1
00:00:00,000 --> 00:00:02,520
The following content is
provided under a Creative

2
00:00:02,520 --> 00:00:03,970
Commons license.

3
00:00:03,970 --> 00:00:06,360
Your support will help
MIT OpenCourseWare

4
00:00:06,360 --> 00:00:10,660
continue to offer high quality
educational resources for free.

5
00:00:10,660 --> 00:00:13,350
To make a donation or
view additional materials

6
00:00:13,350 --> 00:00:17,190
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:17,190 --> 00:00:18,320
at ocw.mit.edu.

8
00:00:25,575 --> 00:00:26,700
JEFFREY C. GROSSMAN: Right?

9
00:00:26,700 --> 00:00:27,370
So here we are.

10
00:00:27,370 --> 00:00:27,870
See?

11
00:00:27,870 --> 00:00:33,570
Because what I want you
to do is feel your oneness

12
00:00:33,570 --> 00:00:36,600
with modeling using
computational quantum

13
00:00:36,600 --> 00:00:38,490
mechanics and the sun's energy.

14
00:00:38,490 --> 00:00:39,210
Right?

15
00:00:39,210 --> 00:00:43,110
That's really what
I've been wanting

16
00:00:43,110 --> 00:00:45,540
to try to bridge, the
applications of using

17
00:00:45,540 --> 00:00:52,230
the sun in renewables,
and how that connects

18
00:00:52,230 --> 00:00:54,070
to ways of modeling electrons.

19
00:00:54,070 --> 00:00:54,720
OK?

20
00:00:54,720 --> 00:00:58,110
So I really want to
get to this other topic

21
00:00:58,110 --> 00:01:00,390
that we started which is
using the sun's energy

22
00:01:00,390 --> 00:01:01,430
to make electricity.

23
00:01:01,430 --> 00:01:01,930
OK?

24
00:01:01,930 --> 00:01:06,180
So I want to cover that today
and a little bit on Thursday.

25
00:01:06,180 --> 00:01:06,900
OK.

26
00:01:06,900 --> 00:01:10,140
Now, since I've made
PSET 6 optional,

27
00:01:10,140 --> 00:01:14,340
I wanted to just discuss a few
points from it to start with.

28
00:01:14,340 --> 00:01:14,850
OK?

29
00:01:14,850 --> 00:01:18,900
Because that's like closing the
loop on our modeling of solids,

30
00:01:18,900 --> 00:01:22,832
and then we'll do a little
more motivation on solar PV,

31
00:01:22,832 --> 00:01:24,540
and then we'll talk
about how computation

32
00:01:24,540 --> 00:01:25,860
can play a role, as I said.

33
00:01:25,860 --> 00:01:26,460
OK?

34
00:01:26,460 --> 00:01:30,660
So actually, I'm going to
bring up the PSET here,

35
00:01:30,660 --> 00:01:31,670
and I just want to--

36
00:01:31,670 --> 00:01:33,420
I'm not going to do
the problems with you,

37
00:01:33,420 --> 00:01:36,390
but I just want to ask
you how you might approach

38
00:01:36,390 --> 00:01:38,400
these problems,
just to make sure

39
00:01:38,400 --> 00:01:44,130
that, as we close this field
this simulation of solids

40
00:01:44,130 --> 00:01:48,450
with quantum mechanics, we have
a sense of what you might do.

41
00:01:51,270 --> 00:01:55,080
The equilibrium
lattice constant,

42
00:01:55,080 --> 00:01:58,450
how would you calculate the
equilibrium lattice constant?

43
00:01:58,450 --> 00:01:59,830
Somebody tell me.

44
00:01:59,830 --> 00:02:00,830
Yeah?

45
00:02:00,830 --> 00:02:01,930
AUDIENCE: Looking for the
lattice which minimizes

46
00:02:01,930 --> 00:02:02,664
[INAUDIBLE]

47
00:02:02,664 --> 00:02:04,872
JEFFREY C. GROSSMAN: And
how would you look for that?

48
00:02:04,872 --> 00:02:06,800
AUDIENCE: Change the
lattice [INAUDIBLE]..

49
00:02:06,800 --> 00:02:07,390
JEFFREY C. GROSSMAN: OK.

50
00:02:07,390 --> 00:02:07,930
Yeah.

51
00:02:07,930 --> 00:02:10,270
So you could just like
change the lattice right,

52
00:02:10,270 --> 00:02:16,660
and I have like
lattice and the energy.

53
00:02:16,660 --> 00:02:22,480
And I do like 4 and 5
and 6, or maybe that's

54
00:02:22,480 --> 00:02:23,530
not a good enough grid.

55
00:02:23,530 --> 00:02:25,620
Maybe I need 4.5, 5.5.

56
00:02:25,620 --> 00:02:26,240
Right?

57
00:02:26,240 --> 00:02:27,657
And what would my
curve look like,

58
00:02:27,657 --> 00:02:30,100
if I have an energy,
energy versus lattice?

59
00:02:30,100 --> 00:02:31,950
What's it going to look like?

60
00:02:31,950 --> 00:02:33,220
Yeah, something like that.

61
00:02:33,220 --> 00:02:35,970
I don't know, and the minimum,
the equilibrium is here.

62
00:02:35,970 --> 00:02:38,010
Now, is that converse
with respect to k-points?

63
00:02:38,010 --> 00:02:39,390
Well, that's
something you can do

64
00:02:39,390 --> 00:02:41,223
the curve for different
numbers of k-points.

65
00:02:41,223 --> 00:02:43,380
Right?

66
00:02:43,380 --> 00:02:46,410
But now, what if I wanted to
calculate the binding energy,

67
00:02:46,410 --> 00:02:50,610
or in a solid, we call
it the cohesive energy?

68
00:02:50,610 --> 00:02:51,450
Why?

69
00:02:51,450 --> 00:02:53,550
It sounds cool.

70
00:02:53,550 --> 00:02:54,720
It's different.

71
00:02:54,720 --> 00:02:55,890
Yeah.

72
00:02:55,890 --> 00:02:57,210
You know what I'm saying.

73
00:02:57,210 --> 00:02:58,980
You were waiting for that.

74
00:02:58,980 --> 00:03:01,758
We got a big cohesive
energy [INAUDIBLE] or no,

75
00:03:01,758 --> 00:03:03,300
that didn't have
anything to do with.

76
00:03:03,300 --> 00:03:04,380
Yeah.

77
00:03:04,380 --> 00:03:07,380
10 seconds, do you call?

78
00:03:07,380 --> 00:03:10,650
Do you call a foul with 10
seconds left in a playoff game?

79
00:03:10,650 --> 00:03:13,030
That's all I want to know.

80
00:03:13,030 --> 00:03:14,760
10 seconds left.

81
00:03:14,760 --> 00:03:16,560
You play it out.

82
00:03:16,560 --> 00:03:17,060
Right?

83
00:03:17,060 --> 00:03:19,050
I don't care if it was like
the worst flagrant foul

84
00:03:19,050 --> 00:03:20,450
of the year, you
let it play out.

85
00:03:20,450 --> 00:03:23,720
You don't call a foul with 10
seconds left in a playoff game.

86
00:03:23,720 --> 00:03:25,280
It's just not the
right way to lose.

87
00:03:25,280 --> 00:03:26,225
AUDIENCE: [INAUDIBLE]

88
00:03:26,225 --> 00:03:27,600
JEFFREY C. GROSSMAN:
What's that?

89
00:03:27,600 --> 00:03:28,800
You're feeling it?

90
00:03:28,800 --> 00:03:30,360
Yeah.

91
00:03:30,360 --> 00:03:31,360
Unbelievable.

92
00:03:31,360 --> 00:03:33,060
Unbelievable.

93
00:03:33,060 --> 00:03:33,840
All right.

94
00:03:33,840 --> 00:03:36,070
How would you calculate
that cohesive energy,

95
00:03:36,070 --> 00:03:37,800
the binding energy in a solid?

96
00:03:40,480 --> 00:03:41,680
Yeah.

97
00:03:41,680 --> 00:03:44,420
AUDIENCE: So after you find the
equilibrium lattice constant,

98
00:03:44,420 --> 00:03:45,940
you have the [INAUDIBLE] energy.

99
00:03:45,940 --> 00:03:46,940
JEFFREY C. GROSSMAN: OK.

100
00:03:46,940 --> 00:03:48,680
That's an energy versus
lattice constant.

101
00:03:48,680 --> 00:03:51,013
AUDIENCE: Then you can make
the lattice constant really,

102
00:03:51,013 --> 00:03:53,680
really, really large, and
then you'll get, basically,

103
00:03:53,680 --> 00:03:56,240
so it's like an isolated
atom, get that energy,

104
00:03:56,240 --> 00:03:58,905
and you've got two [INAUDIBLE]

105
00:03:58,905 --> 00:04:00,280
JEFFREY C. GROSSMAN:
I like that.

106
00:04:00,280 --> 00:04:02,320
Did everybody see that?

107
00:04:02,320 --> 00:04:03,460
Yeah.

108
00:04:03,460 --> 00:04:04,240
Let's see.

109
00:04:04,240 --> 00:04:09,091
If I look over here, because
you know you love the nanoHUB,

110
00:04:09,091 --> 00:04:12,460
and you want to see
something on it today.

111
00:04:12,460 --> 00:04:12,960
Ah!

112
00:04:16,300 --> 00:04:18,250
That's because I
didn't turn on my--

113
00:04:18,250 --> 00:04:19,980
wait a second.

114
00:04:19,980 --> 00:04:21,149
OK.

115
00:04:21,149 --> 00:04:21,720
Turn it on.

116
00:04:24,790 --> 00:04:28,410
So now, if I ask
you how to converge

117
00:04:28,410 --> 00:04:33,540
the k-points, Sam, for the
cohesive energy calculation,

118
00:04:33,540 --> 00:04:35,598
how would you do that?

119
00:04:35,598 --> 00:04:37,140
AUDIENCE: Do exactly
what I just did,

120
00:04:37,140 --> 00:04:38,850
and change the
number of k-points,

121
00:04:38,850 --> 00:04:41,190
and see if cohesive
energy changed.

122
00:04:41,190 --> 00:04:43,860
JEFFREY C. GROSSMAN: And is
the number of k-points that you

123
00:04:43,860 --> 00:04:47,100
need going to be--

124
00:04:47,100 --> 00:04:49,950
what's that going
to be for the atom?

125
00:04:49,950 --> 00:04:52,020
AUDIENCE: Small, probably
really just [INAUDIBLE]

126
00:04:52,020 --> 00:04:54,367
isolate that, just
not really periodic.

127
00:04:54,367 --> 00:04:55,450
JEFFREY C. GROSSMAN: Yeah.

128
00:04:55,450 --> 00:04:57,120
If it's really
isolated, you shouldn't

129
00:04:57,120 --> 00:04:59,430
need more than one
k-point for the atom,

130
00:04:59,430 --> 00:05:01,950
but the solid is
where you might need

131
00:05:01,950 --> 00:05:03,150
to look at the convergence.

132
00:05:03,150 --> 00:05:04,780
Right?

133
00:05:04,780 --> 00:05:10,200
So you see, here's our
tool, and here is--

134
00:05:10,200 --> 00:05:12,540
it's actually already
defaulted to silicon.

135
00:05:12,540 --> 00:05:13,230
Right?

136
00:05:13,230 --> 00:05:16,920
And so you can just press--

137
00:05:16,920 --> 00:05:22,960
why isn't it-- this
is just torturous.

138
00:05:22,960 --> 00:05:24,570
OK, there we go, Simulate.

139
00:05:24,570 --> 00:05:29,290
There's two scroll bars,
and I can't handle it.

140
00:05:29,290 --> 00:05:30,820
Maybe it's me.

141
00:05:30,820 --> 00:05:31,600
I don't know.

142
00:05:31,600 --> 00:05:34,330
I like to blame the nanoHUB.

143
00:05:34,330 --> 00:05:35,510
So this is silicon.

144
00:05:35,510 --> 00:05:36,010
Right?

145
00:05:36,010 --> 00:05:38,005
And I did silicon,
and I got a crystal.

146
00:05:38,005 --> 00:05:40,870
And I forgot to click to
show me the structure,

147
00:05:40,870 --> 00:05:42,320
but it's this
beautiful structure.

148
00:05:42,320 --> 00:05:43,030
OK?

149
00:05:43,030 --> 00:05:51,740
And in the output, it
gave me 214.6 eV, 214.6.

150
00:05:51,740 --> 00:05:54,260
So my energy here,
let's just say

151
00:05:54,260 --> 00:05:58,280
that was the right
lattice is 214.6 eV,

152
00:05:58,280 --> 00:06:01,006
and now what did you say to do?

153
00:06:01,006 --> 00:06:01,900
AUDIENCE: [INAUDIBLE]

154
00:06:01,900 --> 00:06:02,900
JEFFREY C. GROSSMAN: OK.

155
00:06:02,900 --> 00:06:04,460
AUDIENCE: The
lattice really large.

156
00:06:04,460 --> 00:06:06,080
JEFFREY C. GROSSMAN:
Lattice constant.

157
00:06:06,080 --> 00:06:10,050
AUDIENCE: Maybe say, I
don't know, 15 or 20?

158
00:06:10,050 --> 00:06:15,070
JEFFREY C. GROSSMAN:
20, let's apply it too.

159
00:06:15,070 --> 00:06:18,313
20 angstroms, and
then anything else?

160
00:06:18,313 --> 00:06:19,270
AUDIENCE: [INAUDIBLE]

161
00:06:19,270 --> 00:06:20,270
JEFFREY C. GROSSMAN: OK.

162
00:06:20,270 --> 00:06:22,760
Except I don't really need
four k-points, but let's

163
00:06:22,760 --> 00:06:24,990
just leave it in.

164
00:06:24,990 --> 00:06:27,160
OK?

165
00:06:27,160 --> 00:06:29,640
Is there a way to tell after
looking at the simulation

166
00:06:29,640 --> 00:06:32,430
one second if I actually
should have had more--

167
00:06:32,430 --> 00:06:34,200
if I choose one k-point--

168
00:06:34,200 --> 00:06:35,160
AUDIENCE: [INAUDIBLE]

169
00:06:35,160 --> 00:06:37,230
JEFFREY C. GROSSMAN:
And if it's wiggly.

170
00:06:37,230 --> 00:06:38,460
AUDIENCE: [INAUDIBLE]

171
00:06:38,460 --> 00:06:40,960
JEFFREY C. GROSSMAN: I should
try to converge the k-points.

172
00:06:40,960 --> 00:06:42,870
Because if the band
structure is wiggly,

173
00:06:42,870 --> 00:06:46,500
it means there's something
going on versus k.

174
00:06:46,500 --> 00:06:49,180
So there's a dependence
on k, and if that's true,

175
00:06:49,180 --> 00:06:52,650
then I shouldn't be settling
for just not testing

176
00:06:52,650 --> 00:06:53,340
the convergence.

177
00:06:53,340 --> 00:06:53,840
Right?

178
00:06:53,840 --> 00:06:56,730
I should at least try to
increase the number of k-points

179
00:06:56,730 --> 00:06:58,230
to see if [INAUDIBLE]
so here it is.

180
00:06:58,230 --> 00:07:01,840
There is my crystal of silicon
with really large lattice

181
00:07:01,840 --> 00:07:02,340
constants.

182
00:07:02,340 --> 00:07:05,790
So you see, those atoms aren't
even trying to bond together.

183
00:07:05,790 --> 00:07:06,840
Is it large enough?

184
00:07:10,590 --> 00:07:11,370
How can you tell?

185
00:07:11,370 --> 00:07:12,670
Well, you could look
at the bone structure.

186
00:07:12,670 --> 00:07:13,920
Yeah, and what should they be?

187
00:07:13,920 --> 00:07:20,010
If these are really atoms,
these are really atoms,

188
00:07:20,010 --> 00:07:23,700
then the band structure--
oh, look at that.

189
00:07:23,700 --> 00:07:25,680
Those are beautiful
silicon states.

190
00:07:25,680 --> 00:07:26,370
Right?

191
00:07:26,370 --> 00:07:28,530
These are S&P states of silicon.

192
00:07:28,530 --> 00:07:30,240
It's an atom.

193
00:07:30,240 --> 00:07:31,990
Because I just made
them really far apart,

194
00:07:31,990 --> 00:07:34,120
so they didn't feel anything
from their neighbors.

195
00:07:34,120 --> 00:07:34,630
Right?

196
00:07:34,630 --> 00:07:38,170
Now, I got an energy here
of-- and I'm coming back

197
00:07:38,170 --> 00:07:40,000
to your question
here in a second--

198
00:07:40,000 --> 00:07:41,840
but I got an energy here of 203.

199
00:07:41,840 --> 00:07:42,610
OK.

200
00:07:42,610 --> 00:07:49,670
So I got an atom was 203,
and let's just round it.

201
00:07:49,670 --> 00:07:54,920
I'm just going to truncate
214 and 203 is the solid.

202
00:07:54,920 --> 00:07:55,460
OK?

203
00:07:55,460 --> 00:07:57,331
So Sam, what do I do next?

204
00:07:57,331 --> 00:08:00,770
AUDIENCE: [INAUDIBLE]

205
00:08:00,770 --> 00:08:05,130
JEFFREY C. GROSSMAN: And the
cohesive energy is 11 eV.

206
00:08:05,130 --> 00:08:08,790
And then I look it up, and I
actually see that it's 4.6.

207
00:08:08,790 --> 00:08:12,240
Because I looked it up, and it's
been measured in experiments.

208
00:08:12,240 --> 00:08:14,870
What did I do wrong?

209
00:08:14,870 --> 00:08:17,500
The cohesive energy of
silicon, from my calculations

210
00:08:17,500 --> 00:08:22,520
I did the solid minus the
atom, and I find it's 11 eV.

211
00:08:25,047 --> 00:08:27,630
AUDIENCE: Making sure you've got
the right lattice [INAUDIBLE]

212
00:08:27,630 --> 00:08:29,088
JEFFREY C. GROSSMAN:
For the solid?

213
00:08:29,088 --> 00:08:33,659
Maybe not, maybe I was
here, but I wasn't 6 eV off.

214
00:08:33,659 --> 00:08:35,547
I know that, but it's
a really good point.

215
00:08:35,547 --> 00:08:36,630
That's one thing to check.

216
00:08:36,630 --> 00:08:37,140
What else?

217
00:08:41,659 --> 00:08:42,169
Yeah?

218
00:08:42,169 --> 00:08:43,730
AUDIENCE: Did you converge
to your basis set?

219
00:08:43,730 --> 00:08:45,438
JEFFREY C. GROSSMAN:
Did I conversion it?

