Mechanisms of de-icing by surface rayleigh and plate lamb acoustic waves

Mechanisms o De-icing by Su ace Rayleigh and Pla e Lamb Acous ic Wa es Shilpi Pandey,* Jaime del Mo al,* S e an Jacob, Lau a Mon es, Jo ge Gil-Ros a, Alejand o F echilla, A e eh Ka imzadeh, Vic o J. Rico, Raul Kan e , Niklas Kandelin, Ca men López-San os, Heli Koi uluo o, Luis Angu el, And eas Winkle ,* Ana Bo ás, and Agus in R. González-Elipe 1. In oduc ion Ice acc e ion on ma e ials ope a ing ou doo s o in sub-ze o en i onmen s is a c i ical issue ha significan ly impac s e ficiency, main e- nance, and secu i y in a ious indus ies, including a ia ion, ene gy gene a ion, and he unc ionali y o senso and came a windows and sc eens. To add ess his p ob- lem, wo p ima y app oaches ha e been p oposed: 1) Passi e an i-icing based on applying su ace enginee ing p ocesses o educe o delay ice acc e ion; [1–5] e y o en, hese app oaches ely on supe hyd ophobic o liquid-in used su aces. [6–9] 2) Ac i e de-icing/an i-icing sys ems applying di e - en s a egies o ice emo al o p e en ing i s o ma ion; ecen ad ances in he field include nano echnology-based mel ing by Joule hea ing, [10] using ma e ials such as g aphene, [11–13] o pho o he mal de-icing using plasmonic o magne ic nanopa - icles. [14–18] An eme gen de-icing app oach in ol es he ac i a ion o su aces by high- equency acous ic wa es (AW) wi h Acous ic wa es (AW) ha e ecen ly eme ged as an ene gy-e ficien ice- emo al p ocedu e compa ible wi h unc ional and indus ial- ele an subs a es. Howe e , c i ical aspec s a undamen al and expe imen al le els ha e ye o be disclosed o op imize hei ope a ional condi ions. Iden i ying he p ocesses and mechanisms by which di e en ypes o AWs induce de-icing a e some o hese issues. He ein, using model LiNbO 3 sys ems and wo ypes o in e digi a ed ansduce s, he e-icing and an i-icing e ficiencies and mechanisms d i en by Rayleigh su ace acous ic wa es (R-SAW) and Lamb wa es wi h 120 and 510 μm wa eleng hs, espec i ely, a e analyzed. Th ough he expe imen al analysis o de-icing and ac i e an i-icing p ocesses and he fini e elemen simula ion o he AW gene - a ion, p opaga ion, and in e ac ion wi h small ice agg ega es, i is disclosed ha Lamb wa es a e mo e a o able han R-SAWs o induce de-icing and/o p e en he eezing o small ice d ople s. P ospec s o applica ions o his s udy a e suppo ed by p oo o concep expe imen s, including de-icing in an icing wind unnel, demons a ing ha Lamb wa es can e ficien ly emo e ice laye s co e ing la ge LN subs a es. Resul s indica e ha he de-icing mechanism may di e o Lamb wa es o R-SAWs and ha he wa eleng h mus be conside ed as an impo an pa ame e o con olling he e ficiency. S. Pandey, [+] S. Jacob, A. Ka imzadeh, A. Winkle Leibniz IFW D esden, SAWLab Saxony Ins i u e o Eme ging Elec onic Technologies (IET), G oup “Acous ic Mic osys ems” Helmhol zs . 20, 01069 D esden, Ge many E-mail: shilpi.pandey@ um.de; [email p o ec ed] The ORCID iden ifica ion numbe (s) o he au ho (s) o his a icle can be ound unde h ps://doi.o g/10.1002/adem.202401820. [+] P esen add ess: Ma e ials Science and En i onmen al Enginee ing, Facul y o Enginee ing and Na u al Sciences, Tampe e Uni e si y, 589, 33104 Tampe e, Finland © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH. This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modifica ions o adap a ions a e made. [Co ec ion added on 17 Decembe 2024, a e fi s online publica ion: The 9 h au ho name has been co ec ed in his e sion.] DOI: 10.1002/adem.202401820 S. Pandey, Heinz-Nixdo -Chai o Biomedical Elec onics School o Compu a ion, In o ma ion and Technology Technical Uni e si y o Munich T anslaTUM, 80333 Munich, Ge many J. del Mo al, L. Mon es, J. Gil-Ros a, V. J. Rico, C. López-San os, A. Bo ás, A. R. González-Elipe Nano echnology on Su aces and Plasma Lab Ma e ials Science Ins i u e o Se ille Consejo Supe io de In es igaciones Cien íficas (CSIC) Ame ico Vespucio 49, 41092 Se illa, Spain E-mail: [email p o ec ed]ic.es S. Jacob Ge man Na ional Me ology Ins i u e (PTB) Bundesallee 100, 38106 B aunschweig, Ge many A. F echilla, L. Angu el Ins i u o de Nanociencia y Ma e iales de A agón (INMA) CSIC-Uni e sidad de Za agoza 50018 Za agoza, Spain RESEARCH ARTICLE www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (1 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH nano-scale ampli udes, a p ocedu e compa ible wi h indus ial- el- e an su aces. In his con ex , a key dis inc ion should be made be ween he localized applica ion o su aces o comme cially a ailable ul asound gene a o s [19,20] and he inno a i e con- cep s based on he inco po a ion o piezoelec ic ac i e suppo s o laye s, as well as elec odes (la e al field exci a ion (LFE) o in e digi a ed ansduce s (IDTs)), as in eg al pa s o he su - ace o be ac i a ed. [21–26] The p esen wo k fi s wi hin his sec- ond app oach, dealing wi h he in eg a ion o su ace o bulk acous ic wa es (SAWs and BAWs) o de-icing and ice sensing. Among he a icles add essing his opic, he wo k o Yang e al. (2021), which p oposes a s a egy o weaken ice adhesion using Rayleigh su ace acous ic wa es (R-SAW), [21] is pa icula ly no ewo hy. Fo his pu pose, IDTs wi h di e en wa eleng hs (100, 200, 300, and 400 μm) we e manu ac u ed on a 5 μm hick ZnO piezoelec ic hin film deposi ed on an aluminum sub- s a e, p oducing SAWs wi h equencies comp ised be ween 7.38 and 27.84 MHz. The au ho s s a e ha SAWs p oduce ib a ions compa able o “nano-scale ea hquakes”,which esul in he gene a ion o acous ic-hea ing e ec s, he de elop- men o mic o-c acks, and he dis u bance o ice nuclea ion h ough local ib a ions and ene gy s eaming in o he o med liquid phase. The con ibu ion o Joule hea ing e ec s di ec ly s emming om ohmic losses in he IDTs canno be disca ded in hese expe imen s since ice d ople