The lowest-lying electronic singlet and triplet potential energy surfaces for the HNO-NOH system: energetics, unimolecular rate constants, tunneling and kinetic isotope effects for the isomerization and dissociation reactions.

Abstract

The lowest-lying electronic singlet and triplet potential energy surfaces (PES) for the HNO-NOH system have been investigated employing high level ab initio quantum chemical methods. The reaction energies and barriers have been predicted for two isomerization and four dissociation reactions. Total energies are extrapolated to the complete basis set limit applying focal point analyses. Anharmonic zero-point vibrational energies, diagonal Born-Oppenheimer corrections, relativistic effects, and core correlation corrections are also taken into account. On the singlet PES, the (1)HNO → (1)NOH endothermicity including all corrections is predicted to be 42.23 ± 0.2 kcal mol(-1). For the barrierless decomposition of (1)HNO to H + NO, the dissociation energy is estimated to be 47.48 ± 0.2 kcal mol(-1). For (1)NOH → H + NO, the reaction endothermicity and barrier are 5.25 ± 0.2 and 7.88 ± 0.2 kcal mol(-1). On the triplet PES the reaction energy and barrier including all corrections are predicted to be 7.73 ± 0.2 and 39.31 ± 0.2 kcal mol(-1) for the isomerization reaction (3)HNO → (3)NOH. For the triplet dissociation reaction (to H + NO) the corresponding results are 29.03 ± 0.2 and 32.41 ± 0.2 kcal mol(-1). Analogous results are 21.30 ± 0.2 and 33.67 ± 0.2 kcal mol(-1) for the dissociation reaction of (3)NOH (to H + NO). Unimolecular rate constants for the isomerization and dissociation reactions were obtained utilizing kinetic modeling methods. The tunneling and kinetic isotope effects are also investigated for these reactions. The adiabatic singlet-triplet energy splittings are predicted to be 18.45 ± 0.2 and 16.05 ± 0.2 kcal mol(-1) for HNO and NOH, respectively. Kinetic analyses based on solution of simultaneous first-order ordinary-differential rate equations demonstrate that the singlet NOH molecule will be difficult to prepare at room temperature, while the triplet NOH molecule is viable with respect to isomerization and dissociation reactions up to 400 K. Hence, our theoretical findings clearly explain why (1)NOH has not yet been observed experimentally.

Full text

Page 1

THE JOURNAL OF CHEMICAL PHYSICS 136, 164303 (2012)
The lowest-lying electronic singlet and triplet potential energy surfaces
for the HNO–NOH system: Energetics, unimolecular rate constants,
tunneling and kinetic isotope effects for the isomerization and dissociation
reactions
U˘ gur Bozkaya,1,a)Justin M. Turney,2Yukio Yamaguchi,2and Henry F. Schaefer III2
1Department of Chemistry, Atatürk University, Erzurum 25240, Turkey
2Center for Computational Quantum Chemistry, University of Georgia, Athens, Georgia 30602, USA
(Received 25 February 2012; accepted 4 April 2012; published online 25 April 2012)
The lowest-lying electronic singlet and triplet potential energy surfaces (PES) for the HNO–NOH
system have been investigated employing high level ab initio quantum chemical methods. The re-
action energies and barriers have been predicted for two isomerization and four dissociation reac-
tions. Total energies are extrapolated to the complete basis set limit applying focal point analyses.
Anharmonic zero-point vibrational energies, diagonal Born-Oppenheimer corrections, relativistic ef-
fects, and core correlation corrections are also taken into account. On the singlet PES, the1HNO
−→1NOH endothermicity including all corrections is predicted to be 42.23 ±0.2 kcal mol−1. For
the barrierless decomposition of1HNO to H + NO, the dissociation energy is estimated to be 47.48
± 0.2 kcal mol−1. For1NOH −→ H + NO, the reaction endothermicity and barrier are 5.25 ± 0.2
and 7.88 ±0.2 kcal mol−1. On the triplet PES the reaction energy and barrier including all correc-
tions are predicted to be 7.73 ± 0.2 and 39.31 ±0.2 kcal mol−1for the isomerization reaction3HNO
−→3NOH. For the triplet dissociation reaction (to H + NO) the corresponding results are 29.03
± 0.2 and 32.41 ±0.2 kcal mol−1. Analogous results are 21.30 ±0.2 and 33.67 ±0.2 kcal mol−1
for the dissociation reaction of3NOH (to H + NO). Unimolecular rate constants for the isomeriza-
tion and dissociation reactions were obtained utilizing kinetic modeling methods. The tunneling and
kinetic isotope effects are also investigated for these reactions. The adiabatic singlet–triplet energy
splittings are predicted to be 18.45 ±0.2 and 16.05 ±0.2 kcal mol−1for HNO and NOH, respec-
tively. Kinetic analyses based on solution of simultaneous first-order ordinary-differential rate equa-
tions demonstrate that the singlet NOH molecule will be difficult to prepare at room temperature,
while the triplet NOH molecule is viable with respect to isomerization and dissociation reactions
up to 400 K. Hence, our theoretical findings clearly explain why1NOH has not yet been observed
experimentally. © 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4704895]
I. INTRODUCTION
The nitrosyl hydride HNO has considerable importance
in combustion chemistry as an intermediate in the formation
of the air pollutant NO.1Most recently, the HNO molecule
has been of great interest due to its pharmacological ef-
fects and potential physiological functions in mammalian
systems.2–4The HNO molecule has been widely investi-
gated both theoretically and experimentally.5–43Beginning in
1958, Dalby5reported the first experimental observation of
HNO. Subsequently, Brown and Pimentel6observed the HNO
molecule by infrared spectroscopy in an argon matrix. The
HNO molecule has also been suggested as an important inter-
mediate in the catalytic decomposition of ozone in the strato-
sphere by reaction with HNOH.44In 1977, HNO was ob-
served in interstellar media by radio astronomy.45Fourteen
years later in 1991, Turner46detected the presence of HNO
in Sagittarius and confirmed its existence in the interstellar
media.
a)Electronic mail: ugur.bozkaya@atauni.edu.tr.
In contrast to HNO, there appear to be no experimental
data for the NOH isomer, whereas there are only a few the-
oretical studies on the NOH molecule. In 1979, Bruna and
Marian10predicted relative energy of the? X3A??and ˜ a1A?
43.2 kcal mol−1, respectively, using a multireference doubly
excited configuration interaction method with the DZP basis
set. In their 1995 study, Luna et al.24predicted the singlet–
triplet energy splittings for HNO (? X1A?– ˜ a3A??) and NOH
ploying a complete active-space self-consistent field method
with second order perturbation theory.47–50
In this study, the accurate reaction energies and barrier
heights are determined for the following two isomerization
(A and B) and four dissociation (C–F) reactions of HNO and
NOH:
states of NOH with respect to HNO (? X1A?) as 23.2 and
(? X3A??– ˜ a1A?) as 13.5 and 21.8 kcal mol−1, respectively, em-
(A)
(B)
(C)
(D)
(E)
HNO (? X1A?)?
HNO (? X1A?) −→ H (2S) + NO (X2?),
NOH (˜ a1A?) −→ H (2S) + NO (X2?)
?NOH (˜ a1A?),
?NOH (? X3A??),
HNO (˜ a3A??)?
HNO (˜ a3A??) −→ H (2S) + NO (X2?),
0021-9606/2012/136(16)/164303/15/$30.00© 2012 American Institute of Physics
136, 164303-1
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 2

164303-2Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
(F) NOH (? X3A??) −→ H (2S) + NO (X2?).
ization and dissociation reactions at various temperatures. For
these purposes, high level ab initio quantum chemical meth-
ods with large basis sets are employed and focal point anal-
yses are applied. Anharmonic zero-point vibrational energies
(AZPVE), diagonal Born-Oppenheimer corrections (DBOC),
relativistic effects, and core correlation corrections are con-
sidered as well. The rate constants for the isomerization reac-
tionsareestimatedemploying severaltheoreticalkineticmod-
eling methods. Tunneling and kinetic isotope effects of DNO,
TNO, NOD, and NOT are also investigated. Finally, the adia-
batic singlet–triplet energy splittings for HNO (? X1A?– ˜ a3A??)
Unimolecular rate constants are reported for the isomer-
and NOH (? X3A??– ˜ a1A?) are predicted.
II. THEORETICAL METHODS
Geometrical structures of all open-shell species were
optimized using the coupled-cluster singles and doubles
method with perturbative triple excitations [CCSD(T)]
(Refs. 51–53) from a restricted open-shell Hartree-Fock (HF)
reference. For closed-shell molecules, geometry optimiza-
tions were carried out using the CCSD(T) theory from a
restricted closed-shell Hartree-Fock reference. Dunning’s
correlation-consistentpolarized-core-valence
basis set (cc-pCVQZ) (Ref. 54) was used without frozen core
approximations for correlated procedures. The CCSD(T)/cc-
pCVQZ level of theory is known to be highly accurate for
equilibrium geometries of single-reference molecules.55–57
Harmonic vibrational frequencies were determined at the
optimized geometries via numerical differentiation of ana-
lytic gradients at the CCSD(T) level. For all relevant species,
fundamental (anharmonic) vibrational frequencies were com-
puted at the CCSD(T)/cc-pCVQZ level.
Focal point analyses58–62were performed using HF,
CCSD, CCSD(T), CCSDT, and CCSDT(Q) (Refs. 63 and 64)
methods with cc-pVXZ (Refs. 65 and 66) (X = D, T, Q, 5,
and 6) basis sets with the frozen core approximation at the cc-
pCVQZCCSD(T)geometries.Thelargestbasisset,cc-pV6Z,
included 371 contracted Gaussian functions. In the focal point
analyses, for the basis set extrapolation of Hartree-Fock ener-
gies, the functional form of Feller67was used:
quadruple-ζ
EHF= a + be−cX,
(1)
where a, b, and c are the fitting parameters and X is the cardi-
nal number of the cc-pVXZ basis set. For the extrapolation of
correlation energies, the functional form of Helgaker et al.68
was used:
Ecorr= d + eX−3,
(2)
where d and e are the fitting parameters and X is the cardi-
nal number of the cc-pVXZ basis set. In the main focal point
analyses, only valence electrons were correlated. Therefore,
a core correlation correction (?core) was determined by dif-
ferencing all-electron and frozen-core CCSD(T)/cc-pCVQZ
single point energies.
Diagonal Born-Oppenheimer corrections69,70(?DBOC)
were evaluated at the Hartree-Fock level using the cc-pCVQZ
basis set. Relativistic effects (?rel) were also taken into ac-
count by appending mass-velocity and Darwin one-electron
terms71,72computed with the CCSD(T)/cc-pCVQZ method.
Final reaction energies and barriers heights were deter-
mined according to
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
(3)
where ?Efinal is the final energy including all corrections,
?ECBSis the classical relative energy at the complete basis
set (CBS) limit, and ?AZPVEis the anharmonic ZPVE correc-
tion.
The CFOUR program package73was utilized for all
theoretical treatments. CCSDT(Q) computations were ac-
complished by using Kállay’s multireference coupled-cluster
method interfaced with the CFOUR package.64,74–77The non-
linear transformation of the quartic force field from symmetry
internal coordinates to Cartesian coordinates78was performed
using INTDER2005.79–81The ANHARM82program provided
the subsequent second-order vibrational perturbation theory
(VPT2) analysis.62For the transition structures (TSs) funda-
mental frequencies (including the imaginary frequency) were
obtained using the formulas of Miller et al.83
III. THEORETICAL KINETIC MODELING
Rate constants for the isomerization and dissociation
reactions were computed using both canonical transition-
state theory (CTST) and Rice-Ramsperger-Kassel-Marcus
(RRKM) theory.84,85The CTST (kCTST) and RRKM (kuni, ∞)
rate constant computations were performed using the MULTI-
WELL program package.86–88The CTST rate constant is de-
fined as
kCTST=kBT
h
exp
?
−?G‡
RT
?
,
(4)
where kBis the Boltzmann constant, h is Planck’s constant,
?G‡is the Gibbs free energy of the TS relative to the reac-
tant, R is the ideal gas constant, and T is the temperature. The
RRKM rate constant is expressed as
?∞
where Q(T) is the partition function for the reactant’s internal
degrees of freedom (vibration and two-dimensional active ro-
tation), ρ(E) is the density of states of the reactant molecule,
E0is the reaction threshold energy, and k(E) is the micro-
canonical rate constant. Finally, the microcanonical rate con-
stant is given by
kuni,∞=
1
Q(T)
E0
k(E)ρ(E)exp
?
−
E
kBT
?
dE,
(5)
kE=m‡
m
σext
σ‡
ext
g‡
e
ge
1
h
G‡(E − E0)
ρ(E)
,
(6)
where m‡and m are the numbers of optical isomers (chiral
stereoisomer); σ‡
try numbers; g‡
the TS and the reactant, respectively, and G‡(E − E0) is the
sum of states for the TS. The value for kuni, ∞is formally iden-
tical to that given by CTST, however, RRKM rate constant
extand σextare the external rotation symme-
eand geare the electronic state degeneracies of
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 3

