Hemp cellulose-based aerogels and cryogels: From waste biomass to sustainable absorbent pads for food preservation

Hemp cellulose-based ae ogels and c yogels: F om was e biomass o sus ainable abso ben pads o ood p ese a ion Lau a Cab e a-Villamiza a , J´ essica Fe nanda Pe ei a a , Ma ía Cas anedo a , Ampa o L´ opez-Rubio a,b , Ma ía Jos´ e Fab a a,b,* a Food Sa e y and P ese a ion Depa men , Ins i u e o Ag ochemis y and Food Technology (IATA), CSIC, Valencia, Spain b In e disciplina y Pla o m o Sus ainable Plas ics Towa ds a Ci cula Economy- Spanish Na ional Resea ch Council (SusPlas -CSIC), Mad id, Spain ARTICLE INFO Keywo ds: Hemp cellulose Ae ogels C yogels Abso ben pads Food p ese a ion ABSTRACT This s udy p esen s a ci cula economy app oach u ilizing hemp s ems and ice s aw, ypically pe cei ed as low- alue ag icul u al was e, o de elop a sus ainable al e na i e o adi ional plas ic abso ben pads o ood packaging. The de elopmen o an ac i e ma e ial was achie ed h ough he u iliza ion o hemp cellulose and a bioac i e ex ac isola ed om ice s aw. In addi ion o educing plas ic pollu ion, his ma e ial demons a es he po en ial o enhance ood p ese a ion. This esea ch p o ides e idence o he bene i s o epu posing ag icul u al by-p oduc s o c ea e aluable and en i onmen ally- iendly p oduc s. Hemp cellulose was ex ac- ed, cha ac e ized, and p ocessed o de elop s able ae ogels and c yogels h ough supe c i ical CO 2 d ying and eeze-d ying. The wa e s abili y and in e nal s uc u e o he ma e ials we e guided ia TEMPO-media ed oxida ion and high-p essu e homogeniza ion. Bo h ma e ials showed e sa ile physicochemical and mechani- cal p ope ies. Ne e heless, wi h highe wa e so p ion (2.20 mL/g), minimal dimensional changes, and lowe sh inkage, c yogels we e sui able o mea abso ben pad applica ion. To enhance he c yogels unc ionali y, hey we e imp egna ed wi h a ice s aw bioac i e ex ac in wo di e en concen a ions. The inco po a ion o he ex ac did no a ec he s uc u e o he c yogels, imp o ed hei mechanical p ope ies and he an ioxidan ac i i y emained s able a e d ying (63.89–78.96 %). Finally, he pe o mance o he de eloped ma e ials was compa ed o comme cial plas ic pads and p is ine mea p ese a ion challenge es du ing 9 days a e ige a ion condi ions. The inco po a ion o ice s aw ex ac imp o ed mea colo p ese a ion. While mode a e ex ac concen a ions (75 mg/g) showed a p o ec i e e ec agains lipid oxida ion, highe le els (187.5 mg/g) induced p o-oxidan eac ions. This esea ch highligh s he po en ial o hemp cellulose-based c yogels as sus ainable and unc ional packaging ma e ials o mea p oduc s. 1. In oduc ion Ae ogels and c yogels a e po ous ma e ials ha can be p epa ed om o ganic, ino ganic and e en composi es o di e en ma e ials wi h a po en ial use o a wide ange o applica ions, such as liquid so p ion, he mal insula ion, il a ion and as ca ie s o di e en ac i e mole- cules (Long e al., 2018). Howe e , wo ies abou sus ainabili y, biodeg adabili y and long- e m en i onmen al impac ha e d i en esea ch o ind en i onmen ally- iendly al e na i es. In his ega d, he use o biopolyme s (i.e. polysaccha ides, gums and p o eins) has been gaining a en ion o ae ogel p epa a ion (Ni a e al., 2020) due o hei biodeg adable, biocompa ible and enewable cha ac e is ics. Ae ogels a e ma e ials wi h excep ional p ope ies, including ul a- low densi y, high po osi y and a iable mechanical and he mal insu- la ion depending on he composi ion. These cha ac e is ics make hem highly e sa ile o di e se applica ions (Liebne e al., 2008). The d ying me hod employed signi ican ly in luences he in e nal s uc u e and unc ional p ope ies o ae ogels. Supe c i ical CO₂ d ying (SC-CO₂) and eeze-d ying a e commonly used o p oduce ae ogels and c yogels, espec i ely (Da pen igny e al., 2020). While SC-CO₂ d ying o e s ad- an ages such as p ese ing he o iginal gel s uc u e, comple e sol en emo al, and a highe speci ic su ace a ea, wi h smalle po e sizes sui able o con olled elease, i comes wi h inc eased p oduc ion cos s and equi es mo e p ocessing s eps o ensu e hyd ogel s abili y and * Co esponding au ho a : Food Sa e y and P ese a ion Depa men , Ins i u e o Ag ochemis y and Food Technology (IATA), CSIC, Valencia, Spain. E-mail add esses: [email p o ec ed] (L. Cab e a-Villamiza ), [email p o ec ed] (J.F. Pe ei a), [email p o ec ed] (M. Cas anedo), ampa o. [email p o ec ed] (A. L´ opez-Rubio), [email p o ec ed] (M.J. Fab a). Con en s lis s a ailable a ScienceDi ec Ca bohyd a e Polyme s jou nal homepage: www.else ie .com/loca e/ca bpol h ps://doi.o g/10.1016/j.ca bpol.2024.122887 Recei ed 3 Sep embe 2024; Recei ed in e ised o m 14 Oc obe 2024; Accep ed 15 Oc obe 2024 Ca bohyd a e Polyme s 348 (2025) 122887 A ailable online 19 Oc obe 2024 0144-8617/© 2024 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license ( h p://c ea i ecommons.o g/licenses/by- nc-nd/4.0/ ). sol en eplacemen (Yang & C ans on, 2014). In con as , c yogels a e cha ac e ized by lowe p oduc ion cos s, highe mechanical s eng h, and as e compound elease (Ko honen & Bud o a, 2020). Howe e , hey exhibi lowe speci ic su ace a ea and may expe ience s uc u al deg ada ion due o ice c ys al o ma ion du ing he d ying p ocess (Buch o ´ a e al., 2019). By in es iga ing bo h SC-CO₂ d ying and eeze- d ying in he con ex o ood-abso ben pad de elopmen , his s udy aims o explo e hei espec i e impac s and sui abili y o indus ial applica ions. Fu he mo e, he ma e ial om which hey a e made is o g ea impo ance in de e mining hei applica ions. Polysaccha ides and, in pa icula , lignocellulosic ma e ials ha e al eady demons a ed hei po en ial o de elop ae ogels (Beni o-Gonz´ alez e al., 2020; Beni o- Gonz´ alez e al., 2021; de Oli ei a e al., 2019; Fon es-Candia e al., 2019; Zaman e al., 2020a). Cellulosic ma e ials can be ex ac ed om ag o- indus ial esidues and o es y-de i ed was e by-p oduc s such as husks, lea es, s ems and seeds (Vallejo e al., 2021), and he quali y and p ope ies o he ob ained cellulose will be also in luenced by i s sou ce and p ocessing. One sus ainable sou ce o cellulose is hemp (Cannabis sa i a) and i s de i a i es. Hemp cellulose possesses excellen mechan- ical p ope ies, p o iding s i ness and s eng h o composi e ma e ials (Momeni e al., 2021). Fu he mo e, hemp is a mo e sus ainable solu ion han o he wood c ops, as i is a as -g owing plan whose esidues can be alo ized (Ahmed e al., 2022). In addi ion, hemp cellulose has p o en o be one o he bes ma e ials o pape making due o i s s eng h and ligh ness, asc ibed o he acili y o o m an s uc u al ne wo k and he p esence o longe ibe s han o he plan s, making i s onge (Gibson, 2006). P e ious epo s ha e shown he p oduc ion o c yogels om hemp cellulose o he mal insula ion (Beluns e al., 2021; Zhu e al., 2023); oil so p ion in seawa e (Paulauskiene e al., 2024); con olled elease o e ilize s (Kau e al., 2023), and was ewa e o ai cleaning (Lyu e al., 2023). While p e ious s udies ha e demons a ed he e sa ili y o hemp cellulose c yogels in a ious applica ions, hei po en ial o ood- ela ed uses, such as ood p ese a ion pads and SC-CO 2 d ying ech- niques, emains unexplo ed. This esea ch seeks o add ess his knowl- edge gap by in es iga ing he easibili y and e ec i eness o hemp- cellulose c yogels and ae ogels in his speci ic con ex . In he pa ic- ula case o mea packaging, non-pe meable/non-s ick syn he ic poly- me s con aining a hyd ophilic non-wo en bo om laye illed wi h ac i e compounds (i.e., sodium bica bona e o ci ic acid) a e used o p oduce abso bing pads o use hem as pa o he packaging s uc u e. They abso b he excess o liquids eleased upon ood s o age, imp o e he isual pe cep ion, p e en bac e ial g ow h and o he ac o s ha could al e p oduc quali y (Cas ica e al., 2020). To educe en i onmen al impac , esea ch e o s a e being ocused on he use o mo e sus ainable bio based and enewable ma e ials (i.e., cellulose) o p oduce ood-g ade abso bing pads (Iskanda e al., 2022). Ad an ageously, po ous cellu- losic ma e ials can be also imp egna ed wi h ac i e compounds wi h p o ed an imic obial/an ioxidan p ope ies o enhance ood p ese a- ion pe o mance (Fon es-Candia e al., 2019). Fo example, al e na i e app oaches o he de elopmen o ac i e an imic obial ae ogels ha e in ol ed he inco po a ion o me als and hei oxide nanopa icles, which can be sa ely inco po a ed in o ood packaging (Wu, Zhou, e al., 2023). Ne e heless, consume s and esea che s end o be mo e ecep i e o he u iliza ion o plan -de i ed compounds. In he con ex o alo iza ion o lignocellulosic was e in he ansi ion o ci cula econ- omy, Cab e a-Villamiza , Eb ahimi, e al. (2024) ha e ecen ly de el- oped g eene s a egies o ice s aw bio e ine y using combined alkali, ozone and enzyme ea men s o ob ain added- alue by-p oduc s, no only cellulose- ich ac ions bu also aluable an ioxidan compounds o de elop ac i e ma e ials. Pa icula ly, he alkaline ac ion exhibi ed a high concen a ion o lignin de i a i es and polyphenols. This bioac i e ex ac , wi h an ioxidan p ope ies, is a p omising candida e o de eloping bioac i e ma e ials while minimizing was e, educing he use o syn he ic addi i es, and mee ing consume demands o na u al and plan -based p oduc s. In he p esen wo k, hemp cellulose ich ac ions we e combined wi h he polyphenol- ich ex ac ( he mos impo an by-p oduc ob- ained om ice s aw bio e ine y p ocess) o ob ain an ioxidan abso bing pads. I is hypo hesized ha no only he d ying condi ions bu also he composi ion o he ac i e ex ac will a ec he in e nal mo phology and, hus, physicochemical and unc ional p ope ies o he hemp cellulose-based ae ogels/c yogels. The e o e, his wo k aimed a unde s anding how d ying p ocess (SC-CO₂ s. eeze-d ying) a ec ed he physicochemical p ope ies o hemp cellulose-based ae ogels and how a side-s eam, ich in polyphenols, ob ained om ice s aw bio- e ine y impac ed on he unc ionali y and p ope ies o he de eloped abso bing pads o ood packaging applica ions. Mo eo e , his s udy p o ides aluable insigh s in o hei impac on he ma e ials s uc u e and unc ionali y. Addi ionally, he indings cons uc a baseline o de elopmen o sus ainable e ec i e abso ben pads using plan -based ma ix and bioac i es. By add essing he gap in hemp cellulose-based ae ogel esea ch and compa ing he wo d ying me hods, his s udy con ibu ed o he de elopmen o sus ainable and e ec i e packaging solu ions o he ood indus y. Fu he mo e, he p oo -o -concep ali- da ion in bee s o age demons a es he eal-wo ld applicabili y o his inno a i e app oach. 2. Ma e ials and me hods 2.1. Ma e ials The d y biomass used in he p esen s udy we e hemp s ems (Cannabis sa i a), which we e g own du ing 2022 in Spain (Bu gos) and p o ided by a local con e e . All eagen s used we e o analy ical g ade, including: hexane (Lab- Kem, 95 %), oluene (VWR Chemicals, ≥98 %), e hanol (PanReac AppliChem, 99.5 %), NaClO2 (Sigma-Ald ich, 80 %), glacial ace ic acid (LabKem, 99.8 %), KOH (Sigma-Ald ich, 90 %), TEMPO (Sigma-Ald ich, 98 %), NaB (Sigma-Ald ich, ≥99 %), NaClO sodium hypochlo i e so- lu ion (Sigma-Ald ich, 10–15 %), NaOH (PanReac AppliChem, ≥98 %), NaCl (PanReac AppliChem, 99 %), HCl (LabKem, 37 %), ci ic acid (Sigma-Ald ich, ≥99.5 %), β-ca o ene (Sigma-Ald ich, ≥97 %), chlo o- o m (Ca lo E ba Reagen s, 99.9 %), linoleic acid (Sigma-Ald ich, ≥99 %), Tween ® 40 (Sigma-Ald ich), bu yla ed hyd oxy oluene (Sigma- Ald ich, ≥99 %), po assium phospha e bu e (NaH2PO4 (Sigma- Ald ich, ≥99 %) and Na2HPO4 (Sigma-Ald ich, ≥99 %)), HCl (Sigma- Ald ich, 37 %), e hanolic p opyl galla e (Sigma-Ald ich, ≥98 %), EDTA (Sigma-Ald ich, 99.4–100.6 %), silicone an i oaming liquid (PanReac AppliChem), 2- hioba bi u ic acid (Sigma-Ald ich, ≥98 %). 