Kazuo Takahashi

The reactions of O(P-3) atoms with furan, 2-methylfuran (2-MF), and 2,5-dimethylfuran (2,5-DMF) were studied at elevated temperatures by using a shock tube technique coupled with atomic resonance absorption spectroscopy (ARAS). The rate coefficients (k) determined from the time profiles of O(P-3) atoms were expressed by the relation of k(O+2,5-DMF) > k(O+2-MF) > k(O+FURAN). The molecular orbital calculations with CBS-QB3 level were also performed and the possible product channels of these reactions were discussed.

The thermal decomposition reactions of 2,3,3,3- and trans-1,3,3,3-tetrafluoropropenes (TFPs) have been studied both experimentally and computationally to elucidate their kinetics and mechanism. The experiments were performed by observing the temporal profiles of HF produced in the decomposition of the tetrafluoropropenes behind shock waves at temperatures of 15401952 K (for 2,3,3,3-TFP) or 1525-1823 K (for trans-1,3,3,3-TFP) and pressure of 100-200 kPa in Ar bath. The reaction pathways responsible for the profiles were explored based on quantum chemical calculations. The decomposition of 2,3,3,3-TFP was predicted to proceed predominantly via direct 1,2-HF elimination to yield CHCCF3, while trans-1,3,3,3-TFP was found to decompose to HF and a variety of isomeric C3HF3 products including CHCCF3, CF2CCHF, CCHCF3, and CF2CHCF. The C3HF3 isomers can subsequently decompose to either CF2, + CHCF or CF2CC + HF products. Multichannel RRKM/master equation calculations were performed for the identified decomposition channels. The observed formation rates and apparent yields of HF are shown to be consistent with the computational predictions.

A fast and sensitive broadband absorption technique for measurements of high-temperature chemical kinetics and spectroscopy has been developed by applying broadband cavity-enhanced absorption spectroscopy (BBCEAS) in a shock tube. The developed method has effective absorption path lengths of 60-200 cm, or cavity enhancement factors of 12-40, over a wavelength range of 280-420 nm, and is capable of simultaneously recording absorption time profiles over an similar to 32 nm spectral bandpass in a single experiment with temporal and spectral resolutions of 5 its and 2 nm, respectively. The accuracy of the kinetic and spectroscopic measurements was examined by investigating high-temperature reactions and absorption spectra of formaldehyde behind reflected shock waves using 1,3,5-trioxane as a precursor. The rate constants obtained for the thermal decomposition reactions of 1,3,5-trioxane (to three formaldehyde molecules) and formaldehyde (to HCO + H) agreed well with the literature data. High-temperature absorption cross sections of formaldehyde between 280 and 410 nm have been determined at the post-reflected-shock temperatures of 955, 1265, and 1708 K. The results demonstrate the applicability of the BBCEAS technique to time-and wavelength-resolved sensitive absorption measurements at high temperatures.

During the past decade, we have examined the surface improvement of PVF, FEP, PFA, and PTFE films by APGD system. Their adhesive strength with an epoxy resin could be improved that attained by different kind gas introduction to the plasma system. Last year, new idea using NH4/H2O as the treating gas was proposed in the low pressure glow discharge (LPD) to treat PTFE by T. Yajima et al. So we examined the surface effects in both process, APGD and LPD, using NH4/H2O mixture. APGD treatment could not have any chemical effect on the PTFE surface. On the other hand, LPD, we got the proof of super hydrophilicity on the PTFE. It seemed that in the low pressure, a slow sputtering will be a preceding reaction on the PTFE even used the NH4/H2O. The difference of the kinetics was considered in the evaluation of both processes, APGD and LPD.

B2H6/He plasma and B2H6/H-2/He plasma treatments of PTFE were performed. Concentration of the B2H6 was varied through this experiment. Then defluorination degree and adhesive strength were measured. While B2H6 and hydrogen were effective for defluorination of PTFE surface, oxygen functional groups generated by post-oxidation improved adhesive strength. The atmospheric pressure glow plasma treatment with lower diborane concentration more modifies PTFE surface.

