Apparent Dresselhaus coefficient in (001) GaAs quantum wells: Correlation-time renormalization in D'yakonov-Perel' spin relaxation

Jun Ishihara; Satoshi Iba; Yuzo Ohno

REVIEW 4 major objections 4 minor 16 references

Apparent Dresselhaus coefficient in (001) GaAs quantum wells: Correlation-time renormalization in D'yakonov-Perel' spin relaxation

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Spin relaxation in (001) GaAs quantum wells measures an apparent Dresselhaus coefficient of about 9.2 eV Å³, not the bulk cubic value of 12.4 eV Å³.

desk verdict Useful Monte Carlo correlation times, but the apparent-coefficient formula is imposed rather than derived. read the letter →

arxiv 2608.03128 v1 pith:GOLWGZTI submitted 2026-08-04 cond-mat.mtrl-sci

Yuzo Ohno , Jun Ishihara , Satoshi Iba This is my paper · ORCID

classification cond-mat.mtrl-sci PACS 72.25.Rb71.70.Ej

keywords Dresselhausspin-orbitcouplingD'yakonov-Perel'spinrelaxationGaAsquantumwellsMonteCarlosimulationspin-orbit-fieldcorrelationkernelLO-phononscatteringapparentcoefficientpersistenthelix

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The authors claim that the Dresselhaus coefficient inferred from D'yakonov-Perel' (DP) spin relaxation in (001) GaAs quantum wells is not the intrinsic bulk cubic coefficient but an apparent coefficient tied to the correlation kernel of the projected two-dimensional spin-orbit field. In bulk GaAs, their inelastic LO-phonon Monte Carlo reproduces the measured excitation-density dependence with gamma_3D≈12.4 eV Å^3 and a cubic-field correlation time of about 150 fs. The standard projection k_z^2 -> turns the leading Dresselhaus field from cubic to linear in the in-plane momentum, which lengthens the DP correlation kernel so that its effective time becomes about 235 fs. Equating the relaxation rate written with the cubic-field kernel to the rate written with the projected kernel gives gamma_2D^app = gamma_3D sqrt(tau_3^{2D}/tau_{1+3}^{2D}) ≈9.2 eV Å^3, matching quantum-well experiments. If this is right, the apparent reduction of the two-dimensional coefficient is a correlation-time renormalization, not a change in the material parameter.

What carries the argument

The central object is the DP correlation kernel K = ∫ <F(t)·F(0)> dt, where F is the wave-vector part of the Dresselhaus spin-orbit field with the material coefficient factored out. Because the relaxation rate is Gamma = (2γ/ħ)^2 K, the coefficient and the kernel are inseparable: a projected field with a longer effective correlation time must carry a smaller apparent coefficient to give the same rate. The paper separates the cubic bulk field F_3D, which has third-order angular correlations and tau≈150 fs, from the projected two-dimensional field F_2D, whose leading term is linear in in-plane momentum after k_z^2 -> <k_z^2>; this hybridized kernel K_{1+3}^{2D} has tau_{1+3}^{2D}≈235 fs. The r

What would settle it

Measure the spin relaxation time and the static linear Dresselhaus coefficient beta1 on the same series of (001) GaAs quantum wells spanning E1 from about 5 to 80 meV. The paper predicts the DP-inferred gamma_2D^app(L_w) tracks gamma_3D sqrt(tau_3^{2D}/tau_{1+3}^{2D}(L_w)), reaching about 9.2 eV Å^3 in narrow wells, while the static beta1 extraction tracks gamma_3D≈12.4 eV Å^3; if the two agree at every width, the correlation-time renormalization is falsified.

