Phosgene-free decomposition of dimethylhexane-1,6-dicarbamate over ZnO

Article abstract of DOI:10.1007/s11164-015-2224-x

In catalytic decomposition of dimethylhexane-1,6-dicarbamate (HDC) into hexamethylene-1,6-diisocyanate (HDI), Zn-containing homogeneous (i.e., zinc acetate) and heterogeneous (i.e., ZnO) catalysts were active among a number of catalysts tested, due to the

Full text of DOI:10.1007/s11164-015-2224-x

Res Chem Intermed
DOI 10.1007/s11164-015-2224-x
Phosgene-free decomposition of dimethylhexane-1,
6-dicarbamate over ZnO
Min Jeong Hyun1 Mi Shin1 Yong Jin Kim2
Young-Woong Suh1
•
•
•
Received: 11 July 2015 / Accepted: 11 August 2015
Ó Springer Science+Business Media Dordrecht 2015
Abstract In catalytic decomposition of dimethylhexane-1,6-dicarbamate (HDC)
into hexamethylene-1,6-diisocyanate (HDI), Zn-containing homogeneous (i.e., zinc
acetate) and heterogeneous (i.e., ZnO) catalysts were active among a number of
catalysts tested, due to the great electron withdrawing ability of Zn ions. Particu-
larly, when polyethylene glycol dimethyl ether was used as a solvent, ZnO was
found to be relatively robust, because the catalytic performance was maintained up
to the third use (HDC conversion of 93 % and HDI yield of 67 % at 180 °C for 1 h).
Through investigation of a HDC/ZnO mixture at elevated temperatures by IR
spectroscopy, a possible reaction scheme of ZnO-catalyzed decomposition of HDC
was proposed. The H atom is removed from the N–H group of HDC by hydrogen
bonding with an O site on the ZnO surface, followed by coordination of an O–C=O
group in monodentate mode to a Zn site. The C–O group in the O–C=O linkage is
then cleaved yielding the isocyanate and surface methoxide species. Finally,
methanol is released from ZnO by a reaction between the surface methoxide and the
hydroxyl species.
Keywords Catalytic decomposition Á Hexamethylene-1,6-diisocyanate Á
Dimethylhexane-1,6-dicarbamate Á ZnO
&
Young-Woong Suh
ywsuh@hanyang.ac.kr
1
2
Department of Chemical Engineering, Hanyang University, Seoul 133-791, Republic of Korea
Green Process Material Research Group, Korea Institute of Industrial Technology,
Cheonan 331-822, Republic of Korea
123

M. J. Hyun et al.
Introduction
Diisocyanates are important intermediates in the manufacture of polyurethanes, where
the typical chemicals are toluene diisocyanate (TDI) and methylene diphenyl
diisocyanate (MDI) to produce flexible and rigid polyurethane foams, respectively [1].
In addition to aromatic diisocyanates, aliphatic ones such as hexamethylene
diisocyanate (HDI) are useful for resistance to ultraviolet light and outdoor
weathering, even thoughthey are used in smaller volumes [2]. The global consumption
of these diisocyanates has been gradually increasing due to higher demand in versatile
commodity products including cushioning material in furniture and bedding,
insulation material in construction and refrigeration, synthetic leather, etc.
Diisocyanates have been typically synthesized by the reaction of amines with
phosgene or its derivatives. The phosgene route, however, has serious drawbacks
such as handling of toxic feedstock and formation of corrosive HCl as a byproduct;
thus, phosgene-free routes are now being developed in industry and academia [3, 4].
The reductive carbonylation of nitro compounds was introduced as a promising and
sustainable pathway [3], but the presence of highly toxic residues in the final
products was reported [5]. Additional attractive routes are the conversion to
carbamates, including the reductive carbonylation in alcohols [6] and the other
reaction of primary or secondary amine with dialkyl carbonates, particularly
dimethyl carbonate [7]. Although the decomposition of carbamates into isocyanates
is essential in both routes, few reports are yet available in the literature compared to
the synthesis of carbamates.
