Protonation-induced ligand switching within NO2+ bound clusters

Article abstract of DOI:10.1016/S0009-2614(97)01338-9

An intracluster proton-promoted reaction leading to a ligand switching in mixed NO₂⁺ complexes of nitric acid and its derivatives is described. The proton affinity and the NO₂⁺ binding energy are demonstrated to be the factors affecting the process, the trend of which is shown to reflect the peculiar properties of nitric acid and its derivatives.

Full text of DOI:10.1016/S0009-2614(97)01338-9

2
7 March 1998
Chemical Physics Letters 285 Ž1998. 366–372
q
Protonation-induced ligand switching within NO bound clusters
2
)
Giulia de Petris , Federico Pepi
Dipartimento di Studi di Chimica e Tecnologia delle Sostanze Biologicamente AttiÕe, UniÕersit a` degli Studi ‘La Sapienza’, Piazzale Aldo
Moro, 5-00185 Rome, Italy
Received 2 July 1997; in final form 4 November 1997
Abstract
q
An intracluster proton-promoted reaction leading to a ligand switching in mixed NO complexes of nitric acid and its
2
q
derivatives is described. The proton affinity and the NO2 binding energy are demonstrated to be the factors affecting the
process, the trend of which is shown to reflect the peculiar properties of nitric acid and its derivatives. q 1998 Elsevier
Science B.V.
1
. Introduction
Ionic processes such as clustering and ligand-switching reactions are known to occur in the D region of the
q
ionosphere, involving in particular weakly bound NOx hydrated complexes w1x. Mixed-cluster are produced that
incorporate neutral species existing in the mesosphere and stratosphere Že.g. HNO , CH OH, NH ., to an extent
3
3
3
depending on their relative concentrations, their binding energies in the cluster and their proton affinities w2x. In
particular, studies of intracluster reactivity of nitric acid and dinitrogen pentoxide w3–9x have emphasized the
q
role played by NO2 ions in the chemistry of small clusters, and in determining the final ionic composition of
q
the atmosphere. The effect of solvation on the structural rearrangement observed in NO ŽH O. clusters has
2
2
n
q
been demonstrated and correlated to the cluster size w10x. Recently, the NO ion affinities of a number of
2
q
molecules have been determined from the MIKE spectrometry of L ŽNO .L ion-bound dimers, obtained by
1
2
2
NO rchemical ionization ŽCI. of suitable gaseous mixtures of the ligands in the high-pressure source of a mass
2
q
spectrometer w11x. Most of these ŽL , L , NO . clusters obey simple dissociation kinetics, decomposing
1
2
2
q
q
exclusively to L NO and L NO fragments, which indicates that the precursor ion has the structure of a
1
2
2
2
nitronium ion-bound dimer ŽNBD.. A different reactivity is observed when a nitrated ligand Že.g. L sCH OH.
1
3
can act, in addition, as a proton donor inside the complex:
)
Corresponding author.
0
009-2614r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
PII S0009-2614Ž97.01338-9

G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
367
In this case detection of LHq fragments is strongly suggestive of the accompanying presence of a proton-bound
dimer ŽPBD..
In this work, a still different intracluster reaction is reported, namely a protonation-induced ligand-switching
process, the extent of which is critically affected by the nature of the ligands, in particular by the balance
q
q
between their NO2 binding energies ŽNO rBE. and their proton affinities ŽPA..
2
2
. Experimental
All chemicals were research-grade products from Aldrich Chemical Co., and were used without further
purification. Methyl and ethyl nitrate were synthesized and purified according to standard procedures. Nitramide
was synthesized by methanolysis of N-nitroacetamide, prepared in turn by NO BF nitration of acetamide w12x.
2
4
Dinitrogen pentoxide was synthesized by the technique of Davidson et al. w13x, which involves oxidation of NO
by O , generated by passing UHP grade oxygen ŽMatheson 99.95%. through a commercial ozonizer. The gases
3
used in the chemical ionization ŽCI. were purchased from Matheson Gas Products Inc., with a stated purity in
excess of 99.95 mol%.
