Light-assisted cyclohexane oxidation catalysis by a manganese(III) porphyrin immobilized onto zinc hydroxide salt and zinc oxide obtained by zinc hydroxide salt hydrothermal decomposition

Article abstract of DOI:10.1016/j.apcata.2020.117708

Two manganese(III) porphyrin complexes prepared with the tetra anionic free base porphyrin (Na₄[H₂(TDFSPP)]) and two different manganese(II) salts (manganese(II) acetate - MnP(OAc) complex and manganese(II) chloride - MnP(Cl) complex

Full text of DOI:10.1016/j.apcata.2020.117708

Journal Pre-proof
Light-assisted cyclohexane oxidation catalysis by a manganese(III)
porphyrin immobilized onto zinc hydroxide salt and zinc oxide obtained
by zinc hydroxide salt hydrothermal decomposition
K a´ tia C.M. Westrup (Conceptualization) (Methodology)
(
(
Investigation) (Writing - original draft), Renaldo M. da Silva Junior
Methodology) (Investigation) (Validation) (Writing - original draft),
Karen M. Mantovani (Methodology) (Investigation) (Writing - original
draft), Larissa Bach (Investigation) (Formal analysis), Jo a˜ o F. Stival
(Methodology) (Investigation)<ce:contributor-role>data curation),
Patr ´ı cio G.P. Zamora (Conceptualization) (Writing - review and
editing) (Visualization), Fernando Wypych (Conceptualization)
(
(
(
(
Writing - review and editing), Guilherme S. Machado (Investigation)
Writing - review and editing), Shirley Nakagaki (Conceptualization)
Project administration) (Funding acquisition)
Supervision)<ce:contributor-role>data curation) (Writing - review
and editing)
PII:
S0926-860X(20)30301-X
DOI:
https://doi.org/10.1016/j.apcata.2020.117708
APCATA 117708
Reference:
To appear in:
Applied Catalysis A, General
Received Date:
Revised Date:
Accepted Date:
9 March 2020
20 May 2020
16 June 2020
Please cite this article as: Westrup KCM, da Silva RM, Mantovani KM, Bach L, Stival JF,
Zamora PGP, Wypych F, Machado GS, Nakagaki S, Light-assisted cyclohexane oxidation
catalysis by a manganese(III) porphyrin immobilized onto zinc hydroxide salt and zinc oxide
obtained by zinc hydroxide salt hydrothermal decomposition, Applied Catalysis A, General

(2020), doi: https://doi.org/10.1016/j.apcata.2020.117708
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
©
2020 Published by Elsevier.

Light-assisted cyclohexane oxidation catalysis by a manganese(III) porphyrin
immobilized onto zinc hydroxide salt and zinc oxide obtained by zinc hydroxide salt
hydrothermal decomposition
1
1
1
2
Kátia C. M. Westrup , Renaldo M. da Silva Junior , Karen M. Mantovani , Larissa Bach ,
1
2
3
4
João F. Stival , Patrício G. P. Zamora , Fernando Wypych , Guilherme S. Machado , Shirley
Nakagaki1*
1
2
Laboratório de Bioinorgânica e Catálise, Laboratório de Química Analítica e Ambiental,
3Laboratório de Química de Materiais Avançados - 1,2,3Departamento de Química, Centro
Politécnico, Universidade Federal do Paraná (UFPR), Curitiba/PR, Brazil
4Campus Pontal do Paraná - Centro de Estudos do Mar, Universidade Federal do Paraná
(
UFPR), Pontal do Paraná/PR, Brazil
Graphical abstract
Highlights

Metalloporphyrin immobilized onto ZnO prepared by hydrothermal
methodology.

Unusual selectivity for the ketone observed using MnP(OAc)-ZHN
as catalyst.

 Observation of the influence of both counterions and support in the
catalytic activity of metalloporphyrin.
Abstract
Two manganese(III) porphyrin complexes prepared with the tetra anionic free base porphyrin
(
Na
4
[H (TDFSPP)]) and two different manganese(II) salts (manganese(II) acetate -
2
MnP(OAc) complex and manganese(II) chloride - MnP(Cl) complex) were immobilized onto
zinc hydroxide nitrate (ZHN), a layered hydroxide salt, and also into zinc oxide (ZnO),
obtained via hydrothermal decomposition of the ZHN. The resulting solids, MnP(OAc)-ZHN,
MnP(Cl)-ZHN, MnP(OAc)-ZnO and MnP(Cl)-ZnO, were characterized by X-ray powder
diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy
(
(
SEM), energy dispersive spectroscopy (EDS), Fourier transform infrared spectroscopy
FTIR), electronic paramagnetic resonance (EPR) and ultraviolet-visible spectroscopy (UV-
Vis) (solid samples). Manganese and consequently the immobilized manganese porphyrin
complexes were quantified by ICP OES. The materials were applied as catalysts for the
oxidation of cyclohexane, using iodosylbenzene as the oxygen donor, under conventional
conditions (absence of light) as well as using light assisted conditions. Good catalytic results
were obtained and, in absence of light, an unusual selectivity for the ketone was observed
when using MnP(OAc)-ZHN as catalyst different from what can be observed when the
catalyst is the solid MnP(Cl)-ZHN, that present selectivity to alcohol, suggesting an influence
of both axial ligand and support in the catalytic activity.
Keywords: Porphyrin, Zinc Oxide, Zinc Hydroxide Salt, Heterogeneous Catalysis Assisted
by Light, Cyclohexane Oxidation
1. Introduction
In the last decades, synthetic metalloporphyrins (MPs) have been extensively
investigated as catalysts in a wide variety of organic reactions [1-3] and among them, one of
the most important is the oxidation of organic substrates, in which the MPs mimetizes
biological functions performed by cytochrome P450 enzymatic family [4,5].

In homogeneous mediums, MPs have presented excellent results concerning the
catalytic efficiency and selectivity in the studies of oxidation and hydroxylation of several
alkenes and alkanes [6-9]. However, the impossibility of reusing them through several
catalytic cycles together with the possible deactivation of the catalytic species are obstacles
for potential technological applications [5,10], mainly when the structure of the porphyrin
ligand is simpler, and consequently cheaper, than other more elaborated ones containing
protective substituents against oxidative degradation.
The immobilization of MPs in inorganic and organic supports for use as catalysts in a
heterogeneous medium has been presented as an alternative to overcome such problems
and to obtain more robust, selective and efficient results [11-14].
Several inorganic and organic supports have been investigated for the immobilization
of MPs [15-20], e.g. amorphous or mesoporous silicas, zeolites, natural and synthetic
layered materials, among others. In addition, catalytic species have been immobilized in
polymeric matrixes [21] such as porous organic polymers (POPs) or metal organic
frameworks (MOFs) solids with catalytic purposes [21-26].
Recently, layered supports have been used by our research group [16,27-31] and
others [32-34] to produce catalysts studied in the oxidation of organic substrates such as
linear and cyclic alkanes in heterogeneous medium yielding promising results.
Machado et al. [28] reported the first example of an anionic iron(III) porphyrin family
immobilized onto zinc hydroxide nitrate (ZHN), a layered hydroxide salt, and the investigation
of the solids as catalysts towards the oxidation of cyclooctene, cyclohexane, and n-heptane.
In the oxidation of the cyclohexane, the results revealed an unusual selectivity for
cyclohexanone suggesting that ZHN solid was a non-innocent support and may have
influenced the catalytic mechanism favoring a radical pathway. Besides having the ability to
immobilize anionic porphyrins, ZHN has been shown to be a good precursor to obtain high
purity ZnO from its hydrothermal decomposition.
In a following work [35], we demonstrated that different iron metalloporphyrins can be
immobilized onto ZnO obtained by ZHN hydrothermal decomposition. The obtained catalysts
led to good catalytic performance in the oxidation of alkanes and alkenes showing selectivity
to alcohol superior than that presented by the respective metalloporphyrins in homogeneous
medium.
Metal oxides are used in various fields of science and technology, such as solar cells,
gas sensors and photocatalysis [36-41]. The conjugation between photosensitizers, such as
porphyrins (P) and metalloporphyrins (MP), and metal oxides has been leading to interesting
results in the photodynamic therapy (PDT) area [42], in the photooxidation of organic
substrates [43] and in the hydrogen production [44]. Recently, Mantovani et al.[45] reported
the study of catalytic oxidation reactions assisted by light using as catalyst a material

