0:00:00.949,0:00:02.773
I'm thrilled to be here tonight

0:00:02.773,0:00:05.152
to share with you something[br]we've been working on

0:00:05.152,0:00:07.242
for over two years,

0:00:07.242,0:00:09.796
and it's in the area[br]of additive manufacturing,

0:00:09.796,0:00:12.513
also known as 3D printing.

0:00:12.513,0:00:14.231
You see this object here.

0:00:14.231,0:00:18.039
It looks fairly simple,[br]but it's quite complex at the same time.

0:00:18.549,0:00:21.800
It's a set of concentric[br]geodesic structures

0:00:21.800,0:00:24.795
with linkages between each one.

0:00:24.795,0:00:30.797
In its context, it is not manufacturable[br]by traditional manufacturing techniques.

0:00:31.343,0:00:35.290
It has a symmetry such[br]that you can't injection mold it.

0:00:35.290,0:00:38.879
You can't even manufacture it[br]through milling.

0:00:39.470,0:00:42.117
This is a job for a 3D printer,

0:00:42.117,0:00:46.598
but most 3D printers would take between[br]three and 10 hours to fabricate it,

0:00:46.598,0:00:50.824
and we're going to take the risk tonight[br]to try to fabricate it onstage

0:00:50.824,0:00:53.401
during this 10-minute talk.

0:00:53.401,0:00:55.440
Wish us luck.

0:00:56.350,0:00:59.624
Now, 3D printing is actually a misnomer.

0:00:59.624,0:01:03.399
It's actually 2D printing[br]over and over again,

0:01:03.919,0:01:07.761
and it in fact uses the technologies[br]associated with 2D printing.

0:01:08.401,0:01:13.360
Think about inkjet printing where you[br]lay down ink on a page to make letters,

0:01:13.360,0:01:18.346
and then do that over and over again[br]to build up a three-dimensional object.

0:01:18.346,0:01:20.417
In microelectronics, they use something

0:01:20.417,0:01:22.737
called lithography to do[br]the same sort of thing,

0:01:22.737,0:01:24.945
to make the transistors[br]and integrated circuits

0:01:24.945,0:01:26.997
and build up a structure several times.

0:01:26.997,0:01:29.399
These are all 2D printing technologies.

0:01:30.099,0:01:33.987
Now, I'm a chemist,[br]a material scientist too,

0:01:33.987,0:01:36.711
and my co-inventors[br]are also material scientists,

0:01:36.711,0:01:39.010
one a chemist, one a physicist,

0:01:39.010,0:01:41.936
and we began to be[br]interested in 3D printing.

0:01:41.936,0:01:47.531
And very often, as you know,[br]new ideas are often simple connections

0:01:47.531,0:01:51.274
between people with different experiences[br]in different communities,

0:01:51.274,0:01:52.751
and that's our story.

0:01:53.591,0:01:56.122
Now, we were inspired

0:01:56.122,0:02:00.893
by the "Terminator 2" scene for T-1000,

0:02:00.893,0:02:05.836
and we thought, why couldn't a 3D printer[br]operate in this fashion,

0:02:06.426,0:02:10.362
where you have an object[br]arise out of a puddle

0:02:11.052,0:02:13.520
in essentially real time

0:02:13.520,0:02:15.749
with essentially no waste

0:02:15.749,0:02:18.071
to make a great object?

0:02:18.071,0:02:19.488
Okay, just like the movies.

0:02:19.488,0:02:22.877
And could we be inspired by Hollywood

0:02:22.877,0:02:26.384
and come up with ways[br]to actually try to get this to work?

0:02:26.384,0:02:28.450
And that was our challenge.

0:02:28.450,0:02:31.817
And our approach would be,[br]if we could do this,

0:02:31.817,0:02:35.671
then we could fundamentally address[br]the three issues holding back 3D printing

0:02:35.671,0:02:38.086
from being a manufacturing process.

0:02:38.086,0:02:40.617
One, 3D printing takes forever.