s we e placed di ec ly a op he IDTs. The same au ho s ha e ecen ly claimed he possibil- i y o moni o ing ice o ma ion using his ype o SAWs. [22] Simila ly, Zeng e al. [23] demons a ed a dec ease in adhesion be ween ice and subs a e due o in e ace hea ing and a educ ion in elec os a ic o ces and mechanical in e lock- ing on SAW-ac i a ed subs a es. Del Mo al e al. [24] and Jacob e al. [25] ha e ecen ly ex ended he applica ion o acous ic wa e de-icing o anspa en subs a es and suppo s. In he fi s wo k, he au ho s demons a ed ha ex ended elec- odes on piezoelec ic pla es in an LFE configu a ion gene a e hickness shea mode - bulk acous ic wa es (TSM - BAWs) in he 3–4 MHz ange. These wa es e ficien ly induced de-icing and a educ ion o ice acc e ion (i.e., ac i e-an i-icing e ec ), as well as a dec ease in ice adhesion. A d awback o his app oach is he necessi y o a as -elec onic uning o he exci a ion due o he e y na ow bandwid h o BAW modes. The possibili y o using he same de ice o ice de ec ion unde ealis ic ope a ion condi ions (i.e., expe imen s in icing wind unnels (IWT)) and a p ofi able syne gy o educe powe con- sump ion upon he applica ion o an i-icing coa ings we e also demons a ed in Del Mo al e al. wo k. [24] On he con a y, Jacob e al. [25] demons a ed ha R-SAW echnology can be applied o la ge a eas and anspa en subs a es beyond he cen ime e scale, p o ing ha R-SAWs can be used o de-icing indus ially ele an su aces. In ha wo k, he au ho s also demons a ed ha R-SAWs gene a ed on piezoelec ic pla es and piezoelec ic hin films induced he de-icing o glaze ice h ough a pu e acous ic mechanism, disca ding he mel ing h ough di ec elec o he mal (Joule) hea ing induced by ohmic losses in he IDTs on subs a es wi h low he mal conduc i i y. Recen ly, he same au ho s demons a ed ha a hyb id ope a- ional mode is possible by combining R-SAW o de-icing and BAW o sensing by ca e ully con olling he exci a ion mode o he wo finge combs in eg a ed in o he IDTs. [26] O he ele an s udies on he implemen a ion o AW ice senso s and de-icing o ime ice ha e also been ecen ly published. [27–30] Despi e he significance o hese con ibu ions, de-icing wi h subs a e-in eg a ed AW sys ems aces subs an ial undamen al unce ain ies ha s ill hinde he implemen a ion o his echnol- ogy in eal-wo ld applica ions. C i ical bo lenecks encompass he assessmen o ac o s a ec ing he ansmission o AW ene gy o ice o wa e p esen on he de ice su ace, unde s anding he mechanisms behind ice-c acking and o he p ocesses con ibu - ing o ice emo al wi hou comple e mel ing, and defining ac i a ion condi ions o a oid ice o ma ion unde subze o em- pe a u es, he eby enabling an e ficien an i-icing mode. Addi ional ea u es p e en ing an unequi ocal compa ison o e ficiencies and de-icing mechanisms a e he di e ences in he ype o AWs, ei he SAWs, Lamb wa es, o TSM-BAWs, [21–26] he e ec o he subs a e ma e ials (e.g., anspa en and black LN, ZnO hin films deposi ed on di e en subs a es such as Al oils, glass and used silica, he hyd ophobic o hyd ophilic su ace modifica ion o hese subs a es, e c.), o he ple ho a o condi- ions used in he expe imen s ( ype o ice, liquid wa e con en (LWC) when wo king in IWTs, s a ic and windy condi ions, empe a u es and ela i e humidi y, wa e eezing o ice acc e- ion, e c.). To shed some ligh on his wiligh scena io, we he ein sys- ema ically in es iga e he de-icing and ac i e an i-icing ac i a ion o he same subs a e wi h ei he R- SAW o Lamb wa es. Fo his pu pose, we use LiNbO 3 (LN) piezoelec ic chips ac i a ed wi h IDTs ope a ing a di e en d i ing equencies. The selec ion o LN o p epa ing specific model sys ems o s udy de-icing wi h AWs exceeds his opic, as his piezoelec ic ma e ial is widely u ilized in o he applica ions such as high- empe a u e senso s o op ical wa eguide ape ed an ennas. [31–33] He ein, we ha e elec o-acous ically cha ac e ized a se ies o IDT configu a ions o op imize hei pe o mance, ei he o ac i a ing R-SAWs o Lamb wa es. Then, o hese wo wa e modes and a ious sub- ze o empe a u es, we pe o med de-icing and ac i e an i-icing expe imen s o small ice agg ega es and wa e d ople s, moni- o ing he h eshold powe equi ed in each case. Addi ionally, an analysis o he ope a ional condi ions o he chips using fini e elemen model (FEM) simula ions in COMSOL Mul iphysics is used o model he gene a ion o he R-SAW and Lamb wa e modes and accoun o hei in e ac ion wi h ice. Finally, o alida e hese expe imen al and simula ion esul s o Lamb wa es, we ha e ca ied ou some p oo -o - concep de-icing expe imen s p o ing he sui abili y o he Lamb wa e ac i a ion o emo e la ge a eas o acc e ed ice. The o e all assessmen o he de-icing o small ice agg ega es and la ge ice laye s wi h Lamb wa es has pe mi ed he p oposal o a specific mechanism o de-icing using Lamb wa es ha di e s om ha al eady known o R-SAWs. [25,30] We a e con- fiden ha all hese findings will significan ly con ibu e o es ablishing a a ional pa hway o exploi ing AWs as he basis o a nex -gene a ion de-icing sys em. R. Kan e , N. Kandelin, H. Koi uluo o Ma e ials Science and En i onmen al Enginee ing Facul y o Enginee ing and Na u al Sciences Tampe e Uni e si y 589, 33104 Tampe e, Finland www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (2 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License 2. Expe imen al Sec ion 2.1. Fab ica ion and Elec oacous ic Cha ac e iza ion o AW Chip De ices The selec ed piezoelec ic ma e ial was a 0.5 mm hick black Li hium Nioba e 128° YX wa e cu pu chased om CSIMC- F eqcon ol, China. The 4 inch wa e s we e cleaned wi h ace one and IPA using ul asonic cleaning o 5 min and O 2 plasma ea - men a 100 W o 10 min. IDTs we e pa e ned using pho o esis spin-coa ing, a maskless lase w i e (Heidelbe g ins umen s MLA 100), and he li -o echnique. Ti (5 nm)/Al (295 