Le Li,Yansheng Zhao,Wenhao Lian,Chun Han,Qian Zhang,*,Wei Huang,4,*

1 Key Laboratory of Coal Science and Technology of Ministry of Education and Shanxi Province,Taiyuan University of Technology,Taiyuan 030024,China

2 School of Chemistry and Chemical Engineering,Taiyuan University of Technology,Taiyuan 030024,China

3 School of Chemical Engineering and Technology,North University of China,Taiyuan 030024,China

4 Taiyuan Coal Conversion Technology Engineering Co.,Ltd.,Taiyuan 030024,China

Keywords:Internals Hydrodynamics Heat exchanger tube Bubble/slurry bubble column reactors

ABSTRACT Bubble/Slurry bubble column reactors(BCR/SBCR)are intensively used as multiphase reactors for a wide range of application in the chemical,biochemical and petrochemical industries.Most of these applications involve complicate gas–liquid/gas–liquid–solid flow behavior and exothermic process,thus it is necessary to equip the BCR/SBCR with heat exchanger tubes to remove the heat and govern the performance of the reactor.Amounts of experimental and numerical studies have been carried out to describe the phenomena taking place in BCR/SBCRs with heat exchanger tubes.Unfortunately,little effort has been put on reviewing the experiments and simulations for examining the effect of internals on the performance and hydrodynamics of BCR/SBCR.The objective of this work is to give a state-of-the-art review of the literature on the effects of heat exchanger tubes with different types and configurations on flow behavior and heat/mass transfer,then provide adequate information and scientific basis for the design and the development of heat exchanger tubes in BCR/SBCR,ultimately provide reasonable suggestions for better comprehend the performance of different heat exchanger tubes on hydrodynamics.

1.Introduction

The growing environmental problems have attracted widely attention to the access of clean fuels from coal–based technologies.Gas–liquid–solid processes of Fischer–Tropsch (F–T) synthesis,alcohol and ether synthesis and coal liquefaction process have gained much attention from both academic and industrial interests[1–7].Great efforts have been devoted to reactor development to satisfy the requirements of different gas–liquid–solid processes.Different types of reactors were developed:the multi-tubular fixed bed,gas–liquid–solid fluidized bed and slurry reactor [7–10],as shown in Fig.1.The slurry reactor and multi-tubular fixed bed were used for the low–temperature process,while the circulating fluidized bed was used for the high–temperature process [11].Among them,the slurry reactor plays a prominent role in all types of gas conversion processes,due to their advantages of simple construction,large interface area,excellent mass transfer and heat transfer characteristics,plug-free operation,online catalyst addition and withdrawal ability.The advantages and disadvantages of slurry reactors are summarized in Table 1.The slurry reactors are of considerable industrial importance as proved by their widespread use and popularity in various industries,which results in intensive traditional theoretical analyses[12,13]and experimental investigations on them in the past [14–17].Their scale-up and design difficulties have been the subject of several studies in recent years,but have yet to be tackled systematically enough to optimize their performance.

Table 1Advantages and disadvantages of slurry reactors

Fig.1.Reactor types used for gas–liquid–solid processes:(a) fixed bed reactor,(b) fluidized bed reactor,(c) slurry reactors.

Slurry reactors can be divided into stirred tank reactor,bubble column,internal-loop airlift reactor and external-loop airlift reactor according to different structure [11],whose design and scaleup depend on a great quantity of empirical correlations obtained from different scale experiments,as shown in Fig.2.The aspect ratio (height-to-diameter ratio) varies from 1–20 (typically 3–10).In their simplest form,the slurry reactors are basically cylindrical vessels equipped with a gas distributor at the bottom and tube banks or coils for heat transfer at the middle.More complicated structures with internal or external down-comers are used to provide recirculation of the liquid.The gas is sparged in the form of bubbles into vessel containing either a liquid phase or a liquid–solid suspension.According to the size of the slurry reactors and the superficial gas velocity,three flow regimes [18–20] exist:homogeneous regime or bubble flow,transition regime,heterogeneous regime or churn turbulent flow.At low to moderate gas flow rates,homogeneous flow is produced,in which the bubbles are small and the radial distribution of gas holdup is uniform [21].The size of bubbles depends mainly on the nature of the gas distribution and the physical properties of the liquid.At high gas flow rates,the heterogeneous flow occurs,in which the effect of bubble coalescence and breakage is significant,and the radial distribution of gas holdup is uneven [21].Hence,the bulk liquid circulation exists in this flow regime [18].Transition flow regime is observed in between bubbly flow regime and churn turbulent flow regime.Observed flow regimes in slurry reactors are shown in Fig.3.Current industrial interest is in the heterogeneous flow regime because it allows higher gas throughputs and a more homogeneous catalyst suspension for higher volumetric productivities.Therefore,the study of heterogeneous flow has direct industrial application value.

