J. Chen et al.
Journal of Solid State Chemistry 291 (2020) 121652
1
=2
3.3. Photocatalytic performances
Q
t
¼ Kt ⋅ t þ C
(2)
ꢀ
where K
(
t
(mg⋅g 1⋅min1/2) is the intraparticle diffusion rate and C
To confirm whether MoO
ability for low concentration dye, RhB (10 mg/L) is chosen as the
degradation target. As shown in Fig. 7a, MoO @MoS nanorods exhibit
higher degradation efficiency than MoO nanorods. The plots of ln (C
) vs reaction time t were shown in Fig. 7b. It can be seen that the plots
show good linear relationship, indicating that the photocatalytic degra-
dation reactions of RhB in MoO and MoO @MoS systems conform to
the pseudo-first-order kinetic model. The rate constant k (0.01796 min )
3 2
@MoS nanorods have photocatalytic
ꢀ
1
1/2
t
mg⋅g ) is a constant. According to this model, if the plot of Q vs t is
linear and passes through the origin, the intraparticle diffusion is the sole
rate-limiting step. If the plot is multi-linear, two or more steps control the
3
2
3
0
/
1
/2
adsorption process [37]. The plot of Q
t
vs t
was shown in Fig. 6c and
C
t
the result indicated that there are two linear phases corresponding to two
diffusion stages. The first line stage does not pass through the origin,
indicating that intraparticle diffusion was not the only rate-controlling
step, but the stage corresponds to the diffusion of RhB molecules to the
pores. In the second linear stage, the diffusion remains fairly constant in
which RhB is bound onto the actives sites [38]. Thus, the diffusion of RhB
3
3
2
-1
of MoO
3
@MoS
2
nanorod is 6 times that of MoO
3
nanorod (0.00282 min-
1
), indicating that the formation of core-shell structure can significantly
improve the photocatalytic ability.
into the pores of MoO
3
@MoS
2
nanorods, as well as adsorption on the
In order to explore the photocatalytic mechanism, trapping agents
have been added into the degradation system to study the effects on
degradation rates to confirm what active particles control the photo-
available surface is probably responsible for the adsorption process.
The adsorption isotherm in a solid-liquid equilibrium system can be
fitted by Langmuir (eq (3)) [39] and/or Freundlich (eq (4)) [40] formulas
respectively.
2
catalytic reaction. Na EDTA, benzoquinone (BQ) [41], isopropanol
(i-PrOH) and potassium bromate (KBrO
3
) [42] are used to capture hole
þ
ꢀ
(
h ), superoxide anion radical (⋅O
2
), hydroxy radical (⋅OH) and electron
C
Q
e
1
C
e
-
(
e ), respectively. If the degradation reaction is significantly weakened,
¼ Q þ Q
(3)
(4)
e
m
b
m
the captured active particle plays an important role in the photocatalytic
process. The capture experiment results were shown in Fig. 7c and the
1
þ
results indicated that h and ⋅OH play major roles in the degradation of
ln Q
e
¼ ln k
f
þ n ln C
e
RhB. Moreover, the addition of BQ resulted in a moderate decrease in the
ꢀ
degradation rate, implying an indispensable role of ⋅O
2
in the degrada-
where C
mg/g) is the adsorption capacity of RhB, and b (L/mg) is the Langmuir
isotherm parameter related to the adsorption affinity of the binding site.
e m
(mg/L) is the concentration of dye solution at equilibrium. Q
tion. The addition of potassium bromate slightly increased the degrada-
(
-
tion rate, implying the neglected contribution of e to the degradation.
þ
Therefore, in MoO
degradation of RhB, meanwhile, ⋅O
3
@MoS
2
system h and ⋅OH play critical roles in the
-
1
1
e f
Q (mg⋅g ) is the amount of dye adsorbed at equilibrium. k (L/g) and n
ꢀ
2
play an indispensable role in the
are the Freundlich isotherm constant related to adsorption capacity and
adsorption strength. The Langmuir isothermal model (eq (3)) assumes
that the adsorption process is a monolayer adsorption on a homogeneous
adsorbent surface that the adsorption occurs at specific homogeneous
sites within the adsorbent. The Freundlich isothermal model (eq (4))
represents a particular adsorption system when the adsorption takes
place on a heterogeneous surface with interaction between the adsorbed
ions.
degradation.
