Direct three-body dynamics govern ion–atom recombination and barrierless termolecular reactions Rian Koots, Marjan Mirahmadi, Jesús Pérez-Ríos Abstract For over a century, termolecular (third-order) chemical reactions have been explained by the Lindemann-Hinshelwood mechanism, assuming sequential stabilization via bimolecular encounters. Here, we demonstrate that barrierless termolecular reactions are fundamentally governed by direct three-body dynamics. Using classical trajectory calculations in hyperspherical coordinates, we quantitatively reproduce ion–atom recombination kinetics across a wide temperature range without invoking intermediate complexes or steady-state assumptions. Our results not only resolve longstanding discrepancies between theory and experiment, but also establish a general mechanistic framework for barrierless termolecular reactions, with implications spanning atmospheric chemistry, plasma physics, and ultracold chemistry. I Introduction Termolecular or third-order chemical reactions A {}+{} B {}+{} C {}\mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits}{} Products are ubiquitous in nature 1 1 1 It is important to emphasize that in this work, we treat termolecular reactions as third-order reactions, independently of the reaction mechanism behind them. , from plasma physics to the fate of a single ion in an ultracold atomic bath, thus spanning over 10 decades of temperature and covering a multitude of essential systems in physics and chemistry. Termolecular reactions are often the only possible route to a product, such as for ozone formation in the stratosphere through O {}+{} \text{O}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} {}+{} M {}\mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits}{} \text{O}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{3}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{3}}\text{}} {}+{} M , where M is generally \text{N}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , or the formation of primordial stars where molecular hydrogen, required as a coolant, appears as a consequence of the reaction H {}+{} H {}+{} H {}\mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits}{} \text{H}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} {}+{} H . Trautz and Bodenstein pioneered the study of termolecular reactions 35 ; 33 ; 34 ; 6 , both of whom were the main figures in the development of modern chemical kinetics. As early as 1914, Max Trautz, after studying NO + NO + Cl \rightarrow NOCl + NOCl, and developing the first ideas on molecular collision theory, proposed that this third-order reaction occurred sequentially, as two concatenated bimolecular reactions 35 . The reasoning was that it is easier for two molecules to find each other than for three molecules to find each other simultaneously. In 1922, Bodenstein, studying a different third-order chemical reaction, proposed the same mechanism to explain its kinetics 6 . Bodenstein’s reasoning was very similar to that of Trautz. However, these mechanisms were rather speculative and lacking any theoretical foundation, except for Bodenstein, who proposed a model for the third-order reaction rate. The idea of a sequential mechanism was finally formalized by Lindemann and Hinshelwood 22 ; 15 , yielding the so-called Lindemann-Hinshelwood mechanism for termolecular and unimolecular reactions. The Lindemann-Hinshelwood mechanism assumes that a third-order reaction, such as occurssequentially . First , twoofthereactantsmeettoformacomplexas A + B \mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits^{\mkern 9.0muk_{2}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu\hphantom{k_{diss}}\mkern 9.0mu$}}}} \mathrel{\mathop{\makebox[0.0pt]{\leftarrow\displaystyle}}\limits^{\mkern 9.0mu\hphantom{k_{2}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0muk_{diss}\mkern 9.0mu$}}}} AB * , where k_2 denotestherateofformationofAB ^* complexesand k_ diss standsforitsdissociationrate.Inasecondstep,thethirdreactantstabilizesthecomplex\begin{aligned} {}\mathrm{AB}{\vphantom{\mathrm{X}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{}}\mathrm{\ast}}{}+{}{}\mathrm{M}&{}\mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits^{\mkern 5.0muk_{stab}\mkern 9.0mu}}{}{}\text{AB}{}+{}{}\text{M},\end{aligned}where k_ stab referstothestabilizationrateduetoacollisionwithathirdreactant.Assumingthatcomplexformationreachesasteadystate,thetermolecularreactionrateofthereaction\ref{r1}isgivenby\begin{equation}k_{3}=\frac{k_{2}k_{\text{stab}}}{k_{\text{diss}}+k_{\text{stab}}[\text{M}]},\end{equation}where[M]isthenumberdensityoftheparticleM.Therefore,thenatureoftheintermediatecomplexandtheconcentrationofthethirdreactantestablishtheoutcomeofthereaction.Thisargumentiscompellingtothepointthatthecommunityhasbeenusingthisandvariantstodealwithtermolecularreactionseversince~\cite[cite]{\@@bibref{Authors Phrase1YearPhrase2}{Xie_Poirier_Gellene_2003,Russell_1985,Russell_1986,Hauser2015,Troe2005,Herbst1980,Bates1979,Brahms2011,Bates1979bis,Luther2005,RBCmechanism}{\@@citephrase{(}}{\@@citephrase{)}}}.However,despiteitswidespreaduse,theLindemann\textendash Hinshelwoodmechanismfailstoquantitativelydescribemanybarrierlessrecombinationprocesses,particularlyatlowtemperatures~\cite[cite]{\@@bibref{Authors Phrase1YearPhrase2}{Perez-Rios2018,Koots2025,Sulfur}{\@@citephrase{(}}{\@@citephrase{)}}}.Thisraisesafundamentalquestion:arebarrierlesstermolecularreactionsinherentlysequential,oraretheygovernedbydirectmany-bodydynamics?