164303-3 Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
computations include tunneling effects via an unsymmetrical
Eckart barrier.89
In order to better appreciate the singlet and triplet
potential energy surfaces (PESs), simultaneous first-order
ordinary-differential rate equations are solved numerically to
obtain time-dependent concentrations, hence the product dis-
tributions at a given temperature. The MATLAB 7.0.4 program
package90is employed to solve the coupled rate equations. Si-
multaneous first-order ordinary-differential rate equations can
be written as follows:
dyi
dt
= −kiyi +
?
j?=i
kjiyj,
(7)
ki=
?
j?=i
kij,
(8)
where yiis the concentration of the ith species, kijis the rate
constant for the i → j reaction, and t is the time.
IV. RESULTS AND DISCUSSION
We carried out T1-diagnostics91–93to see whether there
is a need for multireference methods for any structure con-
sidered in this research. The present T1-diagnostic results are
0.013 HNO (? X1A?); 0.013 HNO (˜ a3A??); 0.021 NOH (˜ a1A?);
0.015 HNO/NOH (3A??) (TS for B); 0.010 HNO/H+NO (3A??)
(TS for D); 0.032 NOH/H+NO (1A?) (TS for E); and 0.013
NOH/H+NO (3A??) (TS for F). These T1-diagnostics values
indicate that the CCSD(T) level of theory should perform well
fortheHNO–NOHsystem.Although,thetransitionstructures
HNO/NOH (1A?) and NOH/H+NO (1A?) have slight multiref-
erence characters, we can still trust on CCSD(T) as previously
discussed by Alikhani et al.23
0.010 NOH (? X3A??); 0.029 HNO/NOH (1A?) (TS for A);
A. Geometries
Our prediction for the diatomic NO (X2?) bond distance
is reported in Table I, while the geometrical structures for the
HNO(? X1A?),HNO(˜ a3A??),NOH(˜ a1A?),NOH(? X3A??),and
optimized geometries for the TSs at the cc-pCVQZ CCSD(T)
level of theory are depicted in Figures 1–5.
TSs for the reactions (A)–(F) are presented in Table II. The
1. NO (X2?)
We first consider structures that are well established from
experiment, in order to assess the reliability of the methods
TABLE I. Predicted total energy (in hartrees), bond distance (in Å), har-
monic and fundamental (anharmonic) vibrational frequencies (in cm−1), and
zero-point vibrational energies (HZPVE and AZPVE in kcal mol−1) for the
NO (X2?) molecule at the cc-pCVQZ CCSD(T) level of theory.
Molecule
Etot
re
ωe
νe
HZPVEAZPVE
NO
Experimenta
−129.873 0361.1507
1.1508
1919
1904
1893
1876
2.74
2.72
2.73
2.71
aHuber and Herzberg.94
TABLE II. Predicted total energies (in hartrees) and geometries (equilib-
rium bond distances in Å, angles in degrees) for the HNO (? X1A?), HNO
(1A?), NOH (? X3A??), NOH/H+NO (1A?), and NOH/H+NO (3A??) stationary
Molecule
Etot
(˜ a3A??), HNO/NOH (1A?), HNO/NOH (3A??), HNO/H+NO (3A??), NOH
points at the cc-pCVQZ CCSD(T) level of theory.
rNH
rNO
θHNO
1HNO
−130.456 588 1.0524
1.0508a
1.0484b
1.0557c
1.0535d
1.0518e
1.0509f
1.0628
1.0903
1.0246
1.2722
1.2085
1.2075a
1.2136b
1.2044c
1.2107d
1.2075e
1.2097f
1.2116
1.2090
1.2220
1.3319
108.08
108.01d
108.19e
108.30f
108.60
108.05
119.30
49.99
Experimentg
Experimenth
3HNO
1HNO/1NOH
(TS–A)
−130.427 592
−130.339 225
1.2711f
1.2731i
1.2146
1.3335f
1.3350i
1.3278
50.20f
62.10i
55.53
3HNO/3NOH
(TS–B)
3HNO/H+NO
(TS–D)
−130.360 536
−130.367 7841.7879 1.1581 116.74
rOH
rNO
θNOH
110.42
110.30f
109.90j
107.47
119.96
1NOH
−130.388 8540.9854
0.9837f
0.9860j
0.9676
1.4281
1.2626
1.2653f
1.2760j
1.3255
1.1582
3NOH
1NOH/H+NO
(TS–E)
−130.416 518
−130.368 437
1.3711f
1.4969
1.1647f
1.1841
119.30f
118.19
3NOH/H+NO
(TS–F)
−130.353 471
aCCSD(T)/CBS level from Shepard et al.42
bMRCISD/CBS level from Shepard et al.42
cMRAQCC/CBS level from Shepard et al.42
dCCSD(T)/cc-pVQZ level from Dateo et al.22
eCCSD(T)/CBS level from Demaison et al.40
ficMRCI+Q/aug-cc-pVQZ level from Li et al.43
gDalby.5
hOgilvie.95
iCCSD(T)/cc-pVQZ level from Alikhani et al.23
jCCSD(T)/TZ2P level from Lee.21
N
O
1.272
1.332
1.102
FIG. 1. Transition state for the unimolecular isomerization
−→1NOH at the cc-pCVQZ CCSD(T) level (bond distances are in Å). The
energy of this structure lies 70.1 kcal mol−1above1HNO. Thus, a more fa-
vorable mechanism is1HNO −→ H + NO −→1NOH. See Figure 6.
1HNO
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 4

164303-4 Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
N
O
1.428
120.0°
1.158
FIG. 2. Transition state for1NOH −→ H + NO at the cc-pCVQZ CCSD(T)
level (bond distances are in Å). The energy of this structure lies 7.9 kcal
mol−1above1NOH. See Figure 6.
used here. Our prediction for the bond length of NO (re
= 1.1507 Å) is only 0.0001 Å shorter than the experimental
value94of 1.1508 Å. Thus, the CCSD(T)/cc-pCVQZ method
providesanexceptionallyaccurateresultforthebonddistance
of NO (X2?).
2. HNO (? X
HNO(? X1A?)geometrywithexperimentaldeductionsarepro-
of HNO is 0.0104 and 0.0379 Å shorter than Dalby’s experi-
mental value5of 1.0628 Å and Ogilvie’s experimental value95
of 1.0903 Å, respectively. The experimental values were ob-
tained from the rotational analyses. However, in his paper,
Ogilvie determined the N–D bond distance as 1.0795 Å for
the DNO molecule and noted that unusually large difference
in the rNHand rNDbond distances likely indicates the partial
failure of some approximation in that analysis.
In their theoretical study of 2008, Shepard et al.42
predicted the N–H bond distance of HNO as 1.0508,
1.0484, and 1.0557 Å at the CBS limit of CCSD(T),
multireference configuration interaction singles and doubles
(MRCISD), and multireference averaged quadratic coupled-
cluster (MRAQCC) methods, respectively. Our prediction for
theN–HbonddistanceofHNOis0.0016and0.0040Ålonger
than their CBS CCSD(T) and MRCISD results and 0.0033 Å
1A?)
Comparisons of the present theoretical results for the
vided in Table II. The N–H bond distance (rNH= 1.0524 Å)
N
O
1.189
1.215
1.328
FIG. 3. Transition state for the unimolecular isomerization
−→3NOH at the cc-pCVQZ CCSD(T) level (bond distances are in Å). The
energy of this structure lies 39.3 kcal mol−1above3HNO. See Figure 7.
3HNO
N
O
1.788
116.7°
1.158
FIG. 4. Transition state for3HNO −→ H + NO at the cc-pCVQZ CCSD(T)
level (bond distances are in Å). The energy of this structure lies 32.4 kcal
mol−1above3HNO. See Figure 7.
shorter than the CBS MRAQCC result. Further, our rNHvalue
is only 0.0006 Å longer than CBS CCSD(T) result of De-
maison et al.40Thus, our N–H bond distance is in reasonable
agreement with both Dalby’s experiment and with high level
theoretical predictions. Ogilvie’s experimental deduction of
1.0903 Å for rNHappears to be not as reliable.
The N–O bond distance (rNO= 1.2085 Å) of HNO
is 0.0031 and 0.0005 Å shorter than Dalby’s experimental
value5of 1.2116 Å and Ogilvie’s experimental value95of
1.2090 Å, respectively. Further, our rNOvalue is 0.0010 and
0.0041 Å longer than the CBS CCSD(T) (Shepard et al.42
and Demaison et al.40) and CBS MRAQCC results (Shep-
ard et al.42). However, our present rNO value is 0.0051 Å
shorter than the CBS MRCISD estimation of Shepard et al.42
Our prediction for the H–N–O bond angle (θHNO= 108.08◦)
differs by 0.52◦and 0.03◦from the experimental values of
108.60◦(Dalby) and 108.05◦(Ogilvie).
3. NOH (˜ a1A?)
Our result for the O–H bond distance (rOH= 0.9854
Å) of1NOH is only 0.0006 Å shorter than TZ2P CCSD(T)
N
O
1.184
1.497
118.2°
FIG. 5. Transition state for3NOH −→ H + NO at the cc-pCVQZ CCSD(T)
level (bond distances are in Å). The energy of this structure lies 33.7 kcal
mol−1above3NOH. See Figure 7.
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 5

