the formation of neutral ccco2h and hccco2 molecules from anionic precursors in the gas phase: a...

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2005; 19: 3705–3712

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/rcm.2245

The formation of neutral CCCO2H and HCCCO2

molecules from anionic precursors in the gas phase:

a joint experimental and theoretical study

Mark Fitzgerald1, John H. Bowie1*, Detlef Schroder2 and Helmut Schwarz2

1Department of Chemistry, The University of Adelaide, South Australia, Australia 50052Institut fur Chemie, Technische Universitat Berlin, D-10623 Berlin, Germany

Received 29 July 2005; Revised 19 October 2005; Accepted 19 October 2005

Calculations at the CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G(d) level of theory indicate that the anions�CCCO2H and HCCCO2

� are stable species in their singlet states. Upon collision-induced, vertical

one-electron oxidation under neutralisation-reionisation (�NRþ) conditions, they produce the neu-

tral molecules CCCO2H and HCCCO2, respectively. Some of the CCCO2H neutrals should be stable

for the duration of the neutralisation-reionisation experiment (10�6 s), while others will dissociate

to CCCO and OH (requires 125kJmol�1). In contrast, neutral HCCCO2 is expected to be much less

stable, and dissociate to HCC and CO2 (37kJmol�1). Neither CCCO2H nor HCCCO2 is expected to

interconvert, or to rearrange to other isomers. The anions �CCCO2H and HCCCO2� have been

formed in the ion source of the mass spectrometer by the reactions between (CH3)3Si–C:C–CO2H

and F� and HC:C–CO2Si(CH3)3 and F�, respectively. The �NRþ spectrum of �CCCO2H

shows a recovery signal and also indicates that the lowest energy dissociation pathway of neutral

CCCO2H corresponds to the loss of OH. The �NRþ spectrum of HCCCO2 displays little or no recov-

ery signal, and the spectrum is dominated by the [CO2]þ ion. The experimental observations are in

agreement with the predictions of the extensive theoretical studies. Copyright # 2005 John Wiley &

Sons, Ltd.

Several carboxylic acids, including HCO2H, CH3CO2H, and

NH2CH2CO2H, are listed amongst the 123 molecules that

have been detected in the interstellar medium,1–7 and have

been found to be more abundant within UV-shielded envir-

onments such as dense molecular clouds and hot cores (star-

forming regions). Acetic acid has been detected in one such

hot core, the ‘Large Molecular Heimat’ (LMH), which is

within the giant molecular cloud complex Sagittarius B2.8

Its formation within this region, as well as the formation of

other complex oxygenated molecules, has been rationalised

using gas-phase chemistry involving the evaporation of

dust grain mantles.9 This rationale is supported by the

discovery of formic acid in interstellar ice; it was found that

formic acid incorporation into the ice that encapsulates dust

grains of selected objects of the high mass protostar W 33A

was at approximately 3% of the solid H2O.10,11

Carboxylic acids have also been detected in meteorites,

with the study of the Murchison and EET960.20 carbonaceous

chondrites revealing>50 simple-structured carboxylic acids,

including the straight-chained series from formic acid to

octanoic acid.12,13 Analysis of several of these carboxyl

species revealed isotopic abundances consistent with

interstellar origins, providing insight into the formation

processes and survival of dust grains and their constituents in

the interstellar medium.12,14,15

Currently, neither CCCO2H nor HCCCO2 has been

detected in interstellar or circumstellar media. However,

HCCCO2 may be an interstellar molecule since the reaction

between HCC and CO2 (both known interstellar molecules)5

to yield HCCCO2 is thermodynamically favourable (�37 kJ

mol�1, CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G(d) level of

theory).

The anion, HCCCO2�, has been investigated as a potential

decarboxylation product of aqueous acetylenedicarboxylic

acid using Fourier transform infrared (FT-IR) spectroscopy

and density functional theory.16 HCCCO2�, prepared on a

TiO2 surface, has been examined using noncontact atomic

force microscopy.17 The geometry, electronic structure and

energy of the cation [HCCCO2]þ have been determined by ab

initio self-consistent field calculations.18 A neutralisation-

reionisation study has been reported on the isomer

[OCCHCO]þ.19

This study is an extension of our previous work on

cumulenes and polycarbon monoxides of interstellar

significance. It was found that, upon formation by collision-

induced vertical one-electron oxidation, cumulenic CCCHO,

CCCCHO, HCCCCHO, and CCCCCHO undergo hydrogen

rearrangement, and that the species HCnO (n¼ 3–5) are more

stable than their aldehydic counterparts.20–23 It is the aim of

Copyright # 2005 John Wiley & Sons, Ltd.

