Downloaded 233 times
More Related Content
photolithography
- 1. PHOTOLITHOGRAPHY AND DIFFUSION By Kumar Gaurav,B.Tech(3rd year) 1
- 2. INTRODUCTION Photolithography Temporarily coatphotoresist on wafer Transfers designed pattern to photoresist Most important process in IC fabrication 40 to 50% total wafer process time Determines the minimum feature size 2
- 3.
- 4. OVERVIEW OF PHOTOLITHOGRAPHY(CTND.) Lithography consists of patterning substrate by employing the interaction of beams of photons or particles with materials. Photolithography is widely used in the integrated circuits (ICs) manufacturing. The process of IC manufacturing consists of a series of 10-20 steps or more, called mask layers where layers of materials coated with resists are patterned then transferred onto the material layer.
- 5. OVERVIEW OF PHOTOLITHOGRAPHY(CTND.) A photolithography system consists of a light source, a mask, and a optical projection system. Photoresists are radiation sensitive materials that usually consist of a photo-sensitive compound, a polymeric backbone, and a solvent. Resists can be classified upon their solubility after exposure into: positive resists (solubility of exposed area increases) and negative resists (solubility of exposed area decreases).
- 6. APPLICATIONS OF PHOTOLITHOGRAPHY Main application: IC patterning process Other applications: Printed electronic board, nameplate, printer plate, and et al. 6
- 7. PHOTORESIST Photo sensitivematerial Temporarily coated on wafer surface Transfer design image on it through exposure Very similar to the photo sensitive coating on the film for camera 7
- 8. 8 Photoresist Negative Photoresist • Becomesinsoluble after exposure • When developed, the unexposed parts dissolved. • Cheaper Positive Photoresist • Becomes soluble after exposure • When developed, the exposed parts dissolved • Better resolution
- 9. 9 Mask/reticle Exposure After Development Negative Photoresist UV light Positive Photoresist Substrate Substrate Substrate Photoresist Negative andPositive Photoresists Substrate Photoresist
- 10. PHOTORESIST COMPOSITION Polymer Solvents Sensitizers Additives 10 POLYMER Solid organic material Transfers designed pattern to wafer surface Changes solubility due to photochemical reaction when exposed to UV rays.
- 11. SOLVENT Dissolves polymersinto liquid Allow application of thin PR layers by spinning. 11 SENSITIZERS Controls and/or modifies photochemical reaction of resist during exposure. Determines exposure time and intensity ADDITIVES Various added chemical to achieve desired process results, such as dyes to reduce reflection.
- 12. NEGATIVE RESIST Mostnegative PR are polyisoprene type Exposed PR becomes cross-linked polymer Cross-linked polymer has higher chemical etch resistance. Unexposed part will be dissolved in development solution. 12
- 13.
- 14. NEGATIVE PHOTORESIST 14 Disadvantages • Polymerabsorbs the development solvent • Poor resolution due to PR swelling • Environmental and safety issues due to the main solvents xylene.
- 15. POSITIVE PHOTORESIST Exposedpart dissolve in developer solution Image the same that on the mask Higher resolution Commonly used in IC fabs 15
- 16. POSITIVE PHOTORESIST Novolacresin polymer Acetate type solvents Sensitizer cross-linked within the resin Energy from the light dissociates the sensitizer and breaks down the cross- links Resin becomes more soluble in base solution 16
- 17. REQUIREMENT OF PHOTORESIST High resolution Thinner PR film has higher the resolution Thinner PR film, the lower the etching and ion implantation resistance High etch resistance Good adhesion Wider process latitude Higher tolerance to process condition change 17
- 18. PHOTORESIST PHYSICAL PROPERTIES 18 •Photoresist must be able to withstand process conditions • Coating, spinning, baking, developing. • Etch resistance • Ion implantation blocking
- 19. PHOTORESIST PERFORMANCE FACTOR Resolution Adhesion Expose rate, Sensitivity and Exposure Source Process latitude Pinholes Particle and Contamination Levels Step Coverage Thermal Flow 19
- 20. PHOTORESIST CHARACTERISTICS SUMMARY 20 Parameter NegativePositive Polymer Polyisoprene Novolac Resin Photo-reaction Polymerization Photo-solubilization Sensitizer Provide free radicals for polymer cross- link Changes film to base soluble Additives Dyes Dyes
- 21.
- 22. CONTACT PRINTING Themask is directly in contact with the wafer Advantages Simple Low Cost Disadvantages Poor for small features Mask damage may occur from contact Defects from contaminants on mask or wafer due to contacting surfaces
- 23. PROXIMITY PRINTING Themask is above the wafer surface Advantages Mask damage is minimal Good registration possible Disadvantages Poorer resolution due to distance from the surface Defects from contaminants on mask or wafer due to contacting surfaces Diffraction errors
- 24. PROJECTION PRINTING Anoptical system focuses the light source and reduces the mask image for exposure on the surface Advantages Higher resolution Lens system reduces diffraction error Disadvantages Errors due to focus of lens system may occur Limiting factor in resolution can be due to optical system
- 25.
