Isospin

Isospin is the approximate $SU(2)$ flavor symmetry rotating the up and down quarks. It is accurate because their masses are close and both are small compared with the QCD scale.

Two pions, each of isospin one, have a nine-dimensional charge-state space. Clebsch-Gordan coefficients decompose it into an isospin-two quintet, an isospin-one triplet and an isospin-zero singlet. Under exchange of the two particle slots, the $I=2,0$ states are symmetric and the $I=1$ states are antisymmetric. Because pions have zero spin and obey bosonic exchange symmetry, the $I=1$ two-pion channel requires odd relative orbital angular momentum, while $I=0,2$ require even orbital angular momentum.

Nuclear quantum states in different nuclear isobars belonging to the same isospin multiplet, with equal total isospin and approximately the same spatial and spin structure. An isospin raising or lowering operator connects the states in the limit of exact isospin symmetry. Nuclear Coulomb energy and other isospin-breaking effects shift their energies and modify this ideal relationship.

Suppose a transition operator obeys $[I_3,H]=H/2$ and $[I_{+},H]=0$, and acts on the lower component of an isospin-half kaon. Its output has third component zero, and $I_{+}^{2}H|K⟩=H{I}_{+}^2|K⟩=0$. This excludes an isospin-two output. A spinless parent decaying to two spinless pions has zero relative orbital angular momentum, so Bose-Einstein statistics excludes the antisymmetric isospin-one sector. The output is therefore the two-isovector singlet. In the equal-mass isospin limit, charged and neutral two-pion decay widths have ratio two, including the identical-particle counting.

Two spin-one isospin multiplets have a unique normalized singlet. In the displayed formula each ket on the right records the two third components, not their individual total isospins. The total raising operator annihilates this combination because the coefficients of $|1,0⟩$ and $|0,1⟩$ cancel. Its exchange symmetry is positive; the five-dimensional total-isospin-two sector is also symmetric and the three-dimensional total-isospin-one sector antisymmetric.

An isospin multiplet is an irreducible finite-dimensional SU(2) representation with labels $I,M$, where $I\ge 0$, $2I$ is an integer and $M=I,I-1,\dots ,-I$. With Hermitian generators, the lowering coefficient is $\sqrt{(I+M)(I-M+1)}$. The quadratic Casimir operator has value $I(I+1)$, fixing all ladder norms and the termination at $M=-I$.

An isospin-invariant decay amplitude is an intertwining map from the initial isospin multiplet into the final tensor product. The Clebsch-Gordan decomposition for SU2 contains each allowed final isospin once. Commutation with the third generator and the quadratic Casimir operator places the image in the matching coupled state. Commutation with the ladder operators makes its scalar coefficient independent of the magnetic component. Orthogonality of the Clebsch-Gordan coefficients then gives a summed squared amplitude $|a{|}^2$. A physical decay rate additionally includes phase space and spin sums, which can be absorbed into a common reduced coefficient when exact isospin symmetry makes them identical between channels.

With positive ladder coefficients, the second-axis half-turn is $e^{-i\pi I_2}|I,M⟩=(-1{)}^{I-M}|I,-M⟩$. Conjugation sends $I_{\pm }$ to $-I_{\mp }$, making consecutive phases differ by a minus sign. The highest-state rotation formula, expanded through its last nonzero lowering term, fixes the highest-state phase to one.

Weak isospin labels representations of the SU(2) factor in the electroweak gauge group. A left-handed lepton doublet has $T=1/2$ with third-component eigenvalues $+1/2,-1/2$; a right-handed charged-lepton singlet has $T=0$. Along with electric charge, the third component determines the weak neutral current through $T^3-Q{\sin }^2{\theta }_W$.

The product of charge conjugation and a half-turn in isospin is G parity. Both conjugations act on the isospin generators by $(J_1,J_2,J_3)\mapsto (-J_1,J_2,-J_3)$, so their product commutes with every generator. The Schur lemma therefore gives a common G-parity value throughout each irreducible isospin multiplet on which the operator acts internally. It can classify a charged multiplet even though individual charged states are not eigenstates of charge conjugation.

For integer isospin $I$ and neutral-state charge conjugation value $\eta$, the entire isospin multiplet has $G=\eta (-1{)}^I$. Both charge conjugation and the second-axis half-turn reverse the first and third generators, so their product commutes with the full isospin algebra. Evaluate it on the $M=0$ state and use ladder connectivity to obtain the value for every $M$.

Each pion has G parity minus one, so an $n$-pion state has $G=(-1{)}^n$. In the isospin-symmetric strong interaction, a parent with positive G parity can decay only to an even pion multiplicity, and a negative parent only to an odd multiplicity. The rule does not forbid small isospin-breaking or electromagnetic amplitudes.

The neutral pion has positive charge conjugation eigenvalue and its spin-one isospin half-turn has eigenvalue minus one. Hence the G parity eigenvalue is minus one. Since G parity commutes with isospin, both charged members have the same value. Charged-state phase changes alter individual charge-conjugation and rotation formulas together, leaving this eigenvalue unchanged.

An isorotation is a global transformation generated by isospin. In the pion Skyrme model it acts as $U\mapsto AU{A}^{†}$, with constant $A\in SU(2)$, rotating the three pion components while preserving the scalar component. The matrices $A$ and $-A$ induce the same classical rotation, so this action factors through $SO(3)$; quantum states may carry half-integer isospin on its $SU(2)$ cover. This global internal rotation is distinct from a spatial rotation or a local gauge transformation.