*Correspondence to: J. H. Bowie, Department of Chemistry, TheUniversity of Adelaide, South Australia, Australia 5005.E-mail: [email protected]

this study to extend this work further to include the

formation, from the anionic precursors �CCCO2H and

HCCCO2�, of the transient neutral carboxyl species CCCO2H

and HCCCO2, respectively, and to determine if these species

display similar rearrangement.

EXPERIMENTAL AND COMPUTATIONAL

Theoretical approachGeometry optimisations were carried out with the Becke

3LYP method24,25 using the 6-31G(d) basis set within the

GAUSSIAN 98 suite of programs.26 Stationary points were

characterised as minima (no imaginary frequencies). The cal-

culated frequencies were also used to determine zero-point

vibrational energies which were used as a zero-point correc-

tion for the electronic energies. These were scaled by 0.9804.27

We have previously reported the success of the B3LYP

method in predicting geometries of unsaturated chain struc-

tures, and that this method produces optimised structures, at

low computational cost, that compare favourably with higher

level calculations.28 More accurate energies for the B3LYP

geometries were determined using the couple cluster

CCSD(T) method29 using the Dunning aug-cc-pVDZ basis

set.30–32 All calculations were carried out on the Alpha Server

at the Australian Partnership for Advanced Computing

(APAC) National Facility (Canberra).

Mass spectrometric methodsThe experiments were performed with a modified VG ZAB/

HF/AMD 604 four-sector mass spectrometer of BEBE config-

uration (B stands for magnetic sector and E for electric

sector).33 In brief, the neutral precursors [(CH3)3Si–C:C–CO2H and HC:C–CO2Si(CH3)3 (commercial samples)]

were introduced into a chemical ionisation (CI) source in

the presence of F� (from SF6) as reagent. A SN2(Si) reaction

between the neutral substrates and F� gave the required pre-

cursor anions �CCCO2H and HCCCO2�.34 After acceleration

to 8 keV kinetic energy, the anions of interest were mass-

selected using B(1) for the collision-induced dissociation

(CID) experiment and B(1) and E(1) for the charge reversal

(CR) and neutralisation-reionisation (NR) experiments. For

CID, helium was used at 80% transmittance (T) of the incident

ion beam. In �CRþ experiments, the ions were collided with

molecular oxygen at 80% T. For �NRþ experiments, ions were

collided with molecular oxygen at 80% T in the first collision

cell, residual ions were removed from between the two colli-

sion cells via the application of a deflector electrode, then neu-

trals were reionised by a second collision process in the

second collision cell (again oxygen at 80% T). The resulting

product ions were mass-analysed using E(1) for CID experi-

ments (negative ion), and B(2) for the CR and NR experiments

(positive ions).

RESULTS AND DISCUSSION

The theoretical evidence

The anions �CCCO2H and HCCCO2�

It has been reported that singlet anion �CCCCHO rearranges

via H migration to singlet HCCCCO�, with an energy barrier

of only some 28 kJ mol�1.21 Is it possible to form HCCCO2 and

CCCO2H from anionic precursors? Before addressing this

problem experimentally, it is first necessary to determine if

the precursor anions �CCCO2H and HCCCO2� are stable

with respect to rearrangement and/or dissociation. The

possible rearrangement of �CCCO2H to HCCCO2� has been

investigated at the CCSD(T)/aug-cc-pVDZ//B3LYP/

6-31G(d) level of theory. This reaction coordinate is shown

in Fig. 1; full data of structures and energies are listed in

Table 1. Triplet states of �CCCO2H (A), �CCCO2H (B), and

HCCCO2� have been excluded from further study, as they

lie some 349, 369, and 320 kJ mol�1 higher in energy than their

respective singlet states (see Table 1). Dissociation energies

for singlet anions �CCCO2H (B) and HCCCO2� are recorded

in Table 2.

