Particle Formation of Ductile Materials using the
PGSS Technology with Supercritical Carbon
Dioxide
Particle Formation of Ductile Materials using the
PGSS Technology with Supercritical Carbon
Dioxide
Proefschrift
Ter verkrijging van de graad van doctor aan de Technische Universiteit Delft,
op gezag van Rector Magnificus prof.dr.ir. J. T. Fokkema voorzitter van het College van Promoties,
in het openbaar te verdedigen op vrijdag 16 december 2005 om 10.30 uur
door
Perize MÜNÜKLÜ
Diplom Chemie Ingenieur, University of Applied Science Münster, Duitsland Geboren te Sivas / Imranli / Turkije
Dit proefschrift is goedgekeurd door de promotor: Prof. dr. ir. P.J. Jansens
Prof. dr. G.J. Witkamp
Samenstelling promotiecommissie:
Rector Magnificus Voorzitter
Prof. dr. ir. P.J. Jansens Technische Universiteit Delft, Promotor Prof. dr. G.J. Witkamp Technische Universiteit Delft, Promoter Prof. dr.-Ing. M. Wessling University of Twente
Prof. dr. ir.L.P.B.M. Janssen Rijksuniversiteit Groningen Dr. Dipl.-Ing. J.Gross Technische Universiteit Delft Dr. ir. T. W. de Loos Technische Universiteit Delft
Dr. ir. F. E. Wubbolts FeyeCon Development & Implementation B.V.
Dr. ir. F. E. Wubbolts heeft als begeleider in belangrijke mate aan de totstandkoming van het proefschrift bijgedragen.
Cover design & layout by: Gerard van de Sande (SEM picture by Perize Münüklü)
ISBN: 90-9020280-3
Copyright © 2005 by P. Münüklü Printed by Febodruk B.V., Enschede
All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by means, electronic or mechanical, including photocopy, recording or by any information storage and retrieval system, without written permission from the publisher.
To my parents
Canim Babama, hayatimin en aci ve en büyük kayibi, bana değeri ölçülmez sevgi ve hayat dersi verdiği için Canim Anneme, hayatini ve tüm sevgisini babama, kardeşlerime ve bana adadiği için
"İnsanları yasa ve ceza ile yönetirseniz, onlar bir daha yanlış yapmayacaklar, ancak şeref ve utanma duygularına da sahip olmayacaklardır. İnsanları erdemle ve ahlak kuralları ile
yönetirseniz, o zaman onlar hem utanma duygusuna sahip olacaklar, hem de doğruyu yapmaya çalışacaklardır."
Contents
T
able of Contents
Chapter 1: Aim and Scope of the Thesis
1 Background
1
1.1 Aim and Focus
3
1.2 Outline of the Thesis
References
Chapter 2: Introduction to Supercritical Melt Micronization
Outline
2 Supercritical Fluids
2.1 Physical background
2.2 Superictical fluids and applications
2.2.1 Rapid Expansion from Supercritical Solutions, (RESS)
2.2.2 Anti – Solvent Techniques
2.2.3 Particles from Gas-saturated Solutions, (PGSS)
2.2.3.1 State of the Art and PGSS
2.2.3.2 Characteristics of PGSS
3
6
7
8
8
11
11
12
12
13
16
2.3 Spray Drying
2.3.1 Spray Drying below Boiling Point Temperature
2.3.2 Spray Drying above Boiling Point Temperature
2.4 Prilling
2.5 Mechanism of PGSS
2.5.1 Atomization
2.5.2 Disintegration of liquid jets from capillary nozzles
2.5.3 Disintegration of liquid jets from swirl nozzles
2.6 Particle Solidification
References
Chapter 3: Phase Behavior from Systems of Supercritical CO
2and Propane with Edible Fats
Abstract
Introduction
3 Materials and Methods
3.1 Results and Discussions
3.2 Modeling
Conclusion
References
16
17
17
18
18
19
19
20
20
22
25
26
26
27
33
36
37
Contents
Chapter 4: Experimental & Analytical Facilities at TU-Delft
Outline
4.1 Batch experimental Pilot Plant
4.2 Continuous Experimental Pilot Plant
4.2.1 Design Criteria
4.2.2 Design and Dimensioning
4.2.3 Melt Vessel
4.2.4 Filter
4.2.5 Melt Pump / CO
2Pump
4.2.6 Cooling Unit
4.2.7 Heat Exchangers E-C01 and E-B01
4.2.8 Static Mixer
4.2.9 Nozzle
4.2.10 Additional Cooling Unit
4.3 Process control
4.3.1 Pressure and flow control
4.3.2 Temperature control
4.4 Analytical Facilities
4.4.1 Particle Size Distribution
4.4.2 Powder Bulk Density
4.4.3 Structure / Morphology of Particle
4.4.4 Identification of crystalline phases
39
40
41
41
41
44
45
45
46
47
48
48
49
50
51
51
51
51
55
55
56
4.4.5 DSC (Differential Scanning Calorimetry)
4.4.6 XRD-PD (X-ray Diffraction for Powder)
References
Chapter 5: Development of a Batch process for Particle
Formation of Edible Fats
Abstract
Introduction
5 Experimental
5.1 Set-up for Solubility Measurements
5.2 Set-up for batch experiments
5.3 Results and Discussions
5.3.1 Solubility Measurements
5.3.2 Fat Micronization Experiments
5.3.3 Particle Formation mechanism
Conclusion
References
56
58
59
61
62
63
63
64
65
65
68
75
78
79
Contents
Chapter 6: Particle Formation of an Edible fat (RP70) using
the PGSS process
Abstract
6.1 Introduction
6.2 Materials and Method
6.3 Experimental
6.4 Results and Discussion
6.5 a) Effect of melt temperature (0 wt% CO
2)
6.5 b) Effect of melt temperatures (20 wt% CO
2)
6.6 a) Effect of atomization pressure (0 wt% CO
2)
6.6 b) Effect of atomization pressure (20wt% CO
2)
6.7 Effect of CO
2concentration
6.8 Mechanism for particle formation
Conclusions
References
Chapter 7: Particle Formation of Hydrogenated Castor Oil
(HCO) using the PGSS process
Abstract
7.1 Introduction
7.2 Materials and Methods
81
82
83
83
84
84
89
93
96
99
102
105
107
109
110
110
7.3 Phase behavior of HCO and CO
2system
7.4 Experimental & Analytical
7.5 Results and Discussions HCO
7.6 Effect of the CO
2concentration
7.7 Effect of atomization pressure (P
at.)