Polytetrafluoroethylene (PTFE) sheet surface was treated by the diborane/H-2/He plasma for the deflurination to improve the adhesive strength. Diborane was generated by the H-2/He plasma treatment of a boron plate and was transported to the plasma zone for PTFE sheet treatment. And then, the defluorination plasma treatment with this diborane/H-2/He mixture gas performed PTFE sheet. Fluorine atom content of treated PTFE surface was decreased to about twentieth of that of untreated PTFE. The adhesive strength between treated PTFE and epoxy glue became stronger enough for practical use.

Polytetrafluoroethylene (PTFE) sheet surface was treated by the diborane/H-2/He plasma for the deflurination to improve the painting performance. Diborane was generated by the H-2/He plasma treatment of a boron plate. And then, the defluorination plasma treatment with this diborane/H2/He mixture gas performed to PTFE sheet. Fluorine atom content of treated PTFE surface was decreased to about one hundredth of that of untreated PTFE. The water contact angle of treated PTFE was also decreased from 120 degrees to 50 degrees. A paint coated on treated PTFE adhered to its surface strongly.

The thermal decompositions of 1,1,1-trifluoroethane (CH3CF3) and pentafluoroethane (CHF2CF3) were studied by using a turbulent flow reactor at atmospheric pressure over the temperature ranges of 1213-1333 K and 1273-1373 K. The rate coefficients for these thermal decompositions were determined from the first-order decays of the reactants to be k(CH3CF3) = 10(12.9 +/- 1.2) exp[-d276 +/- 29) kJ center dot mol(-1)/RT] and k(CHF2CF3) = 10(13. 9 +/- 1.4) exp [-(324 +/- 36) kJ center dot mol(-1)/RT]s(-1). To examine the reaction paths, we identified the decomposition products using gas chromatography-mass spectrometry (GC-MS) and performed ab initio MO calculations. These results showed that the sequential HF elimination reactions, CH3CF -> CH2CF2 + HF and CH2CF2 -> CHCF + HF, were favorable for the CH3CF3 pyrolysis. On the other hand, four initial steps: CHF2CF3 -> CF3CF + HF, CHF2CF3 -> CF2CF2+ HF, CHF2CF3 -> CHF3 + CF2 and CHF2CF3 -> CHF2 + CF3, were possible for the CHF2CF3 pyrolysis. Such a difference in the reaction paths between the CH3CF3, and CHF2CF3 pyrolyses can be explained by fluorine hyperconjugation and by repulsion between the fluorine atoms on the 1- and 2-carbons of each hydrofluoroethane.

Toluene decomposition is examined using a corona discharge, a UV lamp, and a TiO2 catalyst complex system. The toluene decomposition reaction occurs only with OH and O (D-1) radical. High decomposition rations are obtained: 98% for the toluene/H2O/O-2-corona/UV system and 70% for the toluene/H2O/air-corona/UV systems. Moreover, if the TiO2 catalyst is combined with the corona-UV system for 1 000 ppm of toluene, this value increases to 78% (0.31 Wh.m(-3).ppm(-1)), which is higher than that obtained in a normal corona discharge system.

The reactions of dimethyl ether (CH3OCH3, DME) with O(P-3) and H atoms have been studied at high temperatures by using a shock tube apparatus coupled with atomic resonance absorption spectroscopy (ARAS). The rate coefficients for the reactions CH3OCH3 + O(P-3) -> CH3OCH2 +OH (1) and CH3OCH3 + H -> CH3OCH2 + H-2 (2) were experimentally determined from the decay of O(P-3) and H atoms as: k(1) = 10(-9.33) (+/- 0.35) exp[-(31 +/- 7) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (910-1210K) k(2) = 10(-9.15) (+/- 0.48) exp[-(37 +/- 10) kJ mot(-1)/RT] cm(3) molecule(-1) s(-1) (1030-1210K) These results show that DME can react with O(P-3) atoms more easily than with H atoms. By combining these results with the previous lower temperature data, we obtained the following modified Arrhenius expressions applied over the wide temperature range between 300 and 1500 K: k(1) = 10(-16.35)T(2.0) exp[-11.0 kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) k(2) = 10(-22.74)T(4.0) exp[-7.7 kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) Both reactions of DME are faster than those of ethane, because the dissociation energy of the C-H bond in DME is smaller. Furthermore, the rate coefficients for reactions (1) and (2) were calculated with the transition-state theory (TST). Structural parameters and vibrational frequencies of the reactants and the transition states required for the TST calculation were obtained from the MP2(full)/6-31G(d) ab initio molecular orbital (MO) calculation. The energy barrier, E-0(double dagger), was adjusted until the TST rate coefficient most closely matched the observed one. The fitting results of E-0(double dagger)(1) = 23 kJ mol(-1) and E-0(double dagger)(2) = 34 kJ mol(-1) were in agreement with the G2 energy barriers, within the expected uncertainty, demonstrating that the experimentally determined rate coefficients were theoretically valid. (c) 2006 Wiley Periodicals.