2.2. Raw ma e ial cha ac e iza ion The hemp s ems we e washed wi h ap wa e o emo e any adhe ing di o deb is and d ied a oom empe a u e o one week. The d ied s ems we e subsequen ly milled wice wi h mesh sizes o 3 mm and 1.5 mm o ob ain a homogeneous powde be o e hei cha ac e iza ion. Speci ically, lignin con en was de e mined acco ding o he s anda d me hod TAPPI T211 om-07. The monosaccha ide composi ion was de e mined by anion exchange ch oma og aphy wi h pulsed ampe o- me ic de ec ion (HPAEC-PAD) wi h an ICS-3000 sys em (Dionex, The mo Scien i ic, USA) a e sul u ic hyd olysis as desc ibed by Cab- e a-Villamiza , Eb ahimi, e al. (2024). The lipid con en was es ima ed ollowing a Soxhle ex ac ion p ocedu e, which in ol ed he use o 5 g o aw ma e ial wi h 80 mL o hexane o 2 h (Sox ec 8000, Foss Ana- ly ics, Denma k) and he p o ein con en was de e mined using he Kjeldahl me hod, which measu es he o al ni ogen con en in he samples. L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 2 2.3. Cellulose ex ac ion and cha ac e iza ion Cellulose ex ac ion was pe o med wi h sligh modi ica ions based on he me hod desc ibed by Ma ínez-Sanz e al. (2018). The me hod consis ed o h ee main s eps: i) Dewaxing wi h Soxhle ex ac ion, ii) Lignin emo al using an oxida i e ea men and iii) Hemicellulose emo al by means o an alkaline ea men . B ie ly, 7 g o milled hemp s ems we e subjec ed o a h ee-s ep Soxhle ex ac ion using Sox ec TM 8000 appa a us (Foss Analy ics, Denma k) o emo e waxes, lipids, and pigmen s. A 2:1 ( / ) mix u e o oluene and e hanol (75 mL o al ol- ume) was used as ex ac ion sol en . Each ex ac ion cycle consis ed o a 30-min boiling phase a 210 ◦C, a 1.5-h ex ac ion phase and a 10-min sol en eco e y s ep. The esul ing ac ion was placed in a ume hood o allow he sol en e apo a ion o 24 h, ollowed by d ying a 60 ◦C un il cons an weigh was achie ed. Subsequen ly, he dewaxed biomass was placed in o 700 mL o a solu ion o 1.4 % (w/ ) (NaCLO 2 ) adjus ed o pH 3 using glacial ace ic acid and hen hea ed o 70 ◦C. The eac ion occu ed unde magne ic s i ing a 200 pm o 5 h. Then, he eac ion was s opped using an ice ba h, and he mix u e was il e ed using a 15 μ m nylon clo h. The ob ained ac ion was washed wi h dis illed wa e un il neu ali y. Finally, he ob ained biomass was added o 400 mL o a 5 % (w/ ) KOH solu ion. The ea men was pe o med unde magne ic s i ing a 200 pm a oom empe a u e o 24 h, ol- lowed by an addi ional 2 h a 90 ◦C. The eac ion was s opped wi h an ice ba h, and hen i was il e ed o eco e he cellulosic ex ac and was washed un il neu al. The ob ained cellulose ac ion was s o ed in a hyd ogel o ma unde e ige a ion un il u he use. The cellulose and hemicellulose concen a ion we e de e mined by ion exchange ch oma og aphy, as desc ibed in Sec ion 2.2. Addi ionally, he cellulosic ma e ial’s he mal s abili y and decomposi ion pa e ns we e analyzed by he mog a ime ic analysis (TGA), and he iden i i- ca ion o unc ional g oups p esen in he sample was eco ded using Fou ie - ans o m in a ed spec oscopy (FTIR) analysis. 2.4. Cellulose oxida ion, homogeniza ion, and conduc ime ic i a ion Cellulose TEMPO-media ed oxida ion was conduc ed acco ding o Isogai e al. (2011). Subsequen ly, high-p essu e homogeniza ion and ca boxyl g oup de e mina ion we e pe o med as desc ibed by Cab e a- Villamiza , Campano, e al. (2024) wi h some modi ica ions. B ie ly, he cellulose was oxidized using a solu ion (1 mmol TEMPO and 10 mmol NaB ), ollowed by he addi ion o 10 g o cellulose- ich ac ion. The oxida ion was ini ia ed by adding 50 mmol o NaClO while main aining he pH a 10.0 wi h 0.5 M NaOH un il he pH s abilized, indica ing ha he eac ion was comple e. The oxidized cellulose was hen eco e ed h ough ex ensi e washing wi h deionized wa e un il a neu al pH was achie ed, ollowed by il a ion using a nylon clo h wi h a 15 μ m po e size, and i was s o ed a 4 ◦C. To p epa e cellulose nano ibe s (CNF), a 1 % (w/ ) dispe sion o he oxidized cellulose was sonica ed (Ul asonic P ocesso UP400S, Hielsche , Ge many). The suspension was placed in an ice ba h, unde con inuous sonica ion wi h 300 W/cm 2 o acous ic powe densi y and 175 μ m o ampli ude, un il no isible clumps emained. The mo phology o he cellulosic ibe s was analyzed h ough con ocal lase scanning mic oscopy (CLSM) and ansmission elec on mic oscopy (TEM) using a Nikon Eclipse 90i wide- ield mic oscope (Nikon Co po a ion, Japan) and a TEM JEOL jem 1010 100 kV (JEOL L d., Japan), espec i ely. Fo he CLSM analysis, samples o he sus- pension (10 μ L) we e dyed wi h 0.01 % Calco luo Whi e. In p epa a ion o TEM isualiza ion, he samples we e p e-s ained wi h a 2 % u anyl ace a e solu ion. The deg ee o oxida ion was e alua ed h ough conduc ime ic i a ion (Eu ech Cond 6+, The mo Scien i ic, USA). B ie ly, 0.15 g o d ied pulp was mixed wi h 5 mL o 0.01 M NaCl and deionized wa e o a inal olume o 55 mL. The pH adjus ed o 2.8 using 0.05 M HCl and hen inc emen al addi ions o 0.1 mL o 0.05 M NaOH we e made while measu ing conduc i i y, con inuing un il a consis en inc ease was obse ed, which indica ed comple e dep o ona ion o ca boxylic acid g oups. The olume o NaOH consumed be ween he ini ial s able baseline and he in lec ion poin in he conduc i i y cu e e lec ed he concen a ion o hese unc ional g oups in he oxidized pulp sample (Sanchez-Sal ado e al., 2022). 