Unprecedented Catalysis of Cs+Single Sites Confined in y Zeolite Pores for Selective Csp3-H Bond Ammoxidation: Transformation of Inactive Cs+Ions with a Noble Gas Electronic Structure to Active Cs+Single Sites

Shankha S. Acharyya, Shilpi Ghosh, Yusuke Yoshida, Takuma Kaneko, Takehiko Sasaki,*

Research Article

Unprecedented Catalysis of Cs Single Sites Conﬁned in Y Zeolite

Pores for Selective C −H Bond Ammoxidation: Transformation of

sp3

Inactive Cs Ions with a Noble Gas Electronic Structure to Active Cs

Single Sites

∥

,∥

∥

and Yasuhiro Iwasawa *

Cite This: ACS Catal. 2021, 11, 6698 −6708

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: We report the transformation of Cs ions with an

inactive noble gas electronic structure to active Cs single sites

chemically conﬁned in Y zeolite pores (Cs /Y), which provides an

unprecedented catalysis for oxidative cyanation (ammoxidation) of

C −H bonds with O and NH , although in general, alkali and

sp3

alkaline earth metal ions without a moderate redox property cannot

activate C −H bonds. The Cs /Y catalyst was proved to be highly

sp3

eﬃcient in the synthesis of aromatic nitriles with yields >90% in the

selective ammoxidation of toluene and its derivatives as test

reactions. The mechanisms for the genesis of active Cs single sites

and the ammoxidation pathway of C −H bonds were rationalized

sp3

by density functional theory (DFT) simulations. The chemical

conﬁnement of large-sized Cs ions with the pore architecture of a Y

zeolite supercage rendered the highest occupied molecular orbital (HOMO)−lowest unoccupied molecular orbital (LUMO) gap

reduction, HOMO component change, and preferable coordination arrangement for the selective reaction promotion, which

provides a trimolecular assembly platform to enable the coordination-promoted concerted ammoxidation pathway working closely

on each Cs single site. The new reaction pathway without involvement of O -dissociated O atom and lattice oxygen diﬀers from the

traditional redox catalysis mechanisms for the selective ammoxidation.

KEYWORDS: catalytic C −H activation, conﬁned Cs single site in Y zeolite pore, ammoxidation of toluene and its derivatives,

sp3

selective nitrile synthesis, catalysis mechanism

. INTRODUCTION

surfaces, making Cs−O bonds and reactive coordination.

Ammoxidation of methyl C −H bonds of aromatics with O

sp3

Alkali metal ions with inactive noble gas electronic structures

have been conceived as inactive species for selective

functionalization of C −H and C −H bonds to useful

NH is an avenue for the synthesis of organic nitriles, which

have been commercially used as common building blocks for

high-performance rubbers, polymers, and molecular electronics

and also as integral parts for producing pharmaceuticals,

agrochemicals, and ﬁne chemicals, such as vitamins, hetero-

sp2

sp3

chemicals in industry. Recently, we found a single alkali metal

ion site platform in a β zeolite pore that selectively promoted

benzene C −H activation toward phenol synthesis. The

sp2

catalytic performance of alkali metal ion sites incorporated in

zeolite pores depended strongly on the kind of zeolites, and

ZSM-5, mordenite, and Y zeolites were almost inactive as

cycles, and various carboxylic acid derivatives. Generally,

organic nitriles were synthesized by cyanation of aldehydes

using toxic hydrogen cyanide and metal cyanides, which caused

supports. No C −H activation catalysis of alkali metal ions

and oxides has been achieved to date, although a variety of

3,6−9

sp3

environmental disasters.

Industrially, the vapor-phase

eﬀects of alkali metal additives on metal oxide catalysis are

known. Here, we report the ﬁrst example of the catalytic

Received: February 28, 2021

Revised: April 26, 2021

Published: May 24, 2021

selective activation of C −H bonds to produce C−N bonds

sp3

on Cs single ion sites in Y zeolite pores (Cs /Y) due to

transformation of inactive Cs ions with a noble gas electronic

structure to active Cs sites with HOMO(O 2p)−LUMO(Cs

s) by chemical conﬁnement of Cs ions at Y zeolite pore

2021 American Chemical Society

698

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

Research Article

Table 1. Catalytic Performances of Various Alkali and Alkaline Earth Metal Ion Single Sites in Diﬀerent Zeolite Pores and

a b

Hydrotalcite Interlayers and on a SiO ·Al O Surface for the Toluene Ammoxidation to Benzonitrile (PhCN) at 623 K

entry

catalyst

Y zeolite

toluene conv./%

PhCN selec./%

PhCN syn. rate/μmol s⁻g⁻¹

NH /PhCN

cat

2.2

5.1

9.0

11.2

4.9

17.6

47.1

77.1

31.3

1.0

Mg (2 wt %)/Y

97.8

97.5

97.0

97.5

98.9

98.5

98.1

98.5

97.3

79.3

86.1

0.017

0.04

0.08

0.09

0.04

0.14

0.40

0.60

0.23

0.006

18.1

19.4

10.0

14.6

2.8

2.1

1.2

1.0

2.1

Ca (2 wt %)/Y

Sr (2 wt %)/Y

Ba (2 wt %)/Y

Na (2 wt %)/Y

K (2 wt %)/Y

Rb (2 wt %)/Y

Cs (2 wt %)/Y

Cs (2 wt %)/β

Cs (2 wt %)/ZSM-5

173

411

Cs (2 wt %)/Mor

1.0

0.2

Cs (2 wt %)/Hydr

Cs (2 wt %)/SiAlO

0.03

34.8

93.7

94.6

Cs (2 wt %)/Y

98.9

96.6

98.0

0.27

0.71

0.36

2.0

5.0

1.3

Cs (2 wt %)/Y

Y = Y zeolite with SiO /Al O molar ratio of 30, purchased from Zeolyst Co. Conv. = conversion; selec. = selectivity; NH /PhCN = molar ratios

of consumed NH to synthesized PhCN; zeolites = Y, β, ZSM-5, and mordenite (Mor); Hydr = hydrotalcite; SiAlO = SiO ·Al O with the same

SiO /Al O molar ratio (30) as the Y zeolite. Performance values were averaged during 30−240 min time-on-stream. Reaction conditions:

toluene/O /NH /He = 0.20/1.0/1.8/5.0 (mL min ); cat. = 0.2 g. Reaction conditions: 603 K. Reaction conditions: 673 K. Reaction

conditions: 0.4 g, 623 K. Reaction conditions: toluene + O ; toluene/O /He = 0.20/1.0/5.0 (mL min ). Reaction conditions: toluene + NH ;

toluene/NH /He = 0.20/1.8/5.0 (mL min ). Reaction conditions: O + NH ; O /NH /He = 1.0/1.8/5.0 (mL min ). NH conversion/%.