High-temperature kinetics of several important reactions for flame suppression by 2H-heptafluoropropane (CF3CHFCF3, HFC-227ea) have been studied using a shock tube-atomic resonance absorption spectroscopy technique. The rate coefficients for the initial reactions CF3CHFCF3+H -> CF3CFCF3+H-2 (1a) and CF3CHFCF3 + O(P-3)-> CF3CFCF3+OH (2a) were determined from the decays of H and 0(3 P) atoms to be k(1a)=10(-9.15 +/- 0.66) exp[-(63 +/- 14) kJ(.)mol(-1) /RT] cm(3) molecule(-1)s(-1) (1000-1180 K) and k(2a),= 10(-10.27 +/- 0.67) exp[-(56 +/- 13) kJ(.)mol(-1)/RT] cm(3) molecule(-1)s(-1) (880-1180 K), respectively. The TST calculation showed that these experimental results were theoretically reasonable. Subsequently, the reactions of CF3CFCF3 radicals, which are the product of reactions (1a) and (2a), were studied. The rate coefficients for the reactions CF3CFCF3+H -> CF3CFCF2+HF (3a) and CF3CFCF3 +0 O(P-3)-> CF3CFO+CF3 (4a) had no temperature dependence and were obtained as k(3a)= 10(-9.98 +/- 0.83) cm(3) molecule(-1)s(-1) (1010-1050 K) and k(4a)= 10(-10.84 +/- 0.84) cm(3) molecule(-1)s(-1) (930-1050 K). These results mean that H atoms have higher reactivity with CF3CFCF3 than 0(P-3) atoms, similarly to the kinetic results for the reactions of other fluoroalkyl radicals such as CF3. At temperatures higher than 1050 K, the thermal decomposition of CF3CFCF3 radicals was found to be dominant over reactions (3a) and (4a). So the rate coefficient for the reaction CF3CFCF3 -> 4CF(3)CF+CF3 (5a) was also measured to be k(5a)= 10(10.85 +/- 0.91) exp[-(163 +/- 20) kJ(.)mol(-1)/RT] s(-1) (1050-1300 K).

The reactions of 2H-heptafluoropropane (CF3CHFCF3, HFC-227ea) with O(P-3) and H atoms have been studied at high temperatures by using a shock tube technique coupled with atomic resonance absorption spectroscopy. Electronically ground-state oxygen and hydrogen atoms were produced by the laser photolysis of sulfur dioxide and the thermal decomposition of ethyl iodide, respectively. The rate coefficients for the reactions CF3CHFCF3 + O(P-3) --> i-C3F7 + OH (1a) and CF3CHFCF3 + H --> i-C3F7 + H-2 (2a) were experimentally determined from the decay of O(P-3) and H atoms as k(1a) = 10(-10.27+/-0.67) exp[-(56 +/- 13) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (880-1180 K) and k(2a) = 10(-9.15+/-0.66) exp[-(63 +/- 14) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (1000-1180) K). These results showed that reaction 2a was faster than reaction 1a by a factor of 7-8 over the present experimental temperature range. Both rate coefficients were much smaller than the previous kinetic data for the reactions of propane with O(P-3) and H atoms, because of an electron-attracting effect of fluorine atoms. To compare the reactivities between isomers, the rate coefficients for the reactions of 1H-heptafluoropropane, CHF2CF2CF3 + O(P-3) --> n-C3F7 + OH (3a) and CHF2CF2CF3 + H --> n-C3F7 + H-2 (4a), were also determined by using the same technique as k(3a) = 10(-10.13+/-0.52) exp[-(55 +/- 10) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (880-1180 K) and k(2a) = 10(-9.44+/-0.32+/-) exp[-(57 +/- 7) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (1000- 1180 K). Furthermore, the rate coefficients for reactions 1a and 2a were calculated with the transition state theory (TST). Structural parameters and vibrational frequencies of the reactants and the transition states required for the TST calculation were obtained from the MP2(full)/6-31G(d) ab initio molecular orbital (MO) calculation. The energy barrier, E-0(double dagger), was adjusted until the TST rate coefficient most closely matched the observed one. The fitting results of E-0(double dagger)(1a) = 51 W mol-I and E-0(double dagger)(2a) = 41 kJ mol(-1) were in agreement with the G2(MP2) energy barriers, within the expected uncertainty.