220
00:08:45,438 --> 00:08:46,550
Very unlikely.

221
00:08:46,550 --> 00:08:48,230
Excellent other point to check.

222
00:08:48,230 --> 00:08:53,867
Probably could get me a half an
eV even, or more, maybe an eV,

223
00:08:53,867 --> 00:08:54,950
but it's not getting me 6.

224
00:08:58,560 --> 00:09:01,480
This is actually a really
simple problem here.

225
00:09:01,480 --> 00:09:02,820
What is my cohesive energy?

226
00:09:02,820 --> 00:09:03,480
AUDIENCE: [INAUDIBLE]
divided by 2.

227
00:09:03,480 --> 00:09:03,810
JEFFREY C. GROSSMAN: Yeah.

228
00:09:03,810 --> 00:09:04,428
Why?

229
00:09:04,428 --> 00:09:06,220
AUDIENCE: Because
there's two [INAUDIBLE]..

230
00:09:06,220 --> 00:09:08,040
JEFFREY C. GROSSMAN:
Don't forget that.

231
00:09:08,040 --> 00:09:08,560
Yeah.

232
00:09:08,560 --> 00:09:11,620
See, you have a
certain number of atoms

233
00:09:11,620 --> 00:09:13,690
that you made your
crystal out of that

234
00:09:13,690 --> 00:09:15,553
went into the simulation.

235
00:09:15,553 --> 00:09:16,460
Right?

236
00:09:16,460 --> 00:09:21,050
That's actually, when I
give this problem, that's

237
00:09:21,050 --> 00:09:24,440
a very common mistake.

238
00:09:24,440 --> 00:09:27,290
Well, I guess I'm
giving this problem,

239
00:09:27,290 --> 00:09:28,595
since it is a problem set.

240
00:09:28,595 --> 00:09:29,720
So don't make that mistake.

241
00:09:29,720 --> 00:09:33,650
You don't need to,
because now, you know.

242
00:09:33,650 --> 00:09:39,990
When you go to the input here,
remember, this is my basis.

243
00:09:39,990 --> 00:09:42,300
My basis had a certain
number of atoms in it,

244
00:09:42,300 --> 00:09:45,270
and so then that's
what gets repeated.

245
00:09:45,270 --> 00:09:48,000
So the number of atoms, or
electrons from those atoms,

246
00:09:48,000 --> 00:09:53,040
in the simulation is for
two silicon atoms, not one.

247
00:09:53,040 --> 00:09:58,130
So the cohesive energy I
calculated was per two atoms.

248
00:09:58,130 --> 00:10:00,860
So if I really want the
cohesive energy per atom,

249
00:10:00,860 --> 00:10:03,060
you've got to divide by 2.

250
00:10:03,060 --> 00:10:05,390
Another way to get the
isolated atom would just

251
00:10:05,390 --> 00:10:09,470
be to take-- a lot of people
did this last year, is

252
00:10:09,470 --> 00:10:11,220
they just take this atom out.

253
00:10:11,220 --> 00:10:11,720
OK?

254
00:10:11,720 --> 00:10:13,880
And they change that to one,
and they give a large lattice

255
00:10:13,880 --> 00:10:15,880
constant, and that's a
perfectly legitimate way.

256
00:10:15,880 --> 00:10:18,322
It's the same thing, actually.

257
00:10:18,322 --> 00:10:20,030
You're basically just
getting those atoms

258
00:10:20,030 --> 00:10:21,320
to not see each other.

259
00:10:21,320 --> 00:10:22,790
But then, you had
a cohesive energy

260
00:10:22,790 --> 00:10:25,998
with two atoms in the unit
cell and an atom energy

261
00:10:25,998 --> 00:10:27,290
with one atom in the unit cell.

262
00:10:27,290 --> 00:10:28,748
You just got to
keep track of that.

263
00:10:28,748 --> 00:10:29,600
Right?

264
00:10:29,600 --> 00:10:31,340
So energy per what?

265
00:10:31,340 --> 00:10:32,880
Per your basis.

266
00:10:32,880 --> 00:10:34,190
OK?

267
00:10:34,190 --> 00:10:37,530
Basic stuff of
simulating solids.

268
00:10:37,530 --> 00:10:38,030
OK.

269
00:10:38,030 --> 00:10:39,730
Any other questions?

270
00:10:39,730 --> 00:10:42,380
I may come back to the PV
question in a little bit,

271
00:10:42,380 --> 00:10:48,780
but I just wanted to make
sure we feel our oneness.

272
00:10:48,780 --> 00:10:51,270
OK.

273
00:10:51,270 --> 00:10:53,076
There was a question up here.

274
00:10:53,076 --> 00:10:56,640
AUDIENCE: What is the principle
definition of cohesive energy?

275
00:10:56,640 --> 00:11:04,266
JEFFREY C. GROSSMAN: So
the cohesive energy really,

276
00:11:04,266 --> 00:11:08,910
the feeling is the
gluon between the atoms.

277
00:11:08,910 --> 00:11:10,590
That's the feeling definition.

278
00:11:10,590 --> 00:11:13,290
It's how strongly
they're glued together.

279
00:11:13,290 --> 00:11:15,930
The actual
quantitative definition

280
00:11:15,930 --> 00:11:24,740
is it's the energy of the
solid minus the energy

281
00:11:24,740 --> 00:11:29,330
of the constituent parts
which in this case are atoms.

282
00:11:29,330 --> 00:11:33,065
But you could be looking
at say in a molecular solid

283
00:11:33,065 --> 00:11:35,190
and then you might want to
know the cohesive energy

284
00:11:35,190 --> 00:11:36,700
between molecules.

285
00:11:36,700 --> 00:11:37,200
Right?

286
00:11:37,200 --> 00:11:38,670
In which case, this would be--

287
00:11:38,670 --> 00:11:42,060
and then, you've got to
be right with your ends,

288
00:11:42,060 --> 00:11:43,290
and that's what I'm saying.

289
00:11:43,290 --> 00:11:43,790
Right?

290
00:11:43,790 --> 00:11:45,630
If you did your solid
with two, then you

291
00:11:45,630 --> 00:11:50,240
need to have N times
the energy of an atom.

292
00:11:50,240 --> 00:11:53,480
Or you can think of it as
just energy of all the atoms

293
00:11:53,480 --> 00:11:55,210
together that made
up that solid.

294
00:11:55,210 --> 00:11:55,710
Right?

295
00:11:55,710 --> 00:11:58,250
So just make sure
the numbers work.

296
00:11:58,250 --> 00:12:00,080
OK?

297
00:12:00,080 --> 00:12:04,600
Now, that's actually
the atomization energy.

298
00:12:04,600 --> 00:12:08,690
There lots of fun names for it.

299
00:12:08,690 --> 00:12:13,150
Now, I ended the lecture before
the quiz with this motivation,

300
00:12:13,150 --> 00:12:16,480
which I think is amazing that
the Earth's total resources

301
00:12:16,480 --> 00:12:18,550
of oil, the 2
trillion we've used

302
00:12:18,550 --> 00:12:21,970
and the other trillion
we want to use,

303
00:12:21,970 --> 00:12:25,660
is energy you get from
the sun in two days.

304
00:12:25,660 --> 00:12:27,550
I also very quickly
said, well, I'm not

305
00:12:27,550 --> 00:12:29,680
going to talk about
climate change,

306
00:12:29,680 --> 00:12:30,890
but I can't help myself.

307
00:12:30,890 --> 00:12:33,670
I have to say a few words, just
a few words on climate change,

308
00:12:33,670 --> 00:12:40,717
because it's really important.

309
00:12:40,717 --> 00:12:42,550
And so I'm just going
to spend five minutes.

310
00:12:42,550 --> 00:12:43,467
That's all, I promise.

311
00:12:45,950 --> 00:12:46,450
OK.

312
00:12:46,450 --> 00:12:48,640
So how hot are
things going to get?

313
00:12:48,640 --> 00:12:49,420
Here's an article.

314
00:12:49,420 --> 00:12:50,920
There's lots of
articles since here,

315
00:12:50,920 --> 00:12:52,340
but this is a nice picture.

316
00:12:52,340 --> 00:12:55,270
So I took this
from Time Magazine,

317
00:12:55,270 --> 00:12:57,100
and there's lots of predictions.

318
00:12:57,100 --> 00:13:00,940
And there's a really big range,
and that's part of the problem.

319
00:13:00,940 --> 00:13:04,420
Part of the problem is also that
it's not about just warming.

320
00:13:04,420 --> 00:13:07,173
That is what's happening
on average over the Earth,

321
00:13:07,173 --> 00:13:08,590
but that's why we
really shouldn't

322
00:13:08,590 --> 00:13:10,150
call it global warming.

323
00:13:10,150 --> 00:13:12,970
We should really call
it climate change,

324
00:13:12,970 --> 00:13:16,520
because the changes that
happen are very local.

325
00:13:16,520 --> 00:13:17,020
Right?

326
00:13:17,020 --> 00:13:19,312
They're very different in
different parts of the world.

327
00:13:19,312 --> 00:13:21,010
Some places are
going to get colder.

328
00:13:21,010 --> 00:13:23,270
OK?

329
00:13:23,270 --> 00:13:25,360
OK, and so there's
lots of examples,

330
00:13:25,360 --> 00:13:28,630
and you can see things
melting the size of Rhode

331
00:13:28,630 --> 00:13:31,630
Island and lots of really
great images of what

332
00:13:31,630 --> 00:13:33,640
the changes in temperature do.

333
00:13:33,640 --> 00:13:36,160
Here's one I like,
because you don't always

334
00:13:36,160 --> 00:13:38,200
think about the
warming in this sense.

335
00:13:38,200 --> 00:13:40,600
But the changes in
climate also change

336
00:13:40,600 --> 00:13:43,900
when things get eaten, like
animals that eat trees,

337
00:13:43,900 --> 00:13:44,947
like this beetle.

338
00:13:44,947 --> 00:13:46,780
That's a really bad
thing for this tree that

339
00:13:46,780 --> 00:13:50,590
can live a few weeks either
before or later than it usually

340
00:13:50,590 --> 00:13:53,920
lives, and therefore, there are
no predators around to kill it.

341
00:13:53,920 --> 00:13:57,980
And already, in Canada, this
is just an enormous disaster.

342
00:13:57,980 --> 00:14:03,950
So 411 million cubic feet
of trees have been killed,

343
00:14:03,950 --> 00:14:07,210
and that's double the take
by all loggers in Canada.

344
00:14:07,210 --> 00:14:10,913
So enormous problems can
arise from climate change,

345
00:14:10,913 --> 00:14:12,580
and you've seen these
kinds of pictures.

346
00:14:16,630 --> 00:14:19,810
So does anyone know
what the P stands for?

347
00:14:22,360 --> 00:14:26,880
CO2 emissions, how much we're
putting out into the world.

348
00:14:26,880 --> 00:14:31,200
PgC per year, what's the P?

349
00:14:31,200 --> 00:14:34,070
Peta, peta, that's a lot.

350
00:14:34,070 --> 00:14:35,710
How much is that?

351
00:14:35,710 --> 00:14:38,487
AUDIENCE: [INAUDIBLE]

352
00:14:38,487 --> 00:14:39,820
JEFFREY C. GROSSMAN: Big number.

353
00:14:39,820 --> 00:14:41,800
AUDIENCE: [INAUDIBLE]

354
00:14:41,800 --> 00:14:44,950
JEFFREY C. GROSSMAN: 15th,
10 to the 15th, right?

355
00:14:44,950 --> 00:14:53,410
That's a whole lot, and this
is from that fairly famous IPCC

356
00:14:53,410 --> 00:14:55,540
report which is
actually a good read.

357
00:14:55,540 --> 00:14:59,590
It's kind of boring,
but it's like Smil.

358
00:14:59,590 --> 00:15:00,220
Right?

359
00:15:00,220 --> 00:15:02,555
Has anybody read Smil's books?

360
00:15:02,555 --> 00:15:03,430
You should all read--

361
00:15:03,430 --> 00:15:04,570
anybody?

362
00:15:04,570 --> 00:15:07,680
Nobody's even heard of him?

363
00:15:07,680 --> 00:15:15,470
So Vaclav Smil has written
some wonderful books on energy.

364
00:15:15,470 --> 00:15:17,570
OK?

365
00:15:17,570 --> 00:15:18,530
It's a little bit dry.

366
00:15:18,530 --> 00:15:20,660
I'm not saying
it's a page-turner.

367
00:15:20,660 --> 00:15:23,310
It's not dialogue-driven.

368
00:15:23,310 --> 00:15:24,020
It's not a novel.

369
00:15:24,020 --> 00:15:27,980
It's not a romance, but
it's very interesting,

370
00:15:27,980 --> 00:15:31,880
very well put
together, some of them.

371
00:15:31,880 --> 00:15:35,290
Now, if you want something
a little lighter,

372
00:15:35,290 --> 00:15:36,290
there's Without Hot Air.

373
00:15:36,290 --> 00:15:38,510
Has anybody read
that, Without Hot Air?

374
00:15:38,510 --> 00:15:40,790
Anybody heard of that?

375
00:15:40,790 --> 00:15:41,290
OK.

376
00:15:41,290 --> 00:15:41,840
That's free.

377
00:15:41,840 --> 00:15:44,660
You can just download that off
the internet, Without Hot Air,

378
00:15:44,660 --> 00:15:45,160
I think.

379
00:15:48,250 --> 00:15:52,750
I'll remember the
author's name, in England.

380
00:15:52,750 --> 00:15:56,255
Anyway, this is how
much CO2 we put out,

381
00:15:56,255 --> 00:15:58,630
and these are just what they
did is they just said, well,

382
00:15:58,630 --> 00:16:00,910
if we turn it over quickly--

383
00:16:00,910 --> 00:16:03,310
you see, because the amount
of CO2 in the atmosphere

384
00:16:03,310 --> 00:16:04,330
is different.

385
00:16:04,330 --> 00:16:06,070
It's still going
to keep increasing,

386
00:16:06,070 --> 00:16:07,555
even if we change our ways.

387
00:16:07,555 --> 00:16:09,430
And so there's going to
be still an increase,

388
00:16:09,430 --> 00:16:10,510
and then it's
going to come down.

389
00:16:10,510 --> 00:16:12,230
This would be the CO2
in the atmosphere.

390
00:16:12,230 --> 00:16:14,630
And if we turn it over a little
later, then we get the green.

391
00:16:14,630 --> 00:16:16,672
And if we don't turn
anything over until the red,

392
00:16:16,672 --> 00:16:20,290
we go up to 1,000 parts per
million in the atmosphere,

393
00:16:20,290 --> 00:16:22,440
and that's a lot.

394
00:16:22,440 --> 00:16:23,630
What does that mean?

395
00:16:23,630 --> 00:16:25,310
Well, it's hard to predict.

396
00:16:25,310 --> 00:16:27,760
It's very unlikely,
in my opinion,

397
00:16:27,760 --> 00:16:30,070
we'll do anything
but the red, if that.

398
00:16:30,070 --> 00:16:32,738
Right?

399
00:16:32,738 --> 00:16:33,780
You can see these things.

400
00:16:33,780 --> 00:16:34,630
It's not that hard.

401
00:16:34,630 --> 00:16:36,930
You can look at the data,
and you can actually

402
00:16:36,930 --> 00:16:42,940
see these are the famous
measurements from Keeling.

403
00:16:42,940 --> 00:16:44,630
You can see the
seasonal breathing,

404
00:16:44,630 --> 00:16:46,030
but you can also see the trends.

405
00:16:46,030 --> 00:16:49,000
And you can even see where, for
example, oil production slowed.

406
00:16:49,000 --> 00:16:50,820
You can actually see
it in this curve.

407
00:16:50,820 --> 00:16:51,700
OK?

408
00:16:51,700 --> 00:16:56,440
So it's not hard to see
anthropogenic connections

409
00:16:56,440 --> 00:16:56,950
to CO2.

410
00:16:56,950 --> 00:16:58,060
That's actually obvious.

411
00:16:58,060 --> 00:17:01,023
I think the connection
between that and temperatures

412
00:17:01,023 --> 00:17:02,440
has been debated
more, but I don't

413
00:17:02,440 --> 00:17:08,410
think many people disagree
anymore that there's

414
00:17:08,410 --> 00:17:09,710
a very strong correlation.

415
00:17:09,710 --> 00:17:11,200
And this goes back
400,000 years,

416
00:17:11,200 --> 00:17:13,119
where now I think a
million years of data--

417
00:17:13,119 --> 00:17:17,020
if you look at the ice cores,
there are bubbles in the ice.

418
00:17:17,020 --> 00:17:20,319
I promise, I'm almost done,
but I couldn't help myself,

419
00:17:20,319 --> 00:17:21,819
because it's important.

420
00:17:21,819 --> 00:17:22,408
Right?

421
00:17:22,408 --> 00:17:23,950
There are bubbles
in the ice, and you

422
00:17:23,950 --> 00:17:26,390
can see in the bubble
how much CO2 there was.

423
00:17:26,390 --> 00:17:27,108
Right?

424
00:17:27,108 --> 00:17:28,150
These things get trapped.

425
00:17:28,150 --> 00:17:29,420
It's really cool.

426
00:17:29,420 --> 00:17:29,920
Right?

427
00:17:29,920 --> 00:17:32,020
Bubbles of air get
trapped in ice,

428
00:17:32,020 --> 00:17:34,420
and they're there from
a million years ago.

429
00:17:34,420 --> 00:17:37,030
And we can take them out
and do really, really

430
00:17:37,030 --> 00:17:39,460
careful measurements.

431
00:17:39,460 --> 00:17:41,650
And we can also get
the temperature,

432
00:17:41,650 --> 00:17:45,760
because we can look at the
ratios of different isotopes

433
00:17:45,760 --> 00:17:48,010
of the water at that time.

434
00:17:48,010 --> 00:17:51,430
Because the ratios of the
isotopes of oxygen and hydrogen

435
00:17:51,430 --> 00:17:54,940
in the water at that part
of the ice tell you how warm

436
00:17:54,940 --> 00:17:56,830
the water was on average.

437
00:17:56,830 --> 00:17:58,030
Right?

438
00:17:58,030 --> 00:18:01,090
Fairly robust, and you can
see very strong correlations.

439
00:18:01,090 --> 00:18:02,320
This is where we are here.

440
00:18:02,320 --> 00:18:06,400
This is 373 from 10 years
ago, and 1,000 is very likely

441
00:18:06,400 --> 00:18:07,750
to be where we'll be.

442
00:18:07,750 --> 00:18:09,960
OK?

443
00:18:09,960 --> 00:18:10,680
OK.

444
00:18:10,680 --> 00:18:13,690
Now, so here's the thing.