\par Ontheotherhand,itispossiblethatatermolecularreactionindeedoccursinasinglestep,withthethreereactantsmeetingalmostsimultaneously(three-bodycollision),knownasthedirectmechanism.Inprinciple,inlightoftheargumentsofTrautz,Bodenstein,Lindemann,andHinshelwood,suchareactionmechanismshouldbediscarded,sinceathree-bodycollisionishighlyunlikely.Nevertheless,theargumentdoesnotaccountforthefactthatmosttermolecularreactionsarebarrierlessandoftenexhibitinverse-temperaturescalingcharacteristicoflong-rangecapturedynamics.Incontrast,mostbimolecularreactionsrequirethereactantstoovercomeanenergybarriertoformproducts.Thus,termolecularreactionsaremoreefficientatlowtemperatures,whereasbimolecularreactionsareathightemperatures.Asaconsequence,dependingonthereactionandthesystem^{\prime}sphysicalconditions,thedirectapproachisunavoidableforbarrierlesstermolecularreactions.Indeed,theLindemann-Hinshelwoodmechanismhasalreadybeenshowntofailtodescribehalogenandraregasatomrecombination~\cite[cite]{\@@bibref{Authors Phrase1YearPhrase2}{Koots2025}{\@@citephrase{(}}{\@@citephrase{)}}},orion-atomrecombinationreactionsbetweencoldionsandultracoldatoms~\cite[cite]{\@@bibref{Authors Phrase1YearPhrase2}{Perez-Rios2018,Mirahmadi_2022}{\@@citephrase{(}}{\@@citephrase{)}}}.\par Ion-atomrecombination{}\mathrm{A}{\vphantom{\mathrm{X}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{}}\mathrm{+}}{}+{}{}\mathrm{A}{}+{}{}{}\mathrm{A}\rightarrow\mathrm{Products}isoneofthemoststudiedtermolecularreactions,especially{}\mathrm{He}{\vphantom{\mathrm{X}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{}}\mathrm{+}}{}+{}{}\mathrm{He}{}+{}{}{}\mathrm{He}\rightarrow\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}{}+{}{}\mathrm{He}and{}\mathrm{Ar}{\vphantom{\mathrm{X}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{}}\mathrm{+}}{}+{}{}\mathrm{Ar}{}+{}{}{}\mathrm{Ar}\rightarrow\mathrm{Ar}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}{}+{}{}\mathrm{Ar},bothexperimentallyandtheoretically.Thisreactionisessentialincoldchemistrysinceitestablishesthesurvivalofasingleioninabathofultracoldatoms,orinexcimerlaserphysics~\cite[cite]{\@@bibref{Authors Phrase1YearPhrase2}{Excimer}{\@@citephrase{(}}{\@@citephrase{)}}}.Inplasmaphysics,ion-atomrecombinationcanaffecttheyieldofmolecularionsandhenceaffectthemomentumtransferoftheplasmatothewallsofthereactor.ThevastmajorityoftheoreticaleffortstounderstandthesereactionshavebeeninfluencedbythesequentialapproachunderlyingtheLindemann-Hinshelwoodmechanism.Thetreatments,evenatthefullquantumlevel,failtoexplaintheexperimentalfindings.Aswedemonstrateinthispaper,itisindeedthedirectmechanismthatcorrectlyexplainsion-atomrecombinationkinetics.Ourtreatmentusesclassicaltrajectorycalculationsinhypersphericalcoordinateswithoutinvokingsteady-stateassumptionsorintermediatecomplexes.Ourresultsquantitativelyexplaintheexperimentaldataforion-atomrecombinationreactionsoverawiderangeoftemperatures,describingthemintermsoftheunderlyingpotentialenergysurface.Moreover,ourresultssupportpreviousattemptstoextrapolateexperimentaldataintothecoldregime,whichareotherwisequestionablebasedontheoreticaltreatmentsthatemploytheLindemann-Hinshelwoodmechanism.Thesefindingsredefinethemechanisticdescriptionofbarrierlesschemicalprocessesindiverseenvironments,includingozoneformationintheatmosphere,sulfurrecombinationingeochemicalcycles,ionchemistryinplasmas,andmolecularionformationincoldenvironments.\par\par II Methodology Here, we briefly review the method used for obtaining the temperature-dependent ion-atom recombination rate coefficients, which has been thoroughly described in Refs. 31 ; 23 . We use a classical trajectory method that relies on a coordinate transformation from the Cartesian 3D space to a hyperspherical 6D (after removing the center of mass degrees of freedom) space in order to define the three-body recombination cross section as \sigma_{\text{rec}}(E_{c})=\frac{8\pi^{2}}{3}\int_{0}^{b_{\text{max}}(E_{c})}\mathcal{P}_{\text{rec}}(E_{c},b)b^{4}db, (2) where E_{c} represents the collision energy, \mathcal{P_{\text{rec}}} is the opacity function which defines the probability of recombination for the initial conditions (E_{c},b) , and b is the impact parameter which is now defined as a 5-dimensional vector embedded in the 6D space. The opacity function for several collision energies is plotted in Figure 1 , which visually demonstrates the relationship between the collision energy, the probability of reaction, and the impact parameter. As the collision energy decreases, the probability of the reaction increases, as expected for a barrierless reaction. More importantly, we observe that the maximum impact parameter