Acta Chim. Slov. 2005, 52, 1-12 Review Article Progress in the Chemistry of Dihydrogen Trioxide (HOOOH) Božo Plesničar Department of Chemistry, Faculty of Chemistry and Chemical Technolog,;, University of Ljubljana, P.O.Box 537, 1000 Ljubljana, Slovenia, E-mail: bozo.plesnicar@uni-lj.si Received 23-02-2005 Abstract During the past 15 years, there has been considerable progress in imderstanding of the chemistry of dihydrogen trioxide (HOOOH). In this account, the preparation, characterization and structure (XH, 170 NMR, IR, quantum chemical methods), kinetics and mechanism of decomposition, and reactivity of this simplest of polyoxides are discussed. Dihydrogen trioxide is, in organic oxygen bases as solvents, far more stable than previously believed. For example, the half-life of HOOOH in acetone-fi^ is 16 + 2 min; hovvever, it decomposes rather quickly in water solutions (t1/2 « 20 ms) at room temperature. HOOOH is most likely involved in oxidation processes that span atmospheric, environmental, and biological systems. Key words: ozone, low-temperature ozonation, dihydrogen trioxide HOOOH, density-functional theory, ab-initio, molecular-structure, hydrogen-bonds, HOOO anion, HOOO radical, alkyl hydrotrioxides ROOOH, thermal-decomposition Contents Introduction...............................................................................................................1 Origins of Dihydrogen Trioxide Chemistry.............................................................1 Direct Spectroscopic Detection of Dihydrogen Trioxide ......................................2 Methods of Preparation of Dihydrogen Trioxide in Organic Solvents. NMR (JH and 170) and IR Spectra.........................................................................3 Structure, Aciditv, and Thermochemical Data.......................................................5 Decomposition of Dihvdrogen Trioxide..................................................................8 Reactivity of Dihydrogen Trioxide...........................................................................9 Conclusions and Outlook.......................................................................................10 References ...............................................................................................................10 Introduction Polyoxides are compounds of the general formula ROnR, where R is hydrogen or other atoms or groups, and n > 3. These species may be regarded as higher homologues of hydrogen peroxide, alkyl hydroperox-ides (ROOH), and dialkyl peroxides (ROOR), and are believed to be key intermediates in atmospheric chemistry, the chemistry of combustion and flames as well as in biochemical oxidations.1 Because the chemistry of alkyl hydrotrioxides (ROOOH) and dialkyl trioxides (ROOOR) has been rather well documented in recent years,J we will con-centrate here mainly on the formation, characterization, decomposition, and reactivity of dihydrogen trioxide (HOOOH), i.e., the simplest of polyoxides. This is to the best of our knowledge the first review devoted entireh/ to dihydrogen trioxide. The proposed IUPAC name for HOOOH is trioxidane. Origins of Dihydrogen Trioxide Chemistry Although hydrogen peroxide has been known since 1818 (L. J. Thenard), the existence of higher ox-ides of hydrogen was a subject of heated controversy and speculation ever since 1880, when M. Berthelot proposed the possibility of transient existence of dihy-drogen trioxide as an intermediate in the decomposition of hydrogen peroxide.2 Fifteen years later, Mendeleev Plesničar Progress in the Chemistry of Dihydrogen Trioxide 2 Acta Chim. Slov. 2005, 52, 1-12 suggested dihydrogen tetraoxide (HOOOOH) as stili another unstable intermediate in the same reaction.3 Ali these suggestions were soon forgotten because there were no methods available at that time to test these hypotheses. In 1917 Rothmund and Burgstaller were the first to investigate the kinetics of the reaction betvveen ozone and hydrogen peroxide.4 Several groups of workers later suggested the free radical nature of this reaction with the involvement of HO", HOO", the superoxide anion (02*~), and the ozonide radical anion (03*~) as interme-diates. The potential involvement of HOOOH, resulting from recombination of HO" and HOO" radicals, was also considered.5~7 In the 1950s and 1960s several groups of research -ers studied the dissociation of water vapor by means of an electrical discharge.8 When the discharged vapor stream was chilled quickly to -190 °C, a glassy solid was obtained, which upon warming yielded large amounts of oxygen and hydrogen peroxide. This was, together with the heat evolved, taken as a proof for the involvement of H203 and H204 in these reactions. Direct Spectroscopic Detection of Dihydrogen Trioxide UV and IR spectra. The modern chemistry of HOOOH begun in 1963, when Czapski and Bielski reported the production of an intermediate, believed to be dihydrogen trioxide (on kinetic grounds), in the pulse radiolysis of acidified air-saturated aqueous solutions.9 Later Bielski and Schwartz provided the first UV spectroscopic evidence for the existence of HOOOH in the pulse radiolysis of air-saturated perchloric acid solutions.1011 A UV absorption spectrum very similar to, but more intense than, that of HOOH was reported. These authors also estimated the maximum half-life for HOOOH in aqueous solutions of 17 seconds in 0.027 M acid at 0 °C. The first attempts to detect hydrogen polyoxides by IR were made in the late 1960s and 1970s, when Nekrasov et al. studied the low-temperature (-198 °C) reaction of ozone with atomic hydrogen in liquid ozone as well as the dissociation of water and hydrogen per-oxide vapor with subsequent freezing of the products into the matrix. The