Elementary Reaction Kinetics of Fluorine Atoms, FO, and NF Free

1 Elementary Reaction Kinetics of Fluorine Atoms, F O , and NF Free Radicals E . H. A P P E L M A N Argonne National Laboratory, Argonne, IL 60439

Downloaded by 80.82.77.83 on May 16, 2018 | https://pubs.acs.org Publication Date: June 1, 1978 | doi: 10.1021/bk-1978-0066.ch001

M. A. A. CLYNE Department of Chemistry, Queen Mary College, London E1 4NS, England

The study of reactions of free fluorine atoms in their ground 2p5 P ,1/2 states has progressed rapidly in the last several years. Major technique difficulties have been gradually overcome, and several methods of determining F atom concentrations are now available. Elementary reaction kinetics in low pressure flow systems have received particular attention, probably because of the relative ease of interfacing flow systems to F-atom detection devices such as mass spectrometers and vacuum-uv atomic resonance spectrometers. Photolytic production of F atoms is less easy F , for instance, absorbs only very weakly in the near ultraviolet or visible regions. Similarly,detection of F P atoms in static systems presents problems that have not been fully solved at the present time. Therefore, the work described in this article has mostly been carried out in flow systems. Current interest in fluorine-based chemical lasers extends from HF and DF lasers based on F atom reactions into the area of elementary processes involving electronically excited radicals such as NF. Also, the chemistry of NF and OF radicals is of con siderable interest, and we summarize here selected aspects of the kinetic behaviour of these species. 2

3/2

2

2

J

2

Production of F P atoms J

2

Ground state P halogen atoms may be produced by direct dissociation of the molecular halogens, F , Cl , B r and I . A convenient and commonly-used technique i s to pass a mixture of molecular halogen diluted with argon or helium through a microwave discharge at a t o t a l pressure near 100-200 Ν m"". This i s the simplest technique for forming F P j atoms (1-4); the typical degree of dissociation of F using an uncoated s i l i c a discharge tube i s 70-80%(1). Mixtures of fluorine diluted with inert gas can be handled i n a conventional glass flow vacuum system. A J

2

2

2

2

2

2

©

0-8412-0399-7/78/47-066-003$10.00/0

Root; Fluorine-Containing Free Radicals ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

2

4

FLUORINE-CONTAINING

F R E E

RADICALS

problem with microwave-excited dissociation of F i s that SiF^ and molecular and atomic oxygen are produced as a result of attack on the S i 0 material of the discharge and flow tubes (1, 2, 5) . Rosner and Allendorf (3_) have replaced the s i l i c a discharge tube by one fabricated from pure fused alumina, which i s inert to attack by fluorine atoms. However, this reduces the efficiency of dissociation somewhat. The pyrex or s i l i c a flow tube may be pro tected from attack by F atoms by application of a thin coating of teflon or Kel-F fluorocarbon polymer (2). A technique for coating with a thin fused layer of teflon has been described (6). The presence of undissociated F i n the discharge products causes complications i n kinetic studies involving hydrogen atoms or alkyl radicals, which react rapidly with F . However, F appears to be unreactive towards such other common atoms as Ο P, Ν **S, CI P and Br P (2,7) . d J — — Production of F P j atoms can also be achieved by microwave dissociation of fluorides such as S F and CT% (8). These sources are satisfactory for spectroscopic studies such as the epr spect roscopy of F Ρ 3 / 2 ^ i / 2 (§)r or for the formation of ex cited BrF and IF from the recombination of Br + F or I + F i n the presence of singlet oxygen (9). However, they are in some degree suspect for systematic quantitative kinetic studies, i n as much as the nature and reactivity of the discharge products other than F P j are very incompletely known. In principle, a cleaner source of F P j atoms than any of the preceding ones i s the rapid reaction of Ν **S atoms with N F radicals produced by thermolysis of N F ^ (10,11). This reaction occurs either directly to give F atoms Ν + N F -> 2F + N (10) or via the formation of NF radicals; Ν + N F -> 2NF (11), followed by NF + NF -> N + 2F or Ν + NF -»· N + F. In practice, the i n s t a b i l i t y of N ^ L J . and i t s expense have precluded extensive use of the Ν + N F method for forming F atoms. 2

2

2


2

2

3

2

2

T

T

2

6

2

a n

F

2 p

2

2

2

2

2

2

2

2

2

2

2

The formation of F P j atoms from the bimolecular reaction of NO with F i s extensively used i n HF, DF chemical lasers: NO + F -> FNO + F. The kinetics of this reaction have been studied (12,13). 2

2

Measurement of Atom Concentrations : absolute concentrations. The range of methods that have been employed to measure atom concentrations i s wide, and includes thermal and diffusion methods which are not considered here. Most of the physical methods a v a i l able are suitable only for relative concentration measurements, and require calibration by chemical means in order to y i e l d absolute concentrations. This need not always be a disadvantage; for instance, i n kinetic studies of simple atom reactions under pseudo-first order conditions i t i s sometimes only necessary to monitor relative changes i n atom concentrations.


1.

A P P E L M A N

AND

Elementary Reaction Kinetics of F Atoms

CLYNE

5

Titration reactions based on simple stoichiometry, and pro ceeding extremely rapidly, have been developed for the measurement of absolute concentrations of ground state atoms i n flow systems. Well-known examples for Ν, Ο and Η atoms include the reactions, Ν + NO -* N + Ο, Ο + N0 -> NO + Ο , H + ClNO -> HCl + NO and Η + N0 -> OH + NO. These reactions permit the determination of absolute atom concentrations by the addition of known quantities of a stable reagent. Several similar t i t r a t i o n reactions have been characterized and used to determine F atom concentrations, namely 2

2

2

F + ClNO -*· NO + GIF

(1) ,

F + C l -* C1F + CI

(2) ,

F + Br

(3),


2

F + H

2

2

+ BrF + Br -> HF + H

(4) .