164303-5 Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
value of Lee,21while 0.0017 Å longer than aug-cc-pVQZ
icMRCI+Q result of Li et al.43On the other hand, our rNO
value of1NOH, 1.2626 Å, is 0.0134 and 0.0027 Å shorter
than predictions of Lee21and Li et al.,43respectively. There-
fore, the optimized geometry of the NOH (˜ a1A?) species is in
general agreement with previous theoretical studies of Lee21
and Li et al.43
4. Transition states
The optimized geometries of the isomerization transition
state1HNO/1NOH (1A?) (TS–A) is in agreement with theoret-
ical studies of Alikhani et al.23and Li et al.43Our rNHvalue,
1.2722 Å, is 0.0009 Å shorter than the cc-pVQZ CCSD(T)
value of Alikhani et al.,23while 0.0011 Å longer than the
aug-cc-pVQZ icMRCI+Q result of Li et al.43Further, our
rNOvalue of1NOH, 1.3319 Å, is 0.0031 and 0.0016 Å shorter
than the predictions of Alikhani et al.23and Li et al.,43re-
spectively. For the dissociation transition state NOH/H+NO
(1A?) (TS–E) the rOHbond distance, 1.4281 Å, is 0.0570 Å
longer than the aug-cc-pVQZ icMRCI+Q result of Li et al.,43
while our rNOvalue, 1.1582 Å, is 0.0065 Å shorter than their
prediction.43
B. Harmonic and fundamental vibrational frequencies
For the diatomic NO (X2?) molecule our theoretical har-
monic and fundamental vibrational frequencies are reported
in Table I. The harmonic and fundamental frequencies of
NO, 1919 and 1893 cm−1, differ by 15 and 17 cm−1from
the experimental94values of 1904 and 1876 cm−1, respec-
tively. Thus, the predicted vibrational frequencies for the NO
molecule are in reasonable agreement with experiments.
For the HNO (? X1A?), HNO (˜ a3A??), HNO/NOH (1A?),
(? X3A??), and NOH/H+NO (3A??) stationary points theoreti-
ported in Table III. For the HNO (? X1A?) molecule predicted
ω2, and ω3, respectively, while fundamental frequencies are
2668, 1511, and 1576 cm−1for ν1, ν2, and ν3, respectively.
The theoretical fundamental frequencies differ by only 16, 10,
and 11 cm−1from the experimental values of 2684,301501,29
and 1565 (Ref. 29) cm−1for ν1, ν2, and ν3, respectively.
The present harmonic frequencies are in reasonable agree-
ment with computations of Keçeli et al.,41?ω1= 6 and 51,
?ω2= 4 and 20, and ?ω3= 5 and 19 cm−1at CCSD(T)/cc-
pVQZ and CCSD(2)T/aug-cc-pVQZ levels, respectively.
Moreover, our harmonic frequencies of NOH (˜ a1A?) (ω1
= 3317, ω2= 1289, and ω3= 1473 cm−1) are in reasonable
agreement with computations of Lee21at the CCSD(T)/TZ2P
level, ?ω1= 2, ?ω2= 51, and ?ω3= 11 cm−1. For the
HNO/NOH (1A?) (TS–A) frequencies, the agreement with
Alikhani et al.23at the CCSD(T)/cc-pVQZ level is ?ω1
= 13, ?ω2= 27i, and ?ω3= 13 cm−1.
For isotopic species harmonic and fundamental fre-
quencies are reported in Tables S1 and S2 of the sup-
porting information.96For the DNO molecule fundamental
HNO/NOH (3A??), HNO/H+NO (3A??), NOH (˜ a1A?), NOH
cal harmonic and fundamental vibrational frequencies are re-
harmonic frequencies are 2960, 1551, and 1609 cm−1for ω1,
TABLE III. Harmonic and fundamental (anharmonic) vibrational frequencies (in cm−1) and zero-point vibra-
tional energies (HZPVE and AZPVE) (in kcal mol−1) for the HNO (? X1A?), HNO (3A??), HNO/NOH (1A?),
stationary points at the cc-pCVQZ CCSD(T) level of theory.
HNO/NOH (3A??), HNO/H+NO (3A??), NOH (1A?), NOH (? X3A??), NOH/H+NO (1A?), and NOH/H+NO (3A??)
Molecule
ω1
ω2
ω3
ν1
ν2
ν3
HZPVEAZPVE
1HNO2960
2954a
2909b
1551
1547a
1531b
1609
1604a
1590b
2668151115768.75 8.23
2683c
2679d
2684e
2928
1503c
1499d
1501f
1238
1572c
1568d
1565f
1430
Experiment
1NOH 3317
3319g
2807
2820h
3347
3728
2606
895i
890i
1816i
1289
1238g
2228i
2201ih
1046
1138
2041i
403
1082
674
1473
1462g
1192
1174h
1590
1247
1239
1845
1804
1656
8.698.44
1HNO/1NOH (TS–A) 27402278i
11645.72 5.65
3HNO
3NOH
3HNO/3NOH (TS–B)
3HNO/H+NO (TS–D)
1NOH/H+NO (TS–E)
3NOH/H+NO (TS–F)
aCCSD(T)/cc-pVQZ level from Dateo et al.22
bCCSD(2)T/aug-cc-pVDZ level from Keçeli et al.41
cCC-CBS/hybrid/VCI level from Keçeli et al.41
dCC-CBS/QFF/VCI level from Keçeli et al.41
eJohns et al.30
fJohns et al.29
gCCSD(T)/TZ2P level from Lee.21
hCCSD(T)/cc-pVQZ level from Alikhani et al.23
3126
3523
2529
860i
774i
1781i
1011
1108
2083i
427
1144
689
1557
1215
1214
1816
1772
1629
8.56
8.74
5.50
3.21
4.13
3.33
8.40
8.60
5.44
3.16
4.07
3.26
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 6

164303-6 Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
TABLE IV. Focal point analysis for the HNO (? X1A?) −→ NOH (˜ a1A?) isomerization reaction (A) (energies in
Reaction energy
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
+33.18
cc-pVTZ
+32.92
cc-pVQZ
+32.94
cc-pV5Z
+32.94
cc-pV6Z
+32.94
CBS limit[+32.93]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 41.88 + 0.21 + 0.01 + 0.01 + 0.12 = +42.23 kcal mol−1
Reaction barrier
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
+85.75
cc-pVTZ
+85.16
cc-pVQZ
+85.28
cc-pV5Z
+85.29
cc-pV6Z
+85.29
CBS limit[+85.29]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 72.45 − 2.58 + 0.00 − 0.08 + 0.27 = +70.06 kcal mol−1
aThe symbol δ denotes the increment inthe energy difference (?E) with respect tothe previous level of theory. Bracketed numbers
are the results of basis set extrapolations, whereas unbracketed numbers were explicitly computed. For CCSDT and CCSDT(Q)
additivities are assumed using the cc-pCVQZ CCSD(T) data.
kcal mol−1).aAll single point energetics were carried out at the cc-pCVQZ CCSD(T) geometries.
δ[CCSDT]
−0.46
−0.59
[−0.59]
[−0.59]
[−0.59]
[−0.59]
δ[CCSDT(Q)]
+0.22
+0.19
[+0.19]
[+0.19]
[+0.19]
[+0.19]
?E[CCSDT(Q)]
[+43.42]
[+42.11]
[+41.97]
[+41.91]
[+41.90]
[+41.88]
+9.89
+8.82
+8.64
+8.57
+8.56
[+8.54]
+0.58
+0.77
+0.79
+0.80
+0.80
[+0.80]
δ[CCSDT]
−0.50
−0.45
[−0.45]
[−0.45]
[−0.45]
[−0.45]
δ[CCSDT(Q)]
−0.59
−0.71
[−0.71]
[−0.71]
[−0.71]
[−0.71]
?E[CCSDT(Q)]
[+72.38]
[+71.72]
[+72.20]
[+72.31]
[+72.37]
[+72.45]
−9.20
−8.99
−8.59
−8.48
−8.42
[−8.34]
−3.08
−3.29
−3.33
−3.34
−3.35
[−3.35]
frequencies are 2022, 1158, and 1561 cm−1for ν1, ν2, and
ν3, respectively. The theoretical fundamental frequencies dif-
fer by only 3, 5, and 14 cm−1from the experimental values of
2025,291153,28and 1547 (Ref. 29) cm−1for ν1, ν2, and ν3,
respectively. Our predicted fundamental frequencies are thus
in good agreement with the experiments.28,29
C. Energetics and kinetics
All considered reactions have activation barriers, except
for the reaction (C), which takes place with a barrierless path,
as previously discussed by Walch and Rohlfing.9
1. The HNO (? X
a. Energetics.
tion energy and barrier are reported in Table IV. The focal
point analysis for the classical endothermicity yields a HF
limit of ?E = +32.93 kcal mol−1. At correlated levels, the
?E contribution limits are δ[CCSD] = +8.54, δ[CCSD(T)]
= +0.80, δ[CCSDT] = −0.59, and δ[CCSDT(Q)] = +0.19
kcal mol−1. The magnitude of the subsequent δ[CCSDT] and
δ[CCSDT(Q)] increments point to convergence in the direc-
tion of the full configuration interaction (FCI) limit. The final
extrapolated value is ?E = +41.88 kcal mol−1for the classi-
cal reaction energy.
The HF limit for the unimolecular HNO −→ NOH reac-
tion barrier is ?E = +85.29 kcal mol−1. The necessity for
the treatment of electron correlation is seen in the succes-
sive CBS correlation corrections to the classical barrier. These
are δ[CCSD] = −8.34, δ[CCSD(T)] = −3.35, δ[CCSDT]
= −0.45, and δ[CCSDT(Q)] = −0.71 kcal mol−1. Our fo-
cal point analysis yields ?E = +72.45 kcal mol−1for the
1A?)?
?NOH (˜ a1A?) isomerization
reaction
Focal point analyses for the isomeriza-
classical reaction barrier. The classical reaction barrier in
the present study is in reasonable agreement with Alikhani
et al.’s23prediction of 73.4 kcal mol−1at the CCSD(T)/cc-
pVQZ level.
The AZPVE (?AZPVE), DBOC (?DBOC), relativistic
(?rel), and core (?core) corrections are 0.21, 0.01, 0.01, and
0.12 kcal mol−1for the reaction energy, respectively. For the
reaction barrier, the analogous corrections are −2.58, 0.00,
−0.08, and 0.27 kcal mol−1. Hence, our best prediction for
the isomerization reaction energy is 42.23 ±0.2 kcal mol−1
and that for the barrier is 70.06 ± 0.2 kcal mol−1. Note below
that the latter transition state lies much higher than that for the
HNO −→ H + NO −→ NOH two step process.
b. Kinetics.
carried out at 100–1500 K temperature range. Predicted rate
constants for the forward and reverse reactions are reported
in Table V. For the forward reaction the CTST (kCTST) and
RRKM (kuni) rate constants are 9.1 × 10−38and 3.9 × 10−31
s−1at 300 K. At up to 1200 K both rate constants are lower
than unity and the forward isomerization reaction is not favor-
able. As the temperature increases, 1200 K and higher, both
rate constants become large enough for the forward isomer-
ization reaction to occur.
For the reverse isomerization reaction the CTST (kCTST)
and RRKM (kuni) rate constants are 5.3 × 10−7and 9.7
× 10−3s−1at 300 K. At up to 500 K both rate constants are
lower than unity and the reverse isomerization reaction is not
favorable. As the temperature increases to 500 K and higher,
both rate constants become large enough for the reverse iso-
merization reaction to occur, confirming that the reverse iso-
merization reaction is kinetically far more favorable than the
forward isomerization reaction.
Kinetic modeling computations have been
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 7