IR spectra of the products showed some new bands, which were tentatively assigned to H203 and H204, but no unambiguous interpretation of the IR spectra could be made at that time.12 Giguere et al. were the first to report in a series of papers fundamental skeletal vibrations of H203 and H204 and their deuterated analogs by studying IR and Raman spectra of the products from electrically dissoci-ated mixture of water, hydrogen peroxide and oxygen.13 However, the highest concentration of H203 produced in the glassy solid was below 5% (only one tenth as much for H204). They also reported that H203 begins to decompose at about -55 °C (H204 at about -100 °C), and estimated that at room temperature it could survive only a small fraction of a second. A recent report by Engdahl and Nelander pre-sented the IR spectroscopic identification of dihydrogen trioxide in argon matrices obtained by photolyzing the ozone-hydrogen peroxide (or DOOD and HOOD) mixtures.14 Ali fundamental vibrations of HOOOH were observed, and they agree well with the MP4/6-31G(d,p)14 and B3LYP/6-31G(d,p)15 calculations (Table 1). Several bands of DOOOD, HOOOD, and H16016018OH (1803 and H202) have also been measured. The strongest bands are the torsions, the OH stretches, and the antisymmetric HOO bands. The antisymmetric O-O stre teh at 776 cm4 may be used to detect HOOOH even in the presence of high water concentrations. Biographical Sketch Božo Plesničar studied chemical technology at the University of Ljubljana, from which he received his Ph.D. degree (Chemistry) in 1966. In 1966/1967, he spent one year as a postdoctoral research associate with Professor N. A. Milas at the Massachusetts Institute of Technology (MIT) (Cambridge, USA), and again in 1970 half a year as a NSF postdoctoral fellow with Professor Glen A. Russell at the Iowa State University (Ames, USA). He started at the University of Ljubljana in 1974 as a Dozent, and was promoted to Full Professor in 1985. His present research interests in the field of physical organic chemistry include the characterization of reactive intermediates formed in the oxidation of various organic substrates with ozone (ROOOH, HOOOH). In 2004, he received the Zois Award for research excellence in the field of Organic Chemistry. Plesničar Progress in the Chemistry of Dihydrogen Trioxide Acta Chim. Slov. 2005, 52, 1–12 3 Table 1. Observed and calculated ((MP4/6-31G(d,p))a and [B3LYP/6-31G(d,p)]b) fundamentals of HOOOH (trans) and DOOOD in wave-numbers (cm–1).c HOOOH DOOOD vi (symmetric OH stretch) 3529.6 (3529.6) 2610.4 [3533.1] [2582.2] v2 (antisymmetric OH stretch) 3529.6 (3529.6) 2610.4 [3528.4] [2579.0] v3 (antisymmetric HOO bend) 1359.1 (1359.2) 1007.3 [1353.6] [1013.6] v4 (symmetric HOO bend) 1347.4 (1347.6) (NR)rf [1349.6] [1004.2] v5 (symmetric OO stretch) 821.0 (820.4) (NR)rf [899.5] [892.8] v6 (antisymmetric OO stretch) 776.3 (776.1) 762.6 [770.2] [769.8] v7 (OOO bend) 509.1 (509.2) (NR)rf [506.5] [492.6] v8 (antisymmetric torsion) 387.0 (387.0) 301.6 [410.4] [317.1] v9 (symmetric torsion) 346.4 (345.9) 273.5 [346.9] [252.8] " Ref. 14. b Ref. 15. c Calculated fundamentals are listed in pa-rentheses. d NR; not reported. Methods of Preparation of Dihydrogen Trioxide in Organic Sol- vents. NMR (1H and 170) and IR Spectra Ali the samples of HOOOH obtained by one of the above-mentioned procedures, contained very small amounts of the trioxide (< 5%), and were thus unsuit-able for studying its chemistry in detail. Therefore, a method was needed to prepare relatively highly concen-trated solutions of HOOOH in various organic solvents in order to allow an unambiguous spectroscopic iden-tification and characterization, and to enable the study of the decomposition and reactivity of this polyoxide. Basically, three methods emerged for the preparation of HOOOH in solutions of organic solvents: (a) reduction of ozone; (b) decomposition of hydrotrioxides; and (3) the reaction of ozone with hydrogen peroxide. (a) Reduction of ozone. Several compounds (re-ducing agents-AH2), i.e., l,2-diphenyihydrazines,1617 2-e thylanthrahydroquinone,16 isopropyl alcohol,18 isopropyl methyl ether,18 and cumene19 were found to react with ozone in various organic solvents at low temperatures to form HOOOH. The results of theoretical studies (the gas-phase and the solution-phase) showed that the reaction proceeds by either a step-wise radical process (path a) involving the abstraction of the hydrogen atom from the substrate by ozone to form the hydrotrioxyl radical, HOOO",17'19 or an ionic process (path b) that involves the abstraction of the hydride ion to form the hydrotrioxide anion, HOOO.20'21 These hydrotrioxide species, i.e., HOOO"22 and HOOO 23, the existence of which was recently confirmed by a combination of experimental and quantum chemical studies,22'23 react further by abstracting the hydrogen atom and the proton, respectivelv, to form HOOOH (Scheme 1). AH OOOH AH QOOOH ' HOOOH + A Scheme 1 The ozonation of l,2-diphenylhydrazine (hydra-zobenzene) in acetone-Je at -78 °C is, at least at present, the method of choice for the preparation of relatively highly concentrated solutions of HOOOH (up to 0.1 M) without the interfering presence of organic hydrot-rioxides (ROOOH) (Scheme 2).1617 Besides 1,2-diphe-nyldiazene (azobenzene), l,2-diphenyldiazene-N-oxide (azoxybenzene), and hydrogen peroxide are also formed in this reaction. This method allows the use of com-mercially available l,2-diphenylhydrazine without any further purification. Methyl acetate or tert-butyl methyl ether can also be used as solvents. JH and 170 NMR,17 and IR24 spectra of dihydrogen trioxide obtained in this way are shown in Figure 1. H a o3 (-78 °C) solvent HOOOH + a Scheme 2 Most recently, Wentworth et al. prepared