# Oscillatory Reflection Against Surfaces Explains Phenomena in Interfacial Water --- ## Abstract Molecular oscillations in a liquid reflect off any solid surface - the dielectric echo, formalized as the image dipole by Lennard-Jones (1928). Classical treatments assumed that each molecule couples individually to its own reflection but fluctuates incoherently with its neighbors. But a hydrophilic surface anchors an entire plane of molecules simultaneously. These molecules synchronize through the aether medium, creating a coherent oscillating sheet - a collective effect historically neglected. Unlike individual image-dipole interactions (1/h^3, nanometer range), a coherent plane has no geometric attenuation and can orient successive layers over micrometers. The echo biases existing oscillations toward the surface normal without adding energy, lowering the local standard chemical potential udeg. Wherever udeg < udeg_bulk, particles are excluded by osmotic pressure. This accounts for exclusion zones, their growth under infrared radiation, their shrinkage under heating, and the sign reversal of the interfacial potential with surface chemistry. The solvent chemical potential u_w = udeg + RT ln(gamma_w x_w) + VP also contains a composition term; conventional explanations operate through nablax_w but require different sources for each context and do not account for the infrared effect. The measured electrostatic field near hydrophilic surfaces (~140-200 mV) is the DC component of the oriented oscillations - rectified through anharmonicity from a much larger nonpolar oscillatory coupling. Water adsorbed at the surface constitutes a third layer between the solid and the exclusion zone, with ice-like permittivity. This three-layer geometry (solid | adsorbate | bulk) is under compression - the Lifshitz geometry, but with the coherent planar oscillation of the adsorbate taken into account, yielding forces orders of magnitude stronger than the incoherent Lifshitz estimate. This compression may drive oxidation of the adsorbate, producing van der Waals stacked (H3O2)_n monolayers that retain their radical character - corresponding to either one of Chen et al.'s computed H2O-OH configurations, or an oxidized version of Pollack's proposed fourth phase of water. This oxidation accounts for substrate degradation observed near hydrophilic surfaces where hydrolysis alone is insufficient. --- ## 1. Introduction: Interfacial water phenomena Near hydrophilic surfaces, colloidal particles are expelled from a region extending hundreds of micrometers (Zheng & Pollack, 2003; Chai & Pollack, 2010). This exclusion zone grows up to fourfold within 10 minutes under infrared radiation at the solvent's absorption band (Chai et al., 2009). Global heating of similar magnitude shrinks it. The measured electrostatic field reaches ~140-200 mV near the surface - attributed variously to deprotonation, substrate degradation, or ion exchange depending on conditions, with no single mechanism accounting for all observations. The solvent chemical potential u_w = udeg + RT ln(gamma_w x_w) + VP contains three spatial terms. Ion-exchange and diffusiophoretic mechanisms (Florea et al., 2014; Schurr, 2013; Esplandiu et al., 2020) operate through the composition term nablax_w. They account for charge-dependent particle motion and sqrtt growth kinetics but require different nablax_w sources for each context - ion exchange in salt solutions, substrate degradation in deionized water - and do not address infrared sensitivity or the opposite sign of thermal heating. This paper proposes that the image-dipole interaction of Lennard-Jones (Lennard-Jones & Dent, 1928), extended to account for the coherent planar oscillation that a surface imposes on its first layer, provides a unified account. The surface echoes the liquid's molecular oscillations. The first layer, anchored by hydrogen bonds, synchronizes in parallel through the aether medium, forming a coherent oscillating plane - a collective effect not treated in previous work. Unlike the individual image-dipole interaction (1/h^3, nanometer range), a coherent plane has no geometric attenuation and orients successive layers over micrometers through Bjerknes coupling. This lowers udeg and excludes particles by osmotic pressure. Hydrophilic surfaces carry an adsorbed water layer of ~1 nm (Stoneham & Tasker, 1985), sitting between solid and bulk in a three-layer Lifshitz geometry under coherent compression (S8). Substrate degradation is observed near such surfaces at rates exceeding hydrolysis (Bunkin et al., 2013), implying a reductive process at the solid. If the adsorbate provides the electrons, a plausible product is van der Waals stacked (H3O2)_n monolayers - corresponding to either one of Chen et al.'s computed H2O-OH configurations (Chen et al., 2025), or an oxidized version of Pollack's proposed fourth phase. Chen et al. showed that such monolayers are ferromagnetic semiconductors (0.79 uB); with residual OH^- among the radical sites, the phase may also be electrically conducting. These properties make it a candidate building block in biological interfacial processes. Covalent O-O stacking demonstrated by Chen et al. (Chen et al., 2025) is also possible, which pairs the electrons and eliminates the magnetism. Chen et al. demonstrated that external electric fields can control the transition between these stackings. --- ## 2. Selective cohesion There is a contribution to liquid cohesion that is already contained in the framework of quantum mechanics but is rarely described in physical terms. Within that framework, the force is typically handled as energy bookkeeping - electron volts, kilojoules per mole - without reference to the underlying physical mechanism: oscillatory coupling through the aether medium. It is a general property of any medium that oscillating bodies within it attract when in phase and repel when out of phase - a principle demonstrated by the Norwegian scientist Carl Anton Bjerknes in 1875 with macroscopic pulsating spheres, and formalized by his son Vilhelm Bjerknes - who later served as assistant to Heinrich Hertz - in *Fields of Force* (Bjerknes, 1906). London (London, 1930) showed in 1930 that the same mechanism operates between molecules - quantum-mechanical fluctuations make every molecule a pulsating oscillator, and the in-phase correlation between neighbors is the van der Waals attraction. This attraction is frequency-selective. Two molecules couple most strongly at frequencies where both respond - electronic (UV), vibrational (IR), librational (THz) - and less at frequencies where their responses differ. Lifshitz (Lifshitz, 1956) formalized this in 1956 as an integral of the coupling over all shared frequencies. Van der Waals attraction is, in this sense, selective cohesion: molecules attract most strongly those that oscillate at the same frequencies - which, in practice, means molecules of the same kind. --- ## 3. Oscillatory reflection at the surface When a molecule oscillates near a solid surface, the surface reflects that oscillation back - the dielectric echo. This is the image dipole, calculated by Lennard-Jones (Lennard-Jones & Dent, 1928) and generalized by Lifshitz (Lifshitz, 1956). The echo returns at the molecule's own frequency. It acts as a phantom neighbor behind the surface - one that oscillates at exactly the frequency that holds the liquid together. The surface selectively reinforces the liquid's own cohesive mode. The echo orients existing oscillations. It does not add energy. Each molecule already has kT per mode. The surface biases the direction. Traditional treatments (Lennard-Jones & Dent, 1928; Lifshitz, 1956) calculated the echo for each molecule independently - incoherent fluctuations, uncorrelated from neighbor to neighbor. In this picture, the echo is a short-range, single-molecule effect that decays as 1/h^3 and is negligible beyond a few nanometers. This is the standard textbook result. But at any surface, every molecule in the first layer is echoed by the same surface at the same time. They synchronize - not through direct contact, but through the aether medium - like metronomes on a shared shelf falling into phase. This collective effect always exists. It was simply not considered. Its consequences are qualitatively different from the single-molecule picture: a coherent plane has no geometric attenuation, and its field can orient successive layers far beyond the nanometer range. The echo's sign depends on dielectric contrast: when the liquid responds more strongly than the solid (epsilon_liquid > epsilon_solid), the echo is out of phase - repulsive. Water, with its exceptionally high