Effects of surfactants on the preparation of MnO2 and its capacitive performance

Synthesis of MnO₂ by liquid co-precipitation

Solutions of 0.3 M MnCl₂ and 0.2 M KMnO₄ were prepared by dissolving MnCl₂·4H₂O (5.94 g) and KMnO₄ (3.16 g) of analytical grade, respectively, in de-ionized water. Then, MnCl₂ solution (with the addition of 0.5 g PVP) was added into KMnO₄ solution at 60°C with a speed of 60 drop/min under continuous stirring. After full reaction and subsequent ageing process for 8 h, the suspension was filtered, and washed with de-ionized water and ethanol repeatedly by vacuum filtration. The obtained sample was dried at 100°C for 12 h in an oven, and then ground thoroughly by the mortar to produce MnO₂ powder, which was marked as Sample d. For comparison, the MnCl₂ solutions without surfactant and with the same mass fraction of PEG or SDBS were also used to prepare MnO₂ powder by the same process, and the produced samples were marked as Sample a, Sample b and Sample c, respectively.

Structural characterization

Crystalline structure of the samples was tested by x-ray diffraction (XRD, Rigaku, DMAX^- Ultima⁺ diffraction meter) with Cu Kα radiation (λ = 1.5404 Å). The morphology was observed using field emission scanning electron microscopy (FE-SEM, SUPRA 55 SAPPHHIRE) and transmission electron microscope (TEM, JEOL JEM-2100). The Brunauer-Emmett-Teller (BET) surface area was measured by WBL-8XX BET surface analyzer.

Electrochemical performance

The as-prepared MnO₂ powder, activated carbon (XC-72) and binder were mixed according to a weight ratio of 75:15:10. The binder was the mixture of PVDF and N-methyl-2-pyrrolidone (NMP) with a weight ratio of 1:10. The mixture was smeared into a nickel foam, and then dried in vacuum at 100°C for 8 h. Subsequently, the foam was pressed at 10 MPa to obtain a wafer of 20 mm in diameter. Electrochemical test was conducted on the VMP3 (EG&G) electrochemical workstation using a three-electrode system, with the platinum net, a saturated calomel electrode (SCE) and the prepared foam as the counter electrode, reference electrode, and the working electrode, respectively. All electrochemical measurements were carried out in 6 M KOH solution and the galvanostatic charge/discharge test was at the current density of 120 mA/g.

XRD patterns and BET specific surface area analysis

Figure1 shows the XRD patterns of samples prepared with different surfactants. As can be seen, the four samples have similar diffraction patterns, and present strong diffraction peaks at 37.2° and 66.7° and relatively weak diffraction peaks at 56.0° and 78.9°, which, respectively, correspond to (100) and (110) crystal planes and (102) and (200) planes of α-MnO₂ according to the standard cards (JCPDS No. 30-0820). However, the intensity of diffraction peaks is relatively weak and severely broadened. The results indicated that the addition of different surfactants during preparation had no effects on the crystalline structure of the samples. According to the XRD patterns, the prepared products are amorphous α-MnO₂·nH₂O (21, 22). The amorphous structure is beneficial to the processes of de-embedding and embedding of the proton, which can result in a rapid, reversible chemical absorption/desorption or oxidation/reduction reaction on the electrode surface or within the bulk phase, thereby generating high pseudocapacitance. Because of the amorphous structure, MnO₂ is more suitable for the application concerning the supercapacitor electrode material (23).

Figure 2 shows the specific surface areas of MnO₂ prepared by different surfactants. The specific surface areas for Samples a, b, c and d are 102.32 m²/g, 169.92 m²/g, 137.40 m²/g and 196.64 m²/g, respectively. It can be seen that adding surfactants can effectively promote the dispersion of the samples and thus increase the specific surface area. Compared with Sample a, the specific surface areas of Samples b, c and d increase by 66.1%, 32.3% and 92.2%, respectively. From the perspective of increasing specific surface area, PVP has the best effect, followed by PEG, and SDBS has the worst effect.

Morphology and microstructure analysis

In order to observe the microstructure of the obtained samples by adding different surfactants, the morphology was observed by FE-SEM and HRTEM. As Figure 3 shows, MnO₂ particles tend to have smaller particle size, more regular shape and better dispersion after adding surfactants during preparation. The sample without surfactants (Sample a) has irregular particles and presents a serious agglomeration, with an average particle size of about 400 nm. By contrast, the sample with PVP surfactant (Sample d) has irregular spherical particles, and their average particle size is around 200 nm. From the energy-dispersive spectroscopy (EDS) pattern of Sample d (see supplementary figure S1, available online as supplementary material at www.jab-fm.com), only the peaks of Mn and O can be observed. Combined with the XRD patterns, it can be further determined that the product is amorphous MnO₂. Figure 4 illustrates the TEM morphology of Sample d. It can be seen that there are many long burrs of about 30-50 nm in length on the surface of the spherical particles. The corresponding selected area electron diffraction (SAED) pattern (the inset shown in Fig. 4) shows a few broad halos without diffraction spots, once again confirming the amorphous structure and low crystallinity of the obtained material. In the process of preparation, the added surfactant could form micelles, reverse micelles, vesicles and other organized assemblies, which could be used as a “micro reactor” to inhibit the agglomeration of nanoparticles (24).