−

−1

−

−1

ammoxidation of toluene to benzonitrile has been extensively

studied with metal and metal oxide catalysts of V, Cr, Mo,

2. EXPERIMENTAL SECTION

.1. Catalyst Synthesis. Zeolite Y-supported alkali metal

ion and alkaline earth metal ion catalysts were prepared by a

0−25

etc.,

whose catalysts possess moderate redox potentials

and suﬃcient M−O bond strengths to provide active lattice

reliable ion-exchange method, followed by a controlled

treatment at 573 K, using the corresponding metal nitrates as

oxygen and oxygen atoms at the catalyst surfaces for the redox

catalysis. The Cs /Y catalyst without a beneﬁcial redox

precursors. A typical Cs (2 wt %)/Y sample was synthesized as

property showed a high benzonitrile yield (92.7% yield;

follows. CsNO

(0.88 g) (Kanto Chemical Co.) was dissolved

in 10 mL of deionized water in a 50 mL Erlenmeyer ﬂask, to

4.6% conversion and 98.0% selectivity at 623 K) and a high

which 2.91 g of NH −Y with SiO /Al O = 30 (purchased

NH utilization eﬃciency (almost no extra consumption) in

from Zeolyst Co.) was added with stirring. For the ion-

exchange of NH₄⁺ions with Cs ions, the solution was heated

from room temperature to 353 K and maintained at 353 K for

toluene ammoxidation employed as a test reaction; to the best

of our knowledge, this is the highest yield without NH loss for

the benzonitrile synthesis from toluene with O + NH in a

2 h with stirring, followed by washing with deionized water,

single-step gas-phase reaction. There is always a competition

between the ammoxidation to benzonitrile and undesirable

centrifuging, and ﬁnally drying at 353 K for 8 h. Other zeolite

Y-supported alkali and alkaline earth metal ion catalysts were

prepared similarly. Cs (2 wt %)/Y samples with diﬀerent

−3,11,23

NH oxidation to N and NO on catalyst surfaces,

whereas the Cs /Y was found to catalyze no NH + O

SiO /Al O ratios (5.2, 5.6, 12, and 115) were also prepared in

reaction when toluene was not present (Table 1). The simple

a similar way. Furthermore, other supports for Cs (β, ZSM-5,

mordenite (Mor), hydrotalcite (Hydr), and silica−alumina

ion-exchange method for the chemically conﬁned Cs /Y

catalyst fabrication in this study is also advantageous in

(SiO ·Al O )) were also used to prepare diﬀerent supported

2 2 3

Cs catalysts for comparison.

.2. Catalytic Reactions and Product Analysis. Vapor-

industrial usage. The Cs /Y catalyst was also eﬀective in the

synthesis of other organic nitriles, thereby proving its

versatility. The genesis and mechanism of the unprecedented

phase ammoxidation of toluene was carried out at atmospheric

pressure using a ﬁxed-bed ﬂow reactor (Pyrex glass tube) with

an inner diameter of 6 mm and externally heated by electric

resistances. The reaction temperature was measured inside the

catalyst bed by a thermocouple, whose tip was placed inside

the upper side of the catalyst bed. Catalyst powders were

pressed to pellets, crushed, and sieved (typically, 0.2 g, sieved

into 500−850 μm particles), and the sieved samples were

packed between quartz wool in the ﬂow reactor. Toluene was

fed continuously into the gas stream by using a syringe pump.

catalysis by the Cs single ion site/Y, which diﬀers from the

traditional redox catalysis mechanisms, has been characterized

by X-ray photoelectron spectroscopy (XPS), scanning trans-

mission electron microscopy−energy-dispersive X-ray spec-

troscopy (STEM-EDS), X-ray diﬀraction (XRD), temperature-

programmed desorption (TPD), in situ X-ray absorption near-

edge structure (XANES) and extended X-ray absorption ﬁne

structure (EXAFS), and DFT calculations.

699

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

Research Article

The reactor was fed with a toluene/NH /O /He mixture in

the molar ratio of 0.20/1.8/1.0/5.0 mL min typically. The

and the curve ﬁttings were performed in the R-space (0.18−

0.32 nm). The ﬁtting parameters for each shell were

coordination number (CN), interatomic distance (R),

−

ﬂow ratio of NH to toluene was deliberately kept ∼9 times

larger in order to increase the nitrile selectivity because the

formation of CO₂decreases considerably with increasing

Debye−Waller factor (σ ), and correction-of-edge energy

(ΔE ). The phase shifts and amplitude functions for Cs−O

amounts of NH in the feed. The conversion and selectivity in

and Cs−Cs were calculated from the FEFF 8.4 code. Error

ranges of the curve-ﬁtting analysis of EXAFS Fourier

transforms were based on the deﬁnition of the Larch code.

2.5. DFT Calculations. All calculations were conducted

the current system were calculated by using NH as an internal

standard, taking into account the calibration factors for NH in

the ﬂame ionization detector (FID) and thermal conductivity

detector (TCD) GCs in a similar way to that in our previous

reports, assuming no loss of nitrogen in the material balance

0,31

with Dmol3 ver. 2018 (Biovia) on Materials Studio 2018.

A numerical basis set with polarization function (DNP), of

which quality is comparable to 6-31G*, was adopted. All

electrons were explicitly included. All possible multiplicities of

electronic states were compared in the calculations, and

optimized multiplicity was determined in each system. The

during the catalytic reactions. The outlet stream was sampled

with six-way sampling valves heated at 463 K using an online

Shimadzu GC-2014 with a FID using a ZB-WAX plus capillary

column (30 m, Phenomenex CA) for NH , tolue,ne and

benzonitrile and a Shimadzu GC-2014 with a TCD using a

WG-100 column (GL Science Japan) for O , N , CO, CO ,

Perdew−Wang 91 functional (PW91) was used in all of the

calculations. Transition states were searched by Dmol3 with

the complete quadratic synchronous transit (QST)/linear

synchronous transit (LST) option, where the LST max-

imization was performed for the coordinates interpolated

between a reactant and a product, followed by repeated

conjugated gradient minimizations and the QST max-

NH , and H O. Helium was used as the carrier gas. The

column temperature for TCD was 323 K. The column

temperature for FID was held at 413 K for the ﬁrst 2 min and

−

then increased to 473 K at a heating rate of 298 K min . For

the ammoxidation of toluene in the present system, the

nitrogen-containing products were N₂and benzonitrile

33,34

imizations until a transition state was located.

(

PhCN) with a negligible amount of PhCONH . GC-TCD

To include the HY (Faujasite: FAU) zeolite framework, a

cluster containing a 6-membered ring with 12 neighboring 4-

membered rings was taken from the crystal structure of the

FAU zeolite with hydrogen atom termination (Si−H). One Si

atom in the 6-memebered ring was replaced by an Al atom by

adding a H atom on the neighboring oxygen atom. Positions of

the Al atom, the added H atom, and neighboring oxygen atoms

were relaxed in the structural optimization. Then the H atom

was replaced by a Cs atom, and the positions of the Cs atom,

the Al atom, and neighboring O atoms were relaxed. The

model (Cs Si Al O H ) obtained in this way was used to

analysis showed no formation of N O. The values of the

elemental carbon and nitrogen mass balances were estimated

to be between 97% and 100% in most of the experimental runs.

.3. Temperature-Programmed Desorption. All tem-

perature-programmed desorption (TPD) experiments were

carried out in a ﬂow reaction system in a ﬂow of helium carrier

−

gas (50 mL min ). For each experiment in a ﬁxed-bed reactor

6 mm i.d.), 0.2 g of catalysts with similar grain sizes (0.2−0.4

mm) were used. The catalysts were heated under He ﬂow (4

(

−

mL min ) at 573 K for 30 min and cooled down to

adsorption temperature (300 K). Respective ﬂows of CO and

37 23

examine the adsorption of toluene, O , NH , and subsequent

2 3

−

CH CN were turned on at 5 mL min and kept for 15 min to

intermediates and transition states. Plots for densities of state

were drawn with respect to the Fermi level corresponding to

the midpoint between HOMO and LUMO as the origin.

complete their adsorption. The samples after the adsorption

−

were ﬂushed with helium (40 mL min ) for 15 min and then

−

heated at a ramping rate of 28.3 K min . The analysis of

reactor eﬄuents (CO and CH CN) was performed on gas

. RESULTS AND DISCUSSION

chromatographs with a thermal conductivity detector (TCD)

and a ﬂame ionization detector (FID), respectively, calibrated

by the peak area of known pulses of the respective chemicals.