HOOOH free of the contamination products, i.e., 1,2-diphenyldiazene and l,2-diphenyldiazene-N-oxide, by the low-temperature ozonation (-78 °C) of the resin-bound l,2-diphenylhydrazine in acetone-Je or tetrahy-drofuran-d,,,.25 The oxidized resin can be removed by N Plesničar Progress in the Chemistry of Dihydrogen Trioxide 4 Acta Chim. Slov. 2005, 52, 1–12 -20 °C, acetone-d6 1H NMR (a) 13.1 HOOOH 10.0 HOOH ¦lili. i 1 ppm 12 10 8 6 4 2 0 -60 °C, acetone (c) IK 1.00- 0.95- 0.90- 0.85- HOH ^ 0.80- 0.75- rv /hOOOH \ HOOH II 3700 3500 3300 3100 -1 cm Figure 1. (A) 1H, (B) 170 NMR, and (C) IR spectra of dihydro-gen trioxide generated by the low-temperature ozonation of l,2-diphenylhydrazine in acetone-^. Absorptions of dihydro-gen trioxide completely disappeared at +20 °C. (B) 170 NMR chemical shifts in parentheses refer to the calculated values (GIAO/CCSD(T)/qz2p). (Ref. 17, 24). filtration under argon, so that the remaining solutions contained only HOOOH, hydrogen peroxide, and wa-ter. However, it appears that considerable amounts of HOOOH are lost in this way. Namely, we found that HOOOH forms relatively strongly hydrogen-bonded complexes with l,2-diphenyldiazene, thus stabilizing the polyoxide.17'26 (b) Decomposition of hydrotrioxides (ROOOH). In the decomposition of dimethylphenylsilyl hydrotri-oxide in acetone-Je at -60 °C (generated by the low-temperature ozonation of the corresponding silane), it was obvious from JH NMR spectra that besides other decomposition products, stili another polyoxide species was formed.27 Namely, the disappearance of OOOH (13.8 + 0.1 ppm) and CH3 (0.55 ppm) absorptions of the hydrotrioxide (ROOOH) was accompanied by a simul-taneous appearance of another low field absorption at 13.4 + 0.1 ppm. Since this OOOH absorption belonged to a species with exchangeable protons, it was, in the absence of any other absorption that could be attributed to an organic intermediate, assigned to HOOOH. This absorption also disappeared at higher temperatures and did not reappear upon re-cooling. The assigne-ment was later also confirmed by comparing the NMR spectra with those of an authentic sample of HOOOH (JH and 170 NMR).17 Similar observations were made in the decomposition of the dimethylphenylsilyl hydrotri-oxide in some other solvents (methyl acetate, dimethyl ether), as well as in the decomposition of trimethylsilyl hydrotrioxide and some trialkylgermyl hydrotrioxides. More recently, we found that some hydrotrioxides of l,3-dioxolanes also decompose to form HOOOH.21b Although the mechanism of the formation of HOOOH from the hydrotrioxides (ROOOH) is stili under investigation, our preliminary results show that water, and, in the čase of organometallic hydrotrioxides, the coresponding silanols and germanols, most likely play an important role in these reactions (Scheme 3). R3x ^o—o — OH R13XOR ' + HOOOH /H 'cr R2 (R1= CH3, Ph ; R2= H, R3X; X = Si Ge) Scheme 3 (c) The reaction of ozone with hydrogen peroxide (the “peroxone process”). As already mentioned before, Engdahl and Nelander showed that mixing ozone and hydrogen peroxide in an argon matrix produced a com-plex that, when photolyzed at 266 nm, yielded significant concentrations of HOOOH.14 The reaction betvveen 03 and HOOH is known for almost a century. It has been named both the “peroxone process” and an “advanced oxidation process”.4'6'28 This mixture is far more reactive than either component alone in treating soil, groundwater, and wastewater contaminated with various volatile organic compounds, polycyclic aromatic hydrocarbons, polychlorinated bi- T Plesničar Progress in the Chemistry of Dihydrogen Trioxide Acta Chim. Slov. 2005, 52, 1–12 5 phenyls (PCBs), benzene, toluene, ethylbenzene, and xylene (BTEX), trinitrotoluene (TNT), and some other waste constituents.28b More recently, the reaction between 03 and HOOH has become of biological interest since a research group from the Scripps Research Institute (La Jolla, California), led by Lerner, Wentworth, and Eschenmoser, has discovered that aH antibodies are capable of catalyzing the oxidation of water by singlet oxygen (A^j) to generate HOOH and an oxidant with the chemical signature of 03. These authors postulated that antibodies carry the reaction through HOOOH as a key reaction intermediate. Ali these discoveries have important implications for pathological processes in living systems.29 Wentworth et al. have recenth/ reported that HOOOH is actually formed during the thermal reaction of HOOH with 03 in acetone-Je or tetrahydrofuran-J;o at -78 °C in concentrations of approximately 2.9 mM (acetone-Je). The trioxide was characterized by JH NMR spectroscopic analysis (OOOH absorption at 8 = 13.6 + 0.2 ppm).25 By using quantum chemical methods, Xu and God-dard have proposed that in the gas phase the reaction betvveen HOOH and 03 proceeds via a planar seven-membered-ring triplet biradical intermediate (with the triplet and singlet states being almost degenerate) to yield HOOOH and 302 (Scheme 4).15 The enthalpy of the overall reaction was found to be exothermic by 32.3 kcal/mol, with the highest enthalpy of activation of 4.8 kcal/mol (B3LYP/6-31G(d,p)). Alternatively, Lesko et al. have recently, in order to explain hydrogen isotope effects on the reaction of HOOH with 03 in water, concluded that HOOO anion is involved in these re-actions. The conjugate base of HOOOH is most likely formed from a reaction betvveen the HOO anion and 03.30 Clearly, further studies are needed to clarify the mechanistic details of this interesting and important reaction. HOOH + 03 O—O,* Scheme 4 HOOOH + 302 Structure, Acidity, and Thermochemical Data Structure. Because the structure of HOOOH has not been determined by direct methods yet, we have to rely on the results of