Thermochemistry of organic and heteroorganic species. Part VIII. Ketene and structurally related species

Journal of Molecular Structure 524 (2000) 259–281

www.elsevier.nl/locate/molstruc

Thermochemistry of organic and heteroorganic species.

Part VIII. Ketene and structurally related species^ଝ

a

b,

c

*

A.V. Golovin , D.A. Ponomarev , V.V. Takhistov

^aDepartment of Photonics, Institute of Physics, St. Petersburg State University, 1, Ulyanovskaya str. Petrodvoretz, St. Petersburg 198904, Russia

^bDepartment of Chemical Engineering, St. Petersburg Forest Technical Academy, Institutskii per.5, 194018 St. Petersburg, Russia

^cCentre of Ecological Safety of Russian Academy of Sciences, Korpusnaya 18, 197042 St. Petersburg, Russia

Received 26 April 1999; accepted 15 October 1999

Abstract

Photoionization mass spectrometry was used to obtain the enthalpies of formation of CH₂yCyO Յ Ϫ20:5 and

Ϫ24.0 kcal mol^Ϫ1(from 3-phenylcyclobutanone and diketene, respectively), CH₃CxO^ϩion 147.3 and 152.3 kcal mol^Ϫ1

(from CH₃COOCHyCH₂and CH₃CONH₂molecules, respectively) and PhCH₂CxO^ϩ175.5 kcal mol^Ϫ1(from PhCH₂COOMe

molecules). The enthalpic shift procedure was applied for the estimation of the enthalpies of formation of ketene and related

molecules. The following DH_f⁰values were found CH₂yCyO: Ϫ(22–25), CH₂ySyCH₂(67), CH₂ySyS (60), SySyS (51),

HCxCOH (10 kcal mol^Ϫ1). The low value of DH_f⁰ꢀketene Ϸ Ϫ23 kcal mol^Ϫ1as compared with the currently used value

Ϫ11.4 kcal mol^Ϫ1was supported by the literature data, which have been revised in the present work. Using the new value for

⅐

ketene’s enthalpy of formation, those for ten substituted ketenes and HCxCO (48.5 kcal mol^Ϫ1) free radical were obtained with

the help of macroincremental schemes and introduction of correction terms. Computation of the enthalpies of formation of eight

AyByC molecules by MNDO, AM1, PM3 and MINDO/3 methods revealed that in most cases the latter gives the results closest

to the experimental values or to those gained from the enthalpic shift procedure. The brief analysis of possible sources of errors

in deducing the thermochemical values from appearance energies measurements has been made. Among those the isomeriza-

tion processes occurring in molecular ions are considered the most important, which could lead to incorrect values of the

enthalpies of formation of fragment ions, free radicals and molecules. With many examples it has been demonstrated that the

application of the series of isodesmic reactions could be an effective tool for verifying, correcting and ﬁnding new values of

Keywords: Thermochemistry; Ketene; Isodesmic reactions

Ketene CH₂yCyO is an important species widely

of the processes of combustion of fuels [3] and photo-

lytic reactions [4,5]. Much attention was paid to ther-

mochemistry of ketene and substituted ketenes in

theoretical works [6–13]. The importance of the

knowledge of the DH_f⁰value for ketene and the recent

publication of Aubry et al. [14] on the thermochem-

istry of ketene and methylsubstituted ketenes impelled

us to make: (i) a thorough analysis of the known

experimental data from which the gas enthalpies of

formation DH_f⁰ꢀg for ketene could be obtained; (ii) to

used in organic synthesis [2]. The knowledge of its

⅐

enthalpy of formation and its fragments C₂HO ,

C₂HO^ϩ, C₂O, CH₂is important for the description

ଝ

For part VII see Ref. [1].

* Corresponding author. Tel.: ϩ7-812-555-8798; fax: ϩ7-812-

550-0815.

E-mail addresses: golovin@photonics.phys.spbu.ru

(A.V. Golovin), woodcoal@mailbox.alkor.ru (D.A. Ponomarev).

PII: S0022-2860(99)00452-4

260

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

represent our own experimental and computed data on

RCHyCyO molecules and related species. (The

enthalpy of formation DH_f⁰ꢀ298:15 K (ideal gas) for

all species is substituted in the text by shorter form

DH_f⁰: Another objective of this paper is to ﬁnd out

how reliable are the values of the enthalpies of forma-

tion for neutral and charged species related to the

thermochemistry of ketene if those values are gained

from appearance energies (AEs) measurements by

mass spectrometric methods. The possible sources

of errors in this aspect are discussed. Finally, we

will show that the thermochemical data should be

checked using simple and accessible macroincre-

mental schemes and enthalpic shifts procedures for

molecules and isodesmic reactions for radical cations,

cations and free radicals.

DH_f⁰ꢀketene Ͻ Ϫ22 kcal mol^Ϫ1: Therefore we do

not foresee any chemical reasons for violation of

Eqs. (1) and (2). At the same time C₃O₂is a stable

liquid, safely treated, [6] whereas ketene is a highly

reactive species. Therefore, one can expect difﬁculties

in ﬁnding the experimental value of DH_f⁰for ketene

rather than for carbonsuboxide.

The CO₂molecule also could be used for estimation

of ketene’s enthalpy of formation

DH_f⁰ꢀCH₂yCyO

DH_f⁰ꢀOyCyO ϩ DH_f⁰ꢀCH₂yCyCH₂ : 2 ϩ D

Ϫ94:015 ϩ 45:615 : 2 ϩ D

Ϫ24:2 kcal mol^Ϫ1ϩ D

ꢀ3

D is the correction term revealing the possible addi-

tional stabilization of CO₂caused by the presence of

two oxygen atoms at the carbon atom. Such an effect

is called sigma-conjugation [27], incremental geminal

stabilization [28,29], anomeric effect [30] or

combined polarization [31] and is related to the

compounds possessing several OR, F, or NR₂groups

at the saturated carbon atom. This effect is a small one

(e.g. 1–2 kcal mol^Ϫ1for F-atoms [26]) when those

atoms or groups are bound to the unsaturated carbon

atom. If such an effect exists in the CO₂molecule this

should make the DH_f⁰value for ketene several kilo-

calories more positive (Ϫ21 to Ϫ23 kcal mol^Ϫ1) (Eq.

(3)). However, we take the currently used value

Ϫ11.4 kcal mol^Ϫ1for DH_f⁰(ketene) (see above) and

it will give an improbably high gem-stabilization of

Ϸ 25 kcal mol^Ϫ1in CO₂. Thus, from the application of

macroincremental estimation schemes it comes out

that the currently used value of DH_f⁰(ketene) Ϫ11.4

is doubtful.

1. The enthalpy of formation of ketene is from

structurally related compounds

The currently used value of DH_f⁰(ketene) is

Ϫ11.4 kcal mol^Ϫ1[15–19]. Some reference books

give the value Ϫ14.2 [20,21] or Ϫ14.6 kcal mol^Ϫ1

[22]. The enthalpy of formation of ketene CH₂yCyO

could be calculated by the macroincremental additive

scheme [23–25] from the known DH_f⁰value

Ϫ23.4 kcal mol^Ϫ1[15,16] for carbonsuboxide OyCy-

CyCyO, which is closely related in its structure to

ketene:

DH_f⁰ꢀOyCyCyCyO

2DH_f⁰ꢀCH₂yCyO ϩ DH_f⁰ꢀCH₂yCyCH₂

Ϫ 2DH_f⁰ꢀCH₂yCH₂

ꢀ1

DH_f⁰ꢀCH₂yCyO

DH_f⁰ꢀC₃O₂ Ϫ DH_f⁰ꢀCH₂yCyCH₂

ϩ 2DH_f⁰ꢀC₂H₄ : 2

Earlier, Benson noticed [32] that DDH_f⁰

DH_f⁰ꢀR₂CyO ! R₂CyS or DDH_f⁰ꢀRBr ! RSH

have more or less constant values. Later [25] the

permanency of the DDH_f⁰values for RX ! R⁰X or

RX ! RY replacement was found for many classes

of organic and elementorganic compounds and

systematically used for correcting known or deducing

new values of the enthalpies of formation for different

compounds. Recently, this procedure called the

“enthalpic shift” method [25] was used by some

other workers [1,33–36]. We used Eqs. (4) and (5)

Ϫ23:415 Ϫ 45:615 ϩ 2ꢀ12:515 : 2

Ϫ22:0 kcal mol^Ϫ1

In C₃O₂, the repulsion between two electron-with-

drawing CO-groups is expected (see Ref. [26]),

which is however absent in ketene. Hence,

ꢀ2

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

261

for the estimation of the enthalpy of formation of

ketene (DH_f⁰and DDH_f⁰values in kcal mol^Ϫ1; deduced

values are in square brackets)

(see also Eqs. (20)–(22)) gives Ϫ(21–23) kcal mol^Ϫ1

for DH_f⁰(ketene). The usage of Ϫ11.4 kcal mol^Ϫ1for

the latter gives improbably low DDH_f⁰39 Ϫ

ꢀϪ11:4 50:4 kcal mol^Ϫ1for the yCyO ! yCyS

shift, observed nowhere. Thus, the application of the

enthalpic shift procedure also gives much lower DH_f⁰

values for ketene than is currently accepted

(Ϫ11.4 kcal mol^Ϫ1).

From the schemes similar to Eqs. (4) and (5) we

found DH_f⁰values of some other molecules related to

ketene CH₂ySyCH₂, CH₂ySyS and SySyS (see

later).

ꢀ4

2. DH_f⁰of ketene from some AE measurements

Support for a lower value of DH_f⁰(CH₂yCyO)

can be found in photoionization (PI) studies of

some molecules giving neutral or ionized

ketenes. From fragmentation of 3-phenylcyclobuta-

none (I) we obtained the value Ϫ20.4 kcal mol^Ϫ1for

DH_f⁰(ketene) (Scheme 1) (for details see Appendix

A).

Since IE 8.83 and AE (M–CH₂yCyO) 8.86 eV

are nearly identical if the values of the accuracies

are taken into consideration (0.03 eV for IE and

ꢀ5

The close values for DH_f⁰of ketene obtained from

Eqs. (4) and (5) reveal the absence of the noticeable

stabilization of CO₂due to sigma-conjugation (see

above) as compared with CH₂CO. The value

39 kcal mol^Ϫ1[17] for the thioanalog of ketene

CH₂yCyS gained from the ionization energy (IE) of

CH₂yCyS [38] and DH_f⁰CH₂yCyS^1⅐ϩ[17] combined

with the known value of the enthalpic shift DDH_f⁰for

yCyO ! yCyS replacement 61–62 kcal mol^Ϫ1[32]

⅐ϩ

0.05–0.08 eV for AEs) the thermochemical threshold

⅐ϩ

for [M–CH₂yCyO]

formation could be even

smaller than measured AE, thus shifting DH_f⁰(ketene)

to lower values than Ϫ20.4 kcal mol^Ϫ1

.

Scheme 1.

Scheme 2.

262

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

Scheme 3.

Scheme 4.

Similarly, our present analysis of the thermochem-

The authors claimed [40,41] that in all cases (12

compounds) they obtained ionized phenol according

to Scheme 5. The thermochemistry of PhOH^1⅐ϩand

alkenes is well known and the enthalpies of formation

of initial molecules could be safely estimated.

However, the AEs for M–alkene^1⅐ϩions were found

to be 8–28 kcal mol^Ϫ1higher than expected. The

largest deviation 27.7 kcal mol^Ϫ1was registered for

AE PhOH^1⅐ϩ(ϩC₂H₄) from PhOCH₂CH₂–H. Only

for two molecules, PhOCMe₃and PhOCMe₂Et

AE_exp AE_calcfor PhOH^1⅐ϩϩ alkene. For all other

molecules the observed deviations were explained

[40,41] by the occurrence of excess energy necessary

for the production of the transition state. However

it looks very strange that the elimination of C₂H₄

from PhOCH₂CH¹₃^⅐ϩ-ions requires additional

28(!) kcal mol^Ϫ1whereas the elimination of alkenes

from molecules comprising bulky substituents whose

methyl groups interact with ortho-hydrogens of the

Ph-group takes place without excess energy. We

also mention that the Woodward–Hofmann rule is

not valid for ionized molecules [43,44]. This elimi-

nates the possibility of using this rule for the explana-

tion of high E_cr( E_activation) values. It is notable that

istry of [M–C₂H₂]^⅐ϩion formation from compound II

[38] (Scheme 2) gives DH_f⁰(ketene) Ϫ(20–

23) kcal mol^Ϫ1(see Appendix A).

In our earlier investigation [39] we measured AEs

CHO^Ϫ₂ions from compound IV and showed that frag-

mentation occurs following Schemes 3 and 4.

The values of DH_f⁰bisketene III Ϫ40.0 and

Ϫ37.2 kcal mol^Ϫ1were obtained from the reactions

in Schemes 3 and 4, respectively, which gave DH_f⁰

(CH₂yCyO) Ϫ21.5 and Ϫ20.0 kcal mol^Ϫ1, respec-

tively (for details see Appendix A).

In past years the accurate methods of mass spectro-

metry (MS) (among those PI MS and MS with mono-

energetic electrons) have been applied for

determination of the molecules’ enthalpies of forma-

tion in the gas phase. The possible sources of errors in

such determinations are common with those for ions’

enthalpies of formation. Some of them were discussed

above, the others are considered here in relation to

ketene’s thermochemistry. The sources of errors

discussed here are of general character; therefore,

we analyze them in detail.

Before we analyze the MS data of Aubri et al. [14]

dealing with thermochemistry of ketene eliminated

from PhOCOCH₃molecules, we consider in detail

the process of alkene elimination from structurally

close alkylphenyl ethers [40–42] (Scheme 5).

for the process PhOCOCH¹₃^⅐ϩ! PhOH^1⅐ϩ

ϩ

CH₂yCyO the authors of Ref. [14] do not suggest

the occurrence of excess energy for the process (as

they did for PhOCH₂CH₃molecules) although the

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

263

Scheme 5.

⅐ϩ

transition states for the fragmentation of both mole-

cules PhOCH₂CH₃and PhOCOCH₃are similar.

Our interpretation of the events with PhOR

isomerization and fragmentation is not a new one.

Recently, we analyzed similar cases in detail in

special Account [43].