445
00:18:13,690 --> 00:18:14,190
OK?

446
00:18:14,190 --> 00:18:15,190
I'm almost done.

447
00:18:15,190 --> 00:18:18,660
This is the guy who
ran for governor.

448
00:18:18,660 --> 00:18:20,610
Anybody remember him?

449
00:18:20,610 --> 00:18:22,290
A couple.

450
00:18:22,290 --> 00:18:25,590
I'm not taking a political side
here, but what I don't like

451
00:18:25,590 --> 00:18:27,660
is when this issue
gets politicized,

452
00:18:27,660 --> 00:18:29,820
which is ridiculous levels.

453
00:18:29,820 --> 00:18:33,180
It does on both sides, by the
way, both sides of whatever you

454
00:18:33,180 --> 00:18:36,150
want to call it, the debate.

455
00:18:36,150 --> 00:18:38,940
Do you debate hard,
scientific fact?

456
00:18:38,940 --> 00:18:41,220
Anyway, I can get eight
professors from MIT

457
00:18:41,220 --> 00:18:43,415
on both sides of this issue,
and no one in this room

458
00:18:43,415 --> 00:18:44,790
will walk away
understanding what

459
00:18:44,790 --> 00:18:46,463
they said about climate change.

460
00:18:46,463 --> 00:18:48,447
AUDIENCE: [INAUDIBLE]

461
00:18:48,447 --> 00:18:49,780
JEFFREY C. GROSSMAN: Well, yeah.

462
00:18:49,780 --> 00:18:54,310
So I'm not trying
to make a statement

463
00:18:54,310 --> 00:18:58,090
about this particular
politician,

464
00:18:58,090 --> 00:18:59,920
but I just want to
make this statement

465
00:18:59,920 --> 00:19:06,640
to show how this gets distorted
for political purposes.

466
00:19:06,640 --> 00:19:08,910
This is total nonsense.

467
00:19:08,910 --> 00:19:09,410
Right?

468
00:19:09,410 --> 00:19:14,050
This statement is
absolute nonsense,

469
00:19:14,050 --> 00:19:17,620
and what he's talking
about is there is actually

470
00:19:17,620 --> 00:19:21,610
someone at MIT who
has argued in actually

471
00:19:21,610 --> 00:19:26,890
very strong ways that you
need to pay attention to

472
00:19:26,890 --> 00:19:32,440
against a connection between
anthropogenic activities

473
00:19:32,440 --> 00:19:35,530
and climate change.

474
00:19:35,530 --> 00:19:37,150
There's one person.

475
00:19:37,150 --> 00:19:42,770
Most people I think do
not fall into that camp.

476
00:19:42,770 --> 00:19:48,920
But this is the way that
we tend to then just avoid

477
00:19:48,920 --> 00:19:51,020
the whole problem and
not look at it anymore.

478
00:19:51,020 --> 00:19:51,520
Right?

479
00:19:51,520 --> 00:19:56,360
So I just think that's
detestable, detestable.

480
00:19:56,360 --> 00:19:57,860
Yeah, why not?

481
00:19:57,860 --> 00:19:59,660
We'll go for it, and
this is the reason.

482
00:19:59,660 --> 00:20:00,160
Right?

483
00:20:00,160 --> 00:20:01,627
So now, this is
on the other side.

484
00:20:01,627 --> 00:20:02,710
This is on the other side.

485
00:20:02,710 --> 00:20:06,080
So one side is forget about it.

486
00:20:06,080 --> 00:20:08,450
Nobody at MIT even knows
what they're saying.

487
00:20:08,450 --> 00:20:08,950
Right?

488
00:20:08,950 --> 00:20:12,140
It's total nonsense,
and then the other side

489
00:20:12,140 --> 00:20:14,450
is we're all going to die.

490
00:20:14,450 --> 00:20:16,790
Right?

491
00:20:16,790 --> 00:20:20,250
Now, there's a lot of--

492
00:20:20,250 --> 00:20:20,750
OK.

493
00:20:20,750 --> 00:20:21,930
So this is from Climate Works.

494
00:20:21,930 --> 00:20:24,430
There's some actually really
good studies that have gone on.

495
00:20:24,430 --> 00:20:26,480
What happens when the
temperature does go up?

496
00:20:26,480 --> 00:20:27,860
Because it will.

497
00:20:27,860 --> 00:20:28,787
Right?

498
00:20:28,787 --> 00:20:31,370
It may take a little while, may
take longer than that, shorter

499
00:20:31,370 --> 00:20:32,930
than that, but it's
going to go up.

500
00:20:32,930 --> 00:20:35,120
Right?

501
00:20:35,120 --> 00:20:37,490
And as I said, it's not going
up anywhere, everywhere.

502
00:20:37,490 --> 00:20:39,650
It's going to go up in
some places a lot more,

503
00:20:39,650 --> 00:20:41,970
and in other places,
it will go down.

504
00:20:41,970 --> 00:20:43,100
OK?

505
00:20:43,100 --> 00:20:47,270
And so species extinction,
storms, droughts, all this,

506
00:20:47,270 --> 00:20:49,490
but whether you
believe this side

507
00:20:49,490 --> 00:20:51,980
and you look into those
studies, some of which

508
00:20:51,980 --> 00:20:55,250
are actually very well done
and careful, others of which

509
00:20:55,250 --> 00:20:56,810
are not.

510
00:20:56,810 --> 00:20:59,210
What you realize very
quickly is this is not

511
00:20:59,210 --> 00:21:01,930
an experiment we want to run.

512
00:21:01,930 --> 00:21:02,890
Right?

513
00:21:02,890 --> 00:21:05,540
This is not an experiment
we want to run.

514
00:21:05,540 --> 00:21:06,110
Right?

515
00:21:06,110 --> 00:21:10,790
So that should be one of
the underlying messages that

516
00:21:10,790 --> 00:21:13,520
comes across that makes us
at least, at the very least,

517
00:21:13,520 --> 00:21:15,300
take this very seriously.

518
00:21:15,300 --> 00:21:17,300
OK?

519
00:21:17,300 --> 00:21:22,070
So I could go on, and
I won't, but I just

520
00:21:22,070 --> 00:21:24,510
had to say a few things.

521
00:21:24,510 --> 00:21:25,070
OK.

522
00:21:25,070 --> 00:21:28,550
Now, back to solar PV.

523
00:21:28,550 --> 00:21:34,280
Any questions about my
little climate change rant?

524
00:21:34,280 --> 00:21:35,560
OK.

525
00:21:35,560 --> 00:21:38,680
I'd be happy to talk more
about that part of the problem

526
00:21:38,680 --> 00:21:43,860
as well, which I'm passionately,
obviously, interested in.

527
00:21:43,860 --> 00:21:44,700
Now, OK.

528
00:21:44,700 --> 00:21:49,890
So I said the motivation is that
we get a whole lot of energy--

529
00:21:49,890 --> 00:21:51,120
I said 140.

530
00:21:51,120 --> 00:21:53,640
That's 130, but anyway,
it's around there--

531
00:21:53,640 --> 00:21:57,480
from the sun, and we don't use
much of it compared to that.

532
00:21:57,480 --> 00:22:00,190
And yet, here's how we
get our energy today.

533
00:22:00,190 --> 00:22:00,960
Look at that.

534
00:22:00,960 --> 00:22:04,890
Solar wind, wood,
and waste that's bio

535
00:22:04,890 --> 00:22:07,440
is this yellow sliver.

536
00:22:07,440 --> 00:22:08,760
That's how we use our energy.

537
00:22:08,760 --> 00:22:09,720
Right?

538
00:22:09,720 --> 00:22:11,880
Most of it comes
from, of course,

539
00:22:11,880 --> 00:22:15,390
coal, natural gas, and oil.

540
00:22:15,390 --> 00:22:18,210
And yet with solar, you
wouldn't need much land.

541
00:22:18,210 --> 00:22:21,230
It's actually a pretty
land-efficient technology.

542
00:22:21,230 --> 00:22:21,750
OK?

543
00:22:21,750 --> 00:22:24,810
So if 2% of the US
were covered with PV,

544
00:22:24,810 --> 00:22:27,870
with an efficiency of 10%, you
could supply all the US energy

545
00:22:27,870 --> 00:22:29,040
needs.

546
00:22:29,040 --> 00:22:32,010
If 0.3% of the
land were covered,

547
00:22:32,010 --> 00:22:34,840
you'd get all the electricity
needs in the US met.

548
00:22:34,840 --> 00:22:38,940
0.3%, that's compared
with 40% for agriculture.

549
00:22:38,940 --> 00:22:39,720
OK?

550
00:22:39,720 --> 00:22:41,340
So it's not a land problem.

551
00:22:41,340 --> 00:22:42,540
It's not a land problem.

552
00:22:42,540 --> 00:22:43,040
OK?

553
00:22:43,040 --> 00:22:46,260
Just what's in between
highways is pretty good.

554
00:22:46,260 --> 00:22:50,070
Solar PV has a nice advantage
over some other renewables,

555
00:22:50,070 --> 00:22:53,680
in that you can scale all
the way down, all the way up.

556
00:22:53,680 --> 00:22:57,200
So that's nice, and
if you add it up,

557
00:22:57,200 --> 00:23:01,130
you can just take a small
amount of land in Nevada.

558
00:23:01,130 --> 00:23:04,130
They won't mind.

559
00:23:04,130 --> 00:23:05,150
Right?

560
00:23:05,150 --> 00:23:06,920
Who here is from Nevada?

561
00:23:06,920 --> 00:23:08,510
Would you mind?

562
00:23:08,510 --> 00:23:09,500
AUDIENCE: [INAUDIBLE]

563
00:23:09,500 --> 00:23:10,820
JEFFREY C. GROSSMAN: Thank you.

564
00:23:10,820 --> 00:23:13,815
You see what-- but come on,
there are some things there.

565
00:23:13,815 --> 00:23:14,690
AUDIENCE: [INAUDIBLE]

566
00:23:14,690 --> 00:23:16,730
JEFFREY C. GROSSMAN: There's all
kinds of-- there's arsenic life

567
00:23:16,730 --> 00:23:17,600
there.

568
00:23:17,600 --> 00:23:19,130
There's all kinds of amazing--

569
00:23:19,130 --> 00:23:27,660
so now, if you took spots,
if you had six Nevadas that

570
00:23:27,660 --> 00:23:30,420
were willing to donate
just a small amount of land

571
00:23:30,420 --> 00:23:32,370
around the world,
and you put them

572
00:23:32,370 --> 00:23:34,350
in places that are
sunny, like Nevada,

573
00:23:34,350 --> 00:23:37,080
you get 20 terawatts,
20 terawatts.

574
00:23:37,080 --> 00:23:42,550
That's more than half of what
we think will need in 2050.

575
00:23:42,550 --> 00:23:43,706
Yeah.

576
00:23:43,706 --> 00:23:47,200
AUDIENCE: [INAUDIBLE]
give an estimate something

577
00:23:47,200 --> 00:23:47,977
like this costs.

578
00:23:47,977 --> 00:23:49,060
JEFFREY C. GROSSMAN: Yeah.

579
00:23:49,060 --> 00:23:50,860
There you go, $50 trillion.

580
00:23:53,380 --> 00:23:55,810
AUDIENCE: Is that just
[INAUDIBLE] or upkeep as well?

581
00:23:55,810 --> 00:23:58,580
JEFFREY C. GROSSMAN: That's
just getting it on the ground.

582
00:23:58,580 --> 00:23:59,080
All right?

583
00:23:59,080 --> 00:24:00,455
Well, you got a
20 year warranty.

584
00:24:03,430 --> 00:24:04,060
Yeah.

585
00:24:04,060 --> 00:24:08,650
So this is the PV problem.

586
00:24:08,650 --> 00:24:12,490
This is the PV
problem, and this is

587
00:24:12,490 --> 00:24:15,010
why you don't see
solar installation

588
00:24:15,010 --> 00:24:17,710
on the top of every
roof in town and why

589
00:24:17,710 --> 00:24:20,320
we're not taking advantage
of this enormous resource.

590
00:24:20,320 --> 00:24:22,690
It's too expensive, and
compared to other renewables

591
00:24:22,690 --> 00:24:24,970
and compared certainly
to non-renewables,

592
00:24:24,970 --> 00:24:27,680
it's way too expensive.

593
00:24:27,680 --> 00:24:30,940
All other renewables,
all of the renewables,

594
00:24:30,940 --> 00:24:34,050
except for like ocean thermal--

595
00:24:34,050 --> 00:24:36,900
we're not going to do that--

596
00:24:36,900 --> 00:24:39,430
are much less expensive
than solar PV.

597
00:24:39,430 --> 00:24:39,930
OK?

598
00:24:39,930 --> 00:24:41,370
So that's the problem.

599
00:24:41,370 --> 00:24:44,250
Now it's not just--

600
00:24:44,250 --> 00:24:46,840
you got to look at it a
little bit more carefully.

601
00:24:46,840 --> 00:24:49,250
So sometimes, you see these
broadbrush statements,

602
00:24:49,250 --> 00:24:51,000
but you've got to
realize that, of course,

603
00:24:51,000 --> 00:24:56,610
this is the issue here
is one of grid parity.

604
00:24:56,610 --> 00:25:01,550
So it's about the cost of the
electricity where you are,

605
00:25:01,550 --> 00:25:04,880
and that is quite variable.

606
00:25:04,880 --> 00:25:08,410
The cost per watt at
peak hours is over here,

607
00:25:08,410 --> 00:25:11,140
and what you can
see is that there

608
00:25:11,140 --> 00:25:14,560
are some places where
there's a lot of sun

609
00:25:14,560 --> 00:25:16,360
and electricity prices are high.

610
00:25:16,360 --> 00:25:19,930
We're basically
already at grid parity

611
00:25:19,930 --> 00:25:24,658
up there, like those
places I used to live.

612
00:25:24,658 --> 00:25:26,200
And then there are
other places where

613
00:25:26,200 --> 00:25:28,900
we think that by
within 10 years we

614
00:25:28,900 --> 00:25:31,800
may reach grid parity in
a whole lot more places.

615
00:25:31,800 --> 00:25:34,210
So it depends, because
solar PV isn't just about--

616
00:25:36,880 --> 00:25:40,167
it's not just about the
sun's average over the Earth.

617
00:25:40,167 --> 00:25:41,500
It's about really where you are.

618
00:25:41,500 --> 00:25:43,130
How much sun is there?

619
00:25:43,130 --> 00:25:50,650
That's why you can get I think
the payback time in California

620
00:25:50,650 --> 00:25:53,600
is like seven years, and here,
it's double that or something.

621
00:25:57,200 --> 00:25:57,860
OK.

622
00:25:57,860 --> 00:26:00,980
So that's an important
point, and so I just

623
00:26:00,980 --> 00:26:03,410
want to show this, because you
see sometimes these like it

624
00:26:03,410 --> 00:26:05,340
would cost $50 trillion,
so let's not do it,

625
00:26:05,340 --> 00:26:07,933
or it's going to be--

626
00:26:07,933 --> 00:26:09,600
or we can do it,
because it's all there.

627
00:26:09,600 --> 00:26:10,625
We should just do it.

628
00:26:10,625 --> 00:26:11,750
We see these big arguments.

629
00:26:11,750 --> 00:26:13,670
I think it's sometimes
important to come down

630
00:26:13,670 --> 00:26:16,580
to more like practical ideas.

631
00:26:16,580 --> 00:26:18,080
Right?

632
00:26:18,080 --> 00:26:23,510
And one way to look at this
is just what happened there?

633
00:26:23,510 --> 00:26:25,370
I did not put that
into the graph.

634
00:26:28,310 --> 00:26:32,150
If you look at, say, the
percentage of electricity

635
00:26:32,150 --> 00:26:36,200
sold in the US at
or below some price,

636
00:26:36,200 --> 00:26:38,090
and you look at-- let's
see if this works.

637
00:26:38,090 --> 00:26:43,688
Yes-- and you look at just the
10% most expensive of that,

638
00:26:43,688 --> 00:26:45,230
there are all kinds
of questions that

639
00:26:45,230 --> 00:26:49,130
come into this argument
concerning what would happen

640
00:26:49,130 --> 00:26:51,260
to these higher priced markets.

641
00:26:51,260 --> 00:26:53,780
But let's just suppose
you could grab them

642
00:26:53,780 --> 00:26:55,400
at their current price.

643
00:26:55,400 --> 00:26:57,470
Then, the top 10%
would get you down

644
00:26:57,470 --> 00:27:01,100
to $0.14, $0.14 a kilowatt hour.

645
00:27:01,100 --> 00:27:02,910
That's the price target.

646
00:27:02,910 --> 00:27:04,940
So the question is,
could we replace

647
00:27:04,940 --> 00:27:10,080
10% of electrical energy in the
US with PV at a cost of $0.14?

648
00:27:10,080 --> 00:27:11,080
That becomes the target.

649
00:27:11,080 --> 00:27:11,580
Right?

650
00:27:11,580 --> 00:27:15,400
Actually, now that starts
to sound reasonable,

651
00:27:15,400 --> 00:27:18,400
not easy in any way,
but more reasonable

652
00:27:18,400 --> 00:27:22,000
than replacing all energy
or all electricity.

653
00:27:22,000 --> 00:27:24,970
One important point is it
actually could be done.

654
00:27:24,970 --> 00:27:26,550
It could be deployed.

655
00:27:26,550 --> 00:27:27,580
OK?

656
00:27:27,580 --> 00:27:29,080
But another really
important point

657
00:27:29,080 --> 00:27:32,200
is actually, if anyone says I'm
going to replace all of the US

658
00:27:32,200 --> 00:27:34,750
electricity needs with
solar, ask them how they plan

659
00:27:34,750 --> 00:27:37,100
to store 90% of that energy.

660
00:27:37,100 --> 00:27:37,930
OK?

661
00:27:37,930 --> 00:27:40,190
The grid can't take it.

662
00:27:40,190 --> 00:27:42,650
The grid is not a battery.

663
00:27:42,650 --> 00:27:44,960
It sort of is a little
bit, but it's not a lot.

664
00:27:44,960 --> 00:27:46,130
It can't do it.

665
00:27:46,130 --> 00:27:50,610
It can't handle the
storage that you

666
00:27:50,610 --> 00:27:54,720
would need right now, if you
replaced everything with PV.

667
00:27:54,720 --> 00:27:56,820
And in fact there's
no storage technology

668
00:27:56,820 --> 00:28:00,730
right now that exists
that could handle it.

669
00:28:00,730 --> 00:28:01,720
Right?

670
00:28:01,720 --> 00:28:07,030
So keep that in mind, when your
friend says from say Harvard,

671
00:28:07,030 --> 00:28:07,960
well, I have a plan.

