Dual-Template Pore Engineering of Whey Powder-Derived Carbon as an Efficient Oxygen Reduction Reaction Electrocatalyst for Primary Zinc-Air Battery

Article abstract of DOI:10.1002/asia.202000399

Cost-effective and high-performance electrocatalysts for oxygen reduction reactions (ORR) are needed for many energy storage and conversion devices. Here, we demonstrate that whey powder, a major by-product in the dairy industry, can be used as a sustainable precursor to produce heteroatom doped carbon electrocatalysts for ORR. Rich N and S compounds in whey powders can generate abundant catalytic active sites. However, these sites are not easily accessible by reactants of ORR. A dual-template method was used to create a hierarchically and interconnected porous structure with micropores created by ZnCl₂ and large mesopores generated by fumed SiO₂ particles. At the optimum mass ratio of whey power: ZnCl₂ : SiO₂ at 1 : 3 : 0.8, the resulting carbon material has a large specific surface area close to 2000 m² g^?1, containing 4.6 at.% of N with 39.7% as pyridinic N. This carbon material shows superior electrocatalytic activity for ORR, with an electron transfer number of 3.88 and a large kinetic limiting current density of 45.40 mA cm^?2. They were employed as ORR catalysts to assemble primary zinc-air batteries, which deliver a power density of 84.1 mW cm^?2 and a specific capacity of 779.5 mAh g^?1, outperforming batteries constructed using a commercial Pt/C catalyst. Our findings open new opportunities to use an abundant biomaterial, whey powder, to create high-value-added carbon electrocatalysts for emerging energy applications.

Full text of DOI:10.1002/asia.202000399

CHEMISTRY
AN ASIAN JOURNAL
www.chemasianj.org
Accepted Article
Title: Dual template pore engineering of whey powder derived carbon
as an efficient oxygen reduction reaction electrocatalyst for
primary zinc-air battery
Authors: Shunyao Yan, Zixun Yu, Chang Liu, Ziwen Yuan, Chaojun
Wang, Junsheng Chen, Li Wei, and Yuan Chen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Asian J. 10.1002/asia.202000399
Link to VoR: https://doi.org/10.1002/asia.202000399
A Journal of
A sister journal of Angewandte Chemie
and Chemistry – A European Journal

10.1002/asia.202000399
Chemistry - An Asian Journal
Dual template pore engineering of whey powder derived carbon as an efficient oxygen
reduction reaction electrocatalyst for primary zinc-air battery
Shunyao Yan†, Zixun Yu†, Chang Liu, Ziwen Yuan, Chaojun Wang, Junsheng Chen, Li Wei* and
Yuan Chen*
School of Chemical and Biomolecular Engineering, The University of Sydney, New South Wales,
Australia, 2006
Author email: l.wei@sydney.edu.au (L.Wei) and yuan.chen@sydney.edu.au (Y. Chen)
† These authors contribute equally
1
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Abstract
Cost-effective and high-performance electrocatalysts for oxygen reduction reactions are needed for
many energy storage and conversion devices. Here, we demonstrate that whey powder, a major by-
product in the dairy industry, can be used as a sustainable precursor to produce heteroatom doped
carbon electrocatalysts for ORR. Rich N and S compounds in whey powders can generate abundant
catalytic active sites. However, these sites are not easily accessible by reactants of ORR. A dual-
template method was used to create a hierarchical and interconnected porous structure with micropores
created by ZnCl2 and large mesopores generated by fumed SiO2 particles. At the optimum mass ratio
of whey power: ZnCl2: SiO2 at 1:3:0.8, the resulting carbon material has a large specific surface area
close to 2000 m2 g–1, containing 4.6 at.% of N with 39.7% as pyridinic N. This carbon material shows
superior electrocatalytic activity for ORR, with an electron transfer number of 3.88 and a large kinetic
limiting current density of 45.40 mA cm–2. They were employed as ORR catalysts to assemble primary
zinc-air batteries, which deliver a power density of 84.1 mW cm−2 and a specific capacity of 779.5
mAh g−1, outperforming batteries constructed using a commercial Pt/C catalyst. Our findings open
new opportunities to use an abundant biomaterial, whey powder, to create high-value-added carbon
electrocatalysts for emerging energy applications.