Thermal decomposition of aryl- and alkyl-carbamates has to be conducted at high
temperatures due to its endothermic reaction, resulting in the occurrence of many
side reactions and the lower selectivity to the desired isocyanates. These problems
can be overcome by an efficient catalyst operated at relatively lower temperatures. It
has been reported that Zn-containing catalysts, such as zinc powder [8], zinc acetate
[8], ZnO [9], Zn/ZnO [10], Zn-incorporated berlinite [7], and Bi-Zn composite
oxide [11], as well as di-n-butyltin oxide [12, 13], montmorillonite K-10 [14, 15],
and nano Cu2O [16] are active in carbamate decomposition. Among these,
ZnAlPO4, Bu2SnO, and K-10 catalysts were applied for decomposition of
dimethylhexane-1,6-dicarbamate (HDC) into HDI, where the mono-isocyanate
(HMI) is an intermediate formed by the decomposition of one carbamate end of
HDC as follows [17]:
In this study, the catalyst screening work was conducted to find a highly active
catalyst for HDC decomposition. Particularly, our attention was paid to the
development of a heterogeneous catalyst due to its easy handling in separation and
reuse. The effects of several reaction variables such as reaction solvent,
temperature, and time were subsequently investigated in order to achieve the
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
highest HDI yield under the optimized reaction condition. The experiments for
catalyst recycling were also carried out. Consequently, the change of characteristic
amide bands of HDC in the absence or presence of the catalyst was investigated
using infrared spectroscopy, thus proposing a possible reaction scheme.
Experimental
Chemicals and materials
HDI (99 %, TCI), nitrobenzene (99.5 %, Junsei), and polyethylene glycol dimethyl
ether (avg. Mn & 250, Sigma-Aldrich) were purchased and used without purifica-
tion. HDC was synthesized by the reaction of HDI with methanol at 50 °C for 6 h,
where the methanol/HDI molar ratio was 10. The compound was finally obtained by
1
purification, yielding the purity of [99 %, confirmed by H NMR spectroscopy.
Prior to use, the sample was kept in a vacuum desiccator.
Homogeneous catalysts used for catalyst screening were Ca(CH3COO)2ÁnH2O
(99 %, Sigma-Aldrich), Cr(CH3COO)2ÁnH2O (95 %, Sigma-Aldrich), La(NO3)3-
6H2O (98 %, Junsei), Mg(CH3COO)2Á4H2O (99 %, Sigma-Aldrich), Mn(CH3COO)2-
4H2O (99 %, Yakuri), Ni(CH3COO)2Á4H2O (98 %, Junsei), Zn(CH3COO)2Á2H2O
(99 %, Sigma-Aldrich), ZrCl4 (98 %, Alfa Aesar), ZrOCl2Á8H2O (98 %, Sigma-
Aldrich), and ZrO(NO3)2ÁnH2O (100 %, Strem). These are used after pretreatment at
105 °C for 12 h. For heterogeneous catalysts, activated carbon (250–350 mesh,
Yakuri), c-Al2O3 (97 %, Strem), BaO (99.99 %, Sigma-Aldrich), and MgO (99.99 %,
Sigma-Aldrich) were used as received. In case of zeolite catalysts, NH4-ZSM-5 (SiO2/
Al2O3 = 30, CBV3024E, Zeolyst) was transformed into H-ZSM-5 by calcination at
600 °C for 3 h, while H-Y (SiO2/Al2O3 = 30, CBV720, Zeolyst) and H-Beta (SiO2/
Al2O3 = 28, 931HOA, Tosoh) were pretreated at 500 °C with O2 for 3 h.
Preparation of ZnO
ZnO catalyst was prepared by the precipitation method. In a typical preparation, an
aqueous Zn(NO3)2 solution (1.0 M, 100 ml) was added with the rate of 5 ml/min
into an aqueous NaHCO3 solution (1.2 M, 200 ml) at 70 °C under vigorous stirring.
