2025 Volume 93 Issue 10 Pages 101004
There is a concept called Ecofitting technology. It is a concept that making industrial processes more environmentally friendly leads to a transformation into a more economical industrial process as a result. I have developed several new chemical processes based on this research concept. One of them is the radical vapor reactor (RVR), which is already in practical use (commercially available) as a process device. I found and developed a reaction (called the oxygen P/L reaction or RVR reaction) that can generate high concentrations of hydroxyl radicals and singlet oxygen, which are highly oxidizing among reactive oxygen species, and developed a device that can expose objects to the high-concentration reactive oxygen atmosphere generated by the reaction. I have then put it into practical use as a reactive oxygen species generation and exposure device that can be used in various chemical industrial processes. In this paper, the author, who discovered the RVR reaction, developed the RVR device, and further demonstrated various applications of the RVR, outlines the research and development of the RVR reaction (device design and commercialization), and examples of applications of the commercialized RVR device.
Many chemical processes are based on oxidation or reduction. In these chemical reaction processes, chemicals such as acids and bases are used, and the reaction energy must be increased by heating. Suppose some of these chemical processes can be converted to processes that do not use chemicals. The chemical manufacturing and subsequent waste disposal processes will become unnecessary, creating benefits and advantages such as reduced costs, environmental loads, and carbon dioxide emissions. The authors reasoned that if a chemical process could be constructed that generated high concentrations of reactive oxygen and exposed it, a chemical-free chemical process could be realized. This idea led to our research into the high-concentration generation reaction of reactive oxygen, its application to the exposure process, and the development of a Radical Vapor Reactor (RVR) to convert the process into a device. In this comprehensive paper, we provide an overview of reactive oxygen species, outline how they are generated in environment, and explain the reaction mechanism of the plasma/liquid reaction (RVR reaction) between oxygen and water, which the authors have discovered. We then discuss examples of process applications of the RVR and how its widespread use can lead to reduced environmental impacts, measures against global warming, and the achievement of SDGs.
Most people have heard the term “reactive oxygen,” even if they are not chemists or biologists. Because many types of reactive oxygen exist, the term “reactive oxygen species” (a general phrase for active and radical species generated from oxygen) is used here.
Figure 1 illustrates the main reactive oxygen species defined in the broad sense. The phrase “broad sense” is used because it encompasses both reactive oxygen species that are radicals with unpaired electrons (shown in Fig. 1 red frames) and non-radical reactive oxygen species (shown in Fig. 1 orange frames). The word “main” is used because other types of reactive oxygen species exist, such as atomic oxygen (O), ozone (O3), and molecular oxygen ions (O2+), which are dissociated or excited species of the ground-state triplet oxygen (3O2) that we breathe; these species are not shown in this figure.
Relationships among oxygen, active oxygen, and water in terms of oxidation and reduction.
All these reactive oxygen species have strong oxidizing power (proton abstraction activity); those that are free radicals with unpaired electrons especially so. Due to their high reactivity, in a normal atmosphere, they react with other molecules and atoms (and, of course, with other reactive oxygen species), so their existence as reactive oxygen species is a brief time (less than milliseconds). However, in the gas phase, where the molecular separations are comparatively large, their existence time is longer. From the viewpoint of engineering and industrial physicochemistry, reactive oxygen possesses some noteworthy traits. As shown in Fig. 1, reactive oxygen is generated by reducing oxygen or by oxidizing water. During their use in reduction or oxidation reactions, the species are converted to oxygen and water. In other words, reactive oxygen species, which are strong oxidizing agents, can be generated from only oxygen and water, and after their use, only oxygen and water remain. If a reactive oxygen oxidizing agent can be generated from oxygen and water, no chemical waste needs to be treated after use. This possibility has the potential to realize significant reductions in the costs of operating chemical processes and the resulting environmental load (carbon dioxide emissions). A potential advantage is worth noting here.
Looking at active oxygen (broad definition) from the perspective of industrial use, hydrogen peroxide, which is relatively stable and has a moderate oxidizing power, is used in bleaching processes (waste paper recycling,1 textile manufacturing,2 dentistry3), semiconductor manufacturing processes (etching and cleaning),4 food manufacturing (cleaning and sterilization), and medicine (hydrogen peroxide disinfectant). However, current hydrogen peroxide production relies on the anthraquinone oxidation method, a multi-step synthetic reaction that heavily utilizes organic solvents, resulting in that differ from the conceptual formation of active oxygen, which is “made from oxygen and water and returns to oxygen and water.”
