Molecular Engineering of MnII Diamine Diketonate Precursors for the Vapor Deposition of Manganese Oxide Nanostructures

Article abstract of DOI:10.1002/chem.201703423

Molecular engineering of manganese(II) diamine diketonate precursors is a key issue for their use in the vapor deposition of manganese oxide materials. Herein, two closely related β-diketonate diamine Mn^II adducts with different fluorine contents in the diketonate ligands are examined. The target compounds were synthesized by a simple procedure and, for the first time, thoroughly characterized by a joint experimental–theoretical approach, to understand the influence of the ligand on their structures, electronic properties, thermal behavior, and reactivity. The target compounds are monomeric and exhibit a pseudo-octahedral coordination of the Mn^II centers, with differences in their structure and fragmentation processes related to the ligand nature. Both complexes can be readily vaporized without premature side decompositions, a favorable feature for their use as precursors for chemical vapor deposition (CVD) or atomic layer deposition applications. Preliminary CVD experiments at moderate growth temperatures enabled the fabrication of high-purity, single-phase Mn₃O₄ nanosystems with tailored morphology, which hold great promise for various technological applications.

Full text of DOI:10.1002/chem.201703423

DOI: 10.1002/chem.201703423
Full Paper
&
Chemical Vapor Deposition
Molecular Engineering of MnII Diamine Diketonate Precursors for
the Vapor Deposition of Manganese Oxide Nanostructures
Chiara Maccato,[a] Lorenzo Bigiani,[a] Giorgio Carraro,[a] Alberto Gasparotto,[a]
Roberta Seraglia,[b] Jiyeon Kim,[c] Anjana Devi,[c] Gloria Tabacchi,*[d] Ettore Fois,[d]
Giuseppe Pace,[b] Vito Di Noto,[e] and Davide Barreca*[b]
Abstract: Molecular engineering of manganese(II) diamine
diketonate precursors is a key issue for their use in the
vapor deposition of manganese oxide materials. Herein, two
closely related b-diketonate diamine MnII adducts with differ-
ent fluorine contents in the diketonate ligands are exam-
ined. The target compounds were synthesized by a simple
procedure and, for the first time, thoroughly characterized
by a joint experimental–theoretical approach, to understand
the influence of the ligand on their structures, electronic
properties, thermal behavior, and reactivity. The target com-
pounds are monomeric and exhibit a pseudo-octahedral co-
ordination of the MnII centers, with differences in their struc-
ture and fragmentation processes related to the ligand
nature. Both complexes can be readily vaporized without
premature side decompositions, a favorable feature for their
use as precursors for chemical vapor deposition (CVD) or
atomic layer deposition applications. Preliminary CVD experi-
ments at moderate growth temperatures enabled the fabri-
cation of high-purity, single-phase Mn3O4 nanosystems with
tailored morphology, which hold great promise for various
technological applications.
to anodes of lithium-ion batteries and pseudocapacitors,[1b,2b,3]
up to electrochromic systems,[4] magnetic media,[5] and gas
sensors.[1d,6] In this widespread context, the fabrication of
Mn3O4 nanostructures with tailored morphology (nanoparticles,
nanorods, nanofractals, etc.) has been performed by a variety
of synthetic techniques, encompassing microwave irradiation,
hydrothermal/solvothermal routes, chemical bath deposition,
and chemical vapor deposition (CVD).[1a,b,2a,c,4–7] In particular,
the latter process, along with atomic layer deposition (ALD), is
compatible with current processing standards, thanks to the
capability of achieving in situ, large-area growth of thin films
and nanostructured materials with controlled properties.[8] In
this regard, the development of suitable precursor compounds
endowed with high volatility, thermal stability, and clean de-
composition pathways is a very challenging research area,[9]
which would ideally guide, in a “molecular engineering” ap-
proach, the modulation of material properties in view of the
desired functional applications.
Introduction
Manganese oxide nanomaterials are of considerable impor-
tance for many technological applications thanks to their di-
versified structures and variety of appealing chemical and
physical properties.[1] In particular, Mn3O4, a mixed-valence
oxide with a tetragonal structure, has received attention
thanks to its durability, low cost, and attractive performances
for a variety of end uses, spanning from (photo)catalysts,[1a,b,2]
[a] Prof. C. Maccato, Dr. L. Bigiani, Dr. G. Carraro, Prof. A. Gasparotto
Department of Chemical Sciences, Padova University and INSTM
35131 Padova (Italy)
[b] Dr. R. Seraglia, Dr. G. Pace, Dr. D. Barreca
CNR-ICMATE and INSTM, Department of Chemical Sciences,
Padova University, 35131 Padova (Italy)
E-mail: davide.barreca@unipd.it
[c] Dr. J. Kim, Prof. A. Devi
Inorganic Materials Chemistry
Faculty of Chemistry and Biochemistry
Ruhr-University Bochum, 44801 Bochum (Germany)
To date, the most used CVD and ALD Mn precursors are
mainly based on b-diketonate derivatives,[1e,10] some of which
suffer from poor shelf-life and/or unfavorable thermal properti-
es,[9e] especially if they contain MnII. In fact, MnII complexes
with nonfluorinated b-diketonate ligands are reported to readi-
ly decompose into MnIII derivatives,[10d,11] yielding poor control
of the phase composition of the product. As a consequence,
the obtainment of single-phase Mn3O4 nanomaterials with con-
trolled crystallinity and morphology[12] requires the tailoring of
b-diketonate compound properties at the molecular level. In
this regard, the use of fluorinated ligands, such as 1,1,1,5,5,5-
hexafluoro-2,4-pentanedionate (hfa), is favorable for the ob-
[d] Prof. G. Tabacchi, Prof. E. Fois
Department of Science and High Technology
University of Insubria and INSTM
22100 Como (Italy)
E-mail: gloria.tabacchi@uninsubria.it
[e] Prof. V. Di Noto
Department of Industrial Engineering
Chemical Technology Section
Department of Chemical Sciences, Padova University and INSTM
35131 Padova (Italy)
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under:
https://doi.org/10.1002/chem.201703423.
Chem. Eur. J. 2017, 23, 1 – 11
1
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
tainment of metal complexes with improved shelf-life, thermal
and mass-transport properties relative to those of nonfluorinat-
ed compounds.[9e] The hfa ligand has two ꢀCF3 groups, which
enhance volatility through a decrease in van der Waals inter-
molecular forces[10d] and result in enhanced Lewis acidity of the
metal center,[9e] enabling the effective binding of diamine
Lewis bases, such as N,N,N’,N’-tetramethylethylenediamine
(TMEDA). The introduction of the latter enables a complete sat-
uration of the metal coordination sphere, yielding b-diketo-
nate-diamine compounds with the general formula
M(hfa)2·TMEDA, which feature higher stability towards hydroly-
sis and provide improved thermal/mass-transport properties,[13]
important characteristics for their use as CVD precursors.
formances in the CVD of Mn3O4 nanomaterials. Experimental
results presented herein for the two complexes are validated
and integrated with detailed density functional theory (DFT)
modeling, with the aim of providing a theoretical basis[23] for
the interpretation of similarities and differences in their struc-
ture, bonding, and chemical behavior. Finally, preliminary data
concerning the low-pressure CVD validation of both com-
pounds as Mn molecular sources for high-purity Mn3O4 nano-
deposits on different substrates are also reported.