obtained from the simultaneous immobilization of two metalloporphyrins, one positively and
the other negatively charged, in hydrozincite (zinc hydroxide carbonate
Zn (OH) CO •nH O), containing titanium oxide. The material analyses indicated the presence
–
5
8
3
2
of dispersed titanium oxide on the surface of the hydrozincite, without the insertion of Ti4+
ions into the layered structure. The solid was used as catalyst in the oxidation of cyclohexane
in the absence and presence of light. The results indicated a synergistic effect between the
metalloporphyrin, support and the radiation used, leading to a significant increase in the
production of cyclohexanol and cyclohexanone in presence of light when compared to
reactions carried in the dark [45].
Inspired by these recent results and exploring both capabilities of the ZnO –
photoactivity and act support to MPs immobilization – in the present work we report the
preparation and characterization of two manganese(III) porphyrin complexes from the ligand
tetra anionic free base porphyrin (Na
4 2
[H (TDFSPP)]) using two different manganese(II) salts
(
manganese(II) acetate resulting in the complex named by MnP(OAc) and manganese(II)
chloride - MnP(Cl) complex – Figure 1). The two complexes were immobilized onto zinc
hydroxide nitrate (ZHN), a layered hydroxide salt, and also into zinc oxide (ZnO), obtained
via hydrothermal decomposition of the ZHN. The resulting materials had their catalytic
properties investigated in the cyclohexane oxidation under conventional conditions (absence
of light) as well as under light assisted conditions.
Figure 1. Schematic representation of the structure of the tetra anionic metalloporphyrin
[
5,10,15,20-tetrakis-(2,6-difluor-3-sulfonatephenylporphyrinate) manganese(III)].
2. Experimental
2.1 General information
All chemicals used in this study were purchased from Sigma-Aldrich or Merck and
were of analytical grade. Iodosylbenzene (PhIO) was synthesized by hydrolysis of

iodobenzene diacetate [46]. The solid was carefully dried under reduced pressure and kept
at 5 °C; its purity was assessed by iodometric titration [46,47].
2
.2 Preparation of the anionic free base porphyrin {[5,10,15,20-tetrakis-(2,6-difluor-3-
sulfonatophenyl)porphyrin] tetrasodium} - Na [H (TDFSPP)]
The starting free base porphyrin [H (TDFPP)] ([5,10,15,20-tetrakis(2,6-
difluorphenyl)porphyrin]) was synthesized and purified following the Lindsey methodology
48].
4
2
2
[
The anionic free base porphyrin {Na
reaction of the starting free base porphyrin [H
49]. [H (TDFPP)] (159 mg, 0.21 mmol) and concentrated sulfuric acid (95-97%, 6.2 mL)
4 2
[H (TDFSPP)]} was obtained by the sulfonation
2
(TDFSPP)] by modifying Turk's methodology
[
2
were placed in a round-bottom flask (50 mL) and the resulting dark green solution was kept
at 180 °C for 7 hours. Then, the solution was cooled to room temperature and kept under
magnetic stirring for an additional 48 hours. After that, the flask was left in an ice bath and
-1
the solution was slowly neutralized with an aqueous NaOH solution (2 mol∙L ). The solvent
was removed in the rotary evaporator and the resulting light brown solid was extracted with
ethanol. Then, the ethanol was removed at low pressure and the final purple solid (P) was
1
dried at 60 °C (153 mg, 0.13 mmol, 62% yield). H NMR data (200 MHz, MeOD, 298 K) 
ppm: 9.03 (s, br, 8 H, -pyrrole-H), 8.48 (m, 4.3 H, p-Ph-H), 7.60 (m, 4.3 H, m-Ph-H). max
-1
(
nm, methanol): 408 (Soret band), 504 and 580. FTIR (cm ): 3454 (O–H stretch, broad band
from water), 3313 (N–H stretch, shoulder band), 3097 (=C–H stretch), 1616 and 1581 (C=C
aromatic stretches), 1431 (C–N stretch), 1250-1180 (S=O stretches), 1060–1000 (C–F
multiple bands). The spectra can be found in the Supplementary Information (Figures S1, S2
and S10).
2
.3 Preparation of the anionic metalloporphyrin (MnP(OAc))
The free base porphyrin {Na [H (TDFSPP)]} was metallated with the manganese(II)
acetate salt, Mn(CH COO) 4H O (Mn(OAc) ), according to the metalloporphyrin preparation
literature [50,51]. Mn(OAc)
(233 mg, 9.5 x 10-4 mol) was added to a solution of
Na [H (TDFSPP)] (76 mg, 6.5 x 10 mol) in 40 mL of methanol/acetic acid mixture (1:4, v/v),
4
2
3
2
2
2
2
-5
4
2
previously heated at 120 °C. The reaction was kept under magnetic stirring and heating at
the same temperature for 24 hours. The metallation reaction progress was monitored by TLC
and UV-Vis techniques. The solvent was removed using a high vacuum pump and the
complex was purified by sephadex 25G column (H
2
O). The final metalloporphyrin MnP(OAc)
72 mg, 5.6 x 10-5 mol, 86%) was characterized by UV-VIS and FTIR spectroscopy
Supplementary Information; Figures S3 and S10) and EPR technique that was used to
(
(
confirm the oxidation state of manganese(III) (EPR-silent) and also to be sure of the absence

of manganese(II) acetate (which would be EPR-active) in the sample. max (nm, water): 324,
4
-1 -1
3
68, 392, 458 (Soret band,  = 8.4x10 cm·mol ·L ) and 552. IR spectrum showed similar
bands of free porphyrin except by the presence of the C=O stretch bands of the acetate
group at 1541 and 1384 cm-1 and the absence of the shoulder band at 3313 cm−1
characteristic of N–H stretching.
2
.4 Preparation of the anionic metalloporphyrin (MnP(Cl))
The metallation reaction of the porphyrin {Na [H (TDFSPP)]} using the manganese(II)
chloride salt, MnCl 4H O, was carried out following the aforementioned methodology for
obtaining metalloporphyrins [50,52]. MnCl
solution of the free base porphyrin [H (TDFSPP)] (69 mg, 5.9 x 10 mol) in DMF, previously
4
2
2
2
-4
2
4H O (117 mg, 5.9 x 10 mol) was added to a
2
-5
2
heated at 95 °C. The reaction was kept under magnetic stirring and heating at the same
temperature for 24 hours. The metallation reaction progress was monitored by TLC and UV-
Vis techniques. The solvent was removed using a high vacuum pump and the complex was
purified by Sephadex 25G column (H O). The final metalloporphyrin MnP(Cl) (65 mg, 5.1 x
2
1 -
0
5
mol, 88%) was characterized by UV-VIS spectroscopy (Supplementary Information;
Figures S4 and S10) and EPR technique that was used to confirm the oxidation state of
manganese(III) (EPR-silent) and also to certify the absence in the sample of manganese(II)
4
-
acetate (EPR-active). max (nm, water): 324, 370, 392, 458 (Soret band,  = 7.1 x 10 cm∙mol
1·L-1) and 554. IR spectrum showed similar bands of MnP(OAc) porphyrin except by the
absence of the acetate bands.
2.5 Preparation of the layered hydroxide salt Zinc Hydroxide Nitrate (ZHN)
ZHN was synthesized according to formerly described methodology [28]. A freshly
-1
prepared aqueous solution of NaOH (1 mol∙L ) was dropped into a freshly prepared aqueous
-1
solution of Zn(NO
3
)
2
6H O (0.29 mol∙L ) until the dispersion reached pH 7. During this
2
procedure the mixture was kept under magnetic stirring and the pH change was monitored
with a pHmeter. At the end of the reaction, the white solid was centrifuged and washed with
deionized water (3 portions of 50 mL) and dried at 60 °C for two days.
2
.6 Preparation of the Zinc Oxide (ZnO) obtained by the hydrothermal decomposition
of ZHN
ZnO was prepared according to the previously described methodology [35]. ZHN
1.00 g) was dispersed in 100 mL of distilled water and refluxed at 100 °C for 5 hours. The
(
white solid was centrifuged (4000 rpm for 5 min.), repeatedly washed with distilled water and
dried at 60 °C for two days.