The hydrodynamics in slurry reactors,which is the key factor for the development and scale-up of the reactor,determines the phase mixing,heat and mass transfer in the column.However,due to the fact that the multiphase flow behaviors in slurry reactors involve gas–liquid or gas–liquid–solid multiphase turbulent motion,and various complicated internal components are often installed in the tower,which makes the flow pattern complicated.Moreover,the reactions in the slurry reactors are always performed at conditions with high pressure and high temperature,making the direct observation of the internal flow behaviors become impossible.To develop design tools for engineering purposes,a large amount of research has been carried out in the area of computational fluid dynamics (CFD) modeling that has clearly emerged as a promising tool for the simulation of local hydrodynamics in slurry reactors [3,22–30].CFD is an approach based on the first principles in which the governing equations of continuity,momentum and energy for the gas–liquid–solid three phases in slurry reactors are solved.The detailed information provided by CFD,such as flow pattern,turbulence,interfacial area density,heat transfer characteristics,etc.can be used for the design objectives of the reactor and its internal components.Moreover,the proper use of CFD modelling is helpful in developing scale-up strategies and further optimizing the reactor.Nevertheless,at present,the basic problems about the closure relationship,interphase force,bubble coalescence and breakup,energy dissipation mechanism of multiphase turbulence have not been fully understood.In addition,the computational complexity of CFD is large,and the convergence and stability of the algorithm are affected by many factors.Therefore,establishing a simple and reliable hydrodynamics model is the primary problem to be solved in CFD study of the BCR/SBCR.

Fig.2.Types of slurry reactors:(a) stirred tank reactor,(b) bubble column,(c) internal-loop airlift reactor,(d) external-loop airlift reactor.

Fig.3.Observed flow regimes in slurry reactors (adapted from Ref.[21]).

Recent experimental and simulation researches with slurry reactors frequently focuses on the following topics:flow pattern investigations[31–34],gas holdup studies[35–38],bubble characteristics [36,39,40],interfacial forces and turbulence models studies [41–43],local and average heat transfer measurements[20,44,45],and mass transfer studies[46–48].The effects of superficial gas velocity,solid sizes and concentration,column dimensions,operating condition,i.e.temperature and pressure are commonly investigated in these studies [30,49].However,most of the above-cited studies were performed in empty columns,which unfortunately leaves uncertainties about the impact of internals.Generally,BCR/SBCR can be equipped with two types of internals:horizontal internals,such as gas distributor or horizontal tube bundles,and vertical internals.Horizontal internals are generally used to control the flow behavior to achieve higher productivities,while vertical internals are preferred as means of heat removal because they provide higher heat transfer area per reactor volume.Unfortunately,little effort has been put on reviewing the experiments and simulations for examining the effect of internals on the performance and hydrodynamics of BCR/SBCR,although they are essential in industrial setups.Even fewer studies describe the configuration of internals and their geometrical details.Given the limitations on information in the opening literature,it is clear that the internals types and structures required in industrial applications need to be further studied to accurately account for their effect on flow parameters determining BCR/SBCR performance.Moreover,there are no generally accepted procedures for the proper design of internals.

To overcome these gaps,we summarize and analyze the effect of heat exchanger tubes on hydrodynamics and heat/mass transfer inside the BCR/SBCR.The objective of this work is to give a state-ofthe-art review of the literature over the last 20 years on the effects of heat exchanger tubes with different types and configurations on flow behavior,heat transfer and mass transfer,then provide adequate information and scientific basis for the design and the development of heat exchanger tubes in BCR/SBCR,ultimately provide some reasonable suggestions for potential research to better comprehend the effect of heat exchanger tubes on hydrodynamics.In Section 2,the necessity of using heat exchanger tubes and several characteristics in the design of heat exchanger tubes are given.In Section 3,the types of different heat exchanger tubes are compared and the effects of heat exchanger tubes with different configurations (the geometrical arrangement,the pitch,the inter-tube gap,the occluded CSA,the diameter of the tubes)on flow behavior,heat transfer and mass transfer are illustrated.In Sections 4 and 5,the conclusions and some reasonable suggestions for potential research to better comprehend the effect of heat exchanger tubes on flow behavior and reaction performance are put forward.