After complete addition (final pH of about 7), the resulting precipitate suspension
was stirred at the same temperature for 30 min and aged for 60 min. Then, the
precipitate was filtered and washed with deionized water (ca. 1000 ml). The
washing step was repeated at least three times in order to completely remove
residual ions. The filter cake was dried in an oven at 105 °C overnight, crushed, and
sieved for obtaining the particle size of smaller than 200 lm. Finally, the
calcination was carried out in static air at 350 °C for 3 h.
Characterization
XRD analysis was conducted with a Rigaku D/Max RINT 2000 diffractometer using
Cu Ka (k = 0.1541 nm) as a radiation source (40 kV and 60 mA) in the 2h range of
123

M. J. Hyun et al.
10°–90°. The infrared spectra of the samples were recorded using a Nicolet 6700
FT-IR spectrometer equipped with a MCT-A detector in the range of
4000–650 cm-1 with the resolution of 4 cm-1. For the preparation of IR pellets,
HDC was premixed with KBr and ZnO at the mixing weight ratio of 1/100 and
1/200, respectively, followed by pressing the sample (100 mg for HDC/KBr and
50 mg for HDC/ZnO) at 2700 psi for 90 s. The resulting pellet was loaded inside
the IR cell (Specac HTHP cell with ZnSe windows) working in the transmission
mode [18] and purged at an ambient temperature with N2 (99.999 %, 50 ml/min) for
0.5–1 h. Finally, as the temperature was increased to 200 °C at the ramping rate of
5 °C/min under N2 atmosphere, IR spectra were obtained at certain temperatures,
where KBr and ZnO were used to take a background spectrum for HDC/KBr and
HDC/ZnO pellets, respectively.
Activity test
The reaction was carried out in a three-neck round-bottom flask (100 ml) to which
the reaction solvent (5 g) and HDC (0.5 g) were charged. Then, the reaction flask
was purged with N2 (99.999 %) at the flow rate of 150 ml/min for 30 min to make
sure of the complete removal of inner oxygen and heated at the ramping rate of
10 °C/min. Note that any HDC is not lost at this purge step. As soon as the
temperature approached the desired value, the catalyst of 30 mg was added to the
reaction mixture and the reaction was started (reaction time = 0). Note that N2 flow
of 150 ml/min in the reaction flask was kept until the end of the reaction in order to
continuously remove methanol produced by the decomposition of carbamate to
isocyanate. After the reaction was complete, the sample (ca. 0.2 g) was taken out,
filtered through a syringe filter (0.45 lm), and mixed with the GC calibration
solvent N-methyl-2-pyrrolidone (NMP). This mixture was finally analyzed with a
Younglin YL6500 GC equipped with a FID detector and an HP-5 column
(30 m 9 0.25 mm 9 0.25 lm). Through quantitative measurements of HDC and
HDI, the HDC conversion and HDI yield were calculated based upon the quantities
of the converted HDC and the produced HDI. For calculating the HMI yield, the GC
conversion factor (CF) of HDI was utilized assuming that the CF values of HMI and
HDI were the same, since its authentic sample was not commercially available. In
this case, the mole balance was measured to be almost near 100 % with the
experimental error of 3 % at all reaction runs.
Results and discussion
Catalyst screening on the catalytic decomposition of HDC
For the conversion of HDC into the corresponding isocyanate at 160 °C, a number
of catalysts, homogeneous or heterogeneous, were tested in this work (Fig. 1).
Among homogeneous catalysts, zinc acetate (Zn(OAc)2) showed the best catalytic
performance, where the yields of HMI and HDI were 41 and 17 %, respectively.