Keywords: whey powder; porous carbon; electrocatalyst; oxygen reduction reaction; zinc-air battery
2
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Introduction
Oxygen reduction reaction (ORR) is an essential reaction in renewable energy storage and
conversion devices, such as metal-air batteries and fuel cells.[1] ORR has sluggish kinetics, which is
often a performance-limiting factor for these devices.[2] Platinum (Pt)-based noble metal
electrocatalysts have the best catalytic activity for ORR. However, their applications are limited by
their high cost and poor stability. It is desirable to have more cost-effective and high-performance
ORR electrocatalysts.[3] Heteroatom doped carbon materials is a potential alternative.[3a, 4] The dopant
atoms, for example, nitrogen (N), oxygen (O), boron (B), phosphorus (P), and sulfur (S), can alter
electronic structures of their neighboring carbon (C) atoms and make these C atoms active catalytic
sites for ORR.[5] In general, ORR carbon electrocatalysts should contain abundant heteroatom dopants
to create high-density active catalytic sites and have a large specific surface area (SSA) for
accommodating these sites.[6] Further, a hierarchical pore structure with interconnected micropores
and mesopores is desirable because micropores and small size mesopores (<5 nm) can provide a large
SSA and large-size mesopores (>5 nm) can facilitate the mass transfer.[7] It is essential to have an
optimum balance between pores in different sizes.[8]
Carbon nanomaterials, such as graphene and CNTs, have been used as precursors to produce ORR
carbon electrocatalysts. However, carbon nanomaterials are usually costly, and their synthesis often
produces environmentally harmful by-products and wastes.[9] Because some biomass materials are at
low cost, sustainable, and enriched with N, S, and P, using them as precursors to synthesize ORR
carbon electrocatalysts has drawn lots of interest.[9-10] One of such biomass materials is whey powders,
which is a major by-product in the dairy industry with a global production capacity of over 1.8 million
tons per annum.[11] Whey powders contain roughly 80 wt.% protein, 10 wt.% carbohydrate, 5 wt.% fat,
and 1 wt.% minerals. They are much cheaper (US$0.01 per gram) than commercial electrocatalysts;
for example, Pt catalysts supported on carbon substrates (Pt/C) are ~US$100 per gram. Most of the
whey powders are currently discharged directly to farmlands, and only a small portion is further
3
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
processed as a food-grade ingredient in infant formula or animal feeds.[12] Carbon materials can be
synthesized by direct carbonization of whey powders. However, such carbon materials have a small
SSA without pores. Potential active ORR catalytic sites created by abundant N, S, and P compounds
in whey powder derived carbon materials are not accessible to electrolyte ions and reaction
intermediates. To the best of our knowledge, no previous studies have demonstrated high-performance
ORR electrocatalysts using whey powder derived carbon materials.
Herein, we report a high-performance ORR carbon electrocatalyst by using whey powders as the
precursor. A dual template pore engineering method using both ZnCl2 and fumed silica SiO2 as
templates enabled us to tune hierarchical pore structures of the carbon electrocatalysts. The mass ratio
between the two types of pore-forming templates was systematically optimized. Detailed material
characterization shows that at an optimal mass ratio of mWP : mZnCl : mSiO , the carbon electrocatalyst
2
2
exhibits a large SSA and a hierarchical pore structure, and it has excellent catalytic performance for
ORR, which can compete with commercial Pt electrocatalysts. The optimized carbon electrocatalyst
was used to assemble primary zinc-air batteries (ZABs), which deliver excellent energy storage
characteristics, outperforming ZABs assembled using a commercial Pt electrocatalyst.
Results and Discussion
Figure 1a illustrates the dual template pore engineering method used to create hierarchical pores in
whey powder-derived carbon materials. The details of the synthesis method were described in the
Experimental Section. We chose ZnCl2 and fumed SiO2 particles (~10 nm in size, see Figure S1 in the
Supporting Information (SI)) as templates because ZnCl2 is a commonly used template to create
micropores in carbon materials.[13] During the carbonization, Zn2+ ions are reduced to Zn metal, which
evaporates and leaves behind abundant micropores and small mesopores. Meanwhile, SiO2 particles
can serve as hard templates to create voids similar to their size.[14] Further, amorphous fumed SiO2
particles can be easily washed away by 3 M KOH solution without using the more toxic HF. We
4
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
systematically varied the mass loadings of the two templates, and the resulting carbon materials were
denoted as WC-x-y, where x or y is the mass ratio between ZnCl2 or SiO2 and whey powders,
respectively. In a group of samples, we fixed the mass loading of ZnCl2 (x value) at 3 based on our
previous study.[8d] The mass loading of SiO2 (y) was also systematically varied from 0, 0.2, 0.4, 0.8,
and 1.2. Alternatively, several samples without adding ZnCl2 were also prepared. Our synthesis
method is highly scalable, and as a demonstration, we synthesized one batch of carbon materials in
100 g. Figure 1b shows a photo of the synthesized carbon materials.
Figure 1. (a) Schematic illustration of the synthesis of carbon materials by the dual template method
from whey powders. (b) A photo of a large batch of whey powder derived carbon materials.
We first examined the morphology of different carbon materials by scanning electron microscope
(SEM). Figure 2a shows that WC-3-0.8 has a rough surface with many dents and pores. In comparison,
WC-0-0.8 (Figure 2b) prepared by using SiO2 as the sole template has a similar morphology,
suggesting that the observed dents/pores originate from SiO2 particles. Further, Figure S2 in SI shows
SEM images of WC-3-0.2, WC-3-0.4, and WC-3-1.2. They all have rough surfaces with dents/pores.
In contrast, Figure 2c shows that WC-3-0 prepared by using only the ZnCl2 template has a much
smoother surface with smaller pores. When no templates were used, Figure 2d shows that WC-0-0 has
a flat surface. In summary, the SEM images indicate that the morphology of carbon materials
5
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
carbonized from whey powders can be changed significantly by using different pore formation
templates.
Figure 2. Microscopic analysis of carbon materials derived from whey powders using the dual
template pore engineering method. SEM images of (a) WC-3-0.8 (dual templates), (b) WC-0-0.8 (sole
SiO2 template), (c) WC-3-0 (sole ZnCl2 template), and (d) WC-0-0 (non-template). The scale bar is
500 nm. (e-g) TEM images at different magnifications and (h) a STEM image and the corresponding
EDX elemental mappings of WC-3-0.8.
6
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Table 1. Physiochemical properties of carbon materials synthesized in this study.