2.Necessity of Using Heat Exchanger Tubes in BCR/SBCR

In the BCR/SBCR design,there are several following characteristics to be considered in the different gas–liquid–solid processes mentioned in Introduction:(1)In the industrial BCR/SBCR,the variation and distribution of the gas holdup and the bubble size have direct significant effects on the gas–liquid–solid mass transfer and heat transfer,which further affects the reaction conversion and productivity.Hence,the distribution of gas holdup and bubble size are particularly vital.(2) The reactions are strong exothermic reaction with a large thermal effect.Keeping the temperature under control in the reaction process is very important to keep the activity and stability of the catalyst and the selectivity of the product.(3)If the reaction conditions are not well controlled,more side reactions may occur.Maintaining a uniform temperature distribution is helpful to avoid the side reactions caused by local overheating.Corresponding to the above characteristics,the BCR/SBCR must not only realize a high gas holdup,a more uniform bubble size distribution,and high mass transfer rates,but also remove the accompanying large amount of reaction heat efficiently and maintain a uniform temperature profile.Moreover,the existence of small bubbles should be preferred and the presence of large bubbles should be avoided to improve the mass transfer rates.Sample reactive systems are listed in Table 2 to illustrate the necessity of utilizing proper heat exchanger tubes in multiphase reactors.Therefore,the authors concluded that,since the effective region of the gas distributor is limited to only a certain height,the structure of heat exchanger tubes has a vital effect on the flow field in its region.The first consideration in the reactor design is to equip the BCR/SBCR with a reasonably designed heat exchanger tubesto obtain a higher gas holdup and a more uniform bubble size distribution.Then removing the reaction heat effectively and maintaining a uniform temperature profile by constructing reasonable arrangements of heat exchanger tubes.Therefore,special structural of heat exchanger tubes are needed to decrease the bubble size and increase the gas holdup by increasing bubble breakup,and to enhance the heat transfer by increasing liquid turbulence and circulation.Finally,the stable operation of the reaction and efficient performance require the optimal combination of gas distributor and heat exchanger tubes.

Table 2Applications of slurry reactors

3.Effect of Internals in Slurry Reactors–Heat Exchanger Tubes

BCR/SBCR are widely equipped with heat exchanger tubes to remove the excess heat and to maintain the desired conditions.In the process of liquid methanol synthesis,heat exchanger tubes occupying 5% of the CSA are equipped to maintain an isothermal operation,while the internals covering about 22%–25% of the CSA are required in a highly exothermic process of F–T synthesis.The existence of heat exchanger tubes strongly impacts and alters the fluid dynamics by affecting the phases distributions of the reactants.Therefore,understanding the effect of heat exchanger tubes and their arrangements on the phase distribution is crucial to improving the design of these heat exchanger tubes to enhance mass transfer and heat transfer in these reactors.

3.1.Type and design of heat exchanger tubes

Indirect and direct heat transfer is an important aspect in the design of slurry reactors used for many industrial organic and inorganic processes.Longitudinal flow or cross-flow tube bundle heat exchangers,jacket cooling,direct evaporative cooling or circulation cooling are possible methods for this purpose[59,60].Fig.4 shows some examples of indirect heat transfer in slurry reactors.The manufacturing process,maintenance cost and features of different types of heat exchanger tubes are summarized in Table 3.Generally,different types of heat exchanger tubes,such as jacket coolers(applicable for small reactors only),external heat exchanger tubes and horizontal or vertical internal heat exchanger tubes are typically used.The latter ones are the most common types as they provide a direct excess heat removal and higher heat transfer area per reactor volume[61].Large specific heat exchanging surfaces can be installed by using tube bundles.Therefore,in the case of highly exothermic reactions,longitudinal flow tube bundles are suitable for producing high-pressure steam.

Table 3Comparison of different types of heat exchanger tube

However,design aspects of these heat exchanger tubes remain been ignored,such as the specifications of the diameter,number and gap of tubes.Various operation and design problems,such as non–uniformity of gas holdup profile and low heat transfer efficiency,may arise due to improper selection or design of heat exchanger tubes.Apart from these tribulations,some practical concerns exist.For instance,(a)the location of heat exchanger tubes to gas distributor,(b) the heat transfer area,(c) the selection of the possible number of pipes/rings or the CSA occluded by the tubes,which are decided by the surface area required for heat transfer,(d) the costs.

3.2.Effect of heat exchanger tube on flow

So far,no systematic method has been reported to guide how to select different parameters to design heat exchanger tubes in BCR/SBCR.Although the lack of systematic research on heat exchanger tubes,some guidelines on their design can be extracted from the information available in the current literature.The properly design of heat exchange tubes requires the fully understanding of the effect of the following three main parameters on hydrodynamics in bubble columns:the ratio of occluded CSA by internals,tube configuration (including geometrical arrangement,pitch and inter-tube gap),and tube diameter.In what follows,the effect of these main parameters on hydrodynamic,heat transfer and mass transfer will be summarized.

Fig.4.Different types of heat exchanger tube used in slurry reactors:(a)serpentine heat exchanger,(b)single tube heat exchanger,(c)squirrel-cage heat exchanger,(d)tubebundle heat exchanger.

