Hydrodynamics and mass transfer characteristics of asymmetric rotary

Jan 4, 2018 - In the present study, hydrodynamics of lab scale asymmetric rotating impeller column (ARIC) and asymmetric rotating disc column (ARDC) ...
1 downloads 0 Views 2MB Size
Subscriber access provided by READING UNIV

Article

Hydrodynamics and mass transfer characteristics of asymmetric rotary agitated columns Nilesh Virbhadra Hendre, Vaishali Venkatasubramani, Raosaheb Ananda Farakte, and Ashwin Wasudeo Patwardhan Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.7b04239 • Publication Date (Web): 04 Jan 2018 Downloaded from http://pubs.acs.org on January 4, 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 free 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 accessible to all readers and 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.

Industrial & Engineering Chemistry Research 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 42 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

Industrial & Engineering Chemistry Research

Hydrodynamics and mass transfer characteristics of asymmetric rotary agitated columns

Nilesh V. Hendre, Vaishali Venkatasubramani, Raosaheb A. Farakte, Ashwin W. Patwardhan*

Department of Chemical Engineering, Institute of Chemical Technology, Matunga, Mumbai-400019, India [email protected], [email protected], [email protected], [email protected]

*Corresponding author: Tel: 91-22-33612018; Fax: 91-22-33611020 Email address: [email protected]

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

Abstract In the present study, hydrodynamics of lab scale asymmetric rotating impeller column (ARIC) and asymmetric rotating disc column (ARDC) have been studied. Effect of physical properties and operating parameters on drop size, dispersed phase holdup and axial mixing have been investigated. Correlations for prediction of mean drop size and holdup have been developed in terms of power consumption per unit volume.

The AARE values in the

prediction of drop size and holdup using these correlations are 18% and 14% respectively. Further, the hydrodynamic characteristics of ARIC have been compared with that of ARDC. Mass transfer performance of ARIC for the extraction of metal ions from phosphoric acid has been investigated. Effect of impeller speed on percentage extraction and continuous phase overall mass transfer coefficient has been examined. This work provides an insight into the performance of asymmetric rotary agitated extraction columns, useful in the design of such columns. Keywords: Asymmetric rotating impeller column, asymmetric rotating disc column, metal extraction, Power consumption

ACS Paragon Plus Environment

Page 2 of 42

Page 3 of 42 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

Industrial & Engineering Chemistry Research

1. Introduction Solvent extraction is one of the important separation processes widely practiced in chemical industry. Several agitated extraction units are in operation industrially such as mixer-settler, pulsed column, rotating disc column (RDC), Scheibel column, Karr column, Oldshue-Rushton, multi-impeller column and Kuhnni column.1

Mixer-settlers have

advantages of stable operation and high efficiencies.2 However, they occupy large floor area and require huge solvent inventory. Rotating disc columns (RDC) are one of the rotary agitated extraction units in which agitation is induced by discs.

They find immense

applications in petroleum, pharmaceutical and hydrometallurgical industry on account of simplicity in construction, high throughput and low power consumption.3 Nonetheless, their performance is reduced by severe axial mixing.4 Asymmetric rotating disc column (ARDC) was introduced by Misek5 as a modification to RDC. Agitation is induced by discs mounted on a vertical shaft positioned asymmetrically in the column. ARDC is reported to have lower backmixing as compared to RDC.6 The column is divided into mixing zones and settling zones by an asymmetrically positioned vertical baffle.

Further, horizontal stators divide the column into different

compartments. The stator and baffle arrangement in the column causes both continuous and dispersed phases to flow in a helical path. In the mixing zone, fresh droplets of the dispersed phase are created and both phases are mixed intimately by the agitator. In the settling region, partial coalescence of droplets takes place and the two phases disengage.6 Knowledge of hydrodynamic parameters namely drop size (d32) and holdup (εD) are important in the design of extraction columns. Few researchers have reported correlations for d32 and εD in ARDC. Table 1 and 2 lists the correlations for dispersed phase holdup and drop size in ARDC from previous work. Misek7 developed correlations for maximum drop size in

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

Page 4 of 42

ARDC for the case of low and high agitation. Kumar and Hartland8,9 reported a unified correlation for drop size and hold up accounting for external agitation in terms of power dissipation per unit mass. Kadam et al.4 related drop size and hold up to the operating parameters in terms of power consumption per unit volume and physical properties of the system.

