---
title: 'Leak-Out Spectroscopy: Cold-Ion Action Method'
url: https://www.emergentmind.com/topics/leak-out-spectroscopy-los
type: topic
---

# Leak-Out Spectroscopy: Cold-Ion Action Method

Leak-Out Spectroscopy (LOS) is an action spectroscopy method in which photon absorption by trapped, mass-selected ions is inferred from a state-dependent increase in trap loss rather than from direct absorption, fluorescence, fragmentation, or chemical product formation. In its formalized usage, LOS is implemented in cryogenic RF multipole ion traps: ions are laser-excited, collisions with a neutral buffer gas convert part of the internal excitation into translational kinetic energy, and ions that overcome a tuned axial potential barrier “leak out” and are counted as a function of laser frequency. This scheme was explicitly introduced as “a universal method of action spectroscopy in cold ion traps” and has since been applied to linear cations, open-shell ions, and hydronium, yielding rotationally resolved rovibrational spectra, band origins, and effective spectroscopic constants under cryogenic conditions [2209.13660].

## 1. Definition, scope, and terminological boundaries

In the cold-ion literature, LOS denotes an action spectroscopy scheme in which the observable is the number of parent ions that escape from a cryogenic RF trap after resonant excitation and subsequent collision-induced vibration-to-translation transfer. The defining feature is that the parent ion mass is retained: unlike IRMPD, photodissociation, messenger-tagging, or laser-induced reactions (LIR), LOS does not require fragmentation, complex formation, or a specific reactive channel. The action observable is trap loss of the resonantly excited species itself [2209.13660].

This usage was established experimentally for linear C\(_3\)H\(^+\), where Schmid and co-workers demonstrated rotationally resolved spectroscopy of the C–H stretching vibration \(\nu_1\), and argued that LOS is “universal” in the sense that any internal excitation capable of generating sufficient kinetic energy after collisions can, in principle, be converted into a detectable leak-out signal [2209.13660]. Later work extended LOS to the antisymmetric C–H stretch of C\(_2\)H\(_2^+\), to near-infrared overtone spectroscopy of H\(_3\)O\(^+\), and to vibronic and combination bands of HCN\(^+\), with direct comparisons against LIR in several cases [2310.10186], [2509.16009], [2512.13476].

The term is not uniform across all branches of physics. In plasmonics, a leak-out or leakage-based measurement can refer to leakage radiation from surface plasmon polaritons (SPPs) on thin metal films, collected and interferometrically analyzed in the far field; the ultrafast leakage-imaging framework of [1510.04810] is leak-out based but belongs to near-field optics rather than cold-ion action spectroscopy. In the LiquidO context, the exact phrase “Leak-Out Spectroscopy” does not appear in the cited work; the provided interpretation treats LOS only as an implicit extension of stochastic light-confinement measurements and leak-out tails in an opaque scintillator [2503.02541]. Accordingly, the formal encyclopedia meaning of LOS is the cryogenic ion-trap technique, with other usages best regarded as related leakage-based measurement concepts.

## 2. Physical mechanism of trap-loss generation

LOS begins from a thermal ensemble of trapped ions with kinetic temperature \(T_\text{kin}\) and an axial barrier \(E_A\) chosen so that unexcited ions remain trapped for long times. For safe long-term trapping, the barrier is set such that
\[
E_A \gtrsim 10\,kT.
\]
In this regime the off-resonance leak-out rate is governed by the Maxwell–Boltzmann tail and may be written approximately as
\[
R_\text{off} \approx \frac{v}{d}\,P(E_\text{kin} > E_A),
\]
where \(v\) is a characteristic thermal velocity and \(d\) is a characteristic linear trap dimension [2209.13660].

On resonance, an ion absorbs a photon and is promoted to an internally excited rovibrational state. Subsequent collisions with a neutral buffer gas convert part of that internal energy into translational kinetic energy. This vibration-to-translation transfer is the essential microscopic step: once the ion’s kinetic energy exceeds the barrier, or its trajectory is sufficiently altered, it escapes the trap and is detected downstream. In the general hydronium discussion, the LOS signal is described conceptually by
\[
R(\nu) \propto N_\text{ion} \times P_\text{exc}(\nu) \times P_\text{coll} \times P_\text{escape},
\]
with \(N_\text{ion}\) the number of trapped ions, \(P_\text{exc}(\nu)\) the excitation probability, \(P_\text{coll}\) the probability of effective collisional energy conversion, and \(P_\text{escape}\) the probability of surmounting the exit barrier [2509.16009].