Over the years,various techniques have been used in the experimental studies of empty bubble columns or slurry reactors,but only limited studies have been focused on the effects of heat exchanger tubes on the gas–liquid or gas–liquid–solid hydrodynamics.Bashaet al.[62]reviewed and summarized current experimental and simulations conducted in bubble/slurry bubble columns,and recommended further experimental studies be conducted to address the effect of the heat exchanger tubes arrangement on the hydrodynamics of these reactors.

Chenet al.[63],using γ ray computed tomography (CT) and computer automated radioactive particle tracking(CARPT)quantitatively assessed the effect of the column internals on the gas holdup profiles,liquid recirculation,and turbulent stresses for a 0.44 m diameter bubble column with vertical internals covering 5% of column’s CSA.The experiment setup and configuration of the internals were shown in Fig.5.A slight increase (about 10%)in gas holdup and a significant decrease in turbulent stresses and the eddy diffusivities in the radial direction were reported when the internals inserted,as shown in Fig.6.The presence of vertical tube bundles will lead to the hindrance of radial bubble motion.Nevertheless,no significant difference was observed on liquid recirculation for the columns with and without internals.Therefore,the above all results indicate that the presence of the vertical internals is conducive for improving the gas holdup and physically reducing the length scales of turbulence in the radial direction.Moreover,they do not affect the gas–liquid recirculation flow pattern for the rang of superficial gas velocities (0.02,0.05 and 0.10 m·s-1)studied.In this work,the range of superficial gas velocities covered was low and there is no further study on the effect of different CSA of the vertical internals on the hydrodynamics.Thus,it is not possible to fully evaluate the effect of internals at highsuperficial gas velocity that would guarantee a high volumetric productivity as desired especially in the F–T process.Further,the low CSA internals cannot effectively remove the generated heat from highly exothermic processes,hence evaluating the impact of dense internals is necessary.

The effects of the occluded CSA by the heat exchanger tubes on the bubble dynamics in bubble/slurry bubble columns have already been investigated in many studies.Zhanget al.[64] measured the profiles of gas holdup and liquid velocity in a 0.5 m diameter slurry bubble column with internals covering 11%(N=40)and 5.3%(N=18)of the CSA.The experimental results showed that the vertical internals would remarkably enhance the gas holdup and the large-scale liquid circulation,while impede the radial turbulent motion of the liquid and bubbles.Similarly,Forretet al.[65] studied liquid mixing in a 1 m diameter bubble column with vertical cooling tubes covering 22% of column’s CSA,and found that the presence of vertical heat exchanger tubes significantly affected both large-scale liquid recirculation and local dispersion.The main effect of internals is to reduce radial dispersion and to increase large scale recirculation,as shown in Fig.7.Youssef and Al-Dahhan [66] examined the effect of vertical bundles that mimic the process used in methanol synthesis (5% covered CSA) and the F–T process (22% covered CSA) on the local gas holdup,gas–liquid interfacial area,bubble chord length,and the bubble velocity distributions in an air–water system,as shown in Fig.8.The gas holdup and interfacial area increases for the case of dense internals and the significantly decreased bubble chord length were obtained with insertion of dense internals,as shown in Fig.8 and Table 4.Nevertheless,the influences of the sparse internals on the gas holdup and the large-scale liquid circulation were insignificant,which was in agreement with Chenet al.[63].Subsequently,Youssefet al.[67] developed the research on gas holdup and bubble characteristics in a 0.45 m diameter bubble column with vertical cooling tubes covering 5%,10%,15%,20% and 25% of column’s CSA and at superficial gas velocities in a wide range 0.03–0.45 m·s-1covering bubbly through churn turbulent flow regimes,as shown in Fig.9.They reported enhanced overall gas holdup with increased percentage coverage of column CSA and decreases bubble chord length by internals.A vigorous recirculation behavior is also obtained as a result of the insertion of the internals in a larger-diameter column.Recently,Al Mesferet al.[68] further investigated the impact of dense internals on gas holdup distribution in a 0.14 m diameter bubble column,and at superficial gas velocities in range of 0.05–0.45 m·s-1covering homogeneous to heterogeneous flow regimes.In their work,in addition to the presence of internals progressively increases the overall gas holdup and local gas holdup,other interesting phenomena were observed,including that the effect of dense tubes became insignificant at a high superficial velocity,and the shape of the gas holdup profiles near the wall were affected by the vertical internal arrangements.Using the same sized tubes and the same occluded CSA,the arrangements of the vertical bundles significantly influence the gas holdup distribution over the CSA of the column.

Fig.5.Schematic of the bubble column and internals:(a) experimental setup,(b) configuration of the internals (adapted in [63],1′′=2.54 cm).

Fig.6.Comparison of various parameters at Ug=0.1 m·s-1:(a) gas holdup,(b) turbulent stresses,(c) axial eddy diffusivities,(d) radial eddy diffusivities (adapted in [63]).