2.5. Fab ica ion o ae ogels and c yogels The oxidized and homogenized suspension was used o p epa e ae ogels and c yogels. To ab ica e ae ogels, 2 g o he suspension we e pou ed in o each well on a 12-well pla e (diame e 15 mm, heigh 20 mm, olume 3.53 cm 3 ). Then, 2 mL o 0.5 M ci ic acid was added ca e ully o he suspension o 10 min as a c oss-linking agen o induce gela ion as p e iously epo ed by Wang and Kim (2022). Then, he hyd ogels we e washed wi h dis illed wa e . Followed by a sol en ex- change by imme sing he hyd ogels in absolu e e hanol o 5 d, wi h a sol en exchange e e y 24 h. The alcogels we e placed in a supe c i ical CO 2 (SC-CO 2 ) ex ac o (Ex a eX, F ance). The d ying condi ions we e 121 ba (12.1 MPa) wi h a p essu e inc ease amp o 10 ba /min up o 80 L/min o CO 2 low a e, a 70 ◦C, o 4 h (iso he m). Finally, he con olled empe a u e was u ned o and he equipmen was g adually dep essu ized a a a e o 1 ba /min (0.1 MPa/min). On he o he hand, o ob ain c yogels 2 g o he suspension we e pou ed in o each well on a 12-well pla e. To emo e dissol ed ai , he solu ion unde wen a d y blo ing s ep. Subsequen ly, he pla e was subjec ed o ul a- eezing a −80 ◦C, ollowed by eeze-d ying (LyoBe a, Tels a , USA). A calipe was used o measu e he diame e o each sample in ip- lica e o quan i y he sh inkage o he ma e ials in hei p oduc ion. The sh inkage was de e mined ac oss he di e en s a es o he ma e ials: liquid, alcogel, d ied, and a e wa e apo so p ion. The esul ing ma e ials we e hen s o ed in a desicca o un il u he analysis o minimize sample mois u e up ake. 2.6. Rice s aw ex ac p epa a ion and inco po a ion The ice s aw was milled wi h 3 mm and 1.5 mm mesh. The g ounded biomass was hen subjec ed o mild alkali hyd olysis, ac- co ding o Cab e a-Villamiza , Eb ahimi, e al. (2024). B ie ly, he ice s aw was imme sed in o a solu ion o 1 M NaOH a pH 13 o 3 h a 121 ◦C in an au ocla e. Then, he ma e ial was il e ed using a nylon clo h o 15- μ m po e size. The liquid ex ac was eco e ed and eeze- d ied. Finally, he ex ac was dialyzed o 5 d in dis illed wa e , and he dialysa e was changed wice daily o educe he sal concen a ion while keeping he lignin-de i ed compounds. The ex ac was inco po a ed be o e he d ying p ocess. Thus, he ice s aw ex ac in powde was added in concen a ions o 75, 100, and 187.5 mg/g in o he oxidized-homogenized suspension. Then, he d y- ing p ocesses we e eplica ed as he con ol ma e ials, as desc ibed in Sec ion 2.5. 2.7. An ioxidan capaci y The an ioxidan capaci y o he ice s aw ex ac was de e mined using a modi ied me hod desc ibed by Olszowy and Dawidowicz (2016). B ie ly, 4 mg o β-ca o ene we e dissol ed in 10 mL o chlo o o m. In a ound-bo om lask, 50 mg o linoleic acid and 400 mg o Tween 40 we e weighed. 2 mL o he β-ca o ene solu ion we e added, and he chlo o- o m was e apo a ed o d yness using a o a y e apo a o . Subse- quen ly, 50 mL o ae a ed dis illed wa e was added o he lask, and he mix u e was igo ously s i ed. Samples we e p epa ed in dis illed wa e a a inal concen a ion o 2 mg/mL. Bu ylhyd oxy oluene (BHT) se ed as he posi i e con ol, p epa ed a a concen a ion o 2 mg/mL in ab- solu e e hanol. Dis illed wa e was used as he nega i e con ol. A 96- well pla e was used o he assay. Aliquo s o 10 μ L o samples and con ols we e ca e ully dispensed in o he wells, ollowed by he addi- ion o 250 μ L o he β-ca o ene solu ion. The pla e was hen incuba ed a 45 ◦C o 120 min. Abso bance measu emen s we e aken a 470 nm a 0 L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 3 min and a e 120 min using a CLARIOs a pla e eade (BMG LABTECH, Ge many). The bleaching a e o β-ca o ene and he pe cen age o an ioxidan ac i i y we e calcula ed using he ollowing equa ions: Bleaching a e =Ln Abs0 Abs120*1 120 An ioxidan ac i i y% =(BRcon ol −BRsample BRcon ol )*100 whe e Abs0 is he abso bance alue in ime 0, Abs120 is he alue o he abso bance a 120 min and BR is he bleaching a e. 2.8. Ma e ial cha ac e iza ion 2.8.1. Visual appea ance and mic oscopic analyses The cellulose ibe s, a e chemical and mechanical modi ica ions, we e obse ed using lase FV 1000-IX81 con ocal scanning mic oscope (CLSM, Olympus, Japan) and Calco luo Whi e s ain. Images we e analyzed and p ocessed by using FV10-ASW Ve sion 4.02.03.06 (Olympus Co po a ion, Tokio, Japan). Fu he mo e, he su ace mo phology o he ma e ials and cellulose nano ib il dis ibu ion we e e alua ed using digi al mac opho og aphy. Samples we e imaged a magni ica ions o 2×and 20×using an EVOCAM-II mac oscope (Vision Enginee ing, Woking, UK). Image acquisi ion was pe o med wi h ViPlus 1.00.82 so wa e (2018, Vision Enginee ing), and subsequen image analysis was conduc ed using Nis Elemen s BR 3.2 so wa e (Nikon Co po a ion, Japan). Addi ionally, ligh mic oscopy was used o u he elucida e he appea ance o he su ace o he ma e ials, he o ganiza ion o he cellulosic ibe s, and he po osi y o he ae ogels and c yogels. Images we e acqui ed using a Nikon Eclipse 90i wide- ield mic oscope (Nikon Co po a ion, Japan) equipped wi h a cooled, 5- megapixel digi al colo came a (DS-5Mc; Nikon Co po a ion; Japan). The mic os uc u e o he ae ogels and c yogels was in es iga ed using SEM. Small samples (app oxima ely 5 mm 2 ) we e spu e -coa ed wi h a gold‑palladium mix u e unde acuum o 3 min. Mo phology was examined using a Hi achi S-4800 scanning elec on mic oscope (Hi achi High-Technologies Co po a ion, Tokyo, Japan) a an accele a ing ol age o 10 kV and a wo king dis ance o 8–16 mm. 2.8.2. Densi y and po osi y The densi y o each ae ogel and c yogel sample was de e mined g a ime ically by measu ing i s weigh and olume. The weigh o each sample was measu ed using an analy ical balance (model MS105DU, Me le -Toledo GmbH, Spain). The dimensions o each sample we e measu ed using a digi al calipe (VWR In e na ional, USA). Addi ion- ally, he speci ic su ace a ea was de e mined using he B unaue - Emme -Telle me hod (BET), acco ding o M´ endez e al. (2023). The de e mina ions we e ca ied ou in iplica e o densi y and duplica e o BET analysis. 