Improvement of baffle type Rotating Packed Bed’s packing by visual study

Chemical and P ocess Enginee ing: New F on ie s, 2023, 44(3), e28 DOI: 10.24425/cpe.2023.146730 POSTER COMMUNICATION Imp o emen o ba le ype Ro a ing Packed Bed’s packing by isual s udy Dawid Zawadzki1, Małgo za a Majdzik1, Ondřej Hájek2, Milan Malý2, Michał Bla kiewicz1∗ 1Lodz Uni e si y o Technology, Facul y o P ocess and En i onmen al Enginee ing, Wolczanska 213, 93-005 Lodz, Poland 2B no Uni e si y o Technology, Facul y o Mechanical Enginee ing, Technicka 2, 616-69 B no, Czech Republic ∗Co esponding au ho : e-mail: [email p o ec ed]dz.pl P esen ed a 8 h Eu opean P ocess In ensi ica ion Con e ence, 31.05–2.06.2023, Wa saw, Poland. A icle in o: Recei ed: 27 July 2023 Re ised: 3 Augus 2023 Accep ed: 5 Sep embe 2023 Abs ac P ocess in ensi ica ion is one o he key b anches o p ocess enginee ing. High g a i y equipmen achie es in ensi ica ion by subs i u ing g a i y wi h much highe cen i ugal o ce. Ro a ing Packed Bed is he leading example o high g a i y solu ions, s ongly acili a ing gas-liquid mass ans e . Howe e , cylind ical packings come wi h ce ain d awbacks, such as d y spo s, ha can be o e come wi h new solu ions, such as ba le-based packing geome ies. Howe e , when ba les a e a anged oo close o each o he , liquid b idges a e o med be ween hem, which may lead o dec ease in mass ans e e iciency. This wo k is conce ned wi h imp o emen o a Zickzack-like in e nal by he means o isual s udies wi h he use o high-speed came a. Acco ding o measu ed ligamen b eak-up leng h, wo new packings we e designed o pa icula o a ional speeds and es ed expe imen ally o e ec i e mass ans e a ea and we p essu e d op. Keywo ds hyd odynamics, isual s udy, Ro a ing Packed Bed, packing, mass ans e a ea 1. INTRODUCTION Ro a ing Packed Bed (RPB) is a no el appa a us o con ac - ing wo phases o di e en densi ies, e.g. liquid and gas o liquid and apo , whe e he g a i a ional o ce is subs i u ed wi h much highe cen i ugal o ce (Ramshaw and Mallinson, 1981). The main pa o an RPB uni is a o a ing packing, esponsible o de eloping high mass ans e a ea and p o id- ing in ense mic o-mixing o he phases. A schema ic ep esen- a ion o a coun e -cu en RPB uni is shown in Figu e 1. The mos commonly used packing ypes o RPBs include wi e mesh (Jassim e al., 2007) o me al oam (Zheng e al., 2016). Howe e , such po ous packings o en pose d awbacks, such as high p essu e d op and limi ed ope a ional window, as well as d y zones due o adial di ec ion o cen i ugal accele a ion (G oß e al., 2019). These d awbacks can be o e come wi h he use o s uc u al packings. The i s example o a s uc u al RPB packing was p esen ed by Qamma e al. (2019). The Zickzack in e nal, inspi ed by he o a ing zigzag bed appa a us (Wang e al., 2008) is made o wo solid pla es wi h al e na ing, concen ically a anged ba les, each equipped wi h an addi ional wei o p o ide highe liquid holdup. Recen ly, Zawadzki e al. (2023) used compu a ional luid dynamics o op imiza ion o he ba le shape o RPB s uc u al in e nals in o de o minimize he p essu e d op. The op imiza ion allowed o dec ease he p essu e d op by 50%, while inc easing he e ec i e mass ans e a ea by up o 30% in compa ison wi h he o iginal Zickzack. Al hough he shape and inclina ion angle o he ba les ha e been success ully op imized, he packing s ill lea es oom o op imiza ion. In pa icula , ba le spacing inside he pack- ing is also key o i s pe o mance, as oo dense a angemen may c ea e e y high low esis ance, while oo spaced ba - les may dec ease he mass ans e e iciency due o limi ed mic o-mixing. In his wo k, isual s udies ha e been used o op imize ba le spacing, based on ligamen b eakup dis ance a a ying liquid low a es, ba le adii and o a ional speeds. 