Choice of Polysaccharide
The choice of polysaccharides is a key aspect of the synthesis of Starbon. Not all polysaccharides have been successfully expanded and carbonised using the Starbon approach, notable examples which have proved impossible being cellulose and chitosan. On the other hand, various starches, alginic acid [10], and pectin [11] all readily form mesoporous structures under the appropriate conditions, and these mesostructures can be successfully translated into carbonaceous materials. Recently, Attard et al. [12] have demonstrated that chitosans can be successfully incorporated into Starbon materials via mixing with alginic acid over a range of compositions, maintaining porosity while, at the same time, incorporating nitrogen.
3.2.1 Synthetic Procedure
3.2.1.1 Gelation
The formation of Starbon, irrespective of the nature of the polysaccharide, has three distinct stages (gelation, drying, and pyrolysis). First, gelation of the polysaccharide takes place in water to swell the material (starch, for example, exists in nature as densely packed granules and functions as an energy source) to produce a hydrogel. The gelation is promoted by a combination of temperature and shear forces, which aid in breaking up the tightly packed grain, allowing some of its contents to move outwards, creating porosity. A subsequent step is a retrogradation, where the gel is allowed to ‘rest’ at 4 °C for approximately 24 hours. This causes slight structural changes and stabilisation.
3.2.1.2 Drying of the Hydrogel
A critical stage is the subsequent drying step, as direct drying of the hydrogel leads to extensive collapse. This is due to the material being relatively soft and structurally weak compared to the significant capillary forces generated on the evaporation of water from the surface of a meniscus (Figure 3.1). These forces act to pull the walls of the pore together, causing collapse if the material is mechanically weak, which is the case with soft polysaccharide gels.
The fact that water has high surface tension and that it forms menisci in nanopores means that direct drying of the hydrogel puts the structure under great stress as the evaporation causes the pore walls to be pulled together – these forces are too great for the porous starch network to survive and structural collapse ensues. Roundabout methods, therefore, need to be utilised in order to avoid these forces and maintain the porosity.
The first methodology utilised solvent exchange – water has a much higher surface tension (72.74 mN m−1 at 20 °C [13]) than most organic liquids (e.g. acetone 23.70 mN m−1, ethanol 22.27 mN m−1, and methanol 22.60 mN m−1 [14]).
Therefore, exchanging water first for ethanol by simple mixing and filtration (×3) led to approximately 12% water in ethanol‐filled gel, which had to be further exchanged with acetone using the same methodology in order to provide an aerogel with porosity intact. What is interesting here is that the gel produced after the ethanol exchanges contains a liquid (c. 12% water in ethanol) which has a surface tension (22.6 mN m−1) [15] very close to that of acetone (23.70 mN m−1) [14], yet fails to lead to a satisfactory material. While this may be due to the evaporation of an ethanol‐rich vapour (meaning that the remaining liquid is enriched in water, leading to the surface tension of the remaining liquid increasing and causing damage), more extensive ethanol washing does not lead to better materials. Similarly, direct exchange with acetone does not provide acceptable materials. Thus, both solvents must be used for a successful material.
Figure 3.1 Role of capillary forces in the collapse of soft porous materials.
A major downside to this approach (apart from cost) is the large volume of mixed solvents that are very difficult to purify and reuse. While the materials obtained are very good, the process lacks environmental credentials.
In order to develop a more efficient, less solvent‐intensive process, Borisova et al. developed a route involving freeze‐drying of hydrogels doped with t‐butanol, a molecule often utilised in freeze‐drying to control the freezing process [16]. This approach avoids surface tension issues as the phase transition is from solid to gas, and the capillary forces that plague direct drying are thus avoided.
The optimal porosity characteristics were achieved at the eutectic points of the water – t‐BuOH system, with a more distinct optimum found at the water‐rich eutectic (c. 23% t‐butanol). At this solvent composition, the macroporosity that dominates in freeze‐drying of the pure water hydrogels is strongly reduced and considerable mesoporosity is developed. While all three polysaccharides display qualitatively similar behaviour, the values of mesopore volume (pectin: 3.1 cm3 g−1; alginic acid: 1.8 cm3 g−1; starch: 1.1 cm3 g−1) are quite different, but nonetheless impressive, in each case (Figure 3.2).
Therefore, compared with a solvent exchange, the quality of the materials (in terms of mesopore content) is excellent, and much less solvent is required as the t‐butanol can be recovered and reused.
3.2.1.3 Pyrolysis of the Expanded Aerogel
The final stage of pyrolysis involves a slow temperature ramp from room temperature to the desired temperature. This drives off water, a complex bio‐oil (typically formed between 250 and 400 °C) and smaller gases (CO2, CO, etc.) which are the dominant species at higher temperatures. With increasing temperature, the nature of the surface changes from polysaccharide‐like, gradually increasing the C : O ratio from 1 : 1 to around 30 : 1 at 800 °C. The increasingly carbon‐like nature of the material underlies profound changes in chemistry. While these changes are complex and not fully understood, globally it can be said that, for starch at least, initial dehydrations occur which lead to unsaturation (C=C and C=O functionality) leading to an enhanced reactivity. These groups react further, cross‐linking and releasing small organic molecules into the gas phase to form a broadly aromatic matrix [9]. This is somewhat of an oversimplification as significant aliphatic functionality is always seen by 13C NMR. This may be due, in part at least, to the early dehydration of sugars, which can lead initially to –CH2–C(=O)– units, although the reactivity of such groups suggests that more structural features of greater stability are initially formed. After the initial substantial mass loss, further heating causes much more gradual mass loss, predominantly in the form of small molecules such as water, CO, and CO2.
Figure 3.2 Evolution of porosity as a function of water : butanol composition. Eutectic points are at 25 wt% t‐BuOH and at 95 wt% t‐BuOH.
Source: Original data from Borisova et al. [16].
Thus, the chemical functionality of the surface of the Starbon materials represents a continuum of changing functionality ranging from hydroxylic to highly functional (hydroxylic, unsaturated) to a more aromatic, low oxygen structure.
The mesoporosity of the materials compared to the total (meso +