MIKE spectra were recorded using a ZAB-2F mass spectrometer from VG Micromass Ltd., of which the CI
source, fitted with a specially built cooling system, has operated at temperatures ranging from 408C to 508C,
utilizing L rL rNO mixtures, the composition of which was optimized to obtain the highest intensity of the
1
2
2
q
L ŽNO .L dimers. Typical operation conditions were as follows: accelerating voltage 8 kV, emission current 1
1
2
2
mA, repeller voltage 0 V, electron energy 50 V. The FT-ICR experiments were performed utilizing a 47e APEX
spectrometer from Bruker Spectrospin AG, equipped with an external ion source, operated at total pressures not
y5
exceeding 7P10 Torr.
3
. Results and discussion
3
.1. MIKE spectrometric results
A number of nitronium ion-bound dimers of nitric acid and its derivatives ŽL . with suitable ligands ŽL .
2
1
was generated in the source of a reverse geometry mass spectrometer:
q
HX NO X NO
2
1
L2
2
L
1
X and X denote different groups, such as OH, OCH , OC H , NH , so that XNO and X NO represent the
1
3
2
5
2
2
1
2
conjugate bases of two different Brønsted acids. In each pair of X and X groups, we shall denote X as the one
1
giving HX as the best leaving group, e.g. XsOH if X sOCH , XsOCH if X sNH , and so on. In
1
3
3
1
2
principle, two distinct nitronium ion-bound dimers, NBD1 and NBD2,
HX NOq2 X NO
HX NO XNO2
q
1
2
1
2
NBD1
NBD2

3
68
G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
Table 1
q
MIKE spectra of different ŽL , NO , L . nitronium ion-bound dimers
1
2
2
X
X1
NBD1
NBD2
L1
a
a
L1
L2
CI
mrz
L2
HONO2 c
CI
mrz
b
2
q
q
OH
OH
OCH3
H O
CH ONO
NO
123
CH OH
NO
123
2
3
2
2
3
2
7
8Ž10%.
137
c
q
HONO2 c
HONO2 c
q
OC H
H O
C H ONO
NO
137
C H OH
NO
2
5
2
2
5
2
2
5
2
9
2 Ž3%.
q
2
q
2
q
2
q
2
OH
OCH3
NH2
NH2
H O
NH NO
NO
NO
108
108
NH3
NH3
NO
NO
108
108
63Ž14.8%.
108
2
2
2
2
b
2
CH OH
NH NO
CH ONO
3
2
3
2
6
3 Ž12.2%.
108
3 Ž6.6%.
2 Ž6.8%.
q
c
q
OC H
NH2
C H OH
NH NO
NO
NH3
C H ONO
NO
2
5
2
5
2
2
2
2
5
2
6
9
63 Ž13%.
a
Where not otherwise specified the abundances of the metastable peaks are 100%.
b
c
The same results have been obtained by N O5 CH FrCI.
2
3
N O CH rCI, especially used to avoid aqeous solutions, i.e. when LsHONO .
2
5
4
2
differing as to the presence of a preformed HX molecule can be originally formed, by varying the composition
q
of the gaseous mixtures of neutral ligands utilized in the NO rCI. For istance, for XsOH and X sOCH
2
1
3
q
q
H O—NO —CH ONO
CH OH—NO —HONO
2
2
3
2
3
2
2
NBD1
NBD2
q
one could expect that the NBD1 complex is generated from the NO rCI of a H OrCH ONO mixture,
2
2
3
2
q
whereas the NBD2 complex is formed from the NO rCI of a CH OHrHNO mixture.
2
3
3
All the neutral ligands pairs utilized in the CI experiments are shown in Table 1 where the metastable
q
decomposition pattern of each NBD is reported. The PA and NO rBE of the HXŽX . and XŽX .NO species
2
1
1
2
of interest are summarized in Table 2 w11,14–20x.