0:02:40.617,0:02:45.841
There are mushrooms that grow faster[br]than 3D printed parts. (Laughter)

0:02:47.281,0:02:49.417
The layer by layer process

0:02:49.417,0:02:52.319
leads to defects[br]in mechanical properties,

0:02:52.319,0:02:56.266
and if we could grow continuously,[br]we could eliminate those defects.

0:02:56.266,0:03:01.398
And in fact, if we could grow really fast,[br]we could also start using materials

0:03:01.398,0:03:06.042
that are self-curing,[br]and we could have amazing properties.

0:03:06.042,0:03:10.151
So if we could pull this off,[br]imitate Hollywood,

0:03:10.151,0:03:12.912
we could in fact address 3D manufacturing.

0:03:14.702,0:03:17.953
Our approach is to use[br]some standard knowledge

0:03:17.953,0:03:20.553
in polymer chemistry

0:03:20.553,0:03:27.152
to harness light and oxygen[br]to grow parts continuously.

0:03:27.152,0:03:30.099
Light and oxygen work in different ways.

0:03:30.099,0:03:33.141
Light can take a resin[br]and convert it to a solid,

0:03:33.141,0:03:35.295
can convert a liquid to a solid.

0:03:35.295,0:03:38.829
Oxygen inhibits that process.

0:03:38.829,0:03:42.080
So light and oxygen[br]are polar opposites from one another

0:03:42.080,0:03:44.588
from a chemical point of view,

0:03:44.588,0:03:48.001
and if we can control spatially[br]the light and oxygen,

0:03:48.001,0:03:49.948
we could control this process.

0:03:50.288,0:03:53.739
And we refer to this as CLIP.[br][Continuous Liquid Interface Production.]

0:03:53.739,0:03:55.615
It has three functional components.

0:03:56.465,0:04:00.326
One, it has a reservoir[br]that holds the puddle,

0:04:00.326,0:04:02.205
just like the T-1000.

0:04:02.205,0:04:04.621
At the bottom of the reservoir[br]is a special window.

0:04:04.621,0:04:06.112
I'll come back to that.

0:04:06.112,0:04:09.892
In addition, it has a stage[br]that will lower into the puddle

0:04:09.892,0:04:12.481
and pull the object out of the liquid.

0:04:12.481,0:04:16.285
The third component[br]is a digital light projection system

0:04:16.285,0:04:18.305
underneath the reservoir,

0:04:18.305,0:04:21.578
illuminating with light[br]in the ultraviolet region.

0:04:22.048,0:04:25.271
Now, the key is that this window[br]in the bottom of this reservoir,

0:04:25.271,0:04:28.150
it's a composite,[br]it's a very special window.

0:04:28.150,0:04:31.796
It's not only transparent to light[br]but it's permeable to oxygen.

0:04:31.796,0:04:34.455
It's got characteristics[br]like a contact lens.

0:04:35.435,0:04:37.716
So we can see how the process works.

0:04:37.716,0:04:41.130
You can start to see that[br]as you lower a stage in there,

0:04:41.130,0:04:45.309
in a traditional process,[br]with an oxygen-impermeable window,

0:04:45.309,0:04:47.148
you make a two-dimensional pattern

0:04:48.008,0:04:51.370
and you end up gluing that onto the window[br]with a traditional window,

0:04:51.370,0:04:54.922
and so in order to introduce[br]the next layer, you have to separate it,

0:04:54.922,0:04:58.451
introduce new resin, reposition it,

0:04:58.451,0:05:00.910
and do this process over and over again.

0:05:01.400,0:05:03.234
But with our very special window,

0:05:03.234,0:05:06.563
what we're able to do is,[br]with oxygen coming through the bottom

0:05:06.563,0:05:07.816
as light hits it,

0:05:09.256,0:05:11.926
that oxygen inhibits the reaction,

0:05:11.926,0:05:14.550
and we form a dead zone.