nm) was selec ed o IDT me alliza ion. Fu he , 100 nm SiO 2 , [34] was deposi ed by magne on spu e ing on op o he wa e . This laye ac s as a passi a ion laye and p o ec s he aluminum IDTs om co osion. This SiO 2 hin film con e s a hyd ophilic cha ac e o he su ace o p epa ed chips (we ing con ac angle ≈35°). The con ac pads we e opened o elec ical con ac by a d y e ching p ocedu e. Figu e S1 in he Suppo ing In o ma ion sec ion (SI) shows he u ilized mask design and a manu ac u ed wa e aimed a ab ica ing IDTs wi h wa eleng hs a ying om 120 o 510 μm and an ape u e o 9.95 mm, including a ious designs wi h a di e en numbe o finge pai s. Chips ope a ing a di e en wa eleng hs we e diced in o indi idual chips, as shown in Figu e S2, Suppo ing In o ma ion. Keeping he same ape u es and numbe o finge pai s, di e en IDTs layou s and sizes o diced LN pla es we e used o he chips u ilized o p oo -o - concep expe imen s, as explained in he co esponding sec ion. Chip sizes o 15 20, 15 25, and 15 30 mm 2 ha e been p e- pa ed on he wa e , depending on wa eleng h. The de ices wi h 120 and 510 μm wa eleng h (hence o h called 120 chip and 510 chip, espec i ely), used o he de-icing expe imen s o small ice agg ega es as done in his wo k, we e p epa ed on 15 20 and 15 30 mm 2 LN pieces, espec i ely. A 420 μm chip was also p epa ed o a compa a i e de-icing es wi h big ice agg ega es. This ensu ed a minimum a ea o 15 10 mm 2 wi hou IDTs o de-icing expe imen s. Fo he elec oacous ic cha ac e iza ion o he chips as a unc- ion o finge pai s, on-wa e elec ical measu emen s we e ca - ied ou using a ec o ne wo k analyze (VNA, Keysigh E5080B) and a wa e p obe (Figu e S3, Suppo ing In o ma ion). Be o e RF measu emen , a iscous pho o esis was applied be ween he IDTs o a enua e he wa es and neglec eflec ions on neighbo - ing IDTs, hus a oiding undesi ed e ec s by measu ing he eflec ion coe ficien (S 11 ) cu es. 2.2. De-icing and Ac i e An i-icing Tes s wi h Small Ice Agg ega es De-icing and an i-icing expe imen s we e ca ied ou in a cus om- made icing chambe , whe e empe a u es down o a minimum alue o 20 °C can be con olled (see Figu e S4, Suppo ing In o ma ion, showcasing a iew o his chambe , he chip holde , and he wa e dosing sys em). An ad-hoc wa e dosing sys em consis ing o hea able ubing inse ed ia he wall o he cooling chambe was employed o enable he dosing o d ople s o bi-dis illed wa e on he su ace o he chips o p oduce icepa icles.Inside hechambe , he ubewaskep a a empe a u e o 2 °C o p e en wa e eezing inside he ube. The empe a u e o he subs a e was con olled independen ly om ha o he cooling chambe by a Pel ie pla e on which he holde and p in ed ci cui boa d (PCB) we e placed. Se e al he mocouples we e moun ed on he s age and he d ople hose o p ecisely con ol he empe a u e o he chip and wa e d op- le dispense . A USB came a wi h an a ached C-moun objec- i e was placed ou o he cooling chambe o ollow he e olu ion o ice agg ega es and d ople s. De-icing and ac i e an i-icing expe imen s we e conduc ed on chips wi h IDTs o 120 and 510 μm in his cus om-buil cooling chambe . The icing es s we e pe o med wi h he bes -sui ed configu a ion o a ailable chips (acco ding o he numbe o finge pai s) as de e mined by hei elec o-acous ical cha ac e iza ion. Expe imen s we e ca ied ou a h ee empe a u es (5, 10, and 15 °C, as con olled on he Pel ie pla e) wi h he s age in a ho izon al posi ion. A hese empe a u es, he po osi y and na u e o ice o med upon acc e ion p ocesses om he impac o supe cooled wa e d ople s change om glaze (a 5 °C) o mix (a 10 °C) o ime (a 15 °C). [35] Howe e , unde he expe - imen al condi ions o he ein- epo ed expe imen s, he ice did no o m by acc e ion bu by cooling sessile wa e d ople s. These condi ions a e p omp o gene a e li le po osi y. The s age empe a u e was con inuously moni o ed du ing he expe i- men s using a he mocouple moun ed on an aluminum holde in di ec con ac wi h he Pel ie pla e. Once he empe a u e was s abilized, a 22 μL milliQ wa e d ople a þ2 °C was dispensed on he su ace o he chips a a dis ance o ≈3 mm om he IDTs. A e a while, he dispensed wa e d ople oze in o a small agg ega e o ice, and hen he chip was elec ically exci ed un il he ice was comple ely mel ed. The minimum ac i a ion powe (we deno e his de-icing powe as DP T , a magni ude ha depends on empe a u e T) has been aken as a ele an pa ame e o compa ing he de-icing e ficiency o he 120 and 510 μm chips. Fo he ac i e an i-icing expe imen s, he same olume o wa e was d ipped on an elec o-acous ically ac i a ed chip, and he minimum powe equi ed o p e en eezing (an i-icing powe , AP T ) was aken as he ele an pa ame e o compa ing e ficien- cies. In hese expe imen s, he su ace was pe manen ly ac i a ed o a oid ice nuclea ion and he subsequen eezing o he depos- i ed wa e d ople . The RF AW ac i a ion o he chips inside his cooling chambe was done by con ac ing he diced chips ia a 50 Ωimpedance PCB p o ided wi h gold-coa ed sp ing-pin connec o s as shown in Figu e S2, Suppo ing In o ma ion oge he wi h he chip holde adap ed o conduc he de-icing expe imen . An RF sup- ply, BSG F20 (Belek oniG GmbH, Ge many), has been used o elec oacous ic exci a ion, and he powe supplied by he ins u- men has been aken as a e e ence pa ame e o he ac ual applied powe . 