The reaction of ClNO with F has been used to determine F atom concentrations by mass spectrometric measurement of the C1F produced after the addition of excess ClNO. The C1F ion peak was calibrated by the addition of measured quantities of ClF from a cylinder (14). To eliminate the need to work with the reactive ClF gas, a variant of this method was developed i n which the F atoms were titrated with ClNO to an endpoint identified as the amount of added ClNO beyond which the C1F peak no longer incr eased (1). The ClNO methods are less well established than some other, inasmuch as the rate constant of reaction (1) has not been determined. Also, at high concentrations, the reaction NO + F ->· FNO + F, which has a rate constant of 8 χ 10~ cm molecule" s a t 300 K, may occur to an appreciable extent, lead ing to an erroneous value for [ F ] . +

+

15

3

2

1

- 1

The reaction of F atoms with C l appears to be an excellent means of determining absolute F atom concentrations. Reaction (2) has a forward rate constant around 1 χ ΙΟ" cm molecule" s" at 300 Κ and an equilibrium constant greater than 5 (1_, 7,15) . Hence the reverse reaction i s unlikely to be important under the correct conditions. Furthermore, the possible secondary reaction 2

10

CI + F

2

3

1

1

•> ClF + F ll+

3

1

1

has a rate constant less than 5 χ 10~ cm molecule"" s" at 300 Κ (1_) and may be safely neglected. Reaction (2) has been used to determine F atom concentrations mass spectrometrically by addition of a measured excess concentration of C l and subsequent deter mination of either the amount of C l consumed (7_) or the amount of ClF produced (4) . In the l a t t e r case the C1F ion current was calibrated by measuring i t s intensity when an exces-s of F atoms was added to a known amount of C l . Measurement of ClF production i s more sensitive than measurement of C l consumption and i s also free from errors that might result from recombination of CI atoms or reaction of C l with other species such as H or O. Reaction (2) has also been u t i l i z e d as a t i t r a t i o n of F atoms, with the 2

2

+

2

2

2


6

FLUORINE-CONTAINING

F R E E

RADICALS

endpoint estimated to be the point of c r i t i c a l extinction of the F atom fluorescence (7) . In an alternative t i t r a t i o n procedure, the chemiluminescenee accompanying the recombination of the CI atoms formed i n reaction (2) i s monitored, and C l i s added u n t i l the intensity of this emission reaches a plateau. At low pressures, this method yields a sharp endpoint (16). 2

Concentrations of F atoms have also been determined by t i t r a t i o n with Br to the point of extinction of the F atom fluor escence (7). Reaction(3) has a rate constant of about 2 χ 10" cm molecule" sec" at 298 K, and an equilibrium constant i n ex cess of 10 (7). Hence i t s back reaction i s of no concern and i n this respect i t should be superior to reaction (2) as a means of determining F atom concentrations. On the other hand, the rate of the reaction (5), 2

10

3

1

1


9

Br + F

2

-> BrF + F

(5)

does not appear to be known. In practice reactions (2) and (3) appear to be equally satisfactory for the determination of F atom concentrations. 11

3

Reaction (4) has a rate constant of about 2 χ 10" cm molecule" s" at 300 Κ (10) and has been used to determine F atom concentration. However, the secondary reaction 1

1

F

2

+ H + HF + F

(6) 12

3

1

1

has a rate constant of about 3 χ 10" cm molecule" s" at 300 K. (17). Inasmuch as reaction (6) regenerates the F atoms consumed in reaction (4), when a substantial amount of F i s present the added H w i l l f i r s t consume the F and only subsequently w i l l i t consume the F. Advantage has been taken of this effect to t i t r a t e both F and F with hydrogen mass spectrometrically, using an i n l e t system that incorporates an inhomogeneous magnetic f i e l d to permit monitoring of both the F and F (2). 2

2

2

2

2

Measurement of Atom Concentrations: Atomic Resonance The method of atomic resonance spectrometry i n the vacuum ultraviolet, with detection of either absorption or fluorescence, has become one of the most useful direct methods for the measure ments of reaction rates of ground (18)(and metastable excited (19/20) ) state atoms. The sensitivity and scope of atomic res onance in this respect r i v a l s , and possibly surpasses, that of other methods such as epr and mass spectrometry. Recently, i t has been shown that atomic resonance may be used to measure fluorine atom concentrations in a flow system (7,21). The sensitivity for F i s not yet as high as for atoms such as CI, whose resonance transitions l i e i n the Schumann region of the vacuum ultraviolet. The essence of the method i s as follows. The source of atom i c resonance radiation, usually a microwave-excited discharge i n a low-pressure flowing gas, i s incident on the reaction vessel, either static or flowing, i n which ground state atoms are present.


1.

A P P E L M A N

A N D


C L Y N E

7

On account of the coincidence of the emission line source and the absorption line of ground state atoms, specific and very intense absorption of the exciting radiation occurs. Either the fraction a l absorption intensity, h-I^ /I , or the corresponding fluor escence intensity, I , may be measured and related to the concent ration of absorbing atoms, N. Figure 1 shows a schematic diagram of the atomic resonance experiment. p

The central problem of relating light intensity to concentra tion of absorbers, N, has been discussed (22-24). The fluorescen ce intensity varies in direct proportion to Ν at sufficiently low values of Ν (typically ^ 1 0 cm"" (24)) . This proportionality clearly f a c i l i t a t e s kinetic studies. However, absorption intensities at such low atom concentrations cannot normally be measured, and the dependence of I A upon Ν at measureable absorbances i s complex (22_,25) . 8n the other hand, the magnitude of detected fluorescence intensity, Ι , i s always several orders of magnitude less than that of I , which can lead to low signalto-noise ratios when short integrating times are employed for the measurement of light intensity. In such cases, e.g. i n flash photolysis experiments (26), i t i s not generally possible to use both time resolution and wavelength resolution when atomic res onance fluorescence i s being used to measure the kinetics of elementary reactions. On account of the long integrating times possible i n discharge-flow systems, both time resolution (using linear displacement) and wavelength resolution may routinely be used. There i s then no d i f f i c u l t y i n working i n the region where I varies proportionally with Ν (27). F The most usual approach i n resonance absorption and fluores cence work has been to use microwave lamps, and to calibrate resonance absorption or fluorescence intensity by chemical t i t r a t i o n (see above). Extensive studies of this type on I and Br Ρ3/2Ί/2 atoms (19_, 20_, 27,28) and on CI Ρ3/2Ί/2 (26 ~ 30) have been described. Detection of F atoms by resonance absorption using a conventional microwave lamp has been successful (7,21) . In this case, the fluorine atom lamp developed was based on a dilute mixture of F with He. (Argon lamps gave extremely low intensities of F atom l i n e s ) . Figures 2 and 3 show the rele vant energy levels of F and the spectrum of the F + He microwaveexcited lamp. Because the f i r s t resonance transition of F(3s-2p ) l i e s well below the lithium fluoride absorption cut-off (^ 105 nm), the usual windows were replaced by collimated hole structures in order to separate the lamp from the absorbing atoms i n the flow tube and from the spectrometer. The arrangement used (7) i s shown in Figure 4.


