In-Process Characterization of Antisolvent Crystallization - METTLER TOLEDO

In-Process Characterization of Antisolvent Crystallization

  • Ensure fast and efficient scale-up by optimizing crystallization early in development
  • Target particle size specifications to speed up downstream processing
  • In-process tools improve productivity and safety by eliminating the need for sampling and offline measurement

In-process tools can provide a deeper understanding and allow for faster characterization and development of crystallization processes. Effective implementation of in-process tools allows crystallization processes to be monitored in real-time with no need for sampling. Sampling can be time consuming and non-representative - especially for crystallization systems where samples can change due to nucleation, growth, agglomeration and/or breakage once removed from the crystallization vessel. In-process tools may also be implemented at a range of scales making it possible to compare crystallization experiments performed in the laboratory with batches run in the pilot plant or at the full productions scale. This is especially important for anti-solvent systems where changes in mixing conditions between scales can make effective scale-up and subsequent optimization of manufacturing processes difficult.

In this study, the impact of antisolvent addition rate on crystal size and morphology is assessed through experiments based on the unseeded crystallization of benzoic acid from ethanol-water mixtures using water as the antisolvent. The results of two experiments are presented in a summarized form, one at "fast" antisolvent addition rate – 0.2 gs-1 – and one at "slow" antisolvent addition rate – 0.1 gs-1. Supersaturation, monitored using ReactIRTM, was approximately twice as high for the fast addition rate compared to the slow addition rate. Lasentec® FBRM® distributions indicate that a fast addition rate produces significantly smaller crystals compared to the slow addition rate. Lasentec®  PVM® images confirm these findings and show that the crystals tend to be finer needle-shaped crystals with significant agglomeration at the faster addition rate.

Undersaturated solutions of benzoic acid in aqueous ethanol solutions were prepared and held at a constant temperature of 25ºC.  Benzoic acid is an organic compound, poorly soluble in water but soluble in ethanol with no known polymorphs cited in the literature. Water was added at a fixed rate of 0.1 gs-1 and 0.2 gs-1 respectively and the resulting crystallization was monitored using in-situ process analytical tools. Lasentec® FBRM® and PVM® were used to track the crystalline solid phase and ReactIRTM (ATR-FTIR spectroscopy) was used to monitor the liquid phase i.e. the supersaturation The experiments were performed in a LabMax® vessel allowing the temperature and anti-solvent addition rate to be programmed and controlled automatically.
Results and Discussion

Figure 2 (a) shows concentration profiles for both addition rates as well as the solubility curve for benzoic acid in ethanol-water. The concentration profile shows that the solution begins in the undersaturated region and as water is added becomes supersaturated. In each case, the Metastable Zone Width (MSZW) is narrow with a slightly wider MSZW observed at the fast addition rate. The concentration decreases upon crystal nucleation, stays close to the solubility curve, and at the end of antisolvent addition period drops to the solubility curve. The prevailing level of supersaturation is clearly seen in Figure 2(b) where anti-solvent concentration vs. supersaturation is displayed. Evidently, at the faster addition rate, supersaturation is higher. This is important information as supersaturation is the driving force for crystallization and influences the nucleation and growth rates. Differing levels of supersaturation typically lead to different final product size.  In this case with Lasentec® FBRM® and PVM®, it is possible to quantify the difference in the product distribution.

Figure 3 shows Lasentec® FBRM® distributions at the end of the experiment for the fast and slow addition rates. Clearly, the fast addition rate has produced significantly more fine material and the slow addition rate has produced larger crystals. The difference can be quantified by examining the mean of the FBRM® distributions. Operating at the slow addition rate results in a mean value of 39μm compared to 30μm for the fast addition rate, an increase of almost 30%. As expected, the slower addition rate and resultant low level of supersaturation has resulted in growth being favored over nucleation and a larger final crystal size. The fast addition rate and subsequent high levels of supersaturation favors nucleation over growth, resulting in a smaller average crystal size in this case.
Additional information can be gathered from the same experiment by studying the Lasentec® PVM® images. Figure 4 shows PVM®  images at both addition rates and it is clear that the fast addition rate results in agglomerated fine material whereas the slow addition rate results in large well formed needle-like crystals. This information is valuable as excessive agglomeration can lead to purity issues as well as difficulties in the filtration stage where agglomerate breakage is possible.

This application note highlights the high quality information that may be gathered in real time for the characterization of crystallization systems. In this case the impact of addition rate on particle size and morphology was studied and the mechanism for the difference was confirmed; elevated supersaturation at higher addition rates. The impact of other process parameters including operating temperature, mixing regime, antisolvent used, addition location, initial solute concentration and seeding protocol may all be studied in a similar fashion for a comprehensive development of a robust crystallization process. By utilizing in-process tools at the small scale, crystallization development is facilitated and scale-up can be expedited.


  1. Monitoring and Feedback Control of Supersaturation Using ATR-FTIR to Produce an Active Pharmaceutical Ingredient of a Desired Crystal Size. Liotta & Sabesan, 2004, Organic Process Research and Development, 8, 3, 488-494
  2. Paracetamol Crystallization Using Laser Backscattering and ATR-FTIR Spectroscopy: Metastability, Agglomeration, and Control, Fujiwara et al, (2002), Crystal Growth and Design, 2, 5, 363-370
  3. Process Control of Seeded Batch Cooling Crystallization of the Metastable -Form Glycine Using an in Situ ATR-FTIR Spectrometer and an in SituFBRM Particle Counter, Doki et al, (2004), Crystal Growth and Design, 4, 5, 949-953
  4. Iron Oxy-hydroxide Crystallization in a Hydrometallurgical Residue, Loan et al, (2002) Journal of Crystal   Growth, 235, 1, 482-488

METTLER TOLEDO would like to thank Dr. Brian Glennon and the School of Chemical and Bioprocess Engineering, University College Dublin, Ireland.

Thank you for visiting We have tried to optimize your experience while on the site, but we noticed that you are using an older version of a web browser. We would like to let you know that some features on the site may not be available or may not work as nicely as they would on a newer browser version. If you would like to take full advantage of the site, please update your web browser to help improve your experience while browsing