2. MATERIALS AND METHODS 2.1. Visual s udy o hyd odynamics The measu emen o ligamen b eak-up was isualised using high-speed came a in a anspa en RPB a B no Uni e si y o Technology. Despi e he ac ha he o iginal packing (see Figu e 5a) was manu ac u ed wi h he use o a anspa en esin, he une enness o he p in ed su aces did no allow o make quan i a i e measu emen s o he liquid ligamen s. I was he e o e necessa y o simpli y he packing. Hence, ou o o s o a ious diame e s we e manu ac u ed wi h only one ba le placed on he ou e diame e . A p essu e swi l a omize discha ged a con inuous liquid shee o ensu e uni o m liquid dis ibu ion on he packing’s in e nal pa . A high-speed came a FASTCAM SA-Z (Pho on, Japan), equipped wi h a long-dis ance mic oscope 12X Zoom lens (NAVITAR, New Yo k, USA), was used o backligh eco d- ings. The lens was composed o a 2X F-moun adap e ( ype 1-62922), a 12 mm F.F zoom lens ( ype 1-50486) oge he ©2023. The Au ho (s). This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion (CC-BY 4.0, h ps://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. 1o 4 D. Zawadzki, M. Majdzik, O. Hájek, M. Malý, M. Bla kiewicz Chem. P ocess Eng., 2023, 44(3), e28 wi h 0.25X lens ( ype 1-50011). A con inuous LED ligh model HPL3-36DD18B (Ligh speed Technologies, USA) il- lumina ed he eco ded a ea o 1024 ×1024 px wi h a spa ial esolu ion o 18 µm/px. A shu e speed o all eco dings was equal o 3.75 µs. The ob ained eco dings showed o ma ion o ligamen s on he ba les, as well as hei b eakup in o d ople s, as shown in Figu e 3. This igu e also p esen s he me hodology o lig- amen leng h analysis. Ligamen leng h was de ined as he maximum leng h he ligamen can sp ead be o e i b eaks up in o d ople s. The magni ude o ligamen leng h was de- e mined using dimensions in he xand ydi ec ions. Ap- p oxima ely, 20–30 ligamen s we e measu ed o each expe - imen al poin o de e mine he a e age alue and s anda d de ia ion. 2.2. Measu emen o e ec i e mass ans e a ea and we p essu e d op The measu emen s o we p essu e d op and e ec i e mass ans e a ea we e pe o med on he RPB de ice owned by Lodz Uni e si y o Technology. The expe imen al ig was equipped wi h a liquid pump, an ai blowe , ai humidi ie , a ba e y o comp essed CO2bo les, h ee CO2in a ed mea- su emen de ices (sampling om inle , ou le , and inside he RPB casing), a hyd aulic lock, and a di e en ial p essu e gauge o measu ing p essu e d op be ween he inle and he ou le o he RPB. The eac ion sys em consis ed o 0.2 mol.% CO2in ai , and 1M NaOH solu ion. This sys em was chosen because i ul ils he equi emen s o he me hodology o di ec mass ans- e a ea calcula ion posed by Cha pen ie (1981). Wi h his me hod, he ae can be de e mined wi h Equa ion (1). ae =Φ “k ·DCO2·C∗ CO2|a g 2·CNaOH”0:5(1) The expe imen al de e mina ion o he e ec i e mass ans e a ea and we p essu e d op was conduc ed unde he gas low o 20 m3/h, wi h h ee liquid low a es o 45, 90, 180 L/h, and a ou di e en o a ional speeds o 300, 600, 900 and 1200 pm. 3. RESULTS AND DISCUSSION 3.1. Visual s udy The pu pose o he isual s udy was o de e mine liquid liga- men b eakup leng h a a ying liquid low a es, ba le adii, and o a ional speeds. A p e iously p epa ed anspa en ba - le packing (Figu e 5a) was i s in es iga ed wi h he as came a, and liquid b idges could be obse ed be ween he ba les (see Figu e 2). Howe e , he assembled packing was oo opaque o make quan i a i e measu emen s. Thus, single ba le packings we e ins alled in he RPB and he liquid s uc- u es we e obse ed a hei ou e im. An exempla y pho o o ligamen s o ming a he ba le is p esen ed in Figu e 3. I was obse ed ha liquid low a e did no in luence liga- men geome y, as i s inc ease only led o inc eased amoun o o med ligamen s. Howe e , he o a ional speed and ba - le adius had a signi ican e ec on ligamen leng h, as i u ned ou o be a unc ion o he angula eloci y o he liq- uid a he ba le. The i ing o he unc ion and i s equa ion a e p esen ed in Figu e 4. 3.2. In es iga ion o ailo ed packings Wi h he knowledge om he esul s desc ibed in Sec ion 3.1, wo packings we e made speci ically o he desi ed o a ional speeds: 600 pm (“600”, Figu e 5b) and 1200 pm (“1200”, Figu e 5c). The dis ances be ween he ba les we e dic a ed by he ligamen b eakup leng h modeled by he unc ion shown in Figu e 4, hus p e en ing o ma ion o liquid b idges be ween subsequen ba les a he desi ed o a ional speeds. As lowe o a ional speeds co espond o longe ligamen s, he ailo ed packing o 600 pm is cha ac e ized by la ge dis ances be ween ba les. In all cases, he dis ances dec ease quad a ically wi h he adius. Thus, he packings ailo ed o 600 pm and 1200 pm ha e di e en numbe s o ba les. The esul s o he mass ans e measu emen s a e