The conformational isomers 11�(A) and 11�(B) are sepa-

rated in energy by only some 10 kJ mol�1, so both species

should be formed in the mass spectrometer. The barriers for

the rearrangements of 11�(A) to 12� and 11�(B) to 12� are

calculated to be 117 and 127 kJ mol�1, respectively. The

barrier for the rearrangement of 12� to 11� [(A) or (B)] is

207 kJ mol�1. These barriers are large, indicating that ener-

gised �CCCO2H and HCCCO2� should be stable with respect

to rearrangement. Unimolecular rearrangements of negative

ions of this type are the exception rather than the rule; those

that have been confirmed have barriers �55 kJ mol�1.35 Note,

however, that bi- or termolecular rearrangements via proton

catalysis can be quite facile. Dissociation pathways for 11�(B)

and 12� are listed in Table 2. These are all high-energy

processes, with the exception of the loss of CO2 from 12�

(HCCCO2�!HCC�þCO2, 120 kJ mol�1).

The theoretical data indicate that under the collisional

conditions used for the conversion of an anion into a neutral:

(1) �CCCO2H is stable to both rearrangement and dissocia-

tion, and (2) HCCCO2� is stable to rearrangement, but, if the

anion has an excess energy of 120 kJ mol�1, dissociation to

HCC� and CO2 is possible.

C3 C2 C1O1

O2 H

C3 C2 C1O1

O2

H

C3 C2 C1O1

O2

H

C3 C2 C1O1

O2H

H C3 C2 C1O1

O2

11- (A)

TS11- (A)/11

- (B)

11- (B)

TS11- (B)/12

-

12-

(0.0)

(35)

(-10)

(117)

(-90)

Figure 1. Rearrangement of �CCCO2H to HCCCO2�.

CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G(d) level of theory. For

geometries and energies of minima and transition structures,

see Table 1. Energy units are in kJ mol�1.

Copyright # 2005 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2005; 19: 3705–3712

3706 M. Fitzgerald et al.

Table

1.

Energ

ies

and

geom

etr

ies

of

sin

gle

tand

trip

let

anio

ns

Sta

te1A

01A

01A

01A

1

Sy

mm

etry

CS

C1

CS

CS

C2V

C1

C1

C1

En

erg

y(H

artr

ees)

a�

264.

6878

5�

264.

6746

1�

264.

6916

3�

529.

3794

7�

264.

7220

6�

264.

5547

1�

264.

5509

4�

264.

6001

6D

ipo

lem

om

ent

(Deb

ye)

b6.

564.

453.

504.

583.

200.

83B

on

dle

ng

th(A

)bo

ran

gle

(8)b

C1C

21.

405

1.40

41.

419

1.49

31.

523

1.38

61.

387

1.37

7C

2C

31.

255

1.25

71.

256

1.26

61.

217

1.31

11.

309

1.27

3O

1C

11.

235

1.22

51.

223

1.22

51.

251

1.29

21.

287

1.03

7O

2C

11.

400

1.44

51.

404

1.31

91.

251

1.42

01.

425

1.30

7H

O2

0.97

50.

970.

973

1.44

00.

975

0.97

2H

C3

1.83

91.

065

1.09

8C

1C

2C

317

8.9

178.

017

4.4

166.

618

0.0

169.

016

7.2

169.

1O

1C

1C

212

9.5

129.

513

0.4

129.

511

4.2

120.

612

1.0

126.

1O

2C

1C

211

3.7

113.

511

2.7

101.

611

4.2

119.

011

9.7

126.

0H

O2C

110

1.8

105.

310

3.4

78.1

101.

110

4.1

HC

3C

218

0.0

129.

9C

1C

2C

3H

180.

0�

179.

9O

1C

1C

2C

30.

097

.918

0.0

180.

018

0.0

129.

414

3.1

�91

.8O

2C

1C

2C

318

0.0

�82

.80.

00.

00.

0�

70.5

�56

.791

.6H

O2C

1C

218

0.0

93.2

0.0

0.0

�15

3.8

15.6

aC

CS

D(T

)/au

g-c

c-p

VD

Z/

/B

3LY

P/

6-31

G*

lev

elo

fth

eory

incl

ud

ing

zero

-po

int

ener

gy

(B3L

YP

/6-

31G

*,sc

aled

by

0.98

04).

