音頻
中英對照文本
Computers used to be as big as a room.
計算機過去有一個房間那麼大。
But now they fit in your pocket,
但現在它們裝在你口袋裡了，
on your wrist
在你的手腕上
and can even be implanted inside of your body.
甚至可以植入體內。
How cool is that?
多酷啊？
And this has been enabled by the miniaturization of transistors,
並且這已經通過晶體管的小型化而得以實現，
which are the tiny switches in the circuits
它們是電路中的小開關
at the heart of our computers.
在我們電腦的中心。
And it's been achieved through decades of development
它是通過幾十年的發展而實現的
and breakthroughs in science and engineering
科學和工程上的突破
and of billions of dollars of investment.
以及數十億美元的投資。
But it's given us vast amounts of computing,
但它給了我們大量的計算，
huge amounts of memory
巨大的記憶
and the digital revolution that we all experience and enjoy today.
以及我們今天所經歷和享受的數字革命。
But the bad news is,
但壞消息是，
we're about to hit a digital roadblock,
我們要撞上一個數字路障，
as the rate of miniaturization of transistors is slowing down.
因為晶體管的小型化速度正在減慢。
And this is happening at exactly the same time
這是在同一時間發生的
as our innovation in software is continuing relentlessly
因為我們在軟件方面的創新在不懈地進行
with artificial intelligence and big data.
有人工智能和大數據。
And our devices regularly perform facial recognition or augment our reality
我們的設備定期進行面部識別或增強我們的真實感
or even drive cars down our treacherous, chaotic roads.
或者開車在我們危險混亂的道路上行駛。
It's amazing.
太神奇了。
But if we don't keep up with the appetite of our software,
但如果我們不能跟上我們軟件的胃口，
we could reach a point in the development of our technology
我們的技術可以發展到一定程度
where the things that we could do with software could, in fact, be limited
實際上，我們可以用軟件做的事情是有限的
by our hardware.
我們的硬件。
We've all experienced the frustration of an old smartphone or tablet
我們都經歷過舊智能手機或平板電腦帶來的挫敗感
grinding slowly to a halt over time
隨著時間的推移慢慢地停下來
under the ever-increasing weight of software updates and new features.
在軟件更新和新特性的不斷增加的重量下。
And it worked just fine when we bought it not so long ago.
不久之前我們買的時候它還挺好的。
But the hungry software engineers have eaten up all the hardware capacity
但是飢餓的軟件工程師已經耗盡了所有的硬件能力
over time.
久而久之。
The semiconductor industry is very well aware of this
半導體行業非常清楚這一點
and is working on all sorts of creative solutions,
並且正在研究各種創造性的解決方案，
such as going beyond transistors to quantum computing
例如從晶體管到量子計算
or even working with transistors in alternative architectures
或者甚至在另一種結構中與晶體管一起工作
such as neural networks
例如神經網絡
to make more robust and efficient circuits.
以製造更魯棒和更有效的電路。
But these approaches will take quite some time,
但是這些方法需要相當長的時間，
and we're really looking for a much more immediate solution to this problem.
我們真的在尋找一個更直接的方法來解決這個問題。
The reason why the rate of miniaturization of transistors is slowing down
晶體管的小型化速度減慢的原因
is due to the ever-increasing complexity of the manufacturing process.
這是由於製造工藝的日益增加的複雜性。
The transistor used to be a big, bulky device,
晶體管曾經是一個大而笨重的器件，
until the invent of the integrated circuit
在集成電路發明之前
based on pure crystalline silicon wafers.
基於純結晶硅晶片。
And after 50 years of continuous development,
在50年的持續發展之後，
we can now achieve transistor features dimensions
我們現在可以實現晶體管特徵尺寸
down to 10 nanometers.
低至10納米。
You can fit more than a billion transistors
你可以安裝超過十億個晶體管
in a single square millimeter of silicon.
在一平方毫米的硅中。
And to put this into perspective:
為了正確地看待這個問題：
a human hair is 100 microns across.
人的頭髮有100微米寬。
A red blood cell, which is essentially invisible,
一個基本不可見的紅細胞，
is eight microns across,
是8微米寬，
and you can place 12 across the width of a human hair.
你可以在一根頭髮的寬度上放12個。
But a transistor, in comparison, is much smaller,
但與之相比，晶體管要小得多，
at a tiny fraction of a micron across.
直徑只有一微米的微小部分。
You could place more than 260 transistors
你可以放置超過260個晶體管
across a single red blood cell
穿過一個紅細胞
or more than 3,000 across the width of a human hair.
或者超過3000人的頭髮寬度。
It really is incredible nanotechnology in your pocket right now.
它真的是令人難以置信的納米技術在您的口袋中現在。
And besides the obvious benefit
另外還有明顯的好處
of being able to place more, smaller transistors on a chip,
能夠在芯片上放置更多更小的晶體管，
smaller transistors are faster switches,
更小的晶體管是更快的開關，
and smaller transistors are also more efficient switches.
而較小的晶體管也是更有效的開關。
So this combination has given us
所以這個組合給了我們
lower cost, higher performance and higher efficiency electronics
更低的成本，更高的性能和更高效率的電子
that we all enjoy today.