Watch

Extended reading notes

Core claim

The paper's central discovery is a ratio identity connecting bulk and two-dimensional Dresselhaus coefficients through DP correlation times: gamma_2D^app = gamma_3D sqrt(tau_3^{2D}/tau_{1+3}^{2D}). This follows from demanding that the same physical relaxation rate be reproduced whether it is written with the cubic-field kernel or with the projected two-dimensional kernel. The Monte Carlo results fix the ingredients: in bulk GaAs gamma_3D≈12.4 eV Å^3 and tau_3^{3D}≈150 fs; in the projected (001) quantum well tau_3^{2D}≈130 fs while the hybridized first/third-order kernel tau_{1+3}^{2D}≈235 fs. With tau_{1+3}^{2D}≈1.8 tau_3^{2D}, the apparent coefficient becomes gamma_3D/1.34≈9.2 eV Å^3, close

Load-bearing premise

The argument assumes that the physical spin relaxation rate in the quantum well is correctly represented by gamma_3D^2 tau_3^{2D}, the cubic-field kernel, so that defining gamma_2D^app by equating this rate to (gamma_2D^app)^2 tau_{1+3}^{2D} is legitimate; if the true rate is controlled by the total projected kernel tau_{1+3}^{2D}, the apparent coefficient is an artifact of that choice.

Editorial extensions

If this is right

Using the unchanged bulk gamma_3D in a projected 2D DP model overestimates the spin relaxation rate by roughly the ratio tau_{1+3}^{2D}/tau_3^{2D} ≈1.8 in the narrow-well limit.
A smooth well-width crossover is predicted: gamma_2D^app(L_w) = gamma_3D sqrt(tau_3^{2D}(L_w)/tau_{1+3}^{2D}(L_w)), from gamma_3D in wide wells toward gamma_3D/1.34≈9.2 eV Å^3 in the strict 2D limit.
Static spin-splitting measurements, using beta1 = -gamma<k_z^2>, and DP spin relaxation probe different correlation kernels, so they need not return the same effective gamma.
In persistent-spin-helix analyses, linear and cubic Dresselhaus terms should not be forced to share a single common gamma when spin lifetimes are involved.
Elastic-scattering models capture the same crossover after recalibrating the coefficient, giving gamma≈8.8 eV Å^3 in the quantum-well elastic model, so the renormalization is not an artifact of the inelastic treatment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same correlation-kernel logic should apply to any spin-orbit term whose dimensional reduction changes angular order, including higher subbands and other III-V or II-VI quantum wells, so reported 2D Dresselhaus coefficients may carry a kernel-dependent correction.
A clean within-sample test—static beta1 and DP lifetime on the same quantum well—would sharpen the factor sqrt(tau_{1+3}^{2D}/tau_3^{2D})≈1.34 that the present comparison of separate bulk and quantum-well datasets leaves somewhat indirect.
If the angular-order dependence of correlation times holds generally, comparisons of Rashba and Dresselhaus strengths extracted from spin lifetimes should renormalize each term by its own kernel ratio before assigning persistent-spin-helix balance conditions.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The Monte Carlo correlation-time calculation is the real content here. The finding that the total projected 2D Dresselhaus-field kernel tau_{1+3} is about 1.8 times the cubic-only kernel tau_3, and that it crosses over from ~150 fs in wide wells to ~235 fs in the 2D limit, is concrete and useful. The paper also makes a fair conceptual point: the DP relaxation rate is controlled by the correlation integral of the effective field, so the coefficient you extract depends on which field you put in the Hamiltonian.

The trouble is the headline relation. The paper defines gamma_2D^app = gamma_3D sqrt(tau_3/tau_{1+3}) by equating gamma_3D^2 tau_3^{2D} to (gamma_2D^app)^2 tau_{1+3}^{2D}. That equality is asserted, not derived. In the projected model of Sec. IVA, the transverse field is exactly F_1+F_3, so if you use the bulk coefficient gamma_3D in that Hamiltonian, the DP rate is (2 gamma_3D/hbar)^2 K_{1+3}^{2D}. The paper's own Sec. IVC shows that this rate is faster than experiment. The renormalization then chooses tau_3 as the reference kernel to bring the rate back down. That makes gamma_2D^app a fitted quantity, not a consequence of the correlation-time calculation. Unless there is a microscopic reason why the cubic-only kernel controls the physical relaxation in the QW, the central claim does not follow.

