Scribd
Upload
8 views4 pages

hw4 2017 Soln

Uploaded by

Anand Gupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views4 pages

hw4 2017 Soln

Uploaded by

Anand Gupta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Homework #4 Solutions, EE/MSE 486, Spring 2017

Problem 1
From simulation result of problem 3 of homework 3, we can get the average oxidation rate for the first
half hour, dxO/dt=13.6nm/0.5hr=0.027um/h; Also for second hour the oxidation rate is dxO/dt=(31.6nm-
13.6nm)/(2hr-0.5hr)=0.012um/h.
From the Fig. 11 in the supplement notes for oxidation, we have for
𝑑𝑥𝑂 0.027𝜇𝑚/ℎ 𝐷 4.3 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 30 𝑚𝑖𝑛𝑠
={ → ∗={
𝑑𝑡 0.012𝜇𝑚/ℎ 𝐷 3.9 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑛𝑎𝑙 90 𝑚𝑖𝑛𝑠
The above diffusion enhancement is for phosphorus, so we have
𝐷 𝐷𝑃 𝐶 𝐶
𝐷∗
= 𝐷𝑝∗
= 𝑓𝐼𝑃 𝐶 𝐼∗ + 𝑓𝑉𝑃 𝐶𝑉∗ , 𝑤ℎ𝑒𝑟𝑒 𝑓𝐼𝑃 + 𝑓𝑉𝑃 = 1 (1)
𝐼 𝑉

Since the I/V recombine is assumed near equilibrium, so


𝐶𝑉 𝐶𝐼∗
𝐶𝐼 ∗ 𝐶𝑉 = 𝐶𝐼∗ ∗ 𝐶𝑉∗ → = (2)
𝐶𝑉∗ 𝐶𝐼

Combine (1) and (2), we can get


𝐶𝐼 4.5 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 30 𝑚𝑖𝑛𝑠
∗ = {4.1 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑛𝑎𝑙 90 𝑚𝑖𝑛𝑠
𝐶𝐼
For Arsenic, fIAs=0.4, so we have
𝐷𝐴𝑠 𝐶𝐼 𝐶𝐼∗ 1.93 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 30 𝑚𝑖𝑛𝑠
∗ = 0.4 ∗ + 0.6 ={
𝐷𝐴𝑠 𝐶𝐼 𝐶𝐼 1.79 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑛𝑎𝑙 90 𝑚𝑖𝑛𝑠

For Antimony, fIAs=0.05, so we have


𝐷𝐴𝑠 𝐶𝐼 𝐶𝐼∗ 0.44 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 30 𝑚𝑖𝑛𝑠
∗ = 0.05 ∗ + 0.95 ={
𝐷𝐴𝑠 𝐶𝐼 𝐶𝐼 0.44 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑓𝑖𝑛𝑎𝑙 90 𝑚𝑖𝑛𝑠

In Sentaurus simulation, the equilibrium interstitial concentration at 1000C is 7.7*1011 cm-3.


The output of timedependent interstitial concentration during oxidation is shown below:
For first 30 minutes oxidation at 1000 C, the average interstitial concentration is about 2.2*1011 cm-3, and
CI/CI* =2.85.
For final 90 minutes oxidation at 1000 C, the average interstitial concentration is about 1.6*1011 cm-3, and
CI/CI* =2.08.

Problem 2
According to the Range Statistics plot, we can get Rp and ΔRp for SiO2 and photoresist for B
implantation with 5keV.
For SiO2, Rp=18 nm, and ΔRp=11nm
For photoresist, Rp=40nm, and ΔRp=11nm
The relation between penetrated dose and film thickness is
𝑄 𝑥𝑚 − 𝑅𝑃
𝑄𝑃 = 𝑒𝑟𝑓𝑐( )
2 √2Δ𝑅𝑃
In this problem, Q=1014cm-2, and QP=2*1012cm-2.
𝑥𝑚 − 𝑅𝑃 𝑥𝑚 − 𝑅𝑃
𝑒𝑟𝑓𝑐 ( ) = 0.04 => ( ) = 1.45
√2Δ𝑅𝑃 √2Δ𝑅𝑃
For SiO2, the above equation can be solved to get xm=40.5nm
For photoresist, the above equation can be solved to get xm=62.5nm
The Sentaurus input is as follows:
line x location=0 spacing= 0.001 tag=SiDevTop
line x location=1 spacing=0.01
line x location=5 spacing=0.01 tag=SiDevBot
region silicon xlo=SiDevTop xhi=SiDevBot
init concentration=1e14<cm-3> field=Phosphorus wafer.orient=100

deposit material=photoresist type=isotropic thickness=0.0974


grid remesh

implant Boron energy=5<keV> dose=1e14<cm-2> tilt=7

strip photoresist
diffuse temperature=750<C> time=1<s>
#diffuse temperature=1000<C> time=0.5<s>

struct tdr=p1
select z=BTotal
layers

The simulated thickness to obtain required penetrated dose is 42.5 nm for oxide and 97.4 nm for
photoresist.

Problem 3
If the beam is aligned with the channels, then nuclear stopping will be minimized and electronic
stopping will dominate. The range will be

1 E dE 1 E dE
N 0 Se E  N 0 k E
R 

3
The value of k is given by k  0.2  10 eV cm while N  5  10 cm is the atom density in
15 12 2 22

silicon, so that the range is given by

𝑅 ≈ 20 √𝐸 𝐴𝑛𝑔𝑠𝑡𝑟𝑜𝑚𝑠 ≈ 282 𝑛𝑚
This is about four times the range from the range tables (Rp=74 nm and delta Rp=31nm),
because there is a nuclear stopping component which contributes along with the electronic
stopping. However, the calculation does indicate the range of a well-channeled ion.
For a 1014cm-2 dose implantation, the implantation profile with 50% ions channeled is shown
below:

You might also like