The authors examining PhOR molecules showed

that collisional activation (CA) spectra of fragment

We suggest the interpretation of the events taking

place with PhOR molecules under EI which differs

from that given in Refs. [14,40,41]. As shown in detail

in Refs. [43,44] all rearrangement processes in a

particular ion have to be analyzed only in conjunction

with the energetically lowest process of simple bond

cleavage in this ion. In PhOCH₂CH₃such a simple

bond cleavage is CH₃-loss which requires ϳ2.5 eV

(ϳ58 kcal mol^Ϫ1) whereas in PhOCMe₃Me-loss

requires only 0.8 eV (ϳ18 kcal mol^Ϫ1) (see Appendix

A). Therefore, the PhOC₂H¹₅^⅐ϩmolecular ion—in the

large energy gap ꢀE_cr 2:5 eV can rearrange

(through H-shifts, ring opening, skeletal rearrange-

ments) to the molecular ion(s) of other structure(s)

possessing higher enthalpy of formation. These new

⅐ϩ

ions C₆H₆O from all 17 compounds [14,40,41]

were identical with the CA spectrum of ionized

⅐ϩ

phenol. This result together with AE (C₆H₆O )

measurements was interpreted in the sense that

C₆H₆O^⅐ϩions from all compounds studied, possessed

the structures of ionized phenol at the threshold of

their formation. Here we express our opinion about

the validity of application of CA spectra to the ther-

mochemistry of ionized and neutral fragments

including ketene.

The identity of the CA spectra does not necessarily

⅐ϩ

isomers when eliminating C₂H₄would give C₆H₆O

conﬁrm the structure of ionized phenol for C₆H₆O

species from PhOR molecules since other C₆H₆O

⅐ϩ

fragments with the structures differing from the

⅐ϩ

expected ionized phenol. These C₆H₆O isomers

isomers (which are abundant) could have similar CA

spectra. To give one_ϩexample, the CA spectra of

CH₃CxO^ϩand CH₂y C OH isomers are similar [45].

However, what is even more important, is that a struc-

turally pure species may, upon CA, isomerize rapidly

will be produced at higher AEs. For the PhOCMe₃

molecule E_crfor Me-loss is only 0.8 eV and the initial

⅐ϩ

M

cannot give isomers with high enthalpies of

formation without decomposing. Consequently, the

ionized phenol is obtained at the expected threshold.

Similar interpretation could be applied to fragmenta-

tion of PhOCOCH₃and PhOCOCHMe₂molecules.

Although simple bond cleavage with lowest energy,

i.e. formation of CH₃CxO^ϩor Me₂CHCxO^ϩions,

respectively, gives close AE values 9.94 and

9.88 eV, respectively, in PhOCOCHMe¹₂^⅐ϩmolecular

and then fragment [46]. Such isomerizations are well

ϩ

documented: HCxO^ϩto H O yC : ions [47], diverse

C₂H₃S^ϩisomers to CH₃CxS^ϩions [48], the skeleton

of many R-fragments changes in pyridinium ions

ϩ

[49,50] C₅H₄N –R ! C₅H₄N : ϩR^ϩ₁: Other exam-

ples can be found in Ref. [46]. Despite the high barrier

⅐ϩ

for HCxCOH^1⅐ϩ! CH₂yCyO

isomerization

ion there appears a possibility of obtaining the iso-

ϩ

79 kcal mol^Ϫ1CA spectrum of ketene contains the

_

merized molecular ion Ph O ꢀHCOCMe₂: This iso-

merized molecular ion is much more stable than its

intense peak of the rearrangement ion with m/z 29

ϩ

ꢀHOC =HCxO whereas the CA spectrum of

⅐ϩ

ϩ

analog Ph O ꢀHCOCH₂(from PhOCOCH¹₃^⅐ϩ). The

_

⅐ϩ

ethynol HCxCOH gives the signal of the peak of

former decomposes to give PhOH^1⅐ϩϩ Me₂CyCyO

the CH₂rearrangement ion [51]. In general, this is

whereas PhOCOCH¹₃^⅐ϩinstead of producing

not surprising. In spite of the differences in the excita-

tion methods used to obtain the EI and CA spectra of

molecules, there are marked similarities between the

two phenomena [46]. It is hardly possible to ﬁnd out

the compounds lacking in their EI mass spectra, the

ϩ

_

Ph O ꢀHCOCH₂species gives an isomer, formed

through the H-shift process to the benzene ring. This

⅐ϩ

isomer after fragmentation could give C₆H₆O ion

differing from PhOH^1⅐ϩ

.

264

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

Scheme 6.

peaks of rearrangement ions [25,43,44] and the wide

occurrence of rearrangement processes under CA is

expected as well. Therefore, the initially formed

[54]. The enthalpy of formation for the HCxCOH

isomer can be safely estimated using the enthalpic

shifts procedure (see above). The systematic applica-

tion of this procedure to CH₂yCHX ! PhX systems

allowed Luo et al. [35] not only to demonstrate the

reliability of such a procedure but to deduce many

unknown data. Similarly, we take six couples of

compounds RCHyCH₂=RCxCH with well-estab-

lished values of the enthalpies of formation (see

Appendix A) and arrive at 40:6 ^ 2 kcal mol^Ϫ1for

RCHyCH₂! RCxCH replacement. Then, taking

the known value Ϫ30:6 ^ 1:1 kcal mol^Ϫ1[55] for

DH_f⁰(CH₂yCHOH) we obtain DH_f⁰ ꢀHCxCOH

Ϫ30:6 ϩ 40:6 10:0 kcal mol^Ϫ1: Using theoretical

values for DDH_f⁰ꢀCH₂yCyO ! HCxCOH35:1 [53]

and 36.3 [54] we get the values Ϫ25.7 and

Ϫ26.9 kcal mol^Ϫ1, respectively, for DH_f⁰ꢀCH₂yCyO:

In the attempt to ﬁnd out the enthalpy of formation

of ketene by PI MS we selected diketene V expecting

the formation of ionized and neutral ketene (Scheme

⅐ϩ

C₆H₆O ions from phenylalkylethers PhOR [40,41]

or phenylesters PhOCOR’ [14] may possess one struc-

ture (or the mixture of structures) whereas in course of

⅐ϩ

the CA process they rearrange to the PhOH isomer

⅐ϩ

as the most stable among the C₆H₆O species. In

general, many problems exist in the interpretation of

the CA spectra [52] so the latter should be applied for

the interpretation of the thermochemical data with

great caution.

3. DH_f⁰(ketene) from computation

The range of the computed values of the enthalpies

of formation for ketene is extremely large (for details

see Appendix A) (kcal mol^Ϫ1):19 [6], Ϫ4.3 [6], Ϫ9.7

[7], Ϫ10.2 [6], Ϫ10.8 [9], Ϫ11.1 [9], Ϫ12.1 [8],

⅐ϩ

6, route a). We obtained IE 9.53 eV and AE (C₂H₂O )

Ϫ12.4 [10], Ϫ12.6 [9], Ϫ17.2 [6], Ϫ24.0 (from

10.98 eV (253.21 kcal mol^Ϫ1). Taking DH_f⁰V Ϫ

45:46 kcal mol^Ϫ1[15,16] and IE ꢀCH₂yCyO

9.614 eV [38] (221.71 kcal mol^Ϫ1), we expected to

get DH_f⁰(ketene) from Eqs. (6)–(8)

computed DH_f⁰CH₂yCyO

[9] and IE [38]

⅐ϩ

CH₂yCyO. What is most disturbing is that in the

same publication, [6] using the same computational

method MP2/6-31G^ءϩ ZPVE//MP2/6-31G^ءءbut

different model reactions, the authors obtained 19,

Ϫ4.3 and Ϫ10.2 (or Ϫ17.2) kcal mol^Ϫ1for DH_f⁰

(ketene). In another publication [13] it was computed

that bisketene III is by 3.2 kcal mol^Ϫ1less stable than

its cyclic isomer IIa. This leads (for details see

Appendix A) to 1.6 kcal mol^Ϫ1for DH_f⁰(CH₂yCyO).

Our computation using MNDO, AM1, PM3 and

MINDO/3 methods gives DH_f⁰(CH₂yCyO) values

⅐ϩ

AEꢀCH₂yCyO

DH_f⁰ꢀCH₂yCyO^⅐ϩϩ DH_f⁰ꢀCH₂yCyO

Ϫ DH_f⁰ꢀV

ꢀ6

ꢀ7

DH_f⁰ꢀCH₂yCyO

AEꢀCH₂yCyO^⅐ϩϪ DH_f⁰ꢀCH₂yCyO

ϩ DH_f⁰ꢀV

⅐ϩ

Ϫ6.83, Ϫ5.71, Ϫ9.27 and Ϫ20.74 kcal mol^Ϫ1

,

respectively, the latest being close to the value

deduced in this work.

The accuracy of computation increases or, better to

say, the deviation from experiment decreases when

the relative values for isomers enthalpies of formation

are computed. The computed values of DDH_f⁰for

CH₂yCyO ! HCxCOH are 35.13 [53] or 36.3

since

DH_f⁰ꢀCH₂yCyO

DH_f⁰ꢀCH₂yCyO ϩ IEꢀCH₂yCyO

⅐ϩ

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

265

DH_f⁰ꢀCH₂yCyO

we ﬁnally have:

DH_f⁰ꢀCH₂yCyO

AEꢀCH₂yCyO^⅐ϩϪ IEꢀCH₂yCyO

ϩ DH_f⁰ꢀV : 2

AEꢀCH₂yCyO^⅐ϩϪ IEꢀCH₂yCyO

Ϫ DH_f⁰ꢀHCxCOH ϩ DH_f⁰ꢀV

253:21 Ϫ 221:71 Ϫ 10:0 Ϫ 45:46

Ϫ23:96 kcal mol^Ϫ1

253:24 Ϫ 221:71 Ϫ 45:46

Ϫ6:98 Ϸ Ϫ7:0 kcal mol^Ϫ1

ꢀ12

for

ꢀ8

There

are

two

theoretical

values

DDH_f⁰ꢀCH₂yCyO ! ꢀHCxCOH 36.0 [51]

⅐ϩ

and 45.2 kcal mol^Ϫ1[58]. Taking DH_f⁰ꢀHCxCOH

⅐ϩ

The latter value appeared to be doubtful since most

of the experimental and theoretical data give much

lower values Ϫ(11–26) kcal mol^Ϫ1(see above). We

suggested, instead—in agreement with our recent

publication [43] on regularities of rearrangement

235 kcal mol^Ϫ1[57] (gained from AE measurements)

we have DH_f⁰ꢀCH₂yCyO^⅐ϩ199 and 189.8 kcal mol^Ϫ1

,

respectively. Subtracting from these values

IEꢀCH₂yCyO 221.7 kcal mol^Ϫ1(9.614 eV [38]) we

obtain Ϫ22.7 and Ϫ32 kcal mol^Ϫ1for the enthalpy of

formation for the CH₂yCyO molecule.

⅐ϩ

processes in ions—that molecular ion V rearranges

⅐ϩ

to Va in the energy gap E_cr AE-IE for route “a”.

⅐ϩ

Since ionized enol Va has smaller DH_f⁰value than

⅐ϩ

Ketene’s enthalpy of formation could be deduced

from the related free radicals thermochemistry. The

measured value of the reaction enthalpy DH_r35:3 ^

2:5 kcal mol^Ϫ1[59] Eq. (13) could be used for estima-

tion of the ketene’s enthalpy of formation

its tautomer V (see Ref. [56]) the isomerization

through route “b” followed by expected fragmen-

⅐ϩ

tation of newly born molecular ion Va is quite

accessible. IEest ꢀHCxCOH is 9.75 eV from DH_f⁰

ꢀHCxCOH 10 kcal mol^Ϫ1

(see

above)

and

i.e.

⅐

_

235 kcal mol^Ϫ1[57] for HCxCOH^1⅐ϩ

;

CH₂yCyO ϩ F ! FCH₂CyO

ꢀ13

Taking DH_f⁰values 19 [17,18] (F ) and

⅐

IEꢀHCxCOH is higher IEꢀCH₂yCyO9:614 eV [38]

and this coincides with route b: according to the

Stevenson–Audier rule the charge resides on the fragment

with lower IE (see Ref. [43]). From Eqs. (9)–(12)

following from Scheme 6 (route “b”) we have

Ϫ43 kcal mol^Ϫ1(FCH₂CyO—see Appendix A) we

_

arrive at the value Ϸ Ϫ26.5 kcal mol^Ϫ1for

DH_f⁰ꢀCH₂yCyO:

DH_f⁰ꢀCH₂yCyO Ϫ 24 kcal mol^Ϫ1

:

4. Ketene’s enthalpy of formation from its proton

⅐ϩ

afﬁnity and DH_f⁰CH₃CxO^ϩion

AEꢀCH₂yCyO

DH_f⁰ꢀCH₂yCyO^⅐ϩϩ DH_f⁰ꢀHCxCOH

Ϫ DH_f⁰ꢀV

One of the possible ways to deﬁne the ketene’s

enthalpy of formation is to use ketene’s proton afﬁnity

(PA) (see later) from the process CH₂yCyO ϩ H^ϩ

!

ꢀ9

CH₃CxO^ϩϩ PA for which the value of

DH_f⁰ꢀCH₃CxO^ϩ should be unambiguously deter-

⅐ϩ

DH_f⁰ꢀCH₂yCyO

mined.

The

currently

used

value

for

DH_f⁰ꢀCH₃CxO^ϩ is 156 kcal mol^Ϫ1[17,60]. The

higher value 156:67 ^ 0:19 kcal mol^Ϫ1is given in

Ref. [61]. In Reference Edition [62] the value 150.5

is recommended. We collected in Table 1 some data

on the enthalpies of formation for C₂H₃O^ϩions

(CH₃CxO^ϩand other isomers) gained from their AEs

measured by accurate methods of PI, PI-photoelectron

coincidence spectroscopy (PIPECO) or electron

DH_f⁰ꢀCH₂yCyO ϩ IEꢀCH₂yCyO

ꢀ10

⅐ϩ

AEꢀCH₂yCyO

DH_f⁰ꢀCH₂yCyO ϩ IEꢀCH₂yCyO

ϩ DH_f⁰ꢀHCxCOH Ϫ DH_f⁰ꢀV

ꢀ11

266

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

Table 1

The enthalpies of formation of C₂H₃O^ϩions from AE measurements (in kcal mol^Ϫ1

)

N

Precursor molecule

Neutral partner (DH_f⁰)

DH_f⁰C₂H₃O^ϩ

Structure of C₂H₃O^ϩion

Reference

a

_

1.

