Baryon Chiral Perturbation Theory

• BChPT is the baryonic extension of chiral perturbation theory, incorporating nucleons and other baryons into the effective field theory framework built on spontaneous chiral symmetry breaking.
• It employs effective Lagrangian methods and tailored renormalization schemes, such as the EOMS approach, to systematically restore chiral power counting despite the baryon mass scale.
• BChPT has been successfully applied to calculate nucleon masses, analyze pion–nucleon scattering, and advance SU(3) phenomenology, demonstrating improved convergence over other formulations.

Searching arXiv for recent and foundational BChPT sources to ground the article. arXiv search results relevant to BChPT:

• "Nucleon mass in covariant baryon chiral perturbation theory at leading two-loop order" (Liang et al., 4 Aug 2025)
• "Pion-Nucleon Scattering in Baryon Chiral Perturbation Theory combined with the 1/Nc Expansion" (Jayakodige et al., 5 Jun 2025)
• "Chiral perturbation theory" (Meißner, 2024)
• "Meson-baryon scattering up to the next-to-next-to-leading order in covariant baryon chiral perturbation theory" (Lu et al., 2018)
• "Baryon chiral perturbation theory" (Scherer, 2011)
• "Recent developments in SU(3) covariant baryon chiral perturbation theory" (Geng, 2013) Baryon chiral perturbation theory (BChPT) is the baryonic extension of chiral perturbation theory, the low-energy effective field theory of QCD built from the symmetry pattern $$\mathrm{SU}(N)_L\times \mathrm{SU}(N)_R \to \mathrm{SU}(N)_V$$, with the pseudoscalar octet identified as the Nambu–Goldstone bosons of spontaneous chiral symmetry breaking. In this framework the Goldstone bosons remain the light degrees of freedom, while the lowest-lying baryons are added as explicit matter fields and observables are organized in an expansion in small momenta and quark masses. Relative to the purely mesonic theory, the central complication is that the baryon mass is $\mathcal O(p^0)$: it does not vanish in the chiral limit, and relativistic loop graphs therefore generate analytic terms that violate naive chiral power counting unless a specific prescription is adopted (Meißner, 2024, Geng, 2013).

1. Symmetry basis and effective-field-theory construction

For the light quarks $u,d,s$, the QCD Lagrangian can be written as

$${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$

and in the chiral limit $m_u=m_d=m_s=0$ it has global $SU(3)_L\times SU(3)_R$ symmetry. The spontaneous breaking pattern

$SU(3)_L \times SU(3)_R \to SU(3)_V$

produces eight Goldstone bosons, identified with $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$. Explicit symmetry breaking from nonzero light-quark masses generates their small masses, with the leading relation $M_\pi^2 = B(m_u+m_d)$ and $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$ (Meißner, 2024).

The EFT construction follows the standard chiral logic: one writes the most general Lagrangian built from the relevant low-energy fields and consistent with the symmetries. In the meson sector,

$\mathcal O(p^0)$0

with chiral assignments $\mathcal O(p^0)$1, $\mathcal O(p^0)$2, and $\mathcal O(p^0)$3. BChPT preserves this symmetry structure while extending the field content to include baryons as matter fields. It is therefore the low-energy EFT of QCD for processes involving both Goldstone bosons and baryons, rather than a separate theory with different symmetry principles (Meißner, 2024).

This construction also clarifies why BChPT is richer than purely mesonic CHPT. Baryons carry spin and isospin, introduce additional low-energy constants, and open nontrivial processes such as $\mathcal O(p^0)$4 scattering, pion photoproduction, axial form factors, baryon mass expansions, and low-energy theorems. At the same time, the hard baryon mass scale makes renormalization and counting substantially subtler than in the mesonic sector (Meißner, 2024).

2. One-baryon sector and chiral Lagrangians

In the two-flavor theory, the nucleon isodoublet $\mathcal O(p^0)$5 transforms nonlinearly under chiral rotations through a compensator field $\mathcal O(p^0)$6. The covariant derivative and axial building block are

$\mathcal O(p^0)$7

The leading pion–nucleon Lagrangian is

$\mathcal O(p^0)$8

or equivalently, in the one-nucleon presentation,

$\mathcal O(p^0)$9

Its parameters are the nucleon mass and axial coupling in the chiral limit (Meißner, 2024, Scherer, 2011).

In the three-flavor theory, the corresponding leading meson–baryon Lagrangian is

$u,d,s$0

with $u,d,s$1 the octet mass in the chiral limit and $u,d,s$2 the axial couplings. Explicit symmetry breaking enters at next order through terms such as

$u,d,s$3

which generate the leading quark-mass dependence of the octet masses and underlie relations such as the Gell-Mann–Okubo relation (Meißner, 2024).