12

3

b

S

ρ

2

2

a t o m s

2

2

5

For reasons already given, resonance fluorescence would be a preferred technique for kinetic studies of F atom reactions. Bemand and Clyne (7) detected resonance fluorescence i n a number of lines of the 3s-2p multiplet, using the apparatus of figure 4. 5


F R E E


FLUORINE-CONTAINING

Figure 2. Atomic resonance transitions for F(I). Wavelengths and wave numbers of the lower resonance transitions of F (I).


RADICALS

1.

A P P E L M A N

A N D C L Y N E

3s'-2p

4s-2p 2

P -

2

5

2


9

D £ P

5

P

ΓΤΠ


w

5

3s - 2p

ι ι—m Figure 3. Emission spectrum of F atom resonance lamp. F (l mol %) + He, 25-W incident microwave power. Lower spec trum: ' Pj — Pj multiplets of the 3s — 2p transition using 5μτη X 10 mm slits. De tected photon flux rate at 95.85 nm = 1.8 kHz. Upper spectrum shows the much weaker 3s' Dj - 2p Pj and 4s Pj - 2p Pj multiplets, using 13μm X 10 mm slits. Detected photon count at 80.70 nm = 50 Hz. Note that the lamp spectrum shows negligible continuum background. Noise (photomultiplier dark count -f- scattered light) count rate = 2.5 Hz. 2

2 4

2

2

5

5 2

2

2

hi Κ

ME

m 'Μ

H«/F ^

SE

2

Κ

7^

f

f-*t

1

3i-

0 ι

txhouit - to fwrr.p

reaction zone

ι e x h a u s t - : ο pump

Figure 4. Windowless system for resonance spectrometry of F Pj atoms. B, buffer chamber; C, collimated hole structures, two of which were mounted on specially fabricated glass discs, D; K, inlet to lamp L; M, M , differential manometers; R, section of flow tube; S, spectrometer slit sealed via an O-ring seal to the silica apparatus; V, vacuum spectrometer. 2

1


5

10

FLUORINE-CONTAINING

F R E E

RADICALS

Ζ

Although weak fluorescence i n the **Ρ- Ρ lines was observed, from a practical standpoint only the P - P lines were found intense enough for resonance fluorescence work at low atom concentrations. The fluorescence count rates using the whole of the fully-allowed P - P transition of F were typically 2 counts s" at [F] = 1 χ 10 cm . These data set a lower concentration l i m i t of 1 χ 10 cm for the smallest detectable concentration of F P3/2 atoms (7). The corresponding lower limits for CI, Br and I atoms (27,28) are appreciably less, and are continually being improved by better attention to collimation and detection. Because of the low count rates observed i n the F atom resonance fluorescence studies, a, better approach i s to use resonance absorption with a non-reversed line source (67). 2


2

2

2

1

12

3

11

-3

2

The sensitivity of resonance fluorescence by F atoms was, however, sufficient for kinetic studies of the rapid t i t r a t i o n reaction of F with Br : F + Br -> BrF + Br (7). It was necessary to make corrections both for non-linearity of fluorescence intensity with [F] , and for deviations from pseudo-first-order kinetic conditions, both due to the relatively high F P ^ atom concentrations used (> 1 χ Ι Ο cm" ). The f i n a l rate constant for the F + Br reaction was (1.4 ± 0.5) χ 10~ at 298 K, i n f a i r agreement with results from the comparably d i f f i c u l t mass spectro metry study of Appelman and Clyne (5) (k = (3.1 ± 0.9) χ 1 0 ~ ) . 2

2

2

2

1 1

3

10

2

10

Measurement of atom concentrations:

EPR and chemiluminescence

Although electron paramagnetic resonance has been used as a very successful technique for kinetic studies of elementary react ions, including those of ground state Ο P , H S, and Ν atoms and of OH Π radicals (31-35), l i t t l e effort appears to have been expended on the use of HTis~~technique to study reactions of halo gen atoms. However, the epr spectra of the P / and Ρ χ / states of F, CI, Br and I have a l l been described (36-41), with the except ion of the I Ρ χ / state. 3

2

2

2

2

3

2

2

2

2

A completely different technique for measuring relative F atom concentrations makes use of the intensity of the red chemi luminescence emitted i n the third-order reaction, F + NO + M •> FNO* + M (42). The intensity of emission i s reported to vary directly with [ F f and with [NO] . Kinetic studies using this technique, which appears simple and convenient to use, have been described (43) . More work remains to be done, however, to c l a r i f y the nature of the emitter and the detailed kinetics of the emission process. Measurement of atom concentrations:

Mass spectrometry.

In principle, mass spectrometry provides a specific and highly sensitive method for the detection of gaseous free radic als . Foner (41) has reviewed the techniques available for sampl ing free radicals, and for their mass spectrometrie detection. Unless the radical source i s at pressures comparable to atmospheric


1.

A P P E L M A N


A N D C L Y N E

11

pressure, sampling of radicals through o r i f i c e s usually occurs under conditions approximating to effusive flow, or intermediate between effusive and supersonic flow, thus minimizing collisions within the sampling system. Most of the sampling systems i n which radicals were produced i n a discharge-flow system (near 100 N m" total pressure) have been designed u t i l i z i n g the principles laid down by Foner (44). In a typical system the mass spectrometer i s separated from the flow tube by two in-line, orifices of 0.7 mm dia meter spaced 1 cm apart (45) . The space between the o r i f i c e s i s maintained at a pressure of 0.2 Nm 2 less by d i f f e r e n t i a l pump ing with a large diffusion pump. Ion source pressures are of the order of 10"% m" . A typical system i s shown i n Figure 5. 2

o r


2

Although i t i s sometimes possible to detect radicals i n the presence of large concentrations of a parent molecule by the use of low energy ionizing electrons, the ready dissociation of the F molecule makes i t impractical to detect F atoms mass spectrometr i c a l l y i n the presence of excess F . Hence, direct mass spectrometric detection of F atoms i s limited to those situations i n which l i t t l e or no F i s present. This i s the case when F atoms are generated by a microwave discharge i n a very dilute mixture of F with an inert carrier gas (< 0.1 mol-% F ) . It i s also the case when F atoms are produced by a discharge i n compounds such as CFi±, or by such reactions as 2

2

2

2

2

or

NF

2

+ Ν •> N

NF

2

+ Ο -> NO + 2F

2

+ 2F

NO (excess) + F

2

-> FNO + F.

