Abstract
Noncommuting conserved quantities have recently launched a subfield of quantum thermodynamics. In conventional thermodynamics, a system of interest and an environment exchange quantities—energy, particles, electric charge, etc.—that are globally conserved and are represented by Hermitian operators. These operators were implicitly assumed to commute with each other, until a few years ago. Freeing the operators to fail to commute has enabled many theoretical discoveries—about reference frames, entropy production, resourcetheory models, etc. Little work has bridged these results from abstract theory to experimental reality. This paper provides a methodology for building this bridge systematically: we present a prescription for constructing Hamiltonians that conserve noncommuting quantities globally while transporting the quantities locally. The Hamiltonians can couple arbitrarily many subsystems together and can be integrable or nonintegrable. Our Hamiltonians may be realized physically with superconducting qudits, with ultracold atoms, and with trapped ions.
Similar content being viewed by others
Introduction
One of thermodynamics’ most fundamental and ubiquitous interactions is the exchange of quantities between a system of interest and an environment. Example quantities include energy, particles, and electric charge. As the quantities are conserved globally, we call them ‘charges.’ (We call even the local quantities ‘charges’ for convenience, even though the quantities are not conserved locally). Such exchanges happen, for example, in electrochemical batteries, in a cooling cup of coffee, and when a few spins flip to align with a magnetic field. Given such exchanges’ pervasiveness, studying their quantum facets is essential for (i) developing the field of quantum thermodynamics^{1,2} and (ii) discovering nonclassical features of quantum manybody thermalization in condensed matter; atomic, molecular and optical (AMO) physics; highenergy physics; and chemistry. One important quantum phenomenon is operators’ failure to commute with each other: Noncommutation underlies uncertainty relations, measurement disturbance, and more. Therefore, studying exchanges of noncommuting charges is crucial for understanding quantum thermodynamics. As a result, noncommuting charges have been enjoying a heyday^{3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29} in quantuminformationtheoretic (QIT) thermodynamics.
Lifting the assumption that exchanged charges commute^{3,4,5,13,14,27,30,31} has led to discoveries of truly quantum thermodynamics. Example discoveries include a generalization of the microcanonical state^{5}, resource theories^{3,4,5,6,7,8,32}, a generalization of the majorization preorder^{9}, a reduction of entropy production by charges’ noncommutation^{10}, and referenceframe designs^{11,12}. These discoveries and others have turned noncommuting thermodynamic charges into a growing subfield.
Most of the discoveries have, until recently, belonged to QIT thermodynamics. However, given their fundamental and nonclassical nature, exchanges of thermodynamic noncommuting charges call for bridges to experiments and to manybody physics. Building these bridges requires Hamiltonians that transport noncommuting observables locally while conserving them globally: as stated in the quantumthermodynamics review^{1}, ‘an abstract view of dynamics, minimal in the details of Hamiltonians, is often employed in quantum information’ and so in QIT thermodynamics. In contrast, experiments, simulations, and manybody theory require microscopic Hamiltonians.
Before the present work, it was unknown (i) whether Hamiltonians that transport noncommuting observables locally, while conserving them globally, exist; (ii) how such Hamiltonians look, if they exist; (iii) how to construct such Hamiltonians for given noncommuting charges; and (iv) for which charges such Hamiltonians can be constructed. We answer these questions, enabling the systemandenvironment exchange of noncommuting charges to progress from its QITthermodynamic birthplace to manybody physics and experiments. Example predictions that merit experimental exploration include (i) the emergence of the quantum equilibrium state in refs. ^{3,4,5}, (ii) the decrease in entropy production by noncommuting charges^{10}, (iii) applications of the entropy decrease to quantum engines^{33}, (iv) the conjecture that noncommuting charges hinder thermalization^{14}, and (v) the conjecture’s application to quantum memories. We open the door to experiments by prescribing how to construct the needed Hamiltonians. Our construction also enables the generalization, to noncommuting charges, of manybody–thermalization tools in condensed matter, AMO physics, and highenergy. Examples include the eigenstate thermalization hypothesis, outoftimeordered correlators, and random unitary circuits (e.g., refs. ^{34,35,36,37,38,39,40,41,42}).
This paper introduces a prescription for constructing Hamiltonians that overtly move noncommuting charges between subsystems while conserving the charges globally. The charges form a finitedimensional semisimple complex Lie algebra. The Hamiltonians can couple arbitrarily many subsystems together and can be integrable or nonintegrable. The prescription also produces a convenient basis for the algebra—a basis of charges explicitly transported locally, and conserved globally, by the Hamiltonian. The prescription is general, being independent of any physical platforms. Consequently, the Hamiltonians can be realized with diverse physical systems, such as superconducting circuits, ultracold atoms, and trapped ions.
In a special case, the charges form the Lie algebra \({\mathfrak{su}}(D)\), N identical subsystems form the global system, and each subsystem corresponds to the Hilbert space \({{\mathbb{C}}}^{D}\). In this example the SchurWeyl duality describes the Hamiltonians’ forms^{43,44}: Let the global system (formed from the system of interest and the environment) be many copies of the system of interest. The Hamiltonians are the linear combinations of the permutations of the copies. (Hamiltonians have also been engineered to have SU(D) symmetry without regard to whether noncommuting charges are transported^{45,46}). Our results are more general than the SchurWeyl duality and elucidate the dynamics’ physical interpretation. First, our prescription governs a much wider class of algebras: all finitedimensional, semisimple Lie algebras in which the Killing form induces a metric. Many physically significant algebras satisfy these assumptions—for example, the simple Lie algebras, which include \({\mathfrak{su}}(D)\). Second, our results are not restricted to systems whose Hilbert spaces are \({{\mathbb{C}}}^{D}\). Finally, the Hamiltonian form specified by the SchurWeyl duality—a linear combination of permutations—is an abstract construct. How to implement an arbitrary linear combination of permutations is not obvious. In contrast, our Hamiltonians have a clear physical interpretation, manifestly transporting noncommuting charges between subsystems. To our knowledge, no other class of Hamiltonians that transport charges locally and conserve them globally, comparably general to our class, is known.
