1
00:00:21,920 --> 00:00:26,079
Speaker: Pick up your phone, tap
the glass.

2
00:00:26,079 --> 00:00:29,440
It responds flawlessly.

3
00:00:29,440 --> 00:00:34,240
We communicate through them and
they tell us where to drive.

4
00:00:34,240 --> 00:00:37,200
We use them as flashlights.

5
00:00:37,200 --> 00:00:43,600
They function so seamlessly
that we stop wondering.

6
00:00:43,600 --> 00:00:48,079
We accept them as polished
little black boxes.

7
00:00:48,079 --> 00:00:54,479
Generally, unless you like to
take electronics apart, you

8
00:00:54,479 --> 00:00:57,119
don't ever see into your
devices.

9
00:00:57,119 --> 00:00:59,840
But you've heard of Silicon
Valley.

10
00:00:59,840 --> 00:01:03,759
You know there's silicon in
there, somewhere.

11
00:01:03,759 --> 00:01:06,640
And silicon is in sand.

12
00:01:06,640 --> 00:01:09,200
Quartz sand.

13
00:01:09,200 --> 00:01:14,799
But you can't shovel sand off
the beach and turn it into the

14
00:01:14,799 --> 00:01:16,799
hardware that runs your phone.

15
00:01:16,799 --> 00:01:20,640
The physical reality of the
digital world requires us to

16
00:01:20,640 --> 00:01:23,200
turn sand into something
flawless.

17
00:01:23,200 --> 00:01:28,400
It requires light that is
barely possible.

18
00:01:28,400 --> 00:01:32,879
And it all started with a
mistake.

19
00:01:32,879 --> 00:01:35,760
I'm Daina Bouquin.

20
00:01:35,760 --> 00:01:38,959
This  is Found in the Machine.

21
00:01:38,959 --> 00:01:44,560
The year is 1916.

22
00:01:44,560 --> 00:01:49,280
The world outside is tearing
itself apart in the trenches of

23
00:01:49,280 --> 00:01:50,879
the First World War.

24
00:01:50,879 --> 00:01:56,799
But inside a metallurgy
laboratory in Berlin, it is

25
00:01:56,799 --> 00:01:57,439
quiet.

26
00:02:03,599 --> 00:02:05,200
Jan Czochralski is working late. A
Polish man.

27
00:02:05,200 --> 00:02:08,159
He is a self-taught chemist.

28
00:02:08,159 --> 00:02:13,520
He has spent the entire day and
most of the evening staring at

29
00:02:13,520 --> 00:02:18,879
metals, heating them, cooling
them, trying to understand the

30
00:02:18,879 --> 00:02:21,840
exact speed at which they
crystallize.

31
00:02:21,840 --> 00:02:26,719
The kind of repetitive
observation that blurs the lines

32
00:02:26,719 --> 00:02:27,919
of your vision.

33
00:02:27,919 --> 00:02:31,599
On his desk there is an
inkwell.

34
00:02:31,599 --> 00:02:36,479
Beside it, there is a small
crucible left over from a

35
00:02:36,479 --> 00:02:37,680
previous experiment.

36
00:02:37,680 --> 00:02:42,879
It contains a cooling pool of
molten tin.

37
00:02:42,879 --> 00:02:46,479
Jan reaches for his fountain
pen.

38
00:02:46,479 --> 00:02:50,960
Without looking away from his
papers in front of him, he dips

39
00:02:50,960 --> 00:02:55,280
the nib of the pen, but his hand
misjudges the distance.

40
00:02:55,280 --> 00:02:58,960
He dips the pen into the
crucible.

41
00:02:58,960 --> 00:03:01,840
It goes straight into the
liquid tin.

42
00:03:01,840 --> 00:03:06,000
Realizing his mistake, he
quickly pulls the pen upward.

43
00:03:06,000 --> 00:03:10,719
Then he stops, because
something catches the flickering

44
00:03:10,719 --> 00:03:11,680
light.

45
00:03:11,680 --> 00:03:17,360
A silvery, hair-thin,
shimmering wire is hanging from

46
00:03:17,360 --> 00:03:18,560
the tip of his pen.

47
00:03:18,560 --> 00:03:22,560
He doesn't just wipe it away.

48
00:03:22,560 --> 00:03:24,240
He looks.

49
00:03:24,240 --> 00:03:25,919
He wonders.

50
00:03:25,919 --> 00:03:29,360
Then he does it again.

