Study on the Structure of Cu/ZrO2 Catalyst and the Formation Mechanism of Disodium Iminodiacetate and Sodium Glycine

Article abstract of DOI:10.1007/s10562-019-02989-z

Abstract: A new method to prepare Cu/ZrO₂ catalysts by reducing CuO/ZrO₂ with hydrazine hydrate is reported, and the prepared catalysts were used to synthesize disodium iminodiacetate by diethanolamine dehydrogenation. Hydrazine hydrate can rapidly reduce the CuO/ZrO₂ precursor powder in an alkaline environment at a moderate temperature. The ratio of Cu⁰/Cu⁺ at the Cu/ZrO₂ surface was controlled by the amount of hydrazine hydrate and the reduction reaction time. The formation mechanism of disodium glycine as the main byproduct and iminodiacetate were deduced by investigating the product yield, the reaction time, and the presence of acetaldehyde in the evolved gas. It has been shown that the ratio of Cu⁰/Cu⁺ in Cu/ZrO₂ significantly affects the dehydrogenation of diethanolamine into disodium iminodiacetate. Cu⁰ and Cu⁺ are the catalytic activity centers in the dehydrogenation of diethanolamine which respectively produce intermediate aldehydes and an ester via nucleophilic addition reactions. The formation mechanism of sodium glycinate is related to the tautomerism of intermediate products and Schiff base hydrolysis. Graphic Abstract: The formation mechanism of disodium iminodiacetate and sodium glycine producing by the dehydrogenation of diethanolamine over the Cu/ZrO₂ catalysts which were prepared by a new reduction method.

Full text of DOI:10.1007/s10562-019-02989-z

Catalysis Letters
https://doi.org/10.1007/s10562-019-02989-z
Study on the Structure of Cu/ZrO Catalyst and the Formation
2
Mechanism of Disodium Iminodiacetate and Sodium Glycine
Yongsheng Wang · Hongwen Zhu · Zhengkang Duan · Zhenzhen Zhao1,2 · Yunlu Zhao1,2 · Xiaolin Lan1 ·
1
,2
1
1,2
1
1
Li Chen · Dongjie Guo
Received: 2 July 2019 / Accepted: 3 October 2019
Springer Science+Business Media, LLC, part of Springer Nature 2019
©
Abstract
A new method to prepare Cu/ZrO catalysts by reducing CuO/ZrO with hydrazine hydrate is reported, and the prepared
2
2
catalysts were used to synthesize disodium iminodiacetate by diethanolamine dehydrogenation. Hydrazine hydrate can
0
rapidly reduce the CuO/ZrO precursor powder in an alkaline environment at a moderate temperature. The ratio of Cu /
2
+
Cu at the Cu/ZrO surface was controlled by the amount of hydrazine hydrate and the reduction reaction time. The forma-
2
tion mechanism of disodium glycine as the main byproduct and iminodiacetate were deduced by investigating the product
0
+
yield, the reaction time, and the presence of acetaldehyde in the evolved gas. It has been shown that the ratio of Cu /Cu
0
+
in Cu/ZrO signiꢀcantly aꢁects the dehydrogenation of diethanolamine into disodium iminodiacetate. Cu and Cu are the
2
catalytic activity centers in the dehydrogenation of diethanolamine which respectively produce intermediate aldehydes and
an ester via nucleophilic addition reactions. The formation mechanism of sodium glycinate is related to the tautomerism of
intermediate products and Schiꢁ base hydrolysis.
Graphic Abstract
The formation mechanism of disodium iminodiacetate and sodium glycine producing by the dehydrogenation of diethanola-
mine over the Cu/ZrO catalysts which were prepared by a new reduction method.
2
Keywords Copper-zirconium catalysts · Diethanolamine · Disodium iminodiacetate · Reaction mechanism · Hydrazine
hydrate
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10562-019-02989-z) contains
1
Introduction
supplementary material, which is available to authorized users.
*
Zhengkang Duan
Glyphosate is widely used as a broad spectrum, eꢂcient,
and environmentally friendly pesticide [1] that is typically
obtained by one of two main industrial routes: (I–III) the
dzk0607@163.com
Extended author information available on the last page of the article
Vol.:(0
1
123456789)
3

Y. Wang et al.
glycine (Gly) route and (IV–V) the iminodiacetic acid (IDA)
route [2]. The iminodiacetic acid route usually involves
transforming IDA, formaldehyde, phosphoric acid, and
other reactants into N-(phosphonomethyl) iminodiacetic acid
hydrate (PMIDA), which is then oxidized to obtain glypho-
sate. Currently, using diethanolamine (DEA) dehydrogena-
tion to synthesize IDA is the most environmentally-friendly
and safe process [3].
main species present, but when the Cu content is low, the
amount of Cu in the ZrO carrier is high, and the nearby
2
+
0
oxygen vacancy makes it diꢂcult to reduce Cu to Cu [22,
0
23]. The type of carrier also has a certain eꢁect on the Cu /
+
Cu pair. As the amount of zinc oxide content increases in
+
Cu/SiO catalysts, the Cu content also increases [24]. Cu O
2
2
provides active acidic sites (Lewis acid sites) on the surface
of the catalysts which help promote the chemical adsorption
capacity of intermediate aldehydes, thereby increasing the
ꢀ
ꢁ
ꢀ
ꢁ
NH CH COOH + CH O → HOCH NCH COOH
2
reaction rate and conversion of organic acids [15, 25]. Cu O/
2
2
2
2
2
n
2
(I)
MgO has a better catalytic activity than Cu/MgO for the
ꢀHOCH
ꢁ
ꢀ
ꢁ
catalytic reaction of cyclohexanol for cyclohexanone pro-
NCHCOOH + CH O P(O)H
3
+
0
2
duction [26], suggesting that Cu and Cu have diꢁerences
in their catalytic activities.
