Skip to content

Momentum-Transfer Cross Section

Summarize an elastic scattering angle distribution by weighting each event for its loss of incident-direction momentum.

Version
v1 · 2026-10-03 · History
Domain-specific #
13441
Domain group
Natural Sciences
Origin domain
Physics
Subdomains
Scattering Theory, Kinetic Transport → Physics

Core Idea

For a specified elastic scattering channel and incident energy \(E\), the momentum-transfer cross section is the angular moment

\[ \sigma_{\mathrm{mt}}(E)=\int_{4\pi}(1-\cos\theta)\,\frac{d\sigma_{\mathrm{el}}(E)}{d\Omega}\,d\Omega, \]

where \(\theta\) is the outgoing projectile's deflection from its incident direction. If the differential cross section is azimuthally symmetric, this becomes \(2\pi\int_0^\pi(1-\cos\theta)(d\sigma_{\mathrm{el}}/d\Omega)\sin\theta\,d\theta\). The result has area units and is generally energy dependent. It is an effective area for directional momentum relaxation, not the physical size of the scatterer.[1][2]

The defining choice is the weight \(1-\cos\theta\). In the elastic fixed-speed or heavy-target approximation, outgoing momentum projected onto the incoming direction is \(p\cos\theta\), so the fractional longitudinal loss is \(1-\cos\theta\). Nearly forward scattering contributes little; a near reversal contributes almost twice the unweighted event contribution. The ordinary total elastic cross section instead uses weight $1$ and counts all scattering events equally. NIST calls its calculated \(1-\cos\theta\) quantity a transport cross section, while the LXCat team distinguishes it from both the total elastic and viscosity moments. The label “transport” alone is therefore not enough: one must inspect the stated weight and channels.[2][1]

The scalar is useful precisely because it throws away most angular detail. It retains the first angular moment relevant to certain transport approximations, but not every higher angular feature, energy-transfer channel or spatial complication. It does not, by itself, calculate a universal collision frequency, mobility, diffusion coefficient or solid-state attenuation length; each requires a specified kinetic or electron-transport model and sometimes other cross sections.[1][2]

Structural Signature

Sig role-phrases: declared elastic channel and incident energy → differential angular cross section → incident direction and scattering angle → longitudinal weight \(1-\cos\theta\) → solid-angle integration → momentum-relaxation effective area; transport prediction is downstream.

  • Typed collision and energy. Specify projectile, target, elastic channel and incident energy. Otherwise differently tabulated cross sections cannot be compared. The simple longitudinal fractional-loss reading presumes equal incoming/outgoing projectile momentum magnitude or an appropriate heavy-target approximation.[2]
  • Differential cross section. \(d\sigma_{\mathrm{el}}/d\Omega\) describes how elastic scattering is distributed over outgoing directions. A total cross section alone cannot recover the moment without an angular assumption.[1][2]
  • Angle and reference axis. \(\theta\) is measured between initial and final projectile directions in the chosen frame. “Forward” and “backward” have meaning only relative to that axis.[2]
  • Specified weight. \(1-\cos\theta\) retains longitudinal directional loss. Weight $1$ gives the total elastic cross section, while \(1-\cos^2\theta\) produces the distinct viscosity cross section used in some transport datasets.[1]
  • Integration and effective area. Integrate the weighted DCS over solid angle for a scalar indexed by energy and species. For an azimuthally symmetric DCS the azimuthal integral supplies \(2\pi\); without that symmetry, the full solid-angle integral remains the safe definition.[2]
  • Use boundary. A solver may use this value to estimate a transport property, but the property also depends on density, speed or energy distribution, other channels, geometry and approximation. That use is not part of the cross section's identity.[1][2]

What It Is Not

It is not the total elastic scattering cross section, \(\sigma_{\mathrm{el}}=\int(d\sigma_{\mathrm{el}}/d\Omega)d\Omega\). Total cross section counts events without directional weighting. Two angular distributions can share \(\sigma_{\mathrm{el}}\) yet have different \(\sigma_{\mathrm{mt}}\) if one is more forward-peaked. For nonnegative DCSs, \(0\leq\sigma_{\mathrm{mt}}\leq2\sigma_{\mathrm{el}}\) follows because \(0\leq1-\cos\theta\leq2\). Isotropic angular scattering gives equality \(\sigma_{\mathrm{mt}}=\sigma_{\mathrm{el}}\); this limiting equality does not make the definitions identical.[1][2]

It is not a backscattering cross section. That nearby live node concerns scattering observed or normalized in the reverse direction, possibly with wave-specific \(4\pi\) conventions. The present quantity integrates all directions, even though reverse events receive more weight. It is not a viscosity cross section, whose \(1-\cos^2\theta\) factor treats forward and backward deflection differently. It is not generic scattering itself, which is the physical redistribution; nor is it a collision frequency, which is a rate built from cross sections and a kinetic population model.[1]

It is not every quantity called “transport cross section.” NIST uses that name for the elastic \(1-\cos\theta\) moment; LXCat also records effective momentum-transfer values that approximately add inelastic process cross sections for a specified two-term transport treatment. Other nuclear or radiative transport conventions may include removal or absorption terms. The channel and weighting must be named before equating labels.[2][1]