To measure the global warming of recent years and achieve progress toward the Sustainable Development Goals (SDGs), we need technology to generate active oxygen species and process application technology that can realize “made from oxygen and water and returns to oxygen and water.” These needs served as the motivation for developing an RVR, which was the subject of this study.
However, little research exists on reaction technologies for artificially generating reactive oxygen species. Many in vitro systems exist for generating reactive oxygen species, utilizing metal catalysts or biocatalysts (enzymes) in aqueous solutions or organic solvents. Examples include the hydroxyl radical generation method using aminocarboxylate iron(III) and hydrogen peroxide5 and the generation of superoxide anions by enzyme reactions (xanthine oxidase),6 all of which are intended for application in biochemical or medical experiments. There has been no research aimed at expanding industrial processes.
The author began to study the development of a reactive oxygen species-generating reaction that is “made from oxygen and water and then returns to oxygen and water” for the purpose of applying it to industrial processes. There is a concept that when using a catalyst, such as an enzyme, the selectivity for generating reactive oxygen species is high. This is not necessarily correct; even if only one type of reactive oxygen species is present, a disproportionation reaction or chain reaction with other molecules or atoms will immediately proceed and generate a different type of reactive oxygen species. However, considering their use in oxidation processes, since all reactive oxygen species possess strong oxidizing power, it can be assumed that no problem exists even if various reactive oxygen species are generated simultaneously as long as they can be generated cheaply and in large quantities.
Looking back at Fig. 1 here, we notice that reactive oxygen species can be generated by reducing oxygen or oxidizing water. This insight led to the discovery of the plasma/liquid (P/L) reaction (RVR reaction), which is discussed in the next section.
To repeat, when considering reproducing the conceptual process of “being made from oxygen and water and then returning to oxygen and water” as a chemical reaction (and building a system that can expose this to the target of treatment as a chemical process), we thought that if we dissociated oxygen molecules in advance and reacted them with water, the dissociated oxygen would extract protons from the water molecules and be reduced to produce active oxygen (hydroxyl radical). This would provide a method to reduce oxygen without a catalyst. The challenge was to dissociate the oxygen molecules.
The dissociation energy of oxygen is 493.7 kJ/mol. This energy amount can be considered small for a double bond. In comparison, the dissociation energy of the C=C bond is 610.8 kJ/mol, and that of the C=O bond is 744.7 kJ/mol. Therefore, it can be thought of as possible to dissociate oxygen molecules with relatively small energy input. The motivation for the research was to “generate high concentrations of active oxygen species from oxygen (or water)” and “expose the active oxygen species to the process object.” In other words, the dissociation is best performed in a gas-phase system. This insight led to the idea of dissociating oxygen molecules using a gas-phase discharge.
Ozone is often thought to be produced when oxygen is discharged; however, this is not accurate. Of course, the manner in which the discharge occurs (and its effect on the gas phase) varies depending on the discharge form and the amount of energy input during the process, but in the case of oxygen gas phase discharge, O2+, O2(W), O(1D), O, O2(b1Σg+), O−, O2(a1Δg), and O2− are produced. The same atomic oxygen can have different energy levels (in which case they are indicated by a subscript to the quantum symbol), and the generation ratio changes depending on the conditions. The atomic oxygen (O(1D), O) generation evaluation can be performed by qualitative analysis and relative amount using the plasma emission spectrum at the discharge locus. Using this method, the discharge conditions were specified, the discharged electric phase was brought into contact with water under conditions of a high amount of atomic oxygen, and an oxidation reaction analysis (colorimetric evaluation using redox indicators) was performed at the gas–liquid interface (a Petri dish filled with pure water was placed in the discharged oxygen gas phase). The oxidation reaction was found to have proceeded. Because it was suggested that reactive oxygen species were generated at the gas–liquid interface under these conditions, we decided to quantify the amount of reactive oxygen species generated using spin-trapping electron spin resonance spectroscopy (st-ESR) analysis.