51
00:03:29,360 --> 00:03:34,159
The pen goes back into the tin
and slowly, deliberately this

52
00:03:34,159 --> 00:03:36,879
time, he pulls it up again.

53
00:03:36,879 --> 00:03:38,960
The thread grows.

54
00:03:38,960 --> 00:03:45,439
The tiny slit in his pen's nib
is acting like a capillary.

55
00:03:45,439 --> 00:03:49,360
The liquid metal is being
pulled up and is cooling the

56
00:03:49,360 --> 00:03:52,960
moment it touches the air,
clinging to itself as it rises.

57
00:03:52,960 --> 00:03:56,879
And this isn't just a wire.

58
00:03:56,879 --> 00:04:01,120
As Jan pulled the pen, the
atoms in the pool of tin didn't

59
00:04:01,120 --> 00:04:04,400
scramble into a chaotic jumble
as they cooled.

60
00:04:04,400 --> 00:04:06,560
They aligned.

61
00:04:06,560 --> 00:04:12,879
This wire is a single,
unbroken, continuous crystal

62
00:04:12,879 --> 00:04:13,680
lattice.

63
00:04:13,680 --> 00:04:16,160
It is perfect.

64
00:04:16,160 --> 00:04:21,759
Jan Czochrolski has just
invented a way to grow flawless

65
00:04:21,759 --> 00:04:22,720
crystals.

66
00:04:22,720 --> 00:04:26,240
He wrote a paper about it.

67
00:04:26,240 --> 00:04:30,319
He called it a new method for
measuring the crystallization

68
00:04:30,319 --> 00:04:31,439
rate of metals.

69
00:04:31,439 --> 00:04:35,439
But he had no idea what he had
actually done.

70
00:04:35,439 --> 00:04:37,680
He would never know.

71
00:04:37,680 --> 00:04:46,480
There are two streams of
materials that must collide to

72
00:04:46,480 --> 00:04:48,399
make a microchip possible.

73
00:04:48,399 --> 00:04:52,079
First, there is the silicon
that will become the chip

74
00:04:52,079 --> 00:04:52,720
itself.

75
00:04:52,720 --> 00:04:57,759
It begins its life as
metallurgical grade silicon,

76
00:04:57,759 --> 00:05:02,399
forged by taking high-purity
quartz, which is a rock mostly

77
00:05:02,399 --> 00:05:06,959
mined in deposits from China,
Norway, and Brazil, and throwing

78
00:05:06,959 --> 00:05:10,160
it into an electric arc furnace
with carbon.

79
00:05:10,160 --> 00:05:15,040
It is blasted at up to 2,000
degrees Celsius.

80
00:05:15,040 --> 00:05:21,439
Then it is chemically purified
over and over again until it

81
00:05:21,439 --> 00:05:24,879
reaches what engineers call the
9N level.

82
00:05:24,879 --> 00:05:33,920
That's 99.9999999% pure.

83
00:05:33,920 --> 00:05:39,439
Out of one billion atoms, only
one can be an impurity.

84
00:05:39,439 --> 00:05:44,160
But having the purest liquid
silicon on earth is only the

85
00:05:44,160 --> 00:05:45,120
first step.

86
00:05:45,120 --> 00:05:48,639
You also need to hold that
perfect liquid without

87
00:05:48,639 --> 00:05:50,079
contaminating it.

88
00:05:50,079 --> 00:05:54,399
You need a very, very special
crucible.

89
00:05:54,399 --> 00:05:59,439
The crucible you need does not
come from a global supply chain.

90
00:05:59,439 --> 00:06:04,160
Overwhelmingly, it comes from a
single town in the Blue Ridge

91
00:06:04,160 --> 00:06:04,959
Mountains.

92
00:06:04,959 --> 00:06:08,639
Spruce Pine, North Carolina.

93
00:06:08,639 --> 00:06:13,759
Through a geological anomaly,
Spruce Pine holds the purest

94
00:06:13,759 --> 00:06:17,279
natural quartz ever discovered
on Earth.

95
00:06:17,279 --> 00:06:21,920
So you have the ultra-pure
silicon gathered from across the

96
00:06:21,920 --> 00:06:22,560
globe.

97
00:06:22,560 --> 00:06:27,040
You have the perfect quartz
crucible pulled from the

98
00:06:27,040 --> 00:06:28,959
mountains of North Carolina.

99
00:06:28,959 --> 00:06:35,040
The silicon is molten inside
that pristine vessel.

100
00:06:35,040 --> 00:06:38,079
Now, how do you get it out?