The balance between excitation, collisional quenching, thermalization, and escape is central. In the C\(_3\)H\(^+\) demonstration, N\(_2\) at \(n \approx 10^{12}\,\text{cm}^{-3}\) produced collision rates of order \(10^3\,\text{s}^{-1}\), with a typical time between collisions of about \(1\,\text{ms}\); the buffer-gas density therefore had to be high enough to make rare V–T events occur within the trapping time, but not so high that fast ions were immediately rethermalized before escape [2209.13660]. This establishes the experimentally important optimum-density condition.

The role of the neutral collision partner is not reducible to simple mass matching. In the H\(_3\)O\(^+\) overtone experiment, LOS functioned with both N\(_2\) and He, and the paper explicitly notes that He is only about \(10\%\) of the mass of H\(_3\)O\(^+\) yet still works well. The authors interpret this as evidence that vibrational energy transfer is more important than simple momentum exchange based on mass alone [2509.16009].

## 3. Experimental implementations in cryogenic multipole traps

LOS has been implemented primarily in cryogenic 22-pole RF ion traps integrated into tandem mass-spectrometric beamlines. The generic sequence comprises ion production, upstream mass selection, injection into the trap, collisional cooling, irradiation with a narrowband laser, and downstream detection of the leaked parent ions after a second mass-selection stage [2209.13660], [2512.13476].

In the original LIRtrap realization for l-C\(_3\)H\(^+\), ions were generated in a storage ion source by electron bombardment of allene, selected at \(m/z=37\), and trapped at about \(40\,\text{K}\). A He pulse was used for stopping and cooling, N\(_2\) was continuously present as buffer gas, and a Toptica Topo OPO probed the \(3160\)–\(3175\,\text{cm}^{-1}\) region with step sizes of about \(0.0005\,\text{cm}^{-1}\). During a \(200\,\text{ms}\) irradiation window, the exit barrier was tuned so that only fast ions escaped, and the leaked parent ions were counted downstream [2209.13660].

The COLTRAP implementation for C\(_2\)H\(_2^+\) operated nominally at \(4\,\text{K}\). Several tens of thousands of ions were injected per cycle, continuous He provided cryogenic cooling, and a pulsed Ne/He mixture was added at the beginning of each cycle to furnish a heavier collision partner effective for V–T transfer while avoiding extensive freeze-out on cold surfaces. A cw OPO near \(3\,\mu\text{m}\), with power around \(200\,\text{mW}\), irradiated the ions for \(100\,\text{ms}\) per \(1\,\text{Hz}\) cycle, and the leak-out signal was recorded as counts of C\(_2\)H\(_2^+\) ions escaping during irradiation [2310.10186].

The CCIT cryogenic 22-pole trap was used for both H\(_3\)O\(^+\) and HCN\(^+\)/HNC\(^+\). For H\(_3\)O\(^+\), a temperature-variable trap operated at \(125\,\text{K}\) and \(20\,\text{K}\), with helium cooling pulses and continuous N\(_2\) inflow at neutral number density \(\sim 10^{11}\,\text{cm}^{-3}\); below the freezing point of N\(_2\), He served as the LOS collision partner. An Agilent 8164B light system, calibrated with a Bristol 671A wavemeter, covered \(1370\)–\(1495\,\text{nm}\) with linewidth below \(100\,\text{kHz}\), and the beam was passed three times through the trap cell to increase interaction length [2509.16009]. For HCN\(^+\) and HNC\(^+\), the same CCIT platform operated at \(90\) or \(125\,\text{K}\), employed He for cooling and N\(_2\) for LOS collisional transfer, and detected leaked ions with a Daly detector after a second quadrupole mass filter [2512.13476].

## 4. Spectroscopic observables and representative measurements

LOS spectra are acquired by scanning laser frequency and recording leaked-ion counts at each point. The resulting line profiles are then fitted, typically with Doppler or Gaussian functions, to extract accurate transition wavenumbers. The downstream Hamiltonian analysis depends on molecular symmetry and electronic structure: linear-molecule effective Hamiltonians were used for l-C\(_3\)H\(^+\) and C\(_2\)H\(_2^+\), an oblate symmetric top Hamiltonian with \(l\)-doubling and inversion interactions was used for H\(_3\)O\(^+\), and PGOPHER-based \(^{2}\Pi\) and \(^{2}\Sigma^+\) Hamiltonians were used for HCN\(^+\) [2209.13660], [2310.10186], [2509.16009], [2512.13476].