Two softer but real issues: gamma_3D = 12.4 eV A^3 is calibrated from the authors' own prior bulk measurement, so the absolute scale is not independent. And the comparison with QW data in Fig. 3 is qualitative—no error bars or systematic fit, just a claim that the renormalized curve looks better. The paper is honest about the limited experimental coverage (Sec. VF), which I credit.

Who is this for? People working on DP spin relaxation in 2D semiconductors will find the tau_{1+3}/tau_3 ratio and the crossover discussion worth knowing. But the paper would need a derivation or a falsifiable prediction that distinguishes the apparent-coefficient picture from a simple re-fitting of gamma before I'd take the central claim as established.

I'd send it to a referee. The MC machinery and the kernel comparison deserve scrutiny, and the flawed central relation is the kind of thing a good referee can push on. I would not accept it as is.

Referee Report

4 major / 4 minor

Summary. The paper uses a fixed-time-step ('nonballistic') ensemble Monte Carlo method to compute D'yakonov–Perel' (DP) spin-orbit-field correlation functions and spin relaxation in bulk GaAs and (001) GaAs quantum wells. In bulk GaAs, fitting to the authors' previous experimental data yields a cubic Dresselhaus coefficient γ₃D ≈ 12.4 eV Å³ and a cubic-field correlation time τ₃^{3D} ≈ 150 fs. In the projected two-dimensional model, the replacement k_z² → ⟨k_z²⟩ converts the dominant Dresselhaus field into a linear in-plane term, and the paper computes a hybridized kernel K_{1+3}^{2D} with correlation time τ_{1+3}^{2D} ≈ 235 fs against τ₃^{2D} ≈ 130 fs. The central claim is that the coefficient entering a projected 2D DP model is an apparent coefficient γ₂D^{app} = γ₃D √(τ₃^{2D}/τ_{1+3}^{2D}) ≈ 9.2 eV Å³, reconciling bulk and quantum-well Dresselhaus coefficients via correlation-time renormalization.

Significance. If the central claim were established, the paper would provide a physical explanation for why Dresselhaus coefficients extracted from two-dimensional DP spin relaxation differ from bulk values, and would sharpen the interpretation of linear and cubic Dresselhaus parameters in quantum wells. The Monte Carlo computation of the correlation-kernel crossover from 3D to 2D is a useful technical contribution. However, the key relation defining γ₂D^{app} is imposed, not derived, and the numerical value 9.2 eV Å³ is therefore not an independent prediction. The paper's main conclusion is unsupported as written.

major comments (4)

[Sec. IVC (also Sec. I, Eq. defining γ₂D^{app})] The equality Γ_s = γ₃D² τ₃^{2D} = (γ₂D^{app})² τ_{1+3}^{2D} is asserted, not derived. In a single-subband (001) QW, the transverse Dresselhaus field is the full projected field F_{1+3}^{2D} = F₁^{2D} + F₃^{2D}; its DP correlation kernel is K_{1+3}^{2D}. The rate obtained when the bulk coefficient γ₃D is used in the projected model is (2γ₃D/ℏ)² K_{1+3}^{2D}, not (2γ₃D/ℏ)² K₃^{2D}. Thus the physical QW rate is not γ₃D² τ₃^{2D}; the reference to the cubic-only kernel is chosen by hand. Consequently γ₂D^{app} = γ₃D √(τ₃^{2D}/τ_{1+3}^{2D}) is an algebraic identity, not a derived renormalization. The paper must either provide a microscopic justification for excluding the linear term from the DP kernel, or reframe γ₂D^{app} as a phenomenological mapping rather than a prediction.
[Sec. III, bulk calibration] The value γ₃D = 12.4 eV Å³ is obtained by fitting to the authors' own previous experiment (Ref. [10]) with no reported uncertainty or sensitivity analysis. The central conclusion scales directly with γ₃D; for instance, if γ₃D = 11 eV Å³ (within the usual literature spread), then γ₂D^{app} ≈ 8.2 eV Å³, which is also consistent with the quantum-well data cited. The paper should give a range for γ₃D and propagate the uncertainty into γ₂D^{app}.
[Sec. IVC and Fig. 3] The comparison with quantum-well experiments is qualitative. The calculated spin-relaxation curves have no error bars, and the experimental points from Refs. [13–16] scatter by up to a factor of 2–3. The statement that the renormalized coefficient reproduces the data 'more consistently' is based on visual inspection. A quantitative goodness-of-fit measure, or at least an estimate of the statistical uncertainty in the Monte Carlo time constants, is needed before the consistency claim can be evaluated.
[Sec. V.F, Limitations] The limitations section does not list the most important caveat: that the central equality (Secs. I and IVC) is an assumption rather than a result. The authors present the equality as self-evident and only discuss limitations of scattering processes, well-width data, and the static/probe distinction. The missing justification of the reference kernel should be explicitly acknowledged as the main limitation.

