, *Surface* 

distribution in the slow sliding regime: Part I − Elastic contact and heat transfer

Sarid, D. (1191). *Scanning Force Microscopy*, Oxford University Press, New York.

*Journal of Colloid Science on Science* Vol. 77: 91-101.

Surfaces, *Journal of Tribology* Vol. 114: 666–674.

*Journal of Applied Physics* Vol. 102(No.5): 3510-7.

Process, *Nature* Vol. 271: 431–434.

Thomas, T. R. (1982). *Rough Surfaces*, Longman, New York.

analysis, *Journal of Tribology* Vol. 116: 812–823.

*Science* Vol. 2: 277-287.

Vol. 58: 2-13.

251: 393–400.

304:281-284.

*of Applied Physics* Vol. 84: 3617.

*Technology* Vol. 93: 398–407.

forces on the deformation of anelastic sphere and its sticking to a rigid plane,

Imaging, Roughness Analysis and Contact Mechanics of Magnetic Tape and Head

37–60.

Pittsburgh.

26: 305–333.

York.

In the area of Micro Electro Mechanical Systems (MEMS), bulk micromachining and surface micromachining are two main technologies. Bulk micromachining defines structures by selectively etching inside a substrate while surface micromachining uses a succession of thin film deposition and selective etching on top of a substrate. The two technologies are quite different, resulting in different dimensions and different mechanical properties. Although bulk micromachining is usually considered to be the older technology, the two developments run parallel. This is due to the fact that the two different approaches have trade-offs. Taking an example of MEMS capacitance accelerometers, surface micromachined structures use smaller chip area, thus leaving more space for the electronics. On the other hand, in bulk micromachining the larger mass gives greater sensitivity for accelerometers and the larger area leads to larger capacitances for easy read out, which are extremely useful in the inertial device fabrication.

#### **1.1 Wet anisotropic etching in MEMS**

Bulk micromachining technology relies on isotropic or anisotropic liquid phase (wet) etching as well as by plasma phase (dry) etching in single crystalline silicon for the formation of functional shapes and patterns in many major applications. Although dry etching has penetrated the traditional territory of wet etchants, the high cost in dry etching and the difficulty in the etch rate uniformity on the whole wafer in wet isotropic etching still make wet anisotropic etching the most affordable method for the reliable production, if wet anisotropic etching can be used to deliver a similar intermediate or final structure.

The formation of crystal facets due to etching is referred to as faceting. When the primary flat of Si {100} is along the [110] direction, rectangular structures with concave corners are easily made with four (111) sidewalls and a (100) plane as the bottom. If the slow etching (111) planes meet, etching will be self-limiting to result in inverted pyramids. Etched grooves, trenches, wells and other basic structures of diaphragms (membranes), beams, and cantilevers exemplify the features of crystal plane-dependent etching. Combined with the use of mask patterns, the etch rate anisotropy becomes a most valuable property as it provides a low-cost, precise means for the production of three-dimensional shapes delimited by smooth, shiny facets, leading to complex structures with multiple functionalities, as shown in Fig. 1.

Advanced Surfactant-Modified Wet Anisotropic Etching 133

With different etchants or in the conditions of different concentrations and temperatures, people could get different etching characteristics, which can be clarified as etch rate anisotropy, surface roughness and mask-corner undercut. Surface roughness improvement is important when considering the optical and tribological applications. On the other hand, the conventional design of MEMS structures fabricated by wet silicon bulk micromachining on Si {100} has sharp edge convex and concave corners. This design exhibits stress concentration at the concave corners when a load is applied, which may initiate micro cracks. By providing rounded concave corners instead of sharp ones, the stress can be reduced, thus improving the mechanical efficiency of the microstructures. In pure TMAH solutions, however, both surface roughness and mask-corner undercut are not good enough in regard to above applications. Fig. 2 illustrates anisotropic etched cantilever beam shaped patterns with mask-corners in 10 wt% TMAH at 60 °C. The etched surface is full of hillocks with an average roughness of 110nm. Therefore, many other factors have to be thought over.

Fig. 2. Anisotropic etched cantilever beam shaped patterns with mask-corners in 10 wt% TMAH at 60 °C (etching depth = 34 µm). The upper right figure is 100 µm x 100 µm.

