Subscriber access provided by University of Sunderland
Full Paper
Evaluation of Njs correlations in lab scale tanks with varying baffle configurations Benjamin M Cohen, Bahar Inankur, Kathleen T Lauser, Jennifer Lott, and Wei Chen Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00244 • Publication Date (Web): 01 Oct 2018 Downloaded from http://pubs.acs.org on October 7, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Organic Process Research & Development
Evaluation of Njs correlations in lab scale tanks with varying baffle configurations Benjamin M. Cohen1*, Bahar Inankur1, Kathleen T. Lauser1, Jennifer Lott1, Wei Chen2 1
Bristol-Myers Squibb, Chemical and Synthetic Development, 2 Bristol-Myers Squibb, Drug Product Science and Technology *Corresponding author email:
[email protected] ACS Paragon Plus Environment
Organic Process Research & Development 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Table of contents figure
ACS Paragon Plus Environment
Page 2 of 20
Page 3 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Organic Process Research & Development
Abstract Suspension of insoluble solids in stirred tanks is important for many operations in the pharmaceutical and specialty chemical industries. A common goal in solid suspension is to achieve the just suspended speed, Njs, where all particles are suspended off the tank bottom. Operating at Njs results in high mass transfer between the solid and liquid phases, which is important because many solid-liquid operations rely on adequate mass transfer to achieve the goals of the operation (e.g. reaction completion in a heterogeneous reaction mixture). Njs can be predicted based on physical properties and tank geometry, often through the use of the Zwietering or GMB (Grenville, Mak, Brown) correlations. These correlations use impeller and tankgeometry specific constants, S for Zwietering and Z for GMB, which typically have been obtained in larger scale tanks with four flat baffles. Glass-lined tanks commonly used in the pharmaceutical and specialty chemical industries often have reduced baffling. This paper evaluates the effect of reduced baffling on Njs and determines the S and Z constants as a function of baffling. It was found that Njs can be substantially reduced by removing baffles from a tank. Also, the S and Z constants for the Njs models obtained in the lab scale were comparable to those reported in the literature from larger scale studies. Keywords: Mixing, Scale-up, Multiphase reactions, CFD
ACS Paragon Plus Environment
Organic Process Research & Development 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Introduction Effective suspension of insoluble solids in stirred tanks is a common challenge in the chemical and pharmaceutical industries. Unit operations such as dissolution of solids,1 solid-catalyzed2 or inorganic base containing reactions,3 and crystallizations4 all involve mass transfer, and the success of these and many other operations critically depends on effective mixing and appropriately designed mixing systems.5 The key objective of solid-liquid mixing is to create and maintain a solid-liquid slurry, and to promote and enhance the rate of mass transfer between the solid and liquid phases by maximizing the contact area between the solids and liquid. For instance, a reaction such as heterogeneous catalytic hydrogenation relies on adequate suspension of the solid catalyst in order to achieve the expected reaction rate. If the catalyst is not adequately suspended, it will tend to remain near the bottom of the tank where it will be inaccessible to the reactants in other parts of the solution, and the reaction rate will suffer. Specialty chemical and pharmaceutical processes are often developed in small, laboratory scale stirred tanks (typically ≤ 10 L). They often need to be scaled up in equipment many orders of magnitude larger in order to generate the necessary product quantities to support development and commercial demands. Because of the numerous unit operations that rely on solid suspension to achieve sufficient mass transfer, it is important to have a reliable method to predict mixing conditions that will provide sufficient solid suspension at scale.6 Production scale equipment is designed to have maximum flexibility with respect to chemical compatibility and unit operations. Therefore, these vessels are typically cylindrical glass-lined steel tanks with centrally positioned mixing shaft, with one or two baffles off-set from the vessel wall. From this perspective, an additional challenge is introduced to predict solid suspension in such tanks, since most solid-liquid suspension research and modeling is performed on flat bottom tanks with standard flat baffles installed along the flat tank walls.7-10 An important criterion in the selection of a vessel configuration and agitator speed for a solid-liquid mixing process is the just suspended speed, Njs. At the impeller speed below the Njs, a portion of the particles reside on the tank bottom and mass transfer will be limited by diffusion through the solid bed. As the impeller speed is increased, greater proportions of the solids are suspended, increasing the mass transfer. By further increasing the impeller speed to the Njs, mass transfer is increased because all particles are suspended off the tank bottom. Once the impeller speed is increased above Njs there is only a modest further improvement in the mass transfer; this increase is due to the renewal of the boundary layer around the solid particles.11 Thus, the just suspended speed is the optimum balance between solid-liquid mass transfer and power input to the agitator. Zwietering pioneered the study of solid suspension in stirred tanks, measuring Njs with a selection of tank base geometries, tank diameters, impeller geometries and solid and liquid properties.12 All experiments were conducted in tanks with a standard baffle setup of four flat baffles mounted along the tank walls. In this case, Njs is defined as the agitation rate at which no solids are motionless on the tank bottom for more than 1 or 2 seconds. Zwietering performed a dimensional analysis to develop a correlation for Njs (Eqn.1): =
. . . .