2
2
ꢀ
ꢁ
ꢀ
ꢁ
(II)
(III)
(IV)
→
CH O P(O)CH N CH OH CH COOH
3
2
2
2
2
In the present paper, Cu/ZrO catalysts were prepared by
2
ꢀ
ꢁ
ꢀ
ꢁ
reducing CuO/ZrO precursors with hydrazine hydrate. The
CH O P(O)CH N CH OH CH COOH
3
2
2
2
2
2
ꢀ
ꢁ
best results obtained with this catalyst were a 98% yield of
disodium iminodiacetate in a 2.3 h reaction time. There-
fore, hydrazine hydrate is an excellent candidate to produce
→
(HO) P(O)CH N CH OH CH COOH
2 2 2 2
ꢀ
ꢁ
NH CH COOH + HCHO + PCl3
2
0
+
2
Cu/ZrO catalysts. The eꢁect of the Cu /Cu pair on the
2
ꢀ
ꢁ
→
(HO) OPCH N CH COOH
2
catalytic reaction was studied by controlling the reduction
2
2
2
degree of CuO/ZrO precursors. It was found that increas-
ꢀ
ꢁ
2
0
(
HO) OPCH N CH COOH → (HO) OPCH NHCH COOH
2
2
2
2
2
2
2
ing the Cu content favored the dehydrogenation of DEA to
+
(V)
form intermediate aldehydes, and Cu promoted the reac-
DEA reacts with NaOH in an autoclave at 160–180 °C
tion of an aldehyde intermediate to form iminodiacetate. The
formation mechanism of sodium glycinate is related to the
tautomerism of the intermediate products and the hydrolysis
of a Schiꢁ base.
and a pressure of 1–2 MPa over Cu-based catalysts to pro-
duce disodium iminodiacetate, which is then puriꢀed by
acidiꢀcation to obtain IDA [4–7]. The catalytic mechanism
of DEA dehydrogenation has rarely been reported and is
generally thought to be similar to the dehydrogenation
mechanism of other fatty alcohols where esters are produced
using Cu-based catalysts. The Cu-based catalysts catalyze
the dehydrogenation of diethanolamine to form an interme-
diate aldehyde, followed by two molecules of aldehyde com-
bining to form a carboxylic acid and an alcohol [8].
2
Experimental
2.1 Catalyst Preparation
CuO/ZrO precursor was synthesized by a co-precipitation
2
Cu-based catalysts are widely used in the dehydroge-
nation of alcohols to produce esters or aldehydes [9–14].
Excellent results for the selective conversion alcohols to
method. Brieꢃy, ZrOCl ·8H O and Cu(NO ) ·3H O were
2
2
3 2
2
dissolved in deionized water to obtain Cu/Zr in a molar ratio
of 2:1. Then, 1.0 mol/L NaOH solution was added drop-
wise under continuous stirring to adjust the pH to 12. The
resulting precipitate was aged at room temperature for 4 h.
Next, the precipitate was ꢀltered and washed by deionized
water several times in a Buchner funnel until the pH value
of the supernatant was neutral. The composite precursor was
dehydrated and dried at 90 °C for 12 h. The resulting powder
was calcined in a muꢄe furnace from room temperature to
esters have been achieved using Cu/ZrO catalysts [15–20],
2
and interfacial metal-support interactions between Cu and
ZrO enhances the dehydrogenation of ethanol to form ethyl
2
acetate [18]. A CuO/ZrO precursor was prepared by co-
2
precipitation, followed by high-temperature reduction under
an H ꢃow to obtain Cu/ZrO catalysts [15, 19–21]. In addi-
2
2
0
+
tion to Cu , a small amount of Cu in Cu/ZrO catalysts
2
plays an essential role in the selectivity of ethyl acetate in
5
50 °C at a ramp rate of 2 °C/min and a dwell time of 4 h to
the dehydrogenation of ethanol [15]. The Cu content in CuO/
obtain the CuO/ZrO precursor.
2
ZrO and CuO/SiO during programmed temperature reduc-
2
2
CuO/ZrO precursor powder (10 g) was added to 100 ml
2
tion were studied by X-ray absorption near edge structure
of a 2 M NaOH solution in a round-bottom ꢃask, and then
heated to 50 °C under mechanical stirring. Subsequently,
0.75–15 mL hydrazine hydrate was added slowly into
(
XANES), which showed that the copper in CuO under-
2
+
+
0
went reduction via Cu → Cu → Cu under an H ꢃow.