27

bB

3LY

P/

6-31

G*

lev

elo

fth

eory

.

Experimental and theoretical study of CCCO2H and HCCCO2 molecules 3707


The neutral molecules CCCO2H and HCCCO2

Having identified �CCCO2H and HCCCO2� as potentially

suitable precursors from which to form their respective neu-

tral counterparts, the stability and interconversion energy of

neutral CCCO2H and HCCCO2 were investigated at the

CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G(d) level of theory.

Neutral CCCO2H [(A) and (B)] and HCCCO2 are stable, and

their closest valence bond structures are represented in

Scheme 1. Full details for these species are listed in Table 3.

Neutral HCCCO2 (22) is the most stable of these species, being

95 and 105 kJ mol�1 lower in energy than 21(A) and 21(B),

respectively .

The reaction coordinate for the interconversion of neutral21(A), 21(B), and 22 is shown in Fig. 2. For full structural and

energy data of the neutral species, see Table 3 and for relevant

dissociation energies see Table 4.

Neutral 21(A) requires 44 kJ mol�1 to convert into 21(B),

which then requires an additional 151 kJ mol�1 to rearrange

to HCCCO2 (22), in a process exothermic by 95 kJ mol�1.

Rearrangement over such an energy barrier is feasible for an

energised neutral, but since 21(A) requires only 125 kJ mol�1

Table 2. Dissociation pathways of singlets �CCCO2H (B)

and HCCCO2�

�CCCO2H (B)!CC�þCO2H 508 kJ mol�1

!CCCOþ�OH 339!CCCO2

�þH 383!Cþ�CCO2H 791

HCCCO2� !HCC�þCO2 120

!HCCCOþO� 625!HþCCCO2

� 463!HCþCCO2

� 755

Table 3. Energies and geometries of neutrals included in Fig. 2

State 2A0 2A0 2A0 2A0

Symmetry CS C1 CS CS CS

Energy (Hartrees)a �264.54975 �264.53311 �264.54585 �264.47543 �264.58601Dipole moment (Debye)b 2.06 2.21 3.70Bond length (A)b or angle (8)b

C1C2 1.439 1.434 1.443 1.472 1.439C2C3 1.259 1.269 1.257 1.275 1.209O1C1 1.216 1.208 1.212 1.188 1.259O2C1 1.340 1.368 1.341 1.374 1.259HO2 0.997 0.972 0.974 1.361HC3 1.347 1.068C1C2C3 166.0 163.8 163.5 133.4 179.0O1C1C2 120.5 120.7 119.3 138.8 123.0O2C1C2 114.4 115.4 118.4 92.5 123.0HO2C1 106.9 110.9 109.8 102.2HC3C2 179.0C1C2C3H 180.0O1C1C2C3 180.0 178.9 180.0 180.0 0.0O2C1C2C3 0.0 �4.1 0.0 0.0 180.0HO2C1C2 180.0 90.0 0.0 0.0

a CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G* level of theory including zero-point energy (B3LYP/6-31G*, scaled by 0.9804).27

b B3LYP/6-31G* level of theory.

C3 C2 C1O1

O2 H

C3 C2 C1O1

O2

H

C3 C2C1

O1

O2

H

C3

C2 C1O1

O2H

H C3 C2 C1O1

O2

21 (A)

TS 21 (A)/21 (B)

21 (B)

TS 21 (B)/22

22

(0.0)

(44)

(10)

(195)

(-95)

(125)

CCCO + OH

(-58)HCC + CO2

Figure 2. Rearrangement of CCCO2H to HCCCO2.




C C C

O

OH

CC C

O

O

H

H C C C

O

O

21(A) 21(B) 22

Scheme 1.



to dissociate to CCCO and OH, 21(A) would be expected to

dissociate rather than to rearrange via this pathway.

Conversion of 22 into 21 involves a barrier of 290 kJ mol�1

(see Fig. 2), so this rearrangement in energetically unfavour-

able. However, 22 needs only 37 kJ mol�1 to effect loss of CO2,

so in this case energised 22 will dissociate to HCC and CO2.