0:05:14.550,0:05:18.869
This dead zone is on the order[br]of tens of microns thick,

0:05:18.869,0:05:22.096
so that's two or three diameters[br]of a red blood cell,

0:05:22.096,0:05:24.627
right at the window interface[br]that remains a liquid,

0:05:24.627,0:05:26.577
and we pull this object up,

0:05:26.577,0:05:28.969
and as we talked about in a Science paper,

0:05:28.969,0:05:33.682
as we change the oxygen content,[br]we can change the dead zone thickness.

0:05:33.682,0:05:37.374
And so we have a number of key variables[br]that we control: oxygen content,

0:05:37.374,0:05:40.439
the light, the light intensity,[br]the dose to cure,

0:05:40.439,0:05:42.401
the viscosity, the geometry,

0:05:42.401,0:05:45.817
and we use very sophisticated software[br]to control this process.

0:05:46.697,0:05:49.460
The result is pretty staggering.

0:05:49.460,0:05:53.196
It's 25 to 100 times faster[br]than traditional 3D printers,

0:05:54.336,0:05:56.170
which is game-changing.

0:05:56.170,0:06:00.506
In addition, as our ability[br]to deliver liquid to that interface,

0:06:00.506,0:06:04.246
we can go 1,000 times faster I believe,

0:06:04.246,0:06:07.803
and that in fact opens up the opportunity[br]for generating a lot of heat,

0:06:07.803,0:06:11.866
and as a chemical engineer,[br]I get very excited at heat transfer

0:06:11.866,0:06:16.045
and the idea that we might one day[br]have water-cooled 3D printers,

0:06:16.045,0:06:18.437
because they're going so fast.

0:06:18.437,0:06:22.500
In addition, because we're growing things,[br]we eliminate the layers,

0:06:22.500,0:06:24.474
and the parts are monolithic.

0:06:24.474,0:06:26.564
You don't see the surface structure.

0:06:26.564,0:06:29.057
You have molecularly smooth surfaces.

0:06:29.057,0:06:33.297
And the mechanical properties[br]of most parts made in a 3D printer

0:06:33.297,0:06:37.593
are notorious for having properties[br]that depend on the orientation

0:06:37.593,0:06:41.354
with which how you printed it,[br]because of the layer-like structure.

0:06:41.354,0:06:43.699
But when you grow objects like this,

0:06:43.699,0:06:47.368
the properties are invariant[br]with the print direction.

0:06:47.368,0:06:50.317
These look like injection-molded parts,

0:06:50.317,0:06:53.729
which is very different[br]than traditional 3D manufacturing.

0:06:53.729,0:06:57.259
In addition, we're able to throw

0:06:57.259,0:07:00.835
the entire polymer[br]chemistry textbook at this,

0:07:00.835,0:07:04.826
and we're able to design chemistries[br]that can give rise to the properties

0:07:04.826,0:07:07.868
you really want in a 3D-printed object.

0:07:07.868,0:07:09.205
(Applause)

0:07:09.205,0:07:12.439
There it is. That's great.

0:07:14.049,0:07:17.627
You always take the risk that something[br]like this won't work onstage, right?

0:07:18.177,0:07:21.056
But we can have materials[br]with great mechanical properties.

0:07:21.056,0:07:23.494
For the first time, we can have elastomers

0:07:23.494,0:07:25.955
that are high elasticity[br]or high dampening.

0:07:25.955,0:07:29.368
Think about vibration control[br]or great sneakers, for example.

0:07:29.368,0:07:31.978
We can make materials[br]that have incredible strength,

0:07:32.828,0:07:36.404
high strength-to-weight ratio,[br]really strong materials,

0:07:36.404,0:07:38.517
really great elastomers,

0:07:38.517,0:07:41.242
so throw that in the audience there.

0:07:41.242,0:07:43.878
So great material properties.

0:07:43.878,0:07:47.293
And so the opportunity now,[br]if you actually make a part

0:07:47.293,0:07:50.973
that has the properties[br]to be a final part,

0:07:50.973,0:07:54.073
and you do it in game-changing speeds,

0:07:54.073,0:07:56.860
you can actually transform manufacturing.