The reactions of methane, ethylene, ethane, and propane with chlorine atoms have been studied at high temperatures by using a laser photolysis-shock tube apparatus coupled with atomic resonance absorption spectroscopy. Chlorine atoms were produced by the laser photolysis of silicon tetrachloride. The overall rate coefficients were experimentally determined from the decay of chlorine atoms as: k(CH4 + Cl) = 10(-9.26+/-0.26)exp[-(35.7 +/- 6.6) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (1100-1550 K) k(C2H4 + Cl) = 10(-9.67+/-0.16)exp[-(35.8 +/- 3.9) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (1100-1550 K) k(C2H6 + Cl) = 10(-9.96+/-0.05) cm(3) molecule(-1) s(-1) (1100-1400 K) k(C3H8 + Cl) = 10(-9.77+/-0.06) cm(3) molecule(-1) s(-1) (1100-1400 K) where the error limits are given at the two standard deviation level. With systematic errors also considered, the reliabilities, Delta[log k(RH + Cl)], were evaluated to be +/-0.2 for reactions CH4 + Cl and C2H4 + Cl and to be +/-0.15 for reactions C2H6 + Cl and C3H8 + Cl. These results are in good agreement with the extrapolations of the kinetic data measured up to 800 K by Pilgrim et al. To examine the product channels of these reactions, an ab initio molecular orbital calculation was performed at the QCISD(T)/6-311G (d,p)//MP2(full)/6-31G(d) level. The calculation shows that the energetically favorable channels of CH4 + Cl, C2H6 + Cl, and C3H8 + Cl are H abstraction by Cl atoms. In the C2H4 + Cl system, the measured rate coefficient has no total density dependence, so that the Cl addition to ethylene is negligible in this experimental temperature range, although the addition channel has no energy barrier. Based on transition state theory, the rate coefficients were also calculated and discussed. The reactions of hydrocarbons with Cl atoms are suggested to follow the Evans-Polanyi law.

The reaction of Sn(P-3(o)) atoms with O-2 has been studied over the temperature range 1300 to 2250 K and the total density range 2.5 x 10(18) to 5.4 x 10(18) molecules cm(-3) by using a shock tube equipped for atomic resonance absorption spectroscopy (ARAS). Tetramethyltin was used as a precursor of Sn(P-3(o)) atoms. The overall rate coefficient was determined from the pseudo-first-order decay of Sn(P-3(o)) atoms to be k(Sn + O-2) = 10(-9.41 +/- 0.03) exp[[(11.5 +/- 1.1) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (error limits at the two standard deviation level). This result is in good agreement with previous data of Fontijn and Bajaj (J. Phys. Chem., 1996, 100, 7085), but about twice that of Zaslonko and Smirnov (Kinet. Catal., 1980, 21, 602). To examine the product channel of the Sn(P-3(o)) + O-2 reaction experimentally, O-atom ARAS measurements were performed. Two types of O-profiles were found, one reaching a peak around 40 mus and the other displaying a two-step increase, depending on the composition of the mixtures. When the reaction channel for forming O(P-3) atoms was assumed, numerical calculation was found to reproduce both these measured profiles, demonstrating that the main products are SnO(X(1)Sigma (+)) and atomic O(P-3).