This paper begins with our setup, detailed in the below section. The section ‘Pedagogical explanation’ introduces the Hamiltonianconstruction prescription pedagogically. We also review mathematical background and illustrate the prescription with an example familiar in quantum information, the Lie algebra \({\mathfrak{su}}(2)\). The section ‘Prescription for constructing the Hamiltonians’ synopsizes the prescription, crystallizing the main result, and presents two properties of the prescription. A richer example provides intuition in the section ‘su(3) example’: Hamiltonians that transport and conserve charges in the Lie algebra \({\mathfrak{su}}(3)\). The section ‘Discussion’ concludes with potential realizations of our Hamiltonians in condensed matter, AMO, and highenergy and nuclear physics.
Results
Setup
Consider a global closed quantum manybody system, as in recent thermalization experiments^{47,48,49,50,51,52,53,54,55,56,57,58}. As in conventional statistical mechanics, the global system is an ensemble of N identical subsystems. (We use the term ‘ensemble’ in the traditional sense of statistical physics: a collection of many identical copies of a system of interest. Such ensembles are often invoked to determine equilibrium probability distributions^{59}, p. 62). A few of the subsystems form the system of interest; and the rest, an effective environment. Each subsystem corresponds to a Hilbert space \({{{\mathcal{H}}}}\) of finite dimensionality d.
We will construct global Hamiltonians, H^{tot}, that conserve extensive charges defined as follows. Let Q_{α} denote a Hermitian operator defined on \({{{\mathcal{H}}}}\). We denote by \({Q}_{\alpha }^{(j)}\) the observable defined on the j^{th} subsystem’s \({{{\mathcal{H}}}}\). Each global observable
will be conserved by design:
Although the local \({Q}_{\alpha }^{(j)}\) are not conserved, we will sometimes call them, and the Q_{α}, ‘charges’ for convenience. One might know, initially, of only \(c^{\prime}\) charges’ existence.
These \(c^{\prime}\)Q_{α}’s generate a complex Lie algebra \({{{\mathcal{A}}}}\), which we assume to be finitedimensional. \({{{\mathcal{A}}}}\) consists of all the charges (as well as nonHermitian operators, which we ignore). Lie algebras describe many conserved physical quantities: particle number, angular momentum, electric charge, color charge, weak isospin, and our spacetime’s metric^{44,60,61}. We focus on nonAbelian Lie algebras, motivated by quantum thermodynamics that highlights noncommutation: the commutator exemplifies the Lie bracket, [Q_{α}, Q_{β}].
We assume four more properties of the algebra, to facilitate our proofs. \({{{\mathcal{A}}}}\) is finitedimensional and semisimple. Representing an observable, \({{{\mathcal{A}}}}\) is over the complex numbers. Also, on \({{{\mathcal{A}}}}\) is defined a Killing form (reviewed below) that induces a metric. Many physically significant algebras satisfy these assumptions—for example, the simple Lie algebras (see the Supplementary Note 1 and refs. ^{44,60,61}).
Pedagogical explanation
This section describes the prescription for constructing Hamiltonians H^{tot} that conserve noncommuting charges globally [Eq. (2)] while transporting them locally:
for some site j. (In every such commutator throughout this paper, one argument implicitly contains tensor factors of \({\mathbb{1}}\), so that both arguments operate on the same Hilbert space.) We construct twobody interaction terms, then combine them into manybody terms. This explanation provides a pedagogical introduction; the prescription is synopsized in the section ‘Prescription for constructing the Hamiltonians’. Here, we illustrate each step with an algebra familiar in quantum information, \({\mathfrak{su}}(2)\), which describes spin1/2 angular momentum.
Table 1 lists the simple Lie algebras. Every Cartesian product of simple Lie algebras yields a semisimple Lie algebra \({{{\mathcal{A}}}}\). Such an algebra generates a semisimple Lie group \({{{\mathcal{G}}}}\). For example, if \({{{\mathcal{A}}}}\) consists of angular momentum, \({{{\mathcal{A}}}}={\mathfrak{su}}(D)\). The corresponding \({{{\mathcal{G}}}}\) consists of rotations: \({{{\mathcal{G}}}}=\,{{\mbox{SU}}}\,(D)\).
An algebra has two relevant properties, a dimension and a rank (Table 1). The dimension, c, equals the number of generators in a basis for the algebra.(We chose the notation c to evoke the c introduced in^{14}. There, c was defined as the number of charges. As explained in the section ‘Setup’ under Results, those charges would form a Lie algebra. Infinitely many charges would therefore exist, the c in^{14} would equal infinity, and results in^{14} would be impractical. We therefore define c as the Lie algebra’s finite dimension.) For example, \({\mathfrak{su}}\)(2) has the Paulioperator basis {σ_{x}, σ_{y}, σ_{z}} and so has a dimension c = 3. The rank, r, has a significance that we will encounter shortly.
A representation of \({{{\mathcal{A}}}}\) is a Liebracketpreserving map from \({{{\mathcal{A}}}}\) to a set of linear transformations. The adjoint representation maps from \({{{\mathcal{A}}}}\) to linear transformations defined on \({{{\mathcal{A}}}}\). If \(x\in {{{\mathcal{A}}}}\), the adjoint representation ad(x) acts on \(y\in {{{\mathcal{A}}}}\) as ad(x)(y) : = [x, y]. The adjoint representation features in the Killing form, which we review now. The definition of \({{{\mathcal{A}}}}\) involves a vector space V defined over a field F. A map V × V → F is a form. The Killing form is the symmetric bilinear form
We say that x and y are Killingorthogonal if (x, y) = 0. We say that subalgebras \({{{{\mathcal{A}}}}}_{1}\) and \({{{{\mathcal{A}}}}}_{2}\) are Killingorthogonal if, for all \(x\in {{{{\mathcal{A}}}}}_{1}\) and \(y\in {{{{\mathcal{A}}}}}_{2}\), (x, y) = 0. We will use the Killing form to construct the preferred basis of charges for \({{{\mathcal{A}}}}\).
Our construction begins with another basis: every finitedimensional semisimple complex Lie algebra \({{{\mathcal{A}}}}\) has a Cartan–Weyl basis. In fact, \({{{\mathcal{A}}}}\) has infinitely many. Convention may distinguish one CartanWeyl basis. We use the conventional \({\mathfrak{su}}(2)\) basis for concreteness. We use this basis, in our example, for concreteness. In general, one selects an arbitrary CartanWeyl basis. The basis contains generators of two types: Hermitian operators and ladder operators.