164303-7 Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
TABLE V. Rate constants (in s−1) from canonical transition-state theory (CTST) (kCTST) and Rice-Ramsperger-
Kassel-Marcus (RRKM) theory (kuni, ∞) for the HNO (? X1A?) −→ NOH (˜ a1A?) and NOH (˜ a1A?) −→ HNO
HNO (? X1A?) −→ NOH (1A?)
Temperature
(? X1A?) isomerization reactions (A) (temperatures in K).
NOH (1A?) −→ HNO (? X1A?)
kCTSTa
kuni, ∞b
3.8 × 10−31
3.9 × 10−31
3.9 × 10−31
8.7 × 10−24
3.5 × 10−17
1.9 × 10−12
5.9 × 10−9
2.7 × 10−6
3.4 × 10−4
1.6 × 10−2
4.0 × 10−1
5.6 × 100
5.4 × 101
3.8 × 102
1.9 × 103
kCTSTa
kuni, ∞b
8.5 × 10−12
2.2 × 10−5
9.7 × 10−3
1.1 × 100
1.1 × 102
4.8 × 103
9.3 × 104
9.5 × 105
6.0 × 106
2.7 × 107
9.4 × 107
2.6 × 108
6.4 × 108
1.3 × 109
2.6 × 109
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
4.9 × 10−138
6.8 × 10−63
9.1 × 10−38
3.6 × 10−25
1.4 × 10−17
1.6 × 10−12
6.9 × 10−9
3.7 × 10−6
4.9 × 10−4
2.4 × 10−2
6.0 × 10−1
8.7 × 100
8.4 × 101
5.8 × 102
3.1 × 103
9.7 × 10−46
9.6 × 10−17
5.3 × 10−7
4.2 × 10−2
3.9 × 101
3.8 × 103
1.0 × 105
1.2 × 106
8.0 × 106
3.8 × 107
1.3 × 108
3.8 × 108
9.2 × 108
2.0 × 109
3.8 × 109
aWithout tunneling effects.
bWith tunneling effects.
c. Tunneling effects.
(kCTST) and RRKM (kuni) rate constants, at low temperatures
(below 300–400 K) kCTSTand kuniare quite different. How-
ever, as temperature increases kCTSTand kunibecome more
consistent. The reason for this disparity is that the RRKM
method includes some tunneling effects while CTST does not.
Thus, at low temperatures both the forward and reverse iso-
merization reactions are considerably influenced by tunneling
effects.
When we compare the CTST
d. Isotope effects.
effects, kinetic modeling computations for the DNO (? X1A?)
ization reactions were conducted. Estimated rate constants for
theisomerizationreactionsarereportedinTablesS7,S11,and
S12. With CTST the estimated rate constants are 6.5 × 10−39
and 1.5 × 10−8s−1for the forward and reverse isomerization
reactions of DNO at 300 K, respectively. With RRKM the
predicted rate constants are 3.9 × 10−31and 1.6 × 10−6s−1
for the forward and reverse isomerization reactions of DNO at
300 K, respectively. As in the case of HNO, at high tempera-
tures the rate constants of the two methods are in better agree-
ment due to tunneling effects becoming negligible. However,
at low temperatures tunneling effects are still important.
Both the DNO (in Table S7) and TNO (in Tables S11
and S12) species present slightly smaller rate constants than
HNO (in Table V). This is due to the fact that DNO and TNO
have smaller zero-point vibrational energies. As a result, ki-
netic isotope effects do not change the reaction rates signifi-
cantly and decrease the possibility of the forward and reverse
isomerization reactions.
In order to consider kinetic isotope
?
? NOD (˜ a1A?) and TNO (? X1A?) ?
? NOT (˜ a1A?) isomer-
2. The NOH (˜ a1A?) −→ H (2S) + NO (X2?)
dissociation reaction
a. Energetics.
energy and barrier are reported in Table VI. The focal point
Focal point analyses for the dissociation
analysis for the reaction energy yields a HF limit of ?E
= −2.40 kcal mol−1, but this is quite misleading (see below),
as the reaction in fact is endothermic. At correlated levels the
?E contribution limits are δ[CCSD] = +13.49, δ[CCSD(T)]
= −0.14, δ[CCSDT] = +0.45, and δ[CCSDT(Q)] = −0.07
kcal mol−1. The magnitude of the subsequent δ[CCSDT] and
δ[CCSDT(Q)] increments indicates convergence toward the
FCI limit. The final extrapolated value is ?E = +11.33 kcal
mol−1for the classical reaction energy.
The HFlimitforthe
=
ments are
δ[CCSD] = −22.61,
δ[CCSDT] = +0.30,
mol−1. Our focal point analysis presents ?E = +12.30 kcal
mol−1for the classical reaction barrier. With corrections in
Eq. (3), our best prediction for the reaction energy is 5.25
± 0.2 kcal mol−1and that for the barrier height is 7.88 ± 0.2
kcal mol−1.
reaction
again
barrieris
?E
+40.89kcal mol−1,misleading.
δ[CCSD(T)] = −6.63,
δ[CCSDT(Q)] = −1.33
Incre-
and kcal
b. Kinetics.
reaction are reported in Table VII. The CTST (kCTST) and
RRKM(kuni)rateconstantsare1.4×10−5and2.2×10−3s−1
at 100 K. At 200 K and higher temperatures, both rate con-
stants are significantly greater than unity and the dissociation
reaction is quite favorable. Hence, at these temperatures the
NOH (˜ a1A?) molecule is not stable and readily decomposes
to H + NO. Therefore, the synthesis of this species does not
appear to be possible under room temperature experimental
conditions.
Predicted rate constants for the dissociation
c. Tunneling effects.
(kuni) rate constants are significantly different at 100 K again
due to the tunneling effects. Thus, at 100 K the dissocia-
tion reaction is considerably affected by the tunneling effects.
However, at higher temperatures the tunneling effects are neg-
ligible.
The CTST (kCTST) and RRKM
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 8

164303-8Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
TABLE VI. Focal point analysis for the NOH (˜ a1A?) −→ H (2S) + NO (X2?) dissociation reaction (E)
(energies in kcal mol−1).aAll single point energetics were carried out at the cc-pCVQZ CCSD(T) geometries.
Reaction energy
δ[CCSD(T)]
−0.36
−0.25
−0.19
−0.16
−0.15
[−0.14]
Basis set
cc-pVDZ
cc-pVTZ
cc-pVQZ
cc-pV5Z
cc-pV6Z
CBS limit
?E[HF]
−4.71
−2.89
−2.60
−2.44
−2.41
[−2.40]
δ[CCSD]
+7.70
+11.87
+12.92
+13.25
+13.35
[+13.49]
δ[CCSDT]
+0.33
+0.45
[+0.45]
[+0.45]
[+0.45]
[+0.45]
δ[CCSDT(Q)]
−0.11
−0.07
[−0.07]
[−0.07]
[−0.07]
[−0.07]
?E[CCSDT(Q)]
[+2.84]
[+9.11]
[+10.51]
[+11.03]
[+11.17]
[+11.33]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 11.33 − 5.71 + 0.08 − 0.33 − 0.12 = +5.25 kcal mol−1
Reaction barrier
?E[HF]
δ[CCSD]
δ[CCSD(T)]
+40.89
+40.89
+40.89
+40.89
+40.89
[+40.89]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 12.30 − 4.37 + 0.00 − 0.10 + 0.05 = +7.88 kcal mol−1
aThe symbol δ denotes the increment inthe energy difference (?E) with respect tothe previous level of theory. Bracketed numbers
are the results of basis set extrapolations, whereas unbracketed numbers were explicitly computed. For CCSDT and CCSDT(Q)
additivities are assumed using the cc-pVTZ data.
Basis set
cc-pVDZ
cc-pVTZ
cc-pVQZ
cc-pV5Z
cc-pV6Z
CBS limit
δ[CCSDT]
+0.15
+0.30
[+0.30]
[+0.30]
[+0.30]
[+0.30]
δ[CCSDT(Q)]
−1.23
−1.33
[−1.33]
[−1.33]
[−1.33]
[−1.33]
?E[CCSDT(Q)]
[+6.43]
[+10.85]
[+11.75]
[+12.13]
[+12.21]
[+12.30]
−27.15
−23.90
−23.09
−22.79
−22.71
[−22.61]
−6.23
−6.59
−6.62
−6.62
−6.62
[−6.63]
d. Isotope effects.
effects, we also carried out kinetic modeling computations for
the NOD (˜ a1A?) −→ D (2S) + NO (X2?) and NOT (˜ a1A?)
−→ T (2S) + NO (X2?) dissociation reactions. Estimated
rate constants for the dissociation reactions are reported in
Tables S8 and S13. The predicted rate constants are 1.2 × 106
and 5.0 × 104s−1for the dissociation reaction of1NOD at
300 K with CTST and RRKM, respectively. As distinct from
1NOH, the rate constants of the two methods are in better
In order to consider kinetic isotope
TABLE VII. Rate constants (in s−1) from canonical transition-state the-
ory (CTST) (kCTST) and Rice-Ramsperger-Kassel-Marcus (RRKM) theory
(kuni, ∞) for the NOH (˜ a1A?) −→ H (2S) + NO (X2?) dissociation reaction
(E) (temperatures in K).
Temperature
kCTSTa
kuni, ∞b
2.2 × 10−3
2.7 × 104
1.2 × 107
2.8 × 108
2.0 × 109
7.6 × 109
2.0 × 1010
4.1 × 1010
7.2 × 1010
1.1 × 1011
1.6 × 1011
2.2 × 1011
2.8 × 1011
3.4 × 1011
4.1 × 1011
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1.4 × 10−5
1.2 × 104
1.3 × 107
4.9 × 108
4.4 × 109
2.0 × 1010
5.8 × 1010
1.3 × 1011
2.5 × 1011
4.2 × 1011
6.5 × 1011
9.3 × 1011
1.3 × 1012
1.6 × 1012
2.0 × 1012
aWithout tunneling effects.
bWith tunneling effects.
agreement even at 100 K, indicating that the tunneling effects
are negligible in the case of1NOD. This is due to the fact that
for1NOH the potential energy curve along the reaction coor-
dinate is steeper than those of1NOD with a higher imaginary
frequency.
Both the1NOD (in Table S8) and1NOT (in Table S13)
species present slightly smaller rate constants than1NOH (in
Table VII), since the1NOD and1NOT have lower zero-point
vibrational energies than1NOH. As a result, kinetic isotope
effects do not change reaction rate significantly and decrease
the possibility of the dissociation reaction. Clearly,1NOD
should be easier to observe than1NOH.
3. The HNO (˜ a3A??)?
reaction
?NOH (? X
Focal point analyses for the isomeriza-
3A??) isomerization
a. Energetics.
tion energy and barrier are reported in Table VIII. The
classical reaction energy at the HF limit is ?E = −10.39
kcal mol−1. At correlated levels the ?E contribution limits
are δ[CCSD] = +15.67, δ[CCSD(T)] = +1.83, δ[CCSDT]
= −0.04, and δ[CCSDT(Q)] = +0.35 kcal mol−1. The final
extrapolated value is ?E = +7.41 kcal mol−1for the triplet
endothermicity.
The HF limit for the reaction barrier is ?E = +49.16
kcal mol−1. The necessity for the treatment of elec-
tron correlation is seen in the successive CBS correla-
tion corrections to the classical barrier. These are δ[CCSD]
= −5.51, δ[CCSD(T)] = −1.50, δ[CCSDT] = −0.04, and
δ[CCSDT(Q)] = −0.08 kcal mol−1. Our focal point analy-
sis provides ?E = +42.03 kcal mol−1for the classical re-
action barrier. Final energetic predictions were carried out
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 9