2.3. P oo o Concep : Expe imen s o De-icing La ge Ice Laye s wi h Lamb Wa es Two p oo -o -concep expe imen s using Lamb wa es ha e been ca ied ou o de e mine he de-icing mechanisms o his ype o wa e applied o la ge-a ea ice laye s o agg ega es. www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (3 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License 2.3.1. Expe imen 1 In he 510 μm chip used o his expe imen , he IDT and con- nec ion pads we e s uc u ed on he LN ace ha was no exposed o ice acc e ion. This means he 510 μm IDTs and PCB boa d we e on he opposi e side o he ice- o ma ion su ace. The con- nec ing boa d was adequa ely p o ec ed o p e en sho age o mal unc ion wi h he IDT layou in he cen e o he LN pla e and he connec ing pads a each side o he chip (Figu e S5, Suppo ing In o ma ion shows he chip, PCB, and p o ec ing boa d). In his configu a ion, he connec ing pad a he op side o he chip was he elec ically pola ized one, while ha a he bo om was g ounded. The icing a ea was equi alen o he 15 30 mm 2 chip size. Ice acc e ion was done in he IWT acili y a ailable a he TAU labo a o y (de ailed in o ma ion abou hese Ice Labo a o y acili ies can be ound in Koi uluo o e al. [36] ). A maximum wind speed o 25 m s 1 , wi h a liquid wa e con en o 0.5 g m 3 a a empe a u e o 10 °C was used o acc e e he ice in his expe imen . The o ma ion o a mixed ice ype is expec ed om hese expe imen al condi ions. An ice film wi h a hickness be ween 0.5 and 1 mm, depending on he loca ion on he chip su ace, was acc e ed and kep in he wind unnel chambe a 10 °C be o e being subjec ed o AW ac i a ion. 2.3.2. Expe imen 2 The expe imen was pe o med wi h a 510 μm chip o 15 30 mm 2 wi h he IDT s uc u ed on he ice-exposed side o he LN pla e. In his case, a la ge olume o wa e o ≈180 μL was deposi ed in ambien ai on he chip su ace ou side he zone o IDTs. The wa e co e ed an a ea o ≈0.9 cm 2 , sepa a ed by a dis- ance o ≈2 mm om he IDT. Then, he chip and connec ing boa d we e placed inside he cooling chambe , and he empe a- u e dec eased up o 15 °C o induce ime ice o ma ion. A e wa d, he chip and boa d we e emo ed om he cooling chambe , and he AW de-icing was induced in ambien condi- ions immedia ely a e connec ing wi h he elec onics ( o al ime sho e han 10 s). The p ocess was ollowed by placing a ideo came a ou side. Du ing his ime, he sample empe a u e ne e su passed 0 °C wi hou AW ac i a ion. These condi ions gene a ed some os on op o he su ace o he chip due o wa e apo condensa ion, as isible du ing he expe imen s. Fo compa a i e pu poses, a simila expe imen o he one done in Expe imen 2 was also ca ied ou wi h a 420 μm chip gene - a ing Rayleigh SAWs. Fo he expe imen s, he elec onic con ol and ac i a ion sys ems we e placed ou side ei he he cooling chambe (expe imen 2) o IWT (expe imen 1, al hough in his case, he ambien empe a u e was 10 °C). Elec onics consis ed o a ec o ne wo k analyze (VNA) o de e mine he wo king e- quency du ing and a e ice acc e ion and a swi ch allowing o au oma ically connec he chip o a signal gene a o sys em con- sis ing o a signal gene a o (Keysigh 33210A) and an amplifie (Mini-Ci cui s LZY-22þ). Fu he de ails o his elec onic con- ol and ac i a ion sys em can be ound in Del Mo al e al. [24] The epo ed de-icing expe imen s we e ca ied ou applying an RF signal a a ound 7.3 MHz wi h sligh a ia ions in he o de en hs o kHz depending on he chip and peak- o-peak ol ages o 53 V (expe imen 1) and 70 V (expe imen 2), equi alen o app oxima e powe alues o 1.7 and 4.8 W, espec i ely. 2.4. Fini e Elemen Model (FEM) Calcula ions Fini e elemen (FE) simula ion o he 120 and 510 μm chips was conduc ed using COMSOL Mul iphysics e sion 6.0. This simu- la ion desc ibes he gene a ion and p opaga ion o AWs and hei in e ac ion wi h ice, e i ying he easiness o mechanical ene gy ansmission h ough he piezoelec ic LN and he ice in e ace. Fo his pu pose, we ha e calcula ed he displacemen field (u)in he subs a e and he ice agg ega e unde he assump ion ha la ge displacemen s in he ice agg ega e mean a be e ansmis- sion o he mechanical ene gy o he AW om he ac i a ed LN o ice. The ice agg ega e was loca ed a he same dis ance om he edge o he las finge o he IDT as in he expe imen s, ying o ep oduce hei ac ual condi ions. To keep he compu a ional cos s wi hin easonable limi s, he olume o he simula ed d op- le has been se led a 4 μL. Ice agg ega es we e assumed o p es- en a hemisphe ical shape, as i ypically happens a e eezing wa e d ople s deposi ed on su aces. The AW ib a ions o he LN pla e wi h IDTs ha e been mod- eled using a se ies o elec odynamic and mechanic equa ions, as p e iously desc ibed by Fakh ou i e al. [37] Since R-SAW and Lamb wa es on 128° X–Y LiNbO 3 a e mainly pola ized in he median plane, [38] a wo-dimensional (2D) model was simula ed using a gene alized plane s ain assump ion. This assump ion has been adop ed o simula e bo h he 120 and 510 μm chips. The 0.5 mm hick subs a e has been desc ibed wi h he elas ic- i y, piezoelec ic, and elec ic pe mi i i y enso s o LN, [39] which we e adequa ely o a ed o accoun o he ac ual 128° X–Y c ys al cu o he subs a es used in he expe imen s. The ice was mod- eled as an iso opic linea -elas ic ma e ial in he o m o a semi- ci cle geome y, using he ice p ope ies p o ided by Vic o and Whi wo h, [40] co esponding o a compac ype o ice simila o glaze ice. All his in o ma ion, including he desc ip ion o he cha ac e is ics o he IDTs, dimensions, applied ol age, ime s ep used o he simula ions and dimension, and o he compu- a ional pa ame e s, can be ound in he SI sec ion, SI6. No ably, he ice–subs a e in e ace be ween he ice and he LN subs a e has been simula ed as fla , an assump ion ha fi s well wi h he condi ions o he eal sys em o med by a single c ys alline sub- s a e wi h e y li le su ace oughness. An ideal sys em consis - ing o an R-SAW wi h a long wa eleng h (510 μm) in e ac ing wi h he ice agg ega e was also simula ed o compa a i e and discussion pu poses. The pa ame e s used o compu ing he 510 μm Lamb wa e sys em we e kep o his ideal sys em, excep o he assump ion o a fini e hickness o he LN subs a e. This semi-infini e hickness bounda y condi ion educes eflec ion o he AW a he bo om ace o he pla e o a minimum, ha is, imposing a si ua ion ypical o he gene a ion o R-SAWs. This simula ed model will be designa ed