2.7 Immobilization of MnP(OAc) or MnP(Cl) onto ZHN solid
The metalloporphyrins immobilization process onto ZHN was performed according to
the methodology formerly described [28]. Manganese(III) porphyrin, MnP(OAc) or MnP(Cl)
-6
(
ca. 8 x 10 mol), and 0.50 g of the ZHN support were dispersed in 50 mL of ethanol and the
resulting light orange mixture was kept under magnetic stirring at room temperature for 5
hours. The solid was centrifuged (4000 rpm for 5 min.) and washed exhaustively with ethanol
until the electronic spectrum of the supernatant did not exhibit the absorption bands related
to the manganese porphyrin, which leached during this procedure. The light orange solids
obtained were named by MnP(OAc)-ZHN or MnP(Cl)-ZHN and were dried at 60 °C for two
days.
2.8 Immobilization of MnP(OAc) or MnP(Cl) onto ZnO
The MPs immobilization process onto ZnO was obtained from hydrothermal
decomposition of ZHN by a method already described [35]. The whitish orange aqueous
dispersion (100 mL) containing the manganese(III) porphyrins, MnP(OAc) or MnP(Cl), (ca.
-6
1
6 x 10 mol), and the ZHN (1.00 g) was magnetically stirred at 40 °C for 1 hour. After this
period, the temperature was allowed to stabilize at 100 °C and the system remained at reflux
for 5 hours. The solid was separated by centrifugation (4000 rpm for 5 min.) and washed
exhaustively with distilled water until the electronic spectrum of the supernatant did not
exhibit the absorption bands related to the manganese porphyrin. The obtained solids
(
MnP(OAc)-ZnO or MnP(Cl)-ZnO) were dried at 60 °C for two days.
2.9 Immobilization of MnP(OAc) or MnP(Cl) onto commercial ZnO (ZnOcom
)
Attempts were made to immobilize the metalloporphyrins MnP(OAc) or MnP(Cl)
directly onto the commercial ZnO (ZnOcom) using the same procedure described previously
for the immobilization on ZnO synthesized by ZHN hydrothermal decomposition. The
obtained solids (MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom) were dried at 60 °C for two days.
All the solids described in the items 2.5 to 2.9 were characterized by X-ray powder
diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy
(
SEM) coupled to energy dispersive spectroscopy (EDS), Fourier transform infrared
spectroscopy (FTIR), electronic paramagnetic resonance (EPR) and ultraviolet-visible
spectroscopy (UV-Vis). The manganese concentration in the supports was also quantified by
ICP-OES.
2.10 Cyclohexane oxidation reactions in the dark and assisted by controlled light

The catalyst/oxidant/substrate molar ratio used in all the catalytic reactions was
:20:2000, and it was chosen due to its customary application in similar systems of homo
1
and heterogeneous catalysis [35]. The solid catalyst and the oxidant (iodosylbenzene) were
added to a reaction vial (1.5 mL) and subsequently the solvent acetonitrile (0.24 mL) and the
substrate cyclohexane (0.01 mL). The catalytic reactions were conducted under magnetic
stirring at room temperature in a dark chamber for one hour. Thereafter, the reaction was
inhibited by the addition of a saturated solution of Na
2
SO . The catalyst was separated,
3
washed with methanol and a mixture of acetonitrile and dichloromethane (1:1) and then both
the supernatant and washing solution were mixed quantitatively and analyzed by gas
chromatography to determine the amount of product formed, using n-octanol or
bromobenzene as internal standard.
For the cyclohexane oxidation reactions assisted by light, the reaction vial was
exposed to the radiation of a halogen lamp (250 W; lamp dimensions: 91 mm in length; 14
mm in diameter; quartz bulb) at a distance of 9 cm from the radiation source. The reaction
system was cooled by a water jacket to keep it at 25 ± 1 °C. All the experimental
apparatuses used in these reactions were fit in a homemade white reaction box (0.65 m
height, 0.55 m width and 0.55 m depth).
The corresponding MnP in homogeneous medium were also investigated as catalysts
using similar reagent molar ratio. The experimental procedure in this case was similar to that
used for the heterogeneous catalysis.
Catalysts reuse tests were also carried out with the most promising solids. The
catalysts recovered at the end of the first use were exhaustively washed with methanol and
acetonitrile, dried at 60 °C, and reused in further reactions, adopting the same procedure
described before and using cyclohexane as substrate.
2.11 Materials characterization
For the X-ray diffraction (XRD) analysis, the solids were placed on glass sample
holders and the measurements were performed in reflection mode using a Shimadzu XRD-
000 diffractometer operating at 30 kV and 40 mA (Cu Kα radiation λ = 1.5418 Å) in a θ-2θ
6
-1
geometry Bragg-Brentano with a scan rate of 1°∙min using values of 2θ from 3 to 70°.
FTIR spectra were recorded on a Bomen Michelson MB100 spectrophotometer in the
-1
range of 400-4000 cm , using KBr pellets. KBr was grounded to fine dust together with a
small amount of the solids (ca. 1% m/m), and the spectra were collected with a resolution of
-1
2
cm and accumulation of 64 scans.
For all the solids obtained from the immobilization of the MnP on ZHN and ZnO
(
prepared or commercial), the manganese concentration was quantitatively analyzed by
optical emission spectroscopy with inductively coupled plasma (ICP-OES) in order to

determine how much manganese porphyrin was immobilized after the reaction and the
washing process. The measurements were carried out with an iCAP 6500 from Thermo
Scientific, using diluted solutions prepared from the stock solution of the solid (ca. 26 mg)
dissolved in a 5% (v/v) HNO solution (25 mL). The analyses were performed at least in
3
duplicates and average values were used.
The transmission electron microscopy (TEM) measurements were performed by using
a JEOL JEM-1200 microscope operating at 110 kV. The solid samples were dispersed in
propanone and deposited on FCF300-CU grids (300 mesh) with formvar/carbon film.
Scanning electron microscopic images (SEM) and Energy dispersive spectra (EDS)
were obtained with a Tescan VEGA3 LMU device, using tension of 10 KV. The samples were
deposited on aluminum stubs and the EDS spectra were collected with an SDD detected with
2
8
0 mm and in the range of 0 to 10 KeV. In the sequence, the samples were gold sputtered
and the SEM images were obtained.
The electronic absorption spectra (UV-Vis) of the solutions were obtained with an HP
8452A Diode Array UV-Vis spectrophotometer, in the 200-800nm range. The UV-Vis spectra
of the solid samples were obtained using a Varian Cary 100 BIO by placing the samples in a
Teflon sample holder and measuring the absorbance in the 200 to 800 nm range.
1H NMR spectra (200.13 MHz) were acquired at 298 K on a Bruker AVANCE 200
NMR spectrometer operating at 4.7 T and equipped with a 5 mm tetranuclear direct detection
1
13
19
31
probe ( H, C, F and P). Methanol-d
4
(CD OD) was supplied by Aldrich.
3
The textural property analysis experiments were performed with a NOVA 1000
analyzer (Quantachrome) and the samples were degasified under vacuum for a period of 6
hours (ZnO samples were maintained at 150 °C during this period and those of ZHN at 90
°
C) and later analyzed in liquid nitrogen. The adsorption/desorption isotherms for N were
2
obtained in a relative pressure range from 0.05 to 0.99, and the specific surface area was
calculated using the multi-point Brunauer-Emmet-Teller method (B.E.T.), with a relative
pressure range from 0.05 to 0.3.
Products from oxidation reactions were quantified using an Agilent 6850 gas
chromatograph (flame ionization detector) equipped with a 30 m long DB-WAX capillary
column with 0.25 mm internal diameter (J&W Scientific). The products were quantified using
n-octanol or bromobenzene as internal standard.
Electron paramagnetic resonance (EPR) measurements were accomplished on an
EPR Bruker EMX microX spectrometer (frequency X, band 9.5 GHz) at room temperature
and 77 K (using liquid N ), in the perpendicular microwave polarization X-band.
2
-1
The Zeta potential for the dispersed solids ZHN, ZnO and ZnOcom (0.1 mg m L ) in
water (18 Ω) was determined using a Microtrac Stabino Particle Charge Titration Analyzer
(
Particle Metrix GmbH, Meerbusch, Germany).