164303-9Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
TABLE VIII. Focal point analysis for the HNO (˜ a3A??) −→ NOH (? X3A??) isomerization reaction (B) (energies
Reaction energy
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
−10.42
cc-pVTZ
−10.50
cc-pVQZ
−10.40
cc-pV5Z
−10.38
cc-pV6Z
−10.39
CBS limit[−10.39]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 7.41 + 0.20 − 0.09 − 0.04 + 0.25 = +7.73 kcal mol−1
Reaction barrier
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
+48.45
cc-pVTZ
+48.86
cc-pVQZ
+49.11
cc-pV5Z
+49.15
cc-pV6Z
+49.16
CBS limit[+49.16]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 42.03 − 2.96 + 0.01 − 0.13 + 0.36 = +39.31 kcal mol−1
aThe symbol δ denotes the increment inthe energy difference (?E) with respect tothe previous level of theory. Bracketed numbers
are the results of basis set extrapolations, whereas unbracketed numbers were explicitly computed. For CCSDT and CCSDT(Q)
additivities are assumed using the cc-pVTZ data.
in kcal mol−1).aAll single point energetics were carried out at the cc-pCVQZ CCSD(T) geometries.
δ[CCSDT]
+0.03
−0.04
[−0.04]
[−0.04]
[−0.04]
[−0.04]
δ[CCSDT(Q)]
+0.30
+0.35
[+0.35]
[+0.35]
[+0.35]
[+0.35]
?E[CCSDT(Q)]
[+5.15]
[+6.26]
[+6.98]
[+7.21]
[+7.29]
[+7.41]
+14.02
+14.84
+15.35
+15.50
+15.57
[+15.67]
+1.22
+1.60
+1.73
+1.78
+1.80
[+1.83]
δ[CCSDT]
−0.14
−0.04
[−0.04]
[−0.04]
[−0.04]
[−0.04]
δ[CCSDT(Q)]
−0.05
−0.08
[−0.08]
[−0.08]
[−0.08]
[−0.08]
?E[CCSDT(Q)]
[+39.57]
[+40.70]
[+41.60]
[+41.84]
[+41.92]
[+42.03]
−7.31
−6.51
−5.86
−5.68
−5.61
[−5.51]
−1.38
−1.53
−1.53
−1.51
−1.51
[−1.50]
according to Eq. (3). Our best prediction for the reaction en-
ergy is 7.73 ± 0.2 kcal mol−1and that for the isomerization
barrier is 39.31 ± 0.2 kcal mol−1.
b. Kinetics.
reverse isomerization reactions are reported in Table IX. For
the forward reaction the CTST (kCTST) and RRKM (kuni) rate
constants are 1.8 × 10−16and 1.5 × 10−12s−1at 300 K.
At up to 700 K both rate constants are lower than unity and
Predicted rate constants for the forward and
the forward isomerization reaction is not favorable. As tem-
perature increases, 700 K and higher, both rate constants be-
come large enough for the forward isomerization reaction to
occur.
For the3NOH −→3HNO reaction the CTST (kCTST) and
RRKM (kuni) rate constants are 7.3 × 10−11and 6.9 × 10−8
s−1at 300 K. At up to 600 K both rate constants are lower
than unity and the reverse isomerization reaction is not favor-
able. As temperature increases, 600 K and higher, both rate
TABLE IX. Rate constants (in s−1) from canonical transition-state theory (CTST) (kCTST) and Rice-
Ramsperger-Kassel-Marcus (RRKM) theory (kuni, ∞) for the HNO (˜ a3A??) −→ NOH (? X3A??) and NOH (? X3A??)
HNO (3A??) −→ NOH (? X3A??)
Temperature
−→ HNO (˜ a3A??) isomerization reactions (B) (temperatures in K).
NOH (? X3A??) −→ HNO (3A??)
kCTSTa
kuni, ∞b
3.5 × 10−31
5.4 × 10−20
1.5 × 10−12
5.9 × 10−8
1.7 × 10−4
6.8 × 10−2
5.8 × 100
1.6 × 102
2.6 × 103
2.2 × 104
1.3 × 105
5.7 × 105
1.9 × 106
5.5 × 106
1.3 × 107
kCTSTa
kuni, ∞b
3.9 × 10−31
1.0 × 10−14
6.9 × 10−8
6.6 × 10−4
4.0 × 10−1
4.4 × 101
1.5 × 103
2.2 × 104
1.9 × 105
1.1 × 106
4.3 × 106
1.4 × 107
3.7 × 107
8.5 × 107
1.7 × 108
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
3.2 × 10−74
5.8 × 10−31
1.8 × 10−16
3.4 × 10−9
8.3 × 10−5
7.1 × 10−2
8.9 × 100
3.4 × 102
5.7 × 103
5.5 × 104
3.7 × 105
1.7 × 106
6.2 × 106
1.9 × 107
5.1 × 107
2.4 × 10−57
1.5 × 10−22
7.3 × 10−11
5.4 × 10−5
1.9 × 10−1
4.3 × 101
2.1 × 103
4.0 × 104
3.9 × 105
2.5 × 106
1.1 × 107
3.9 × 107
1.1 × 108
2.8 × 108
6.2 × 108
aWithout tunneling effects.
bWith tunneling effects.
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 10

164303-10Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
TABLE X. Focal point analysis for the HNO (˜ a3A??) −→ H (2S) + NO (X2?) dissociation reaction (D)
(energies in kcal mol−1).aAll single point energetics were carried out at the cc-pCVQZ CCSD(T) geometries.
Reaction energy
δ[CCSD(T)]
−1.30
−1.06
−0.92
−0.86
−0.85
[−0.83]
Basis set
cc-pVDZ
cc-pVTZ
cc-pVQZ
cc-pV5Z
cc-pV6Z
CBS limit
?E[HF]
+21.93
+22.83
+22.95
+23.06
+23.06
[+23.05]
δ[CCSD]
+7.26
+11.22
+12.09
+12.30
+12.34
[+12.40]
δ[CCSDT]
+0.15
+0.24
[+0.24]
[+0.24]
[+0.24]
[+0.24]
δ[CCSDT(Q)]
−0.12
−0.14
[−0.14]
[−0.14]
[−0.14]
[−0.14]
?E[CCSDT(Q)]
[+27.90]
[+33.08]
[+34.21]
[+34.60]
[+34.66]
[+34.73]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 34.73 − 5.66 + 0.12 − 0.39 + 0.23 = +29.03 kcal mol−1
Reaction barrier
?E[HF]
δ[CCSD]
δ[CCSD(T)]
+32.04
+32.93
+33.02
+33.12
+33.12
[+33.11]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 37.48 − 5.24 + 0.05 − 0.13 + 0.25 = +32.41 kcal mol−1
aThe symbol δ denotes the increment inthe energy difference (?E) with respect tothe previous level of theory. Bracketed numbers
are the results of basis set extrapolations, whereas unbracketed numbers were explicitly computed. For CCSDT and CCSDT(Q)
additivities are assumed using the cc-pVTZ data.
Basis set
cc-pVDZ
cc-pVTZ
cc-pVQZ
cc-pV5Z
cc-pV6Z
CBS limit
δ[CCSDT]
+0.06
+0.15
[+0.15]
[+0.15]
[+0.15]
[+0.15]
δ[CCSDT(Q)]
−0.17
−0.19
[−0.19]
[−0.19]
[−0.19]
[−0.19]
?E[CCSDT(Q)]
[+32.72]
[+36.66]
[+37.30]
[+37.49]
[+37.49]
[+37.48]
+2.45
+5.35
+5.82
+5.89
+5.88
[+5.88]
−1.66
−1.58
−1.50
−1.48
−1.48
[−1.47]
constants become large enough so that the reverse isomeriza-
tion reaction occurs. Thus, these results indicate that the re-
verse isomerization reaction3NOH −→3HNO is kinetically
more favorable than the forward isomerization reaction.
c. Tunneling effects.
(kCTST) and RRKM (kuni) rate constants, at low temperatures
kCTSTand kuniappear to be different due to the tunneling ef-
fects. However, as temperature increases kCTSTand kunibe-
come more consistent. Thus, at low temperatures both the
forward and reverse isomerization reactions are considerably
under influence of the tunneling effects, whereas at high tem-
peratures the tunneling effects are practically negligible.
When we compare the CTST
d. Isotope effects.
effects, we also carried out kinetic modeling computations for
the DNO (˜ a3A??)?
?NOD (? X3A??) and TNO (˜ a3A??)?
theisomerizationreactionsarereportedinTablesS9,S14,and
S15. With CTST the estimated rate constants are 9.7 × 10−17
and 2.0 × 10−11s−1for the forward and reverse isomeriza-
tion reactions of3DNO at 300 K, respectively. With RRKM
the corresponding rate constants are 2.2 × 10−15and 4.6
× 10−10s−1, respectively. As in the case of3HNO, at high
temperatures the rate constants of the two methods are in bet-
ter agreement, because the tunneling effects become negligi-
ble. However, at low temperatures tunneling effects are still
important.
Both the3DNO and3TNO species present slightly lower
rate constants than3HNO, owing to their lower zero-point vi-
brational energies than3HNO. Consequently, kinetic isotope
effects do not change reaction rate significantly and decrease
In order to consider kinetic isotope
?NOT
(? X3A??) isomerization reactions. Estimated rate constants for
the possibility of the forward and reverse isomerization reac-
tions.
4. The HNO (˜ a3A??) −→ H (2S) + NO (X2?)
dissociation reaction
a. Energetics.
reaction energy and barrier are reported in Table X. The
focal point analysis for the classical reaction energy yields
a HF limit of 23.05 kcal mol−1. At correlated levels the
?E contribution limits are δ[CCSD] = +12.40, δ[CCSD(T)]
= −0.83, δ[CCSDT] = +0.24, and δ[CCSDT(Q)] = −0.14
kcal mol−1. The final extrapolated value is ?E = +34.73 kcal
mol−1for the classical reaction energy.
The HF limit for the dissociation barrier is ?E
= +33.11 kcal mol−1. Increments are δ[CCSD] = +5.88,
δ[CCSD(T)] = −1.47,
δ[CCSDT(Q)] = −0.19
analysis yields ?E = +37.48 kcal mol−1for the dissociation
barrier. With corrections in Eq. (3), our best prediction for
the reaction energy is 29.03 ± 0.2 kcal mol−1and that for the
dissociation barrier is 32.41 ± 0.2 kcal mol−1.
Focal point analyses for the dissociation
δ[CCSDT] = +0.15,
kcal mol−1.
and
pointOurfocal
b. Kinetics.
tion reaction are reported in Table XI. The CTST (kCTST) and
RRKM (kuni) rate constants are 3.1 × 10−11and 1.2 × 10−11
s−1at 300 K. At up to 600 K both rate constants are signifi-
cantly small and the dissociation reaction is not favorable. As
temperature increases, 600 K and higher temperatures, both
rate constants become large enough for the dissociation reac-
tion to occur. Especially, at 900 K and higher temperatures
rate constants for the dissociation reaction become large.
Predicted rate constants for the dissocia-
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 11