p esen ed in Figu e 6a. I can be seen ha he ailo ed packings do no show s ong inc eases in e iciency a hei desi ed o a ional speeds. In all cases, including he ini ial packing, he mass ans e a ea inc eases s eadily wi h he o a ional speed, ex- cep o he cases a 300 pm, whe e he Ini ial and 1200 packings ope a e unde condi ions close o looding, which causes sudden spikes in e ec i e mass ans e a ea. I can be also seen ha wi hin he 600–1200 pm ange, he Ini ial packing and he 1200 packing beha e e y simila ly, while he e ec i e mass ans e a eas gene a ed by packing 600 a e signi ican ly lowe . The inc easing numbe o ba les o ces he phases o low h ough na owe channels, hus inc easing he low esis- ances. In Figu e 6b we p essu e d ops o he h ee packings a e shown as unc ions o o a ional speed. Wi h he excep- ion o 1200 packing ope a ing a 300 pm ( looding), e y s eady ends can be seen o all h ee packings. Bo h o a- ional speed and numbe o ba les ha e posi i e e ec on he we p essu e d op. 4. CONCLUSIONS Wi h he assump ion ha he design o he packings p e- en s liquid b idge o ma ion a pa icula o a ional speeds, and gi en he da a shown in Figu e 6, i can be concluded ha liquid b idges a e no he limi ing ac o o he mass ans e a ea in he ba le- ype s uc u al packing. I is he e- o e possible ha liquid ilms o ming a he solid elemen s o 2o 4h ps://jou nals.pan.pl/cpe Chem. P ocess Eng., 2023, 44(3), e28 Imp o emen o ba le ype Ro a ing Packed Bed’s packing by isual s udy he packing con ibu e o he mass ans e e iciency mo e signi ican ly han he ee liquid s uc u es be ween he ba - les. F om he expe imen al da a i can be seen ha inc ease in he numbe o ba les leads o inc ease in mass ans e a ea as well as p essu e d op, and igh ly a anged ba les in- c ease he isk o looding a low o a ional speeds. In u u e wo k, gas low may also be aken in o accoun o isual s ud- ies, as i may also in luence he o ma ion o liquid s uc u es wi hin he packing. ACKNOWLEDGMENTS This wo k was suppo ed by he join Polish-Czech p ojec : no. UMO-2020/39/I/ST8/01381 inanced by he Na ional Science Cen e, Poland, and no. GA 21-45227L inanced by he Czech Science Founda ion. SYMBOLS ae e ec i e mass ans e a ea, m2/m3 Ddi usi i y, m2/s Cmola concen a ion, kmol/m3 k eac ion a e cons an , 1/s G eek le e s Φabso p ion a e, kmol/s Supe sc ip s ∗gas-liquid equilib ium Subsc ip s a g a e age be ween casing and ou le CO2ca bon dioxide NaOH sodium hyd oxide REFERENCES Cha pen ie J.-C., 1981. Mass- ans e a es in gas-liquid ab- so be s and eac o s. In Ad . Chem. Eng., 11, 1–133. DOI: 10.1016/S0065-2377(08)60025-3. G oß K., Biebe le A., Gladyszewski K., Schube M., Ham- pel U., Skibo owski M., Gó ak A. 2019. Analysis o low pa - e ns in high-g a i y equipmen using gamma- ay compu ed omog aphy. Chem. Ing. Tech., 9, 1032–1040. DOI: 10.1002/ ci e.201800085. Jassim M.S., Rochelle G., Eime D., Ramshaw C., 2007. Ca bon dioxide abso p ion and deso p ion in aqueous mo- noe hanolamine solu ions in a o a ing packed bed. Ind. Eng. Chem. Res., 46, 2823–2833. DOI: 10.1021/ie051104 . Qamma H., Gładyszewski K., Gó ak A., Skibo owski M., 2019. Towa ds he de elopmen o ad anced packing design o dis- illa ion in o a ing packed beds. Chem. Ing. Tech., 91, 1663– 1673. DOI: 10.1002/ci e.201900053. Ramshaw C., Mallinson R.H., 1981. Mass ans e p ocess. US Pan en No. US4283255A. Wang G.Q., Xu Z.C., Yu Y.L., Ji J.B., 2008. Pe o mance o a o- a ing zigzag bed – A new HIGEE. Chem. Eng. P ocess. P ocess In ensi ., 47, 2131–2139. DOI: 10.1016/j.cep.2007.11.001. Zawadzki D., Bla kiewicz M., Jaskulski M., Pią kowski M., Koop J., Loll R., Gó ak A., 2023. Design and op imisa ion o s uc- u al packings o o a ing bed abso be s using compu a ional luid dynamics simula ion. Chem. Eng. Res. Des., 195, 508– 525. DOI: 10.1016/j.che d.2023.06.002. Zheng X.H., Chu G.W., Kong D.J., Luo Y., Zhang J.P., Zou H.K., Zhang L.L., Chen J.F., 2016. Mass ans e in ensi ica- ion in a o a ing packed bed wi h su ace-modi ied nickel oam packing. Chem. Eng. J., 285, 236–242. DOI: 10.1016/j.cej. 2015.09.083. h ps://jou nals.pan.pl/cpe 3o 4 D. Zawadzki, M. Majdzik, O. Hájek, M. Malý, M. Bla kiewicz Chem. P ocess Eng., 2023, 44(3), e28 Chem. P ocess Eng., 2023, 44(3), e28 4o 4h ps://jou nals.pan.pl/cpe