2
0
When the Cu content in the catalysts is high, Cu is the
1
3

Study on the Structure of Cu/ZrO Catalyst and the Formation Mechanism of Disodium…
2
Fig. 1 Catalyst precursor reduction image
the above mixture, which continued to react at 50 °C for
1
0–60 min. Finally, the resulting composite was obtained
after being ꢀltered, washed with deionized water and ethanol
several times, and dried in a vacuum oven at 40 °C for 12 h.
Figure 1 is the reduction diagram of catalyst precursor.
Fig. 2 XRD patterns of reduced Cu/ZrO catalysts (X in CZ-X rep-
2
.2 Catalytic Test
2
resents the volume of hydrazine hydrate used in the reduction of the
precursors)
The dehydrogenation reaction of DEA was carried out in
a stainless steel autoclave. The initial reagents, including
3
1.5 g DEA, 6 g Cu/ZrO catalyst, 25.2 g NaOH (dissolved
2
participate in transforming ZrO from a monoclinic to a
2
in 64.8 g deionized water), were added to the autoclave.
tetragonal phase. The CuO/ZrO precursor exhibits CuO
2
After the autoclave was purged with N , it was pressurized
2
0
diꢁraction peaks without Cu characteristic peaks. As the
to 0.5 MPa, heated to 160 °C at a rate of 4 °C/min at a
speed of 400 rpm, and the volume of gas generated during
the reaction was measured by a rotor ꢃow meter. When the
pressure reached 1.5 MPa, the exhaust valve was opened to
release the gas until the pressure decreased to 1 MPa, and the
temperature and volume of each exhaust was recorded. The
pressure increased by no more than 0.5 MPa within 20 min,
which indicated that the reaction was complete.
amount of hydrazine hydrate increased, the main CuO dif-
fraction peaks (2θ=35.5°, 38.7°) in Cu/ZrO weakened, and
2
0
characteristic Cu peaks appeared at 2θ=43.3°. When the
amount of hydrazine hydrate increased to 12 mL, the main
CuO diꢁraction peak (2θ = 35.5°) disappeared, indicating
that a higher amount of hydrazine hydrate led to a higher
+
reduction degree of Cu/ZrO . The Cu diꢁraction peaks
2
were not detected in the XRD spectra, possibly because the
+
Cu content was lower than the XRD detection limit [27].
The catalyst morphology was characterized by scan-
ning electron microscopy (SEM) and transmission electron
microscopy (TEM). Figure 3a, b shows the SEM images
3
Results and Discussion
3.1 Characterization and Analysis of Cu/ZrO2
of the unreduced CuO/ZrO catalyst precursor, in which
2
Catalysts
needle-like CuO is inserted into an irregular bulk ZrO sup-
2
port. Figure 3c, d are the SEM images of the Cu/ZrO cata-
2
Figure 2 shows the X-ray diꢁraction (XRD) patterns of
pure ZrO supports, the CuO/ZrO precursor, and the Cu/
lyst after hydrazine hydrate reduction, and Fig. 3e, f are the
corresponding TEM images. Rod-shaped CuO was reduced
to spherical Cu particles using hydrazine hydrate. Cu nano-
particles with a particle size of about 15 nm were uniformly
2
2
ZrO catalysts reduced using diꢁerent amounts of hydrazine
2
hydrate. In this ꢀgure, X in Cu/ZrO -X represents the addi-
2
tion of X ml of 80 wt% hydrazine hydrate aqueous solution.
distributed on the surface of the irregular bulk ZrO support,
2
The preparation method is described in Sect. 2.1, using a
indicating that hydrazine hydrate reduced CuO to Cu to suc-
reduction time of 60 min. ZrO was prepared through the
cessfully prepare the Cu/ZrO catalyst.
2
2
same procedure for the CuO/ZrO precursor, but without the
To study the surface composition of the catalysts, Cu/
2
addition of copper salt solution.
ZrO catalyst samples were characterized by Fourier-trans-
2
As shown in Fig. 2, pure ZrO exhibits diꢁraction peaks
form infrared spectroscopy (FTIR). Figure 4 is the FTIR
2
−
1
which correspond to tetragonal (t−) and monoclinic (m−)
image of the sample. The band at 3419 cm was attributed
to the –OH stretching vibration of adsorbed water molecules,
ZrO crystal phases. However, the CuO/ZrO precursor
2
2
−
1
only contains t-ZrO , indicating that the zirconium salt and
the band at 1624 cm was attributed to the –OH bending
vibration of free water molecules, whereas the small bands
2
copper salt solutions in the above co-precipitation reaction
1
3

Y. Wang et al.
Fig. 3 SEM images of unreduced CuO/ZrO (a, b), Cu–ZrO (c, d); and TEM images of Cu–ZrO (e, f)
2
2
2
(
CZ-3) hydrazine hydrate are shown in Fig. 5a and b, respec-
tively. The preparation method was as described in Sect. 2.1,
and the reduction time was 60 min.
