Where Imperfections Lead To Opportunity

By Evelyn Hu

It is natural to hold a bias that assumes that the highest-quality devices are those formed from the most perfect materials (crystalline, well-ordered, stoichiometric). Therefore, it is ironic, and perhaps counterintuitive, that particular kinds of defects, such as vacancies (missing atoms) in semiconductor materials, can form the building blocks of a new quantum information technology. Such defects often produce the brilliant colors in otherwise transparent materials, such as crystalline diamond, and, thus, we often refer to these defects as “color centers.” The new opportunities made possible by these imperfections relate to the creation of quantum mechanical bits of information, or “qubits.”

Bits and qubits

Today’s powerful semiconductor information and computing technologies build upon the idea of “bits,” basic units of information that generally have one of two possible values (for example, “0” and “1”). Often those two values are embodied by the state of a transistor, whether it is “on” (conducting) or “off” (non-conducting). The concatenation of billions of logical “0s” and “1s” represents tremendous information complexity, but it is a complexity that maps back to the numerous, small-scale, densely-packed physical components (transistors) that embody those bits.

A quantum mechanical bit or “qubit” may be regarded as the analogue of the classical bit. Like a classical bit, the qubit is generally a two-level system, with a “0” (often represented as |0⟩) and “1” (often represented as |1⟩). The physical embodiment of a qubit might be a superconducting junction, a trapped ion, or the spin of an electron. Beyond just |0⟩ and |1⟩, the quantum mechanical description allows for a qubit state that is a linear combination or superposition of |0⟩ and |1⟩.

Quantum mechanics also allows entanglement of qubits. This relates to a correlation of the states of the qubits, so that the description of one qubit in an assembly cannot be made independently of the state of the other qubits. For example, the spin of an electron is a quantum mechanical descriptor, and if a pair of entangled spins is generated with net zero spin, and one spin is measured to have a given polarization or direction, then the other spin (qubit) must have the opposite polarization. Superposition and entanglement endow an assembly of qubits with far greater information than their classical counterparts, and this forms part of the enormous allure of the emerging areas of quantum information technologies.

What are the conditions under which the information in arrays of qubits can be preserved and controlled? The major challenge relates to maintaining the coherence of the qubits; preparing particular quantum mechanical states (for example, a particular spin polarization), and preserving that spin coherence over times long enough to ensure that various operations can be performed on the qubit assembly. In recent years, the research community has made profound advances in the control and analysis of a wide variety of qubits, with a parallel development in quantum algorithms and error correction. Superconducting qubits are most mature in manufacturability and scaling up, and these form the basis of quantum computer developments by major companies such as IBM, Google and Intel. Other qubit implementations are also in nascent commercialization, including those based on trapped ions and cold atoms. In this first flowering of quantum information systems, it is too early to predict which physical qubit systems, if any, will be dominant.

Semiconductor defects as qubits

The tremendous impact of semiconductor-based integrated circuit technology has hallmarked the advantages of compact, “solid-state” computing and communications platforms. Color center or “defect qubits,” formed in materials such as diamond and silicon carbide (SiC), may be able to leverage the infrastructure of semiconductor device processing and technology for the creation of new quantum information technologies.

The essential elements of a defect qubit relate to atom-like structures, with well-defined electron spin states, and long spin coherence lifetimes, even at room temperature. The best-studied defect qubit has been the negatively-charged “nitrogen-vacancy (NV −) center” in single crystal diamond: a substitutional nitrogen atom in proximity to a carbon-vacancy. More recently, other types of defects have been explored in diamond, in addition to defect qubits in materials such as SiC. The correlation between a photon and spin states of the defects has allowed preparation, control, and “readout” of the quantum state of the qubit, using optical means. This provides an important advantage in both initial understanding of the qubits themselves and in the longer-term buildup of a technology.

To better appreciate the attributes of defect qubits, it may be useful to adopt a “half-full” picture: rather than thinking of an imperfect structure within an otherwise perfect single-crystal host, envision an atomic-scale entity with a characteristic energy situated within a semiconductor energy gap, relatively isolated from valence or conduction bands. The surrounding semiconductor material then serves to isolate and protect the defect state. Moreover, the highly localized electron/spin states of the defect determine the distinctive spin-photon coupling of the defect qubits, and as well influence their spin coherence lifetimes. The exceptional sensitivity of the defect qubits to their local environment provides both advantage and challenge. Such defect qubits can serve as exquisitely sensitive probes of their local spin/magnetic environment, and the use of such qubits as sensors is already being pursued. However, such sensitivity, even to the nuclear spins in the qubit’s local environment, can be the source of decoherence of the qubit.

Beyond limiting decoherence, there are numerous challenges to the creation of these defect qubits by semiconductor processing. At the foundational level, these challenges include the controlled fabrication with spatial precision of the defects. In the creation of defects, we must avoid excessive collateral damage, which would negatively influence the coherence times of the qubit. At more complex device levels, there are profound challenges in how to engender entanglement, create robust and stable qubit signals, and achieve low-loss signal transmission. At the same time, there is a considerable palette of other possible defect qubits that are as yet unexplored. Defects in other “wide bandgap” materials, such as gallium nitride (GaN) and zinc oxide (ZnO), may enable photon signals with different spectral signatures and stabilities, different and perhaps tunable sensitivities to local strain or electro-optic properties, giving rise to new, integrative quantum device functionality.

Balancing the promise of semiconductor qubits that are as yet unexplored with addressing the challenges surrounding better-developed qubits offers an unparalleled opportunity to make substantial innovations for future quantum information systems.