in he ex as 510 μm R-SAW o di e en ia e i om he simula ed model o he 510 μm Lamb wa es. Figu e 1 p esen s a scheme wi h he defini ion o sys em coo - dina es (x,y,z) used o calcula ions. Since he COMSOL Mul iphysics calcula ions ha e been ca ied ou in wo dimen- sions (2D), he ele an coo dina es a e xand z(c. . Figu e 1b). www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (4 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License The o me co esponds o he p opaga ion di ec ion o he wa e, while he la e co esponds o he pe pendicula o he LN pla e. Fo calcula ions, figu es, and da a p esen a ion, we ha e adop ed he ollowing defini ions: 1) Volume ic displacemen s: he local, wa e-induced displacemen field unde gone by he la - ice o he LN pla es. They a e cha ac e ized by a la ice displace- men ec o u. The ele an magni udes o da a calcula ion and p esen a ion a e i s modulus in 2D |u| and he wo componen s u x and u z . 2) Su ace displacemen s: he displacemen field unde gone by he ou e su ace o he pla e, which in he icing expe imen is in con ac wi h he ice and is di ec ly in ol ed in he ansmission o he mechanical ene gy o he ice. This defi- ni ion simplifies he no a ion and ema ks on he impo ance o powe confinemen on he su ace o he chip. Su ace displace- men s a e cha ac e ized by a su ace displacemen ec o , s. The ele an magni udes o p esen da a and o calcula ions a e i s modulus in 2D |s| and he wo componen s s x and s z . The geome y o he whole sys em has been disc e ized wi h an uns uc u ed iangula mesh, and he acous ic modes ha e been desc ibed wi h a leas 10 elemen s pe wa eleng h conside ing he ollowing sound eloci ies: in LN subs a e 3981 m s 1 , [39] in ice 1850 m s 1 . [40] The esul ing g id had app oxima ely h ee million elemen s (equi alen o ≈18.3 mil- lion deg ees o eedom). Elec ical exci a ion was modeled using an elec ic po en ial ha ollows he pe iodici y o he IDT fin- ge s, which we e loca ed on he le hal o he model geome y, neglec ing he mechanical damping o he me al elec odes. To a oid undesi ed eflec ions o he AWs a he edges o he chips, he AWs we e abso bed by inco po a ing pe ec ly ma ched laye s (PML) as defined in he COMSOL so wa e documen a ion. [41] The model was sol ed in bo h he equency domain and he ime domain using a equency sol e o a ime-dependen sol e , espec i ely. In he equency domain, he IDTs we e defined as an RF po whe e powe is explici ly defined as an inpu pa ame e (1 W was a bi a ily selec ed o he calcula ions since simula ions a e linea and p o ide S 11 spec a and ha - monic solu ions o he displacemen and elec ic fields, no dependence o esul s is expec ed on he used inpu powe ). Simula ions we e also done in he ime domain o ob ain eal- ime es ima ion o he ib oacous ic esponse o he LN pla e and he LN/ice sys ems upon elec oacous ic exci a ion. In his case, he elec ical exci a ion is applied o he IDTs as a sinusoidal ol age signal wi h defined ampli ude and wo king equency as inpu pa ame e s. Wo king equencies we e selec ed based on he equency domain esul s ollowing maximum powe ans- mission c i e ia. To gene a e equi alen mechanical powe s o he AWs in he simula ion o he 120 and 510 μm chips, he selec ed ol age ampli udes in each case we e di e en o com- pensa e o he di e ences in elec oacous ic coupling. The e o e, he esul ing elec ic powe o each wa eleng h was sligh ly di - e en o ensu e ha he gene a ed mechanical powe was equi a- len in all simula ed models. 3. Resul s and Discussion 3.1. RF Cha ac e iza ion o AW Chips The adio- equency (RF) elec ical cha ac e iza ion o he chips was pe o med by measu ing he S 11 ( e u n loss) and |S 11 | 2 ( eflec ion coe ficien o powe ) pa ame e s as a unc ion o e- quency. The measu ed esonance equencies ( equency o he S 11 minima) closely ma ch he calcula ed esonance equencies o he chips, acco ding o he well-known exp ession ¼ s λ(1) whe e , s, and λa e, espec i ely, he wo king equency, he wa e eloci y on 128° YX LN subs a e (3981 m s 1 ) [39] and he wa eleng h, his la e defined by he IDT layou (i.e., he dis- ance be ween finge s). The se ies o e u n loss spec a plo ed as a unc ion o equency eco ded o he di e en chips p e- pa ed in his wo k a e epo ed as SI7. These spec a show ha he S 11 minima deepen o a highe numbe o finge pai s. To minimize he eflec ed powe , de-icing es s we e ca ied ou wi h 120 and 510 μm chips wi h he la ges numbe o finge pai s es ed du ing he elec oacous ic e alua ion, ha is, 15 and 26, espec i ely. To es ima e he powe e ficiency, he eflec ion coe ficien o powe |S 11 | 2 is he ele an pa ame e , which can be calcula ed om Equa ion (2) and (3): RLðdBÞ¼10 log10 Pi P  (2) jS11j2¼P Pi (3) whe e RL(dB) is he e u n loss o he chips and P and P i a e eflec ed and inciden powe , espec i ely. Figu e 2 shows a se ies o plo s o he eflec ion coe ficien o powe |S 11 | 2 e sus equency as eco ded o he 120 and 510 μm chips wi h di e en numbe s o finge pai s. A pho og aph o hese wo chips is also included in he figu e. The ull se o Figu e 1. Schema ic desc ip ion o he geome y o chips wi h displace- men field and axis defini ions; a) 3D diag am o he chip; b) 2D diag am o he chips as used o he calcula ions. www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (5 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License |S 11 | 2 cu es eco ded o all manu ac u ed chips is shown in Figu e S7, Suppo ing In o ma ion. To ensu e he maximum e ficiency o he de ices, hey should p esen |S 11 | 2 alues close o ze o o minimize he eflec ed powe upon elec ical ac i a ion. The plo s in Figu e 2 show ha he eflec ed powe is highly dependen on he numbe o finge pai s and dec eases when i s numbe inc eases. Fo he longes wa eleng h chip, ha is, he 510 μm chip, powe losses eached sa u a ion a |S 11 | 2 alues a ound 0.25 o 26 finge pai s. Fo he 120 