2-
-OH
-CO3
CZ-0.75
CZ-1
All Cu 2p spectra contain a large satellite peak near
-
OH
2+
9
40–944 eV, which suggests there is a large amount of Cu
on the surface of the catalysts [29–31]. Depending on the eꢁec-
tive symmetry, copper ions may transition to 3d. Therefore,
a complex satellite structure composed of at least two peaks
was clearly visible in the Cu 2p spectrum may be due to the
distortion of the octahedral structure of the copper compound
CZ-3
0
+
[
28]. Peaks at 932.2 eV were assigned to Cu and Cu , and
cannot be eꢁectively distinguished from the XPS Cu 2p peak
since they are so close to each other. Additionally, character-
2+
4
000 3500 3000 2500 2000 1500 1000 500
istic Cu peaks appear at 934.4 eV. As the amount of hydra-
-1
0
+
Wavenumber(cm )
zine hydrate was increased, the peak intensity of Cu /Cu
increased, indicating that the reduction degree of the catalysts
also increased, which is consistent with the XRD spectra.
Figure 5b shows the Auger electron spectra of the Cu/
Fig. 4 FTIR spectra of the Cu/ZrO catalysts
2
ZrO catalysts. The three peaks with Auger electron kinetic
2
at 1450 and 853 cm−1 were due to the various vibrational
modes of carbonate. In addition, the stretching vibration of
energies of 916.3 eV, 917.4 eV, and 918.4 eV correspond
+
2+
0
to Cu , Cu , and Cu , respectively [29]. Table 1 shows the
surface copper components based on Cu LMM deconvolu-
tion and surface atomic ratios of Zr and O versus Cu based
on XPS analysis. It can be concluded that increasing the
2
−
the CO group in the CZ-1 sample was the most intense,
3
indicating a large amount of CO had adsorbed on the Lewis
2
basic sites of CZ-1 to form a bicarbonate.
2
+
+
The surface composition and oxidation state of the cata-
lyst components were investigated by XPS. The XPS spectra
amount of hydrazine hydrate lowered the amount of Cu
0
on the catalyst surface, while the proportion of Cu and Cu
of Cu 2p and Auger electron spectra of Cu/ZrO samples
increased. Combined with the catalyst activity test, when
2
reduced by 0.75 mL (CZ-0.75), 1 mL (CZ-1), and 3 mL
1
3

Study on the Structure of Cu/ZrO Catalyst and the Formation Mechanism of Disodium…
2
b
a
Cu /Cu0 Cu
+
2+
Cu 2p
Cu Auger
+
2+
Cu
Cusa
2+
Cu
0
CZ-0.75
Cu
CZ-0.75
CZ-1
2+
+
Cu
Cu
0
Cu
CZ-1
CZ-3
2
+
+
Cu
Cu
0
Cu
CZ-3
932
936
940
944
905
910
915
BE(eV)
920
925
BE(eV)
Fig. 5 XPS spectra of Cu 2p (a), and Auger electron spectra (b) of reduced Cu/ZrO samples
2
Table 1 Surface copper components based on Cu LMM deconvolu-
tion and surface atomic ratios of Zr and O versus Cu based on XPS
analysis
1 ml of hydrazine hydrate was used, the catalyst showed the
0
+
best activity when the ratio of Cu /Cu was 1:0.89.
Figure 6 shows the preparation of Cu/ZrO catalysts with
2
Samples Kinetic energies
Peak area ratioa Atomic content /%
b
1 mL hydrazine hydrate with reduction times of 20 min
(
eV)
(
t-20), 30 min (t-30), and 40 min (t-40). As the reduc-
0
+
2+
0
+
2+
Cu Cu Cu
Cu :Cu :Cu
Cu Zr O
0
+
tion time increased, the peaks representing Cu and Cu
increased ꢀrst and then weakened. According to the Auger
CZ-0.75 918.4 916.0 917.4
1:5.27:0.68
1:0.89:0.27
1:1.44:0.15
1:2.17:0.49
1:0.97:0.15
1:7.49:1.43
24.1 15.2 60.7
28.6 11.8 59.6
26.4 12.3 61.3
28.5 12.1 59.4
24.1 13.9 62.0
23.3 14.8 61.9
0
electron spectra, the proportion of Cu increased ꢀrst and
CZ-1
CZ-3
t-20
918.4 916.0 917.4
918.5 915.8 917.3
918.4 916.0 917.4
918.4 916.0 917.4
918.4 916.0 917.4
2
+
then decreased, while the proportion of Cu decreased ꢀrst
+
0
and then increased. It is presumed that Cu and Cu were
2
+
produced by the reduction of Cu with hydrazine hydrate,
and the reduction degree was the highest at a reduction time
of 30 min. As the reaction continued, the amount of hydra-
t-30
t-40
a
2+
+
0
Peak area ratios between Cu , Cu , and Cu were calculated by the
deconvolution of Cu LMM XAES
+
0
zine hydrate in the solution decreased, and Cu and Cu was
2+
oxidized to form Cu at a moderate temperature of approxi-
b
Atomic ratios of Cl to Cu were obtained by XPS analysis
mately 50 °C. According to the catalyst activity test, when
a
+
0
2+
b
2+
Cu /Cu
Cu
+
Cu
Cu Auger
Cu
Cu 2p
t-20
2+
Cu
0
Cusa
t-20
0
Cu
+
Cu
2+
Cu
t-30
t-40
t-30
t-40
2
+
+
Cu
Cu
0
Cu
932
936
940
944
905
910
915
920
925
BE(eV)
KE(eV)
Fig. 6 XPS spectra of Cu 2p (a), and Auger electron spectra (b) of reduced Cu/ZrO samples
2
1
3

Y. Wang et al.
the hydrazine hydrate reduction time was 30 min, the cata-
10
8
100
+
lyst activity was best when the ratio of Cu0/Cu was 1:0.97.