+

However, formation of F i n the ion source by fragmentation of CF^, NF , or FNO must be considered. The formation of F by fragment ation of spin-paired molecules can be avoided by placing a barrier between the molecular beam i n l e t and the ionizer and then using an inhomogeneous magnetic f i e l d to bend paramagnetic molecules around the barrier. In this way spin-paired molecules are largely pre vented from reaching the ionizer, and F may be detected i n the presence of F . Removal of the barrier allows z ke monitored (46). When possible,direct monitoring of F as F+ ions represents one of the most successful approaches to quantitative kinetic studies of F atom reactions, as demonstrated by the work of Warnatz, Wagner and Zetzsch (10,14). 2

F

t o

2

Mass spectrometry may also be used for the indirect detection of F atoms by the attenuation of the F peak when a microwave dis charge i s struck i n a mixture of F with inert carrier gas. This method obviously makes no allowance for reactions that consume F atoms. Indirect detection of F atoms i n the mass spectrometer may also be effected by monitoring either the consumption of reagents that react rapidly with F atoms or the formation of products i n such reactions. These techniques have already been discussed i n the section on measurement of absolute F atom concentrations. 2

2



12

FLUORINE-CONTAINING

F R E E

RADICALS

Figure 5. Schematic of flow system for on-line mass spectrometric studies of F atom reactions. A, flow of F atoms in He carrier from discharge; B, silica flow tube; C, first pinhole (0.7 mm) separating flow tube from differentially-pumped chamber; D, second pinhole (0.7 mm) separating differentially-pumped chamber from quadrupole mass spectrometer Q(E.A.I. 150/A); E, access to discharge tube; F, furnace; G, 50 mm gate valve; H, connections to total pressure manometer; IG, ion gage; J, connection to cold traps and large pump (60 Is' ); M, multiplier; P P , pumping lines for quadrupole chamber and for differentially-pumped chamber; R, reagent inlets. 1

ly

2


1.

A P P E L M A N

A N D

13


C L Y N E

Addition of Fluorine Atoms to Spin-Paired Inorganic Molecules. There has been very l i t t l e study of reactions i n which fluorine atoms add to spin-paired inorganic molecules with the ultimate formation of stable products (addition of F atoms to olefins i s discussed elsewhere i n this monograph). We may expect such reactions to be rather slow termolecular processes and hence to be rather d i f f i c u l t to investigate. The reaction of F with excess CO has been studied i n a discharge-flow system (4). Both COF and C0 were identified as products by mass spectrometry, and the F atom concentration was measured by mass spectrometric monitoring of the ClF formed when an excess of C l was added at the end of the reaction zone. The results were consistent with a rate determining step k CO + F + M -> COF + M, (7) although the dependence of the rate constant on [M] was not uniquely determined. Values of 3.4 ί 1 and 5.7 ± 2 χ 10"" cm molecule" s" , respectively, were obtained for k with helium and argon as third bodies at 298 K. A more recent study of reaction (7) has used epr monitoring of the F atoms to give values of 6.5 ± 1 and 7 ± 1 χ Ι Ο " cm molecule" sec" i n helium and n i t rogen, respectively at 293 Κ (47). 2

2


2

32

2

6

1

32

6

2

2

A study of reaction (7) by flash photolysis of mixtures of N F^, CO, and N resulted in the observation of a transient u l t r a v i o l e t spectrum that was attributed to COF (48) . The decay of the transient produced the spectrum of (COF)2(48)· (C0F did not ab sorb i n the spectral region that was examined). 2

2

2

Formation of FCO has been observed to result from the reaction of CO with photolytically formed F atoms i n a solid matrix at l i q u i d helium temperature (49). The authors of this a r t i c l e conclude that the reaction of CO with F has l i t t l e or no activat ion energy. The COF reacts further i n the matrix to form both C0F and (C0F) . The reaction of F atoms with excess xenon was studied i n the same discharge-flow apparatus used for the F + CO reaction, and XeF was identified as a product (|) . The rate determining step was assumed to be k Xe + F + M -> XeF + M , (8) and on the basis of a single measurement k was estimated to be about 2.3 χ l o " cm molecule" s" at 298 Κ with argon as the third body. In the same discharge-flow system no evidence was found for the reaction of F atoms with krypton i n helium at 298 K, and an upper l i m i t of 2 χ 1θ" cm molecule' s" was set for the rate constant of the reaction 2

2

2

3 3

6

2

3Ι+

6

Kr + F + M

1

2

KrF + M

1

(4)


14

FLUORINE-CONTAINING

F R E E

RADICALS

These differences i n the rate of reaction of F atoms with Κι, Xe and CO correlate with the fact that the fluorine atoms become progressively more strongly bound as one goes from KrF2 to XeF to COF . Both XeF and KrF have been produced by the reaction of photolytically produced F atoms with Xe and Kr i n low temperature matrices,(50,51). 2

2

It i s not known what subsequent stepsproduce the f i n a l prod ucts from the reactions of F atoms with CO and Xe. The most plausible process appears to be disproportion tion, e.g., 2C0F -> CO + C0F .


2

Also possible are additions of a second F atom by a more rapid termolecular process, e.g. COF + F + M -> C0F + M, 2

or reaction with F

2

and F i n schemes such as COF + F

2

-> C0F + F 2

COF + F -> CO + F . 2

In the case of CO, the reaction 2C0F -> (COF) involved.

2

i s apparently also

We may note that C0F and XeF can be prepared by the thermal reactions of F with CO and Xe respectively. (52,53). These reactions very probably proceed via the formation of F atoms from the readily dissociated F molecule, followed by reactions (7) and (8) to form the Xe-F and C-F bonds. 2

2

2

2

Warnatz (54) looked at the addition of F to HCN and (CN) , obtaining bimolecular rate constants of 1.5 χ 1 0 and 5 χ 10" cm molecule s" at 300 Κ respectively for the reactions 2

_10

13

3

-1

1

F + HCN -> HFC Ν a n

F + (CN)

2

-> F(CN) . 2

These reactions were measured mass spectrometrically in a discharge flow system at pressures of 470Ν " , and they thus appear to be true bimolecular processes. Both reactions have apparent negative activation energies (-5 kJ/mole for F + HCN and -2 kJ/mole for F + (CN) ). The reaction of F with HCN i s followed by 2

m

2

F + HFCN -> HF + FCN

(54) .

Bimolecular Reactions of F Atoms. a)

Hydrogen Abstraction

These reactions are discussed elsewhere i n this monograph and have been reviewed by Foon and Kaufman (55) and by Jones and Skolnik (43). The reactions are generally very fast, with rate constants that are within a factor of 10 of the hard-sphere bimolecular c o l l i s i o n frequency. Notable exceptions are


1.