The baryon propagator introduces the defining difference from the meson theory. In the one-baryon sector, meson masses and derivatives count as $u,d,s$4, while the baryon mass is $u,d,s$5. This implies that tree graphs start at $u,d,s$6 and one-loop graphs enter at $u,d,s$7 once power counting is restored, but only after a suitable renormalization prescription has removed the analytic lower-order pieces generated by relativistic loops (Geng, 2013, Meißner, 2024).

3. Power counting, renormalization, and competing formulations

The baryonic power-counting formula for a diagram with a single baryon line is

$u,d,s$8

with $u,d,s$9 the number of loops, $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$0 mesonic vertices of order $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$1, and $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$2 meson–baryon vertices of order $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$3. Formally this is the baryonic analogue of the mesonic counting, but in relativistic perturbation theory loop integrals produce analytic contributions of too-low chiral order because the baryon mass remains nonzero in the chiral limit (Meißner, 2024).

A standard illustration is the nucleon mass at $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$4. After $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$5 subtraction, the relativistic one-loop self-energy still contains an analytic $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$6 term in addition to the expected nonanalytic $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$7 piece. Since the unwanted term is analytic in the quark mass, it can be absorbed into a finite renormalization of a low-energy constant such as $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$8. The general lesson is that, in BChPT, renormalization must remove not only ultraviolet divergences but also specific finite analytic terms that spoil counting (Scherer, 2011).

Three standard remedies are used. Heavy-baryon CHPT rewrites the baryon momentum as $${\cal L}_{\rm QCD} = {\cal L }_{\rm QCD}^0 - \bar q {\cal M} q, \qquad q=\begin{pmatrix}u\ d\ s\end{pmatrix},$$9, splits the baryon field into large and small components, and integrates out the latter. This restores power counting because the large baryon mass disappears from the leading propagator, but the $m_u=m_d=m_s=0$0 expansion can distort analytic structure and recoil effects. Infrared regularization preserves covariance by decomposing each loop into an infrared singular part $m_u=m_d=m_s=0$1, which carries the nonanalytic low-energy physics, and a regular part $m_u=m_d=m_s=0$2, which is analytic and absorbed into low-energy constants; however, the discarded regular part contains more than just power-counting-breaking terms and can generate unphysical cuts. The extended-on-mass-shell scheme keeps the full covariant loop result and subtracts only the analytic pieces that violate chiral order,

$m_u=m_d=m_s=0$3

thereby preserving both power counting and the relativistic analytic structure (Geng, 2013, Scherer, 2011).

Within the literature summarized here, the covariant EOMS formulation is repeatedly singled out as the most consistent practical scheme. It is described as satisfying analyticity and symmetry constraints, recovering power counting by subtracting only PCB terms, and tending to converge faster than heavy-baryon and infrared approaches, particularly in observables sensitive to SU(3) breaking (Geng, 2013).

4. Covariant EOMS at two loops: the nucleon mass as a precision benchmark

A recent benchmark development is the calculation of the nucleon mass in manifestly relativistic BChPT up to leading two-loop order, $m_u=m_d=m_s=0$4, using the EOMS renormalization scheme (Liang et al., 4 Aug 2025). In this formulation the nucleon mass is defined by the pole of the full propagator,

$m_u=m_d=m_s=0$5

so that

$m_u=m_d=m_s=0$6

Here $m_u=m_d=m_s=0$7 is the nucleon mass in the chiral limit, $m_u=m_d=m_s=0$8 the self-energy, and $m_u=m_d=m_s=0$9 the wave-function renormalization constant.

The perturbative expansion is organized as

$SU(3)_L\times SU(3)_R$0

with

$SU(3)_L\times SU(3)_R$1

The use of $SU(3)_L\times SU(3)_R$2 is chosen so that mass insertion diagrams are automatically included. Up to $SU(3)_L\times SU(3)_R$3, the calculation contains 2 tree diagrams, 5 one-loop diagrams, and 12 two-loop diagrams, for a total of 19 Feynman diagrams. Integration-by-parts identities reduce the loop integrals to 3 one-loop and 10 two-loop master integrals (Liang et al., 4 Aug 2025).

The central technical issue is the classification and removal of power-counting-breaking terms. The two-loop analysis uses dimensional counting analysis, equivalent to the method of regions, to separate loop contributions into non-local PCB terms, local one-loop PCB terms, and local two-loop PCB terms. The non-local terms cancel against one-loop sub-diagrams, while the local ones are absorbed into counterterms. EOMS renormalization is then implemented in two steps: ultraviolet poles are absorbed into bare low-energy constants in dimensional regularization, including double poles at two loops, and finite PCB subtraction terms are added to define EOMS-renormalized couplings. The resulting $SU(3)_L\times SU(3)_R$4 nucleon-mass representation preserves analyticity, obeys correct power counting, and is renormalization-scale independent (Liang et al., 4 Aug 2025).