3. Results and discussion
3.1 Characterization of solid supports and the MnP-support catalysts
All materials used in this work and their respective name codes are detailed in the
Supplementary Information file, Table S1. Table 1 shows the manganese amounts obtained
by the ICP-OES analyses as well as MnP immobilization degrees and loadings calculated
from these values.
The immobilization of the metalloporphyrins on ZHN, that resulted in the solids named
by MnP(OAc)-ZHN and MnP(Cl)-ZHN, occurred via strong electrostatic interactions between
the negative charges present in the MPs structure and the positive charges on the surface
[
28,53]. The Zeta potential measured for the ZHN solid was +5.0 mV, confirming the
presence of positive charges in this material justifying the immobilization of the anionic MnP.
The solids obtained in the attempt to immobilize the MnP on solid ZnOcom
(
(
-
MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom) do not exhibit manganese in their composition
Table 1), indicating no immobilization. The Zeta potential measured for the ZnOcom solid was
2.0 mV, indicating that this solid support presented residual negative charges on its surface
which makes the electrostatic interaction with the anionic MnP used in this work impossible.
Regardless, when the MnP complexes interacted with ZnO from hydrothermal
decomposition of ZHN, the ICP-OES result indicated that there was a reasonable amount of
Mn on the solids MnP(OAc)-ZnO and MnP(Cl)-ZnO (Table 1). These results suggested that
the immobilization process took place, even with a negative zeta potential (-9.0 mV).
Probably, the immobilization in the zinc oxide’s network generated in situ via hydrothermal
decomposition of the ZHN was possible due to the fact that there were some residual
positive charges on the surface of the ZnO material formed under the conditions used for the
immobilization process (hydrothermal process on the medium slightly acidic to neutral pH)
[
35] and also due to some positive charges present in the starting material ZHN.
Alternatively, the anionic MnP could have been immobilized in the solid ZHN and was kept
trapped in the generated voids of the oxide structure when thermal decomposition of the
ZHN took place.
Table 1. Immobilization degree and loading of the materials in relation to the dosage of
manganese obtained by the ICP-OES technique and SBET values of all the solidsa.
Immobilization / % b
Loading / mol g
c
-1
S
BET/ m g
2
-1
Solid
ZHN
–
–
10.207
12.290
MnP(OAc)-ZHN
11.6
2.8x10-6

MnP(Cl)-ZHN
ZnOcom
12.6
–
2.7x10-6
8.112
7.125
11.463
10.582
6.414
8.284
8.109
–
ZnO
–
–
MnP(OAc)-ZnO
MnP(OAc)-ZnOcom
MnP(Cl)-ZnO
MnP(Cl)-ZnOcom
11.3
3.7x10-6
–
–
3.3x10-6
–
6.8d
–
aConcentration of manganese identified in the solutions analyzed by ICP-OES. Calibration
-1
curve used for the dosage of manganese: 0.01 to 2 mg L . Experimental values below 0.01
mg L-1 were considered as absence of manganese in the sample trace quantities (-).
bImmobilization percentage (%) = mass of manganese in the final solid in relation to the
c
mass of manganese used in the immobilization process. Loading value = amount of
d
manganese (mol) per gram (g) of prepared material. Part of the material was lost during the
washing process, because the immobilization degree is smaller for this compound.
All materials were characterized by X-ray diffraction (XRD) (Figures 2 and 3). The
XRD diffraction patterns of the ZHN (Figure 2a), presented a typical diffraction peak around
9.6 Å (calculated from the third basal peak) which corresponds to basal distance of
Zn (OH) (NO 2H O [28,53,54]. The maintenance of the ZHN original basal distance
5
8
3
)
2
2
observed in the solids MnP(OAc)-ZHN (Figure 2b) and MnP(Cl)-ZHN (Figure 2c) in the
range of 9.6-9.7 Å indicated that the immobilization of MnP onto ZHN took place on the
surface of the solid. This same immobilization behavior had already been observed by our
research group [28].
c
b
a
10
20
30
40
50
60
70
2 / Degrees

Figure 2. XRD patterns of the ZHN support (a), MnP(OAc)-ZHN (b) and MnP(Cl)-ZHN (c).
The diffraction pattern presented by the solids ZnO, MnP(OAc)-ZnO and MnP(Cl)-
ZnO (Figure 3) indicated that the ZHN matrix was completely converted into ZnO after
hydrothermal treatment (Figure 3b). The diffraction peaks observed (Figure 3b) were the
same ones presented by commercial zinc oxide, ZnOcom (as observed in the card JCPDS:
36-1451), which is characteristic of the hexagonal wurtzite crystal structure [55,56] (Figure
3a). For the solids MnP(OAc)-ZnOcom (Figure 3d) and MnP(Cl)-ZnOcom (Figure 3f), the same
XRD pattern presented by the commercial ZnO was obtained, even after the reflux process
in the attempt to immobilize of the MnP in this support (Figure 3c and 3e).
The surface area (SBET) measured for all solids using the BET method are shown in
Table 1. The SBET values obtained for all the synthesized solids are in agreement with
surface areas already reported [57]. Immobilization of MnP did not cause any significant
change in the area of the solids indicating that the support remained unchanged,
corroborating the XRD results.
f
e
d
c
b
a
10
20
30
2
40
50
60
70
 / Degrees
Figure 3. XRD patterns of the ZnO (obtained by hydrothermal decomposition of ZHN) (a),
ZnOcom (b), MnP(OAc)-ZnO (c), MnP(OAc)-ZnOcom (d), MnP(Cl)-ZnO (e) and MnP(Cl)-
ZnOcom (f).
The morphology of all materials was examined by TEM and Figure 4 presents the
TEM images of MnP(Cl)-ZHN and MnP(Cl)-ZnO while further images can be seen in the

Figures S5-S7 from the supplementary Information. The TEM analysis of the support ZHN
Figure S5a) as well as materials obtained after immobilization of MnPs (solids MnP(OAc)-
(
ZHN and MnP(Cl)-ZHN, Figures S5b and 4(left), respectively), exhibit respectively platelet
and board-like particles which present a pseudo-orthorhombic orientation [53] according to
the results published [28].
Figure 4. TEM images of some selected particles of MnP(Cl)-ZHN (left) and MnP(Cl)-ZnO
(
right).
The solids resulting from the use of commercial zinc oxide MnP(OAc)-ZnOcom and
MnP(Cl)-ZnOcom - Figures S6b and S6c, respectively), have the same morphology as
observed to the ZnOcom (Figure S6a), indicating that the heating and stirring process during
the tentative of immobilization of the MnP complexes in this support did not alter its rod
shaped morphology [58]. Nevertheless, the morphology of the zinc oxide obtained from
hydrothermal decomposition of the ZHN (ZnO) and the solids MnP(OAc)-ZnO and MnP(Cl)-
ZnO resulting from the immobilization of the MPs (Figure S7a, S7b and 4(right), respectively)
resembles that observed for commercial oxide. However, it is observed that these materials
have less defined structures than the ZnOcom, probably due to stirring and reflux process
which should inhibit the growth of well-shaped crystals.
All materials were also examined by scanning electron microscopy (SEM) in order to
evaluate the surface morphology (Figure 5 and Figures S8 and S9, Supplementary
Information).
The typical platelet-like morphology was observed in SEM images for solids ZHN,
MnP(OAc)-ZHN and MnP(Cl)-ZHN (Figures S8a, S8b and Figure 5(left) respectively) [59-
61].
As already indicated by the TEM analyses, the SEM images have shown that the
solids MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom (Figures S9c and S9e, respectively) have the
same morphology as the ZnOcom (Figure S9a), presenting rectangle, radial hexagonal and
rod shapes [62,63]. The materials from the hydrothermal decomposition of the ZHN – ZnO,
MnP(OAc)-ZnO and MnP(Cl)-ZnO (Figure S9b, S9d and Figure 5(right), respectively) –