A P P E L M A N

15


A N D C L Y N E

12

3

1

1

F + NH (k o = 1 x 10" cm molecule" s" ) and F + trihalogenated methanes, such as CHF , CHC1 F and CHC1F (k o 2 χ 10" cm molecule" s " ) , and C H C l ( k = 5 χ 10" cm molecule"" s" ) . (42_, 1_, 54_, 68). 3

30

K

3

12

3

1

2

2

12

3

1

30

1

3

300

1

We have already discussed the reaction of F with H as a means of determining F atom concentrations. A novel hydrogen abstract ion reaction i s the reaction of F atoms with the recently pre pared hypofluorous acid: 2

F + HOF -> HF + OF.


10

3

1

A lower limit of 2 χ 10" cm molecule"" s^has been set for the rate constant of this reaction at 300 Κ on the basis of a massspectrometric discharge-flow study (4). I t i s one of the two reactions known to produce OF radicals. Inasmuch as the l a t t e r decompose to reform F atoms (45): 20F -> 0

2

+ 2F,

we have, i n effect, an F-atom catalyzed decomposition of HOF. These reactions may contribute to the spontaneous decomposition of HOF at room temperature, a decomposition that i s enchanced by F and inhibited by such F-atom scavengers as H and CO (56). 2

b)

2

Oxygen Abstraction

Abstraction of oxygen by fluorine atoms requires a reagent molecule containing a relatively weakly bonded oxygen atom, inas much as the OF product i s only bound by 213 kJ/mole (57) . The only well established case i s the reaction with ozone, which i s important as a source of OF radicals: F + 0

3

-> OF + 0 . 2

A mass spectrometrie discharge flow study gives for the rate constant k = 2.8 χ 10"" exp(-450 ± 400 c a l mol^/RT) cm molecule"" - s" , corresponding to a value of 1.3 χ 10" cm molecule" s" at 300 K. In this study, [θ], [F] and \OF] were a l l monitored, and conditions ranged from excess Ο to excess F. The F atom concentration tends to remain constant even with excess 0, because of the regeneration of F atoms i n the subsequent decompo sition of OF. Hence a pseudo-first order kinetic analysis can be applied (58). 11

1

3

1

1

11

3

1

Inasmuch as the energy of a xenon-oxygen bond i s only about 88 kJ/mole (53), F atoms should be able to extract oxygen from such molecules as XeO^ and XeOF^. These reactions do not, however, appear to have been investigated. c)

Halogen Abstraction.

The reaction of F atoms with C l and B r have already been discussed as means for determining F atom concentrations. The reaction F + C l -> ClF + CI 2

2

2


16

FLUORINE-CONTAINING

F R E E

RADICALS

has been the subject of three mass spectrometric discharge-flow studies, one with excess C l and the others with excess F. (1_,4_, 14). The studies using excess F monitored the disappearance of C l ; the study with excess C l monitored the disappearance of F and the formation of ClF and C l . The reported rate constants range from 0.86 to 1.6 χ 10~ cm molecule" s at 300 K. A mean value of 1.2 χ lo"* ^ may be as good a choice as any. The reaction 2

2

2

10

3

1

- 1

1

F + B r -> BrF + Br, 2

has been studied i n discharge-flow systems with mass spectromet r i c monitoring of Br in the presence of excess F, (4) and by monitoring F atom resonance fluorescence i n the presence of excess Br (7). The two studies yielded respective 300 Κ rate constants of 3.1 ± 0.9 and 1.4 ± 0.5 χ 10~" cm molecule" s" . The higher value may be a l i t t l e more reliable inasmuch as the fluorescence study was subject to uncertain attenuation of F atoms on surfaces and also involved a significant correction for s e l f reversal of the fluorescence. Warnatz reported a preliminary value of 0.8 χ Ι Ο " for this constant (54).


2

2

10

3

1

1

10

The reactions of F atoms with C l and Br can constitute convenient sources of Cl and Br atoms, respectively. It i s often possible to generate higher concentrations of such atoms i n this way than by the passage of a discharge through chlorine or bromine (4). 2

2

The reactions F + I

2

•> IF + F

(9) ,

F + ICI -> IF + Cl

(10) ,

F + ICI -> ClF + F

(11) ,

have been studied by mass spectrometric monitoring of the con sumption of I and ICI and the formation of ClF i n a discharge flow system i n the presence of excess F (4). Respective rate constants of 4.3 ± 1.1, 3.8 ± 1.5, and 1.2" ± 0.7 x l O " cm molecule" s" were obtained at 300 K. The rate constant for reaction (10) sets a lower l i m i t to the binding energy of IF and allows a choice to be made among the several estimates of this quantity that have been put forth. 2

10

1

3

1

The reactions of F atoms with Br , I and ICI are complicated by the fact that BrF and IF are unstable to disproportionation, with the subsequent formation of higher fluorides and B r and I . As a result, the reaction of an excess of F atoms with Br forms BrF rapidly and then, on a longer time scale, goes on to form BrF . Surprisingly, there i s no evidence for the formation of substantial amounts of the stable intermediate fluoride BrF3. When an excess of F atoms reacts with I or ICI, formation of IF5 appears to be nearly competitive with IF formation, particularly at higher con centrations. I t i s d i f f i c u l t to imagine a homogeneous process 2

2

2

2

2

5

2


1.

A P P E L M A N

A N D


C L Y N E

17

that could form IF5 so rapidly, and i t seems that surface react ions are involved. A mechanism has been proposed i n which the IF polymerizes reversibly on the walls of the reaction tube and I F i s formed by the subsequent heterogeneous reaction of F atoms with the polymer (4). It seems l i k e l y that the formation of BrF takes place by a similar process. 5

5

Inasmuch as the F molecule i s only bound by about 155 kJ/ mole, fluorine abstraction by F atoms requires a reagent with a very weakly bound fluorine. One p o s s i b i l i t y i s the reaction 2

F + 0F

-> F

2

2

+ OF.

A kinetic analysis of the thermal decomposition of 0F yields for this reaction a rate constant k = 8 χ 10~ exp(-57.3 ± 4 k j/mole" /RT) cm molecule" s" and implies a negligibly slow reaction rate at room temperature (32) .