101
00:06:38,079 --> 00:06:43,920
How do you turn it into a
pristine canvas for microchips?

102
00:06:43,920 --> 00:06:48,879
You use the method that Jan
Czochrolski accidentally

103
00:06:48,879 --> 00:06:52,319
invented when he was too tired
to notice where he put his pen.

104
00:06:52,319 --> 00:06:59,040
You see, in the late 1940s and
early 1950s, as researchers

105
00:06:59,040 --> 00:07:03,439
tried to make reliable
transistors, they kept running

106
00:07:03,439 --> 00:07:04,879
into the same wall.

107
00:07:04,879 --> 00:07:09,759
They needed perfectly ordered
semiconductor material. A single

108
00:07:09,759 --> 00:07:13,519
flaw, a single microscopic
misalignment in the

109
00:07:13,519 --> 00:07:17,040
crystal lattice, and the
electrical signals would scatter

110
00:07:17,040 --> 00:07:18,000
and fail.

111
00:07:18,000 --> 00:07:21,120
They needed a perfect canvas.

112
00:07:21,120 --> 00:07:24,480
And they did not know how to
make one.

113
00:07:24,480 --> 00:07:29,600
Until they found an obscure
paper from 1916.

114
00:07:29,600 --> 00:07:34,000
Today, nearly all silicon
wafers on Earth are made using

115
00:07:34,000 --> 00:07:38,480
Czochrolski's method, often
referred to as the C Z method.

116
00:07:38,480 --> 00:07:44,560
Inside those ultra-pure Spruce
Pine quartz crucibles, we dip a

117
00:07:44,560 --> 00:07:50,560
tiny seed of crystal into the
molten 9N silicon, and slowly,

118
00:07:50,560 --> 00:07:52,079
we pull it upward.

119
00:07:52,079 --> 00:07:54,879
Spinning as it cools.

120
00:07:54,879 --> 00:08:01,199
We grow giant, flawless crystal
cylinders, hundreds of pounds

121
00:08:01,199 --> 00:08:05,040
heavy, atoms aligned in perfect
lockstep.

122
00:08:05,040 --> 00:08:11,040
But Jan did not spend the rest
of his life being celebrated for

123
00:08:11,040 --> 00:08:11,360
it.

124
00:08:11,360 --> 00:08:16,959
He left Germany in 1928,
returning to a newly independent

125
00:08:16,959 --> 00:08:17,600
Poland.

126
00:08:17,600 --> 00:08:22,160
He became a professor at the
Warsaw University of Technology.

127
00:08:22,160 --> 00:08:26,240
He was a brilliant, wealthy
patron of the arts.

128
00:08:26,240 --> 00:08:29,360
Then came 1939.

129
00:08:29,360 --> 00:08:31,680
The invasion.

130
00:08:31,680 --> 00:08:35,919
The universities were violently
shut down.

131
00:08:35,919 --> 00:08:41,840
Academics were arrested, sent
to camps, executed.

132
00:08:41,840 --> 00:08:47,600
Jan, leaning on his pre-war
connections in Germany, and his

133
00:08:47,600 --> 00:08:51,600
German wife, got permission from
the occupying forces to keep

134
00:08:51,600 --> 00:08:53,360
his laboratory open.

135
00:08:53,360 --> 00:08:57,519
He rebranded it as a materials
research facility.

136
00:08:57,519 --> 00:09:03,039
To the people on the street, to
his former colleagues, he

137
00:09:03,039 --> 00:09:04,879
looked like a collaborator.

138
00:09:04,879 --> 00:09:09,919
A man who had traded his
country's dignity for his own

139
00:09:09,919 --> 00:09:10,960
survival.

140
00:09:10,960 --> 00:09:16,320
But they did not know that the
lab was a front.

141
00:09:16,320 --> 00:09:20,080
Jan was secretly employing
members of the Polish

142
00:09:20,080 --> 00:09:23,840
resistance, providing them with
documents to keep them out of

143
00:09:23,840 --> 00:09:24,639
the camps.

144
00:09:24,639 --> 00:09:28,960
He was using the lab's
machinery to reverse engineer

145
00:09:28,960 --> 00:09:31,840
German V-1 and V2 rockets.

146
00:09:31,840 --> 00:09:36,559
He was manufacturing grenade
shells for the underground home

147
00:09:36,559 --> 00:09:37,200
army.

148
00:09:37,200 --> 00:09:42,320
His work had the approval of
the Polish underground state.

149
00:09:42,320 --> 00:09:45,759
There's evidence that he
sheltered two Jewish women in

150
00:09:45,759 --> 00:09:47,120
his own home.

