Virtual Spray Drying
Alex Read & Simon Lo, CD-adapco

Introduction

Throughout the Chemical, Pharmaceutical and Food industries, Computational Fluid Dynamics (CFD) is increasingly being embraced as part of the Virtual Product Development (VPD) process. This article details how CD-adapco has used its CFD expertise, in partnership with the application excellence of NIZO Food Research. The result is an easy-to-use, upfront CFD product for calculating the flow in spray driers: es-spraydry. es-spraydry enables Design Engineers - who are not CFD 'gurus' - to evaluate design critical parameters in a virtual environment, saving time and money.

Spray drying is used to turn a liquid into a powder: examples being detergents, herbal extracts, instant coffee and milk. These systems have design critical parameters which historically have been optimized through expensive and time consuming physical testing. The material residence time is paramount: it must be long enough to remove moisture but not so long as to cause over-heating. Optimizing the drying process is non-trivial, since it is dependant on a number of factors. These include: flow regime, temperature distribution in the system, humidity, initial particle temperature and moisture content, and particle drying characteristics.

A key strength of CFD is the ability to carry out “what-if” and optimization analyses quickly. As an example, a dryer with a given set of feed conditions was considered. CFD simulations were carried out with the aim to find the optimum condition for the drying air. Key information of interest to the plant operator was extracted from the CFD results and presented in percentages of particles leaving the particle and air exits, and the particle conditions at these exits in terms of mean diameter, temperature and moisture content. From these results, the operator of the dryer can easily select the optimum operating condition, which allows him to achieve the desired product quality at minimum cost.

Behind es-spraydry

Historically, companies have been reluctant to attempt such calculations since the initial investment – both in terms of personnel and hardware – was deemed prohibitively great. Complex, multiphase calculations were the exclusive domain of PhD educated specialist Engineers with large, expensive computers at their disposal. These requirements have recently become less and less stringent. Even relatively complex calculations can now be carried out on hardware obtainable on the high street for just a few hundred Dollars: such as the examples given here. But what about the knowhow required performing such calculations?

CD-adapco has long identified industry’s requirement for process specific tools: termed ‘es’ (Expert System) solutions. The model for these tools is that they encapsulate the CFD set-up, running and post-processing process, in an environment that is accessible to design engineers without specialist training. Thus, the requisite knowledge-transfer to get new starters up, running and adding value to the design process is greatly reduced.

The prerequisites to produce such a tool are twofold. First, you must be familiar with the design process in question. Second, you must be familiar with the modeling software and algorithms. This is where the partnership between CD-adapco and NIZO is so important.
CD-adapco is a CAE company – now in its 25th year – of unparalleled experience and expertise. NIZO has been researching spray drying since the 1950s and has extensive experimental, pilot plant and real plant operating experience. In this, experimental data used for validating CFD results is included Together, they have produced es-spraydry.

The modeling process

In every CFD analysis it is necessary to go through several steps to build the computational model, carry out the computation and analyze the results as follow:

1. Define the shape and dimensions of the dryer;
2. Specify the inlet and outlet configurations;
3. Define the atomizer and the spray characteristics;
4. Specify the flow rates of feed and drying air;
5. Build a computational mesh;
6. Run the solver to obtain a converged solution;
   
7. Analyze the solution and produce a performance report of the dryer.
   

This sequence of steps provides us with a well defined methodology for applying CFD in spray drying analyses. This methodology has been encapsulated, with the steps defined above automated, paying particular attention to ensure that the inputs and outputs are engineering values rather than CFD-speak. This was achieved through close collaboration with NIZO design engineers. Therefore, the end result from an es-spraydry simulation is not only detailed flow visualization images, but also an automatically generated report, providing engineers with the all important design metrics.

Validation

Key to any modeling process is a validation stage. Here again, users benefit from CD-adapco and NIZO’s years of experience and validation work. Just one example is the validation of spray models in es-spraydry and STAR-CD. In Figure 2, a direct comparison is made between STAR-CD results and Schlieren spray images at discrete time increments after injection in order to analyze the spray development and spreading. Due to the temporal delay of the swirl flow development after injection, the injection starts with a straight pre-jet and subsequently turns into a hollow cone spray. This effect is clearly seen in the visualization of analysis results.

Figure 2 also demonstrates the level of detailed understanding yielded by CFD analyses. As shown, such information can be obtained from experimental work, but only on highly idealized geometries and even then at great cost.

Industrial example

We have a simple dryer, shown in Figure 3, typical of that used in industry and evaluated by NIZO for producing dried milk powder. The outer diameter of the dryer is 9.5 m and an overall height of 14 m. There is one central air inlet and four circular air outlets. A rotary wheel atomizer spinning at 2600 rpm is used for the atomization of the feed.

The droplets produced are assumed to have a log-normal size distribution and a mean diameter of 100 micron with a geometric standard deviation of 0.6.

The dryer is required to process a feed at a rate of 4990 kg/h, a temperature of 25ºC and a solid content of 55% w/w (weight/weight). Drying air is to be supplied at 185ºC with a moisture content of 1%. We need to find the optimum air flow rate which will satisfy the following requirements:

1. 90% of particles exit the bottom particle exit.
   
2. Mean particle moisture content is less than 9%.
   
3. Mean particle temperature is less than 100ºC.
   

From the es-spraydry results we monitored the particle conditions exiting the dryer at the particle and air exits. Several es-spraydry calculations were performed with different air flow rates.

The results were further analyzed against the operation requirements, shown graphically in Figures 4 to 6. From the analysis we would select an air flow rate of 55,000 kg/h for minimum operating cost in terms of supplying the drying air.

The es-spraydry model used in the analysis has 20902 cells. 100 parcels of droplets were used to represent the spray. Converged solutions for all cases were obtained within 200 iterations. The CPU times for the cases range from 4000 to 7555 seconds on an Intel P3, 1.2GHz computer.

Conclusion

Historical barriers to running design-enhancing CFD simulations are quickly becoming a thing of the past. Through dedicated CFD software such as es-spraydry, the CFD analysis process has been simplified and automated. Close collaboration between CAE code vendors (CD-adapco) and experienced ‘in the field’ organizations (NIZO) has yielded an easy-to-use upfront CFD analysis tool, capable of adding value to any spray drying design process.

With this simulation tool it is now possible for the spray dryer operators to carry out systemic analyses of their dryers to ensure they operate at optimal conditions.