我們今天都很享受。
To manufacture these integrated circuits,
為了製造這些集成電路，
the transistors are built up layer by layer,
這些晶體管是逐層構建的，
on a pure crystalline silicon wafer.
在純結晶硅晶片上。
And in an oversimplified sense,
在過於簡化的意義上，
every tiny feature of the circuit is projected
電路的每一個微小特徵都被投射出來
onto the surface of the silicon wafer
在硅晶片的表面上
and recorded in a light-sensitive material
並記錄在感光材料中
and then etched through the light-sensitive material
然後蝕刻穿過光敏材料
to leave the pattern in the underlying layers.
以將圖案留在下面的層中。
And this process has been dramatically improved over the years
多年來這一過程得到了極大的改善
to give the electronics performance we have today.
為我們今天的電子錶演做準備。
But as the transistor features get smaller and smaller,
但是隨著晶體管特徵變得越來越小，
we're really approaching the physical limitations
我們真的接近生理極限了
of this manufacturing technique.
關於這種製造技術。
The latest systems for doing this patterning
完成這種模式的最新系統
have become so complex
變得如此複雜
that they reportedly cost more than 100 million dollars each.
據說他們每個人的花費都超過1億美元。
And semiconductor factories contain dozens of these machines.
半導體工廠裡有幾十臺這樣的機器。
So people are seriously questioning: Is this approach long-term viable?
因此，人們正在嚴肅地質疑：這種方法長期可行嗎？
But we believe we can do this chip manufacturing
但我們相信我們可以做芯片製造
in a totally different and much more cost-effective way
以一種完全不同且更具成本效益的方式
using molecular engineering and mimicking nature
利用分子工程學和模仿自然
down at the nanoscale dimensions of our transistors.
在我們晶體管的納米尺度上。
As I said, the conventional manufacturing takes every tiny feature of the circuit
就像我說的，傳統的製造需要電路的每一個微小的特徵
and projects it onto the silicon.
把它投射到硅上。
But if you look at the structure of an integrated circuit,
但是如果你看看集成電路的結構，
the transistor arrays,
晶體管陣列，
many of the features are repeated millions of times.
許多特徵被重複了數百萬次。
It's a highly periodic structure.
它是一個高度週期性的結構。
So we want to take advantage of this periodicity
所以我們想利用這種週期性
in our alternative manufacturing technique.
在我們的替代製造技術中。
We want to use self-assembling materials
我們要用自組裝材料
to naturally form the periodic structures
自然形成周期性結構
that we need for our transistors.
我們需要的晶體管。
We do this with the materials,
我們用材料做這個，
then the materials do the hard work of the fine patterning,
然後材料進行精細圖案化的艱苦工作，
rather than pushing the projection technology to its limits and beyond.
而不是將投影技術推向極限甚至更遠。
Self-assembly is seen in nature in many different places,
在自然界許多不同的地方都可以看到自我組裝，
from lipid membranes to cell structures,
從脂質膜到細胞結構，
so we do know it can be a robust solution.
所以我們知道它可以是一個強大的解決方案。
If it's good enough for nature, it should be good enough for us.
如果對大自然足夠好，對我們也應該足夠好。
So we want to take this naturally occurring, robust self-assembly
所以我們想把這種自然發生的，強健的自我組裝
and use it for the manufacturing of our semiconductor technology.
並將其用於製造我們的半導體技術。
One type of self-assemble material --
一種自組裝材料--
it's called a block co-polymer --
它被稱為嵌段共聚物--
consists of two polymer chains just a few tens of nanometers in length.
由兩條長度僅幾十納米的聚合物鏈組成。
But these chains hate each other.
但這些鎖鏈互相憎恨。
They repel each other,
它們互相排斥，
very much like oil and water or my teenage son and daughter.
很像油和水或者我十幾歲的兒子和女兒。
(Laughter)
（笑聲）
But we cruelly bond them together,
但我們殘忍地把他們綁在一起，
creating an inbuilt frustration in the system,
在系統中產生固有的挫折感，
as they try to separate from each other.
當他們試圖彼此分離的時候。
And in the bulk material, there are billions of these,
在大塊材料中，有數十億這樣的物質，
and the similar components try to stick together,
並且類似的部件試圖粘在一起，
and the opposing components try to separate from each other
相對的部件試圖彼此分離
at the same time.
同時.
And this has a built-in frustration, a tension in the system.
這是一種內在的挫敗感，一種系統的張力。
So it moves around, it squirms until a shape is formed.
所以它四處移動，蠕動直到形成一個形狀。
And the natural self-assembled shape that is formed is nanoscale,
並且所形成的自然自組裝形狀是納米級的，
it's regular, it's periodic, and it's long range,
它是有規律的，有周期性的，而且是遠距離的，
which is exactly what we need for our transistor arrays.
這正是我們需要的晶體管陣列。
So we can use molecular engineering
所以我們可以利用分子工程學
to design different shapes of different sizes
設計不同大小的不同形狀
and of different periodicities.