151
00:09:47,120 --> 00:09:50,480
But he looked like a
collaborator.

152
00:09:50,480 --> 00:09:55,519
And when the war ended, the new
communist regime arrested him

153
00:09:55,519 --> 00:09:57,519
and tried him for treason.

154
00:09:57,519 --> 00:09:59,840
He was acquitted.

155
00:09:59,840 --> 00:10:03,840
An investigation proved that he
did not collaborate with the

156
00:10:03,840 --> 00:10:04,720
Nazis.

157
00:10:04,720 --> 00:10:08,000
He had never been a traitor.

158
00:10:08,000 --> 00:10:11,519
But it did not matter.

159
00:10:11,519 --> 00:10:14,960
The reputational damage was
done.

160
00:10:14,960 --> 00:10:18,399
He was stripped of his
professorship.

161
00:10:18,399 --> 00:10:23,279
His name was meticulously and
deliberately erased from the

162
00:10:23,279 --> 00:10:24,799
university's records.

163
00:10:24,799 --> 00:10:28,720
A man who would later be listed
among the fathers of modern

164
00:10:28,720 --> 00:10:34,559
electronics died in 1953,
running a small operation in his

165
00:10:34,559 --> 00:10:38,320
hometown, making cosmetics and
sneezing powder.

166
00:10:38,320 --> 00:10:44,000
Here is what he never saw.

167
00:10:44,000 --> 00:10:48,639
By the time Jan died, the
canvas was solved.

168
00:10:48,639 --> 00:10:53,039
His crystals would become every
microchip we would ever build.

169
00:10:53,039 --> 00:10:56,960
The brains in every
supercomputer, satellite, and

170
00:10:56,960 --> 00:10:58,240
smart toaster.

171
00:10:58,240 --> 00:11:02,720
But you still have to draw the
pathways for the electricity to

172
00:11:02,720 --> 00:11:03,360
flow.

173
00:11:03,360 --> 00:11:07,759
In the early days, we drew
those pathways using visible

174
00:11:07,759 --> 00:11:12,080
light, shining it through a
stencil-like mask onto the

175
00:11:12,080 --> 00:11:14,720
silicon to etch the circuits.

176
00:11:14,720 --> 00:11:19,519
But as we demanded smaller and
faster machines, the stencils

177
00:11:19,519 --> 00:11:20,639
had to shrink.

178
00:11:20,639 --> 00:11:25,360
Today there are billions of
transistors on a single chip.

179
00:11:25,360 --> 00:11:29,360
Those transistors are only a
few nanometers wide.

180
00:11:29,360 --> 00:11:34,080
Five nanometers is about the
length your fingernail grows in

181
00:11:34,080 --> 00:11:35,519
five seconds.

182
00:11:35,519 --> 00:11:39,679
You cannot etch something that
small with ordinary light.

183
00:11:39,679 --> 00:11:43,200
The wavelength of visible light
is too wide.

184
00:11:43,200 --> 00:11:47,519
It is like trying to paint a
microscopic masterpiece with a

185
00:11:47,519 --> 00:11:48,960
heavy push broom.

186
00:11:48,960 --> 00:11:54,320
To work at that scale, you need
light that is roughly 100 times

187
00:11:54,320 --> 00:11:56,399
smaller than visible light.

188
00:11:56,399 --> 00:12:00,720
You need extreme ultraviolet
light.

189
00:12:09,679 --> 00:12:12,240
The machines that make and harness EUV light are among the most complex devices humanity has ever built. One
machine costs hundreds of

190
00:12:12,240 --> 00:12:16,559
millions of dollars and contains
more than 100,000 parts.

191
00:12:16,559 --> 00:12:20,720
Inside, EUV light bounces
through a series of mirrors

192
00:12:20,720 --> 00:12:22,720
before passing through the
stencil.

193
00:12:22,720 --> 00:12:26,399
More mirrors focus it onto the
silicon.

194
00:12:26,399 --> 00:12:31,200
Those mirrors are polished so
perfectly that if you expanded

195
00:12:31,200 --> 00:12:35,440
one to be the size of Germany,
the tallest bump on its surface

196
00:12:35,440 --> 00:12:37,840
would be less than a millimeter
high.

197
00:12:37,840 --> 00:12:44,080
Those mirrors a blueprint for
billions of transistors onto the

198
00:12:44,080 --> 00:12:48,159
silicon that Jan's method grew
with unthinkable precision.