164303-11Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
TABLE XI. Rate constants (in s−1) from canonical transition-state the-
ory (CTST) (kCTST) and Rice-Ramsperger-Kassel-Marcus (RRKM) theory
(kuni, ∞) for the HNO (˜ a3A??) −→ H (2S) + NO (X2?) dissociation reaction
(D) (temperatures in K).
Temperature
kCTSTa
kuni, ∞b
3.9 × 10−31
3.1 × 10−23
1.2 × 10−11
1.0 × 10−5
4.1 × 10−2
1.1 × 101
5.9 × 102
1.2 × 104
1.3 × 105
8.7 × 105
4.1 × 106
1.5 × 107
4.2 × 107
1.0 × 108
2.1 × 108
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
5.7 × 10−59
3.0 × 10−23
3.1 × 10−11
3.5 × 10−5
1.6 × 10−1
4.6 × 101
2.7 × 103
5.7 × 104
6.3 × 105
4.3 × 106
2.1 × 107
7.7 × 107
2.3 × 108
6.1 × 108
1.4 × 109
aWithout tunneling effects.
bWith tunneling effects.
c. Tunneling effects.
(kuni) rate constants are significantly different at 100 K again
due to the tunneling effects. Thus, at 100 K the dissocia-
tion reaction is considerably affected by the tunneling effects,
whereas at higher temperatures the tunneling effects are neg-
ligible. Further, for the isomerization reactions on the singlet
and triplet potential energy surfaces, the kCTSTand kuniare in
reasonable agreement at 500–600 K and higher temperatures.
For the HNO (˜ a3A??) −→ H (2S) + NO (X2?) dissociation
The CTST (kCTST) and RRKM
reaction, both rate constants are similarly in reasonable agree-
ment at 200 K and higher temperatures. These findings indi-
cate that the tunneling effects are more important for the iso-
merization reaction than the dissociation reaction of3HNO.
d. Isotope effects.
(˜ a3A??)−→D(2S)+NO(X2?)andTNO(˜ a3A??)−→T(2S)
+ NO (X2?) dissociation reactions are reported in Tables
S10 and S16. The predicted rate constants are 3.7 × 10−12
and 1.1 × 10−12s−1for the dissociation reaction of3DNO at
300 K with CTST and RRKM, respectively. As in the case of
3HNO, at 200 K and higher temperatures the rate constants of
the two methods are in better agreement due to the negligible
tunneling effects.
Both the3DNO and3TNO species present slightly lower
rate constants than3HNO due to lower zero-point vibrational
energies than3HNO. As a result, isotopic substitutions of H
with D or T on the3HNO species decrease the probability of
the dissociation reaction.
Estimated rate constants for the DNO
5. The NOH (? X
a. Energetics.
ation reaction energy and barrier are reported in Table
XII. The classical reaction energy at the HF limit is ?E
= +33.45 kcal mol−1. At correlated levels the ?E contribu-
tion limits are not too large, δ[CCSD] = −3.26, δ[CCSD(T)]
= −2.66, δ[CCSDT] = +0.28, and δ[CCSDT(Q)] = −0.49
kcal mol−1. Our focal point analysis shows ?E = +27.32
kcal mol−1for the classical reaction energy.
3A??) −→ H (2S) + NO (X2?)
dissociation reaction
Focal point analyses for the dissoci-
TABLE XII. Focal point analysis for the NOH (? X3A??) −→ H (2S) + NO (X2?) dissociation reaction (F)
Reaction energy
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
+32.35
cc-pVTZ
+33.32
cc-pVQZ
+33.34
cc-pV5Z
+33.45
cc-pV6Z
+33.45
CBS limit[+33.45]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 27.32 − 5.87 + 0.21 − 0.34 − 0.02 = +21.30 kcal mol−1
Reaction barrier
Basis set
?E[HF]
δ[CCSD]
δ[CCSD(T)]
cc-pVDZ
+59.32
cc-pVTZ
+60.42
cc-pVQZ
+60.44
cc-pV5Z
+60.50
cc-pV6Z
+60.49
CBS limit[+60.49]
?Efinal= ?ECBS+ ?AZPVE+ ?DBOC+ ?rel+ ?core
= 38.88 − 5.34 + 0.19 − 0.10 + 0.04 = +33.67 kcal mol−1
aThe symbol δ denotes the increment in the energy difference (?E) with respect to the previous level of theory. Bracketed
numbers are the results of basis set extrapolations, whereas unbracketed numbers were explicitly computed. For CCSDT and
CCSDT(Q) additivities are assumed using the cc-pVTZ data.
(energies in kcal mol−1).aAll single point energetics were carried out at the cc-pCVQZ CCSD(T) geometries.
δ[CCSDT]
+0.11
+0.28
[+0.28]
[+0.28]
[+0.28]
[+0.28]
δ[CCSDT(Q)]
−0.43
−0.49
[−0.49]
[−0.49]
[−0.49]
[−0.49]
?E[CCSDT(Q)]
[+22.75]
[+26.83]
[+27.23]
[+27.40]
[+27.37]
[+27.32]
−6.76
−3.63
−3.26
−3.20
−3.23
[−3.26]
−2.52
−2.67
−2.64
−2.65
−2.65
[−2.66]
δ[CCSDT]
−0.10
+0.04
[+0.04]
[+0.04]
[+0.04]
[+0.04]
δ[CCSDT(Q)]
−0.46
−0.54
[−0.54]
[−0.54]
[−0.54]
[−0.54]
?E[CCSDT(Q)]
[+36.56]
[+39.13]
[+39.09]
[+39.07]
[+38.99]
[+38.88]
−19.03
−17.19
−17.16
−17.21
−17.26
[−17.34]
−3.17
−3.61
−3.69
−3.73
−3.74
[−3.77]
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 12

164303-12 Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
TABLE XIII. Rate constants (in s−1) from canonical transition-state the-
ory (CTST) (kCTST) and Rice-Ramsperger-Kassel-Marcus (RRKM) theory
(kuni, ∞) for the NOH (? X3A??) −→ H (2S) + NO (X2?) dissociation reac-
Temperature
kCTSTa
tion (F) (temperatures in K).
kuni, ∞b
3.6 × 10−31
3.1 × 10−20
5.4 × 10−11
1.0 × 10−5
2.6 × 10−2
5.9 × 100
3.1 × 102
6.3 × 103
6.7 × 104
4.5 × 105
2.2 × 106
7.8 × 106
2.3 × 107
5.7 × 107
1.2 × 108
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
7.9 × 10−62
9.9 × 10−25
2.8 × 10−12
5.1 × 10−6
3.1 × 10−2
1.1 × 101
7.2 × 102
1.7 × 104
2.0 × 105
1.5 × 106
7.4 × 106
2.9 × 107
9.1 × 107
2.5 × 108
5.8 × 108
aWithout tunneling effects.
bWith tunneling effects.
TheHF limitforthereactionbarrier is
?E
= +60.49 kcal mol−1. Increments with the correlated
methods are δ[CCSD] = −17.34, δ[CCSD(T)] = −3.77,
δ[CCSDT] = +0.04,
mol−1. The extrapolated value is ?E = +38.88 kcal mol−1
for the classical reaction barrier. With various corrections in
Eq. (3), our best prediction for the reaction energy is 21.30
± 0.2 kcal mol−1and that for the dissociation barrier is 33.67
± 0.2 kcal mol−1.
and
δ[CCSDT(Q)] = −0.54 kcal
b. Kinetics.
reaction are reported in Table XIII. The CTST (kCTST) and
RRKM (kuni) rate constants are 2.8 × 10−12and 5.4 × 10−11
s−1at 300 K. At up to 600 K both rate constants are lower
than unity and the dissociation reaction is not favorable. As
temperature increases, 600 K and higher temperatures, both
rate constants become large enough that the dissociation re-
action is kinetically possible. Especially, at 900 K and higher
temperatures the rate constants for the dissociation reaction
are substantially large.
Predicted rate constants for the dissociation
c. Tunneling effects.
action is considerably influenced by the tunneling effects,
whereas at higher temperatures the tunneling effects are neg-
ligible.
Below 500 K the dissociation re-
d. Isotope effects.
(˜ a3A??) −→ D (2S) + NO (X2?) and NOT (˜ a3A??) −→ T
(2S) + NO (X2?) dissociation reactions are reported in
Tables S10 and S17. The predicted rate constants are
2.2 × 10−13and 6.5 × 10−13s−1for the dissociation reac-
tion of3NOD at 300 K with CTST and RRKM, respectively.
Similar to3NOH, at 300 K and higher temperatures, the rate
Estimated rate constants for NOD
TABLE XIV. Product distribution for the singlet PES (initial molecule is
1HNO, temperatures in K, simulation time is 1000 s).
Temperature% (1HNO) % (1NOH)% (H + NO)
0a
0b
0c
0d
4e
100
100
100
100
100
100
100
100
100
100
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
100
100
100
100
96
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
aAt t ∼ 1032s it becomes 100%.
bAt t ∼ 1032s it becomes 100%.
cAt t ∼ 1020s it becomes 100%.
dAt t ∼ 1011s it becomes 100%.
eAt t ∼ 106s it becomes 100%.
constants of the two methods are in better agreement, because
the tunneling effects become negligible. However, for3NOH
the tunneling effects are important below 500 K, while for
3NOD they are important below 300 K. Thus, at low temper-
atures tunneling effects are more important for3NOH, due to
the fact that for3NOH the potential energy along the reaction
coordinate is steeper than that for3NOD (larger imaginary vi-
brational frequency).
Both the3NOD and3NOT species present slightly lower
rate constants than3NOH, owing to their lower zero-point
vibrational energies than3NOH. As a result, kinetic isotope
effects do not change the reaction rates significantly and de-
crease the possibility of the dissociation reaction.
TABLE XV. Product distribution for the singlet PES (initial molecule is
1NOH, temperatures in K, simulation time is 1000 s).
Temperature% (1HNO)% (1NOH) % (H + NO)
89a
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
aAt t ∼ 104s it becomes 100%.
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 13

