Right-Handed Neutrinos: The Ghost Particles of the Standard Model

The Standard Model has a peculiar asymmetry. For every charged fermion — electron, muon, tau, quarks of every flavour — there exists both a left-handed component and a right-handed component. The two are independent fields with different gauge transformations under SU(2) × U(1), but both are physically real. We measure them, count them, see them in every reaction.

For the neutrino, only the left-handed component appears. The right-handed neutrino — if it exists at all — has never been observed in any process. Its absence is so striking that, in the original Standard Model written down in the 1970s, the field ${\nu }_R$ was simply omitted. The neutrino was a Weyl fermion: massless, with only one chirality, eternal partner of the electron in the SU(2) doublet.

The 1998 Super-Kamiokande discovery of neutrino mass changed this. Mass requires either a right-handed partner (Dirac mass) or a Majorana mass term that makes the particle its own antiparticle. Either way, the simple “no right-handed neutrinos” picture had to give. Some kind of right-handed neutrino — heavy, hidden, sterile — has to exist if the neutrino is massive.

Twenty-eight years later, we still haven’t directly observed one. Searches range from LHC collider signatures to kiloton-scale neutrinoless double beta decay experiments to precision flavour-changing-decay measurements. Every search has, so far, returned consistent with no signal. The right-handed neutrino remains the ghost particle of the Standard Model — required by what we know, but invisible to everything we can do.

This post lays out what right-handed neutrinos would be, why they are theoretically required, what we’ve ruled out, and where the searches stand.

What “right-handed” means in this context

A standard Dirac fermion is described by a four-component spinor field $\psi$. The chirality projectors $P_L=(1-{\gamma }^5)/2$ and $P_R=(1+{\gamma }^5)/2$ split it into two independent components: ${\psi }_L=P_L\psi$ (left-chirality) and ${\psi }_R=P_R\psi$ (right-chirality). For massless particles, these are independent helicity eigenstates; for massive ones, they are mixed by the mass term, but the chirality decomposition still applies algebraically.

In the Standard Model, the gauge transformations are different for the two chiralities. Left-chiral fields are organised into SU(2) doublets — for example, the electron field $L_e=({\nu }_e,e{)}_L$ — and have non-trivial weak isospin. Right-chiral fields are SU(2) singlets — $e_R$ alone, with no neutrino partner.

The “right-handed neutrino”, ${\nu }_R$, would be the missing partner: a Standard Model singlet (no charge under SU(2), U(1), or colour) whose only role is to pair with ${\nu }_L$ to form a complete Dirac fermion or to participate in Majorana mass generation.

Because ${\nu }_R$ has no gauge charges, it interacts with nothing in the standard Lagrangian except (potentially) the Higgs. It cannot be produced in any reaction governed by the Standard Model gauge couplings. It is invisible — sterile in the modern terminology — to every electromagnetic, weak, and strong process.

Why mass demands a right-handed neutrino

The Dirac mass term for a fermion is ${\mathcal{L}}_{\text{Dirac}}\supset m{\bar{\psi }}_L{\psi }_R+\text{h.c.}$ The mass term has a specific chirality structure: it couples a left-chiral field to a right-chiral one. For the electron, ${\psi }_L=e_L$ from the doublet, ${\psi }_R=e_R$ as the singlet. Both exist; the mass term works.

For the neutrino, ${\psi }_L={\nu }_L$ from the doublet exists. But ${\psi }_R={\nu }_R$ does not, in the minimal Standard Model. The corresponding mass term simply cannot be written.

Three options to give the neutrino a mass anyway:

Option A: Add the right-handed neutrino as a new field. Introduce ${\nu }_R$ as a Standard Model singlet. Couple it to the Higgs and the left-handed lepton doublet through the Yukawa interaction $y_{\nu }\bar{L}\tilde{\varphi }{\nu }_R$. After symmetry breaking, this gives a Dirac mass $m_{\nu }=y_{\nu }v/\sqrt{2}$. Required Yukawa: $y_{\nu }\sim 10^{-12}$ — small, perhaps unnaturally so.