以及不同的週期。
So for example, if we take a symmetrical molecule,
例如，如果我們取一個對稱的分子，
where the two polymer chains are similar length,
在兩個聚合物鏈長度相似的情況下，
the natural self-assembled structure that is formed
所形成的自然自組裝結構
is a long, meandering line,
是一條長長的蜿蜒的線，
very much like a fingerprint.
很像指紋。
And the width of the fingerprint lines
以及指紋線的寬度
and the distance between them
他們之間的距離
is determined by the lengths of our polymer chains
是由我們的聚合物鏈的長度決定的
but also the level of built-in frustration in the system.
同時也反映了系統內在的挫折感程度。
And we can even create more elaborate structures
我們甚至可以創造出更復雜的結構
if we use unsymmetrical molecules,
如果我們使用不對稱分子，
where one polymer chain is significantly shorter than the other.
其中一個聚合物鏈明顯短於另一個。
And the self-assembled structure that forms in this case
以及在這種情況下形成的自組裝結構
is with the shorter chains forming a tight ball in the middle,
其中較短的鏈在中間形成緊密的球，
and it's surrounded by the longer, opposing polymer chains,
它被相對的長聚合物鏈包圍著，
forming a natural cylinder.
形成一個自然的圓柱體。
And the size of this cylinder
以及這個圓柱體的大小
and the distance between the cylinders, the periodicity,
圓柱體之間的距離，週期性，
is again determined by how long we make the polymer chains
又是由我們製造聚合物鏈的時間決定的
and the level of built-in frustration.
以及內在挫折感的程度。
So in other words, we're using molecular engineering
換句話說，我們用的是分子工程學
to self-assemble nanoscale structures
納米尺度結構的自組裝
that can be lines or cylinders the size and periodicity of our design.
這可以是線條或圓柱的大小和週期性我們的設計。
We're using chemistry, chemical engineering,
我們用的是化學，化學工程，
to manufacture the nanoscale features that we need for our transistors.
以製造我們晶體管所需的納米級特徵。
But the ability to self-assemble these structures
但自我組裝這些結構的能力
only takes us half of the way,
只花了我們一半的時間，
because we still need to position these structures
因為我們仍然需要定位這些結構
where we want the transistors in the integrated circuit.
我們需要集成電路中的晶體管。
But we can do this relatively easily
但我們可以相對容易地做到
using wide guide structures that pin down the self-assembled structures,
使用固定自組裝結構的寬引導結構，
anchoring them in place
把它們固定在合適的位置
and forcing the rest of the self-assembled structures
強迫剩下的自組裝結構
to lie parallel,
平行的躺著，
aligned with our guide structure.
與我們的導向結構一致。
For example, if we want to make a fine, 40-nanometer line,
例如，如果我們想製作一條40納米的細線，
which is very difficult to manufacture with conventional projection technology,
這很難用傳統的投影技術來製造，
we can manufacture a 120-nanometer guide structure
我們可以製造一個120納米的導向結構
with normal projection technology,
使用普通的投影技術，
and this structure will align three of the 40-nanometer lines in between.
並且這種結構將使其間的40納米線中的三條對準。
So the materials are doing the most difficult fine patterning.
因此這些材料正在進行最困難的精細圖案化。
And we call this whole approach "directed self-assembly."
我們稱之為"直接的自我組裝"。
The challenge with directed self-assembly
直接自我組裝的挑戰
is that the whole system needs to align almost perfectly,
整個系統需要幾乎完美地對齊，
because any tiny defect in the structure could cause a transistor failure.
因為任何微小的結構缺陷都可能導致晶體管失效。
And because there are billions of transistors in our circuit,
因為我們的電路中有數十億個晶體管，
we need an almost molecularly perfect system.
我們需要一個幾乎分子完美的系統。
But we're going to extraordinary measures
但我們會採取非常措施
to achieve this,
為了達到這個目的，
from the cleanliness of our chemistry
從我們化學上的清潔
to the careful processing of these materials
仔細處理這些材料
in the semiconductor factory
在半導體工廠
to remove even the smallest nanoscopic defects.
以去除甚至最小的納米缺陷。
So directed self-assembly is an exciting new disruptive technology,
所以定向自組裝是一種令人興奮的新的顛覆性技術，
but it is still in the development stage.
但它仍處於發展階段。
But we're growing in confidence that we could, in fact, introduce it
但我們越來越有信心，事實上，我們可以把它介紹給
to the semiconductor industry
為了半導體產業
as a revolutionary new manufacturing process
作為一個革命性的新制造過程
in just the next few years.
在接下來的幾年裡。
And if we can do this, if we're successful,
如果我們能做到，如果我們成功了，
we'll be able to continue
我們就能繼續
with the cost-effective miniaturization of transistors,
隨著晶體管成本有效的小型化，
continue with the spectacular expansion of computing
繼續進行大規模的計算
and the digital revolution.
以及數字革命。
And what's more, this could even be the dawn of a new era
更重要的是，這甚至可能是一個新時代的開始
of molecular manufacturing.
分子製造。
How cool is that?
多酷啊？
Thank you.
謝謝你。
(Applause)
（掌聲）
閱讀更多 一起看演講 的文章