An alternative route to the NBDs is the protonation and alkylation of N O , performed by the CH rCI
2
5
4
Žprotonation and ethylation. and by CH FrCI Žmethylation. of dinitrogen pentoxide. As shown by recent
3
q
q
q
theoretical and experimental results w18x, the N O H , N O CH and N O C H ions produced are
2
5
2
5
3
2
5
2
5
ion–molecule complexes of NOq2 with HNO , CH ONO and C H ONO , respectively, and give nitronium
3
3
2
2
5
2
ion-bound adducts with HX neutral ligands. The MIKE spectra of the NBD1 complexes show the typical
dissociation pattern of the nitronium ion-bound dimers, the competitive dissociation processes of which reflect
q
the NO rBE of the two ligands.
2
q
As a consequence of the notation adopted, and consistent with the data of Table 2, the NO rBE of any given
2
q
X NO ligand is always larger than that of each HX and, hence in most cases, the X NO NO fragment is the
1
2
1
2
2
q
only one observed, without any detectable amount of HXNO .
2
HX NOq2 X NO ™X NO NO qHX
q
Ž .
1
1
2
1
2
2
NBD1
q
A small fraction of C H OHNO is only formed from the NBD1 of C H OH and NH NO , the closer
2
5
2
2
5
2
2
y1
q
q
Žwithin 0.8 kcal mol . NO rBE values of which allow the HXNO ion to be detected as well.
As regards NBD2 complexes, this fragmentation process accounts for a small fraction of their decomposition
pattern
2
2
Ž2-3.

G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
369
Table 2
q
NO rBE and PA values of HXŽX . and XŽX .NO compounds
2
1
1
2
q
a
PAa
NO rBE
2
Žkcal moly1.
Žkcal moly1.
HONO2
18.3
19.6
20.7
21.5
22.2
22.8
23.0
27.2
182.0
165.0
178.0
181.7
188.3
178.4
175.4
203.5
H O
2
CH ONO
3
2
CH OH
3
C H OH
2
5
C H ONO
2
2
5
NH NO
2
2
NH3
a
q
The NO rBEs of XŽX .NO species are all referred to the nitrogroup side, their PAs are all referred to the X group Žthe most basic site,
2
1
2
y1
except for NH NO , the PA of which at the nitro group amounts to 182 kcal mol . w14–18x.
2
2
q
q
q
The NO rBE of HX is now higher than that of each XNO , and HX NO ŽL NO . will be the only
2
1
2
1
2
1
2
nitrated ligand detectable from this fragmentation pathway. To our surprise, we found that the most abundant
fragment is the X NO NO ion corresponding to the loss of a HX molecule, which is not preformed within a
q
1
2
2
NBD2 complex, e.g.
q
q
CH OH NO HONO2 ™CH ONO NO qH O
3
2
3
2
2
2
In order to account for this fragmentation process, extensive rearrangement must occur within the NBD2
complex, yielding the same fragment as from a NBD1 type complex. A simple explanation for this peculiar
behavior can be based on the following considerations.
q
As mentioned before, in the case of L sCH OH, the CH OHNO ion can be viewed both as nitrated
1
3
3
2
methanol and protonated methyl nitrate, which help accounting for the intracomplex rearrangement
q
q
CH OH NO L2 l CH ONO H L2
3
2
3
2
q
especially when L ligands of suitable PA, e.g. CH CN, are involved. Then the appearance of L H fragments
2
3
2
q
Že.g. CH CNH . reveals the presence of a precursor having the structure of a proton-bound donor, formed
within the cluster from a low-barrier process. However, in a NBD2 type complex, the developing L H
3
q
2
fragment ion has again the 2-fold character of a Brønsted acid and a nitrated ligand. When L is nitric acid, the
2
y1
PA of which is f4 kcal mol higher than that of CH ONO w14,15x, the incipient product of dissociation of
3
2
the PBD,
q
q
q
CH OH NO HONO2 l CH ONO H HONO2 l CH ONO NO H O
3
2
3
2
3
2
2
2
NBD2
PBD
NBD1
q
q
the H ONO ion, is as well a weakly bound nitrated ligand, the NO rBE of which is lower than that of
2
2
2
CH ONO . In other words, as the proton moves to the most basic site, a better nitronium ion-donor is
3
2
simultaneously generated. The complex then reaches the configuration we have denoted as NBD1, which easily
q
dissociates into a CH ONO NO fragment. This mechanism also accounts for the absence of a significant
3
2
2
amount of kinetic energy released in the rearrangement process since, as in the simple dissociation channel,
weakly bound species are formed and dissociated within the cluster.