3.1. Catalytic Performances of Cs /Y and Other

Supported Alkali and Alkaline Earth Metal Ions. We

have examined catalytic performances of alkali and alkaline

earth metal ions, which were chemically conﬁned in various

zeolite pores and hydrotalcite interlayers and supported on a

SiO

cyanation) of toluene with O

623 K as a test reaction. The results are shown in Table 1.

Alkaline earth metal ions (Mg , Ca , Sr , Ba )/Y samples

(Y zeolite: SiO /Al molar ratio = 30) were incapable of

activating Csp3−H bonds as expected, and alkali metal ions

(Na and K )/Y samples also showed insuﬃcient activity for

benzonitrile (PhCN) synthesis, whereas Rb /Y and partic-

ularly Cs /Y possessed great potential to functionalize a

toluene C −H bond toward PhCN. The toluene conversions

were 47.1% and 73.0%, respectively, and the PhCN selectivities

were as high as 98.1% and 98.5%, respectively; the only

NH TPD experiments on Cs (2 wt %)/Y and Cs (2 wt %)/β

at 353−923 K were also conducted in a similar way to the

earlier TPD experiments to examine the adsorption behavior

and strength of NH in relation to the ammoxidation catalysis

·Al

surface, in gas-phase ammoxidation (oxidative

mechanism.

and NH on 0.2 g of catalysts at

.4. In Situ X-ray Absorption Fine Structure Measure-

ments and Data Analysis. In situ X-ray absorption ﬁne

structure (XAFS) spectra at the Cs L -edge for the as-

2 3

synthesized Cs (2 wt %)/Y sample and after the pretreatment

+ +

and catalysis of the sample without exposure to air were

measured at 15 K in a ﬂuorescence mode at a BL36XU station

in SPring-8. XANES and EXAFS spectra were analyzed in a

,26

similar way to that in previous reports, using the Larch code

containing the IFEFFIT Package ver. 2 (Athena and

sp3

Artemis). Background subtraction in the EXAFS analysis

was performed using Autobk. The Victoreen function was

employed for the background subtraction, and the spline

byproduct was CO , and no other liquid byproducts were

observed. It is noteworthy that the Rb /Y and Cs /Y catalysis

showed a remarkable suppression of undesirable NH₃

oxidation to N and NO ; the ratios of NH (consumed)/

smoothing method with Cook and Sayers criteria was used as

the μ method. The extracted k -weighted EXAFS oscillations

−

were Fourier-transformed to R-space over k = 25−80 nm ,

PhCN(synthesized) were 1.2 and 1.0, respectively, which are

700

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

Research Article

Figure 1. Catalytic activity of Cs /Y vs Cs loading (0.2 g of cat) (A) and catalyst weight (B).

nearly stoichiometric in the ammoxidation reaction equation,

more detail on the inﬂuence of Cs loadings on the structure of

C H CH + 3/2O + NH → C H CN (PhCN) + 3H O.

conﬁned Cs sites and the kinetic regime of ammoxidation

To examine the inﬂuence of zeolite compositions (SiO /

over diﬀerent Cs /zeolites under the reaction conditions used

Al O molar ratios) on the Cs /Y catalysis, toluene

ammoxidation on Cs (2 wt %)/Y samples with diﬀ erent Si/

in this study. Instead, we have provided Figures S1 and S2

(Supporting Information), which display the inﬂuence of Cs

loadings (1, 2, 3, 5, and 10 wt %) on the steady-state time-on-

Al ratios was performed. The results are shown in Table S1

(

Supporting Information). It was observed that the Cs (2 wt

stream for the toluene ammoxidation, where all the Cs /Y

)/Y with SiO /Al O = 12 showed a similar selectivity of

catalysts showed stable performances under the present

conditions. The Cs /Y catalyst with 2.0 wt % Cs loading was

4.8−98.0% for the Cs (2 wt %)/Y with SiO /Al O = 30

while keeping a high toluene conversion (98.5%), but

undesired extra NH oxidation to N became about double

the superior one. The conversion and yield increased with an

increase of the catalyst weight up to 0.4 g (for example, the

conversion increased from 73.0% to 94.6% with a catalyst

increase from 0.2 to 0.4 g, respectively, while keeping a high

selectivity of 98.0% (Table 1, Table S3, and Figure 1)), and

then the value was saturated, where most of the fed toluene

molecules were transformed to benzonitrile (94.6% con-

version), thereby bringing almost no supply to the remaining

catalyst layer located close to the reactor exit (e.g., 0.1 g in 0.5

g of catalyst) (Figure 1 and Table S3). When the catalyst

under the reaction conditions in Table S1. By decreasing the

SiO /Al O ratio to 5.2−5.6, the selectivity toward benzonitrile

decreased to 80.0−81.3% at 97.2−93.1% conversion, which

indicates that strong Brønsted acidic sites in the Y zeolite

promote the formation of undesired products (mainly CO₂)

and undesired NH oxidation. When the SiO /Al O ratio was

as high as 115, the toluene conversion abruptly decreased to

4.2% (Table S1 ), indicating the necessity of suﬃcient proton

sites for ion-exchange with Cs ions favorably near surface

layers of zeolite particles. Among the samples examined in this

increased above 0.4 g, the undesirable extra NH oxidation

(consumed NH /synthesized PhCN) increased (Table S3).

study, Cs (2 wt %)/Y with SiO /Al O molar ratio = 30

proved to be superior.

Neither bicomponent reactions of toluene + O (without

NH ) (eq 1), toluene + NH (without O ) (eq 2), nor NH +

The performance (PhCN yield) of Cs /β was 40% of that of

O (without toluene) (eq 3) on the Cs /Y proceeded as shown

Cs /Y, as shown in Table 1. Other zeolites such as ZSM-5 and

mordenite (Mor) were not eﬀective as support for Cs (0.79%

and 0.86% yields, respectively; Table 1). Cs /hydrotalcite

in Table 1 as follows.

C H CH + O → no oxidation products

(1)

(2)

(3)

(

Hydr), where Cs ions are located in the interlayers of

C H CH + NH → no amination products

hydrotalcite, showed no activity for the ammoxidation. Cs /

SiO ·Al O with the same SiO /Al O molar ratio (30) as the

NH + O → no oxidation products

Y zeolite, where Cs ions were supported on the SiO ·Al O

surface, was inactive (Table 1). These results suggest the

C H CH + 3/2O + NH

importance of the pore architecture of the Y zeolite to conﬁne

Cs chemically at the pore surface; thereby inactive Cs ions

with a noble gas electronic structure were transformed to active

Cs ion sites. The eﬀect of Cs loading in 0.2 g of Cs /Y on the

catalytic performance at 623 K was examined in the Cs loading

range of 0.5−10.0 wt % as shown in Figure 1 and Table S2 .

