
Zerofield nuclear magnetic resonance (NMR) provides complementary analysis
modalities to those of highfield NMR and allows for ultrahighresolution
spectroscopy and measurement of untruncated spinspin interactions. Unlike for
the highfield case, however, universal quantum control  the ability to
perform arbitrary unitary operations  has not been experimentally
demonstrated in zerofield NMR. This is because the Larmor frequency for all
spins is identically zero at zero field, making it challenging to individually
address different spin species. We realize a compositepulse technique for
arbitrary independent rotations of $^1$H and $^{13}$C spins in a twospin
system. Quantuminformationinspired randomized benchmarking and state
tomography are used to evaluate the quality of the control. We experimentally
demonstrate singlespin control for $^{13}$C with an average gate fidelity of
$0.9960(2)$ and twospin control via a controllednot (CNOT) gate with an
estimated fidelity of $0.99$. The combination of arbitrary singlespin gates
and a CNOT gate is sufficient for universal quantum control of the nuclear spin
system. The realization of complete spin control in zerofield NMR is an
essential step towards applications to quantum simulation,
entangledstateassisted quantum metrology, and zerofield NMR spectroscopy.

Superposition, arguably the most fundamental property of quantum mechanics,
lies at the heart of quantum information science. However, how to create the
superposition of any two unknown pure states remains as a daunting challenge.
Recently, it is proved that such a quantum protocol does not exist if the two
input states are completely unknown, whereas a probabilistic protocol is still
available with some prior knowledge about the input states [M. Oszmaniec
\emph{et al.}, Phys. Rev. Lett. 116, 110403 (2016)]. The knowledge is that both
of the two input states have nonzero overlaps with some given referential
state. In this work, we experimentally realize the probabilistic protocol of
superposing two pure states in a threequbit nuclear magnetic resonance system.
We demonstrate the feasibility of the protocol by preparing a families of input
states, and the average fidelity between the prepared state and expected
superposition state is over 99%. Moreover, we experimentally illustrate the
limitation of the protocol that it is likely to fail or yields very low
fidelity, if the nonzero overlaps are approaching zero. Our experimental
implementation can be extended to more complex situations and other quantum
systems.

Quantum computers promise to outperform their classical counterparts in many
applications. Rapid experimental progress in the last two decades includes the
first demonstrations of smallscale quantum processors, but realising
largescale quantum information processors capable of universal quantum control
remains a challenge. One primary obstacle is the inadequacy of classical
computers for the task of optimising the experimental control field as we scale
up to large systems. Classical optimisation procedures require a simulation of
the quantum system and have a running time that grows exponentially with the
number of quantum bits (qubits) in the system. Here we show that it is possible
to tackle this problem by using the quantum processor to optimise its own
control fields. Using measurementbased quantum feedback control (MQFC), we
created a 12coherence state with the essential control pulse completely
designed by a 12qubit nuclear magnetic resonance (NMR) quantum processor. The
results demonstrate the superiority of MQFC over classical optimisation
methods, in both efficiency and accuracy. The time required for MQFC
optimisation is linear in the number of qubits, and our 12qubit system beat a
classical computer configured with 2.4 GHz CPU and 8 GB memory. Furthermore,
the fidelity of the MQFCprepared 12coherence was about 10% better than the
best result using classical optimisation, since the feedback approach
inherently corrects for unknown imperfections in the quantum processor. As MQFC
is readily transferrable to other technologies for universal quantum
information processors, we anticipate that this result will open the way to
scalably and precisely control quantum systems, bringing us closer to a
demonstration of quantum supremacy.

To exploit a given physical system for quantum information processing, it is
critical to understand the different types of noise affecting quantum control.
Distinguishing coherent and incoherent errors is extremely useful as they can
be reduced in different ways. Coherent errors are generally easier to reduce at
the hardware level, e.g. by improving calibration, whereas some sources of
incoherent errors, e.g. T2* processes, can be reduced by engineering robust
pulses. In this work, we illustrate how purity benchmarking and randomized
benchmarking can be used together to distinguish between coherent and
incoherent errors and to quantify the reduction in both of them due to using
optimal control pulses and accounting for the transfer function in an electron
spin resonance system. We also prove that purity benchmarking provides bounds
on the optimal fidelity and diamond norm that can be achieved by correcting the
coherent errors through improving calibration.