minor comments (4)

[Sec. V.F] Typo: 'narrw' should be 'narrow'.
[Ref. [12]] Author name 'Ferrira' should be 'Ferreira'.
[Sec. IVA] The decomposition F_{2D,n} = F₁^{2D} + F₃^{2D} is stated without derivation; adding a few lines showing the trigonometric identities would improve readability.
[Abstract and Sec. II] The term 'nonballistic' is used for the fixed-time-step ensemble Monte Carlo method; a brief definition or a different term (e.g., 'fixed-time-step') would avoid confusion with the common meaning of ballistic vs scattering-dominated evolution.

Circularity Check

2 steps flagged · score 6.0 of 10

γ2D^app is defined by the same equation that is presented as its derivation; the 9.2 eV Å^3 value is a rescaled fitted γ3D, not an independent prediction.

self definitional [Sec. I (Introduction, central point); Eq. Γs = γ3D^2 τ3^{2D} = (γ2D^app)^2 τ1+3^{2D}]
"If the same physical relaxation rate is written either in terms of the cubic-field correlation kernel or in terms of the projected two-dimensional kernel, Γs = γ3D^2 τ3^{2D} = (γ2D^app)^2 τ1+3^{2D}, then γ2D^app ≅ γ3D sqrt(τ3^{2D}/τ1+3^{2D})."

The equality is imposed, not derived. In a single-subband (001) QW, the transverse Dresselhaus field of the projected Hamiltonian is F1+F3, so the DP rate computed with the bulk coefficient is (2γ3D/ℏ)^2 K_{1+3}^{2D}. Equating this to γ3D^2 τ3^{2D} requires the unproved assertion that the physical QW rate is controlled by the cubic-only kernel τ3^{2D}. With the natural projected-field kernel K_{1+3}^{2D} as reference, one would obtain γ2D^app = γ3D identically. The advertised reduction sqrt(τ3/τ1+3) is thus a choice of reference kernel, i.e., the definition of the apparent coefficient, not a physical prediction.
fitted input called prediction [Sec. IVC (Well-width dependence of spin relaxation time with the effective two-dimensional coefficient); also Sec. V.A]
"the projected model should be written using an apparent coefficient γ2D^app defined by equating the relaxation rate written in terms of the cubic-field kernel and that written in terms of the projected two-dimensional kernel: γ3D^2 τ3^{2D} = (γ2D^app)^2 τ1+3^{2D}."