7.8
Comparison between experimental results of HCO and Rapeseed
70
Conclusions
References
Chapter 8: Proposed Models for describing the Solidification
of Fat Droplets due to Expansion
Abstract
Introduction
Assumption
8.1 Model Assuming Equilibrium Solidification
8.1.1 Droplet Velocity
8.1.2 CO
2Concentration
8.1.3 Gas-Phase Temperature
8.1.4 Droplet Temperature
8.2 Model accounting for solidification kinetics
8.2.1 Nucleation and Growth of solid particles
8.2.2 Growth Rate (G(∆T))
111
112
113
113
118
120
125
126
129
130
131
131
131
132
136
137
138
139
140
Contents
8.2.3 Nucleation Rate
8.2.4 Strategy for Determining a Plausible Nucleation Rate
8.3 Results and Discussions
8.4 Thermal images and analyses of the spray jet
Conclusions
References
Chapter 9: Manufacture of Powderous Food ingredients
Abstract
Introduction
9.1 Description of the supercritical melt micronization technology
9.2 Method for process design
9.3 Results and Discussions and Basis for the design
9.3.1 Product design and selection of process conditions
alone” plant
9.3.3 Preparation and mixing of raw materials
9.3.4 Spray and product recovery
9.3.5 Product packaging & storage
9.3.6 Mass and Energy balances
9.3.7 Equipment Design and Sizing
9.4 Process Control
141
142
144
155
159
163
167
168
169
170
170
171
9.3.2 Process structure and description for base case study: “stand 173
173
174
175
175
176
176
9.5 HAZOP
9.6 HACCP
9.7 Economics
9.7.1 Sensitivity analysis and possible costs reduction
9.7.2 Costs benchmark
9.7.3 Energy consumption reduction
Conclusions and Remarks
References
Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Summary
Samenvatting
Özet
Zusammenfassung
Acknowledgements
About the Author
177
177
177
182
183
183
184
185
189
197
207
211
215
219
223
227
231
237
241
Background
1
A
im and scope of
this thesis
Micronized powder can be used to put structure into final products. Micronized wax, for example, is added to coatings, sealants and pastes to influence the rheology. Micronized edible fat gives structure to spreads, like margarine. Margarine is usually a water-in-oil emulsion that is stabilised by a fine network of fat crystals.
Currently, these fat crystals are produced by crash cooling of the entire mixture of oil, water and fat in a scraped surface heat exchanger. The disadvantages of this technology are that it is difficult to control the particle size distribution and morphology of the fat crystals. Further more for the crash cooling of the mixture ammonia is used, which may contaminate the product due to leakage. Also high energy consumptions are expected due to the cooling of the entire product.
The conventional method to produce a fine powder of a ductile material is through cryogenic milling. With this technology the material is first cooled
are that, the low temperature does not only make the material brittle, but also prevents aggregation of the particles [1, 2]. Further more, the cryogenic method is an expensive technique, because of the low temperature needed to solidify the material. Therefore fine powders of materials like fats are not produced on a significant scale with cryogenic milling using liquid nitrogen.
It would be preferable to produce spreads by simply blending already micronized fat powder into an emulsion of water and oil. Just as in the conventional process, the fat crystals would form a layer around the water droplets and stabilise the emulsion. This raises the question of, how to produce micronized powder of ductile materials in general and micronized powder of edible fat in particular.
A promising method is called ScMM (Supercritical Melt Micronization), also known as PGSS (Particles from Gas Saturated Solutions) [3]. In this technology, a melt of the edible fat or the wax is mixed with CO2 and expanded over a nozzle, so that a fine powder is
obtained without using any solvents harmful to the environment. ScMM is expected to produce spherical particles, hollow or solid, with a relatively narrow size distribution. The size, shape and structure of the micronized material, like an edible fat or a wax, determines the possibilities for various applications, such as in the food and cosmetics industry. In the food industry, micronization is an important process stage in the treatment of food and dairy products. It provides improved product stability, shelf life, digestibility, and taste. Thus micronization effects can also significantly reduce the amount of additives required. In the cosmetic industry, micronization is essential to the quality of beauty creams, lotions nail varnishes, shampoos, and toothpastes. Micronization also ensures the most efficient use of the active agents, thus reducing product costs.
The type of equipment and the process conditions, chosen for the ScMM technology, determine the properties of the product. Fundamental knowledge of the interplay between the spraying, cooling and solidification of droplets enables the fine-tuning of the process to suit the needs of the various applications. ScMM is relatively new technology [4] and has never before been applied to edible fats and waxes. Therefore physical data and thermodynamic properties were not available and needed to be measured.
Background
This thesis addresses the challenge to produce micronized edible fat or wax using this novel ScMM technology.
1.1 Aim and Focus
The aim of this study is to develop and optimize a process, and to design and construct equipment for the micronization of edible fats and waxes based on ScMM (Supercritical Melt Micronization) technology. This was done by collecting data about the physical properties of the model components and thermodynamic data from the measurement of high-pressure phase equilibria of the CO2 / melt systems. Next, these thermodynamic data
and the physical properties of the model components were used to design a batch and continuous operated experimental pilot plant for particle formation. Further, these data were essential in order to be able to choose the right process conditions for the powder production of the edible fat and the wax. These process conditions were of great interest for the production of particles with different morphologies, particle sizes and densities. After producing such powder from the edible fat and the wax, numerical modeling was needed to develop and to understand the mechanism of heat and mass transfer of the droplet during the process of droplet cooling and solidification.
To validate the results of the model, an infrared camera was employed to be able to show the temperature profile of the droplet during solidification. This technique provided very valuable information for a better understanding of the temperature change in the spray. Based on the pilot plant, a new plant with a capacity of 1000 tons/year of the edible fat RP70, including economical evaluation, was designed.
1.2 Outline of the Thesis
Chapter 2 describes the properties of supercritical fluids and present technologies for particle formation, like RESS (Rapid Expansion of Supercritical Solutions), ASES, PCA (anti solvent techniques) and PGSS. It discusses the characteristics and the principles of the
PGSS technology and its state of the art. Also it addresses to technologies similar to PGSS, like spray drying and prilling.
Chapter 3 represents the measurement of high-pressure phase equilibria of the edible fat (RP70), hydrogenated castor oil (HCO) and tripalmitin with CO2. The experimental data of
the CO2/tripalmitin system are modelled with the Peng-Robinson equation of state.
Chapter 4 describes, the experimental facilities, designed and constructed at the Delft University of Technology, followed by analytical facilities and methods. First an introduction to a batch wise operated plant is given. Later a detailed description of a newly designed continuously operated pilot plant is given. Furthermore the analytical facilities and methods are described.
Chapter 5 shows the first results of particle formation of the edible fat (Rp70) in the batch wise operated experimental unit. In this chapter SEM pictures of micron-sized particles of Rp70 with different morphologies under various process conditions are presented. The effects of the dominant operating conditions (CO2 concentration, pre-expansion
temperature and atomization pressure) are elucidated.
Chapter 6 is discusses the experimental results of an edible fat with the continuously operated pilot plant. This chapter provides a detailed description of the influence of governing process variables, such as CO2 concentration, atomization pressure and feed
temperature on different characteristic of the final product. Chapter 7 looks at the experimental results of a wax with the continuously operated pilot plant. This chapter provides a detailed description of the influence of governing process variables, such as CO2
concentration, atomization pressure and feed temperature on different characteristic of the final product in comparison with RP70 and batch experiments presented in chapter 5.
Chapter 8 presents two mathematical models that describe the solidification and cooling of the particles after atomization. The basic model assumes equilibrium solidification. A more
Background
sophisticated model incorporates nucleation and growth kinetics as well as equation of moment for the particle size distribution. Both models predict the solidification behaviour of the droplets as a function of temperature and time after expansion. Moreover, thermal images of the expansion were made to measure the temperature profile of the droplets as a function of temperature and time after expansion. The experimental results of the thermal images are compared with the predictions of the two models.