2

11+

3

1

1

Xenon-fluorine bonds have energies around 125 kJ/mole, (53.) and we might expect F atoms to abstract fluorine from the xenon fluorides. However, upper limits of about 7 χ 10" cm molecule s" have been set for the reaction of F with XeF and XeFi+ at 300 Κ on the basis of mass spectrometric studies in a dis charge-flow system (4). It i s possible that removal of the f i r s t fluorine from either of these compounds may require considerably more than 125 kJ/mole. Johnston and Woolfolk (60) found that NO and N0 reacted much more slowly with XeF and XeF^ than with F . By correlating reaction rates with bond energies,they concluded that the reactions 16

1

3

1

2

2

2

2

XeF ·> XeF + F 2

and

XeF^ -> XeF + F 3

were endoergic by 226 and 201 kJ/mole, respectively. The reactions of F atoms with perhalogenated alkanes have been reviewed by Foon and Kaufman (55) and by Jones and Skolnik (43). As might be expected, those reactions that are exoergic, such as CCl3Br + F -> BrF + CCI3 and CF3I + F -> IF + CF , proceed at rates close to the bimolecular c o l l i s i o n frequency. Endoergic abstraction reactions are very much slower, and in some cases they may have to compete with fluorine substitution. There are wide discrepancies in the rates and mechanisms reported for these latter reactions. Thus the reported rate constants for the reaction of F with CCl^ span a range of 3 χ 10 ! Production and reactions of FO radicals. 3

7

The gaseous CIO, BrO and 10 radicals have a l l been observed by uv absorption spectroscopy in both static and flow systems. However, no optical or epr spectrum of the gaseous FO radical has


1

18

FLUORINE-CONTAINING

F R E E

RADICALS

been reported, even though definite evidence for this species was obtained i n Ar and N matrices at 4 Κ using infrared spectrometry (61). This i s surprising i n view of the ready identification of FO radicals by mass spectrometry after their production i n a d i s charge-flow system by the rapid reaction, F + 0 -> FO + 0 , (57,58). It i s possible that the FO analogue of the well-known Π - Χ Π uv band absorption systems of CIO, BrO and 10 (62) i s diffuse and therefore d i f f i c u l t to identify. Franck-Condon factors for discrete vibronic transitions from ν" = Ο of FO Χ Π may be low, with the main fraction of the o s c i l l a t o r strength of the ν" = Ο transitions occurring i n the continuum region. I t i s also possi ble to rationalize the failure to observe the epr spectrum of FO. According to unpublished experiments of Clyne and Sales, any epr spectrum of FO must be at least a factor of 20 weaker than that of BrO, for the same concentrations of FO and BrO. This i s most l i k e l y due to the FO radical having a very small dipole moment because of the similar electronegativities of Ο arid F. 2

3

2

2

2


2

Figure 6 shows the flow apparatus, i n l e t system, and mass spectrometer used by Clyne and Watson for kinetic studies of FO radical reactions (45). The source of FO was the rapid reaction F + 0 -> FO + 0 . Absolute concentrations of FO were determined mass spe ctrome t r i cal1y by addition of excess NO and monitoring of the N0 produced i n the rapid reaction: 3

2

2

FO + NO + F

+ N0 , 2

12

3

1

which has a rate constant i n excess of 5 χ 10" cm molecule" s" at 298 K. The concentration of FO at any time i s equated to the concentration of N0 produced by reaction with excess NO at that time, the FO + NO reaction being essentially instantaneous under the conditions used. The F0 ion current can thus be related via the N 0 ion current to the absolute [FCQ concentration. This method i s also applicable to the determination of CIO and BrO con centrations (45) .

1

2

+

+

2

Production of FO radicals suitable for quantitative kinetic studies required careful attention to the F P atom source employed. For example, use of the Ν + N F reaction to generate F led to the formation of significant quantities of NO, and hence of N0 when the F was subsequently converted to FO by means of the F + Ο3 reaction. This N0 background complicated the measurement of absolute FO concentrations. The source of NO was either impurity N0NF i n the N F^ used to generate NF , or the reaction of N F with Ο P atom impurities i n the Ν S atom stream, or possibly both. The use of a discharge i n dilute F + He mixtures to produce the F atoms led to an acceptably low concentration of N0 . 2

2

2

2

2

2

3

2

2

4

2

2

3

Ο Ρ atoms are known to be formed to the extent of £0.l|Y]when F i s dissociated i n a s i l i c a microwave discharge tube. These atoms arise from dissociation of 0 liberated by attack of F on S i 0 . Ο P atoms should not greatly affect the measurement of absolute concentrations of FO by the N0 method inasmuch as most 2

2

3

2

2



1.

APPELMAN

A N D CLYNE


19

Figure 6a. Flow apparatus, inlet system, and mass spectrometer used by Clyne and Watson for kinetic studies of FO radical reactions. Block diagram of mass spectro metric system. Pumping lines as follows: A, differential chamber (between pinholes X and Y); B, mass spectrometer ion source; C, flow tube; D, flow tube diffusion pump (used only for overnight clean-up of flow tube); X and Ύ, sampling pinholes. System shown is for study of FO radical reactions.


20

F R E E


FLUORINE-CONTAINING

»

* t cm

Figure 6b, Flow apparatus, inlet system, and mass spectrometer used by Clyne and Watson for kinetic studies of FO radical reactions. Scale diagram of sampling system, showing pinholes X and Y, crossed-beam inlet Z, and connections to flow tube and pumping lines A and B.


RADICAL

1.

APPELMAN

AND

CLYNE


21

d

of the Ο P would be expected to be removed by excess FO radicals before reaching the point of addition of NO: Ο + FO

0

2

+ F.

Preliminary work on the FO + FO reaction, indicating this reaction to be kinetically second-order i n \jo\ (57) was confirmed by later work of Clyne and Watson (45). The second order rate constant was determined to be (8.5 ± 2.8) χ 10" cm molecule" s a t 298 K. The mechanism of this reaction i s believed to i n volve regeneration of F atoms and to be analogous to the similar second-order disproportionation reaction of CIO + CIO and BrO + BrO, i . e . : 12

3

1


_ 1

FO + FO -> 2F +

0. 2

Although the reactions of FO with NO, H atoms and Ο atoms are known to be fast (k £ 5 χ 10~ cm molecule" s" at 298 Κ), no quantitative kinetic data on these reactions appear to be a v a i l able. Since the F-0 bond energy i s only (215 ± 17) kJ mole" (57), there are many exoergic elementary reactions of FO radicals possible. Further investigation of such reactions should be rewarding. 12

3

1

1

1

Production and reactions of NF The NF radical i s isoelectronic in structure with 0 . Its electronic states are closely similar to those of 0 . The ground state i s NF Χ Σ", whilst the low-lying a Δ and b E metastable states, closely analogous to the singlet states of 0 , are expect ed and known. The singlet states of NF are somewhat more energetic than those of 0 . Douglas and Jones (62) were the f i r s t to report NF. They observed the forbidden b-X and a-X band systems i n emission, using a low-pressure discharge in NF as the source of excited NF radicals. Subsequently, Clyne and White (63) reported strong NF b-X and a-X emission from the reactions H + NF and Ο + NF . The 0-0 band of NF a-X near 874.2 nm, and the 0-0 (and 0-1) bands of the b-X system near 528.8 nm, were reported (63). These workers (63) also reported intense Β Π - Α Σ F i r s t Positive emission bands of N i n the Η ? NF reaction zone. Since the observed N Β-A bands had the same vibrational energy distribution to that from the Ν + Ν + M recombination, Clyne and White (63) proposed that Ν *+S ground state atoms were a product of the H + NF reaction. A l i k e l y reaction mechanism would comprise the following steps (1) and (2):2

2

3

1

l

+

2

2

3

2

2

3

3

2

+

2

2

2

Η + NF

2

-> NF + HF

Η + NF -> HF + Ν

(1) , (2) .