CH₃CO–OCHyCH₂

OCHyCH₂$ CH₂CHO

147.04

CH₃CxO^ϩ

(3.7) [63,64]

CH₃(34.8) [25,64]

⅐

2.

CH₃CO–CH₃

148.48

151.48

152.63

149.6

CH₃CxO^ϩ

[65]

[66]

[67]

[69]

[70]

[69]

_

3.

4.

5.

6.

7.

8.

CH₃CO–CH₂CH₂OH

CH₃CO–OC₂H₅

CH₃CO–COOCH₃

CH₃CO–OH

CH₃CO–NH₂

CH₃CO–C₂H₅

CH₂CH₂OH (Ϫ7.4) [64,68]

_

OC₂H₅ꢀϪ4:1 [63]

151.6

_

COOCH₃(Ϫ40.4) [63]

152.74

153.28

152.3

CH₃CxO^ϩ

[67]

b

_

OH (9.4) [63]

a

_

NH₂(45.1) [63,64]

CH₃CxO^ϩ

⅐

b

C₂H₅(26.5) [64]

153.98

154.44

155.67

150.17

155.46

149.46

145.96

154.77

155.69

157.53

152.1

CH₃CxO^ϩ

[66]

[71]

[69]

CH₃CxO^ϩ

b

CH₃CxO^ϩ

b

_

9.

CH₃CO–COOH

CH₃CO–COCH₃

COOH (Ϫ47.3) [25,64]

⅐

CO₂ϩ H (42.0) [18]

CH₃CxO^ϩ

b

CH₃CxO^ϩ

[61]

_

10.

CH₃CyO (Ϫ5.8) [63]

_

CH₂CHO (0.2) [72]

CH₃CxO^ϩ

_

CH₂CHO (3.7) [73]

CH₃CxO^ϩ

[67]

[38]

⅐

b

11.

12.

CH₃CO–H

H (52.1) [18]

b

CH₃CxO^ϩ

b

⅐

CH₃CO–CxCH

HCxC (138.7) [9]

(122) [64]

CH₃CxO^ϩ

b

169

CH₃CxO^ϩ

ϩ

⅐

13.

14.

15.

16.

17.

CH₂yCHO–H

CH₂yCHO–CH₃

Propyleneoxide

Ethyleneoxide

1,4-dioxane

H

178.55

183

CH₂y C Ϫ OH

[74]

ϩ

⅐

CH₃

CH₂y C OH^b

[75,76]

[65]

ϩ

CH₂y C OH^b

ϩ

⅐

CH₃

185.66

202.29

204.4

191.5

⅐

H

C H₂CHO

[38]

ϩ

_

OC₂H₅

C H₂CHO

ϩ

HCxC O H₂

[77]

⅐

[CH₂yO ϩ Me ] (Ϫ8.8)

a

This work.

b

For these ions the contribution of the ions with higher values of enthalpies of formation is suggested (see text).

Table 2

The enthalpies of formation of C₂H₅CxO^ϩions from AE measurements

Precursor molecule

Neutral partner

DH_f⁰(C₂H₅CxO^ϩ) in

Reference

kcal mol^Ϫ1

C₂H₅CO–CH₃

CH₃

144

142.45

136.45

142

138.25

141.17

Յ 139.67

138.53

138.30

138.09

135.88

[60]

[66]

[78]

[86]

[78]

[38]

[78]

Cyclo-propanol

C₂H₅CO–OCH₃

H

_

CH₃

_

CH₂OH

Cyclopentanol

C₂H₅C(yO)–H

C₂H₅

H

C₂H₅CO–C₂H₅

C₂H₅CO–OH

C₂H₅

OH

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

267

ϩ

impact (EI) with monoenergetic electrons. We recal-

culated and reinterpreted all the data using the experi-

mental AE values (for details see Appendix A). The

introduction by Traeger et al. [61,71,78,79] about the

thermal energy correction term 3–7 kcal mol^Ϫ1to the

AE_expvalues which increases the enthalpies of forma-

tion of fragment ions was questioned by Holmes [80].

We also question, whether or not, ab initio MO calcula-

tions of those correction terms for complex organic ions

can give precise values. The obvious source of errors in

such calculations is the usage of frequencies of parent

molecules, for example, RCHO instead ofRCxO^ϩions,

which are unknown. Therefore, we used the experi-

mental AE values of the authors (Tables 1 and 2).

Several conclusions can be made from analysis of

the data represented in Table 1. For popular molecules

like acetaldehyde CH₃CHO and acetone CH₃COCH₃

there are several values of AEs for the CH₃CxO^ϩion

and hence several values for DH_f⁰ꢀCH₃CxO^ϩ 148.5–

152.6 from acetone and 154.8–155.7 kcal mol^Ϫ1from

CH₃CHO. The variety of the values from acetone can

be explained by differences in sensitivities of the

instruments and methods. It is reasonable to select

the lowest value since the CH₃CxO^ϩion is the most

stable among C₂H₃O^ϩisomers and one cannot explain

the lower value by contribution of the isomers with

lower enthalpies of formation. The higher values

Ͼ153–155 kcal mol^Ϫ1manifest the contribution of

the ions with higher DH_f⁰values to the ion efﬁciency

curve. For example, it was shown that for fragmenta-

tion of the CH₃CO–OH molecule, the CH₂yC^ϩ–OH

isomer contributes to the formation of CH₃CxO^ϩions

[45] thus increasing the DH_f⁰value for the latter by

shifting the threshold of its formation. The minor

contribution of the isomer with higher enthalpy of

formation was noticed for acetone [81].

mated [25]. Other high energy isomers CH₃O yC :

and cyclic oxiranyl ions were considered in Ref.

[45].

Table

1 illustrates the contribution of the

isomers with higher DH_f⁰values than acetyl ion

to the formation of the latter. The correct inter-

pretation of the thermochemical data could be

further complicated by the possible variation in

stoicheometry of the fragmentation processes

which could change the fragment ions DH_f⁰values

and their structures (items 9 and 17). This can

happen, when C₂H₃O^ϩions are formed in either

one-step or two-step processes. An additional difﬁ-

culty in gaining the correct thermochemical data

for fragment ions can arise from possible isomer-

ization of the part of the molecular ion, which

gives free radical. Thus, the production of

⅐

_

CH₃CHOH instead of C₂H₅O (or mixture of

˙

both) (items 4 and 17) or instead of CH₂CH₂OH

(item 3) will increase the resulting value of the

enthalpy of formation for the C₂H₃O^ϩion. Simi-

larly, partial or complete isomerization of the free

radical part in 2,3-butanedione (item 10) will

decrease the DH_f⁰value of the CH₃CxO^ϩion.

One of the possible ways for molecular ions

CH₃COR^1⅐ϩisomerization is its enolization. The

lower value of the DH_f⁰CH₂yC(OH)R^1⅐ϩas

compared with its tautomer CH₃COR^1⅐ϩ[56] is the

driving force for such isomerization. The newly born

molecular ion can produce, after further isomerization,

any of the C₂H₃O^ϩisomers or their mixtures [43].

To verify the value of DH_f⁰CH₃CxO^ϩion we

measured AE(C₂H₃O^ϩ) 9.80 eV from CH₃COO-

CHyCH₂molecules (IE 9.43 eV, DH_f⁰ꢀM Ϫ

75:26 kcal mol^Ϫ1[15,16]] and taking DH_f⁰

⅐

3:7 kcal mol^Ϫ1[63,64] for the CH₂yCHO $

CH₂CHO radical we obtain DH_f⁰ꢀCH₃CxO^ϩ

_

The partial or complete isomerization of the

molecular ions before fragmentation is a well-

known fact. The detailed analysis of the problem

of increasing of experimental AE values and,

hence, the enthalpies of formation of fragment ions

is given elsewhere [43]. The detailed analysis of

experimental and theoretical data made in Ref.

147:04 kcal mol^Ϫ1: If the lower values 3.0 [25] or

0.2 kcal mol^Ϫ1[72] for the OCHyCH₂free radical

_

are applied, then DH_f⁰CH₃CxO^ϩwill shift to 147.7

or 150.5 kcal mol^Ϫ1, respectively.

We also measured by PI MS IE 8.87 eV, AE (M–

⅐ϩ

OMe)^ϩ10.87 eV and AEꢀM Ϫ MeOH Յ 10:97 ^

[75] leads to the following values of the enthalpies

0:1 eV from PhCH₂COOMe molecules. The ion

current for the M–MeOH^1⅐ϩion was very weak so

we could not get reliable values for the PhCHy-

ϩ

of formation for C₂H₃O^ϩisomers: 149.5 (CH₃CxO ),

ϩ

178.5 (CH₂yC OH; 204 kcal mol^Ϫ1ꢀC H₂CHO: In

CyO^1⅐ϩenthalpy of formation. From AE M–OMe^1ϩ

,

ϩ

addition DH_f⁰ꢀHCxC O H₂ 193 kcal mol^Ϫ1was esti-

DH_f⁰(PhCH₂COOMe) Ϫ71.13 (estimated by the

268

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

additive scheme [25,82]) and DH_f⁰(OMe) 4.2 k-

cal mol^Ϫ1[63] we obtained DH_f⁰(PhCH₂CxO^ϩ)

175.33 kcal mol^Ϫ1. When we involve this value in

the isodesmic reaction (Eq. (14)), which reﬂects the

relative stabilities of CH₃CxO^ϩand PhCH₂CxO^ϩ

ions we ﬁnd out that PhCH₂CxO^ϩion is by

Ϸ2.6 kcal mol^Ϫ1more stable than the CH₃CxO^ϩion

if the value 149.5 kcal mol^Ϫ1[25] for the latter is

used. At higher values for DH_f⁰CH₃CxO^ϩion Q

values in Eq. (14) become 9–10 kcal mol^Ϫ1(DH_f⁰

DH_f⁰for the CH₂yCyO molecule than is currently

accepted.

˙

Now we make a brief analysis of the values of the

enthalpies of formation for the RCxO^ϩions (R

alkyl). The known data for the C₂H₅CxO^ϩions are

given in Table 2. We recalculated all the data using

experimental AE values (for details see Appendix A).

The Q values in isodesmic reactions (Eqs. (15)–

(17)) reﬂect the tendencies in the stabilization of the

RCxO^ϩions at CH₃! C₂H₅; C₂H₅! n-Pr and

C₂H₅! ꢀCH₃₂CH replacement (the derived values

are given in square brackets).

and Q in kcal mol^Ϫ1

)

CH₃CxO^ϩϩ PhCH₂CHO

DH_f⁰CH₃CxO^ϩϩ C₂H₅CHO

Ϫ11:31est

149:525

ꢀkcal mol^Ϫ1

15660;17

156:6761

15660;17

149:525

149

Ϫ44:3615

! CH₃CHO ϩ PhCH₂CxO^ϩϩ Q

ꢀ14

! C₂H₅CxO^ϩϩ CH₃CHO ϩ

Q

12:9–9:4

6:4

ꢀ15

ꢀ16

ϩ2:6

ϩ9:1

ϩ9:8

Ϫ39:7315

175:33

138:5–142

138:578

140:5

Ϫ39:7315

3:9

Since the H ! Ph replacement in the RCH₂CxO^ϩ

ion takes place in a non-conjugated and relatively

remote b-position to positive charge, one expects a

small stabilization of the cationic center caused only

by polarizability of the Ph-group. If we compare, for

example, the values of proton afﬁnities PAs of some

bases B: we shall realize that the DPAs values

at H ! Ph replacement in the b-position to the

protonation site, which reﬂect the relative stabilities

C₂H₅CxO^ϩ! n Ϫ C₃H₇CxO^ϩϩ Q

Ϫ0:6

ϩ8:0

ϩ1:4

ϩ9:9

138:578

140:5

134:566

125:9487

134:566

125:987

C₂H₅CxO^ϩ! ꢀCH₃₂CHCxO^ϩϩ Q

ꢀ17

Ϫ0:42

ϩ1:58

2:3

138:5

140:5

131:766

131:7

131

ϩ

of the BH ions are rather small: DPAꢀH–CH₂NH₂!

The low value of DH_f⁰(n-C₃H₇CxO^ϩ) obtained

from PEPICO studies [87] of PrCOOH molecules

could be explained, possibly, by the contribution of

the low energy ion pair formation PrCxO^ϩϩ HO^Ϫ.

Finally, we suggest the following values of the enthal-

pies of formation for acyl ions as a compromise to the

different literature values (Eq. (18)) (DH_f⁰and Q

values in kcal mol^Ϫ1).

Ϫ1

Ph Ϫ CH₂NH₂ 2:7 kcal mol (from PAs 214.1 and

216.8 kcal mol^Ϫ1, respectively [83]) or 3.0 kcal mol^Ϫ1

for H Ϫ CH₂NMe₂! Ph Ϫ CH₂NMe₂replacement

(PAs 225.1 and 228.1 kcal mol^Ϫ1, respectively [83]).

Hence, the experimental value of DH_f⁰(PhCH₂CxO^ϩ)

supports the low (149.5 kcal mol^Ϫ1) value for DH_f⁰

CH₃CxO^ϩ(Eq. (14)).

Taking the DH_f⁰value 149.5 kcal mol^Ϫ1[25] for the

CH₃CxO^ϩion and the known PA values for ketene

CH₂yCyO 192:9 ^ 2 [84], 195:3 ^ 0:7 [85] or

198.0 kcal mol^Ϫ1[17,83] we obtain DH_f⁰(CH₂yCyO)

Ϫ23.3, Ϫ20.9 or Ϫ18.2 kcal mol^Ϫ1, respectively, from

the known equation [83]

The tendencies in stabilization of acyl ions at H !

Me replacement in the b-position to the cationic center

Eq. (18)aresupportedbythetendenciesinthe IEchange

(DIE values) in C₆H₅R, CH₂yCHR and HCxCR

compounds (R Me, Et, n-Pr, Me₂CH) [38,62].