Note that HMI is a more abundant product than HDI, meaning the facile
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
Fig. 1 Activities of homogeneous (a) and heterogeneous (b) catalysts tested for HDC decomposition at
160 °C. The asterisk stands for the use of 180 °C for ZnO catalyst
decomposition of HDC to mono-isocyanate rather than to diisocyanate. Thus,
several homogeneous catalysts with an acetate ligand were applied for HDC
decomposition; however, their catalytic activities were not superior to that of
Zn(OAc)2 (Fig. 1a). The activities of several heterogeneous catalysts were also
investigated, in which their character depends upon acidity, basicity, and
amphotericity. However, other catalysts except ZnO did not exhibit any noticeable
isocyanate yields (Fig. 1b). The HMI and HDI yields were 22 and 7 % over ZnO,
respectively. It should be noted here that the HDI selectivity was not high even if the
reaction temperature of 180 °C was employed in order to compensate for lower
activities of heterogeneous catalysts. The higher activities of Zn(OAc)2 and ZnO are
presumably due to greater electron withdrawing ability of Zn ions, which was
similarly observed in the methoxycarbonylation of diamine with dimethyl carbonate
[19, 20].
Effect of reaction variables on the catalytic decomposition of HDC
Since the above results were obtained using nitrobenzene as a solvent, a new
reaction solvent would be required for enhancing the catalytic activity and
producing the diisocyanate HDI. It was found from the literature that there are the
two key properties of a reaction solvent in carbamate decomposition, that is, a high
boiling point and a proton withdrawing (electron donating) character. The first
property comes from the fact that the conversion of carbamate needs to be operated
at elevated temperatures because it is an endothermic reaction [7], while the latter is
associated with an E1CB mechanism in which the hydrogen atom is first removed
from the N–H group of carbamate [21, 22]. Thus, we tested polyethylene glycol
dimethyl ether (PGDE) as a solvent, since Yang et al. [23] proposed a favorable
production of dimethyl carbonate from urea and methanol owing to both properties
of PGDE (i.e., a high boiling electron donor). Figure 2 shows the activities of
Zn(OAc)2 and ZnO in nitrobenzene and PGDE, where HDC decomposition over the
former and latter catalysts was conducted at 160 and 180 °C, respectively.
Remarkably, the activities of both catalysts in PGDE were superior to those in
123

M. J. Hyun et al.
Fig. 2 Activity comparison of Zn(OAc)2 and ZnO catalysts in the HDC decomposition using the solvent,
nitrobenzene or PGDE for 1 h. The reaction temperature was 160 °C for Zn(OAc)2 and 180 °C for ZnO
nitrobenzene, where the major product was HDI in PGDE over the two catalysts.
Even in a blank experiment (i.e., without a catalyst) at 180 °C, PGDE exhibited the
HDC conversion and HDI yield of 23 and 8 %, respectively, whereas any isocyanate
products were not detected in nitrobenzene. This originates from the electron donor
character of PGDE to abstract the H atom from the N–H group of carbamate.
Therefore, it is believed that PGDE is a powerful solvent for efficient decomposition
of HDC into the corresponding diisocyanate.
For further work, ZnO was chosen as a catalyst, although Zn(OAc)2 showed the
higher HDI yield. This is due to difficulty in the separation and reuse of
homogeneous catalysts, as well as decomposition of acetate ligand above 160 °C
observed by TGA. The effects of reaction variables (mainly, reaction temperature
and time) were investigated for ZnO-catalyzed decomposition of HDC in PGDE.
Figure 3 presents the HDC conversion and product yields obtained at the
temperature of 160–200 °C, because the catalytic activity was hardly achieved
below 160 °C and the mole balance at higher temperatures than 200 °C was far
from 100 % due to occurrence of side reactions discussed below. At 160 °C, ZnO
appeared to become active (HDC conversion = 28 %) with the HDI selectivity of
50 %. The activity of ZnO substantially increased at 180 °C affording the HDC
conversion of 96 % and the total isocyanates yield of 89 % (74 % HDI and 15 %
HMI). The difference between the two values was more evident at 200 °C
(16 % = 100 % HDC conversion - 84 % HDI yield). This indicates further
conversions of some isocyanates at higher temperatures, such as the self-addition of
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
Fig. 3 Effect of reaction temperature on the HDC decomposition over ZnO catalyst in PGDE for 1 h
isocyanate (products: uretidinedione, isocyanurate, carbodiimide), condensations
between unconverted carbamate and isocyanate (product: allophanate) and between
carbodiimide and isocyanate (product: uretoneimine), and further polymerization
[7]. Since polymerized products were visually observed inside the reactor after the
reaction, a temperature higher than 200 °C should be avoided to produce HDI at the
maximum yield.