| System | Measured band(s) | Representative outcome |
|---|---|---|
| l-C\(_3\)H\(^+\) | \(\nu_1\), \(\nu_1+\nu_5 \leftarrow \nu_5\) | \(\tilde{\nu}_1 = 3169.97769(16)\,\text{cm}^{-1}\) |
| C\(_2\)H\(_2^+\) | \(\nu_3\) (\(^2\Pi \leftarrow {}^2\Pi\)) | \(v_3 = 3135.97376(48)\,\text{cm}^{-1}\) |
| H\(_3\)O\(^+\) | \(2\nu_3^{2+}\), \(2\nu_3^{2-}\) | \(6845.610(14)\) and \(6878.393(13)\,\text{cm}^{-1}\) |
| HCN\(^+\) | X~\(^2\Pi~(000)^1-(210)^0\mu\), X~\(^2\Pi~(000)^1-\mathrm{A}~^2\Sigma^+~(10^00)\) | \(6284.9187(19)\) and \(6330.5924(36)\,\text{cm}^{-1}\) |

For l-C\(_3\)H\(^+\), LOS yielded a rotationally resolved spectrum of the \(\nu_1\) C–H stretch with clear P- and R-branches, line spacings characteristic of a linear rotor, and a hot band at \(\tilde{\nu}_{\nu_1+\nu_5\leftarrow \nu_5} = 3168.76964(12)\,\text{cm}^{-1}\). The \(\ell\)-type doubling constant for \(v_5=1\) was reported as \(q_5 = 68.348(96)\,\text{MHz}\), and Gaussian fits gave \(T_\text{kin}=62(8)\,\text{K}\), while rotational intensities implied \(T_\text{rot}=50\,\text{K}\) [2209.13660].

For C\(_2\)H\(_2^+\), LOS revisited the \(\nu_3\) antisymmetric C–H stretch of an open-shell \(^2\Pi\) ion under mass-selective cryogenic conditions and resolved \(\Lambda\)-doubling for higher \(J\). The fitted band origin was \(3135.97376(48)\,\text{cm}^{-1}\), with refined constants including \(B_0 = 1.104636(16)\,\text{cm}^{-1}\), \(B_3 = 1.099521(23)\,\text{cm}^{-1}\), \(A_0 = -30.871(16)\,\text{cm}^{-1}\), and \(A_3 = -30.42744(79)\,\text{cm}^{-1}\) [2310.10186].

For H\(_3\)O\(^+\), LOS enabled rotationally resolved overtone spectroscopy in the \(6750\)–\(6950\,\text{cm}^{-1}\) region. The two strongest first overtone bands of the asymmetric stretching mode \(\nu_3\), namely \(2\nu_3^{2+}\) and \(2\nu_3^{2-}\), were fit with a standard oblate symmetric top Hamiltonian and additional off-diagonal matrix elements accounting for \(l\)-doubling, giving band origins \(6845.610(14)\,\text{cm}^{-1}\) and \(6878.393(13)\,\text{cm}^{-1}\), respectively. The analysis used 39 transitions for \(2\nu_3^{2+}\) and 29 for \(2\nu_3^{2-}\), and the lower-temperature \(20\,\text{K}\) spectra simplified assignment by depopulating the \(0^{-}\) ground state and suppressing transitions to \(2\nu_3^{2-}\) [2509.16009].

For HCN\(^+\), LOS was the main route to high-resolution spectra of the X~\(^2\Pi~(000)^1-(210)^0\mu\) combination band and the X~\(^2\Pi~(000)^1-\mathrm{A}~^2\Sigma^+~(10^00)\) vibronic band in the \(6200\)–\(6800\,\text{cm}^{-1}\) region. Effective fits yielded \(T_{(210)^0\mu}=6284.9187(19)\,\text{cm}^{-1}\), \(B_{(210)^0\mu}=1.346376(71)\,\text{cm}^{-1}\), \(p=0.03685(51)\,\text{cm}^{-1}\), \(q=0.002387(27)\,\text{cm}^{-1}\), and for the A-state band \(T_{A(10^00)}=6330.5924(36)\,\text{cm}^{-1}\), \(B_{A(10^00)}=1.38624(13)\,\text{cm}^{-1}\), and \(\gamma_{A(10^00)}=0.01305(40)\,\text{cm}^{-1}\) [2512.13476].

## 5. Relation to other action methods, strengths, and limitations

LOS occupies a distinct position among ion action spectroscopies. Relative to direct absorption, it is effective at ion numbers of only \(10^4\)–\(10^5\) per trap load, a regime explicitly described as far below what is needed for classical absorption spectroscopy [2509.16009]. Relative to LIR, LOS does not require a chemically suitable endothermic reaction channel. Relative to messenger spectroscopy or LIICG, it does not require complex formation or tagging. Relative to IRMPD, it does not require multiphoton absorption or fragmentation. In the original formulation, these features motivate the description of LOS as a universal, highly sensitive, nearly background-free action method for cold ions [2209.13660].