No
1
WC-x-y-T,[a]
WC-3-0.2
N, at.% [b] S, at.% [b] O, at.% [b]
SSA, m2 g–1
896.9
Pore volume, cm3 g–1 D/G ratio
4.7
4.3
4.6
4.2
5.0
5.3
5.2
9.7
1.9
1.1
1.2
1.0
1.0
1.1
0.9
1.1
2.2
0.5
9.0
9.6
0.322
0.458
0.635
0.497
0.047
0.147
0.0022
0.412
0.621
1.04
1.06
1.10
1.05
1.03
1.00
1.01
1.18
1.06
2
WC-3-0.4
1232.4
1944.7
1475.7
676.6
3
WC-3-0.8-900
WC-3-1.2
8.9
4
9.7
5
WC-3-0
8.1
6
WC-0-0.8
10.0
9.2
251.4
7
WC-0-0
76.5
8
WC-3-0.8-700
WC-3-0.8-1100
15.9
1359.6
9
7.2
1895.1
[a] Except for the samples 8 and 9, all others were carbonized at 900 oC.
[b] Determined by SEM-EDX.
Next, we measured elemental compositions of the carbon materials by energy-dispersive X-ray
spectroscopy (EDX), as shown in Figure S3 in SI. As a representative example, Figure 2h shows that
C, N, S, and O are the main compositing elements in WC-3-0.8. The abundances of N, S, and O of all
carbon materials determined by EDX are listed in Table 1, showing that the abundances of N and S
o
are similar in all samples carbonized at 900 C. This suggests that although the two templates can
affect the surface morphology substantially, they do not influence the elemental composition strongly.
It should be noted that most of Zn added evaporated during the thermal annealing, and the remnant Zn
was further removed by washing in 3 M HCl for 3 h. EDX measurement showed no-detectable Zn.
We further utilized the inductively coupled plasma atomic emission spectroscopy to analyze WC-3-
0.8, and no Zn was found.
Further, a transmission electron microscope (TEM) was utilized to investigate the pore structures
of the carbon materials. Under the lower magnification, Figure 2e shows that WC-3-0.8 has a porous
structure. Under the higher magnification, Figure 2f reveals many mesopores of 8–10 nm in diameter,
which are created by SiO2 particles. Well-resolved lattice fringes with a d-spacing of ~0.35 nm can be
observed in Figure 2g, indicating a high level of graphitization in carbon materials, which is beneficial
for providing a high electrical conductivity. Figure 2h shows that EDX mappings obtained under the
7
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
scanning TEM mode (STEM) are consistent with the uniform elemental distribution found in SEM
images (Figure S3 in SI).
Figure 3. (a) N2 physisorption isotherms, (b) the corresponding pore size distribution, (c) cumulative
adsorption volumes, (d) magnified micropore regions of (c), and (e) Raman spectra of whey powder
derived carbon materials. (f) Comparison of three physiochemical structural properties of the carbon
materials: SSA, pore-volume, and Raman D/G ratio.
Nitrogen physisorption was used to study the pore structure of the carbon materials. The isotherms
are shown in Figure 3a. Their specific surface areas are calculated by the Brunauer–Emmett–Teller
(BET) method and listed in Table 1. Their pore size distributions were determined by the non-local
8
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
density functional theory (NL-DFT) model and plotted in Figure 3b for comparison. We first
compared the effectiveness of ZnCl2 and SiO2 templates in creating pores. Without using any templates,
WC-0-0 has an SSA of 76.5 m2 g–1 and a negligible pore volume of 0.0022 cm3 g–1. When ZnCl2 was
used, the line shape of the isotherm of WC-3-0 exhibits a combination of Type I and Type IV in the
low pressure (P/P0<0.05) and high pressure (P/P0~0.8) regions, respectively, indicating the co-
existence of both micropores and mesopores.[15] The pore-size distribution displayed in Figure 3b
confirms abundant micropores with a diameter of ~1.5 nm and mesopores of 2.8 nm. The SSA of WC-
3-0 increases nearly 10 times to 676.6 m2 g–1 in comparison to that of WC-0-0. SiO2 particles generate
pores like their size in WC-0-3 with large mesopores (~10 nm). These large mesopores increase the
pore volume to 0.147 cm3 g–1 (Figure 3c); however, they contribute less to the SSA. WC-0-3 has an
SSA of 251.4 m2 g–1.
The dual template method provides an efficient tool for engineering the pore structure and SSA of
whey powder derived carbon materials. As shown in Figure 3c, both the SSA and pore volume increase
with the increase of the mass loadings of SiO2. At an optimal mass loading of SiO2 particles to whey
power at 0.8, WC-3-0.8 has the maximum SSA of 1944.7 m2 g–1 and the largest pore volume of 0.635
cm3 g–1 among all the carbon materials. There is an increasing trend of the S-curve slopes in the
micropore/small-mesopore region (1–3 nm), as displayed in Figure 3d. This suggests that large
mesopores created by SiO2 particles may have formed interconnected pores, which expose buried
micropores and small mesopores formed by ZnCl2. However, further increasing the mass loadings of
SiO2 deteriorates the pore structure of WC-3-1.2 due to the aggregation of SiO2 particles. WC-3-1.2
exhibits a reduced SSA of 1475.7 m2 g–1 and a smaller pore volume of 0.497 cm3 g–1 compared to WC-
3-0.8. The pore size distribution curve of WC-3-1.2 exhibits a broadened peak of large mesopores and
a new peak at around 20 nm (marked by * in Figure 3b). Its S-curve in Figure 3c also exhibits an
extended increasing trend, indicating the formation of larger pores.