Asadollahzadeh et al.10 formulated correlations for d32 and εD in terms of

dimensionless groups by using experimental data of ARDC, RDC and perforated rotating disc column from literature. Performance of industrial extraction columns is assessed based on the mass transfer characteristics, which in turn is affected by axial mixing, holdup and drop size. The physical properties of the liquid system and operating conditions significantly influence the hydrodynamics and mass transfer behavior. A comprehensive analysis of these parameters in industrial columns is not feasible because of their size and requirement of huge inventory. A better approach to understand the performance of such columns is to carry out experiments in scaled-down columns having geometrical similarity to that of industrial columns.

This

approach has been used in the present work to study the hydrodynamic characteristics of an industrial asymmetric rotating impeller column (ARIC). The internal diameter of industrial column is 2 m with an effective height of 6 m. Agitation is induced by pitched blade-disc impellers in ARIC. Other internal components such as vertical baffle and stators are similar in both ARDC and ARIC. The flow generated by the discs in ARDC is mainly in radial direction. In ARIC, the presence of disc induces flow in radial direction while the pitched blades tend to generate axial flow in downward direction.11 Thus, the transport phenomena, which affect the hydrodynamics, would be significantly different in both the columns because of different flow patterns.

ACS Paragon Plus Environment

Page 5 of 42 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

Industrial & Engineering Chemistry Research

The hydrodynamic characteristics of RDC and ARDC have been studied by several researchers. Moreover, performance of RDC with modifications in discs, such as perforated rotary disc contactor (PRDC) has been investigated in the past. However, the performance of asymmetric rotary agitated columns with modified discs has not been explored. Further, as discussed in the previous paragraph, the aim of this work is to understand the hydrodynamics of industrial asymmetric rotary impeller column (ARIC) with 45o pitched blade-disc impellers. For this purpose, a lab scale ARIC has been fabricated maintaining geometrical similarity with that of industrial ARIC. Also, it is necessary to compare the hydrodynamic characteristics of ARIC and ARDC owing to their different flow patterns. In view of this, the main objectives of the present work are as follows 1. To study the effect of agitator speed, phase velocities and physical properties on mean drop size, hold-up and axial mixing in ARIC and ARDC 2. To develop correlations for predicting drop size and holdup in ARIC and ARDC 3. Comparative study between ARIC and ARDC 4. Investigation of mass transfer characteristics in ARIC

2. Experiments 2.1. Experimental set up The schematic of experimental setup is shown in Figure 1. It consists of a cylindrical glass column (0.1 m internal diameter and 0.75 m effective height) equipped with internals of SS 316. Figure 2a depicts the internals of lab scale ARIC. Further, Figure 2b and 2c illustrates the agitator used in ARDC and ARIC respectively. The pitched blade-disc (PBD) impeller used in ARIC consists of four blades attached to a central disc at an angle of 45o. The diameter of PBD impeller is 0.052 m. The vertical shaft is driven by an electric motor of 0.25 HP via a variable speed gearbox. The continuous phase enters the column at top while

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

dispersed phase is introduced at bottom end of the column through sparger. Both the phases are transported from their storage tanks to the column using magnetic pumps. The flow rates of both the phases are controlled by a calibrated rotameter. Disengagement sections are located at the top and bottom part of the column, which allows complete separation of the phases. The dispersed phase is allowed to leave the column via overflow. 2.2. Chemicals and reagents Two liquid systems have been used in the present work, namely system-1: water – heavy normal paraffin (HNP) and system-2: phosphoric acid (29 % by wt) - organic solvent (5 % v/v TBP+45 % v/v D2EHPA+50% v/v HNP). Phosphoric acid, organic solvents, hydrogen peroxide (30% w/v), Potassium chloride (KCl), Sudan III and Remazol Red 3BS were purchased from S D Fine-Chem Limited, Mumbai. The initial concentration of metal ions in phosphoric acid was 110 ppm. Viscosity of the chemicals was measured using Ostwald viscometer. Interfacial tension of the liquid systems was measured by drop volume method.12 All the measured physical properties of both the systems are listed in Table 3. 2.3. Experimental procedure The aqueous and organic phases were saturated before performing experiments to avoid change in physical properties during the experiment. The column was initially filled with continuous phase (aqueous) and the agitation was started at desired rotation speed. Later, the dispersed phase (organic) was introduced at the required flow rate. The interface was maintained by adjusting the height of the U-loop and the system was allowed to attain steady state. Following parameters were measured at various operating conditions.

