VLSI DESIGN:

Theory & Labs.

• Picture above shows a Si wafer to be checked under micro-scope.

• The different color on the wafer is due to reflection.
• The surface shows a repetition of pattern.
• Each pattern is a die when scribed and packaged one IC chip. Here approximately
36 chips.
• Each pattern is an electronic circuit with millions of devices
• Two types of design hierarchy
o Top Down Approach suitable for digital design
o Bottom Up Approach suitable for analog design
• Device used here could be FETs leading to CMOS technology
• Could be bipolar or MESFET
• In VLSI Circuit course and in the lab CMOS technology will be discussed

CMOS Æ Complimentary Metal Oxide Semiconductor

NMOS +PMOS complimentary of NMOS

NMOS 3D VIEW:

1
Schematic:

Biasing:

Creating a Channel for Current Flow

VDS

Current-Voltage Characteristics
• Subthreshold region: Gate voltage < Threshold voltage channel is depleted.
Channel current or drain current is negligible.
• Linear region: Gate votage – Threshold voltage > Drain Voltage
• Saturation Region: Gate Voltage – Threshold voltage < Drain Voltage.

2
PMOS Cross-Section:

3
Creating a Channel for Current Flow

Depletion Mode Devices

• Used in NMOS technology.

• Obsolete technology.
• Easier fabrication with depletion mode devices
• By doping the channel is made n-type. The device is normally on.

4
Schematic Design by SEDIT from TANNER Tools
CMOS Inverter
Operating point

W=12u

Include models L=5u

in out
W=8u

V=3.0 V=1.0 L=5u C=800ff

• SEDIT Circuit Description with symbols.

• TSPICE Circuit description in language.
• Can convert an SEDIT file to a TSPICE file.
• Simulation can be done.

W=20u W=20u

L=5u L=5u

A Out
W=22u

L=2u

nand1
B W=22u

L=2u

5
Device Cross-sections within
Fabrication Process

6
CMOS Process with Poly Gate Process
or Self Aligned Gate Process

Growth of gate dielectric in the window created by

active (gate+source+drain) mask

Poly etching by poly mask

NMOS whenever a
poly crosses active

7
Layout in LEDIT for NAND Gate

Growing Silicon Crystals

Czochralski method of crystal growth.

8
RCA Clean

Removal of:

• Organic contaminants e.g. grease

• Sacrificial oxide
• Ionic or heavy metal contaminants

Chemical Storage

Solution Preparation with DI water

9
Solution Bath

10
Set-Up Bubbler Rinse

Turn on the deionized water flow valve to the bubbler rinse station. Be careful that the
valve is not opened too far, as this could cause overflowing of the station.

Turn on the nitrogen flow to the rinse station. When correctly adjusted, there should be a
gentle bubbling of nitrogen through the rinse station.

Organic Clean

Once all the solutions are prepared and have been heated to there proper temperature, the
cleaning process can begin.

11
First the wafer are submerged in the Organic Solution for 15 minutes.

Then the wafers are placed in Bubbler Rinse for 5 minutes.

The Organic clean phase removes surface contaminants such as dust, grease and silica gel
from the wafer.

Oxide Clean

Next, the wafers are submerged in the Oxide Stripping solution for 15 seconds.

Then, the wafers are immediately placed into the Bubbler Rinse, where they are agitated
for 30 seconds.

12
The oxide clean removes any thin oxide layer which may have formed on the wafer
surface.

It is important that the wafers remain in the Bubbler Rinse no more than 30 seconds as
this could cause a new oxide layer to form.

Ionic Clean

The wafers are submerged in the ionic solution for 15 minutes.

Then, the wafers are placed in Bubbler Rinse for 5 minutes.

The ionic clean removes any heavy metal ions from the surface of the wafer.

Transporting Wafers

Following the ionic clean, the wafers must be transported to the dryer.

13
Before leaving the clean station, the teflon carrier containing the wafers is placed in a
black box and covered. This prevents the wafers from being re-contaminated.

The laminar flow of nitrogen under the clean station fume hood makes this area cleaner
than the general lab area.

Wafers should always be handled with teflon tweezers which will not scratch them or
leave deposits.

Drying

The final step in the RCA clean procedure is to dry the wafers.

First the wafers are loaded into the dryer.