672
00:28:10,840 --> 00:28:12,190
I have a business strategy.

673
00:28:12,190 --> 00:28:14,200
We're going to replace
everything with PV.

674
00:28:14,200 --> 00:28:16,120
Ask them how they're
going to store it.

675
00:28:16,120 --> 00:28:17,660
Right?

676
00:28:17,660 --> 00:28:18,800
OK.

677
00:28:18,800 --> 00:28:21,800
But 10% you could
squeeze in there

678
00:28:21,800 --> 00:28:26,620
without inventing a
completely new battery.

679
00:28:26,620 --> 00:28:27,120
OK?

680
00:28:31,420 --> 00:28:32,107
Oh, yeah.

681
00:28:32,107 --> 00:28:32,940
This is a side note.

682
00:28:35,670 --> 00:28:37,620
We think a lot about
our developing--

683
00:28:37,620 --> 00:28:38,850
I think maybe I showed this.

684
00:28:38,850 --> 00:28:39,840
I don't remember.

685
00:28:39,840 --> 00:28:43,740
We think a lot about the CO2
in the developed countries,

686
00:28:43,740 --> 00:28:45,480
but this is really
where the problems

687
00:28:45,480 --> 00:28:47,880
are going to be, is how
they choose to do things.

688
00:28:47,880 --> 00:28:49,110
OK?

689
00:28:49,110 --> 00:28:51,470
OK.

690
00:28:51,470 --> 00:28:52,700
OK.

691
00:28:52,700 --> 00:28:56,570
Now, we're going to talk
about how a solar cell works

692
00:28:56,570 --> 00:28:59,780
and how computational
quantum mechanics can

693
00:28:59,780 --> 00:29:05,660
impact this problem, this
problem of cost really.

694
00:29:05,660 --> 00:29:08,060
And just to set the
stage, I want to be sure.

695
00:29:08,060 --> 00:29:11,690
So this is what you
already know this that--

696
00:29:11,690 --> 00:29:13,280
let me get back to this.

697
00:29:13,280 --> 00:29:17,150
The sun, when it's
above the atmosphere--

698
00:29:17,150 --> 00:29:21,440
because you had these spectra
in the second homework,

699
00:29:21,440 --> 00:29:24,170
and we talked about this.

700
00:29:24,170 --> 00:29:30,440
If you're up here, then
that's how much power

701
00:29:30,440 --> 00:29:32,090
you get from the sun.

702
00:29:32,090 --> 00:29:35,187
And if you're now
when you're down here,

703
00:29:35,187 --> 00:29:37,520
then you have the atmosphere
which absorbs some of that.

704
00:29:37,520 --> 00:29:39,500
Remember that causes
dips in the spectrum,

705
00:29:39,500 --> 00:29:41,557
because you've got
certain molecules

706
00:29:41,557 --> 00:29:43,640
in the atmosphere that
just take all the light out

707
00:29:43,640 --> 00:29:44,630
of that frequency.

708
00:29:44,630 --> 00:29:45,320
Right?

709
00:29:45,320 --> 00:29:47,720
And then what we
tend to do, usually

710
00:29:47,720 --> 00:29:51,200
when we talk about solar
PV, is we say, well,

711
00:29:51,200 --> 00:29:53,570
that's actually not
what you really get.

712
00:29:53,570 --> 00:29:55,100
What you really
get on average is

713
00:29:55,100 --> 00:29:59,120
something like the sun
coming through at an angle.

714
00:29:59,120 --> 00:30:01,090
Now, why is that important?

715
00:30:01,090 --> 00:30:04,820
So this is the standard solar
PV spectrum that you use.

716
00:30:04,820 --> 00:30:07,430
It's called AM1.5, Air Mass.

717
00:30:07,430 --> 00:30:09,630
So you take the air
mass into account,

718
00:30:09,630 --> 00:30:14,360
and it's 1.5, because
it's at 45 degrees.

719
00:30:14,360 --> 00:30:16,750
Why does that matter?

720
00:30:16,750 --> 00:30:17,530
Yeah?

721
00:30:17,530 --> 00:30:21,680
AUDIENCE: It's at
an angle [INAUDIBLE]

722
00:30:21,680 --> 00:30:22,763
JEFFREY C. GROSSMAN: Yeah.

723
00:30:22,763 --> 00:30:25,120
Well, so definitely there's
an angular dependence.

724
00:30:25,120 --> 00:30:26,980
That's absolutely right.

725
00:30:26,980 --> 00:30:29,860
There is an angular
dependence of the power.

726
00:30:29,860 --> 00:30:30,430
Right?

727
00:30:30,430 --> 00:30:31,720
That's the cosine.

728
00:30:31,720 --> 00:30:35,410
You've got a cosine function,
basically, and interdependence

729
00:30:35,410 --> 00:30:37,990
of the power that the solar
cell generates with respect

730
00:30:37,990 --> 00:30:40,553
to the position of the sun.

731
00:30:40,553 --> 00:30:42,970
But what else is going on,
when I put the sun at an angle?

732
00:30:45,958 --> 00:30:47,250
AUDIENCE: It's more [INAUDIBLE]

733
00:30:47,250 --> 00:30:47,880
JEFFREY C. GROSSMAN: Yeah.

734
00:30:47,880 --> 00:30:49,110
It's more air to go through.

735
00:30:49,110 --> 00:30:49,800
Right?

736
00:30:49,800 --> 00:30:54,720
So that's why very
often this is the power.

737
00:30:54,720 --> 00:30:56,472
This is the average.

738
00:30:56,472 --> 00:30:57,930
So there is more
air to go through,

739
00:30:57,930 --> 00:30:59,940
less air to go through,
more air to go through.

740
00:30:59,940 --> 00:31:00,440
Right?

741
00:31:00,440 --> 00:31:02,940
And so we just have
some standards.

742
00:31:02,940 --> 00:31:06,280
A lot of times, you take
AM1.5 as the standard.

743
00:31:06,280 --> 00:31:09,750
It's what you use when you
buy like a solar simulator,

744
00:31:09,750 --> 00:31:11,830
and you do the tests in the lab.

745
00:31:11,830 --> 00:31:12,730
That what we use.

746
00:31:12,730 --> 00:31:13,230
OK.

747
00:31:13,230 --> 00:31:15,022
So now, we'll go through
every single curve

748
00:31:15,022 --> 00:31:20,080
here and have a quiz on it.

749
00:31:20,080 --> 00:31:20,660
Yeah.

750
00:31:20,660 --> 00:31:24,840
No, this is actually-- so
has anybody seen this graph?

751
00:31:24,840 --> 00:31:28,380
One person, two people, OK.

752
00:31:28,380 --> 00:31:33,905
This is actually, if you
meet somebody in a dark alley

753
00:31:33,905 --> 00:31:34,405
someday.

754
00:31:37,160 --> 00:31:40,370
And you've got your
panel, your solar panel,

755
00:31:40,370 --> 00:31:43,560
and you pull it out, and
they pull theirs out.

756
00:31:43,560 --> 00:31:46,520
And you want to know who's
got a more efficient panel,

757
00:31:46,520 --> 00:31:47,810
who decides?

758
00:31:47,810 --> 00:31:48,440
Right?

759
00:31:48,440 --> 00:31:50,840
NREL, that's where you take it.

760
00:31:50,840 --> 00:31:52,050
You take it to NREL.

761
00:31:52,050 --> 00:31:52,550
Right?

762
00:31:52,550 --> 00:31:54,890
So you say, take it outside.

763
00:31:54,890 --> 00:31:56,590
Take it to NREL.

764
00:31:56,590 --> 00:31:57,690
NREL is the National--

765
00:31:57,690 --> 00:31:59,950
Does anybody know who NREL is?

766
00:31:59,950 --> 00:32:02,380
National Renewable
Energy Lab, it's over

767
00:32:02,380 --> 00:32:05,680
in Colorado, not a bad state.

768
00:32:05,680 --> 00:32:09,430
And they do this
very careful testing

769
00:32:09,430 --> 00:32:15,670
on all kinds of
solar technologies.

770
00:32:15,670 --> 00:32:18,660
One of the cool things
here about this to me

771
00:32:18,660 --> 00:32:22,980
is, first of all, look at how
many materials make electricity

772
00:32:22,980 --> 00:32:24,430
from the sun.

773
00:32:24,430 --> 00:32:26,320
Isn't that amazing?

774
00:32:26,320 --> 00:32:28,300
I get really excited about that.

775
00:32:28,300 --> 00:32:31,263
Now, when you buy
a solar cell today,

776
00:32:31,263 --> 00:32:32,680
there's a little
bit of variation,

777
00:32:32,680 --> 00:32:36,470
but most of the time,
it's made out of what?

778
00:32:36,470 --> 00:32:38,890
Silicon.

779
00:32:38,890 --> 00:32:41,050
Silicon, and in
fact, look at here.

780
00:32:41,050 --> 00:32:43,090
There's different kinds,
but actually, it's

781
00:32:43,090 --> 00:32:45,010
really multicrystalline.

782
00:32:45,010 --> 00:32:47,080
Single crystalline is
a little more efficient

783
00:32:47,080 --> 00:32:47,980
but more expensive.

784
00:32:47,980 --> 00:32:50,560
So most of the cells you
buy are multicrystalline.

785
00:32:50,560 --> 00:32:53,650
So you can find it, and
that's these squares here.

786
00:32:53,650 --> 00:32:57,610
And you can see that silicon
has gotten better with time.

787
00:32:57,610 --> 00:33:00,970
The efficiency used to be there,
and it keeps on getting better.

788
00:33:00,970 --> 00:33:04,810
And now, if you want to
buy silicon solar cells,

789
00:33:04,810 --> 00:33:06,220
you can buy them.

790
00:33:06,220 --> 00:33:08,110
I can't get 20%--

791
00:33:08,110 --> 00:33:12,610
multicrystalline silicon, you
can't get that on the market

792
00:33:12,610 --> 00:33:16,537
easily, but I can
get about 18% cells.

793
00:33:16,537 --> 00:33:17,620
Actually, that's not true.

794
00:33:17,620 --> 00:33:20,370
Multicrystalline I
can get lower cells.

795
00:33:20,370 --> 00:33:26,968
So these are like the efficiency
that you could get at most,

796
00:33:26,968 --> 00:33:28,760
and then there's like
what you can actually

797
00:33:28,760 --> 00:33:30,630
get in your product.

798
00:33:30,630 --> 00:33:31,230
Right?

799
00:33:31,230 --> 00:33:32,700
And there's a big difference.

800
00:33:32,700 --> 00:33:33,200
Right?

801
00:33:33,200 --> 00:33:36,590
So these are like
what the record has

802
00:33:36,590 --> 00:33:40,460
been in research labs,
tested and verified by NREL,

803
00:33:40,460 --> 00:33:42,770
and then there's
like now I buy it.

804
00:33:42,770 --> 00:33:45,290
Who can sell me it, and
what's your efficiency?

805
00:33:45,290 --> 00:33:46,790
And it's always lower.

806
00:33:46,790 --> 00:33:49,610
And for monocrystalline
silicon, which

807
00:33:49,610 --> 00:33:52,220
is like the best you
can do, that's here,

808
00:33:52,220 --> 00:33:54,800
you see that we're
pushing what I'll

809
00:33:54,800 --> 00:33:58,130
show you is close to the
theoretical limit of how

810
00:33:58,130 --> 00:34:01,160
efficient this can
even do this process.

811
00:34:01,160 --> 00:34:05,180
But that's 27%, but
if I buy it, 18%

812
00:34:05,180 --> 00:34:07,980
is the best you can get on
the market of monocrystalline.

813
00:34:07,980 --> 00:34:08,480
Right?

814
00:34:08,480 --> 00:34:10,520
Now, why is that?

815
00:34:10,520 --> 00:34:13,489
Why is there such a difference
between the research cell

816
00:34:13,489 --> 00:34:17,034
and what I can buy today?

817
00:34:17,034 --> 00:34:21,052
AUDIENCE: [INAUDIBLE]
But then you also

818
00:34:21,052 --> 00:34:22,094
have the bidding process.

819
00:34:22,094 --> 00:34:25,949
You have interim [INAUDIBLE]
mass production [INAUDIBLE]

820
00:34:25,949 --> 00:34:29,820
JEFFREY C. GROSSMAN: Scale
up, scale up is hard.

821
00:34:29,820 --> 00:34:32,550
So I can make like a
little solar cell like this

822
00:34:32,550 --> 00:34:33,780
and send it off to NREL.

823
00:34:33,780 --> 00:34:37,080
And they'll mark it at 28%,
and now I get $100 million

824
00:34:37,080 --> 00:34:41,320
from some happy-go-lucky VC.

825
00:34:41,320 --> 00:34:44,469
And I build my plant,
and all of a sudden,

826
00:34:44,469 --> 00:34:47,080
there are regions like
this in the cells I make,

827
00:34:47,080 --> 00:34:49,719
but most of the
cell is like 10%.

828
00:34:49,719 --> 00:34:50,270
Right?

829
00:34:50,270 --> 00:34:50,770
Why?

830
00:34:50,770 --> 00:34:55,239
Well, manufacturability, scale
up, huge issues with solar PV,

831
00:34:55,239 --> 00:34:56,790
huge issues.

832
00:34:56,790 --> 00:34:57,340
OK?

833
00:34:57,340 --> 00:35:00,010
There was a spin
off that I won't

834
00:35:00,010 --> 00:35:03,160
name names, where it was really
exciting, because they were

835
00:35:03,160 --> 00:35:05,410
literally like putting hard
drives in the toaster oven

836
00:35:05,410 --> 00:35:07,210
and getting out solar cells.

837
00:35:07,210 --> 00:35:11,620
And it was such a cool
idea using old material

838
00:35:11,620 --> 00:35:15,040
from other processes, and
little parts of those cells

839
00:35:15,040 --> 00:35:16,000
were so efficient.

840
00:35:16,000 --> 00:35:18,220
It was really promising,
but with the homogeneity

841
00:35:18,220 --> 00:35:19,930
was terrible.

842
00:35:19,930 --> 00:35:23,680
And that's because you
need processes that can

843
00:35:23,680 --> 00:35:27,520
get you uniformity at scale.

844
00:35:27,520 --> 00:35:29,760
This does not tell
you about that.

845
00:35:29,760 --> 00:35:31,860
Right?

846
00:35:31,860 --> 00:35:34,350
Now, I love this
area down here, where

847
00:35:34,350 --> 00:35:36,790
you have all kinds of
new materials happening.

848
00:35:36,790 --> 00:35:38,730
So in my group, we
work on the bottom

849
00:35:38,730 --> 00:35:42,300
of the barrel in efficiency,
but it's really exciting.

850
00:35:42,300 --> 00:35:44,370
Because these are all
different materials, where

851
00:35:44,370 --> 00:35:48,570
I think there could be
big changes in efficiency,

852
00:35:48,570 --> 00:35:52,680
like materials made out
of carbon, quantum dots.

853
00:35:52,680 --> 00:35:54,820
You can make solar cells
out of quantum dots.

854
00:35:54,820 --> 00:35:55,860
OK?

855
00:35:55,860 --> 00:35:59,560
Why would I care about those
kinds of other materials?

856
00:35:59,560 --> 00:36:02,460
What might they offer me?

857
00:36:02,460 --> 00:36:03,263
Yeah?

858
00:36:03,263 --> 00:36:06,498
AUDIENCE: [INAUDIBLE]

859
00:36:06,498 --> 00:36:08,790
JEFFREY C. GROSSMAN: Let's
say, I can get them up here.

860
00:36:08,790 --> 00:36:10,457
Let's say, I can get
a quantum dot cell.

861
00:36:10,457 --> 00:36:13,040
Let's look at my
little legend there.

862
00:36:13,040 --> 00:36:14,870
Where are my quantum
dots, the triangles?

863
00:36:14,870 --> 00:36:19,170
Oh, they've only been
around since 2010.

864
00:36:19,170 --> 00:36:20,120
How cool is that?

865
00:36:20,120 --> 00:36:21,660
And look at that curve.

866
00:36:21,660 --> 00:36:23,420
They're doing well.

867
00:36:23,420 --> 00:36:26,030
If you have two points,
you can make a line.

868
00:36:26,030 --> 00:36:26,880
Right?

869
00:36:26,880 --> 00:36:30,230
And if you extrapolate
this, in just six years,

870
00:36:30,230 --> 00:36:33,840
they're going to be as
efficient as silicon.

871
00:36:33,840 --> 00:36:35,140
Why do I care?

872
00:36:35,140 --> 00:36:39,030
Why do I want all
these materials?

873
00:36:39,030 --> 00:36:40,080
Yeah?

874
00:36:40,080 --> 00:36:41,820
AUDIENCE: [INAUDIBLE]
less expensive

875
00:36:41,820 --> 00:36:44,032
or more easy to mass produce.

876
00:36:44,032 --> 00:36:45,240
JEFFREY C. GROSSMAN: Exactly.

877
00:36:45,240 --> 00:36:45,740
Yeah.

878
00:36:45,740 --> 00:36:47,250
It's all about cost.

879
00:36:47,250 --> 00:36:48,150
It's all about cost.

880
00:36:48,150 --> 00:36:53,290
So new materials might
give me different ways,

881
00:36:53,290 --> 00:36:58,660
especially like low temperature
solution processing is a key.

882
00:36:58,660 --> 00:36:59,160
Right?

883
00:36:59,160 --> 00:37:03,810
So silicon PV, you got to bake
this stuff at like 11 times.

884
00:37:03,810 --> 00:37:06,330
You put it in the
oven at 1,000 Celsius.

885
00:37:06,330 --> 00:37:07,230
Right?

886
00:37:07,230 --> 00:37:10,980
I'm just making that up, but
it's not too far from that.

887
00:37:10,980 --> 00:37:14,490
Low temperature solution
processing is much cheaper.

888
00:37:14,490 --> 00:37:15,000
OK.

889
00:37:15,000 --> 00:37:16,958
Now, the other thing that
you need to remember,

890
00:37:16,958 --> 00:37:19,590
and I'll get to computation
in a moment, in a little bit,

891
00:37:19,590 --> 00:37:24,542
but I want to finish this
overview of solar PV.

892
00:37:24,542 --> 00:37:26,250
The other thing that
you have to remember

893
00:37:26,250 --> 00:37:33,300
is that the cost of solar
cells is not just the material.

894
00:37:33,300 --> 00:37:36,050
So it is the material,
but notice that--

895
00:37:36,050 --> 00:37:37,300
so here's crystalline silicon.

896
00:37:37,300 --> 00:37:40,030
Most of the solar cells
today are still made of this.