Fig.7.Schematic of liquid recirculation in bubble columns without (a) and with internals (b) (adapted in [65]).

Fig.8.(a)Schematic of bubble column setup.(b)Internals configuration:(A)5%covered CSA and(B)22%covered CSA.(c)Effect of internals on the local gas holdup.(d)Effect of internals on the interfacial area (adapted in [66],1′′=2.54 cm).

Table 4Quantities of the chord length distribution for different internals arrangements(adapted in [66])

Thus far,the effect of heat exchanger tubes configurations on the hydrodynamics of the bubble column has also been investigated.Sultanet al.[69] employed three arrangements (i.e.,hexagonal,circular without a central internal,and circular with a central internal) to investigate the impact of the vertical internal configurations in a 0.1524 m diameter bubble column,as shown in Fig.10.A remarkable increase in the gas holdup values and a better spread of the gas spread over the entire CSA of the column were obtained with the hexagonal configuration,as shown in Fig.11.The bubble column with tubes configured in a circular arrangement with an extra central tube displayed distinctly asymmetric gas holdup distributions.The experimental data in this work are helpful to evaluate the three-dimensional CFD simulations to better predict the hydrodynamics parameters involved in these types of reactors.

In addition,some numerical simulation studies had been conducted to address the effect of the vertical heat exchanger tubes arrangement on the hydrodynamics.Larachiet al.[70] built twofluid Euler continuum transient 3D computational fluid dynamics(CFD)simulations for five pilot-scale plant bubble columns:vessels of uniform filling(dense and sparse),vessels of non-uniform filling with large core and wall clearances,and equal CSA hollow vessels,as shown in Fig.12.Their simulation revealed that the arrangements of the tube bundles significantly affected the liquid circulation pattern of the bubble columns.The liquid gross flow structure follows a core–annulus structure with the insertion of a uniform implantation of vertical internals.On the contrary,non-uniform internals arrangements implied more complex flow patterns with even liquid downward circulation in the core region when the tubes were larger than near the wall,as shown in Fig.13.However,these results for bubble columns with internals were not validated against any experimental data because there are little experimental studies for bubble columns with bundle of heat exchanger tubes.Guo and Chen [71] investigated the impacts of longitudinal flow tube-bundle heat exchangers on gas–liquid hydrodynamic of bubble columns.The study was simulated using an Eulerian two fluid model coupled with a population balance method (TFM–PBM).The numerical results showed that more bubbles with smaller bubble size were predicted in the bubble column with the vertical internals inserted,and the turbulent dissipation rates increased significantly in the gaps between the internal walls.Meanwhile,the gas holdup increased when the dense internals insertion and the vertical internals and the configurations also affect the overall liquid circulation,which was in agreement with the above-mentioned experimental results.

Fig.9.Configurations of internals bundles covering (a) 5%,(b) 10%,(c) 15%,(d) 20% and (e) 25% of the CSA (adapted in [67]).

Fig.10.Schematic and photos of the top view of the configurations of the vertical internal tubes:(a)circular design,(b)hexagonal design,(c)circular design with one tube at the center (adapted in [69]).

Fig.11.Effect of the vertical internal tube configurations on time-averaged CSA gas holdup distributions:(a)without internals,(b)hexagonal configuration,(c)circular design without central tube configuration,(d) circular design with one tube at the center (adapted in [69]).

Fig.12.Numerical mesh used:(a) empty vessel,(b) dense arrangement of internals,(c) sparse arrangement of internals,(d) star arrangement:wall clearance,(e) star arrangement:core clearance (adapted in [70]).

Nevertheless,it is insufficient to study the influence of the vertical heat exchanger tubes covering different column’s CSA merely on hydrodynamics of bubble column.Much attention should also be paid on the effect of the types of internal tubes and the internals size on the flow field behavior when designing the heat exchanger tubes.The effect of the tubes’ diameter on bubble dynamics has not yet been properly addressed.Kagumba and Al-Dahhan [72] examined the impact of dense internals size encountered in F–T synthesis on bubble properties in a 0.14 m inner diameter bubble column.Two different configurations,namely,30 tubes of 2.54 cm diameter and 8 tubes of 1.27 cm diameter equally covering 25% of the CSA,were tested in a wide range 0.03–0.45 m·s-1covering bubbly flow regime through the churn turbulent flow regimes,as shown in Fig.14.The results suggested that the tubes with 2.54 cm diameter gave consistently higher overall and local gas holdup than that of the tubes with 1.27 cm or empty column,as shown in Fig.15.The effect of the internals diameter was insignificant in the churn turbulent flow regime.It is obvious therefore that,whereas the presence of internals affects the bubble dynamics in bubble columns,the gas holdup enhancement attributable to higher breakup rates due to dense internals size is negligible at high superficial gas velocity.However,the tube bundles in this work were configured differently for each case.For instance,the internals with smaller diameter were configured in a hexagonal-like shape,while the larger ones were configured in a circular shape.Therefore,these changes in the local gas holdup cannot be determined by the change in tube size or the arrangement of the tube bundles.Further investigations are necessary to properly determine which variable was responsible for the results.