present morphologies that resemble rectangle, radial hexagonal and rod shapes, including
needle shape, but have less defined structures. This result indicates that the stirring and
reflux process can influence the final morphology of the obtained materials.
Figure 5. SEM images of MnP(Cl)-ZHN (left) and MnP(Cl)-ZnO (right).
Figures 6 and 7 illustrate the FTIR spectra of all solids prepared in this work, whereas
Figure S10 (Supplementary Information) shows only the porphyrin and metalloporphyrins
spectra for comparison purposes. As can be observed in Figure S10, the spectra of both
MnP are very similar showing small differences in the regions highlighted by the dotted
rectangles. The presence of the band at 1384 cm-1 in the MnP(OAc) spectrum can be
attributed to the asymmetric vibrations of the C=O bond in the acetate group. Although the
-1
region around 1500 cm is a bit crowded and the bands overlapped, the symmetric vibrations
-1
of the C=O bond can be tentatively assigned to the band at 1541 cm . The difference
-1
between these bands ( = 157 cm ) could suggest that the acetate is found as a counterion
64,65].
[

e
d
c
b
a
4000 3500 3000
1500
1000
500
-
1
Wavenumber / cm
Figure 6. FTIR spectra of MnP(Cl) (a), MnP(OAc) (b), MnP(Cl)–ZHN (c), MnP(OAc)–ZHN
-1
(
d) and ZHN support (e). (Omitted region of 2800-1850 cm region, due to the absence of
bands).
In the spectrum of the ZHN support, as well as those of the MnP(OAc)-ZHN and
-1
MnP(Cl)-ZHN (Figure 6), the broad band in the 3487 cm region can be attributed to the
stretching of the O-H bond, probably referring to the presence of water molecules retained in
-1
the layers of the material [28,66-68]. In addition, two bands at 3575 and 3313 cm are also
observed in this same region, which can be attributed to the O-H group stretching linked with
the layer and with the nitrate group, respectively [28,66,67]. Whereas the band in 1637 cm-1
is attributed to the angular vibration of the O-H bond of interlayer water molecules [28,66,67].
The N-O vibrations, relative to the nitrate groups, are observed at 1384, 1016, 838 and 763
-1
cm [28,53,68]. Furthermore, the vibrations of the Zn-O are observed at 638, 524 and 470
-1
cm [28,69].
-1
In the ZnO systems, all materials present bands only below the 600 cm region which
is characteristic of the Zn-O bond (Figure 7) [70-73]. The registered spectra for the samples
with commercial ZnO support (ZnOcom, MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom), present
-1
three bands of similar intensity in this region at 538, 495 and 428 cm (Figure 7h, 7f and 7d
respectively). Nonetheless, in the materials coming from the hydrothermal decomposition of
the ZHN (ZnO, MnP(OAc)-ZnO and MnP(Cl)-ZnO) a different pattern is observed, a more
intense band at 412 cm-1 and a very broad band at 532 cm-1 (Figure 7g, 7e and 7c
respectively).This same spectral pattern was also observed in a work previously published
[
35], which was attributed to the different morphologies observed for the ZnOcom and the ZnO
obtained by the hydrothermal decomposition of ZHN [72,73]. In some spectra of the

-1
materials containing the ZnO as a support band is observed at 3455 cm (Figure 7) which is
attributed to the O-H stretch relative to the presence of residual water in the material due to
the low drying temperature (60 °C) [70]. Since there is a low concentration of MnP in solids
MnP(OAc)-ZHN, MnP(Cl)-ZHN, MnP(OAc)-ZnO and MnP(Cl)-ZnO, as shown on Table 1,
bands related to the porphyrin ring were not be observed. These materials have a light
orange color, which is a visual evidence of MPs presence. This result is commonly observed
for other materials from the immobilization of MPs in inorganic supports [74-76].
h
g
f
e
d
c
b
a
4000 3500 3000
1500
Wavenumber / cm
1000
500
-
1
Figure 7. FTIR spectra of MnP(Cl) (a), MnP(OAc) (b), MnP(Cl)-ZnO (c), MnP(Cl)-ZnOcom
(
d), MnP(OAc)-ZnO (e), MnP(OAc)-ZnOcom (f), ZnO (g) and ZnOcom (h) (omitted region of
-1
2
800-1850 cm region, due to the absence of bands).
The UV-Vis spectra observed for the solids which present metalloporphyrins in their
structures are shown in Figure 8, while the spectra recorded for the supports and for
MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom are shown in Figure S11.
The spectra for MnP(OAc)-ZHN and MnP(Cl)-ZHN (Figure 8d and 8e, respectively),
the Soret and Q bands can all be observed in the same region of the spectra as for the
corresponding MPs, MnP(OAc) and MnP(Cl) (Soret at 467 nm and Q band at 565 nm,
Figure 8b and 8c, respectively), indicating that the immobilization process does not lead to
complexes demetalation nor structural distortions. In addition, the bands related to Mn-L
charge transfer were also observed in the 400-300 nm region [77,78]. The Mn-L charge
transfer bands are overlaid by the highly intense ZnO bands that appears below 400 nm
[
77,78].

g
f
e
d
c
b
a
200
300
400
500
600
700
Wavelength / nm
Figure 8. UV-Vis spectra of the solids: free base porphyrin P (a), (MnP(OAc) (b), MnP(Cl)
c), MnP(OAc)-ZHN (d), MnP(Cl)-ZHN (e), MnP(OAc)-ZnO (f) and MnP(Cl)-ZnO (g). For the
analysis pellets of P and MnP was diluted in MgSO
(
4
.
The characteristic Soret bands and Q bands of the MPs were also observed in the
UV-Vis spectra of the solids MnP(OAc)-ZnO and MnP(Cl)-ZnO (Figure 8f and 8g
respectively), confirming the presence of the MPs in the support. However, bands at 427 nm
(
MnP(OAc)-ZnO) and at 423 nm (MnP(Cl)-ZnO) were also observed in the spectra of these
compounds, which are significantly shifted when compared to the Soret band of the
-
-
Mn(III)P(X) (X= Cl or OAc , around 460 nm) or free base porphyrin at 414 nm (P), suggesting
that the species immobilized on ZnO solid that are related to this band are not the Mn(III)P(X)
or the free base porphyrin P (Figure 8). Blue shift of the Soret band of Mn(III)P are frequently
reported when the porphyrins species containing Mn(II), Mn(IV)-oxo or Mn(V)-oxo are formed
[
79-82]. The formation of the species Mn(IV)-oxo and Mn(V)-oxo were frequently observed
as intermediate species in organic substrate oxidation reaction catalyzed by Mn(III)P resulted
from the reaction of Mn(III)P and excess of an oxygen donor reactant such as
iodosylbenzene, H
2
O
2
, organic peracids (e.g. mCPBA), sodium persulphate or hypochlorite
[
81-85]. Since the solids MnP(OAc)-ZnO and (MnP(Cl)-ZnO were not treated with any of
these oxidant agents mentioned, it is reasonable to suggest that the observation of this
bands around 420 nm can be attributed to the presence of manganese(II) species metalating
the porphyrin ligand on the solids, which may have been formed via reduction of the partial
quantity of Mn(III)P complex during the immobilization and hydrothermal decomposition of

the ZHN, in order to obtain ZnO [83,86,87]. Corroborating this, after the immobilization
process, two additional bands (418 and shoulder at 430 nm) were also observed in the
supernatant solution spectrum, resulting from the immobilization process (Figure S12,
Supplementary Information). The presence of these two new bands may indicate a process
III
II
of partial reduction of the Mn /Mn metal center during the immobilization process in the ZnO
matrix at high temperature, resulting in the presence of the Mn(II)P complex and also some
residual free-base porphyrin species [88,89]. In addition, the EPR spectrum of the
supernatant solution resulted from the immobilization process (Figure S13) shows signals in
the 3300 G region characteristic of manganese(II) ions [90,91].
In spite of Mn(III) presenting a more stable oxidation state in metalloporphyrins in
solution, it was reported that some reduction extension was observed in different solvents
such as ethanol, toluene and water under a light irradiation [82]. In the last solvent, the
reduction reaction showed dependence of the pH. It was better at higher pH values due to
the high concentration of hydroxide ions that cause the deprotonation of possible coordinated
water molecules to Mn(III) ion, contributing to reduce the metal in an intramolecular
mechanism. The basic form with OH coordinated to manganese(III) porphyrin
(
2
(OH)(H O)Mn(III)P) was considerably more reactive than its protonated analogue with water
molecules as axial ligands ((H
2
O) MnP) [92].
2
Another possible way to explain the observation of the band around 420 nm in the
solids MnP(OAc)-ZnO and MnP(Cl)-ZnO spectra is the possible transmetallation of the
Mn(III) by Zn(II) ion during the process of thermal decomposition of the ZHN to produce ZnO,
considering it is well know that Zn(II)P presents Soret band at 418-420 nm [93]. In fact,
transmetallation was investigated by Spasojević et al. [94], which in an in vivo study showed
that a highly positively charged manganese(III)porphyrin presented antioxidant therapeutic
III
properties in mitochondria. They developed a sensitive method for Mn P determination in
tissues based on the exchange of porphyrin Mn(II) with Zn(II) (in situ transmetallation)
II
followed by the HPLC/fluorescence detection of Zn P [94]. However, the transmetallation
hypothesis is not the focus of our investigation since, as before mentioned,, in all supernatant
of the thermal decomposition reaction (MnP immobilization on ZnO) the EPR signal was
observed around g = 2.0, suggests paramagnetic manganese species, which corroborate the
reduction to Mn(III) (Figure S12).
To prove the instability of Mn(III) porphyrin in water under reflux (100 °C, same
temperature used in the reaction of the thermal decomposition of ZHN to ZnO), the aqueous
solution of Mn(III)P(OAc) was heated and the solution was analyzed by UV-Vis for 8 h
(
Figure S14, top). It was observed the appearance of a new band at the 420 nm region with
the concomitant decrease of the Soret band in the region of 460 nm suggesting the formation
of the Mn(II)P. Furthermore, it was also observed in the solution after 8 h of heating the same