Chapter 9 focuses on the design of a plant with a capacity of 1000 tons per year of Rapeseed 70, including an economic evaluation of the added cost per ton associated with the micronization of the fat. The main equipment of this plant is evaluated for size and costs, following a rigorous design method. Further this chapter provides a flow scheme of the process. This chapter also discusses aspects like the hazard and operability (HAZOP), fire and explosion index and hazard analysis of critical control point (HACCP) and evaluates these to assess the main risks that the process presents to operators and consumers.
References
[1] B. Edward Brooker, R. Ivor Tomlis, Patent EP 128 55 84
[2] T.L. Roggers, A.C. Nelsen, J.Hu, J.N. Brown, M. Sarkari, T.J. Young, K.P. Johnston, R.O. Williams III, A novel particle engineering technology to enhance dissolution of poorly water soluble drugs: Spray freezing into liquid. European Journal of pharmaceutics and Biopharmaceutics, Vol.54, page 271-280, 2002
[3] E. Weidner, S. Steiner, Z. Knez, Powder Generation from
Polyethylene glycols with Compressible Fluids, High Pressure Chemical Engineering, page 223-228, 1996
[4] P. Münüklü, F.E.Wubbolts, G.J.Witkamp, P.J.Jansens. “Supercritical Melt Micronisation Using the PGSS Process”, in “Supercritical Carbon Dioxide”, ACS-Books, University Press, Oxford, 2003
Outline
2
I
ntroduction to Supercritical Melt
Micronisation (ScMM)
This chapter provides background information to Supercritical Melt Micronization Processes (ScMM). First, an introduction to supercritical fluids with a detailed description of their physical properties is given. Secondly an overview of supercritical fluid technology on particle formation is presented.
Third the state of the art of ScMM, also known as PGSS (Particles from Gas Saturated Solutions), is described. Fourth, ScMM is compared to technologies like spray drying and prilling and the similarities between those technologies are pointed out. Finally two important steps of the ScMM process, hydraulic atomization and solidification, are evaluated.
2 Supercritical Fluids
A supercritical fluid is a substance above its critical temperature and critical pressure. In Figure 1, a Pressure - Temperature phase diagram of CO2 is presented.
Above its critical temperature a vapour it cannot be condensed into a liquid by compression. The physical properties of such a supercritical vapour are in between those of a liquid and a gas.
2.1 Physical Background
To have a better understanding of the properties of a supercritical fluid, a more detailed view of its physical background is given and discussed below.
At absolute zero, molecular motion ceases and all material is present in a solid state. By increasing the temperature, the thermal energy increases and the solid melts and transforms into liquid and vapour phase, as shown in the PT diagram for CO2 in (Figure 1).
Figure 1: P/T diagram of a pure species showing the supercritical region
Pr es su re [ M Pa ] Vapour (gas) Critical point Tc Pc Temperature [K] Vapour (ScF) Triple point Liquid Solid
Outline
In the liquid phase, the molecules are free to move, but the attraction between the molecules is so large that the molecule is still fixed to a position for over 90% of the time.
Above the critical temperature, the thermal energy of the molecules is always high enough to overcome the potential energy provided by the neighbouring molecules. No pressure can therefore be high enough to condense the vapour into the liquid state.
As a dense Supercritical Fluid, the molecules have a longer mean free path than as a liquid of comparable density. This gives the supercritical fluid the unique combination of a low gas-like viscosity and a high, liquid-like density. In Figure 2, the generalised diagram for supercritical fluids is given.
Near the 0.1 1.0 10.0 0 1.0 2.0 0.9 1.0 1.1 1.2 1.55 CP
ρ
rρ
ρ
c=
P
rP
P
c=
T = T T =
r c 0.8 criticAs the solvent power of a supercritical fluid is exponentially proportional to the density, a reduction in pressure very rapidly decreases its solubility. This feature is used mainly for micronization of solids. The low surface tension of a supercritical fluid allows it to penetrate even the smallest pores. This makes supercritical fluid especially suitable for rinsing, dyeing and impregnation.
SCFs have a high density, which leads to an appreciable solvent power, whilst the viscosity of SCFs is lower than liquids and the diffusivity is higher, which enhances mass transfer. SCFs demonstrate significant variations in thermodynamic and transport properties as a function of pressure and temperature. The unique properties of supercritical fluid have led to some industrial applications. For example, compressed gases, mainly CO2 and nitrogen,
are interesting for application in foods or pharmaceuticals, as they leave no residue. In table 1 a list of well know supercritical fluids is presented [1].
Table1: Supercritical Fluids
Tc [K] Pc [MPa] Carbondioxide Helium Nitrogen Water Nitrous Oxide Ammonia Perfluorobutane Ethane Propane Butane 304 3.31 126 647 310 405 287 305 370 425 7 0.11 3.4 22.1 7.2 11.4 2.3 4.9 4.3 3.8
Outline
Methanol Ethanol Ethyl Acetate Acetone Dimethyl Ether Diethyl Ether 1,4 Dioxane CH4 512 514 523 508 400 400 587 190.9 8.1 6.1 3.8 7.4 5.2 5.2 5.2 4.62.2 Supercritical Fluids and Applications
Apparently the supercritical fluid technology has been spoken in the last few years as a new green technology. The potential of supercritical fluid technology to prepare micro particles has mainly been demonstrated using two approaches. The compressed gas, mostly carbon dioxide, is used either as a solvent (RESS) or as an antisolvent (GAS and PCA, SAS, ASES) with regard to the material to be micronized. The latest approach, PGSS (particles from gas saturated solution), uses CO2 as a processing aid, and CO2 is dissolved in the melt
of the component to be micronized.
2.2.1 Rapid Expansion from Supercritical Solutions, (RESS)
In the RESS process, the material is first dissolved in compressed gas with a high density. This solution is rapidly depressurized to a lower density, which induces a rapid build-up of a high supersaturation. Due to the high supersaturation, the solute precipitates as fine particles with a narrow size distribution also due to the plug-flow character of RESS. Processing challenges are nozzle blockage coupled with the fact that most materials exhibit
mainly in the polymer and pharmaceutical field. Many publications are related to atomization of pharmaceutical components, either to produce fine particles, or to encapsulate an active component into a carrier. With the RESS process, typically micro particles are obtained, but there are some examples reporting the formation of nano sized particles using appropriate nozzles, for instance of benzoic acid [3] or cholesterol [4].
2.2.2 Anti-Solvent Techniques
Although the solvent power of a compressed gas is high compared to a low-density gas, it is still much less than that of liquid solvents. Because a compressed gas can usually be mixed with most common solvents, it can be used as an anti-solvent to precipitate a solute from a solution in the solvent. In batch GAS techniques, a solution is pressurized with a gas. As the pressure increases the gas dissolves and the solute precipitates in the bulk solution. Other varieties (PCA, SAS, and ASES) involve the spraying of a solution, as fine droplets, into a vessel filled with compressed gas. Depending on the pressure, the main driving force for crystallization is either the dissolution of the gas in the droplets or the extraction of solvent. Beside many patents in GAS, SAS and ASES, numerous publications show their use in recrystallization of many different products, like explosives, polymers, pharmaceuticals and coloring materials in order to make fine powders of an active component embedded into a carrier. Further reviews and applications of supercritical anti solvent precipitation have been published by Reverchon and Foster et al. [5-7].
