Insight into the mechanism of 3,3′-dimethyl-4,4′-diaminodiphenylmethane synthesis over Hβ zeolite

Article abstract of DOI:10.1016/j.molcata.2014.10.028

Heterogeneous synthesis of 3,3′-dimethyl-4,4′-diaminodiphenylmethane (MDT) from o-tolylamine and formaldehyde on acidic zeolites HY, HZSM-5, Hβ and TS-1 was investigated. Hβ zeolite with strong Bronsted acidity and proper textural property demonstrated be

Full text of DOI:10.1016/j.molcata.2014.10.028

Accepted Manuscript
Title: Insight into the mechanism of
3,3ꢀ-dimethyl-4,4ꢀ-diaminodiphenylmethane synthesis over
H␤ zeolite
Author: Yingkuo Yang Yingxian Zhao Jianhua Shao
Shengjian Zhang Rongqiang Yang
PII:
S1381-1169(14)00483-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2014.10.028
MOLCAA 9330
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
20-9-2014
26-10-2014
29-10-2014
Please cite this article as: Y. Yang, Y. Zhao, J. Shao, S. Zhang, R. Yang,
Insight into the mechanism of 3,3ꢀ-dimethyl-4,4ꢀ-diaminodiphenylmethane synthesis
over Hrmbeta zeolite, Journal of Molecular Catalysis A: Chemical (2014),
http://dx.doi.org/10.1016/j.molcata.2014.10.028
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. 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.

H
H
C O
CH
3
CH
CH
3
3
H
Formaldehyde
N
CH
Hβ
2
NH
N
CH
+ H2O
2
2
OH
o-Tolylamino-methanol
Methylene-o-tolyl-amine
o-Tolylamine
CH3
Hβ
NH2
H3C
CH3
H2
C
Hβ
H C
3
N
N
H
NH
2
H
H
2
H C
3
C
N
H
N,N'-Di-o-tolyl-methanediamine
Hβ
Hβ
2-Methyl-6-(o-tolylamino-methyl)-phenylamine
Hβ
H C
3
H C
3
H
2
C
H
H C
3
H N
2
2
C
N
H
N
H
2
C
H
CH
3
H C
3
N
H N
2
H
2-Methyl-5-(o-tolylamino-methyl)-phenylamine
2-Methyl-3-(o-tolylamino-methyl)-phenylamine
H N
2
CH
3
2-Methyl-4-(o-tolylamino-methyl)-phenylamine
Hβ
CH3
NH2
H2
C
H3C
H2N
3,3'-Dimethyl-4,4'-diaminodiphenylmethane (MDT)
Overview of reaction sequence for MDT synthesis over Hβ zeolite
Page 1 of 39

Highlights
ꢀ
ꢀ
ꢀ
ꢀ
A novel method for synthesis of MDT catalyzed by acidic zeolites replacing mineral acids
Hβ with strong Bronsted acidity and proper textural property showed good catalytic performance
Synthetic process went through multiple steps involving several reaction intermediates
The detailed interpretation to the mechanism of various intermediates reaction was presented.
Page 2 of 39

Insight into the mechanism of 3,3’-dimethyl-4,4’-diaminodiphenylmethane
synthesis over Hβ zeolite
Yingkuo Yanga, Yingxian Zhaoa,*, Jianhua Shaob, Shengjian Zhanga, Rongqiang Yangb
a
Ningbo Institute of Technology, Zhejiang University, Ningbo 315100, Zhejiang, China
b
Jiangsu Qingquan Chemical Co. Ltd., Yancheng 224555 , Jiangsu, China
ABSTRACT
Heterogeneous synthesis of 3,3’-dimethyl-4,4’-diaminodiphenylmethane (MDT) from o-tolylamine
and formaldehyde on acidic zeolites HY, HZSM-5, Hβ and TS-1 was investigated. Hβ zeolite with strong
Bronsted acidity and proper textural property demonstrated better catalytic performance than other three
zeolites for this synthetic reaction. Yield and selectivity of product MDT from the catalytic reaction over
Hβ strongly depended on temperature, and optimal reaction temperature appeared to be about 433 K. Based
on analyses of LC/MS and 1H NMR for reaction intermediates, it was suggested that the synthetic process
went through multiple steps involving intermediates o-tolylamino-methanol, methylene-o-tolylamine,
N,N’-di-o-tolyl-methanediamine and 2-methyl-4-(o-tolylamino-methyl)-phenylamine. Interpretation to the
mechanism of various intermediates formation and conversion was presented.
Keywords: MDT; Catalytic synthesis; Hβ-zeolite; Intermediates; Mechanism
1. Introduction
The condensation reaction of aromatic amines with aldehydes is a useful vehicle for production of
diaminodiphenylmethanes and derivatives, which find a variety of applications in plastic and paint industry.
As well known, 4,4’-diaminodiphenyl methane (MDA) is an important starting materials for production of
polyurethanes and thermoplastics [1,2], 3,3’-dichloro-4,4’-diaminophenyl methane (MOCA) is used as
curing agent of polyurethane rubber and solidifying agent of epoxide resin [3], and
3,3’-dimethyl-4,4’-diaminodiphenylmethane (MDT) is an important curing agent and polymer additives
[4-6] in the preparation of electric cables, wires and electromagnetic strings with good heat-resisting,
chemical-resisting and electric-insulating property [7,8]. For example, isocyanate prepared from MDT can
be used as room temperature curing agent with good adhesion property [9].
*
Corresponding author. Tel: +86 57488229075; Email address: zyx@nit.zju.edu.cn (Y.X. Zhao)
Page 3 of 39

