17 Photolithography

Dr. Ayushi Paliwal and Dr. Monika Tomar

Introduction

Microfabrication is becoming progressively more important to modern science and technology. The ability to fabricate new types of microstructures or to engineer smaller versions of the existing structures offers many opportunities for technological advancements. These microstructures also provide the prospects to study basic scientific phenomenon that occur at smaller dimensions like in case of quantum confinement which is observed in nanostructures. However, microfabrication has its basic use in microelectronics and its application ranges from microanalysis to microelectromechanical systems (MEMS). In the era of miniaturization, performing chemical or biochemical reactions and analysis require cavities, channels, pumps, valves, storage containers, electrodes, and many more. The typical dimensions of these components are in the range of a few micrometers to several millimeters in length or width, and between a few nm to few µm in depth and height. Due to microfabrication, some of the advantages of microsystem based devices are as follows:

Lesser material usage Reduction in power

Better functionality per unit area

Ease of access to regions that are forbidden to larger products

Among various microfabrication techniques, photolithography is one of the most commonly used technique. In the following sections, we begin with the description of photolithography, then focus on the methodology involved in photolithography.

1. 1. Photolithography

1.1 Introduction

The fabrication of an integrated circuit (IC) requires multiple physical and chemical processes to be executed on a semiconductor (e.g., silicon) substrate. In general, the various processes used to manufacture an IC fall into three categories: film deposition, patterning, and semiconductor doping. Conducting (such as polycrystalline silicon and aluminum) as well as insulating films (various forms of silicon dioxide, silicon nitride, and others) are used to connect and isolate transistors and their components. The conductivity of the silicon can be altered with the application of voltage by selectively doping various regions of the silicon. By creating structures of these different components, millions of transistors can be fabricated and assembled together to form the complex circuitry of a modern microelectronic device. The most fundamental of these processes is lithography, i.e., the formation of three-dimensional images for subsequent transfer of the image pattern to the substrate.

Photolithography as the name suggest , consist of word “Photo” and “lithography” where “photo” means light and “lithography” is further another word from greek where “lithos” means stone and “graphia” means to write. So we can say that lithography is to engrave some pattern or write something on stone but the same word is used for incising some pattern on desired substrate like Silicon, Corning, Quartz etc. and Photolithography is to incise the desired pattern on desired substrate using light (for example Ultra Violet rays). The patterns are formed using photo resist which will be discussed later in detail. Also termed as optical lithography or UV lithography, it is the processes used in micro fabrication to pattern parts of a thin film or a substrate. Light
is used to transfer a geometric pattern from a photo mask to a light-sensitive chemical “photo resist”, or
simply “resist,” on the substrate. In this way a photo resist pattern is incised on substrate which may useful in incising pattern of any desired material like Aluminium, gold, copper, platinum etc. on desired substrate.

1.2 Process steps of Lithography

In typical Photolithography basically we start with the wafer already coated with the material to be patterned, photo resist is coated using spin coating technique and allowed to bake for some time such that the solvent can be evaporated, followed by the UV exposure. Figure 1 presents the schematic for following the process steps in photolithography. On UV exposure the properties of the photoresist got changed and exposed part get dissolved in developer for the positive photoresist whereas unexposed part gets dissolved in developer for the negative photoresist. Since, we have transferred the pattern from mask to substrate. Now the part which got dissolved in developer can be etched easily using chemical etching or plasma etching keeping the undissolved part unaffected (since it is protected by the photoresist). The photoresist which was protecting the patterns can be easily removed using chemical treatment or plasma etching. There are some materials which cannot be etched easily are patterned using lift off technique. In this technique negative image of the pattern is incised on the wafer before depositing the material. The material is deposited after this such that the material gets deposited on the desired places (where there is no photoresist) and deposited on the photoresist layer (where the pattern is not required). On dipping the wafer in suitable chemical photoresist got dissolved, taking with it the material layer from the unwanted places (Hence the process is called as lift off). So, next we will be discussing about the photolithography steps in detailed. Here the etching part is discussed in detail. Since lift off is same as etching only (except the ordering of steps), so only a brief introduction of lift off is given in the end. Here is the flow chart of steps followed by the detailed description of each step.