199
00:12:48,159 --> 00:12:52,559
It is akin to standing on Earth
and shooting an arrow through

200
00:12:52,559 --> 00:12:54,960
an apple on the surface of the
moon.

201
00:12:54,960 --> 00:12:57,120
Then doing it again.

202
00:12:57,120 --> 00:12:58,639
And again.

203
00:12:58,639 --> 00:13:03,440
Thousands of chips can be
etched onto a single silicon

204
00:13:03,440 --> 00:13:04,320
wafer.

205
00:13:04,320 --> 00:13:07,919
And the light itself does not
want to exist.

206
00:13:07,919 --> 00:13:12,480
Extreme ultraviolet light is
swallowed by air.

207
00:13:12,480 --> 00:13:16,000
It does not occur naturally on
Earth.

208
00:13:16,000 --> 00:13:21,200
So, deep within the EU V
machine, there is a vacuum

209
00:13:21,200 --> 00:13:22,159
chamber.

210
00:13:22,159 --> 00:13:27,120
In that airless space, a
generator ejects a microscopic

211
00:13:27,120 --> 00:13:29,919
droplet of liquid metal into the
dark.

212
00:13:29,919 --> 00:13:31,600
Then another.

213
00:13:31,600 --> 00:13:36,720
50,000 droplets, every single
second.

214
00:13:36,720 --> 00:13:42,159
As each droplet falls, a
high-powered laser fires a weak

215
00:13:42,159 --> 00:13:46,480
pulse, just enough to flatten
the tiny sphere into a tiny

216
00:13:46,480 --> 00:13:47,360
pancake.

217
00:13:47,360 --> 00:13:53,519
A microsecond later, a second
pulse fires with devastating

218
00:13:53,519 --> 00:13:54,159
force.

219
00:13:54,159 --> 00:13:56,240
The metal vaporizes.

220
00:13:56,240 --> 00:14:00,639
It becomes a plasma burning 40
times hotter than the surface of

221
00:14:00,639 --> 00:14:01,279
the sun.

222
00:14:01,279 --> 00:14:08,240
And in that violent,
microscopic death, it flashes,

223
00:14:08,240 --> 00:14:12,240
throwing off a burst of extreme
ultraviolet light.

224
00:14:12,240 --> 00:14:17,200
And the metal, the drop that
dies in the dark to light the

225
00:14:17,200 --> 00:14:18,399
digital age.

226
00:14:18,399 --> 00:14:20,080
It is tin.

227
00:14:20,080 --> 00:14:22,879
Molten tin.

228
00:14:22,879 --> 00:14:33,840
In 1916, a tired man dipped his
pen into a crucible of tin and

229
00:14:33,840 --> 00:14:37,039
pulled out the foundation of the
digital world.

230
00:14:37,039 --> 00:14:43,039
More than a century later, we
blast falling drops of that same

231
00:14:43,039 --> 00:14:46,960
metal to make light so fragile
it can barely exist.

232
00:14:46,960 --> 00:14:52,159
That light is used to draw
microchips onto the perfect

233
00:14:52,159 --> 00:14:54,879
crystals that power our devices.

234
00:14:54,879 --> 00:15:00,559
And we use those devices as
flashlights.

235
00:15:00,559 --> 00:15:04,879
I'm Daina Bouquin.

236
00:15:04,879 --> 00:15:08,159
And this is Found in the
Machine.

237
00:15:08,159 --> 00:15:13,679
Before you go, if you'd like to
support the show, rate and

238
00:15:13,679 --> 00:15:16,159
review this podcast wherever you
listen.

239
00:15:16,159 --> 00:15:18,799
Or better yet, share it with a
friend.

240
00:15:18,799 --> 00:15:22,480
I'm really trying to turn this
into something sustainable, and

241
00:15:22,480 --> 00:15:24,720
I'd love your help getting the
word out.

242
00:15:24,720 --> 00:15:27,679
Another way you can support the
show is by going to

243
00:15:27,679 --> 00:15:32,639
bookshop.org/shop/found in the
machine.

244
00:15:33,519 --> 00:15:37,360
Your purchases support
independent bookstores across

245
00:15:37,360 --> 00:15:41,440
the US, and they also help make
this podcast possible.

246
00:15:41,440 --> 00:15:45,200
And if you'd like to dive
deeper into any of these

247
00:15:45,200 --> 00:15:48,639
stories, you can sign up for my
newsletter at

248
00:15:48,639 --> 00:15:49,279
notes.foundinthemachine.com. Thanks for listening.