theoretical studies. A number of ab initio studies on the monomeric dihydrogen triox- ide were published.3137 Ali of them indicate a zig-zag skew-chain structure with C2 symmetry so that both HOOO dihedral angles are equal and opposite. The calculated geometry parameters for the monomeric form of HOOOH are, together with, for comparison, the experimental values for HOOH shown in Figure 2.37'38 0.965 Dihedral angle (HOOH) = 111.8 I5l 0.966 Dihedral anale (HOOO) = 81.2 Figure 2. (A) Experimental geometry parameters for HOOH (IR, microwave spectroscopy)(Ref 38). (B) CCSD(T)/6-311++G(3df,3pd) geometrical parameters for HOOOH (Ref. 37). Bond lengths and angles are in Angstroms and degrees, respectively. The dipole and the hyperconjugation effects were found to be predominant factors governing the conformational behavior of HOOOH. The shortening of the O-O in this polyoxide relative to the O-O bond in HOOH was also explained by hyperconjugation arguments.32'36 Theoretical studies indicate that self-association, i.e., dimerization/oligomerization, is most likely the characteristic structural feature of HOOOH. Our pre-liminary studies indicate that various dimeric27'39 and trimeric (as well as tetrameric) structures of HOOOH (Figure 3) are plausible in “inert”, i.e., non-basic solvents.39 However, basic solvents (B = oxygen base) are able to complex with such aggregates, and, if basic enough, disrupt the hydrogen bonds in these aggregates to form 1:1 and/or 1:2 hydrogen-bonded complexes with HOOOH.17bl8b For an example, see Scheme 5. Plesničar Progress in the Chemistry of Dihydrogen Trioxide 6 Acta Chim. Slov. 2005, 52, 1-12 Figure 3A Figure 3B (3a) (3b) Figure 3C Figure 3. (A) The dihydrogen trioxide dimer 2a: equipotential surfaces. (B) The two most stable structures (among several distinct enti-ties) of HOOOH dimers (2a, BE = 9.2 kcal/mol, per H-bond, 4.6 kcal/mol; 2b, BE = 9.0 kcal/mol, per H-bond, 4.5 kcal/mol). (C) The two most stable structures of HOOOH trimers (3a, BE = 21.1 kcal/mol, per H-bond, 5.2 kcal/mol; 3b, BE = 21.5 kcal/mol, per H-bond, 5.3 kcal/mol; B3LYP/6-31G(d,p), BSSE corrected). Plesničar Progress in the Chemistry of Dihydrogen Trioxide Acta Chim. Slov. 2005, 52, 1–12 7 o—o h—o y H 6—H O — O H-O—O— O-H.....(5) ® (5).....H-O—O—O-H.....(S) Scheme 5 Because water and hydrogen peroxide were always present in solutions of HOOOH (prepared by either of the methods mentioned above), theoretical studies on HOOOH-HOH (BE = 8.2 kcal/mol; B3LYP/6-31G(d,p), BSSE corrected) and HOOOH-HOOH (BE = 8.0 kcal/mol) as well as HOOOH-HOH-HOOH complexes were also of interest.17'39 As will be discussed later, these structures have a considerable effect on the decomposition of HOOOH. It is interesting to mention that, the dihydrogen trioxide-water dimerization energy is 0.8 kcal/mol (B3LYP/6-31G(d,p), BSSE corrected) smaller than that for the dihydrogen trioxide dimer, suggesting that the self-association of HOOOH is slightly favored with respect to its solvation by one molecule of water. This, together with a relatively good solubility of HOOOH in less polar solvents, might indicate that HOOOH is more lipophilic than water (and most likely more lipophilic than HOOH as well). There is at present no direct evidence for the various self-associated HOOOH entities and for intermolecularly hydrogen-bonded HOOOH-oxygen base complexes. However, a study of the temperature dependence of the OOOH absorption of HOOOH in acetone-Je, methyl acetate, and terf-butyl methyl ether revealed a small but definitive upfield shift of the OOOH absorption that was observed with increasing temperature1813 (see Table 2). This corresponds to the appearance of larger clusters of HOOOH both at low temperatures and at higher HOOOH concentrations. Interchange betvveen aH these forms, which contributes to narrow time-averaged features, must be fast since no exchange broadening of the OOOH absorption was observed even at the lowest temperature investigated (-100 °C, dimethyl ether). The observed OOOH chemi-cal shift thus reflects both the hydrogen-bond strength and their extent in these entities.18b Acidity. The proton affinity of HOOOH (AH(298) = 365.7 kcal/mol) clearly indicates that this polyox-ide is more acidic than HOOH (AH(298) = 376.6 kcal/mol) and HOH (AH(298) = 391.3 kcal/mol).23'37 A peculiar structure of the HOOO anion is worth mentioning here. Namely, according to CCSD(T)/6-311++G(3df,3pd) calculations, the singlet state of the HOOO anion possesses a strong HOO-O bond with some double bond character, while a weakly covalent Table 2. Proton chemical shifts, rate constants, frequency factors, and activation energies for the decomposition of the dihydrogen trioxide (HOOOH) formed in the ozonation of l,2-diphenylhy-drazine in various solvents." solvent T(°C) 8 (ppm) W4k(T),^ Ea (kcalmor1) log A acetone-<4 -20 13.11 0.1 -10 13.02 0.25 0 12.93 1.4 10 12.84 2.5 20 12.75 6.8 30 12.67 17.4 methyl -10 12.47 0.4 acetate 0 12.38 1.43 10 12.29 4.3 15 12.25 7.2 20 12.21 12.4 30 12.14 29.0 fert-butyl -10 12.62 0.38 methyl 0 12.55 0.95 ether 10 12.50 2.4 20 12.41 5.7 25 12.39 11.4 30 12.36 17.0 16.7± 1.5 9.4 ±1.0 16.5 ±1.5 9.5 ±1.1 15.3 ±1.3 8.3 ±0.9 " c(HOOOH) = 0.002-0.01 M. Standard deviations < ±10%. By following the decay of the HOOOH XH NMR absorption. HO-OO bond is extraordinarily long (1.80 A, i.e., the longest O-O bond ever found for a peroxide). In aque-ous solution, the HOOO" anion adopts a geometry that closefy resembles that of HOOOH (HOO-O = 1.388 A, HO-OO = 1.509 A,x(HOOO) = 78.3°), and can thus be regarded as the conjugate base of HOOOH. The other metastable form, i.e., the triplet form of the HOOO" anion (10.9 kcal/mol more stable than the singlet), was actually found to be a weakly bounded (AH(298) = 2.1 