897
00:37:40,030 --> 00:37:42,480
But notice the blue here
is the cost of the module.

898
00:37:42,480 --> 00:37:46,080
That's like the material
and the glass stuff,

899
00:37:46,080 --> 00:37:49,340
and then the green is
the installation cost.

900
00:37:49,340 --> 00:37:51,230
The installation
cost is actually

901
00:37:51,230 --> 00:37:53,750
more than the
material module cost.

902
00:37:53,750 --> 00:37:55,970
So in solar cells,
actually, that

903
00:37:55,970 --> 00:38:02,620
has become only recently
the biggest problem is

904
00:38:02,620 --> 00:38:04,410
the installation costs.

905
00:38:04,410 --> 00:38:07,350
Even just 10 years ago, the
module cost was twice as much

906
00:38:07,350 --> 00:38:08,040
as installation.

907
00:38:08,040 --> 00:38:10,590
Today, it's half as much.

908
00:38:10,590 --> 00:38:12,200
OK?

909
00:38:12,200 --> 00:38:17,450
So you can see
that, if I go nano,

910
00:38:17,450 --> 00:38:18,720
this means a lot of things.

911
00:38:18,720 --> 00:38:22,640
I can make cells that have
much cheaper materials costs,

912
00:38:22,640 --> 00:38:24,900
but see they're
not very efficient.

913
00:38:24,900 --> 00:38:27,770
So if I want to normalize
my cost per watt,

914
00:38:27,770 --> 00:38:31,260
I got to install a
whole lot more of them.

915
00:38:31,260 --> 00:38:32,880
So my green curve
is really high.

916
00:38:32,880 --> 00:38:33,530
You see?

917
00:38:33,530 --> 00:38:34,940
Does everybody see that?

918
00:38:34,940 --> 00:38:35,960
So that's the balance.

919
00:38:35,960 --> 00:38:36,460
Right?

920
00:38:36,460 --> 00:38:39,470
Now, the interesting
thing is that, you see,

921
00:38:39,470 --> 00:38:43,430
one of the reasons why
silicon PV is so expensive

922
00:38:43,430 --> 00:38:45,545
is, why the insulation
is expensive,

923
00:38:45,545 --> 00:38:49,440
is because it has to be put
on glass, and glass is heavy.

924
00:38:49,440 --> 00:38:50,330
Right?

925
00:38:50,330 --> 00:38:53,210
There are laws in this
country about how much

926
00:38:53,210 --> 00:38:58,080
a worker can actually carry
on a roof, and that limits,

927
00:38:58,080 --> 00:39:00,870
literally, that limits how
big you can make a panel.

928
00:39:00,870 --> 00:39:01,557
OK?

929
00:39:01,557 --> 00:39:03,390
The weight is a real
problem, and then there

930
00:39:03,390 --> 00:39:05,265
are other costs
as well on they're

931
00:39:05,265 --> 00:39:08,850
called balance of systems that
go into that installation cost.

932
00:39:08,850 --> 00:39:12,270
But what if you
could make materials

933
00:39:12,270 --> 00:39:18,380
that could be put on
templates other than glass?

934
00:39:18,380 --> 00:39:20,090
Maybe then you
could really change

935
00:39:20,090 --> 00:39:23,420
the installation, the
paradigm on the installation.

936
00:39:23,420 --> 00:39:26,060
That's a big, big
reason why we're

937
00:39:26,060 --> 00:39:29,150
interested in all of these.

938
00:39:29,150 --> 00:39:29,650
Right?

939
00:39:29,650 --> 00:39:31,640
So these are all thin film.

940
00:39:31,640 --> 00:39:33,470
These are all thin film.

941
00:39:33,470 --> 00:39:35,600
This is thick film.

942
00:39:35,600 --> 00:39:42,670
Now, what can you compute
because of this class

943
00:39:42,670 --> 00:39:45,160
to tell me why this is
thick, and these are thin?

944
00:39:48,080 --> 00:39:50,270
Is it the band gap?

945
00:39:50,270 --> 00:39:51,792
What about the band gap?

946
00:39:51,792 --> 00:39:53,120
AUDIENCE: [INAUDIBLE]

947
00:39:53,120 --> 00:39:55,850
JEFFREY C. GROSSMAN:
Direct or indirect?

948
00:39:55,850 --> 00:39:58,490
And as a first thing, as
a first approximation,

949
00:39:58,490 --> 00:40:00,850
is that band gap
direct or indirect?

950
00:40:00,850 --> 00:40:04,730
indirect here, direct
here, therefore, you

951
00:40:04,730 --> 00:40:07,950
can make very thin films.

952
00:40:07,950 --> 00:40:10,530
You can make amorphous
silicon at a micron.

953
00:40:10,530 --> 00:40:13,530
You've got to make
this at 100 microns

954
00:40:13,530 --> 00:40:14,650
to absorb all the light.

955
00:40:14,650 --> 00:40:16,317
If you don't make it
that thick, it just

956
00:40:16,317 --> 00:40:18,272
won't absorb enough light.

957
00:40:18,272 --> 00:40:19,480
Why isn't it absorbing light?

958
00:40:19,480 --> 00:40:21,580
The band structure tells you.

959
00:40:21,580 --> 00:40:23,315
OK?

960
00:40:23,315 --> 00:40:24,940
The band structure
would also tell you,

961
00:40:24,940 --> 00:40:27,700
or these electrons
states would also

962
00:40:27,700 --> 00:40:30,010
tell you, that for
nanostructure PV,

963
00:40:30,010 --> 00:40:33,730
you can make them as thin
as 50 to 100 nanometers

964
00:40:33,730 --> 00:40:35,390
and absorb all the light.

965
00:40:35,390 --> 00:40:38,030
That's how efficient they
can be at absorbing light.

966
00:40:38,030 --> 00:40:38,530
OK?

967
00:40:38,530 --> 00:40:41,883
So that is a really
important metric,

968
00:40:41,883 --> 00:40:43,300
because if you can
make them thin,

969
00:40:43,300 --> 00:40:46,060
then you can put them maybe
on different templates that

970
00:40:46,060 --> 00:40:47,380
are flexible.

971
00:40:47,380 --> 00:40:51,140
Maybe you don't need a
thick glass template.

972
00:40:51,140 --> 00:40:54,190
The solution processing
is, obviously,

973
00:40:54,190 --> 00:40:55,852
equally or more important.

974
00:40:58,630 --> 00:41:00,570
Any questions?

975
00:41:00,570 --> 00:41:03,470
There's another issue that
I just want to mention that

976
00:41:03,470 --> 00:41:07,610
people, happy-go-lucky
PV people--

977
00:41:07,610 --> 00:41:08,710
there are a lot of them--

978
00:41:08,710 --> 00:41:11,630
talk about, which is just--

979
00:41:11,630 --> 00:41:14,540
I should say, don't talk
about, which is just again

980
00:41:14,540 --> 00:41:16,890
this scale, this
production scale.

981
00:41:16,890 --> 00:41:17,390
Right?

982
00:41:17,390 --> 00:41:19,250
We talk about how
we're going to get

983
00:41:19,250 --> 00:41:23,930
to 81 kilowatt hours a day per
person of solar electricity.

984
00:41:23,930 --> 00:41:28,180
That's from Without Hot Air,
an analysis done in that book.

985
00:41:28,180 --> 00:41:29,180
That should be a target.

986
00:41:29,180 --> 00:41:29,660
OK.

987
00:41:29,660 --> 00:41:31,077
So but the thing
is that, you see,

988
00:41:31,077 --> 00:41:33,747
if you want to meet that target,
then what you're going to need

989
00:41:33,747 --> 00:41:35,330
is you're going to
need to get to like

990
00:41:35,330 --> 00:41:38,930
100% of the US electricity
generated by solar PV

991
00:41:38,930 --> 00:41:39,740
or close to that.

992
00:41:39,740 --> 00:41:41,448
And that would require
being able to make

993
00:41:41,448 --> 00:41:47,295
14,000 acres per day, and
today, we make 14 acres per day.

994
00:41:47,295 --> 00:41:48,670
So this is the
kind of thing, you

995
00:41:48,670 --> 00:41:52,060
got to look at these
problems as well,

996
00:41:52,060 --> 00:41:54,550
when you think about
the solar PV problem.

997
00:41:54,550 --> 00:41:57,100
You can't just dream up
a material and say, aha.

998
00:41:57,100 --> 00:41:57,640
Right?

999
00:41:57,640 --> 00:41:59,290
You got to really
ask how it's going

1000
00:41:59,290 --> 00:42:02,970
to fit in to the full problem.

1001
00:42:02,970 --> 00:42:04,470
OK?

1002
00:42:04,470 --> 00:42:08,040
So anything, any
new technology, has

1003
00:42:08,040 --> 00:42:11,400
to accelerate over the
silicon production rate,

1004
00:42:11,400 --> 00:42:12,750
has to be better than silicon.

1005
00:42:12,750 --> 00:42:15,650
Silicon is a moving
target, and it

1006
00:42:15,650 --> 00:42:18,800
has to retire efficiencies
and/or lower installation

1007
00:42:18,800 --> 00:42:22,440
costs for different reasons,
as I just mentioned.

1008
00:42:22,440 --> 00:42:22,940
OK.

1009
00:42:22,940 --> 00:42:24,398
Now, this is one
I'm going to skip,

1010
00:42:24,398 --> 00:42:25,852
but I put it here for you.

1011
00:42:25,852 --> 00:42:28,060
Because this is what I don't
want to talk about which

1012
00:42:28,060 --> 00:42:29,990
is the device view.

1013
00:42:29,990 --> 00:42:32,610
This is what you'd learn
in like a EE class.

1014
00:42:32,610 --> 00:42:33,950
OK?

1015
00:42:33,950 --> 00:42:35,930
And it's great, and
it's a really good way

1016
00:42:35,930 --> 00:42:37,130
to look at solar
cells, and I've got

1017
00:42:37,130 --> 00:42:38,338
two slides that summarize it.

1018
00:42:38,338 --> 00:42:44,980
Basically, a solar cell is a
diode that you shine light on.

1019
00:42:44,980 --> 00:42:47,500
I don't want to talk
about that picture,

1020
00:42:47,500 --> 00:42:52,790
happy to do so offline,
after class, whenever.

1021
00:42:52,790 --> 00:42:54,430
So this is the typical
kind of picture,

1022
00:42:54,430 --> 00:42:56,110
because this does not--

1023
00:42:56,110 --> 00:42:58,630
this picture does
not let me take

1024
00:42:58,630 --> 00:43:01,160
the viewpoint of electrons.

1025
00:43:01,160 --> 00:43:05,860
That's where we're coming
from in this class is, well,

1026
00:43:05,860 --> 00:43:08,470
what does an electron do in
this material at the scale

1027
00:43:08,470 --> 00:43:11,750
of the atom, and that's the
picture I want to start with.

1028
00:43:11,750 --> 00:43:15,430
So I'm going to go down
the level from these device

1029
00:43:15,430 --> 00:43:18,250
models which are
extremely useful,

1030
00:43:18,250 --> 00:43:21,300
and I'm going to
look at this picture.

1031
00:43:21,300 --> 00:43:22,260
OK?

1032
00:43:22,260 --> 00:43:25,110
So this is the picture.

1033
00:43:25,110 --> 00:43:29,310
This is what an electron,
our precious electrons,

1034
00:43:29,310 --> 00:43:30,270
our electrons.

1035
00:43:30,270 --> 00:43:33,660
We now feel that they
are ours in this class.

1036
00:43:33,660 --> 00:43:35,970
Do we own electrons?

1037
00:43:35,970 --> 00:43:37,590
Yeah?

1038
00:43:37,590 --> 00:43:38,430
OK.

1039
00:43:38,430 --> 00:43:41,820
I saw some doubt, but it's OK.

1040
00:43:41,820 --> 00:43:45,360
We still have Thursday.

1041
00:43:45,360 --> 00:43:49,560
We simulate electrons, and
that's what we've been doing.

1042
00:43:49,560 --> 00:43:51,600
This whole part of
the class how can you

1043
00:43:51,600 --> 00:43:55,350
model the behavior of electrons,
and in particular, what

1044
00:43:55,350 --> 00:43:57,870
are the energies of
electrons in a material?

1045
00:43:57,870 --> 00:44:00,280
That's really been our key goal.

1046
00:44:00,280 --> 00:44:01,320
OK?

1047
00:44:01,320 --> 00:44:04,620
And this is how an
electron in that landscape

1048
00:44:04,620 --> 00:44:05,940
sees a solar cell.

1049
00:44:05,940 --> 00:44:08,520
Light shines on the material,
and an electron's just

1050
00:44:08,520 --> 00:44:11,580
cruising around, and
then it's like, oh man,

1051
00:44:11,580 --> 00:44:13,970
I just got hit by a photon.

1052
00:44:13,970 --> 00:44:14,630
right?

1053
00:44:14,630 --> 00:44:17,600
And then it's like, I
better take that energy in,

1054
00:44:17,600 --> 00:44:21,530
but we know that for an
electron, when it takes energy

1055
00:44:21,530 --> 00:44:22,730
in--

1056
00:44:22,730 --> 00:44:26,070
Oh, David MacKay is the guy
that wrote Without Hot Air.

1057
00:44:26,070 --> 00:44:30,065
OK-- when it takes that
energy in, what happens?

1058
00:44:32,610 --> 00:44:33,940
Let's say these are filled.

1059
00:44:33,940 --> 00:44:34,440
Right?

1060
00:44:34,440 --> 00:44:36,780
This is my Fermi energy.

1061
00:44:36,780 --> 00:44:38,460
It's going to jump up.

1062
00:44:38,460 --> 00:44:41,220
That's what electrons do,
when they take energy in.

1063
00:44:41,220 --> 00:44:42,370
We've talked about that.

1064
00:44:42,370 --> 00:44:46,160
So that's what happened, and
then, what did it leave behind?

1065
00:44:46,160 --> 00:44:50,300
A hole which is the
same as an electron,

1066
00:44:50,300 --> 00:44:53,720
but not, but it's positive.

1067
00:44:53,720 --> 00:44:54,470
Right?

1068
00:44:54,470 --> 00:44:57,290
So that's a positive charge,
and there's my negative charge.

1069
00:44:57,290 --> 00:45:03,386
And that's what happened when
sun shined, showned, shown,

1070
00:45:03,386 --> 00:45:05,360
shown on it.

1071
00:45:05,360 --> 00:45:08,554
And then what happens here?

1072
00:45:08,554 --> 00:45:09,990
AUDIENCE: [INAUDIBLE]

1073
00:45:09,990 --> 00:45:11,850
JEFFREY C. GROSSMAN: Yeah.

1074
00:45:11,850 --> 00:45:12,570
It relaxes.

1075
00:45:12,570 --> 00:45:14,130
What does that mean?

1076
00:45:14,130 --> 00:45:15,605
It's not taking it easy.

1077
00:45:15,605 --> 00:45:17,336
AUDIENCE: You go
back down a couple.

1078
00:45:17,336 --> 00:45:18,890
JEFFREY C. GROSSMAN: To where?

1079
00:45:18,890 --> 00:45:19,557
AUDIENCE: There.

1080
00:45:19,557 --> 00:45:20,640
JEFFREY C. GROSSMAN: Here?

1081
00:45:20,640 --> 00:45:21,320
AUDIENCE: Yeah.

1082
00:45:21,320 --> 00:45:22,310
JEFFREY C. GROSSMAN: Right here?

1083
00:45:22,310 --> 00:45:23,030
AUDIENCE: Yeah.

1084
00:45:23,030 --> 00:45:23,840
JEFFREY C. GROSSMAN:
Can it go here?

1085
00:45:23,840 --> 00:45:25,520
AUDIENCE: [INAUDIBLE]
the one below it.

1086
00:45:25,520 --> 00:45:26,603
JEFFREY C. GROSSMAN: Here.

1087
00:45:26,603 --> 00:45:29,070
It can't go in
between, we know that.

1088
00:45:29,070 --> 00:45:32,390
So it goes here, and that
is the valence band--

1089
00:45:32,390 --> 00:45:38,960
no, Conduction Band Minimum,
CBM, of the material.

1090
00:45:38,960 --> 00:45:41,720
That's where it lands.

1091
00:45:41,720 --> 00:45:44,120
And so just like
in the solar field,

1092
00:45:44,120 --> 00:45:47,300
what I wanted you to see in that
third problem of the solar fuel

1093
00:45:47,300 --> 00:45:54,290
homework is that you
can excite electrons up,

1094
00:45:54,290 --> 00:45:57,500
but you're only going
to get out delta H.

1095
00:45:57,500 --> 00:45:59,600
So you lose whatever energy.

1096
00:45:59,600 --> 00:46:03,230
When you hold something
out in the sun,

1097
00:46:03,230 --> 00:46:06,920
photons of all energies from the
sun are hitting the something.

1098
00:46:06,920 --> 00:46:07,940
Right?

1099
00:46:07,940 --> 00:46:14,370
And most of them are going
to just heat, wasted heat.

1100
00:46:14,370 --> 00:46:17,790
Most of that sun's
energy is wasted heat,

1101
00:46:17,790 --> 00:46:20,930
and that's because
of this right here.

1102
00:46:20,930 --> 00:46:22,870
Right?

1103
00:46:22,870 --> 00:46:25,030
So this is just heat here.

1104
00:46:25,030 --> 00:46:31,150
It's just heating the
material, and then this

1105
00:46:31,150 --> 00:46:33,010
is the energy, or
the voltage if you

1106
00:46:33,010 --> 00:46:35,440
want, that you can
pull the electron out,

1107
00:46:35,440 --> 00:46:36,820
that conduction band.

1108
00:46:36,820 --> 00:46:40,940
All the other energy that
you might have had is lost.

1109
00:46:40,940 --> 00:46:42,330
Right?

1110
00:46:42,330 --> 00:46:44,760
And so why don't I just make
this really, really big,

1111
00:46:44,760 --> 00:46:45,300
the gap?

1112
00:46:48,518 --> 00:46:49,577
AUDIENCE: [INAUDIBLE].

1113
00:46:49,577 --> 00:46:50,660
JEFFREY C. GROSSMAN: Yeah.

1114
00:46:50,660 --> 00:46:53,930
Because if it's too big,
if the gap is too big, then

1115
00:46:53,930 --> 00:46:57,900
anything less than the
gap doesn't get absorbed.

1116
00:46:57,900 --> 00:46:59,460
If I make it too
small, then boy,

1117
00:46:59,460 --> 00:47:01,080
do I waste a lot of
the sun's energy.

1118
00:47:01,080 --> 00:47:01,650
Right?

1119
00:47:01,650 --> 00:47:03,095
Then, I have almost no voltage.