Results and Discussion
Synthesis and characterization of MnL2·TMEDA compounds
In our previous studies, we devoted our attention to
M(hfa)2·TMEDA complexes of various elements, in particular,
Cu,[13a,14] Co,[13b] Fe,[15] and Zn.[16] Although all of these molecu-
lar systems present a common structural motif, that is, a
pseudo-octahedral MO4N2 geometry, investigations have pro-
vided evidence that specific physicochemical properties, as
well as the features of the obtained CVD products, significantly
depend on the nature of the metal center. Even in the case of
manganese, stable MnII compounds can be obtained by using
fluorinated diketonate ligands, such as hfa.[9e,11,17] Now, the
question arises as to whether the presence of only one ꢀCF3
group for each diketonate could be sufficient to endow the di-
amine adducts with the stability, volatility, and clean decompo-
sition properties required for CVD/ALD applications. If the
weakest bonds of the complex, that is, the first to be broken,
depend on the metal center,[14b,15c,16] the effect of the ligand is
equally important. Notably, a diketonate with a single ꢀCF3
group (indicated hereafter as tfa=1,1,1-trifluoro-2,4-pentanedi-
onate) could be formally obtained by replacing one of the hfa
fluorinated moieties with a methyl group. How would such a
ligand modify the chemistry of these precursors, and to what
extent would their CVD performances be affected? Literature,
unfortunately, offers no clear answer to these questions.
Though M(tfa)2 complexes have been reported for M=Co,[18]
Ni,[19] and Cu,[20] M(tfa)2·TMEDA adducts have been much less
studied than their hfa-containing counterparts. In fact, only
one study of Cu(tfa)2·TMEDA has been reported so far,[21] and
no direct connections between ligand properties and precursor
behavior have been investigated in detail.
Herein, MnL2·TMEDA adducts were synthesized through a pro-
cedure different from that reported in the literature for
Mn(hfa)2·TMEDA,[17] involving the reaction in aqueous mixtures
between MnII chloride and L ligands in the presence of TMEDA
(Scheme 1). The process, which was carried out at room tem-
Scheme 1. The synthesis of MnL2·TMEDA derivatives.
perature with no need for heating at reflux, at variance with a
previous study,[11] yielded the target adducts, which could be
readily manipulated in the presence of air, moisture, and light
without any detrimental degradation. Apart from a shelf-life of
several months, an important feature for CVD applications, the
present MnL2·TMEDA compounds possessed an appreciable
volatility (m.p.=86 and 998C for L=hfa and tfa, respectively[22]
)
and could be readily sublimed under vacuum (ꢁ10ꢀ3 mbar).
The melting point of Mn(hfa)2·TMEDA at atmospheric pressure
was higher than that previously obtained by some investiga-
tors,[11,17] but in line with that reported in a patent quoting the
use of this compound as a gasoline additive.[22]
To elucidate such inter-relations, we investigate herein the
structure/property interplay for two MnII complexes with either
hexa- or trifluorinated diketonate ligands, namely,
Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA. It is worthwhile noting
that, despite the preparation of Mn(hfa)2·TMEDA has already
been reported,[22] only some data on its structure and thermal
behavior are available in the literature,[11,17] whereas a detailed
theoretical–experimental characterization of this compound is
completely missing. The need for these studies is even more
relevant for Mn(tfa)2·TMEDA, which, so far, has been mentioned
only once in a patent as an antiknock additive.[22]
The molecular structures of the two complexes are displayed
in Figure 1, whereas crystallographic and structural refinement
data, as well as geometrical parameters of DFT-calculated
structures, are presented in Tables S1 and S2 in the Supporting
Information. Selected bond lengths and angles are listed in
Table 1. At variance with other cases, such as that of Mn
bis(N,N’-diisopropylacetamidinate)[24] or variously substituted
dialkylmanganese(II) complexes,[8d] both compounds were
monomeric both in the solid state and in solution [see below
for electrospray ionization-mass spectrometry (ESI-MS) results],
which indicated that the use of TMEDA was effective in satu-
Herein, our main aim is to disclose how the degree of fluori-
nation of the ligand in MnL2·TMEDA precursors affects their
physicochemical features, including stability, volatility, and gas-
phase fragmentation, with particular attention to their per-
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the b-diketonate used, the mean MnꢀO and MnꢀN bond
lengths were in agreement with those obtained for coordina-
tion complexes of 2-(4-quinolyl)nitronyl nitroxide[27] and 2,2’-bi-
pyridine[28] with Mn(hfa)2 and for various MnIIꢀhfa compounds,
including Mn(hfa)2·TMEDA,[9e,10d,17,26] although this study con-
tains a better quality of structure refinement for the latter
complex. As observed in Figure 1, both compounds presented
a sixfold coordination around the MnII centers, resulting in a
MnO4N2 distorted octahedral environment, in line with previ-
ous reports for homologous complexes available in the Cam-
bridge Structural Database.[26] In comparison to other
ML2·TMEDA adducts (M=Fe,[15a] Co,[13b] Cu,[13a] Zn[29]), the O-M-
O, O-M-N, and N-M-N bond angles (Table 1) are slightly lower
(up to ꢁ58), whereas MꢀO and MꢀN distances are longer. Sim-
ilarly to the Fe homologue,[15a] the OꢀC bond lengths of b-di-
ketonate ligands were all close to 1.25 ꢁ, a value that suggests
a double-bond character (typical OꢀC single bonds ꢁ1.40 ꢁ).
For both compounds, the atomic distances between O(1)ꢀC(7)
and O(4)ꢀC(14) are slightly longer than those of O(2)ꢀC(9) and
O(3)ꢀC(12), due to ꢀCF3 electron-withdrawing groups being di-
rectly bonded to C(7) and C(14) atoms. In addition, MnꢀO(2)
and MnꢀO(3) distances were slightly longer than those trans
Figure 1. Molecular structures of a) Mn(hfa)2·TMEDA and b) Mn(tfa)2·TMEDA.
Displacement ellipsoids are shown at the 50% probability level. Hydrogen
atoms and rotational disorder from ꢀCF3 groups are omitted for clarity.
Table 1. Selected bond lengths and angles for Mn(hfa)2·TMEDA and
Mn(tfa)2·TMEDA.