2.2.3 Particles from Gas-saturated Solutions, (PGSS)
This study deals with another process under the acronym ´PGSS´, which stands for 'Particles from Gas-Saturated Solutions'. A gas, such as carbon dioxide, is first dissolved under pressure in a melt. The mixture then passes through a nozzle to a lower pressure area, where the rapid expansion of the gas beaks the melt into small particles or droplets. Suspensions may be treated in the same way [8].
Outline
2.2.3.1 State of the Art and PGSS
PGSS technology can be used in different fields, ranging from inorganic to pharmaceutical products. Looking at the history of this process, the basis of it was developed in 1979 by Graser [9], who prepared crystallized fine organic pigments in recrystallizing it using a supercritical fluid under high temperature and pressure. In 1991, Karasawa [10] developed a process for quick drying pollution free adhesives, where an adhesive prepolymer is compressed with CO2 and sprayed on a surface in order to obtain dry particles. Later, Union
Carbide patented a process called UNICARB for powder coating, where a solid material was suspended in a liquid carrier, mixed with a supercritical fluid and sprayed on a surface for coating. Most of the applications here are related to powder coating applications. Nordson et al. [11] also developed a very similar process to UNICARB. Both used supercritical carbon dioxide as a solvent and atomization agent.
In 1995, Weidner et al. [12] patented the batch wise PGSS process, mentioning the possibility of the use of a wide range of components. In the same year, Sievers et al. [13] also patented a process very much related to PGSS where he calls an aerosolization process. The active substance is dissolved into an aqueous solution and mixed under high pressure with supercritical CO2 and an aerosol spray is obtained.
A lot of experience on the PGSS process therefore exists and can be found in the open literature, like a review from Weidner et al. [14] In this review, he discusses the thermo- and fluid dynamic aspects and advantages of such a technology, which is also considered and studied by Knez et al. [15] for micronization of polymers. Kappler et al [16] show his work in producing polymer powders with the PGSS technology the possibility to obtain different morphologies under variation of process variables, for instance the CO2
concentration.
Furthermore, Jun Li et al. [17] developed a model for particle formation of hydrogenated palm oil with PGSS, which describes the change of the atomization pressure, the temperature, velocity and density along the nozzle. Petra Sencar –Bozic et al. [18] used the PGSS technology in order to improve the poorly water-soluble nifedipine and studied the
particle size distribution under the influence of pre-expansion conditions. Later, Kerc et al. [20] studied the micronization of practically insoluble calcium channel blockers nifedipine and felodipine and the hypolipidemic agent fenofibrate. The aim of this study was to increase the dissolution rate and hence the bioavability of these drugs. Rodrigues et al. [19] applied the PGSS process to particle formation of hydrogenated palm oil for new controlled-release carriers of theophyline and studied the effect of pressure on the particle size and morphology.
Only a few applications for the PGSS process and some reviews of the PGSS and other particle formation technology are available, which are also discussed by Jacques Fages et al. [20] and Jennifer Jung et al. [21].
However, no literature sources for the particle formation of edible fats with the PGSS process are available. In table 2, components, which have been produced with the PGSS technique so far, are presented.
Table 2: Components atomized with the PGSS process using CO2 Organic and inorganic
components
Rapeseed 70 (edible fat)
Hydrogenated Castor oil Polyethylenglycol
Glucose
Glycerides
Benzoic Acid
Results and Observation
Spherical hollow, distorted and sponge like particle
Particle size: 60-90µm Particle size: 150-400
Aerosolization of an aqueous solution Particle Size: 20-50 µm
Aerosolization of an aqueous solution
Aerosolization, dehdratisation and pyrolysis of a metal acetate or nitrate aqueous solutio
Reference This thesis This thesis Weidner, 1996 Sievers,1999 Novak,1993 Sievers,1999
Outline
Metal oxide Phosphors Plastic Additives Powder Coating Powder Coating Powder Lacquer Pharmaceuticals Albuterol sulfate Alkaline phosphatase Cromolyn sodium DL-alanine Glucose oxidaseAerosolization, dehydratisation and pyrolysis of an aqueous solution
Not available
Particle size: 1-30 µm
Not available
Aerosolization of an aqueous solution , Particle size :0.69µm
Aerosolization of an aqueous solution Aerosolization of an aqueous solution, Particle size: 1µm
Aerosolization of an aqueous solution, Particle size: 0.3-0.5
Aerosolization of an aqueous solution Aerosolization of an aqueous solution
Aerosolization of an aqueous solution
Aerosolization of an aqueous solution Particle size: 15.4µm Sievers, 199 Xu,1998 Prince,1993 Mandel,1997 Weidner,1999 Kieser,1998 Sievers,1998 2000 Sievers,1999 Sievers,1998 2000 Sievers,1995 Sievers, 1999
Glutathione Horseradish Peroxidase Na2Fe(DTPA) Nifedipine RhDNase Tobramycin
Aerosolization of an aqueous solution, Particle size: 1-2µm Not available Not available Not available Not available Sievers, 1999 Sievers, 1999 Sievers, 1995 Weidner, 1994 Sievers, 1999 2.2.3.2 Characteristics of PGSS
The PGSS technique has similar features to show characteristics of two conventional techniques: spray drying and prilling. Spray drying is the evaporation of a solvent due to hot air and prilling is the solidification of a melt due to cooling by cold air. Below, these processes are discussed in detail.
2.3 Spray Drying
The structure of materials produced with PGSS can best be compared with that from spray drying. Spray drying is divided into spray drying in air below and above the boiling point temperature of the solution. Spray drying involves the atomization of feed to contact the solution droplets and hot drying medium. As a result, the solvent evaporates. The solidification in this process is mostly governed by heat transfer. There is an essential difference in the products when the ambient temperature is higher or lower than the boiling point of the solution.
Outline
2.3.1 Spray Drying below Boiling Point Temperature
During spray-air contact, droplets meet hot air and moisture evaporation takes place from the droplet surfaces. The evaporation is rapid due to the large surface area of droplets in spray. With the contact between atomized droplets and drying air, heat is transferred from air to the droplets, providing the heat for evaporation. The vaporized moisture is transported into the air by convection. This first period of drying ceases, when the droplet moisture contents fall to a critical value, characterized by the initial presence of a solid phase forming at the droplet surface. Then follows a period of drying in which the average rate of evaporation falls. Moisture movement to the droplet surface is reduced, due the increasing resistance to mass transfer caused by the solid phase becoming more extensive. When the rate of heat transfer exceeds that of mass transfer the droplet begins to heat up. Sub-surface evaporation occurs if the heat transfer is sufficiently high to cause vaporization within the droplet.
2.3.2 Spray Drying above Boiling Point Temperature
For spray drying operations, where the air temperature is above the boiling point of the droplet solution, the first part of the drying process is the same. However, when the droplet is heated up above boiling point temperatures of the moisture, the liquid inside the droplet reaches its boiling point and vapor is formed. The pressure can build up inside when a crust forms around each droplet. The effect of pressure depends upon the nature of the crust. Vapors will be released if the crust is porous, but if it is non-porous, the droplet may rupture or even disintegrate.
Looking at the spray drying technique and the PGSS technique, the best way to describe the PGSS process is spray drying in air above the boiling point temperature. The carbon dioxide, which is dissolved in a substrate, acts like the moisture in the droplets when dried in air above its boiling point. When sprayed over the nozzle, the carbon dioxide in the droplets expands due to its high vapor pressure [22].