We take notice that ϪDIE ϪIEꢀC₆H₅Me Ϫ

IEꢀC₆H₅Et Q for the corresponding isodesmic

DH_f⁰ꢀCH₂yCyO DH_f⁰ꢀCH₃CxO^ϩ

reaction with radical ions C₆H₅Me^1⅐ϩ! C₆H₅Et^1⅐ϩ

:

Ϫ DH_f⁰ꢀH^ϩ ꢀ365:7 82 kcal mol^Ϫ1

ϩ PAꢀCH₂yCyO

˙

Since the IE of any free radical, i.e. RCyO gives the

main numerical contribution to the total value of

0

ϩ

0

_

Weagain see that the thoroughinspectionofthe thermo-

DH_fR Ϫ C yO DH_fꢀR Ϫ CyO ϩ IER Ϫ

0

chemistry of the CH₃CxO^ϩion leads to a lower value of

CyO the comparison of the tendencies in the DDH_f

_

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

269

ꢀ18

shift at CH₃! C₂H₅! n-C₃H₇(or iso-C₃H₇) repla-

cement at the ionization center in the free radical and

in the neutral molecule is justiﬁed. Those shifts in D

IE are small 0.1–0.2 eV (2–4 kcal mol^Ϫ1) [38,62]

since the substitution takes place at the b-position to

the ionization center. Hence, the value 156 kcal mol^Ϫ1

forthe CH₃CxO^ϩion isdoubtfulsince itgives veryhigh

Q values (9–3 kcal mol^Ϫ1) for b-ꢀH ! CH₃ substitu-

tion in the CH₃CxO^ϩ! C₂H₅CxO^ϩisodesmic

reaction. It is possible that the DH_f⁰(CH₃CxO^ϩ)

ably low value of 18 kcal mol^Ϫ1for the HNyCH₂mole-

cule whereas other workers gained from experiments,

values of 22–26 (see Ref. [90]) or even 32 kcal mol^Ϫ1

[17]. The DH_f⁰Ϫ (23–24) kcal mol^Ϫ1for MeCHyCyO

does not coincide with the value Ϫ23 kcal mol^Ϫ1for

CH₂yCyO, obtained in this work. If Benson’s additive

scheme is applied [82] then we get DH_f⁰

ꢀMeCHyCyO Ϸ Ϫ30:5 kcal mol^Ϫ1based on DH_f⁰

(CH₂yCyO) Ϫ23 kcal mol^Ϫ1

.

Now we consider the thermochemistry of the

Me₂CyCyO molecule. We reinterpreted the data

of Aubry et al. [14] (see Appendix A) and gained

DH_f⁰(Me₂CyCyO) Ϫ37.3 kcal mol^Ϫ1from AE

PhOH^1⅐ϩbeing produced in the process

PhOCOCHMe¹₂^⅐ϩ! PhOH^1⅐ϩϩ Me₂CyCyO: Simi-

larly, from AE Me₂CyCyO^1⅐ϩ(in the process

PhOCOCHMe₂! Me₂CyCyO^1⅐ϩϩ PhOH) and

IE (Me₂CyCyO) 8.38 [17] or 8.45 eV [91] we

value should be decreased to 148–147 kcal mol^Ϫ1

.

Consequently, the DH_f⁰values of other RCxO^ϩions

in Eq. (18) will be shifted down as well.

5. Thermochemistry of substituted ketenes and its

heteroanalogs

The data for DH_f⁰(MeCHyCyO) is a matter of

controversy. Its enthalpy of formation could be

derived from PA(MeCHyCyO) and DH_f⁰(EtCxO^ϩ)

140.5 kcal mol^Ϫ1(see above). Although two publica-

tions give close values of PA MeCHyCyO 201.96

[84] and 201.24 [85] leading to DH_f⁰(MeCHyCyO)

Ϫ23.24 and Ϫ23.96 kcal mol^Ϫ1, respectively, the PA

values for other compounds measured in these works

are doubtful thus making the PA data on MeCHyCyO

questionable. Thus, the measured PA(Me₂CyCH₂)

197.9 kcal mol^Ϫ1is given in Ref. [84] whereas the

recently obtained values are 192.83 [88] and

obtained

DH_f⁰

(Me₂CyCyO)

Ϫ34.10

or

Ϫ35.71 kcal mol^Ϫ1, respectively. We also suggest

⅐ϩ

the different stoicheometry of the Me₂CyCyO

ion formation at its threshold from 2,2,4,4-tetra-

methylcyclobutanedione-1,3 (VI) which coincides

⅐ϩ

with CO elimination from the molecular ion VI

[14] (Scheme 7).

Taking AE(Me₂CyCyO ) 9.41 eV [92] from

⅐ϩ

compound VI and IEs 8.38 [17] or 8.45 eV [91] for

Me₂CyCyO we have DH_f⁰(Me₂CyCyO) Ϫ36.31 or

Ϫ37.92, respectively. Finally, we take DH_f⁰

(Me₂CyCyO) Ϫ37 kcal mol^Ϫ1

.

191.68 kcal mol^Ϫ1

[89].

The

PA

value,

205.5 kcal mol^Ϫ1for HNyCH₂[85], combined with

DH_f⁰(C^ϩH₂NH₂) 178 kcal mol^Ϫ1[17] gives the improb-

Now we consider in detail the effect of Me-substi-

tution in ketene. There are two sets of data on DDH_f⁰

Scheme 7.

270

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

Table 3

releasingMeO-groupdestabilizessuchacharge.Methyl

is an electron-releasing group, therefore, in alkanes

lacking the atoms with high electronegativities (ENs),

The thermochemistry of substituted alkenes at H ! Me; (DH_f⁰in

kcal mol^Ϫ1

)

the DDH_f⁰ꢀH ! Me shift is only Ϫ5 kcal mol^Ϫ1

.

DH_f⁰

12.55 [16]

DDH_f⁰

[15,16,82] as compared, with DDH_f⁰Ϫ 7:77 [15,16]

or Ϫ10.35 kcal mol^Ϫ1[15,16] at CH₂yCH₂!

MeCHyCH₂and HCxCH ! MeCxCH replacement,

respectively, owing to the higher ENs of atoms in sp₂

and sp hybridization states. We think that computa-

tion [10] correctly reveals these tendencies at least

qualitatively giving low Ϫ(4–6) kcal mol^Ϫ1for

DDH_f⁰values at H ! Me substitution in ketene. It is

obvious that Me would destabilize the negative charge

in ketene. For example, compare electron afﬁnities

(EAs) [17] 1.83(OH) and 1.58(OMe) or 2.31(SH)

and 1.87 eV (SMe). Taking Ϫ23 and Ϫ37 kcal mol^Ϫ1

for DDH_f⁰values for CH₂yCyO and Me₂CyCyO

CH₂yCH₂

CH₂yCHBr

CH₂yCHCHO

CH₂yCHCOOH

MeCHyCH₂

Z–MeCHyCHBr

E–MeCHyCHBr

Z–MeCHyCHCHO

E–MeCHyCHCOOH

18.93 [16]

Ϫ16.84 [25]

Ϫ79.04 [94]

4.78 [16]

Ϫ7.77

Ϫ9.18

Ϫ8.70

Ϫ9.38

Ϫ9.03

9.75 [16]

10.49 [16]

Ϫ26.22 [94]

Ϫ88.07 [94]

values in

DDH_f⁰ꢀ1

DDH_f⁰ꢀ2

CH₂yCyO ÿÿÿ! MeCHyCyO ÿÿÿ! Me₂CyCyO

molecules. Aubry et al. [14] and Deming et al. [93]

advocate the high values for DDH_f⁰(1) Ϫ9.8 [14] or

Ϫ13.6 [93] and DDH_f⁰(2) Ϫ10.0 [14] or

Ϫ12 kcal mol^Ϫ1[93], respectively. Scott et al. [10]

computed Ϫ3.85 and Ϫ5.75 kcal mol^Ϫ1, respectively,

for DDH_f⁰(1) and DDH_f⁰(2). Our analysis of thermo-

chemistryofmethylatedketenesandrelatedcompounds

gives the following results. We found a few but quite

clear examples that H ! Me substitution in alkenes

possessing strong electron-withdrawing substituents

provokes larger stabilization than in alkenes lacking

such substituents (Table 3).

Electron withdrawing groups (Br, CHO, COOH)

induce the positive charge at the terminal C_datom.

Therefore Me group response is higher than at the

C_datom in molecules lacking such groups, i.e.

MeCHyCH₂. SinceyCyO in ketene is a strong elec-

tron-withdrawing group the authors of Refs. [14,93]

suggest an even stronger response of the Me group in

ketenes than in alkenes.

molecules, respectively (see above) and followi_ꢀn₂g

ꢀ1

the tendencies in CH₂yCH₂! MeCHyCH₂!

Me₂CyCH₂DDH_f⁰ꢀ1 and DDH_f⁰ꢀ2 are Ϫ7.77

[15,16] and Ϫ8.78 kcal mol^Ϫ1[15,16], respectively]

we get DH_f⁰ꢀMeCHyCyO Ϫ29.5 kcal mol^Ϫ1with

Ϫ6.5 and Ϫ7.5 kcal mol^Ϫ1for DDH_f⁰ꢀ1 a_ꢀn₁d

DDH_f⁰ꢀ2;

respectively,

MeCHyCyO ! Me₂CyCyO substitution.

for

CH₂yCyO !

ꢀ2

Taking the new value for DH_f⁰(CH₂yCyO)

Ϫ23 kcal mol^Ϫ1we estimated the enthalpies of forma-

tion for some substituted ketenes (for details see

Appendix A). We used both macro- and microincre-

mental (Benson’s) schemes introducing minor correc-

tion terms for repulsion of electron-withdrawing

halogen and the CO-group. These correction terms

were estimated using the isodesmic reactions like

Eq. (19) (DH_f⁰values—from Refs. [15,16]).

2F₂CyCH₂! F₂CyCF₂ϩ CH₂yCH₂

DH_f⁰ꢀkcal mol^Ϫ1

2×ꢀϪ80:4

Ϫ173:3 calc:

Ϫ157:5 exp:

12:5

However, the structure of ketene and its substituted

analogs differs drastically from that of alkenes [95]:

ꢀ19

Eq. (19) displays the destabilization of Ϫ157:5 Ϫ

ꢀϪ173:5 16 kcal mol^Ϫ1owing to the repulsion

between two CF₂-groups. The electron-withdrawing

properties of CO- and CF₂-groups are similar:

compare ENs (covalent potentials) 2.83 and 2.84 for

the OCH- and F₂CH-groups, respectively [96]. Hence,

taking the CF₂-group as a model for the CO-group we

used the known values of the enthalpies of formation

for X(Y)CyCH₂and X(Y)CyCF₂ethylenes for

ϩ

ꢀ

R₂CyCyO $ R₂C– C yO : $ R₂C–CxO

VII

VIIa

VIIb

The contribution of resonance forms VIIa $ VIIb to

ketene’s molecule VII was supported by IR frequencies

shifts in (4–XC₆H₄)₂CyCyO molecules, where

X MeO, H or CN and by some other data [95]. The

electron-withdrawing CN group stabilizes the negative

charge in the VIIa $ VIIb form whereas the electron-

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

271

calculation of the enthalpies of formation for target

X(Y)CyCyO molecules.

23 kcal mol^Ϫ1we have only Ϫ12 kcal mol^Ϫ1for

H ! F replacement: compare DDH_f⁰Ϫ45 and

Ϫ47 kcal mol^Ϫ1for CH₂yCH₂! FCHyCH₂!

F₂CyCH₂substitution, respectively (from Refs.

[15,16]). We do not see any chemical reasons

for such high destabilization of ﬂuorinated ketenes

as compared with ﬂuorinated ethylenes. For example,

repulsion in tetraﬂuoroethylene C₂F₄destabilizes the

molecule as compared with F₂CyCH₂only by

16 kcal mol^Ϫ1(Eq. (19)).

When we compare the DH_acidvalues (deprotonation

enthalpies) for PhCH₃(381), Ph₂CH₂(361.5), CH₂F₂

(390), CH₃Cl (396), CH₂Cl₂(374),CH₃Br (393) and

CH₃I (386.5 kcal mol^Ϫ1) (all data—from Ref. [17])

we see that most of the compounds are very weak

acids and the participation of the resonance form

VIIa is hardly expected.

For Ph₂CyCyO and Cl₂CyCyO we introduced

the small Ϫ(1–1.5) correction terms reﬂecting the

possible contribution of the VIIa $ VIIb form.

Finally, we obtained the following new values of

the enthalpies of formation DH_f⁰(in kcal mol^Ϫ1):

For unknown reasons the computation of the

enthalpies of formation for ﬂuorinated molecules

very often gives unreliable results. In addition to

the examples considered above we represent such

a simple molecule as FCxCH. The computation of

its enthalpy of formation gives the values 24.7,

(25.3) [97] or 25.3 kcal mol^Ϫ1[98] whereas the

enthalpic shift procedure from either FCHyCH₂or

HCxCOH molecules using DDH_f⁰ꢀRCHyCH₂!

PhCHyCyO

FCHyCyO

ClCHyCyO

(Ϫ0.2),

(Ϫ60.0),

(Ϫ22.5),

Ph₂CyCyO

F₂CyCyO

Cl₂CyCyO

(21.5),

(Ϫ100.5),

(Ϫ25),

BrCHyCyO (Ϫ11), ICHyCyO (2). The values

for phenylated ketenes differ from earlier calcu-

lated values 6 [93] or 17.3 kcal mol^Ϫ1[11] for

RCxCH ϳ 40

[15,16]

or

DDH_f⁰ꢀROH !

PhCHyCyO

and

29 kcal mol^Ϫ1

[93]

for

RFϳ Ϫ4 kcal mol^Ϫ1[36] leads to the values 6.5 or

Ph₂CyCyO molecules. Our values for FCHyCyO

and F₂CyCyO values differ drastically from

recently computed values [7] Ϫ35.2 and

Ϫ69.4 kcal mol^Ϫ1, respectively. If we take the

authors’ values for ketene Ϫ12.4 (or Ϫ9.7) we

get DDH_f⁰shifts Ϫ22.8 (or Ϫ25.5) and

Ϫ34.2 kcal mol^Ϫ1for CH₂yCyO ! FCHyCyO !