Because reactive oxygen species have a short existence time, spin-trapping agents are dissolved in the aqueous phase in advance, and spin adducts are generated when reactive oxygen species are generated at the air–liquid interface. When spin adducts are generated, the half-life of the reactive oxygen species, which normally exist for less than milliseconds, can be extended to minutes or even hours depending on the reactive oxygen species. Spin-trapping agents also exhibit selectivity toward reactive oxygen species. St-ESR analysis revealed that hydroxyl radicals and singlet oxygen were generated at high concentrations. Figure 2 shows ESR spectra of the hydroxyl radicals and singlet oxygen measured by st-ESR analysis and the structure of each spin adduct. The spin numbers of the hydroxyl radicals and singlet oxygen were calculated by comparing the relative peak area of the ESR spectrum and the ESR peak area of a spin standard substance (Tempol), from which the molar concentration of each reactive oxygen species generated was obtained.
Representative spin-trapping reagents and their spin adducts formed with hydroxyl radical (·OH) and singlet oxygen (1O2) and their typical st-ESR spectra.
In an attempt to generate reactive oxygen species and expose them to heat in the reaction space, the water phase was treated as water vapor, and the amount of reactive oxygen species generated at the interface between the gas phase and the water vapor phase (water phase) in which oxygen was discharged was measured by changing the temperature in the reaction system within the range 25–40 °C. The amount of reactive oxygen species generated in the water vapor phase was analyzed using st-ESR, and the results are shown in Figs. 3 and 4.7 The amount of hydroxyl radicals increased with increasing temperature in the reaction system, peaked at 3–4 min, and then decreased (Fig. 3). The half-life of DMPO-OH, a spin adduct formed using the spin-trapping agent 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) and hydroxyl radicals, was 30 min. Therefore, measurements on this time axis can be quantified (albeit by using a one-handed stopwatch). However, because the half-life of this DMPO-OH spin adduct is shortened at higher temperatures, it is possible that measurements after 2 to 3 minutes of reaction time may not reflect the correct hydroxyl radical amount and may be low. Therefore, we investigated the temperature dependence of singlet oxygen generation using 2,2,5,5-tetramethyl-3-pyrroline-3-carboxamide (TPC) as a spin trap (Fig. 4). The spin adduct of TPC-1O2 has a long half-life (3 h) and is relatively thermally stable, so we reasoned that there should be no problems with the analysis within the experiment temperature range (25–40 °C). The singlet oxygen generation increased depending on both the temperature and reaction time, suggesting that the reactive oxygen generation reaction at the interface between the gas phase and the water vapor phase (aqueous phase) can continuously generate both hydroxyl radicals and singlet oxygen within a time range of up to 10 min. Furthermore, from the results of attempts to irradiate the gas phase/water vapor phase with ultraviolet light, we concluded that a reactive oxygen species generation reaction proceeds in which atomic oxygen extracts a proton from water and reduces it, as shown in the following reaction formula.7
Dependence of UV irradiation time on amount of ·OH radical in RVR. The RVR was operated with RVR standard conditions but with different UV irradiation times, as shown in the graph. (Reprint source: Ref. 7)
DPI reaction time and 1O2 production in the RVR. Ozone was injected for 2 min at 4 dm3/min and UV irradiation was carried out with the DPI of water for the indicated times on the x-axis. (Reprint source: Ref. 7)
Reactions (1) and (2) are well-known reactions for ozone generation by discharge (electrons in the reaction formulas are donated by electron collisions from discharge). Reactions (3)–(6) are the plasma/liquid (P/L) reactions. This P/L reaction is the world’s first achievement not only of the reduction of oxygen with water (generation of reactive oxygen species) but also of the reduction of nitrogen with water, that is, the reductive fixation of nitrogen (ammonia synthesis) without the need for hydrogen (H2).8–10 Our P/L research results are pioneering a new field of reaction chemistry, an ammonia synthesis method that does not require hydrogen (using water as a direct hydrogen source), and have opened up a new field in which many researchers around the world are involved.11
The P/L reaction of oxygen and water is also called the radical vapor reaction because its purpose is radical production and exposure. However, to transform this radical vapor reaction into an actual process technology, a reaction device (reactor) that can stably and continuously proceed with this chain reaction is required, as well as a process device that can expose the target object to the high concentration of active oxygen species produced there. We believed that the only way to provide such multiple functions, that is, to simultaneously proceed with the production reaction and exposure process, is to design it as a device rather than as an experimental instrument. Therefore, we built a device that can simultaneously achieve radical vapor production and radical vapor exposure.