Figure 1: Schematic for Photolithography process

1. Surface Preparation

Substrate preparation is the important and most crucial step to start with the photolithography. It is aimed to improve the adhesion of the photoresist material to the substrate. This is achieved by one or more of the following process involving substrate cleaning, and dehydration bake. The substrate cleaning is used to remove contamination. Substrate contamination can be in the the form of particulates or a film and can be either organic or inorganic. Defects in the final resist pattern are a consequence of particulates, whereas film contamination can cause poor adhesion and subsequent loss of line width control. Particulates generally arise from airborne particles or contaminated liquids (e.g., dirty adhesion promoter). Therefore, substrate cleaning is very much necessary before photolithography. For chemical cleaning, RCA1, RCA2 and pirana cleaning are used to remove all the organic and inorganic impurities. Also, various other solvents like tetra chloroethylene, acetone and iso-propyl alcohol are also used for cleaning different substrates like glass, silicon and sapphire etc.

Pirana cleaning is firstly used to clean the wafers generally which is a mixture of Sulphuric Acis and Hydrogen Peroxide in 1:1 ratio. It removes the organic and inorganic contaminations from the wafer. RCA1 is a mixture of De-ionised Water:Ammonium Hydroxide: Hydrogen Peroxide in the ratio of 5:1:1. Wafers are dipped in the RCA1 solution at 80oC typically for 10 minutes. RCA1 cleaning helps to remove the organic residues and is also very effective in removing particles from the surface. Subsequently, De-ionised water is used and the wafers are dried with ajet of dry nitrogen gas. Whereas, RCA2 cleaning removes ionic and metallic contaminations present on the surface of the wafer. RCA solution consists of the mixture of DI water: Hydrochloric Acid: Hydrogen Peroxide in 5:3:3 ratio at 70oC. The cleaning is done for 20 minutes and finally the wafers are rinsed in DI water and dried finally. These treatments results in the formation of thin silicon dioxides layer in case of silicon wafers. It can lead to recontamination since the bare silicon surface is very effective. So, to remove the native oxide from the silicon wafer, wafers are dipped in the solution of Hydrogen Fluoride: De-ionized water in the ratio of 1:20 for 5 to 10 seconds. Again, the wafers are to be rinsed in DI water and dry with dry nitrogen gas.

2. Spin coating of photoresist

Resist formulation may generally consists of 1) film-forming resin, 2) solvent, 3) sensitizer or photoinitiator or photoacid generator and 4) additives. The dried film coated form a solution containing the above ingredients undergoes changes when exposed to radiation of a certain wavelength or electrons or even ions. This exposure alters the solubility in the developer of the exposed areas relative to the unexposed areas. If the resist prints in a negative tone, the exposed areas are hardened, whereas positive tone resists are made more soluble in the developer solvent by the exposure. The changes in the exposed areas are due to some chemical reactions brought about by the absorbed radiation.

When the surface is cleaned, the photo resist is applied using spin coating technique. A viscous, liquid solution of photo resist is dispensed onto the substrate, and the substrate is spun rapidly, while being coated with the photo resist such that it bonds uniformly to the surface. This uniformity can be accounted for using a fluid-mechanical model, which shows that the resist moves much faster at the top of the layer than at the bottom, where viscous forces bind the resist to the wafer surface. Thus, the top layer of resist is rapidly expelled from the wafer’s edge while the bottom layer still moves slowly and radially along the substrate. In this extra resist is removed, leaving a very smooth and uniform layer of photoresist.

3. Pre-Bake

The pre-bake is a simple process where the surface is heated in a convection oven or through a hot plate placed below it. The purpose of this step is to evaporate the excess coating solvent and to compact and harden the photo resist such that the adhesion of photo resist increases and film got less susceptible to contamination.

4. Alignment and Exposure

After baking the wafer must be aligned correctly with respect to mask such that we get the pattern of the mask on wafer at desired position. This procedure is accomplished by micromechanical controller provided on the Mask aligner using certain marks preprinted on the mask or by using an automatic pattern recognition device.

Mask aligner

Mask aligner is one of the important instruments that performs one to one exposure for the micro structuring in the lithography process. The first Mask Aligner was developed by Karl Süss in 1963 for the fabrication of transistors at Siemens in Munich, Germany. [http//www.suss.com].