1120
00:47:03,095 --> 00:47:05,220
There's a sweet spot, and
that's what problem three

1121
00:47:05,220 --> 00:47:07,850
was about in PSET 5.

1122
00:47:07,850 --> 00:47:08,350
Right?

1123
00:47:11,050 --> 00:47:13,030
Now, then the electron
has to get out.

1124
00:47:15,608 --> 00:47:17,150
If it's a certain
kind of solar cell,

1125
00:47:17,150 --> 00:47:20,090
it can actually recombine
with the hole, which we really

1126
00:47:20,090 --> 00:47:20,810
don't want.

1127
00:47:20,810 --> 00:47:24,860
See, in some materials,
like in Silicon,

1128
00:47:24,860 --> 00:47:28,070
the electron and the hole really
don't like each other much.

1129
00:47:28,070 --> 00:47:31,910
They're Facebook
friends, or I don't know.

1130
00:47:31,910 --> 00:47:33,500
What do you kids
call it these days?

1131
00:47:33,500 --> 00:47:34,220
They tweet.

1132
00:47:34,220 --> 00:47:36,620
No, no, because nobody tweets.

1133
00:47:36,620 --> 00:47:38,580
Who tweets?

1134
00:47:38,580 --> 00:47:41,330
Does anybody tweet?

1135
00:47:41,330 --> 00:47:42,980
I'm amazed by that.

1136
00:47:42,980 --> 00:47:45,110
I have no clue what
I'm talking about.

1137
00:47:45,110 --> 00:47:47,600
How many of you have
a Facebook page?

1138
00:47:47,600 --> 00:47:49,840
Right.

1139
00:47:49,840 --> 00:47:50,500
OK.

1140
00:47:50,500 --> 00:47:51,380
I'm just learning.

1141
00:47:51,380 --> 00:47:52,960
It's a learning thing for me.

1142
00:47:52,960 --> 00:47:54,160
Google Plus?

1143
00:47:54,160 --> 00:47:56,050
Google Plus?

1144
00:47:56,050 --> 00:47:56,920
It's gotten better?

1145
00:47:56,920 --> 00:47:58,540
Has it gotten better?

1146
00:47:58,540 --> 00:47:59,600
Sort of.

1147
00:47:59,600 --> 00:48:00,810
Yeah.

1148
00:48:00,810 --> 00:48:01,770
OK.

1149
00:48:01,770 --> 00:48:07,100
So fact of the matter
is, where was I?

1150
00:48:07,100 --> 00:48:12,533
Anyway, they don't
like each other much.

1151
00:48:12,533 --> 00:48:13,950
They like each
other a little bit,

1152
00:48:13,950 --> 00:48:18,360
like a millielectron
volt or two of binding

1153
00:48:18,360 --> 00:48:20,430
between that electron
and the hole in silicon.

1154
00:48:20,430 --> 00:48:25,318
So basically, as soon as that
electron and hole form, and one

1155
00:48:25,318 --> 00:48:27,110
is in the connection
band, and the other is

1156
00:48:27,110 --> 00:48:30,110
in the valence band,
they're like, eh, whatever,

1157
00:48:30,110 --> 00:48:31,980
and they just wander off.

1158
00:48:31,980 --> 00:48:32,780
OK?

1159
00:48:32,780 --> 00:48:35,330
Now, that's a
weakly bound exciton

1160
00:48:35,330 --> 00:48:37,720
is another way of saying
that, or really not really

1161
00:48:37,720 --> 00:48:39,470
much of an exciton at
all, because there's

1162
00:48:39,470 --> 00:48:40,670
exciton binding.

1163
00:48:40,670 --> 00:48:42,500
An exciton is nothing
more than a fancy way

1164
00:48:42,500 --> 00:48:46,280
of saying electron hole
pair, but in some materials,

1165
00:48:46,280 --> 00:48:49,840
like polymer solar cells,
which I will talk about,

1166
00:48:49,840 --> 00:48:51,460
that's not true.

1167
00:48:51,460 --> 00:48:55,450
You see, in some materials,
the electron and the hole

1168
00:48:55,450 --> 00:48:57,760
are very strongly
attracted to each other.

1169
00:48:57,760 --> 00:48:59,420
They are BFF.

1170
00:48:59,420 --> 00:49:01,230
Did I say that right?

1171
00:49:01,230 --> 00:49:03,630
Best friends forever,
except it's not forever,

1172
00:49:03,630 --> 00:49:06,580
because you get them
apart in a solar cell.

1173
00:49:06,580 --> 00:49:08,370
You do get them
apart, and it's sad.

1174
00:49:08,370 --> 00:49:11,470
But then they go, and they do
work, and then they come back.

1175
00:49:11,470 --> 00:49:15,180
So it's a happy ending.

1176
00:49:15,180 --> 00:49:19,820
So now, when it's really
weak, like in Silicon, well,

1177
00:49:19,820 --> 00:49:23,740
you don't need much
to pull them away.

1178
00:49:23,740 --> 00:49:24,240
Right?

1179
00:49:24,240 --> 00:49:25,670
You've got to pull them away.

1180
00:49:25,670 --> 00:49:28,640
You got to transport the holes
and transport the electrons

1181
00:49:28,640 --> 00:49:29,960
over to some metals.

1182
00:49:29,960 --> 00:49:32,300
And so actually, how
do you pull them away?

1183
00:49:32,300 --> 00:49:36,376
How do you get them out on the
right side, different sides?

1184
00:49:36,376 --> 00:49:38,017
AUDIENCE: Electric
field [INAUDIBLE]

1185
00:49:38,017 --> 00:49:39,100
JEFFREY C. GROSSMAN: Yeah.

1186
00:49:39,100 --> 00:49:40,400
Yes, and yes.

1187
00:49:40,400 --> 00:49:41,620
Right?

1188
00:49:41,620 --> 00:49:44,590
A PN junction will do
it, and even there, there

1189
00:49:44,590 --> 00:49:47,290
can be whole regions where
there's no field really.

1190
00:49:47,290 --> 00:49:50,080
But as long as they finally get
to some region where there is,

1191
00:49:50,080 --> 00:49:55,370
and they can get off on
the right side, we're good.

1192
00:49:55,370 --> 00:49:57,950
OK?

1193
00:49:57,950 --> 00:50:00,080
But in other solar
cells, and this

1194
00:50:00,080 --> 00:50:01,520
is all within this picture.

1195
00:50:01,520 --> 00:50:03,170
This is all within
the same picture.

1196
00:50:03,170 --> 00:50:05,250
We can talk about
all these concepts.

1197
00:50:05,250 --> 00:50:08,330
So in a silicon solar
cell, recombination here

1198
00:50:08,330 --> 00:50:12,690
is really not a problem,
but this part is a problem.

1199
00:50:12,690 --> 00:50:16,830
That's really hard, because
the indirect band gap.

1200
00:50:16,830 --> 00:50:19,410
And then because you've
got to make it thick,

1201
00:50:19,410 --> 00:50:21,990
because that's bad,
this part is a problem,

1202
00:50:21,990 --> 00:50:24,840
because they got to transport
for a long distance.

1203
00:50:24,840 --> 00:50:27,090
And because they got to
transport for a long distance,

1204
00:50:27,090 --> 00:50:29,010
you got to make
the material pure.

1205
00:50:29,010 --> 00:50:31,745
So that it has good
highways to get out.

1206
00:50:31,745 --> 00:50:32,245
OK?

1207
00:50:35,110 --> 00:50:39,010
But in a plastic solar cell
or an organic solar cell,

1208
00:50:39,010 --> 00:50:41,890
well, these are
really strongly bound.

1209
00:50:41,890 --> 00:50:45,720
They really want
to be best friends.

1210
00:50:45,720 --> 00:50:47,730
There are half an
electron volt bound

1211
00:50:47,730 --> 00:50:49,800
instead of millielectron volts.

1212
00:50:49,800 --> 00:50:52,020
Those are strongly
bound excitons.

1213
00:50:52,020 --> 00:50:54,690
So a PN junction won't do it.

1214
00:50:54,690 --> 00:50:58,080
You can't do that, because
they're going to be together,

1215
00:50:58,080 --> 00:51:00,800
unless you rip them apart.

1216
00:51:00,800 --> 00:51:01,310
Right?

1217
00:51:01,310 --> 00:51:02,810
And because they're
going to be together,

1218
00:51:02,810 --> 00:51:05,420
what's going to happen is, if
you don't rip them apart soon,

1219
00:51:05,420 --> 00:51:08,317
they're just going to collapse,
and you're going to lose that.

1220
00:51:08,317 --> 00:51:10,400
All the photons energy,
you're just going to lose,

1221
00:51:10,400 --> 00:51:12,980
and the electron and
hole will recombine.

1222
00:51:12,980 --> 00:51:16,390
That's this recombination,
and they'll emit a photon.

1223
00:51:16,390 --> 00:51:16,890
Right?

1224
00:51:16,890 --> 00:51:18,060
That's bad.

1225
00:51:18,060 --> 00:51:19,860
That's a bad thing.

1226
00:51:19,860 --> 00:51:20,430
Right?

1227
00:51:20,430 --> 00:51:22,800
So in those kinds
of solar cells,

1228
00:51:22,800 --> 00:51:25,050
the challenges are different.

1229
00:51:25,050 --> 00:51:27,750
They're actually
really good at this.

1230
00:51:27,750 --> 00:51:30,232
Remember, I said 100
nanometers of thickness

1231
00:51:30,232 --> 00:51:32,190
is all you need to absorb
all the light in some

1232
00:51:32,190 --> 00:51:34,430
of these materials, even less.

1233
00:51:34,430 --> 00:51:35,360
OK?

1234
00:51:35,360 --> 00:51:37,790
But where there is a challenge
is how do you get them--

1235
00:51:37,790 --> 00:51:39,200
how do you pull them apart?

1236
00:51:39,200 --> 00:51:44,300
And the way you do it is
you create interfaces,

1237
00:51:44,300 --> 00:51:48,110
where when this electron and
hole pair get to the interface,

1238
00:51:48,110 --> 00:51:50,900
one of them really
sees the grass is

1239
00:51:50,900 --> 00:51:53,120
really greener on the other
side of that interface.

1240
00:51:53,120 --> 00:51:55,790
And the other one is like,
I don't like that material,

1241
00:51:55,790 --> 00:51:59,780
and then they split, and that's
called a type 2 heterojunction.

1242
00:51:59,780 --> 00:52:02,300
Which I'll show you a picture
in a sec, in a little bit,

1243
00:52:02,300 --> 00:52:03,570
or maybe Thursday morning.

1244
00:52:03,570 --> 00:52:05,360
Thursday at the
beginning of class,

1245
00:52:05,360 --> 00:52:08,270
I think I might have time.

1246
00:52:08,270 --> 00:52:14,120
So this is the level that we can
really understand solar cells

1247
00:52:14,120 --> 00:52:16,280
from the fundamental
electron viewpoint,

1248
00:52:16,280 --> 00:52:18,430
and this is the
level where you know

1249
00:52:18,430 --> 00:52:22,020
how to compute relevant things.

1250
00:52:22,020 --> 00:52:24,080
So tell me what you
can compute that's

1251
00:52:24,080 --> 00:52:25,150
relevant to this picture.

1252
00:52:27,838 --> 00:52:28,977
AUDIENCE: [INAUDIBLE]

1253
00:52:28,977 --> 00:52:30,060
JEFFREY C. GROSSMAN: Yeah.

1254
00:52:30,060 --> 00:52:30,780
AUDIENCE: Fermi energy.

1255
00:52:30,780 --> 00:52:31,740
JEFFREY C. GROSSMAN:
OK, Fermi energy.

1256
00:52:31,740 --> 00:52:32,240
OK.

1257
00:52:32,240 --> 00:52:33,700
So that would be for the metal.

1258
00:52:33,700 --> 00:52:34,560
Right?

1259
00:52:34,560 --> 00:52:38,190
That's the level that's going to
try to line up with that level,

1260
00:52:38,190 --> 00:52:41,460
and you'd like it
to be pretty close.

1261
00:52:41,460 --> 00:52:42,300
OK?

1262
00:52:42,300 --> 00:52:42,930
What else?

1263
00:52:42,930 --> 00:52:44,670
Somebody said band gap.

1264
00:52:44,670 --> 00:52:45,420
Very good.

1265
00:52:45,420 --> 00:52:47,740
That's this.

1266
00:52:47,740 --> 00:52:49,000
Right?

1267
00:52:49,000 --> 00:52:49,540
What else?

1268
00:52:52,430 --> 00:52:54,155
What about the band gap?

1269
00:52:54,155 --> 00:52:55,030
AUDIENCE: [INAUDIBLE]

1270
00:52:55,030 --> 00:52:59,420
JEFFREY C. GROSSMAN: OK, direct
or indirect, very important.

1271
00:52:59,420 --> 00:53:00,730
What else?

1272
00:53:00,730 --> 00:53:01,630
AUDIENCE: [INAUDIBLE]

1273
00:53:01,630 --> 00:53:03,790
JEFFREY C. GROSSMAN:
Mobility which is this part.

1274
00:53:03,790 --> 00:53:06,890
How efficiently is
it going to do this?

1275
00:53:06,890 --> 00:53:07,390
Right?

1276
00:53:07,390 --> 00:53:11,210
Well, you can calculate that
now from the band structure.

1277
00:53:11,210 --> 00:53:11,710
Right?

1278
00:53:11,710 --> 00:53:14,215
What gives you the mobility?

1279
00:53:14,215 --> 00:53:15,550
AUDIENCE: [INAUDIBLE]

1280
00:53:15,550 --> 00:53:17,770
JEFFREY C. GROSSMAN:
Curvature of the band.

1281
00:53:17,770 --> 00:53:18,790
OK?

1282
00:53:18,790 --> 00:53:22,100
You know how to do that.

1283
00:53:22,100 --> 00:53:23,777
And the last thing
is that I'll just

1284
00:53:23,777 --> 00:53:25,360
tell you because I'm
going to show you

1285
00:53:25,360 --> 00:53:29,473
in a second is that when
you need to pull electrons

1286
00:53:29,473 --> 00:53:32,140
and holes apart from each other,
because they're really strongly

1287
00:53:32,140 --> 00:53:35,290
bound, you need two different
materials with a very

1288
00:53:35,290 --> 00:53:38,290
particular way of lining
their energy levels up.

1289
00:53:38,290 --> 00:53:40,930
So again, it comes back to--

1290
00:53:40,930 --> 00:53:47,680
there, it comes back to having
one material with levels.

1291
00:53:47,680 --> 00:53:50,110
This is, let's say,
the HOMO in the LUMO,

1292
00:53:50,110 --> 00:53:55,360
like that, and another
material, or if you

1293
00:53:55,360 --> 00:53:58,360
want to think about this
is the conduction band

1294
00:53:58,360 --> 00:53:59,500
and the valence band.

1295
00:54:02,460 --> 00:54:04,770
You need to have--

1296
00:54:04,770 --> 00:54:08,230
if you have a really
strongly bound electron hole,

1297
00:54:08,230 --> 00:54:10,682
there it is.

1298
00:54:10,682 --> 00:54:11,640
You're with me on this.

1299
00:54:11,640 --> 00:54:11,820
Right?

1300
00:54:11,820 --> 00:54:12,920
You see what I mean here.

1301
00:54:12,920 --> 00:54:14,040
This is a level.

1302
00:54:14,040 --> 00:54:15,270
That was where it was.

1303
00:54:15,270 --> 00:54:19,350
It's thermalized down, and
this is strongly bound.

1304
00:54:19,350 --> 00:54:21,300
So I got to rip them
apart, and then that

1305
00:54:21,300 --> 00:54:23,670
means I need another
material with levels

1306
00:54:23,670 --> 00:54:26,580
that come in at a very
particular way with very

1307
00:54:26,580 --> 00:54:28,020
particular alignment.

1308
00:54:28,020 --> 00:54:32,900
And that's called the
type 2 heterojunction,

1309
00:54:32,900 --> 00:54:37,860
and if it's not type 2, you're
not going to split them apart.

1310
00:54:37,860 --> 00:54:38,640
OK?

1311
00:54:38,640 --> 00:54:41,320
Now, how can you calculate that?

1312
00:54:41,320 --> 00:54:43,800
Well, you can
calculate these levels.

1313
00:54:43,800 --> 00:54:45,690
This is what you know how to do.

1314
00:54:45,690 --> 00:54:49,320
So you can actually, with
this same kind of methodology,

1315
00:54:49,320 --> 00:54:52,890
you can calculate the
levels across an interface.

1316
00:54:52,890 --> 00:54:54,690
Exactly the same
with the SIESTA tool

1317
00:54:54,690 --> 00:54:57,510
you have, you can look
at what those levels look

1318
00:54:57,510 --> 00:54:59,260
like on one side
of an interface,

1319
00:54:59,260 --> 00:55:00,987
and when you change
the material, what

1320
00:55:00,987 --> 00:55:02,070
they look like on another.

1321
00:55:02,070 --> 00:55:04,560
And you can tell me if
it's a type 2 interface

1322
00:55:04,560 --> 00:55:08,340
or say a type 1 or a broken
junction or other kinds that

1323
00:55:08,340 --> 00:55:09,960
wouldn't be good for PV.

1324
00:55:09,960 --> 00:55:11,120
OK?

1325
00:55:11,120 --> 00:55:12,230
OK.

1326
00:55:12,230 --> 00:55:15,177
So this is a very rich
picture, and really, you

1327
00:55:15,177 --> 00:55:16,760
can see that there
are a lot of things

1328
00:55:16,760 --> 00:55:18,260
you know how to
compute now that are

1329
00:55:18,260 --> 00:55:19,640
directly relevant to solar PV.

1330
00:55:19,640 --> 00:55:21,650
So let's take a closer look.

1331
00:55:21,650 --> 00:55:22,150
OK.

1332
00:55:22,150 --> 00:55:26,450
So that's what we want
to do is take a closer

1333
00:55:26,450 --> 00:55:27,800
look at some of these.

1334
00:55:27,800 --> 00:55:31,428
Any questions?

1335
00:55:31,428 --> 00:55:36,087
AUDIENCE: What is
the [INAUDIBLE]

1336
00:55:36,087 --> 00:55:38,420
JEFFREY C. GROSSMAN: That's
actually the conduction band

1337
00:55:38,420 --> 00:55:42,200
minimum and the
valence band maximum,

1338
00:55:42,200 --> 00:55:46,820
and because sometimes
the conduction band

1339
00:55:46,820 --> 00:55:48,830
can be thought of
as a band of states.

1340
00:55:51,980 --> 00:55:54,830
But often, we only care
about the very lowest

1341
00:55:54,830 --> 00:56:00,378
one, which is where the
electron is going to wind up.

