Prof. Junbo Gong
Professor, School of Chemical Engineering and Technology
Tianjin University

KEY WORDS: Crystal habit control, Mechanism of crystal and particle growth, Additive and Solvent effect, Molecular dynamic simulations, Spherical crystallization, Design strategy

In the solution crystallization, organic crystals are generally assembled based on hydrogen bonds between molecules, as hydrogen bonds have directionality, organic crystals often grow into needle-like or lamellar crystal habits, with low bulk density and poor flowability. However, due to the lack of in-depth understanding of crystal growth mechanism and the impeccable crystal habit prediction model, crystal habit optimization relies heavily on empirical operation with high cost and low efficiency.

Aiming at the problems of poor crystal habit and low purity of D-PC, the crystal habit regulation of D-PC was realized based on the crystal form control. The crystal forms and habits of D-PC at different solvents were fully characterized, and calcium Dpantothenate tetramethanol monohydrate (D PC·4MeOH·H2O) with block habit and low solvent removal temperature, was considered as the suitable crystal form for industry production. The phase transformation from Calcium D-pantothenate methanol solvate (D-PC·MeOH) to·D-PC·4MeOH·H2O was monitored by in situ ATR-FTIR and Raman spectroscopy, the results demonstrated that the rate-controlling step was the nucleation and growth of D-PC·4MeOH·H2O. The desolvation process of DPC·4MeOH·H2O was studied by variable-temperature PXRD and Hot Stage Polarized Microscopy, and the solvent-free amorphous D-PC with block habit was obtained. The molecular dynamics simulations were performed to reveal the temperature role in the phase transformation process at the molecular level. This work provides a layout to optimize the crystallization of D-PC in the future.

The effects of the concentration and the chain length of sodium alkylsulfate (SDS), sodium alkylsulfonate (SLS), and sodium alkyl benzenesulfonate (SDBS) on the growth of VB1 were studied by single crystal growth experiments. It was found that SDS and SDBS exhibited significant inhibitory effects on VB1 crystal growth along the axial than radial axis, therby modifying VB1 from long rod to block; while SLS had a very slight inhibition on VB1 crystal growth, and exhibited almost no effect on VB1 habit. The mechanism of additive in regulating VB1 crystal habit was proposed by combining experiment and molecular dynamics simulation; all the three kinds of additives showed slight inhibitory effects on VB1 crystal growth by hindering the diffusion of solute. However, compared with SLS, SDS and SDBS can selectively adsorb on the VB1 crystal surface along axial axis and inhibiting its growth through strong electrostatic and hydrogen bonding.

The crystal growth of VB6 in the absence and presence nonionic and ionic surfactants was investigated by single crystal growth experiments and theoretical calculations. In experiments, VB6 crystals exhibited block-like habit in aqueous and non-ionic surfactant solution, while needle-like VB6 crystals were observed in ionic surfactant solutions. The mechanism of surfactants in modulating VB6 crystal morphology was proposed by calculating the electrostatic potential of the surfactants; the preferential adsorption of sodium dodecyl sulfate (12 SDS) on the radial (010) surface inhibiting its growth over the axial (100) surface results in high aspect ratio of VB6 crystals. Hexadecyl trimethyl ammonium bromide (CTAB) enhances both (100) and (010) surfaces growth by promoting Cl- ions integration to the surfaces. Dodecyl dimethyl betaine (DDMB) hinders (010) surface growth by surface adsorption while promotes (100) surface growth by accelerating Cl- integration. In addition, Tween 80 has a slight inhibitory effect on (100) and (010) crystal growth by hindering solute transport through steric hindrance.

The constant chemical potential dynamics simulations (CμMD) was constructed to quantitatively investage the crystal growth rate of INH in different solvents, and uncover the growth mechanism of INH. Experiments showed isoniazid grew as needlelike crystals in water, while in alcohols such as methanol, ethanol and isopropanol, it exhibited a rod-like crystal habit. The simulation results revealed a rough growth mechanism for the fast growing (110) surface along the of INH crystal, and bulk transport of the solute was the limiting-step, the relative growth rate of this surface decreased from methanol, ethanol to isopropanol. On the other hand, the slow growing (002) surface along the axial direction appeared to follow a stepwise growth mechanism, with a surface integration step chiefly controlling the growth. The relative growth rate of this surface increased from methanol to ethanol and isopropanol.

Spherical crystals are special crystalline products with enhanced performance,unique functions and extra values. It has become a hot spot in the high-tech fields of pharmaceutical, food, daily chemical, fertilizer, and military industries. However, it is still challenging for preparing the given crystals into spheres with desired performance.To solve this issue, three universal strategies were developed for designing spherical crystallization processes, i.e. spherical agglomeration, spherulitic growth and multicluster growth.

Strategies of selecting ternary solvent systems and process control for spherical agglomeration were developed. The strategy of selecting solvent systems emphasizes wettability as the key parameter which is quantitative estimated by the Lifshitz-van der Waals acid-base approach. Valid systems therefore are selected in the first place while invalid ones are removed effectively according to the wettability difference. With the aid of the strategy, 4 valid solvent systems for cefotaxime sodium from 720 solvent combinations and 24 valid ones for benzoic acid from 2184 solvent combinations were selected successfully without missing the reported systems. The strategy of process control is designed with the independent size space and shape space, so that the quantitative relationship between process parameters and morphology of spheres can be clearly established. This strategy successfully tuned the spherical agglomerate of benzoic acid from 1000-5000 μm avoiding wide size distribution and fragments by the conventional strategy.

A strategy of designing spherulitic growth process was developed. It predicts the critical superstation of spherulitic growth with the rough growth model. Based on the strategy, spherulites of clopidogrel hydrogen sulfate (Form I) and fructose were successfully prepared. In the case study of clopidogrel hydrogen sulfate, it indicated the theoretical drawbacks of the single solvent system designed by trial-and-error method in literature, and designed the mixture system achieving the targets of spherical shape, no crystal form transformation, and no jelly-like phase.

A novel spherical crystallization method was developed. According to crystalline cluster growth on crystal surface and attrition effect of agitation, it prepares spherical crystals in a single solvent (water) system under a common condition of supersaturation. Its strategy was developed with multi-cluster growth model and agglomerate size model.Spherical crystals of potassium chloride, sodium chloride and cesium iodide were prepared successfully by the strategy.