The activity of the Cs ions varied with the amount of Cs ions

in the Y zeolite. The conversion and selectivity gradually

decreased to a saturated level after a maximum at 2.0 wt % Cs

→

C H CN + 3H O with good yield

(4)

The toluene ammoxidation (benzonitrile synthesis) pro-

ceeded only in the tricomponents reaction (toluene + O +

NH ) (eq 4), indicating the contribution of NH to activation

of both toluene C −H and O and no contribution of lattice

sp3

oxygen atoms of the catalyst to the ammoxidation. Thus, the

Cs /Y catalyst possesses a unique property: toluene oxidation

loading. The decrement of the catalytic activity of Cs /Y

can be promoted only under the NH coexistence, no toluene

catalysts with increasing Cs loadings above 2.0 wt % may be

attributed to the decreasing accessibility of Cs sites at deeper

layers of zeolite particles. At the moment we cannot discuss in

amination proceeds, and undesirable NH oxidation with O is

suppressed. The activation of both toluene C −H and O by

sp3

coadsorbed NH and the whole ammoxidation pathway on the

701

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

conversion (%) and benzonitrile selectivity (%) on 0.2 g of

Research Article

Cs /Y will be discussed in the Coordination-Promoted

Concerted Ammoxidation Pathway section. The catalytic

comparison. The conversion increased with an increase of

alkali and alkaline earth metal ion radii up to ∼0.14 nm, but

the performances of their alkali and alkaline earth metal ions/Y

performance of Cs (2 wt %)/Y did not change during 240

min time-on-stream (Figure S2), and the pore volume (0.990

samples were as low as 2.2−17.6% conversions. The Rb /Y

−1

cm g ) of the Cs (2 wt %)/Y remained unchanged after 240

min of reaction. After 300 min the catalytic conversion at 623

and Cs /Y samples with ion radii larger than ∼0.14 nm,

particularly Cs /Y (Cs ion diameter = 0.334 nm), showed

K began to decrease gradually, probably due to coking.

remarkable conversions with high selectivities (98.1−98.5%).

However, the pore volume did not reduce signiﬁcantly. The

performance almost recovered by treatment of the 300-min

spent catalyst at 673 K in air.

Figure 2 shows the plots of the toluene single-path

The single Cs sites cannot activate toluene, O , and NH

when they get adsorbed independently and also in the case of

their bicomponents such as toluene + O , toluene + NH , and

NH + O (Table 1). However, when the three reactants

(

toluene, O , and NH ) coadsorbed together, the trimolecular

2 3

concerted reaction working closely on the Cs single metal ion

site proceeded eﬃciently, causing toluene C −H activation

sp3

toward benzonitrile synthesis.

To gain further insight into the nature of accessible sites in

the Cs /Y catalyst, the ammoxidation of toluene derivatives

with increasing bulkiness was compared with the toluene

ammoxidation (Table 2). The Cs /Y catalyst showed good

performances in achieving the ammoxidation of o-, m-, and p-

xylene (81.5−88.8% conversion; 98.0−99.0% selectivity),

mesitylene (76.2% conversion; 99.8% selectivity), and o-, m-,

and p-chlorotoluene (77.6−91.6% conversion; 94.3−94.6%

selectivity). With two bulky alkyl benzene molecules, 1,3,5-tri-

t-butylbenzene and 1,3,5-tris(2,4,6-trimethylphenyl)benzene,

no reactions occurred due to the bulkier sizes than the

accessible aperture (0.74 nm) of the Y zeolite with a 3-

dimensional network of the open aperture and supercage (1.2

Figure 2. Relation between the catalytic performances (toluene one-

path conversion and benzonitrile selectivity) on 0.2 g of the alkali (red

mark) and alkaline earth (blue mark) metal ions (2 wt %)/Y catalysts

and the ionic radii of alkali and alkaline earth metal ions. The

nm), in which Cs ions are conﬁned and bound chemically

to the pore surface lattice oxygen atoms (as discussed

conversion values for Cs /β, Cs /mordenite (Mor), Cs /hydrotalcite

(

Hydr), and Cs /SiO ·Al O are also plotted for comparison. For

2 2 3

hereinafter).

reaction conditions, see Table 1.

3.2. Characterization of Cs (2 wt %)/Y by XRD, STEM-

EDS, and XAFS. To get an insight into the superiority of

the alkali (red mark) and alkaline earth (blue mark) metal ion

Cs (2 wt %)/Y, we integrated in situ and ex situ character-

(2 wt %)/Y catalysts versus the ionic radii of alkali and alkaline

ization of the Cs (2 wt %)/Y catalysts by XRD (Figure S3),

earth metal ions to reveal a factor to generate the catalytic

STEM-EDS (Figure S4), XPS (Figure 3A), in situ XANES

(Figure 3B), in situ EXAFS (Figure 3 C), DFT simulations

(Figures 4, S5, and S6 ), and TPD (Figure S7). XRD patterns

performance. In Figure 2 the conversion values for Cs /β, Cs /

Mor, Cs /Hydr, and Cs /SiO ·Al O are also plotted for

Table 2. Catalytic Performances of Cs (2 wt %)/Y for Selective Ammoxidation of Substituted Toluenes at 623 K

entry

reactant

reactant conv./%

85.4

nitrile selec./%

nitrile syn. rate/μmol s⁻g⁻¹

NH /nitrile

cat

o-xylene

98.5

0.294

1.0

(

86.6)

98.0

m-xylene

p-xylene

mesitylene

81.5

88.8

76.2

0.275

0.300

0.228

1.1

78.8)

99.0

0.91

1.1

92.8)

99.8

52.5)

94.5

94.3

94.6

96.0

98.0

99.0

o-chlorotoluene

m-chlorotoluene

p-chlorotoluene

o-nitrotoluene

m-nitrotoluene

p-nitrotoluene

1,3,5-tri-tert-butylbenzene

1,3,5-Tris(2-methylphenyl)benzene

81.7

77.6

91.6

42.0

38.0

3.5

0.288

0.274

0.322

0.148

0.134

0.012

0.93

0.98

0.83

1.8

1.9

2.2

Conv. = conversion; selec. = selectivity. NH /nitrile = molar ratios of consumed NH to synthesized nitriles. Performance values were averaged

−

during 30−240 min time-on-stream. Reaction conditions: toluene/O /NH /He = 0.20/1.0/1.8/4.0 (mL min ); catal. = 0.4 g; reaction temp =

23 K. Sum of the selectivity (%) of the corresponding mononitrile and dinitrile produced in each entry. The values in the parentheses are the

selectivity (%) of the corresponding mononitriles.

702

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

determined by EXAFS curve-ﬁ tting analysis. The EXAFS data were noisy, but the ﬁ tting analysis was su ﬃ cient for a structural argument.

Research Article

Figure 3. (A) Ex situ XPS Cs 3d peaks and binding energies for as-synthesized (fresh) Cs (2 wt %)/Y (a; red), spent Cs (2 wt %)/Y (b; blue),

and CsNO (commercial Cs ion salt) (c; black). (B) In situ Cs L -edge XANES spectra for as-synthesized (fresh) Cs (2 wt %)/Y (1; red), spent

Cs (2 wt %)/Y (2; blue), and Cs (2 wt %)/β (3; green), and ex situ XANES spectrum for CsCl as reference (4; black). (C) In situ Cs L -edge

EXAFS data and analysis for as-synthesized (fresh) Cs (2 wt %)/Y; (C-1) EXAFS oscillation (black, obs; red, ﬁtting); (C-2) EXAFS Fourier

transform associated with (C-1) (black, obs; red, ﬁtting; gray, imaginary part; orange, ﬁtting); (C-3) structural parameters for the pretreated Cs/β

Figure 4. Changes of the energy levels of HOMO and LUMO and the HOMO−LUMO gap by chemical conﬁnement of Cs in Y zeolite pore

making Cs−O bonds with pore surface lattice oxygen atoms. Ball−stick model ﬁgures for the HOMO and LUMO of Cs single site/Y are also

shown.