The number of Hermitian operators is the algebra’s rank, r. These operators commute with each other. If r > 1, we rescale the operators to endow them with unit HilbertSchmidt norms:
We include these operators, Q_{α=1,2, …, r}, in our preferred basis. In the \({\mathfrak{su}}\)(2) example, r = 1; and Q_{1} = σ_{z}, whose eigenstates \(\left\pm z\right\rangle\) correspond to the eigenvalues ±1. The Q_{α}’s generate a subalgebra, a Cartan subalgebra.
The CartanWeyl basis contains, as well as Hermitian operators, ladder operators. They form pairs L_{±β}, for \(\beta =1,2,\ldots ,\frac{cr}{2}\): Since the CartanWeyl basis has c elements, and r of them are Hermitian, there are c − r ladder operators. Each β corresponds to two ladder operators, one raising (+β) and one lowering (−β). Hence β runs from 1 to \(\frac{cr}{2}\). Each L_{±β} raises or lowers at least one Q_{α}. In the \({\mathfrak{su}}\)(2) example, the ladder operators \({\sigma }_{\pm z}=\frac{1}{2}({\sigma }_{x}\pm i{\sigma }_{y})\) raise and lower σ_{z}: \({L}_{\pm z}\left\mp z\right\rangle =\left\pm z\right\rangle\). In other algebras, an L_{±β} can raise and/or lower multiple Q_{α}’s. Examples include \({\mathfrak{su}}(3)\) (explained in the section ‘su(3) example’).
From each ladderoperator pair, we construct an interaction that couples subsystems j and \(j^{\prime}\). Let \({J}_{\beta }^{(j,j^{\prime} )}\) denote a hopping frequency. An interaction that transports all the charges between j and \(j^{\prime}\), while conserving each charge globally, has the form
We assemble the other terms in \({H}^{(j,j^{\prime} )}\) from other CartanWeyl bases, constructed as follows. Let U denote a general element of the group \({{{\mathcal{G}}}}\). We conjugate, with U, each element of our first CartanWeyl basis: For α = 1, 2, …, r and \(\beta =1,2,\ldots ,\frac{cr}{2}\),
We include the new Q_{α}’s (for which α = r + 1, r + 2, …, 2r) in our preferred basis for the algebra.
We constrain U such that each new Q_{α} is Killingorthogonal to (i) each other new charge Q_{β} and (ii) each original charge Q_{γ}:
for all α, β = r + 1, r + 2, …, 2r and all γ = 1, 2, …, r. This orthogonality restricts U, though not completely. The new Q_{α}’s generate a Cartan subalgebra Killingorthogonal to the original Cartan subalgebra. The new ladder operators contribute to the interaction:
In the \({\mathfrak{su}}\)(2) example, U can be represented by \(\left[\begin{array}{ll}a&{b}^{* }\\ b&{a}^{* }\end{array}\right]\ ,\) wherein \(a,b\in {\mathbb{C}}\) and ∣a∣^{2} + ∣b∣^{2} = 1. The prescription restricts U only via the Killingorthogonality of U^{†}σ_{z}U to U. We enforce only this restriction in the Supplementary Note 2. Here, we choose a U for pedagogical simplicitly: \(U=({\mathbb{1}}+i{\sigma }_{y})/\sqrt{2}\), such that Q_{α+r} = Q_{2} = σ_{x}. The new ladder operators, \({\sigma }_{\pm x}:=\left(\frac{{\mathbb{1}}+i{\sigma }_{y}}{\sqrt{2}}\right){\sigma }_{\pm z}\left(\frac{{\mathbb{1}}+i{\sigma }_{y}}{\sqrt{2}}\right)\), create and annihilate quanta of the xcomponent of the angular momentum. The interaction becomes
We repeat the foregoing steps: write out the form of a general \(U\in {{{\mathcal{G}}}}\). Conjugate each element of the original CartanWeyl basis with U. Constrain U such that the new Q_{α}’s are orthogonal to each other and to the older Q_{α}’s. Include the new Q_{α}’s in our preferred basis for the algebra. Form a term, in \({H}^{(j,j^{\prime} )}\), from the new ladder operators L_{±β}.
Each CartanWeyl basis contributes r elements Q_{α} to the preferred basis. The basis contains c elements, so we form c/r mutually orthogonal CartanWeyl bases. c/r equals an integer for the finitedimensional semisimple complex Lie algebras, according to Proposition 1 in the section ‘Prescription for constructing the Hamiltonians’. Table 1 confirms the claim for the simple Lie algebras. Our algebra’s finite dimensionality ensures that our prescription halts. The twobody interaction is now
Why is the preferred basis {Q_{α}} preferable? First, the basis endows the Hamiltonian with a simple physical interpretation: \({H}^{(j,j^{\prime} )}\) transports all these charges locally while conserving them globally. Second, the basis is (Killing)orthogonal.
In the \({\mathfrak{su}}\)(2) example, c/r = 3/1 = 3. Hence we construct three CartanWeyl bases, using two SU(2) elements. If the first unitary was \(({\mathbb{1}}+i{\sigma }_{y})/\sqrt{2}\), the second unitary is \(({\mathbb{1}}i{\sigma }_{x}+i{\sigma }_{y}+i{\sigma }_{z})/2\), to within a global phase. Consequently, Q_{3} = σ_{y}, the preferred basis for \({{{\mathcal{A}}}}\) is {σ_{z}, σ_{x}, σ_{y}}, and
Next, we constrain the interaction to conserve every global charge:
The commutation relations (14) constrain the hopping frequencies \({J}_{\alpha }^{(j,j^{\prime} )}\). The frequencies must equal each other in the \({\mathfrak{su}}(2)\) example: \({J}_{\alpha }^{(j,j^{\prime} )}\equiv {J}^{(j,j^{\prime} )}\) for all α. The Hamiltonian simplifies to^{19}
This Heisenberg model is known to have SU(2) symmetry and so to conserve each global spin component \({\sigma }_{\alpha }^{{{{\rm{tot}}}}}:=\mathop{\sum }\nolimits_{j = 1}^{N}{\sigma }_{\alpha }^{(j)}\). But the Hamiltonian is typically written in the dotproduct form (15), as
or in the zbiased form \({H}^{(j,j^{\prime} )}\propto 2({\sigma }_{+z}^{(j)}{\sigma }_{z}^{(j^{\prime} )}+{\sigma }_{z}^{(j)}{\sigma }_{+z}^{(j^{\prime} )})+{\sigma }_{z}^{(j)}{\sigma }_{z}^{(j^{\prime} )}\). None of these three forms reveals that the Heisenberg model transports noncommuting charges between subsystems. Our expression (13) and our prescription do. In relativistic field theories, making the action manifestly Lorentzinvariant is worthwhile; analogously, making the Hamiltonian manifestly transport noncommuting charges locally, while conserving them globally, is worthwhile. Furthermore, our prescription constructs Hamiltonians that overtly transport noncommuting charges locally and conserve the charges globally not only in this simple \({\mathfrak{su}}(2)\) example, but also for all finitedimensional semisimple complex Lie algebras on which the Killing form induces a metric—including algebras for which this prescription does not produce the Heisenberg Hamiltonian. Supplementary Note 3 discusses a generalization of the simple form (15).