In spray drying, the temperature of the feed is lower than that of the surrounding air. In the PGSS process, the feed contains CO2 and is sprayed into air at a lower temperature. Due to
its high vapor pressure, the CO2 boils off during the expansion and influences the
morphology of the final product, which is ranging from hollow spheres to sponge like particles. These morphologies can also be obtained by spray drying.
2.4 Prilling
Prilling is another process of particle formation, which is more similar to the PGSS-process than spray drying, while in spray drying a solvent is evaporated, in prilling a melt solidifies by cooling. In the prilling process, a melt is sprayed from the top of a tower. The droplets fall against a countercurrent stream of cooling air and solidify into mostly spherical particles. The particle size obtained with this technique is usually bigger compared to particles produced by PGSS or spray drying [23]. In the prilling process, only solid particles can be obtained. The reason for this is that in a prilling no solvent is used, while in PGSS the dissolved CO2 not only lowers melt temperature and viscosity but also produces
hollow or sponge like particles.
2.5 Mechanism of PGSS
In this paragraph the mechanism of the PGSS technology will be explained. The most important aspects of it, which are responsible for cooling and particle formation, are atomization at the nozzle, escape of CO2 and cooling by the surrounding air. In the PGSS
process aspects like atomization, heat effects and solidification need to be discussed. Under atomization the formation of droplets from a melt by (hydraulic) atomization is described. The escape of the dissolved gas results in heat effects that are important for the initial solidification. A further important point to be considered is the ambient air, which is entrained into the jet and provides a major part of the cooling power that is needed for full solidification.
Outline
2.5.1 Atomization
2.5.2 Disintegration of liquid jets from capillary nozzles
The character of liquid jet disintegration depends on the velocity of discharge from the nozzle, Figure 3.Generally speaking, there are three characteristic forms of disintegration, caused by axisymmetric waves, asymmetric waves, and aerodynamic forces [4]. These forms of disintegration apply to velocities of order 1, 10, and 100 m/s, respectively, and therefore to increasing influence of aerodynamic forces.
Figure 3: jet disintegration a) varicose b) sinusoidal c) atomization [4]
At relatively low velocities, jet break-up is characterized by drop wise disintegration. In this velocity regime, a liquid stream becomes unstable with increasing distance from the nozzle. Swelling and contractions of growing amplitude form on the liquid stream, and the stream eventually breaks up into separate droplets. The diameter of the droplets that are formed during this regime is similar to that of the original liquid stream. This is a regime where surface tension dominates the break up process.
increases with increasing distance from the atomizer, and continues until ribbons finally break free and form fine liquid ligaments. The ligaments further disintegrate into droplets as the waves that form on their surface reach some critical amplitude. A small proportion of the droplets that are formed in this velocity regime are of a diameter that is smaller than that of the original liquid stream.
2.5.3 Disintegration of liquid jets from swirl nozzles
For the PGSS-process in this project, a swirl nozzle is used to achieve finer disintegration of liquid jets. With this sort of a nozzle, the jet breaks up easier and faster. A swirl atomizer consists of a conical swirl chamber with a small orifice at the vortex. During operation, a liquid is introduced into a chamber through tangential ports and allowed to swirl. If the liquid pressure is sufficiently high, a high annular velocity is obtained and an air-cored vortex is created. The swirling liquid then flows through the outlet of the swirl chamber and spreads out of the orifice under the action of both, axial and radial forces are forming a tulip shaped or conical sheet beneath the orifice. The sheet subsequently disintegrates into droplets. The liquid-air interaction, liquid surface tension and viscous forces are the primary factors governing liquid breakup process.
2.6 Particle Solidification
In melt crystallization, the solute material is present in a very high concentration and the maximum solidification rate is normally governed by the rate at which the heat of fusion can be removed [26, 36]. PGSS is a melt crystallization process. A very fast removal of the heat of fusion would result in an amorphous material, because the molecules become frozen in a fixed position without the possibility to orient themselves in a crystal lattice. When the rate of heat removal is slow, compared to the rate of crystallization, the solid phase forms by a nucleation and growth process similar to solution crystallization.
The rate of appearance of new particles by nucleation and the growth rate of existing ones both depend on the driving force, which is called supersaturation or supercooling. The primary nucleation rate increases exponentially with the supersaturation, whereas the growth rate does not show such a strong dependency.
Outline
The supersaturation can be increased by external factors such as the temperature and solvent- or anti-solvent concentration. Growth on the surface of existing material reduces the supersaturation, counteracting the effect of the external factors [24, 25].The PGSS process is a fast solidification process including heat and mass transfer due to evaporation of CO2from the melt and ambient air temperature.
References
[1] F.E. Wubbolts, Supercritical Crystallization, Volatile Components as (Anti-) Solvents, Thesis, Technical University of Delft, 2000
[2] PabloG, Debenedetti, Supercritical Fluids as particle formation media, Supercritical Fluids, Fundamentals for Application, edited by Erdogan Kiran, Kluwer Academic publisher, 1994
[3] C. Domingo, E. Berends, G.M. van Rosmalen, Precipitation of Ultrafine Organic Crystals from the Rapid Expansion of Supercritical Solutions over a Capillary and a Frit Nozzle. Journal of Supercritical Fluids, Vol.10, page 39-55, 1997
[4] H. Kröber, U. Teipel, H. Krause, Formation of Submicron Particles by Rapid Expansion of Supercritical Solutions. GVC-Fachausschuss `High Pressure Chemical Engineering`, 3- 5 March, Karlsruhe (Germany) page 247-250, 1999
[5] E. Reverchon, Supercritical Anti-Solvent precipitation: its Application to Microparticle Generation and Product Fractionation. Proceedings of the 5th Meeting on Supercritical Fluids, Tome1; M. Perrut, P. Subra (Eds.), ISBN 2-905-267-28-3, 23-25 March, page 221-236, Nice, 1998
[6] E. Reverchon, Review: Supercritical Anti-Solvent Precipitation of micro- and nano-particles. Journal of Supercritical Fluids, Vol.15, page 1-21, 1999
[7] N.R. Foster, K.Benzanehtak, M. Charonechaitrakool, G. Combes, F. Dehghani, L. Sze Tu, R. Thiering, B. Warwick, R. Bustami, H.K. Chan, Processing Pharmaceuticals Using Dense Gas Technology. Proceedings of the 5th International Symposium on Supercritical Fluids, 8-12 April, Atlanta, 2000
Outline
[8] E. Weidner, S. Steiner, Z. Knez, Powder Generation from Polyethylene glycols with Compressible Fluids, High Pressure Chemical Engineering, page 223-228, 1996