for the Cs /Y samples with 2, 3, 5, and 10 wt % Cs loadings

exhibited only typical XRD peaks associated with Y zeolite, and

no Cs oxide peaks were observed up to 10 wt % Cs loadings

measured in situ EXAFS data to obtain information on a local

coordination structure around the Cs site. The observed

EXAFS data for Cs (2.0 wt %)/Y (Figure 3C-1 and C-2) were

Figure S3 ). Bright-ﬁeld STEM images and element (Si, O,

noisy, but we could do a suﬃcient ﬁtting analysis of the EXAFS

and Cs) EDS maps for Cs (2 wt %)/Y revealed no feature

except for the Y zeolite lattice (Figure S4 ). Hence, we

data for a structural argument of single Cs sites. The EXAFS

analysis at the Cs L -edge for the Cs (2 wt %)/Y catalyst in

703

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

Research Article

Figure 3C revealed only Cs−O bonds at 0.303 (±0.010) nm

with an approximate coordination number of 2.6, and no Cs−

Cs bonds were observed, which indicates isolated Cs single

sites that are chemically bound to three lattice O atoms of the

zeolite pore surface. The experimental results were essentially

consistent with the theoretically simulated structure around the

are composed of Cs 5p and 6s states, respectively. For the Cs⁺

single site chemically conﬁned in the Y zeolite making Cs−O

bonds, HOMO and LUMO are composed of O 2p state

bonding with the Al atom at the Y pore surface and Cs 6s state

on Cs , respectively, as shown in Figure 4, where the HOMO−

LUMO gap decreased drastically to 4.7115 eV by increasing

the HOMO energy and the decreasing LUMO energy (Figure

Cs ion in the pore (Cs−O = 0.315, 0.316, and 0.366 nm), as

shown in Figure S5, considering that the local structure

determined by EXAFS is an averaged coordination structure at

every Cs site involved in the catalyst. The coordination

structure of Cs sites in Y zeolite pores (Y zeolite component:

4 and Figure S6). The LUMO for Cs /β with much less

activity than Cs /Y constitutes O 2p around the Si atom at the

β pore surface, and the HOMO −LUMO gap (5.1059 eV) was

larger than that for the Cs /Y (Figure S6 ). The energy gap and

SiO /Al O = 30) resembles that of Cs single sites in β zeolite

nature of HOMO and LUMO states are expected to aﬀect the

pores (β zeolite component: SiO /Al O = 30), while the Cs−

catalytic performance of the Cs single sites for C −H bond

sp3

O interatomic distances are shorter than 0.335, 0.356, and

activation. This discussion is not limited to the alkali metal ion

single sites in Y zeolite pores. For example, Rh/ZSM-5, in

which Rh single atoms were attached on ZSM-5 pore surfaces,

.386 nm, respectively, for Cs /β. The Mulliken charges on

Cs in Y and β pores were estimated as +0.773 and +0.327,

respectively, by DFT, which is due to the shorter and stronger

has been reported to show unique, eﬃcient CH activation

Cs−O bonds in the Cs /Y than in the Cs /β. It is suggested

toward HCOOH and CH COOH at 50 bar, although the

that the diﬀerence in the local structures, Mulliken charges,

origin of the unique catalysis of Rh single atom sites/ZSM-5

and pore sizes caused the increased activity of Cs /Y compared

has not been investigated in detail.

to Cs /β in the catalytic ammoxidation (Table 1). As to the

3.3. Coordination-Promoted Concerted Ammoxida-

tion Pathway. Alkali metal ions do not have the moderate

redox property that brings about selective oxidation and

ammoxidation, which is suggested by the unchanged XPS and

XANES spectra in Figure 3. To estimate the basic and acidic

very poor activity of Cs /ZSM-5 and Cs /MOR, proper local

structures around the Cs sites may be obtained for the

systems without adsorbed molecules before the ammoxidation

catalysis; however, the narrower pores and channel structures

(

pore diameter of 0.54 × 0.56 nm for ZSM-5 and one-

strengths of the Cs single sites/Y catalyst, we conducted

dimensional pore of 0.67 × 0.70 nm in diameter for

mordenite), compared to pore diameters of 0.55 × 0.55 nm

and 0.76 × 0.64 nm for β and a supercage diameter of 1.2 nm

and an open aperture of 0.74 nm for Y, may exert steric

constraints for a trimolecular concerted pathway involving

large-sized reaction intermediates and transition states

temperature-programmed desorption (TPD) of CH CN and

CO on Cs (2 wt %)/Y (Figure S7 ). The TPD peak for

CH CN on the Cs /Y catalyst was observed at a low

temperature around 440 K, and the desorbed CH CN amount

per Cs after subtracting the amount desorbed from the bare Y

zeolite was 66.4% of the Cs quantity. These results indicate a

weak Lewis acidity of the Cs single sites in Y zeolite pores,

which is a little stronger than the Lewis acidity of the Cs sites

(

discussed later), resulting in almost no progress of the

ammoxidation reaction on Cs /ZSM-5 and Cs /Mor (Table

in β pores (TPD peak around 360 K). In the TPD of CO on

Conversely, the Cs /β was active for benzene C −H

Cs (2 wt %)/Y to estimate the basic strength, most of the

sp2

functionalization toward phenol, but the Cs /Y was inactive for

the phenol synthesis. It is to be noted that the activation of

adsorbed CO desorbed below 323 K and the desorbed

amount was negligible (0.1% of Cs ), indicating a negligible

basicity of the Cs single sites in Y pores. Thus, no signiﬁcant

C−H bonds by alkali metal ion single sites is very sensitive to

the local structure around the alkali metal ion. Further, the

valency of Cs was proved by XPS 3d spectra and in situ

XANES (Figure 3A and B). The XPS 3d₅_/₂binding energies

acidity−basicity contribution to the toluene C −H ammox-

sp3

idation catalysis of Cs (2 wt %)/Y is needed in the current

study.

for the fresh and spent Cs /Y catalysts were 724.7 and 724.6

Although it is generally agreed that activation of the methyl

group is the rate-determining step during the toluene

eV, respectively, which are situated in the values for Cs⁺

,38,39

monocations; 724.3 eV for CsNO ; 724.5 eV for CsOH; and

ammoxidation,

the nature of activated species and

24.8 eV for Cs CO on Ag. The white line peak intensity in

intermediates is still under debate. We examined the eﬀect of

electron-donating and -withdrawing side groups on the

aromatic ring (Table 2). The nitrile yields in the ammoxidation

of xylene and chlorotoluene varied in the order p- > o- > m-

the in situ Cs L -edge XANES spectra for the fresh and spent

Cs /Y catalysts was much larger than the ex situ XANES

spectrum for CsCl with a formal valency of Cs . The white line

peak intensity in the XANES spectra at the Cs L -edge reﬂects

substituents, unlike that reported previously, which might be

the multielectron photoexcitation feature involving electrons

from outer subshells 6s to 4p, and it is also related to Cs

due to a heterolytic C−H rupture mechanism forming a

carbocationic intermediate and involving electronic eﬀects.

valency, which resembles the large white line intensity for Cs /

In the case of nitrotoluene, p-nitrotoluene was not converted

signiﬁcantly to p-nitrobenzonitrile, whereas o-nitrotoluene was

transformed to the corresponding nitrile with a similar yield to

that for m-nitrotoluene (Table 2). Further, the carbocationic

intermediate should be accompanied by hydride ion formation,

thereby reducing Cs⁺sites partially or fully, which is

β with direct Cs−O bonds. Cs ions with a noble gas

electronic structure Coulombically supported on SiO and

SiO ·Al O surfaces were inactive for toluene ammoxidation

(

Table 1). Inactive Cs ions were transformed to remarkable

Cs single sites conﬁned in Y zeolite pores making direct Cs−

O bonds with the Y pore surface lattice oxygen atoms (Table 1,

Figure 3C, and Figure S5 ). The density of states (HOMO and

LUMO) for Cs /Y by DFT calculations using the optimized

structural model of Cs single site/Y (Figure S5 ) is shown in

inconsistent with the in situ Cs L -edge XANES spectra that

were unchanged during the catalysis (Figure 3B). Hence, the

association of the superior catalytic property of Cs single sites

to other factors such as the largest size of Cs ion (0.334 nm)

