Intensification of convective heat transfer in a stator–rotor–stator spinning disc reactor

A stator–rotor–stator spinning disc reactor is presented, which aims at intensification of convective heat‐transfer rates for chemical conversion processes. Single phase fluid‐rotor heat‐transfer coefficients h_r are presented for rotor angular velocities rad s^?1 and volumetric throughflow rates m³s^?1. The values of h_r are independent of and increase from 0.95 kWm^?2K^?1 at ω = 0 rad s^?1 to 34 kWm^?2K^?1 at ω = 157 rad s^?1. This is a factor 2–3 higher than values achievable in passively enhanced reactor‐heat exchangers, due to the 1–2 orders of magnitude larger specific energy input achievable in the stator–rotor–stator spinning disc reactor. Moreover, as h_r is independent of , the heat‐transfer rates are independent of residence time. Together with the high mass‐transfer rates reported for rotor–stator spinning disc reactors, this makes the stator–rotor–stator spinning disc reactor a promising tool to intensify heat‐transfer rates for highly exothermal chemical reactions. © 2015 American Institute of Chemical Engineers AIChE J, 61: 2307–2318, 2015     

Forced convection boiling in a stator–rotor–stator spinning disc reactor

Boiling of a pure fluid inside the rotor–stator cavities of a stator–rotor–stator spinning disc reactor (srs‐SDR) is studied, as a function of rotational velocity ω, average temperature driving force and mass flow rate . The average boiling heat transfer coefficient h_b increases a factor 3 by increasing ω up to 105 rad s^?1, independently of and . The performance of the srs‐SDR, in terms of h_b vs. specific energy input ?, is similar to tubular boiling, where pressure drop provides the energy input. The srs‐SDR enables operation at Wm , yielding values of h_b not practically obtainable in passive evaporators, due to prohibitively high pressure drops required. Since h_b is increased independently of the superficial vapor velocity, h_b is not a function of and the local vapor fraction. Therefore, the srs‐SDR enables a higher degree of control and flexibility of the boiling process, compared to passive flow boiling. © 2016 American Institute of Chemical Engineers AIChE J, 62: 3763–3773, 2016     

Multistage rotor‐stator spinning disc reactor

The scale up of a rotor‐stator spinning disc reactor by stacking single stage rotor‐stator units in series is demonstrated. The gas‐liquid mass transfer per stage is equal to the mass transfer in a single stage spinning disc reactor. The pressure drop per stage increases with increasing rotational disc speed and liquid flow rate. The pressure drop is more than a factor 2 higher for gas‐liquid flow than for liquid flow only, and is up to 0.64 bar at 459 rad s^?1. The high mass and heat transfer coefficients in the (multistage) rotor‐stator spinning disc reactor make it especially suitable for reactions with dangerous reactants, highly exothermic reactions and reactions where selectivity issues can be solved by high mass transfer rates. Additionally, the multistage rotor‐stator spinning disc reactor mimics plug flow behavior, which is beneficial for most processes. © 2011 American Institute of Chemical Engineers AIChE J, 2012     

Residence time distribution in a single‐phase rotor–stator spinning disk reactor

A reactor model for the single‐phase rotor–stator spinning disk reactor based on residence time distribution measurements is described. For the experimental validation of the model, the axial clearance between the rotor and both stators is varied from 1.0 × 10^?3 to 3.0 × 10^?3 m, the rotational disk speed is varied from 50 to 2000 RPM, and the volumetric flow rate is varied from 7.5 × 10^?6 to 22.5 × 10^?6 m³ s^?1. Tracer injection experiments show that the residence time distribution can be described by a plug flow model in combination with 2–3 ideally stirred tanks‐in‐series. The resulting reactor model is explained with the effect of turbulence, the formation of Von Kármán and Bödewadt boundary layers, and the effect of the volumetric flow rate. © 2013 American Institute of Chemical Engineers AIChE J, 59: 2686–2693, 2013     