1342
00:56:00,378 --> 00:56:01,920
It's got plenty of
time to get there.

1343
00:56:01,920 --> 00:56:07,160
This thermalization process
is super fast, super fast.

1344
00:56:07,160 --> 00:56:07,660
OK?

1345
00:56:11,010 --> 00:56:11,510
OK.

1346
00:56:11,510 --> 00:56:13,427
So this is again what
I've been talking about,

1347
00:56:13,427 --> 00:56:15,960
but I'll give you a few
more specific examples.

1348
00:56:15,960 --> 00:56:17,810
So in crystalline
silicon which is still--

1349
00:56:17,810 --> 00:56:19,040
I don't know if it's 80%.

1350
00:56:19,040 --> 00:56:20,390
Maybe it's between 70% and 80%.

1351
00:56:24,160 --> 00:56:24,700
Right?

1352
00:56:24,700 --> 00:56:28,780
The band structure, the band
gap, and the electron hole

1353
00:56:28,780 --> 00:56:31,900
mobilities, this is all
something we just talked about.

1354
00:56:31,900 --> 00:56:32,770
OK?

1355
00:56:32,770 --> 00:56:34,993
And it all comes from
the band structure.

1356
00:56:34,993 --> 00:56:36,910
So you get the band
structure of the material,

1357
00:56:36,910 --> 00:56:39,100
and these properties
of the picture

1358
00:56:39,100 --> 00:56:41,090
we just showed fall out.

1359
00:56:41,090 --> 00:56:41,590
OK?

1360
00:56:44,900 --> 00:56:48,140
In amorphous silicon,
see amorphous silicon

1361
00:56:48,140 --> 00:56:55,820
is actually 3% of the market,
and actually light absorption

1362
00:56:55,820 --> 00:57:00,200
is really pretty good, and so
you can make them very thin.

1363
00:57:00,200 --> 00:57:04,310
That picture was--
where was it, there--

1364
00:57:04,310 --> 00:57:05,480
that's crystalline silicon.

1365
00:57:05,480 --> 00:57:06,897
That's a crystalline
silicon cell,

1366
00:57:06,897 --> 00:57:09,080
and that's an
amorphous silicon cell.

1367
00:57:09,080 --> 00:57:12,470
And you can see one is curving,
because you can make it really,

1368
00:57:12,470 --> 00:57:13,700
really thin.

1369
00:57:13,700 --> 00:57:14,510
OK?

1370
00:57:14,510 --> 00:57:17,090
Whereas, if we tried
to curve 100 microns,

1371
00:57:17,090 --> 00:57:18,388
it's really hard to do.

1372
00:57:18,388 --> 00:57:19,430
You're going to crack it.

1373
00:57:19,430 --> 00:57:19,930
Right?

1374
00:57:23,100 --> 00:57:24,930
The problem with
amorphous silicon,

1375
00:57:24,930 --> 00:57:26,970
so electrons and
holes also separate.

1376
00:57:26,970 --> 00:57:28,500
That's not a problem.

1377
00:57:28,500 --> 00:57:30,060
They separate easily.

1378
00:57:30,060 --> 00:57:32,250
They're not strongly
bound to each other,

1379
00:57:32,250 --> 00:57:35,010
but here, you have the
transport is the problem,

1380
00:57:35,010 --> 00:57:36,750
holes in particular.

1381
00:57:36,750 --> 00:57:40,500
Positive charges move
very, very slowly

1382
00:57:40,500 --> 00:57:42,160
through amorphous silicon.

1383
00:57:42,160 --> 00:57:43,320
OK?

1384
00:57:43,320 --> 00:57:47,250
So they're actually,
both electrons and holes,

1385
00:57:47,250 --> 00:57:49,710
are a lot slower in
amorphous silicon

1386
00:57:49,710 --> 00:57:52,560
than in crystalline silicon,
but holes are 200 times

1387
00:57:52,560 --> 00:57:56,170
slower than electrons,
200 times slower.

1388
00:57:56,170 --> 00:57:56,670
Right?

1389
00:57:56,670 --> 00:57:58,337
Whereas, in crystalline
silicon, they're

1390
00:57:58,337 --> 00:57:59,920
only like two or
three times slower.

1391
00:57:59,920 --> 00:58:03,690
So that's a big discrepancy,
and that causes big problems

1392
00:58:03,690 --> 00:58:07,470
and a big drop in efficiency.

1393
00:58:07,470 --> 00:58:11,180
Now, what you can do with the
calculations you know how to do

1394
00:58:11,180 --> 00:58:13,250
is you can calculate
the hole mobility,

1395
00:58:13,250 --> 00:58:15,530
and actually, you
can also calculate

1396
00:58:15,530 --> 00:58:19,760
whether the amorphous material
wants to trap a positive charge

1397
00:58:19,760 --> 00:58:21,630
or not.

1398
00:58:21,630 --> 00:58:24,950
And the way you do that is
you simply, in the code,

1399
00:58:24,950 --> 00:58:28,330
you simply calculate
the structure

1400
00:58:28,330 --> 00:58:30,230
with and without a
positive charge in it,

1401
00:58:30,230 --> 00:58:31,610
and you get an energy.

1402
00:58:31,610 --> 00:58:32,920
Right?

1403
00:58:32,920 --> 00:58:35,420
And then you can compare that
from one structure to another,

1404
00:58:35,420 --> 00:58:39,260
and try to understand something
about why this material is

1405
00:58:39,260 --> 00:58:40,370
trapping holes.

1406
00:58:40,370 --> 00:58:42,800
You see, in an amorphous
material-- do I have a picture?

1407
00:58:42,800 --> 00:58:46,040
I think I had a picture of--

1408
00:58:46,040 --> 00:58:49,047
maybe I have a picture later.

1409
00:58:49,047 --> 00:58:50,630
Why would we want
an amorphous silicon

1410
00:58:50,630 --> 00:58:53,310
over a crystalline silicon?

1411
00:58:53,310 --> 00:58:53,950
It's cheaper.

1412
00:58:53,950 --> 00:58:55,428
Why is it cheaper?

1413
00:58:55,428 --> 00:58:56,670
AUDIENCE: [INAUDIBLE]

1414
00:58:56,670 --> 00:58:57,774
JEFFREY C. GROSSMAN: Why?

1415
00:58:57,774 --> 00:59:00,356
AUDIENCE: [INAUDIBLE]

1416
00:59:01,877 --> 00:59:02,960
JEFFREY C. GROSSMAN: Yeah.

1417
00:59:02,960 --> 00:59:05,680
It's just a mess.

1418
00:59:05,680 --> 00:59:08,580
Amorphous silicon is a mess.

1419
00:59:08,580 --> 00:59:09,630
Right?

1420
00:59:09,630 --> 00:59:10,920
It's a messy material.

1421
00:59:10,920 --> 00:59:13,427
Now, there are still
very important ways

1422
00:59:13,427 --> 00:59:15,510
of doing the processing
that can make a difference

1423
00:59:15,510 --> 00:59:16,570
on the performance.

1424
00:59:16,570 --> 00:59:17,070
Right?

1425
00:59:17,070 --> 00:59:19,440
Hydrogen concentration,
how you anneal it,

1426
00:59:19,440 --> 00:59:21,720
but it's a much easier
material to make,

1427
00:59:21,720 --> 00:59:23,100
since it's inherently messy.

1428
00:59:23,100 --> 00:59:25,380
Now, that poses
a real challenge,

1429
00:59:25,380 --> 00:59:29,280
because now you have a
problem that you want to fix,

1430
00:59:29,280 --> 00:59:32,430
the transport of charges
in a messy material,

1431
00:59:32,430 --> 00:59:35,730
and it's really hard to
characterize messy materials.

1432
00:59:35,730 --> 00:59:37,270
Right?

1433
00:59:37,270 --> 00:59:41,540
Beautiful problem for
computation, beautiful problem.

1434
00:59:41,540 --> 00:59:44,000
And we're talking about
holes and electrons,

1435
00:59:44,000 --> 00:59:46,392
so the only game in town
is quantum mechanics.

1436
00:59:46,392 --> 00:59:47,600
That's the only game in town.

1437
00:59:47,600 --> 00:59:48,140
Right?

1438
00:59:48,140 --> 00:59:51,800
You got to get to
the description

1439
00:59:51,800 --> 00:59:55,010
that we know how to do now of
where electrons and holes are

1440
00:59:55,010 --> 00:59:55,820
in the material.

1441
00:59:55,820 --> 00:59:58,850
And in the computer, you
can make messy this way,

1442
00:59:58,850 --> 01:00:03,170
messy that way, add a little
of this, put a defect in,

1443
01:00:03,170 --> 01:00:07,310
and you can calculate what
that does all the way down

1444
01:00:07,310 --> 01:00:09,920
to the scales of these
atoms and electrons.

1445
01:00:09,920 --> 01:00:11,970
And that would be
extremely hard,

1446
01:00:11,970 --> 01:00:14,720
especially in a material
that's disordered,

1447
01:00:14,720 --> 01:00:15,950
to do experimentally.

1448
01:00:15,950 --> 01:00:18,740
So this is a great
problem for computation.

1449
01:00:18,740 --> 01:00:19,700
I love this problem.

1450
01:00:19,700 --> 01:00:24,170
Amorphous silicon has a
wonderful problem called

1451
01:00:24,170 --> 01:00:25,820
the Stabler-Wronski effect.

1452
01:00:25,820 --> 01:00:30,500
Stabler-Wronski in
1977, they wrote an APL,

1453
01:00:30,500 --> 01:00:32,750
where they showed that you
make an amorphous silicon

1454
01:00:32,750 --> 01:00:34,130
solar cell.

1455
01:00:34,130 --> 01:00:37,070
And let's say you make it,
and you test it, and it's 10%.

1456
01:00:37,070 --> 01:00:40,520
And now you hold it out
under the sun for two hours,

1457
01:00:40,520 --> 01:00:42,920
and after that, it's 7%.

1458
01:00:42,920 --> 01:00:43,455
OK?

1459
01:00:43,455 --> 01:00:46,080
And it doesn't come back, unless
you heat it up, in which case,

1460
01:00:46,080 --> 01:00:47,340
it does.

1461
01:00:47,340 --> 01:00:48,510
What a great problem.

1462
01:00:48,510 --> 01:00:49,010
Right?

1463
01:00:49,010 --> 01:00:53,810
So you get an immediate 30%
hit on your performance, when

1464
01:00:53,810 --> 01:00:56,660
you have an amorphous solar
cell, amorphous silicon

1465
01:00:56,660 --> 01:01:03,110
solar cell, and that is a
30, 40 year problem, 30 year

1466
01:01:03,110 --> 01:01:07,217
problem that is unsolved.

1467
01:01:07,217 --> 01:01:08,300
I love problems like that.

1468
01:01:08,300 --> 01:01:09,273
Why?

1469
01:01:09,273 --> 01:01:11,690
You hold it out under the sun,
and because of the sunlight

1470
01:01:11,690 --> 01:01:13,670
exposure, something's--

1471
01:01:13,670 --> 01:01:17,390
the mess is getting
messy in a bad way.

1472
01:01:17,390 --> 01:01:20,250
Not all messes
are created equal.

1473
01:01:20,250 --> 01:01:20,750
Right?

1474
01:01:20,750 --> 01:01:21,590
How cool is that?

1475
01:01:21,590 --> 01:01:24,050
At the atomic
scale, the disorder

1476
01:01:24,050 --> 01:01:27,380
is different, when
you shine light on it,

1477
01:01:27,380 --> 01:01:29,600
and that difference,
the different disorder,

1478
01:01:29,600 --> 01:01:33,050
means something really
important to positive charges.

1479
01:01:33,050 --> 01:01:35,030
What?

1480
01:01:35,030 --> 01:01:38,240
Unsolved, very cool
problem, computation

1481
01:01:38,240 --> 01:01:40,470
can play a big role.

1482
01:01:40,470 --> 01:01:40,980
OK.

1483
01:01:40,980 --> 01:01:45,840
Now, in organic PV, as I already
mentioned, I mentioned this.

1484
01:01:45,840 --> 01:01:48,015
So let me get to that,
but look at this.

1485
01:01:48,015 --> 01:01:50,070
This is that solar
spectrum, and often we

1486
01:01:50,070 --> 01:01:54,360
like to plot just several ones.

1487
01:01:58,113 --> 01:02:00,530
So but these are just the sun,
whether you're on the Earth

1488
01:02:00,530 --> 01:02:04,050
or how you measure it,
but now here's the key.

1489
01:02:04,050 --> 01:02:07,550
This is the absorption
of a polymer

1490
01:02:07,550 --> 01:02:09,350
that can make an organic cell.

1491
01:02:09,350 --> 01:02:14,570
Now, you can see that one
issue with organic materials

1492
01:02:14,570 --> 01:02:17,450
is they can have really good
absorption but in a fairly

1493
01:02:17,450 --> 01:02:19,610
narrow range.

1494
01:02:19,610 --> 01:02:23,770
So I can make this
cell very thin,

1495
01:02:23,770 --> 01:02:28,750
and theoretically, I could get
a decent efficiency out of it.

1496
01:02:28,750 --> 01:02:32,080
I don't by the way, but
one would think I could.

1497
01:02:32,080 --> 01:02:34,930
But you can see that
I'm missing a whole lot

1498
01:02:34,930 --> 01:02:36,640
of the solar spectrum.

1499
01:02:36,640 --> 01:02:37,140
Right?

1500
01:02:37,140 --> 01:02:40,190
I'm missing a whole lot
of the solar spectrum,

1501
01:02:40,190 --> 01:02:42,400
and so that's one
issue with polymers.

1502
01:02:42,400 --> 01:02:44,650
Is they're very
good at absorbing,

1503
01:02:44,650 --> 01:02:47,080
very efficient at
absorbing light,

1504
01:02:47,080 --> 01:02:49,390
but usually in a fairly
narrow part of the spectrum.

1505
01:02:49,390 --> 01:02:51,637
So that's one challenge,
and that's something

1506
01:02:51,637 --> 01:02:54,220
that you now know a lot about,
because you've been integrating

1507
01:02:54,220 --> 01:02:57,720
this from like here back.

1508
01:02:57,720 --> 01:02:58,830
That was fun.

1509
01:02:58,830 --> 01:03:00,340
Wasn't that fun?

1510
01:03:00,340 --> 01:03:02,070
Yeah.

1511
01:03:02,070 --> 01:03:03,950
I felt the pain.

1512
01:03:03,950 --> 01:03:06,900
Sam, help me feel the pain.

1513
01:03:06,900 --> 01:03:08,790
Inform me about the pain.

1514
01:03:08,790 --> 01:03:10,740
OK.

1515
01:03:10,740 --> 01:03:14,490
Now, so that's one issue,
and you know how to do that.

1516
01:03:14,490 --> 01:03:18,810
You can take a polymer, put
it in your nanoHUB tool,

1517
01:03:18,810 --> 01:03:21,550
and integrate the
sun from its gap.

1518
01:03:21,550 --> 01:03:22,200
Right?

1519
01:03:22,200 --> 01:03:24,990
And that'll give you this
now, and you can even

1520
01:03:24,990 --> 01:03:26,940
do it more fancy than that.

1521
01:03:26,940 --> 01:03:29,310
You can even use the
density of states

1522
01:03:29,310 --> 01:03:32,400
to get a more realistic
estimate of how

1523
01:03:32,400 --> 01:03:37,620
much energy versus wavelength
that that polymer can absorb.

1524
01:03:37,620 --> 01:03:39,960
But another really
important problem

1525
01:03:39,960 --> 01:03:43,960
that comes with polymers
is that, again--

1526
01:03:43,960 --> 01:03:46,380
and this is exactly that
picture I just showed--

1527
01:03:46,380 --> 01:03:48,570
you have to have two materials.

1528
01:03:48,570 --> 01:03:54,210
Because in a polymer, when you
excite an electron and a hole,

1529
01:03:54,210 --> 01:03:56,380
they're usually strongly bound.

1530
01:03:56,380 --> 01:03:56,880
Right?

1531
01:03:56,880 --> 01:04:00,060
So that exciton is a strong
bound exciton, and therefore,

1532
01:04:00,060 --> 01:04:02,850
you can't split them
apart, unless you

1533
01:04:02,850 --> 01:04:06,250
have an interface that has
levels that align like this.

1534
01:04:06,250 --> 01:04:09,150
So another really important
part of polymer PV

1535
01:04:09,150 --> 01:04:13,860
is to design the two
kinds of materials

1536
01:04:13,860 --> 01:04:15,520
that will give you
those interfaces

1537
01:04:15,520 --> 01:04:18,180
and then control them in
some way to blend them.

1538
01:04:18,180 --> 01:04:21,502
It's actually a really
pretty hard problem.

1539
01:04:21,502 --> 01:04:23,460
There's a company, there
are several companies,

1540
01:04:23,460 --> 01:04:31,910
selling P3HT that is
poly-3-hexylthiophene solar

1541
01:04:31,910 --> 01:04:33,260
cells.

1542
01:04:33,260 --> 01:04:35,570
Konarka, has anybody
heard of Konarka?

1543
01:04:35,570 --> 01:04:38,230
You can buy them, Konarka.

1544
01:04:41,960 --> 01:04:45,590
A brave, bold move into
polymer solar cells

1545
01:04:45,590 --> 01:04:49,790
was made with $120
million by Konarka.

1546
01:04:49,790 --> 01:04:54,720
I would say too early, but
they're trying to push it.

1547
01:04:54,720 --> 01:04:58,340
Now, the problem is
efficiencies are very low still.

1548
01:04:58,340 --> 01:05:01,280
So you have these like they'll
give you one of their products.

1549
01:05:01,280 --> 01:05:04,340
It's like this size,
and it's all packed in.

1550
01:05:04,340 --> 01:05:06,590
And you can unfold it to
the size of this table,

1551
01:05:06,590 --> 01:05:08,630
because it's just
flexible and thin,

1552
01:05:08,630 --> 01:05:11,280
and you put it on
plastic instead of glass.

1553
01:05:11,280 --> 01:05:11,780
Right?

1554
01:05:11,780 --> 01:05:13,610
P3HD is a plastic.

1555
01:05:13,610 --> 01:05:17,030
It's a cheap, cheap solution
[? possible ?] plastic,

1556
01:05:17,030 --> 01:05:19,760
and the films only need to
be 100 nanometers thick.

1557
01:05:19,760 --> 01:05:22,010
So you can put it on this
flexible thing that folds up

1558
01:05:22,010 --> 01:05:23,360
into a wallet,
and you unfold it,

1559
01:05:23,360 --> 01:05:24,652
and it's the size of the table.

