Rapid Free Electron Reduction in Plasma Irradiated Microscale Water Droplets

Abstract

Many reaction studies show that microdroplets can provide a new avenue for green chemistry by enabling the electrochemical activity of water molecules. The observation of enhanced chemical reaction rates in gas-phase microdroplets compared to bulk liquids, often by orders of magnitude, has sparked considerable research interest. A number of factors may be involved, including evaporation enhanced reactant concentration, partial solvation at the surface, high surface to bulk number ratio, enhanced surface rate constants, pH gradients, gas-phase reactions and mass transfer, electric field enhancement and surface charging. Plasma interactions with liquids involve the mass transfer and accommodation of reactive radicals and other plasma chemical species, often occurs in the presence of high electric field and temperature gradients, UV flux and electric currents.We have investigated the use of low temperature atmospheric pressure plasma irradiated liquid microdroplets. [1] Picolitre microdroplets are totally surrounded by plasma and their high surface area relative to volume receives chemical, photon and charge flux during flight leading to chemical reactions in the liquid at low temperature. With the inclusion of precursors, the small volume provides an excellent basis for gas phase chemical microreactors that can deliver products continuously and almost instantaneously downstream for applications such as plasma medicine and specialist chemical or nanomaterial printing.In particular, low temperature plasmas are a copious source of free electrons, up to 106 times greater than corona discharges, which can interact with the liquid surface to promote rapid reduction reactions and possibly on-water catalysis. We have observed the reduction of metal salts in flight to produce nanoparticles at rates many orders of magnitude greater than standard solution chemistry or via radiolysis. [2] We have measured the plasma gas temperature in the presence of microdroplet streams with droplet rates up to 5 x 104 s-1 and observed no significant increase in gas temperature, which is typically ~300 K, thus keeping the evaporation limited. [3] Recently, we carried out plasma simulations, coupled with downstream chemical flux measurements, to determine the evolution of gas-phase chemistry in the plasma and effluent. [4] We have also performed the first measurements of plasma charging of particles, at atmospheric pressure for diameters > 1 um. The determined average droplet charge per droplet was 2.5 x 106 electrons (400 fC). Using a number of plasma particle charging models we estimate the electron flux ranged from 5 x 1022 m-2 s-1 to 4 x 1025 m-2 s-1.In [2], at least 50% Au3+ to Au0 reduction of the droplet precursor (HAuCl4) was observed over a ~120 µs plasma flight time and the equivalent Au0 generation rate is ~1013 atoms s-1. The ratio of electron flux per droplet to metal generation rate provides a dimensionless figure of merit, e-/Au0, of ~100 - 1000. Using plasma chemistry simulations along with estimates of electron flux to the droplet surface, we then simulated the subsequent water reactions using a 1-D radial reaction – diffusion scheme. For 15 µm diameter droplets without precursor, surface H2O2 concentrations ranged from 0.1 M to 100 µM whereas surface OH· concentrations were almost constant (~1 mM), as electron flux density (m-2 s-1) increased. However the OH· concentration decayed rapidly with depth, reaching 1 nM within 1 µm. As the droplet size increased, surface concentration and penetration depth both decreased. For electron flux densities up to ~5 x 1022 m-2 s-1, the solvated electron surface concentration was ~0.1 mM and the penetration depth was ~250 nm but at higher fluxes, the penetration depth increases significantly likely due to the reduced concentration of sub-surface H2O2 and OH·, which scavenge electrons. When 1 mM HAuCl4 precursor is added, the surface electron concentration and penetration depth is significantly reduced (< 50 nm), even at very high flux densities. We also observe the loss of surface Au3+ and the increase in Au0 concentration. However, the maximum overall conversion rate is 15%, compared to the experimental observation of >50%. Enhanced conversion would likely require greater precursor transport from droplet centre to surface, either by internal fields or convection. We carried out preliminary simulations of (Au0)N (N: 2 → 20) cluster growth via simple collisions in this environment and observed almost complete conversion of Au0 to cluster (Au0)N with the concentration of (Au0)20 >> (Au0)2.[1] PD Maguire et al., Appl. Phys. Lett. 106, 224101 (2015); doi: 10.1063/1.4922034[2] PD Maguire et al., Nano Lett., 17, 1336–1343 (2017) doi: 10.1021/acs.nanolett.6b03440[3] N Hendawy et al 2020 Plasma Sources Sci. Technol. 29 085010 doi: 10.1088/1361-6595/aba2aa[4] H McQuaid et al 2022, in-press. See preprint - doi: 10.48550/arXiv.2207.04518