2.2.2. Preparation of Sol-Gel pH-Sensitive Membranes
The preparation steps of the pH-sensitive membranes were as follows:
(1) A certain amount of cellulose acetate was dissolved in DMF, stirred thoroughly, and placed in an ultrasonic cleaner for 2 min, before the solution was placed on a magnetic stirrer and continued stirring at 60 °C for 5 h.
(2) A mixed solution was prepared using a certain volume of TEOS, absolute ethanol, hydrochloric acid, and deionized water with a fixed molar ratio.
(3) The solutions prepared in (1) and (2) were mixed, and the four indicators (congo red, bromophenol blue, cresol red, and chlorophenol red) were added; the mixed solution was placed in an ultrasonic cleaner to vibrate for 5 min so as to dissolve the indicator completely.
(4) The mixed solution obtained in (3) was stirred at 50 °C on a magnetic stirrer until a homogeneous transparent sol was formed, and the sol was left for 12 h to allow foaming and aging.
(5) The sol was dried in the vacuum oven at 70 °C.
2.2.3. Optimization of Sol-Gel pH-Sensitive Membrane
While preparing the sol–gel pH-sensitive membranes, the reaction conditions for the sol-gel precursor and the amounts of each component of the sensitive membrane, including cellulose acetate, DMF, indicators, hydrochloric acid, and precursor glue needed to be optimally determined.
(i) Reaction Conditions of Sol Precursor
The hydrolysis and polycondensation reaction of TEOS under acid catalysis was mainly influenced by several factors, such as the amounts of ethanol, water, and hydrochloric acid, as well as the dropping speed and the temperature.
(1) Amount of Ethanol
The sol-gel precursor solution was prepared using TEOS as the raw material. Because of the limited solubility of TEOS in water, a certain amount of co-solvent was needed. The co-solvent needed to be soluble with TEOS and water; as such, ethanol was chosen as the co-solvent. The ternary miscibility diagram of TEOS, C
2H
5OH, and H
2O is shown in
Figure 1. From
Figure 1, it can be concluded that the addition of less ethanol would lead the ternary solution system to be immiscible, but the addition of more ethanol would immobilize the hydrolysis reaction of the system. In this paper, the molar ratio of ethanol to TEOS was 2:1.
(2) Amount of Water
When the molar ratio (R0) of water to TEOS was less than 4, the gelation time decreased with increasing R0. TEOS could not be fully hydrolyzed while R0 was less than 4. As R0 increased, so did the concentration of Si–OH groups formed after hydrolysis of TEOS. At this time, the polycondensation reaction proceeded more quickly which decreased the gelation time. On the contrary, if R0 was more than 4, TEOS was be completely hydrolyzed and the concentration of Si–OH groups formed after hydrolysis of TEOS decreased with increasing R0. Hence, the reaction between Si–OH groups through collision was attenuated, resulting in an increased gelation time. Furthermore, as one of the polycondensation products, more water being added resulted in the polycondensation reaction being hindered. The relationship between the water–silica ratio and the system’s gelation time is shown in
Figure 2.
Moreover, the amount of water added also affected the drying process of the sensitive membrane, which aggravated its cracking phenomenon. In our work, the final molar ratio of water to TEOS was 4:1.
(3) Dropping Rate and Amount of Hydrochloric Acid
In the sol-gel reaction system, hydrochloric acid can accelerate the rates of hydrolysis and polycondensation, and promote the expansion of TEOS, ethanol, and water from group A in the ternary mixed zone to group B, as shown in
Figure 3. In this work, the exact amount of hydrochloric acid was determined using the orthogonal test [
21].
During the reaction of hydrochloric acid with TEOS, the titration rate of hydrochloric acid also had a certain influence on the sol-gel.
If the dropping rate of hydrochloric acid is too fast, a white precipitate, instead of the homogeneous gel, is formed in the TEOS solution due to the fast local hydrolysis. By means of dropwise addition, the speed of the hydrolytic polycondensation of TEOS was slowed down so that the reaction could be carried out steadily, which was beneficial for the generation of a homogeneous gel. In this work, deionized water and hydrochloric acid were mixed thoroughly, and the homogeneous precursor solution was prepared by means of dropwise addition.
(4) Temperature
The relationship between the temperature and gelation time is shown in
Figure 4, where it is shown that the gelation time decreased with an increase in temperature.
As the temperature increased, the hydrolytic activity of TEOS was enhanced and the average kinetic energy of the water molecules increased, which increased the probability of collisions between TEOS molecules and water molecules. As a result, the hydrolysis reaction was accelerated, and the gelation time of the mixed solution was reduced. In this work, the temperature was set at 50 °C.
(ii) Optimization of Material Dosage
In this work, the orthogonal test was used to optimize the dosage of the sensor-sensitive membrane materials, including cellulose acetate, DMF, indicator, hydrochloric acid, and precursor glue. With fewer trials, better production conditions or better preparation technology can be obtained using the orthogonal test table (i.e., the optimal ratio of sensitive film materials can be quickly and efficiently obtained).
The orthogonal experiment scheme for the composition ratio of the materials is shown in
Table 1, and the optical pH sensors based on sol–gel were prepared accordingly. In total, five factors (cellulose acetate, mixed indicator, DMF, hydrochloric acid, and precursor) were considered and four levels of experiment were conducted. The optimal recipe was determined by comparing the slope of the fitted straight lines of different pH values.
The orthogonal experimental results are shown in
Table 2, where the slope averages of fitting lines for each level K
i (i = 1, 2, 3, 4) are listed at the last four lines. The level corresponding to the max average slope was selected as the optimal value for that factor; thus,
was determined as the optimal combination. However, too much DMF would hinder the drying of the sensitive membrane, leading to easy cracking. The experimental results showed that the preparation of the sensitive membrane with a volume of DMF between 15 mL and 20 mL had a higher successful rate. Hence, the final recipe included 0.1 g of cellulose acetate, 8 g of mixed indicator (congo red, bromophenol blue, cresol red, and chlorophenol red, 2 g each), 15 mL of DMF, 50 µL of hydrochloric acid, and 7.5 mL of precursor. The final optimized sensitive membranes are shown in
Figure 5.