Lane cc pn - #98
Open
beykyle wants to merge 11 commits into
Open
Lane cc pn#98beykyle wants to merge 11 commits into
beykyle wants to merge 11 commits into
Conversation
Workspace.tmatrix() and .xs() take optional U1_central and U1_spin_orbit. When None, each defaults independently to -(U_n - U_p) * isovector_factor; when given, the array is used as-is in the radial integral. Add tests for default equivalence, independent defaults, linear scaling of T with U1, and shape validation. Add a channel-energy vs midpoint-energy U1 comparison to the CHEX validation notebook (not asserted against CHEX).
Channels in one partial wave share a physical channel radius but may have different wavenumbers, reduced masses and Sommerfeld parameters (e.g. the p and n channels of Lane (p,n)_IAS). Previously every channel's Coulomb-Hankel functions were evaluated at k_0 a, surface derivatives were taken in s = k_0 r without the mu_0/mu_i Bloch factor, and S was the raw amplitude ratio. - get_partial_wave_channels tabulates each channel at rho_i = k_i a, keyed by (eta, rho) - Solver.solve scales channel derivatives by v_i/v_0, returns the flux-normalized S and R in the rho-derivative convention All changes are the identity for a single channel or equal k and mu. Add tests: uncoupled different-k channels reproduce single-channel solves, unitary/symmetric S for real coupling, and channel-radius independence.
jitr.xs.lane_pn.Workspace solves the 2x2 proton/neutron coupled channels for each (l, j), coupled by the isovector transition potential U1, with the incoming wave in the proton channel. It returns the full R- and flux-normalized S-matrices (rsmatrix) and builds the differential and angle-integrated (p,n) cross sections from S_np. The API mirrors the DWBA workspace, including the optional U1_central/U1_spin_orbit. Factor the potential validation, default U1 recipe, isovector factor and spin-1/2 CG/Y_lm transition geometry out of quasielastic_pn into shared helpers; DWBA output is unchanged (max rel diff 3e-15). Tests: CC(eps U1)/eps^2 converges to DWBA as O(eps^2), partial-wave and angle-integrated cross sections agree, S is unitary and symmetric for real potentials, j-independence without spin-orbit, default/explicit U1.
Add a section computing the Lane coupled-channels (p,n)_IAS cross section with jitr.xs.lane_pn.Workspace, with the same potentials and kinematics as the DWBA, plotted against DWBA, CHEX and data (not asserted against CHEX). For 48Ca(p,n) at 35 MeV, CC is ~1% below DWBA (sigma_pn 7.55 vs 7.65 mb).
Review of the coupled-channels branch found that solve(..., wavefunction=True) returned wrong interior coefficients for every multichannel solve: the outgoing term in the Bloch surface source contracted the S-matrix with the channel index instead of the incoming weights. Interior and exterior wavefunctions disagreed by O(100%); single-channel results were unaffected. Wavefunctions.uext() also raised TypeError unconditionally (Channels has no __len__), passed a float l to the Coulomb functions, evaluated every channel in the channel-0 coordinate, and used the flux-normalized S in the raw-S asymptotic form. All fixed; it now takes the S returned by solve. Add tests/test_wavefunction_continuity.py pinning interior/exterior continuity at the channel radius for 1 and 3 channels with several incoming weights, and for two channels with different k and mu. Also from review: raise on the N - Z = 1 divergence of isovector_factor, replace the four-deep selection loop in quasielastic_pn.xs with the einsum it was equivalent to (DWBA output unchanged, 1.5e-15), hoist Y_lm out of the j loop, name the (eta, rho) table keys, and document the unused Elab/Ecm parameters of get_partial_wave_channels.
48Ca(p,n)48Sc(IAS) DWBA at Elab(p) = 25 and 35 MeV, from decks contributed by Jin Lei (Tongji University) and re-run locally with Frescox 7.2-20-ga7f491. These are the first non-elastic regression cases and the first external check on the spin-flip terms of the (p,n) amplitude: Frescox does its own channel-spin algebra, so it is independent of jitR and CHEX. jitR reproduces both to better than 1.3e-4 at every angle; with the pre-fix Clebsch-Gordan argument order both cases fail. Harness: build_case dispatches on observable_type, BuiltCase grows a dsdo extractor (the (p,n) workspace returns a bare array, elastic an ElasticXS), and _build_quasielastic_pn_case builds both channels from the deck's own lab energies. U1_central is left to jitR's default isovector difference, which equals the deck's tabulated form factor to 6e-9; U1_spin_orbit is zero because a Frescox KIND=1 form factor is central only. tools/make_pn_formfactor.py regenerates the form factor from the KD02 parameters, and documents two Frescox conventions: the header must carry explicit LOP/DER or the form factor is dropped silently and the cross section comes out identically zero, and FSCALE = sqrt(2)*sqrt(4 pi) cancels against Frescox's own KIND=1 coupling coefficient rather than being a free normalization.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add Lane coupled (p,n)