The theoretical data suggest that CCCO2H and HCCCO2

are unlikely to interconvert (under the conditions required

for the charge-stripping process), but that HCCCO2 may

dissociate to yield HCC and CO2.

The next question to be answered is whether either

CCCO2H or HCCCO2 can rearrange via other low-energy

pathways. Potential surfaces illustrating other possible

rearrangements of CCCO2H and HCCCO2 are shown in

Figs. 3 and 4, respectively. Structural and dissociation data

for these systems are recorded in Tables 5 and 6.

Possible rearrangements of 21 are shown in Fig. 3. Con-

version of 21 into CCOCOH (23) is energetically unfavour-

able. The rearrangement of 21 to HOCCCO (25) is exothermic

by 106 kJ mol�1. This process has a barrier of 134 kJ mol�1, a

figure higher than that required to cause dissociation of 21 to

CCCO and OH (125 kJ mol�1).

Finally, the rearrangement of 22 shown in Fig. 4 is unlikely

to occur because 22 requires only 37 kJ mol�1 to effect

dissociation to HCC and CO2.

In summary, the theoretical study suggests that: (1)�CCCO2H and HCCCO2

� may be successfully converted into

neutrals CCCO2H and HCCCO2 by vertical one-electron

oxidation during a neutralisation-reionisation (�NRþ)

experiment, (2) CCCO2H should be a stable species, and (3)

HCCCO2 should be stable, but if it has an excess energy of

�37 kJ mol�1, it may dissociate to give HCC and CO2.

The experimental evidence

The anions �CCCO2H and HCCCO2�

These ions were formed by chemical ionisation in the source

of a modified VG ZAB/HF/AMD four-sector mass spectro-

meter as follows.

ðCH3ÞSi--C------C--CO2H þ F� ! ðCH3Þ3SiF þ �CCCO2H ð1Þ

HC------C--CO2SiðCH3Þ3 þ F� ! ðCH3Þ3SiF þ HCCCO�2 ð2Þ

The CID negative ion mass spectrum of �CCCO2H (m/z 69,

100%) shows the loss of H (m/z 68, 55%), and minor losses of

HO (m/z 54), CO2 (m/z 25), and CO2H (m/z 24), consistent

with the bond connectivity CCCO2H. The loss of CO2 is

likely to occur by the process �CCCO2H! [(CC) �CO2H]!�C2HþCO2. The CID spectrum of HCCCO2

� (m/z 69, 100%)

displays the loss of H (m/z 68, 30%) and CO2 (m/z 25, a weak

ion), in agreement with the connectivity HCCCO2.

The charge reversal (�CRþ) mass spectra of �CCCO2H

and HCCCO2� are recorded in Figs. 5(A) and 6(A), respec-

tively. These spectra indicate that any interconversion of

[CCCO2H]þ and [HCCCO2]þ is minimal. Figure 6(A) shows

major losses of O, HC2 and CO2, diagnostic of [HCCCO2]þ;

Fig. 5(A) shows major losses of HO, C2 and CO2H, diagnostic

of [CCCO2H]þ.

Table 4. Dissociation pathways of neutrals CCCO2H (A)

and HCCCO2

CCCO2H (A)!CCþCO2H 380 kJ mol�1

!CCCOþOH 125!CCCO2þH 356!CþCCO2H 598

HCCCO2 !HCCþCO2 37!HCCCOþO 601!HþCCCO2 451!HCþCCO2 561

C3 C2 C1O1

O2 H

C3C2

O1

C1O2

H

C3

H

C2

O1

C1 O2

C3

C2 C1 O1

O2

H

C3

C2C1 O1

O2

H

H

O2C2

C3

C1

O1

H

O2C2 C3

C1O1

21 (A)

TS 24/25 (B)

23

TS 21 (A)/23

25

(0.0)

(320)

(254)

(-106)

(134)

(-46)

(-40)

TS 21 (A)/24

24

[Pathway 1]

[Pathway 2]

(125)

CCC + OH

Figure 3. Rearrangement of CCCO2H to CCOCOH and

HOCCCO. CCSD(T)/aug-cc-pVDZ//B3LYP/6-31G(d) level of

theory. For geometries and energies of minima and transition

structures, see Table 5. Energy units are in kJ mol�1.