Traditionally, MDT has been prepared by the condensation reaction of o-tolylamine with
formaldehyde in the presence of a liquid protonic acid such as HCl. The synthetic process goes through a
complex mechanism involving salinization, condensation, transposition rearrangement and neutralization
as showed in Scheme 1 [10]. Since MDT is yielded in the hydrochloride form, it has to be neutralized with
NaOH to obtain the final product. Obviously, the use of HCl and NaOH poses several problems such as
corrosion, high toxicity and the disposal of a large amount of aqueous waste containing NaCl contaminated
with traces of aromatic amines, which have to be removed before discharging to pollute the residual water.
Although a completely green solution to the existing problems might involve the replacement of this
chemistry with an inherently less toxic combination of reagents, the replacement of current homogeneous
catalysts with heterogeneous catalysts might eliminate the pollution of a waste aqueous phase [11].
Therefore, there is great academic interest and practical importance to develop easily separable, highly
active and reusable solid acids to replace mineral acids for the synthetic process of
diaminodiphenylmethanes [1,12]. When the uses of heterogeneous catalysts for MDT synthesis is scarcely
disclosed in the literature, several solid acids including ion-exchanged resins [13], metallic compounds
[14-17], clays [11,18], and zeolites [1,19-22] have been tested for the synthesis of
4,4’-diaminodiphenylmethane (MDA) by condensation of anilines and formaldehyde. According to the
research results published in literature so far, zeolites are among the most promising candidates for choice
of such solid catalysts due to their strong Brönsted acidity, high hydrothermal stability, tailorable acid site
density and porosity.
In our recent studies, the catalytic activities of several zeolites in the synthesis of
3,3’-dimethyl-4,4’-diaminodiphenylmethane (MDT) from o-tolylamine and formaldehyde have been
scanned, and Hβ zeolite exhibited good performance to catalyze this synthetic reaction [23]. Therefore,
there are further interests to determine the mechanism of MDT synthesis over Hβ zeolite and optimize
reaction conditions. In this paper, the effects of catalyst property and reaction conditions on MDT synthesis
over Hβ zeolite are deeply discussed, and the detailed interpretation to reaction mechanism of various
intermediates are presented, which will establish the foundation of developing and applying a green
Page 4 of 39

