Effects of acid hydrolysis conditions on cellulose nanocrystal yield and properties: A response surface methodology study

Cellulose nanocrystals (CNCs) are frequently prepared by sulfuric acid hydrolysis of a purified cellulose starting material. CNC yields, however, are generally low, often below 20%. This study employs response surface methodology to optimize the hydrolysis conditions for maximum CNC yield. Two experimental designs were tested and compared: the central composite design (CCD) and the Box–Behnken design (BBD).

The three factors for the experimental design were acid concentration, hydrolysis temperature, and hydrolysis time. The responses quantified were CNC yield, sulfate group density, ζ-potential, z-average diameter, and Peak 1 value. The CCD proved suboptimal for this purpose because of the extreme reaction conditions at some of its corners, specifically (1,1,1) and (–1,–1, –1). Both models predicted maximum CNC yields in excess of 65% at similar sulfuric acid concentrations (~59 wt %) and hydrolysis temperatures (~65 °C).

With the BBD, the hydrolysis temperature for maximum yield lay slightly outside the design space. All three factors were statistically significant for CNC yield with the CCD, whereas with the BBD, the hydrolysis time in the range 60–150 min was statistically insignificant. With both designs, the sulfate group density was a linear function of the acid concentration and hydrolysis temperature and maximal at the highest acid concentration and hydrolysis temperature of the design space. Both designs showed the hydrolysis time to be statistically insignificant for the ζ-potential of CNCs and yielded potentially data-overfitting regression models. With the BBD, the acid concentration significantly affected both the z-average diameter and Peak 1 value of CNCs.

However, whereas the z-average diameter was more strongly affected by the hydrolysis temperature than the hydrolysis time, the Peak 1 value was more strongly affected by the hydrolysis time. The CCD did not yield a valid regression model for the Peak 1 data and a potentially data-overfitting model for the z-average diameter data. A future optimization study should use the BBD but slightly higher hydrolysis temperatures and shorter hydrolysis times than used with the BBD in this study (45–65 °C and 60–150 min, respectively).