In order to analyze the Lewis basic sites of the catalyst
8
6
4
2
0
0
0
0
0
surface, the characterization of CO -TPD was performed on
2
samples reduced with diꢁerent amounts of hydrazine hydrate
(
Fig. 7). All three catalyst samples exhibited a CO desorp-
2
6
tion peak at 99 °C, which corresponded to the weakly basic
sites [32]. In addition to the weakly basic sites, the desorp-
tion peak at 260 °C can be assigned to medium-strength
Lewis basic sites (MSLB) [33, 34]. In the CZ-1 sample, the
MSLB dominated the sample, which indicates that the use
of diꢁerent amounts of hydrazine hydrate can change the
amount of Lewis basic sites on the catalyst. Combined with
the catalyst performance test, it can be seen an increase in
the MSLB improved the eꢂciency of the dehydrogenation
of diethanolamine.
4
2
CZ-0.75 CZ-1 CZ-1.5 CZ-3
Catalyst
CZ-6 CZ-12
Fig. 8 The relationship between the reducing agent used to prepare
the catalysts and the IDA yield and the reaction time (X in CZ-X rep-
resents the volume of hydrazine hydrate used in catalyst reduction)
3.2 Catalytic Activity
Figure 8 shows that the higher the yield of IDA, the shorter
the required reaction time. As the reduction degree of the
catalysts increased, the IDA yield ꢀrst increased and then
decreased, reaching a maximum yield of 92.7% when using
diꢁerent, as is the catalytic eꢁect. Combined with Fig. 9, it
can be seen that the sample reduced by hydrazine hydrate for
+
0
30 min had the highest Cu /Cu , the shortest reaction time,
and the highest IDA yield.
1
mL of reductant.
The ester preparation process by the dehydrogenation of
fatty alcohol occurred in two steps: alcohol dehydrogena-
tion to produce an aldehyde, and ester production from the
reaction between an aldehyde and an alcohol [16, 17]. It
can be seen from Figs. 5 and 6 that as the amount of hydra-
The relationship between the IDA yield and reaction time
and the reduction time of the catalysts is shown in Fig. 9.
The preparation method is as described in Sect. 2.1, and 1
mL of hydrazine hydrate was used.
0
+
The IDA yield ꢀrst increased and then decreased with
the catalyst reduction time, and the maximum yield of IDA
zine hydrate increased, the Cu and Cu content gradually
increased.
(
98.1%) was obtained at t-30. The XPS analysis of the cat-
Figure 10 shows the relationship between the amount
of hydrazine hydrate used to prepare the catalysts and the
reaction temperature at the ꢀrst exhaust. The lowest tem-
perature of the ꢀrst exhaust was recorded at CZ-1, and then
the initial exhaust temperature rose, and then decreased as
+
0
alysts showed that the ratio of Cu /Cu had an important
eꢁect on the reaction yield. Figure 6c shows that when dif-
+
0
ferent reduction times are used, the ratio of Cu /Cu is also
CZ-0.75
CZ-1
CZ-3
3.6
3.4
3.2
3.0
2.8
2.6
2.4
2.2
98
96
94
92
t-20
t-30
t-40
t-50
t-60
Catalyst
1
00
200
300
400
C)
500
600
Temperature(
°
Fig. 9 The relationship between the reduction time of the prepared
catalyst and the reaction time and yield of IDA (X in t-X represents
the reduction time of the prepared catalysts)
Fig. 7 CO -TPD proꢀles for the samples
2
1
3

Study on the Structure of Cu/ZrO Catalyst and the Formation Mechanism of Disodium…
2
more hydrazine hydrate was added. When more than 3 mL of
hydrazine hydrate was added, the initial temperature of the
exhaust gas rapidly decreased, but the overall reaction time
was longer. When 1 mL of hydrazine hydrate was used, the
initial exhaust temperature was the lowest, and the reaction
dehydrogenation reaction is endothermic, while the reaction
between the ester and NaOH is exothermic. When the rate
of the entire reaction is very fast, the heat provided by the
autoclave is not suꢂcient to drive the alcohol dehydrogena-
tion reaction, and the temperature rapidly decreases. The
ester formation rate simultaneously increased, and NaOH
reacted with it immediately to release a signiꢀcant amount
of heat which increased the temperature from its set value of
160 °C to almost 190 °C. These processes may be the cause
of the sudden temperature increase.
0
+
rate was the fastest, as shown in Fig. 6. The Cu and Cu
content in the catalysts was not the highest compared with
0
+
other catalysts, indicating that the Cu /Cu pair reached the
optimum ratio when 1 mL hydrazine hydrate was used.