technology for the synthesis of MDT and similar compounds.
2. Experimental
2.1. Materials
Reactants o-tolylamine and formaldehyde (37 wt% aqueous solution) were purchased from Sinopharm
Chemical Reagent Corporation Ltd., China. Chemicals methanol and methyl cyanide were purchased from
Tedia Corporation Ltd, United States, and potassium dihydrogen phosphate from Sinopharm Chemical
Reagent Corporation Ltd., China. Zeolites Hβ was supported by Tianjin NCIC catalyst Corporation Ltd.,
HY and HZSM-5 were supported by Shandong Qilu Huaxin Gaoke Corporation Ltd., China.
2.2. Reaction procedures
In a typical run, a certain amount of formaldehyde, o-tolylamine purified by distillation, and zeolite
catalyst activated at 823 K for 3 h were added into a 500 ml autoclave made in Weihai Zhengwei
Machinery Equipment Corporation Ltd, China. The sealed reactor was purged with nitrogen and then
heated up to the designed temperature with vigorous stirring. After the reaction went through the scheduled
times, the product mixture was collected and sampled for HPLC analysis.
2.3. Sample analysis
The textural properties of zeolites were characterized by BET method using an ASAP2020M analyzer
of Micromeritics Co., USA, and the acidic properties was measured by NH3 temperature programmed
desorption(TPD) analysis. The NH3-TPD experiments were conducted in a flow apparatus (PX200
analyzer of Tianjin Golden Eagle Technology Lit. Co., China) with helium as carrier gas, and a thermal
conductivity detector (TCD) was used for evolved gas detection. A small amount of the granulated zeolite
(100 mg) was heated at 823 K in an He flow (30 mL min-1) for 1h, and then cooled to 373 K. The activated
sample was saturated with ammonia at 373 K for 1 h, and after being purged with pure He for 1.5 h at the
same temperature, was heated at the rate of 10 K min-1 up to 973 K under helium flow (20 mL min-1).
The samples of reactants and products were analyzed by EX1600SM High Pressure Liquid
Chromatography (HPLC) with the conditions: detection wavelength of UV254 nm, C18 column, room
temperature, methanol solvent, mobile phase of acetonitrile / water (1:1 volume ratio) and potassium
dihydrogen phosphate (0.5%) at flow rate of 1.00 mL min-1. The content of each component was
Page 5 of 39

determined by corresponding calibration factor.
1
Reaction intermediates were identified by Agilent 6520 Q-TOF LC/MS, and H NMR spectra were
generated from Brucker Avance 400 NMR using DMSO-d6 as solvent and TMS as internal standard.
3. Results and discussion
3.1. Effects of zeolite acidity and textural properties on MDT synthesis
For a heterogeneous reaction over molecular sieve catalysts, the reaction activity and selectivity
depended on the acidity and textural property of zeolites. In general, the acidity of zeolite refers to acidic
type, strength and density. For the synthesis of MDT from o-tolylamine and formaldehyde in the presence
of zeolites, there was the prime interest to find out if either BrÖnsted acid or Lewis acid or both would be
active in reaction. For this purpose, three comparable runs of MDT synthesis were first carried out in the
absence of a catalyst and in the presence of TS-1 or Hβ zeolite at 413 K, respectively, and the HPLC
chromatograms of product samples were presented in Fig. 1. One can see that besides the large peak of
reactant o-tolylamine, no peak of final product MDT and a very small peak appeared at retention time of
10.5 min in the absence of catalyst, and a negligible peak of MDT and a visible overlapping peak at 10 min
appeared in the presence of TS-1 zeolite. Meanwhile, the HPLC picture of the sample from the reaction in
the presence of Hβ zeolite showed a large peak of product MDT, two small peaks of intermediates M1 and
M2 at 6 and 7 min, and a very large peak of intermediate M3 at 9 min as well as a very small peak of
by-product P at 12.3 min. As well known, Hβ as a aluminum-silicon zeolite contains both BrÖnsted acid
sites (proton-donors) and Lewis acid sites (electron-acceptors), while TS-1 as a titanium-silicon zeolite
contains Lewis acid sites alone [24,25]. Therefore, we conclude that BrÖnsted acid sites may play the
principal role in catalyzing formation of the intermediates and consequent product MDT.
There were more interests to investigate how the acid strength and density of zeolite affect the MDT
synthesis. Therefore, three kinds of aluminum-silicon zeolites Hβ, HY and HZSM-5 as well as a
titanium-silicon zeolite TS-1 were comparatively tested for MDT synthesis from o-tolylamine and
formaldehyde at 433K. As showed in Fig. 2, TS-1 has very poor catalytic performance regarding HCHO
Page 6 of 39