Among those methods for improving surface roughness and mask-corner undercut, for instance, metal impurities, alcohols, diffusion, light, pressure, microwave irradiation, corner compensation and ultrasonic irradiation et al., surfactant-modified wet anisotropic etching is more outstanding for the effects in both surface smoothness and undercut decrease, also

Various ionic (e.g. anionic SDSS, cationic ASPEG, etc.) and non-ionic (e.g. PEG, NC series, Triton X-100, etc.) surfactants have been investigated. Although anionic surfactants exhibit the highest etching rate, the cationic and non-ionic surfactants are suitable for TMAH solutions to improve the roughness of the etched surface owing to the excellent capacity to wet the silicon wafer. TMAH solutions with cationic and non-ionic surfactants are ICcompatible process. Furthermore, adding non-ionic surfactant to TMAH solutions can efficiently reduce undercutting at mask-corners. Such an addition is preferred when accurate profiles are required without very deep etching. Therefore, non-ionic surfactantmodified etching process attracts researchers' attention. With regard to easily handling and

**1.2 Motivation of surfactant-modified etchants** 

for its stabilization of anisotropy change.

less toxicity, in this study, Triton X-100 is selected.

Fig. 1. Wet anisotropic etched structures for various kinds of applications: (a) microneedles for bio-medical applications, Shikida et al. 2006; (b) deep grooves, Sato et al. 1998; (c) mass-spring systems for accelerometers, Butefisch et al. 2000; (d) grooves for the optical fibre alignment, Hoffmann et al. 2002; (e) cantilever beam; (f) suspended filament beams as MEMS heaters, Lee et al. 2009.

A number of alkaline etchants have been tried for wet anisotropic etching of single crystal silicon. Some main features of wet etchants are compared in Table 1. EDP (also referred to as EPW for ethylene diamine, pyrocathecol and water) is not wildly used because of occupational safety and health hazards. Therefore, potassium hydroxide (KOH) and tetra methyl ammonium hydroxide (TMAH) are the most commonly used anisotropic etchants. Based on the ability to withstand the chemical attack by these etchants, silicon oxide (SiO2), silicon nitride (Si3N4), and other metal layers (e.g. Cr, Au) have been used as masking materials. KOH is non-toxic, easy to use, provides excellent etching profiles and has a good selectivity between Si and Si3N4, although poor for SiO2. In addition, KOH is incompatible with CMOS processing due to the presence of an alkali metal. Although TMAH has a lower etch rate, it has outstanding characteristics, such as a high selectivity between Si and SiO2, and the absence of harmful ions. Hence, TMAH solutions are preferred in recent research and production.


+:Good 0:Fair -:Poor

Table 1. Comparison of some main features of wet etchants.

Fig. 1. Wet anisotropic etched structures for various kinds of applications: (a) microneedles

(c) mass-spring systems for accelerometers, Butefisch et al. 2000; (d) grooves for the optical fibre alignment, Hoffmann et al. 2002; (e) cantilever beam; (f) suspended filament beams as

A number of alkaline etchants have been tried for wet anisotropic etching of single crystal silicon. Some main features of wet etchants are compared in Table 1. EDP (also referred to as EPW for ethylene diamine, pyrocathecol and water) is not wildly used because of occupational safety and health hazards. Therefore, potassium hydroxide (KOH) and tetra methyl ammonium hydroxide (TMAH) are the most commonly used anisotropic etchants. Based on the ability to withstand the chemical attack by these etchants, silicon oxide (SiO2), silicon nitride (Si3N4), and other metal layers (e.g. Cr, Au) have been used as masking materials. KOH is non-toxic, easy to use, provides excellent etching profiles and has a good selectivity between Si and Si3N4, although poor for SiO2. In addition, KOH is incompatible with CMOS processing due to the presence of an alkali metal. Although TMAH has a lower etch rate, it has outstanding characteristics, such as a high selectivity between Si and SiO2, and the absence of

harmful ions. Hence, TMAH solutions are preferred in recent research and production.

