1 Introduction

Mesoporous carbons (MCs), owing to their low mass density, tunable pore size, high surface area, high electrical conductivity, high chemical stability and other interesting properties [1], have been used in lots of fields such as storage of hydrogen and methane [2], electrochemical capacitors [3, 4] and lithium ion battery [5], molecular adsorption and separation [6], catalyst support [7], etc. Conventionally, MCs were successfully prepared through the so-called nanocasting technique [8–11]. Although the nanocasting pathway is a useful tool for the creation of nanostructured carbon materials, the disadvantages (such as the long process and using toxic chemicals to remove the templates) are also very obvious.

Recently, organic–organic self-assembly was used to prepare MCs with tunable pore size and structure. Many scientists make substantive efforts and have obtained significant achievements [12–21]. For example, Ueyama et al. [18] synthesized MCs by an organic–organic method using resorcinol (R)/formaldehyde (F) and two copolymers ((PO: propyleneoxide; EO: ethyleneoxide) such as F127(EO106PO70EO106) and P123(EO20PO70EO20)) under basic condition, and the pore size could be tuned from 4.4 to 6.8 nm by changing the ratio of R/F or using mixed triblock copolymers. Yuan et al. [19] prepared MCs with ultra-large pore size with the use of triblock copolymers and phloroglucinol/formaldehyde polymer as filler using 37 % HCl as catalyst, and the pore size of MCs could be tailored from 11.5 to 14.7 nm when using P123 as template with the decane as a swelling agent. Furthermore, the MCs exhibit large pore at 19.2 nm with F127 was used as a template to be a substitute for P123. Hayashi et al. [15] also synthesized MCs with pore size from 7.0 to 12.5 nm using different ratios of R/F to F127 and polymerization time. The common drawback in all the above mentioned studies is the need of an acid or basic catalyst due to the low reactivity of phenol or resorcinol with formaldehyde, what will limit their potential application. Liu et al. [20] even pointed out that the concentration of acid and the reactivity of phenols are two key factors which determine the polymerization rate of phenolic resins. It is well-known the phloroglucinol has three hydroxyl groups [22–24], so herein we try to utilize the acidity of phloroglucinol itself to catalyze the reaction with formaldehyde producing resin with more hydroxyl groups which can facilitate a higher degree of cross-linking network. Furthermore, lots of works have been done on the synthesis of MCs with tunable pore size and structure by using mixed copolymers or a pore expanding agent. However, it is still difficult to tailor the pore size of MCs by an individual copolymer template without the addition of small organic molecules such as trimethylbenzene or long-chain n-alkanes as a swelling agent. In this work, we report the direct preparation of MCs using phloroglucinol–formaldehyde composites as carbon precursor and reverse amphiphilic triblock copolymer PO97EO186PO97 as template through soft-template method as shown in Fig. 1.

Fig. 1
figure 1

Scheme for the preparation of MCs

Full size image

2 Chemicals and experimental

PO97-EO186-PO97 (MW = 19,500, purchased from Nanjing Co. Limit. Chem. Com). Phloroglucinol, (C6O3H6; 99 %), formaldehyde (HCHO; 37 %), purchased Tianjin Co. Limit. Chem. Com.

Typically, PO97-EO186-PO97 (5 g) was dissolved in ethanol (24 g) under magnetic stirring at 30 °C. Then, the resin solution was synthesized as follows, 3 g of phloroglucinol were dissolved in ethanol (24 g), the pH of solution was about 5–6, followed by addition of a certain amount of 37 % formaldehyde and 8 g of ethanol with stirring at 30 °C for 2 h. Subsequently, the resin solution was added into the ethanol solution containing PO97-EO186-PO97, and stirred for 2 h to form a homogeneous solution. A brown film was obtained by pouring the solution into a dish and allowing the ethanol to evaporate at room temperature for about 24 h, then heating in an oven at 100 °C for 24 h. The removal of template in this manuscript is by direct carbonization at 700 °C just like the references [14, 25] because the template remains only negligible 3.1 % residue at 400 °C. MCs were obtained by direct carbonization of the as synthesized samples at 700 °C for 3 h at a heating rate of 1 Kmin−1 under N2. (here the samples can be denoted as MCs-X, where X is the weight of the formaldehyde, X = 1.0, 1.2, 1.4 1.6 and 2.0, respectively).