2.8.3. An ioxidan capaci y o he c yogels The an ioxidan capaci y o he c yogels was e alua ed using he β-ca o ene bleaching me hod acco ding o Fon es-Candia e al. (2019) wi h some modi ica ions. In pa icula , he same condi ions and calcu- la ions desc ibed in Sec ion 2.10 we e used, bu in his case, he olume was inc eased. Fo his pu pose, 2.5 mL o he β-ca o ene solu ion was placed in Py ex ubes, and hen bo h he whole an ioxidan -imp egna ed and he con ol c yogels we e imme sed. A ime 0 and 120 min a e incuba ion a 45 ◦C, 1 mL o he supe na an was sampled o measu e he abso bance a 470 nm in a UV–Vis spec opho ome e (Agilen Tech- nologies, USA). The calcula ions we e pe o med as desc ibed in Sec ion 2.7. 2.8.4. Wa e apo so p ion capaci y The wa e apo so p ion capaci y o he ma e ials was assessed acco ding o Cab e a-Villamiza , Campano, e al. (2024) using a closed sys ems o e alua e wa e up ake un il sa u a ion. Ae ogel and c yogel samples we e p e- ea ed by hea ing a 60 ◦C o 2 h o emo e adso bed mois u e, ollowed by cooling in a desicca o wi h silica gel o comple e d ying. The d y weigh o each sample was measu ed using an analy ical balance (model MS105DU, Me le -Toledo GmbH, Spain). Samples we e hen placed in a desicca o wi h dis illed wa e o c ea e a 100 % ela i e humidi y en i onmen , moni o ed by a he mohyg ome e . Weigh s we e pe iodically eco ded un il wa e equilib ium (W e ) and sa u a ion we e achie ed, enabling he de e mina ion o he ma e ials’ wa e apo so p ion capaci y. 2.8.5. Wa e so p ion capaci y The wa e so p ion capaci y was de e mined by imme sing a p e- weigh ed sample o 1 ×1 cm o he ma e ials. The ma e ials we e equilib a ed (d ied) a 60 ◦C o 2 h and in a desicca o wi h silica gel un il eaching oom empe a u e o elimina e esidual wa e . Subse- quen ly, hei weigh s we e measu ed using an analy ical balance (model MS105DU, Me le -Toledo GmbH, Spain). Then, he samples we e imme sed in 15 mL o dis illed wa e a oom empe a u e (22 ± 2 ◦C), and hei mass was de e mined a p ede e mined ime in e als (0.5, 1, 1.5, 2, and 4 h). The abso bed wa e o e ime was calcula ed based on he di e ence o he mass on a de e mina e ime poin minus he ini ial d y weigh o he ma e ial. 2.8.6. Comp essibili y Comp ession es s we e pe o med on he ma e ials using an INS- TRON 34TM-5 Dual Column Table op Model (Ins on Co po a ion, USA) a a s ain a e o 5 mm/min a he empe a u e o 22 ±2 ◦C and ela i e humidi y o 60 ±5 %. A comp ession de o ma ion es was conduc ed wi h a minimum o ce o 0.05 N applied o he samples using he pa allel pla es accesso y. The es s we e ca ied ou using ae ogels and c yogels wi h de e mined dimensions (10 ×5 mm) and (14 ×10 mm), espec- i ely. To ensu e da a accu acy, a minimum o wo samples we e es ed. 2.8.7. The mog a ime ic analysis (TGA) The mog a ime ic analysis (TGA) was pe o med using a TGA 550 ins umen (TA Ins umen s, USA). Samples we e hea ed om 30 ◦C o 850 ◦C a a hea ing a e o 20 ◦C/min unde an ai a mosphe e. De i - a i e he mog a ime ic (DTG) cu es we e analyzed o iden i y weigh loss e en s as a unc ion o empe a u e. The he mal deg ada ion beha io o he samples was de e mined, and he pe cen age o ash con en was calcula ed as he esidual mass a e he comple ion o he empe a u e p og am. 2.8.8. Fou ie - ans o m in a ed spec oscopy (FTIR) FTIR analyses we e pe o med in a enua ed o al e lec ance (ATR) mode using a The mo Nicole Nexus (The mo Fishe Scien i ic, Wal- ham, MA, USA) spec ome e (The mo Fishe Scien i ic, Wal ham, MA, USA). Samples we e placed on he ATR c ys al o analysis a oom empe a u e (22 ±2 ◦C). The spec a we e collec ed wi h a esolu ion o 4 cm −1 in he wa enumbe ange o 4000–600 cm −1 , co-adding a min- imum o 32 scans. 2.9. Challenge es The easibili y o using c yogels o mea s o age was e alua ed h ough an expe imen compa ing he s o age o minced bee o 9 d alone, wi h a comme cial plas ic pad, wi h c yogels and ice-s aw- imp egna ed c yogels. All he ea men s we e pe o med in iplica e. Two concen a ions o he ice s aw we e es ed, 75 and 187.5 mg/g, and, he addi ion o he ex ac was pe o med in powde be o e d ying. B ie ly, 10 g o he ob ained homogenized suspensions we e pou ed in o 60 mm Pe i dishes. The suspensions we e hen ozen a −80 ◦C and subsequen ly eeze-d ied. The an ioxidan capaci y o he esul ing c yogels was de e mined as desc ibed in Sec ion 2.8.3. Then, all he L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 4 es ed ma e ials we e placed in Pe i pla es, and UV s e ilized o 30 min. The esh bee was g ound using a mea g inde o achie e homo- genei y, and 12 g o minced bee was placed on op o he ma e ials unde asep ic condi ions. The pla es we e sealed wi h Pa a ilm o p e- en desicca ion and mic obial con amina ion. The p epa ed samples we e s o ed a 4 ◦C o 9 d. To assess he impac o c yogel s o age on mea quali y, colo ime ic alues, malondialdehyde (MDA) concen a- ion, and lipid oxida ion we e measu ed in he mea samples. Mea- su emen s we e ob ained bo h be o e and a e he 9-d s o age pe iod. 2.9.1. Colo ime y Colo measu emen s we e aken di ec ly om he mea su ace using a a ge mask (MAV, measu emen a ea 8 mm diame e ) a ached o a spec opho ome e (CM26dG, Konica Minol a, Japan). The measu e- men s we e ca ied ou in iplica e, and each eplica e consis ed o h ee subsamples. CIE-L*a*b* coo dina es, Ligh ness (L*), hue (h* ab ), ch oma (C* ab ) and he o al colo di e ences (ΔE) we e ob ained om he e lec ion spec a o he samples using D65 illuminan /10◦obse e . 