F₂CyCyO substitution. Taking DH_f⁰ꢀCH₂yCyO Ϫ

6 kcal mol^Ϫ1, respectively. The reference book [99]

gives DH_f⁰(FCxCH) 10.0 kcal mol^Ϫ1

.

From the value Ϫ23 kcal mol^Ϫ1for DH_f⁰

(ketene) the unknown values of the enthalpies of

formation for some structurally related compounds

could be derived (Eqs. (20)–(22)) with the help of

the enthalpic shift procedure (see Eqs. (4) and (5))

(DH_f⁰and DDH_f⁰—in kcal mol^Ϫ1

)

the derived

ꢀ20

ꢀ21

272

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

values are in square brackets):

6. Thermochemistry of ketene’s fragments

From the known values of the deprotonation enthal-

pies DH_acidfor PhCH₃and HCxCCH₃380.7 [17] and

381.8 kcal mol^Ϫ1, respectively (the latter value was

calculated from DH_acid380.5 kcal mol^Ϫ1[17] for

allene CH₂yCyCH₂and EA 0.893 eV for

_

CH₂yCyCH $ CH₂CxCH free radical-from Ref.

[17]) we suggest the similar DH_acidvalues for PhOH

and HCxCOH molecules. Taking the DH_acidfor

PhOH 349.2, 350.4 kcal mol^Ϫ1[17] we estimated

DH_acidꢀHCxCOH ϳ 350 kcal mol^Ϫ1: The electron-

withdrawing properties of HCxCϪ as compared

with the Ph-group are higher (compare ENs [96]

3.00 and 2.74, respectively, s_p, [100] 0.23 and

Ϫ0.01 or F(ﬁeld) [100]—constants 0.22 and 0.12,

respectively). However, for negative ions the

exclusively high importance of the polarizability

effect [101,102] (which is larger for Ph than for the

HCxC-group) makes DH_acidvalues for PhXH

ꢀ22

It is noteworthy that the ySyO ! ySyS substitution

gives nearly identical enthalpic shift DDH_f⁰with that for

yCyO ! yCyS substitution of 61–62 kcal mol^Ϫ1

.

yCy ! ySy substitution gives 21–23 kcal mol^Ϫ1

enthalpic shift. The latter could be more certain if the

lower value Ϫ24 kcal mol^Ϫ1is taken for DH_f⁰

(CH₂yCyO). It is noteworthy that the linear molecule

SySyS is less stable than its cyclic isomer S₃

(DH_f⁰34 ^ 2 kcal mol^Ϫ1[17]). In Table 4 we represent

our computed data on DH_f⁰values for ketene and its

heteroanalogs. We see that AyByC molecules are yet

difﬁcult for computation. Among the computational

methods being tried, MINDO/3 appeared to be the

most reliable. It is demonstrated that the enthalpic

shift procedure could be an alternative to computation.

and

HCxCXH

nearly

identical.

Taking

DH_acidꢀHCxCOH 350 kcal mol^Ϫ1we have DH_f⁰

ꢀHCxCO^Ϫ DH_acidꢀHCxCOH Ϫ DH_f⁰ꢀH^ϩ Ϫ

DH_f⁰ꢀHCxCOH

350

Ϫ

365:7ϩ 10:0

Ϫ5:7 kcal mol^Ϫ1and combining the latter value

Ϫ1

_

with EA, ꢀHCxCO 2:350 eV [17] (54.2 kcal mol

)

Table 4

Computed values of the enthalpies of formation DH_f⁰(gas) of ketene and its heteroanalogs (in kcal mol^Ϫ1

)

Molecule

Method

MNDO

Experimental

AM1

46.05

Ϫ5.71

Ϫ79.85

Ϫ47.03

(107.9Њ)

28.22

PM3

46.98

Ϫ9.27

Ϫ85.08

Ϫ50.79

(106.08Њ)

31.20

MINDO/3

H₂CyCyCH₂

H₂CyCyO

OyCyO

43.87

Ϫ6.83

Ϫ75.08

4.43^b

41.89

Ϫ20.74

Ϫ95.73

Ϫ70.72

(130.6Њ)

Ϫ57.81

(180ЊC)

Ϫ50.61

Ϫ0.01

45.5 [15,16]

Ϫ23^a

Ϫ94.1 [82]

Ϫ70.9 [82]

OySyO

(106.8Њ)^b

120.26

58.85

OySyO (linear, 180Њ)

OySyO (with ﬁxed angle)

Ϫ70.9

3.05

0.99

(154.4Њ)

Ϫ22.91

44.71

(158.6Њ)

Ϫ28.98

9.94

(156.9Њ)

Ϫ23.77

14.55

SyCyO (linear)

SySyO

Ϫ34 [82]

Ϫ10 [82]

H₂CySyO

H₂CySyCH₂

SySyS

29.56

61.81

73.69

Ϫ8.92

37.47

32.16

Ϫ2.42

52.10

44.01

ϩ14.03

82.04

73.1

Ϫ1.9 [37]

60^a

51^a

(114.5Њ)

110.98

(180Њ)

32.16

(180Њ)

44.01

(180Њ)

SySyS (ﬁxed, 180Њ)

73.1

a

This work.

The central angle ЄAyByC is given.

b

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

273

0

⅐

_

we get DH_fꢀHCxCO $ HCyCyO Ϫ5:7 ϩ

O-centered free radicals RO (R H, Me, Ph) (Eq.

54:2 48:5 kcal mol^Ϫ1: The literature values for

(23)) are less stable than C-centered free radicals

RCH₂owing to the larger electronegativity of the

0

⅐

_

DH_fꢀHCxCO $ CHyCyO are 28:8 ^ 2:0 [103] or

42:2 ^ 0:7 kcal mol^Ϫ1[104]. To make the choice for

oxygen as compared with carbon. Therefore, O-

centered radicals extract larger stabilization from the

same substituent, here Me or Ph, than the more stable

C-centered radicals (compare the Q values 6.0 and 14.9

or 12.0 and 15.8 kcal mol^Ϫ1). The detailed discussion of

Eq. (23)was necessary before weintroduce to the reader

0

_

the DH_fꢀHCxCO value among the quoted ones

we use the isodesmic reactions (Eqs. (20)–(22)).

We would like to stress that the submillimeter-

wave spectra, [105] of species obtained by two reac-

⅐

tions HCxCH ϩ O ! HCxCO and H₂CyCyO ϩ

⅐

the estimation procedure of DH_f⁰ꢀHCxCO $

_

F ! HCyCyO are identical supporting the struc-

⅐

_

ture of the C₂HO free radical as CHyCyO $

HCyCyO free radicals, which is given in Eqs. (24)

⅐

HCxCO :

and (25) (DH_f⁰values of free radicals in kcal mol^Ϫ1are

given below the formulae and Q values—at the arrows;

the derived values are in square brackets)

In our recent publication [64] the general princi-

ples and rules of quantiﬁcation of structure–stability

relationships for free radicals using series of

isodesmic reactions were represented. Using many

examples it was demonstrated that the thermodyna-

mically less stable free radical extracts a larger stabi-

lizing effect from the same substituent (say, Me, Ph,

HCxC) than the more stable one. For example (from

Ref. [64] (Eq. (23)) (DH_f⁰values of free radicals in

kcal mol^Ϫ1are given under the formulae, Q values—

at the arrows):

ꢀ24

ꢀ23

ꢀ25

The isodesmic reactions are given in the short form.

The negative Q⁰⁰⁰value Ϫ2.5 kcal mol^Ϫ1for the

⅐

For example, CH₃! HO ꢀQ Ϫ14:6 kcal mol^Ϫ1 is

PhCH₂! HCxCCH₂isodesmic reaction reveals the

well-known fact that the HCxC-group is an electron-

withdrawing group as compared with Ph. For example,

s_p-constants are ϩ0.23 and Ϫ0.01 for HCxC-and Ph-

groups, respectively, and R-(resonance) constants are

0.01 and Ϫ0.13, respectively [100].Therefore one

⅐

the short form of the isodesmic reaction CH₃ϩ

0

_

0

_

H₂O ! CH₄ϩ OHϩQ; and hence, Q DH_fꢀCH₃ ϩ

0

_

DH_fꢀH₂OϪDH_fꢀCH₄ϪDH_fꢀOH 34:8Ϫ57:8 Ϫ

ꢀϪ17:8 Ϫ 9:4 Ϫ14:6 kcal mol^Ϫ1

: We use the

short form of the isodesmic reaction, omitting

the molecules whose structures are obvious. The

positive sign ꢀQ Ͼ 0 reveals stabilization, i.e.

⅐

expects that Q⁰Ͻ 0 (Eq. (24)) as well for the PhO !

⅐

HCxCO isodesmic reaction. The negative Q⁰⁰value of

⅐

MeCH₂is by 6 kcal mol^Ϫ1more stable than the

Ϫ4.6 kcal mol^Ϫ1for HCxCO ! HCxCCH₂is

doubtful because is skeptical whether the O-centered

free radical is more (by 4.6 kcal mol^Ϫ1) stable than the

_

⅐

CH₃free radical whereas the negative sign

⅐

manifests destabilization, i.e. HO is by

14.6 kcal mol^Ϫ1less stable thermodynamically than

C-centered free radical (compare with Eq. (23)). The

⅐

0

value 48.5 kcal mol^Ϫ1for DH_fꢀHCxCO is further

_

the CH₃free radical in the framework of isodesmic

reactions.

supported by a series of isodesmic reactions (Eq. (26))

274

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

(DH_f⁰and Q in kcal mol^Ϫ1) (the data from Ref. 64)

Ϫ15 kcal mol^Ϫ1for the HCxCCH^ϩ₂! HCxCO^ϩ

isodesmic reaction if DH_f⁰values 281 and 262 kcal

mol^Ϫ1, respectively, are taken for these ions [60].

ꢀ26

7. Conclusion

From our own experimental data and thorough

analysis of literature sources, most of which have

not been earlier considered in relation to ketene’s

thermochemistry, the new value DH_f⁰ꢀCH₂yCyO ϳ

Ϫ23 kcal mol^Ϫ1is obtained as compared with the

currently used Ϫ11.4 kcal mol^Ϫ1[15–18]. This

lower value was gained from the thermochemical

properties of about 20 different species (molecules,

free radicals, ions). Unfortunately, the process of

getting the correct value for the enthalpy of formation

for the CH₂yCyO molecule by computation appeared

to be a difﬁcult task—the computed values range

from ϩ19 to Ϫ32 kcal mol^Ϫ1(from about ten litera-

ture sources analyzed). Our computation for DH_f⁰

⅐

It is obvious that the DH_f⁰ꢀHCxCO value

28.8 kcal mol^Ϫ1[103] is doubtful since it gives an

improbably high positive Q value of ϩ17.6 for the

⅐

Ph ! HCxCO isodesmic reaction.

From DH_f⁰ꢀHCxCO^Ϫ$ HCyCyO Ϫ 5:4 (see

ꢀ

above) and the new DH_f⁰value Ϫ23 kcal mol^Ϫ1for

CH₂yCyO molecules we get DH_acidꢀH Ϫ CHyCyO

DH_f⁰ꢀHCxCO^Ϫ ϩ DH_f⁰ꢀH^ϩ Ϫ DH_f⁰ꢀCH₂yCyO

Ϫ5:7 ϩ 365:7 Ϫ ꢀϪ23 383 kcal mol^Ϫ1(compare

with the earlier found value 365 kcal mol^Ϫ1[17]). The

resonance stabilization of the anions of the type

HCxCO^Ϫ$ HCyCyO was experimentally demon-

(ketene) ranges from Ϫ6.8 to Ϫ20.7 kcal mol^Ϫ1

.

ꢀ

strated in our previous work [106]. The problem, still

remains with the enthalpy of formation value for

carbene: CH₂. The range of known values is 60–

100 kcal mol^Ϫ1[107,108]. Althoughthe valueofenergy

It is demonstrated that the enthalpic shift procedure

and macroincremental scheme for molecules, and the

usage of isodesmic reactions for cations, radical

cations and free radicals (elaborated in our recent

publications and in the works of some other

workers—see text) could be treated as the general

methodology of verifying and correcting the known

and ﬁnding the new values of the enthalpies of forma-

tion for both charged and neutral species. An addi-

tional advantage of application of isodesmic

reactions is that the well-known properties of the

molecules or substituents like s-constants, ionization

energies, proton afﬁnities, ENs and polarizabilities, IR

frequencies and some other characteristics could be

used for estimation of the enthalpies of these reactions

and consequently for estimation of the enthalpies of

formation of different species.

The brief analysis of possible sources of errors in

getting the correct thermochemical data from AE

measurements is made. Among those sources the

most important (and mostly often ignored) is the

concealed isomerization processes taking place in

molecular ions and leading to the reaction products

(either charged or neutral) with the structures differing

from the fragments of initial molecule [43,111]. The

rearrangement of molecular ion is a very serious

gap between singlet ¹A and triplet B₁carbenes CH₂is

3

determined by spectroscopy methods with high accu-

racy [109] 9:023 ^ 0:014 kcal mol^Ϫ1; the value of the

enthalpy of formation of either carbene is still under

question. Whatever is the accurate determination of

the bond dissociation energy in ketene CH₂yCyO !:

CH₂ϩ CO ꢀD₀ 86:1 kcal mol^Ϫ1[4]) DH_f⁰ꢀ: CH₂

still depends on the value of the enthalpy of formation

for the former. Taking the value DH_f⁰ꢀCH₂yCyO Ϫ

23 kcal mol^Ϫ1suggested in the present work and

using the data from [109,110] we obtain

DH_f⁰ꢀ298 K : CH₂ꢀ A₁ and :CH₂(³B₁) 90.5 and

1

81.5 kcal mol^Ϫ1, respectively (the data for OK were

recalculated for 298 K using Refs. [109,110]).