Fig.13.Time-averaged contour plots of the axial liquid velocity:(a) empty vessel,(b) dense arrangement of internals,(c) sparse arrangement of internals,(d) star arrangement for the 5 mm bubble size:wall clearance,(e)star arrangement for the 19 mm bubble size:wall clearance,(f)star arrangement:core clearance.Color scale was voluntarily exaggerated to distinguish up-flowing (red) and down-flowing (blue) regions (adapted in [70]).

Fig.14.Internals configurations covering 25% CSA:(a) 1.27 cm diameter,(b) 2.54 cm diameter (adapted in [72]).

Pradhanet al.[73] investigated the effect of types of internals on overall gas holdup and bubble characteristics in a bubble column of 2.5 m height and 0.102 m diameter provided with helical coils and a vertical straight tube bundle.The helical coils provided higher gas holdup than vertical tubes,and a gas holdup enhancement of up to 55% was achieved when the helical coil were used.The difference was attributed to larger inter-tube gaps for the vertical internals that provided more space for larger bubbles to escape,thus decreasing the gas holdup,unlike the helical coil in which only smaller gaps were present.Further,the smaller gaps would provide more obstruction to the gas flow,resulting in longer residence time for the gas phase.Balamuruganet al.[74]examined the effect of vertical internals such as vibrating helical springs on gas holdup in bubble columns.The internals,whose structure was similar to that in the Fig.4(a),were coiled structures positioned vertically along the vertical axis of the bubble column,as shown in Fig.16.Enhanced the overall gas holdup was observed with the insertion of vibrating helical spring internals in bubble columns.Additionally,they concluded that the vibrating springs breakup the gas into fine bubbles,which effectively reduces their rise velocity and enhances their residence time in the columns.

To sum up,increased gas holdup,liquid axial velocity and the large-scale liquid circulation with vertical internal tubes inserted in a larger-diameter column were observed on the one hand,while hinder the lateral turbulent motion of liquid and bubbles on the other hand.The effect results in steeper distributions of gas holdup and axial velocity,and enhancement of liquid circulation and gas phase back-mixing,which increased the risk of scale up for slurry bubble column reactor with dense vertical internal bundles.In addition,although a large amount of evidence has qualitatively shown that the heat exchanger tubes affected the bubble dynamics,and particularly the gas holdup and the bubble size distributions,very few quantitative analyses of the effects of internals on hydrodynamics of bubble/slurry columns were conducted in literature.

3.3.Effect of heat exchanger tube on heat transfer

Fig.15.Effect of size of internals on radial profiles of local gas holdup:(a)Ug based on free CSA,(b) Ug based on total CSA (adapted in [72]).

The effect of heat exchanger tubes on the heat-transfer coefficient in BCR/SBCR has been studied by only a few researchers[75–77].Table 5 summarizes recent reported heat transfer studies in bubble/slurry bubble columns with multiple internals.The presence of internals could change the hydrodynamic conditions,leading to an increase in turbulence and heat transfer.However,in these studies,only a few attempts conducted by Saxena and coworkers cover the effect of heat exchanger tubes on heat transfer in the BCR/SBCR.Saxenaet al.[75] reported the effect of heat exchanger tubes on the heat-transfer coefficient is insignificant in a large diameter column (0.3 m),while the effect is significant in a smaller diameter column (0.108 m).The heat-transfer coefficients obtained in the small diameter column with a seven-tube bundle are similar to those obtained in a large diameter column.The presence of heat exchanger tubes can help in promoting the better mixing in small diameter columns by limiting the maximum stable size of the bubbles.Schl¨uteret al.[76] presented measurements of heat-transfer coefficients in columns with longitudinal flow and cross-flow tube bundles.The authors observed that the tube pitch has no significant effect on the heat-transfer coefficient in low viscosity and it has only a small effect in the case of highly viscous liquid.There is need for additional work to investigate the role of heat exchanger tubes design of different configurations on heat transfer in different size columns.Abdulmohsin and Al-Dahhan [77] investigated the effect of heat exchanger tubes on the heat transfer coefficient in a 0.19 m diameter bubble column.The operating conditions,the CSA occluded by the tubes,and the configuration of the tubes were in accordance with Youssefet al.[66].Fig.17 shows the effect of internals on the radial profiles of heat transfer coefficient.An increase in the heat–transfer coefficient was obtained with the presence of a high percentage of internals at the same actual gas velocity on the basis of the open CSA.In bubbly flow regime (0.03 m·s-1),as shown in Fig.17(a),a noticeable increase in the heat–transfer coefficient has been observed.Nevertheless,in heterogenous flow regime (0.2 m·s-1),the effect of dense internals (22%) is smaller,as shown in Fig.17(b).This is because the bubble coalescence and breakup rates come to a balance and hence bubble size and their bubble rise velocity reach close to plateau in heterogeneous flow regime.This finding can be attributed to the impact of internals on the bubble dynamics and properties,as discussed and supported by Youssef and Al-Dahhan[66]and Wu[78].In addition,the heat transfer coefficients in the center are larger than those near the wall of the column for all of the studies.