An RVR is a device that simultaneously performs radical vapor production and radical vapor exposure. As shown in Scheme 1, the P/L reaction between oxygen and water generates active oxygen through multiple reactions that proceed in a chain. In this study, this chain reaction is referred to as the radical vapor reaction. It is not enough to create a reaction system in which this reaction proceeds continuously; it is also necessary to expose the target object to water vapor that contains a high concentration of the generated active oxygen species. Therefore, we considered carrying out the reaction shown in Scheme 1 in a water vapor atmosphere and defining the space where the amount of radical vapor production (the amount of active oxygen species produced) was the highest as the radical vapor exposure locus.
The reactions in the P/L reaction with oxygen gas and water (RVR).
Reactions (1)–(3) in Scheme 1 are well-known oxygen dissociation and ozone generation reactions caused by oxygen discharge. A discharger was used to promote the reaction. Atomic oxygen and ozone generated by discharging pure oxygen were introduced into an insulated container containing a water vapor generator. The water vapor generated from this reaction reacted with atomic oxygen, and reactions (4) and (5), shown in Scheme 1, occurred. However, the hydroxyl radicals generated in reaction (4) were consumed in reaction (5), lowering the concentration, and singlet oxygen was not generated. Therefore, to advance reaction (6), the reaction space was irradiated with ultraviolet light with wavelengths of 185 and 254 nm. As shown diagrammatically in Fig. 5, the discharged oxygen (atomic oxygen and ozone, a), water vapor (b), and ultraviolet light (c) intersect at the sample stage (exposure field, d) in an insulated sealed container, and reactions (4), (5), and (6) of Scheme 1 proceed to the maximum extent within this system. With this configuration, the generated reactive oxygen species (hydroxyl radicals and singlet oxygen) can be exposed to the target object on the sample stage; that is, we have succeeded in constructing a process device, the “RVR”, that simultaneously performs radical vapor production and radical vapor exposure.7 The RVR, which the author originally handcrafted from a collection of materials, has subsequently been successfully commercialized by EBARA JITSUGYO Co., LTD. (Tokyo, Japan) as an RVR (EKBIO-1100), which is an oxygen radical generation and exposure device with excellent safety and operability (Fig. 6).
Instrumental setup for the radical vapor reactor (RVR). (Reprint source: Ref. 7)
Commercially available Radical Vapor Reactor (EKBIO-1100). Manufactured and sold by EBARA JITSUGYO CO., LTD. (Tokyo, Japan).
The applications of RVR in chemical processes include “surface modification of materials surface,” “organic matter decomposition,” “advanced sterilization,” and “organic chemical synthesis.” Several examples are presented in this section.
5.1 Decomposition of organic matterA Si substrate was prepared by chemically modifying polyethylene glycol (PEG) via silane coupling, and the unmodified Si substrate and PEG-modified Si substrate after the RVR treatment were observed by atomic force microscopy (AFM) (Fig. 7). The AFM image shows that the surface of the bare Si substrate is flat even at the nanometer scale (Fig. 7a). The flat surface was chemically modified with PEG via silane coupling and the substrate was ultrasonically cleaned. When the substrate was observed using AFM, modified polymer molecules were observed on the surface (Fig. 7b). The Si substrate, which was chemically modified with PEG, was treated with oxygen radicals via RVR, ultrasonically cleaned, and observed using AFM. Most of the chemically modified polymers on the surface had disappeared (Fig. 7c). This result indicated that the chemically modified PEG on the Si substrate was decomposed and removed by the RVR treatment.12
AFM image of Si substrate modified with PEG: (a) Si substrate after washing by sonication (28 kHz, 5 min) in deionized water; (b) Si substrate modified with 10 % PEG solutions (mixture of 2-[methoxy(poly-ethyleneoxy)propyl]trichlorosilane, tech-90, and toluene, 1 : 9) after washing under sonication (28 kHz, 5 min) in toluene, ethanol, and ethanol, in that order; (c) substrate of (b) after RVR treatment. The measurement ranges were 5 µm × 5 µm × 10 nm. In these measurements, a PPP-NCHAu probe (42 N/m, Nano World) was used for imaging. Imaging was carried out in tapping mode and under atmospheric conditions. (Reprint source: Ref. 12)