Scope of Application

This entry is most direct where an angle-resolved elastic projectile–target scattering description is available and directional momentum degradation is at issue. In weakly ionized gases, electron–neutral elastic momentum-transfer cross sections are part of the “complete” collision sets used by plasma Boltzmann or Monte Carlo solvers; the same sets also need separately treated excitation, ionization and attachment channels. LXCat warns that fits inferred from swarm properties need not be unique, so a tabulated scalar should not be mistaken for a uniquely measured underlying angular distribution.[1]

For electron transport in solids, the NIST database supplies calculated elastic differential, total and transport cross sections for electrons scattering from constituent free atoms of elements across its documented energy range. These are inputs to radiation-physics, electron-lithography and surface-analysis modeling. The free-atom potential is a modeling approximation for a solid, not a proof that condensed-matter scattering exactly duplicates isolated-atom scattering. The mathematical angular moment is the same; the physical input model and uncertainty are not.[2]

Clarity

Three words often conceal three different measurements: differential gives an angle-resolved distribution; total integrates it without a weight; momentum-transfer integrates it with \(1-\cos\theta\). Report species, collision channel, energy, frame and normalization. In particular, a printed table titled “elastic cross section” may mean total or momentum-transfer in different collections, and LXCat explicitly labels which it tabulates.[1]

The momentum language must also be precise. For equal-magnitude incoming and outgoing projectile momentum, the projection along the original direction changes by \(p(1-\cos\theta)\). In an inelastic process or appreciable recoil, the outgoing momentum magnitude need not equal the incoming one; inserting \(1-\cos\theta\) without a channel convention can then misrepresent actual momentum exchange. The entry's simple formula is the elastic angular moment, not a general theory of every collision's vector momentum balance.[2]

Manages Complexity

A full DCS is a function of angle and energy. The momentum-transfer cross section compresses the angle dimension at each energy to one scalar chosen for a transport purpose. Instead of tracking every deflection individually, a suitable kinetic approximation can use this first angular moment to represent how collisions degrade a directed population. This is a purpose-designed loss of information: forward events receive near-zero influence on directional relaxation even though they still count as collisions.[1][2]

The compression is not sufficient for every calculation. LXCat notes that using elastic momentum-transfer and inelastic total cross sections as angular inputs is a common assumption in homogeneous solvers, while full differential data remain relevant to testing its accuracy, especially outside those conditions. NIST's solid-electron context also uses model-specific atomic scattering data. A transport result inherits these approximations and cannot be inferred from \(\sigma_{\mathrm{mt}}\) alone.[1][2]

Abstract Reasoning

At one energy, start with \(d\sigma/d\Omega\). If the question is “how often does elastic scattering occur?”, integrate with weight $1$. If it is “how rapidly is incident-direction momentum randomized under this elastic model?”, integrate with \(1-\cos\theta\). A forward-biased DCS may yield a large total event cross section but a much smaller momentum-transfer area; a backward-biased one can reverse that relation. This conclusion follows directly from the two integrals, not from a claim that total cross section is measured incorrectly.[1][2]

Then ask whether the derived scalar is the one the downstream solver actually needs. Does its collision set also account for inelastic channels? Is the beam monoenergetic or does it have an energy distribution? Is a two-term angular approximation defensible? Do electron–atom values adequately model the solid of interest? Such questions determine whether a formally correct \(\sigma_{\mathrm{mt}}\) supports a valid physical inference.[1][2]

Knowledge Transfer

The exact weighted-moment operation transfers from gas-electron scattering data to calculated electron–atom data used in solid-electron transport: the projectile, target, elastic DCS, angle and first-moment weight can all be mapped. The gas case plugs into electron-swarm/Boltzmann models; the solid case plugs into electron-trajectory and surface-signal models. The shared area is a derived scattering statistic, not a guarantee that the same total transport equation, energy range or free-particle potential applies.[1][2]

The broader lesson is a legitimate aggregation: select a weight that preserves the functional aspect relevant to the question and recognize what was discarded. Here the weight is physically fixed by longitudinal momentum projection; elsewhere a weighted integral might preserve a different feature. That analogy explains a proposed edge to Aggregation without promoting every weighted statistic to a momentum-transfer cross section.[1][2]

Examples

Electron–neutral gas transport

LXCat's complete electron–neutral collision sets include an elastic momentum-transfer cross section and separate angle-integrated cross sections for inelastic channels. For a specified gas species and electron energy, an elastic DCS where available is reduced with \(1-\cos\theta\); a Boltzmann or Monte Carlo treatment can then use that scalar alongside other channel data when predicting swarm transport. The dataset may be calculated or inferred under a model, and different detailed sets may fit the same swarm evidence. This example is not a claim that one scalar alone fixes mobility.[1]

Mapped back: electron and neutral at specified energy → elastic DCS → angle from electron incidence → \(1-\cos\theta\) → energy-indexed effective area → one input to gas-plasma transport model.