We have constructed a twobody interaction \({H}^{(j,j^{\prime} )}\) that couples subsystems j and \(j^{\prime}\). We construct kbody terms \({H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(k)}\right)}\) by multiplying twobody terms (12) together, constraining the couplings such that \([{H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(k)}\right)},{Q}_{\alpha }^{{{{\rm{tot}}}}}]=0\), and subtracting off any fewerbody terms that appear in the product. Section II C details the formalism. In the \({\mathfrak{su}}(2)\) example, a threebody interaction has the form (see Supplementary Note 2)
wherein \({J}^{(j,j^{\prime} ,j^{\prime\prime} )}\in {\mathbb{R}}\).
The Hamiltonian we constructed may be integrable. For example, the onedimensional (1D) nearestneighbor Heisenberg model is integrable^{62}. Integrable Hamiltonians have featured in studies of noncommuting charges in thermodynamics^{21}. But one might wish for the system to thermalize as much as possible, as is promoted by nonintegrability^{37,63}. Geometrically nonlocal couplings, manybody interactions, and multidimensional lattices tend to break integrability. Hence one can add terms \({H}^{(j,j^{\prime} )}\) and \({H}^{(j,j^{\prime} ,\ldots ,{j}^{(k)})}\) to the global Hamiltonian H^{tot}, and keep growing the lattice’s dimensionality, until H^{tot} becomes nonintegrable. Nonintegrability may be diagnosed with, e.g., energygap statistics^{37}. In the \({\mathfrak{su}}(2)\) example, one can break integrability by creating nextnearestneighbor couplings or by making the global system twodimensional^{19}.
Prescription for constructing the Hamiltonians
Here, we synopsize the prescription elaborated on in the section ‘Pedagogical explanation’. Then, we present two results pertinent to the prescription. We construct, as follows, Hamiltonians that transport noncommuting charges locally and conserve the charges globally:

1.
Identify an arbitrary CartanWeyl basis for the algebra, \({{{\mathcal{A}}}}\).

2.
The CartanWeyl basis contains r Hermitian operators that commute with each other. Scale each such operator such that it has a unit HilbertSchmidt norm [Eq. (5)]. Label the results Q_{α=1,2, …, r}. Include them in the preferred basis for the algebra.

3.
The other CartanWeylbasis elements are ladder operators that form raisingandlowering pairs: L_{±β}, for β = 1, 2, …c − r. From each pair, form one term in the twobody interaction, \({H}^{(j,j^{\prime} )}\) [Eq. (6)].

4.
Write out the form of the most general element \(U\in {{{\mathcal{G}}}}\) of the Lie group \({{{\mathcal{G}}}}\) generated by \({{{\mathcal{A}}}}\). Conjugate each charge Q_{α} and each ladder operator L_{±β} with U [Eq. (7)]. The new charges and new ladder operators, together, form another Cartan–Weyl basis.

5.
Constrain U such that every new charge Q_{α} is Killingorthogonal to (i) each other new charge and (ii) each charge already in the basis [Eq. (9)].

6.
Include each new Q_{α} in the basis for \({{{\mathcal{A}}}}\).

7.
From each new pair L_{±β} of ladder operators, form a term in the twobody interaction \({H}^{(j,j^{\prime} )}\) [Eq. (10)].

8.
Repeat steps 47 until having identified c/r CartanWeyl bases, wherein c denotes the algebra’s dimension. Each CartanWeyl basis contributes r elements Q_{α} to the preferred basis for \({{{\mathcal{A}}}}\). The basis is complete, containing \(r\cdot \frac{c}{r}=c\) elements.

9.
Constrain the twobody interaction to conserve each global charge [Eq. (14)], for all α = 1, 2, …, c. Solve for the frequencies \({J}_{\beta }^{(j,j^{\prime} )}\) that satisfy this constraint.

10.
If a kbody interaction is desired, for any k > 2: Perform the following substeps for ℓ = 3, 4, …, k: Multiply together ℓ unconstrained twobody interactions (12) cyclically:
$${H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(\ell )}\right)}={H}^{(j,j^{\prime} )}{H}^{(j^{\prime} ,j^{\prime\prime} )}\ldots {H}^{\left({j}^{(\ell 1)},{j}^{(\ell )}\right)} \times {H}^{({j}^{(\ell )},j)}.$$(19)Constrain the couplings so that \([{H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(\ell )}\right)},{Q}_{\alpha }^{{{{\rm{tot}}}}}]=0\) for all α. If \({H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(\ell )}\right)}\) contains fewerbody terms that conserve all the \({Q}_{\alpha }^{{{{\rm{tot}}}}}\), subtract those terms off.

11.
Sum the accumulated interactions \({H}^{\left(j,j^{\prime} ,\ldots ,{j}^{(k)}\right)}\) over the subsystems \(j,j^{\prime} ,\ldots\) to form H^{tot}.

12.
If H^{tot} is to be nonintegrable, add longerrange interactions and/or largekkbody interactions until breaking integrability, as signaled by, e.g., energygap statistics.
Having synopsized our prescription, we present two properties of it. The first property ensures that the prescription runs for an integer number of iterations (step 8).
Proposition 1
Consider any finitedimensional semisimple complex Lie algebra. The algebra’s dimension, c, and rank, r, form an integer ratio: \(c/r\in {{\mathbb{Z}}}_{ \,{ > }\,0}\).
We prove this proposition in the Supplementary Note 4. The second property characterizes the prescription’s output.