μm chip, a |S 11 | 2 alue smalle han 0.1 is eached wi h 15 finge pai s. In gene al, al hough he eflec ion coe ficien o powe could s ill sligh ly dec ease by inc easing he numbe o finge pai s, he benefi s would be e y small and likely compensa ed by he elec ical powe losses ha also inc ease wi h his pa ame e . The e o e, we ha e selec ed 15 and 26 finge pai s o he 120 and 510 μm chips o de-icing and an i-icing expe imen s. This choice is a good comp omise be ween low eflec ion and an accep able ee chip a ea o de-icing expe imen s (see below). These plo s in Figu e 2 also e eal a lowe ing in baseline o he 120 μm chip (ideally, |S 11 | 2 should equal one ou side o he esonance peak), which is pa icula ly no iceable o he 120 μm chip. This indica es he exis ence o some elec ical powe losses, which a e no due o he esonance exci a ion o he c ys al and a e likely caused by ac o s such as pa asi ic capaci ances in he IDT, ohmic esis ance o he hin film me alliza ion, o con ac esis ances. These pa asi ic losses a e in he o de o 10–20% o he cha ac e ized chips, a pe cen age o he inciden powe ha will no con ibu e o he gene a ion o AW ib a ions h ough elec omechanical coupling. The plo s in Figu e 2 also show ha , as expec ed, hese elec ical powe losses inc ease wi h he num- be o finge pai s. Acco ding o Figu e 2, S7/S8, Suppo ing In o ma ion, he elec oacous ic esponse o he 510 μm chip is di e en . The beha io o his chip is cha ac e ized by wo esonance minima ins ead o one in he equency egion o in e es . We a ibu e hese wo esonances o Lamb pla e wa es ins ead o Rayleigh Figu e 2. Elec ical cha ac e iza ion o AW chips: a1,b1) |S 11 | 2 e sus equency plo s o ab ica ed AW de ices wi h wa eleng hs (λ) o 120 μm (a1) and 510 μm (b1). No e he di e en span o he xscales used in each case. The a ow and ellipse highligh he p og essi e inc ease o powe losses in he se ies o 120 μm de ices as he numbe o finge pai s inc eases. Pho og aphs o he 120 μm a2) and 510 μm b2) chip de ices wi h he maximum numbe o finge pai s. www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (6 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License wa es. Acco ding o he well-es ablished heo y o SAW gene a- ion in piezoelec ic pla es, [42] an R-SAW can be ega ded as he high- equency limi o he supe posi ion o symme ic and an i- symme ic Lamb wa es. These wo coun e pa s s a o sepa a e o pla e hicknesses o a ound 1.5 λ. F om he |S 11 | 2 plo s in Figu e 2, i appea s ha when he wa eleng h o he AW app oaches he hickness o he pla e, he condi ions o pu e Rayleigh-SAW a e no longe ulfilled. Acco ding o hese consid- e a ions and Equa ion (1), he esonance a 7.36 MHz co e- sponds o he A0 Lamb wa e mode, whe eas he ≈8 MHz esonance mus be a ibu ed o he S0 Lamb wa e mode. In he nex sec ion, we u he suppo his a ibu ion by he sim- ula ions ob ained using COMSOL Mul iphysics analysis o he AWs in he 510 μm chip. RF chips wi h a simila configu a ion bu a wa eleng h la ge han 510 μm beha e simila ly and gene a e wo AWs a di e en equencies (see Figu e S8, Suppo ing In o ma ion o he |S 11 | 2 cu es expe imen ally de e mined o a600μm chip, which is epo ed o compa ison). Table 1 includes expe imen al and calcula ed (i.e., Equa ion (1)) esonance equencies o he manu ac u ed chips wi h he maxi- mum numbe o finge pai s in each case, as well as he ype o gene a ed AW. 3.2. FEM Simula ion o AW Simula ion esul s o he AW gene a ion and p opaga ion in he 120 and 510 μm chips we e ca ied ou as desc ibed in he Expe imen al and Me hods sec ion. Figu e 3a,b p esen selec ed esul s o his simula ion analysis o each ype o wa e. Videos (V1_S10) and (V2_S11) p o ided as SI showcase he ime- dependen a ia ion o olume ic displacemen s ep esen ed in he o m o colo maps wi h he same colo code as o he snapsho s in Figu e 3a2,b2. The plo s in Figu e 3a1,b1 compa e he expe imen al and cal- cula ed |S 11 | 2 spec a o he 120 and 510 μm chips. The conco - dance be ween cu es ob ained wi h expe imen s and simula ions confi ms ha simula ions ep oduce he expe imen ally eco ded signals, excep o he elec ical IDT powe losses e idenced by he non-ze o backg ounds o he expe imen al cu es, a ea u e pa - icula ly no iceable o he 120 μm R-SAW. This p o es ha , excep o his de ia ion a ibu able o ohmic losses and/o capaci i e e ec s in he eal chips no included in he analysis, he simula- ions o he wo chips ag ee wi h hei ac ual beha io . The analysis in Figu e 3a2 o he olume ic displacemen s de e mined in he ime domain o he 120 μm chip confi ms ha ib a ions only a ec he ou e su ace laye s o he LN pla e, as expec ed o a ypical R-SAW ype (see ideo (V2_S11), Suppo ing In o ma ion). This is e idenced by he colo maps in his figu e showcasing snapsho s o he modulus (|u|) and u x and u z componen s o olume ic displacemen s o his wa e ha confi m he confinemen in he ou e su ace egions o he pla e. The plo s in he same figu e ep esen ing he alue o |u| along he pla e hickness as a unc ion o he coo dina e z show ha he displacemen field goes o ze o in he bulk. Figu e 3a3 p esen s he plo s o su ace displacemen compo- nen s s x ,s z , and he modulus o he displacemen ec o a he su - ace |s| along he di ec ion o wa e p opaga ion (x). No ably, hese pa ame e s co espond o he AW-induced displacemen s in he pla e ou e plane o he chip pla e, ha is, he su ace ha will be in con ac wi h he ice. Figu e 3a3 e eals ha maximum dis- placemen s along xand za e almos equi alen and a e sepa a ed by a phase shi o 93°. This esul means ha he su ace displace- men field app oaches a ci cula pola iza ion, as shown by he plo in Figu e 3c ep esen ing he componen s x e sus s z (no e ha he di e en ci cles in he plo e e o calcula ions a di e en imes). The “in he ime domain”simula ion o he elec oacous ic ac i a ion o he 510 μm chip in Figu e 3b 1 gi es ise o wo well-defined