The full EOMS result can also be expressed as the small-$SU(3)_L\times SU(3)_R$5 series

$SU(3)_L\times SU(3)_R$6

with $SU(3)_L\times SU(3)_R$7 and $SU(3)_L\times SU(3)_R$8 scheme independent by renormalization-group arguments. The paper stresses, however, that truncating the full covariant expression to a finite chiral series alters the analytic structure; the truncated series is therefore not identical to the full EOMS result (Liang et al., 4 Aug 2025).

At the physical pion mass the decomposition

$SU(3)_L\times SU(3)_R$9

shows that the leading two-loop $SU(3)_L \times SU(3)_R \to SU(3)_V$0 contribution is about $SU(3)_L \times SU(3)_R \to SU(3)_V$1 MeV. The same calculation finds excellent agreement with lattice QCD data for $SU(3)_L \times SU(3)_R \to SU(3)_V$2, and the full EOMS curve performs better than truncated EOMS or truncated IR representations as the pion mass increases because truncation changes the analytic structure. Within the terms of that work, this establishes two-loop relativistic BChPT as a precision tool for the nucleon mass and a robust foundation for chiral extrapolation (Liang et al., 4 Aug 2025).

5. SU(3), decuplet degrees of freedom, and phenomenology

Three-flavor BChPT is both phenomenologically necessary and technically more demanding. Kaon and eta masses are sizable, odd and even chiral orders both appear at low order, and the number of low-energy constants grows rapidly. The literature therefore places particular emphasis on covariant SU(3) BChPT with EOMS, where analyticity is preserved and convergence is generally improved relative to heavy-baryon and infrared formulations (Geng, 2013).

A recurrent theme is the role of the lowest-lying decuplet resonances, especially the $SU(3)_L \times SU(3)_R \to SU(3)_V$3. Because the $SU(3)_L \times SU(3)_R \to SU(3)_V$4 splitting is about $SU(3)_L \times SU(3)_R \to SU(3)_V$5 MeV, explicit decuplet degrees of freedom can be numerically important, and their inclusion is usually organized with $SU(3)_L \times SU(3)_R \to SU(3)_V$6-counting and consistent spin-$SU(3)_L \times SU(3)_R \to SU(3)_V$7 couplings to remove unphysical spin-$SU(3)_L \times SU(3)_R \to SU(3)_V$8 components. In $SU(3)_L \times SU(3)_R \to SU(3)_V$9 phenomenology, a covariant EOMS analysis including the $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$0 in $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$1-counting yielded $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$2 from phase-shift fits up to $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$3 GeV, while the same modern framework was used to argue that a relatively large $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$4 MeV is not incompatible with a small strange scalar content, with $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$5 in the quoted SU(3) analysis (Camalich, 2013).

Octet baryon masses and sigma terms form another major application. Covariant SU(3) BChPT with EOMS has been pushed to $(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$6 and N$(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$7LO, with self-consistent finite-volume corrections. A representative N$(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$8LO analysis employed 19 low-energy constants, achieved clear order-by-order improvement from NLO to NNLO to N$(\pi^0,\pi^\pm,K^0,\bar K^0,K^\pm,\eta)$9LO, and found $M_\pi^2 = B(m_u+m_d)$0 for a restricted set of $M_\pi^2 = B(m_u+m_d)$1 lattice data. Using the fitted mass dependence and the Feynman–Hellmann relations, it predicted $M_\pi^2 = B(m_u+m_d)$2 and $M_\pi^2 = B(m_u+m_d)$3 (Ren et al., 2012). Related analyses emphasize that finite-volume corrections are crucial, that different lattice collaborations can be mutually consistent when analyzed in a common covariant EFT, and that the fitted low-energy constants are natural in size (Geng, 2013).

Not all observables display equally good convergence. Manifestly covariant SU(3) BChPT applied to QCDSF octet-mass fan plots showed that certain symmetry-breaking combinations of low-energy constants, notably $M_\pi^2 = B(m_u+m_d)$4 and $M_\pi^2 = B(m_u+m_d)$5, are cleanly constrained near the SU(3)-symmetric point, whereas singlet-sector quantities remain poorly determined. The same study stressed slow convergence around the physical strange-quark mass and argued that reliable sigma-term extraction is not possible in that setup without more data at smaller average quark masses (Bruns et al., 2012).

