Introduction

Abstract

In the development of surface micromachining it has been necessary to be able to remove a section of the substrate or a deposited layer. Although this layer is not present in the final device it is an essential part of the processing sequence. Ease of etching and high selectivity over the etch-rate of the mechanical layer are essential for MEMS fabrication. This issue contains four special papers covering sacrificial etching. Three of these involve the etching itself and the fourth considers the problems of drying the structure after releasing. Of course this is not claimed to be a comprehensive handling of the whole subject. It is intended to discuss the issues for a selection of the available materials, although many of these issues are also valid for other sacrificial layers not discussed here. The first paper involves the use of porous silicon as a sacrificial layer where part of the substrate is removed to leave free-standing structures. This is an electrochemical process in which part of the substrate is selectively made porous, usually using HF as the etchant. Due to its high surface area, porous silicon can be removed extremely easily, which is a desirable property for a sacrificial layer. Recent developments have also been made in alternative materials and this is the subject covered in two of the articles. The first examines sacrificial etching techniques for non-silicon semiconductors, which leads to the possibility of micromachining piezoelectric materials. The second deals with the use of aluminium as a sacrificial layer. Aluminium is a standard layer used in electronic devices and its use as a sacrificial layer is therefore an interesting option. The paper examines four etchants for aluminium sacrificial etching. The final paper covers an important issue for all the above mentioned sacrificial etching techniques, that of sticking. After the mechanical structures are released in the etchant they have to be dried. If this is not done correctly the structures will stick to the substrate where they will remain. This drying process has presented many problems in the past, and several solutions have been proposed. These solutions along with the theoretical limitations on structural dimensions are discussed. This issue is by no means closed. Further developments are extending the use of new materials and improving the use of materials already available. I would like to thank the people who have contributed to making this issue possible; the authors and the referees, and of course the staff of IOP Publishing. P J French Delft University of Technology