H C3 C2 C1O1

O2

H C3 C2 C1O2

O1

HC3

C2O1

C1O2

22

TS 22/26

(0.0)

(141)

(58)

26

(37)HCC + CO2

Figure 4. Rearrangement of HCCCO2 to HCCOCO.






Table

5.

Energ

ies

and

geom

etr

ies

of

neutr

als

inF

igs.

3and

4

Sta

te2A

02A

02A

02A

02A

0

Sy

mm

etry

C1

C1

C1

CS

CS

CS

CS

CS

En

erg

y(H

artr

ees)

a�

264.

4275

4�

264.

4529

2�

264.

4985

3�

264.

5671

1�

264.

5649

5�

264.

5899

5�

264.

5324

1�

264.

5641

1D

ipo

lem

om

ent

5.07

3.99

2.55

1.86

(Deb

ye)

b

Bo

nd

len

gth

(A)b

or

ang

le(8

)b

C1C

21.

189

2.29

51.

345

1.46

51.

693

1.78

72.

328

C1C

32.

609

1.44

31.

383

1.29

3C

2C

31.

290

1.28

41.

313

1.34

71.

348

1.30

81.

230

1.20

2O

1C

11.

321

1.40

81.

160

1.19

81.

176

1.17

71.

358

1.38

8O

1C

21.

403

1.30

11.

390

1.32

2O

2C

11.

298

1.30

81.

883

1.18

31.

179

O2C

21.

802

1.31

91.

314

1.31

0H

O2

0.98

90.

974

0.97

90.

977

0.98

30.

983

HC

31.

067

1.06

5C

1C

2C

315

4.2

147.

915

8.0

61.6

52.6

156.

2C

1C

3C

263

.276

.615

1.4

C1O

1C

250

.211

8.4

O1C

1C

250

.030

.716

8.6

148.

0O

1C

1C

316

7.0

172.

8O

1C

2C

315

5.1

175.

7O

2C

1C

215

8.0

132.

365

.4O

2C

2C

315

3.4

142.

713

5.1

O2C

1O

113

3.0

124.

0H

O2C

111

1.4

106.

710

5.6

109.

2H

O2C

210

9.9

111.

3H

C3C

216

2.5

179.

3C

1C

2C

3H

0.0

0.0

O1C

1C

2C

317

7.3

152.

0�

0.9

180.

018

0.0

180.

018

0.0

180.

0O

2C

1C

2C

312

3.7

151.

7�

178.

40.

018

0.0

O2C

2C

3C

118

0.0

180.

018

0.0

HO

2C

1C

242

.8�

179.

990

.0H

O2C

2C

118

0.0

180.

00.

0

aC

CS

D(T

)/au

g-c

c-p

VD

Z/

/B

3LY

P/

6-31

G*

lev

elo

fth

eory

incl

ud

ing

zero

-po

int

ener

gy

(B3L

YP

/6-

31G

*,sc

aled

by

0.98

04).

27

bB

3LY

P/

6-31

G*

lev

elo

fth

eory

.



The neutral molecules CCCO2H and HCCCO2

The �NRþ spectrum of �CCCO2H (one-electron oxidation of�CCCO2H to neutrals in the first collision cell, then a second

one-electron oxidation to form cations in the second collision

cell) is shown in Fig. 5(B). This should be compared with the�CRþ spectrum of the same anion (see Fig. 5(A)). The ions in

the �NRþ spectrum that appear similar to their correspond-

ing ions in the �CRþ spectrum include: the parent ion

[CCCO2H]þ, the fragment ions due to [CC]þ (m/z 24),

[CCH]þ (m/z 25), [CCC]þ (m/z 36), [CCO]þ (m/z 40), and the

losses of neutral CO (m/z 41) and O (m/z 53). The presence

of a recovery signal in the �NRþ spectrum shows that vertical

one-electron oxidation of �CCCO2H forms some neutral

CCCO2H species which are stable for the microsecond dura-

tion of the NR experiment.

The �NRþ and �CRþ spectra of �CCCOOH display the

same peaks, but they have different relative peak heights.

Whereas the base peak in the �CRþ spectrum is at m/z

45 [CO2H]þ, the largest peak in the �NRþ spectrum is due

to the loss of OH [the lowest energy dissociation pathway

for neutral CCCO2H (see Table 4)]. This is consistent with

the formation of some unstable neutral CCCO2H species.