conversion and MDT production. For the three aluminum-silicon zeolite catalysts, HCHO conversion
decreased from 80.7 to 75.2 to 73.6%, MDT yield from 53.3 to 42.9 to 30.7%, and MDT selectivity from
66.1 to 57.0 to 41.7% with shifting catalyst from Hβ to HY to HZSM-5. By comparison, Hβ demonstrated
the best performance, while HZSM-5 gave the disappointing performance to this reaction system.
On the other hand, the NH3 temperature programmed desorption (TPD) analysis showed the
difference of acid site density and strength distribution among the four zeolites. As illustrated in Fig. 3, one
peak of NH3 desorption was detected in 375-550 K for each zeolite, and second peak detected in 700-850
K for HZSM-5 and Hβ which indicated the existence of stronger acid sites. As presented in Table 1, the
amount of NH3 absorbed on HZSM-5, Hβ, HY and TS-1 decreased from 643 to 517 to 381 to 233μmol/g,
indicating HZSM-5 possessed the highest acid density while TS-1 possessed the lowest acid density.
Imaginably, the catalytic activity of zeolites should follow the order of HZSM-5 > Hβ > HY if the acid
strength would have played a dominant role, and also the HZSM-5 should give the better performance than
Hβ if the amount of acid site would have played a critical role in MDT synthesis. The data in Fig. 2 ,
however, declared the best catalytic performance for Hβ and the better performance of HY than HZSM-5,
suggesting that the pore structure and surface area of zeolites might have significant effects on this reaction.
As well known, Hβ, HY and HZSM-5 are the protonated (H+)form of parent zeolites β, ZSM-5 and Y
respectively. The high-silica zeolite β has an unique crystal structure with 12-membered ring channels (6.6
× 6.7 Å)crossed by a second type of 12-membered ring channels (5.6 × 5.6 Å) [26], and such
interconnected pore (channel) system can promote reactants to easily pass from one pore to another. As one
of the high-silica zeolites with MFI-type structure, ZSM-5 has also a two-dimensional pore structure with
straight 10-membered ring channels (5.1 × 5.5 Å) intersected by a second type of 10-membered ring
channels (5.3 × 5.6 Å) [27]. The low-silica zeolite Y is a member of the structural group named after the
mineral Faujasite [28], which has a framework structure with the 12-membered ring pores to be arranged
perpendicular to each other. Based on the data in Table 1, HZSM-5 with smaller pore opening (0.56 nm)
and volume (0.19 cm3/g) might significantly hinder the diffusion of product MDT and rearrangement of
reaction intermediates, while HY with larger pore diameter (0.74 nm) and volume (0.36 cm3/g) was less
active due to low acid density and weaker acid strength. By comparison, the Hβ with enough strong acidity,
Page 7 of 39

moderate pore size (0.66 nm), pore volume (0.28 cm3/g) and surface area (461.3 m2/g) exhibited the best
overall performance to catalytic synthesis of MDT.
Table 1
Surface area, pore diameter, pore volume and acidity of Hβ, HY, HZSM-5 and TS-1 zeolites
Surface area
(m2/g)
Pore diameter
(nm)
Pore volume
(cm3/g)
Acidity
( μmol/g)
Zeolites
Hβ
Si/Al
40
461.3
0.66
0.28
517
HY
4.8
684.0
0.74
0.36
381
HZSM5
TS-1
45
352.7
359.6
0.56
0.54
0.19
0.16
643
233
―
3.2. Identification of key reaction intermediates
There might be several active intermediates in MDT synthesis by condensation of o-tolylamine and
formaldehyde over Hβ zeolite. The identification of these intermediates is useful for us to understand the
reaction mechanism and optimize the reaction conditions. Some efforts have been spent for determination
of the three intermediates M1, M2, and M3 in Fig. 1(c). LC-MS analysis of the product sample was first
conducted and indicated that the three intermediates had the same molecular weight as final product MDT
(226g/mol). In other words, these intermediates and MDT were isomers with chemical formula C15H18N2.
Unfortunately, the chemical structures of three isomeric intermediates could not be convincingly
determined by LC-MS alone, due to lack of the information of standard reference compounds. Considering
that there might be some differences of H position and vibration among the isomers, the three intermediates
M1, M2 and M3 were carefully isolated each other by preparative chromatography, purified and then
1
1
analyzed by H nuclear magnetic resonance (NMR). Figure 4 illustrated the H NMR spectra of the final
product MDT (purity > 98%) and intermediates (purity > 95%) with the indication of H number and
vibration frequency. The patterns of H in Fig. 4(a) well match the structural features of H in MDT
molecule. By carefully interpreting the patterns of H in Fig. 4(b), we could confidently assign this 1H NMR
spectrum of the intermediate M3 to N,N’-di-o-tolyl-methanediamine. Based on the theoretical analysis of H
shift and the follow-up discussion for temperature effects on the reaction mechanism, we have also
deduced that the 1H NMR spectra of intermediates M1 and M2 in Fig. 4(c) and (d) belonged to
Page 8 of 39