follows ( b_{\mathrm{max}}(E_{c})\propto E_{c}^{-1/4} ) below E_{c}=10 K, accordingly to the classical threshold law 29 as a result of the preponderance of the charge-induced dipole interaction (ion-atom) over the induced dipole-induce dipole interaction (atom-atom). This integral is evaluated via the Monte Carlo technique, which involves random sampling of the trajectory space over an appropriate range of impact parameters up to the maximum impact parameter, b_{\text{max}} . FIG. 1: Opacity function of \text{He}{\vphantom{\text{X}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{}}\text{+}} {}+{} He {}+{} He {}\mathrel{\mathop{\makebox[0.0pt]{\rightarrow\displaystyle}}\limits}{} \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} {}+{} He for various collision energies. Lower collision energies result in larger values of b_{\mathrm{max}} . The cross section is evaluated via the integral of these curves, as shown in Eq. 2 . The thermally averaged recombination rate coefficient is easily obtained by integrating the energy-dependent cross section over the three-body Maxwell-Boltzmann energy distribution as k_{\text{rec}}(T)=\frac{1}{2(k_{b}T)^{3}}\int_{0}^{\infty}\sqrt{\frac{2E_{c}}{\mu}}\sigma_{\text{rec}}(E_{c})E_{c}^{2}e^{-E_{c}/(k_{b}T)}dE_{c} (3) where k_{b} is the Boltzmann constant, T is the temperature of the system, and \mu=\sqrt{m_{1}m_{2}m_{3}/(m_{1}+m_{2}+m_{3})} is the three-body reduced mass. For \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination, we include a factor of 1/2 due to the symmetry-induced isotope effect arising from collisions of the same isotope 10 . For \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination, we use a factor of 2/3 to account for the splitting of \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} electronic states. Potential energy surfaces calculated with spin-orbit coupling, where four out of the six states are attractive 9 . All the calculations have been performed using Py3BR software 20 running batches of 50,000 trajectories per impact parameter and including up to 60 impact parameters per collision energy, yielding about 10^{6} trajectories per collision energy. The initial hyperradius R for \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination was chosen as 500\pm 50 a 0 , and 5,000\pm 500 a 0 for \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination. II.1 Potential energy surfaces It has been shown that for atom and ion-atom recombination reactions the role of the three-body interaction term is negligible 36 . Thus, we use a pairwise additive potential to describe the interaction among the three colliding atoms, such that V(r_{1},r_{2},r_{3})=V(r_{12},r_{23},r_{31}) . The \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} interaction is described by the following potential of Ref. 8 , whereas \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} is described by the analytic Hartree-Fock-dispersion potential HFD-B3-FCI1 4 . In addition we use Lennard-Jones potentials ( LJ(r)=C_{m}/r^{m}-C_{n}/r^{n} ) to explore the role of the short-range interactions. For the ion-induced dipole interaction LJ_{He_{2}^{+}}(r) , we use C_{4}=0.69 a.u., C_{8}=1.32 a.u., and for the van-der Waals interaction LJ_{{He_{2}}}(r) , we use C_{6}=1.35 a.u., C_{12}=1.38\times 10^{4} a.u.. For ion-atom recombination in argon, we employed Lennard-Jones potentials for the atom-atom interaction ( C_{6}=63.5 a.u., C_{12}=2.07\times 10^{6} a.u.) and the ion-atom interaction ( C_{4}=43.87 a.u., C_{8}=9434 a.u.), where the short-range interaction follows r^{-8} . More details about the six model potentials are given in Table 1 . TABLE 1: Potential parameters used for ion-atom recombination in helium and argon, with the dissociation energy ( D_{e} ) and equilibrium distance ( r_{e} ) across models. All parameters are in atomic units. Type Model D_{e} r_{e} Ion - Neutral V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}} 0.09027 2.0473 LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}} 0.09017 1.3993 LJ_{{}\mathrm{Ar}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}} 0.05099 4.5532 Neutral - Neutral V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} 3.4695\times 10^{-5} 5.6094 LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} 3.3016\times 10^{-5} 5.2287 LJ_{{}\mathrm{Ar}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} 4.8722\times 10^{-4} 6.3435 III Results and Discussion III.1 Helium ion recombination FIG. 2: Temperature-dependent formation rate of \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} via three-body recombination. The results of the present work, based on the direct mechanism, are shown in blue, with the uncertainty indicated by the shaded region. The blue dashed line represents the classical power-law behavior, \propto\mathrm{T}^{-3/4} . The experimentally determined rate coefficients are taken from Ref. 7 (circles), Ref. 16 (squares), Ref. 26 (x), Ref. 32 (triangles), Ref. 17 (inverted triangles), Ref. 5 (stars), and Ref. 11 (diamonds). The black dashed line represents the current