Based on this evidence, it is clear that the one-electron

oxidation of �CCCO2H, during the NR experiment, forms

some stable (for at least 10�6 s), and some decomposing,

neutral CCCO2H species.

The �NRþ and �CRþ spectra of HCCCO2� are shown in

Fig. 6. The two spectra show the same fragmentation peaks

but with some differences in relative abundances. The

major differences in the spectra are: (1) There is no parent

(recovery) signal at m/z 69 in the �NRþ spectrum, and (2)

the major peaks m/z 44 [CO2]þ and 28 [CO]þ are more

pronounced in the �NRþ than in the �CRþ spectrum. This is

consistent with the vertical one-electron oxidation of

HCCCO2� forming energised HCCCO2, which requires only

37 kJ mol�1 to effect dissociation to HCC and CO2. The energy

Figure 5. (A) �CRþ spectrum of �CCCO2H and (B) �NRþ

spectrum of �CCCO2H. Modified VG ZAB/HF/AMD 604 four-

sector mass spectrometer. See Experimental section for

details.

Figure 6. (A) �CRþ spectrum of HCCCO2� and (B) �NRþ

spectrum of HCCCO2�. Modified VG ZAB/HF/AMD 604 four-

sector mass spectrometer. See Experimental section for

details.

Table 6. Additional neutral dissociation pathways

CCOCOH !CCOCOþH 234 kJ mol�1

!CCOCþOH 298!CCOþCOH 70!CCþOCOH 133

cyc-C3OHO! cyc-C3OHþO 533! cyc-C3OþOH 441! cyc-C3O2þH 322!CCOHþCO 90!CCOþCOH 370

HOCCCO !OCCCOþH 13!HOþCCCO 230!HOCCþCO 150!HOCCCþO 616!HOCþCCO 430

HCCOCO !HþCCOCO 526!HCCOCþO 605!HCCOþCO �136!HCCþCO2 �21



required for this to occur is accounted for by the 39 kJ mol�1 of

Franck-Condon energy involved in the neutralisation of

HCCCO2� (the difference in energy between stable neutral

doublet HCCCO2 and HCCCO2 with the singlet anion

geometry on the neutral reaction coordinate). The enhanced

abundance of m/z 28 [CO]þ can be explained by dissociation

of [CO2]þ by the high-energy pathway [CO2]þ! [CO]þþO

(505 kJ mol�1).

SUMMARY AND CONCLUSIONS

1. Anions �CCCO2H and HCCCO2� have been formed

from precursors (CH3)3Si–C:C–CO2H and HC:C–

CO2Si(CH3)3, respectively. Calculations predict the anions

should be stable; this is confirmed experimentally from

their CID and �CRþ spectra.

2. Neutralisation-reionisation of �CCCO2H gives some

stable neutral CCCO2H, and some neutrals which

are energised and decompose to produce CCCO and

OH.

3. Neutralisation-reionisation of HCCCO2� gives energised

HCCCO2 which dissociate to HCC and CO2.

AcknowledgementsWe thank the Australian Research Council for the award of an

international link grant that enabled MF to spend three

months at the Technical University, Berlin, as an exchange

postgraduate student. We also acknowledge the Australian

Partnership for Advanced Computing (APAC) National

Facility (Canberra) for a generous allocation of time on their

Alpha Server. The Berlin group is grateful to the Deutsche

Forschungsgemeinschaft and the Fonds der Chemischen

Industrie for continuous support.

REFERENCES

1. Olofsson H. Molecules in the Stellar Environment, LectureNotes in Physics. Springer: Heidelberg, 1994.

2. Bernath PF, Hinkle KH, Keady JJ. Science 1989; 244: 562.3. Hinkle KH.Molecules in the Stellar Environment, Lecture Notes

in Physics. Springer: Heidelberg, 1994.4. Blanksby SJ, Bowie JH. Mass Spectrom. Rev. 1999; 18: 81.5. Wootten HA. http://www.cv.nrao.edu/�awooten/allmols.

html.6. Ohishi M, Suzuki H, Ishikawa S, Yamada C, Kanamori H,

Irvine WM, Brown RD, Godfrey PD, Kaifu N. Astrophys.J. 1991; 380: L39.