9
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
We further collected Raman spectra of the carbon materials to study the impact of pore-engineering
on their graphitic structure. Figure 3e shows all carbon materials have two major peaks. The peak at
1587 cm–1 (G-band) is originated from the sp2-hybridized carbon atoms in the graphite lattice while
the peak at 1329 cm–1 (D-band) can be assigned to the defect-induced phonon scatterings. Hence, the
intensity ratio between the D- and G-band peaks (D/G ratio) can be utilized to compare the crystallinity
of the carbon materials.[16] The calculated D/G ratio is listed in Table 1 for comparison. The results
show that the addition of the SiO2 template has a minor impact. WC-0-0.8 has a D/G ratio of 1.01,
which is similar to that of WC-0-0 at 1.00 without adding any templates. WC-3-0 has an increased
D/G ratio of 1.03. Considering the elemental analysis results show that heteroatom abundances are
similar in WC-3-0 and WC-0-0, the higher D/G ratio of WC-3-0 can be attributed to defects introduced
by micropores. Figure 3f shows that the changes in SSA and pore volumes exhibit a good agreement
with the changes in Raman D/G ratios, further confirming the capability of the dual template method
in tuning the structures of whey powder derived carbon materials.
Besides the pore structures, the concentration of N and S dopants and their configurations can also
strongly influence the ORR catalytic activity of carbon electrocatalysts.[17] While EDX elemental
analysis results show that the pore-engineering does not change the abundances of N and S
significantly, we further probed their detailed configurations by high-resolution N1s and S2p X-ray
photoelectron spectroscopy (XPS). Figure 4a shows that the N1s spectra of the carbon materials can
be deconvoluted into four different peaks at 398.4, 400.5, 401.5, and 403.7 eV, which can be assigned
to pyridinic-, pyrrolic-, graphitic-, and oxidized-N species, respectively (as illustrated in Figure 4c).[18]
Further, the S2p spectra in Figure 4b can be split into two peaks. The peaks at 163.9 and 165.2 eV can
be assigned to S2p3/2 and S2p1/2 of S-C bonds (as illustrated in Figure 4c).[19]. The broad peak at a
higher binding energy of 168.4 eV can be attributed to oxidized S-O bonds. The relative abundances
of N and S in different configurations are listed in Table S1 in SI for comparison.
10
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Figure 4. High-resolution XPS spectra of (a) N1s and (b) S2p of carbon materials. (c) Schematic
illustration of N and S atoms in different configurations. The abundances of N atoms in various
configurations in carbon materials with (d) different templates and (e) changing the mass ratios of the
SiO2 template.
Figure 4d shows that WC-0-0 without using any template has the highest abundance of pyridinic N.
ZnCl2 and SiO2 templates can convert some pyridinic-N into pyrrolic-N. For example, SiO2 templated
WC-0-0.8 has a high abundance of pyrrolic-N (~39.7 %). The pyridinic-N content in the dual-
templated WC-3-0.8 was found to be higher over that in WC-0-0.8 and WC-3-0. Besides, more S in
the S-C configuration was also found in the dual-templated WC-3-0.8 (68.3 %, Table S1 in SI).
We further explored potential correlations between the mass loading of SiO2 particles and the
configuration of N- and S-dopants. WC-3-y samples (y = 0.2, 0.4, 0.8 and 1.2) shown in Figure S4 in
SI were deconvoluted. The abundances of N in different configurations are displayed in Figure 4e.
11
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
While the abundances pyridinic- and pyrrolic-N remain largely unchanged, the abundance of graphitic-
N drops slightly from WC-3-0.2 to WC-3-0.8, and then increases in WC-3-1.2. This trend is inverse
to the trend of the SSA shown in Figure 3f. Initially, more SiO2 particles generate more pores, which
have more edges to accommodate pyridinic- and pyrrolic-N species. Thus, WC-3-0.8 has the highest
SSA of 1944.7 m2 g–1 and the lowest abundance of graphitic-N at 22.9 %. However, further increasing
the mass loading of SiO2 reduces the SSA to 1475.7 m2 g–1 in WC-3-1.2, and its abundance of
graphitic-N rises to 28.4 %. In contrast, Table S1 in SI shows that the abundance of S-C has minor
changes, suggesting that the configuration of S is not affected by the mass loading of SiO2 particles.
Figure 5. Impact of carbonization temperatures on (a) SSA and Raman D/G ratio and (b) N and S
configurations of whey powder derived carbon materials.
The impact of carbonization temperature on the pore structure and heteroatom dopant configurations
was also investigated. Two additional WC-3-0.8 samples were prepared by carbonization at 700 and
1100 ºC (denoted as WC-3-0.8-700 and WC-3-0.8-1100, respectively). First, SEM images in Figure
S5 in SI show that the two samples have minor morphology changes compared to WC-3-0.8-900. Their
adsorption isotherms in Figure S6 in SI shows that WC-3-0.8-700 has a reduced SSA and pore volume,
while the parameters of WC-3-0.8-1100 are similar to those of WC-3-0.8-900, as summarized in Table
1 and Figure 5a. The pore size distributions of the three samples are also similar except that WC-3-
12
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
0.8-700 has a slightly lower micropore volume, which may be due to the fact that the carbonization at
700 ºC is not high enough to remove Zn efficiently. Overall, when the carbonization temperature is
above 900 ºC, the pore structure of whey powder derived carbon materials has minor changes. Next,
EDX elemental analysis of N, S, and O listed in Table 1 show that abundances of heteroatoms decrease
with the increase of carbonization temperature. Further, Raman spectra in Figure S7 and the
corresponding D/G ratios shown in Figure 5a indicate that the defect density is lower at higher
carbonization temperatures. Last, Figure S8 in SI shows N1s and S2p XPS spectra of the three samples.
