Ultraviolet-photon exposure stimulates negative current conductivity in amorphous ice below 50 K

Highlights

• Water ice can be a conductor carrying a negative current by UV irradiation with supplying electrons to the surface.

• The negative current can be sensitively controlled by UV irradiation.

• Negative current conductivity can be preserved even at temperatures below 50 K.

• Quantum chemical calculation suggests that the negative current is carried by proton-hole transfer.

Abstract

Water ice is known to carry positive charge via transfer of excess protons through a hydrogen-bonded system. In contrast, constant negative charge conductivity in ice has neither been confirmed experimentally nor theoretically. Here, we report that ultraviolet-photon irradiation stimulates negative current conductivity in amorphous ice at temperatures below 50 K, where hitherto known proton transfer from the surface through the bulk ice is inhibited. Quantum chemical calculations imply that the negative current conductivity results from almost barrier-less proton-hole transfer processes that neither requires intrinsic defects, doped species, nor reorientation of H₂O molecules in ice.

Introduction

Water ice is well known to deliver positive current similar to a “p-type” semiconductor via the proton transfer of excess hydronium ions [1]. Because of its importance in various chemical and biological processes [1], [2], [3], proton transfer in both liquid water [4], [5], [6] and ice [7], [8], [9] has been extensively studied. Although the detailed mechanism, especially at low temperatures [7], [10], is still under debate, the positive current conductivity in ice can be well described by the relay of proton transfer, the so-called “Grotthuss mechanism” [1]. For the negative current conductivity, the concept of “a proton-hole transfer (PHT)” was proposed about 100 years ago [11]. The PHT is the relay of proton-abstraction of OH⁻ ion from neighboring H₂O, and therefore was considered as a “mirror image” concept of the Grotthuss mechanism. However, to date the efficient negative current conductivity has not been found for both liquid water [12] and ice [13], [14]. In water, the mechanism of OH⁻ migration was found to differ from the PHT concept [15] and recent years, migration of OH⁻ in water was reported to be much less effective than the proton transfer [12]. In water, H⁺ exists as single complex of tricoordinate H₃O⁺(H₂O)₃, and the proton transfer is conducted by the re-formation of this tricoordinate form via intermediate H₅O₂⁺ (i.e. the Grotthuss mechanism). In contrast, OH⁻ in water exists to alternate between two configurations, which are a hypercoordinated form with four acceptor hydrogen bonds to neighbor water molecules and a nearly tetrahedral form with three acceptor hydrogen bonds [15]. Since the proton abstraction only occurs in the tetrahedral form, transformation from the hypercoordinated form to the tetrahedral form is necessary before the proton abstraction [15]. Thus, migration of OH⁻ in water is not the mirror image of the Grotthuss mechanism. This difference may lead to the low efficiency of the OH⁻ relay in aqueous medium. Furthermore, a recent ab-initio molecular dynamics simulation have shown that the multiple proton jumps that enhance the efficiency of the proton relay, namely the concerted proton transfers, are strongly suppressed in the OH⁻ system due to the higher activation barrier than that in H₃O⁺, [12]. This also contributes to low efficiency of the proton abstraction in the OH⁻ in aqueous medium. Eventually, suppression of PHT in water would originate from the hypercoordination form, which is shaped by four H₂O molecules sterically surrounding OH⁻.

In ice, the hypercoordination complex may not be formed due to the limited motion of H₂O molecules unlike in liquid phase. However, no theoretical studies about OH⁻ migration in ice have been reported to date. Thus, the PHT phenomenon in ice still remains unexplored, although it is a fundamental electrochemical property of ice. In experiments, negative current conductivity of ice has been examined by electron bombardment of ice or using base-doped ice, which results in OH⁻ ions in bulk ice. For the doped ice experiment, a small negative charge delivery was observed above 140 K, and was reported to be the result of OH⁻ Brownian migration rather than the PHT [14]. For the electron bombardment of ice, although a transient negative current probably due to the transport of solvated electrons was observed, and it immediately diminished below 50 K [16]. In addition, the negative current conductivity has never been examined when OH⁻ ions on the surfaces of ice. The PHT at the surface may control the negative current conductivity as the first step of PHT relay in bulk. This phenomenon is also relevant to chemistry of ice dust surfaces in space and planetary atmosphere, where it is important to know whether the negative charge on OH can be localized on the surface [17].

Here, we report that ultraviolet-photon (UV) irradiation stimulates negative current conductivity in amorphous ice below 50 K, where positive current by proton transfer from the surface to the bulk ice is significantly suppressed [8]. We found that the negative constant current can be controlled by UV irradiation. No temperature dependence or isotope effect was observed in the range of 10–50 K. Quantum chemical calculations suggest that the negative current results from the PHT in ice, triggered by the surface OH⁻ ions, which are created by photodissociation of H₂O and subsequent electron attachment onto OH radical at the surface. The computed barrier of the proton abstraction by OH⁻ is very low, in accordance with the experimentally observed temperature independence of conductivity. This process requires neither intrinsic defects nor reorientation of the H₂O molecules in bulk ice, in contrast to the well-known positive current conductivity by the transfer of excess protons. The present paper strongly suggests the occurrence of the PHT in ice and may lead to a new aspect of the electrochemical nature of ice, which has been commonly considered as a positive current semiconductor.