standard fit to the experimental data from Refs. 7 ; 32 ; 11 , given by 1.4\times 10^{-31}(300/\mathrm{T})^{0.6} . The purple curve shows the theoretically predicted rate reported in Ref. 37 Ion-atom recombination is a barrierless exothermic process, suggesting an increasing trend of the reaction rate coefficient with decreasing temperatures. In Figure 2 , we see that this trend is present in the experimentally measured termolecular formation rate of \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} between 10 K and 500 K. The temperature-dependent direct recombination rate coefficient is also presented in this temperature range, which accurately captures the experimentally measured data in both temperature trend and magnitude. In addition, we plot the most recent theoretical results from a quantum dynamical study under the Lindemann-Hinshelwood framework 37 . Here, we see a comparison between the termolecular one-step, direct recombination mechanism and the two-step bimolecular Lindemann mechanism. Although there is general agreement between the two theories above \sim 30 K, they demonstrate a different behavior at lower temperatures. In particular, the two-step approach presents a maximum near 30 K before rapidly decreasing, which is indicative of an activation barrier in the reaction since still this temperature is far beyond the Wigner threshold limit (constant reaction rate at very low collision energies). On the other hand, the direct mechanism yields a power law behavior where ion atom recombination dominates at low temperatures, as described by the classical threshold law 29 and expected for a barrierless reaction. The Lindemann-Hinshelwood mechanism fails because not all intermediate states can be considered, due to the complexity of the energy landscape. In practice, the number of complexes and the nature of those considered make a difference. The best example is the results of Smirnov and Dickenson et al., both using the Lindemann-Hinshelwood mechanism but yielding different results for the same ion-atom recombination reaction 30 ; 28 . On the contrary, the direct mechanism naturally includes all the most relevant states, thereby improving its description of the chemical kinetics of barrierless termolecular reactions. In addition to the quantal results, the current standard for the rate coefficient of \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination is based on the best fit to the experimental data of Refs. 7 ; 32 ; 11 : k_{\mathrm{fit}}(T)=1.4\times 10^{-31}(300/T)^{0.6}, (11) as shown in Figure 2 . The obvious disagreement has raised questions about whether the experimental fitting can be extrapolated to lower temperatures, or if the rate rapidly decreases as the two-step approach suggests 12 . Now, we have a third look at this reaction, which agrees with the low-temperature behavior of the experimental data while showing excellent agreement throughout the experimental temperature range. Therefore, the direct mechanism yields reaction kinetics compatible with experimental findings. FIG. 3: Thermal rate coefficient of \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination within the direct approach, calculated with varying potential functions to describe the \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} and \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} interaction. The blue curve is the same as presented in Figure 2 , such that the ( \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} , \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} ) potentials are (V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} , V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}}) . The orange (LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} , LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}}) , green (V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} , LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}}) , and red (LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} , V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}}) curves are the results of different combinations of potentials. The blue dashed line corresponds to the T^{-3/4} classical threshold law shifted to match the blue curve at low temperature. The inset shows the log-scaled absolute value of the four potentials, V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} (solid brown), V_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}} (solid black), LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}} (dotted brown), LJ_{{}\mathrm{He}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{+}}} (dotted black). The dissociation energies of each molecule are shown as horizontal dashed lines in their respective colors, and the repulsive parts of the potentials are 50% opaque. At low temperatures, the direct three-body recombination rate is controlled by the long-range interaction for which the classical threshold law behavior of ion-atom recombination has been derived within a capture model 29 ; 25 as k_{3}(T)\propto T^{-3/4} . This behavior