Reaction (.2) would be the only known example of a chemical react ion where Ν atoms could be produced exoergically. The details of reaction (2) are of interest. The reacting NF radical can be i n either the Χ Σ" ground state, or the a. à or iΣ"*" metastable excited states. H i s the Is S ground state, and 3

1

1

2


22

FLUORINE-CONTAINING

1

F R E E

RADICALS

3

HF i s X !' (y £ O) . If the reacting NF i s Χ Σ , spin and orbital angular momentum conservation between the reactants and products determine that the Ν atom formed i n reaction (2) shall be ^S or S. The latter i s energetically impossible? so we conclude that m NF Χ Σ ~ gives Ν **S ground state atoms. On the other hand, H S + NF a A must give Ν D, i.e. metastable excited atoms. It i s interesting that H S+ NF b E must give Ν S which i s energetic a l l y impossible. Hence we would expect that H S does not react withNF k ^ at a significant rate. In addition, formation of Ν D atoms i n the H + NF system would be a diagnostic for the occur rence of reaction between Η and NF a. à. Experimental work along these l^nes to investigate the relative reactivity of NF Χ Σ ,a A and b ^ , i s currently being carried out by Clyne and Whitefield.

2

3

2

x

2

2

1

+

2

2

1

2

2

1


3

x

The relative reactivity of the electronic states of NF i s important in considering proposals for an NF electronic transition laser based on the a Δ or b ^ states. According to Herbelin and coworkers (64), their experimental results are consistent with a product yield of 100% NF a â i n the H + NF reaction. The b ^ state i s believed to be formed by pumping of the a state by vibrationally-excited HF: - NF a à + HF(v ) -> NF b ^ + HF(v-l) . It i s not certain whether the apparent high yields of metastable singlet NF i n the H + NF reaction might in fact be due to the superior reactivity of the t r i p l e t NF Χ Σ~. 1

x

2

+

1

2

3

1

Herbelin (65) has suggested that NF a Δ should be less react ive towards atoms and radicals than NF Χ Σ~. He assumed that NF can be considered as analogous to an Ν atom - i . e . , the F atom in NF behaves as an inert spectator. The NF a and b states would then both correlate with F + Ν D excited atoms, whilst NF X would correlate to F + Ν **S. The bimolecular NF + NF reaction (3) was considered i n l i g h t of this model. If both the NF radicals are a à or b Z , N would be formed i n the a ^ excited state. For reaction between two NF a Δ radicals, formation of N a E i s endo ergic. Hence, this reaction was argued to be very slow at 298 K. On the other hand, NF Χ Σ~ + NF Χ Σ~ could give % X Z + 2F by exoergic reaction, consistent with the assumption of a high rate constant for reaction between two ground state NF radicals. 3

2

1

l

2

u

1

1

2

3

3

u

l

g

3

The failure so far to unequivocally detect NF Χ Σ i n H + NF systems using epr, even though NF Δ was detected i n the same systems (66), could be interpreted as support for the greater reactivity of NF Χ Σ" which has been suggested above. Further work on this question i s required, and i t appears that detection mass spectrometrically of NF, with atomic resonance for detection of atoms, w i l l be a promising approach. 2

Χ

3

Acknowledgement: We wish to thank Dr M Kaufman and Dr W Ε Jones for making available to us preprints of their reviews. The prep aration of this a r t i c l e was p a r t i a l l y supported by the U.S. Energy Research and Development Adminstration.


I.

APPELMAN

AND

CLYNE

Elementary

Reaction

Kinetics

of F Atoms

23


References

1.Μ.Α.A.Clyne, D.J. McKenney and R.F. Walker, Canad.J.Chem., (1973), 51 3596. 2.C.E.Kolb and M. Kaufman,J.Phys.Chem.,(1972), 76, 947. 3. D.E. Rosner and H.D. Allendorf, J.Phys.Chem., (1971), 75, 308. 4. B.R. Zegarski, T.J. Cook and T.A. Miller, J.Chem.Phys., (1975), 62, 2952. 5. Ε.Η. Appelman and M.A.A. Clyne, J.C.S. Faraday I, (1975), 71, 2072. 6. H.C. Berg and D. Kleppner, J.Sci.Instr., (1962), 33, 248. 7. P.P. Bemand and M.A.A. Clyne, J.C.S.Faraday II,(1976), 72, 191. 8. A. Carrington and D.H. Levy, J.Chem.Phys.,(1966), 44, 1298; ibid., (1970), 52, 309. 9. M.A.A. Clyne, J.A. Coxon and L.W. Townsend, J.C.S.Faraday II, (1972), 68, 2134. 10. K.H. Homann, W.C. Solomon, J. Warnatz, H.Gg. Wagner and C. Zetzsch,Ber.Bunsenges.Phys.Chem., (1970), 74, 585. 11. M.A.A. Clyne and I.F. White, Chem.Phys.Letters,(1970), 6, 465. 12. H.S. Johnston and D. Rapp, J.Chem.Phys.,(1960),33,695. 13.C.E.Kolb, J.Chem.Phys., (1976), 64, 3087. 14. J. Warnatz, H.Gg. Wagner and C. Zetzsch,Ber.Bunsenges.Phys. Chem., (1971), 75, 119. 15. P.Kim, D.I. MacLean and W.G. Valence, NTIS AD 751456, Abstracts of 164th Am.Chem.Soc.Meeting (1972). 16. P.C. Nordine, J.Chem.Phys., (1974), 61, 224. 17. R.G. Albright, A.F. Dodonov, G.K. Lavrovskaya, I.I.Morosov and V.L. Tal'roze, J.Chem.Phys., (1969), 50, 3632. 18. M.A.A. Clyne in Physical Chemistry of Fast Reactions (Ed. B.P. Levitt), (1973), 1, 245 (Plenum Press, N.Y.). 19. R.J. Donovan and H.M. Gillespie, Spec.Per.Repts., Reaction Kinetics, Vol.1 (The Chemical Society, London 1975), Ch.2. 20. D. Husain and R.J. Donovan, Adv.Photochem.,(1971), 8,1. 21. P.P. Bemand and M.A.A. Clyne, Chem.Phys.Letts.,(1973), 21,555. 22. W Braun and T. Carrington, J.Quant.Spectr.Rad.Trans.,(1969), 9, 1133. 23. F. Kaufman and D.A. Parkes, Trans.Faraday Soc.,(1970),66,1579. 24. P.P. Bemand and M.A.A. Clyne, J.C.S.Faraday II, (1973),69, 1643.