2.9.2. De e mina ion o myoglobin, oxymyoglobin and me myoglobin Myoglobin con en (% Mb), oxymyoglobin (% O₂Mb), and me - myoglobin (% Me Mb) we e de e mined acco ding o a modi ied me hod desc ibed by Ca lez e al. (1995). B ie ly, 2 g o mea samples we e homogenized wi h 20 mL o 0.04 mol/L phospha e bu e solu ion (PBS) (pH 6.8) using an Ul a-Tu ax homogenize un il homogeneous suspensions we e ob ained. The homogena es we e hen kep on ice o 1 h, ollowed by cen i uga ion a 4200 pm o 30 min a 10 ◦C. The supe na an s we e il e ed h ough 0.45 μ m il e s and hen hey we e olume up o 25 mL wi h PBS. Aliquo s o 250 μ L we e pipe ed in o a 96-well pla e, and abso bance was measu ed a 525, 545, 565, and 572 nm using a CLARIOs a pla e eade (BMG LABTECH, Ge many). The ollowing equa ions we e used o calcula e he pe cen ages o Mb, O₂Mb, and Me Mb: %Mb = (0.369 R1+1.140 R2 −0.941 R3 +0.015)*100 %O2Mb = (0.882 R1−1.267 R2 +0.809 R3 −0.361)*100 %Me Mb = ( − 2.541 R1+0.777 R2 +0.800 R3 +1.098)*100 whe e R1, R2 and R3 a e espec i ely he abso bance (A) a io A572/ A525, A565/A525, A545/A525. 2.9.3. Lipid oxida ion de e mina ion by hioba bi u ic acid eac i e subs ances (TBARS) Lipid oxida ion in he samples was e alua ed be o e and a e he 9- d s o age pe iod using he TBARS assay as desc ibed by Fon es-Candia e al. (2019) wi h mino modi ica ions. B ie ly, 5 g o mea was ho- mogenized wi h 15 mL o dis illed wa e using an Ul a-Tu ax ho- mogenize . The homogena e was hen ans e ed o a 1 L olume ic lask along wi h 32.5 mL o dis illed wa e . The pH was adjus ed o 1.5 using HCl. Subsequen ly, 0.5 mL o 10 % (w/ ) EDTA, 0.5 mL o 10 % (w/ ) e hanolic p opyl galla e, wo d ops o an i oam silicone, and i e glass pea ls we e added o he lask. The lask was connec ed o a Soxhle appa a us, and he mix u e was e luxed un il 50 mL o dis illa e was collec ed. Fo TBARS de e mina ion, 5 mL o he dis illa e was combined wi h 5 mL o 0.02 M TBA solu ion in 90 % ace ic acid wi hin a sc ew-capped es ube and incuba ed in a boiling wa e ba h o 35 min. A con ol sample con aining 5 mL o dis illed wa e and 5 mL o TBA eagen was incuba ed alongside he samples. A e cooling o oom empe a u e, he abso bance o all samples and he con ol was measu ed a 535 nm using a spec opho ome e . TBARS concen a ion was calcula ed by mul i- plying he abso bance alues by a ac o o 7.8 and exp essed as mg malondialdehyde (MDA)/kg mea . 2.10. S a is ical analysis S a is ical analyses we e pe o med o assess signi ican di e ences be ween g oups. Analysis o a iance (ANOVA) was employed o es he null hypo hesis o no di e ence be ween ea men g oups. Following a signi ican ANOVA esul (p <0.05), Tukey’s Hones ly Signi ican Di - e ence (HSD) pos -hoc es was conduc ed o iden i y which speci ic g oups di e ed om each o he . 3. Resul s and discussion 3.1. Cellulose ex ac ion Composi ional analysis was i s pe o med on he aw biomass (C. sa i a s ems) and cellulosic ac ion and he esul s a e summa ized in Table 1. The cellulosic ex ac ion yield was 65.89 ±7.33 %, which is simila o he alues epo ed in he li e a u e o cellulosic hemp ma- e ials (Ba bash e al., 2016). As expec ed, he cellulosic ac ion was en iched a e he ex ac ion p ocess, cellulose being p edominan and eaching alues o 79.47 %. Al hough he lignin and he hemicellulose con en s we e signi ican ly educed in he cellulose- ich ex ac , amoun s o 3.50 and 14.02 %, espec i ely, we e s ill p esen , indica ing he p esence o ecalci an lignin. The cellulosic ex ac was u he cha ac e ized by FTIR and TGA. Typical lignocellulosic bond ib a ional bands we e obse ed. Fo ins ance, he FTIR analysis shown in Fig. S1A e ealed a b oad peak a 3600–3000 cm −1 , cha ac e is ic o O – H s e ching, a ibu ed o hyd ogen bonding be ween hyd oxyl g oups in cellulose and hemicel- lulose (Geminiani e al., 2022). Addi ionally, peaks a ound 1730 and 1500 cm −1 indica ed he p esence o ca bonyl and a oma ic s uc u es, espec i ely. These unc ionali ies a e associa ed wi h es e linkages in hemicellulose and lignin (Le T oedec e al., 2008); howe e , hei low in ensi y sugges s a low lignin con en , as expec ed. Finally, he peak a 1030 cm −1 , cha ac e is ic o s e ching ib a ions, was assigned o he py anose ing skele al ib a ion o cellulose (Mhlongo e al., 2022). The mog a ime ic analysis, as shown in Fig. S1B, e ealed a h ee- s ep deg ada ion p o ile. An ini ial weigh loss o 7 % occu ed be ween 60 and 100 ◦C, a ibu ed o wa e e apo a ion. Subsequen ly, an exo he mic peak a 280 ◦C co esponded o cellulose decomposi ion, accoun ing o 86 % o he o al weigh loss. The inal deg ada ion s age, be ween 300 and 400 ◦C, was indica i e o lignin decomposi ion, consis en wi h p e ious s udies on hemp lignocellulosic ex ac s (Rachini e al., 2009). 