Theorem 1
The charges Q_{1}, Q_{2}, …, Q_{c} produced by the prescription form a basis for the algebra \({{{\mathcal{A}}}}\).
Proof
The charges are Killingorthogonal by construction: (Q_{α}, Q_{β}) = 0 for all α, β. The Killing form induces a metric on \({{{\mathcal{A}}}}\) by assumption. Therefore, the Q_{α} are linearly independent according to this metric.
The prescription produces c charges (step 8). c denotes the algebra’s dimension, the number of elements in each basis for \({{{\mathcal{A}}}}\). Hence every linearly independent set of c\({{{\mathcal{A}}}}\) elements forms a basis for \({{{\mathcal{A}}}}\). Hence the Q_{α} form a basis.
\(\boldsymbol{\mathfrak{su}}(\mathbf{3})\) example
Section II B illustrated the Hamiltonianconstruction prescription with the algebra \({\mathfrak{su}}(2)\). The \({\mathfrak{su}}(2)\) example offered simplicity but lacks other algebras’ richness: In other algebras, each Cartan–Weyl basis contains multiple Hermitian operators and multiple ladderoperator pairs. We demonstrate how our prescription accommodates this richness, by constructing a twobody Hamiltonian that transports \({\mathfrak{su}}(3)\) elements locally while conserving them globally. Such Hamiltonians may be engineered for superconducting qutrits, as sketched in Discussion. However, this \({\mathfrak{su}}(3)\) example only illustrates our more general prescription, which works for all finitedimensional semisimple complex Lie algebras on which the Killing form induces a metric.
Each basis for \({\mathfrak{su}}(3)\) contains c = 8 elements. The most famous basis consists of the Gellmann matrices, λ_{k=1,2, …, 8}^{64}. The λ_{k} generalize the Pauli matrices in certain ways, being traceless and Killingorthogonal. From the Gellmann matrices is constructed the conventional CartanWeyl basis^{65}, reviewed in the Supplementary Note 5. The r = 2 Hermitian elements are Gellmann matrices:
Q_{1} and Q_{2} belong in the preferred basis of charges for \({\mathfrak{su}}(3)\). For pedagogical clarity, we will identify all the charges before addressing the ladder operators.
A general element U ∈ SU(3) contains eight real parameters. In the Euler parameterization^{66},
The parameters ϕ_{1}, ϕ_{3}, ϕ_{5}, ϕ_{7} ∈ [0, 2π); ϕ_{2}, ϕ_{4}, ϕ_{6} ∈ [0, π]; and \({\phi }_{8}\in [0,2\sqrt{3}\pi )\). We now constrain U, identifying the instances U_{i} that map the first charges to \({Q}_{3}={U}_{{{{\rm{i}}}}}^{{\dagger} }{Q}_{1}{U}_{{{{\rm{i}}}}}\) and \({Q}_{4}={U}_{{{{\rm{ii}}}}}^{{\dagger} }{Q}_{2}{U}_{{{{\rm{ii}}}}}\) that are Killingorthogonal to each other and to the original charges. Supplementary Note 5 contains the details. We label with a superscript (i) the parameters used to fix U_{i}: \({\phi }_{1}^{({{{\rm{i}}}})}\), \({\phi }_{3}^{({{{\rm{i}}}})}\), \({\phi }_{7}^{({{{\rm{i}}}})}\), \({\phi }_{8}^{({{{\rm{i}}}})}\), and n^{(i)}. For convenience, we package several parameters together: \({a}^{({{{\rm{i}}}})}:=\frac{1}{2}\left({\phi }_{3}^{({{{\rm{i}}}})}{\phi }_{7}^{({{{\rm{i}}}})}\sqrt{3}{\phi }_{8}^{({{{\rm{i}}}})}+\pi {n}^{({{{\rm{i}}}})}+\frac{\pi }{2}\right)\), and \({b}^{({{{\rm{i}}}})}:={a}^{({{{\rm{i}}}})}+{\phi }_{7}^{({{{\rm{i}}}})}\). In terms of these parameters, the new charges have the forms (see Supplementary Note 5)
Q_{3} has the same form as Q_{5} and Q_{7}, which satisfy the same Killingorthogonality conditions. Similarly, Q_{4} has the same form as Q_{6} and Q_{8}. The later charges’ parameters a^{(ℓ)} and b^{(ℓ)} are more restricted, however (see Supplementary Note 5). We have identified our preferred basis of charges.
Let us construct the ladder operators and Hamiltonian. Each Cartan–Weyl basis contains c − r = 8 − 2 = 6 ladder operators. The conventional Cartan–Weyl basis contains ladder operators formed from Gellman matrices:
Transforming these operators with unitaries U_{ii,iii,iv} yields L_{±4} through L_{±12}, whose forms appear in the Supplementary Note 5. From each ladder operator, we form one term in the twobody Hamiltonian (6).
Finally, we determine the hopping frequencies \({J}_{\alpha }^{(j,j^{\prime} )}\), demanding that \([{H}^{(j,j^{\prime} )},{Q}_{\alpha }^{{{{\rm{tot}}}}}]=0\) for all α. For all possible values of the a^{(ℓ)}, b^{(ℓ)}, and n^{(ℓ)}, if all the frequencies are nonzero, then all the frequencies equal each other. We set \({J}_{\alpha }^{(j,j^{\prime} )}\equiv \frac{4}{3}\ {J}^{(j,j^{\prime} )}\), such that
The Hamiltonian collapses to a simple form analogous to the \({\mathfrak{su}}(2)\) example’s Eq. (16) (see Supplementary Note 3).
Discussion
We have presented a prescription for constructing Hamiltonians that transport noncommuting charges locally while conserving the charges globally. The Hamiltonians can couple arbitrarily many subsystems together and can be integrable or nonintegrable. The prescription produces, as well as Hamiltonians, preferred bases of charges that are (i) overtly transported locally and conserved globally and (ii) Killingformorthogonal. This construction works whenever the charges form a finitedimensional semisimple complex Lie algebra on which the Killing form induces a metric. Whether there exists any Hamiltonians that transport charges locally, while conserving the charges globally, outside of those found by our prescription, is an interesting open question for theoretical exploration.