Basic principle: The principle of a Mask Aligner is based on the illumination of a photomask which is brought into the contact or kept at a small proximity distance from a photoresist coated wafer. This technique can be classified as contact or proximity printing depending on whether the mask is in direct contact to the wafer or there exists a proximity gap between both. Furthermore, the contact mode is divided into two modes: (i) soft contact and (ii) hard contact. In soft contact, the substrate is brought into contact with the mask by a predetermined force during the time of exposure whereas, in hard contact an additional upward force is given to the wafer by inletting N2 purge into the system to provide better contact. Exposure of UV radiation to the wafers can be carried in different contact modes as shown in Fig. 2.

Hard contact is preferred over soft contact in doing alignment for pattern/linewidth less than 10 m. Different ways adopted for alignment are shown in Fig. 2.

Two most commonly used alignment types are: (i) Top side alignment (TSA) and (ii) Back side alignment (BSA) as shown in Fig. 3. In TSA, at least one time patterned wafer is aligned with new photomask from the microscopes given on the top side of the system. On the other hand in BSA, the patterned wafer is aligned from the microscopes given at the bottom of the system. In BSA, an image of the photomask (which has to be aligned) is captured first and then the back side of the patterned wafer is aligned with the grabbed image. After alignment a high intensity UV lamp is used for exposure of photoresist coated on the wafer through the photomask to imprint the pattern given on the photomask.

After alignment UV light is allowed to fall on the substrate as shown in figure 5 through the mask (containing the desired pattern) for the optimized time which will be different for different substrates.

5. Development

A post-exposure bake (PEB) is carried out before developing the wafer so that the standing wave phenomena caused by the destructive and constructive interference patterns of the incident light can be reduced in order to get a smooth pattern.

Upon exposure to light, a chemical change in the photoresist occurs that allows some of the photoresist to be removed by a special solution, called “developer”. During the development step, chemicals are applied to the surface causing either a reaction with photoresist. Positive photoresist, becomes soluble in its corresponding developer when exposed while in the negative photoresist, unexposed regions dissolve in its corresponding developer as shown in Fig. 6.

6. Post Bake

The post bake is done to stabilize and harden the photo resist. It also removes any remnant developer. The resulting substrate is then “hard-baked”, typically at 140°C for 20 minutes. The remaining photo resist solidifies on hard bake, so that it is able to withstand etching and doesn’t get damaged during etching process.

7. Etching

Etching is an essential step during microfabrication. The process of etching involves both chemical and mechanical mechanisms leading to the removal of the material which is not protected by the photoresist. Etching can be done either in the form of wet etching or dry etching. Wet etch, is perhaps one of the simplest form of etching method. It requires an etchant solution such as an acid that removes the underlying film reacting chemically leaving the photoresist intact. This etching is isotropic and thus can lead to damage of pattern below the photoresist also. So anisotropic etching can be used as subsitute, in which directionality is induced into the etch process. In this case, wet etching does not serve the purpose, then dry etching or plasma etching comes into picture. In plasma etching, the etchant is replaced using plasma i.e. an ionized gas. An electric field is applied that causes the ions to be accelerated downwards towards the wafer. The resulting etch is a mixture of both chemical etching that occurs due to reaction of the film with the plasma and physical sputtering, due to the bombardment of the ions hitting the wafer. The chemical character of the etch leads to the etching sensitivity of the film with respect to the resist and with respect to the substrate below the film. Plasma etching is very directional, but not very selective. Reactive ion etching is another way of etching that merges both effects to give better selectivity and directionality.

8. Stripping (removal of photo resist)

After the imaged wafer has been pattern transferred, the remaining photoresist must be removed. There are two classes of resist stripping technique: 1) wet stripping that is done using organic or inorganic solutions, and 2) dry stripping. In the end since we have got our desired pattern of the desired material beneath photo resist which is no longer required, it must be removed from the substrate.

1.3 Lift Off Process

There are some materials which are not easy to be etched in that cases we patterned the photo resist on the wafer where we don’t need the pattern of desired material such that it leaves the desired places uncoated with photo resist using dark field masks. Now we can coat the substrate with desired material and material from the undesired places can be removed using lift off technique. On dipping the substrate in acetone it will dissolve photo resist beneath the undesired places removing them.

Summary:

1. The complete process of photolithography for the development of microstructures has been discussed along with the detailed description of each and every step followed.