Figure 4 and Figure S6 . HOMO and LUMO for free Cs ion

among the alkali metal ions and the unaﬀected Sanderson

704

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

Research Article

Figure 5. Catalytic cycle (ball−stick model ﬁgures) of the ammoxidation of toluene C −H bonds via the inter-coadsorbate reaction pathway

sp3

without O dissociation and lattice oxygen involvement on a Cs single site in the Y zeolite pore rationalized by DFT calculations involving the

reaction intermediates and transition states. For bond distances, see Figures S8 and S9 .

electropositivity of the Cs⁺ion is envisioned.¹^,⁴³This

hypothesis is supported by the trend of the ion-size eﬀect in

the catalytic performance, as shown in Figure 2, and by the

methyl cyanation experiment of 3-methylpyridine, where

neither aldehyde nor cyanide products were produced because

idation on the Cs single site in the Y zeolite pore are shown in

Figure 5 (catalytic cycle outline), Figure S8 (reaction

pathway), and Figure S9 (detailed mechanism and energy

proﬁles). The signiﬁcantly large diameter of the Cs ion

enables all three reactant molecules (C H CH , O and NH )

to get adsorbed on a Cs single site. It was found that an NH

molecule adsorbed/coordinated on a Cs site promotes the

perhaps the electropositivity of Cs was quenched by the lone-

pair electrons of 3-methylpyridine. The larger electropositivity

and size of the Cs ions stem from the lower polarizabilty or

toluene C −H activation and O activation accompanied by

sp3

higher softness of the Cs ions, which are further ensconced by

the C−O bonding on the Cs site to produce hydroperoxy

their conﬁnement within the zeolite pores. The polarizabilty

intermediate (Int1 = C H CH OOH) via transition state TS1,

shuﬄes with the high electropositivity in the Cs /Y, which, in

followed by the formation of gem-diol intermediate (Int2 =

turn, eﬃciently copes with the energy penalty to overcome the

C H CH(OH) ), gaining a bond-rearrangement energy of

−

ammoxidation reaction barriers. There was no change in the

233.5 kJ mol (Figure S9). NH plays an important role in the

Cs oxidation state during the catalysis, and thus, the

coadsorption of toluene and O on the Cs single site, where

mechanism diﬀers from the traditional redox catalysis (e.g.,

donation from NH to Cs is essential for the interaction

Mars−van Krevelen mechanism) and also acid−base

between the toluene methyl group and the O molecule on a

catalysis involving clearly deﬁned interaction modes.

Cs single site. The adsorption behavior and strength of NH

To shed light on the origin of the selective C −H

were examined by NH₃TPD (Figure S10 ). The NH

sp3

adsorption behavior at the Cs⁺single sites in the zeolite

cyanation performance of the Cs single site/Y catalyst and to

understand the ammoxidation mechanism at the molecular

level, we performed periodic self-consistent DFT calculations

for the reaction pathway. The reaction mechanism, coordina-

tion arrangements, and energy proﬁles involving reaction

intermediates and transition states for the toluene ammox-

pores cannot be evidenced from the NH TPD curves, but the

NH TPD data are compatible with the trend of Mulliken

charges for Cs /Y and Cs /β by DFT and the fact that NH

promotes the toluene ammoxidation more with Cs /Y than

Cs /β. Then, Int2 was transformed to the aminohydroxyto-

705

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

https://pubs.acs.org/doi/10.1021/acscatal.1c00946.

Research Article

luene intermediate (Int3 = C H CH(NH )(OH)) with a C−

initial stage of the reaction, NH plays an important role in the

N bond via a benzaldehyde-type transition state (TS3) that

interacts with the coadsorbed NH and H O. The hydrogen

coadsorption of toluene and O on Cs single sites, where

donation from NH to Cs is essential for the interaction

atom of the C−OH end of Int2 shifts to the other O−H group

between the toluene methyl group and the O molecule to

to form TS3, which consists of distorted benzaldehyde, H O,

form a C−O bond on a Cs single site according to the DFT

and NH . In TS3, the benzaldehyde molecule exhibits a

calculations. Then, the neighboring NH molecule joins the

nonplanar conformation, where the plane of the phenyl group

and the plane of the carbonyl group form an angle of 41.3°. It

is noted that the stable planar benzaldehyde molecule does not

appear in the whole reaction pathways. Int3 was dehydrated by

the interligand bond rearrangement involving N−H bond

scission to form an imine intermediate (Int4 = C H CH

ammoxidation reaction, which proceeds successively via several

reaction intermediates and transition states with C−N bond,

CN bond, and ﬁnally CN bond on a large-sized Cs single

site conﬁned in the Y zeolite pore. Our synthetic strategy

potentially increases the scope of on-demand organic nitrile

synthesis by rationally incorporating Cs ions in Y zeolite pores

NH) via the C−O breaking transition state (TS4), which is

simply by an industrially advantageous facile ion-exchange

process.

stabilized on the Cs single site. The imine N atom interacts

with the Cs 6s LUMO as an electron acceptor, resulting in

the promotion of the reaction with the second adsorbed O at

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

■

the N position (Int5 = C H CHNH(O ) and Int6 =

C H CHNH(OO)), and the interligand rearrangements

occur between the two imine H atoms and oxygen successively

to produce the C H CHNOOH intermediate (Int7) via

transition state TS5 and benzonitrile C H CN (F) via

Lists of catalytic performance data for Cs /Y with

diﬀerent Si/Al ratios, Cs loadings, and catalyst weights;

transition state TS6 accompanied by H O (readily decom-

posed to H O at 623 K). These interligand bond rearrange-

ments are possible with the large-sized (0.334 nm) Cs ion.

time-on-stream performances (conversion and selectiv-

ity) of Cs /Y at 623 K; XRD patterns and bright-ﬁeld

Finally, the adsorbed benzonitrile (F) desorbs to the gas phase

with a desorption energy of 71.0 kJ mol⁻in a gas-phase,

single-path ﬂow reaction. The rate-determining steps in the

whole reaction pathway constitute three C−H and N−H

bond-dissociation steps (Int1 → Int2, Int3 → Int4, and Int7 →

STEM-EDS maps; DFT simulations of the optimized

structure model of Cs /Y and HOMO−LUMO states;

TPD curves of CH CN and CO ; DFT computational

reaction pathways and energy proﬁles; and TPD curves

of NH (PDF)

F) after the ﬁrst dissociation of toluene C −H bonds (Figure

sp3

S9 ). In the catalysis mechanism the Cs single-ion sites

conﬁned in the Y zeolite pores provide a trimolecular assembly

platform to enable the coordination-promoted concerted

AUTHOR INFORMATION

■

Corresponding Authors

ammoxidation pathway without involvements of O -dissoci-

Takehiko Sasaki − Graduate School of Frontier Science, The

University of Tokyo, Kashiwa, Chiba 277-8561, Japan;

orcid.org/0000-0001-6190-4087; Email: takehiko@k.u-

tokyo.ac.jp

ated O atom and lattice oxygen.