EPR signal typical of Mn(II) species confirming the reduction of the metal ion under this
temperature/water condition (Figure S14, bottom).
As expected, no bands related to the presence of porphyrin or metalloporphyrin were
observed in the spectra recorded for the solids MnP(OAc)-ZnOcom and MnP(Cl)-ZnOcom
,
(
(
Figure S11), in agreement with the metal quantification obtained by ICP-OES analysis
Table 1).
Figure 9 shows the recorded EPR spectra for all solid materials. The absence of
paramagnetic species in the ZHN support causes this material to be EPR-silent [28].
However, the MnP(OAc)-ZHN solid presents a set of low intensity signals in the 3300 G
region, also suggesting the presence of some small quantity of EPR active of manganese
impurities [95,96]. This result may suggest that even at lower temperatures, such as the
room temperature used in the immobilization of MnP in ZHN solid, it can cause the reduction
of manganese in a smaller proportion.
i
h
g
f
e
d
c
b
a
2500
3000
3500
4000
4500
Magnetic Field / Gauss
Figure 9. EPR spectra of the solids: ZHN (a), MnP(OAc)-ZHN (b), MnP(Cl)-ZHN (c), ZnOcom
d), MnP(OAc)-ZnOcom (e), MnP(Cl)-ZnOcom (f), ZnO (g), MnP(OAc)-ZnO (h) and MnP(Cl)-
ZnO (i).
(
The spectra registered for the samples supported in zinc oxide (ZnO, ZnOcom
,
MnP(OAc)-ZnOcom, MnP(Cl)-ZnOcom, MnP(OAc)-ZnO and MnP(Cl)-ZnO) presented a
signal around 1.95 G, probably relative to defects on the ZnO solid lattice that produce EPR
signal typical of radical species [97-99]. Possible species of manganese(II) in these materials

were not observed in the spectra since the intense radical signal is located in the same
region expected for the manganese(II) species.
3.2 Catalytic activity studies
The catalytic activity of the prepared materials was investigated in the cyclohexane
oxidation, a reasonably inert saturated hydrocarbon [21,100-102], which allows to evaluate
the catalytic efficiency and selectivity since cyclohexanol and/or cyclohexanone are typically
the major products formed in reactions catalyzed by MPs [6,7,16,100,103].
The catalytic reactions were investigated using iodosylbenzene as oxidant in both
absence and presence of an artificial light (exposure to the radiation of a 250 W halogen
lamp with a quartz bulb). This lamp can simulate solar radiation since it has an intense
emission in the visible and the infrared region, with a maximum between 600 and 1000 nm
(
Figure S15, Supplementary Information). In addition, the bulb of the lamp is made of quartz,
which is transparent to the UV-C radiation, as illustrated by Figure S16 (Supplementary
Information). The catalytic results obtained are shown in Table 2.
As already observed by our group [28,29,35,45,102] and other systems reported in
the literature [83,87,91,104-107] neither the supports used (ZHN, ZnO or ZnOcom) nor the
PhIO oxidant are capable of promoting the oxidation of cyclohexane under dark conditions
(
entries 10,11 and 12, Table 2). That being said, when assisted by light and in the presence
of oxidant, the ZHN, ZnO and ZnOcom supports exhibited modest production of both
cyclohexanol (< 10%) and cyclohexanone (< 10%), as can be seen in Table 2. The PhIO
oxidant alone was not capable of promoting such reaction [91].
The photochemical properties of ZnO are already reported in the literature [56,108-
110]. These results indicate that the amount of emission in the UV-A region (320-390 nm), on
which the highly intense ZnO band at 365 nm is present, is enough to promote the oxidation
of cyclohexane in low yield. Moreover, as can be seen from Figure S16 (Supplementary
Information), at the distance used to perform the experiments (9 cm), a small emission of
radiation in the region of UV-C (254 nm) was also registered. This small amount may be
enough to form oxidation products via photocatalysis promoted by the ZHN support,
considering this compound has an absorption band at 225 nm (Figure S11, Supplementary
Information).
In a homogeneous medium, the complexes MnP(OAc) and MnP(Cl) show catalytic
efficiency and high selectivity for alcohol under dark conditions (Table 2, entry 1 and 2), as
already observed in other metalloporphyrin systems [8,105,111].
In solution, the effects of the axial ligand (X) on the catalytic activity have already
III
been studied by using first generation ironporphyrins, XFe (TPP) (with PhIO as the oxygen
III
atom donor) [106] and XMn TTP (with O
2
as the oxygen atom donor, in Lion system) [112].

In these works, the authors reported better catalytic results with the stronger electron
withdrawing axial ligands, and, the metalloporphyrin with acetate ion as axial ligand
presented a slightly better catalytic performance when compared to the one containing
chloride. In our studies, in spite of the fact that it is not possible to affirm that the chloride ion
is present in the MnP(Cl) complex as an axial coordinated ligand (X), using a second
generation porphyrin [XMn(TDFSPP)], the MnP(Cl) presented a better catalytic activity than
the MnP(OAc) (about 43% versus 32% of alcohol yield respectively, Table 2, entries 1 and
2) in dark conditions. As the FTIR analysis suggested that the acetate ion is presented in the
solid MnP(OAc) as counterion, it is not possible to exclude that methanol solvent (the
metallation process solvent) or acetonitrile (catalytic reaction solvent) is present in the
compound MnO(OAc) as an axial coordinator ligand, and consequently, in the catalytic
active species in homogeneous system [113]. These possible solvent molecules acting as
axial ligands on the MnP(OAc) complex seems not to act as positively in the catalytic results
as the chloride ion in MnP(Cl). A possible explanation for the best catalytic result presented
-
by the catalyst MnP(Cl) in comparison to MnP(OAc) is that the axial ligand (X = Cl ) in the
V
intermediate catalytic species (XMn O(porphyrin) (see explanation below, Figure 10, species
C) can increase the electrophilicity of the trans oxo ligand, leading to better catalytic results
to MnP(Cl) in comparison to MnP(OAc) [106,112].
Although the cyclohexanol and cyclohexanone have been the only monitored
products in the investigated catalytic reactions, it is well know that some cyclohexyl chloride
is also a possible product formed when chloride ion is the axial ligand of the
metalloporphyrins [104,107,114,115]. This product is directly related with the alkane radical
cage escape (see explanation in the sequence, Figure 10). In this way, the product total yield
percentage of about 43% observed for the catalytic reaction involving MnP(Cl) might be
underestimated (entry 2, Table 2).
When in the presence of light, the inverse tendency of the catalytic results was
observed in comparison to dark conditions, and the total product yield for MnP(OAc) was
higher than MnP(Cl) (about 47% versus 36% respectively, Table 2, entries 1 and 2).
Besides, MnP(Cl) exhibits a slight increase in the ketone production with a corresponding
decay in alcohol yield. Interestingly, the observation of cyclohexanone as a product in the
reaction catalyzed by this MnP suggests that under light condition, some carbon radicals can
escape from the solvent cage, according to the mechanism proposed to this catalytic
reaction (Figure 10) [101,104,107,114,116]. The solvent cage is the intermediary set of
●
species represented by [(C) + R-C ] (Figure 10), on the admitted mechanism of oxidation of
alkanes catalyzed by metalloporphyrins and oxygen donors. The alkyl radical species R-C●
resulted from the hydrogen atom abstraction of the generic substrate R-C-H by a
(
X)manganese(V)oxo porphyrin intermediate considered as catalytic active species (species