14
Then the door is shut, and the start button is pressed.

Following the dry, the wafers are ready for the next process step.

It is important that whatever this next step is be performed immediately to ensure that the
wafers are as clean as possible.

Oxidation
Silicon Dioxide on the surface
Oxide Types:
• Gate Oxide
o Low leakage current
o High breakdown voltage
• Field Oxide
• Protective Oxide

An RCA Clean should be performed immediately prior to any oxidation process.

During the RCA Clean, the oxidation furnace should be set-up.

15
Furnace Preparation

First, verify that the Gas Flow Control Panel is set in the "Idle" configuration.

The Gas Flow Control Panel controls gas flow to three furnaces - the anneal furnace, the
boron furnace, and the oxidation furnace.

In the "Idle" configuration, nitrogen is flowing into all three furnaces in order to keep
them clean and dry.

16
Next, turn on the oxidation furnace by adjusting the Oxidation Furnace Temperature
Controller to 900 degrees C. This is accomplished by adjusting the furnace controller in
the back of the furnace clean hood.

[Note: This step is only required for Field Oxidation!] Then, fill the Bubbler Flask with
deionized water and turn the controller knob up to 50, then immediately back to 25.

This will start heating the water in the Bubbler Flask to a temperature between 90 and
100 degrees C.

Quartz Boat Removal

First, remove the Quartz End Cap from the end of the Quartz Diffusion Tube.

17
The Quartz Push Rod has a white mark about halfway down the rod. To avoid
introducing contamination into the furnace, Never touch the rod below the marking.

Loading Wafers Into Boat

Using teflon or teflon-tipped tweezers, carefully load the wafers into the Quartz Boat.

Avoid passing your arms or head over cleaned wafers.

Loading Boat Into Oven

With boat loader, return wafers on boat to the oxidation furnace.

18
In order to prevent thermal stress which could cause the wafers to fracture, allow the
wafers to sit just inside the mouth of the open furnace for two minutes.

Very slowly push the Quartz Boat into the mouth of the furnace using the Quartz Push
Rod. Push the quartz rod just to the point where the white mark on the Quartz Push Rod
is flush with the metal furnace door. This indicates that the center of the Quartz Boat is at
the center of the furnace.

Now place the Quartz End Cap on the end of the Quartz Duffusion Tube.

Then shut oven door.

Field Oxidation Example

The process is:

a. Load in N2 (Idle gas)

b. Dry O2 5 min.
c. Wet O2 120 min.
d. Dry O2 5 min.
e. Unload in N2 (Idle gas)

Gate Oxidation Example

The process is:

a. Load in N2 (idle gas)

b. Dry O2 45 minutes
c. Unload in N2 (idle gas)

Ramp the furnace controller up to the oxidation temperature of 1100 degrees C. Wait
until the temperature has stabilized (ramp mode is set at 15 degrees per minute).

Once the oxidation temperature has been reached, we will be performing 45 minutes of
dry oxidation. This should give us about 1000A of oxide.

To begin dry oxidation, set the gas control panel to the "Dry Oxidation" configuration.

19
In this configuration, dry oxygen is flowing into the oxidation furnace.

Check the Flow Meter, and adjust the flow rate to 2.5 scfh.

When 45 minutes have elapsed, turn oxide furnace valve to N2. The gas delivery system
is back to "Idle" configuration.

Finally, set the furnace controller to ramp back down to 900 degrees C.

Once again, this will take about 20 minutes.

Unloading Wafers
At this point, the oxidation has been completed.

20
Oxide Thickness Measurement

Once the wafers have cooled, the thickness of the oxidation layer can be measured using
an ellipsometer.

First, the wafer is placed on the chuck. The stage is adjusted so that the beam is reflecting
from the polarizer toward the analyzer for an accurate reading.

Wafer Storage
Upon completion of the field oxidation process, the cooled wafer should be stored in the
Nitrogen Flow Storage Area (assuming that no other processes are to be performed
during this session).

21
The Nitrogen Flow Storage Area is an enclosed cabinet which has nitrogen gas
circulating through it.
The circulating gas prevents dust from settling on the wafer surface during storage.

Photolithography
Photolithography is the process of transferring geometric shapes on a mask to the surface
of a silicon wafer. The steps involved in the photolithographic process are wafer
cleaning; barrier layer formation; photoresist application; soft baking; mask alignment;
exposure and development; and hard-baking.