Hydrodynamical particle containment in a rotor‐stator spinning disk device

A novel type of rotor‐stator spinning disk device is proposed which allows for the entrapment of solid particles solely by hydrodynamic means. In this new configuration, the solid rotating disk is replaced with two conjoined rotors with a variable gap spacing. Liquid is fed through the top stator and can flow out again through the rotor‐rotor interior and the hollow rotation axis. Moreover, the volume between the two rotors is optionally filled with a highly porous reticulated carbon foam. It was found that particle containment was strongly improved by the presence of this reticulated foam as it hinders the buildup of centripetal boundary layer flow near the disks in the interior of the rotor‐rotor assembly. These centripetal boundary layers drag along particles resulting in a loss of containment. Experiments utilizing glass beads showed that particles with a diameter down to 17.8 µm can be completely entrapped when a carbon foam is placed between the two conjoined disks at rotor speeds up to the maximum investigated value of 178 rad s^?1. Additionally, the rotor‐rotor gap did not have an effect on the particle entrapment level when the reticulated carbon foam was omitted and can be ascribed to the build‐up of boundary layers, which is independent of rotor‐rotor distance. © 2015 American Institute of Chemical Engineers AIChE J, 61: 3656–3665, 2015     

Classical cationic polymerization of styrene in a spinning disc reactor using silica‐supported BF3 catalyst

The carbo‐cationic polymerization of styrene has been studied in a Spinning Disc Reactor (SDR) and the results were compared to those observed in a conventional Stirred Tank Reactor (STR). Addition of styrene to a slurry of silica‐supported boron trifluoride (BF₃/SiO₂) in 1,2‐dichloroethane led to uncontrollable reactions in the STR at monomer concentrations > 25%w/w and initial temperatures of 20–25°C. By comparison, monomer concentrations of 75% w/w were safely and controllably polymerized in the SDR at 40°C to yield polymers with molecular weights comparable to those reported in the literature for polymer prepared at ?60°C. Exceptional heat transfer rates achieved in the SDR are sufficient to deal with the heat evolved when styrene is polymerized at concentrations as high as 75% w/w, the reaction proceeding under essentially isothermal conditions. In the present study, the effects of monomer/solvent feed rates, monomer concentrations, disc size, and disc speed on monomer conversions, polymer molecular weights, and polydispersities achieved in the SDR are investigated. Speculative explanations of the observed results are presented in terms of enhanced mixing effects on the polymerization mechanisms in the SDR. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 101: 8–19, 2006     

Preparation of sorbitol from D‐glucose hydrogenation in gas–liquid–solid three‐phase flow airlift loop reactor

A new process for D ‐glucose hydrogenation in 50 wt% aqueous solution, into sorbitol in a 1.5 m³ gas–liquid–solid three‐phase flow airlift loop reactor (ALR) over Raney Nickel catalysts has been developed. Five main factors affecting the reaction time and molar yield to sorbitol, including reaction temperature (T_R), reaction pressure (P_R), pH, hydrogen gas flowrate (Q_g) and content of active hydrogen, were investigated and optimized. The average reaction time and molar yield were 70 min and 98.6% under the optimum operating conditions, respectively. The efficiencies of preparation of sorbitol between the gas–liquid–solid three‐phase flow ALR and stirred tank reactor (STR) under the same operating conditions were compared. Copyright © 2004 Society of Chemical Industry     

An evaluation of the effectiveness of continuous thin film processing in a spinning disc reactor for bulk free-radical photo-copolymerisation

This paper reports on UV-initiated free-radical copolymerisation of vinyl acetate with n-butyl acrylate (VAc-BA) under conditions of thin film flow in a spinning disc reactor (SDR). Almost 40% overall monomer conversion can be achieved in under 5 s under optimised operating conditions in the SDR, with controlled molecular weight properties of the copolymer, highlighting the good levels of mixing in the film. Residence time on the SDR is a limiting factor in the extent of conversion achievable in a single pass. Comparison with a static film demonstrates the superiority of the SDR in maintaining a high overall rate of polymerisation. Composition of the copolymer formed in the SDR indicates that, due to its plug flow behaviour, the SDR cannot address the inherent problem of compositional drift.We have shown that efficiency of light absorption is dictated by conditions favouring longest UV exposure times, rather than thinner films on the disc. Initiator decomposition efficiency, an important consideration in the overall rate of the co-polymerisation, is enhanced by lower fluid flowrates. This study highlights the promising technology offered by the SDR in combination with UV irradiation for the exploitation of photo-copolymerisation as a viable method for bulk copolymer synthesis.     