B, Figure 10). The (X)manganese(V)oxo is generated by the reaction of XMn(III)P (species
A, Figure 10) with PhIO. Besides R-C● there is also in the solvent-cage the
(
X)manganese(IV) hidroxo porphyrin species (species C, Figure 10) [28,101,104,116]. The
more frequent occurrence of cage escaping of the alkyl radical in manganese(III) porphyrin-
catalyzed reactions in comparison to Fe(III) porphyrin in solution has been attributed to the
IV
IV
higher stability of the intermediate XMn OH(porphyrin) in comparison to XFe OH [101,107].
When in assisted by light condition the MnP(OAc) catalyst exhibited high efficiency
and selectivity for the alcohol product (entries 1,Table 2). This result suggested that the “in
cage” reaction (also known as cage collapse reaction), which forms alcohol [101,104,116], is
favored when the catalyst MnP(OAc) was used (Figure 10, reaction 1).
Figure 10. Schematic representation of the reaction of generic metallo(III)porphyrin (A) and
iodosylbenzene (PhIO) to form the catalytic active intermediate species (B) (high valent
metal oxo species [117,118]) and the cage solvent species high valent hydroxo-
metallo(IV)porphyrin (C) and the alkyl radical species (R-C•) that came from the hydrogen
abstraction from the substrate generic represented by R-C-H.
Nevertheless, using MnP(Cl) as catalyst, some cyclohexanone yield was observed
(
entry 2, Table 2) which may indicate that the intermediary set of species formed in solvent
IV
C
●] can be stable enough to allow the escape of some
cage [(Cl)Mn OH(porphyrin)+ H11
5
organic radical species. Since the reactions were carried out in aerobic conditions, the
organic radicals can be trapped by “out of cage” dioxygen, resulting in the cyclohexanone in
the reaction catalyzed by MnP(Cl) (Figure 10, reaction 2).
The radical escape way in the mechanism of catalytic oxidation of alkanes by (X)MnP
is also responsible for the formation of cyclohexyl chloride product in the oxidation of
cyclohexane catalyzed by metalloporphyrins when the X axial ligand is the chloride ion
(
Figure 10, reaction 3) [101,104,107,114,115]. Yet, similar to the reactions carried out in dark
conditions, this product was not monitored and, once again the product total yield of about
6% observed for MnP(Cl) under light condition (entry 2, Table 2) can be underestimated,
3

justifying the contradictory result (in comparison to dark conditions) that MnP(Cl) presents
lower product total yield under light condition in comparison to MnP(OAc) catalyst.
The heterogeneous catalysis using the solid MnP(OAc)-ZnOcom shows catalytic
results similar to those observed for the ZnOcom support both in the dark and illuminated
conditions. This result agrees with the characterization of these materials, which indicates
that there is no MPs immobilized in the ZnOcom
.
The heterogeneous catalysis promoted by the solids MnP(OAc)-ZHN and MnP(Cl)-
ZHN presented very interesting results and quite different from the homogeneous catalysis.
Under dark conditions, the catalyst MnP(Cl)-ZHN showed selectivity to alcohol
product as observed in homogeneous conditions but the efficiency to the products was lower
as expected for heterogenous catalysis (product total yield = 19.8%, Table 2 entry 4) in
comparison to the homogenous one (product total yield = 42.7%, Table 2 entry 2). In this
heterogeneous catalytic condition using this solid, it is not possible to disregard that some
amount of cyclohexyl chloride could have been produced but not quantified.
Under the same conditions, catalyst MnP(OAc)-ZHN showed catalytic efficiency
(
Table 2, entry 3, product total yield = 31.2%) quite similar to homogeneous conditions (entry
, total yield = 31.8%) with the selectivity favoring the production of the ketone, despite the
1
alcohol product (alcohol/ketone ratio = 0.2). This selectivity result was completely opposed to
the result observed in homogeneous catalysis using this MnP (Table 2, entry 1,
alcohol/ketone ratio = 32).
The ketone production higher than alcohol has already been observed in previous
studies in catalytic oxidation of cyclohexane using a family of iron porphyrins [28]. In that
work the high selectivity to ketone, a product resulting from free radical escape of the solvent
cage (Figure 10, reaction 2), was attributed to the influence caused by the how the
immobilization of the iron porphyrin in the ZHN support took place. As observed for the solid
FeP-ZHN reported before [28], the MnPs in the solids MnP(OAc)-ZHN are predominantly
immobilized onto the surface of ZHN. The complex MnP(OAc)/support environment created
IV
by this interaction might be more stable (solvent cage species [(X)Mn OH(porphyrin) + R-
●
C ]) than in solution in dark catalytic reaction conditions. This can make the OH transfer to
●
the substrate radical R-C to form cyclohexanol in the “in cage” reaction harder (Figure 10,
reaction 1), thereby causing radical escape favoring the formation of radical escape
products.
Although the total yield of products for the solid MnP(Cl)-ZHN (19.8%, entry 4, Table
2) is smaller than the observed to the catalyst MnP(OAC)-ZHN (31.2% entry 3, Table 2), just
analyzing the alcohol product, the solid MnP(Cl)-ZHN can be considered a better catalyst for
hydroxylation reaction than MnP(OAc)-ZHN, agreeing with the results observed for
homogeneous catalysis under dark conditions and the positive catalytic effect of the axial

ligand in the complex MnP(Cl). The combination of these results suggests that the ZHN
support and the axial ligand ion chloride actively influence the oxidation mechanism and the
solid ZHN is not just an innocent support for the metalloporphyrin immobilization.
It is difficult to understand the exact influence of the support and how it is modifying
the product distribution beyond the role of isolating the Mn centers to prevent the formation of
inactive species and to facilitate the recovery and reuse.
As the same tetra anionic manganese porphyrin is present in both complexes named
by MnP(Cl) and MnP(OAc), originally we expected the immobilization on the support by
strong electrostatic interactions for the two complexes [28]. However, the fact that the
chloride ion is possibly axially coordinated in MnP(Cl) and the acetate ion is the counter ion
of MnP(OAc), it can lead to different modes of immobilization of these two MnP on the ZHN
as illustrated in the comparison of the schematic representation of MnP(Cl) immobilization
(
Figure 11, model A) and MnP(OAc) immobilization, where a possible axial ligand X (a
solvent molecule, for example) cannot be discarded (Figure 11, models B and C).
Because of the presence of the axial ligand chloride ion in MnP(Cl), this complex
could preferably be immobilized on ZHN by electrostatic interaction by one or two positions
of the meso phenyl sulfonate porphyrin substituents, as recently discussed by our group [11],
with the chloride ion remaining axially coordinated to manganese after immobilization [113]
(
Figure 11, model A). Still, as the MnP(OAc) does not have an anionic axial ligand (except
for a possible solvent molecule), it may be interacting with ZHN with unsaturated bonds of
some hydroxyl group from the ZHN border or some OH groups occupying the water positions
in the apical positions of the tetrahedral zinc in addition to electrostatic interactions [119]
(
Figure 11, model B and C). These different possible modes of immobilization can lead to
III
greater or lesser stabilization of the species generated from the reaction between Mn P and
iodosylbenzene (Figure 10, species B and C) which can reflect on the selectivity of the
reaction for different products.