Wafer Cleaning, Barrier Formation and Photoresist Application

In the first step, the wafers are chemically cleaned to remove particulate matter on the
surface as well as any traces of organic, ionic, and metallic impurities. After cleaning,
silicon dioxide, which serves as a barrier layer, is deposited on the surface of the wafer.
After the formation of the SiO2 layer, photoresist is applied to the surface of the wafer.
High-speed centrifugal whirling of silicon wafers is the standard method for applying
photoresist coatings in IC manufacturing. This technique, known as "Spin Coating,"
produces a thin uniform layer of photoresist on the wafer surface.

Positive and Negative Photoresist

There are two types of photoresist: positive and negative. For positive resists, the resist is
exposed with UV light wherever the underlying material is to be removed. In these
resists, exposure to the UV light changes the chemical structure of the resist so that it
becomes more soluble in the developer. The exposed resist is then washed away by the
developer solution, leaving windows of the bare underlying material. In other words,
"whatever shows, goes." The mask, therefore, contains an exact copy of the pattern which
is to remain on the wafer.

22
Negative resists behave in just the opposite manner. Exposure to the UV light causes the
negative resist to become polymerized, and more difficult to dissolve. Therefore, the
negative resist remains on the surface wherever it is exposed, and the developer solution
removes only the unexposed portions. Masks used for negative photoresists, therefore,
contain the inverse (or photographic "negative") of the pattern to be transferred. The
figure below shows the pattern differences generated from the use of positive and
negative resist.

Negative resists were popular in the early history of integrated circuit processing, but
positive resist gradually became more widely used since they offer better process
controllability for small geometry features. Positive resists are now the dominant type of
resist used in VLSI fabrication processes.

Soft-Baking

Soft-baking is the step during which almost all of the solvents are removed from the
photoresist coating. Soft-baking plays a very critical role in photo-imaging. The
photoresist coatings become photosensitive, or imageable, only after softbaking.
Oversoft-baking will degrade the photosensitivity of resists by either reducing the
developer solubility or actually destroying a portion of the sensitizer. Undersoft-baking
will prevent light from reaching the sensitizer. Positive resists are incompletely exposed
if considerable solvent remains in the coating. This undersoft-baked positive resists is
then readily attacked by the developer in both exposed and unexposed areas, causing less
etching resistance.

23
Mask Alignment and Exposure

One of the most important steps in the photolithography process is mask alignment. A
mask or "photomask" is a square glass plate with a patterned emulsion of metal film on
one side. The mask is aligned with the wafer, so that the pattern can be transferred onto
the wafer surface. Each mask after the first one must be aligned to the previous pattern.

Once the mask has been accurately aligned with the pattern on the wafer's surface, the
photoresist is exposed through the pattern on the mask with a high intensity ultraviolet
light. There are three primary exposure methods: contact, proximity, and projection. They
are shown in the figure below.

Contact Printing

In contact printing, the resist-coated silicon wafer is brought into physical contact with
the glass photomask. The wafer is held on a vacuum chuck, and the whole assembly rises
until the wafer and mask contact each other. The photoresist is exposed with UV light
while the wafer is in contact position with the mask. Because of the contact between the
resist and mask, very high resolution is possible in contact printing (e.g. 1-micron
features in 0.5 microns of positive resist). The problem with contact printing is that
debris, trapped between the resist and the mask, can damage the mask and cause defects
in the pattern.

Proximity Printing

The proximity exposure method is similar to contact printing except that a small gap, 10
to 25 microns wide, is maintained between the wafer and the mask during exposure. This
gap minimizes (but may not eliminate) mask damage. Approximately 2- to 4-micron
resolution is possible with proximity printing.

Projection Printing

Projection printing, avoids mask damage entirely. An image of the patterns on the mask
is projected onto the resist-coated wafer, which is many centimeters away. In order to
achieve high resolution, only a small portion of the mask is imaged. This small image
field is scanned or stepped over the surface of the wafer. Projection printers that step the

24
mask image over the wafer surface are called step-and-repeat systems. Step-and-repeat
projection printers are capable of approximately 1-micron resolution.

Development

One of the last steps in the photolithographic process is development. The figure below
shows response curves for negative and positive resist after exposure and development.