3.2. Ae ogels and c yogels de elopmen Cellulosic ae ogels and c yogels, despi e hei ini ially obus po ous s uc u e, exhibi ins abili y when subme ged in wa e . This ins abili y mani es s as dispe sion o esuspension wi hin he aqueous medium. This beha io is a ibu ed o cellulose’s inhe en hyd ophilici y, he ma e ial’s highly po ous s uc u e acili a ing apid wa e pene a ion, and he dis up ion o weak in e molecula o ces, such as hyd ogen bonds, by wa e molecules. P e ious esea ch has add essed his issue by applying polyme ic coa ings o enhance he hyd ophobici y o cellulosic Table 1 Composi ional analysis o he aw ma e ial (C. sa i a s ems) and o he ligno- cellulosic ex ac . Componen (%) C. sa i a s ems Cellulosic ac ion Cellulose 41.21 ±1.89 79.47 ±0.70 Hemicellulose 21.43 ±3.01 14.02 ±0.75 Lignin 22.06 ±2.96 3.50 ±0.16 P o ein 8.01 ±0.05 * Lipid 1.34 ±0.01 * Ash 2.96 ±0.03 3.00 ±0.06 * Unde e mined alues. L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 5 ma e ials (Beni o-Gonz´ alez e al., 2020), he eby p e en ing wa e ab- so p ion and p ese ing s uc u al in eg i y in aqueous en i onmen s. Howe e , o simpli y end-o -li e managemen and disposal, a mono- ma e ial app oach is p e e ed. Consequen ly, chemical modi ica ion eme ges as a necessa y s a egy o p e en he dispe sion o disin e- g a ion o cellulosic ae ogels and c yogels. To add ess his, a wo-s ep p ocess was implemen ed: TEMPO-oxida ion o s abiliza ion and high-p essu e homogeniza ion o de ib illa ion. This app oach aimed o c ea e ae ogels and c yogels wi h enhanced s abili y and well-de ined po ous s uc u es. TEMPO-media ed oxida ion (TEMPO-NaB -NaClO) is known o gene a e TEMPO-oxidized cellulose nano ib ils, which, when ans o med in o ae ogels and c yogels, exhibi unique he mal conduc i i ies, il a ion beha io s and mechanical p ope ies (Puangsin e al., 2017). Addi ionally, TEMPO-oxidized cellulose coupled wi h high-p essu e homogeniza ion, in con as o he unoxidized coun e - pa s, has p o en o imp o e mechanical p ope ies, inc ease he su ace a ea, allowed be e dispe sion and s abili y o he nano ib ils suspen- sion, p oduces a wide ange o po ous ma e ials, enhance wa e so p ion capaci y (Da pen igny e al., 2020; Lee e al., 2009; Pi ozzi e al., 2023). The e ec i eness o he oxida ion ea men was e alua ed by quan i ying he ca boxyl g oup con en in he ea ed cellulose compa ed o he un ea ed ma e ial. The concen a ion signi ican ly inc eased om 0.11 mmol/g o 0.69 mmol/g a e oxida ion. Addi- ionally, con ocal mic oscopy and ansmission elec on mic oscopy (Fig. S2) con i med he success ul isola ion o cellulose ibe s. In e es - ingly, hemp cellulose only equi ed h ee high-p essu e homogeniza ion passes o achie e he desi ed le el o ib illa ion, compa ed o eigh passes equi ed o ice s aw cellulose (Cab e a-Villamiza , Campano, e al., 2024). This highligh s he ease o oxida ion and de ib illa ion o hemp s em cellulose compa ed o o he biomass sou ces, po en ially educing p ocessing s eps, ene gy consump ion and p oduc ion ime. Using es ablished d ying me hods, he s abilized cellulose suspen- sion was success ully used o p oduce bo h ae ogels and c yogels. Ae - ogel p oduc ion in ol ed a sol en exchange p ocess, whe e wa e was eplaced wi h e hanol o acili a e subsequen dissolu ion and emo al ia SC-CO 2 . To enable his exchange, he oxidized cellulose suspension unde wen c oss-linking wi h ci ic acid, inducing a sol-gel ansi ion. Ci ic acid is a well-known c osslinking agen o a ious cellulose de- i a i e sys ems (Demi i e al., 2008). As p e iously men ioned, he c osslinking concen a ion was 0.5 M. This concen a ion could p oduce s able hyd ogels wi hou b i leness. As p e iously epo ed, i he concen a ion o ci ic acid used was oo low, he b i le hyd ogel migh be suscep ible o shape-changing du ing he demolding phase (Nasu ion e al., 2022). Addi ionally, ci ic acid has p o en o be a non- oxic, non- ca cinogenic, mo e sus ainable al e na i e o adi ional c oss-linke s, enhancing he hyd ogel’s pe o mance and biodeg adabili y. The esul ing s able hyd ogels we e hen manipula ed o sol en exchange, ollowed by SC-CO 2 d ying o he ob ained alcogels (Fig. S3). As shown in Fig. 1A, ae ogels and c yogels exhibi ed dis inc mac o- and mic oscopic ea u es. Ae ogels displayed signi ican ly smoo he su aces compa ed o he i egula su aces obse ed in c yogels. C yo- gels also p esen ed shee -like s uc u es and possessed la ge po es han ae ogels. This aligns wi h p e ious epo s (Buch o ´ a & Bud o a, 2016), demons a ing ha eeze-d ying leads o a shee -like cellulose ne wo k wi h la ge and in e connec ed po es (se e al mic ome e s) due o ice c ys al g ow h du ing wa e eezing. Consequen ly, he po osi y o he ae ogels was expec edly lowe han ha o he c yogels, as i will be shown below. 3.3. Ae ogels and c yogels cha ac e iza ion BET analysis con i med he an icipa ed di e ences in po osi y be- ween c yogels and ae ogels. Due o he la ge po es wi hin hei s uc- u e, c yogels could no e ain he ni ogen gas equi ed o su ace a ea measu emen . F eeze-d ying, a s ep in c yogel p epa a ion, eplica es he sublima ed ice c ys als wi hin he hyd ogel ne wo k, esul ing in a highly po ous s uc u e wi h po es anging om mic ons o hund eds o mic ons (Buch o ´ a e al., 2019). This ice c ys al o ma ion can comp ess he cellulose ib ils, po en ially sac i icing some nanos uc u e in he po e walls. Con e sely, he ae ogels displayed a ypical mesopo ous s uc u e wi h a BET su ace a ea o 295.13 m 2 /g, an a e age po e diame e o 7.17 nm, and an a e age po e wid h o 14.02 nm (BJH adso p ion). No ably, he ob ained po e wid h is signi ican ly smalle compa ed o o he cellulose ae ogels. Fo ins ance, Da pen igny e al. (2020) epo ed an a e age po e size