In general, two factors affect the whole process: Ži. the larger local PA of the X site of any given XNO with
2
q
respect to that of each X NO , which drives the proton to X, and Žii. the larger NO rBE of any given X NO
1
2
2
1
2
with respect to that of each HX, which favors the ligand switching process.
q
q
q
HX NO XNO2 l X NO H XNO2 l X NO NO HX
4
Ž .
1
2
1
2
1
2
2
NBD2
PBD
NBD1

3
70
G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
The occurrence of this rearrangement implies that the barriers to the isomerization of NBD2 ions are lower
1
than the dissociation threshold of the cluster. Actually, a rough estimate of the energetics of the competing
q
fragmentation processes indicates that the observed decomposition to the X NO NO ion is in all cases the
1
2
2
least endothermic one, which nicely explains why a similar isomerization is never observed from a NBD1
complex. The interplay between kinetic and thermodynamic factors in the control of the intracluster rearrange-
ment suggests that the barriers for the interconversion of all the complexes are relatively low — lower, in any
cases, than the decomposition threshold. Furthermore, dissociation into the most stable fragments occurs
irrespective of the specific way of formation of the cluster. Reinforcing evidence for this view, and for the
intermediacy of PBDs in the process, can be founded on the following considerations.
In the case of a NBD2 complex where L sNH and L sCH ONO , the energetically favourable proton
1
3
2
3
2
q
q
shift from NH NO to CH ONO causes formation of NBD1, that decomposes to NH NO NO ion,
3
2
3
2
2
2
2
mrzs108. However, a distinctive feature of NH NO is that the local PA of the X site is lower than that at
2
2
y1
the nitro group Ž175.4 vs. 182 kcal mol . w11,17x, whereas the local PA of the metoxy group of CH ONO
3
2
w15x happens to be intermediate Ž178.0 kcal moly1 . between the above values. Therefore, the CH OHNO ion
q
3
2
before dissociation can transfer the proton to the nitro group of NH NO , free of rotating within the complex to
2
2
q
allow proton transfer to its most basic site Žcluster 2. w21,22x. In practice, the CH OHNO2 ion can act as an
effective proton carrier, from the less basic to the more basic site of NH NO .
3
2
2
The same result is obtained on the energy surface of a NBD1 complex Žsee Table 1., or of a PBD, as shown
q
by experiments involving CH rCI of a NH NO rCH ONO mixture. In addition to the expected NH NO H
4
2
2
3
2
2
2
q
ion Žmrz 63, 15%., the MIKE spectrum of the dimer, shows the NH NO NO ion as the most abundant
fragment. In other words, irrespective of the specific formation route of the cluster, the system evolves to the
same products through a series of intermediate PBDs.
2
2
2
Coming back to NBD2, an alternative formation route of the fragment of mrz 63 is conceivable, i.e. the
q
q
simple dissociation to HX NO Žnamely NH NO ., previously described as a minor fragmentation channel of
1
2
3
2
NBD2 complexes, and corresponding to the nitramide protonated at the X site. Even if this process cannot be
excluded, several considerations militate in favor of a mechanism involving PBDs: Ži. the larger endothermicity
1
q
q
H 8 of XŽX .NO NO can be estimated from the NO rBE values of XŽX .NO and their heats of formation. The same procedure
f
1
2
2
2
1
2
q
can be followed for H 8 of HXŽX .NO , otherwise the PAs values can be used.
f
1
2

G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
371
of the process leading to the less stable isomer of protonated nitramide, Žii. the similarity of the MIKE spectra of
precursor ions other than NBD2, and Žiii. the inefficiency of this fragmentation in those cases where the PA of
the ligands is not intermediate between those of the two sites of nitramide. A typical example is HONO , which
2
hence cannot play the role of a catalyst as CH ONO or C H ONO do.