Electron transport in a solid-modeling context

NIST tabulates calculated electron elastic DCSs, total cross sections and transport cross sections for isolated atoms of selected elements; these data are used as inputs to modeling electrons moving through solids and surface-analysis signals. For an element and chosen energy, its transport tabulation is the angular momentum-loss moment of its atom DCS, not the unweighted total. The setting is unlike a low-temperature gas plasma: its target model, energy scale and downstream observable differ, while the angular weighting relation survives.[2]

Mapped back: electron plus specified element/energy → calculated atom elastic DCS → angle from incidence → \(1-\cos\theta\) → NIST transport area → input to solid-electron trajectory/signal model.

Rejected near miss

Integrate a DCS with no weight, call the result a “transport cross section,” then infer directional relaxation from the name. The calculation yields total elastic cross section. Without the weighted angular moment—or an explicit theorem equating it under special isotropy—the defining role is missing.[1]

Structural Tensions

Compact first moment versus angular detail. The scalar is economical for approximations whose response depends chiefly on longitudinal directional relaxation. It cannot reconstruct side scattering, higher angular moments or detailed trajectories, and those losses matter when a solver's symmetry assumptions fail. Diagnostic: Which transport equation and symmetry or closure actually make this first moment sufficient?[1][2]

Event incidence versus directional degradation. Total elastic cross section answers how much scattering occurs; the momentum-transfer moment asks how effectively those events change the forward component. A strongly forward-peaked process can score high on the first and low on the second. Diagnostic: Is the target inference about collision incidence or momentum relaxation, and is the chosen angular weight aligned with that inference?[1][2]

Structural–Framed Character

Its character: the integral is a structural mathematical operation within a framed physical model. Once channel, energy, DCS, frame and elastic kinematic convention are fixed, the weighted area is determinate. The word “momentum” points to a physical quantity, not an evaluative preference; the decision to retain longitudinal relaxation rather than all angular information is purpose-specific. Gas-plasma and solid-electron researchers can recognize the same formula without importing one institution's vocabulary, while a raw “transport” label must still be checked against local conventions.[1][2]

Across the five criteria: vocabulary travels among scattering subfields but not as a generic social metaphor; evaluative weight is absent; institutional origin of LXCat or NIST is not constitutive; human-practice dependence lies in measurement/model choice rather than the angular identity; and import versus recognize requires the actual \(1-\cos\theta\) moment, not merely a similar-sounding rate. This supports a domain-specific physics entry below a broader aggregation pattern.[1][2]

Structural Core vs. Domain Accent

The core is the differential elastic cross section integrated against the incident-direction loss factor. Electron or other projectile, gas or atom-in-solid model, angular coordinate system, computational solver and tabulation units are accents or parameter choices—provided the same collision channel and kinematic assumptions are declared. NIST's name “transport cross section” and LXCat's name “momentum transfer” can denote this same core, but LXCat's effective value that includes inelastic channels is a conventionally different input. The generic “transport cross section” surface should not be added as an unconditional alias.[1][2]

This entry is a kind of Aggregation. The angular differential cross section is deliberately reduced to one momentum-weighted summary.

Relationships to Other Abstractions

Local relationship map for Momentum-Transfer Cross SectionParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Momentum-TransferCross SectionDOMAINPrime abstraction: Aggregation — is a kind ofAggregationPRIME

Current abstraction Momentum-Transfer Cross Section Domain-specific

Parents (1) — more general patterns this builds on

  • Momentum-Transfer Cross Section is a kind of Aggregation Prime

    The angular differential cross section is deliberately reduced to one momentum-weighted summary.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Momentum-Transfer Cross Section sits in a sparse region of the domain-specific corpus (69th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Domain-Specific Measurement Parameters (36 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With

  • Total elastic cross section: angular weight $1$; collision incidence rather than directional loss.[1]
  • Viscosity cross section: angular weight \(1-\cos^2\theta\) in the LXCat convention.[1]
  • Backscattering cross section: reverse-direction response under its own normalization, not an integral over all angles.
  • Effective momentum-transfer cross section: may add inelastic channels approximately under a model; not the pure elastic angular moment.[1]
  • Differential cross section with momentum transfer as an axis: a DCS reparameterized by momentum-change variable is not automatically the integrated \(\sigma_{\mathrm{mt}}\).[2]
  • Collision frequency or mobility: rates or transport coefficients derived with density, distribution and model assumptions; not cross-section areas.[1]

References

[1] Leanne C. Pitchford and coauthors, “LXCat: an Open-Access, Web-Based Platform for Data Needed for Modeling Low Temperature Plasmas”, Plasma Processes and Polymers 14 (2017), §3 Table 1 and §§3.1–3.2. Original article by LXCat platform contributors; use the paragraphs specifying total, momentum-transfer, viscosity and effective weights and the modeling limits. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y ↩z ↩27 ↩28

[2] A. Jablonski, F. Salvat and C. J. Powell, NIST Electron Elastic-Scattering Cross-Section Database, Version 4.0: User's Guide, NIST NSRDS 64 (2016), §1 p. 1, §2.2 pp. 9–12 and §3.2 pp. 18–19, especially eqs. (14)–(15). The database's numerical contents are not reproduced here. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y