. (
ACS Paragon Plus Environment
(1)
Page 4 of 20
Page 5 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Organic Process Research & Development
where S is a dimensionless constant which must be measured for each impeller and vessel geometry tested as it is dependent on impeller diameter, tank diameter and impeller height as well as degree of baffling and base shape including dish, flat and conical bases. Zwietering reported S values for some agitator and vessel configurations.12 Subsequent research reported S values for additional agitator and tank configurations, though most of this work is done on standard-baffled tanks.13 One of the few works to examine the effect of Njs and baffling was conducted by Brucato et al.14, which examined differences between baffled and unbaffled tanks with a Rushton impeller. The study showed that unbaffled tanks required significantly lower power at Njs than baffled tanks. In another study with similar findings, Tamburini et al., showed unbaffled systems gave more efficient particle suspension than corresponding baffled systems by an order of magnitude.15 In their investigation, Tamburini et al. used single system of particles and water with flat bottom tanks equipped with Rushton, Pitched-Blade turbine (PBT) and Lightnin turbine A310 impellers. Despite these findings, there is very limited published data of measured S values for prediction of Njs in glass-lined tanks with reduced baffling, which are commonly found in the pharmaceutical industry. Another limitation of the Zwietering correlation is that it does not adequately account for the effect of scale. In a later study, Mak investigated Njs for sand in water with pitched blade turbines in geometrically similar tanks, with tank diameters, T, of 0.31 and 0.61 m. Mak found that using the S value measured at 0.31 m scale to predict Njs in the 0.61 m tank resulted in significant under prediction of the Njs.13 Following these studies, Grenville, Mak, and Brown attempted to address the effects of scale by using 3 different vessel scales, ranging from 0.3 to 1 m in diameter, and using a variety of tank sizes and impeller geometries.16 All experiments were conducted in dish bottom tanks that were fully baffled with four standard baffles. Grenville et al. also expanded the experimental systems from Zwietering’s work to include a broader range of physical properties, such as viscosity and solids concentration. To create a more comprehensive equation accounting for the effect of liquid viscosity on Njs, Grenville et al. related turbulent eddy dissipation, ε, to the Archimedes number, Ar, which is the dimensionless ratio of buoyant forces to inertial forces. It has been proposed that the particle suspension is primarily driven by turbulent eddies.16,17 For a given tank configuration, the turbulent dissipation ε of these eddies, at Njs can be described by Eqn. 2: ∝
/
(2)
The Reynolds number of the eddies, ReE, is represented in Eqn. 3. "#$ =
%& / '() -/ + , *
(3)
ReE is the minimum turbulence required to suspend particles at a given set of conditions, and as derived by Grenville is related to Ar by Eqn. 4: 4
5 6.766
"#$ = ./01 23 + ,
(4)
When this relationship is rearranged, the “GMB” correlation for Njs is obtained: 8
= .
.99: +; %
6.766