The kinetics of the thermal decomposition of tin tetrachloride has been studied experimentally and theoretically. Ab initio MO calculations showed that SnCl4 finally decomposed into Sn(P-3) and four chlorine atoms through four subsequent Sn-CI bond dissociation channels. Two sets of kinetic experiments were performed using a shock tube equipped with atomic resonance absorption spectroscopy (ARAS). Chlorine atoms were first measured over the temperature range of 1250-1700 K and the total density range of 1.7 x 10(18) to 8.9 x 10(18) molecules cm(-3). The rate coefficient for the initial reaction step, SnCl4 (+M) --> SnCl3((2)A(1)) + Cl (+M) (eq 4a), was found to be in the falloff region fairly close to the low-pressure limit under the present conditions. The second-order rate coefficient based on the Cl-atom measurements was determined to be k(1a)(2nd) = 10(-5.37+/-0.62) exp[-(285 +/- 18) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (error limits at the 2 standard deviation level). The second group of experiments was carried out by detecting tin atoms over the temperature range of 2250-2950 K and at a total density of 3.2 x 10(18) molecules cm(-3). The second-order rate coefficients for the subsequent reaction steps: SnCl2((1)A(1)) (SM) --> SnCl((2)Pi) + Cl (+M) (eq 3a) and SnCl((2)Pi) (+M) --> Sn(P-3) + Cl (+M) (eq 4a) were obtained to be kg(3a)(2nd) = 10(-8.36+/-0.86) exp[-(310 +/- 42) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) and k(4a)(2nd) = 10(-9.50+/-0.78) exp[-(265 +/- 40) kT mol(-1)/RT] cm(3) molecule(-1) s(-1), respectively. The Rice-Ramspcrger-Kassel-Marcus (RRKM) calculations including variational transition state theory were also applied for reactions la and 3a. Structural parameters and vibrational frequencies of the reactants and transition states required for the RRKM calculations were obtained from the ab initio MO calculations. Energy barriers of the reactions, Eo's, which are the most sensitive parameters in the calculations, were adjusted until the RRKM rate coefficients matched the observed ones. These fittings yielded E-0,E-1a = 326 kJ mol(-1) for reaction la and E-0,E-3a = 368 kJ mol(-1) for reaction 3a, in good agreement with the Sn-Cl bond dissociation energies of SnCl4 and SnCl2, demonstrating that the experimental data for k(1a) and k(3a) were theoretically reasonable and acceptable.

Cl and Si concentration measurements were performed in highly diluted SiHCl3/Ar mixtures. The reflected shock method in combination with sensitive atomic resonance absorption spectroscopy was applied. In the literature, the initial decomposition step of SiHCl3 has been suggested to be SiHCl3 reversible arrow (k1) SiCl2 + HCl (R1) but this has not yet been established. At 1730 K less than or equal to T less than or equal to 3300 K, 0.8 bar less than or equal to p less than or equal to 4.0 bar, and mixtures of 0.5 to 5 ppm SiHCl3 in argon, a strongly temperature-dependent formation of Cl atoms was observed. The signals were evaluated in terms of the rate coefficient k(2): SiCl2 + M reversible arrow (k2) SiCl+ Cl+ M (R2) k(2) = 5.4(-2.3)(+3.8) x 10(15) exp[-(39,300 +/- 1200)K/T] cm(3) mol(-1) s(-1) Si2H6 is known to be a rapid silicon source at T greater than or equal to 1550 K. In highly diluted Si2H6/Ar mixtures, a very fast consumption of the initial Si atom concentrations was measured when a few parts per million SiHCl3 was added. The signals obtained at 1570 K less than or equal to T less than or equal to 2030 K and p = 1.7 bar were kinetically analyzed and assigned to the reaction of Si atoms with SiCl2. Si + SiCl2 reversible arrow (k4) products (R4) k(4) = 5.6 +/- 0.97 +/- 10(14) cm(3) mol(-1) s(-1) All results were verified and discussed based on computer simulations.