Mn(hfa)2·TMEDA
Mn(tfa)2·TMEDA
to the
O
atoms of
L
ligands [MnꢀO(1) and MnꢀO(4),
Table 1[9e]]. A similar trans effect has already been observed for
M(hfa)2·TMEDA compounds with M=Mg,[30] Fe,[15a] Co,[13b] and
Zn.[16,29] Finally, it is worth noting that, for both complexes,
MnꢀN bonds were longer than those of MnꢀO. This effect,
which was particularly evident for Mn(tfa)2·TMEDA, anticipated
an easier opening of the TMEDA ring with respect to that of
the b-diketonate one, as suggested by the calculated bond
orders, electronic population analyses, and decomposition en-
ergies for the two precursors. In both complexes, especially for
Mn(tfa)2·TMEDA, the diketonate is a stronger electron donor
towards Mn, compared with that of the diamine (Table S3 in
the Supporting Information). Accordingly, MnꢀN bonds are sig-
nificantly weaker than those of MnꢀO (Table S4 in the Support-
ing Information), which suggests that, at least in the gas
phase, the TMEDA ligand should be more easily released than
the diketonate one. On this basis, we calculated the precursor
decomposition energy (DE) for the pathways shown in Equa-
tions (1) and (2), with L=hfa/tfa, in vacuum and in methanol,
that is, the solvent used in the present ESI-MS experiments:
bond lengths [ꢁ]
MnꢀO(1)
2.1472(14)
2.1743(14)
2.1546(14)
2.1493(14)
2.2984(17)
2.2989(17)
1.251(2)
1.245(2)
1.244(3)
1.248(3)
2.1481(14)
2.1629(14)
2.1525(14)
2.1265(14)
2.3428(18)
2.3116(17)
1.261(2)
1.255(2)
1.252(2)
1.260(2)
MnꢀO(2)
MnꢀO(3)
MnꢀO(4)
MnꢀN(1)
MnꢀN(2)
O(1)ꢀC(7)
O(2)ꢀC(9)
O(3)ꢀC(12)
O(4)ꢀC(14)
bond angles [8]
O(1)ꢀMnꢀO(2)
O(3)ꢀMnꢀO(4)
N(1)ꢀMnꢀN(2)
O(1)ꢀMnꢀO(4)
O(3)ꢀMnꢀN(1)
O(2)ꢀMnꢀN(2)
MnꢀO(1)ꢀC(7)
MnꢀO(2)ꢀC(9)
MnꢀO(3)ꢀC(12)
MnꢀO(4)ꢀC(14)
82.07(5)
82.60(5)
79.81(6)
171.03(5)
167.48(5)
166.34(6)
130.02(13)
129.06(13)
128.78(13)
129.13(13)
83.21(6)
83.52(6)
78.66(6)
173.85(6)
166.57(6)
166.49(6)
127.70(13)
130.70(13)
131.12(13)
127.72(13)
MnL2 ꢂ TMEDA ! MnL ꢂ TMEDAþ þ Lꢀ
MnL2 ꢂ TMEDA ! MnL2 þ TMEDA
ð1Þ
ð2Þ
rating the MnII coordination sphere. In addition, though the
synthesis was carried out in aqueous mixtures, no water mole-
cules were present in the MnII environment and no classical hy-
drogen bonds formed in the solid-state structure. The latter
feature is of key importance in view of CVD/ALD utiliz-
ation[13,15a] (see below for thermoanalytical data). In contrast,
for Mn(hfa)2·2H2O[10d] and other MnII b-diketonates, such as ad-
ducts of Mn(hfa)2 with substituted nitronyl nitroxides,[25] the
occurrence of hydrogen bonding has been observed.
The geometries of MnL·TMEDA+ and MnL2 fragments were
initially optimized in vacuum. The loss of one ligand strongly
desaturates the manganese coordination sphere, and all frag-
ments exhibit a tetrahedral coordination, as depicted in Fig-
ure S1 in the Supporting Information. For L=hfa, calculations
yielded DE1 =125.4 kcalmolꢀ1 and DE2 =38.0 kcalmolꢀ1, where-
as the corresponding values for L=tfa were DE1 =131.3 kcal
molꢀ1 and DE2 =31.0 kcalmolꢀ1. Hence, in the gas phase, the
loss of a hfa/tfa moiety would be significantly disfavored with
respect to the loss of TMEDA, in line with previously discussed
For both structures shown in Figure 1, X-ray crystal-structure
determination provided evidence of a cis geometry,[26] as also
reported for M(hfa)2·TMEDA with M=Fe,[15a] Co,[13b] and Cu,[13a]
with a twofold axis bisecting the TMEDA ligand. Irrespective of
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
3
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
data. Nevertheless, when the same quantities are calculated in
methanol,[31] the difference substantially decreases, which indi-
cates that the energies involved in the two decomposition
routes become comparable. This is particularly evident for L=
hfa, for which DE1 =30.3 kcalmolꢀ1 and DE2 =28.8 kcalmolꢀ1,
whereas for L=tfa we found values of DE1 =32.9 kcalmolꢀ1
and DE2 =23.3 kcalmolꢀ1. The reaction medium therefore has a
key influence on the fragmentation pathways. The fragmenta-
tion route in Equation (1), highly disfavored in vacuum, be-
comes viable in a polar solvent due to stabilization of the re-
sulting ionic species. This might be particularly important for
ESI-MS experiments, in which the first fragmentation of the
complex occurs in the solvent (see below). On the other hand,
the route in Equation (2) should be favored in the gas phase,
such as in thermal CVD experiments.
moments of the complexes. The results showed that
Mn(hfa)2·TMEDA had a dipole moment considerably larger
than that of Mn(tfa)2·TMEDA, due to the net charge separation
between hfa and diamine ligands, as depicted in the electro-
static potential maps (Figure 3). Indeed, whereas the electro-
UV/Vis optical spectra of the compounds are displayed in
Figure 2. The broad band at lꢁ300 nm, due to electronic
states mostly localized on the diketonate ligands (Figure 2b,
inset), arises from p–p* ligand-to-ligand excitations (see Figur-
es S2–S4 in the Supporting Information). The calculated spec-
tra reproduce the experimental trend, with a spectral shift to
higher wavelengths for Mn(hfa)2·TMEDA compared to
Mn(tfa)2·TMEDA, and the agreement is further improved by
taking into account the solvent contribution. The same trend
is found for the p–p* transitions in isolated hfa and tfa (see
Figure S4 in the Supporting Information), which indicates that
the compound electronic excitation and optical properties are
significantly influenced by the ligand nature.
Figure 3. Electrostatic potential maps for Mn(hfa)2·TMEDA and
Mn(tfa)2·TMEDA. Regions of high (positive) potential (in blue) are electron
poor, whereas regions of low (negative) potential (in red) are electron rich.
White/gray colors represent intermediate electrostatic potential values.
Arrows mark the direction and magnitude of electric dipole moments, m.
Atom color codes: Mn=pink; F=green; O=red; N=blue; C=cyan;
H=white.
static potentials of TMEDA and hfa are positive and negative,
respectively, the tfa ligand exhibits both positive and negative
regions, which are localized on the ꢀCH3 and ꢀCF3 groups, re-
spectively. Hence, the application of external electric fields, as
in ESI-MS experiments (see below), might have different effects
on the two compounds. Calculations indicated that both com-
plexes were slightly stabilized by a moderate electric field, and
showed a slight dipole moment increase, especially in the case
of Mn(hfa)2·TMEDA (see Table S5 in the Supporting Informa-
tion). Such an electric field would therefore favor a preferential
orientation of the complexes, with an enhanced effect for
Mn(hfa)2·TMEDA due to its more asymmetric charge distribu-
tion (Figure 3).