The method’s strengths are most apparent under cryogenic conditions. Low trap temperatures suppress thermal leak-out, enhance contrast between on- and off-resonance conditions, and narrow rotational populations. In C\(_3\)H\(^+\), off-resonance loss rates of \(2.3\times10^{-2}\,\text{s}^{-1}\) increased to \(6.6\times10^{-1}\,\text{s}^{-1}\) on resonance for the R(5) line, with off/on enhancement of about 30 and strong lines giving more than \(2000\) counts per cycle [2209.13660]. In C\(_2\)H\(_2^+\), mass selection and cryogenic operation produced “uncontaminated spectra” with mostly resolved \(\Lambda\)-doublets, and LOS gave higher ion counts and better signal-to-noise ratio than LIR for that system [2310.10186].

At the same time, LOS is intrinsically dependent on collisional quenching efficiency. The HNC\(^+\)/HCN\(^+\) comparison makes this explicit: for an HNC\(^+\) overtone transition, LIR and LOS signals differed by a factor of 50 under otherwise identical conditions, indicating that collisional quenching of the X~\(^2\Sigma^+~(20^00)\) state of HNC\(^+\) by N\(_2\) is substantially less efficient than the corresponding Langevin reaction used for LIR. For HCN\(^+\), by contrast, LOS intensities were only about three times lower than LIR on the same transitions, and LOS was operationally preferable because LIR with Kr suffered from secondary chemistry involving impurities [2512.13476].

Other limitations recur across applications. LOS intensities depend on collisional energy-transfer probability, trap-potential settings, and ion number per trap load; they therefore cannot be interpreted solely in terms of thermal population and transition dipole strength [2509.16009]. Weak high-\(J,K\) lines may remain inaccessible, limiting higher-order distortion analysis, and nearby combination bands can produce strong Coriolis or Fermi interactions that perturb line positions and complicate effective Hamiltonian fits [2509.16009], [2512.13476]. The method also requires careful selection of buffer gases and temperatures to avoid freeze-out while maintaining efficient V–T transfer [2209.13660].

A distinctive application of LOS is state-selective sample preparation. Because the ions that absorb at a chosen frequency are the ions that leave the trap, LOS can be used to selectively expel one structural isomer, one isobaric species, or one nuclear-spin isomer, thereby preparing cleaner residual ensembles for subsequent spectroscopy or reaction studies [2209.13660]. This is not merely a by-product of the method; it is one of the explicit broader uses claimed for LOS.

## 6. Extensions and related leak-out concepts outside cold-ion spectroscopy

A separate lineage of leak-out-based measurements exists in ultrafast plasmonics. In “Tracking surface plasmon pulses using ultrafast leakage imaging,” leakage radiation microscopy is implemented in the time domain to measure both group and phase velocities of SPP wave packets propagating on thin Au films. The leakage angle satisfies
\[
\sin\alpha = \frac{\mathrm{Re}[k_{\text{SPP}}]}{n_g k_0},
\]
and the interferogram
\[
\mathcal{I} (x,\tau) \propto e^{-\frac{\Gamma}{2} |x|} \, e^{-\sigma_0^2\left(\tau - \tau_g\right)^2} \, \cos \left[\kappa (\omega_0)\cdot |x| + \omega_0\tau\right]
\]
encodes damping, group delay, and phase information in the far field. The provided synthesis describes this as “essentially ultrafast LOS,” because the leaked field acts as a spectroscopic probe of the guided mode, but the underlying publication introduces ultrafast leakage imaging rather than the ion-trap action method [1510.04810].

In LiquidO, the supplied interpretation uses “LOS” even more loosely. The cited work demonstrates stochastic light confinement in an opaque scintillator and measures a non-zero tail of photons detected outside the main light ball, with \(50\%\), \(80\%\), and \(90\%\) of the detected light confined within approximately \(2.0\), \(4.0\), and \(5.0\,\text{cm}\), respectively. The diffusion–absorption model gives a confinement width
\[
\sigma = \sqrt{\frac{2a}{b}} = \sqrt{\frac{2}{3}}\left(\lambda_{\text{scatter}}\lambda_{\text{absorption}}\right)^{1/4},
\]
and the provided discussion proposes that systematic use of the leak-out tail in radius, wavelength, and time could amount to a form of leak-out spectroscopy. However, the paper itself does not use the phrase “Leak-Out Spectroscopy,” so this should be treated as an interpretive extension rather than a settled technical term [2503.02541].

These broader usages clarify a recurrent conceptual pattern: “leak-out” denotes conversion of otherwise localized or trapped excitation into a measurable escape channel. In the formal cold-ion sense, the escaping entity is the parent ion. In plasmonics it is leakage radiation from a guided mode. In LiquidO it is the non-confined tail of diffusively transported photons. The commonality is methodological rather than disciplinary, and the established meaning of LOS remains the action spectroscopy technique of cryogenic ion traps [2209.13660].

Source: https://www.emergentmind.com/topics/leak-out-spectroscopy-los