At low-exposure energies, the negative resist remains completely soluble in the developer
solution. As the exposure is increased above a threshold energy Et, more of the resist film
remains after development. At exposures two or three times the threshold energy, very
little of the resist film is dissolved. For positive resists, the resist solubility in its
developer is finite even at zero-exposure energy. The solubility gradually increases until,
at some threshold, it becomes completely soluble. These curves are affected by all the
resist processing variables: initial resist thickness, prebake conditions, developer
chemistry, developing time, and others.

Hard-Baking

Hard-baking is the final step in the photolithographic process. This step is necessary in
order to harden the photoresist and improve adhesion of the photoresist to the wafer
surface.

Pre-Bake
Remove the black box containing the wafer from the Nitrogen Flow Storage Area.

25
Place the wafer in a tin plate.

Preheat the Dehydration Oven to between 200 - 250 degrees C. Place the tin plate
containing the wafer into the Dehydration Oven for 30 minutes to "Pre-Bake" it.

The purpose of this step is to ensure that the wafer is completely dry.

Any moisture on the wafer surface would interfere with the photolithography process,
causing it to yield poor results.

Photoresist Application
Remove the tin plate containing the wafer from the Dehydration Oven, and place the tin
plate in the black box for transportation to the Photoresist Spinner.

Using teflon tweezers, place the wafer on the wafer chuck in the center of the Photoresist
Spinner.

After making sure the wafer is centered on the wafer chuck, press the button to lock on
the vacuum line.

The vacuum will secure the wafer to the chuck.

26
Using the Nitrogen Gun, blow the wafer surface to remove any dust particles.

Using an eye dropper, flood the wafer surface with primer.

Wait ten seconds, letting the primer dwell on the wafer surface so that it will allow good
adhesion between the photo resist and the wafer surface.

27
Then, press the spin button which will cause the wafer to spin at 5000 RPM.

Centrifical force will cause excess primer to move in waves away from the center of the
wafer. Hold the spin button down until the waves disappear.

At this point, the wafer is primed and ready to accept photo resist.

Note: We cannot photograph the photoresist actual being applied because white light
causes it to be exposed.

It is worth noting that all lighting in the lab is covered by yellow filters expressly to
prevent interference with the photolithography process.

In a manner similar to applying the primer, an eye dropper is also used to place
photoresist on the wafer. Apply the resist so that about 2/3 of the wafer surface is
covered.

It is not necessary to completely cover the surface because spinning will cause the
photoresist to distribute over the entire wafer, and the primer will ensure that it adheres
evenly.

Spin the wafer for 30 seconds.

Then, release the vacuum, and use Teflon tweezers to carefully remove the wafer and
place it back in the tin plate.

Soft Bake

Pre-heat the Soft-Bake Oven to the temperature specified by the type of photoresist you
are using. Typically this will be between 90-110 degrees C.
28
Place the tin plate containing the wafer into the Soft-Bake Oven for 30 minutes.

The purpose of the soft bake is to semi-harden the photoresist.

Mask Aligner Set-Up

While the wafer is in soft bake, you may set up the mask aligner.

For each substrate layer requiring photolithography, there is a seperate mask.

A mask is a patterned glass plate.

A set of masks cost between $500-$1500, depending on the number of layers and the
required precision.

Locate the appropriate mask for this process step.

29
Place the mask in the wafer holder so that the emulsion side of the mask will be facing
downward toward the wafer.

Slide the wafer holder into the mask aligner, and tighten the two knobs to secure it in
place.

Wafer Loading
After the wafer has been in the Soft-Bake Oven for 30 minutes, remove the tin plate
containing the wafer.

30
Using teflon tweezers, carefully place the wafer on the wafer chuck of the Aligner.

Always orient the wafer so that the flat side of the wafer is facing the same direction (to
the left, for example).

This is important because each mask layer after the first one will be precisely aligned
with the pattern formed by previous mask layers.

31
Carefully slide the wafer chuck into the aligner so that the wafer is positioned under the
mask.

Slowly raise the lever so that the wafer comes into contact with the mask.

A light will come on indicating that the wafer is in contact.

Then, press the button labeled "Seperation".

This will cause the wafer to move 0.5 microns away from the mask, far enough that it can
be moved without scratching the wafer.