0
The results of the above reactions show that Cu catalyzes
the ꢀrst step in the DEA dehydrogenation reaction to form
+
an intermediate aldehyde, and Cu has excellent catalytic
3.3 Formation Mechanism of Disodium
Iminodiacetate and Sodium Glycine
activity for the reaction of the intermediate aldehyde. As the
0
Cu content increased, the temperature of the ꢀrst exhaust
gas decreased, the speed of the early exhaust accelerated,
but the rate of the subsequent reaction quickly decreased. It
3.3.1 Formation Mechanism of Disodium Iminodiacetate
0
should be noted that Cu does not catalyze the entire reaction
In the dehydrogenation of homogeneous alcohols, organic
bases or other basic compounds are often used as additives
to accelerate the deprotonation of the hydroxyl group to
achieve high yields of the target product. Similarly, in het-
erogeneous catalytic reactions, basic compounds improve
the catalytic activity of alcohol transfer dehydrogenation.
In the Cu–ZrO catalyst, the ZrO carrier has a Lewis basic
to prepare disodium iminodiacetate from DEA as described
0
in the literature [5]. Increasing the Cu content accelerated
the initial reaction rate and decreased the temperature of
the primary exhaust. However, as the reaction proceeded,
the reaction rate decreased since the intermediate aldehyde
0
could not be transformed via Cu catalysis. When 1 mL of
2
2
0
+
hydrazine hydrate was used, the Cu /Cu pair in the catalysts
site, and the Cu–ZrO catalyst reduced by 1 ml of hydrazine
2
+
reached the optimum ratio, and Cu catalyzed the intermedi-
hydrate had the most Lewis basic sites and exhibited better
ate aldehyde generated in the ꢀrst step to produce disodium
iminodiacetate. There was no accumulation of intermediate
aldehyde, and the ꢀrst step of the reaction was rapid, which
decreased the overall reaction time.
diethanolamine dehydrogenation performance. Therefore,
the transfer eꢂciency of diethanolamine in Cu–ZrO catalyst
2
is related to its surface Lewis basic sites.
The mechanism for the preparation of disodium imino-
diacetate from DEA has rarely been reported. Similar to the
reaction of monoethanolamine during the preparation of
glycine [35], it is generally believed that DEA is dehydro-
genated to produce aldehyde intermediates using Cu-based
catalysts. Then, two aldehyde molecules form sodium imi-
nodiacetate in an alkaline medium [8]. However, the active
During the initial stage of diethanolamine dehydrogena-
tion using CZ-1 catalyst, the temperature inside the kettle
rose rapidly and reached a peak at 230 °C and then rapidly
dropped to 120 °C. In contrast, the temperature was con-
stant at 165 °C when using CZ-6 and CZ-12. The alcohol
center of Cu/ZrO catalysts and the ester formation process
2
160
155
150
145
140
135
130
125
120
115
from aldehydes are unclear.
The Cu/ZrO catalyzed dehydrogenation of DEA occurs
2
in two steps, similar to the mechanism of fatty alcohol
dehydrogenation over Cu-based catalysts. The hydroxyl
functional groups on the DEA are ꢀrst removed via nucleo-
philic attack of the Lewis basic sites at the ZrO interface
2
to form an alkoxide intermediate, and an alkaline solution
can accelerate the rate of deprotonation [36]. Subsequently,
0
the adjacent Cu catalyzes the formation of the intermediate
aldehyde by α-H cleavage [20, 37, 38].
The reaction process to prepare an ester from aldehydes
has two possible reaction routes: (VI) the Cannizzaro reac-
tion between two aldehydes, and (VII) the nucleophilic addi-
tion of aldehydes and alcohols:
CZ-0.75 CZ-1 CZ-1.5 CZ-3
Catalyst
CZ-6 CZ-12
Fig. 10 Relationship between the amount of reducing agent used to
prepare the catalysts and temperature at the ꢀrst exhaust (X in CZ-X
represents the volume of hydrazine hydrate used in catalyst reduction)
R − CH CHO → R − CH COOCH CH − R
2
2
2
2
(VI)
1
3

Y. Wang et al.
addition of alcohols and aldehydes [15]. The formation
mechanism of disodium iminodiacetate is shown in Fig. 11.
R − CH CHO + R − CH CH OH → R − CH COOCH CH − R
2
2
2
2
2
2
(VII)
In the ꢀrst route, the Cannizzaro reaction cannot produce
the corresponding ester [39]. When the aldehyde in the reac-
tants contains an α-H, it is extracted by the base, and aldol
condensation occurs [16], which prevents the synthesis of
disodium iminodiacetate by DEA dehydrogenation from fol-
lowing this route. In the second route, the α-H on the alde-
hyde of the intermediate is removed by the alkali, causing
the electrophilic α-C to react with an alkoxide to form an
ester [40]. In an alkaline environment, the ester decomposes
to form disodium iminodiacetate and DEA. Compared with
3
.3.2 Formation Mechanism of Sodium Glycine
Sodium glycine is the main by-product in the preparation of
disodium iminodiacetate by DEA dehydrogenation, and its
formation mechanism has not been reported. Gas chroma-
tography analysis revealed that the gas produced during the
reaction contained acetaldehyde. Combined with high per-
formance liquid chromatography to analyze crude products,
the formation mechanism of sodium glycine is divided into
the acetaldehyde route and glycolaldehyde route.
0
+
Cu , Cu has better catalytic activity for the nucleophilic
3
.3.2.1 Acetaldehyde Route Acetaldehyde was identiꢀed
by gas chromatography during the reaction, and the mecha-
nism of the main by-product sodium glycine was speculated.