ꢀ
In order to further verify the generation of ⋅OH and ⋅O
2
, EPR spec-
troscopy using 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as a spin trap
was performed and the result of MoO @MoS nanorod was shown in
Fig. 8a and b. It can be seen that in DMPO-H O system (Fig. 8a), under
3
2
2
dark condition weak signals were detected, and under visible light irra-
diation the strong signals attributed to ⋅OH [43] were detected. Similarly,
ꢀ
in DMPO-MeOH system (Fig. 8b), the strong signals attributed to ⋅O
2
Langmuir and Freundlich isotherm models were used to fit the
adsorption results of RhB onto MoO
[
44] were detected. Hence, under simulated sunlight irradiation porous
3
@MoS
2
nanorods and the plots of
ꢀ
MoO
3
@MoS
2
nanorod can efficiently generate ⋅OH and ⋅O
2
radicals to
C
e
/Q
e
vs C
e
and lnQ
e
vs lnC
e
were displayed in Fig. 6d and e, respectively.
oxidize and reduce pollutant. This result is consistent with the results
coming from capture experiments.
2
For Langmuir isotherm model, the R value extracted by linear fitting of
C /Q vs C is 0.9831 when the C is below 50 mg/L, and the calculated
e e e e
maximum adsorption capacity is 425.5 mg/g which significantly higher
than that of the experimental value (326.83 mg/g). Actually, Langmuir
isotherm model is not matched with the whole adsorption process, that
To investigate the charge separation efficiency, EIS was measured at
open-circuit potential and the EIS Nyquist plots of MoO
MoS samples were presented in Fig. 8c. Obviously, the Nyquist plot
shows a semicircle at high frequency region which characterizes the
charge transfer process. MoO @MoS nanorods show smaller semi-circle
than MoO nanorods, implying lower charge transfer resistance and
3 3
and MoO @-
2
3 2
is, the adsorption of RhB on porous MoO @MoS nanorods is not local-
ized on a monolayer and all the adsorption sites on the adsorbent are not
3
2
2
3
homogeneous [11]. When C
linear fitting of ln Q vs ln C
model is matched very well. This indicates that the adsorption process of
RhB onto porous MoO @MoS nanorods is plausibly controlled by
Freundlich isotherm model under the Ce exceeding 80 mg/L. Therefore,
the initial fast adsorption process of RhB onto porous MoO @MoS
e
exceeds 80 mg/L, the R value extracted by
higher charge transfer rate [45].
Mott-Schottky (M-S) analysis is commonly used to determine both
dopant density and flat band potential at solid-liquid contact [46]. The
e
e
is 0.9983, that is, the Freundlich isotherm
3
2
M–S plots of MoO
The slopes of M-S plots for MoO
3
and MoO
3
@MoS
2
nanorods were shown in Fig. 8d.
@MoS samples are positive,
3
and MoO
3
2
3
2
implying that they are n-type semiconductors. For n-type semi-
conductors, their conduction band potentials are very close to the
flat-band potentials [22]. It can be seen that the conduction band of
nanorods may localized on both homogeneous (monolayer) and hetero-
geneous (multilayer) active sites, and then mainly localized on the het-
erogeneous active sites for the multilayer adsorption, which matches
with the Freundlich isotherm model.
θ
ꢀ
MoO
3
nanorods locates at -0.2 V, less negative than E (O
2
/⋅O
2
) which
locates at -0.33 V vs. NHE, so MoO
3
nanorods do not have the ability to
Desorption is the process of removing adsorbed substance from the
ꢀ
generate ⋅O
2
. For MoS
2
sample, its valence band position (1.42 V) is
adsorbent. The RhB dye adsorbed on the MoO
removed by tetrahydrofuran solution with low-polarity. Fig. 6f showed
the adsorption efficiency of RhB onto MoO @MoS nanorods in 8 cycles
and the result indicated that the removal rate of RhB still reached 92.4%
after eight consecutive cycles, indicating the excellent recycling ability
3
@MoS
2
nanorods was
θ
-
lower than E (⋅OH/OH ) which locates at 2.38 V vs. NHE, therefore it
does not have the ability to generate ⋅OH [47]. However, the EPR results
3
2
ꢀ
(
Fig. 8a, b) showed that MoO
3
@MoS
2
can generate ⋅OH and ⋅O
2
radicals,
implying that MoO
3
@MoS
2
is a Z-scheme photocatalyst [48]. The con-
duction band of MoS
more negative than E (O
2
in MoO
3
@MoS
2
nanorod locates at -0.4 V, which is
for porous MoO
3
@MoS
2
nanorods.
θ
ꢀ
2
/⋅O
2
), so the photogenerated electron in the CB
7