00:14
Logxen has quit [Ping timeout: 240 seconds]
00:26
Logxen has joined #homecmos
00:34
Logxen has quit [Ping timeout: 256 seconds]
00:34
Logxen has joined #homecmos
00:41
Logxen has quit [Ping timeout: 240 seconds]
00:59
Logxen has joined #homecmos
03:13
Logxen has quit [Changing host]
03:13
Logxen has joined #homecmos
06:39
Textmode has joined #homecmos
07:01
osfe has joined #homecmos
08:46
<
whitequark >
It's a metamaterial made by patterning silicone! We can do any amount of these right with very little issues
09:29
chris_99 has joined #homecmos
09:54
Textmode has quit [Quit: "It was one dev, naked in a room with a carton of cigarettes, a thermos full of coffee and bourbon, and all his summoned angels."]
10:07
<
SpeedEvil >
Shiny disco balls.
10:08
<
SpeedEvil >
Also - lolz.
10:08
<
SpeedEvil >
they usually require sophisticated designs with nonspherical surfaces that can be difficult to polish.
10:08
<
whitequark >
I wonder if it would be really easy to pattern that, even DIY
10:08
<
SpeedEvil >
I mean patterning a wafer is entirely trivial.
10:08
<
whitequark >
contamination doesn't matter
10:08
<
whitequark >
the pattern is just fucking circles
10:08
<
whitequark >
that sounds like an awesome idea for a test
10:08
<
SpeedEvil >
An errors mean that you simply lose contrast
10:09
<
SpeedEvil >
50% error means a 50% contrast lens
10:09
<
SpeedEvil >
Also, the above is clearly infrared only
10:09
<
SpeedEvil >
you'd need to work something else out for visible
10:09
<
whitequark >
"Photo of a high contrast transmitarray lens designed for red light (650nm)"
10:10
<
SpeedEvil >
The lens is made of over 140 million 360nm-tall amorphous silicon nano-posts.
10:10
<
whitequark >
oh, lithography
10:10
<
SpeedEvil >
that would make the posts 100nm or so in diameter
10:10
<
SpeedEvil >
50-100nm
10:10
<
SpeedEvil >
That seems challenging to do at home.
10:11
<
SpeedEvil >
In principle, linear features might work I think.
10:11
<
whitequark >
well, you could really settle for infrared, a CMOS camera can capture it
10:11
<
SpeedEvil >
not so much.
10:12
<
whitequark >
not so much to what part?
10:12
<
SpeedEvil >
For obvious reasons, the silicon photodetector doesn't work so well at wavelengths silicon is transparent in.
10:12
<
whitequark >
i have seen it with my own eyes
10:12
<
whitequark >
tell me it doesn't work XD
10:13
<
SpeedEvil >
You've seen imaging with a conventional camera through a silicon lens?
10:13
<
SpeedEvil >
I'm not saying conventional cameras don't pick up IR
10:13
<
whitequark >
yes, I got your point
10:13
<
SpeedEvil >
An if you can back-thin the wafer to 1um as they did in the paper, you can get it to work. That seems almost as challenging as etching
10:14
<
whitequark >
wait, no, if the camera has a silicon sensor that detects IR, how come silicon would be transparent to IR in that same rnage in the lens?
10:15
<
whitequark >
as for back-thinning the wafer, I was thinking of growing epitaxial silicon on something and then etching it
10:15
<
SpeedEvil >
Silicon is used as IR windows on thermal cameras at 10um - it's got decent ttransmissivity
10:15
<
SpeedEvil >
Also - I wonder if it would be possible to CVD actual glass or some transparent oxide into the required form
10:16
<
whitequark >
glass?
10:16
<
whitequark >
isn't it transparent?
10:17
<
whitequark >
weren't you just saying that being transparent would be an issue?
10:17
<
SpeedEvil >
It should work fine for this, though in a different manner
10:18
<
SpeedEvil >
If the silicon is transparent at the wavelength in question, then the light goes right through the detecor
10:21
<
_Sync_ >
whitequark: protip for epi growth you need a substrate
10:21
<
_Sync_ >
SpeedEvil: you can CVD PSG
10:22
<
whitequark >
_Sync_: right. dumb suggestion
10:22
<
whitequark >
what about using sputtering to make a film?
10:24
<
_Sync_ >
with Si? I actually cannot say what would happen
10:24
<
_Sync_ >
I suppose you'd get a mess
10:25
<
whitequark >
yeah, Si
10:25
<
whitequark >
I'm curious.
10:25
<
whitequark >
I'm going to try it, but I'll need to find a SEM somewhere else.
10:25
<
_Sync_ >
a sem is not all that useful, get an AFM
10:26
<
_Sync_ >
that has no problems with atom size resolution
10:26
<
whitequark >
wouldn't that only allow to peek at the surface of the film?
10:28
<
_Sync_ >
yup but after you got a single layer down you will get epi
10:28
<
whitequark >
hm makes sense
10:28
<
whitequark >
actually that sounds like a very interesting method, i bet there's something in the literature
10:29
<
_Sync_ >
well it's not too interesting for everything semi related
10:33
<
whitequark >
hm, from what i could find, people are growing actual high-quality epitaxial films
*with* sputtering
10:33
<
whitequark >
350°C, huh
10:34
<
_Sync_ >
sure but not without a si substrate
10:34
<
_Sync_ >
350°C sounds reasonable
10:37
<
whitequark >
hm, no one seems to be interested in sputtering silicon on anything except monocrystalline substrates
10:38
<
_Sync_ >
that is because it makes no sense
10:38
<
whitequark >
pffff, when did that stop academics
10:39
<
_Sync_ >
when they have to apply for grant money
10:39
<
whitequark >
hm, related question. what are the advantages to RF sputtering? no charge buildup?
10:40
<
whitequark >
or rather, apart from lack of charge buildup
10:41
<
_Sync_ >
you can sputter nonconductive materials
10:42
<
whitequark >
oh, right.
11:40
SpeedEvil has quit [Ping timeout: 272 seconds]
11:41
<
whitequark >
wait they quit
12:27
SpeedEvil has joined #homecmos
16:18
nmz787_i has joined #homecmos
16:24
nmz787_i has quit [Ping timeout: 252 seconds]
18:25
Logxen has quit [*.net *.split]
18:26
kristian1aul has joined #homecmos
18:26
kristianpaul has quit [Ping timeout: 244 seconds]
18:27
Logxen has joined #homecmos
19:57
Textmode has joined #homecmos
21:00
chris_99 has quit [Remote host closed the connection]
21:03
chris_99 has joined #homecmos
21:03
chris_99 has quit [Changing host]
21:03
chris_99 has joined #homecmos
21:18
Logxen has quit [Ping timeout: 256 seconds]
21:24
Logxen has joined #homecmos
21:29
Textmode has quit [Quit: "It was one dev, naked in a room with a carton of cigarettes, a thermos full of coffee and bourbon, and all his summoned angels."]
23:05
chris_99 has quit [Remote host closed the connection]