This work provides a systematic means of bridging noncommuting thermodynamic charges from abstract quantum information theory to condensed matter, AMO physics, and highenergy and nuclear physics. The mathematical results that have accrued^{3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25} can now be tested experimentally, via our construction. This paper’s introduction highlights example results in that merit testing. Such experiments’ benefits include the simulation of quantum systems larger than what classical computers can simulate, the uncovering of behaviors not predicted by theory, and the grounding of abstract QIT thermodynamics in physical reality.
In addition to harnessing controlled platforms to study noncommuting charges’ quantum thermodynamics, one may leverage that quantum thermodynamics to illuminate highenergy and nuclear physics. Such physics includes nonAbelian gauge theories, such as quantum chromodynamics. How to define and measure such theories’ thermalization is unclear^{67}. One might gain insights by using our dynamics as a bridge from quantum thermodynamics to nonAbelian field theories.
As mentioned above, the Heisenberg model (13) can be implemented with ultracold atoms and trapped ions^{68,69,70,71,72,73}. Reference^{19} details how to harness these setups to study noncommuting thermodynamic charges. We introduce a more complex example here: We illustrate, with superconducting qubits, how today’s experimental platforms can implement the \({\mathfrak{su}}(3)\) instance of our general prescription.
Superconducting circuits can serve as qudits with Hilbertspace dimensionalities d ≥ 2^{74}. Qutrits have been realized with transmons, slightly anharmonic oscillators^{75}. The lowest two energy levels often serve as a qubit, but the second energy gap nearly equals the first. Hence the third level can be addressed relatively easily^{76}. Superconducting qutrits offer a tabletop platform for transporting and conserving \({\mathfrak{su}}(3)\) charges as in the section ‘su(3) example’.
Experiments with ≤5 qutrits have been run^{77,78}, Furthermore, many of the tools used to control and measure superconducting qubits can be applied to qutrits^{76,79,80,81,82,83,84,85,86,87,88}. A noncommutingchargesinthermodynamics experiment may begin with preparing the qutrits in an approximate microcanonical subspace, a generalization of the microcanonical subspace that accommodates noncommuting charges^{14}. Such a state preparation may be achieved with weak measurements^{19}, which have been performed on superconducting qudits through cavity quantum electrodynamics^{89}.
\({T}_{2}^{* }\) relaxation times of ~39 μs, for the lowest energy gap, and ~14 μs, for the secondlowest gap, have been achieved^{78}. Meanwhile, twoqutrit gates can be realized in ~10 − 10^{2} ns^{78,90,91}. Some constant number of such gates may implement one threelevel gate that simulates a term in our Hamiltonian. If the number is order10, information should be able to traverse an 8qutrit system ~10 times before the qutrits decohere detrimentally. According to numerics in Yunger et al.^{14}, a small subsystem nears thermalization once information has had time to traverse the global system a number of times linear in N. Therefore, realizations of our Hamiltonians are expected to thermalize the system internally. The states of small subsystems, such as qutrit pairs, can be read out via quantum state tomography^{76,79,80,81,82}. Hence superconducting qutrits, and other platforms, can import noncommuting charges from quantum thermodynamics to manybody physics, by simulating the Hamiltonians constructed here.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Vinjanampathy, S. & Anders, J. Quantum thermodynamics. Contemp. Phys. 57, 545–579 (2016).
Goold, J., Huber, M., Riera, A., del Rio, L. & Skrzypczyk, P. The role of quantum information in thermodynamics — a topical review. J. Phys. A Math. 49, 143001 (2016).
Lostaglio, M., Jennings, D. & Rudolph, T. Thermodynamic resource theories, noncommutativity and maximum entropy principles. New J. Phys. 19, 043008 (2017).
Guryanova, Y., Popescu, S., Short, A. J., Silva, R. & Skrzypczyk, P. Thermodynamics of quantum systems with multiple conserved quantities. Nat. Commun. 7, 12049 (2016).
Yunger Halpern, N. Beyond heat baths ii: framework for generalized thermodynamic resource theories. J. Phys. A Math. 51, 094001 (2018).
Sparaciari, C., Del Rio, L., Scandolo, C. M., Faist, P. & Oppenheim, J. The first law of general quantum resource theories. Quantum 4, 259 (2020).
Khanian, Z. B. From quantum source compression to quantum thermodynamics. Preprint at https://arxiv.org/abs/2012.14143 (2020).
Khanian, Z. B., Bera, M. N., Riera, A., Lewenstein, M. & Winter, A. Resource theory of heat and work with noncommuting charges: yet another new foundation of thermodynamics. Preprint at https://arxiv.org/abs/2011.08020 (2020).
Gour, G., Jennings, D., Buscemi, F., Duan, R. & Marvian, I. Quantum majorization and a complete set of entropic conditions for quantum thermodynamics. Nat. Commun. 9, 5352 (2018).
Manzano, G., Parrondo, J. M. & Landi, G. T. Nonabelian quantum transport and thermosqueezing effects. Preprint at https://arxiv.org/abs/2011.04560 (2020).
Popescu, S., Sainz, A. B., Short, A. J. & Winter, A. Quantum reference frames and their applications to thermodynamics. Philos. Trans. R. Soc. A 376, 20180111 (2018).
Popescu, S., Sainz, A. B., Short, A. J. & Winter, A. Reference frames which separately store noncommuting conserved quantities. Phys. Rev. Lett. 125, 090601 (2020).
Lostaglio, M. The Resource Theory of Quantum Thermodynamics. Master’s thesis (Imperial College London, 2014).
Yunger Halpern, N., Faist, P., Oppenheim, J. & Winter, A. Microcanonical and resourcetheoretic derivations of the thermal state of a quantum system with noncommuting charges. Nat. Commun. 7, 12051 (2016).
Ito, K. & Hayashi, M. Optimal performance of generalized heat engines with finitesize baths of arbitrary multiple conserved quantities beyond independentandidenticaldistribution scaling. Phys. Rev. E 97, 012129 (2018).
Bera, M. N., Riera, A., Lewenstein, M., Khanian, Z. B. & Winter, A. Thermodynamics as a consequence of information conservation. Quantum 3, 121 (2019).
MurPetit, J., Relaño, A., Molina, R. A. & Jaksch, D. Revealing missing charges with generalised quantum fluctuation relations. Nat. Commun. 9, 2006 (2018).
Manzano, G. Squeezed thermal reservoir as a generalized equilibrium reservoir. Phys. Rev. E 98, 042123 (2018).