kcal/mol) van der Waals complex betvveen the HO anion and triplet oxygen.37 It is quite likely that solvation will somewhat reduce the gas-phase acidity differences among these species. Namely, the HOO anion, and particularly the HO anion, possess a more localized negative charge, and can thus be better solvated by water molecules than the HOOO anion. The estimated pK, value for HOOOH is 9.5 + 0.2 in water at 25 °C (pIC, (HOOH) = 11.6 in water at 20 °C).n'37 The calculated bond dissociation energies (BDE) for HOOOH are: BDE(HOO-OH) = 33.9 kcal/mol and BDE(HOOO-H) = 81.3 kcal/mol.36 The enthalpy of formation of HOOOH, AH(° (298), is -22.7 kcal/mol37 (-24 + 3 kcal/mol).22l'22k B Plesničar Progress in the Chemistry of Dihydrogen Trioxide 8 Acta Chim. Slov. 2005, 52, 1–12 Decomposition of Dihydrogen Trioxide The decomposition of HOOOH has been studied in the last decade, both by experiment and theory.1W8 Several experimental facts emerged. Firstly, the kinetics of decomposition of HOOOH, measured by following the decay of the OOOH JH NMR absorption, was found to obey cleanly first-order kinetics over at least 3-4 half-lives in acetone-Jg, methyl acetate, and terf-butyl methyl ether (Table 2). Secondh/, singlet oxygen (A^j) and water were the main decomposition products found. The studies are underway to determine unambiguously, whether hydrogen peroxide and ozone are also formed in these reactions. Several mechanisms for the decomposition of HOOOH were proposed to accommodate the experi-mental data. Nangia and Benson proposed a radical chain mechanism for the decomposition of HOOOH (Scheme 6).40 The activation energy of this chain reac-tion mechanism would be lower than the BDE of the HOO-OH bond (33.9 kcal/mol).36 However, the initially formed HOOO" radical would break down to the HO" radical and triplet oxygen. This, together with relatively low experimental log A values for the decomposition of HOOOH,17a18 seem not to support this mechanistic proposal. HOOOH -=^ HOO* + 'OH HO" + HOOOH — [HOH + •OOOH] — H20 + HO' + 02 HOO' + HOOOH — [HOOH + "OOOH] — HOOH + HO' + 02 measured Arrhenius energies (16.5 + 1.5 kcal/mol)17a (see Table 2). The half-life of HOOOH at room temperature (20 °C) is 16 + 2 min in aH solvents investigated. This is in contrast with the half-life of HOOOH in water, which was estimated to be just a fraction of a second (t1/2« 20 ms). Xu, Muller, and Goddard investigated stili some other plausible decomposition pathways of HOOOH.41 In one of them (Scheme 8), the decomposition is initiated in the dimer of cis-HOOOH (each of the HOOOH molecules adopts the 2.4 kcal/mol less stable Cs-symmetrical cis conformation) to form 102 and two HOOH molecules via intermediate 03 and dihydrogen tetraoxide (H204) complexes. The activation enthalpy of the rate-determining step of the process was calculated to be 19.7 kcal/mol. (a) H „,..o.,„, t O" »o / H AHa(298) = 18 8 kcal/mol 9^\ /^ H 6—H......O h' H203 + 2H20 H I ,o- -h-o \ H 02(1Ag) + 3 H20 (AHR(298) = -9.2 kcal/mol) (b) H Ow'""0"""""0 AHa(298) = / \ 17 7 kcal/mol H *'H H H 6—H......O 4 H H-O H O.....H—6 ^O \ H Scheme 6 H203 + H20 +H202 * 02(1Ag) + 2 H20 + H202 The activation energy for the intramolecular proton-transfer was calculated to be far too high (« 49 kcal/mol)23a for the reaction to proceed on the singlet surface. Therefore, stili some other mechanistic pos-sibilities, including those involving water and hydrogen peroxide were considered. Namely, as stated before, water (and hydrogen peroxide as well) were always present in solutions of HOOOH prepared by either of the methods mentioned in sufficient amounts to com-plex aH the HOOOH present. Water was found to be labeled with "O (ozonation with 1703) indicating that, at least in part, it was formed by the decomposition of HOOOH during the ozonation procedure.17a The kinetic and activation parameters for the decomposition of HOOOH are in accord with the mechanism depicted in Scheme 7 involving (a) two molecules of water, acting as a bifunctional catalyst or (b) hydrogen peroxide and one molecule of water. Namely, the calculated activation enthalpies, i.e., 18.8 and 17.7 kcal/mol,17a respectively, are very similar to the P—oy H H H «-o-°-» Scheme 7 AHa(298) = 16 9 kcal/mol AHa(298) = 19 7 kcal/mol -? r ? H /°-H AHa(298) = 7.2 kcal/mol -H........?' /°-H °A^° Cleavage of 0-0 bond: 10.2 kcal/mol O I H... .0-. / -H AHa(298) = 9.7 kcal/mol / .0 -O H203 + H203 °-o-° O + O—H H-0 —*- 02(1Ag) + 2 H202 Scheme 8 H H H H O O H H O Plesničar Progress in the Chemistry of Dihydrogen Trioxide Acta Chim. Slov. 2005, 52, 1-12 9 Alternatively, dihydrogen trioxide can react with singlet oxygen to form ozone and hydrogen peroxide (Scheme 9). The activation energy of this process is 12.0 kcal/mol.41 .0=0, AHa(298) = O—O,' AHa(298) = O----O, 8.3 kcal/mol^ H' \ 12.0 kcal/mol_ H» \ K0^ <0J ^0J H203 + 02(1Ag) ----------¦- 03 + H202 (AHR(298) = 8.6 kcal/mol) Scheme 9 Ali these mechanistic possibilities for the de-composition seem to explain relatively high losses of HOOOH (up to 50%) during the ozonation procedures for making this polyoxide.1617a18 Some oxygen bases (B), for example ethers,17'18 appear to slow down the decomposition of HOOOH. Namely, somewhat smaller rates for the decomposition of HOOOH in various ethers, compared to the values in other solvents (acetone-Je, methyl acetate), may reflect the lower solubility of water in these solvents and consequently the formation of relatively strongly hydrogen-bonded HOOOH-B17b and/or B-HOOOH-B complexes17b18 (as already discussed above), thus preventing the donation of a hydrogen atom (proton) to the other end of the molecule. In such cases, homo-lytic contributions to the decomposition pathways for HOOOH, particularly at higher temperatures, might