The reaction progress of HDC decomposition over ZnO was monitored at 180 °C
(Fig. 4). In an initial run, HMI and HDI were produced in similar amounts (22 and
20 % at 20 min, respectively). At little longer reaction periods, HDI was the major
product because of the faster conversion of HMI into HDI than the initial HDC-to-
HMI conversion in PGDE. The HDC conversion was complete and the HDI yield
was close to 97 % at 2.5 h. When the experiment was conducted for more than 3 h,
the HDI yield dropped to 85 % due to the above-mentioned side reactions.
Therefore, the reaction time of 2–3 h is considered to be appropriate for the highest
HDI yield.
Reusability of ZnO catalyst on the catalytic decomposition of HDC
The catalytic stability of ZnO was examined by recycling experiments under the
condition such as 2 g HDC in 20 g PGDE, 0.133 g ZnO, 180 °C, and 1 h, which is a
fourfold scale up than that used in the above sections for reproducibility and catalyst
recovery. As shown in Fig. 5, the HDC conversion and HDI yield in the first run
123

M. J. Hyun et al.
Fig. 4 Effect of reaction time on the HDC decomposition over ZnO catalyst in PGDE at 180 °C
Fig. 5 Recycling test of ZnO catalyst on the HDC decomposition in PGDE. Reaction condition: HDC
2 g, PGDE 20 g, ZnO 0.133 g, 180 °C, and 1 h
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
were 95 and 72 %, respectively, where the HDI/HMI molar ratio was 4.9. Since the
result is considered to be comparable to the previous data (HDC conversion of 96 %
and HDI yield 74 %), the catalytic performance of ZnO appeared to be quite
reproducible. Furthermore, the activity of ZnO was found to be maintained up to the
third run, when the spent catalyst was recovered repeatedly through a series of
operations including centrifugation, washing with methanol, and drying at 105 °C.
However, the HDC conversion and HDI yield at the fourth run were decreased to 87
and 61 %, respectively, where the HDI/HMI molar ratio was 3.7. The apparent
activity loss may be due to leaching of ZnO in some amounts into the reaction
mixture and/or transformation of ZnO structure during recycling experiments.
In order to investigate the change of ZnO crystals during recycling experiments,
XRD analysis was conducted for fresh and spent ZnO catalysts. Figure 6 presents
XRD patterns of the fresh ZnO, and the spent ZnO catalysts obtained after the first
and fourth runs. Evidently, all the reflections were identical to those of ZnO (JCPDS
PDF No. 36-1451). Two kinds of differences were, however, observed in the spent
catalysts: the intensities of major reflections of ZnO were lower and the baseline
hump was seen at 2h = 21°. These findings suggest that a certain amount of
crystalline ZnO is transformed into an amorphous form and subsequently leached
out into the reaction mixture.