and dis inguishable |S 11 | 2 peaks ma ching he shape o he expe imen al cu es, as well as he equency alues o he minima (Table 1). Meanwhile, he analysis in Figu e 3b 1 demon- s a es ha olume ic displacemen s a ec he whole pla e hick- ness. The colo plo s in his figu e show ha he fi s peak a 7.4 MHz (no e ha he e is a small di e ence wi h espec o he expe imen al minimum a 7.36 MHz) co esponds o ib a- ions wi h an i-symme ical cha ac e and should be a ibu ed o a A0 ib a ional mode. Meanwhile, olume ic displacemen s co esponding o he second peak a 8.1 MHz ha e a symme ical cha ac e (i.e., ypical o an S0 mode, da a no p esen ed). Meanwhile, plo s in his figu e o |u| as a unc ion o he coo di- na e zwi hin he pla e show ha he displacemen field does no anish in i s in e io and ex ends om one ace o he opposi e o he LN pla e. Figu e 3b 3 shows ha he magni ude o s x and s z componen s is simila , and he e is a phase shi o 91º be ween hem, ha is, he su ace displacemen s a e ci cula ly pola ized. This is confi med by he ep esen a ion o s x e sus s z in Figu e 3c. When compa ing he wa es gene a ed in he 120 and 510 μm chips, i is no ewo hy ha hey p esen su ace and pla e cha - ac e , espec i ely. Ano he significan di e ence is ha he su - ace displacemen s in he 510 μm chip o he A0 mode a e sligh ly smalle han o he 120 μm chip, and hey p esen a ce - ain modula ion due o he supe posi ion o a second con ibu- ion o much la ge wa eleng h in he o de o 4–5 mm. This second con ibu ion is likely due o a weak s anding wa e mode wi h a la ge wa eleng h ha is supe imposed on he main one (see Figu e S9, Suppo ing In o ma ion). 3.3. De-icing Expe imen s o Small Ice Agg ega es To e i y he e ec o he AW cha ac e is ics on he de-icing p o- cess, we ha e sys ema ically compa ed a se ies o de-icing expe i- men s ca ied ou wi h he 120 and 510 μm chips. Acco ding o he p e ious simula ions in Sec ion 3.2, hese chips gene a e Rayleigh SAW and Lamb wa es, espec i ely. Table 1. Calcula ed and expe imen al equencies o he bes IDT configu a ion o he chips in e ms o he numbe o finge pai s. Wa eleng h [μm] Numbe o finge pai Measu ed equency [MHz] Calcula ed equency [MHz] Mode 120 15 32.32 33.25 R-SAW 150 16 25.79 26.60 R-SAW 240 20 16.10 16.62 R-SAW 330 22 11.68 12.09 R-SAW 420 28 ≈9.20 9.50 R-SAW 510 26 ≈7.36 & 8.14 –Lamb wa e www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (7 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License Expe imen s we e done a empe a u es o 5, 10, and 15 °C. A summa y o he expe imen al condi ions is shown in Table 2, including in o ma ion abou empe a u e, ac ual e- quency, and inciden powe alues (i.e., DPT alues), as well as an es ima ion o he e ec i e powe ac ually a ailable o AW exci a ion. The la e ha e been es ima ed a e he co ec ion o he expe imen al DP T alues. A fi s co ec ion conside s ha he elec ical powe ha is no inse ed in o he LN pla e in he o m o AW elec omechanical ac i a ion would be equi alen o 1- |S 11 | 2 . Fo ins ance, o he 120 μm chip wi h 15 finge pai s, i 0 0.1 0.2 0.3 0.4 0.5 0 0.4 0.8 1.2 1.6 = 1/5T = 2/5T = 3/5T = 3/5T = T ) mn ( ||u|| z (mm) 0 0.1 0.2 0.3 0.4 0.5 0 0.4 0.8 1.2 1.6 =1/5T =2/5T =3/5T =4/5T =T )mn( ||u|| z (mm) (b)(a) 28 30 32 34 36 0 0.2 0.4 0.6 0.8 1 F equency (MHz) Expe imen al Simula ion |11S| 2 32.5 MHz = 120 m 6 6.5 7 7.5 8 8.5 0 0.2 0.4 0.6 0.8 1 F equency (MHz) Expe imen al Simula ion | 11S | 2 7.4 MHz 8.1 MHz = 510 m -2 -1 0 1 2 -2 -1 0 1 2 510um BAW 120um SAW sz)m n ( s x (nm) (a2) (a3)(b 3) (a1)(b 1) (b2) 345 -1.5 -1 -0.5 0 0.5 1 1.5 )mn( x(mm) sx sz || s || 0.75 1 1.25 -1.5 -1 -0.5 0 0.5 1 1.5 sx sz || s || )mn( x(mm) Su ace displacemen ield ( ) (c) Volume ic displacemen ield ( ) 9.6 x10 x (mm) )mm( z 6.4 x10 x (mm) 2.6 z (mm) x10 1.3 x (mm) z (mm) 9.6 9.6 x10 x10 6.46.4 x (mm) z (mm) 9.6 9.6 x10 1.31.3 Figu e 3. Simula ion o he AWs: a,b) simula ions o he o he 120 μm (a) and 510 μm (7.4 MHz peak) (b) chips. a 1 ,b 1 ) simula ion in he equency domain o |S 11 | 2 .a 2 ,b 2 ) simula ion in he ime domain o he bulk displacemen s. Simula ions include: i) colo maps snapsho s o he bulk displacemen s (modulus |u| and u x and u z componen s); ii) plo s a a ious imes o he modulus o he bulk displacemen ec o |u| along he coo dina e z( hickness o he pla e). a 3 ,b 3 ) plo s he modulus |s| and s x and s z componen s. No e ha he xscale is no he same o he wo wa es o p ope ly compa e he displacemen fields. c) Plo o he s x e sus s z componen s o assess he pola iza ion o he displacemen field a he su ace in he wo chips. www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (8 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License appea ed ha app oxima ely an 8% o he inciden elec ical powe is no e ficien ly coupled o he de ice. As indica ed in Sec ion 3.1, pa asi ic elec ical powe losses associa ed wi h he IDTs should be summed up in his co ec ion. Pa asi ic losses we e impo an o he 120 μm chip wi h 15 finge pai s, which amoun ed o ≈20% o inciden powe (c. ., Figu e 2). In o al, a ough es ima e o he elec ical powe ac ually con e ed in o mechanical exci a ion wi h his chip migh be in he o de o 72% o he inciden powe ( his pe cen age s ems upon subs a - ing he 20% o pa asi ic losses plus ≈8% due o he coupling issues). Fo he 510 μm chip, a simila analysis gi es an es i- ma ed co ec ing ac o o 69% (no e ha in his case ohmic o capaci i e losses associa ed o he IDTs we e smalle , c. . Figu e 2). Since he es ima ed powe losses esul ing om hese co ec ing ac o s we e a he simila o he 120 and 510 μm chips, compa a i e assessmen s o powe s and e ficiencies would be equi alen using ei he inciden o e ec i e powe alues. No e, howe e , ha hea ing e ec s due o ohmic losses a he IDTs will likely be highe o he 120 μm chip. The de-icing expe imen s p oceeded in he ollowing way: once he chip