play a more important role. A proton transfer from dihydrogen trioxide to the HO anion without any barrier was calculated when HO was brought near HOOOH.36 The base-catalyzed reaction of HOOOH + HO —> HOH + HOOO is exothermic by 26.9 kcal/mol.37 The hydrogen-bonded H20-HOOO complex de-composes quickly into singlet oxygen and HO", most likely without activation energy. (The overall reaction HOOOH + HO ->¦ HO + AJ02 + HOH is exothermic by 12.6 kcal/mol; HOOO (singlet) ->¦ AJ02 + HO" is endothermic by 14.3 kcal/mol). Therefore, the decomposition of HOOOH will be very fast in the presence of a very strong base.36'37 This prediction was actually confirmed by experiment. It is also interesting to note that HOOOH decomposes very fast in various organic solvents in the presence of triphenylphosphine-oxide, which apparently is basic enough to abstract the proton from HOOOH.42 The acid-catalyzed decomposition most likeh/ in-volves conversion of HOOOH into the protonated form, HOOOH2+ (protonation by H30+ of one of the terminal oxygen atoms),35'36 with subsequent fragmen-tation of this species to AJ02 and HOH. This process is most likely heavily assisted by clusters of water mol-ecules, which delocalize the positive charge, and thus stabilize the system.36 The details of the acid-catalyzed decomposition of HOOOH are not known at present. It is interesting to mention that, the protonation energy of HOOOH was calculated to be 160.8 kcal/mol, while the predicted binding energy of Li+ to HOOOH was 33 kcal/mol (QCISD/DZ+P level).35 Therefore, it is quite probable that both adducts, i.e., HOOOH2+ and Li+-HOOOH, might be observable by mass spectrometry. In the trioxide CF3OOOCF343 (the first X-ray structure of a trioxide43a), neither the acid- or base-cata-lyzed nor the solvent-assisted decomposition pathways are possible. The radical fragmentation was found to be a predominant mechanistic feature. The BDE of the CF3O-OOCF3 bond is estimated to be 30.1 kcal/mol, while the experimental value was found to be 30.3 kcal/mol.43c'43d Although this value is slightly smaller than the BDE for the HO-OOH bond (33.7 kcal/mol), the perfluoroalkylated trioxide is much more stable. It actually does not decompose until at or above 25 °C thus convincingly demonstrating that, if proton-transfer pathways are blocked, the stability of the trioxides is increased. Reactivity of Dihydrogen Trioxide Very little of a definitive nature is known about the reactivity of HOOOH against various organic substrates. It was reported that solutions of HOOOH reacted with thianthrene 5-oxide (introduced by Adam et al. as a chemical probe for differentiating betvveen the electrophilic versus nucleophilic nature of various oxidants44) as an electrophilic oxidant, i.e., XNu values of 0.17 + 0.01 were obtained in acetone and methyl acetate, respectively. HOOOH was found to be more electrophilic than ozone in these reactions (03, XNu = 0.24, methylene chloride; 0.59, acetone).16 Oxidation of substituted phenyl methyl sulfides with solutions of HOOOH (competition experiments in acetone or methyl acetate at -40 °C) gave Hammett p values (vs a) of -1.90 + 0.02. Only sulfoxides were formed in these reactions. Control experiments showed that HOOH does not react with the sulfides under the conditions investigated.1617 While they are certainly alternative explanations possible for the oxidation of sulfides to sulfoxides with HOOOH (for example, single-electron-transfer processes, and the involvement of HOOO" or HOOO species), we believe that the mechanism which involves the rate-determining attack of the sulfur atom on the middle oxygen atom in HOOOH might be one of them (Scheme 10). The question of Brensted and Lewis acid ca-talysis in oxidations with HOOOH also awaits to be explored. Plesničar Progress in the Chemistry of Dihydrogen Trioxide 10 Acta Chim. Slov. 2005, 52, 1–12 R R S + H-O—O—O-H HO® R2SOOHl HOH + R2SOOS R2S 1 R2S R 2 S=0 / R Scheme 10 Conclusions and Outlook We have demonstrated that HOOOH, character-ized by JH and 170 NMR, and IR spectroscopy, can be prepared by the low-temperature reduction of ozone with various organic reducing agents in oxygen bases (acetone-de, methyl acetate, terf-butyl methyl ether) as solvents. The ozonation of l,2-diphenylhydrazine (1,2-DPH) in acetone-Je is, at least at present, the method of choice for the preparation of relativeh/ highly concentrated solutions of HOOOH (up to 0.1 M) without the interfering presence of organic hydrot-rioxides (ROOOH). Using a variety of DFT methods in connection with the B3LYP functional, a stepwise mechanism involving the hydrotrioxyl radical (HOOO") formed by the abstraction of the N-hydrogen atom by ozone to form a radical pair, (PhNNHPh" "OOOH), was found. The HOOO" radical than abstracts the remain-ing N(H) hydrogen atom from the PhNNHPh" radical to form HOOOH and the diazene, or recombines with RNNHR" to form the hydrotrioxide, RN(OOOH)NHR. The homolytic decomposition of the latter most likely explains the presence of other reaction products, i.e., l,2-diphenyldiazene-N-oxide and hydrogen peroxide. The use of the resin-supported 1,2-DPH, intro-duced recently by Wentworth et al., enables the preparation of HOOOH free of the contamination products arising from the oxidation of 1,2-DPH. Water, always present in solutions of HOOOH after the ozonation procedure, plays a crucial role as a bifunctional catalyst in a “pericyclic” decomposition process of this polyoxide to produce water and singlet oxygen (A^j). The possibility that HOOH is, besides water, also involved in the decomposition of HOOOH, also cannot be ruled out. The half-life of HOOOH in acetone-Jeis 16 + 2 min at 20 °C, while it is much shorter in water («20 ms). The Scripps researchers found that