Earlier it was found [43] that both free radicals and

evenelectronionsrevealsimilartendenciesintheirstabi-

lities at R₁! R₂replacement at the free radical and

cationic centers R₁A ! R₂A and R₁A^ϩ! R₂A^ϩ;

respectively, although the absolute Q values in both

systems are obviously different. For example, the desta-

⅐

_

bilizationofHCxCO ascomparedwithHCxCCH₂free

radical (Eq. (25)) is supported by the negative Q value

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

275

artifact in EI, PI and CA processes. In the past 30 years

we have measured about 350 AE values from more than

200 compounds by PI MS [112,113] but could extract

reliable thermochemical data from no more than 50–60

values. The reason is that the experimentally found

enthalpies offormation of the fragment ions were incon-

sistent withthe expectedstructuresof either ions or free

radicals. Such inconsistency was established for each

fragment with the help of isodesmic reactions.

its MS spectra the common acetyl ion CH₃CxO^ϩ. Simi-

larly, no traces of PhCH₂COOH, as a possible source of

the PhCH₂CxO^ϩion were observed in GC. EI mass

spectra were run on the Russian mass spectrometer

MH-1303. The PI mass spectra and AEs measurements

were performed on a modiﬁed PI mass spectrometer

MS-1302 (Russia). Earlier experiments (see, for

examples Refs. [33,36,43,44,75,94,106,112,113,119,

121]) on about 200 compounds indicated that the

accuracy of AE values is better than ^0.08 eV.

The recently appeared communication [121]

presents the following DH_f⁰values (in kcal mol^Ϫ1):

CH₃CHyCyO (Ϫ16.25), (CH₃)₂CyCyO (Ϫ22.0),

CH₃CH₂CxO^ϩ(147.7), (CH₃)₂CHCxO^ϩ(139.3),

and PAs values: CH₂yCyO (196.0), CH₃CHyCyO

(201.2) and (CH₃)₂CyCyO (204.3). These justify

once more the necessity of analysis of experimental

and theoretical thermochemical values, related to

ketene which was performed in this publication.

Therefore, it was not surprising that the structure of

⅐ϩ

the M Ϫ CH₂yCyO ion from the PhOCOCH₃

molecule did not possess the expected structure of

⅐ϩ

PhOH as suggested by the authors [14] or that frag-

mentation of many CH₃COR molecules did not lead to

the structurally pure CH₃CxO^ϩion but rather to the

mixture of C₂H₃O^ϩisomers (see text). Systematic appli-

cation of the enthalpic shift procedure and macroincre-

mental schemes for molecules and the series of

isodesmic reactions for ions and free radicals combined

with our own and literature experimental data allowed

us to get many new or corrected values of the enthalpies

offormation for about 20 species, among those (DH_f⁰in

kcal mol^Ϫ1): MeCHyCyO (Ϫ29.5), Me₂CyCyO

(Ϫ37), PhCHyCyO (0), Ph₂CyCyO (21.5),

FCHyCyO (Ϫ60), F₂CyCyO (Ϫ100.5), ClCHyCyO

(Ϫ22.5), Cl₂CyCyO (Ϫ25), BrCHyCyO (Ϫ11),

ICHyCyO (2), (OyCyCH)₂(Ϫ43), CH₂ySyCH₂

(67), CH₂ySyS (60), SySyS (51), HCxCOH (10),

Acknowledgements

Financial support of Foundation INTAS Ref. No.

96-1990 (V.V.T.) and Foundation “Russian Universi-

ties” Grant No. 3538 (D.A.P. and V.V.T.) are grate-

fully acknowledged. One of us (A.V.G.)

acknowledges the ﬁnancial support from INTAS-

RFBR (grant IR-97-471). The authors highly

appreciate the assistance of Mrs T.Yu. Grigor’eva in

preparation of ﬁnal text of the manuscript.

⅐

ϩ

_

HCxCO (Ϫ48.5), FCH₂CyO (Ϫ43), CH₃CxO

(149), PhCH₂CxO^ϩ(175.5). The new values 383

and 350 for kcal mol^Ϫ1DH_acid(H–CHyCyO) and

HCxCO–H molecules, respectively, were derived.

Appendix A

The enthalpies of formation of the molecules exam-

ined or used in isodesmic reactions are taken from Refs.

[15,16]. In addition, other literature sources were used

(DH_f⁰in kcal mol^Ϫ1): CH₂yCHCxCH (69.2 [114]),

PhCxCH (78.4[114]), C₂H₅F (Ϫ62.9[36]), FCH₂CHO

(Ϫ78.7[36]), CH₂yCHOH(Ϫ30.6[55])dimethylenecy-

clobutene (80.2[115]). For DH_f⁰(ClCHyCH₂) we took a

theoretical value 5.2 [8] rather than an experimental one

of 8.9 kcal mol^Ϫ1[15,16] since it was shown [1,26] that

the latter value is incorrect. Benson’s microincremental

scheme using group contributions [25,82] was applied

for estimation of enthalpies of formation of the

following molecules: CH₃COCxCH (9.8), CH₃COCH_2-

CH₂OH (92.7), PhCH₂COOCH₃(Ϫ71.1). For more

8. Experiment

The purchased compounds methyl ester of

phenylacetic acid PhCH₂COOCH₃, vinylacetate

CH₃COOCHyCH₂, acetamide CH₃CONH₂, and

synthesized compounds 3-phenylcyclobutanone (I)

and diketene (V) (details of synthesis are given in

Ref. [33]) were distilled before AE measurements.

The purity of the compounds was controlled by

GC and then by MS (EI and PI) methods. Mass

spectra did not contain any noticeable peaks of

impurities. The compounds CH₃COOCHyCH₂and

CH₃CONH₂were specially controlled by GC in the

possible presence of acetic acid which could give in

276

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

ꢀ

ꢁ

complex compounds their enthalpies of formation were

estimated using either the macroincremental scheme or

the latter combined with the microincremental scheme

(Eqs. (A1)–(A10)) (group contributions were taken

from Refs. [25,82]).

DH_f⁰

ꢀIII

DH_f⁰ p-MeOC₆H₄COOH Ϫ DH_f⁰ꢀC₆H₆

ϩ DH_f⁰ꢀ furane

Ϫ108:01 Ϫ 19:74 Ϫ 8:34 Ϫ136:09 kcal mol^Ϫ1

The enthalpies of formation for BrCHyCH₂,

ICHyCH₂and cyclo-C₃H₅CHyCH₂were estimated

using the observation, [25,35] that DH_f⁰ꢀXCHyCH₂

DH_f⁰ꢀXPh Ϫ 7:2 kcal mol^Ϫ1: Taking 25.2 [15,16],

39.4 [15,16] and 36.0 kcal mol^Ϫ1[117] for PhBr, PhI

and cyclo-C₃H₅Ph molecules, respectively, we obtain

DH_f⁰values 18.0, 32.2 and 28.8 kcal mol^Ϫ1for

XCHyCH₂molecules, where X Br, I and cyclo-

C₃H₅, respectively. The DH_f⁰value for ClCHyCH₂

molecules (5.2 kcal mol^Ϫ1) calculated using DH_f⁰

(ClC₆H₅) 12.4 kcal mol^Ϫ1[15,16] supports the theore-

tical value [8] (see above). The experimental DH_f⁰value

31.2 kcal mol^Ϫ1[116] for cyclo-C₃H₅CHyCH₂is not in

consensus with DH_f⁰(cyclo-C₃H₅Ph) 36.0 kcal mol^Ϫ1

[117] and was replaced by the estimated value

ꢀA2

ꢀ

ꢁ

DH_f⁰

ꢀ

ꢁ

DH_f⁰

Ϫ 2C_dꢀH Ϫ 2C_dꢀC_d₂

2

Ϫ 2C_dꢀHꢀC_d ϩ 2COꢀC_dꢀCO

ϩ 2C_dꢀHꢀCO

29 kcal mol^Ϫ1

.

DH_f⁰

(exp.) cyclo-C₃H₅CxCH

80:2 Ϫ 2ꢀ6:26 Ϫ 2ꢀ4:6 Ϫ 2ꢀ6:78

ϩ 2ꢀϪ27:0 ϩ 2ꢀ6:0 2:9 ϳ 3 kcal mol^Ϫ1

69.8 kcal mol^Ϫ1[116], used here coincides with a

suggested here value for DH_f⁰for cyclo-C₃H₅CHyCH₂.

We calculated the values of the enthalpies of

formation for ﬂuorinated and chlorinated ketenes

X(Y)CyCyO using the known values for DH_f⁰

X(Y)yCF₂molecules, [15,16]. The application of

the additive scheme for the estimation of DH_f⁰

(FCHyCyO) (Eq. (A6)) gives the deviation from the

experimental value 9.1 kcal mol^Ϫ1similar estimation

scheme for ClCHyCyO gives a deviation of

7.7 kcal mol^Ϫ1. They reﬂect the repulsion between

halogen and the CO-group. We extrapolated those

values for BrCHyCyO and ICHyCyO ϳ 6.5 and

5 kcal mol^Ϫ1, respectively, and using the estimation

scheme similar to Eq. (A6) and the correction terms

we obtained DH_f⁰Ϫ11 and 2 kcal mol^Ϫ1for

BrCHyCyO and ICHyCyO molecules.

ꢀA3

DH_f⁰PhO Ϫ COCHMe₂

DH_f⁰PhOCOMe ϩ DH_f⁰MeOCOCHMe₂

Ϫ DH_f⁰MeOCOMe

Ϫ66:85 Ϫ 115:13 Ϫ ꢀϪ98:45

Ϫ83:43 kcal mol^Ϫ1

ꢀA4

DH_f⁰PhOCHMe₂

DH_f⁰Me₂CH–OCHMe₂ Ϫ 2CꢀH₃Ϫ

0

_

DH_fꢀCH₃COCOOCH₃

Ϫ CꢀHꢀC₂ꢀO Ϫ OꢀC ϩ OꢀCꢀC_B

2

DDH_f⁰ꢀCH₃COCOCH₃

ϩ C_BꢀO ϩ 5C_BꢀH

ϩ DH_f⁰ꢀCH₃OCOCOOCH₃ :2

Ϫ76:29 Ϫ ꢀϪ20:16 Ϫ ꢀϪ7:2 Ϫ ꢀϪ23:2

Ϫ 23:0 Ϫ 1:5 ϩ 16:5

ꢀϪ78:18 Ϫ 169:43 : 2 Ϫ123:8 kcal mol^Ϫ1

ꢀA1

Ϫ33:73 kcal mol^Ϫ1

ꢀA5

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

277

DH_f⁰FCHyCF₂

DH_f⁰FCHyCH₂ ϩ DH_f⁰CH₂yCF₂

Ϫ DH_f⁰CH₂yCH₂

Eq. (A5) needs comments. To take into account the

steric interaction in PhOCHMe₂, PhOCMe₃and

similar compounds the estimation scheme was

suggested [25] based on equalizing the steric demands

of certain groups. For example, it was shown [118]

that Ph- and Me₂CH-groups possess similar “sizes”.

Taking as the model, compounds Me₂CHOCHMe₂

and Me₂CHOCMe₃(DH_f⁰:Ϫ76.29 [15,16] and

Ϫ85.47 kcal mol^Ϫ1[15,16], respectively) for

PhOCHMe₂and PhOCMe₃molecules, respectively,

the enthalpies of formation for the latter were esti-

33:46 Ϫ 80:40 Ϫ 12:55 Ϫ126:41

DH_f⁰FCHyCF₂_exp Ϫ117:35 kcal mol^Ϫ1

DDH_f⁰ Ϫ117:35 Ϫ ꢀϪ126:41 ϩ9:06 kcal mol^Ϫ1

;

ꢀA6

mated Ϫ33.73 (Eq. (A5)) and Ϫ42.91 kcal mol^Ϫ1

,

respectively. Since in bisketene (III) the polar inter-

action between the electron-withdrawing CO-groups

(destabilizing the molecule) is expected we introduce

a small correction term ϳ2 kcal mol^Ϫ1like for other

electron-withdrawing groups [25] (Eq. (A10)). Using

estimation schemes similar to Eqs. (A1)–(A10)

the following DH_f⁰values were obtained (see also

text) (in kcal mol^Ϫ1): PhCH₂CHO (Ϫ11.3), cyclo-

DH_f⁰FCHyCyO

DH_f⁰FCHyCF₂ ϩ DH_f⁰CH₂yCyO

Ϫ DH_f⁰CH₂yCF₂

Ϫ117:35 Ϫ 23:0 Ϫ ꢀϪ80:40

Ϫ60:0 kcal mol^Ϫ1

ꢀA7

propanol (Ϫ28.3), 3-phenylcyclobutanone (6.1).

⅐ϩ

DH_f⁰ꢀPhCHyCH₂

(Scheme 1) was calculated

DH_f⁰ClCHyCyO

DH_f⁰ClCHyCF₂ ϩ DH_f⁰CH₂yCyO

Ϫ DH_f⁰CH₂yCF₂ Ϫ 79:7 Ϫ 23:0 Ϫ ꢀϪ80:4

Ϫ22:3 kcal mol^Ϫ1

using the following data: DH_f⁰ꢀPhCHyCH₂

⅐ϩ

DH_f⁰ꢀPhCHyCH₂ ϩ IEꢀPhCHyCH₂ 35:3515 ϩ

195:3338 ꢀ8:47 eV 230:68 kcal mol^Ϫ1: From AE

11:2 ^ 0:05 eV (258.3 kcal mol^Ϫ1) [38] for [M–

C₂H₂]

ion from compound II (DH_f⁰Ϫ

⅐ϩ

29:37 kcal mol^Ϫ1[15,16]) we gain DH_f⁰[M–C₂H₂]

⅐ϩ

(A8)

174.7 kcal mol^Ϫ1. DH_f⁰IIa ꢀScheme 2 DH_f⁰ꢀIIϩ

⅐ϩ

IEꢀII 3:0 (see Eq. (A3)) ϩ225:8 kcal mol^Ϫ1ꢀ9:79

38 eV 228:8 which is much higher than the

experimental value 174.7 kcal mol^Ϫ1(from AE).