The PEG decomposition was analyzed by Fourier transform infrared (FT-IR) analysis. The FT-IR spectra of the PEG-modified Si substrate before and after RVR treatment (PEG/Si) were measured (Fig. 8). Two major peaks were observed for PEG/Si before the RVR treatment (Fig. 8, top). The IR absorption peaks at 3000–2840 cm−1, 1260–1000 cm−1, and 1150–1080 cm−1 correspond to the C–H stretching vibrations of alkanes, C–O stretching vibrations of ethers, and C–O–C antisymmetric stretching vibrations of ethers, respectively. In addition, a small peak at 1600–1400 cm−1 corresponds to the C–H bending vibration of alkanes, indicating the presence of PEG on the substrate. After the PEG-modified surface was exposed to oxygen radicals by RVR, a new IR absorption peak was observed at 1740–1720 cm−1, corresponding to the C=O stretching vibration of the aldehydes. These aldehyde groups were generated by the decomposition of PEG upon exposure to oxygen radicals in the RVR. To clarify the decomposition process, PEG in the RVR was analyzed before and after oxygen radical exposure using MALDI-TOF-MS. As untreated PEG, a single major peak was observed for the untreated PEG. The molecular weight of PEG before the RVR treatment (before decomposition by oxygen radicals) was 379.724 g/mol, based on the peak at m/z 379.724 (Fig. 9a). In contrast, in the mass spectrum of PEG after oxygen radical exposure in the RVR, the intensity of the peak at m/z 379.729 decreased, whereas that of the other peaks increased. The three major peaks at m/z values lower than the major peak (m/z 335.745, 294.762, and 212.896) matched the mass peaks indicated by the corresponding numbers in the spectrum in Fig. 9b and corresponded to the masses of the fragments of the decomposed PEG molecule (Fig. 9c). These results demonstrate that oxygen radical exposure treatment using RVR can decompose organic polymers (break them down into smaller molecules) and that the chemical process of decomposing organic matter (removing organic contaminants) can be achieved using only water and oxygen.12
FT-IR spectra of the Si substrate before and after the RVR treatment. Analysis was performed from 4000 to 500 cm−1. Before the RVR treatment, the Si substrate modified with PEG (black line) showed two major peaks (at approximately 2900 cm−1 from alkane C–H stretching and 1100 cm−1 from ether C–O stretching) and some small peaks (at approximately 1440 cm−1 from alkane C–H bending). After RVR treatment, the PEG-modified Si substrate (orange line) showed an additional peak (at 1700 cm−1) attributed to aldehyde C=O stretching. (Reprint source: Ref. 12)
MALDI-TOF-MS data (a) before and (b) after RVR treatment. Before the RVR treatment, there is a single peak (379.724 m/z) for the PEG raw material. After the RVR treatment, the intensity of the peak of the raw material decreases and that of the other peaks on the low molecular side increases. The three new peaks correspond to the molecular weights when cut at the three locations indicated in (c). (Reprint source: Ref. 12)
Carbon is used as an electrode in electrolytic and battery systems; therefore, surface modification of carbon materials is an interesting application in the field of electrochemistry. The surface of highly oriented pyrolytic graphite (HOPG) was measured by X-ray photoelectron spectroscopy (XPS) using a monochromatic AlKα source before and after oxygen radical exposure treatment by RVR. Wide-range O 1s and C 1s spectra are shown in Fig. 10. Before RVR treatment, no major peaks were observed in the O 1s spectrum; after RVR treatment, peaks due to hydroxyl C–OH and epoxide (533.2 eV) and COOH (533.8 eV) were observed in the O 1s spectrum.13
XPS spectra: (a) wide range, (b) O 1s spectra, and (c) C 1s spectra. (Reprint source: Ref. 13)
Before the RVR treatment, in the binding energy range of 283–294 eV, three distinct peaks were observed in the spectrum, one of which was associated with the sp2 C–C 1s bond (285.0 eV) and two minor peaks attributed to hydroxyl C–OH (286.5 eV), C–O (286.5 eV), epoxide (286.3 eV), and O–C–O (287.9 eV). After oxygen radical exposure treatment by RVR, the spectrum showed four clear peaks in the binding energy range of 283–294 eV, of which one major peak was associated with the sp2 C–C 1s bond (285.0 eV) and three secondary peaks were due to the hydroxyl C–OH (286.5 eV), C–O (286.5 eV), epoxide (286.3 eV), O–C–O (287.9 eV), and COOH (288.8 eV) functional groups. These results revealed that oxygen was introduced (modified) onto the HOPG surface by oxygen radical treatment using RVR.13 Previous studies have reported that the C–C bonds are cleaved by singlet oxygen and hydroxyl radicals.14–17 The results suggest that bond dissociation and oxygen introduction (modification) proceed via chain reactions.