[9] F. Graser, Wickenhauser, G.; Patent US 4,451,654, September 20, 1982
[10] Y. Karasawa, Patent DE 19640027, 1997
[11] D.R. Hastings, Hendricks (Nordson Corporation), Patent US 89/416855, 1989
[12] E.Weidner; Z. Knez, Z. Novak, Process for preparation of particles or powders, patent WO 9521688, 1995
[13] R.E. Sievers, U. Karst, EU Patent 0677332, 1995. US Patent 5,639,441, 1997
[14] E.Weidner, M.Petermann, Z. Knez, Multifunctional composites by high-pressure spray process, Current Opinion in Solid State and Material Science Vol. 7, Page 385-390, 2003
[15] Z. Knez, Zelijko, High Pressure Micronisation Process, ACS Journal, 2004
[16] P. Kappler, W. Leiner, M. Petermann, E. Weidner, Size and Morphology of Particles Generated by spraying polymer melts with Carbon Dioxide, Proceedings of the 6th
International symposium on supercritical Fluids, page 1891-1896, Versailles, France, 2003
[17] Jun Li, Henrique A. Matos, Edmundo Gomez de Azevado, Two-Phase homogeneous model for particle formation from gas-saturated solution processes, Journal of Supercritical Fluids, 2004
[19] M. Rodrigues, N. Peirico, H. Matos, E. Gomes de Azevedo, M.R. Lobato, A.J. Almeida, Microcomposites theophyline/hydrogenated palm oil from a PGSS process for controlled drug delivery systems, Journal of Supercritical Fluids, Vol. 29, page 175-184, 2004
[20] J. Fages, H. Lochard, J-J. Letourneau, M. Sauceau, E. Rodier, Particle generation for pharmaceutical applications using supercritical fluid technology, Powder Technology, Vol. 141, page 219-226, 2004
[21] J. Jung, M. Perrut, Particle design using supercritical fluids: Literature and patent survey. Journal of Supercritical Fluids Vol. 20, page 179-219, 2001
[22] K. Masters, Spray Drying Handbook, 3rd Edition, London 1979
[23] Perry, R.H., Perry’s Chemical Engineers’ Handbook, 6th Edition, 1984
[24] J.W.Mullin, Crystallization, 3rd edition, Butterworth Heinemann, Oxford 1986
[25] D. Kachiew, Nucleation, Basic theory with applications, Butterworth Heinemann, Oxford, 2000
Thermodynamics
3
P
hase Behaviour from
Systems of Supercritical CO
2
and
Propane with Edible Fats
The phase behaviour of the systems CO2 / Rp70 (hardened rapeseed oil), CO2 / tripalmitin,
CO2 / HCO (hydrogenated castor oil) and propane / Rp70 was studied experimentally
according to the synthetic method using a Cailletet apparatus. For the CO2 / Rp70 system
bubble-point curves and solid-liquid curves were determined for CO2 concentrations from
5wt% up to 30wt%. A comparison was made between the phase behavior of these systems and the phase behavior of the systems propane / tripalmitin. The solubility of propane in
tripalmitin is markedly higher as compared to the one of CO2 and tripalmitin. For the
propane / Rp70 system liquid-vapour curves, solid-liquid curves and liquid-liquid lower solution temperature curves were measured at propane concentrations from 10wt% up to 70wt%. An increase in pressure or a decrease in temperature results in a higher solubility of carbon dioxide or propane in the fat melt. The measurements made it possible to predict
CO2 / HCO systems a type III system was observed according to the classification of van
Konynenburg and Scott [1, 2]. For propane / Rp70 a type IV phase diagram was observed.
Introduction
Supercritical melt micronization is considered to be an interesting technology to create fine powders of fats and waxes with a particle size below 50µm [3]. For this process CO2 and
propane appear to be suitable supercritical solvents and reliable information on the phase behavior of these compounds in mixtureswith fats or waxes is needed. Experimental data has been reported already by Coorens et al. for propane/tripalmitin [4]. CO2 / tripalmitin
was studied by Weber et al. [5], and by Bharath et al., who has measured only at 353 K [6]. Therefore a study on the phase behavior of an edible fat mixture (rapeseed 70) and a pure edible fat, (tripalmitin) with CO2 and propane was performed.
Coorens et al. measured the propane / tripalmitin system with the Cailletet apparatus and our results show the same behavior as was observed with our propane / Rp70 system. The CO2 / tripalmitin measured by Weber et al. was measured with a view cell where a sample
of the composition CO2 / tripalmitin in equilibrium was taken in order to obtain the p-x-y
data for the system. Their measurements show a clear temperature influence on L-V isotherms. In this work, we determine liquid-vapour, liquid-liquid and solid-liquid equilibrium data for the above mentioned systems in a temperature range of 328 to 403 K using a Cailletet apparatus.
3 Materials and Methods
The CO2 and propane used was from Messer-Griessheim with a purity of 99.995% for CO2
and 99.95% for propane. Rapeseed 70 was obtained from Quest and it constitutes a mixture of various fatty acids. Tripalmitin was obtained from Sigma-Aldrich with a purity of 99%. The HCO wax (hydrogenated castor oil) was provided from manufacturer, containing up to 87% triglycerides. All components were used without further purifications.
A Cailletet apparatus as schematically illustrated in Figure 1 was used to measure mixture isopleths [7]. Solid-liquid and vapour-liquid curves were determined for the studied systems covering a temperature range of 333-403K. For detection of the bubble-curves (L-
Thermodynamics
Sample Hg level Thermostat Hg Oil Piston pumpV) the mixture is first brought into a heterogeneous state. Then the temperature is kept constant and the pressure is varied until the second phase disappears. To detect the solid-liquid (S-L) line the pressure is kept constant and the temperature is varied until the last solid crystal completely disappears. The temperature is controlled within 0.02 K and measured with an accuracy of 0.01 K using a PT-100 thermometer. The pressure is controlled and measured with a dead-weight gauge to within 0.05 bar. The errors in the weight fractions of the supercritical gas / fat mixtures are less then 0.1 wt%. The accuracy of the determined phase boundary pressures for the rapeseed 70 system is estimated to be better than ± 0.2 bar and for the tripalmitin system better than ± 0.02 bar. This difference in accuracy of Rp70 and tripalmitin can be explained by the fact that tripalmitin is a pure component, which leads to a more clear measurement of L-V and S-L points.
Figure 1: Schematically experimental set-up for solubility measurements
3.1 Results and Discussions
In this section the liquid-vapor (L-V) and the solid-liquid (S-L) phase behavior of CO2 /
tripalmitin, CO2 / Rp70, CO2 / HCO and propane / tripalmitin and propane / Rp70 system
In Figure 2, the phase diagram of the CO2 / Rp70 system is shown. The results from the
solubility measurements are plotted in a P-T diagram as lines of constant composition (isopleths). In this figure the bubble-point curves and solid-liquid lines are displayed. By increasing the weight percentage of CO2 in the sample, higher pressures are needed to
dissolve the CO2 in the fat melt. The pressure ranges between 29 and 307 bar and the
temperature between 325.15 and 367.15 K. Solid-liquid equilibria were measured for 0wt%, 5wt%, 10wt%, and 20wt% of CO2. As can be seen in Figure 2, the vertical S+L→L
phase boundary curves shifts to lower temperatures with increasing CO2 concentration
clearly as the result of melting point depression.
