工程師們剛剛用人工原子制造了一個(gè)非常穩(wěn)定的量子硅芯片

The team created their atoms using a metal surface gate electrode to apply voltage to the silicon, attracting spare electrons from the silicon into the quantum dot.

研究小組用一個(gè)金屬表面柵電極來給硅施加電壓，將硅中的多余電子吸引到量子點(diǎn)中，從而創(chuàng)造出原子。

In a real atom, you have a positive charge in the middle, being the nucleus, and then the negatively charged electrons are held around it in three-dimensional orbits, explained solid state physicist Andre Saraiva of UNSW.

新南威爾士大學(xué)的固體物理學(xué)家安德烈·薩拉瓦解釋說:“在真實(shí)的原子中，正電荷在原子核的中間，然后帶負(fù)電荷的電子在三維軌道中圍繞原子核運(yùn)動(dòng)。”

In our case, rather than the positive nucleus, the positive charge comes from the gate electrode which is separated from the silicon by an insulating barrier of silicon oxide, and then the electrons are suspended underneath it, each orbiting around the centre of the quantum dot. But rather than forming a sphere, they are arranged flat, in a disc.

“在我們的例子中，正電荷不是來自帶正電的原子核，而是來自柵電極，柵電極被氧化硅的絕緣屏障從硅中分離出來，然后電子懸浮在柵電極下，每個(gè)電子都繞著量子點(diǎn)的中心旋轉(zhuǎn)。但它們不是形成一個(gè)球體，而是平展地排列在一個(gè)圓盤里。”

Hydrogen, lithium and sodium are elements that can have just one electron in their electron shell. This is the model used for quantum computing. When the team creates artificial atoms equivalent to hydrogen, lithium and sodium, they can use that single electron as a qubit, the quantum version of a binary bit.

氫、鋰和鈉都是電子殼層中只有一個(gè)電子的元素。這是用于量子計(jì)算的模型。當(dāng)這個(gè)團(tuán)隊(duì)創(chuàng)造出相當(dāng)于氫、鋰和鈉的人工原子時(shí)，他們就可以用那個(gè)電子作為量子位元，也就是二進(jìn)制位的量子版本。

However, unlike binary bits, which process information in one of two states (1 or 0), a qubit can be in the state of a 1, a 0, or both simultaneously - a state called superposition - based on their spin states. This means they can perform parallel computations, rather than do them consecutively, making them a much more powerful computing tool.

然而，與二進(jìn)制位不同的是，二進(jìn)制位以兩種狀態(tài)(1或0)之一處理信息，量子位可以同時(shí)處于1、0或兩者的狀態(tài)——一種稱為疊加的狀態(tài)——基于它們的自旋狀態(tài)。這意味著它們可以執(zhí)行并行計(jì)算，而不是連續(xù)執(zhí)行，這使它們成為更強(qiáng)大的計(jì)算工具。

This is what the team demonstrated previously, but the system wasn't perfect.

這是該團(tuán)隊(duì)之前展示的，但是這個(gè)系統(tǒng)并不完美。

Up until now, imperfections in silicon devices at the atomic level have disrupted the way qubits behave, leading to unreliable operation and errors, said UNSW quantum engineer Ross Leon.

新南威爾士大學(xué)的量子工程師羅斯·利昂說:“到目前為止，原子水平上的硅器件的缺陷已經(jīng)擾亂了量子位元的行為方式，導(dǎo)致了不可靠的操作和錯(cuò)誤。”

So, the team turned up the voltage on their gate electrode, which drew in more electrons; these electrons, in turn, mimic heavier atoms, which have multiple electron shells. In the artificial atoms, just as in real atoms, these shells are predictable and well organised.

因此，研究小組提高了柵極上的電壓，從而吸引了更多的電子;這些電子，反過來，模仿更重的原子，有多個(gè)電子殼層。在人造原子中，就像在真實(shí)原子中一樣，這些殼層是可預(yù)測(cè)的，而且組織良好。

When the electrons in either a real atom or our artificial atoms form a complete shell, they align their poles in opposite directions so that the total spin of the system is zero, making them useless as a qubit. But when we add one more electron to start a new shell, this extra electron has a spin that we can now use as a qubit again, Dzurak said.

“當(dāng)一個(gè)真正的原子或人造原子中的電子形成一個(gè)完整的殼層時(shí)，它們將它們的極向相反的方向排列，因此系統(tǒng)的總自旋為零，使得它們作為一個(gè)量子位毫無用處。”但是當(dāng)我們?cè)僭黾右粋€(gè)電子來開始一個(gè)新的殼層時(shí)，這個(gè)額外的電子就有了一個(gè)自旋，我們現(xiàn)在又可以把它用作量子位元了，”Dzurak說。

This new set-up also appears to compensate for the errors introduced by atomic-scale imperfections in the silicon chip.

這種新的裝置似乎也彌補(bǔ)了硅片原子尺度上的缺陷所帶來的誤差。

Our new work shows that we can control the spin of electrons in the outer shells of these artificial atoms to give us reliable and stable qubits, said Dzurak.

“我們的新研究表明，我們可以控制這些人造原子外層電子的自旋，從而得到可靠而穩(wěn)定的量子位。”祖拉克說。

This is really important because it means we can now work with much less fragile qubits. One electron is a very fragile thing. However an artificial atom with 5 electrons, or 13 electrons, is much more robust.

“這真的很重要，因?yàn)檫@意味著我們現(xiàn)在可以用更少的脆弱量子位來工作。一個(gè)電子是非常脆弱的。然而，一個(gè)擁有5個(gè)或13個(gè)電子的人工原子要健壯得多。”

The research has been published in Nature Communications.

這項(xiàng)研究發(fā)表在《自然通訊》雜志上。