3
2
2
5
2
3
.2. ICR spectrometry results
q
Different HXŽX .NO ions were generated in the external CI ion source of a Fourier-transform ion
1
2
cyclotron resonance ŽFT-ICR. mass spectrometer. The following general reaction has been used to obtain XNO2
species protonated at the X site:
q
q
HXqNO ™HXNO
2
2
The ions formed were isolated by selective-ejection techniques and allowed to react with XŽX .NO
1
2
molecules in the resonance cell. The major processes observed
CH ONO
3
2
q
q
H ONO
™
CH ONO NO qH O
Ž
Ž
Ž
Ž
5
6
7
8
.
2
2
3
2
2
2
C H ONO
2
5
2
q
q
H ONO
™
C H ONO NO qH O
.
.
.
2
2
2
5
2
2
2
CH ONO
3
2
q
q
NH NO
™
NH NO NO qCH OH
3
2
2
2
2
3
C H ONO
2
5
2
q
q
NH NO
™
NH NO NO qC H OH
3
2
2
2
2
2
5
outline two distinct reactivity patterns, i.e.
q
q
HXNO qX NO ™X NO NO qHX
Ž
Ž
a
.
.
2
1
2
1
2
2
q
q
HX NO qXNO ™X NO NO qHX
b
1
2
2
1
2
2
q
Reaction Ža. refers to ions where XNO has a higher PA and a lower NO rBE than the X NO reagent present
2
2
1
2
in the cell, and therefore nitronium ion transfer is the only process observed, e.g. in the experiments Ž5. and Ž6..
Reaction Žb. refers to ions where X NO has a lower PA than XNO , and the occurrence of the proton transfer
1
2
2
q
q
is proved by detection of HXNO2 ions Že.g. CH OHNO ., as shown in Fig. 1. Nevertheless, the most
3
2
q
q
abundant product is the X NO NO ion, namely NH NO NO in the experiments Ž7. and Ž8. reported above.
1
2
2
2
2
2
An intracluster reaction, following the proton transfer to the most basic XNO2 species ŽCH ONO or
3
2
q
q
Fig. 1. Time profile of the ionic abundances in the reaction of NH NO , mrzs63, with CH ONO , yielding NH NO NO ion,
3
2
3
2
2
2
2
q
q
mrzs108. The intensity of ŽCH ONO .H ion, mrzs78, decreases following reaction with CH ONO , yielding CH ONO NO ion,
3
2
3
2
3
2
2
mrzs123.

3
72
G. de Petris, F. PepirChemical Physics Letters 285 (1998) 366–372
q
q
C H ONO ., is required for the formation of the NH NO NO ion. In fact, although the adduct wNH NO
2
5
2
2
2
2
3
2
CH ONO x cannot be detected in the low pressure range typical of the FT-ICR experiments, the observed loss
3
2
of methanol requires that protonated methyl nitrate, formed in the first step of the intracluster process, acts in
turn as a nitronium-ion donor within the complex, before dissociation takes place.
In conclusion, the occurrence of this rearrangement in complexes that are short-lived under ICR conditions,
speaks for facile interconversion of the isomeric forms of long-lived cluster ions. The reaction observed reflects
certain peculiar properties of nitric acid and its derivatives. In fact, protonation at the X sites of these
compounds yields protomers that are particularly stable, due to their character of ion-molecule complexes
w23–28x, and the PA trend NH NO -CH ONO (C H ONO -HONO Žsee Table 2. mirrors the enhance-
2
2
3
2
2
5
2
2
q
ment of this distinctive character. It follows that a high basicity of XNO is contrasted by a low NO rBE of
the HX moiety, peculiar feature that provides the driving force for the intracluster reaction reported.