Alignment
Adjust the microscope so that it focuses clearly on the mask and wafer.

32
There are three knobs which allow you to move the wafer in the horizontal and vertical
direction, and also to adjust the angle.

If this is the first mask layer, it is only important that the mask area completely cover the
wafer so that devices are not lost.

For subsequent layers, the pattern on the wafer must be precisely aligned to the mask.

To make this posible, there are markings on each layer to guarantee proper alignment.

As a general strategy, first align the wafer so that rows of devices on the substrate are
parallel to rows of devices on the mask.

To accomplish this, it will be necessary to iteratively move back and forth accross a row,
adjusting the angular position and the vertical position until the angular component is
correct.

Once the angular component is correct, it should be easier to align the horizontal and
vertical markers.

Exposure
When the wafer has been properly aligned to the mask, press the button labeled
"Contact".

This will cause the wafer to once again be in direct contact with the mask.

33
The exposure time should be set according to the particular type of photoresist and
wattage of the bulb being used.

In our example, the exposure time is 10 seconds.

Gently but firmly press the expose button.

A very bright white light will come on inside the aligner, exposing the wafer through the
mask for the specified amount of time.

For safety reasons, avert your eyes from the aligner during exposure.

It is worth noting that the aligner is sitting on a stabilizer table. Essentially, each leg of
the table has its own nitrogen shock absorber, minimizing the chance that there will be
any disturbing movement during exposure.

Once exposure is complete, the wafer should be released from the contact position.

Then, the chuck should be slid out and the wafer gently placed back in the tin plate and
carried to the chemical sink, where it will be developed.

Development
The type of developer solution used is determined by the type of photoresist chosen.

34
Fill the pyrex beaker about 2/3 full with developer solution. Fill another beaker about 2/3
full with deionized water.

Using teflon tweezers, carefully place the wafer into a chemically resistant carrier.

Be sure to lock the wafer in so that it is firmly secured.

Check your developer for the recommended development time. Typically, this will be
around 30 seconds.

35
Set a timer, then immerse the wafer in the developer and agitate mildly until the time has
expired.

Then, immediately plunge the wafer into the beaker filled with water to stop the
development process.

Pull the wafer out of the deionized water beaker, and use the deionized water gun to
completely rinse the wafer (front, back and also the carrier) for about 30 seconds.

36
Use the nitrogen gun to thoroughly dry the wafer, front and back.

Then, use teflon tweezers to carefully remove the wafer from the chemically resistant
wafer carrier, and place it back in the tin plate.

Hard Bake

The final step is to inspect the wafers under a microscope, to see if the patterns are clear
and well defined.
37
The final step is to inspect the wafers under a microscope, to see if the patterns are clear
and well defined.

Since this is something that takes some experience to know what you are looking for, our
lab is equipped with a dual microscope so that both the instructor and the student can
view the wafer at the same time.

If everything looks okay, the wafer is placed into the Hard Bake oven which should be
preheated to between 120-130 degrees C.

The wafers should remain in the hard bake oven for 30 minutes. This prepares the wafer
for the next processing step.

38
Doping
Doping is the addition of controlled amounts of specific impurity atoms for the purpose
of changing the conductivity of a semiconductor sample by increasing either the electron
or hole concentration. Some common dopants for silicon are listed in the table below.

Dopant atoms can either be "donors" or "acceptors." Donors increase the elctron
concentration in the silicon, whereas acceptors increase the hole concentration. This is
due to the fact that silicon, being an element from Column IV of the Periodic Table of the
Elements, has four valence electrons. In an undoped silicon crystal, each atom shares
these electrons with its four nearest neighbors via covalent bonding (see Figure 1).

When Column V elements such as arsenic or phosphorus are introduced into the silicon
crystal lattice, four of the five valence electrons fit into the silicon bonding structure. The
fifth electron, however, is weakly bound, and at room temperature is free to move
through the crystal (Figure 2a). It is thus said to be "donated" to the semiconductor.

Likewise, Column III elements such as boron have only three valence electrons, and
therefore cannot complete all four of the available bonds when substituted for a silicon
atom. These elements, however, readily "accept" an electron from a nearby silicon-silicon

39
bond. This creates a hole which is also free to wander through the crystal lattice at room
temperature (Figure 2b).