164303-13Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
TABLE XVI. Product distribution for the triplet PES (initial molecule is
3HNO, temperatures in K, simulation time is 1000 s).
Temperature% (3HNO) % (3NOH)% (H + NO)
0a
0b
0c
1d
100
100
100
100
100
100
100
100
100
100
100
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
100
100
100
99
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
aAt t ∼ 1032s it becomes 100%.
bAt t ∼ 1024s it becomes 100%.
cAt t ∼ 1012s it becomes 100%.
dAt t ∼ 106s it becomes 100%.
6. The lowest-lying electronic singlet and triplet
potential energy surfaces
The lowest-lying electronic singlet and triplet PES for the
HNO–NOH system are depicted in Figures 6 and 7. All rela-
tive energies reported, including all corrections in Eq. (3), in
this subsection are defined with respect to the ground elec-
tronic state energy of HNO (? X1A?). On the singlet PES, the
iest path to get from1HNO to1NOH.1HNO decomposes to
H + NO without a barrier in excess of the endothermicity.
The relative energy is 47.48 kcal mol−1for H + NO. Thus,
barrier height for the unimolecular isomerization of1HNO to
1NOH is 70.06 kcal mol−1. However, this is far from the eas-
TABLE XVII. Product distribution for the triplet PES (initial molecule is
3NOH, temperatures in K, simulation time is 1000 s).
Temperature% (3HNO) % (3NOH)% (H + NO)
0a
0b
0c
1d
100
100
100
100
100
100
100
100
100
100
100
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
0
0
0
48
0
0
0
0
0
0
0
0
0
0
0
100
100
100
51
0
0
0
0
0
0
0
0
0
0
0
aAt t ∼ 1032s it becomes 100%.
bAt t ∼ 1024s it becomes 100%.
cAt t ∼ 1012s it becomes 100%.
dAt t ∼ 106s it becomes 100%.
E (kcal/mol)
Reaction Coordinate
1HNO
0.00
70.06
47.48
H + NO
1NOH
42.23
50.11
10
20
30
40
50
60
70
0
FIG. 6. Relative energy profile for the singlet PES for the HNO–NOH
system.
the1HNO molecule will decompose to H + NO upon heating
anditsunimolecularisomerizationreactiononthesingletPES
is not favorable. On the other hand, the1NOH molecule lies
42.23 kcal mol−1above the HNO (? X1A?). The barrier height
mol−1, while the1NOH dissociates to H + NO with an acti-
vation energy of 7.88 kcal mol−1. Therefore, for the1NOH
molecule the most favorable path is again decomposition to
H + NO.
In Table XIV the kinetic analyses starting with1HNO
demonstrate that the unique products are dissociation prod-
ucts (H + NO), and formation of1NOH is not possible at
standard temperature and pressure. Starting with1NOH (in
Table XV) the unique products are again dissociation prod-
ucts (H + NO), and formation of1NOH is unlikely at almost
all temperatures (200–1500 K). These product distributions
based on the simultaneous rate equations are consistent with
the energetic arguments mentioned above.
The relative energies for3HNO and3NOH are 18.45
and 26.18 kcal mol−1, respectively. On the triplet PES, the
for the isomerization reaction of1NOH to1HNO is 27.83 kcal
E (kcal/mol)
Reaction Coordinate
57.76
47.48
H + NO
3NOH
26.18
59.85
3HNO
18.45
50.86
10
20
30
40
50
60
70
0
FIG. 7. Relative energy profile for the triplet PES for the HNO–NOH system
[energies are relative to HNO (? X1A?)].
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 14

164303-14 Bozkaya et al.J. Chem. Phys. 136, 164303 (2012)
barrier height is 39.31 kcal mol−1for the isomerization re-
action of3HNO, while for the decomposition to H + NO is
32.41 kcal mol−1. Thus, on the triplet surface, again the dis-
sociation reaction is more favorable than the isomerization re-
action. In Table XVI the kinetic analyses (solution of time-
dependent rate equations) starting with3HNO demonstrate
that the3HNO molecule should be stable against both isomer-
ization and dissociation reactions at up to 400 K. However,
at higher temperatures the unique products are dissociation
products (H + NO), and the formation of3NOH is far less
probable. These product distributions based on the simultane-
ous rate equations confirm the energetic discussions described
above. On the other hand, for3NOH the barrier heights are
31.58 for the isomerization reaction and 33.67 kcal mol−1
for the dissociation reaction. Hence, at a first glance it may
appear that the isomerization reaction is more favorable than
the dissociation reaction. However, the kinetic analyses (solu-
tion of time-dependent rate equations) starting with3NOH in
Table XVII show that the3NOH molecule is stable at up to
400 K as in the case of3HNO, but, the unique products are
H + NO, and the formation of isomerization product is less
favorable at higher temperatures (T ≥ 500 K).
In summary, the singlet NOH molecule is dissociative
at room temperature, while the triplet NOH molecule is fa-
vored against isomerization and dissociation reactions at up to
400 K. Our theoretical findings help to explain why the1NOH
species has not been observed experimentally. Although the
formation of3NOH molecule from3HNO is not favorable,
3NOH might be observed experimentally at low temperatures
(at up to 400 K); experimental identification of3NOH may
thus be achievable under appropriate experimental conditions.
Such new experiments are certainly encouraged.
V. CONCLUSIONS
In this research, we have investigated the lowest-lying
electronic singlet and triplet potential energy surfaces for
the HNO–NOH system employing high level ab initio quan-
tum chemical methods. Focal point analyses were performed
to extrapolate the total energies to the CBS limit. AZPVE,
DBOC, relativistic effects, and core correlation effects were
also evaluated. The reaction energies and barriers have been
predicted for the two isomerization (A and B) and four disso-
ciation reactions (C–F). Unimolecular rate constants for the
isomerization and dissociation reactions were obtained utiliz-
ing several kinetic modeling methods.
On the singlet PES, the reaction energy and unimolec-
ular barrier, including all corrections, are predicted to be
42.23 ± 0.20 and 70.06 ± 0.20 kcal mol−1for the isomer-
ization reaction of1HNO. For the barrierless decomposition
of1HNO to H + NO the reaction energy is estimated to be
47.48 ± 0.20 kcal mol−1. For the dissociation of1NOH the
reaction energy and barrier are determined to be 5.25 ± 0.20
and 7.88 ± 0.20 kcal mol−1.The kinetic analyses demonstrate
that on the singlet PES the dissociation reaction is more favor-
able than the isomerization reaction and formation of1NOH
from1HNO is not probable.
On the triplet PES, the reaction energy and barrier, in-
cluding all corrections, are predicted to be 7.73 ± 0.20 and
39.31 ± 0.20 kcal mol−1for the isomerization reaction of
3HNO. For the dissociation reaction (products H + NO) anal-
ogous results are 29.03 ± 0.20 and 32.41 ± 0.20 kcal mol−1.
For the dissociation of3NOH the reaction energy and bar-
rier are estimated to be 21.30 ± 0.20 and 33.67 ± 0.20 kcal
mol−1. The kinetic analyses reveal that the dissociation re-
action is more favorable than the isomerization reaction and
formation of3NOH from3HNO is not probable.
Analyses of the kinetic results demonstrate that the iso-
merization and dissociation reactions are strongly affected by
tunneling effects at low temperatures, while at high temper-
atures the tunneling effects are negligible. Further, investiga-
tion of kinetic isotope effects demonstrates that substitutions
of H with D or T do not change reaction rate significantly and
rather decrease the possibility of the isomerization and disso-
ciation reactions.
Kinetic analyses based on solution of simultaneous first-
order ordinary-differential rate equations demonstrate that the
singlet NOH molecule is disfavored at any temperature, while
the triplet NOH molecule is viable against isomerization and
dissociation reactions at up to 400 K. Hence, the experimental
identification of3NOH may be achievable under appropriate
experimental conditions. Such new experiments are certainly
encouraged.
Finally, the lowest-lying singlet–triplet splittings (adia-
batic) are predicted to be 18.45 ± 0.20 and 16.05 ± 0.20
kcal mol−1, for the HNO (? X1A?– ˜ a3A??) and NOH (? X3A??
– ˜ a1A?) molecules, respectively.
ACKNOWLEDGMENTS
This research was supported by the U.S. Department
of Energy, Office of Basic Sciences, Division of Chemistry,
Combustion Program,GrantNo.DEFG02-97-ER14748. U.B.
would like to thank the Scientific and Technological Research
Council of Turkey (TÜB˙ITAK) for supporting his scientific
study at the University of Georgia, B˙IDEB-2214.
1J. A. Miller and C. T. Bowman, Prog. Energy Combust. Sci. 15, 287 (1989).
2K. M. Miranda, Coord. Chem. Rev. 249, 433 (2005).
3J. M. Fukuto, A. S. Dutton, and K. N. Houk, ChemBioChem 6, 612 (2005).
4J. M. Fukuto, C. H. Switzer, K. M. Miranda, and D. A. Wink, Annu. Rev.
Pharmacol. Toxicol. 45, 335 (2005).
5F. W. Dalby, Can. J. Phys. 36, 1336 (1958).
6H. W. Brown and G. C. Pimentel, J. Chem. Phys. 29, 883 (1958).
7H. B. Ellis and G. B. Ellison, J. Chem. Phys. 78, 6541 (1983).
8A. A. Wu, S. D. Peyerimhoff, and R. J. Buenker, Chem. Phys. Lett. 35, 316
(1975).
9S. P. Walch and C. M. Rohlfing, J. Chem. Phys. 91, 2939 (1989).
10P. J. Bruna and C. M. Maria, Chem. Phys. Lett. 67, 109 (1979).
11P. J. Bruna, Chem. Phys. 49, 39 (1980).
12A. Heiberg and J. Almlöf, Chem. Phys. Lett. 85, 542 (1982).
13O. Nomura and S. Iwata, Chem. Phys. Lett. 66, 523 (1979).
14K. Saito and K. Tagaki, J. Mol. Spectrosc. 47, 99 (1973).
15J. L. Bancroft, J. M. Hollas, and D. A. Ramsay, Can. J. Phys. 40, 322
(1962).
16M. J. Y. Clement and D. A. Ramsay, Can. J. Phys. 39, 205 (1961).
17J. K. Cashion and J. C. Polanyi, J. Chem. Phys. 30, 317 (1959).
18A. W. Salotto and L. Burnelle, J. Chem. Phys. 52, 2936 (1970).
19P. E.M. Siegbahn, J. Almlöf, A. Heiberg, and B. O. Roos, J. Chem. Phys.
74, 2384 (1981).
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Page 15