Nitrogen adsorption–desorption isotherms were performed at 77 K on a Micromeritics ASAP-2000 volumetric adsorption system. The specific surface area was calculated from the adsorption data in the relative pressure interval from 0.05 to 0.35 using the Brunauer-Emmett-Teller (BET) method. The pore size distribution curve was gained from adsorption branch by using Barrett-Joyner-Halenda (BJH) method. The total pore volume (Vtotal) was calculated at the relative pressure of 0.99. The micropore volume (Vmicro) was determined by t-plot model, and the mesopore volume (Vmeso) was calculated by the difference of Vtotal and Vmicro. Small angle X-ray scattering (SAXS) recorded using an imaging plate with X-ray wavelength of λ = 1.54Å at beam line 1W2A at the Beijing Synchrotron Radiation station. Transmission electron microscopy (TEM, Hitachi operated at 200 kV).

3 Results and discussion

The SAXS patterns of MCs are shown in Fig. 2. It can be found that all samples, excepting MCs-1.2, do not show well resolved peaks at low angles, indicating the disordered pore structure of samples. It is noticeable that the MCs-1.2 gives a small and weak peak in low angle area (as exhibit in the inset of Fig. 2), suggesting that MCs-1.2 present, in some extent, an ordered mesoporous structure [26].

Fig. 2
figure 2

SAXS patterns for MCs

Full size image

The N2 adsorption/desorption isotherms of the typical samples are shown in Fig. 3a, all samples demonstrate mesoporous character according to the IUPAC classification. It is interesting when the content of formaldehyde varies, so does the shape of the hysteresis loops in MCs. For example, the MCs-1.0, 1.2, 1.4 exhibit a characteristic similar type-IV curve with an obvious H1 hysteresis loop with leap start-points at relative pressures of P/P0 = 0.65, which is typical of uniformly sized mesoporous structure consisting of very large mesopores with an open cylindrical pore geometry [27, 28]. The steep capillary condensation steps indicate that corresponding pore size distributions are narrow as shown in Fig. 3b.

Fig. 3
figure 3

N2 adsorption/desorption isotherms (a) and BJH adsorption pore size distributions of MCs (b)

Full size image

As the content of formaldehyde increases to 1.6 or 2.0, both samples exhibit a characteristic type-IV curve with an H2 hysteresis loops at about relative pressure (P/P0) between 0.4 and 0.8 as demonstrated in Fig. 3a, which implies smaller mesopore size and cage-like type structure [28, 29]. Accordingly, the MC-1.0 has the biggest volume 0.78 cm3/g and MC-2.0 has the smallest volume 0.28 cm3/g as shown in Table 1.

Table 1 Pore parameters of MCs
Full size table

The pore size distributions of MC-1.0, MC-1.2, MC-1.4, MC-1.6 and MC-2.0 are as shown in Fig. 3b. Apparently, it can be found there is a decreasing trend in the pore size distribution with the amount of formaldehyde increasing. For example, the pore size of MCs decreases from 14.8 to 4.9 nm with the content of formaldehyde increasing from 1.0 to 2.0 as shown in Fig. 4. Furthermore, the pore volume also has the similar tendency with the variation of formaldehyde. The mentioned changes in the textural parameters are similar to those observed under basic NaOH condition [25], further confirming our model to tune the pore size and structure.

Fig. 4
figure 4

The pore size and volume variation as a function of the weight content of formaldehyde

Full size image

Figure 5 shows the typical TEM images of MCs. It can be observed that MC-1.2 presents a hexagonal arrangement as demonstrated in Fig. 5a, which is consistent with SAXS result. With the formaldehyde content increasing, the pore structure has transformed into cagelike or ink bottle pore as indicated by the TEM image of MC-2.0 in Fig. 5b. These above results can be explained as follows: the different content of formaldehyde plays a crucial role in the process of formaldehyde/phloroglucinol composite, which further affects the interaction with surfactant through hydrogen bond. The formation of unusual lyotropic phase from reverse triblock copolymer as well as the hydrogen-bonding interaction may also play an important role in the formation of mesoporous structure.

Fig. 5
figure 5

TEM images of a MC-1.2-700 and b for MC-2-700

Full size image

4 Conclusion

In conclusion, MCs with controllable pore size and structure were fabricated using an easy organic–organic process without any acid or basic catalyst. Both the final mesopore size and the mesostructure of MCs can be controlled by changing the amount of formaldehyde. Thus, by increasing formaldehyde ratio, the mesopore size can be tuned from 14.8 to 4.9 nm and the mesostructure symmetry can be changed from hexagonal to cubic. The amount of formaldehyde determines the degree of polymerization, which in turn, determines the interaction resin-Pluronic and the correct replication of the soft-template without a collapse.