2-methyl-6-(o-tolylamino-methyl)-phenylamine
respectively, although it was not entirely sure whether there was any interference of isomers
2methyl-5-(o-tolylamino-methyl)-phenylamine and 2methyl-3-(o-tolylamino-methyl)-phenylamine.
and
2-methyl-4-(o-tolylamino-methyl)-phenylamine
Nevertheless, these 1H NMR spectra gave us great help to understand the mechanism of MDT synthesis.
3.3. Influence of reaction temperature and time
In general, reaction temperature has some significant effects on kinetics, selectivity and mechanism of
a
heterogeneous catalytic reaction. Under the conditions of n(o-tolylamine):n(HCHO)=3/1,
w(Hβ):w(HCHO)=1:1 and 1h, the synthesis of MDT from o-tolylamine and formaldehyde over Hβ was
investigated in temperature range of 373 to 473 K. As shown in Fig. 5, HCHO conversion significantly
increased from 21.4 to 80.7% with increasing temperature from 393 to 433 K, and then stabilized in 433 –
473 K. The selectivity of final product MDT gently increased from 4.5 to 28.7 % as temperature increased
373 to 413 K, rapidly rose up to 66.1% from 413 to 433 K, and then slightly decreased as temperature
continually increased from 443 to 473 K, probably due to the generation of polymeric by-products in
higher
temperature.
On
the
contrary,
the
selectivity
of
key
intermediate
M3
(N,N’-di-o-tolyl-methanediamine) significantly declined from 86.1 to 16.9 % with increasing temperature
from 373 to 433 K, and then continually decreased to 2.4 % at 473 K. Clearly, there was a strong
relationship between MDT production and N,N’-di-o-tolyl-methanediamine consumption indicating that
the synthetic mechanism of MDT must involve formation and conversion of this intermediate. Also as
temperature increased,selectivity of intermediate M2 (2-methyl-4-(o-tolylamino-methyl)-phenylamine)
gradually increased to a maximum of 9.4 % at 393 K, and then decreased to 2.4 % over 453 K, while the
selectivity of intermediate M1 (2-methyl-6-(o-tolylamino-methyl)-phenylamine) slightly reached a small
maximum of 3.5% and then decreased to2.9% over 443 K. Besides, the selectivity of other by-products
such as P in Fig. 1(c) significantly increased from 9.9 to 26.3% with increasing temperature from 433 to
473 K. In view of both higner HCHO conversion and higher MDT selectivity required by practical
application, the optimal reaction temperature was about 433 K.
Page 9 of 39

On the other hand, the effect of reaction time on the yield and selectivity of MDT in the condensation
reaction of o-tolylamine and formaldehyde over Hβ at 433 K was illustrated in Fig. 6. The yield of MDT
slightly increased from 48.5 to 51.5 % with increasing reaction time from 0.5 to 1h, almost remained the
same level up to 4 h, and then decreased to 45.1% after 8 h, while the selectivity of MDT slightly increased
from 64.6 to 66.1 % with increasing time from 0.5 to 1 h, and gradually decreased to 62.4 % after 8 h. In
comparison with reaction temperature, the effect of reaction time during 0.5 to 8 h on this reaction system
was less significant.
3.4. Insight into the mechanism of MDT synthesis over Hβ zeolite
3.4.1. Overview of synthetic sequence
The reaction network and mechanism of MDT synthesis from o-tolylamine and formaldehyde over
solid acid catalysts have, to the best of our knowledge, not been fully interpreted so far. According to the
basic chemistry of amine-aldehyde condensation reaction and the implication of experimental data in above,
an overview of sequence for this synthetic reaction was suggested in Scheme 2. This overall reaction
sequence involves the formation and conversion of four major intermediates, including
(1) an o-tolylamine molecule reacted directly with a formaldehyde molecule to form an intermediate
o-tolylamino-methanol;
(2) the o-tolylamino-methanol was converted into a methylene-o-tolyl-amine ( Schiff base ) by loss of a
water;
(3) the Schiff base reacted with an o-tolylamine to form a N, N’-di-o-tolyl-methanediamine;
(4) N, N’-di-o-tolyl-methanediamine was rearranged to 2-methyl-4-(o-tolylamino-methyl)-phenylamine
or isomers;
(5) 2-methyl-4-(o-tolylamino-methyl)-phenylamine went through rearrangement to form final product
MDT.
Page 10 of 39