Diffusion
P-Type Diffusion

Oxidize Boron Wafers

Before using the BN source wafers the first time, it is necesary to oxidize them. (This is
not necessary for every use, as the sources are compromised if oxidized too frequently.
Previously oxidized wafers are stored in the boron furnace, which should be idling at 600
degrees C).

The instructor will have prepared the souce wafers, if required, using the following
procedure.

40
Oxidize BN wafers at 900 degress C in 25% O2 for approximatedly six hours. This is not
necessary for every use, as the sources are compromised if oxidized too frequently. After
oxidation, the BN wafers need to be stabalized in N2 in the furnace for at least 30
minutes. The source wafers should be stored at the mouth of the furnace to avoid
hydration.

RCA Clean and Furnace Check

An RCA Clean should be performed immediately prior to the diffusion process step.

Verify that that the Boron Furnace temperation is set to 600 degrees C.

41
And that the Gas Flow Control Panel is set in the "Idle" configuration.

The Gas Flow Control Panel controls gas flow to three furnaces - the anneal furnace, the
boron furnace, and the oxidation furnace. In the "Idle" configuration, nitrogen is flowing
into all three furnaces in order to keep them clean and dry.

Load Wafers Into Boat

At completion of RCA clean, load wafers into boat with sources.

Always wear chemically resistant poly gloves and/or high temperature gloves when
handling quartzware

42
Using teflon or teflon-tipped tweezers, carefully load the wafers into the Quartz Boat.

The wafers should be inserted such that the device side of each is facing a BN source
wafer. There are two wafer slots between sources, allowing for a device wafer facing
both sides.

Avoid passing your arms or head over cleaned wafers.

43
Load Boat Into Oven

Return quartz boat, which now contains the wafers, to the boron furnace using the quartz
boat loader.

Load the boat into the mouth of the boron furnace.

Pre-Deposition Example

The boron furnace needs to be heated to the temperature at which the diffusion will be
performed. For our example, turn the furnace controller up to the deposition temperature
of 935 degrees C. Wait until the temperature has stabilized (about 10-15 minutes).

44
Begin nitrogen flow through the tube by setting the gas flow rate to 1000 sccm (standard
cubic centimeter per min)

Wafers should be loaded in teh wafer boat, between each solid source of boron and
placed in the mouth of the furnace tube.

Using the push rod, push the waters into the center of the tube using a continuous
movement until you reach the center zone.

Maintain wafers in the center of the tube at 935 degrees C for 30 minutes.

When 30 minutes have elapsed, boron deposition is completed.

45
Using the pushrod return wafers back to the mouth of the furnace using a continuous
movement.

Finally, set the furnace controller to ramp back down to 600 degrees C. Wait for the
furnace to reach 600 degrees. This will take about 10-11 minutes.

Allow the wafers to cool down in the mouth of the tube for at least five minutes before
removing.

Unload Wafers

Remove the boat from the furnace and allow the wafers to continue to cool before
removing them from the wafer carrier.

The wafers are now covered with borosilicate glass, which served to dope the exposed
source/drain regions and the polysilicon. Other areas were masked from diffusion by the
field oxide.

When wafers are cool, place them in the RCA wafer carrier and dip them back into the
HF bath for 15 seconds and rinse for 5 minutes. This is to remove outer boron film from
surface that is no longer needed.

Note: The sheet resistance will increase when the Boron is removed from the surface
because the boron doping tends to increase conductivity (and therefore reduce resistivity).
When this boron is removed, the conductivity decreases and resistivity (as well as sheet
resistance) goes up.

Wafer Storage

Upon completion of the diffusion process, the cooled wafer should be stored in the
nitrogen flow storage area (assuming that no other processes are to be performed during
this session).
46
The nitrogen flow storage area is an enclosed cabinet which has nitrogen gas circulating
through it. The circulating gas prevents dust from settling on the wafer surface during
storage.

Metallization
Metallization is the fabrication step in which proper interconnection of circuit elements is
made. Aluminum is a popular metal used to interconnect ICs, both to make ohmic contact
to the devices and connect these to the bonding pads on the chip's edge. Aluminum
adheres well to both silicon and silicon dioxide, can be easily vacuum deposited (since it
has a low boiling point), and has high conductivity. In addition to pure aluminum, alloys
of aluminum are used to form IC interconnections for different performance-related
reasons. For example, small amounts of copper are added to reduce the potential for
electromigration effects (in which current applied to the device induces mass transport of
the metal). Small amounts of silicon also are added to aluminum metallization to reduce
the formation of metal "spikes" that occur over contact holes.