0:07:56.860,0:07:59.716
Right now, in manufacturing,[br]what happens is,

0:07:59.716,0:08:02.678
the so-called digital thread[br]in digital manufacturing.

0:08:02.678,0:08:07.717
We go from a CAD drawing, a design,[br]to a prototype to manufacturing.

0:08:07.717,0:08:10.440
Often, the digital thread is broken[br]right at prototype,

0:08:10.440,0:08:12.872
because you can't go[br]all the way to manufacturing

0:08:12.872,0:08:16.587
because most parts don't have[br]the properties to be a final part.

0:08:16.587,0:08:18.978
We now can connect the digital thread

0:08:18.978,0:08:23.227
all the way from design[br]to prototyping to manufacturing,

0:08:23.227,0:08:26.176
and that opportunity[br]really opens up all sorts of things,

0:08:26.176,0:08:31.129
from better fuel-efficient cars[br]dealing with great lattice properties

0:08:31.129,0:08:33.080
with high strength-to-weight ratio,

0:08:33.080,0:08:36.508
new turbine blades,[br]all sorts of wonderful things.

0:08:37.468,0:08:42.623
Think about if you need a stent[br]in an emergency situation,

0:08:42.623,0:08:46.593
instead of the doctor pulling off[br]a stent out of the shelf

0:08:46.593,0:08:48.822
that was just standard sizes,

0:08:48.822,0:08:52.978
having a stent that's designed[br]for you, for your own anatomy

0:08:52.978,0:08:54.789
with your own tributaries,

0:08:54.789,0:08:58.038
printed in an emergency situation[br]in real time out of the properties

0:08:58.038,0:09:01.477
such that the stent could go away[br]after 18 months: really-game changing.

0:09:01.477,0:09:05.633
Or digital dentistry, and making[br]these kinds of structures

0:09:05.633,0:09:08.814
even while you're in the dentist chair.

0:09:08.814,0:09:11.530
And look at the structures[br]that my students are making

0:09:11.530,0:09:13.504
at the University of North Carolina.

0:09:13.504,0:09:16.313
These are amazing microscale structures.

0:09:16.313,0:09:19.309
You know, the world is really good[br]at nano-fabrication.

0:09:19.309,0:09:23.599
Moore's Law has driven things[br]from 10 microns and below.

0:09:23.599,0:09:25.201
We're really good at that,

0:09:25.201,0:09:29.241
but it's actually very hard to make things[br]from 10 microns to 1,000 microns,

0:09:29.241,0:09:31.261
the mesoscale.

0:09:31.261,0:09:34.094
And subtractive techniques[br]from the silicon industry

0:09:34.094,0:09:35.510
can't do that very well.

0:09:35.510,0:09:37.159
They can't etch wafers that well.

0:09:37.159,0:09:39.109
But this process is so gentle,

0:09:39.109,0:09:41.594
we can grow these objects[br]up from the bottom

0:09:41.594,0:09:43.590
using additive manufacturing

0:09:43.590,0:09:45.843
and make amazing things[br]in tens of seconds,

0:09:45.843,0:09:47.932
opening up new sensor technologies,

0:09:47.932,0:09:50.417
new drug delivery techniques,

0:09:50.417,0:09:54.149
new lab-on-a-chip applications,[br]really game-changing stuff.

0:09:55.149,0:09:59.983
So the opportunity of making[br]a part in real time

0:09:59.983,0:10:02.816
that has the properties to be a final part

0:10:02.816,0:10:05.792
really opens up 3D manufacturing,

0:10:05.792,0:10:08.992
and for us, this is very exciting,[br]because this really is owning

0:10:08.992,0:10:15.589
the intersection between hardware,[br]software and molecular science,

0:10:15.589,0:10:19.755
and I can't wait to see what designers[br]and engineers around the world

0:10:19.755,0:10:22.029
are going to be able to do[br]with this great tool.

0:10:22.499,0:10:24.618
Thanks for listening.

0:10:24.618,0:10:29.727
(Applause)