aH antibodies have the ability to catalyze the oxidation of water by singlet oxygen to generate HOOH and ozone. They proposed HOOOH as a key intermediate in these reactions. Ali these species may participate in patho-logical processes in living systems (atherosclerosis, neurodegenerative disorders, cancer). A recent report that HOOOH was found in quantifiable amounts in the thermal reaction of 03 with HOOH (“peroxone process”) for the first time demonstrates that, a link exists betvveen these three oxidants, and that HOOOH might actually be involved in oxidation processes that span atmospheric, environmental, and biological systems. Further studies of reactions of HOOOH with either (a) ozone, hydrogen peroxide, singlet oxygen, superoxide anion, or itself (in dimers and/or oligomers), or (b) reactions of HOOOH with various organic sub-strates (acid- or base-catalyzed), will certainly provide further surprises in this interesting field of research. And finally, the question “How long can you make an oxygen chain?” is stili a very relevant and challenging one. Acknowledgment I gratefully acknowledge the essential contribu-tions to the research reported here from our group of my colleagues Drs. Janez Cerkovnik, Jože Koller, Franci Kovač, and Evgen Eržen. Without their ef-forts this work would not have been done. I especially acknowledge the theoretical contributions of Prof. Dieter Cremer and Prof. Elfi Kraka, and their co-workers Drs. Teli Tuttle and Anan Wu (University of Göteborg, Sweden). The research from our labora-tory has been supported by the Ministry of Educa-tion, Science, and Šport of the Republic of Slovenia. References 1. For revievvs, see: (a) B. Plesničar. In Organic Perooddes; W. Ando, Ed.; Wiley: New York, 1992; pp. 479-533. B. Plesničar. In The Chemistry of Perooddes; S. Patai, Ed.; Wiley: New York, 1983, pp. 483-520. (b) A. de Meijere, F. Wolf. In Organische Peroxoverbindungen; H. Kropf, Ed.; George Thieme Verlag: Stuttgart, 1988; Vol. E13, p. 971-990. (c) V V. Shereshovets, S. L. Khursan, V D. Komissarov, G. A. Tolstikov, Russ. Chem. Rev. 2001, 70, 105-129. (d) A. F. Khalizov, S. F. Khursan, V. V. Shereshovets, Kinet. Catal. 1999, 40, 194-206. 2. M. Berthelot, Compt. Rend. 1880, 90, 656. 3. D. I. Mendeleev, Osnovy Khimii, 6th ed., 1895. 4. V. Rothmund, A. Burgstaller, Monatsh. Chem. 1917, 38, 295. o o Plesničar Progress in the Chemistry of Dihydrogen Trioxide Acta Chim. Slov. 2005, 52, 1–12 11 5. E. Ohara,/. Chem. Soc. Jpn. 1940, 61, 569. 6. H. Taube, W. C. Bray, /. Am. Chem. Soc. 1940, 62, 3357-3373. 7. W. C. Bray, /. ^4m. Chem. Soc. 1938, 60, 82-87. 8. For a review of this earlier work, see: M. Venugopalan, R. A. Jones. Chemistry of Dissociated Water Vapor and Related Systems; Wiley: New York, 1968; Chapter 7. 9. G. Czapski, B. H. J. Bielski, /. Phys. Chem. 1963, 67, 2180-2184. 10. B. H. J. Bielski, H. A. Schvvarz,/ Phys. Chem. 1968, 72, 3836-3841. 11. B. H. J. Bielski,/. Phys. Chem. 1970, 74, 3213-3216. 12. (a) T. V. Yagodovskaya, D. Y. Zhogin, L. I. Nekrasov, Zh. Fiz. Khim. 1976, 50, 2736-2739, and references therein. (b) T. V Yagodovskaya, L. I. Nekrasov, Zh. Fiz. Khim. 1977, 51, 2434-2445. (c) T. V. Yagodovskaya, M. R. Khadzhiogly, L. I. Nekrasov, Zh. Fiz. Khim. 1986, 60, 922-926. 13. (a) P. A. Giguere, K Herman, Can. J. Chem. 1970, 48, 3473-3482. (b) X. Deglise, P. A. Giguere, Can. J. Chem. 1971, 49, 2242-2247. (c) J. L. Arnau, P. A. Giguere, /. Chem. Phys. 1974, 60, 270-273. (d) P. A. Giguere, T. K. K. Srinivasan, Chem. Phys. Lett. 1975, 33, 479-482. 14. A. Engdahl, B. Nelander, Science 2002, 295, 482-483. 15. X. Xu, W. A. Goddard III, Proč. Natl. Acad. Sci. U.S.A. 2002, 99, 15308-15312. 16. J. Cerkovnik, B. Plesničar,/. Am. Chem. Soc. 1993, 115, 12169-12170. 17. (a) B. Plesničar, T. Tuttle, J. Cerkovnik, D. Cremer, /. Am. Chem. Soc. 2003, 125, 11553-11564. (b) A. Wu, D. Cremer, J. Gauss, /. Phys. Chem. A 2003, 107, 8737-8745. 18. (a) B. Plesničar, J. Cerkovnik, T. Tekavec, J. Koller, /. Am. Chem. Soc. 1998, 120, 8005-8006. (b) B. Plesničar, J. Cerkovnik, T. Tekavec, J. Koller, Chem.-Eur. J. 2000, 6, 809-819. (c) For the formation of HOOOH in the ozonation of 2-norcaranol, see: E. Eržen, J. Cerkovnik, B. Plesničar,/. Org. Chem. 2003, 68, 9129-9131. 19. J. Cerkovnik, E. Eržen, J. Koller, B. Plesničar, / Am. Chem. Soc. 2002, 124, 404—409. 20. A. Wu, D. Cremer, B. Plesničar,/ Am. Chem. Soc. 2003, 125, 9395-9402. 21. (a) B. Plesničar, J. Cerkovnik, T. Tuttle, E. Kraka, D. Cremer, / Am. Chem. Soc. 2002, 124, 11260-11261. (b) T. Tuttle, J. Cerkovnik, B. Plesničar, D. Cremer, / Am. Chem. Soc. 2004, 126, 16093-16104. 22. For theoretical studies on HOOO*, see: (a) R.J. Blint, M. D. Newton, /. Chem. Phys. 1973, 59, 6220-6228. (b) K. B. Mathisen, P. E. M. Siegbahn, Chem. Phys. 1984, 90, 225-230. (c) M. Dupuis, G. Fitzgerald, B. Hammond, W. A. Lester, H. F. Schaefer, III. / Chem. Phys. 1986, 84, 2691-2697. (d) M. A. Vmcent, I. H. Hillier, / Phys. Chem. 1995, 99, 3109-3113. (e) M. A. Vmcent, I. H. Hillier, N. A. Burton, Chem. Phys. Lett. 1995, 233, 111-114. (f) M. A. Vincent, N. A. Burton, I. H. Hillier, Mol. Phys. 1996, 87, 945-959. (g) T. P. W. Jungkamp, J. H. Seinfeld, Chem. Phys. Lett. 1996, 257, 15-22. (h) M. Speranza, Inorg. Chem. 1996, 35, 6140-6151. (i) M. Speranza,/ Phys. Chem. A 1998, 102, 7535-7536. (j) H. G. Yu, A. J. C. Varandas, Chem. Phys. Lett. 2001, 334, 173-178. (k) P. A. Denis, M. Kieninger, O. N. Ventura, R. E. Cachau, G. H. F. Diercksen, Chem. Phys. Lett. 2002, 365, 440^-49. Chem. Phys. Lett. 2003, 377, 483^-84. (1) S. Aloisio, J. S. Francisco,/ Am. Chem. Soc. 1999, 121, 8592-8596. (m) For experimental detection, see: F. Cacace, G. de Petris, F. Pepi, A. Troiani, Science, 1999, 285, 81-82. B. Nelander, A. Engdahl, T. Svensson, Chem. Phys. Lett. 2000, 332, 403-408. 