The abundances of N and S in different configurations are shown in Figure 5b. The abundance of S
atoms in the S-C configuration is higher at 900 and 1100 ºC. N atoms also exhibit substantial changes.
o
The abundance of pyridinic N is higher at 900 and 1100 C, while the abundance of pyrrolic N has
minor variations at the three temperatures, while the abundance of graphitic N also has small changes
above 900 oC. Since the total N abundance in WC-3-0.8-1100 is only 1.9 at.%, which is much lower
than that in WC-3-0.8-900 at 4.6 at.%. Thus, WC-3-0.8-900 contains more pyridinic N, which is
beneficial for enhancing the catalytic activity for ORR.[5b]
13
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Figure 6. Electrochemical performance of carbon electrocatalysts tested in 0.1 M KOH electrolyte. (a)
ORR LSV curves of carbon electrocatalysts obtained by (a) using different types of templates and (b)
different mass loadings of SiO2 template. (c) Comparison of ORR kinetic parameters at 0.4 VRHE. (d)
Durability tests by (d) CV cycling and (e) chronopotentiometry methods. The inset in (e) displays the
MeOH tolerance test result.
The whey powder derived carbon materials were used as electrocatalysts for ORR. Their ORR
catalytic activity was tested in O2 saturated 0.1 M KOH electrolyte. Figure 6a compares the LSV
curves of the carbon electrocatalysts synthesized using different templates. The dual templated WC-
3-0.8 exhibits superior performance with the largest onset potential of 0.98 V vs. VRHE and the highest
limiting current density of 6.08 mA cm–2 at 0.2 V. The single templated samples (WC-3-0 and WC-0-
0.8) show inferior performances with more negative onset potentials and lower limiting current
densities. It is noteworthy that even though WC-3-0 has the SSA 10 times higher than the non-porous
WC-0-0, WC-3-0 only exhibits a minor improvement in its ORR activity compared to WC-0-0. And
WC-3-0 is outperformed by WC-0-0.8 with only 1/3 of its SSA but containing more mesopores. These
observations suggest that mass transfer resistances of electrolyte ions and reaction intermediates in
micropores is a critical limiting factor for the ORR catalytic activity of porous carbon electrocatalysts
derived from whey powders.
Table 2. ORR performance parameters of carbon electrocatalysts derived from whey powders.
ORR activity
Sample
E1/2, V
0.764
0.783
0.875
0.801
0.515
0.789
0.521
0.887
n [a]
3.55
3.79
3.88
3.83
3.06
3.28
2.95
3.96
jK, mA cm-2 [a]
31.22
WC-3-0.2
WC-3-0.4
WC-3-0.8
WC-3-1.2
WC-3-0
WC-0-0.8
WC-0-0
Pt/C
38.64
45.40
41.33
23.18
30.17
15.19
43.50
[a] calculated by K-L plots at 0.4 VRHE
.
14
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
We further studied how the pore-engineering influences the ORR catalytic activity of the dual-
templated carbon electrocatalysts. Figure 6b shows that WC-3-0.8 exhibits the highest ORR catalytic
activity among all carbon electrocatalysts. When compared to the activity of the commercial 20 wt%
Pt/C catalyst, WC-3-0.8 can reach the half-wave potential (E1/2) at 0.875 V vs. VRHE, which is similar
to that of Pt/C at 0.887 V. Further, WC-3-0.8 delivers a limiting current density 16% higher than that
of Pt/C. K-L plots of WC-3-0.8 are shown in Figure S9, and they were used to calculate the electron
transfer number (n) and kinetic limiting current density (jK). The results are listed in Table 2 and
compared in Figure 6c. WC-3-0.8 has the largest n of 3.88 and jK of 45.40 mA cm–2 among all carbon
electrocatalysts, which is also comparable to those of Pt/C at 3.96 and 43.50 mA cm–2, respectively.
Figure 6c shows that the changes in the catalytic activities of carbon electrocatalysts exhibit a similar
trend as those in their structure properties, including SSA and pore volume. We also compared the
catalytic activity of WC-3-0.8 obtained at different carbonization temperatures, and the results are
shown in Figure S10 in SI. WC-3-0.8-900 shows higher activity than WC-3-0.8-700 and WC-3-0.8-
1100. Besides, the performance of WC-3-0.8 one of the best among recently reported biomass-derived
carbon electrocatalysts (see a detailed comparison in Table S2 in SI).
The durability of WC-3-0.8 was also assessed. As shown in Figure 6d, the LSV curve of WC-3-0.8
exhibits a shift of ~14 mV after 10000 cyclic voltammetry (CV) cycles scanned between 1 to 0 V,
which is much smaller than that of Pt/C at 36 mV. Besides, Figure 6e shows that in the durability test
carried out by the chronopotentiometric method, WC-3-0.8 can retain ~92% of its initial current
intensity after 6 under 0.4 V, which is higher that of Pt/C at 73%. Moreover, WC-3-0.8 also exhibits
much better tolerance toward methanol crossover than Pt/C, as shown in the inset of Figure 6e. The
surface morphology of WC-3-0.8 has negligible alternations after the stability test (Figure S11 in SI).