7. Matthews HE, Irvine WM, Friberg P, Brown D, Godfrey PD.Nature 1984; 310: 125.

8. Mehringer DM, Snyder LE, Miao Y. Astrophys. J. 1997; 480:L71.

9. Rodgers SD, Charnley SB. Astrophys. J. 2001; 546: 324.10. Schutte WA. Astron. Astrophys. 1999; 343: 966.11. Ehrenfreund P. Space Sci. Rev. 1999; 90: 233.12. Epstein S, Krishnamurthy RV, Cronin JR, Pizzarello S, Yuen

GU. Nature 1987; 326: 477.13. Blagojevic V, Petrie S, Bohme DK. Mon. Not. R. Astron. Soc.

2003; 339: L7.14. Cronin JR. Adv. Space Res. 1989; 9: 59.15. Huang Y, Wang Y, Alexandre MR, Lee T, Rose-Petruck C,

Fuller M, Pizzarello S. Geochim. Cosmochim. Acta 2005; 69:1073.

16. Li J, Brill TB. J. Phys. Chem. A 2002; 106: 9491.17. Sasahara A, Uetsuka H, Ishibashi T, Onishi H. Appl. Surf.

Sci. 2002; 188: 265.18. Maier WF, Reetz MT. J. Am. Chem. Soc. 1975; 97: 3787.19. Westmore J, Buchannon WD, Plaggenborg L, Wenclawiak

BW. J. Am. Mass Spectrom. 1998; 9: 29.20. Peppe S, Blanksby SJ, Dua S, Bowie JH. J. Phys. Chem. A

2000; 104: 5817.21. Fitzgerald M, Dua S, Bowie JH, McAnoy AM. Int. J. Mass

Spectrom. 2003; 228: 467.22. Fitzgerald M, Bowie JH, Dua S. Org. Biomol. Chem. 2003; 1:

1769.23. Fitzgerald M, Bowie JH, Dua S. Org. Biomol. Chem. 2003; 1:

3111.24. Becke AD. J. Chem. Phys. 1993; 98: 5648.25. Stevens PJ, Devlin FJ, Chablowski CF, Frische MJ. J. Phys.

Chem. A 1994; 98: 11623.26. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb

MA, Cheeseman JR, Zakrzewski VG, Montgomery JA,Stratmann RE, Burant JC, Dapprich S, Millam JM, DanielsAD, Kudin KN, Strain MC, Farkas O, Tomasi J, Barone V,Cossi M, Cammi R, Mennucci B, Pomelli C, Adamo C,Clifford S, Ochterski J, Petersson GA, Ayala PY, Cui Q,Morokuma K, Salvador P, Dannenberg JJ, Malick DK, RabuckAD, Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV,Baboul AG, Stefanov BB, Liu G, Liashenko A, Piskorz P,Komaromi I, Gomperts R, Martin RL, Fox DJ, Keith T,Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M,Gill PMW, Johnson B, Chen W, Wong MW, Andres JL,Gonzalez C, Head-Gordon M, Replogle ES, Pople JA.Gaussian 98, Gaussian Inc.: Pittsburgh PA, 2001.

27. Wong MW. Chem. Phys. Lett. 1996; 256: 391.28. Blanksby SJ, Dua S, Bowie JH. J. Phys. Chem. A 1999; 103:

5161.29. Deegan MJO, Knowles PJ. Chem. Phys. Lett. 1994; 1227: 321.30. Dunning TH. J. Chem. Phys. 1989; 90: 1007.31. Woon DE, Dunning TH. J. Chem. Phys. 1993; 98: 1358.32. Dunning TH, Peterson KA, Woon DE. Encyclopedia of Com-

putational Chemistry. John Wiley: Chichester, 1998.33. Schalley CA, Schroder D, Schwarz H. Int. J. Mass Spectrom.

Ion Processes 1996; 153: 173.34. Wenthold PG, Hu J, Squires RR. J. Am. Chem. Soc. 1994; 116:

6961.35. Dua S, Blanksby SJ, Peppe S, McAnoy AM, Bowie JH. Curr.

Org. Chem. 2003; 7: 1545.



the formation of neutral ccco2h and hccco2 molecules from anionic precursors in the gas phase: a...

Documents