CFD Simulation and Experimental Validation of Air-Assisted Atomization in a Büchi Spray Dryer

Abstract

This degree project combines computational fluid dynamics (CFD) simulations with
experimental validation using a Büchi B-290 spray dryer to study particle deposition patterns
during spray drying of a glycerol-based solution. The CFD simulations were performed using
ANSYS Fluent with a monophasic air model, where the liquid phase was represented as
discrete inert particles injected via the Discrete Phase Model (DPM). Particle deposition on
heated walls was modeled using trap boundary conditions with erosion/accretion activation.
Experimental validation was conducted using a solution containing glycerol, trehalose
dihydrate, and colorant, processed under controlled conditions (inlet temperature 138°C,
outlet temperature 80°C). The effects of particle number, particle size, mesh refinement, and
aspirator rate on deposition patterns using the computational model were systematically
investigated. Quantitative analysis using spectrophotometry revealed that the wall adjacent to
the outlet captured the highest colorant concentration (10.70 ± 0.05 % at 50% aspirator).
Temperature analysis of 10,000 particles showed that escaped particles experienced a mean
temperature increase of 9 K, while trapped particles reached a maximum temperature of 380
K. A comparison with the high-fidelity 3D Large-Eddy Simulations of the same spray dryer
by Pralits et al. [10] revealed that the primary deposition zones identified in our experiments
(walls adjacent to outlet and connection sections) are consistent with their findings. However,
our 2D planar RANS model could not capture the inherent three-dimensionality and
unsteadiness of the turbulent flow. Increasing particle size from 1 μm to 6 μm revealed a
critical transition at 5 μm, where the conical section began collecting more particles than the
outlet tube, indicating an inertia-dominated deposition regime for larger particles. The study
model provides a framework for understanding deposition mechanisms in spray drying
processes.

Popular Abstract

How Computers and Experiments Help Us Understand Spray Drying

What is spray drying and why is it important?
Have you ever wondered how instant coffee, powdered milk, or certain medicines are made?
Many of these products start as liquid and end up as a fine powder. The process that makes this
possible is called spray drying.
Imagine a tall glass cylinder where hot air blows downward while a liquid is sprayed in as tiny
droplets. As the droplets fall, the water evaporates, leaving behind dry particles that are collected
as powder. This is exactly how a spray dryer works. The Büchi B-290, the machine used in this
project, is a small-scale version commonly found in research laboratories.
Spray drying is widely used in the food and pharmaceutical industries because it is fast, gentle on
heat-sensitive ingredients, and can produce high-quality powders. However, one common
problem is that some particles stick to the walls of the drying chamber instead of being collected.
This reduces the amount of powder produced and can also cause the stuck particles to overheat
and lose their quality.
Our approach: combining computer simulations with real experiments
To better understand why particles stick to the walls and how to prevent them, we used two
complementary methods: computer simulations and laboratory experiments.
Computer simulations allow us to see inside the spray dryer in ways that are impossible with the
naked eye. Using a software called ANSYS Fluent, we created a virtual model of the Büchi B-
290. We then injected virtual particles (representing the liquid droplets) and tracked their
movement, their speed, their temperature, and whether they escaped or stuck to the walls. The
model was simplified into two dimensions (like a cross-section) to make the calculations faster
while still capturing essential physics.
Laboratory experiments were conducted using a real Büchi B-290 spray dryer. We prepared a
special solution containing glycerol (a common substance that prevents drying), trehalose
dihydrate (a common excipient and stabilizer used in formulation development), and a colorant
(to see where particles deposit). The solution was sprayed under controlled conditions while we
varied the air flow rate and other parameters. After each experiment, we washed four different
zones of the dryer (right wall, left wall, outlet tube, and conical section) and measured how much
colorant had deposited in each zone using a spectrophotometer, a device that measures light
absorption.
What did we discover?
Our experiments showed that the right wall of the drying chamber consistently captured most
particles (up to 11% of the total colorant at the lowest air flow setting). When we reduced the air

flow rate, more particles stuck to the walls because they spent more time inside the chamber
before being carried out. The total amount of particles stuck increased from about 11% at high
air flow to about 20% at low air flow.
The computer simulations gave us even more detailed information. They revealed a fascinating
pattern: particles that escape the chamber do so very quickly, within about 0.001 seconds, while
particles that get stuck can recirculate inside the chamber for up to 0.08 seconds before finally
depositing on a wall. This means that some particles travel in circles inside the dryer, passing
through hot zones multiple times, before they eventually stick.
We also studied how particle size affects where they stick. When we simulated particles of
different sizes (from 1 to 6 micrometers, which is about 50 to 100 times smaller than the
diameter of a human's hair), we found a critical transition at 5 micrometers. Smaller particles
tended to stick in the outlet tube, while larger particles penetrated deeper into the conical section
before sticking. This is because larger particles have more inertia and are harder for the air flow
to redirect.
The temperature analysis of 10,000 simulated particles showed that escaped particles warmed up
only slightly (by about 9 degrees Kelvin), while some trapped particles heated up dramatically,
reaching temperatures up to 87.2 degrees Kelvin above room temperature. This is important
because excessive heat can damage heat-sensitive ingredients in food and medicine.
How do our simulations compare to reality?
We compared our simplified computer model with a much more detailed simulation from the
scientific literature by Pralits and colleagues. Their model was three-dimensional and unsteady
(meaning it captured the chaotic nature of turbulent flow), while ours was two-dimensional and
steady (assuming the flow pattern doesn't change over time). Despite these simplifications, our
model correctly identified the same main deposition zones: the right walls and the outlet tube.
This gives us confidence that even simplified models can be useful for understanding spray
drying, especially when computational resources are limited.
Why does this matter?
This research helps us understand why some particles stick to the walls of spray dryers and how
to prevent it. By identifying the critical particle size of 5 micrometers, we can adjust the
atomization process to produce particles either smaller or larger than this threshold, depending
on where we want them to go. By understanding the relationship between air flow rate and
particle deposition, we can optimize operating conditions to maximize product collection while
minimizing overheating. By recognizing that particles that recirculate for longer periods heat up
more, we can design drying chambers that minimize recirculation zones.
What's next?
This project opens several avenues for future research. A full three-dimensional model would
capture the complex flow patterns more accurately. Including an evaporation model would make

the simulations more realistic by accounting for cooling as the droplets dry. Testing the critical
particle size of 5 micrometers with different materials would confirm whether this threshold is
universal or specific to our solution.
In summary, this project showed that combining computer simulations with laboratory
experiments is a powerful way to understand and optimize spray drying. The insights gained can
help manufacturers produce better quality powders with less waste, whether they are making
instant coffee, powdered milk, or life-saving medicines.

• author: Atetelim Mougoue, Serge Didier
• supervisor: Jesús Enrique Campos Pacheco; Andreas Håkansson
• organization: Pharmaceutical Technology (master); Food Technology and Nutrition (M.Sc.); Biotechnology (M.Sc.Eng.)
• course: KLGM16 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: ANSYS Fluent, Spray drying, DPM, Discrete Phase Model, Particle deposition, Büchi B-290, Glycerol, LES, Large-Eddy Simulation, RANS, Reynolds-Averaged Navier-Stokes, Residence time distribution, Pharmaceutical formulation
• language: English
• id: 9235909
• date added to LUP: 2026-06-12 13:37:14
• date last changed: 2026-06-12 13:37:14