2
2
Acknowledgements
This work has been financially supported by the Universit `a ‘La Sapienza’ of Rome and Consiglio Nazionale
delle Ricerche ŽCNR.. The authors wish to thank Professor Dr. Fulvio Cacace for helpful discussions and
continuous and generous support. Mr. F. Angelelli is acknowledged for FT-ICR measurements.
References
w1x D. Smith, P. Spanel, Mass Spectrom. Rev. 14 Ž1995. 255.
w2x D. Smith, N.G. Adams, E. Alge, Planet. Space Sci. 29 Ž1981. 449.
w3x H. Bohringer, D.W. Fahey, F.C. Fehsenfeld, E.E. Ferguson, Planet. Space Sci. 31 Ž1983. 185.
w4x P.J. Crutzen, F. Arnold, Nature 324 Ž1986. 651.
w5x C.M. Nelson, M. Okumura, J. Phys. Chem. 96 Ž1992. 6112.
w6x Y. Cao, J.-H. Choi, B.-M. Haas, M.S. Johnson, M. Okumura, J. Chem. Phys. 99 Ž1993. 9307.
w7x X. Zhang, E.L. Mereand, A.W. Castleman Jr., J. Phys. Chem. 98 Ž1994. 3554.
w8x H. Wincel, E.L. Mereand, A.W. Castleman Jr., J. Phys. Chem. 98 Ž1994. 8606.
w9x A.J. Stace, J.F Winkel, S.R. Atrill, J. Chem. Soc. Faraday Trans. 90 Ž1994. 3469.
w10x Y. Cao, J.-H. Choi, B.-M. Haas, M. Okumura, J. Phys. Chem. 98 Ž1994. 12176.
w11x F. Cacace, G. de Petris, F. Pepi, F. Angelelli, Proc. Natl. Acad. Sci. USA 92 Ž1995. 8635.
w12x S.A. Andrev, B.A. Lebedev, I.V. Tselinskii, Zh. Org. Khim. 14 Ž1978. 2513.
w13x J.A. Davidson, A.A. Viggiano, C.J. Howard, I. Dotan, F.C. Fehsenfeld, D.L. Albriton, E.E. Ferguson, J. Chem. Phys. 68 Ž1978. 2085.
w14x F. Cacace, M. Attin `a , G. de Petris, M. Speranza, J. Am. Chem. Soc. 116 Ž1994. 6413.
w15x A. Ricci, Org. Mass Spectrom. 29 Ž1994. 55.
w16x M. Aschi, F. Cacace, G. de Petris, F. Pepi, J. Phys. Chem. 100 Ž1996. 16552.
w17x M. Attin a` , F. Cacace, E. Ciliberto, G. de Petris, F. Grandinetti, F. Pepi, A. Ricci, J. Am. Chem. Soc. 115 Ž1993. 12398.
w18x F. Bernardi, F. Cacace, G. de Petris, F. Pepi, I. Rossi, submitted.
w19x S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, W.G. Mallard, J. Phys. Chem. Ref., 17 Ž1988. Data Suppl. 1.
w20x J.E. Szulejko, T.B. McMahon, J. Am. Chem. Soc. 155 Ž1993. 7839.
w21x R.D. Bowen, Acc. Chem. Res. 24 Ž1991. 364.
w22x P. Longevialle, Mass Spectrom. Rev. 11 Ž1992. 157.
w23x F. Cacace, M. Attin `a , G. de Petris, M. Speranza, J. Am. Chem. Soc. 111 Ž1989. 548.
w24x F. Cacace, M. Attin `a , G. de Petris, M. Speranza, J. Am. Chem. Soc. 112 Ž1990. 1014.
w25x G. De Petris, Org. Mass Spectrom. 25 Ž1990. 83.
w26x T.J. Lee, J.E. Rice, J. Am. Chem. Soc. 114 Ž1992. 8247.
w27x T.J. Lee, J.E. Rice, J. Phys. Chem. 96 Ž1992. 650.
w28x J.-H. Choi, K.T. Kuwata, Y.-B. Cao, B.-M. Haas, M. Okumura, J. Phys. Chem. 101 Ž1997. 6753.