Spin systems controlled and probed by magnetic resonance have been valuable
for testing the ideas of quantum control and quantum error correction. This
paper introduces an Xband pulsed electron spin resonance spectrometer designed
for highfidelity coherent control of electron spins, including a loopgap
resonator for submillimeter sized samples with a control bandwidth ~ 40 MHz.
Universal control is achieved by a singlesideband upconversion technique with
an IQ modulator and a 1.2 GS/s arbitrary waveform generator. A single qubit
randomized benchmarking protocol quantifies the average errors of Clifford
gates implemented by simple Gaussian pulses, using a sample of gammairradiated
quartz. Improvements in unitary gate fidelity are achieved through phase
transient correction and hardware optimization. A preparation pulse sequence
that selects spin packets in a narrowed distribution of static fields confirms
that inhomogeneous dephasing (1/T2*) is the dominant source of gate error. The
best average fidelity over the Clifford gates obtained here is 99.2%, which
serves as a benchmark to compare with other technologies.

The ability to perform quantum error correction is a significant hurdle for
scalable quantum information processing. A key requirement for multipleround
quantum error correction is the ability to dynamically extract entropy from
ancilla qubits. Heatbath algorithmic cooling is a method that uses quantum
logic operations to move entropy from one subsystem to another, and permits
cooling of a spin qubit below the closed system (Shannon) bound.
Gammairradiated, $^{13}$Clabeled malonic acid provides up to 5 spin qubits: 1
spinhalf electron and 4 spinhalf nuclei. The nuclei are strongly hyperfine
coupled to the electron and can be controlled either by exploiting the
anisotropic part of the hyperfine interaction or by using pulsed
electronnuclear double resonance (ENDOR) techniques. The electron connects the
nuclei to a heatbath with a much colder effective temperature determined by
the electron's thermal spin polarization. By accurately determining the full
spin Hamiltonian and performing realistic algorithmic simulations, we show that
an experimental demonstration of heatbath algorithmic cooling beyond the
Shannon bound is feasible in both 3qubit and 5qubit variants of this spin
system. Similar techniques could be useful for polarizing nuclei in molecular
or crystalline systems that allow for nonequilibrium optical polarization of
the electron spin.

Application of multiple rounds of Quantum Error Correction (QEC) is an
essential milestone towards the construction of scalable quantum information
processing devices. However, experimental realizations of it are still in their
infancy. The requirements for multiple round QEC are high control fidelity and
the ability to extract entropy from ancilla qubits. Nuclear Magnetic Resonance
(NMR) based quantum devices have demonstrated high control fidelity with up to
12 qubits. On the other hand, the major challenge in the NMR QEC experiment is
to efficiently supply ancilla qubits in highly pure states at the beginning of
each round of QEC. Purification of qubits in NMR, or in other ensemble based
quantum systems can be accomplished through Heat Bath Algorithmic Cooling
(HBAC). It is an efficient method for extracting entropy from qubits that
interact with a heat bath, allowing cooling below the bath temperature. For
practical HBAC, coupled electronnuclear spin systems are more promising than
conventional NMR quantum processors, since electron spin polarization is about
$10^3$ times greater than that of a proton under the same experimental
conditions. We provide an overview on both theoretical and experimental aspects
of HBAC focusing on spin and magnetic resonance based systems, and discuss the
prospects of exploiting electronnuclear coupled systems for the realization of
HBAC and multiple round QEC.

Due to its geometric nature, holonomic quantum computation is faulttolerant
against certain types of control errors. Although proposed more than a decade
ago, the experimental realization of holonomic quantum computation is still an
open challenge. In this Letter, we report the first experimental demonstration
of nonadiabatic holonomic quantum computation in liquid NMR quantum information
processors. Two noncommuting holonomic singlequbit gates, rotations about
xaxis and about zaxis, and the twoqubit holonomic controlNOT gate are
realized with high fidelity by evolving the work qubits and an ancillary qubit
nonadiabatically. The successful realization of these universal elementary
gates in nonadiabatic quantum computing demonstrates the experimental
feasibility and the fascinating feature of this fast and resilient quantum
computing paradigm.

Anyons have exotic statistical properties, fractional statistics, differing
from Bosons and Fermions. They can be created as excitations of some
Hamiltonian models. Here we present an experimental demonstration of anyonic
fractional statistics by simulating a version of the Kitaev spin lattice model
proposed by Han et al[Phys. Rev.Lett. 98, 150404 (2007)] using an NMR quantum
information processor. We use a 7qubit system to prepare a 6qubit pseudopure
state to implement the ground state preparation and realize anyonic
manipulations, including creation, braiding and anyon fusion. A $\pi/2\times 2$
phase difference between the states with and without anyon braiding, which is
equivalent to two successive particle exchanges, is observed. This is different
from the $\pi\times 2$ and $2\pi \times 2$ phases for Fermions and Bosons after
two successive particle exchanges, and is consistent with the fractional
statistics of anyons.