15
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Figure 7. The performance of primary ZABs assembled using WC-3-0.8 as ORR electrocatalyst. (a)
Galvanodynamic discharge curves and power density and (b) the galvanostatic specific capacity of
ZABs fabricated using WC-3-0.8 and Pt/C. (c) Galvanostatic discharge curves of the ZAB-WC-3-0.8
collected at different current densities.
Since WC-3-0.8 has excellent ORR catalytic activity, we further demonstrated their practical
applications as ORR catalysts in primary ZABs. A ZAB was assembled using a gas-diffusion electrode
deposited with WC-3-0.8 with a mass loading of 0.5 mg cm–2 (denoted as ZAB-WC-3-0.8). A Zn foil
was used as a metal electrode, and a mixture of 6 M KOH with 0.5 M Zn(CH₃CO₂)₂ was used as the
electrolyte. Another ZAB was assembled as a reference by replacing WC-3-0.8 with Pt/C at the same
mass loading (denoted as ZAB-Pt/C). The battery performances were tested in ambient conditions at
25 °C. As shown in the galvanodynamic charge-discharge curves in Figure 7a, ZAB-WC-3-0.8
delivers an open circuit potential of 1.43 V, which is similar to 1.45 V of ZAB-Pt/C. ZAB-WC-3-0.8
16
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
outperforms ZAB-Pt/C when the discharge current density is higher than ~20 mA cm–2. The maximum
power density of ZAB-WC-3-0.8 at 84.1 mW cm−2 is ~10% higher than that of ZAB-Pt/C at 73.5 mW
cm−2. Further, the specific capacity of ZAB-WC-3-0.8 reaches 779.5 and 746.2 mAh g−1 at the
discharge current density of 5 and 20 mA cm–2, respectively, which are higher than the 735.1 mAh g−1
of ZAB-Pt/C. ZAB-WC-3-0.8 also exhibits robust responses when the discharge current density
changes. As shown in Figure 6c, the voltage of ZAB-WC-3-0.8 can quickly stabilize when the current
density increases from 1 to 50 mA cm–2. Its output voltage returns to its original value when the current
density is set back to 1 mA cm–2, indicating excellent performance with fast responses.
Conclusion
In summary, using ZnCl2 and fumed SiO2 as templates, the dual-template method can effectively tune
pore structures of carbon materials derived from whey powders. ZnCl2 created abundant micropores
and small mesopores, while SiO2 particles produced large interconnected mesopores. The hierarchical
pores provide a large surface area accessible to electrolyte ions and reaction intermediates. The optimal
carbon material (WC-3-0.8) can deliver the highest SSA of 1944.7 m2 g-1 and superior ORR activity
with E1/2 at 0.875 V, n of 3.88, and jK of 45.40 mA cm–2, comparable to the commercial Pt/C catalyst.
The ZAB assembled using WC-3-0.8 has an open circuit potential of 1.43 V, the maximum power
density of 84.1 mW cm−2, the specific capacity of 779.5 mAh g−1 at the discharge current density of 5
mA cm–2, which outperform the ZAB fabricated using commercial Pt/C catalyst. This work opens up
the opportunity to transform abundant biomaterials into high-value-added electrocatalysts for energy
storage and conversion applications.
Experimental section
Material synthesis. Whey powder (NKD Whey Protein Isolate) was obtained from NKD Co., which
contains 83.3 wt.% protein, 10 wt.% carbohydrate, and 6.7 wt% fat. Whey powder derived carbon
17
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
materials were synthesized using the same procedure while the type and amount of template materials
were varied. Typically, 1 g of the whey powders were dissolved in 20 mL acetone under stirring.
Different quantities of ZnCl2 (99.9%, Sigma) and fumed SiO2 (specific surface area of 200 m2 g−1,
Sigma) were added to the acetone solution. The mixture was further stirred for 2 h before it was heated
and dried under stirring to remove all solvent. The solid material was grounded into fine powders, put
in an alumina boat, and placed in the middle of a tube furnace (BTC-1200, BEQ, tube diameter = 50
mm). The furnace was flushed with Ar (99.999%, BOC) for 30 min before the sample was heated to
1
900 °C under Ar flow (200 sccm) at a ramp rate of 5 °C min− . The solid material was held at 900 °C
for 2 h before cooled down to room temperature under Ar protection. The carbonized material was
first refluxed in 3 M KOH to remove SiO2. Next, it was washed by 3 M HCl and a large amount of
deionized water to remove Zn residuals. The resulting carbon materials were denoted as WC-x-y (see
Table 1), where x and y refer to the mass ratios of ZnCl2 and SiO2 to whey powders, respectively. Two
additional carbon materials were synthesized by changing the carbonization temperature to 700 and
1100 °C (denoted as WC-3-0.8-700 and WC-3-0.8-1100, respectively). All carbon materials were
dried in an oven at 80 °C for 12 h before electrochemical catalytic activity tests.