permittivity, produces a repulsive echo against most solid surfaces. --- ## 4. Hydrophilic versus hydrophobic At hydrophobic surfaces, nothing holds the liquid against the repulsive echo. Water retreats, leaving a depletion gap of ~2 A. The gap weakens the echo coupling considerably. Weak echo, weak alignment, no propagation, no exclusion zone. At hydrophilic surfaces, hydrogen bonds (~20 kJ/mol) anchor the first layer despite the repulsive echo (~2-5 kJ/mol at contact). The anchored layer's oriented oscillations couple to the next layer through Bjerknes attraction - the same force between synchronized oscillators described in S2. Neighbor by neighbor, the orientation propagates outward, damped by thermal noise. Lennard-Jones and Lifshitz calculated the interaction of a single molecule against a surface - a point source, decaying as 1/h^3, limited to nanometer range. But the surface synchronizes the first layer in parallel. All molecules in layer 1 are anchored simultaneously, oscillating in phase - like metronomes on a shared shelf, synchronizing not through direct contact but through the medium they share. This makes layer 1 a coherent planar source. A coherent plane has no geometric attenuation - its field propagates without the 1/h^3 loss that limits individual molecular echoes. Layer 2, synchronized by layer 1, becomes a coherent plane itself - and orients layer 3 in turn. Each plane propagates the orientation to the next without geometric loss. This is why the orientation can reach micrometers, not nanometers. --- ## 5. The polarization field Most oscillations are nonpolar - electronic fluctuations carry no net dipole. A DC electrode cannot detect them. What it detects is the small polar fraction, rectified through anharmonicity: the time-averaged net polarization from polar modes - OH vibrations, librations, molecular bending - rectified by anharmonicity. This is the -140 mV measured by Chai and Pollack (Chai & Pollack, 2010). The DC shadow of a much larger oscillatory orientation. The sign confirms this: at anionic surfaces, water orients H inward, O outward - negative potential. At metallic surfaces, O inward, H outward - positive potential (Chai et al., 2012). The sign follows the surface chemistry, not the water. Expected for oriented oscillations anchored by surface groups. --- ## 6. Why IR grows the exclusion zone IR at the solvent's absorption band pumps the same molecular oscillations through which molecules attract via Bjerknes force and which coordinate, via the echo, into a coherent plane at the surface - where the interfacial layer behaves as a single rhythmic unit. Larger amplitude means stronger coupling, and the exclusion zone grows. The growth peak tracks the absorption maximum because that is where the most vibrational energy is deposited. The IR itself carries no directional preference - it pumps amplitude equally in all orientations. The surface provides the orientation. Thermal heating has the opposite effect. It increases amplitudes across all frequencies but randomizes phases. The echo's orientational bias is overwhelmed and the oriented region contracts. Narrowband excitation at specific frequencies adds amplitude the surface can orient; broadband heating adds noise it cannot. Chai et al. (Chai et al., 2009) observed exactly this: growth under IR, shrinkage under bath heating. --- ## 7. Particle exclusion Where udeg < udeg_bulk, water is more cohesive than bulk. The difference Deltaudeg creates osmotic pressure. Particles are pushed out. The exclusion zone boundary is where Deltaudeg vanishes. The force acts on volume, not charge. Chai and Pollack (Chai & Pollack, 2010) found that exclusion-zone size did not depend significantly on microsphere charge. The solvent chemical potential also contains a composition term, RT ln(gamma_w x_w), which can drive particle transport through solute gradients. This term is real but cannot account for the infrared growth or the opposite sign of heating, and requires context-specific explanations - ion exchange in salt solutions, substrate degradation in deionized water - for each experimental scenario. --- ## 8. The adsorbate Water adsorbed at the surface through hydrogen bonding forms an adsorbate - typically ~1 nm of structured water at hydrophilic surfaces - with restricted rotation and ice-like permittivity (epsilon ~ 3). This creates a three-layer system: Solid (epsilon ~ 15) | Adsorbate (epsilon ~ 3) | Bulk water (epsilon ~ 80) The Lifshitz-Hamaker constant: A ~ (epsilon1 - epsilon3)(epsilon2 - epsilon3) = (15 - 3)(80 - 3) > 0. The outer phases attract through the film. The adsorbate is compressed from both sides. Surface chemistry holds it. The adsorbate persists under continuous Lifshitz compression. Standard Lifshitz theory assumes incoherent fluctuations - each molecule fluctuates independently. The compression force scales as N per unit area. But the surface synchronizes the adsorbate layer coherently: all molecules oscillate in phase. For a coherent plane, the force scales as N^2. With ~10^1^4 molecules per cm^2, this is an enhancement of order 10^1^4 over the incoherent Lifshitz estimate - the energy source for the oxidation described in S9. The adsorbate is also a better mirror for bulk water than the solid itself: reflection coefficient 0.93 versus 0.68 for bare Nafion. --- ## 9. Deprotonation and oxidation of the adsorbate The ice-like adsorbate partially deprotonates at the surface - as ice surfaces do (Inagawa et al., 2019) - releasing H^+ and leaving OH^- sites in the network. The coherent Lifshitz compression (S8) can drive electron transfer from the deprotonated sites: OH^- -> OH* + e^- This is thermodynamically easier than oxidizing intact water (Edeg = +1.89 V vs +2.31 V). In the case of Nafion, the electron reduces pendant SO3^- to HSO3^- (Bunkin et al., 2013). The radical OH* remains in the adsorbate. The result is van der Waals stacked (H3O2)_n monolayers - a deprotonated, oxidized water network where approximately half the hydroxyl groups are OH* radicals. This corresponds to either one of Chen et al.'s computed H2O-OH configurations (Chen et al., 2025), or an oxidized version of Pollack's proposed fourth phase of water. In either stacking, the radical character is retained. Alternatively, covalent O-O stacking produces H2O2 between layers, pairing the electrons and eliminating the magnetism (Chen et al., 2025). Chen et al. showed that the monolayer is a ferromagnetic semiconductor (0.79 uB); with residual OH^- among the radical sites, the partially oxidized adsorbate may also be electrically conducting. Evidence for this process: Bunkin et al. (Bunkin et al., 2013) detected HSO3^- near Nafion at concentrations exceeding what hydrolytic C-S cleavage can account for at room temperature, consistent with reductive SO3^- -> HSO3^- conversion. The release oscillates with a period of ~500 s, consistent with a cycle of adsorbate buildup, compression-driven oxidation, disruption by degradation products, and reformation. For comparison, Song et al. (Song et al., 2024) observed the same reaction - OH^- -> OH* + e^- - at ice-water interfaces during freezing, in NaCl solutions, where the growing ice incorporates Na^+ but pushes the larger Cl^- ahead of the freezing front, producing a charge separation and an electric field at the boundary sufficient to oxidize hydroxide. Here, the driving force is the coherent Lifshitz compression. --- ## 10. Discussion Every step in this paper rests on established physics: Bjerknes coupling between oscillators (Bjerknes, 1906), Lennard-Jones image-dipole interaction at surfaces (Lennard-Jones & Dent, 1928), London dispersion (London, 1930), Derjaguin disjoining pressure (Derjaguin, 1936), and Lifshitz three-layer forces (Lifshitz, 1956). What has been historically neglected is the collective behavior. Each of the classical treatments - Lennard-Jones, London, Lifshitz - calculated the interaction of independently fluctuating molecules. The mathematics is correct for that case. But a hydrophilic surface does not interact with one molecule at a time. It anchors an entire plane, and those molecules synchronize through the aether medium into a coherent oscillating sheet. De Ninno (De Ninno, 2017) proposed that water near hydrophilic surfaces is oscillatorically coherent - "QED coherent domains" - and used Lifshitz compression between this coherent phase and the bulk as a reinforcing force. The oscillatory reflection in a plane described here provides a mechanistic explanation for why such domains arise: the surface synchronizes the first plane, which propagates outward through Bjerknes coupling. De Ninno's coherent domains are the