The paper explicitly says γ2D^app is 'defined by equating' the two rates. Since γ3D is itself a fit to bulk spin-relaxation data (Sec. II: 'γ3D is treated as an adjustable parameter and is determined by comparison with the experimentally measured spin relaxation time ... [10]'), the quoted 9.2 eV Å^3 is a rescaled fitted value rather than an independent prediction. The later statement that this value is 'consistent with spin relaxation in (001) GaAs quantum wells' is a consistency check of the defining equation, not a falsifiable test of a separately derived coefficient.

full rationale

The paper's Monte Carlo computation of the correlation times τ3^{2D} ≈ 130 fs and τ1+3^{2D} ≈ 235 fs is a genuine, self-contained numerical result; if the paper had only reported the kernel crossover, no circularity would arise. The circularity enters when this ratio is converted into a 'predicted' Dresselhaus coefficient. The central equation γ3D^2 τ3^{2D} = (γ2D^app)^2 τ1+3^{2D} appears in the Introduction as the central point and in Sec. IVC/V.A as the definition of γ2D^app. No independent microscopic argument is given for why the physical QW DP rate should be written with the cubic-only kernel τ3^{2D} rather than the projected-field kernel τ1+3^{2D}; in the latter case the apparent coefficient would equal γ3D. The 9.2 eV Å^3 number therefore reduces by construction to the chosen reference kernel and to the fitted bulk γ3D. This is a partial circularity: the derived coefficient is a reparameterization, while the genuinely new content is the simulated correlation-time ratio. Self-citation of the authors' prior bulk measurement [10] is not by itself load-bearing here; the issue is that the 'prediction' is definitionally tied to a fitted input.

Assumptions & free parameters 2 free parameters · 5 assumptions · 1 invented entities

The central numerical claim rests on one fitted parameter (gamma_3D), an imposed rate-equality axiom, and the assumption that the projected 2D field captures the relevant physics. The correlation-time ratio is computed rather than fitted, but the apparent coefficient is a calibrated quantity tied to the fitted gamma_3D.

free parameters (2)

gamma_3D (bulk cubic Dresselhaus coefficient) = 12.4 eV Å^3
Adjusted to match bulk GaAs spin relaxation data from the authors' previous measurement (Ref [10]); all later apparent-coefficient values inherit this fit.
Elastic relaxation times for acoustic-phonon and electron-electron scattering = not given
In the bulk model these processes are approximated as elastic relaxation-time channels; the relaxation times are not specified, so the correlation times depend on undisclosed modeling choices.

assumptions (5)

domain assumption D'yakonov-Perel' rate equals the integral of the spin-orbit-field autocorrelation function
Used throughout Sec. II to define the correlation kernel; standard DP theory [4].
domain assumption The projection k_z^2 -> <k_z^2> gives the correct leading spin-orbit field in a (001) QW
Invoked in Sec. IVA following refs [11,12]; it is the operation that creates the linear Dresselhaus term whose correlation time drives the renormalization.
ad hoc to paper The physical QW spin relaxation rate equals gamma_3D^2 tau_3^{2D}, so gamma_3D^2 tau_3^{2D} = (gamma_2D^app)^2 tau_{1+3}^{2D}
Stated in Sec. I ('central point') and Sec. IVC; this equality defines gamma_2D^app and is not derived. It is the load-bearing premise that makes the apparent coefficient well-defined.
domain assumption LO-phonon scattering dominates spin-orbit-field randomization at room temperature and weak excitation in (001) QWs
Sec. IVA and VF assume this to justify omitting other scattering from the primary QW calculation; if false, the tau ratios change.
domain assumption Monte Carlo ensemble of 10^5 trajectories with fixed timestep gives converged correlation times
Sec. II; no convergence or variance analysis is provided.

invented entities (1)

Apparent two-dimensional Dresselhaus coefficient gamma_2D^app
purpose: Absorb the change of DP correlation kernel when the projected 2D model is used, so the projected model reproduces the assumed cubic-field rate.
Defined by the rate-equality equation in Sec. IVC; no independent measurement targets gamma_2D^app directly, and its value follows from the fitted gamma_3D and the computed tau ratio.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Apparent Dresselhaus coefficient in (001) GaAs quantum wells: Correlation-time renormalization in D'yakonov-Perel' spin relaxation." pith.science (2026). https://pith.science/paper/GOLWGZTI