164303-15Bozkaya et al. J. Chem. Phys. 136, 164303 (2012)
20T. J. Lee, J. Chem. Phys. 99, 9783 (1993).
21T. J. Lee, Chem. Phys. Lett. 223, 431 (1994).
22C. E. Dateo, T. J. Lee, and D. W. Schwenke, J. Chem. Phys. 101, 5853
(1994).
23M.E.Alikhani,C.E.Dateo,andT.J.Lee,J.Chem.Phys.107,8208(1997).
24A. Luna, M. Merchán, and B. O. Roos, Chem. Phys. 196, 437 (1995).
25J. Ivanic and K. Ruedenberg, Theor. Chem. Acc. 120, 295 (2008).
26D. B. Hartley and B. A. Thrush, Proc. R. Soc. London, Ser. A 297, 520
(1967).
27P. N. Clough, B. A. Thrush, D. A. Ramsay, and J. G. Stamper, Chem. Phys.
Lett. 23, 155 (1973).
28M. Jacox and D. Milligan, J. Mol. Spectrosc. 48, 536 (1973).
29J. W. C. Johns and A. R.W. McKellar, J. Chem. Phys. 66, 1217 (1977).
30J. W. C. Johns, A. R. W. McKellar, and E. Weinberger, Can. J. Phys. 61,
1106 (1983).
31J. C. Petersen and M. Vervloet, Chem. Phys. Lett. 141, 499 (1987).
32A. W. Salotto and L. Burnelle, Chem. Phys. Lett. 3, 80 (1969).
33P. Botschwina, Mol. Phys. 32, 729 (1976).
34P. Botschwina, Chem. Phys. 40, 33 (1979).
35O. Nomura, Int. J. Quantum Chem. 18, 143 (1980).
36R. N. Dixon, K. B. Jones, M. Noble, and S. Carter, Mol. Phys. 42, 455
(1981).
37R. Guadagnini, G. Schatz, and S. P. Walch, J. Chem. Phys. 102, 774 (1995).
38D. H. Mordaunt, H. Flöthmann, M. Stumpf, H. M. Keller, C. Beck, R.
Schinke, and K. Yamashita, J. Chem. Phys. 107, 6603 (1997).
39J. Demaison, L. Margulés, and J. E. Boggs, Chem. Phys. 260, 65 (2000).
40J. Demaison, A. G. Császár, and A. Dehayem-Kamadjeu, J. Phys. Chem.
A 110, 13609 (2006).
41M. Keçeli, T. Shiozaki, K. Yagi, and S. Hirata, Mol. Phys. 107, 1283
(2009).
42R. Shepard, G. S. Kedziora, H. Lischka, I. Shavitt, T. Müller, P. G. Szalay,
M. Kállay, and M. Seth, Chem. Phys. 349, 37 (2008).
43A. Li, H. Han, and D. Xie, J. Chem. Phys. 135, 104304 (2011).
44R. Patrick and D. M. Golden, J. Phys. Chem. 88, 491 (1984).
45B. L. Ulich, J. M. Hollis, and L. E. Synder, Astrophys. J. 217, L105 (1977).
46B. E. Turner, Astrophys. J., Suppl. Ser. 76, 617 (1991).
47K. Andersson, P.-A. Malmqvist, B. O. Roos, A. J. Sadlej, and K. Wolinski,
J. Phys. Chem. 94, 5483 (1990).
48K. Andersson, P.-A. Malmqvist, and B. O. Roos, J. Chem. Phys. 96, 1218
(1990).
49K. Andersson and B. O. Roos, in Modern Electronic Structure Theory
Part II, Advanced Series in Physical Chemistry Vol. 2, edited by D. R.
Yarkony (World Scientific, London, 1995), pp. 55–109.
50B. O. Roos, in Theory and Applications of Computational Chemistry:
The First Forty Years, edited by C. Dykstra (Elsevier, New York, 2005),
pp. 725–764.
51G. E. Scuseria, Chem. Phys. Lett. 176, 27 (1991).
52K. Raghavachari, G. W. Trucks, J. A. Pople, and M. Head-Gordon, Chem.
Phys. Lett. 157, 479 (1989).
53G. E. Scuseria and T. J. Lee, J. Chem. Phys. 93, 5851 (1990).
54D. E. Woon and T. H. Dunning, J. Chem. Phys. 103, 4572 (1995).
55X. Zhang, A. T. Maccarone, M. R. Nimlos, S. Kato, V. M. Bierbaum,
G. B. Ellison, B. Ruscic, A. C. Simmonett, W. D. Allen, and H. F. Schaefer,
J. Chem. Phys. 126, 044312 (2007).
56K. L. Bak, J. Gauss, P. Jørgensen, J. Olsen, T. Helgaker, and J. F. Stanton,
J. Chem. Phys. 114, 6548 (2001).
57U. Bozkaya, J. M. Turney, Y. Yamaguchi, and H. F. Schaefer, J. Chem.
Phys. 132, 064308 (2010).
58A. L. L. East and W. D. Allen, J. Chem. Phys. 99, 4638 (1993).
59M. S. Schuurman, S. R. Muir, W. D. Allen, and H. F. Schaefer, J. Chem.
Phys. 120, 11586 (2004).
60J. M. Gonzales, C. Pak, R. S. Cox, W. D. Allen, H. F. Schaefer, A. G.
Császár, and G. Tarczay, Chem.-Eur. J. 9, 2173 (2003).
61A. G. Császár, W. D. Allen, and H. F. Schaefer, J. Chem. Phys. 108, 9751
(1998).
62A. G. Császár, G. Tarczay, M. L. Leininger, O. L. Polyansky, J. Tennyson,
and W. D. Allen, in Spectroscopy from Space, edited by J. Demaison and
K. Sarka (Kluwer, Dordrecht, 2007), pp. 317–340.
63J. Bomble, J. F. Stanton, M. Kállay, and J. Gauss, J. Chem. Phys. 123,
054101 (2005).
64M. Kállay and J. Gauss, J. Chem. Phys. 123, 214105 (2005).
65T. H. Dunning, J. Chem. Phys. 90, 1007 (1989).
66A. K. Wilson, T. van Mourik, and T. H. Dunning, J. Mol. Struct. 388, 339
(1996).
67D. Feller, J. Chem. Phys. 98, 7059 (1993).
68T. Helgaker, W. Klopper, H. Koch, and J. Noga, J. Chem. Phys. 106, 9639
(1997).
69N. C. Handy, Y. Yamaguchi, and H. F. Schaefer, J. Chem. Phys. 84, 4481
(1986).
70H. Sellers and P. Pulay, Chem. Phys. Lett. 103, 463 (1984).
71S. A. Perera and R. J. Bartlett, Chem. Phys. Lett. 216, 606 (1993).
72R. D. Cowan and D. C. Griffin, J. Opt. Soc. Am. 66, 1010 (1976).
73CFOUR, a quantum chemical program package written by J. F. Stanton, J.
Gauss, M. E. Harding, P. G. Szalay with contributions from A. A. Auer,
R. J. Bartlett, U. Benedikt, C. Berger, D. E. Bernholdt, Y. J. Bomble,
L. Cheng, O. Christiansen, M. Heckert, O. Heun, C. Huber, T.-C. Jagau,
D. Jonsson, J. Jusélius, K. Klein, W. J. Lauderdale, D. A. Matthews, T.
Metzroth, D. P. O’Neill, D. R. Price, E. Prochnow, K. Ruud, F. Schiffmann,
W. Schwalbach, S. Stopkowicz, A. Tajti, J. Vázquez, F. Wang, J. D. Watts
and the integral packages MOLECULE by J. Almlöf and P. R. Taylor, PROPS
by P. R. Taylor, ABACUS by T. Helgaker, H. J. Aa. Jensen, P. Jørgensen,
and J. Olsen, and ECP routines by A. V. Mitin and C. van Wüllen.
74M. Kállay and J. Gauss, J. Chem. Phys. 120, 6841 (2004).
75M. Kállay, P. G. Szalay, and P. R. Surjan, J. Chem. Phys. 117, 980
(2002).
76M. Kállay, J. Gauss, and P. G. Szalay, J. Chem. Phys. 119, 2991 (2003).
77M. Kállay and J. Gauss, J. Chem. Phys. 121, 9257 (2004).
78K. Sarka and J. Demaison, in Computational Molecular Spectroscopy,
edited by P. Jensen and P. R. Bunker (Wiley, Chichester, 2000), pp. 255–
304.
79W. D. Allen and A. G. Császár, J. Chem. Phys. 98, 2983 (1993).
80INTDER2005 is a general program written by W. D. Allen, which per-
forms sundry vibrational analyses and higher order nonlinear transforma-
tions among force field representations.
81W. D. Allen, A. G. Császár, V. Szalay, and I. M. Mills, Mol. Phys. 89, 1213
(1996).
82ANHARM is a fortran program written for VPT2 analysis by Y. Yamaguchi
and H. F. Schaefer, Center for Computational Quantum Chemistry, Univer-
sity of Georgia, Athens, GA, USA, 2009.
83W. H. Miller, R. Hernandez, N. C. Handy, D. Jayatilaka, and A. Willetts,
Chem. Phys. Lett. 172, 62 (1990).
84K. A. Holbrook, M. J. Pilling, and S. H. Robertson, Unimolecular Reac-
tions, 2nd ed. (Wiley, New York, 1996), pp. 39–77.
85J. I. Steinfeld, J. S. Francisco, and W. L. Hase, Chemical Kinetics and Dy-
namics, 2nd ed. (Prentice-Hall, New Jersey, 1999), pp. 287–382.
86MULTIWELL-2009.3 software, 2009, designed and maintained by John
R. Barker with contributors N. F. Ortiz, J. M. Preses, L. L. Lohr, A.
Maranzana, P. J. Stimac, and L. T. Nguyen, University of Michigan, Ann
Arbor, MI, see http://aoss.engin.umich.edu/multiwell/.
87J. R. Barker, Int. J. Chem. Kinet. 33, 232 (2001).
88J. R. Barker, L. M. Yoder, and K. D. King, J. Phys. Chem A 105, 796
(2001).
89W. H. Miller, J. Am. Chem. Soc. 101, 6810 (1979).
90MATLAB 7.0.4, The Language of Technical Computing, The MathWorks
Inc., Natick, MA, 2005.
91T. J. Lee, J. E. Rice, G. E. Scuseria, and H. F. Schaefer, Theor. Chem. Acc.
75, 81 (1989).
92T. J. Lee and P. R. Taylor, Int. J. Quantum Chem., Symp. 23, 199 (1989).
93D. Jayatilaka and T. J. Lee, J. Chem. Phys. 98, 9734 (1993).
94K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Struc-
ture, Constants of Diatomic Molecules, Vol. 4 (Van Nostrand, Princeton,
1979).
95J. F. Ogilvie, J. Mol. Struct. 31, 407 (1976).
96See supplementary material at http://dx.doi.org/10.1063/1.4704895 for ge-
ometries, harmonic, and anhormonic frequencies, and rate constants for
isotopic species.
Downloaded 25 Apr 2012 to 194.27.49.142. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions