Hollow ni-p amorphous alloy nanospheres: An efficient catalyst for sugars hydrogenation to polyols

Article abstract of DOI:10.1016/j.cattod.2020.04.031

In this paper, hollow Ni ? P nanospheres (NSs) are prepared through Ni electroless plating on the Au-activated silica NSs externally covered by aminopropyl moieties, followed by removing the silica template with sodium hydroxide. With various characterizations, the resulting hollow Ni ? P NSs are identified to be amorphous alloy. During liquid-phase hydrogenation of sugars to sugar alcohols, the hollow Ni ? P amorphous alloy NSs delivered much superior catalytic performances to the commercial Raney Ni catalyst, showing a good potential in practical applications. Of particular interest is the unique hollow chamber structure of the hollow Ni ? P amorphous alloy NSs, which allows for improving catalytic activity and durability relative to those associated with the dense Ni ? P amorphous alloy NSs. This work demonstrated that such hollow Ni materials with nanoporous chamber structure displayed advantages such as easy experimental handling and high accessibility for the reactants in liquid-phase reaction, more Ni active sites, as well as the existence of more electron-enriched inner surface, which is essential to provide highly efficient catalysts for some reactions.

Full text of DOI:10.1016/j.cattod.2020.04.031

Journal Pre-proof
Hollow ni-p amorphous alloy nanospheres: An efficient catalyst for
sugars hydrogenation to polyols
Yong Yang, Huajun Gu, Qingxiao Zhang, Fang Zhang, Hui Li
PII:
S0920-5861(20)30225-X
DOI:
https://doi.org/10.1016/j.cattod.2020.04.031
CATTOD 12810
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
10 January 2020
19 March 2020
8 April 2020
Please cite this article as: { doi: https://doi.org/
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Hollow Ni-P amorphous alloy nanospheres: an
efficient catalyst for sugars hydrogenation to polyols
Yong Yang1, Huajun Gu1, Qingxiao Zhang, Fang Zhang*, Hui Li
Education Ministry Key Lab of Resource Chemistry and Shanghai Key Laboratory of Rare Earth
Functional Materials, Shanghai Normal University, Shanghai 200234, PR China
* Corresponding author. Tel: +86 21 64323578; Fax: +86 21 64322272.
E-mail address: lihui@shnu.edu.cn (H. Li) zhangfang@shnu.edu.cn (F. Zhang)
1 These authors contributed equally to this work.
Graphical abstract
1

Highlights
 ► Hollow Ni−P amorphous alloy nanospheres were prepared
by electroless plating method.
 ► Such hollow catalyst showed a good potential in
hydrogenation of sugars to polyols.
 ► Such hollow catalyst has the advantages of easy handling
and accessibility.
 ► Such hollow catalyst provides more active sites for catalytic
reaction.
 ► Such hollow catalyst possesses more electron-enriched
inner surface.
Abstract
In this paper, hollow Ni−P nanospheres (NSs) are prepared through Ni
electroless plating on the Au-activated silica NSs externally covered by
aminopropyl moieties, followed by removing the silica template with sodium
hydroxide. With various characterizations, the resulting hollow Ni−P NSs are
identified to be amorphous alloy. During liquid-phase hydrogenation of sugars
to sugar alcohols, the hollow Ni−P amorphous alloy NSs delivered much
superior catalytic performances to the commercial Raney Ni catalyst, showing
a good potential in practical applications. Of particular interest is the unique
2

hollow chamber structure of the hollow Ni−P amorphous alloy NSs, which
allows for improving catalytic activity and durability relative to those associated
with the dense Ni−P amorphous alloy NSs. This work demonstrated that such
hollow Ni materials with nanoporous chamber structure displayed advantages
such as easy experimental handling and high accessibility for the reactants in
liquid-phase reaction, more Ni active sites, as well as the existence of more
electron-enriched inner surface, which is essential to provide highly efficient
catalysts for some reactions.
Keywords: Ni−P amorphous alloy catalyst; Hollow nanospheres; Biomass
transformation; Sugar hydrogenation; Polyols
1. Introduction
The production of fuels, chemicals and materials from renewable and
abundant biomass resources has drawn immense attention in recent years as it
possesses the possibility of using carbon neutral technologies to tackle global
energy and environmental problems [1-5]. In viewpoint of atom economy,
biomass materials are more desirable to be converted into oxygen-containing
chemicals rather than hydrocarbon fuels due to their relatively high oxygen
content [6]. In this respect, polyols derived from biomass resources are
versatile molecules those have found a wide variety of applications [7]. In fact,
3

sugars can be facilely obtained from biomass resources, primarily starch [8-10]
and even cellulose [11,12], through enzymatic hydrolysis bioprocesses.
Therefore, the catalytic hydrogenation of naturally occurring sugars is an
environmentally-friendly route to the production of the corresponding polyols.
Among polyols from biomass, sorbitol, a hexitol, one of the top 12 biobased
building blocks listed by the U.S. Department of Energy [13], is a valuable
platform chemical that can be facilely transformed by straightforward methods
into fuels or chemicals [1,14]. Nowadays, almost all existing sorbitol production
processes are usually based on the hydrogenation of glucose catalyzed by
metallic catalysts (Scheme 1) [15-23]. Currently, Raney Ni is mainly adopted in
this industrial process [24]. However, the process is usually ineffective and may
give rise to problems of environmental pollution resulting from the alkali
leaching of Ni−Al alloy. In addition, the leaching of nickel during hydrogenation
and the low activity make this catalyst less profitable, because a refilling of the
catalyst and a purification of product to remove residual Ni are necessary,
which eventually render this process economically less attractive [25].
Amorphous alloys, a kind of non-equilibrium metastable materials in
thermodynamics with long-range disordered but short-range ordered structure,
have attracted growing attention from both academia and industry because of
their superior catalytic properties in comparison with the crystalline
counterparts [26-28]. In general, M−B and M−P amorphous alloys are the most
thoroughly studied [27]. Compared with M−B, M−P amorphous alloys
4

displayed much higher hydrogenation activity, including sugar hydrogenation
to sugar alcohol [29,30], due to more highly unsaturated metal active sites
promoted by alloying with P [27]. Although Ni−P amorphous alloy has proved
to be a potential alternative to Raney Ni in liquid-phase glucose hydrogenation
to sorbitol [31], there is still space to enhance the reactivity through controlling
over its morphology, which will result in enhanced intrinsic activity, increased
surface-active area, and an improved adsorption model of substances on the
surface of catalysts [28].
In recent years, hollow structures have captured extensive attention owing
to their potential applications in adsorption, microelectronics, photonics, as
well as catalysis [32]. In the domain of catalysis, hollow metallic nanospheres
(NSs) with permeable shell represent a new class of efficient catalysts because
they offer some advantages over their dense counterparts in terms of increased
surface area, low density, easy recovery, self-supporting capacity, cost
reduction, and surface permeability [33-35]. Presently, a series of hollow
amorphous alloy NSs have been fabricated based on soft-templating approach,
in which the “soft” templates (e.g., composite vesicle) are coated with metals by
an interfacial reduction reaction [36-38]. However, this method is synthetically
challenging, because the vigorous and exothermic reduction of metal
precursors usually damages the architectures of the formed surfactant-directed
assembly. Strategies based on the mechanisms of galvanic replacement have
also been developed to fabricate hollow metal structures [39], which are more
5

facile and cost-effective since the templates could act as a reducing agents to
produce metal shells and no additional steps are needed to remove the
templates. Nevertheless, the synthesis of hollow metal NSs is limited mainly to
noble metals or precious metal-containing alloys [40] by this synthesis strategy
and thus their practical applications are limited due to the high cost. Adopting
polystyrene beads as hard template to produce hollow Ni−B amorphous alloy
was first introduce by Qiao’s group, which displays advantages in its specific
topological stability, veracity, predictability and controllability [41]. For
achieving hollow metal structures, however, multistep surface activation are
often necessary due to the weak affinity of the hydrophobic polymer host to
nickel cations, complicating the preparation process. Because of surface
hydrophilicity, silica NSs can act as promising hosts for depositing Ni−P
amorphous alloy. Herein, we designed a novel approach to synthesizing hollow
Ni−P amorphous alloy NSs by a combination of silica NSs templating and
electroless plating (Scheme 2). Silica NSs externally covered by aminopropyl
moieties were first fabricated by co-condensation method. The amine-
functionalized host materials were activated with Au and then coated with Ni−P
amorphous alloy nanoparticles (NPs) through electroless plating. Finally,
hollow Ni−P amorphous alloy NSs were achieved by removing silica template.
Such a Ni catalyst exhibits much superior performances to Raney Ni catalyst
during liquid-phase hydrogenation of sugars to sugar alcohols.
6

2. Experimental section
2.1. Catalyst preparation
Chemicals and reagents. All of the chemicals used in this experiment were
obtained from commercial sources and used without any other treatments.
Tetraethoxysilane (TEOS), aminopropyl triethoxysilane (APTES), chloroauric
acid trihydrate (HAuCl4·3H2O), trisodium citrate dihydrate (Na3C6H5O7·2H2O),
sodium borohydride (NaBH4), nickel sulfate hexahydrate (NiSO4·6H2O),
sodium hypophosphite dihydrate (NaH2PO2·2H2O), sodium acetate
(CH3COONa), nickel chloride (NiCl2) and ethanol were purchased from Aladdin
Industrial Co., Ltd. (Shanghai, China).
Synthesis of amine-functionalized silica NSs. Aminopropyl group grafted
silica NSs were synthesized through a modified Stöber method [42], and
designated as SiO2−NH2. In a typical run of synthesis, TEOS (4.7 mL) were
added in a mixture of ethanol (50 mL), deionized water (16 mL), and
concentrated ammonia aqueous solution (28 wt%, 9.0 mL). After being stirred
at 313 K for 2.5 h, APTES (0.2 mL) was added dropwise. After being stirring for
12 h, the solid product was collected and washed thoroughly with deionized
water, followed by drying under vacuum at 353 K for 8 h.
Synthesis of Ni−P/SiO2−NH2 composites. To generate crystal seeds, gold
colloid NPs were synthesized according to the method reported by Jana et al.
[43]. Briefly, a 150 mL aqueous solution containing 2.5 × 10-4 M HAuCl4 and
2.5 × 10-4 M Na3C6H5O7 was firstly prepared. Next, 4.5 mL of 0.1 M NaBH4
7

solution was added dropwise to the solution at 273 K and was stirred
countinuously until no bubbles released. The as-prepared SiO2−NH2 was then
added into the above solution. After being stirred at 298 K for 12 h, the solid
was centrifuged, washed thorougly with deionized water, and dried under
vacuum at 353 K for 8 h. The Au-modified SiO2−NH2 NSs were added into an
electroless plating bath containing 150 mL of plating solution with the
composition as shown in Table 1. High-purity nitrogen was used as the
sheltering gas. With stirring, electroless plating was carried out to generate
Ni−P/SiO2−NH2 composites under the reation conditions (see Table 1). In this
plating process, NiSO4 was used as the source of metal ions; NaH2PO2 was used
as a reducing agent; Na3C6H5O7 was used as a complexing agent for Ni2+ in
solution; CH3COONa was used as buffering agent for pH control. The resultant
Ni−P/SiO2−NH2 composites were centrifuged and washed with deionized
water for several times.
Synthesis of hollow Ni−P amorphous alloy NSs. The silica template was
removed by refluxing the Ni−P/SiO2−NH2 composites in 10 mL of 2 M NaOH
asolution at 313 K for 30 min, followed by washing thoroughly with deionized
water until the pH reached 7. Finally, the solid was washed with ethanol three
times and kept in ethanol until the time of use. The as-received sample is
denoted as Ni−P-H, where the suffix “-H” means hollow structure.
Reference catalysts. For comparison, the reference Ni−P amorphous alloy
NPs was also prepared by the regular method, i.e., by direct reduction of Ni
8

chloride with sodium hypophosphite [44]. In a typical synthesis, 21.0 mmol of
NiCl2, 47.2 mmol of NaH2PO2, and 36.7 mmol of CH3COONa were dissolved in
100 mL of deionized water in a three-necked flask and maintained at 363 K.
The resulting black precipitate was separated by centrifugation and washed
with deionized water. Finally, the solid was washed with ethanol three times
and kept in ethanol until the time of use. The as-received sample is denoted as
Ni−P-S, where the suffix “-S” means solid structure. Raney Ni was commercially
available from a sorbitol production factory in China and was used without
additional activation.
2.2. Catalyst characterization
The bulk compositions were analyzed by means of inductively coupled
plasma optical emission spectrometry (ICP-OES; Varian VISTA-MPX). The
amorphous structure was determined by X-ray diffraction (XRD; Rigaku
D/Max-RB with Cu Kα radiation). Solid-state nuclear magnetic resonance
(NMR) spectra were collected on a Bruker AV-400 NMR spectrometer. Fourier
transform infrared (FTIR) spectra were recorded on a Nicolet Magna 550
spectrometer by KBr technique. Thermogravimetric analysis and differential
thermal analysis (TG/DTA) were conducted on a Shimadzu DTG-60 TG/DTA
analyzer with a ramping rate of 10 K min−1 in 50 mL min−1 of air flow. The
catalyst shapes and morphologies were observed by both field emission
scanning electron microscopy (FESEM; HITACHI S−4800) and transmission
9

electron microscopy (TEM; JEOL JEM2100). The surface electronic states were
investigated by X-ray photoelectron spectroscopy (XPS; ULVAC-PHI PHI5000
VersaProbe using Al Kα radiation), during which all samples were dried and
pretreated in situ in pure Ar atmosphere to avoid oxidation. N2 adsorption-
desorption isotherms were obtained at 77 K using a Quantachrome NOVA
4000e apparatus. By N2 adsorption, the Brunauer-Emmett-Teller (BET)
surface area (SBET) was calculated by using the multiple-point BET method in
the relative pressure range of P/P0 = 0.05−0.2. The pore size distribution curve
(calculated from the adsorption isotherm) was obtained by the Barrett–
Joyner–Halenda (BJH) model. The active surface area (Sact) was measured by
the hydrogen chemisorption at room temperature, which was performed on a
Micromeritics AutoChem II 2920 instrument using a dynamic pulse method.
The sample was purged under an argon flow (purity of 99.997%, treated with
an Alltech Oxy-Trap column) at 523 K for 2 h. The pretreated sample was cooled
down to room temperature under argon atmosphere, and hydrogen pulses were
injected at 303 K until the calculated areas of consecutive pulses were constant.
According to the hydrogenation chemisorption, Sact of the as-prepared catalyst
was calculated assuming Ni/H = 1 and a Ni surface of 6.5 × 10−20 m2 per Ni atom
[45]. Every sample was measured for three times. The reproducibility of the
results was checked by repeating the measurements for three times on the same
catalyst and was found to be within acceptable limits (<  2%). The hydrogen
10

temperature-programmed desorption (H2-TPD) curve was obtained on the
same instrument in Ar flow at a ramping rate of 20 K min−1.
2.3. Catalytic performances test
The liquid-phase hydrogenation of sugar was performed at 4.0 MPa of H2
pressure and 373 K in a 200-mL stainless steel autoclave with a Teflon tube to
avoid metal contamination, in which 0.5 g of Ni catalyst and a sugar aqueous
solution (10 wt.% in 50 mL H2O) were well mixed. According to the drop of the
H2 pressure in the autoclave within 10 min, both the specific activity (the H2
uptake rate per gram of Ni, RHm, mmol h−1 gNi−1) and the intrinsic activity (the
S
H2 uptake rate per m2 of Ni, RH , mmol h−1 mNi−2) were calculated by using the
ideal gas equation. The reaction mixture was sampled at intervals for product
analysis through a liquid-phase chromatograph (Agilent 1200) equipped with a
carbohydrate column (Shodex, SC1011) and a refractive index detector at 333 K
with water as movable phase at 0.50 mL min−1. Preliminary kinetic studies
revealed that there was a plateau in the dependency of the reaction initial rate
upon the stirring rate above 1000 rpm and that the reaction initial rate varied
linearly with catalyst amount from 0.1 to 1.0 g, indicating that the stirring rate
of 1200 rpm was high enough that the hydrogenation rates were independent
of mass transfer. After cooling to room temperature at the end of the reaction,
the catalyst was separated by centrifugation and washed with deionized water
for further characterizations and applications. In order to determine the
11

catalyst durability, the used catalyst was centrifuged and washed thoroughly
with distilled water after each run of the reaction. Then, the used catalyst was
reused with fresh charge of sugar for subsequent recycle runs under same
reaction conditions.
3. Results and discussion
3.1. Catalyst characterization
As shown in FESEM image (Fig. S1), the as-prepared SiO2−NH2 material is
present in the form of uniform spheres with an average diameter around 230
nm, which is similar to the previously reported silica spheres [42]. The FTIR
spectrum in Fig. S2 reveals that, besides the absorbance band at 1088 cm-1
indicative of Si−O stretching vibration, SiO2−NH2 NSs displays two absorbance
bands at 2928, 2849 and 1495 cm-1 characteristic of asymmetric stretching,
symmetric stretching and bending vibrations of aliphatic C−H bonds in the
NH2-CH2-CH2-CH2− group [46]. Meanwhile, the absorbance band at 1522 cm-
1 corresponding to the bending vibration of N−H bond is also clearly observed,
though the N−H stretching vibration peak at 3300 cm-1 is overlapped by O−H
vibration at 3500 cm-1 [46]. These results demonstrated the successful
incorporation of aminopropyl groups into the silica network, which could be
further confirmed by the solid-state NMR spectra. The 29Si CP MAS NMR
spectrum (Fig. S3) of the as-made SiO2−NH2 NSs shows three well-resolved up-
field resonance peaks for the siloxane corresponding to Q4 (δ = − 110 ppm), Q3
12

(δ = − 100 ppm) and Q2 (δ = − 91 ppm), where Qn = Si(OSi)n(OH)4−n (n = 2−4).
Meanwhile, two down-field organosilane signals were observed at − 65 and −
56 ppm, corresponding to T3 and T2, where Tm = RSi(OSi)m(OH)3−m (m = 1−3),
which demonstrated the incorporation of the organic groups into the silica
framework via forming C−Si bond [46]. Additionally, the TGA/DTA curves (Fig.
S4) reveal that, besides the weight loss before 473 K due to the desorption of
physically adsorbed solvents, SiO2−NH2 NSs displays weight loss between 473
and 873 K, corresponding to the oxidation removal of the organic groups
anchored on SiO2−NH2 NSs.
Due to the growth of highly isotropic structure, the Ni–P sample prepared
via conventional chemical reduction method displayed regular spherical
particles with the average size around 70 nm (Fig. S5). By using the surface-
activated host materials, however, selective deposition of Ni–P can be achieved
throught electroless plating method. The TEM image (Fig. 1a) shows that the
Ni−P/SiO2−NH2 composites synthesized through the present method was
monodisperse and with uniformly spherical NPs with an average particle size
around 250 nm. A close view of these NSs by TEM (Fig. 1b) reveals a compact
Ni−P overlayer with 25 nm thickness was readily formed on the surface of
SiO2−NH2, confirming the deposition of Ni−P on the host material. The TEM
morphology (Fig. 1c) demonstrate that Ni−P-H is present in the form of
uniformly spheres with the average diameter around 250 nm, almost the same
as the diameter of SiO2−NH2 NSs. Meanwhile, TEM observation of Ni−P-H
13

shows clear contrast difference between the center and the edge, indicating the
inner of these spheres is hollow. From the high-magnification TEM image of a
single Ni−P-H nanosphere in Fig. 1d, a shell thickness of 28 nm could be
measured. Moreover, one can observe that the outer shell of Ni−P-H was
comprise of uniform NPs with an averae size of 3 nm. Fig. 2a shows N2
adsorption-desorption isotherms of Ni−P-H sample. Notably, Ni−P-H
exhibited a hysteresis loop at the relative pressure range of 0.4–1.0, which was
assigned to mesopores formed through the pileup of Ni−P NPs. Ni−P-H also
displayed a narrow pore size distribution centered around 3.5 nm (Fig. 2b),
which was consistent with the observation from high-magnification TEM image
(Fig. 1d).
ICP ananlysis (Table 1)revealed that the bulk atom composition of Ni−P-H
(Ni79P21) was almost exactly the same as that of the conventionally-prepared
Ni−P-S (Ni80P20), suggesting the different preparation methods had no
significant influence on the P contents in the final samples. Generally, the P
content has great effect on the structural properties of Ni−P sample. More
specifically, the Ni−P was amorphous when the P content is higher than 10
mol%, while lower P content resulted in a mixture of amorphous Ni−P and face-
centered cubic Ni [27]. The SAED image presented in Fig. 1e shows a successive
diffraction halo, suggesting that the Ni−P-H synthesized via the present
method was present in a typical amorphous alloy structure [27]. The
amorphous alloy structure of Ni−P-H can be further confirmed by the XRD
14