The detailed interpretation of the suggested mechanism and the idiographic description to reactions of
various intermediate were presented below.
3.4.2 The reactions of various intermediates
Great efforts have been made to sketch and describe the reaction mechanism of each individual
intermediate in the synthetic process of MDT from o-tolylamine and formaldehyde catalyzed by Hβ
zeolite.
o-Tolylamino-methanol formation
It is commonly accepted that the nucleophilic addition reaction of a carbonyl by an amine may take
place easily to form an unstable carbinolamine, and thus an acidic catalyst could be, but not necessary to
involve in this reaction step. As showed in Scheme 3, the direct attack of o-tolylamine to formaldehyde in
the absence of catalyst went through the transitional state to form the o-tolylamino-methanol.
Methylene-o-tolylamine formation
Stoichiometrically, an o-tolylamino-methanol could convert to a methylene-o-tolylamine (Schiff base)
by loss of a water. In fact, this process might be difficult to take place without the involvement of an acidic
catalyst. Over Hβ zeolite, this reaction was catalyzed by either a BrÖnsted acid or a Lewis acid site as
showed in Scheme 4(a) or Scheme 4(b). However, BrÖnsted acid appeared more active to this reaction
than Lewis acid, partly because the desorption of water from Lewis acid sites could be more difficult.
N,N’-Di-o-tolyl-methanediamine formation
An extra o-tolylamine reacted with a methylene-o-tolylamine absorbed on a BrÖnsted acid or a Lewis
acid site of Hβ to form the key intermediate N,N’-di-o-tolyl-methanediamine, as showed in Scheme 5(a)
and Scheme 5 (b). The data in Fig. 1 declared that this reaction was more favorable to be catalyzed by
BrÖnsted acid instead of Lewis acid. In the absence of an acid catalyst, the formation of N,N’-di-o-tolyl
Page 11 of 39

-methanediamine by the process as showed in Scheme 5 (c) was almost impossible.
2-Methyl-4-(o-tolylamino-methyl)-phenylamine formation
The rearrangement reaction of N,N’-di-o-tolyl-methanediamine almost certainly required BrÖnsted
acid sites on zeolite. As showed in Scheme 6, the amino group of N,N’-Di-o-tolyl-methanediamine was
first protonated by a BrÖnsted acid, nearly causing the dissociation of the aminal into a molecule of
o-tolylamine and a carbocation, which could interact by an electrophilic substitution with para-position
and ortho-position on the aromatic ring of o-toluidine to 2-methyl-4-(o-tolylamino-methyl)-phenylamine
and 2-methyl-6-(o-tolylamino-methyl)-phenylamine respectively, while the electrophilic attack of a
carbocation on the meta-position or δ-position on the ring of o-toluidine to form 2-methyl-5- or
2-methyl-3-(o-tolylamino-methyl)-phenylamine was almost impossible. Also, we assumed that the
formation of 2-methyl-4-(o-tolylamino-methyl)-phenylamine was favored due to the lowest steric
hindrance of the para-position compared to the ortho-position.
MDT formation
When the intermediate 2-methyl-4-(o-tolylamino-methyl)-phenylamine was adsorbed onto a BrÖnsted
acid site of the catalyst, the C-N bond would be weakened by the additional charge. The resulting positive
partial charge on the C-atom made it susceptible to a nucleophilic attack by the free electron pair of the
amino function or the phenyl ring of the substrate itself (intramolecular rearrangement), or one of the other
intermediates (intermolecular rearrangement). Considering the steric hindrance of intramolecular
rearrangement as shown in Scheme 7(a), we suggested that the substrate most likely underwent the
2
intermolecular rearrangement, i.e. an SN -type nucleophilic substitution reaction to form the final product
MDT as illustrated in Scheme 7(b). Obviously, the intermolecular rearrangement was favorable to a higher
concentration of intermediate 2-methyl-4-(o-tolylamino-methyl)-phenylamine.
3.4.3. The formation of by-products
Page 12 of 39