This finding prompted us to investigate more closely the
ligand effect on the electronic structures and electric dipole
An additional insight into the behavior of the two manga-
nese complexes was obtained by means of ESI-MS, a soft ioni-
zation technique that provided important clues about the
compound reactivity. ESI-MS analyses were carried out in both
positive (+) and (ꢀ) ion modes, with the aim of elucidating
the adduct fragmentation pathways and their interplay with
molecular structures. It is worthwhile observing that, to the
best of our knowledge, no such investigation on MnL2·TMEDA
compounds has ever been reported in the literature to date.
In positive-ion mode, the behavior of the two compounds
was qualitatively similar, irrespective of the ligand nature. The
ESI(+) mass spectra (Figure 4) are dominated by single signals
centered at m/z 378 and 324, which correspond to [Mn(hfa)·T-
MEDA]+ and [Mn(tfa)·TMEDA]+, respectively. This result agreed
with those previously obtained for analogous M(hfa)2·TMEDA
compounds, with M=Cu and Co.[13,14]
Figure 2. Experimental (a) and theoretical (b) UV/Vis optical spectra for
Mn(hfa)2·TMEDA (red lines) and Mn(tfa)2·TMEDA (black lines). The orbitals in-
volved in one of the components of the transition state for Mn(hfa)2·TMEDA
are shown in the inset of b) (see Figures S2 and S3 in the Supporting Infor-
mation for graphical representations of all components for the two com-
plexes). Theoretical spectra were calculated both in vacuum (no label) and
with a polarizable continuum model[31] for the solvent ethanol (label ’Solv’).
To gain a deeper insight into the complex fragmentation
pathways, MS2 and MS3 experiments were carried out on
[MnL·TMEDA]+ ions (see Figures S5 and S6 in the Supporting
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
4
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 5. Negative-ion ESI-MS spectra of solutions of a) Mn(hfa)2·TMEDA and
b) Mn(tfa)2·TMEDA in methanol. Calculated optimized structures for the
most abundant ionic species are shown in the insets. Atom color codes are
the same as those used in Figure 3.
Figure 4. Positive-ion ESI-MS spectra of solutions of a) Mn(hfa)2·TMEDA and
b) Mn(tfa)2·TMEDA in methanol. Calculated optimized structures for the
most abundant ionic species are shown in the insets. Atom color codes are
the same as those used in Figure 3.
hancing the probability of a successful TMEDA/hfa ligand ex-
change leading to the observed anion. Finally, it is worth
noting that no dimer/polynuclear species have ever been de-
tected. Considering the ESI-MS soft ionization conditions, this
result suggests that both complexes are monomeric, in agree-
ment with structural analyses (see above).
Information). Irrespective of the ligand nature, MS2 spectra
were characterized by the presence of ions at m/z 190 and
115, corresponding to diamine-related derivatives, with the
first one arising from a ligand-to-metal fluorine transfer pro-
cess. This behavior was directly dependent on the metal
nature, since similar MS2 experiments on [M(hfa)·TMEDA]+ ions
yielded [CuTMEDAꢀH]+, for the Cu derivative,[13a,14a] and
[CoF2·TMEDA+H]+, for the Co one.[13b]
To be successfully employed as CVD/ALD precursors, the
target compounds should possess sufficient stability to ensure
vaporization free from undesired side decomposition, as well
as a constant and reproducible vapor supply. To investigate
the precursor thermal properties as a function of the ligand
nature, thermogravimetric analysis (TGA) was performed on
both MnL2·TMEDA compounds, yielding very similar results for
freshly synthesized and aged sample batches. As observed in
Figure 6a, both target adducts displayed an analogous behav-
ior, characterized by a single-step mass loss for Tꢃ1208C, indi-
cating a high volatility. For Mn(hfa)2·TMEDA, the residual
weight was close to zero for Tꢃ1508C, evidencing the occur-
rence of clean and quantitative vaporization in a narrow tem-
perature range. The latter phenomenon is a key advantage in
view of CVD/ALD applications, especially in comparison with
commonly used Mn precursors, which show either lower vola-
tility [as observed for Mn(dpm)3; dpm=2,2,6,6-tetramethyl-3,5-
heptanedionate] or multi-stage decompositions, with a high
residual weight [as observed for Mn(acac)2(H2O)2; acac=2,4-
pentanedionate].[10c,11] Differential scanning calorimetry (DSC)
analyses (not reported) enabled us to identify the presence of
two endothermic peaks at 84.5 and 100.68C for
Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA, respectively, related to
melting processes. In line with melting point values,
Mn(tfa)2·TMEDA presented a slightly higher volatilization onset
In negative-ion mode, MnL2·TMEDA ESI-MS spectra revealed
a different influence of hfa/tfa ligands on the fragmentation
pathway. The ESI(ꢀ) mass spectrum of Mn(hfa)2·TMEDA (Fig-
ure 5a) was characterized by the presence of ions at m/z 676
and 207, corresponding to [Mn(hfa)3]ꢀ and [hfa]ꢀ, respectively.
Conversely, the corresponding spectrum of Mn(tfa)2·TMEDA
(Figure 5b) displayed only the signal at m/z 153, correspond-
ing to [tfa]ꢀ ions. MS/MS analyses on [Mn(hfa)3]ꢀ ions yielded
the sole hfaꢀ (see Figure S7 in the Supporting Information), in
line with previous results obtained in the ESI-MS analysis of
Fe(hfa)2·TMEDA.[15a] Different behavior of the two complexes
emerging from the results shown in Figure 5 suggested a dif-
ferent binding capacity of hfa and tfa ligands towards MnII
center. Indeed, both [MnL3]ꢀ adducts were predicted to be
stable with respect to the separated Lꢀ and MnL2 fragments,
but the calculated formation energies differed by 2.0 kcalmolꢀ1
in methanol (7.4 kcalmolꢀ1 in vacuum) in favor of [Mn(hfa)3]ꢀ
(see Figure S8 in the Supporting Information). Furthermore,
since the formation of [MnL3]ꢀ adducts would involve the frag-
mentation of at least two MnL2·TMEDA molecules, it might be
argued that the higher Mn(hfa)2·TMEDA dipole moment could
promote a head-to-tail alignment of two such molecules, en-
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 6. a) TGA profiles for MnL2·TMEDA complexes. Isothermal weight
changes recorded for b) Mn(hfa)2·TMEDA and c) Mn(tfa)2·TMEDA.
Figure 7. Glancing-angle XRD patterns of Mn3O4 systems deposited at
4008C: a) on Si(100), from Mn(hfa)2·TMEDA; and b) on SiO2, from
Mn(tfa)2·TMEDA. Vertical bars mark the relative intensities of the Mn3O4
powder spectrum. Inset: representation of the Mn3O4 solid-state structure.[32]
than Mn(hfa)2·TMEDA, in line with the lower fluorine content
of the former compound (see above).[10d] In addition, a non-
zero residual weight, progressively lowering from 130 to
6008C, could be observed.