- 26. BASIC STEPS substratecleaning Dehydration bake Spin coating primer and PR Soft bake Alignment and exposure Development Pattern inspection Hard bake 26 PR coating Development
- 27.
- 28. SUBSTRATE CLEANING Particularlytroublesome grease, oil or wax stains 2-5 min ultrasonic bath in trichloroethylene (TCE) or trichloroethane (TCA), 65-75ºC (carcinogenic) Standard grease removal 2-5 min ultrasonic bath in acetone 2-5 min ultrasonic bath in methanol 2-5 min ultrasonic bath in D.I. H2O Repeat the first three steps 3 times 30 sec rinse under free flowing D.I. H2O Oxide and other material removal 5 min H2O:H2O2:NH3OH 4:1:1 70-80ºC (cleaning Ge) 30 sec 50% HF (Glass or SiO2) D.I. H2O 3 rinses 5 min H2O:H2O2:HCl 5:1:1 70-80ºC D.I. H2O 3 rinses Spin dry (wafer) / N2 blow dry
- 29. PRE-BAKE AND PRIMERVAPOUR 29 P-Well USGSTI Polysilicon Primer
- 30. PRE-BAKE Remove theresist solvent Convection Oven 90-100 ºC, 15-20 min removal starts at surface solvent trapping Conduction (hotplate) 75-85ºC, 40-60 sec removal starts at bottom uniform heating
- 31.
- 32. PHOTORESIST COATING Photoresist (anorganic polymer sensitive to UV light and resistant to attack by acids) is applied to the oxidized wafer using a photoresist spinner. This process uses centrifugal force from high speed rotation of the wafer. The PR is applied as a small puddle in the center of the wafer. When the wafer spins, the PR spreads out over the wafer due to centrifugal force. After spinning, a uniform layer of PR remains on the surface. ECE444 student dispensing photoresist onto an oxidized silicon wafer. Note the yellow cast to the picture – short wavelength light (green, blue, violet, and ultraviolet) exposes PR, so it has been filtered out of the room light, leaving only red, orange, and yellow to see with!
- 33.
- 34. SOFT BAKE Purposeof this is to remove solvents,promote adhesion,and harden the resist. Typical soft-bake temperatures are around 90 degree centigrade for 10-20 min. 34
- 35.
- 36. ALIGNMENT AND EXPOSURE 36 GateMask P-Well USGSTI Polysilicon Photoresist
- 37. ALIGNMENT AND EXPOSURE ThePR coated wafer is placed into a system (mask aligner or ‘stepper’) which allows the mask to be aligned to the wafer. After alignment, the system opens a shutter to allow UV light to illuminate the PR through the mask for a controlled period of time. The PR which is exposed to UV light undergoes a photochemical reaction to make the PR more acidic (indene carboxylic acid is ECE444 student loading PR coated wafer into an Ultratech 1000WF Step and Repeat Projection Alignment system (also known as a ‘stepper’).
- 38.
- 39.
- 40. 40 After the waferis exposed to UV light through the mask, the acidic regions of PR are removed by dipping the wafer into an alkaline (base) developing solution. The acidic PR reacts chemically with the basic developer to form water soluble salts that dissolve in the developer. At this point the mask image can be seen in the PR (remember that the PR was illuminated with UV light through the mask, so only light in the shape of the circuit reaches the PR – the rest of the PR did not change!). Note: the image from the mask has only been transferred to the PR. The PR will be used as a mask for etching the underlying oxide in an acid bath. DEVELOPMENT
- 41.
- 42. HARDBAKE Stabilize thedeveloped resist for subsequent processes Can make removal very difficult Remove residual solvent Not necessary for lift-off Temperature/time can change the profile
- 43. ETCH Etching removesthe SiO2 layer at selected regions where the resist has been removed. 43
- 44. STRIP RESIST Afteretching,the resist coating that remains on the surface must be removed. Stripping is accomplished using either wet or dry techniques. Wet stripping uses liquid chemicals.The nearer to the horizontal the surface the faster it etches.Therefore the protruding cusp is reduced faster than the nearly vertical sidewalls. 44
- 45.
- 47. 47 • Photolithography uses lightenergy passing through a patterned mask • The light is focused onto the photosensitive surface • Chemical changes in the surface coating occur • Subsequent chemical development creates a temporary pattern on the surface.
- 48.