Yunger Halpern, N., Beverland, M. E. & Kalev, A. Noncommuting conserved charges in quantum manybody thermalization. Phys. Rev. E 101, 042117 (2020).
Manzano, G. et al. Hybrid thermal machines: generalized thermodynamic resources for multitasking. Phys. Rev. Res. 2, 043302 (2020).
Fukai, K., Nozawa, Y., Kawahara, K. & Ikeda, T. N. Noncommutative generalized gibbs ensemble in isolated integrable quantum systems. Phys. Rev. Res. 2, 033403 (2020).
MurPetit, J., Relaño, A., Molina, R. A. & Jaksch, D. Fluctuations of work in realistic equilibrium states of quantum systems with conserved quantities. SciPost Phys. Proc. 3, 024 (2020).
Scandi, M. & PerarnauLlobet, M. Thermodynamic length in open quantum systems. Quantum 3, 197 (2019).
Boes, P., Wilming, H., Eisert, J. & Gallego, R. Statistical ensembles without typicality. Nat. Commun. 9, 1–9 (2018).
Mitsuhashi, Y., Kaneko, K. & Sagawa, T. Characterizing symmetryprotected thermal equilibrium by work extraction. Preprint at https://arxiv.org/abs/2103.06060 (2021).
Croucher, T., Wright, J., Carvalho, A. R. R., Barnett, S. M. & Vaccaro, J. A. Information Erasure, 713–730 (Springer International Publishing, 2018).
Vaccaro, J. A. & Barnett, S. M. Information erasure without an energy cost. Proc. Math. Phys. Eng. Sci. 467, 1770–1778 (2011).
Wright, J. S., Gould, T., Carvalho, A. R., Bedkihal, S. & Vaccaro, J. A. Quantum heat engine operating between thermal and spin reservoirs. Phys. Rev. A 97, 052104 (2018).
Croucher, T. & Vaccaro, J. A. Memory erasure with finitesized spin reservoir. Preprint at https://arxiv.org/abs/2111.10930 (2021).
Jaynes, E. T. Information theory and statistical mechanics II. Phys. Rev. 108, 171–190 (1957).
Balian, R., Alhassid, Y. & Reinhardt, H. Dissipation in manybody systems: a geometric approach based on information theory. Phys. Rep. 131, 1–146 (1986).
Yunger Halpern, N., Faist, P., Oppenheim, J. & Winter, A. Microcanonical and resourcetheoretic derivations of the thermal state of a quantum system with noncommuting charges. Nat. Commun. 7, 12051 (2016).
Quan, H.T., Liu, Y.x, Sun, C.P. & Nori, F. Quantum thermodynamic cycles and quantum heat engines. Phys. Rev. E 76, 031105 (2007).
Deutsch, J. M. Quantum statistical mechanics in a closed system. Phys. Rev. A 43, 2046–2049 (1991).
Srednicki, M. Chaos and quantum thermalization. Phys. Rev. E 50, 888–901 (1994).
Rigol, M., Dunjko, V. & Olshanii, M. Thermalization and its mechanism for generic isolated quantum systems. Nature 452, 854 (2008).
D’Alessio, L., Kafri, Y., Polkovnikov, A. & Rigol, M. From quantum chaos and eigenstate thermalization to statistical mechanics and thermodynamics. Adv. Phys. 65, 239–362 (2016).
Brown, W. & Fawzi, O. Decoupling with random quantum circuits. Commun. Math. Phys. 340, 867–900 (2015).
Nahum, A., Vijay, S. & Haah, J. Operator spreading in random unitary circuits. Phys. Rev. X 8, 021014 (2018).
Khemani, V., Vishwanath, A. & Huse, D. A. Operator spreading and the emergence of dissipative hydrodynamics under unitary evolution with conservation laws. Phys. Rev. X 8, 031057 (2018).
HunterJones, N. Operator growth in random quantum circuits with symmetry. Preprint at https://arxiv.org/abs/1812.08219 (2018).
Swingle, B. Unscrambling the physics of outoftimeorder correlators. Nat. Phys. 14, 988–990 (2018).
Goodman, R. & Wallach, N. R. Symmetry, Representations, and Invariants, Vol. 255 (Springer, 2009).
Das, A. & Okubo, S. Lie Groups and Lie Algebras for Physicists (World Scientific, 2014).
Choi, S., Yao, N. Y. & Lukin, M. D. Dynamical engineering of interactions in qudit ensembles. Phys. Rev. Lett. 119, 183603 (2017).
Choi, J. et al. Robust dynamic hamiltonian engineering of manybody spin systems. Phys. Rev. X 10, 031002 (2020).
Ritter, S. et al. Observing the formation of longrange order during boseeinstein condensation. Phys. Rev. Lett. 98, 090402 (2007).
Trotzky, S. et al. Probing the relaxation towards equilibrium in an isolated strongly correlated onedimensional bose gas. Nat. Phys. 8, 325–330 (2012).
Langen, T., Geiger, R., Kuhnert, M., Rauer, B. & Schmiedmayer, J. Local emergence of thermal correlations in an isolated quantum manybody system. Nat. Phys. 9, 640–643 (2013).
Kaufman, A. M. et al. Quantum thermalization through entanglement in an isolated manybody system. Science 353, 794–800 (2016).
Smith, J. et al. Manybody localization in a quantum simulator with programmable random disorder. Nat. Phys. 12, 907 (2016).
Neill, C. et al. Ergodic dynamics and thermalization in an isolated quantum system. Nat. Phys. 12, 1037 (2016).
Tang, Y. et al. Thermalization near integrability in a dipolar quantum newton’s cradle. Phys. Rev. X 8, 021030 (2018).
Malvania, N. et al. BAPS, Vol. 63 (APS, 2018).
Sánchez, C. M. et al. Emergent perturbation independent decay of the loschmidt echo in a manyspin system studied through scaled dipolar dynamics. Preprint at https://arxiv.org/abs/1902.06628 (2019).
Landsman, K. A. et al. Verified quantum information scrambling. Nature 567, 61–65 (2019).
LewisSwan, R., SafaviNaini, A., Bollinger, J. J. & Rey, A. M. Unifying scrambling, thermalization and entanglement through measurement of fidelity outoftimeorder correlators in the dicke model. Nat. Commun. 10, 1–9 (2019).