Isothermal analyses (Figure 6b and c) carried out for 2 h evi-
denced a nearly constant weight loss as a function of time for
both compounds. Such results, in line with previous reports on
Fe, Co, and Cu hfa derivatives,[13,15a] enabled us to rule out det-
rimental decomposition phenomena and confirmed the occur-
rence of clean vaporization, an important feature for CVD/ALD
applications.
ples, characterized by a good adhesion to the substrate, were
preliminarily investigated by X-ray diffraction (XRD, Figure 7),
which revealed the formation of body-centered tetragonal
Mn3O4 [haussmannite; space group: I41/amd;[1a,2a,32] lattice pa-
rameters a=5.762 ꢁ, c=9.470 ꢁ; average crystallite size=
(40ꢅ5) nm], with MnIII and MnII centers in octahedral and tet-
rahedral sites, respectively[2b,5b,7] (Figure 7, inset). Irrespective of
the substrate used, no reflections related to other Mn oxides
or MnII fluoride could be detected, indicating the obtainment
of phase-pure systems, as also confirmed by X-ray photoelec-
tron spectroscopy (XPS; see Figure S9 in the Supporting Infor-
mation). The system morphology, analyzed by means of field
emission-scanning electron microscopy (FE-SEM, Figure 8a and
b), revealed the presence of well interconnected lamellar struc-
tures [average dimensions=(270ꢅ50) nm] uniformly distribut-
ed over the substrate surface. From the mean nanodeposit
thickness [(350ꢅ20) nm], an average growth rate of
6 nmminꢀ1 could be estimated.
CVD depositions from MnL2·TMEDA
An important point of this study is the functional validation of
MnL2·TMEDA compounds to assess their potential as CVD pre-
cursors for the fabrication of manganese oxide nanosystems.
Preliminary deposition experiments were carried out on both
Si(100) and SiO2 substrates, by using vaporization (ꢄ658C) and
growth (4008C) temperatures lower than those previously
adopted in vapor phase processes from conventional manga-
nese precursors, such as Mn(hfa)2 and Mn(dpm)3, and also
from Mn(hfa)2·TMEDA.[1e,10a,b,e,11,17] The obtained brownish sam-
The compositional purity of Mn3O4 systems was confirmed
by EDXS analysis. The obtained spectrum (Figure 8c) showed
the presence of MnKa and MnKb peaks located at 5.90 and
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
from cracks/pinholes. From the line height profile, a root-
mean-square (RMS) roughness of 5 nm could be evaluated.
Finally, optical absorption analyses were carried out (Fig-
ure 9c). The spectral shape was in agreement with that report-
ed for Mn3O4-based materials.[4] As observed, the system was
almost transparent in the IR range, whereas significant absorp-
tion at lower wavelengths (l<600 nm) corresponded to inter-
band transitions. The optical band gap was evaluated by the
Tauc method, by plotting (ahn)n versus hn (Figure 9c, inset),
with n=2 corresponding to direct allowed transitions,[1d,33] and
extrapolating the obtained trend to zero absorption. The esti-
mated value (EG =2.5 eV) was in agreement with that previous-
ly reported for Mn3O4,[4,5] and highlighted the efficient harvest-
ing of visible light, paving the way to the use of the developed
materials in solar-assisted applications.
Conclusion
This study was devoted to the preparation and combined ex-
perimental/theoretical characterization of two different MnII di-
amine diketonate adducts, of interest as molecular precursors
for the vapor deposition of Mn oxide nanomaterials. The target
molecular systems, Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA, de-
signed as alternatives to well-known manganese b-diketonates,
differed by the presence of one ꢀCF3 group in the ligand
chain. The two compounds, developed by a simpler route than
that previously reported for Mn(hfa)2·TMEDA, are monomeric
and water-free, thanks to the complete saturation of the MnII
coordination environment. In particular, the presence of fluo-
rine in the diketonate moieties played a key role in the stabili-
zation of the complexes and in the obtainment of physico-
chemical properties (thermal behavior and gas-phase reactivi-
ty) favorable for CVD/ALD applications.
Figure 8. a) Plane-view and b) cross-sectional FE-SEM images of a Mn3O4
specimen deposited on Si(100) at 4008C from Mn(hfa)2·TMEDA. c) Corre-
sponding energy dispersive X-ray spectroscopy (EDXS) results.
6.50 keV, as well as the OKa signal at 0.52 keV. No evidence of C
or F presence could be obtained, in agreement with the clean
precursor decomposition discussed above. Irrespective of the
analyzed region, in-plane EDXS analyses highlighted a homo-
geneous oxygen and manganese lateral distribution.
Efforts were also devoted to the characterization of systems
supported on silica. In this regard, the surface morphology was
investigated by atomic force microscopy (AFM, Figure 9a and
b), which showed the presence of well interconnected protrud-
ing nanograins. The deposit appeared homogeneous and free
The present results highlight that variations in the fluorine
content of b-diketonate ligands do not affect appreciably the
stability of these precursors to air and moisture. Differences in
the behavior of the two compounds, as highlighted by ESI-MS
fragmentation patterns, could mainly be related to different
charge distributions in their molecular structures, depending
on the nature of the b-diketonate ligand. Both precursors ex-
hibit a higher volatility than that of conventional Mn b-diketo-
nates, paving the way to their successful application in the
vapor phase deposition of Mn oxides. Preliminary CVD experi-
ments enabled the preparation of high purity, single-phase
Mn3O4 nanomaterials endowed with tailored morphology, as
well as an appreciable visible light absorption. These results
suggest that the developed nanosystems are suitable candi-
dates for possible technological end-uses in solar driven pro-
cesses, ranging from photoactivated H2O splitting to wastewa-
ter purification, as well as in the development of solid state
gas sensing devices for the detection of toxic/flammable ana-
lytes (such as CO and CH4). Additional attractive perspectives
for this work will involve the extensive use of both molecular
compounds in CVD/ALD processes, to explore in detail the in-
terplay between processing parameters and the resulting ma-
terial properties. Preliminary studies in these research areas are
currently being carried out within our group.
Figure 9. a) Representative AFM image; b) height profile along the marked
line, and c) optical spectrum and derived Tauc plot for a Mn3O4 deposit ob-
tained on SiO2 at 4008C from Mn(hfa)2·TMEDA.
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
morphs as a function of the adopted preparation route, as ob-
served in the case of Cu(hfa)2·TMEDA.[13a] In addition, a key goal of
this work was a detailed investigation of similarities and differences
in the properties and behavior of MnL2·TMEDA as a function of the
ligand fluorination degree. Since the structure of Mn(tfa)2·TMEDA
has not been reported so far, an investigation of the molecular
structures of both compounds under the same experimental condi-
tions and with a similar refinement quality was performed to attain
a direct comparison of experimental data pertaining to the two
molecular systems. Furthermore, the simulation of compound
properties, involving the electronic excitation analysis, the geome-
tries of ions arising from their fragmentation, as well as the deter-
mination of their spin states, requires as a first step the optimiza-
tion of the compound geometry, which, in turn, is based on the
availability of structural data with a similar quality to compare and
validate the results obtained by computational experiments.
Experimental Section
General
MnCl2·4H2O (98+%), Hhfa (98%), and Htfa (98%) were purchased
from Strem Chemicalsꢂ and TMEDA (ꢃ98%) from Merckꢂ; all were
used without further purification. All manipulations were carried
out under normal laboratory conditions. Melting points (m.p.) were
measured in air on a FALC melting point device at atmospheric
pressure. Elemental analyses were carried out by using a Fisons
Carlo Erba EA1108 apparatus (CHNS version).