Fig. 6 XRD patterns of fresh ZnO (a) and spent ZnO obtained after the first (b) and fourth (c) runs
123

M. J. Hyun et al.
Plausible reaction mechanism for ZnO-catalyzed HDC decomposition
Amide bands of HDC were investigated by FT-IR spectroscopy, where HDC was
pressed with KBr that had no catalytic activity on this reaction (Fig. 7a). Several
characteristic bands were observed at 30 °C; 3340 cm-1 for amide A (m(N–H)),
1693 cm-1 for amide I (m(C=O)), 1537 cm-1 for amide II (d(N–H), m(C–N)),
1264 cm-1 for amide III (d(N–H), m(C=O), m(C–N)), 1195 and 1053 cm-1 for
urethane (m(C–N), m(C=O), d(C–O–C)). These frequency values were similar within
the frequency difference of 1–4 cm-1 to those of a crystal HDC reported by
Suchkova and Maklakov [24]. Upon heating the sample in N2 flow to 200 °C, blue
or red shifts occurred in these amide bands (blue: amide A and amide I, red: amide
II and amide III). This is caused by dissociation of hydrogen bonds at elevated
temperatures. The typical example is variation of amide A band with increasing the
temperature. The intensity of the 3354 cm-1 absorption peak assigned to the bonded
N–H group decreased and, simultaneously, that of the 3443 cm-1 absorption peak
assigned to the free N–H group increased, as observed in N-hexyl hexyl carbamate
[25]. This is an indication of a thermal effect on hydrogen bonding of HDC.
Fig. 7 FT-IR spectra obtained by heating pure HDC (a) and a mixture of HDC and ZnO (b) from 30 to
200 °C (ramping rate: 5 °C/min)
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
A mixture of HDC and ZnO was also characterized with increasing the
temperature up to 200 °C (Fig. 7b). Compared to pure HDC, the absorption of
amide A (3370 cm-1), II (1527 cm-1) and III (1252 cm-1) bands were much
diminished, relative to that of the amide I band (1686 cm-1), in the spectrum
recorded at 30 °C. Since these bands are all associated with the N–H group of HDC,
the N–H bonding was presumed to be weakened due to interaction with ZnO. This is
in accordance with appearance of new bands at 3564 and 3475 cm-1, because Noei
et al. [26] ascribed the bands to hydroxyl species (N–HÁÁÁO) formed by interaction of
the N–H group with oxygen sites of the ZnO surface. Therefore, it is believed that
the N–H group of HDC is coordinated via the H atom to the O site of the ZnO
surface at 30 °C.
Upon heating the HDC/ZnO mixture to 200 °C, remarkable changes were
observed. (1) The intensity of the amide A band at 3370 cm-1 and those of the
bands at 3564 and 3475 cm-1 became negative. (2) The absorption bands at 1485,
1396, and 1352 cm-1, related with das or ds(C–N–H) and m(C–N) [27], gradually
decreased. These changes are explained by complete cleavage of N–H bonding of
HDC and disappearance of the hydroxyl species formed on the ZnO surface. (3) The
absorption of the amide I band at 1686 cm-1 was not shifted while the intensity was
decreasing. When compared to the same band of pure HDC, the frequency
difference was only 6 cm-1. This implies no interaction of the C=O group of HDC
with ZnO. (4) The absorption bands at 1582 and 1373 cm-1 grew up to 160 °C and
then disappeared around 190–200 °C. From IR spectra of CO2-adsorbed ZnO
samples [28, 29], these are considered to be a consequence of carboxylate binding
with Zn sites: 1582 and 1373 cm-1 are assigned to mas(O–C=O) and ms(O–C=O),
respectively. Also, the Dm (=mas(O–C=O) - ms(O–C=O)) of 209 cm-1 indicates
monodentate coordination, because a larger Dm value than 180 cm-1 is typical for
this type of carboxylate binding mode [30]. Therefore, it can be suggested that the
O–C=O group of HDC is coordinated in monodentate mode via the O atom of the
C–O group to the Zn site of the ZnO surface at higher temperatures to 160 °C.
Above this temperature, however, this binding will lessen (5). New absorption
bands at 2201 and 2816 cm-1 were observed in the temperature range of
150–200 °C. For clarification, the spectra in the region 2900–2100 cm-1 were
shown in Fig. 8. The former band suggests formation of isocyanate species on ZnO.