Metal layers are vacuum-deposited onto wafers by one of the following methods:

1. Filament Evaporation
2. Flash Evaporation
3. Electron-beam Evaporation
4. Sputtering

In this lab, a Denton Vacuum DU-502A Evaporator is used to perform Filament

Evaporation.

47
The Evaporation Process

The Evaporation process can be devided into three steps. 1) The solid aluminum metal
must be changed into a gaseous vapor. 2) The gaseous aluminum must be transported to
the substrate. 3) The gaseous aluminum must condense onto the substrate.

This process requires high temperature, and low pressure. The chamber of the Denton
Vacuum must be "pumped down" in order to achieve the low pressure requirement. The
pressure inside of the chamber is determined by the equation below.

Where P is the chamber pressure at time t, Po is the initial pressure, S is the pumping
speed, Q is the rate of "outgassing", and V is the volume of the chamber.

The concentration of gaseous aluminum molecules in the vacuum chamber can be

determined using:

Knowing the chamber pressure, one can calculate the impingement rate of the gaseous
aluminum hitting the substrate, the time required for one monolayer of aluminum to
deposit on the wafer surface, and the approximate final thickness of the aluminum on the
substrate.

The Impingement Rate is given by:

The time required for one monolayer of aluminum to deposit is given by:

The final thickness of the aluminum is given by:

The evaporation process doesn't produce a uniform layer of aluminum across the
substrate. The deposition rate changes as one moves radially from the center of the
48
substrate. This radial dependence of the deposition rate is described by the following
equation:

Filament Evaporation

Filament evaporation is accomplished by gradually heating a filament of the metal to be

evaporated. This metal may come in one of several different forms: pellets, wire, crystal,
etc. Gold, platinum and aluminum are metals typically used. The PMOS process uses
aluminum wire.

The metal is placed in a basket. Electrodes are connected to either side of the basket and a
high current passed through it, causing the basket to heat. As power (and therefore heat)
is increased, the metallic filament partially melts and is eventually vaporized. In this way,
atoms of aluminum break free from the filament and deposit onto the wafers. While
filament evaporation is the simplest of all metallization approaches, problems of
contamination during evaporation preclude its widespread use in IC fabrication.

Flash Evaporation

Flash evaporation uses the principle of thermal-resistance heating to evaporate metals.

Like filament evaporation, flash evaporation offers radiation-free coatings. This
technique does offer some benefits beyond filament evaporation: contamination-free
coatings, speed or good throughput of wafers, and the ability to coat materials or layers
that are composite in nature. Flash evaporation is accomplished by passing a continuous
supply of the material to be evaporated over a ceramic structure that provides heat. The
sources are usually either powder of thin wires. The ceramic "flash" bar is heated
between two positively and negatively charged posts, and the metal evaporates as the
source material is fed onto the bar.

49
Electron-beam Evaporation

Electron-beam evaporation works by focusing an intense beam of electrons into a

crucible, or "pocket," in the evaporator that contains the aluminum. As the beam is
directed into the source area, the aluminum is heated to its melting point, and eventually,
evaporation temperature. The benefits or this technique are speed and low contamination,
since only the electron beam touches the aluminum source material.

Sputtering

Aluminum sputtering is used commonly in IC metallization processes and is popular

because the adhesion of deposited metals is excellent. RF sputtering is done by ionizing
inert gas particles in an electric field (producing a gas plasma) and then directing them
toward the source or target, where the energy of these gas particles physically dislodges,
or "sputters off," atoms of the source material. Sputtering is a versatile tool in that many
materials can be deposited by this technique, using not only RF but also DC power
sources.

PAN Etch

The aluminum layer is etched using a PAN etch solution. PAN stands for "Phosphoric -
Acetic - Nitric." These are the three acids which are mixed togther to form the etch
solution.

The following steps make up the PAN Etch process:

• Prepare Chemical Solution

• Heat Chemical Solution
• Etch Wafer
• Rinse Wafer

50
Prepare Chemical Solution

The acids which make up the PAN etch solution are mixed according to the following
formula (by volume).