pattern (Fig. 3a), where only one broad peak centered at 2θ = 44° was observed
[27]. Heat treatment of the fresh Ni−P-H sample (at 673 K in N2 for 2 h)
resulted in the appearance of several diffraction peaks corresponding to
metallic Ni and the crystalline Ni−P alloy (Fig. 3b). The appearance of Ni−P
crystalline phases during the crystallization process verified the formation of an
alloy between Ni and P for the as-synthesized Ni−P-H. We infer that except for
the crystallization, partial decomposition of a Ni−P amorphous alloy may occur
in this process. Similarly, considering the combined results from both the SAED
(insert in Fig. S5) and the XRD (Fig. S6), Ni−P-S was identified to be
amorphous alloy.
Fig. 4 shows the XPS spectrum of Ni−P-H sample. Only one peak around a
BE at ca. 853.1 eV was observed in the Ni 2p3/2 level (Fig. 4a). In comparison
with the standard pure nickel, no significant change in the BE of the Ni 2p3/2
level was observed, indicating that almost all of the nickel species in Ni−P-H
were present as metallic state. This observation is in agreement with our
previous results [27]. But, the phosphorus species in Ni−P-H sample were
present in both the oxidized state and elemental state, corresponding to BE of
around 133.6 and 130.0 eV (Fig. 4b), respectively, in P 2p level. In comparison
with Ni−P-H, Ni−P-S reveals nearly the same XPS spectrum (Fig. S7). The
surface composition was determined by analysis of relative peak areas in the
XPS spectra. According to the calculation based on XPS peak areas, the surface
atomic ratio of the alloying P to the metallic Ni in Ni−P-H is 16/84, similar to
15

that in Ni−P-S (17/83). This means the nature of Ni active sites in both the
Ni−P-H and the Ni−P-S is identical, regardless of the different preparation
methods.
3.2. Catalytic performances
Initially, the liquid-phase hydrogenation of glucose to sorbitol under 4.0
MPa of hydrogen and 373 K is used for evaluating the performances of the as-
synthesized catalysts. The reaction profiles (Fig. 5) reveal that Ni−P-H was
extremely selective to sorbitol (nearly 100% throughout the reaction process).
It should be noted that, all the Ni-based catalysts displayed 100% selectivity
towards sorbitol under the present reaction conditions.
Table 1 summarizes some structural and catalytic parameters of the
different catalysts during liquid-phase glucose hydrogenation. Although the
commercial Raney Ni possessed a much higher Sact than all the Ni-based
catalysts, it still displayed a much lower RHm and glucose conversion within 4 h
S
S
because of its extremely lower intrinsic activity (RH ). The superior RH of the
as-prepared Ni−P-H and Ni−P-S catalysts could be linked to their unique
amorphous alloy structure, which had proven to be favorable for hydrogenation
processes [27]. More specifically, the unique short-range ordering but long-
range disordering amorphous structure endowed Ni−P amorphous alloy with a
stronger synergistic effect between Ni active sites and more highly unsaturated
Ni active sites than the crystalline catalyst, which may promote the adsorption
16

of reactants and favor hydrogenation activity [27]. This could be also supported
by the fact that an abrupt decrease in activity was observed for the Ni−P-H
treated at 673 K for 2 h in N2 flow (Table 1), which was firstly due to the
significant agglomeration of Ni−P NPs at elevated temperatures and, thus,
decreasing metallic area (Sact) and, secondly, the decrease in RHS resulted from
the crystallization of Ni−P amorphous alloy. In addition, the presence of
phosphorus oxide could further enhance the adsorption for glucose molecule
via abstraction of proton, thereby favoring the subsequent hydrogenation of the
reactant.
m
Based on the RH and glucose conversion data presented in Table 1, it is
obvious that the reactivity of Ni−P-H was more active than Ni−P-S. This could
S
be attributed to both the larger Sact and the enhancement of RH . The larger Sact
of Ni−P-H compared to Ni−P-S could be interpreted in terms of the
promotional effect of the hollow structure. Such nanoporous nature of the
spherical chamber allowed reactant molecules to be adsorbed on both the outer
shell and the interior shell of Ni−P-H, which might increase the accessibility
and provide much more Ni active sites for reactant molecules, apparently
favorable for hydrogenation. Apart from the larger specific surface area, Ni−P-
S
H has larger RH compared to its corresponding dense counterpart (Ni−P-S).
Because both the Ni−P-H and the Ni−P-S catalysts possess similar
compositions, amorphous alloy structure, as well as surface electronic
characteristics, the district difference in morphology should be considered to
17

S
be responsible for the diverse RH . The use of nonconvex-structured materials
has been a hot frontier subject in recently years [47]. Noble-metal nanocrystals
with concave surface have displayed great potential to serve as excellent
catalysts due to their unique surface structure [48-52]. Ding’s group inferred
that the negative curvature of the inner surface in Ni-based catalysts provides
more active centers for the reactant molecules and thus, favoring the catalytic
hydrogenation [49]. To get insight into the differnt surface features between
Ni−P-H and Ni−P-S, the two catalysts were subjected to H2-TPD. Apparently,
both the Ni−P-H and the Ni−P-S catalysts display analogous H2-TPD profiles
(Fig. 6), indicative of the presence of chemisorbed hydrogen species with
similar adsorption energies. Compared with the H2-TPD profile of Ni−P-S, the
desorption peak observed from Ni−P-H shifts markedly to a lower temperature,
suggesing that the adsorption strength of hydrogen over the surface of Ni−P-H
decreases. From this, we could deduce the existence of more electron-enriched
surface, maybe due to the deviation from planarity. Generally, the concave
metal surface possesses more active centers, which results into increase in
electron density and weakens the strength of the hydrogen adsorption.
Beenackers et al. confirmed that, upon adsorption of glucose on metals, an
ionized species is generated (i.e. glucose anion), which was susceptible to attack
by hydrogen adsorbed dissociatively on the neighboring metal sites, initiating
the reaction [53]. Meanwhile, Wit et al. showed the promotional effect of proton
abstraction from the anomeric hydroxyl group on the hydrogenation of glucose
18