For estimation of IE bisketene III (Scheme 2) we

use the principles and estimation schemes described

in detail in Ref. [43]. We have to estimate DIE for the

H ! CHyCyO replacement in H–CHyCyO, whose

IE value is 9.614 eV [38]. The OyCyCH-group could

be qualiﬁed as a group with dual, electron-releasing or

electron-withdrawing, properties in its inﬂuence on

IEs of molecules like X CH₂Cl, COR, CxCH, CN

[43] groups. Such groups increase IEs of the mole-

cules H–R (i.e. destabilize their radical cations)

possessing low IEs and decrease IEs of H–R

molecules ꢀH–R ! X–R; i.e. stabilize their radical

cations if these molecules possess the relatively high

IE values. For example, H ! CHO replacement in

H–CHO (IE 10.88 [38] eV) decreases IE (CHO–

CHO) to 10.52 eV [38], whereas IE slightly rises

DH_f⁰Ph₂CyCyOyDH_f⁰Ph₂CyCH₂

ϩ DH_f⁰CH₂yCyO Ϫ DH_f⁰CH₂yCH₂ ϩ D

58:70 Ϫ 23:0 Ϫ 12:55 Ϫ 1:5ꢀsee text

ϳ21:5 kcal mol^Ϫ1

ꢀA9

DH_f⁰OyCyCHCHyCyOꢀIII

2DH_f⁰ꢀCH₂yCyO Ϫ 2DH_f⁰C_dꢀH

2

ϩ 2C_dꢀHꢀC_d ϩ D

2DH_f⁰CH₂yCyO Ϫ 2ꢀ6:26 ϩ 2ꢀ6:78 ϩ D

2DH_f⁰ꢀCH₂yCyOϪ2ꢀ6:26 ϩ 2ꢀ6:78

ϩ 2ꢀsee text

2DH_f⁰ꢀCH₂yCyOϩ3 kcal mol^Ϫ1

(A10)

278

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

when coming from PhH (9.24 [38]) to PhCHO (9.4–

9.6 eV [38]). Compare also IEs H–CMe₂CHO (9.705

[38]) and CHO–CMe₂CHO (9.80 eV [38]).

using Eq. (A10) we derived DH_f⁰(CH₂yCyO) Ϫ21.5

and Ϫ20.0 kcal mol^Ϫ1, respectively.

0

˙

For the estimation of DH_fFCH₂CyO free radicals

The comparison of neutral groups ENs [96]yCH₂

(2.68), CHyCH₂(2.70), CHO (2.83), CHyCyO

(3.12), yO (3.73) reveals the slightly higher

electron-withdrawing properties of –CHyCyO as

compared with the CHO-group. Taking in considera-

tiontheIEsdatamentionedandIE(CH₂yCyO)9.60 eV

we expect that H ! CHyCyO replacement in ketene

will give IE (OyCyCH–CHyCyO) 9.5–9.7 eV. Close

values could be derived from published data [12] on IEs

of Me₃SiCHyCyO (threshold at ϳ8.6 eV) and

OyCyC(SiMe₃)– (Me₃Si)CyCyO (threshold at

ϳ8.0 eV). H–CHyCyO ! Me₃Si–CHyCyO replace-

ment gives DIE ϳ 1.0 eV, then substitution of two

H-atoms in bisketene OyCyCHCHyCyO (III) by two

strong electron-releasing Me₃Si-groups should give

DIE Ͼ 1:0 eV: Hence, IEꢀIII Ͼ ꢀ8:0 ϩ 1:0; i.e.

Ͼ9.0 eV and the value 9.4–9.7 eV looks quite reason-

able. If we take DH_f⁰(CH₂yCyO) Ϫ11.4 [17] and DH_f⁰

(to use in Eq. (13)) we take the series of isodesmic

reactions (Eq. (A12)) (DH_f⁰and Q values—in

kcal mol^Ϫ1

)

ꢀA12

Since FCH₂CH₂is 2.7 kcal mol^Ϫ1less stable than the

_

⅐

CH₃CH₂free radical, the former will extract a slightly

⅐

_

larger stabilization at FCH₂–CH₂! FCH₂–CyO

replacement, i.e. Q₂Ͼ Q₁(see Eqs. (23)–(25)). Taking

an arbitrary value Q₂ϳ 13.5 kcal mol^Ϫ1we get

0

Ϫ1

_

⅐ϩ

(bisketene) (III ) 174.7 kcal mol^Ϫ1(see above) and

DH_fFCH₂CyO Ϫ 43 kcal mol

:

We revised the interpretation of data on thermo-

chemistry of fragmentation of the CH₃COCH₂CH₂OH

molecule [69]. Taking DH_f⁰(CH₃CxO^ϩ) 156 kcal

using the following equation

⅐ϩ

DH_f⁰ꢀOyCyCHCHyCyO

0

mol^Ϫ1

,

the authors obtained DH_fꢀCH₂CH₂OH

_

DH_f⁰ꢀOyCyCHCHyCyOꢀIII

Ϫ13:8 kcal mol^Ϫ1from AE (CH₃CxO^ϩ) 10.20 eV.

_

The suggested value Ϫ13.8 for the CH₂CH₂OH

ϩ IEꢀOyCyCHCHyCyOꢀIII

ꢀA11

0

_

radical is close to DH_fꢀCH₃CHOH Ϫ 15:2 kcal

mol^Ϫ1[63] whereas the former should be less stable

than the latter one: (i) the secondary free radicals are

more stable than the isomeric primary ones; (ii) the

HO-group in the a-position stabilizes the free radical

center whereas in the b-position it destabilizes such

we

obtain

IEꢀIII DH_f⁰ꢀIII^⅐ϩϪ DH_f⁰ꢀIII

DH_f⁰ꢀIII^⅐ϩϪ 2DH_f⁰ꢀCH₂yCyO Ϫ 3(Eq. (A10))

174:7 Ϫ 2ꢀϪ11:4 Ϫ 3 194:5 kcal mol^Ϫ1or

8.43 eV. The latter value is obviously too low for IE

(III) (see above). Taking DH_f⁰(CH₂yCyO)

Ϫ23 kcal mol^Ϫ1we have IE (III) 217.7 kcal mol^Ϫ1or

9.44 eV.

⅐

center. For example, (Eq. (A13)) the HOCHyCH free

radical in the framework of the isodesmic reaction is

2.6 kcal mol^Ϫ1less stable than the unsubstituted vinyl

radical

For Schemes 3 and

4 we made the fol-

⅐

lowing calculations: DH_f⁰ꢀHCOO^Ϫ DH_f⁰ꢀHCOO

⅐

ϪEAꢀHCOO

Ϫ35:525

Ϫ

74:517

⅐

DH_f⁰CH₂yCH ϩ HOCHyCH₂

Ϫ110:0 kcal mol^Ϫ1and DH_f⁰ꢀ⅐COOHyDH_f⁰ꢀ⅐COOH

ꢀkcal mol^Ϫ1

70:564

Ϫ30:655

_

ϪEAꢀCOOH Ϫ47:364 Ϫ 7:2 Ϫ 54:4 kcal

mol^Ϫ1

:

We took EAꢀ·COOH ഡ EAꢀ·CHO

⅐

! CH₂yCH₂ϩ HOCHyCH ϩ Q

ꢀA13

0:313 eVꢀ7:22 kcal mol^Ϫ1[17]). Taking AEs 0.9 eV

30:0119

Ϫ2:65

12:5515

(20.8 kcal mol^Ϫ1) and 3.4 eV (78.4 kcal mol^Ϫ1) for

0

Ϫ1

HCOO^Ϫand COOH ions, respectively (Schemes 3

Earlier [25], DH_f(CH₂CH₂OH Ϫ 7:0 kcal mol

ꢀ

_

and 4) and DH_f⁰(Me ) 34.8 kcal mol^Ϫ1[64] we

was estimated using Eq. (A13). This value was

⅐

obtained DH_f⁰(OyCyCHCHyCyO) Ϫ40.1 and

Ϫ37.3 kcal mol^Ϫ1, respectively. From these values,

later supported by computation, [68] which gave

0

DDH_f⁰ϩ 7:8 kcal mol^Ϫ1for DDH_fꢀCH₃CHOH !

_

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

279

DH_f⁰

we

(CH₃CHOHϪ

The small Q⁰value could be explained by the

absence of direct conjugation of the Ph group with a

C-cationic center. Here the Ph-substituent displays

_

CH₂CH₂OH:

Taking

15:2 kcal

mol^Ϫ1

[63]

obtain

DH_f⁰

Ϫ1

_

(CH₂CH₂OH Ϫ 7:4 kcal mol (Eq. (A14))

only the polarizability effect. Since the

DH_f⁰CH₃CH₂ϩ CH₃CH₂OH

_

ϩ

CH₃O yCMe₂ion is much (by ϳ27 kcal mol^Ϫ1

)

ꢀkcal mol^Ϫ1

26:564

Ϫ56:2415

ϩ

more stable than the CH₃O yCH₂ion then

Q⁰⁰Ͻ Q⁰: We take an arbitrary value Q⁰⁰ϳ O: Earlier

it was found that in very stable cations, Ph could

behave even as a weak electron-withdrawing group.

_

! CH₃CH₃ϩ CH₂CH₂OH ϩ Q

ꢀA14

ϩ4:1

Ϫ2:1

Ϫ2:5

Ϫ20:0215

Ϫ13:869

Ϫ7:468

Ϫ7:025

0

Taking Ϫ7.4 kcal mol^Ϫ1for the DH_fCH₂CH₂OH

free radical as a partner for the CH₃CxO^ϩion from

the CH₃COCH₂CH₂OH molecule we obtain

(CH₃CxO^ϩ) 149.6 kcal mol^Ϫ1(Table 1).

For _ϩexample, Q Ϫ0:7 kcal mol^Ϫ1for the

_

ϩ

CH₃N HyCH₂! Ph N HyCH₂isodesmic reaction

ϩ

[120]. From the isodesmic reaction CH₃O yCMe₂!

ϩ

Ph O yCMe₂(Eq. (A18)) we obtain DH_f⁰Ph O yCMe₂

We calculated DH_f⁰(CH₂yCyO) values from the

computed DE values for the following reactions (from

Ref. [13])

ϩ

141.5

and

AEPh O yCMe₂ 218:2 kcal mol^Ϫ1

(9.46 eV) from the PhOCMe₃molecule. Taking

IE(PhOCMe₃) 8.66 eV [38] we obtain E_cr 9:46 Ϫ

8:66 0:80 eV for Me elimination. E_crfor Me-elim-

ination from the PhOCH₂–CH₃molecule is 10.66

[AE(M–Me)^ϩ] [120] Ϫ8.20 [IE(PhOEt)] [38]

2.46 eV.

CH₂yCyO ϩ H^ϩ! CH₃CxO^ϩϩ DE

Ϫ199:2 kcal mol^Ϫ1

ꢀA15

ꢀA16

ꢀA17

CH₂yCyO ϩ HO^Ϫ! CH₃COO^Ϫϩ DE

Ϫ106:3 kcal mol^Ϫ1

References

[1] I.N. Domnin, D.A. Ponomarev, V.V. Takhistov, Zh. Org.

Khim. 35 (1999) 35 (in Russian).

[2] A.D. Allen, T.T. Tidwell, J. Org. Chem. 64 (1999) 266.

[3] A.N. Hayhurst, H.R.N. Jones, J. Chem. Soc., Faraday Trans.

2 83 (1987) 1.

[4] E.D. Potter, M. Gruebele, L.R. Khundkar, A.H. Zewail,

Chem. Phys. Lett. 164 (1989) 463.

[5] T.J. Wallington, J.C. Ball, A.M. Straccia, M.D. Hurley, E.W.

Kaiser, M. Dill, W.F. Schneider, Int. J. Chem. Kinet. 28

(1996) 627.

[6] A. Allen, J.D. Colomvakos, I. Eagle, R.A. Liu, J. Ma, R.M.

Marra, M.A. McAllister, T.T. Tidwell, Can. J. Chem. 74

(1996) 457.

[7] D.R. Burgess, M.R. Zachariah, W. Tsang, P.R. Westmore-

land, Prog. Energy Combust. Sci. 21 (1996) 453.

[8] L.A. Curtis, K. Raghavachari, P.C. Redfern, J.A. Pople, J.

Chem. Phys. 106 (1997) 1063.

CH₂yCyO ϩ H₂O ! CH₃COOH ϩ DE

Ϫ41:4 kcal mol^Ϫ1

Taking DH_f⁰(H^ϩ) 365.7 [83], CH₃COOH

(Ϫ103.44), OH^Ϫ(Ϫ33.2), CH₃COO^Ϫ(Ϫ120.5),

H₂O (Ϫ57.8) [17] and CH₃CxO^ϩ(156 [17] or 149

(this work)) kcal mol^Ϫ1we obtain DH_f⁰(CH₂yCyO)

Ϫ10.2 or Ϫ17.2 from Eq. (A15), ϩ19 (Eq. (A16)) and

Ϫ4.3 kcal mol^Ϫ1(Eq. (A15)).

ϩ

The DH_f⁰Ph O yCMe₂ion was estimated using the

series of isodesmic reactions (Eq. (A18)) (DH_f⁰and Q

values—in kcal mol^Ϫ1

)

[9] K. Raghavachari, B.B. Stefanov, L.A. Curtiss, J. Chem.

Phys. 106 (1997) 6764.

[10] A.P. Scott, L. Radom, Int. J. Mass Spectrom. Ion Processes

160 (1997) 73.

[11] N.H. Werstiuk, H.M. Muchall, J. Ma, M.T.H. Liu, Can. J.

Chem. 76 (1998) 1162.

[12] N.H. Werstiak, J. Ma, M.A. McAllister, T.T. Tidwell, D.-c.

Zhao, J. Chem. Soc. Faraday Trans. 90 (1994) 3383.

[13] M.A. McAllister, T.T. Tidwell, J. Am. Chem. Soc. 116

(1994) 7233.

ꢀA18

280

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

[14] C. Aubry, J.L. Holmes, J.K. Terlouw, J. Phys. Chem. 101

(1997) 5958.

[40] M.C. Blanchette, J.L. Holmes, F.P. Lossing, Org. Mass Spec-

trom. 24 (1989) 673.

[15] J.B. Pedley, R.D. Naylor, S.P. Kirby, Thermochemical Data

of Organic Compounds, 2, Chapman and Hall, London,

1986.

[16] J.B. Pedley, Thermochemical Data and Structures of Organic

Compounds, 1, TRC, College St, TX, 1994.

[17] S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, R.D.

Levin, W.G. Mallard, Gas Phase Ion and Neutral Thermo-

chemistry, J. Phys. Chem. Ref. Data 17 (1988) Suppl 1.

[18] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics

CRC Press, Boca Raton, 1997.

[41] D. Harnish, J.L. Holmes, J. Am. Chem. Soc. 113 (1991)

9729.

[42] J.C. Traeger, T.H. Morton, J. Am. Chem. Soc. 118 (1996)

9661.