The Cu/ZrO used to catalyze the dehydrogenation of DEA
2
and the ZrO support contain weakly acidic and basic sites,
2
respectively [21, 41]. Cu/ZrO catalyzes the dehydrogena-
2
tion of alcohol (I) to form oleꢀns, followed by the forma-
tion of a Schiꢁ base via enamine-imide tautomerism (II).
As the reaction proceeds, the basicity of the reaction system
gradually weakens. At high temperatures, the Schiꢁ base
produces acetaldehyde and sodium glycinate by hydrolysis
(III). The formation mechanism is shown in Fig. 12.
3
.3.2.2 Glycolaldehyde Route The dehydrogenation (IV) of
DEA using Cu/ZrO catalysts produces intermediate alde-
2
hydes and forms a Schiꢁ base by keto-enol tautomerism
(
V) and an enamine-imine tautomer (VI). The Schiꢁ base
produces the main glycine and glycolaldehyde byproducts
through a hydrolysis reaction (VII). Due to the catalytic
Fig. 11 Formation mechanism of disodium iminodiacetate
activity of Cu/ZrO towards hydroxyl and aldehyde groups,
2
Fig. 12 The formation mechanism of sodium glycine
1
3

Study on the Structure of Cu/ZrO Catalyst and the Formation Mechanism of Disodium…
2
glycolaldehyde is oxidized to form the glycolic acid and
oxalic acid intermediate byproducts [6, 42]. The formation
mechanism is shown in Fig. 12.
2. Chen D, Li J, Li GR, Chen BH, Yin FX (2013) Chem Ind Eng
Prog 32:1635–1640
3
.
Andreev DV, Sergeev EE, Gribovskii AG, Makarshin LL,
Prikhod’ko SA, Adonin NY, Pai ZP, Parmon VN (2017) Chem
Eng J 330:899–905
4
5
6
7
.
.
.
.
Zheng XJ, Yang GW, Xu XX, Yang G (2001) Fine Chem
1
8:608–610
Wang Y, Zhao Y, Zhao Z, Lan X, Xu J, Xu W, Duan Z (2019) Acta
Chim Sin 71:661–668
4
Conclusion
Hickman DA, Mosner K, Ringer JW (2015) Chem Eng J
278:447–453
In this paper, Cu/ZrO catalysts were rapidly prepared by
2
Andreev DV, Gribovskii AG, Makarshin LL, Adonin NY,
Prikhod’ko SA, Pai ZP, Parmon VN (2013) Catal Ind 5:1–8
hydrazine hydrate reduction at a moderate temperature from
CuO/ZrO precursors. The catalysts could also be used in
2
8. Yang AS, Pan YF, Sun Q, Cheng R, Zheng YP, Xu GM (2010) J
Chem Eng Chin Univ 4:590–595
the oxidative dehydrogenation of diethanolamine to pre-
pare disodium iminodiacetic acid. Adding more hydrazine
9. Lu Z, Gao D, Yin H, Wang A, Liu S (2015) J Ind Eng Chem
31:301–308
0
hydrate in the preparation increased the Cu content in Cu/
1
0. Nagaiah P, Venkat Rao M, Thirupathaiah K, Venkateshwarlu
V, David Raju B, Rama Rao KS (2018) Res Chem Intermed
44:5817–5831
ZrO , but the yield of sodium iminodiacetate synthesized
2
0
via diethanolamine dehydrogenation increased with the Cu
0
+
content in the catalysts. When the Cu /Cu ratio is approxi-
mately 1:0.9, the highest catalyst activity was observed. The
11. Yin H, Yin H, Wang A, Shen L, Liu Y, Zheng Y (2017) J Nanosci
Nanotechnol 17:1255–1266
1
2. Sun D, Misu T, Yamada Y, Sato S (2019) Appl Catal A. https://
doi.org/10.1016/j.apcata.2019.06.007
0
abovementioned results indicate that Cu was the catalytic
active center when diethanolamine dehydrogenation was
used to obtain disodium iminodiacetic acid, which con-
ꢃicts with previous literature. XPS and XRD were used to
13. Ohira M, Liu H, He D, Hirata Y, Sano M, Suzuki T, Miyake T
(2018) J Jpn Petrol Inst 61:205–212
1
4. Gao D, Yin H, Wang A, Shen L, Liu S (2015) J Ind Eng Chem
2
6:322–332
analyze the structure and reaction results of the Cu/ZrO
2
15. Wang LX, Zhu WC, Zheng DF, Yu X, Cui J, Jia MJ, Zhang WX,
Wang ZL (2010) React Kinet Mech Catal 101:365–375
0
catalysts, and showed that Cu played a key role in the ꢀrst
step of the dehydrogenation of diethanolamine to form inter-
16. Inui K, Kurabayashi T, Sato S (2002) J Catal 212:207–215
17. Inui K, Kurabayashi T, Sato S (2002) Appl Catal A 237:53–61
+
mediate aldehydes. Meanwhile, Cu was also essential for
1
8. Ro I, Liu YF, Ball MR, Jackson DHK, Chada JP, Sener C, Kuech
TF, Madon RJ, Huber GW, Dumesic JA (2016) ACS Catalysis
6:7040–7050
transforming intermediate aldehydes into disodium iminodi-
acetic acid. The formation mechanism of glycine, the main
by-product of diethanolamine dehydrogenation, was also
proposed for the ꢀrst time. By analyzing the presence of
acetaldehyde in the gases produced by the reaction, it was
surmised that intermediate aldehydes or alkenes underwent
keto-enol tautomerism and imine-enamine tautomerism, to
generate a Schiꢁ base, followed by hydrolysis to produce the
main by-products sodium glycinate, sodium glycolate, and
sodium silicate. This mechanism is proposed to improve the
yield of the main product, iminodiacetic acid, and guide the
optimization of the process conditions.