The reactions of CS(X(1)A(1)) radicals with O(P-3) and H atoms have been studied by using a shock tube/atomic resonance absorption spectroscopy technique over the temperature ranges 2000-2430 and 1450-1860 K and the total density range 6.1 x 10(18) to 1.2 x 10(19) molecules cm(-3). Nitrous oxide and ethyl iodide were used as precursors of O(P-3) and H atoms, respectively. Electronically ground state CF2(X(1)A(1)) radicals were produced through the thermal decomposition of chlorodifluoromethane. The rate coefficients for the reactions CF(2()X(1)A(1)) + O(P-3) and CF2(X(1)A(1)) + H were obtained from the decay profiles of O and H atom concentrations as k(CF2+O) = 10(-10.39+/-0 07) and k(CF2+H) = 10(-10.18+/-0.21) exp[-(19.0 +/- 6.7) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) terror limits at the two standard deviation level). Neither rate coefficient had any pressure dependence under the present experimental conditions. The G2-level ab initio molecular orbital calculation was also performed to examine the product channels for the CF2(X(1)A(1)) + O(P-3) and CF2(X(1)A(1)) + H reactions. The theoretical calculation showed that the most energetically favorable pathways for CF2(X(1)A(1)) + O(P-3) and CF2(X(1)A(1)) + H systems were the channels producing FCO + F and CF + HF, respectively. The G2 energy of the transition state for the channel CF2(X(1)A(1)) + O(P-3) - FCO + F was 116 kJ mol(-1) lower than that of the reactants CF2(X(1)A(1)) + O(P-3), while the energy of the three-centered transition state for the channel CF2(X(1)A(1)) + H - CF + HF is 45 kJ mol(-1) higher than that of the reactants CF2(X(1)A(1)) + H. These results could qualitatively explain the difference of the temperature dependence observed between k(CF2+O) and k(CF2+H).

The kinetics of the high-temperature reactions of CF3 radicals with O(P-3) and H atoms has been investigated experimentally and theoretically. The product channels of the CF3 + O(P-3) and CF3 + H reactions were examined by calculating their branching fractions with the multichannel Rice-Ramsperger-Kassel-Marcus (RRKM) theory. Structural parameters, vibrational frequencies, and threshold energy required for the RRKM calculation were obtained from an ab initio MO calculation. The theoretical calculation showed that the productions of CF2O + F and CF2((1)A(1)) + HF were the unique possible channels for the CB + O(P-3) and CF3 + H reactions, respectively, and that the other channels such as deactivation were negligible for both the reactions. The rate coefficients for these reactions were experimentally determined by using a shock tube-atomic resonance absorption spectroscopy technique over: the temperature ranges of 1900-2330 and 1150-1380 K and the total density ranges of 8.2 x 10(18)-1.2 x 10(19) and 6.1 x 10(18)-9.8 x 10(18) molecules.cm(-3) Nitrous oxide and ethyl iodide were used as precursors of electronically ground-state oxygen and hydrogen atoms, respectively. Trifluoromethyl radicals were produced through the thermal dissociation of CF3I. The rate coefficients for the reactions CF3 + O(P-3) --> CF2O + F (Ib) and CF3 + H --> CF2((1)A(1)) + HF (2c) were obtained from the decay profiles of O- and H-atom concentrations as k(1b) = (2.55+/-0.23) x 10(-11) and k(2c) = (8.86+/-0.32) x 10(-11) cm(3) molecule(-1) s(-1) (error limits at the one standard deviation level). Neither rate coefficient had any temperature or pressure dependence under the present experimental conditions; the values were in good agreement with some room-temperature data reported previously.

To estimate the inhibition effect of some polyfluorocarbons containing N, O, or S atoms such as (CF3)(3)N, (CF3)(2)NCF2H, (C2F5)(2)O, CF3SF5, and C2F5SF5 on flame propagation, the laminar burning velocity was measured for mixtures of 9.50vol% methane, 90.00vol% air, and 0.50vol% inhibitor and of 1.87vol% n-heptane, 97.63vol% air, and 0.50vol% inhibitor. The five compounds have higher ability as fire suppressants than CF3CHFCF3 and CF3H, which have already been put to practical use.

The reaction of CHF3 (HFC-23) with H atoms has been investigated by using a shock tube-atomic resonance absorption spectroscopy technique over the temperature range 1100-1350 K and the total concentration range 5.5 x 10(18)-8.5 x 10(18) molecules cm(-3). Ethyl iodide was used as a precursor of hydrogen atoms. The rate coefficient for the reaction CHF3 + H --> CF3 + H-2 (1a) was determined from the decay profiles of H-atom concentration to be k(1a) = 10(-9.80+/-0.10) exp[-(64.6 +/- 2.3) kJ mol(-1)/RT] cm(3) molecule(-1) s(-1) (error limits at the 1 standard deviation level), which is 50-50% smaller than the value recommended by the NIST group. The rate coefficient was also calculated with the transition-state theory(TST). Structural parameters and vibrational frequencies of the reactants and transition state required for the TST calculation were obtained from an ab initio molecular orbital (MO) calculation. The energy barrier, E-0(double dagger), which is the most sensitive parameter in the calculation, was adjusted until the TST rate coefficient most closely matched the observed one. This fitting yielded E-0(double dagger) = 59.0 kJ mol(-1), in excellent accord with the barrier of 62.0 kJ mol(-1) calculated with the ab initio MO method at the G2(MP2) level.