- 49. DIFFUSION – CONTINUED Inthe periodic table, the closer elements are to each other, the more similar they are. So the best candidates would come from column V (which have five outer shell electrons). The element closest to silicon in column V is phosphorus. If phosphorus is inserted into the silicon wafer in a certain way, it will take the place of a silicon atom and bond with its four neighbor silicon atoms. After bonding, phosphorus has an electron left over that is not bonded to a silicon atom. It turns out this extra electron is not strongly held by the phosphorus atom any more, so it can be removed easily. This electron then becomes a ‘carrier’ for current – it is free to move around the wafer. So the conductivity of the silicon wafer increases. This type of silicon ‘doped’ with phosphorus is called an n- type semiconductor.
- 50. DIFFUSION Silicon is acolumn IV element – this means there are four electrons in the outermost shell of the atom. It is these electrons that are used when bonding to other atoms. In a wafer, each silicon atom bonds to four other silicon atoms (each Si-Si bond shares one electron). So in an intrinsic (pure) silicon wafer, all the electrons in the outer shell are part of a bond – they are ‘stuck’ between the bonded silicon atoms. In order for current to flow in a material there must be ‘loose’ electrons. But all the electrons in silicon are working at holding the atoms together, which means it is not a good conductor of current. So what can be done to allow the silicon to conduct current more easily? Free ‘carriers’ of current must be added. The goal is to find an element about the same size as a silicon atom so that it fits together well with the silicon, but with more electrons in its outer shell.
- 51. DIFFUSION – CONTINUED Extendingthis idea of inserting an element with a different number of valence electrons, a column III element (such as boron) could be added to the silicon wafer. In this case, the boron will try to bond with four silicon atoms, but it only has three electrons to bond with. This means there is an incomplete bond with one of the silicon atoms – a ‘hole’ where an electron would normally be. This ‘hole’ behaves much like an electron and can move around the wafer, but with an opposite charge (+). So a different type of current carrier is present in the wafer that increases the wafer’s conductivity. This type of silicon ‘doped’ with boron is called a p-type semiconductor. By adding impurities to silicon, the conductivity increases. This conductivity can be adjusted by the amount of impurity added.
- 52. DIFFUSION – CONTINUED Nowfor the interesting part - when n-type silicon comes into contact with p-type silicon. A built-in potential (voltage) develops that must be overcome before current can flow from the n-type to p-type regions. Think of carriers as being able to only move across a flat surface or down a slope. The built in potential is a hill that the carrier can not go up. So in order for the carrier to keep moving, the low part must be pushed up to be level or higher than the top of the hill. In the case of an n-type / p-type junction, the energy to push up the low side comes in the form of a voltage applied to the wafer. The voltage is used to ‘push up’ the ‘ground’ on the low side of the hill before current flows from n-type to p-type regions. But if the voltage is reversed, the energy is used to push the low side lower while keeping the high side at the same height! That means the carrier probably won’t ever make it up the higher hill, so it is stuck (no current flows).
- 53. DIFFUSION – CONTINUED Sowhen n-type silicon is brought into contact with p-type silicon (a pn junction), current can flow in only one direction. This is the fundamental semiconductor device – a pn junction diode – a one way switch for current. The devices used in integrated circuits are specialized combinations of pn junctions. The junctions are formed by the addition of impurity atoms from columns III and V of the periodic table into the silicon wafer through diffusion.
- 54. DIFFUSION -CONTINUED The goalof the dopant predeposition diffusion is to move dopant atoms from a source to the wafer, and then allow the dopants to diffuse into the wafer. The source of dopant can be in several forms – solid (boron nitride and phosphorus oxide ceramic discs), liquid (boron tribromide and POCl3), or gas (diborane or phosphine). In order for the dopants to move into the silicon, they must be given energy, usually in the form of heat. In order for the diffusion to occur in a reasonable time, the temperature must be very high (900ºC <T<1200º). At this temperature the dopant (in the form of an oxide) reacts with the exposed silicon surface to form a highly doped glass. It is from this glass that the dopants can then diffuse into the wafer. ECE444 Diffusion furnace
- 55. DOPANT DIFFUSION -DRIVE After the predeposition diffusion the dopants are situated close to the surface of the wafer. However, they must diffuse even farther to lower the overall concentration in order for some of the devices to work properly. The first diffusion (predeposition) introduces dopants into the wafer. The second diffusion (drive) redistributes the dopants and allow the dopants to diffuse into the wafer more deeply (up to ~3 micrometers) In addition, oxygen and water vapor are introduced during the drive diffusion to grow a new oxide over the areas which were exposed to bare silicon during the photolithography process. This new oxide can be patterned again so that other 100mm diameter wafer fabricated in the ECE444 laboratory following boron predeposition, boron drive, and re-oxidation.
- 56.