Besides the main stream reactions in the above sequence, there were some side-reactions to produce
by-products as indicated in Fig. 1. The data in Fig. 5 have confirmed that the formation of these
by-products could become very significant at higher temperature. These byproducts might mainly consist
of trimeric or polymeric compounds, and the proposed mechanism of trimeric by-products from dimeric
intermediates was presented in Scheme 8. It was believed that the formation and spread of such
by-products could gradually plug the pore channels of catalyst and further cause catalyst deactivation.
More experimental and theoretical studies for intermediates isomerisation, byproducts reaction and catalyst
deactivation are underway. We will present a thorough discussion on the details of by-product reaction and
catalyst deactivation mechanism elsewhere in near future.
4. Conclusions
For the acid-catalyzed heterogeneous synthesis of MDT from o-tolylamine and formaldehyde, an
effective solid acid catalyst should have strong Bronsted acidity and proper textural property. Hβ zeolite
demonstrated to be an good catalyst for this synthesis reaction. Reaction temperature significantly affected
the activity and selectivity of Hβ zeolite. The identification of key reaction intermediates and interpretation
to various intermediates formation and conversion have provided us with an insight into the details of
reaction mechanism.
(1) The whole synthetic chain might start from the formation of an unstable intermediate
o-tolylamino-methanol by the direct nucleophilic attack of o-tolylamine to formaldehyde either in the
presence of catalyst or in the absence of catalyst.
(2) The intermediate methylene-o-tolylamine (Schiff base) was produced from o-tolylamino-methanol by
loss of a water in the presence of Hβ zeolite.
(3) Key intermediate N,N’-di-o-tolyl-methanediamine was mainly formed by the reaction of o-tolylamine
with a methylene-o-tolylamine absorbed on BrÖnsted acid rather than Lewis acid sites.
(4) Rearrangement of N,N’-di-o-tolyl-methanediamine on BrÖnsted acid sites of Hβ generated
2-methyl-4-(o-tolylamino-methyl)-phenylamine or 2-methyl-6-(o-tolylamino-methyl)-phenylamine.
Of the two isomers, the former was more favorably produced than the latter.
(5) Formation of final product MDT was accomplished by intramolecular or intermolecular
Page 13 of 39

rearrangement of 2-methyl-4-(o-tolylamino-methyl)-phenylamine on Hβ. The intermolecular
2
rearrangement similar to an SN -type nucleophilic substitution reaction was more favorable.
Acknowledgment
We are grateful to Xiao Lin, Yi Wang and Shan Yan for assistance in various aspects of experiments.
References
[1] S. Michael, B.F. Matthias, A.L. Johannes, Applied Catalysis A: General 393 (2011) 189-194.
[2] A.L. Silva, J.C. Bordabo, Catal. Rev. 46 (2004) 31-51.
[3] J.K. Yu, Chemical, Propellants & Polymeric Materails 6 (2008) 15-20.
[4] Y.Q. Fang, J. Wang, Q. Zhang, Y. Zeng, Y.H. Wang, European Polymer J. 46 (2010) 1163-1167.
[5] L.H. Wang, J.D. Li, Z.P. Zhao, C.X. Chen, J. Macromolecular Science, Part A: Pure and Applied
Chemistry 43 (2006) 305-314.
[6] X.T. Han, Y. Tian, L.H. Wang, C.F. Xiao, B.Q. Liu, European Polymer J. 43 (2007) 4382-4388.
[7] Q. Tu, H.L. Shen, The proceedings of insulating material and technology conference, Beijing,
China, 2008, 296-301.
[8] S.Q. Zhang, Z.J. Ye, S.Q. Sun, Chemistry and Adhesion 4 (1989) 217-219.
[9] S.Q. Yang, Z.J. Ye, X.Q. Sun, Chemistry and Adhesion 2 (1990) 93-94.
[10] G.Q. Yan, D.C. Wang, Q. Zhao, Chemistry and Adhesion 2 (1997) 105-108.
[11] D. Bahulayan, R. Sukumar, K.R. Sabu, M. Lalithambika, Green Chemistry (1999)191-193.
[12] P. Botella, A. Corma, H.C. Robert, J.M. Christopher, Applied Catalysis A: General 398 (2011)
143-149.
[13] J.L. Nafzinger, L.A. Rader, L.J. Seward, U.S. Patent 4554378 (1985).
[14] B.A. Pablo, K.P.B. Joris, C. Avelino, U.S. Patent 0101801A1 (2005).
[15] W. Heiko, C.M. David, J.G. Robert, U.S. Patent 0144368A1 (2011).
[16] M. Kitano, K. Nakajima, J.N. Kondo, American Chemical Society 132 (2010) 6622-6623
[17] P. Carlo, D.A. Alberto, U.S. Patent 0171619A1 (2003).
[18] F.F. Frulla, A.A. Sayigh, H. Ulrich, P.J. Whitman, U.S. Patent 4092343 (1978).
[19] A. Corma, P. Botella, C. Mitchell, Chemical Communications (2004) 2008-2010
[20] K. Tsuyoshi, H. Shigekazu, N. Seitaro, Catalysis Today 111 (2006) 275-279.
Page 14 of 39