Figure 11. Schematic representation of the immobilization of the complex MnP(Cl) (model A)
and complex MnP(OAc) (model B and C) on the lamella of the solid ZHN. The scales are
close to the real dimensions.
It is well known that the axial coordination of different ligand species to
III
III
Mn Porphyrins (and also Fe Porphyrins) can change the catalytic behavior in terms of
V
efficiency [106,107] and selectivity [91,107,120]. The X axial coordination on the (X)Mn oxo
species (such as the OH from the ZHN solid) can, for instance, keep the metal more inside
the porphyrin ring (in plane configuration). Nappa et al. [107] proposed that due to the
HOMO-LUMO gap relatively small for the hexacoordination geometry of Mn(V)oxo species
(
Figure 10, species B) it is expected a considerable contribution from spin-unpaired
configurations of the two d electron to this species. This fact can contribute to a
oxomanganese intermediate active species with an oxy radical character, resulting
subsequently in products typical from the radical escape of the solvent cage such as ketone,
as observed to the catalyst MnP(OAc)-ZHN. The more in plane configuration of the
manganese porphyrin [(F)Mn(TTP)], caused by instances such as fluoride ion as axial ligand,
also promoted best catalytic results when cyclooctane was oxidized by O to ketone [112].
2
Alternatively, for the catalyst MnP(OAc)-ZHN, the in plane configuration of the
hydroxomanganese(IV)porphyrin generated in sequence (species C in solvent cage shown in
Figure 10 (OH)Mn(IV)(OH)porphyrin) could also be more stable, favoring radical escape of
solvent cage, promoting the formation of more ketone than alcohol.
One can argue that chloride ion can also be present as a sixth coordination axial
ligand in the catalytic intermediate species using the solid MnP(Cl)-ZHN as tentatively
illustrated in Figure 11 (model A). This possibility is not discarded since the reaction with this
solid catalyst also yields some ketone as product and likewise possibly some amount of

cyclohexyl chloride (no quantified). However, due to the steric constrains caused by the
bigger size of the chloride ion (~180 pm) in comparison to hydroxo ion (~130 pm) [121,122]
V
this effect can be minimized and the Mn oxo with choride axial ligand is more out of the
V
plane than in plane when compared to Mn oxo with hydroxo axial ligand. The
oxomanganese(V) species (species B, Figure 10) with the manganese out of the plane of the
porphyrin, results in an increase of the HOMO-LUMO gap and a spin-paired configuration of
the d electrons causing a significantly reduction of the oxy radical character of the
oxomanganese intermediate, favoring the cage collapse that yield more alcohol [107], as
observed for the solid catalyst MnP(Cl)-ZHN.
The selectivity for the radical escape product ketone is not maintained in the catalysis
promoted by MnP(OAc)-ZHN when assisted by light (Table 2, entry 3, A/K ratio = 4 under
light versus 0.2 under dark condition). This result suggests that the catalyst
MnP(OAc)/support ZHN interaction can be changed in the presence of the artificial radiation
and the oxygen transfer in the solvent cage (‘in cage’ reaction) was favored, as well as in the
homogeneous catalysis.
A plausible explanation for these catalytic results is the breakdown of the axial bond
of (ZHN)-hydroxo-manganese promoted by the irradiation of the system. As it is expected
that MnP(OAc) is also immobilized on the support ZHN by electrostatic interactions, the
complex remains attached to the support even without having an axial coordination similar to
that observed for the MnP(Cl)-ZHN solid. In fact, no MnP(OAc) leaching was observed after
the catalytic reaction suggesting that the complex remains attached to the ZHN support.
Moreover, using the MnP(Cl)-ZHN catalyst in the presence of light, the alcohol
production was increased by ca. 15% when compared to the absence of light, confirming that
the MnP(Cl) immobilized or in solution is a better catalyst for cyclohexane oxidation than
MnP(OAc). It was also recently observed [45] the positive effect of the light in catalytic
systems based on two metalloporphyrins immobilized on hydrozincite/TiO anatase, used for
2
oxidation of cyclohexane. In spite of the fact that the ketone yield increased in the light
condition for MnP(Cl)-ZHN, suggesting some radical escape route in comparison to the dark
ones, this result should be viewed with caution since the oxidation reaction using only ZHN
as a catalyst already shows a large production of cyclohexanone.
In the absence of light, although the MnP(OAc)-ZnO and MnP(Cl)-ZnO catalysts
show selectivity for the alcohol product (9.7% entry 5 and 6.2% entry 7, respectively, Table
2), lower yields were observed, comparing to those presented when the MPs were
immobilized on ZHN. This result may be due to the aforementioned possible presence of
manganese(II) species in the catalyst, which were evidenced by UV-Vis analysis (Figure 8).
These lower catalytic activities could also be attributed to the MnP immobilization modes in

the surface of the oxide [11,16]. Besides, the access of the substrate to the catalyst species
should be hindered if the active site is not easily accessible.
Under light, on the other hand, the yield of alcohol in the reactions catalyzed by both
solids MnP(OAc)-ZnO and MnP(Cl)-ZnO were increased by approximately four times,
suggesting some synergistic effect of the light and the prepared catalyst particularly the ZnO
support.
Again, similar to the heterogeneous catalysis using the solids with ZHN as support,
the yield of cyclohexanone originated from the solvent cage radical escape (and possible too
cyclohexyl chloride) could not be appropriately rationalized since the control reaction using
ZnO solid as catalyst also presented cyclohexanone yield.
The reuse capacity of the solids MnP(OAc)-ZnO and MnP(Cl)-ZnO was investigated
in further cyclohexane oxidation reactions (entries 6 and 8 respectively, Table 2). In both
cases (dark or light), the catalytic activity of the solids were maintained but it was reduced in
comparison to the first reaction cycle, mainly for the solid MnP(Cl)-ZnO under light
conditions suggesting some destruction of the catalyst immobilized in this solid. Alternatively
these results can also be attributed to the poisoning of the catalyst consequently leading to
decrease in the product yields, as observed in other works [16, 86], since MPs leaching was not
observed in the catalytic conditions, as confirmed by UV-Vis spectroscopy.

Table 2. Results of the catalytic oxidation of cyclohexane using iodosylbenzene as oxidant without light and assisted by the artificial radiation (exposure to
a halogen lamp - 250W).
Yield /%
Yield /% darknessc,d
Products
total yield /%e
A/K
Products
total yield/%e
A/K
Entry Catalyst
assisted by artificial radiationc,d
ratio
ratio
A
K
A
K
1
2
3
4
5
6
7
8
9
1
1
1
MnP(OAc)b
MnP(Cl)b
31.8 ±3.3
42.7 ±3.3
4.9 ± 0.8
17.5 ± 2.8
9.7 ± 0.5
6.5 ± 0.86
6.2 ± 1.0
5.0 ± 1.5
ndg
< 1
31.8
42.7
31.2
19.8
9.7
6.5
6.2
5.0
-
32
43
0.2
7
46.9± 1.0
32.0 ± 1.0
28.2 ± 3.9
40.9 ± 7.6
36.3 ± 0.6
23.5 ± 5.9
27.1 ± 7.5
15.7 ± 1.1
6.9 ± 0.90
8.8± 2.7
< 1
46.9
36.1
35.1
54.9
47.0
35.4
41.0
15.7
17.3
20.9
15.3
-
47
8
< 1
26.3 ± 3.1
2.3 ± 0.5
< 1
4.1 ± 0.27
6.9 ± 1.2
14.1 ±3.7
10.8 ± 0.7
11.9 ± 0.7
13.9 ± 3.5
< 1
MnP(OAc)-ZHN
MnP(Cl)-ZHN
MnP(OAc)-ZnO
First Reuse
MnP(Cl)-ZnO
First Reuse
MnP(OAc)-ZnOcom
ZHN
4
3
10
6
2
< 1
4
< 1
6
2
< 1
5
16
0.7
0.7
1
nd
-
10.4 ± 1.3
12.1 ± 0.67
7.4 ± 2.2
nd
0
1
2
Nd
nd
-
-
ZnO
Nd
nd
-
-
7.9 ± 1.8
nd
PhIO (non catalyzed)f
Nd
nd
-
-
-
aReaction conditions: solvent – acetonitrile; room temperature; magnetic stirring; one hour; catalyst/oxidant/cyclohexane molar ratio of 1:20:2000.
b
-
-
c
d
homogeneous phase catalysis using MnP(X) as catalysts (X = Cl or OAc ). Yields based on the amount of PhIO used in the reaction. A =
e
f
cyclohexanol and K = cyclohexanone. Products = only the cyclohexanol (A) and cyclohexanone (K) yields were considered. Reaction in the
g
absence of catalyst (control reaction). nd = non-detected.