Synthesis of Mn(hfa)2·TMEDA
The synthesis of the target adduct was performed following a dif-
ferent procedure from that previously reported.[11,17] Hhfa (3.4 mL,
23.30 mmol; d=1.47 gmLꢀ1) was slowly added to a stirred aque-
ous solution of MnCl2·4H2O [2.37 g, 11.73 mmol, in deionized H2O
(50 mL)]. The subsequent dropwise addition of NaOH [0.93 g,
23.50 mmol, in deionized H2O (10 mL)] yielded a clear yellow solu-
tion. TMEDA (1.9 mL, 12.59 mmol; d=0.78 gmLꢀ1) was then slowly
added to the reaction mixture, which turned to a maroon-like
color. After reacting for 150 min in the dark, the obtained product
was repeatedly extracted in dichloromethane until a completely
colorless aqueous phase was obtained. The organic solution was
thoroughly washed with deionized water and the solvent
was removed at room temperature under reduced pressure
(ꢁ10ꢀ3 mbar), ultimately affording a yellow–orange solid (yield:
5.15 g, 75%). M.p. 868C at 1 atm; elemental analysis calcd. (%) for
C16H18O4N2F12Mn (Mw =585.25): C 32.84, H 3.10, N 4.79; found: C
33.60, H 2.90, N 4.78.
XRD data for the synthesized compounds were collected on an
Agilent Technologies SuperNova diffractometer, with an Atlas CCD
detector, by using CuKa radiation (l=1.54184 ꢁ) from multilayer X-
ray optics. The crystals were coated with a perfluoropolyether,
picked up with a glass fiber, and mounted in the nitrogen cold gas
stream of the diffractometer. The obtained data were processed
with CrysAlisPro.[34] An absorption correction based on multiple-
scanned reflections was carried out with ABSPACK in CrysAlisPro.
The crystal structure was solved by direct methods by using
SHELXS-97 and refined with SHELXL-2013.[35] For Mn(hfa)2·TMEDA,
two of the CF3 groups showed rotational disorder. Disordered parts
were modeled with appropriate restraints and constraints on ge-
ometry and atomic displacement parameters (ADPs). Anisotropic
ADPs were introduced for all non-hydrogen atoms. Hydrogen
atoms were placed in geometrically calculated positions and re-
fined with the appropriate riding model.
Synthesis of Mn(tfa)2·TMEDA
CCDC 1585867 (Mn(hfa)2·TMEDA) and 1585868 (Mn(tfa)2·TMEDA)
contain the supplementary crystallographic data for this paper.
These data are provided free of charge by The Cambridge Crystal-
lographic Data Centre
Htfa (2.9 mL, 23.30 mmol; d=1.27 gmLꢀ1) was slowly added to an
aqueous solution of MnCl2·4H2O [2.37 g, 11.73 mmol, in deionized
H2O (50 mL)], maintained under vigorous stirring, resulting in
phase separation. Subsequently,
23.50 mmol, in deionized H2O (10 mL)] was added dropwise, result-
ing in the formation of yellow solution. TMEDA (1.9 mL,
a solution of NaOH [0.93 g,
Analysis techniques
a
12.59 mmol; d=0.78 gmLꢀ1) was then added to the above mix-
ture, which became maroon-like. After reaction in the dark for
150 min, the obtained product was repeatedly extracted in di-
chloromethane until the aqueous phase turned colorless. The or-
ganic solution was washed with deionized water and the solvent
was removed at room temperature (ꢁ10ꢀ3 mbar), yielding a light
yellow solid (yield: 3.7 g, 66%). M.p. 998C at 1 atm; elemental anal-
ysis calcd. (%) for C16H24O4N2F6Mn (Mw =477.31): C 40.26, H 5.07, N
5.87; found: C 40.93, H 5.10, N 6.03.
Optical spectroscopy analyses were carried out by using a Cary 50
spectrophotometer (Varian; spectral bandwidth=1 nm). Measure-
ments were carried out on 10ꢀ6 m solutions of Mn(hfa)2·TMEDA and
Mn(tfa)2·TMEDA in ethanol, using quartz cuvettes (optical path=
0.5 cm).
ESI-MS characterization was carried out using a LCQ Fleet ion trap
instrument (ThermoFisher), operating in both positive- and nega-
tive-ion modes. The used entrance capillary temperature and volt-
age were set at 2508C and 4 kV, respectively. The 10ꢀ6 m solutions
of the target Mn compounds in methanol were introduced by
direct infusion with a syringe pump (flow rate=8 mLminꢀ1). MSn
experiments were performed by applying a supplementary radio
frequency (RF) voltage to the ion-trap end caps (5 V peak-to-peak).
Both Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA were stored at room
temperature and could be easily handled in air without any detri-
mental degradation. The powders were soluble in various solvents,
such as hexane, dichloromethane, acetone, and alcohols. In both
cases, crystals for X-ray analysis were obtained by re-dissolution in
1,2-dichloroethane, followed by slow solvent evaporation.
TGA was performed with a TGA 2950 thermobalance manufactured
by TA Instruments. Measurements were conducted under a pre-
purified nitrogen atmosphere (heating rate=108Cminꢀ1) on sam-
ples that had a mass between 5 and 10 mg. DSC analyses were car-
X-ray crystallography
ried out by using
equipped with a liquid nitrogen cooling system, using a heating
rate of 38Cminꢀ1
a MDSC2920 apparatus (TA Instruments)
Though the structure of Mn(hfa)2·TMEDA has been previously re-
ported,[17] in this work crystallographic data were collected on both
MnL2·TMEDA compounds. In fact, since in the present work the
Mn(hfa)2·TMEDA compound was prepared through a synthesis pro-
cedure different from that previously reported,[17] the first aim was
to verify the possible formation of different Mn(hfa)2·TMEDA poly-
.
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
8
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Simulation
Acknowledgements
DFT calculations on Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA were per-
formed with the PBE functional[36] augmented with the long-range
corrections of Hirao et al.[37] The Gaussian 09 program was adopt-
ed,[38] with Stuttgart–Dresden ECP pseudopotential for Mn and
Stuttgart–Dresden basis set for all atoms.[39] This basis set was en-
hanced with diffuse and polarization functions from the (D95+ +
(d,p)) basis set,[40] which provided a satisfactory description of
other members of the M(hfa)2·TMEDA series.[13a,14a,15a,b,16] All calcu-
lated minima had positive frequencies and were in the high-spin
state (sextet). The spin state was established by optimizing the ge-
ometry of the compounds in the sextet, quartet, and doublet
states. Electronic excitations were calculated on the minimum-
energy structures by time-dependent (TD) DFT. The 50 excitations
at lower energy were considered. The spectra reported in Fig-
ure 2b were obtained by smoothing the TD-DFT excitations with a
2 nm Gaussian broadening. TD-DFT excitations were calculated for
the two complexes also in ethanol, using a polarizable continuum
model for the solvent.[31] Natural bond orbital (NBO) wave function
analyses were performed with NBO 5.0.[41] The compound decom-
position energies (DE) with respect to the fragments take into ac-
count the zero-point-energy contributions, and basis-set superposi-
tion errors were counterpoise-corrected. In addition to those in
vacuum, DE values were also calculated in methanol with a polariz-
able continuum model.[31]
We kindly acknowledge financial support under Padova Univer-
sity ex-60% 2014–2017, Insubria University FAR 2015–2016, P-
DiSC #SENSATIONAL BIRD2016-UNIPD projects and ACTION
post-doc fellowship. Thanks are also due to Dr. Loris Calore
and Filippo Gri (Padova University), as well as Dr. Manuela
Winter (Ruhr-University Bochum) for elemental microanalyses
and useful experimental assistance.