ACS Paragon Plus Environment

Page 6 of 42

Page 7 of 42 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

Industrial & Engineering Chemistry Research

2.3.1. Drop size Drop size measurements were performed by taking photographs of column using high resolution camera (Cannon 1200D).

The captured images were analyzed using image

analysis software, WebPlotDigitizer. The volume to surface mean diameter, known as Sauter mean diameter (d32) was used to characterize the mean drop size and was subsequently used in the calculation of interfacial area. The drop size was measured at stage number 4, 6, 8, 10 and 12. More than 200 droplets were analyzed in each experiment. The Sauter mean diameter was calculated from individual drop diameters using the following equation N

∑n d

3 i i

d32 =

i =1 N

(1)

∑ ni di2 i =1

where ni denotes the number of drops of diameter di The specific interfacial area (a) was calculated using following equation

a=

6ε d d32

(2)

2.3.2. Dispersed phase holdup Holdup is defined as the ratio of volume fraction of dispersed phase to the effective volume (15 stages) of column. Displacement method was used for the measurement of holdup. The inlet and outlet valves of both the phases were closed simultaneously once steady state was attained. Agitation was continued for five minutes to avoid accumulation of dispersed phase under horizontal stators. Fall in interface level was observed as the dispersed phase accumulates at the top. The volume of dispersed phase between working condition interface and shutdown condition interface was used to calculate the dispersed phase holdup.

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

Page 8 of 42

2.3.3. Axial mixing Tracer injection technique was used in determining axial mixing behavior in both continuous and dispersed phase. For system-1, 5 ml of aqueous solution of KCl (25% w/w) and 5 ml of Sudan III dye solution (200 ppm) were used as tracer for aqueous and organic phase, respectively. The concentration of KCl was measured using a conductivity meter located at the outlet of the continuous phase. The concentration of dye in organic phase outlet was measured using UV-Visible Spectrophotometer. The detection wavelength for Sudan III was 501 nm. For system-2, 10 ml of Remazol Red 3BS solution (2000 ppm) and 5 ml of Sudan III dye solution (200 ppm) were used as tracer for aqueous and organic phase, respectively. Tracer concentrations in aqueous and organic phases were analyzed using UVVisible Spectrophotometer. The detection wavelength for Remazol Red 3BS was 514 nm. Residence time distribution curve was plotted using concentration versus time data for each experiment. The value of mean residence time (tm) and variance (σ2) were calculated from this curve. Axial dispersion model with open-open boundary condition was used to quantify the axial mixing in terms of Peclet number (Pe).13 The value of Pe was calculated using following equation,

σ2 t

2 m

=

2 Pe + 8 Pe + 4 Pe + 4 2

(3)

The skewness of residence time distribution curve was calculated as follows ∞

∫ (t − t )

3

m

skewness =

E (t )dt

0

(4)

δ3

2.3.4. Mass transfer In the present work, mass transfer performance of ARIC was investigated for extraction of metal ions from phosphoric acid using an organic solvent. The oxidation

ACS Paragon Plus Environment

Page 9 of 42 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

Industrial & Engineering Chemistry Research

potential of phosphoric acid was adjusted in the range 500-600 mV by adding hydrogen peroxide, which favors the extraction of metal ions into the solvent.

The distribution

coefficient of the metal ion between the two phases was determined by performing batch experiments with phase ratio of 1:1. The two phases were mixed vigorously for 10 minutes and were allowed to disengage in a separating funnel.

Concentration of metal ions in

aqueous phase was measured using inductively coupled plasma technique.