As a typical ionic surfactant, SDBS always takes effect in the form of electrostatic stabilization. According to “DLVO theory” (25), the colloidal particles with negative charges from the sulfonic acid groups in SDBS can be formed. When MnCl₂·4H₂O solution was added as a reductive agent into SDBS solution, electrostatic attraction of the colloidal particles with Mn²⁺ ions with positive charges occurred, thus forming a stable double electrical layer. Then, as a “micro reactor”, it could react with oxidizer and generate nanostructured MnO₂. Under the action of electrostatic repulsive force, it was difficult for MnO₂ particles to agglomerate together. However, PEG and PVP, as non-ionic surfactants, could also produce steric stabilization besides the effect of electrostatic stabilization, and further inhibited the growth of grains, leading to further optimization of the morphology and microstructure.

Electrochemical properties

Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) and galvanostatic charge/discharge tests are considered as important tools to study the electrochemical properties of electrode materials. Figure 5 shows the CV curves of the samples with the addition of different surfactants measured in 6 M KOH solution. As can be seen, there are no apparent redox peaks in the CV curves, which are roughly rectangular at lower scan rate. This indicated that the four electrode materials all had good capacitance and fast charge/discharge property. However, with the increase of scan rate, the CV curves deviated from the roughly rectangular shape to a spindle shape. With the increase of scan speed, electrochemical polarization and concentration polarization occurred in the solution, so that the electrolyte ions could only reach the outer surface of the MnO₂ electrode during extremely brief time. As a result, the effective interaction between the electrolyte and active material of electrode was greatly reduced, which caused the limited utilization rate of active material and poor electrochemical behavior. From the CV curves, the areas of the samples follow the sequence of Sample d > Sample b > Sample c > Sample a. Furthermore, compared with Sample a, the CV curves of other samples (Samples b, c and d) are excellently symmetrical with the zero-current line, and have less deformation with the increase of the scanning rate. The experimental results showed that the reversibility and rate performance of the electrode materials could be improved by adding surfactants.

In order to investigate the influence of different surfactants on the alternating current (AC) impedance, the EIS of the samples were tested. Figure S2 (available online as supplementary material at www.jab-fm.com) shows the Nyquist plots of different samples. All Nyquist plots show an approximate semicircle in the high frequency range and exhibit a linear characteristic in the low frequency range. Compared with Sample a, the radius of the semicircle decreases gradually while the slope of the straight line increases gradually following the order of Sample c, Sample b and Sample d. The results confirmed the decrease of the electrochemical reaction impedance of the electrode, which corresponded to the increase of the ion diffusion ability in the solution (8, 11). The reason can be attributed to the fact that adding surfactants produced particles of smaller size, which increased the specific surface area and inhibited the agglomeration of nanoparticles.

Figure 6 shows the galvanostatic charge/discharge curves of different samples. As can be seen, all the charge/discharge curves show a good linear characteristic and exhibit an approximate isosceles triangle shape. This indicated that the prepared electrode materials had ideal capacitance characteristics and good reversibility. The following formula can be used to calculate specific capacitance:

where Cm represents the specific capacitance of the electrode material (F/g); i is the current of discharge (A); Δt is the time of discharge (s); m is the mass of active materials in the working electrode (g); and ΔV is the voltage drop of discharge (V).

According to the formula, the specific capacitances of Samples a, b, c and d at 120 mA/g are 155.9 F/g, 207.9 F/g, 187.5 F/g and 238.7 F/g, respectively. The results indicated that during the preparation process of MnO₂, adding surfactants could effectively improve their electrochemical performance. In particular, the samples obtained from nonionic surfactants (PEG and PVP) showed better performance than that from ionic surfactant (SDBS). Since nonionic surfactants played dual roles of electrostatic stabilization and steric stabilization (26, 27), the performance improvement may be explained by the following reasons. PVP is a kind of hyperdispersant, and its molecular structure includes two parts: one is the typical anchoring group and the other is the solvent chain. When it was added as a surface dispersant, on the one hand, the multivariate alkyl residue and cyano group (—NCO) as anchoring group in PVP structure could be tightly adsorbed on the surface of particles by chemical bond, hydrogen bond or ionic bond, and played a role of electrostatic stabilization. On the other hand, the lateral polarity groups of pentabasic ring could act as soluble chain, which fully stretched in the solution and formed a steric hindrance layer with an adequate thickness on the surface of particles. Thus, the steric hindrance effect could be produced, which hindered the flocculation and agglomeration of nanoparticles.

In order to further study the stability of Sample d, cyclic charge/discharge tests were carried out at 120 mA/g. Figure 7 shows the specific capacitance of the electrode material keeps stable at the initial stage, and still remains 91.1% of the initial value or 217.46 F/g after 500 cycles. This further confirmed that the prepared amorphous hydrated MnO₂was a good electrode material for supercapacitors.