Conflict of interest
The authors declare no conflict of interest.
Keywords: chemical vapor deposition · manganese oxides ·
Mn b-diketonates
· nanomaterials · single-crystal X-ray
diffraction
[1] a) Z. Chen, Z. Jiao, D. Pan, Z. Li, M. Wu, C.-H. Shek, C. M. L. Wu, J. K. L.
Lai, Chem. Rev. 2012, 112, 3833–3855; b) Z. Bai, B. Sun, N. Fan, Z. Ju, M.
Li, L. Xu, Y. Qian, Chem. Eur. J. 2012, 18, 5319–5324; c) C. E. Frey, P. Kurz,
Chem. Eur. J. 2015, 21, 14958–14968; d) L. Ben Said, A. Inoubli, B. Bouri-
cha, M. Amlouk, Spectrochim. Acta Part A 2017, 171, 487–498; e) O.
Nilsen, H. Fjellvꢄg, A. Kjekshus, Thin Solid Films 2003, 444, 44–51.
[2] a) Z.-Y. Fei, B. Sun, L. Zhao, W.-J. Ji, C.-T. Au, Chem. Eur. J. 2013, 19,
6480–6487; b) K. Wang, X. Ma, Z. Zhang, M. Zheng, Z. Geng, Z. Wang,
Chem. Eur. J. 2013, 19, 7084–7089; c) D. M. Robinson, Y. B. Go, M. Mui,
G. Gardner, Z. J. Zhang, D. Mastrogiovanni, E. Garfunkel, J. Li, M. Green-
blatt, G. C. Dismukes, J. Am. Chem. Soc. 2013, 135, 3494–3501.
[3] M.-K. Song, S. Cheng, H. Chen, W. Qin, K.-W. Nam, S. Xu, X.-Q. Yang, A.
Bongiorno, J. Lee, J. Bai, T. A. Tyson, J. Cho, M. Liu, Nano Lett. 2012, 12,
3483–3490.
[4] D. P. Dubal, D. S. Dhawale, R. R. Salunkhe, V. J. Fulari, C. D. Lokhande, J.
Alloys Compd. 2010, 497, 166–170.
[5] a) D. P. Dubal, D. S. Dhawale, R. R. Salunkhe, S. M. Pawar, C. D. Lokhande,
Appl. Surf. Sci. 2010, 256, 4411–4416; b) H. Y. Xu, S. L. Xu, X. D. Li, H.
Wang, H. Yan, Appl. Surf. Sci. 2006, 252, 4091–4096.
[6] L. Zhang, Q. Zhou, Z. Liu, X. Hou, Y. Li, Y. Lv, Chem. Mater. 2009, 21,
5066–5071.
[7] H. W. Kim, Y. J. Kwon, H. G. Na, H. Y. Cho, C. Lee, J. H. Jung, Microelec-
tron. Eng. 2015, 139, 60–69.
[8] a) H. Van Bui, F. Grillo, J. R. van Ommen, Chem. Commun. 2017, 53, 45–
71; b) S. M. George, Chem. Rev. 2010, 110, 111–131; c) H. Pedersen, S. D.
Elliott, Theor. Chem. Acc. 2014, 133, 1476; d) J. S. Price, P. Chadha,
D. J. H. Emslie, Organometallics 2016, 35, 168–180.
[9] a) A. Devi, Coord. Chem. Rev. 2013, 257, 3332–3384; b) P. Marchand,
C. J. Carmalt, Coord. Chem. Rev. 2013, 257, 3202–3221; c) L. McElwee-
White, Dalton Trans. 2006, 5327–5333; d) C. E. Knapp, C. J. Carmalt,
Chem. Soc. Rev. 2016, 45, 1036–1064; e) S. Mishra, S. Daniele, Chem.
Rev. 2015, 115, 8379–8448.
[10] a) T. Nakamura, Phys. Status Solidi C 2015, 12, 958–963; b) T. Nakamura,
R. Tai, T. Nishimura, K. Tachibana, J. Electrochem. Soc. 2005, 152, C584–
C587; c) S. Dhar, A. Varade, S. A. Shivashankar, Bull. Mater. Sci. 2011, 34,
11–18; d) S. I. Troyanov, O. Y. Gorbenko, A. A. Bosak, Polyhedron 1999,
18, 3505–3509; e) S. Kang, G. Brewer, K. R. Sapkota, I. L. Pegg, J. Philip,
IEEE Trans. Nanotechnol. 2012, 11, 437–440.
[11] Z. Lipani, M. R. Catalano, P. Rossi, P. Paoli, G. Malandrino, Chem. Vap.
Deposition 2013, 19, 22–28.
[12] A. R. Merritt, R. Rajagopalan, J. D. Carter, Thin Solid Films 2014, 556, 28–
34.
[13] a) G. Bandoli, D. Barreca, A. Gasparotto, R. Seraglia, E. Tondello, A. Devi,
R. A. Fischer, M. Winter, E. Fois, A. Gamba, G. Tabacchi, Phys. Chem.
Chem. Phys. 2009, 11, 5998–6007; b) G. Bandoli, D. Barreca, A. Gasparot-
to, C. Maccato, R. Seraglia, E. Tondello, A. Devi, R. A. Fischer, M. Winter,
Inorg. Chem. 2009, 48, 82–89.
CVD synthesis and characterization of Mn3O4 nanomaterials
Manganese oxide depositions were performed by means of a
custom-built cold-wall CVD reactor,[15a] using Mn(hfa)2·TMEDA or
Mn(tfa)2·TMEDA precursors contained in an external glass reservoir.
In this study, the precursor vaporization temperatures were kept at
60 and 658C for Mn(hfa)2·TMEDA and Mn(tfa)2·TMEDA, respectively,
while the substrate temperature was 4008C. Gas lines and valves
connecting the precursor vessel and the reactor were maintained
at Tꢃ1008C for each growth process to prevent precursor conden-
sation. Depositions were carried out in O2-based atmosphere for
1 h on 1ꢃ1 cm2 Si(100) (MEMCꢂ, Merano, Italy) and Herasil silica
(Heraeusꢂ) substrates, which were subjected to suitable pre-clean-
ing procedures before CVD experiments. For silicon substrates, the
native SiOx layer was removed prior to deposition by means of HF
etching. O2 [total flow rate=200 standard cubic centimeters per
minute (sccm)] was used as the carrier and reaction gas. Mass flow
rates were controlled by MKS flow meters (Andover, USA). The
total pressure, measured using a capacitance manometer (BOC Ed-
wards, Crawley, UK), was set at 10.0 mbar.