Material characterization. Surface morphology and elemental composition were characterized by an
SEM (Ultra-Plus, Zeiss) equipped with an EDX spectrometer. TEM and STEM-EDX characterization
were performed on an electron microscope (JEM-2100, Joel). N2 physisorption isotherms were
obtained on a surface analyzer (Autosorbe-1, QuantaChrome). The SSA was calculated by the BET
method, and the pore size distribution was determined by the NL-DFT method. Raman spectra were
collected under 532 nm laser irradiation by a Raman microscope (inVia, Renishaw). The chemical
composition and configuring of different elements were determined by XPS (K-Alpha+, Thermo
Fisher Scientific) with an Al-Kα source (1486.3 eV).
Electrode preparation. Carbon materials were first ground into fine powders. About 5 mg of the
ground carbon powder was added in a water/isopropanol alcohol (IPA) solution (1:9 v/v) with 0.05
18
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
wt. % Nafion 117 (Sigma) at a concentration of 5 mg mL–1. The solution was then homogenized by an
ultrasonic bath for 1 h and served as catalyst ink to prepare electrodes. Working electrodes were
prepared by drop-casting 20 µL of catalyst ink on pre-polished glassy carbon rotating disk electrodes
(RDEs) (5 mm in diameter, Pine Research). The areal mass loading is about 0.5 mg cm–2. The RDEs
were dried at the ambient condition overnight. The commercial Pt/C (20 wt.% Pt on Vulcan XC-72
carbon black, Sigma) was used as a reference catalyst for ORR. They were cast on electrodes using
the same method. At least three electrodes were prepared and tested for every catalyst sample to ensure
result reproducibility.
Electrochemical performance test. The activity for ORR was evaluated using the three-electrode
configuration. A saturated calomel electrode and a Pt mesh were used as the reference and counter
electrode, respectively. All potentials were calibrated to a reversible hydrogen electrode (RHE).
Electrocatalytic tests were performed on an electrochemical workstation (660E, CHI Instrument) in
0.1 M KOH electrolyte. The electrolyte was saturated with either Ar or O2 for 30 min before
performance measurements. Cyclic voltammetry (CV) curves were recorded at 50 mV s−1. Linear
sweep voltammetry (LSV) curves were collected from 400 to 2500 rpm at 5 mV s−1 without iR
compensation. Electron transfer number (n) and the kinetic limiting current density (jK) of different
catalysts were calculated by the Koutechky-Levich (K-L) equation, as shown below:
1/j= 1/j퐿+ 1/j퐾 = 1/퐵휔1/2+ 1/j퐾
퐵=0.62푛퐹퐶0(퐷0)2/3휈−1/6
j퐾=푛퐹푘퐶0
(1)
(2)
(3)
where j is the measured current density, j푘 and jL are the kinetic- and diffusion- limiting current
densities, ω is the angular velocity, n is electron transfer number, F is the Faraday constant, C0 is the
saturated concentration of O2 in 1 M KOH at room temperature, D0 is diffusion coefficient of oxygen
in water, υ is the kinematic viscosity of the electrolyte, and k is the electron-transfer rate constant.
19
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Zn-air battery assembly and tests. Gas diffusion electrodes were prepared by drop-casting the catalyst
ink on a piece of carbon cloth (2×2 cm2 in size) with a mass loading of 0.5 mg cm−2. A Zn foil was
used as the metal electrode, and a mixture of 6 M KOH with 0.2 M ZnCl2 aqueous solution was used
as the electrolyte. Zn-air batteries were assembled by sandwiching the gas diffusion electrode, the Zn-
metal electrode, and the electrolyte together. Several reference batteries were also assembled by using
gas diffusion electrodes prepared by the reference Pt/C catalyst with the same areal loading. The
battery performance was tested on a battery test station (Land).
Acknowledgment
The Australian Research Council supported this work under the Future Fellowships scheme
(FT160100107) and Discovery Programme (DP180102210), and we acknowledge technical assistance
provided by the Sydney Microscopy & Microanalysis, a Core Research Facility of the University of
Sydney.
References
[1] a) G. Wu, A. Santandreu, W. Kellogg, S. Gupta, O. Ogoke, H. Zhang, H.-L. Wang and L. Dai,
Nano Energy 2016, 29, 83-110; b) P. Gu, M. Zheng, Q. Zhao, X. Xiao, H. Xue and H. Pang, J. Mater.
Chem. A 2017, 5, 7651-7666; c) X. Li, F. Dong, N. Xu, T. Zhang, K. Li and J. Qiao, ACS Appl. Mater.
Interfaces 2018, 10, 15591-15601; d) B. C. H. Steele and A. Heinzel, Nature 2001, 414, 345-352.
[2] a) Y. Nie, L. Li and Z. Wei, Chem. Soc. Rev. 2015, 44, 2168-2201; b) S. Sui, X. Wang, X. Zhou,
Y. Su, S. Riffat and C.-j. Liu, J. Mater. Chem. A 2017, 5, 1808-1825; c) M. Liu, Z. Zhao, X. Duan and
Y. Huang, Adv. Mater. 2019, 31, 1802234.
[3] a) X. Liu and L. Dai, Nat. Rev. Mater. 2016, 1, 1-12; b) H. F. Wang, C. Tang and Q. Zhang, Adv.