24

FLUORINE-CONTAINING FREE RADICALS

25. A.C.G. Mitchell and M.W. Zemansky, Resonance Radiation and Excited Atoms, Cambridge Univ.Press, (1971). 26. for example, W. Braun, A.M. Bass and D.D.Davis, Int.J.Chem. Kin., (1970), 2, 101. 27. M.A.A.Clyne and H.W. Cruse, J.C.S.Faraday II,(1972),68,1281. 28. M.A.A.Clyne and H.W. Cruse, J.C.S.Faraday II,(1972),68,1377. 29. R.J. Donovan, D. Husain, W. Braun, A.M. Bass and D.D. Davis, J.Chem.Phys., (1969), 50, 4115. 30. M.A.A. Clyne and H.W. Cruse, Trans.Faraday.Soc.,(1971),67, 2869. 31. A.A. Westenberg, Ann.Rev.Phys.Chem., (1973), 24, 77. 32. A.A. Westenberg, Prog.Reaction Kinetics, (1973), 7, 23. 33. P.Kim, R.B. Timmons, Int.J.Chem.Kin., (1975), 7,77. 34. C.N. Wei and R.B. Timmons, J.Chem.Phys.,(1975), 62, 3240. 35. G.A. Takacs and G.P. Glass, J.Phys.Chem.,(1973),77, 1182. 36. H.E. Radford, V.W. Hughes and V. Beltran-Lopez, Phys.Rev., (1961), 123, 153. 37. V. Beltran-Lopez and H.G. Robinson, Phys.Rev.,(1961),123,161. 38. J.S.M. Harvey, R.A. Kamper and K.R. Lea, Proc.Phys.Soc., (1960), B76, 979. 39. K.D. Bowers, R.A. Kamper andC.D.Lustig, Proc.Phys.Soc., (1957), Β70, 1176. 40. A. Carrington, D.H. Levy and T.A. Miller,J.Chem.Soc., (1966), 45, 4093. 41. P.B. Davies, Β.A. Thrush, A.J. Stone and F.D. Wayne, Chem.Phys.Letters, (1972), 17, 19. 42. T.L. Pollock and W.E. Jones, Canad.J.Chem.,(1973),51, 2041. 43. W.E. Jones and E.G. Skolnik, Chem.Revs.,(1976) (in press). 44. S.N. Foner, Adv.Atomic Mol.Phys.,(1966),2,385. 45. M.A.A. Clyne and R.T. Watson,J.C.S.Faraday I, (1974),70, 1109. 46. M. Kaufman andC.E.Kolb, Chem.Instrumentation,(1971),3,175. 47. V.S. Arutyunov, S.N. Buben and A.M. Chaikin, React.Kinet. Catal.Lett.,(1975), 3, 205. 48. D.K.W. Wang and W.E. Jones, J.Photochem., (1972),1,147. 49. D.E. Milligan, M.E. Jacox, A.M. Bass, J.J. Comeford and D.E. Mann, J.Chem.Phys., (1965), 42, 3187. 50. J.J. Turner andG.C.Pimentel,Science (1963), 140, 974.



1.

APPELMAN

AND

Elementary

CLYNE

Reaction

Kinetics

of

F

25

Atoms

51. J.J. Turner and G.C. Pimentel, in "Noble Gas Compounds", H.H. Hyman, Ed. Univ. of Chicago Press, Chicago (1963), pp 101-105. 52. O. Ruff and Shih-Chang Li, Z. Anorg.Chem.(1939),242, 272. 53. J.G. Malm and Ε.Η. Appelman, Atomic Energy Review, (1969), VII, p.3-48. 54. J. Warnatz, Ph.D. Dissertation, Georg-August Univ., Göttingen (1971). (Discussed in references 43 and 55). 55. R. Foon and M. Kaufman, Progress in Reaction Kinetics, Vol.8, Pt. 2 (1975). 56. Ε.Η. Appelman, Accounts of Chem.Research,(1973),6,113. 57. M.A.A. Clyne and R.T. Watson, Chem.Phys.Letts.,(1971),12,344. 58. H. Gg.Wagner, C.Zetzsch and J. Warnatz, Ber.Bunsenges. Physik.Chem., (1972), 76, 526. see also, Angew.Chem.Intl.Ed. (1971), 10, 564. 59. J. Czarnowski and H.J. Schumacher, Chem.Phys.Letts.,(1972), 17, 235. 60. H.S. Johnston and R. Woolfolk, J.Chem.Phys.,(1964),41,269. 61. A. Arkell, R.R. Reinhard and L.P. Larson, J.Am.Chem.Soc., (1965), 87, 1016. 62. A.E. Douglas and W.E. Jones, Canad.J.Phys.,(1966),44,2251; W.E. Jones, Canad.J.Phys.,(1967), 45, 21. 63. M.A.A. Clyne and I.F. White, Chem.Phys.Letts.,(1970), 6, 465. 64. J.M. Herbelin, D. Spencer and M. Kwok, Report TR-0076(6603)-2 The Aerospace Corporation, (1976), (cited in ref. 65). 65. J.M. Herbelin, Chem.Phys.Letts., (1976), 42, 367. 66. A.H. Curran, R.G. McDonald, A.J. Stone and B.A. Thrush, Proc.Roy.Soc.A,(1973), 332, 355. 67. M.A.A. Clyne and W.S. Nip, (1976) to be published. 68. Recent results according to ref. 67 for k (cm molecule s ) are as follows:- F + CHClF, (1.04 ± 0.22) x 10 ; F + CHClF, (5.3 ± 1.1) x 10 ; F + CHCl , (6.3 ± 1.4) x 10 . Other values for k are:- F +CH ,(7.5 ± 1.1) x 10 ; F +HCl,(1.6 ± 0.4) x 10 . Atomic resonance absorption at 95.48 nm with a non-reversed resonance lamp was used to determine F P atom concentrations. 3

300

-1

-13

-12

2

2

3

300

-11

-11

-1

-12

4

2

J


Elementary Reaction Kinetics of Fluorine Atoms, FO, and NF Free

Recommend Documents