Guglielminotti and Boccuzzi [31] observed a band at 2205 cm-1 assigned to an
isocyanate group formed by adsorption of NO and CO on Ru/ZnO. This is a direct
evidence to show the transformation of the carbamate group of HDC into isocyanate
on ZnO in this temperature range where the activity tests were conducted. On the
other hand, the latter frequency is assigned to ms(C–H) of CH3O adsorbate, which
was confirmed by exposure of methanol vapor on ZnO [32]. This means the
cleavage of the C–O bond in the O–C=O linkage leaving the methoxide (CH3O)
species on the ZnO surface. Therefore, the observation of these two bands suggests
that the carbamate group of HDC is decomposed to isocyanate over ZnO at
150–200 °C. (6) At 200 °C, the spectrum of the HDC/ZnO mixture was very similar
to that of ZnO.
From the aforementioned FT-IR observations, a possible reaction scheme on
ZnO-catalyzed decomposition of carbamate was proposed, as depicted in Fig. 9.
123

M. J. Hyun et al.
Fig. 8 Enlarged version of Fig. 7b in the region of 2900–2100 cm-1
Fig. 9 Possible reaction scheme on ZnO-catalyzed decomposition of carbamate to isocyanate
The H atom is first removed from N–H group by hydrogen bonding with the O site
of the ZnO surface, indicating that the decomposition follows an E1CB mechanism.
Then, the O–C=O group is coordinated in the monodentate mode to the Zn site with
123

Phosgene-free decomposition of dimethylhexane-1,6-dicarbam…
an electron withdrawing character. The C–O group in the O–C=O linkage is
subsequently cleaved yielding an isocyanate product. The resulting surface
methoxide species linked to Zn reacts with the H atom of the surface hydroxyl
species, finally releasing methanol from ZnO.
Conclusion
We have elucidated the method to efficiently decompose HDC to HDI. Catalyst
screening experiments revealed that Zn-containing catalysts Zn(OAc)2 and ZnO
were highly active in the reaction due to the electron withdrawing character of Zn
species. Furthermore, the HDC conversion and HDI yield of the two catalysts were
enhanced using electron-donor PGDE as a solvent. From optimization works over
ZnO selected due to thermal instability of Zn(OAc)2, the reaction condition of
180–200 °C and 2–3 h was appropriate for the highest HDI yield. Additionally,
ZnO was found to be recyclable up to the third use. FT-IR results at elevated
temperatures indicated the removal of the H atom from the N–H group of carbamate
by interaction with the O site of the ZnO surface, the monodentate coordination of
the O–C=O linkage to the Zn site, and the cleavage of the C–O group in O–C=O
linkage yielding the isocyanate product and surface methoxide species that will be
released into methanol from ZnO, thus suggesting a possible reaction scheme.
Although the explanation was developed from results using HDC as the reactant, it
should apply to other aliphatic carbamates since the structural and chemical
characters of ZnO are important factors for decomposition performance.
Acknowledgments This research was supported by the Fusion Research Program for Green
Technologies through the National Research Foundation of Korea (NRF) funded by the Ministry of
Science, ICT & Future Planning (2012M3C1A1054501).