XRD patterns were recorded in glancing-incidence mode (qi =18)
on a Bruker D8 Advance X-ray diffractometer, equipped with a
CuKa X-ray source (40 kV, 40 mA) and a Gçbel mirror. Crystallite di-
mensions were estimated using the Scherrer equation.
FE-SEM analyses were performed by a Zeiss SUPRA 40 VP instru-
ment, equipped with an Oxford INCA x-sight X-ray detector for
EDXS investigation (primary beam voltage=20 kV).
Optical absorption spectra for samples deposited on silica sub-
strates were collected in transmission mode at normal incidence
by means of a Cary 50 spectrophotometer, subtracting the sub-
strate contribution. Tauc plots based on the obtained data were
used to determine the optical band gap.
AFM measurements were performed using a NT-MDT SPM solver
P47H-PRO apparatus, operating in tapping mode. RMS roughness
values were obtained from the analysis of 2ꢃ2 mm2 images after
plane fitting.
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
9
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[14] a) D. Barreca, E. Fois, A. Gasparotto, R. Seraglia, E. Tondello, G. Tabacchi,
Chem. Eur. J. 2011, 17, 10864–10870; b) E. Fois, G. Tabacchi, D. Barreca,
A. Gasparotto, E. Tondello, Angew. Chem. Int. Ed. 2010, 49, 1944–1948;
Angew. Chem. 2010, 122, 1988–1992.
[15] a) D. Barreca, G. Carraro, A. Devi, E. Fois, A. Gasparotto, R. Seraglia, C.
Maccato, C. Sada, G. Tabacchi, E. Tondello, A. Venzo, M. Winter, Dalton
Trans. 2012, 41, 149–155; b) D. Barreca, G. Carraro, A. Gasparotto, C.
Maccato, R. Seraglia, G. Tabacchi, Inorg. Chim. Acta 2012, 380, 161–166;
c) G. Tabacchi, E. Fois, D. Barreca, G. Carraro, A. Gasparotto, C. Maccato
in Advanced Processing and Manufacturing Technologies for Nanostruc-
tured and Multifunctional Materials II, Wiley, Hoboken, 2015, pp. 83–90.
[16] G. Tabacchi, E. Fois, D. Barreca, A. Gasparotto, Phys. Status Solidi A 2014,
211, 251–259.
[17] G. Malandrino, R. G. Toro, M. R. Catalano, M. E. Fragalꢅ, P. Rossi, P. Paoli,
Eur. J. Inorg. Chem. 2012, 1021–1024.
[18] F. A. Cotton, R. H. Holm, J. Am. Chem. Soc. 1960, 82, 2979–2983.
[19] T. Yamada, T. Takai, O. Rhode, T. Mukaiyama, Bull. Chem. Soc. Jpn. 1991,
64, 2109–2117.
[20] A. Lꢆpez-Periago, O. Vallcorba, C. Frontera, C. Domingo, J. A. Ayllon,
Dalton Trans. 2015, 44, 7548–7553.
[21] N. A. Bailey, D. E. Fenton, M. V. Franklin, M. Hall, J. Chem. Soc. Dalton
Trans. 1980, 984–990.
[22] W. B. S. McCormack, UK Patent GB 2022088A, 1979.
[23] G. Tabacchi, E. Fois, D. Barreca, A. Gasparotto, Int. J. Quantum Chem.
2014, 114, 1–7.
[24] B. S. Lim, A. Rahtu, J.-S. Park, R. G. Gordon, Inorg. Chem. 2003, 42, 7951–
7958.
[25] L.-Y. Wang, X.-Q. Wang, K. Jiang, J.-L. Chang, Y.-F. Wang, J. Mol. Struct.
2007, 840, 14–21.
[26] F. H. Allen, Acta Crystallogr. Sect. B 2002, 58, 380–388.
[27] H. Wang, Z. Liu, C. Liu, D. Zhang, Z. Lꢇ, H. Geng, Z. Shuai, D. Zhu, Inorg.
Chem. 2004, 43, 4091–4098.
[32] Pattern No. 024-0734, JCPDS (2000).
[33] B. D. Viezbicke, S. Patel, B. E. Davis, D. P. Birnie, Phys. Status Solidi B
2015, 252, 1700–1710.
[34] CrysAlisPro Software System, version 1.171.36.20; Agilent Technologies
UK Ltd.: Oxford, U. K., 2012.
[35] G. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[36] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865–
3868.
[37] a) H. Iikura, T. Tsuneda, T. Yanai, K. Hirao, J. Chem. Phys. 2001, 115,
3540–3544; b) A. W. Lange, M. A. Rohrdanz, J. M. Herbert, J. Phys. Chem.
B 2008, 112, 6304–6308; c) J.-W. Song, T. Hirosawa, T. Tsuneda, K. Hirao,
J. Chem. Phys. 2007, 126, 154105.
[38] Gaussian 09, Revision D.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, ꢈ. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc., Wallingford, CT, 2009.
[39] A. Bergner, M. Dolg, W. Kꢇchle, H. Stoll, H. Preuß, Mol. Phys. 1993, 80,
1431–1441.
[40] T. H. Dunning, Jr., P. J. Hay in Modern Theoretical Chemistry, Vol. 2,
Chap. 1, Plenum, New York, 1976.
[41] a) L. Goodman, R. R. Sauers, J. Comput. Chem. 2007, 28, 269–275;
b) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–926;
c) E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A.
Bohmann, C. M. Morales, F. Weinhold in NBO 5.0, Theoretical Chemistry
Institute, University of Wisconsin, Madison, USA, 2001.
[28] D. Luneau, C. Stroh, J. Cano, R. Ziessel, Inorg. Chem. 2005, 44, 633–637.
[29] J. Ni, H. Yan, A. Wang, Y. Yang, C. L. Stern, A. W. Metz, S. Jin, L. Wang,
T. J. Marks, J. R. Ireland, C. R. Kannewurf, J. Am. Chem. Soc. 2005, 127,
5613–5624.
[30] L. Wang, Y. Yang, J. Ni, C. L. Stern, T. J. Marks, Chem. Mater. 2005, 17,
5697–5704.
Manuscript received: July 24, 2017
[31] M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys. Lett. 1998, 286,
253–260.
Version of record online: && &&, 0000
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
10
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
Producing perfect precursors: A con-
venient route for the synthesis of MnII
diamine diketonate complexes is pro-
posed. For the first time, the target ad-
ducts are characterized in detail by a
combined experimental-theoretical ap-
proach. The compounds possess a mon-
omeric structure, very favorable mass-
transport properties, and a clean frag-
mentation, all of which make them
promising precursors for the vapor
phase fabrication of manganese oxide
nanosystems (see figure).
&
Chemical Vapor Deposition
C. Maccato, L. Bigiani, G. Carraro,
A. Gasparotto, R. Seraglia, J. Kim, A. Devi,
G. Tabacchi,* E. Fois, G. Pace, V. Di Noto,
D. Barreca*
&& – &&
Molecular Engineering of MnII Diamine
Diketonate Precursors for the Vapor
Deposition of Manganese Oxide
Nanostructures
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
11
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