[54]. The adsorption and hydrogenation of the glucose molecule on the inner
surface of Ni−P-H could be described in the following model schematized in
Scheme 3. Consequently, the Ni active sites in the inner surface of Ni−P-H with
more electron is anticipated to readily abstract proton from glucose. One the
other hand, the higher electron density on Ni active sites might facilitate the
formation of H− species [55], which could promote glucose hydrogenation
activity.
Additionally, the shells of Ni−P-H are nanoporous, which is expected to
enhance the efficiency of mass transport in liquid-phase reaction. To closer look
into the influence on the diffusion of reaction molecules, hydrogenation of
maltose (Scheme 4), a bigger sugar molecule, was performed under the same
reaction conditions as those for glucose hydrogenation. From the results
tabulated in Table 3, one can see that all the Ni−P catalysts displayed 100%
selectivity towards maltitol. Nonetheless, only 75% selectivity was obtained on
the commercial Raney Ni catalyst because of the formation of sorbitol and an
uncertain byproduct in small amount. Sorbitol could be considered as the
hydrogenation product of glucose stemmed from the hydrolysis of maltose [30],
and the uncertain byproduct was possibly formed via the dehydration reaction
between glucose molecules and maltitol molecules [30]. The poor selectivity of
Raney Ni could be attributed to the existence of residual alumina, which served
as acidic sites to catalyze the hydrolysis of maltose and the condensation
between glucose and maltitol [30]. Meanwhile, the commercial Raney Ni
19

m
catalyst displayed much lower RH
and conversion during maltose
hydrogenation compared with those in glucose hydrogenation. This could be
attributed to the presence of micropores in Raney Ni, which is unfavorable for
catalytic activity in liquid-phase reaction owing to the mass transfer. Hardly,
any difference in the catalytic efficiency for glucose hydrogenation and maltose
hydrogenation could be found using Ni−P-S, apparently due to the absence of
mass transfer limitation over Ni−P-S with dense structure. It should be noted
that the run performed using Ni−P-H for maltose hydrogenation delivers
similar catalytic efficiency to that for glucose hydrogenation. Apparently, the
porous shells in Ni−P-H offer a convenient path for the big reactant molecules
crossing into the inner surface for hydrogenation to maltitol.
Industrial application of a heterogenous catalyst greatly relies on its
handling convenience and stability. During the liquid-phase hydrogenation of
glucose, Ni−P-H could be easily separated from the reaction solution via
centrifugation and used repetitively at least for 6 times without significant
decrease of sorbitol yield (Fig. 7a). However, a remarkable loss in acitivity (30%)
was observed for Ni−P-S after six recycles (Fig. 7a). During the liquid-phase
hydrogenation of maltose, likewise, the maltitol yield decreases by 35% over
Ni−P-S after 6 consecutive runs (Fig. 7b), whereas Ni−P-H could be used
repetitively 6 times with slight loss of activity for maltose hydrogenation (9%)
(Fig. 7b). ICP analysis revealed that no leaching of Ni species could be detected
in the reaction mixtures for Ni−P-H and Ni−P-S catalysts during repeated use,
20

implying that both catalysts were stable against the chelating effect of reactant
and product. TEM image of the used catalyst after 6 cycles (Fig. 8a) reveal that
the deactivation of Ni−P-S can be due to a severe agglomeration of Ni−P NPs.
However, the hollow chamber of Ni−P-H is stable, since part of the hollow
sperical structure can be preserved after being used 6 times (Fig. 8b). According
to these observation, the high durability of Ni−P-H can be attributed to its large
diameter of hollow nanospheres, which facilitates separation from solution via
centrifugation and diminishes the loss of catalyst during recycling tests. On the
other hand, its self-supporting capacity is another factor responsible for the
high durability of Ni−P-H, which retards the Ni−P NPs from agglomeration
during the hydrogenation process.
4. Conclusions
In summary, hollow structured Ni−P amorphous alloy nanospheres were
synthesized through Ni electroless plating on aminopropyl-functionalized silica
nanospheres, followed by removing silica template in NaOH solution. In
particular, the present method is more facile and cost-effective. During liquid-
phase hydrogenation of sugars to sugar alcohols, the synthesized hollow Ni−P
amorphous alloy nanospheres delivered much superior catalytic performances
to the dense Ni−P amorphous alloy and Raney Ni catalyst. These results not
only greatly extend the application of amorphous alloy but also highlight the
role of the significant catalysis from the unique concave surface. This strategy
21

can potentially be extended to other hollow metallic nanospheres with different
composition and diameter, providing highly efficient catalysts for some
reactions.
Credit Author Statement
We declare that we have no competing financial interests
or personal relationships that could have appeared to
influence the work reported in this paper.
Acknowledgement
This work was supported by the National Natural Science Foundation of
China (21972093, 21677098), Shanghai government (19SG42, 19520710700
and
18230742500), the Program for Professor of Special Appointment
(Eastern Scholar) at Shanghai Institutions of Higher Learning (TP2016034),
the Ministry of Education of China (PCSIRT_IRT_16R49) and Shanghai
Engineering Research Center of Green Energy Chemical Engineering
(18DZ2254200).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the online
version, at http://dx.doi.org/
References
[1] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411.
22