This paper outlines only a few applications of the RVR. Oxygen radical exposure by RVR is also used in various other ways, including sterilization, mutagenesis, and organic chemical synthesis. For example, persister bacteria (bacteria that have become dormant or stationary and have become antibacterial-resistant and highly resistant to chemical sterilization) cannot be sterilized by chemicals, which poses a challenge in the sterilization of medical equipment; however, oxygen radical exposure by RVR can effectively sterilize persister bacteria.18
Discharge involves gas activation (dissociation, excitation, ionization, or transfer of kinetic energy to molecules and atoms) by electrons. The P/L reaction that we discovered can be considered to have significantly advanced the chemical reactions of gases and their engineering applications. For the P/L reaction of nitrogen and water, we achieved the world’s first ammonia synthesis without using hydrogen (H2 gas). In the P/L reaction of oxygen and water, as described in this paper, we succeeded in generating high concentrations of oxygen radicals and successfully commercialized and launched the world’s first chemical process device, the RVR, which generates and exposes reactive oxygen species.
The P/L and RVR reactions are reactions in which gas is reduced by water. Another chemical reaction between plasma and water that we discovered is noble gas plasma-collisional splitting (NgPCS).19 This involves water splitting and hydrogen generation using the kinetic energy of noble gas plasma particles. Noble gases do not react with water, and their kinetic energy can be used to split water (to generate hydrogen) efficiently, creating a new reaction technology. The research and development of this NgPCS is also ongoing.
This series of research is based on the authors’ fundamental concept of “Turning Air and Water into Resources: Elemental Circulation Chemistry”, and the concept and chemistry/technology can be found in detail in books written by the authors.20
Nitrogen, oxygen, argon, and carbon are the main atmospheric elements. Hydrogen and oxygen are the elements that constitute water. We believe that sustainability can be increased if we can actively utilize industrial processes that actively combine and utilize these elements and then return them to their original air and water. The authors and our research group will continue to develop and commercialize this chemical technology as an industrial technology.
This study required a long time to complete. This was the result of the many people who taught me over a long period and of the many students, graduate students, and postdoctoral researchers who worked diligently on the research. I am deeply grateful to you all. I am also deeply grateful to EBARA JITSUGYO CO., LTD., who led the research on our own prototype to produce a device with excellent operability, safety, and commercialization. At the same time, I will continue to work towards further development of RVR application technologies and P/L reactions, and strive to expand sustainable industries.
The APC for the publication of this paper was supported by The Electrochemical Society of Japan.
Tetsuya Haruyama: Conceptualization (Lead), Investigation (Lead), Methodology (Lead), Project administration (Lead), Writing – original draft (Lead), Writing – review & editing (Lead)
The authors declare no conflict of interest in the manuscript.
T. Haruyama: ECSJ Fellow
Tetsuya Haruyama (Professor, Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology)
He graduated from Graduate School of Science and Engineering, Tokyo Institute of Technology in March 1993, and received Doctor of Engineering, He worked as Assistant Professor in Tokyo Institute Technology in 1993–2001. He moved to Kyushu Institute of Technology as Associate Professor in 2001. Full Professor in 2002. He was awarded Seiyama Award from Japan Association of Chemical Sensors in 2011, Etoh-Hosoya Memorial Award from The Futaba Foundation in 2024, Technical Development Award (Tanahashi Award) from The Electrochemical Society of Japan in 2025. His research interests are Chemical reactions at the interface between different phases including electrochemical reactions.