Water 2 parts
Phosperic Acid 16 parts
Acetic Acid 1 part
Nitric Acid 1 part

After the pan etch solution is made the remaining amount of solution can be stored in this
bottle.

If there is some Pan Etch solution in the Pan Etch bottle, use that before making more.

51
Heat Chemical Solution

Pour the PAN etch solution into the glass containter marked "PAN Etch".

Heat the PAN etch solution to between 40 and 45 degrees C.

52
Etch Wafer

Immerse the wafer in the solution and agitate. You will begin to see metal bubble up and
lift off the surface. Conutinue to agitate for one minute AFTER you no longer see metal
moving away from the wafer surface.

53
Rinse Wafer

Rinse the wafer with water, then blow it dry with nitrogen.

After it is dry, check under the microscope to make sure that all the necessary Aluminum
has been removed to prevent shorting.

54
Sintering

Sintering/Annealing is a very simple step consisting only of a heat cycle. This step is
used to help the aluminum adhere to the silicon substrate and to heal any substrate
damage caused by plasma etching.

Before the Sinter/Anneal procedure is begun, verify that the gases flowing through the
furnaces are in the IDLE CONFIGURATION.

Note: In the idle configuration, nitrogen gas is flowing through all three furnaces.

Set temp to furnace controller marked sintering / anneal at 400 degrees C.

Remove quartz boat from mouth of the funace using sinter push rod.

Carefully slide boat onto boat carrier and place on table under laminar hood.

Mount all wafers that are to be annealed.

Slide wafer carrier carefully back into the mouth of the furnace.

Wait 1 minute for wafers to adjust to furnace temp.

Slide boat using quartz rod up to marked line. This will place the wafer carrier in the
center of the quartz tube.

55
Set timer for 30 minutes.

When time is up, slide quartz boat back to mouth of the furnace.
56
Wait for the wafers to cool in the mouth of the furnace 1 minute.

Carefully remove wafer carrier and unload wafers.

57
Device Characterization

In this module, the devices are tested to compare their performance to expected results.
The following steps are involved in device characterization:

• Connect Probes
• Determine Threshold Voltage and Characterize Transistors
• Connect Probes

Load the wafer onto the chuck, making sure the probes are raised (out of the
way).

58
After the wafer is on the chuck, turn on the vacuum to secure the wafer.

Using the microscope, select a device to characterize. To characterize a resistor

only two probes are needed; for a transistor, three are needed.

59
To lower the probes, turn the knobs on top of the probe counter-clockwise.

To raise the probes, turn the knobs on top of the probe clockwise.

You can tell when the probes are on the pad because they will begin to slide
across it. You don't want to slide to probes across the pad excessively, since this
will ruin the contact pad, possibly causing erroneous results. It also bends the
probe tips.

The probes, labeled "S","D", and "G", can be seen above.

To characterize a resistor, place the probe labeled "S" on one contact pad located
next to the resistor you want to characterize. Place the probe labeled "D" on the
contact pad located on the other side of the resistor. The probe labeled "G" isn't
used in resistor characterization.

To characterize a transistor, place the probe labeled "G" on the gate of the
transistor, and the "D" and "S" probes on the drain and source contact pads of the
transistor (The drain and source are actually interchangeable on discrete
transistors).

60
Measure Resistance

Make sure that the curve tracer is on. To measure resistance, set the dial on step-
gen, and you should see a I-V (current-voltage) curve. Adjust the scales until you
get a typical characteristic.

61
Determine Threshold Voltage

To determine the threshold voltage set the dial to open and increase the power
supply voltage until you can see the transistor turn on. The power supply is
located above the curve tracer.

Once you know the threshold voltage, set the dial on the curve tracer to "Step-
Gen" to get a family of curves for that particular transistor. You may need to
adjust the offset-voltage on the curve tracer if your transistor has a high threshold
voltage. Adjust the scale on the curve tracer to get nice-looking curves.

62
These are typical plots of a I-V characteristics for a PMOS transistor ...

... and a resistor.

63
Print Out Results

Once you get a represenitive plot of the device characteristics, move the probe to
the "printing" curve tracer. Adjust the settings to get a good plot, then print the
results.

64