natural consequence of this synchronization. Ordering in the exclusion zone is not excluded. McGeoch and McGeoch (McGeoch & McGeoch, 2008) demonstrated compressed hexagonal water (lattice spacing 3.73 A versus 3.90 A for ordinary ice) formed at room temperature under confinement. Whether any such ordering in the exclusion zone decays abruptly enough to constitute a sharp dielectric boundary - and thus contribute Lifshitz compression between bulk, exclusion zone, and adsorbate - is unclear, and is not the primary concern. The echo-driven Bjerknes coupling accounts for the exclusion zone without requiring it. The adsorbate acting as a reducing agent could explain substrate degradation observed near hydrophilic surfaces at rates exceeding hydrolysis (Bunkin et al., 2013). The coherent Bjerknes forces between solid and bulk acting on the adsorbate (S8) provide a possible driving force for this electron transfer. The oxidized product - (H3O2)_n with OH* radical sites (S9) - is a structure analogous to Pollack's proposed fourth phase, but oxidized. Chen et al. (Chen et al., 2025) predict that such a phase is ferromagnetic; with residual OH^- among the radical sites, it may also be electrically conducting. A water-derived film at biological interfaces that is both magnetic and conducting is an interesting possible building block for biology. --- ## References Bjerknes, V. (1906). *Fields of Force*. Columbia University Press. Bunkin, N. F., Kozlov, V. A., Ignatiev, P. S., Shkirin, A. V., & Kobelev, A. V. (2013). Study of the phase states of water close to Nafion interface. *WATER Journal*, *4*, 129-154. Chai, B., & Pollack, G. H. (2010). Solute-free interfacial zones in polar liquids. *Journal of Physical Chemistry B*, *114*, 5371-5375. Chai, B., Mahtani, A. G., & Pollack, G. H. (2012). Unexpected presence of solute-free zones at metal-water interfaces. *Contemporary Materials*, *3*, 1-12. Chai, B., Yoo, H., & Pollack, G. H. (2009). Effect of radiant energy on near-surface water. *Journal of Physical Chemistry B*, *113*, 13953-13958. Chen, J., Zhao, Q., Xu, M., et al. (2025). Control of magnetic transitions via interlayer engineering in ferroelectric H2O-OH systems. *Nature Communications*, *16*, 4809. De Ninno, A. (2017). Dynamics of formation of the Exclusion Zone near hydrophilic surfaces. *Chemical Physics Letters*, *667*, 322-326. Derjaguin, B. V. (1936). Theory of the interaction of particles in the presence of electric double layers. *Acta Physicochimica USSR*, *5*, 1-22. Esplandiu, M. J., Reguera, D., & Fraxedas, J. (2020). Electrophoretic origin of long-range repulsion of colloids near water/Nafion interfaces. *Soft Matter*, *16*, 3717-3726. Florea, D., Musa, S., Huber, J. M. R., et al. (2014). Long-range repulsion of colloids driven by ion exchange and diffusiophoresis. *Proceedings of the National Academy of Sciences*, *111*, 6554-6559. Inagawa, A., Harada, M., & Okada, T. (2019). Charging of the ice/solution interface by deprotonation of dangling bonds, ion adsorption, and ion uptake in an ice crystal as revealed by zeta potential determination. *Journal of Physical Chemistry C*, *123*, 6062-6069. Lennard-Jones, J. E., & Dent, B. M. (1928). The change in lattice spacing at a crystal boundary. *Proceedings of the Royal Society A*, *121*, 247-259. Lifshitz, E. M. (1956). The theory of molecular attractive forces between solids. *Soviet Physics JETP*, *2*, 73-83. London, F. (1930). Zur Theorie und Systematik der Molekularkrafte. *Zeitschrift fur Physik*, *63*, 245-279. McGeoch, J. E. M., & McGeoch, M. W. (2008). Entrapment of water by subunit c of ATP synthase. *Journal of the Royal Society Interface*, *5*, 311-318. Schurr, J. M. (2013). Phenomena associated with gel-water interfaces. *Journal of Physical Chemistry B*, *117*, 7653-7674. Song, J., Werner, L., Carreira Mendes Da Silva, Y., Theis, A., Donaldson, D. J., & George, C. (2024). Spontaneous production of H2O2 at the liquid-ice interface: a potential source of atmospheric oxidants. *Environmental Science & Technology*, *58*, 22691-22699. Stoneham, A. M., & Tasker, P. W. (1985). Metal-non-metal and other interfaces: the role of image interactions. *Journal of Physics C*, *18*, L543-L548. Zheng, J.-M., & Pollack, G. H. (2003). Long-range forces extending from polymer-gel surfaces. *Physical Review E*, *68*, 031408.