23. For theoretical studies on HOOCT, see: (a) J. Koller, B. Plesničar, / Am. Chem. Soc. 1996, 118, 2470-2472. (b) E. Kraka, D. Cremer, J. Koller, B. Plesničar, / Am. Chem. Soc. 2002, 124, 8462-8470. (c) P. S. Nangia, S. W. Benson, / Am. Chem. Soc. 1980, 102, 3105-3115. (d) B. Elliott, A. N. Alexandrova, A. I. Boldvrev, / Phys. Chem. A 2003, 107, 1203-1206. (e) For an experimental detection, see: F. Cacace, R. Cipollini, G. de Petris, A. Troiani, Int. J. Mass Spectrom. 2003, 228, 717-722. 24. N. Lendero, J. Cerkovnik, B. Plesničar, unpublished results. 25. P. T. Nvffeler, N. A. Boyle, L. Eltepu, C.-H. Wong, A. Eschenmoser, R. A. Lerner, P. Wentworth, Jr., Angew. Chem. Int. Ed. 2004, 43, 4656-4659. 26. J. Cerkovnik, B. Plesničar, unpublished results. 27. B. Plesničar, J. Cerkovnik, J. Koller, F. Kovač, / Am. Chem. Soc. 1991, 113, 4946^-953. 28. (a) C.-H. Kuo, L. Zhong, M. E. Zappi, A. P. Hong, Can. J. Chem. Eng. 1999, 77, 473^-82. (b) J. Prousek, Chem. Listy 1996, 90, 229-237. (c) R. J. Spanggord, D. Yao, T. Mili, Environ. Sci. Technol. 2000, 34, 450-454. (d) M. S. Elovitz, U. von Gunten, Ozone Sci. Eng. 1999, 21, 239-260. (e) M. S. Elovitz, U. von Gunten, H. P. Kaiser, Ozone. Sci. Eng. 2000, 22, 123-150. (f) J. L. Acero, U. von Gunten, Ozone Sci. Eng. 2000, 22, 305-328. 29. (a) P. Wentworth, Jr., L. H. Jones, A. D. Wentworth, X. Y. Zhu, N. A. Larsen, I. A. Wilson, X. Xu, W. A. Goddard III, K. D. Janda, A. Eschenmoser, R. A. Lerner. Science, 2001, 293, 1806-1811. (b) X. Y. Zhu, P. Wentworth, Jr., A. D. Wentworth, A. Eschenmoser, R. A. Lerner, I. A. Wilson. Proč. Natl. Acad. Sci. U.SA. 2004, 101, 2247-2252. (c ) D. Datta, N. Vaidehi, X. Xu, W. A. Goddard III. Proč. Natl. Acad. Sci. U.SA. 2002, 99, 2636-2641. (d) X. Xu, W. A. Goddard III. Proč. Natl. Acad. Sci. U.SA. 2002, 99, 15308-15312. (e) P. Wentworth, Jr., A. D. Wentworth, X. Y. Zhu, I. A. Wilson, K D. Janda, A. Eschenmoser, R. A. Lerner. Proč. Natl. Acad. Sci. U.SA. 2003, 100, 1490-1493. (f) P. Wentworth, Jr., J. E. McDunn, A. D. Wentworth, C. Takeuchi, J. Nieva, T. Jones, C. Bautista, J. M. Ruedi, A. Plesničar Progress in the Chemistry of Dihydrogen Trioxide 12 Acta Chim. Slov. 2005, 52, 1–12 Gutierrez, K. D. Janda, B. M. Babior, A. Eschenmoser, 35. R. A. Lerner, Science 2002, 298, 2195-2199. (g) B. M. Babior, C. Takeuchi, J. Ruedi, A. Gutirrez, P. Wentworth, 36. Jr. Proč. Natl. Acad. Sci. U.S.A. 2003, 100, 3031-3034. (h) P. Wentworth, Jr., J. Nieva, C. Takeuchi, R. Galve, 37. A. D. Wentworth, R. B. Dilley, G. A. DeLaria, A. Saven, B. M. Babior, K. D. Janda, A. Eschenmoser, R. A. 38. Lerner, Science 2003, 302, 1053-1056. (i) R. A. Lerner, A. Eschenmoser, Proč. Natl. Acad. Sci. U.S.A. 2003, 100, 3013-3015. (j) Q. H. Zhang, E. T. Powers, J. Nieva, M. 39. E. Huff, M. A. Dendle, J. Bieschke, C. G. Glabe, A. 40. Eschenmoser, P. Wentworth, Jr., R. A. Lerner, J. W. Kelly, Proč. Natl. Acad. Sci. U.S.A. 2004, 101, 4752-4757. 41. (k) E. Keinan, A. Alt, G. Amir, L. Bentur, H. Bibi, D. Shosevov, Bioorg. Med. Chem. 2005, 13, 557-562. (1) For 42. an entertaining yet thought-provoking “The Story of O”, 43. see: R. Hoffmann, American Scientist 2004, 92, 23-27. 30. T. M. Lesko, A. Colussi, M. R. Hoffmann, /. Am. Chem. Soc. 2004, 126, 4432-4436. 31. B. Plesničar, S. Kaiser, A. Ažman,/. Am. Chem. Soc. 1973, 95, 5476-5477. 32. D. Cremer,/. Chem. Phys. 1978, 69, 4456-4471. 33. (a) C. F. Jackels, D. H. Phillips, /. Chem. Phys. 1986, 84, 5013-5024. (b) C. F. Jackels, /. Chem. Phys. 1993, 99, 44. 5768-5779. 34. C. Gonzalez, J. Theisen, L. Zhu, H. B. Schlegel, W. L. Hase, E. W. Kaiser, /. Phys. Chem. 1991, 95, 6784-6792. C. Gonzalez, J. Theisen, H. B. Schlegel, W. L. Hase, E. W. Kaiser, /. Phys. Chem. 1992, 96, 1767-1774. Povzetek V zadnjih petnajstih letih je bil narejen zelo pomemben napredek pri poznavanju kemije divodikovega trioksida (HOOOH). Ta pregledni članek opisuje pripravo, karakterizacijo in strukturo (XH, 170 NMR, IR, kvantno- kemijske metode), kinetiko in mehanizem razpada ter reaktivnost tega najenostavnejšega polioksida. Divodikov trioksid je v organskih kisikovih bazah kot topilih precej bolj stabilen kot so menili do sedaj. Tako je, naprimer, razpolovni čas HOOOH v acetonu-^ 16 + 2 minut (20 °C), medtem ko v vodi razpade zelo hitro (tm « 20 ms). Divodikov trioksid najbolj verjetno sodeluje kot reaktivni intermediat pri oksidacijskih procesih v atmosferi, okolju in v bioloških sistemih. T. Fujii, M. Yashiro, H. Tokiwa, /. Am. Chem. Soc. 1997, 119, 12280-12284. D. J. Mckay, J. S. Wright, /. Am. Chem. Soc. 1998, 120, 1003-1013. E. Kraka, D. Cremer, J. Koller, B. Plesničar,/. Am. Chem. Soc. 2002, 124, 8462-8470. (a) J. Koput, /. Mol. Spectrosc. 1986, 115, 438-441. (b) R. H. Hunt, R. A. Leacock, C. W. Peters, K. T. Hecht, /. Chem. Phys. 1965, 42, 1931-1946. S. Prebil, J. Koller, B. Plesničar, unpublished results. P. S. Nangia, S. W. Benson, /. Phys. Chem. 1979, 83, 1138-1142. X. Xu, R. P. Muller, W. A. Goddard III, Proč. Natl. Acad. Sci. U.SA. 2002, 99, 3376-3381. J. Cerkovnik, B. Plesničar, unpublished results. (a) K. I. Gobbato, M. F. Klapdor, D. Mootz, W. Poli, S. E. Ulic, H. Willner, H. Oberhammer,^4wgevu Chem. Int. Ed. 1995, 34, 2244—2245. (b) L. R. Anderson, W. B. Fox,/. Am. Chem. Soc. 1967, 89, 4313-4315. (c) P. G. Thompson,/. Am. Chem. Soc. 1967, 89, 4316^-319. (d) J. Czarnovvski, H. J. Schumacher,/wf. / Chem. Kinet. 1981, 13, 639-649. (e) J. S. Francisco, I. H. Williams,/wf. / Chem. Kinet. 1988, 20, 455^-66. (a) W. Adam, W. Haas, B. B. Lohray, / Am. Chem. Soc. 1991, 113, 6202-6208. W. Adam, M. N. Korb, K. J. Roschmann, C. R. Saha-Möller, / Org. Chem. 1998, 63, 3423-3428. (b) E. Eržen, J. Koller, B. Plesničar, / Org. Chem. 2001, 66, 5155-5162. Plesničar Progress in the Chemistry of Dihydrogen Trioxide