@misc{pith2026260803128,
  author       = {Pith},
  title        = {Pith review of: Apparent Dresselhaus coefficient in (001) GaAs quantum wells: Correlation-time renormalization in D'yakonov-Perel' spin relaxation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GOLWGZTI}},
  note         = {Machine review of arXiv:2608.03128}
}

read the original abstract

We study the Dresselhaus coefficient inferred from D'yakonov-Perel' spin relaxation in bulk GaAs and (001) GaAs quantum wells using nonballistic Monte Carlo simulations. In bulk GaAs, inelastic LO-phonon simulations reproduce the spin relaxation with a cubic Dresselhaus coefficient \gamma_{3D}\simeq12.4 eV{\AA}^3 and a cubic-field correlation time \tau_3^{3D}\simeq150 fs. In a projected two-dimensional quantum-well model, however, k_z^2 is replaced by the static expectation value \langle k_z^2 \rangle, converting the dominant Dresselhaus field into a term linear in the in-plane wave vector. We show that this projection changes the D'yakonov-Perel' correlation kernel from K_3^{2D} to a first- and third-order hybridized kernel K_{1+3}^{2D}, for which the correlation time \tau_{1+3}^{2D} approaches 235 fs in the strict two-dimensional limit, in contrast to the third-order correlation time \tau_3^{2D} of 130 fs. Consequently, the coefficient entering the projected two-dimensional model is not the intrinsic cubic coefficient itself but an apparent coefficient, \gamma_{2D}^{app}=\gamma_{3D} \sqrt{\tau_3^{2D}/\tau_{1+3}^{2D}}. Since \tau_{1+3}^{2D}\simeq1.8\tau_3^{2D} under LO-phonon-dominated scattering, \gamma_{2D}^{app}\simeq9.2 eV{\AA}^3, consistent with spin relaxation in (001) GaAs quantum wells. This correlation-time renormalization clarifies why Dresselhaus coefficients extracted from two-dimensional spin relaxation can differ from the cubic bulk coefficient and provides a useful framework for interpreting linear and cubic Dresselhaus parameters in quantum wells.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[10]

Ohno and S

Y . Ohno and S. Iba, Electron spin relaxation in bulk GaAs at room temperature: A Monte Carlo-based quantitative study, Applied Physics Letters 128, 082401 (2026)

work page 2026
[1]

Meier and B

F. Meier and B. P. Zakharchenya, Optical Orientation (Elsevier, Amsterdam, 1984)

work page 1984
[2]

Fabian and S

J. Fabian and S. Das Sarma, Spin relaxation of conduction electrons, J. Vac. Sci. Technol. B 17, 1708 (1999)

work page 1999
[3]

M. W. Wu, J. H. Jiang, and M. Q. Weng, Spin dynamics in semiconductors, Phys. Rep. 493, 61 (2010)

work page 2010
[4]

M. I. D’yakonov and V . I. Perel’, Spin relaxation of conduction electrons in noncentrosymmetric semiconductors, Sov. Phys. Solid State 13, 3023 (1972)

work page 1972
[5]

M. P. Walser, U. Siegenthaler, V . Lechner, D. Schuh, S. D. Ganichev, W. Wegscheider, and G. Salis, Dependence of the Dresselhaus spin-orbit interaction on the quantum well width, Phys. Rev. B 86, 195309 (2012)

work page 2012
[6]

Dettwiler, J

F. Dettwiler, J. Fu, S. Mack, P.J. Weigele, J.C. Egues, D.D. Awschalom and D.M. Zumbühl, Stretchable persistent spin helices in GaAs quantum wells, Phys. Rev. X 7, 031010 (2017)

work page 2017
[7]

Y . Tojo, S. Yoshida, F. Tsuboi, T. Yamamoto, S. Iba, and Y . Ohno, Consistent description of room-temperature electron spin relaxation in GaAs/AlGaAs (001) quantum wells, Appl. Phys. Lett. 129, 032106 (2026)

work page 2026

Show all 16 references

[8]