occurs in the low-energy regime, or energies below the smallest dissociation energy of the system (in this case \sim 10 K). Above this threshold, short-range interactions begin to play a role in the reaction dynamics, leading to deviations from this power-law behavior. This phenomenon explains the discrepancy between the experimentally derived power law [see Eq. ( 11 )] and the classical threshold law for \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} formation. In Figure 3 , we investigate the role of the interaction potential on the ion-atom recombination rate coefficient between 1 K and 800 K. The calculations were replicated with varying sets of pairwise potentials, where either one or both of the pairwise interactions are described by the generalized Lennard-Jones potentials described in Section II.1 . In the inset, the log-scaled absolute value of the four potentials are plotted, which allows for a direct comparison. Below T=10 K, the whole set of potential pairs shows converged rates that follow the classical threshold law behavior. This is expected, since the dynamics are dominated by the \propto r^{-4} charge-induced dipole long-range interaction below collision energies of 10 K, which stays the same independently of the kind of the pairwise potential employed. Therefore, the temperature dependence may be extrapolated to lower temperatures following the power law behavior k_{3}(T)\propto T^{-3/4} . As the collision energy increases, the repulsive short-range part of the \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} interaction comes into play, and the thermal rate deviates from the threshold law. At higher temperatures, slight variations in the pairwise interaction potential (see inset Fig. 3 ) translate into a different ion-atom recombination rate coefficients. Therefore, the accuracy of our method is such that it is possible to find the best potential that accurately describe the experimental recombination rate. Thus, it is possible to invert the scattering information and transform into an effective potential similar to what it has been done for bimolecular processes 3 ; 2 ; 1 . III.2 Argon ion recombination As the lightest rare gas, the formation of \text{He}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} serves as a benchmark for the study of rare gas ion atom recombination in their parent gas. Here, we extend our calculations to the heavier \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} recombination reaction to demonstrate that the one-step direct mechanism is responsible for all ion-atom recombination reactions. In Figure 4 , the ion-atom recombination rate coefficient is presented from T = 0.01 K to 2000 K, along with experimentally determined values. The predicted rates via the direct mechanism show excellent agreement with both the low-temperature and high-temperature experimental results. In addition, we see that at temperatures below 0.1 K the rate coefficient follows the T^{-3/4} classical threshold law, which is dictated by the long-range charge-induced dipole interaction of \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} . The pairwise interactions for \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} and \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} dictate the temperature dependence of the reaction, and are shown in the inset. FIG. 4: Temperature-dependent formation rate coefficient of \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} via three-body recombination. The ion-atom recombination rate coefficient using a direct approach are presented in blue, with the shaded region representing the statistical error. The blue dashed line corresponds to the T^{-3/4} classical threshold law shifted to match the prediction at low temperature. The experimentally determined rate coefficients are of Ref. 14 (circles), Ref. 16 (squares), Ref. 17 (triangles), Ref. 13 (upside-down triangles) , Ref. 27 (diamonds). The inset shows the Lennard-Jones potentials used to describe the \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{+}} (black axis) and \text{Ar}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} (dark red axis) interactions. IV Conclusion We have shown that ion–atom recombination is governed by a direct three-body mechanism, without the need for intermediate complexes or steady-state assumptions. This result resolves longstanding discrepancies between theory and experiment and challenges the traditional Lindemann–Hinshelwood framework. More broadly, it establishes a unified description of barrierless termolecular reactions, opening new directions for understanding chemical kinetics across a wide range of environments. Previous results on sulfur recombination 18 , halogen atom recombination 19 , ion-atom recombination in the cold regime 21 and ozone recombination 24 , support the same conclusion: barrierless termolecular reactions are mostly the result of the direct mechanism. Therefore, the direct mechanism should be incorporated as a main contributor in the study of the kinetics of termolecular reactions.