Fermion mass diagonalization and the CKM matrix
Derive charged-current mixing from mismatched left-handed mass rotations and separate observable flavor parameters from field conventions.
On this page
The question for today
What you’ll understand
- Derive the CKM matrix with an explicit weak-to-mass basis convention.
- Distinguish singular-value diagonalization from diagonalizing a Hermitian matrix.
- Count physical angles and phases after quark rephasings.
Flavor labels refer to how particles participate in weak interactions, whereas propagation and rest energies identify mass eigenstates. These two descriptions need not use the same basis. The CKM matrix records their mismatch for quarks. Deriving it from the mass terms makes its conjugations unambiguous and explains why arbitrary changes of field phases cannot become experimental signals. Throughout this lesson natural units mean hbar and the speed of light are one; mixing matrices themselves are dimensionless.
Specify the direction of every basis change
Write the weak-basis fields as and , with analogous equations for down-type quarks. The unprimed columns contain mass eigenfields. A general complex mass matrix connects left and right chiralities, so two different unitary matrices are required. Choose , where D has real nonnegative masses on its diagonal. This is a singular-value decomposition, not an assumption that the original mass matrix is Hermitian. Degenerate masses permit further basis freedom inside their common eigenspace.
Insert these definitions into the weak charged current before simplifying. The up-quark conjugate supplies the dagger on its left rotation; the down field supplies its undaggered rotation. Thus multiplies the down-type mass column. Other authors may define their rotation matrices in the opposite direction, producing apparently reversed daggers with identical physics. A useful diagnostic is to reconstruct the current from the stated field definitions, rather than copying the matrix formula without its convention.
Unitarity constrains amplitudes rather than hadronic rates alone
Because both left rotations are unitary, V is unitary. Each row and column therefore has squared magnitudes summing to one, and distinct rows or columns have vanishing inner products. A weak transition amplitude contains a particular CKM entry; an isolated rate contribution contains its squared magnitude. Measured hadron decay rates additionally include phase space, strong-interaction matrix elements, electroweak factors, and sometimes competing amplitudes. Their raw branching fractions do not form a CKM probability table without accounting for these ingredients.
A two-family illustration isolates the geometry. Define a real rotation R whose first row is cosine, sine and whose second row is minus sine, cosine. If the up and down left rotations have angles alpha and beta, the current contains the rotation through beta minus alpha. A common rotation of both sectors disappears. Mixing consequently measures a relative orientation, not the size of either diagonalizing rotation considered separately. This cancellation is also why choosing an up-diagonal weak basis does not remove physical down-sector mixing.
Remove phase conventions before counting CP parameters
A general three-dimensional unitary matrix has nine real parameters: three rotation angles and six phases. Rephasing each of the six quark mass fields changes row or column phases of V. One common phase transformation changes no matrix entry, leaving five useful phase removals. The remaining physical parameters are three angles and one CP phase, provided the relevant masses are nondegenerate. For two families the analogous phase count leaves no charged-current CP phase. Complex-looking entries alone therefore do not demonstrate CP violation.
The right rotations remain essential for constructing positive masses, even though they cancel from the displayed charged current. Likewise, mass hierarchies and mixing angles are separate empirical properties; small masses do not algebraically require small CKM angles. Precision flavor analysis combines several processes to determine the matrix and test unitarity under declared theoretical assumptions. A fitted discrepancy can reflect an incorrect hadronic input, experimental uncertainty, or additional interactions. Identifying its origin requires more information than a single matrix element extracted from one decay channel.
Work it through
Subtract two left-handed angles
In a hypothetical two-family sector, let and . Find V and the relative squared charged-current weight of its first-row, second-column transition when all other factors are equal.
Our weak-to-mass convention gives . Subtracting in the opposite order would reverse the off-diagonal signs.
Thus the first row is and the second is . These rows have unit norm and zero inner product.
The requested amplitude entry is one half, giving a squared weight of one quarter. Its partner in the same row has weight three quarters, so their sum checks unitarity.
The relative mixing angle is 30 degrees and the selected charged-current weight is 1/4 under the equal-dynamics assumption.
The living notebook
Explore the model
Interactive lab · quantum
An amplitude angle determines squared weights
Reset and set BOTH Mixing angle θ = 60° and Relative phase φ = 0°; predict P(1) = 0.25. For a fresh comparison, Reset and set θ = 90° and φ = 0°; predict P(1) = 0.50. Inspect the probability readouts before playback.
- Probability |0⟩
- 71.1%
- Probability |1⟩
- 28.9%
- Bloch (x, y, z)
- 0.78, 0.45, 0.42
A normalized two-level pure state, not an interacting field simulation. Playback adds one radian per second to the relative phase; basis probabilities stay fixed. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.
Look forThe lab uses cos(θ/2) and sin(θ/2), so its angle is twice the real two-component amplitude angle. It illustrates normalization, not a CKM fit or quark mass diagonalization.
Pause & reason
Check your understanding
Your turn
Try it yourself
Take and . Calculate the first-row squared weights. Does multiplying the entire second column by a phase create observable CP violation in this two-family sector?
Take a moment to make a prediction or work it out on paper. Open a hint when you need one.
Give me a hint
Square magnitudes, then identify a phase redefinition of the second down-type mass field.
Compare with the solution
The matrix V equals the stated down rotation. Its first-row weights are and , summing to one. Multiplying a column by a unit-modulus phase leaves both weights and all rephasing-invariant observables unchanged. That column phase can be absorbed into the corresponding mass field. With only two nondegenerate families, the charged-current mixing matrix has no physical CP phase; a complex entry generated by this convention change is not evidence for CP violation.
Keep the idea
Take it with you
- CKM mixing is the mismatch between left-handed quark mass rotations.
- Field definitions fix dagger placement; rephasing cannot change observables.
- Three nondegenerate families permit one physical charged-current CP phase.
Read further
Original sources and deeper explanations to follow at your own pace.
Optional progress, saved in this browser when available. Revisit any lesson at any time.
Lesson not marked complete.