(40 wt%)

Etching of SiO2 mask 0 ++ +

µm/min 1 0.5 <sup>1</sup>

IC process - + +

(toxicity) + + - Cost ++ + +

TMAH (25 wt%)

EDP (80 wt%)

(at 115 °C)

Etchant KOH

Table 1. Comparison of some main features of wet etchants.

for bio-medical applications, Shikida et al. 2006; (b) deep grooves, Sato et al. 1998;

MEMS heaters, Lee et al. 2009.

Rate (at 80 °C)

Compatibility for

Handling

+:Good 0:Fair -:Poor

#### **1.2 Motivation of surfactant-modified etchants**

With different etchants or in the conditions of different concentrations and temperatures, people could get different etching characteristics, which can be clarified as etch rate anisotropy, surface roughness and mask-corner undercut. Surface roughness improvement is important when considering the optical and tribological applications. On the other hand, the conventional design of MEMS structures fabricated by wet silicon bulk micromachining on Si {100} has sharp edge convex and concave corners. This design exhibits stress concentration at the concave corners when a load is applied, which may initiate micro cracks. By providing rounded concave corners instead of sharp ones, the stress can be reduced, thus improving the mechanical efficiency of the microstructures. In pure TMAH solutions, however, both surface roughness and mask-corner undercut are not good enough in regard to above applications. Fig. 2 illustrates anisotropic etched cantilever beam shaped patterns with mask-corners in 10 wt% TMAH at 60 °C. The etched surface is full of hillocks with an average roughness of 110nm. Therefore, many other factors have to be thought over.

Fig. 2. Anisotropic etched cantilever beam shaped patterns with mask-corners in 10 wt% TMAH at 60 °C (etching depth = 34 µm). The upper right figure is 100 µm x 100 µm.

Among those methods for improving surface roughness and mask-corner undercut, for instance, metal impurities, alcohols, diffusion, light, pressure, microwave irradiation, corner compensation and ultrasonic irradiation et al., surfactant-modified wet anisotropic etching is more outstanding for the effects in both surface smoothness and undercut decrease, also for its stabilization of anisotropy change.

Various ionic (e.g. anionic SDSS, cationic ASPEG, etc.) and non-ionic (e.g. PEG, NC series, Triton X-100, etc.) surfactants have been investigated. Although anionic surfactants exhibit the highest etching rate, the cationic and non-ionic surfactants are suitable for TMAH solutions to improve the roughness of the etched surface owing to the excellent capacity to wet the silicon wafer. TMAH solutions with cationic and non-ionic surfactants are ICcompatible process. Furthermore, adding non-ionic surfactant to TMAH solutions can efficiently reduce undercutting at mask-corners. Such an addition is preferred when accurate profiles are required without very deep etching. Therefore, non-ionic surfactantmodified etching process attracts researchers' attention. With regard to easily handling and less toxicity, in this study, Triton X-100 is selected.

Advanced Surfactant-Modified Wet Anisotropic Etching 135

affected by surfactant adding is clearly visible. The etch rates of exact and vicinal {100} planes are almost unaffected when the surfactant is added, while the etch rates of exact and

Fig. 3. Location of different crystallographic orientations on one silicon hemispherical sample and schematic view of surface profile measurement. (Reproduced with permission from IOP)

Fig. 4. Photos of the hemisphere specimen: (a) before etching and (b) after etching

The effect of etching temperature on the etch rates is shown in Fig. 6. The etching anisotropy is influenced by temperature. The orientation of the highest etch rate is shifted toward {350} with the increase in temperature. Contrary to 71 and 81 °C, no single plane prominently appears as the highest etching rate plane at 61 °C. The etch rates of the orientations between {100} and {110} largely depend on the temperature; however, those in between {110} and {111} exhibit less dependence. It can obviously be concluded from these results that the etching anisotropy depends upon the etching temperature. This will result in different etched profiles due to the difference in etching temperatures even if the

(etching temperature = 61 °C).

vicinal {110} planes are reduced significantly.

This chapter starts by providing the completely etch rate anisotropy in surfactant-modified wet etching in section 2; the mechanism behind of the change of etching characters when compared with pure etchants will be analyzed in section 3; several applications for the fabrication of new structures by using this advanced anisotropic wet etching will be presented in section 4.