Option B: Use the Majorana option. A Majorana fermion is its own antiparticle. The mass term has the structure $m{\bar{\psi }}^c\psi$ rather than $m{\bar{\psi }}_L{\psi }_R$. For the SU(2)-doublet ${\nu }_L$, this requires a non-renormalisable dimension-5 operator (the “Weinberg operator”): ${\mathcal{L}}_{\text{Weinberg}}=\frac{1}{\Lambda }(\bar{L}\tilde{\varphi })({\tilde{\varphi }}^{†}L)$ The operator generates a Majorana mass $m_{\nu }\sim v^2/\Lambda$ for the left-handed neutrino. No explicit right-handed field needed at low energy.

Option C: Combine A and B — the seesaw. Add right-handed neutrinos with both Dirac couplings to the Higgs and their own Majorana mass. The resulting low-energy effective theory contains the Weinberg operator with $\Lambda =M_R/y_{\nu }^2\cdot v^2$. This is the seesaw mechanism.

Options A and C explicitly involve right-handed neutrinos. Option B doesn’t, but the dimension-5 Weinberg operator is non-renormalisable; in any complete theory, it must come from integrating out a heavy state. The standard interpretation: that heavy state is the right-handed Majorana neutrino. So even Option B implicitly requires right-handed neutrinos at high energy.

Where the seesaw places them

The Type-I seesaw — the most economical and theoretically attractive variant — places the right-handed neutrino mass at: $M_R\sim \frac{(y_{\nu }v{)}^2}{m_{\nu }}$

Plugging in observed light-neutrino masses ($m_{\nu }\sim 0.1$ eV) and a Yukawa coupling of order unity ($y_{\nu }\sim 1$, comparable to the top quark), the heavy mass scale is: $M_R\sim \frac{(1\cdot 246\text{GeV}{)}^2}{0.1\text{eV}}\sim 6\times 10^{14}\text{GeV}$

This is just below the typical grand-unification scale ($10^{16}$ GeV). The natural location of the right-handed neutrino is therefore at scales comparable to grand-unified theories, far above any currently accessible experiment.

Variants of the seesaw place the heavy state at lower scales:

Type-II seesaw uses an SU(2) triplet Higgs instead of right-handed singlets. The relevant scale is the triplet’s vev rather than $M_R$.

Inverse seesaw uses two pseudo-Dirac states with very small Majorana mass splittings. The light-neutrino mass becomes $m_{\nu }\sim m_D^2\mu /M_R^2$, where $\mu$ is a small Majorana parameter. With small $\mu$, the right-handed states can be at TeV scales without producing too-heavy neutrinos.

Linear seesaw is similar in spirit to inverse but with different parameter dependence. Also allows TeV-scale right-handed states.

These variants are theoretically motivated and experimentally accessible. If realised in nature, the right-handed neutrinos could appear in LHC searches.

The collider hunt

If right-handed neutrinos exist at the TeV scale, they should appear in proton-proton collisions through their mixing with active neutrinos. The classic signature is “lepton number violating” same-sign-lepton final states: $pp\to W^{\ast }\to N{\ell }^{+}\text{followed by}N\to W{\ell }^{+}\to q{q}^{\prime }{\ell }^{+}$ producing two same-sign leptons plus jets, with no missing transverse energy (the heavy Majorana neutrino decays fully observably).

Searches at ATLAS, CMS, and LHCb have looked for this signature in all flavour combinations. The 2024 results put bounds:

• For $m_N\sim 100$ GeV: same-sign-lepton mixing $|V_{eN}{|}^2<10^{-5}$
• For $m_N\sim 1$ TeV: $|V_{eN}{|}^2<10^{-3}$
• For $m_N\sim$ few TeV: rate too small to constrain effectively

These bounds rule out a substantial portion of inverse-seesaw parameter space but leave room for low-mixing scenarios at the TeV scale.

The flavour-violation hunt

Right-handed neutrinos can mediate processes that violate lepton flavour at lower energies. The classic example: $\mu \to e\gamma$. In the seesaw with non-trivial PMNS-like mixing in the heavy sector, the rate for $\mu \to e\gamma$ scales as $|y_e^{\ast }{y}_{\mu }\log (M/M_W)/M^2|$, where $y_{e,\mu }$ are the Yukawa couplings to right-handed neutrinos and $M$ is the heavy mass.