Fig.16.Schematic diagram of bubble column with spring internals:(a)set-up,(b)helical spring holder–isometric view,(c)helical spring internal holder–top view(adapted in [74]).

Table 5Summary of heat-transfer studies conducted in BCR/SBCR with heat exchanger tubes

3.4.Effect of heat exchanger tube on mass transfer

In view of the studies done in BCR/SBCR with heat exchanger tubes,the mass transfer characteristics will most likely be affected by their configurations.However,few studies have reported the effect of heat exchanger tubes on the mass transfer in BCE/SBCR.This is mainly due to the difficulties involved in the measurement of the quantities in the presence of heat exchanger tubes,especially at high superficial gas velocities.The increase of the interfacial area of bubbles [67] in the presence of internals will enhance the overall volumetric mass transfer coefficient,kLa.However,the reduction in the turbulent kinetic energy [70] and fluctuating velocity [65] in the presence of internals may lead to a decrease in the liquid–side mass transfer coefficient,kL.Manjrekaret al.[88] evaluated the impact of vertical cooling tubes on overall gas holdup and volumetric mass transfer coefficient on a 0.45 m diameter pilot scale bubble column and a 0.19 m diameter lab scale bubble column.The configuration of vertical internals is consistent with Youssef and Al-Dahhan[66].Increases in the gas holdup with the insertion of internals were observed,while no significant impact of presence of internals on volumetric mass transfer coefficient was obtained in the churn turbulent flow regime,as shown in Fig.18.The reason is that the decrease in mass transfer coefficient due to the reduction of turbulent intensity in the presence of internals is compensated by the increase of gas–liquid interface area due to increased bubble breakup.However,the effects of internal size and tube configuration have not been addressed at all.Molleret al.[89] systematically investigated different tube arrangements with triangular and square pitch and tube diameter of 0.032 m and 0.045 m at the same CSA in a bubble column of 4.2 m height and 0.392 m diameter.The layout of the tubes and their specifications are summarized in Fig.19 and Table 6.The reactor was operated at bubbly and churn turbulent flow regimes.The authors gave correlations for gas holdup,axial liquid dispersion and the volumetric gas–liquid–solid mass transfer coefficient,which took the internals’ geometry into account.The results revealed that internals reduce the mixing time due to the induction of large-scale liquid circulation and the effect of the internals on the gas–liquid–solid transfer is almost negligible (Fig.20),which is in accordance with Manjrekaret al.[88].Later,they[90]further proposed an advanced recirculation cell model to describes fluid dynamics and mass transfer in bubble columns with and without internals.The model that predicted bubble size distributions,total gas holdup,Sauter mean diameter and volumetric mass transfer coefficients was validated with experimental data of Molleret al.[89].

4.Conclusions

The above-mentioned experimental and numerical studies present a deep sight on the effect of heat exchanger tubes configuration (including geometrical arrangement,pitch and inter-tube gap),diameter of the tubes and the ratio of occluded CSA on bubble dynamics,liquid circulation,heat transfer and mass transfer which are considered among the most important hydrodynamics parameters that govern the performance of BCR/SBCR.The specific alterations in flow pattern,mixing intensities and generally hydrodynamics due to insertion of heat exchanger tubes in BCR/SBCR are presented.Through conducting this review,the level of fundamental understanding of the effect of the design of heat exchanger tubes on the hydrodynamics and heat/mass transfer can be improved.Table 7 provides a summary of the abovementioned studies on heat exchanger tubes.Table 8 provides a summary of the effect of design parameters of the heat exchanger tubes on hydrodynamics and heat/mass transfer in BCR/SBCR.The main findings of the current review on the heat exchanger tubes can be summarized as follows:

Table 6Summary of the internals’ design parameters

Fig.17.Effect of internals on the radial profile heat transfer coefficients:(a) at low superficial gas velocity (Ug=0.03 m·s-1),(b) at high superficial gas velocity(Ug=0.20 m·s-1) (adapted in [77]).

Fig.18.Mass transfer coefficients with and without internals:(a) Dc=19 cm,(b) Dc=45 cm (adapted in [88]).