Meson–baryon scattering provides a complementary test. A covariant SU(3) BChPT analysis through NNLO, with EOMS removal of power-counting-breaking terms, performed the first combined study of $M_\pi^2 = B(m_u+m_d)$6 and $M_\pi^2 = B(m_u+m_d)$7 scattering data and found a reasonable description of the phase shifts. It also showed that experimental baryon masses and $M_\pi^2 = B(m_u+m_d)$8 and $M_\pi^2 = B(m_u+m_d)$9 scattering data can be fitted simultaneously at this order, providing a consistency check on covariant BChPT. In that framework, leading virtual-decuplet contributions improve the $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$0 phase shifts near the $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$1 peak, while their effect on $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$2 phase shifts is negligible (Lu et al., 2018).

Further static and transition observables reinforce the same pattern. Reviews of ground-state baryon properties report that EOMS improves the description of octet magnetic moments relative to heavy-baryon and infrared approaches, that decuplet magnetic moments can be predicted at NLO once a single low-energy constant is fixed from $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$3, and that the hyperon vector coupling $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$4 is especially clean because the Ademollo–Gatto theorem forbids unknown local contributions up to $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$5 (Martin-Camalich et al., 2010, Geng, 2013).

6. Extensions, generalizations, and open directions

BChPT has been extended in several directions beyond the standard one-baryon SU(2) and SU(3) sectors. One major line combines the chiral expansion with the $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$6 expansion. In this framework, baryons obey an emergent spin-flavor symmetry in the large-$B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$7 limit, and the natural linking of the two expansions is the $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$8-expansion, $B = |\langle 0|\bar q q|0\rangle|/F_\pi^2$9. For pion–nucleon scattering, this enforces the inclusion of both $\mathcal O(p^0)$00 and $\mathcal O(p^0)$01 as active degrees of freedom and improves convergence relative to ordinary BChPT without an explicit dynamical $\mathcal O(p^0)$02, which is described there as inconsistent with the constraints of $\mathcal O(p^0)$03 scaling (Jayakodige et al., 5 Jun 2025). A related heavy-baryon large-$\mathcal O(p^0)$04 analysis of the baryon axial current used the Jenkins–Manohar chiral Lagrangian and found large cancellations between octet and decuplet loop contributions, consistent with spin-flavor symmetry expectations (Hernandez-Ruiz, 2014).

Another direction concerns baryon–baryon scattering. Manifestly Lorentz-invariant SU(3) BChPT formulated with time-ordered perturbation theory defines the effective potential as the sum of two-baryon irreducible time-ordered diagrams and leads to a coupled-channel generalization of the Kadyshevsky equation. The resulting integral equations have milder ultraviolet behavior than their nonrelativistic analogues, and at leading order the potentials are perturbatively renormalizable with unique solutions in all partial waves, provided corrections beyond leading order are treated perturbatively (Baru et al., 2019). Applications to $\mathcal O(p^0)$05 scattering further argue that at least part of the higher-order contributions in $\mathcal O(p^0)$06 and $\mathcal O(p^0)$07 must be treated nonperturbatively, while still allowing BPHZ-type subtractive renormalization (Ren et al., 2019).

Precision electroweak extensions also exist. A relativistic SU(2) pion–nucleon Lagrangian with virtual photons and light leptons has been constructed through fourth chiral order, using electromagnetic and weak spurions together with explicit leptonic building blocks. Its purpose is to provide the EFT counterterm structure required for radiative corrections to semileptonic processes such as neutron beta decay and muon capture (Supanam et al., 2010). Other variants embed axions into heavy-baryon ChPT as external axial sources, yielding systematic axion–baryon couplings in SU(2) and SU(3), and extend covariant BChPT to heavy-baryon systems such as spin-$\mathcal O(p^0)$08 doubly charmed baryons, where EOMS again removes power-counting-breaking terms from relativistic loops (Vonk, 2022, Shi et al., 2021).

Two recurring controversies remain visible across these developments. The first is convergence, especially in SU(3), where several analyses stress that the strange sector is more difficult and that quantities such as $\mathcal O(p^0)$09 remain sensitive to higher orders and additional low-energy constants (Camalich, 2013, Bruns et al., 2012). The second is the treatment of explicit resonances and additional degrees of freedom: the decuplet, vector and axial-vector mesons, virtual photons, leptons, or large-$\mathcal O(p^0)$10 multiplets can materially improve phenomenology, but only when incorporated with a counting and renormalization scheme compatible with the EFT structure (Scherer, 2011, Geng, 2013). Taken together, these results define BChPT not as a single fixed computational recipe but as a family of symmetry-based low-energy EFTs whose predictive power depends decisively on the chosen realization of power counting, analytic structure, and active degrees of freedom.