The MEG-II experiment at PSI provides the strongest current bound: $\text{BR}(\mu \to e\gamma )<4.2\times 10^{-13}(90\%\text{ C.L.})$

This excludes substantial regions of the seesaw parameter space, especially for low-scale variants where the right-handed neutrinos are at TeV scales with large Yukawa couplings.

The Mu2e experiment at Fermilab and the COMET experiment at J-PARC push further with $\mu \to e$ conversion in atomic nuclei. These will reach $10^{-17}$ branching ratio sensitivity in the 2030s — an order of magnitude beyond MEG-II.

The cosmological hunt

Heavy right-handed Majorana neutrinos at very high masses would have left fingerprints in the early universe. Two principal signatures:

Leptogenesis: Heavy right-handed neutrinos decay out-of-equilibrium in the early universe with CP-violating asymmetry, generating a lepton-number excess that electroweak sphalerons convert to baryon-number excess. The required parameters give roughly the observed baryon-to-photon ratio of $6\times 10^{-10}$. The mass scale required is $M_R\gtrsim 10^9$ GeV. Direct test of leptogenesis is not possible, but consistency with observed BBN and CMB results is required.

Effective number of relativistic species ($N_{\text{eff}}$). If right-handed neutrinos are light enough to thermalise in the early universe, they contribute extra relativistic degrees of freedom and shift $N_{\text{eff}}$ from 3.04 to higher values. Planck data give $N_{\text{eff}}=2.99\pm 0.17$, ruling out fully thermalised eV-scale sterile neutrinos. CMB-S4 in the 2030s will tighten this constraint by another factor of three.

What would discovery look like

Several experimental signatures would directly establish right-handed neutrinos:

Same-sign-lepton resonance at LHC — would be a clear discovery of TeV-scale Majorana right-handed neutrinos. Decisive but only relevant if $M_R$ is in LHC reach.

$0\nu \beta \beta$ detection — would establish the Majorana nature of light neutrinos, indirectly supporting the seesaw scenario. The most likely first sign of right-handed neutrinos through their effects on the light sector.

$\mu \to e\gamma$ or $\mu \to e$ conversion observation — would constrain the heavy-sector flavour mixing structure and provide quantitative information about the right-handed Yukawa couplings.

Cosmological detection of leptogenesis-relevant CP violation — through precision measurements of ${\delta }_{CP}$ at DUNE/Hyper-K, combined with the requirement that the same physics also generates the cosmic baryon asymmetry. This connection is model-dependent but increasingly tight.

None of these provides a direct “particle ID” of the right-handed neutrino in the way that, say, the W/Z discovery in 1983 directly identified those bosons. The right-handed neutrino, if it exists at the seesaw scale, may never be directly observed. The case for its existence will be built indirectly, through accumulating evidence about neutrino mass, leptogenesis, and the cosmic baryon asymmetry.

A final reflection

The right-handed neutrino occupies an unusual position in particle physics. It is required by the simplest extensions of the Standard Model, expected at scales we can’t currently reach, and indirectly constrained by a variety of experiments — but never observed. The closest analogy is to the weakly interacting massive particle (WIMP) hypothesis for dark matter: a theoretically motivated particle that has so far evaded all direct searches.

The difference is that the right-handed neutrino is much harder to evade theoretically. Once you accept that neutrinos have mass, something has to play the role of the right-handed partner — even if you don’t call it that explicitly. The Weinberg operator from the previous section is what your low-energy theory looks like after integrating it out.

So the right-handed neutrino is, in a sense, theoretically observed — its existence is forced by the consistency of the rest of the framework. We just haven’t yet seen it directly. Whether we ever will depends on which version of the seesaw nature happens to have chosen.

The next two decades will provide the answer indirectly. By the 2040s, with precision $0\nu \beta \beta$, mass-ordering determination, CP-violation measurement, and improved cosmological bounds, the picture should be settled. The ghost in the Standard Model machine will, at last, leave a fingerprint we can read.