empe a u e was s abilized in he cooling chambe , he sys em was kep unde hese condi ions o a leas 30 min o empe a u e homogeniza ion. Then, a liquid d ople (þ2 °C, 22 μL) was deli e ed and ozen on he su ace o he chips, o m- ing a hemisphe ical ice agg ega e defined by a ci cula pe ime e o ≈4.5 mm. Then, an RF signal wi h he inciden (DP T ) powe epo ed in Table 2 was applied o gene a e AWs a he esonance equencies o each chip. The epo ed alues co espond o he minimum inciden powe equi ed o induce mel ing. Bo h DP T and he e ec i e AW powe alues in Table 2 confi m ha mel - ing on he 510 μm chip is mo e a o able han on he 120 μm chip. We en a i ely a ibu e his di e ence o he dis inc cha - ac e is ics o he wa es gene a ed in each case, a ea u e ha will be u he discussed in he ollowing sec ions. The mel ing sequence induced upon applica ion o he epo ed powe s in Table 2 was ideo- eco ded o bo h chips. Figu e 4 shows a se ies o snapsho s ep esen ing d ipping (Figu e 4a1,b1), eezing (Figu e 4a2,b2), pa ial mel ing Table 2. De-icing da a o he 120 and 510 μm chips: empe a u e, ope a ing equencies, inciden (DP T ), and e ec i e AW powe s a e co ec ion by he es ima ed losses. Chip [μm] Tempe a u e [°C] F equency [MHz] Inciden powe a) (DP T ) [W] e o E ec i e powe [W] 120 15 32.33 3.76 0.26 2.70 10 32.40 2.76 0.68 1.98 5 32.44 1.46 0.20 1.05 510 15 7.36 2.02 0.39 1.39 10 7.40 1.46 0.29 1.00 5 7.36 1.45 0.16 1.00 a) The unce ain y limi s o he DP T alues define he a ia ion in he measu emen s in each case. Figu e 4. Compa ison o de-icing p ocess on he a) 120 μm and b) 510 μm chips. A 22 μL d ople is deli e ed on he chips a –15 °C a1,b1); d ople is ozen; a2,b2): d ople s a s mel ing, a e chip ac i a ion wi h he DP T powe s epo ed in Table 2 a3,b3); d ople emains liquid (ac i e an i-icing e ec ) a4,b4). The beha io o he mel ed wa e d ople was di e en o he wo chips: i sp ead o he 120 μm chip bu was pushed owa d he chip edge in he case o he 510 μm chip. The wa e d ople was dispensed a almos simila dis ance om he las finge o he IDTs (i.e., a abou 3 mm). www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (9 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License [22] X. Zeng, H. Ong, L. Hawo h, Y. Lu, D. Yang, M. Rahma i, Q. Wu, H. To un, J. Ma in, X. Hou, X. L , W. Yuan, Y. He, Y. Fu, ACS Appl. Ma e . In e aces 2023,15, 35648. [23] X. Zeng, Z. Yan, Y. Lu, Y. Fu, X. L , W. Yuan, Y. He, Langmui 2021,37, 11851. [24] J. del Mo al, L. Mon es, V. Rico-Ga i a, C. López-San os, S. Jacob, M. Oli a-Rami ez, J. Gil-Ros a, A. Fakh ou i, S. Pandey, M. Gonzalez delVal,J.Mo a,P.Ga cía-Gallego,P.F.Ibá˜nez-Ibá˜nez, M.A.Rod íguez-Val e de,A.Winkle ,A.Bo ás,A.R.González-Elipe, Ad . Func . Ma e . 2023,33, 2209421. [25] S. Jacob, S. Pandey, J. Del Mo al, A. Ka imzadeh, J. Gil-Ros a, A. R. González-Elipe, A. Bo ás, A. Winkle , Ad . Ma e . Technol. 2023,8, 2300263. [26] A. Ka imzadeh, U. Weisske , J. del Mo al, A. Winkle , A. Bo ás, A. R. González-Elipe, S. Jacob, Ad . Ma e . Technol. 2024,9, 2301749. [27] V. I. Anisimkin, N. V. Vo ono a, Ul asonics 2021,116, 106496. [28] P. Schulmeye , M. Weihnach , H. Schmid , Senso s 2024,24, 2292. [29] V. Anisimkin, V. Koleso , A. Kuzne so a, E. Shamsu dino a, I. Kuzne so a, Senso s 2021,21, 919. [30] D. Yang, L. Hawo h, P. Ag awal, R. Tao, G. McHale, H. To un, J. Ma in, J. Luo, X. Hou, Y. Fu, Langmui 2022,38, 11314. [31] M. Rashedul, H. Sa ke , J. L. Sil a, M. Cas aneda, B. Wilbu n, Y. Lin, N. Lo e, J. In ell. Ma e . Sys . S uc . 2018,29, 938. [32] C. Zhaoa, W. Genga, X. Qiaoa, F. Xuea, J. Hea, G. Xuea, Y. Liua, H. Weia, K. Bia, Y. Lib, M. Xunb, X. Choua, Sens. Ac ua o s: A Phys. 2022,333, 113230. [33] H. Lu, Y. Li, J. Zhang, Senso s 2021,21, 3672. [34] A. Winkle , S. Menzel, H. Schmid , in P oc. o SPIE, ol. 7362, SPIE Eu ope Mic o echnologies o he New Millennium, D esden, Ge many 2009, p. 73621Q. [35] G. Fo in, J. Pe on, in 47 h AIAA Ae ospace Sciences Mee ing Including The New Ho izons Fo um and Ae ospace Exposi ion, O lando, FL, Janua y 2009, p. AIAA 2009-0274. [36] H. Koi uluo o, C. S en oos, R. Ruohomaa, G. Bolelli, L. Lus a ghi, P. Vuo is o, in P oc. o he In . Wo kshop on A mosphe ic Icing o S uc u es (IWAIS), Uppsala, Sweden 2015, p. 28. [37] A. Fakh ou i, C. De end an, T. Alb ech , D. Collins, A. Winkle , H. Schmid , A. Neild, Lab Chip 2018,18, 2224. [38] D. Mandal, S. Bane jee, Senso s 2022,22, 820. [39] G. Ko acs, M. Anho n, H. E. Engan, G. Visin ini, C. C. W. Ruppel, IEEE Symp. Ul ason. 1990,1, 438. [40] V. F. Pe enko, R. W. Whi wo h, in Physics o Ice, Ox o d Uni e si y P ess, Ox o d 2002. [41] Pe ec ly Ma ched Laye s, h ps://doc.comsol.com/5.5/doc/com. comsol.help.aco/aco_ug_p essu e.05.106.h ml (accessed: July 2024). [42] E. A. Ash, in Acous ic Su ace Wa es (Ed: A. A. Oline ), Sp inge , Be lin 1978. [43] P. B une , M. Baudoin, O. Bou Ma a , F. Zouesh iag, Phys. Re . E 2010,81, 036315. [44] A. Wix o h, C. S obl, C. Gaue , A. Toegl, E. J. Sc iba, Z. V. Gu enbe g, Anal. Bioanal. Chem. 2004,379, 982. [45] R. J. Shil on, V. Ma oli, M. T a aglia i, M. Agos ini, A. Desii, F. Bel am, M. Cecchini, Ad . Func . Ma e . 2015,25, 5895. [46] H. L. Ong, D. Yang, H. Chen, J. Zhou, L. Hawo h, J. Zhang, D. Gibson, P. Ag awal, H. To un, Q. Wu, X. Hou, Ma e . Chem. Phys. 2023,304, 127842. [47] M. Baudoin, R. Chu ani, F. B e agnol, A. Pe e , US 20240042969A1 2024, h ps://pa en s.google.com/pa en /US20240042969A1/en?oq=. þUSþ2024%2 0042969þA1. www.ad ancedsciencenews.com www.aem-jou nal.com Ad . Eng. Ma e . 2024, 2401820 2401820 (16 o 16) © 2024 The Au ho (s). Ad anced Enginee ing Ma e ials published by Wiley-VCH GmbH 15272648, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/adem.202401820 by Readcube (Lab i a Inc.), Wiley Online Lib a y on [19/02/2025]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License