Mass Transfer Characteristics in a Rotor‐Stator Reactor

Mass transfer characteristics in a rotor‐stator reactor in terms of the overall volumetric mass‐transfer coefficient (K_xa) using the N₂‐H₂O‐O₂ system were investigated. The effects of various operating parameters including rotation speed, liquid volumetric flow rate, and gas volumetric flow rate on K_xa were systematically examined, and a gas‐liquid mass transfer model was established to predict K_xa. Results reveal that K_xa increased with higher rotation speed, liquid volumetric flow rate, and gas volumetric flow rate. The results also confirm that the predicted values of K_xa were in agreement with the experimental values with deviation within 15 %. The results contribute to a better understanding of mass transfer characteristics in rotor‐stator reactors.     

Cyclic mass transport phenomena in a novel reactor for gas–liquid–solid contacting

This study introduces a novel reactor concept, referred to as the Siphon Reactor, for intensified phase contacting of gas–liquid reactants on heterogeneous catalysts. The reactor comprises a fixed catalyst bed in a siphoned reservoir, which is periodically filled and emptied. This serves to alternate liquid–solid and then gas–liquid mass transfer processes. As the duration of each phase can be manipulated, mass transfer can be deliberately harmonized with the reaction. Residence time experiments demonstrate that, in contrast to periodically operated trickle‐bed reactors, the static liquid hold‐up is exchanged frequently and uniformly due to the complete homogeneous liquid wetting. A mathematical model describing the siphon hydrodynamics was developed and experimentally validated. The model was extended to account for a heterogeneously catalyzed gas–liquid reaction and capture the influence of siphon operation on conversion and selectivity of a consecutive reaction. © 2016 American Institute of Chemical Engineers AIChE J, 63: 208–215, 2017     

Liquid‐phase axial dispersion of turbulent gas–liquid co‐current flow through screen‐type static mixers

This article discusses the characteristics of turbulent gas–liquid flow through tubular reactors/contactors equipped with screen‐type static mixers from a macromixing perspective. The effect of changing the reactor configuration, and the operating conditions, were investigated by using four different screen geometries of varying mesh numbers. Residence time distribution experiments were conducted in the turbulent regime (4500 < Re < 29,000). Using a deconvolution technique, the RTD function was extracted to quantify the axial/longitudinal liquid‐phase dispersion coefficient. The findings highlight that axial dispersion increases with an increasing flow rate and/or gas‐phase volume fraction. However, regardless of the number and geometry of the mixing elements, reactor configuration, and/or operating conditions, the recorded liquid‐phase axial dispersion coefficients in the presence of screens was lower than that for an empty pipe. Furthermore, the geometry of the screen was found to directly affect the axial dispersion coefficient in the reactor. © 2016 American Institute of Chemical Engineers AIChE J, 63: 1390–1403, 2017     

Structure development during dry–jet–wet spinning of acrylonitrile/vinyl acids and acrylonitrile/methyl acrylate copolymers

Dry–jet–wet spinning of three copolymers, poly(acrylonitrile/methyl acrylate), poly(acrylonitrile/methacrylic acid), and poly(acrylonitrile/itaconic acid), was performed with a dimethylformamide/water (60:40 v/v) coagulation bath at different temperatures (10–40°C). The fibers were stretched to different levels (1.1–6×) in boiling water, collapsed, and annealed over a heater plate at 130°C. The effects of the polymer composition, coagulation bath temperature, and draw ratio on the cross‐sectional morphology, structure, and tensile properties are reported. The cross‐sectional shape of the gel fibers underwent a transition from a kidney shape to an oval shape, and macrovoids began to appear at higher temperatures. However, F(AN/IA) gel fibers changed from a kidney shape to an irregular shoe type with a gel network of interconnected polymer fibrils. For F(AN/MAA) gel fibers, the diameter increased from 45 to 67 μm when the coagulation bath temperature was increased from 10 to 40°C, and the denier value decreased from 17.5 to 14.3 den/filament. The strength, modulus, and elongation at break decreased with an increase in the coagulation bath temperature. For F(AN/MAA) fibers coagulated at 10°C in a spin bath, the strength increased from 0.43 to 2.213 g/den, the modulus increased from 27 to 76 g/den, and the density increased from 1.177 to 1.196 g cm^?3 when the gel fibers were drawn to 6×. However, 6× drawn F(AN/MA) fibers had a higher strength (3.1 g/den) and elongation (14.6%) in a 40°C coagulation bath. F(AN/IA) fibers could be drawn only to a draw ratio of 4× instead of the 6× draw ratio for F(AN/MAA) and F(AN/MA) fibers. Therefore, the final F(AN/IA) fibers exhibited poor mechanical properties (tenacity = 0.81 g/den, modulus = 22 g/den, and elongation at break = 8%). The crystallinity did not change significantly (χ_c = 61–63%) with the draw ratio, but the crystal size increased from 22.9 to 43.4 Å and orientation factor from 0.41 to 0.78. The dichroic ratio, measured with Fourier transform infrared, decreased with an increase in the draw ratio, but the sonic modulus and crystalline orientation values increased with an increase in the draw ratio. Thermomechanical data show a maximum physical shrinkage of 51.7% for 6× drawn F(AN/MA) and a minimum physical shrinkage of 30.5% for 4× drawn F(AN/IA) fibers. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 773–787, 2002     