[43] V.V. Takhistov, D.A. Ponomarev, Org. Mass Spectrom. 29

(1994) 395.

[44] V.V. Takhistov, D.A. Ponomarev, Eur. Mass Spectrom. 1

(1995) 107.

[45] P.C. Burgers, J.L. Holmes, J.E. Szuleiko A. Mommers,

J.K. Terlouw, Org. Mass Spectrom. 18 (1983) 254.

[46] R.G. Cooks (Ed.), Collisional Spectroscopy, Plenum Press,

New York, 1978, p. 458.

[47] P.C. Burgers, J.L. Holmes, A.A. Mommers, J. Am. Chem.

Soc. 107 (1985) 1099.

[19] M. Frenkel, K.N. Marsh, R.C. Wilhoit, G.J. Kabo, G.N.

Roganov, Thermodynamics of Organic Compounds in the

Gas Phase, 1–2, Thermodynamics Research Center, College

St. TX, 1994.

[48] M.N. Eberlin, T.K. Majumdar, R.G. Cooks, J. Am. Chem.

Soc. 114 (1992) 2884.

[49] A.R. Katritzky, C.H. Watson, Z. Dega-Szafran, J.R. Eyler, J.

Am. Chem. Soc. 112 (1990) 2471.

[50] A.R. Katritzky, C.H. Watson, Z. Dega-Szafran, J.R. Eyler, J.

Am. Chem. Soc. 112 (1990) 2479.

[51] B. Van Baar, T. Weiske, J.K. Terlouw, H. Schwarz, Angew.

Chem. Int. Ed. 25 (1986) 282.

[20] J.D. Cox, G. Pilcher, Thermochemistry of Organic and

Organometallic Compounds, Academic Press, London,

1970.

[21] R. Atkinson, D.L. Baulch, R.A. Cox Jr., R.F. Hampson, J.A.

Kerr, J. Troe, J. Phys. Chem. Ref. Data 18 (1989) 881.

[22] T.E. Daubert, R.P. Danner, Physical and Thermodynamic

Properties of Pure Chemicals, Taylor and Francis,

Washington, 1994.

[52] R. Stolze, H. Budzikiewicz, Tetrahedron 37 (1981) 781.

[53] R.D. Brown, E.H.N. Rice, M. Rolder, Chem. Phys. 99 (1985)

347.

[23] J.D. Cox, Method for Estimating the Enthalpies of Formation

on Benzene Derivatives in the Gas Phase, NPL Rep. Chem.

(1978) 83.

[54] W.J. Bouma, R.H. Nobes, L. Radom, C.E. Woodward, J.

Org. Chem. 47 (1982) 1869.

[24] H.M. Rosenstock, J. Dannacher, J.F. Liebman, Radiat. Phys.

Chem. 20 (1982) 7.

[55] F. Turecek, Org. Mass Spectrom. 27 (1992) 1087.

[56] J.L. Holmes, F.P. Lossing, J. Am. Chem. Soc. 102 (1989)

1591.

[57] J.M. Buschek, J.L. Holmes, F.P. Lossing, Org. Mass

Spectrom. 21 (1986) 729.

[58] W.J. Bouma, P.M.W. Gill, L. Radom, Org. Mass Spectrom.

19 (1984) 610.

[59] J.E. Bartmess, Mass Spectroscopy Rev. 8 (1989) 297.

[60] F.P. Lossing, J.L. Holmes, J. Am. Chem. Soc. 106 (1984)

6917.

[61] J.C. Traeger, B.M. Kompe, Thermochemical data for free

radicals from studies of ions, in: J.A. Martinho Simoes,

A. Greenberg, J.F. Liebman (Eds.), Energetics of Organic

Free Radicals, Blackie Academic, London, 1996, pp. 59–

109.

[62] H.M. Rosenstock, K. Draxl, B.W. Steiner, J.T. Herron, J.

Phys. Chem. Ref. Data 6 (1977) Suppl 1.

[63] D.F. McMillen, D.M. Golden, Annu. Rev. Phys. Chem. 33

(1982) 493.

[64] D.A. Ponomarev, V.V. Takhistov, J. Mol. Struct. 435 (1997)

259.

[65] L. Bombach, J.-P. Stadelmann, J. Vogt, Chem. Phys. 72

(1982) 259.

[25] V.V. Takhistov, Organic Mass Spectrometry, Nauka, Lenin-

grad, 1990 Russ.

[26] D.A. Ponomarev, V.V. Takhistov, J. Chem. Educ. 74 (1997)

201.

[27] M.J.S. Dewar, J. Am. Chem. Soc. 106 (1984) 669.

[28] W.R. Doblier, K.S. Medinger, Tetrahedron 38 (1982) 2415.

[29] J.-D. Wu, W. Kirmse, K.N. Houk, J. Am. Chem. Soc. 112

(1990) 4557.

[30] H.-D. Beckhaus, B. Dogan, S. Verevkin, J. Hadrich, C.

Ruchardt, Angew. Chem. Int. Ed. 29 (1990) 320.

[31] J. Klein, Tetrahedron 44 (1988) 503.

[32] S.W. Benson, Chem. Rev. 78 (1978) 23.

[33] V.M. Orlov, A.A. Krivoruchko, A.D. Misharev, V.V.

Takhistov, Isv. Acad. Nauk SSSR (Russ.) (1985) 2619.

[34] S.T. Graul, R.R. Squires, J. Am. Chem. Soc. 112 (1990)

2517.

[35] J.R. Luo, J.L. Holmes, J. Phys. Chem. 96 (1992) 9568.

[36] V.V. Takhistov, T.A. Pashina, N.G. Ismagilov, V.M. Orlov,

A.A. Rodin, V.G. Barabanov, Russian J. Org. Chem. (Engl.

Transl.) 31 (1995) 1582.

[37] G. Bouchoux, J.J. Salpin, J. Am. Chem. Soc. 118 (1996)

6515.

[38] R.R. Levin, S.G. Lias, Ionization Potential and Appearance

Potential Measurements, 1971–1981, US Government

Printing Ofﬁce, Washington, 1982.

[66] E. Murad, M.G. Ingram, J. Chem. Phys. 40 (1964) 3263.

[67] J.C. Traeger, R.G. McLoughlin, A.J. Nickolson, J. Am.

Chem. Soc. 104 (1982) 5318.

[68] L.A. Curtis, D.L. Lucas, J.A. Pople, J. Am. Chem. Soc. 102

(1995) 3292.

[39] V.V. Takhistov, M.V. Muftakhov, A.A. Krivoruchko, V.A.

Mazunov, Izv. Acad. Nauk SSSR (ser. Khim.)(Engl. Transl.)

(1991) 1812.

A.V. Golovin et al. / Journal of Molecular Structure 524 (2000) 259–281

281

[69] J.L. Holmes, F.P. Lossing, P.M. Mayer, J. Am. Chem. Soc.

113 (1991) 9723.

[97] J.W. Ochterski, G.A. Petersson, K.B. Wiberg, J. Am. Chem.

Soc. 117 (1995) 11 299.

[98] B.J. Persson, P.R. Taylor, J.M.L. Martin, J. Phys. Chem. A

102 (1998) 2483.

[70] L. Frazer-Monteiro, M.L. Frazer-Monteiro, J.J. Butler, T.

Baer, J. Phys. Chem. 86 (1982) 752.

[99] L.V. Gurvich, I. Veyts, C.B. Adcock (Eds.), 4, Thermody-

namic Properties of Individual Substances, 2, Hemisphere,

New York, 1991, p. 2.

[100] C. Hanch, A. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165.

[101] R.W. Taft, J.L.M. Abboud, F. Anvia, M. Berthelot, M. Fujio,

J.-F. Gal, A.D. Headley, W.G. Henderson, I. Koppel, J.H.

Qian, M. Mishima, M. Taagepera, S. Ueji, J. Am. Chem. Soc.

110 (1988) 1797.

[102] P. Burk, I.A. Koppel, Theor. Chim. Acta 86 (1993) 417.

[103] J.L. Holmes, F.P. Lossing, Int. J. Mass Spectrosc. Ion

Processes 58 (1984) 113.

[104] D.L. Osborn, D.H. Mordaunt, H. Choi, R.T. Bise, D.M.

Neumark, C.M. Rohling, J. Chem. Phys. 106 (1997) 10087.

[105] Y. Endo, E. Hirota, J. Chem. Phys. 86 (1987) 4319.

[106] M.V. Muftakhov, Yu.V. Vasil’ev, V.A. Masunov, V.V.

Takhistov, D.A. Ponomarev, J. Mass Spectrom. 30 (1995)

275.

[71] J.C. Traeger, M.J. McAdoo, J. Mass, Spectrosc. Ion

Processes 68 (1986) 35.

[72] J.L. Holmes, F.P. Lossing, J.K. Terlouw, J. Am. Chem. Soc.

108 (1986) 1086.

[73] M. Rossi, D.H. Golden, Int. J. Chem. Kinet. 11 (1979) 715.

[74] F. Turecek, V. Hanus, Org. Mass Spectrom. 19 (1984) 423.

[75] V.V. Takhistov, D.A. Ponomarev, A.D. Misharev, V.M.

Orlov, K. Pihlaja, J. Gen. Chem. (Russ.) 64 (1994) 110.

[76] J.L. Holmes, F.P. Lossing, J. Am. Chem. Soc. 102 (1980)

3732.

[77] M.L. Frazer-Monteiro, L. Frazer-Monteiro, J.J. Butler, T.

Baer, J. Phys. Chem. 86 (1982) 739.

[78] J.C. Traeger, Org. Mass Spectrom. 20 (1985) 223.

[79] J.C. Traeger, Org. Mass Spectrom. 24 (1989) 230.

[80] J.L. Holmes, Org. Mass. Spectrom. 20 (1985) 169.

[81] P.C. Burgers, J.L. Holmes, C.E.C.A. Hop, J.K. Terlouw, Org.

Mass. Spectrom. 21 (1986) 545.

[107] H.Gg. Wagner, F. Zabel, Ber. Bunzen. Gesellsch. 75 (1971)

114.

[108] D.A. Ponomarev, V.V. Takhistov, J. Mol. Struct. 477 (1999)

91.

[109] P.R. Bunker, P. Jensen, W.P. Kraemer, R. Beardswarth, J.

Chem. Phys. 85 (1986) 3724.

[82] S.W. Benson, Thermochemical Kinetics, New York, Wiley,

1976.

[83] S.G. Lias, J.F. Liebman, R.D. Levin, J. Phys. Chem. Ref.

Data 13 (1984) 695.

[84] R.E. March, R.J. Hughes, Quadrupole Storage Mass Spectro-

metry, Wiley, New York, 1989, p. 209.

[110] R. Atkinson, D.L. Baulch, R.A. Cox, R.F. Hampson, J.A.

Kerr, J. Troe, J. Phys. Chem. Ref. Data 21 (1992) 1125.

[111] D.A. Ponomarev, V.V. Takhistov, General fragmentation

rules of organic and elementoorganic compounds under elec-

tron impact, Recent Adv. Anal. Techniques, 2000, in press.

[112] V.V. Takhistov, Dr Sci. thesis, Leningrad, 1989.

[113] D.A. Ponomarev, Dr Sci. thesis, St. Petersburg, 1997.

[114] R. Shaw, in: S. Patai (Ed.), Thermochemistry in the Carbon–

Carbon Triple Bond, 1, Wiley, New York, 1978, pp. 69–73.

[115] W. Roth, private communication, quoted in S.G. Lias, P.

Ausloos, J. Chem. Phys. 82 (1985) 361.

[116] M.P. Kozina, V.S. Mastr’ukov, E.M. Mil’vitskaya, Uspekhi

Khimii (Russian Chemical Reviews) 51 (1982) 1337.

[117] R. Fuchs, J.H. Hallman, M.O. Perlman, Can. J. Chem. 60

(1982) 1832.

[118] H.-D. Beckhaus, Chem. Ber. 116 (1983) 86.

[119] L.-H. Lai, Y.C. Hsu, Y.-P. Lee, J. Chem. Phys. 97 (1992)

3092.

[85] G. Bouchoux, J.-Y. Salpin, J. Phys. Chem. 100 (1996) 16555.

[86] R. Bombach, J. Dannacher, E. Honegger, J.-P. Stadelmann,

R. Neier, J. Chem. Phys. 82 (1983) 459.

[87] J.J. Butler, M.L. Fraser-Monteiro, L. Fraser-Monteiro, T.

Baer, J. Phys. Chem. 86 (1982) 747.

[88] L.W. Sieck, J. Phys. Chem. A 101 (1997) 89140.

[89] E.P. Hunter, S.G. Lias, J. Phys. Chem. Ref. Data, in press.

[90] M.B. Seasholtz, T.B. Thompson, N.G. Rondan, J. Phys.

Chem. 99 (1995) 17835.

[91] H. Bock, T. Hirabayashi, S. Mohmand, Chem. Ber. 114

(1981) 2595.

[92] J.C. Traeger, private communication (quoted in Ref. [14]).

[93] R.L. Deming, C.A. Wulff, The Chemistry of Ketenes,

Allenes and Related Compounds, 1, Wiley, New York,

1980, chap. 4.

[94] Van Chin S’yan, Y.Y. Kochubey, V.V. Sergueev, Yu.A.

Raevsky, S.I. Guerasimchuk, H.Z. Kotovichuk, J. Phys.

Chem. (Russ) 70 (1996) 1932.

[95] M.A. McAllister, T.T. Tidwell, Can. J. Chem. 72 (1994)

882.

[96] C.-R. Han, Acta Chim. Sinica 48 (1990) 627.

[120] D.A. Ponomarev, V.V. Takhistov, A.D. Misharev, V.M.

Orlov, J. Gen. Chem. (Russ.) 63 (1993) 2067.

[121] M.T. Nguyen, H.M.T. Nguyen, Chem. Phys. Lett. 300 (1999)

346.

Article Doi

Thermochemistry of organic and heteroorganic species. Part VIII. Ketene and structurally related species

DOI: 10.1016/S0022-2860(99)00452-4

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S0022-2860(99)00452-4

Full text of DOI:10.1016/S0022-2860(99)00452-4

Products guided by the article

R&D Labs maybe for 463-51-4

Relevant to this article

Hot Product