1560
01:05:24,652 --> 01:05:26,630
And you plug it into
your cell phone,

1561
01:05:26,630 --> 01:05:30,590
and you wait the
whole day, and you get

1562
01:05:30,590 --> 01:05:32,600
like two minutes of talk time.

1563
01:05:32,600 --> 01:05:33,980
They're very low efficiency.

1564
01:05:33,980 --> 01:05:35,913
OK?

1565
01:05:35,913 --> 01:05:37,580
But they're getting
better, and actually

1566
01:05:37,580 --> 01:05:41,360
the research efficiencies
of polymer cells

1567
01:05:41,360 --> 01:05:45,330
are now at 10%,
almost 10%, 9%, 10%.

1568
01:05:45,330 --> 01:05:46,880
That's pretty exciting.

1569
01:05:46,880 --> 01:05:50,640
Remember, research efficiencies,
what you buy is very different.

1570
01:05:50,640 --> 01:05:53,843
It's still very
low, 2%, maybe 3%.

1571
01:05:53,843 --> 01:05:56,135
That's not what they say,
but it's what people measure.

1572
01:05:59,042 --> 01:06:00,500
There's another
issue with polymers

1573
01:06:00,500 --> 01:06:01,625
though, with polymer cells.

1574
01:06:01,625 --> 01:06:05,660
Can anybody tell me what
they think it might be?

1575
01:06:05,660 --> 01:06:07,170
What is that?

1576
01:06:07,170 --> 01:06:09,930
Degrades, why?

1577
01:06:09,930 --> 01:06:14,090
Two reasons, why do
organic things degrade?

1578
01:06:14,090 --> 01:06:15,650
AUDIENCE: You
start pulling apart

1579
01:06:15,650 --> 01:06:18,600
[INAUDIBLE] from structure
[INAUDIBLE] at certain points,

1580
01:06:18,600 --> 01:06:19,895
it's going to become unusable.

1581
01:06:19,895 --> 01:06:21,895
JEFFREY C. GROSSMAN: Well,
that's actually not--

1582
01:06:21,895 --> 01:06:25,370
you get them back, so that's
actually not the problem.

1583
01:06:25,370 --> 01:06:26,620
It's not that you're pulling--

1584
01:06:26,620 --> 01:06:29,140
It's not that you're
depleting them or anything.

1585
01:06:29,140 --> 01:06:30,936
It's a different problem.

1586
01:06:30,936 --> 01:06:33,915
AUDIENCE: [INAUDIBLE]

1587
01:06:36,215 --> 01:06:37,840
JEFFREY C. GROSSMAN:
That is absolutely

1588
01:06:37,840 --> 01:06:42,340
the right answer for
3012, but in 3021-- well,

1589
01:06:42,340 --> 01:06:46,460
it's still-- yes, I can't ever
say no to the second law, ever.

1590
01:06:46,460 --> 01:06:46,960
OK?

1591
01:06:46,960 --> 01:06:50,410
So yes, but what else?

1592
01:06:50,410 --> 01:06:52,330
What happens to polymer?

1593
01:06:52,330 --> 01:06:55,300
There's two badnesses.

1594
01:06:55,300 --> 01:06:56,170
AUDIENCE: Oxidize?

1595
01:06:56,170 --> 01:06:57,170
JEFFREY C. GROSSMAN: OK.

1596
01:06:57,170 --> 01:07:00,310
So oxygen is a tough one,
water, oxygen. Right?

1597
01:07:00,310 --> 01:07:02,170
Chemically they degrade.

1598
01:07:02,170 --> 01:07:03,740
Now, how can I prevent that?

1599
01:07:03,740 --> 01:07:05,580
AUDIENCE: [INAUDIBLE]

1600
01:07:05,580 --> 01:07:07,440
JEFFREY C. GROSSMAN:
Packaging, you got it,

1601
01:07:07,440 --> 01:07:09,240
and what does packaging say?

1602
01:07:09,240 --> 01:07:11,310
What is really, really,
really good packaging

1603
01:07:11,310 --> 01:07:13,200
to really, really make
sure oxygen and water

1604
01:07:13,200 --> 01:07:14,790
doesn't get in?

1605
01:07:14,790 --> 01:07:19,520
That should ring this
in your mind, dollars.

1606
01:07:19,520 --> 01:07:21,980
That's expensive, and
that's the problem,

1607
01:07:21,980 --> 01:07:23,602
is now you're going back.

1608
01:07:23,602 --> 01:07:25,310
Well, you had solution
[? processable, ?]

1609
01:07:25,310 --> 01:07:26,923
cheap materials,
flexible substrates.

1610
01:07:26,923 --> 01:07:28,340
You're hitting it
out of the part,

1611
01:07:28,340 --> 01:07:31,270
and then you've got to spend
all your money on packaging.

1612
01:07:31,270 --> 01:07:32,890
Right?

1613
01:07:32,890 --> 01:07:36,780
And there's another reason
why polymers degrade.

1614
01:07:36,780 --> 01:07:39,342
Anybody know?

1615
01:07:39,342 --> 01:07:40,813
AUDIENCE: [INAUDIBLE]

1616
01:07:40,813 --> 01:07:42,230
JEFFREY C. GROSSMAN:
That could be

1617
01:07:42,230 --> 01:07:45,710
where the degradation
happens, but there's

1618
01:07:45,710 --> 01:07:51,230
something else that happens to
polymers, when you put them out

1619
01:07:51,230 --> 01:07:52,332
in the sun.

1620
01:07:52,332 --> 01:07:53,750
AUDIENCE: [INAUDIBLE]

1621
01:07:53,750 --> 01:07:56,100
JEFFREY C. GROSSMAN: Why
did car paint used to fade?

1622
01:07:56,100 --> 01:07:56,990
It doesn't anymore.

1623
01:07:56,990 --> 01:07:59,450
You guys are too young.

1624
01:07:59,450 --> 01:08:05,660
Back when I was in
school, in the 1930s,

1625
01:08:05,660 --> 01:08:09,560
car paint used to fade.

1626
01:08:09,560 --> 01:08:11,832
Why?

1627
01:08:11,832 --> 01:08:13,270
AUDIENCE: UV radiation?

1628
01:08:13,270 --> 01:08:14,940
JEFFREY C. GROSSMAN: It's UV.

1629
01:08:14,940 --> 01:08:17,960
UV, right?

1630
01:08:17,960 --> 01:08:19,920
UV is tough on us.

1631
01:08:19,920 --> 01:08:20,420
Right?

1632
01:08:20,420 --> 01:08:23,120
And UV is tough
on organic matter,

1633
01:08:23,120 --> 01:08:28,920
and so UV will over time
degrade polymers as well.

1634
01:08:28,920 --> 01:08:32,180
Now, there's still lots of
discussion and debate, how much

1635
01:08:32,180 --> 01:08:36,810
is one, moisture, oxygen,
versus UV important here,

1636
01:08:36,810 --> 01:08:38,029
but both are degrading this.

1637
01:08:40,758 --> 01:08:42,800
Now, when you install a
solar cell in your house,

1638
01:08:42,800 --> 01:08:44,830
what kind of
warranty do you get?

1639
01:08:44,830 --> 01:08:45,710
AUDIENCE: 20 years.

1640
01:08:45,710 --> 01:08:48,640
JEFFREY C. GROSSMAN:
20 years, you got it.

1641
01:08:48,640 --> 01:08:49,580
Right?

1642
01:08:49,580 --> 01:08:52,939
And if it's a
Cad-Tel solar cell,

1643
01:08:52,939 --> 01:08:55,430
they'll offer to buy it back.

1644
01:08:55,430 --> 01:08:57,890
No, they'll offer to safely
dispose of the cadmium

1645
01:08:57,890 --> 01:08:59,160
after 20 years.

1646
01:08:59,160 --> 01:08:59,660
Yeah.

1647
01:08:59,660 --> 01:09:02,590
I'll believe that when I see it.

1648
01:09:02,590 --> 01:09:05,359
That's a very expensive
business plan,

1649
01:09:05,359 --> 01:09:09,069
but that's the kind of
warranty you get, 20 years.

1650
01:09:09,069 --> 01:09:11,770
And if you got stuff that's
degrading in six months, which

1651
01:09:11,770 --> 01:09:14,399
by the way you do, it's
just very hard to compete.

1652
01:09:14,399 --> 01:09:15,189
OK?

1653
01:09:15,189 --> 01:09:17,410
So those are some key
issues degradation.

1654
01:09:17,410 --> 01:09:23,770
Is not really as addressable
with these computational

1655
01:09:23,770 --> 01:09:27,946
quantum mechanics approaches
we've been talking about,

1656
01:09:27,946 --> 01:09:29,529
but I wanted to
mention it, because it

1657
01:09:29,529 --> 01:09:30,439
is a serious problem.

1658
01:09:30,439 --> 01:09:31,460
So if somebody comes
along and says,

1659
01:09:31,460 --> 01:09:34,060
I have this great polymer
that I think won't degrade.

1660
01:09:34,060 --> 01:09:36,700
We can tell them whether it
will still separate charge

1661
01:09:36,700 --> 01:09:37,670
and absorb light.

1662
01:09:37,670 --> 01:09:38,170
Right?

1663
01:09:38,170 --> 01:09:42,960
So we can screen some key
properties for suggestions.

1664
01:09:42,960 --> 01:09:44,850
OK.

1665
01:09:44,850 --> 01:09:46,640
Is that band gap
too high or too low?

1666
01:09:49,450 --> 01:09:50,960
It's pretty high.

1667
01:09:50,960 --> 01:09:51,939
It's pretty high.

1668
01:09:51,939 --> 01:09:52,990
OK?

1669
01:09:52,990 --> 01:09:54,170
Good.

1670
01:09:54,170 --> 01:09:55,420
OK.

1671
01:09:55,420 --> 01:09:59,290
Now, there's one other
I'll talk about today,

1672
01:09:59,290 --> 01:10:03,310
and then I'll finish this up.

1673
01:10:03,310 --> 01:10:04,930
There's not much
more I want to say,

1674
01:10:04,930 --> 01:10:09,160
but I'll finish this up on
Thursday at the beginning.

1675
01:10:09,160 --> 01:10:11,450
But another kind of cell
that is really fun--

1676
01:10:11,450 --> 01:10:15,860
actually in a class I taught
here to 2 1/2 years ago,

1677
01:10:15,860 --> 01:10:17,650
we made these in class.

1678
01:10:17,650 --> 01:10:19,030
We made dye sensitized cells.

1679
01:10:19,030 --> 01:10:21,040
Has anybody made a
dye sensitized cell?

1680
01:10:21,040 --> 01:10:24,040
It's really easy.

1681
01:10:24,040 --> 01:10:27,850
Literally, you just
take some TiO2,

1682
01:10:27,850 --> 01:10:31,090
and we had a contest with fruit.

1683
01:10:31,090 --> 01:10:34,660
We had the team
pomegranate, team raspberry,

1684
01:10:34,660 --> 01:10:37,490
and team blackberry, and
you mash up your fruit,

1685
01:10:37,490 --> 01:10:39,290
and that gives you the dye.

1686
01:10:39,290 --> 01:10:39,790
Right?

1687
01:10:39,790 --> 01:10:42,070
And that's just a light
absorber, and then

1688
01:10:42,070 --> 01:10:46,300
you coat TiO2, you
coat your powder.

1689
01:10:46,300 --> 01:10:47,220
OK?

1690
01:10:47,220 --> 01:10:50,530
And you put it down
on a metal, and then

1691
01:10:50,530 --> 01:10:53,050
you put an electrolyte
in, and I'll tell you

1692
01:10:53,050 --> 01:10:54,280
what's happening in a minute.

1693
01:10:54,280 --> 01:10:59,890
And it's such a simple, cool
fruit power, fruit power.

1694
01:10:59,890 --> 01:11:01,090
Right?

1695
01:11:01,090 --> 01:11:02,500
Solve the world's
energy problems

1696
01:11:02,500 --> 01:11:07,660
with blackberries which
are like $6.99 for a pint.

1697
01:11:07,660 --> 01:11:09,890
So that's scalable, not.

1698
01:11:09,890 --> 01:11:12,340
But anyway, there are
cheaper ways of getting dyes.

1699
01:11:12,340 --> 01:11:16,300
OK, but anyway, the
point is that it's

1700
01:11:16,300 --> 01:11:19,847
really simple and really elegant
and actually pretty efficient.

1701
01:11:19,847 --> 01:11:21,430
These are also known
as Gratzel cells,

1702
01:11:21,430 --> 01:11:25,810
because Gratzel was really
one of the key founders

1703
01:11:25,810 --> 01:11:29,460
of this idea, I think
the key founder.

1704
01:11:29,460 --> 01:11:31,210
And there are companies
making and selling

1705
01:11:31,210 --> 01:11:32,468
dye sensitized cells.

1706
01:11:32,468 --> 01:11:34,510
The reason they're appealing
is they are actually

1707
01:11:34,510 --> 01:11:37,120
even more efficient than
polymer-based cells.

1708
01:11:37,120 --> 01:11:42,650
They're up to I
think even 14%, 15%.

1709
01:11:42,650 --> 01:11:44,760
Again, the issue is packaging.

1710
01:11:44,760 --> 01:11:45,260
Why?

1711
01:11:45,260 --> 01:11:50,390
Because in this case, what
happens is the dye coats

1712
01:11:50,390 --> 01:11:54,750
a particle that is really good
at pulling the electrons off.

1713
01:11:54,750 --> 01:11:56,000
Oh, you love these pictures.

1714
01:11:56,000 --> 01:11:56,750
You know them now.

1715
01:11:56,750 --> 01:11:58,200
You feel them now.

1716
01:11:58,200 --> 01:11:58,700
Right?

1717
01:11:58,700 --> 01:12:02,820
These are the pictures we use
to describe solar technologies.

1718
01:12:02,820 --> 01:12:03,320
Right?

1719
01:12:03,320 --> 01:12:06,530
So you have the dye, where
sunlight shines on it,

1720
01:12:06,530 --> 01:12:08,060
and an electron gets kicked up.

1721
01:12:08,060 --> 01:12:10,460
Now, in this case,
you see, you're

1722
01:12:10,460 --> 01:12:14,090
not going to worry about
moving that electron around

1723
01:12:14,090 --> 01:12:19,610
in some slow polymer
or amorphous material.

1724
01:12:19,610 --> 01:12:21,800
You're going to put it
into a TiO2 particle

1725
01:12:21,800 --> 01:12:26,040
matrix, where actually electrons
zip along really efficiently.

1726
01:12:26,040 --> 01:12:30,880
So you really boost the
transport part of the problem.

1727
01:12:30,880 --> 01:12:33,730
See, TiO2 by itself wouldn't
absorb light very well,

1728
01:12:33,730 --> 01:12:36,740
so the dye helps you kick
the light absorption up.

1729
01:12:36,740 --> 01:12:37,850
So that really helps.

1730
01:12:37,850 --> 01:12:40,220
So the light is absorbed
here, and then the electron

1731
01:12:40,220 --> 01:12:42,770
transfers to the TiO2
and goes out here.

1732
01:12:42,770 --> 01:12:47,000
And here's the problem is you
got to get that electron back,

1733
01:12:47,000 --> 01:12:49,400
and the way you get
it back is you get it

1734
01:12:49,400 --> 01:12:55,445
back from the cathode, through
a dye, through an electrolyte--

1735
01:12:55,445 --> 01:12:57,273
I'm sorry, through
an electrolyte,

1736
01:12:57,273 --> 01:12:58,190
I just read that word.

1737
01:13:03,110 --> 01:13:08,300
Which is basically just shipping
electrons from this metal

1738
01:13:08,300 --> 01:13:10,820
electrode over to the dye.

1739
01:13:10,820 --> 01:13:14,210
It's just an iodine solution,
very cheap, very cheap,

1740
01:13:14,210 --> 01:13:16,100
but it's just an electrolyte.

1741
01:13:16,100 --> 01:13:20,180
Has anybody heard of or
worked with electrolytes?

1742
01:13:20,180 --> 01:13:21,920
Well, even if not, it's OK.

1743
01:13:21,920 --> 01:13:22,430
Right?

1744
01:13:22,430 --> 01:13:26,775
There in lots of uses, and
basically, what they are

1745
01:13:26,775 --> 01:13:27,650
is they're a shuttle.

1746
01:13:27,650 --> 01:13:32,960
It's atoms, I3's, molecules that
are carrying electrons for you

1747
01:13:32,960 --> 01:13:33,972
in a liquid.

1748
01:13:33,972 --> 01:13:35,180
What's the problem with that?

1749
01:13:39,610 --> 01:13:40,750
Yeah.

1750
01:13:40,750 --> 01:13:46,600
It's how are you going to
keep that liquid in there

1751
01:13:46,600 --> 01:13:48,620
for 20 years?

1752
01:13:48,620 --> 01:13:55,000
20 years you need a liquid
to stay perfectly liquid.

1753
01:13:55,000 --> 01:13:55,700
That's hard.

1754
01:13:55,700 --> 01:13:56,200
Right?

1755
01:13:56,200 --> 01:13:57,700
There's a lot of
work that's been

1756
01:13:57,700 --> 01:14:01,370
done on solid-state electrolytes
for other applications.

1757
01:14:01,370 --> 01:14:01,870
Right?

1758
01:14:01,870 --> 01:14:04,090
Lots of work has been done.

1759
01:14:04,090 --> 01:14:08,170
It's an entire field that,
if you had a good enough one,

1760
01:14:08,170 --> 01:14:10,600
could be used here
as well, but so far,

1761
01:14:10,600 --> 01:14:15,100
solid-state electrolytes do
not get you the electrons

1762
01:14:15,100 --> 01:14:16,660
efficiently enough.

1763
01:14:16,660 --> 01:14:17,380
OK?

1764
01:14:17,380 --> 01:14:21,040
But when you use this very
cheap liquid and very cheap dye

1765
01:14:21,040 --> 01:14:24,820
and very cheap nanoparticles,
you get a 14% efficient cell,

1766
01:14:24,820 --> 01:14:29,290
and you can do it in one hour
violating only a few codes,

1767
01:14:29,290 --> 01:14:31,990
safety codes, in a classroom.

1768
01:14:31,990 --> 01:14:33,070
OK?

1769
01:14:33,070 --> 01:14:37,270
So it's actually a
really cool technology.

1770
01:14:37,270 --> 01:14:40,450
I'll pick up here and finish
the solar PV part on Thursday.

1771
01:14:40,450 --> 01:14:44,260
Please, come to class, because
we will fill out evaluations

1772
01:14:44,260 --> 01:14:45,360
and have pizza.

1773
01:14:45,360 --> 01:14:46,910
OK.