19. Zonetti PC, Celnik J, Letichevsky S, Gaspar AB, Appel LG (2011)
J Mol Catal A 334:29–34
2
0. Sato AG, Volanti DP, Freitas IC, Longo El, Bueno JC (2012) Catal
Commun 26:122–126
21. Bai GY, Wang YL, Li F, Zhao Z, Chen GF, Li N, Han X (2012)
Catal Lett 143:101–107
2
2. Freitas IC, Damyanova S, Oliveira DC, Marques CMP, Bueno
JMC (2014) J Mol Catal A 381:26–37
2
3. Chen CQ, Ruan CX, Zhan YY, Lin XY, Zheng Q, Wei KM (2014)
Int J Hydrogen Energy 39:317–324
24. Ji DH, Zhu WC, Wang ZL, Wang GJ (2007) Catal Commun
8
:1891–1895
2
5. Hu Q, Fan GL, Yan L, Cao XZ, Zhang P, Wang BY, Li F (2016)
Green Chem 18:2317–2322
Acknowledgements Support from the National Natural Science
Foundation of China (Grant No. NSFC 21576229) is gratefully
acknowledged.
2
2
6. Shi QJ, Liu N, Liang Y (2007) Chin J Catal 28:57–61
7. Komandur VR, Guggilla VS, Chakravarthula SS, Vattikonda VR
(
2007) J Phys Chem B 11:543–550
2
8. Kim S (1974) J Electron Spectrosc Relat Phenom 3:217–226
Compliance with Ethical Standards
29. Severino F, Brito JL, Laine J, Fierro JL, L´opez Agudo A (1998)
J Catal 177:82–95
3
3
3
0. Acharyya SS, Ghosh S, Bal R (2014) ACS Sustain Chem Eng
Conflict of interest The authors declare that they have no conꢃicts of
2
:584–589
interest.
1. Maiti S, Llorca J, Dominguez M, Colussi S, Trovarelli A, Priolkar
KR, Aquilanti G, Gayen A (2016) J Power Sources 304:319–331
2. Unnikrishnan P, Srinivas D (2012) Ind Eng Chem Res
51:6356–6363
References
3
3
3. Hu Q, Yang L, Fan GL, Li F (2016) Chem Nano Mat 2:888–896
4. Wang J, Lei Z, Qin H, Zhang L, Li F (2011) Ind Eng Chem Res
1
.
Alibhai MF, Stallings WC (2001) Proc Natl Acad Sci USA
8:2944–2946
5
0:7120–7128
9
3
5. Yang YC, Duan ZK, Liu WY, Li GL, Xiong Y (2001) Chem React
Eng Technol 17:210–215
1
3

Y. Wang et al.
3
3
3
3
4
6. Balaraman E, Khaskin E, Leitus G, Milstein D (2013) Nat Chem
41. Manrı
́
quez ME, López T, Gómez R, Navarrete J (2004) J Mol
5
:122–125
Catal A 220:229–237
7. Zhang M, Zhao YJ, Liu Q, Yang L, Fan GL, Li F (2016) Dalton
Trans 45:1093–1102
42. Jiang ZW, Zhang ZR, Song JL, Meng QL, Zhou HC, He ZH, Han
BX (2016) ACS Sustain Chem Eng 4:305–311
8. Wang ZY, Liu XY, Rooney DW, Hu P (2015) Surf Sci
6
40:181–189
Publisher’s Note Springer Nature remains neutral with regard to
9. Takeshita K, Nakamura S, Kawamoto K (1978) Bull Chem Soc
Jpn 51:2622–2627
jurisdictional claims in published maps and institutional aꢂliations.
0. Neurock M, Tao ZY, Chemburkar A, Hibbitts DD, Iglesia E
(
2017) Faraday Discuss 197:59–86
Aꢀliations
Yongsheng Wang · Hongwen Zhu · Zhengkang Duan · Zhenzhen Zhao1,2 · Yunlu Zhao1,2 · Xiaolin Lan1 ·
1
,2
1
1,2
1
1
Li Chen · Dongjie Guo
Yongsheng Wang
Li Chen
8
71585227@qq.com
1062785680@qq.com
Hongwen Zhu
Dongjie Guo
zhw3255805@qq.com
1457555605@qq.com
Zhenzhen Zhao
1
2
College of Chemical Engineering, Xiangtan University,
Xiangtan 411100, China
1
098808414@qq.com
Yunlu Zhao
820955484@qq.com
Hunan Collaborative Innovation Center of New Chemical
Technologies for Environmental Benignity and Eꢂcient
Resource Utilization, Xiangtan 411105, Hunan, China
1
Xiaolin Lan
107300386@qq.com
1
1
3