Elevating temperature to 523 K has improved NOx reduction from burned gas by an electric discharge under fuel-lean conditions; addition of water vapor to the burned gas improved the NOx reduction. We found that NiO - Ni2O3 set in a discharge tube was very effective for NOx elimination in that about 70% of NOx was removed under fuel-lean conditions.

To estimate the inhibition effect of several bromine-free polyfluoroalkylamines such as (CF3CF2)(3)N, (CF3)(2)NCF2CF3, (CF3)(2)NCF2CHF2, and (CF3)(2)NCF=CF2 on flame propagation, the laminar burning velocity was measured for a mixture of 9.5% methane, 90.0% air, and 0.5% inhibitor, at an initial temperature of 298 +/- 2K and a pressure of 1 atm. All of the polyfluoroalkylamines inhibited flame propagation less than CF3Br (Halon 1301), but more than CH3CHFCF3 (FM200). Calculations showed that the inhibition effect of polyfluoroalkylamines was caused not only by physical factors, but also by a chemical process in which fluoric species capture chain carriers (H, O, and OH) of combustion reactions to form stable HF molecules. The inhibition efficiencies of polyfluoroalkylamines were found to be higher than those of polyfluoroalkanes because of polyfluoroalkylamines decomposition to reactive polyfluoroalkyl radicals in the lower temperature region of each flame. The application of polyfluoralkylamines to fire-extinguishers was also discussed.

Ignition delay times were measured in a CH4-O2-HBr-Ar mixture behind reflected shock waves. Hydrogen bromide retarded methane ignition, though methyl bromide accelerated it. Model calculations showed that two different chain cycles, promotion and inhibition cycles, were formed for bromine-containing species. The inhibition cycle was predominant in the HBr addition.

We have been investigating Fourier transform laser-induced photoacoustic spectroscopy (FT-LIPAS) for speciation of actinides in aqueous solution, and ultraviolet, visible and near-infrared photoacoustic spectroscopy (UV-VIS-NIR PAS) and Fourier transform infrared photoacoustic spectroscopy (FT-IR PAS) for speciation in the solid phase. These spectroscopic methods were applied to the speciation of uranium(VI) in NaHCO3/NaClO4 solution/precipitate systems. The results show that FT-LIPAS is capable of speciation of aqueous species at concentrations of 10(-4) M and that UV-VIS-NIR PAS and FT-IR PAS are useful for the speciation of the solid phase, especially for amorphous materials.

An investigation of the influence of obstacles on flame propagation has been performed in an open space confined only by a floor. One wire net or several nets piled on top of each other were laid on the floor as continuous obstacles. In the presence of nets, the flame shape became more slender than without ones. The highest flame speed under the nets-laid condition was about four times as fast as that under the nets-free one, although no transition to detonation was found. The velocity of gas flow ahead of the flame, which seems to be intimately related to the flame propagation, has also been calculated by using a simple model. The calculation proved that "an imaginary center of thermal expansion of gas due to combustion" moved downward to the floor under the nets-laid condition, and it could also explain some data observed for flame propagation.

The surface fluorination of graphite in some glow discharge plasmas was studied. The plasmas contained excited fluorine atoms, and were, generated by various discharges, Direct-Current, Radio-Frequency (13.56MHz) and Micro-Wave (2450MHz), with CF₄ as source gas. The contact angles formed by water, dropped on the treated surfaces, were over 160°, and changes of mass indicated addition of the fluorine.<BR>But the effect of etching appeared as a decrease of mass under some conditions, and such effects were observed also by a scanning electron microscope. The ESCA results on the surfaces treated by plasma were similar to those which were measured on the surfaces treated by F₂ gas at high temperature.