Joshi, M. K. et al. Quantum information scrambling in a trappedion quantum simulator with tunable range interactions. Phys. Rev. Lett. 124, 240505 (2020).
Mandl, F. Statistical Physics (Wiley, 1971).
Iachello, F. Lie Algebras And Applications, Vol. 12 (Springer, 2006).
Gilmore, R. Lie groups, Lie Algebras, And Some Of Their Applications (Courier Corporation, 2012).
Baxter, R. J. Exactly Solved Sodels In Statistical Mechanics (Elsevier, 2016).
Gogolin, C. & Eisert, J. Equilibration, thermalisation, and the emergence of statistical mechanics in closed quantum systems. Rep. Prog. Phys. 79, 056001 (2016).
GellMann, M. Murray GellMann: Selected Papers, 128–145 (World Scientific, 2010).
Cahn, R. N. SemiSimple Lie Algebras and Their Representations (Dover, 2006).
Byrd, M. Differential geometry on su (3) with applications to three state systems. J. Math. Phys. 39, 6125–6136 (1998).
Mueller, N., Zache, T. V. & Ott, R. Thermalization of gauge theories from their entanglement spectrum. Preprint at https://arxiv.org/abs/2107.11416 (2021).
Jané, E., Vidal, G., Dür, W., Zoller, P. & Cirac, J. I. Simulation of quantum dynamics with quantum optical systems. Quantum Inf. Comput. 3, 15–37 (2003).
Barredo, D., de Léséleuc, S., Lienhard, V., Lahaye, T. & Browaeys, A. An atombyatom assembler of defectfree arbitrary twodimensional atomic arrays. Science 354, 1021–1023 (2016).
de Léséleuc, S. et al. Observation of a symmetryprotected topological phase of interacting bosons with rydberg atoms. Science 365, 775–780 (2019).
Zhang, J. et al. Observation of a manybody dynamical phase transition with a 53qubit quantum simulator. Nature 551, 601–604 (2017).
Fukuhara, T. et al. Microscopic observation of magnon bound states and their dynamics. Nature 502, 76–79 (2013).
Viola, L., Lloyd, S. & Knill, E. Universal control of decoupled quantum systems. Phys. Rev. Lett. 83, 4888–4891 (1999).
You, J. Q. & Nori, F. Atomic physics and quantum optics using superconducting circuits. Nature 474, 589–597 (2011).
Koch, J. et al. Chargeinsensitive qubit design derived from the cooper pair box. Phys. Rev. A 76, 042319 (2007).
Bianchetti, R. et al. Control and tomography of a three level superconducting artificial atom. Phys. Rev. Lett. 105, 223601 (2010).
Morvan, A. et al. Qutrit randomized benchmarking. Phys. Rev. Lett. 126, 210504 (2021).
Blok, M. S. et al. Quantum information scrambling on a superconducting qutrit processor. Phys. Rev. X 11, 021010 (2021).
Xu, H. et al. Coherent population transfer between uncoupled or weakly coupled states in laddertype superconducting qutrits. Nat. Commun. 7, 1–6 (2016).
Kumar, K., Vepsäläinen, A., Danilin, S. & Paraoanu, G. Stimulated raman adiabatic passage in a threelevel superconducting circuit. Nat. Commun. 7, 1–6 (2016).
Tan, X. et al. Topological maxwell metal bands in a superconducting qutrit. Phys. Rev. Lett. 120, 130503 (2018).
Vepsäläinen, A., Danilin, S. & Paraoanu, G. S. Superadiabatic population transfer in a threelevel superconducting circuit. Sci. Adv. 5, eaau5999 (2019).
Lu, X.J. et al. Nonleaky and accelerated population transfer in a transmon qutrit. Phys. Rev. A 96, 023843 (2017).
Vepsäläinen, A., Danilin, S., Paladino, E., Falci, G. & Paraoanu, G. S. Photonics, Vol. 3, 62 (Multidisciplinary Digital Publishing Institute, 2016).
Yang, C.P., Su, Q.P. & Han, S. Generation of greenbergerhornezeilinger entangled states of photons in multiple cavities via a superconducting qutrit or an atom through resonant interaction. Phys. Rev. A 86, 022329 (2012).
Shlyakhov, A. et al. Quantum metrology with a transmon qutrit. Phys. Rev. A 97, 022115 (2018).
Danilin, S., Vepsäläinen, A. & Paraoanu, G. S. Experimental state control by fast nonabelian holonomic gates with a superconducting qutrit. Phys. Scr. 93, 055101 (2018).
Shnyrkov, V., Soroka, A. & Turutanov, O. Quantum superposition of three macroscopic states and superconducting qutrit detector. Phys. Rev. B 85, 224512 (2012).
Naghiloo, M. Introduction to experimental quantum measurement with superconducting qubits. Preprint at https://arxiv.org/abs/1904.09291 (2019).
Huang, H.L., Wu, D., Fan, D. & Zhu, X. Superconducting quantum computing: a review. Sci. China Inf. Sci. 63, 1–32 (2020).
Kjaergaard, M. et al. Superconducting qubits: current state of play. Annu. Rev. Condens. Matter Phys, 11, 369–395 (2020).
Acknowledgements
N.Y.H. is grateful to Michael Beverland, Aram Harrow, Iman Marvian, Mark Mueller, and Martin Savage for thoughtprovoking conversations. S.M. would like to thank José Polo Gómez; Jimmy ShihChun Hung; Eduardo MartínMartínez; Erickson Tjoa; and, in particular, Tibra Ali for fruitful discussions. This work was supported by an NSF grant for the Institute for Theoretical Atomic, Molecular, and Optical Physics at Harvard University and the Smithsonian Astrophysical Observatory, as well as by administrative support from the MIT CTP. This work received support from the National Science Foundation (QLCI grant OMA2120757).
Author information
Authors and Affiliations
Contributions
N.Y.H. developed the prescription, managed the project, and led the paper writing. S.M. worked out the su(3) example, proofs, Supplementary Notes, and superconductingqutrit details, in addition to leading the referee revisions.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Yunger Halpern, N., Majidy, S. How to build Hamiltonians that transport noncommuting charges in quantum thermodynamics. npj Quantum Inf 8, 10 (2022). https://doi.org/10.1038/s41534022005164
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41534022005164
This article is cited by

Noncommuting conserved charges in quantum thermodynamics and beyond
Nature Reviews Physics (2023)