References
1. Z. Wirpsza, Polyurethanes: Chemistry, Technology, and Applications (Ellis Horwood, London, 1993)
2. M. Szycher, Szycher’s Handbook of Polyurethanes, 2nd edn. (CRC Press, New York, 2012)
3. O. Kreye, H. Mutlu, M.A.R. Meier, Green Chem. 15, 1431–1455 (2013)
4. H. van den Berg, L. van der Ham, H. Gutierrez, S. Odu, T. Roelofs, J. de Weerdt, Chem. Eng. J.
207–208, 254–257 (2012)
5. E. Drent, P.W.N.M. van Leeuwen, EP Patent 86281 (1983)
6. D. Chaturvedi, Tetrahedron 68, 15–45 (2012)
7. D.-L. Sun, J.-Y. Luo, R.-Y. Wen, J.-R. Deng, Z.-S. Chao, J. Hazard. Mater. 266, 167–173 (2014)
8. X. Zhao, Y. Wang, S. Wang, H. Yang, J. Zhang, Ind. Eng. Chem. Res. 41, 5139–5144 (2002)
9. D. Chen, L.-M. Liu, Y. Wang, J. Yao, G.-Y. Wang, S.-X. Li, Y. Xue, J.-H. Zhan, Chin. J. Catal. 26,
987–992 (2005)
10. X. Guan, H.Q. Li, H.T. Liu, F. Guo, X.X. Yao, J. Beijing Univ. Chem. Technol. (Nat. Sci.) 36, 12–16
(2009)
11. Y. Dai, Y. Wang, Q. Wang, G. Wang, Chin. J. Catal. 30, 1131–1136 (2009)
12. M. Takahito, A. Kyoji, K. Yasushi, JP Patent 6239826 (1994)
13. Y.-S. Dai, Y. Wang, J. Yao, Q.-Y. Wang, L.-M. Liu, L.-L. Cui, Y.-F. Zhao, G.-Y. Wang, Acta Chim.
Sin. 65, 1064–1070 (2007)
14. P. Uriz, M. Serra, P. Salagre, S. Castillon, C. Claver, E. Fernandez, Tetrahedron Lett. 43, 1673–1676
(2002)
123

M. J. Hyun et al.
15. C.M. Serglo, C.C. Cannen, US Patent 6639101 (2003)
16. Q. Wang, W. Kang, Y. Zhang, X. Yang, J. Yao, T. Chen, G. Wang, Chin. J. Catal. 34, 548–558
(2013)
17. X. Li, H. Li, H. Liu, G. Zhu, J. Hazard. Mater. 198, 376–380 (2011)
18. J. Ryczkowski, Catal. Today 68, 263–381 (2001)
19. T. Baba, A. Kobayashi, Y. Kawanami, K. Inazu, A. Ishikawa, T. Echizenn, K. Murai, S. Aso, M.
Inomata, Green Chem. 7, 159–165 (2005)
20. D.-L. Sun, J.-R. Deng, Z.-S. Chao, Chem. Cent. J. 1, 27–35 (2007)
21. B.M. Bergon, N.B. Hamida, J.-P. Calmon, J. Agric. Food Chem. 33, 577–583 (1985)
22. M.B. Berezin, O.M. Chernova, P.A. Shatunov, N.A. Pashanova, D.B. Berezin, A.S. Semeikin,
Molecules 5, 809–815 (2000)
23. B. Yang, D. Wang, H. Lin, J. Sun, X. Wang, Catal. Commun. 7, 472–477 (2006)
24. G.C. Suchkova, L.I. Maklakov, Vib. Spectrosc. 51, 333–339 (2009)
25. Q. Li, H. Zhou, D.A. Wicks, C.E. Hoyle, D.H. Magers, H.R. McAlexander, Macromolecules 42,
1824–1833 (2009)
¨
¨
26. H. Noei, H. Qiu, Y. Wang, E. Loffler, C. Woll, M. Muhler, Phys. Chem. Chem. Phys. 10, 7092–7097
(2008)
27. D.-L. Sun, S.-J. Xie, J.-R. Deng, C.-J. Huang, E. Ruckenstein, Z.-S. Chao, Green Chem. 12, 483–490
(2010)
28. G. Busca, V. Lorenzelli, Mater. Chem. 7, 89–126 (1982)
29. S. Matsushita, T. Nakata, J. Chem. Phys. 36, 665–669 (1962)
ˇ´
´
´
30. V. Zelenak, Z. Vargova, K. Gyo¨ryova, Spectrochim. Acta A 66, 262–272 (2007)
31. E. Guglielminotti, F. Boccuzzi, J. Catal. 141, 486–493 (1993)
32. R.N. Spitz, J.E. Barton, M.A. Barteau, R.H. Staley, A.W. Sieight, J. Phys. Chem. 90, 4067–4075
(1986)
123