Table 1: Experimentally obtained L-V points for the CO2 / Rp70 system
wt% CO2 P [MPa] T=333K T=338K T=343K T=348K T=353K T=358K T=367 K 5 3.15 3.23 3.38 3.58 3.7 10 5.89 6.22 6.55 7.15 7.45 17 26 9.49 17.66 9.96 18.4 10.46 19.14 10.97 20.04 11.99 20.73 21.44 30 28.94 28.9 28.94 29.38 29.44 29.82 30.78
Thermodynamics
0 5 10 15 20 25 30 35 320 330 340 350 360 370 380 Temperature [K] P [ M P a ] 30 wt% CO2 26 wt% CO2 17 wt% CO2 10 wt%CO2 5 wt% CO2 100 wt% fat 5 wt% CO2 10wt% CO2 17wt% CO2Figure 2: P-T cross-section showing bubble-point curves (L+V→L), and melting-point
curves (S+L→L), for the binary system CO2+ Rp70
The phase behavior of CO2 / tripalmitin at indicated wt% CO2 mixtures is depicted in
Figure 3. The bubble- point pressures are lower as compared with the case of Rp70 / CO2,
indicating a better solubility. Rapeseed 70 apparently contains components in which the solubility of CO2 is much lower than in the pure tripalmitin. The high purity of tripalmitin
made it easier to get accurate L-V and S-L data. The point of intersection of the bubble-point curves and the solid - liquid lines for a mixture of the same composition gives a bubble-point of the three phase curve solid + liquid + vapor, which starts in the triple point of the pure fat. According to Figures 2 and 3 this curve is found at temperatures higher than the critical temperature of pure CO2 and shifts with increasing CO2 concentration to pressures higher
then the critical pressure of CO2. This makes it very likely that the CO2 / fat system can be
classified as a system with type III fluid phase behavior and a metastable liquid-liquid phase split [7, 8].
Table 2: Experimentally obtained L-V points for CO2 / tripalmitin system mole fraction of CO2 P [MPa) T= 336K T= 338K T= 343 K T= 348 K T= 353 K T= 359 K T= 363K 0.491 3.24 3.38 3.53 3.7 3.85 0.671 5.83 6.13 6.45 6.78 7.08 0.821 10.31 10.68 11.4 12.13 12.49 0.887 21.4 22.3 23.1 23.9 24.75
Figure 3: P-T cross-section showing bubble-point curves (L+V→L), and melting point curves (S+L→L), for the binary system CO2 + tripalmitin at indicated wt% of CO2
0 5 10 15 20 25 325 330 335 340 345 350 355 360 365 370 Temperature [K] P /M P a 5wt% CO2 10wt% CO2 pure tripalmitin 10wt% CO2 20wt% CO2 20wt% CO2 30 wt% CO2 30 wt%CO2 5wt% CO2
Thermodynamics
In Figure 4, the result of solubility measurements of CO2 in HCO is presented. Compared to
Rp70 and tripalmitin at the same CO2 concentrations much higher pressures are needed
to dissolve CO2 in the HCO. This is expected due to the high molecular mass of HCO,
which has an average carbon number 63 as compared to 51 for tripalmitin.
Table 3: Experimentally obtained L-V points for CO2 / wax system (HCO)
wt% CO2 P [MPa] T= 353K T= 358K T= 363K T= 368K T= 373K T= 383K T= 393K T= 403K 5 4.23 4.58 4.9 5.2 5.53 10 8.2 8.8 9.48 10.13 10.73 20 17.82 18.26 18.74 19.22 25 31.46 31.68 32.02 32.28
Figure 4: P-T cross-section showing bubble-point curves (L+V→L), and melting point curves (S+L→L), for the binary system CO + hydrogenated castor oil (HCO) at
0 5 10 15 20 25 30 35 40 345 355 365 375 385 395 405 Temperature [K] P [ M P a ] 5 wt% CO2 25 wt% CO2 20 wt% CO2 10 wt% CO2 25 wt% CO2 20 wt% CO2 10 wt% CO2 5 wt% CO2
Figure 5 shows the L-V and S-L phase behavior of the propane / Rp 70 system. As can be seen, relatively low pressures are needed to dissolve propane in Rp 70 as compared to CO2 /
Rp70. What can clearly be seen here
is that propane is more effective in terms of solvent capacity. Coorens et al. also found this fact for the propane / tripalmitin system [5].
Table 4: Experimentally obtained L-V and L-L points for Propane / Rp70 system wt% propane P [MPa] T= 328 K T= 333K T= 338K T= 343 K T= 348K T= 353K T= 358K T= 363K 10 0.99 1.078 1.178 1.289 1.403 1.508 1.633 1.747 20 1.668 1.848 2.013 2.208 2.388 2.61 2.848 30 1.772 1.967 2.171 2.406 2.66 2.92 3.198 3.55 40 1.87 2.08 2.317 2.567 2.832 3.132 3.47 4.027 50 1.914 2.124 2.361 2.661 3.284 4.234 5.111 6.027 70 1.97 2.195 2.59 3.567 5.437 6.32 7.177
Thermodynamics
Figure 5: P-T cross-section showing bubble-point curves (L+V→L), and melting-point curves (S+L→L), for the binary system propane / Rp70 at indicated wt% of propane
3.2 Modeling
The Peng-Robinson equation of state has been applied to model the experimental data of CO2/tripalmitin, Appendix A. The other systems were not modelled because of the complex
composition of the heavy components. For the calculation of the binary parameters the quadratic mixing rules were used [9]. The experiments were carried out for isotherms at 343, 353and 359 K, Figure 6. In Table 2 the experimentally observed L-V data measured at constant temperatures between 343 and 359 K are shown. The values to calculate the pure component data for CO2 and tripalmitin are given in Table 5 together with the binary
interaction parameters. The pure component data, 0 2 4 6 8 10 12 14 16 322 327 332 337 342 347 352 357 362 367 372 377 Tem perature [K] P [ M p a ] Pure rapeseed 20% propane 30% propane 10% propane 70% propane 50% propane 40% propane 30% propane 20% propane 10% propane
co-volume (b) and attraction parameter (a) of CO2 were calculated using the critical data of
CO2. The pure component parameters of tripalmitin were not calculated from critical
parameters, but constant values were chosen for the attraction parameter (a) and co-volume (b). The model was fitted to the data, minimising the AARD, by changing the values of a_tripalmitin, kij and lij. The absolute average relative deviation (AARD) in mole fraction of CO2 was found to be 2,2 %.
Table 5: Calculated parameter of CO2 and Tripalmitin and the Deviation (AAD)
Carbon dioxide Tripalmitin Pcritical [MPa] 7.38
Tcritical [K] 304.19
ω[-] 0.225
a 0.41 109
b 2.666*10-5 1.012*10-3
Figure 6: P-x diagram of the CO2 / Tripalmitin system. Comparison of experimental
vapor-liquid equilibrium correlating results from the Peng-Robinson equation of state. 0 5 10 15 20 25 30 0,4 0,5 0,6 0,7 0,8 0,9 1 molefraction CO2 P [M p a ] PR-EOS 343 K PR-EOS 348 K PR-EOS 353 K PR-EOS 358 K Exp.343 K Exp.348 K Exp.353K Exp.359 K
Thermodynamics
The model is showing a good fit of experimentally obtained data. Experimental results obtained by Weber et al. are given in Figure 7 for comparison. Their experimental data is presented as the dashed curves.
Figure 7: Experimental results of CO2 / tripalmitin system from this work and Weber
et al., 1999
The bubble-point curves at 333 K and 353 K from Weber et al. indicate a clear temperature influence on the solubility measurements. In contrast to our data significantly less pronounced temperature dependence is indicated. At this stage we do not have an adequate explanation for this observation. At 0wt% CO2 tripalmitin will show a horizontal straight
line indicating no temperature influence. Similar systems, triolin / CO2 system, measured by
Weber et al. as well did not show a temperature influence.
0 10 20 30 40 50 60 0,2 0,6 1 molefraction CO2 P [ M P a ] this work 343K this work 348K this work 353K this work 359K Weber et al. 333K Weber et al.353K
Conclusion
From the experimental results on the bubble point curves of CO2 / fat and propane / fat
systems it can be concluded that propane dissolves better in the fats than CO2. For the CO2 /
Rp70 system much higher pressures are needed to dissolve CO2 than for CO2 / tripalmitin
system. With increasing carbon numbers from tripalmitin, Rp70 to HCO much higher pressure were needed to dissolve CO2.
In accordance with the classification of van Kyonenburg and Scott experiments show that the systems CO2 / Rp 70, CO2 / tripalmitin and CO2 / HCO behave like a type III system.
For the system propane / Rp 70 a type IV system, with a metastable liquid-liquid phase split was observed. With Peng-Robbinson equation of state a fit of the experimental data was achieved. The absolute average deviation (AARD) in mole fraction of CO2 was 2.2%.
Thermodynamics
References
[1] P.H. Van Konynenburg, R. L. Scott, Critical Lines and Phase
Equilibria in Binary van der Waals Mixtures, Phil.Trans., Vol. 298A, page 495-540, The Netherlands, 1980
[2] Th.W.De Loos, Understanding Phase Diagrams, Supercritical Fluids, Kluwer Academic Publisher, page 65-89, Netherlands, 1994
[3] P. Münüklü, F.E.Wubbolts, G.J.Witkamp, P.J.Jansens. “Supercritical Melt Micronisation Using the PGSS Process”, in “Supercritical Carbon Dioxide”, ACS-Books, University Press, Oxford, 2003
[4] H.G.A. Coorens, C.J. Peters, J.De Swaan Arons, Phase Equilibria in binary mixtures of propane and tripalmitin, Fluid Phase Equilibria, Vol. 40, page 135-151, Amsterdam, 1988
[5] W.Weber; S.Petkov; G. Brunner; Vapour-Liquid-equilibria and calculations using the Reddish –Kwong –Aspen-equation of state for tristearin, tripalmitin and triolein in CO2 and
propane, Fluid Phase Equilibria, page158-160, 695-706, 1999
[6] R. Bharath; S. Yamane; H. Inomata; T. Adschiri; K.Arai; Phase Equilibria of Supercritical CO2-Fatty oil Component Binary Systems, Fluid Phase Equilibria, page
83,183-192, 1993
[7] Th. W. De Loos, I. Kikic, High Pressure Technologies: Fundamentals and Applications, Vol.9, 2001
[8] Th.W. De Loos, H.J. Van Der Kooi; P.L. Ott, Journal of. Chemical Engineering Data, Vol. 31, page 166-168, 1986
[9] O. Pfohl, S. Petkov, G. Brunner, PE 2000, A powerful Tool to Correlate Phase Equilibria, Herbert Utz Verlag, Wissenschaft, München, 2000
Experimental and Analytical Facilities
4
E
xperimental and
Analytical Facilities at TU-Delft
Process development work was performed using two experimental units, which are operated batch wise and continuously. The batch equipment was used to perform the first scouting experiments with model component rapeseed 70 (edible fat). More experiments with rapeseed 70 and a wax were performed with the continuously operated plant. The continuous pilot plant has been developed in this project and is described in detail, considering design criteria and description of the units and the process. Further, the analytical methods and facilities for calculating and determining the properties of the powder, i.e. rapeseed 70 and wax, such as densities, morphologies and particle size distribution are discussed.
4.1 Batch experimental Pilot Plant
In Figure 1 the schematic diagram and in figure 2 picture of the batch experimental pilot plant is shown.
After melting the fat at Tmelt, it is poured into the heated pressure vessel (a) from the top of
the vessel. The CO2 (b) to be dissolved into the fat enters the pressure vessel from the
bottom until equilibrium is reached at Pdissolve, (XCO2). Next, CO2 pressure is built up in the
autoclave with the pressure control valve (d). When the desired pressure is reached, the CO2/fat mixture is pressurized to atomization pressure (Patom) using Helium (j). After the
desired atomisation pressure is reached for dissolving CO2 in the melt, the stirrer is turned
off and valves (e) and (f) are closed. By opening valve (f) in the direction of the nozzle (g), a fine spray of droplets is sprayed in the collection vessel (h). To prevent the particles from being released from the collection vessel into the air, the CO2 outlet of the collection vessel
passes through a filter [1]. By using this unit, the first series of experiments are carried out where the feed temperature, CO2 concentration and atomization pressure of the CO2 / melt
mixture are varied in order to see their influence on particle morphology, size and density.
Figure1: Batch wise operated Experimental Unit
Helium (j) (e) (f) (i) CO2 outlet (a) (c) PC PI TI (g) (h) Compressed CO2 Supply (b) (d)
Experimental and Analytical Facilities
4.2 Continuous Pilot Plant
A continuous pilot plant has been designed and developed at the Laboratory of Process Technology, Figure 3. The objective of this new pilot plant was to produce powder from ductile materials using the PGSS (Particles from Gas Saturated Solutions) process [chapter 2, 2]. Up till now two fatty materials were examined using this pilot plant. The effects of different thermodynamic parameters like pressure, temperature, concentration, flow rates and ambient temperature were studied. The design criteria, the description of the process units, including the specifications and dimensions and the process control are described in the following sections.
4.2.1 Design Criteria
The design of the experimental set-up is based on the results of the solubility measurements, the batch experiments, and the following product and operating criteria:
• The melting point for fats in general, including some polymers, is in the range of 40 to 150 oC, so the upper level of the temperature range was decided to be at least 150 oC.
• The solubility measurements showed that the maximum required pressure is around 400 bar to cover a wide range of process condition.
• The operating temperature is just above the melting point of the fat, so special attention must be paid to cold parts in the equipment.
• Fat and CO2 need to be supplied at constant flows to create a constant weight ratio
between the fat melt and the carbon dioxide.
• The maximum production rate is calculated to be 20 L/h.
4.2.2 Design an Dimensioning
The unit, Figure 2, consists of a supply system for the CO2 and a supply system for the
Figure 2: P&ID of the Continuous Experimental Unit V A -A 0 1 V -2 V -C 0 3 V -8 5 1 6 0 1 6 1 6 0 3 8 0 FI P I C O2 C o o l g e n e ra to r P -C 0 1 E -C 0 1 1 2 0 6 0 2 5 6 0 3 1 0 3 8 0 M -B 0 1 F i – A 0 1 F i-C 0 1 P is a -A 0 1 F a t T i-A 0 1 C O 2 t o a tm o sp h e re M S ta rt -U p l o o p M T I P I P is a-C 0 1 F I E -C 0 2 T i-C 0 1 T I 4 1 6 0 3 8 0 O il T I V A -A 0 1 T i-B 0 1 P i-B 0 1 1 /2 ” 3 /8 ” 3 /8 ” 3 /4 ” O il 3 /8 ” C W V A -A 0 3 O il E -B 0 1 V A -A 0 2 P I V -A 0 1 V -C O 2 K n o ck o u t d ru m V -B 0 1 P -A 0 1 C -B 0 1 T h -A 0 1 M a in s tr e a m s to re d C O2 A d d it io n a l C O2 V -A 0 3 _ _ _ _ P ip in g -O il p ip in g V -C O 1 V A -A 0 2