[21] Kiso Y, Takai T, Hayashi T. European Patent, 329,367
[22] D.F. Jin, Z.Y. Hou, X.M. Zheng, China Patent CN101007767A (2007).
[23] X. Lin, Y.X. Zhao, S.J. Zhang, L.Y. Wang, Y. Wang, Y.N. Guo, Catalysis Communications 45
(2014) 69-73
[24] C. Emeis, J. Catalysis 141 (1993) 347-354.
[25] C. Shi, B. Zhu, M. Lin, J. Lon, Catalysis Letters 133 (2009) 70-75.
[26] C. Baerlocher, L.B. McCusker, D.H. Olson, Atlas of Zeolite Framework Types, Sixth Ed.,
Elsevier (2007).
[27] R.J. Argauer, G.R. Landolt, U.S. Patent 3702886 ( 1972).
[28] M. Damour, Ann. Mines, Ser. 4 (1) (1842) 395–399.
Page 15 of 39

Fig. 1. HPLC chromatograms of product samples from reaction of o‐tolylamine and formaldehyde at 413
K. Reaction
conditions: n(o‐tolylamine):n (HCHO)=3:1, 1 h. (a) no catalyst, (b) w(TS): w(HCHO)=1:1, (c) w(Hβ):
w(HCHO)=1:1
Fig. 2. HCHO conversion, yield and selectivity of MDT in condensation of o‐tolylamine and formaldehyde
over various zeolites. Reaction conditions : n(o‐tolylamine): n (HCHO)=3:1，w(Hβ): w(HCHO)=1:1, 433K, 1 h.
Color annotation: red
(HCHO conversion); blue (MDT yield); green (MDT selectivity)
Fig. 3. NH3‐TPD curves of HY, HZSM‐5, Hβ and TS‐1 zeolites
Fig. 4 (a).
1H NMR spectrum of product MDT
Fig. 4 (b). 1H NMR spectrum of intermediate M3
Fig. 4 (c). 1H NMR spectrum of intermediate M1
Fig. 4 (d). 1H NMR spectrum of intermediate M2
Fig. 5. Effect of reaction temperature on HCHO conversion and MDT selectivity.
Reaction conditions : n(o-tolylamine): n (HCHO)=3:1, w(Hβ）: w(HCHO)=1:1, 1 h.
M3: N, N’-di-o-tolyl-methanediamine; M2: 2-methyl-4-(o-tolylamino-methyl)-phenylamine;
M1: 2-methyl-6-(o-tolylamino-methyl)-phenylamine; P: other byproducts
Fig. 6. Effect of reaction time on the yield and selectivity of MDT.
Reaction conditions: n（o-tolylamine）: n (HCHO)=3:1, w(Hβ）: w(HCHO)=1:1; 433 K.
Scheme 2. Overview of reaction sequence for MDT synthesis over Hβ zeolite
Scheme 3. Formation of o-tolylamino-methanol by the direct reaction of o-tolylamine with formaldehyde
Scheme 4(a). Formation of methylene-o-tolylamine from o-tolylamino-methanol catalyzed by BrÖnsted acid site on Hβ
Scheme 4(b). Formation of methylene-o-tolylamine from o-tolylamino-methanol catalyzed by Lewis acid site on Hβ
Scheme 5(a). Formation of N,N’-di-o-tolyl-methanediamine by reaction of o-tolylamine with o-tolylamino-methanol
absorbed on BrÖnsted acid site
Scheme 5(b). Formation of N,N’-di-o-tolyl-methanediamine by the reaction of o-tolylamine with o-tolylamino-methanol
absorbed on Lewis acid site
Scheme 5(c). Hypothetical process for the formation of N,N’-di-o-tolyl-methanediamine by reaction of o-tolylamine with
o-tolylamino-methanol in the absence of catalyst
Scheme 6. Rearrangement of N,N’-di-o-tolyl-methanediamine over a BrÖnsted acid site on Hβ
Scheme 7(a). Formation of MDT by intra-rearrangement of 2-methyl-4-(o-tolylamine-methyl)-phenylamine adsorbed on a
BrÖnsted acid site
Scheme 7(b). Formation of MDT by inter-rearrangement of 2-methyl-4-(o-tolylamine-methyl) -phenylamine adsorbed on a
BrÖnsted acid site
Scheme 8. Possible pathways of trimeric by-product formation from dimeric intermediates
Page 16 of 39

Page 17 of 39

Page 18 of 39

Page 19 of 39

Page 20 of 39

Page 21 of 39

Page 22 of 39

Page 23 of 39

Page 24 of 39

Page 25 of 39

Page 26 of 39

Page 27 of 39

Page 28 of 39

Page 29 of 39