Funct. Mater. 2018, 28, 1803329; c) D. Liu, L. Dai, X. Lin, J. F. Chen, J. Zhang, X. Feng, K. Müllen,
X. Zhu and S. Dai, Adv. Mater. 2019, 31, 1804863.
20
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
[4] a) K. Gong, F. Du, Z. Xia, M. Durstock and L. Dai, Science 2009, 323, 760; b) C. Hu and L. Dai,
Angew. Chem. Int. Ed. 2016, 55, 11736-11758; c) L. Yang, J. Shui, L. Du, Y. Shao, J. Liu, L. Dai and
Z. Hu, Adv. Mater. 2019, 31, 1804799; d) C. Hu, Y. Xiao, Y. Zou and L. Dai, Electrochem. Energy
Rev. 2018, 1, 84-112; e) S. Zhao, D. W. Wang, R. Amal and L. Dai, Adv. Mater. 2019, 31, 1801526.
[5] a) J. Zhang, L. Qu, G. Shi, J. Liu, J. Chen and L. Dai, Angew. Chem. Int. Ed. 2016, 55, 2230-2234;
b) H. B. Yang, J. Miao, S.-F. Hung, J. Chen, H. B. Tao, X. Wang, L. Zhang, R. Chen, J. Gao and H.
M. Chen, Sci. Adv. 2016, 2, e1501122; c) Q. Lv, W. Si, J. He, L. Sun, C. Zhang, N. Wang, Z. Yang,
X. Li, X. Wang and W. Deng, Nat. Commun. 2018, 9, 1-11.
[6] a) G. A. Ferrero, K. Preuss, A. B. Fuertes, M. Sevilla and M. M. Titirici, J. Mater. Chem. A 2016,
4, 2581-2589; b) H. Yan, Chem. Commun. 2012, 48, 3430-3432.
[7] L. Zhang, X. Yang, F. Zhang, G. Long, T. Zhang, K. Leng, Y. Zhang, Y. Huang, Y. Ma, M. Zhang
and Y. Chen, J. Am. Chem. Soc. 2013, 135, 5921-5929.
[8] a) C. Long, L. Jiang, X. Wu, Y. Jiang, D. Yang, C. Wang, T. Wei and Z. Fan, Carbon 2015, 93,
412-420; b) J. Wang and S. Kaskel, J. Mater. Chem. 2012, 22, 23710-23725; c) Y. Li, H. Zhang, Y.
Wang, P. Liu, H. Yang, X. Yao, D. Wang, Z. Tang and H. Zhao, Energy Environ. Sci. 2014, 7, 3720-
3726; d) L. Wei, H. E. Karahan, K. Goh, W. Jiang, D. Yu, Ö. Birer, R. Jiang and Y. Chen, J. Mater.
Chem. A 2015, 3, 7210-7214; e) J. Pampel and T. P. Fellinger, Adv. Energy Mater. 2016, 6, 1502389.
[9] J. Deng, M. Li and Y. Wang, Green Chem. 2016, 18, 4824-4854.
[10] a) M. Borghei, J. Lehtonen, L. Liu and O. J. Rojas, Adv. Mater. 2018, 30, 1703691; b) L. Wei, H.
E. Karahan, S. Zhai, Y. Yuan, Q. Qian, K. Goh, A. K. Ng and Y. Chen, J. Energy Chem. 2016, 25,
191-198.
[11] G. Chegini; and M. Taheri, Middle-East J. Sci. Res. 2013, 13, 1377-1387.
[12] J. G. Zadow in Whey and Whey Powders | Production and Uses, (Ed. B. Caballero), Academic
Press, Oxford, 2003, pp. 6147-6152.
[13] C. Saka, J. Anal. Appl. Pyrol. 2012, 95, 21-24.
21
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
[14] a) J. Park, Y. Nabae, T. Hayakawa and M. A. Kakimoto, ACS Catal. 2014, 4, 3749-3754; b) X.
R. Wen, D. S. Zhang, L. Y. Shi, T. T. Yan, H. Wang and J. P. Zhang, J. Mater. Chem. 2012, 22, 23835-
23844; c) T. Liu, L. Y. Zhang, W. You and J. G. Yu, Small 2018, 14.
[15] a) Y. Ren, Z. Ma, R. E. Morris, Z. Liu, F. Jiao, S. Dai and P. G. Bruce, Nat. Commun. 2013, 4,
2015-2015; b) J. Shen, A. Liu, Y. Tu, G. Foo, C. Yeo, M. B. Chan-Park, R. Jiang and Y. Chen, Energy
Environ. Sci. 2011.
[16] F. Tuinstra and J. L. Koenig, J. Chem. Phys. 1970, 53, 1126-1130.
[17] C. V. Rao, C. R. Cabrera and Y. Ishikawa, J. Phys. Chem. Lett. 2010, 1, 2622-2627.
[18] L. Ma, J. Liu, S. Lv, Q. Zhou, X. Shen, S. Mo and H. Tong, J. Mater. Chem. A 2019, 7, 7591-
7603.
[19] J. P. Paraknowitsch, A. Thomas and J. Schmidt, Chem. Commun. 2011, 47, 8283-8285.
22
This article is protected by copyright. All rights reserved.

10.1002/asia.202000399
Chemistry - An Asian Journal
Graphitic abstract
Pore structures of whey powder derived carbon materials are engineered by using ZnCl2 and SiO2
1
particles as dual templates. A large surface area of 1944 m2 g− is obtained at an optimal mass ratio of
whey powder: ZnCl2: SiO2 = 1: 3: 0.8, which delivers excellent catalytic activity for the oxygen
reduction reaction.
23
This article is protected by copyright. All rights reserved.