Distribution

coefficient (Kd) was calculated as the ratio of concentration of metal in organic phase to that in the aqueous phase at equilibrium. The percentage extraction (E) of metal ions in ARIC is calculated as

E=

[Ci ] − [Co ] ×100 [Ci ]

(5)

Qc (Ci − Co ) = Koca ∆CLM VE

(6)

The overall mass transfer coefficient based on continuous phase (Koc) was determined assuming plug flow behavior using equation 6. The log-mean driving force in the above equation was calculated as

∆CLM =

∆C1 − ∆C2  ∆C  ln 1   ∆C2 

(7)

wherein ∆C1 and ∆C2 represents the driving force at top and bottom of the column respectively. Further, the true values of Koc were determined corrected for axial mixing effect, using the method developed by Sleicher14, incorporating the effect of axial mixing. The number of transfer unit assuming plug flow (NTUP) was calculated as follows

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

NTUP =

Ci − Co ∆CLM

Page 10 of 42

(8)

The value of true number of transfer units was calculated using the following equation,

Pec Ped NTU P = NTUT Pec Ped + Q ( NTUT )

(9)

Q = a Pec + b Ped + c( Pec Ped )0.5 − d ( Pec + Ped )0.5 + g ( Pec − Ped )exp [ −k ( NTU )T ] (10) The values of constants a, b, c, d, g and k in above equation have been provided by Sleicher14. Further, the true height of transfer unit and the values of true mass transfer coefficient were determined as shown in following equations,

HTUT =

Koc a =

Z NTUT

(11)

Vc HTUT

(12)

3. Results and discussion 3.1. Dispersed phase holdup Figure 3 and 4 illustrates the effect of agitator speed on dispersed phase holdup for two liquid systems in ARIC and ARDC, respectively. With increase in agitator speed, there is an increase in holdup of dispersed phase in both the columns. At lower agitator speed, the shear force is not sufficient to break the drops, resulting in lower values of holdup. Above a certain speed, higher rate of increase in holdup is noticed. The increase in shear with increase in agitator speed causes drop breakage. Increase in droplet population is observed in the column owing to the lower rise velocities of small droplets and increased tortuosity of the flow path. This leads to enhancement in holdup of dispersed phase. The speed above which substantial increase in holdup is observed, is called critical agitator speed (Ncr).15 The value

ACS Paragon Plus Environment

Page 11 of 42 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

Industrial & Engineering Chemistry Research

of Ncr in ARIC for system-1and system-2 is observed to be around 132 and 127 rpm, while the same in ARDC is 473 and 427 rpm for system-1and system-2, respectively. Also, critical speed is found to be independent of phase velocities. With increase in superficial velocities of continuous and dispersed phase, increase in dispersed phase holdup is observed. With increase in superficial velocity of continuous phase (Vc), there is an increase in drag force on the rising droplets. The relative velocity between the drops and continuous phase reduces, leading to the increase in holdup of dispersed phase. Also, it can be inferred from Figure 3 and 4 that the effect of dispersed phase velocity (Vd) on holdup is more pronounced than that of Vc. Similar types of results have been reported in literature for other types of extraction columns including PRDC3, Oldshue-Rushton16 and multi-impeller columns17,18. The holdup for system-2 is found to be higher than that for system-1 in both ARIC and ARDC. This behavior is attributed to the lower interfacial tension of system-2 as compared to that of system-1.

3.1.1. Development of correlation for prediction of holdup (εD) Few correlations are reported in literature for prediction of εD in ARDC as shown in Table 1. The efficacy of these correlations for predicting experimental εD of the present work has been evaluated based on the average absolute value of relative error (AARE).

AARE =

predicted value − experiment al value × 100 experiment al value

(13)

Equation 13 was used to calculate the values of AARE. The value of AARE using correlations of Kumar and Hartland9 and Kadam et al.4 are 77% and 81% respectively (Figure 5a).

Correlation of Kumar and Hartland8 was formulated majorly using data of RDC,

whereas, data from only a single source of ARDC was incorporated, with an AARE value of 23% with their own data. Kadam et al.4 carried out experiments in the velocity range 0.27 –

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 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

Page 12 of 42

0.55 mm/s and hence it is not able to predict the holdup values at higher phase velocities of present work. The values of holdup predicted by correlation of Asadollahzadeh et al.19 are very low (