Jacoboni and P

C. Jacoboni and P . Lugli, The Monte Carlo Method for Semiconductor Device Simulation (Springer, Vienna 1989)

1989
[9]

B. K. Ridley, Quantum Processes in Semiconductors (Oxford University Press, Oxford, 2013)

2013
[11]

D’yakonov and V .Y

M.I. D’yakonov and V .Y . Kacholovskii, Spin relaxation of two-dimensional electrons in noncentrosymmetric semiconductors, Soviet Physics Semiconductors 20, 110 (1986)

1986
[12]

Ferrira and G

R. Ferrira and G. Bastard, Evaluation of some scattering times for electrons in unbiased and MonteCarlo in (001) QW ver3.0 25 biased single- and multiple-quantum-well structures, Physical Review B 40, 1074 (1989)

1989
[13]

Tackeuchi, S

A. Tackeuchi, S. Muto, T. Inata, and T. Fujii, Direct observation of picosecond spin relaxation of excitons in GaAs/AlGaAs quantum wells using spin-dependent optical nonlinearity, Appl. Phys. Lett. 56, 2213 (1990)

1990
[14]

Britton, T

R.S. Britton, T. Grevatt, A. Malinowski, R.T. Harley, P. Perozzo, A.R. Cameron and A. Miller, Room temperature spin relaxation in GaAs/AlGaAs multiple quantum wells, Appl. Phys. Lett. 73, 2140 (1998)

1998
[15]

Terauchi, Y

R. Terauchi, Y . Ohno, T. Adachi, A. Sato, F. Matsukura, A. Tackeuchi, and H. Ohno, Carrier Mobility Dependence of Electron Spin Relaxation in GaAs Quantum Wells, Japanese Journal of Applied Physics 38, 2549 (1999)

1999
[16]

Malinowski, R.S

A. Malinowski, R.S. Britton, T. Grevatt, R.T. Harley, D.A. Ritchie and M.Y . Simmons, Spin relaxation in GaAs/AlxGa1-xAs quantum wells, Phys. Rev. B 62, 13034 (2000)

2000

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [10]

Ohno and S

Y . Ohno and S. Iba, Electron spin relaxation in bulk GaAs at room temperature: A Monte Carlo-based quantitative study, Applied Physics Letters 128, 082401 (2026)

work page 2026

[2] [1]

Meier and B

F. Meier and B. P. Zakharchenya, Optical Orientation (Elsevier, Amsterdam, 1984)

work page 1984

[3] [2]

Fabian and S

J. Fabian and S. Das Sarma, Spin relaxation of conduction electrons, J. Vac. Sci. Technol. B 17, 1708 (1999)

work page 1999

[4] [3]

M. W. Wu, J. H. Jiang, and M. Q. Weng, Spin dynamics in semiconductors, Phys. Rep. 493, 61 (2010)

work page 2010

[5] [4]

M. I. D’yakonov and V . I. Perel’, Spin relaxation of conduction electrons in noncentrosymmetric semiconductors, Sov. Phys. Solid State 13, 3023 (1972)

work page 1972

[6] [5]

M. P. Walser, U. Siegenthaler, V . Lechner, D. Schuh, S. D. Ganichev, W. Wegscheider, and G. Salis, Dependence of the Dresselhaus spin-orbit interaction on the quantum well width, Phys. Rev. B 86, 195309 (2012)

work page 2012

[7] [6]

Dettwiler, J

F. Dettwiler, J. Fu, S. Mack, P.J. Weigele, J.C. Egues, D.D. Awschalom and D.M. Zumbühl, Stretchable persistent spin helices in GaAs quantum wells, Phys. Rev. X 7, 031010 (2017)

work page 2017

[8] [7]

Y . Tojo, S. Yoshida, F. Tsuboi, T. Yamamoto, S. Iba, and Y . Ohno, Consistent description of room-temperature electron spin relaxation in GaAs/AlGaAs (001) quantum wells, Appl. Phys. Lett. 129, 032106 (2026)

work page 2026