. CONCLUSIONS

Yasuhiro Iwasawa − Innovation Research Center for Fuel Cells

and Graduate School of Informatics and Engineering, The

University of Electro-Communications, Chofu, Tokyo 182

Industrial vapor-phase ammoxidation reactions have been

extensively studied over mixed transition metal oxide catalysts

with moderate redox potentials and suﬃcient M−O bond

strengths that bring about active lattice oxygen and oxygen

atoms at the catalyst surfaces to promote redox catalysis.

However, alkali and alkaline earth metal ions with noble gas

electronic structures have been thought to be inactive for

selective oxidation processes before our recent report on

benzene C −H hydroxylation with N O or O toward

8585, Japan; orcid.org/0000-0002-5222-5418;

Email: iwasawa@pc.uec.ac.jp

Authors

Shankha S. Acharyya − Innovation Research Center for Fuel

Cells and Graduate School of Informatics and Engineering,

The University of Electro-Communications, Chofu, Tokyo

sp2

phenol. The Cs single-ion site/Y catalyst without a beneﬁcial

redox property, which was inactive for benzene hydroxylation,

showed unprecedented catalysis with high eﬃciencies toward

182 8585, Japan; orcid.org/0000-0001-7572-3674

Shilpi Ghosh − Innovation Research Center for Fuel Cells and

Graduate School of Informatics and Engineering, The

University of Electro-Communications, Chofu, Tokyo 182

catalytic cyanation (ammoxidation) of C −H bonds of

sp3

toluene and its derivatives with NH as test reactions. The

8585, Japan; orcid.org/0000-0001-6332-0595

molar ratios of [NH consumed]/[nitriles synthesized] were

Yusuke Yoshida − Innovation Research Center for Fuel Cells,

The University of Electro-Communications, Chofu, Tokyo

nearly stoichiometric, minimizing undesired extra oxidation of

NH , which is advantageous industrially. DFT calculations

182 8585, Japan

indicated the reduction of the HOMO−LUMO gap and the

Takuma Kaneko − Innovation Research Center for Fuel Cells,

The University of Electro-Communications, Chofu, Tokyo

variation of the HOMO component by the chemical

conﬁnement of the Cs single-ion sites in the Y zeolite

182 8585, Japan

pores, making Cs−O(lattice) bonds that provided a new

trimolecular assembly platform to enable the coordination-

promoted concerted pathway that worked closely on the large-

Complete contact information is available at:

https://pubs.acs.org/10.1021/acscatal.1c00946

sized Cs single site without any signiﬁcant change of the Cs

charge, which is diﬀerent from the traditional redox catalysis

mechanisms (e.g., the Mars−van Krevelen mechanism). At the

Author Contributions

S.S.A., S.G., T.S., and Y.I. contributed equally.

∥

706

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

(18) Younes, M. K.; Ghorbel, A. Catalytic Nitroxidation of Toluene

Research Article

Notes

into Benzonitrile on Chromia-Alumina Aerogel Catalyst. Appl. Catal.,

A 2000, 197, 269−277.

The authors declare no competing ﬁnancial interest.

19) Chary, K. V. R.; Reddy, K. R.; Bhaskar, T.; Sagar, G. V.

ACKNOWLEDGMENTS

■

Dispersion and Reactivity of Mo/Nb O Catalysts in the Ammox-

The study was supported by Grant-in-Aid for Scientiﬁc

Research of JSPS/MEXT (16F16078 and 16F16383). S.S.A.

and S.G. thank JSPS Postdoctoral Fellowship for foreign

researchers. The XAFS measurements were performed with the

approval of SPring-8 subject numbers 2017A7808, 2017A7841,

idation of Toluene to Benzonitrile. Green Chem. 2002, 4, 206−209.

20) Rombi, E.; Ferino, I.; Monaci, R.; Picciau, C.; Solinas, V.;

Buzzoni, R. Toluene Ammoxidation on α -Fe O -based Catalysts.

(

Appl. Catal., A 2004, 266, 73−79.

(21) Teimouri, A.; Najari, B.; Najafi Chermahini, A.; Salavati, H.;

Fazel-Najafabadi, M. Characterization and Catalytic Properties of

Molybdenum Oxide Catalysts Supported on ZrO -γ -Al O for

018A7840, 2018B7840, and 2019A7840.

Ammoxidation of Toluene. RSC Adv. 2014, 4, 37679−37686.

REFERENCES

■

(22) Goto, Y.; Shimizu, K.; Kon, K.; Toyao, T.; Murayama, T.;

(

1) Ghosh, S.; Acharyya, S. S.; Kaneko, T.; Higashi, K.; Yoshida, Y.;

Ueda, W. NH -Efficient Ammoxidation of Toluene by Hydro-

Sasaki, T.; Iwasawa, Y. Confined Single Alkali Metal Ion Platform in a

Zeolite Pore for Concerted Benzene C-H Activation to Phenol

Catalysis. ACS Catal. 2018, 8, 11979−11986.

thermally Synthesized Layered Tungsten-Vanadium Complex Metal

Oxides. J. Catal. 2016, 344, 346−353.

(23) Hari Babu, B.; Venkateswara Rao, K. T.; Suh, Y. W.; Sai Prasad,

(

2) Acharyya, S. S.; Ghosh, S.; Yoshida, Y.; Kaneko, T.; Sasaki, T.;

P. S.; Lingaiah, N. One-Step Selective Synthesis of 2-Chlorobenzoni-

trile from 2-Chlorotoluene via Ammoxidation. New J. Chem. 2018, 42,

Iwasawa, Y. NH ₃ -Driven Benzene C-H Activation with O ₂ that

Opens a New Way for Selective Phenol Synthesis. Chem. Rec. 2019,

892−1901.

24) Li, X.; Sun, L.; Hu, M.; Huang, R.; Huang, C. Hydrothermal

9, 2069−2081.

3) Cavalli, P.; Cavani, F.; Manenti, I.; Trifiro, F. Ammoxidation of

Synthesis of Urchin-like W-V-O Nanostructures with Excellent

Toluene to Benzonitrile on Vanadium-Titanium Oxides Catalysts

Prepared by Precipitation. The Role of Catalyst Composition. Ind.

Eng. Chem. Res. 1987, 26, 639−647.

Catalytic Performance. Inorg. Chem. 2018 , 57 , 14758 −14763.

(25) Centi, G.; Jiru, P.; Trifiro, F. Catalytic Properties of Zeolites in

Oxidation and Ammoxidation Reaction. Stud. Surf. Sci. Catal. 1989,

4, 247−254.

26) Iwasawa, Y., Asakura, K., Tada, M., Eds. XAFS Techniques for

Catalysts. In Nanomaterials and Surfaces; Springer: 2016; pp 3−10,

3−50, 299−316.

27) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS:

4) Anbarasan, P.; Schareina, T.; Beller, M. Recent Developments

(

and Perspectives in Palladium-Catalyzed Cyanation of Aryl Halides:

Synthesis of Benzonitriles. Chem. Soc. Rev. 2011, 40, 5049−5067.

5) Sundermeier, M.; Zapf, A.; Mutyala, S.; Baumann, W.; Sans, J.;

Weiss, S.; Beller, M. Progress in the Palladium-Catalyzed Cyanation of

Aryl Chlorides. Chem. - Eur. J. 2003 , 9 , 1828 −1836.

6) Hodgson, H. H. The Sandmeyer Reaction. Chem. Rev. 1947, 40,

(

Data Analysis for X-Ray Absorption Spectroscopy using IFEFFIT. J.

Synchrotron Radiat. 2005, 12, 537 −541.

(

51−277.