Hydrodynamic modeling on the external liquid–solid wetting efficiency in a trickling flow reactor

A three‐region model was proposed, which considers the bed cross section being composed of a stagnant liquid region, a liquid film region, and a rivulet flow region. To estimate the fractions of the three regions, the fraction of film flow was evaluated first, by transforming the complex trickling flow texture into pure liquid film flow. Through the measurements of liquid holdup and pressure drop for the film flow, a relationship between relative permeability and gas saturation was established, and from which the fraction of film flow region was obtained. It shows packing size is most important to the faction of rivulet flow. The external wetting efficiency of the packing was correlated as the sum of two‐third power of the liquid film fraction and the rivulet flow fraction, besides, a correlation based on Reynolds and Galileo numbers of the two phases in the form of was proposed. © 2012 American Institute of Chemical Engineers AIChE J, 59: 283–294, 2013     

Aldol condensation of n‐butyraldehyde in a biphasic stirred tank reactor: Experiments and models

To model a biphasic stirred tank reactor, intrinsic reaction kinetics and interfacial area are required. In this study, reactor modeling for n‐butyraldehyde aldol condensation was investigated under industrially relevent conditions. The interfacial area in the reactor was directly measured using a borescope system under appropriate temperature, NaOH concentration and rpm conditions. To estimate the interfacial area, a semiempirical correlation was developed, which provides good estimates within ±15% error. The reactor model based on two‐film theory was developed, combining the interfacial area and intrinsic reaction kinetics reported in our prior work. The model was verified by reaction experiments in the range 0.05–1.9 M NaOH, 80–130°C, and 600–1000 rpm. The prediction errors using the interfacial area from direct measurements and the correlation were ±8% and ±15%, respectively, suggesting that the model accuracy may be improved with better interfacial area estimation. © 2015 American Institute of Chemical Engineers AIChE J, 61: 2228–2239, 2015     

A preliminary evaluation of a continuous‐flow reactor for liquid–liquid–solid phase‐transfer catalyzed synthesis of n‐butyl phenyl ether

This study evaluates the feasibility of using a continuous‐flow stirred vessel reactor (CFSVR) to synthesize n‐butyl phenyl ether (ROPh) from n‐butyl bromide (RBr) and sodium phenolate (NaOPh) by liquid–liquid–solid phase‐transfer catalysis (triphase catalysis). The factors affecting the preparation of triphase catalysts, the etherification reaction in a batch reactor, and the performance in a CFSVR were investigated. The kinetic study with a batch reactor indicated that when the initial concentration of NaOPh or RBr was high, the conversion of RBr would depend on the initial concentration of both RBr and NaOPh. The reaction can be represented by a pseudo‐first‐order kinetic model when the concentration of NaOPh is in proper excess to that of RBr, and the apparent activation energy is 87.8 kJ mol^?1. When the etherification reaction was carried out in the CFSVR, the catalyst particles did not flow out of the reactor, even at a high agitation speed. The conversion of RBr in the CFSVR was, as predicted, lower than that in the batch reactor, but was higher than the theoretical value because the dispersed phase is not completely mixed. Copyright © 2004 Society of Chemical Industry