[2] R. Rinaldi, F. Schüth, Energy Environ. Sci. 2 (2009) 610.
[3] D.A. Simonetti, J.A. Dumesic, Catal. Rev. Sci. Eng. 51 (2009) 441.
[4] J.C. Serrano-Ruiz, J.A. Dumesic, Energy Environ. Sci. 4 (2011) 83.
[5] M. Besson, P. Gallezot, C. Pinel, Chem. Rev. 114 (2014) 1827.
[6] H. Yue, Y. Zhao, X. Ma, J. Gong, Chem. Soc. Rev. 41 (2012) 4218.
[7] X. Liu, X. Wang, S. Yao, Y. Jiang, J. Guan, X. Mu, RSC Adv. 4 (2014) 49501.
[8] J. Kadokawa, Chem. Rev. 111 (2011) 4308.
[9] L. Xu, W. Wei, H.X. Li, H. Li, ACS Catal. 4 (2014) 251.
[10] W. Wei, Y. Zhao, S. Peng, H. Zhang, Y. Bian, H.X. Li, H. Li, ACS Appl. Mater.
Interfaces 6 (2014) 20851.
[11] M. Katz, E.E. Reese, Appl. Microbiol. 12 (1968) 419.
[12] W. Wei, C. Wang, Y. Zhao, S. Peng, H. Zhang, Y. Bian, H.X. Li, X. Zhou, H.
Li, J. Catal. 327 (2015) 78.
[13] T. Werpy, G. Peterson, Top value added chemicals from biomass, vol. 1:
results of screening for candidates from sugars and synthesis gas. US
Department of Energy, Energy efficiency and renewable energy, Battelle,
http://www1.eere.energy.gov/biomass/pdfs/35523.pdf.
[14] P.L. Dhepe, F. Fukuoka, Catal. Surv. Asia 11 (2007) 186.
[15] W.S. Fedor, J. Millar, A.J. Accola, Ind. Eng. Chem. 52 (1960) 282.
[16] W.M. Kruse, L.W. Wright, Carbohydr. Res. 64 (1978) 293.
[172] P. Gallezot, P.J. Cerino, B. Blanc, G. Fleche, P. Fuertes, J. Catal. 146 (1994)
93.
23

[18] S. Kolaric, V. Sunjic, J. Mol. Catal. A 110 (1996) 189.
[19] H. Li, H.X. Li, J.F. Deng, Catal. Today 74 (2002) 53.
[20] B. Kusserow, S. Schimpf, P. Claus, Adv. Synth. Catal. 345 (2003) 289.
[21] S. Schimpf, C. Louis, P. Claus, Appl. Catal. A 356 (2009) 112.
[22] V.N. Sapunov, M.Y. Grigoryev, E.M. Sulman, M.B. Konyaeva, V.G.
Matveeva, J. Phy. Chem. A 117 (2013) 4073.
[23] S. Wang, W. Wei, Y. Zhao, H.X. Li, H. L, Catal. Today 258 (2015) 327.
[24] B.W. Hoffer, E. Crezee, F. Devred, P.R. Mooijman, W.G. Sloof, P.J.
Kooyman, A.D. van Langeveld, F. Kapteijn, J. A. Moulijn, Appl. Catal. A 253
(2003) 437.
[25] I. Toufeili, S. Dziedzic, Food Chem. 47 (1993) 17.
[26] Y. Pei, G.B. Zhou, N. Luan, B.N. Zong, M.H. Qiao, F. Tao, Chem. Soc. Rev.
41 (2012) 8140.
[27] H. Li, W. Wei, Y. Zhao, H.X. Li, Catalysis 27 (2015) 144.
[28] M. Wang, B. Feng, H. Li, H.X. Li, Chem 5 (2019) 805.
[29] H. Li, P. Yang, D. Chu, H.X. Li, Appl. Catal. A 325 (2007) 34.
[30] H. Li, D. Chu, J. Liu, M. Qiao, W. Dai, H.X. Li, Adv. Synth. Catal. 350 (2008)
829.
[31] H.X. Li, W. Wang, J.F. Deng, J. Catal. 191 (2000) 257.
[32] X. Wang, J. Feng, Y. Bai, Q. Zhang, Y. Yin, Chem. Rev. 116 (2016) 10983.
[33] H. Li, J. Liu, S. Xie, M. Qiao, W. Dai, Y. Lu, H.X. Li, Adv. Funct. Mater. 18
(2008) 3235.
24

[34] H. Li, Z. Zhu, J. Liu, S. Xie, H.X. Li, J. Mater. Chem. 20 (2010) 4366.
[35] H. Li, Z. Zhu, H.X. Li, P. Li, X. Zhou, J. Colloid Interf. Sci. 349 (2010) 613.
[36] H. Li, Y. Xu, J. Liu, Q. F. Zhao, H. X. Li, J. Colloid Interface Sci. 334 (2009)
176.
[37] H. Li, D. Zhang, G. Li, Y. Xu, Y. Lu, H. X. Li, Chem. Commun. 46 (2010)
791.
[38] W. Wei, Y. Zhao, S. Peng, H. Zhang, Y. Bian, H. X. Li, H. Li, J. Mater. Chem.
A 2 (2014) 19253.
[39] Y. Sun, Y. Xia, Science 298 (2002) 2176.
[40] H. Li, H. Lin, Y. Hu, H. X. Li, P. Li, X. G. Zhou, J. Mater. Chem. 21 (2011)
18447.
[41] X. Chen, W. Yang, S. Wang, M. Qiao, S. Yan, K. Fan, H. He, New J. Chem.
29 (2005) 266.
[42] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62.
[43] N.R. Jana, L. Gearheart, C.J. Murphy, Langmuir 17 (2001) 6782.
[44] S. Xie, M. Qiao, W. Zhou, G. Luo, H. He, K. Fan, T. Zhao, W. Yuan, J. Phys.
Chem. B 109 (2005) 24361.
[45] J.J.F. Scholten, A.P. Pijers, A.M.L. Hustings, Catal. Rev.: Sci. Eng. 27 (1985)
151.
[46] S. Wang, W. Wei, Y. Zhao, H.X. Li, H. Li, Catal. Today 258 (2015) 327.
[47] Q. Chen, Y. Jia, S. Xie, Z. Xie, Chem. Soc. Rev. 45 (2016) 3207.
25