(28) Rehr, J. J.; Albers, R. C. Theoretical Approaches to X-Ray

7) Jagadeesh, R. V.; Junge, H.; Beller, M. Green Synthesis of

Absorption Fine Structure. Rev. Mod. Phys. 2000, 72, 621−654.

Nitriles using Non-Noble Metal Oxides-Based Nanocatalysts. Nat.

Commun. 2014, 5, 4123.

8) Yang, J.; Karver, M. R.; Li, W.; Sahu, S.; Devaraj, N. K. Metal-

Catalyzed One-Pot Synthesis of Tetrazines Directly from Aliphatic

Nitriles and Hydrazine. Angew. Chem., Int. Ed. 2012, 51, 5222−5225.

9) Pollak, P.; Romeder, G.; Hagedorn, F.; Gelbke, H. P. Ullman’s

Encyclopedia of Industrial Chemistry ; Wiley-VCH Verlag GmbH & Co.

KGaA: Weinheim, 2000; DOI: 10.1002/14356007.a17_363.

10) Wang, L.; Wang, G.; Zhang, J.; Bian, C.; Meng, X.; Xiao, F. S.

Controllable Cyanation of Carbon-Hydrogen Bonds by Zeolite

Crystals over Manganese Oxide Catalyst. Nat. Commun. 2017, 8,

5240.

11) Centi, G.; Marchi, F.; Perathoner, S. Effect of Ammonia

Chemisorption on the Surface Reactivity of V-Sb-oxide Catalysts for

Propane Ammoxidation. Appl. Catal., A 1997 , 149 , 225 −244.

12) Otamiri, J. C.; Andersson, A. Kinetics and Mechanisms in the

Ammoxidation of Toluene over a V O catalyst. Part 1: Selective

Reactions. Catal. Today 1988, 3, 211 −222.

13) Rapolu, R.; Panja, K. R. Highly selective V-P-O/γ -Al O

Catalysts in the Ammoxidation of Toluene to Benzonitrile. J. Chem.

(29) Newville, M.; Ravel, B.; Haskel, D.; Rehr, J. J.; Stern, E. A.;

Yacoby, Y. Analysis of Multiple-Scattering XAFS Data Using

Theoretical Standards. Phys. B 1995 , 208 −209 , 154 −156.

(30) Delley, B. An All-Electron Numerical method for Solving the

Local Density Functional for Polyatomic Molecules. J. Chem. Phys.

990, 92, 508−517.

31) Delley, B. From Molecules to Solids with the DMol3 Approach.

J. Chem. Phys. 2000, 113, 7756−7764.

32) Perdew, J. P.; Wang, Y. Accurate and Simple Analytic

(

Representation of the Electron-Gas Correlation Energy. Phys. Rev.

B: Condens. Matter Mater. Phys. 1992, 45, 13244 −13249.

(33) Halgren, T. A.; Lipscomb, W. N. The Synchronous-Transit

Method for Determining Reaction Pathways and Locating Molecular

Transition States. Chem. Phys. Lett. 1977 , 49 , 225 −232.

(34) Bell, S.; Crighton, J. S. Locating Transition States. J. Chem.

(35) Al-Khattaf, S. Catalytic Transformation of Toluene over a

Phys. 1984, 80, 2464 −2475.

High-Acidity Y-Zeolite Based Catalyst. Energy Fuels 2006, 20, 946−

954.

P.; Groen, J. C.; Pérez-Ramirez, J.; Sels, B. F. Synthesis, Character-

Soc., Chem. Commun. 1993, 1175−1176.

(36) Verboekend, D.; Nuttens, N.; Locus, R.; Van Aelst, J.; Verolme,

isation, and Catalytic Evaluation of Hierarchical Faujasite Zeolites:

Milestones, Challenges, and Future Directions. Chem. Soc. Rev. 2016,

45, 3331−3352.

(37) Tang, Y.; Li, Y.; Fung, V.; Jiang, D.; Huang, W.; Zhang, S.;

Iwasawa, Y.; Sakata, T.; Nguyen, L.; Zhang, X.; Frenkel, A. I.; Tao, F.

Single Rhodium Atoms Anchored in Micropores for Efficient

Transformation of Methane under Mild Conditions. Nat. Commun.

2018, 9, 1231.

14) Sanati, M.; Andersson, A.; Wallenberg, L. R.; Rebenstorf, B.

Zirconia-Supported Vanadium Oxide Catalysts for Ammoxidation and

Oxidation of Toluene: A Characterization and Activity Study. Appl.

Catal., A 1993, 106, 51−72.

15) Martin, A.; Berndt, H.; Lu

cke, B.; Meisel, M. Reaction Pathway

of Benzonitrile Formation during Toluene Ammoxidation on

Vanadium Phosphate Catalysts. Top. Catal. 1996, 3, 377−386.

(16) Zhang, Y.; Martin, A.; Berndt, H.; Lucke, B.; Meisel, M. FTIR

Investigation of Surface Intermediates Formed during the Ammox-

idation of Toluene over Vanadyl Pyrophosphate. J. Mol. Catal. A:

Chem. 1997, 118, 205−214.

(38) Lucke, B.; Narayana, K. V.; Martin, A.; Jähnisch, K. Oxidation

and Ammoxidation of Aromatics. Adv. Synth. Catal. 2004, 346, 1407−

17) Martin, A.; Hannour, F.; Bruckner, A.; Lucke, B. Ammoxidation

̈ ̈

1424.

of Toluene on Vanadyl Polyphosphates-VO(PO ) , 2. Catalytic

(39) Ghosh, S.; Acharyya, S. S.; Tripathi, D.; Bal, R. Preparation of

Silver-Tungsten Nanostructure Materials for Selective Oxidation of

Properties. React. Kinet. Catal. Lett. 1998, 63, 245−251.

707

ACS Catal. 2021, 11, 6698−6708

ACS Catalysis

40) Busca, G.; Cavani, F.; Trifiro,

Research Article

Toluene to Benzaldehyde with Hydrogen Peroxide. J. Mater. Chem. A

014, 2, 15726−15733.

F. Oxidation and Ammoxidation

of Toluene over Vanadium-Titanium Oxide Catalysts: A Fourier

Transform Infrared and Flow Reactor Study. J. Catal. 1987, 106,

(

71−482.

41) Lucke, B.; Martin, A. Catalysis of Organic Reactions; Scaros, M.

G., Prunier, M. L., Eds.; Marcel Dekker, Inc.: 1994; pp 479 −482.

42) Mirth, G.; Cejka, J.; Lercher, J. Transport and Isomerization of

Xylenes over HZSM-5 Zeolites. J. Catal. 1993, 139, 24−33.

43) Keller, T. C.; Desai, K.; Mitchell, S.; Pérez-Ramírez, J. Design of

Base Zeolite Catalysts by Alkali-Metal Grafting in Alcoholic Media.

ACS Catal. 2015, 5, 5388−5396.

(44) Boronat, M.; Corma, A. What Is Measured When Measuring

Acidity in Zeolites with Probe Molecules? ACS Catal. 2019, 9, 1539−

548.

45) Mars, P.; van Krevelen, D. W. Oxidations Carried out by Means

of Vanadium Oxide Catalysts. Chem. Eng. Sci. 1954, 3, 41−59.

46) Zachariasse, K. A.; Grobys, M.; von der Haar, T.; Hebecker, A.;

Il'ichev, Y. V.; Morawski, O.; Ruckert, I.; Kuhnle, W. Photo-Induced

Intramolecular Charge Transfer and Internal Conversion in Molecules

with a Small Energy Gap Between S1 and S2. Dynamics and

Structure. J. Photochem. Photobiol., A 1997, 105, 373−383.

708