1.Vertical heat exchanger tubes promote bubble breakup and produce smaller bubbles with a narrow size distribution,which could greatly improve gas holdup and interfacial area and physical reduce the length scales of turbulence at low superficial velocities.

2.For the ratio of occluded CSA by vertical internals:The effect of the internals is significant only when a large fraction of the column’s CSA is equipped with internals.Dense vertical heat exchanger tubes promote formation of large-scale liquid circulation and impede the radial dispersion and the bubble chord length.With increased percentage of column CSA,the overall gas holdup and the heat-transfer coefficient enhances and the bubble chord length decreases.

3.For heat exchanger tubes configuration:Compared to the circular arrangement of vertical heat exchanger tubes,a remarkable increase in the gas holdup values and a better spread of the gas spread over the entire CSA of the column were obtained with the hexagonal configuration.

4.For heat exchanger tubes diameter and pitch:In homogeneous,thin internals gave consistently higher overall and local gas holdup than that of the thick internals or empty column,while the effect of the internals diameter was insignificant in the churn turbulent flow regime.The tube pitch has no significant effect on the heat-transfer coefficient in low viscosity and it only has insignificant effects in the case of highly viscous liquid.

5.For different heat exchanger tubes types:Compared to the vertical heat exchanger tubes,the helical coils provided higher gas holdup than vertical internals.The difference was attributed to larger inter-tube gaps for the vertical internals that provided more space for larger bubbles to escape,thus decreasing the gas holdup,unlike the helical coil in which only smaller gaps were present.Further,the smaller gaps would provide more obstruction to the gas flow,resulting in longer residence time for the gas phase.

Fig.19.The layout of internals (adapted in [89]).

6.No significant impact of presence of heat exchanger tubes,tube size and tube configuration on volumetric mass transfer coefficient was obtained in the churn turbulent flow regime for the reason that the decrease in mass transfer coefficient due to the reduction of turbulent intensity in the presence of internals is compensated by the increase of gas–liquid interface area due to increased bubble breakup.

5.Recommendations

Although these studies provide useful information about BCR/SBCR with the insertion of heat exchanger tubes,many questions remain unanswered in topics of relevance to this review.Here are few recommendations for potential research to better comprehend the effect of gas distributor and heat exchanger tubes on hydrodynamics.

1.Most of the studies on the effect of the internals configuration and size on hydrodynamics are limited to air water system under ambient temperature and pressure while the reactions in the SBCR are always involved a gas–liquid–solid system and performed at conditions with high temperature and pressure.Hence,under high temperature and pressure,it is necessary to study the influence of the internals configuration and size on the gas–liquid–solid system to mimic standard industrial conditions.

Fig.20.Volumetric mass transfer coefficient and liquid–side mass transfer coefficient:(a) 32 mm tubes,(b) 45 mm tubes (adapted in [89]).

2.The effect of the gap between the gas sparger and the heat exchanger tubes on hydrodynamics in BCR/SBCR has not been reported in the literature.Therefore,it is necessary to investigate the influence of internals height above the gas distributor on the hydrodynamics in bubble columns.

3.Experiments and simulations studies on flow regime identification of BCR/SBCR with different internal structure,internal size and arrangements should also be carried out.Moreover,it is necessary to study the effect of heat exchanger tubes on other related hydrodynamic parameters such as phase residence time,liquid velocity and eddy diffusivity.

4.Understanding of the effect of heat exchanger tubes at wide range of superficial gas velocity on heat transfer of BCR/SBCR is still lacking in the open literature.Hence,it is necessary to investigate the influence of internals with different structure,size and arrangements on the heat transfer in these reactors.

5.The results of the above literature show that dense vertical heat exchanger tubes promote formation of large-scale liquid circulation and impede the radial dispersion,while the influence of the sparse internals on the gas holdup and the large–scale liquid circulation were insignificant.Moreover,the gas holdup and interfacial area increases for the case of dense internals.In addition,the heat exchanger tubes with spiral structure provide higher gas holdup than vertical tubes,for the helical structure breakups the gas into fine bubbles,which effectively reduces their rise velocity and enhances their residence time in the columns.Hence,the gas holdup and liquid circulation flow in the BCR/SBCR can be enhanced by increasing the CSA occluded by the vertical heat exchanger tubes or by using heat exchanger tubes with spiral structure.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We gratefully acknowledge the support of the National Key Research and Development Program of China (2018YFB0604603-03),National Natural Science Foundation of China (21706175,201703151 and 21776195),Key Research and Development Program of Shanxi Province (201803D121043).

Nomenclature

Dcdiameter of bubble column

hheat transfer coefficient,kW·(m2·K)–1

kLthe liquid-side mass transfer coefficient

kLathe overall volumetric mass transfer coefficient

Ppressure,MPa

Rradius of column,cm

rradial position,cm

Ttemperature,°C

Ugsuperficial gas velocity,m·s-1