[48] Y. Zhu, F. Liu, W. Ding, X. Guo, Y. Chen, Angew. Chem. Int. Ed. 45 (2006)
7211.
[49] M. Mo, L. Han, J. Lv, Y. Zhu, L. Peng, X. Guo, W. Ding, Chem. Commun.
46 (2010) 2268.
[50] Y. Jia, Y. Jiang, J. Zhang, L. Zhang, Q. Chen, Z. Xie, L. Zheng, J. Am. Chem.
Soc. 136 (2014) 3748.
[51] L. Zhang, L. T. Roling, X. Wang, M. Vara, M. Chi, J. Liu, S. I. Choi, J. Park,
J. A. Herron, Z. Xie, M. Mavrikakis, Y. Xia, Science, 349 (2015) 412.
[52] Q. Chen, Y. Yang, Z. Cao, Q. Kuang, G. Du, Y. Jiang, Z. Xie, L. Zheng, Angew.
Chem. Int. Ed. 55 (2016) 9021.
[53] G. de Wit, J.J. de Vlieger, A.C.K. Dalen, R. Heus, R. Laroy, A.J. Van
Hengstum, A.P.G. Kieboom, H. Van Bekkum, Carbohydr. Res. 91 (1981) 125.
[54] H. Noller, W.M. Lin, J. Catal. 85 (1984) 25.
[55] J.A.W.M. Beenackers, B.F.M. Kuster, H.S. Van der Baan, Carbohydr. Res.
140 (1985) 169.
26

Figure captions
Scheme 1. Catalytic hydrogenation of glucose to sorbitol.
Scheme 2. Schematic illustration of the formation of hollow Ni−P amorphous
alloy nanospheres.
Scheme 3. Schematic representation of the reaction mechanism between
adsorbed glucose and hydrogen on Ni−P-H amorphous alloy catalyst.
Scheme 4. Catalytic hydrogenation of maltose to maltitol.
Fig. 1. (a) TEM image of Ni−P/SiO2−NH2. (b) High-magnification TEM image
of Ni−P/SiO2−NH2. (c) TEM image of Ni−P-H. (d) High-magnification TEM
image of Ni−P-H. (e) SAED pattern of Ni−P-H.
Fig. 2. (a) N2 adsorption-desorption isotherms and (b) pore size distribution
of Ni−P-H.
Fig. 3. XRD patterns of (a) the fresh Ni−P-H, (b) the Ni−P-H treated at 673 K
in N2 flow for 2 h.
Fig. 4. XPS of Ni−P-H. (a) Ni 2p and (b) P 2p.
Fig. 5. Reaction profile of glucose hydrogenation over Ni−P-H. Reaction
conditions are given in Table 2.
Fig. 6. H2-TPD profiles of (a) Ni−P-H and (b) Ni−P-S. The signal is normalized
based on unit mass Ni.
Fig. 7. Recycling test of Ni−P-H and Ni−P-S catalysts for (a) glucose and (b)
maltose hydrogenation. Reaction conditions are given in Table 2 and Table 3.
Each run was conducted for 4 h in the recycling test.
Fig. 8. TEM image of (a) Ni−P-S and (b) Ni−P-H after six consecutive runs for
maltose hydrogenation. Reaction conditions are given in Table 3. Each run was
conducted for 4 h in the recycling test.
27

Table 1 Electroless-plating bath composition and plating conditions
Concentrations and conditions
NiSO4·6H2O
NaH2PO2·2H2O
Na3C6H5O7·2H2O
CH3COONa
pH
0.076 mol/L
0.19 mol/L
0.017 mol/L
0.061 mol/L
7.0
temperature
time
45℃
30 min
28

Table 2 Structural characteristics and catalytic properties of the as-prepared
catalysts during glucose hydrogenationa
m
m
Composition
Sact
RH
RS
Conversion
(%)
Catalyst
(at.%)
(m2 g−1)
(mmol h−1 g−1)
18.5
(mmol h−1 m−2)
Ni−P-H
Ni−P-S
Ni79P21
Ni80P20
Ni79P21
/
7.8
4.7
2.4
2.0
96
88
46
76
9.42
Crystallized Ni−P-Hb
Raney Nic
3.0
2.08
0.69
0.15
32.2
4.79
a Reaction conditions: 0.5 g catalyst, glucose aqueous solution (10 wt.%, 50 mL), T = 373 K, pH2 = 4.0 MPa,
reaction time = 4 h, stirring rate = 1200 rpm.
b Obtained by treating the fresh Ni−P-H at 673 K under N2 flow for 2 h.
c 1.0 g Raney Ni was used.
29