o 24.7 ±10.4 μ m using TEMPO- oxidized cellulose nano ib ils. This di e ence sugges s ha hemp cel- lulose exhibi s unique p ope ies a e oxida ion, con as ing wi h wood pulp cellulose used in ae ogel de elopmen . Scanning elec on Fig. 1. Mac oscopic and mic oscopic cha ac e iza ion o ae ogels and c yogels. Mac oscopic appea ance unde wo di e en magni ica ions cap u ed using mac opho og aphy equipmen . Mic oscopic s uc u e as isualized by scanning elec on mic oscopy (SEM). L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 6 mic oscopy (SEM) images co obo a ed hese obse a ions ega ding he po osi y o he ma e ials (see Fig. 1). Addi ionally, as shown in Fig. S4, he po e dis ibu ion a ea analyzed using SEM images showed a ange o sizes below 900 μ m 2 o c yogels and below 70 μ m 2 o ae ogels. Densi y analysis e ealed no s a is ically signi ican di e ences be ween he ae ogels (5.0 ×10 −4 ±7.0 ×10 −5 g/cm 3 ) and c yogels (4.7 ×10 −4 ±3.0 ×10 −5 g/cm 3 ). Al hough ae ogels exhibi ed sligh ly highe den- si y, his a ia ion was negligible. Following cha ac e iza ion, he comp essi e s eng h o he ma e- ials was e alua ed (Fig. 2A). Ae ogels displayed signi ican ly highe comp essi e s eng h (40.01 kPa a 50 % comp essi e de o ma ion) compa ed o c yogels (19.10 kPa). This di e ence can be a ibu ed o he dense s uc u e and in e nal po e dis ibu ion o ae ogels. In e - es ingly, bo h ae ogels and c yogels exhibi ed 2–3 imes g ea e esis- ance o comp ession compa ed o ca bon ae ogels de i ed om TEMPO-oxidized cellulose nano ib ils (Lai e al., 2021). This supe io pe o mance can be linked o he inhe en lexibili y and comp ession esis ance o cellulose nano ib il ae ogels (Zhu e al., 2022). Addi ion- ally, as p e iously epo ed by Zheng e al. (2020) one o he echnical s a egies o imp o e s a ch-based ae ogel is he use o lignocellulose nano ib ils o inc ease he mechanical s eng h, making he de eloped ma e ials s onge o comp ession in con as o s a ch-based ae ogels. A c ucial ac o o indus ial applica ions is con ol o e ma e ial size and sh inkage du ing p oduc ion and use. The diame e and sh inkage o he ma e ials we e moni o ed h oughou p ocessing (Table 2). Ae ogels exhibi ed minimal dimensional changes in diame e a e gela ion. Howe e , a signi ican educ ion occu ed ollowing SC- CO 2 d ying and dec eased u he upon exposu e o wa e apo so p- ion es s. The ae ogels migh p esen his sh inkage beha io due o capilla y condensa ion and hyd ogen bonding. Thus, wa e apo con- denses in he nanopo es, c ea ing su ace ension o ces ha pull he po e walls oge he (Paulauskiene e al., 2024). In addi ion, he hyd o- philic cellulose nano ib ils o med new hyd ogen bonds wi h he wa e molecules, leading o s uc u al ea angemen and collapse o he po ous ne wo k (Zaman e al., 2020b). Fu he mo e, p e ious s udies epo ed ha he p essu es abo e 13 MPa du ing SC-CO2 d ying could minimize sh inkage in cellulose nano ib ils (Zhu e al., 2022); howe e , in he p esen s udy he used p essu e was 12.1 MPa which may lead o he uns able s uc u e. In consequence, he sh inkage upon wa e exposu e limi s he po en ial applica ion o ae ogels as mea -abso ben pads. In con as , c yogels displayed minimal dimensional changes ac oss Fig. 2. A) Comp essi e s ess-s ain cu es o ae ogels and c yogels. B) FTIR spec a o he ae ogel and c yogel samples. D) De i a i e he mog a ime ic (DTG) cu e depic ing he a e o weigh loss e sus empe a u e. Table 2 Diame e o he ae ogels and c yogels in hei suspension, gelled, d ied, and a e wa e apo so p ion es s a es. a Ma e ial s a e Ae ogels diame e (mm) C yogels diame e (mm) Suspension 15.00 ±0.00 a 15.00 ±0.00 a Geli ied 14.24 ±0.45 a 15.00 ±0.45 a D ied 10.32 ±0.52 b 14.14 ±0.19 a A e wa e apo so p ion 8.00 ±1.04 c 13.79 ±0.20 a a Di e en le e s ep esen ed signi ican di e ences. L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 7 all p ocessing s ages, demons a ing ema kable s abili y h oughou p ocessing and po en ial use. The s abili y o his ma e ial migh be explained because he la ge po es a e less suscep ible o capilla y condensa ion e ec s and a e he eeze-d ying p ocess, s onge in e - ib illa bonds make he s uc u e mo e esis an o ea angemen (Zhang e al., 2015). The e o e, c yogels may be be e sui ed o mea - abso ben pads applica ion. Addi ionally, ano he impo an ac o du ing he de elopmen o abso ben pads is undoub edly hei capaci y o so b and e ain wa e . The e o e, he wa e apo so p ion capaci y was used o e alua e he capaci y o sa u a ion and wa e so p ion wa e om he closed 100 % RH en i onmen . The ae ogels e ained 1.86 ±0.10 mL/g, and he c yogels e ained 2.20 ±0.004 mL/g a e equilib ium (W e ). The esul s showed signi ican di e ences and illus a ed ha c yogels we e capable o so bing mo e wa e han ae ogels, hus, making hem mo e sui able o so p ion applica ions. Addi ionally, o he ood cellulosic ae ogels showed ha wa e apo so p ion capaci y was signi ican ly lowe han he one epo ed in he p esen s udy due o he use o c ys alline cel- lulose (Ciu a in e al., 2023). The e o e, he use o cellulose ibe s con aining bo h c ys alline and amo phous egions inc eases he wa e apo so p ion capaci y o he ma e ials. The FTIR and TGA analyses o he ma e ials co esponded wi h he composi ional analysis o he hemp cellulose, showing he ypical lignocellulosic ma e ial beha io . Pa icula ly, FTIR analysis e ealed sub le a ia ions in he chemical bonding p o ile o he ae ogels and c yogels (Fig. 2B). In he 3700–3000 cm −1 egion, bo h samples exhibi ed a b oad peak assigned o he in amolecula hyd ogen bonding wi hin cellulose (Cichosz & Masek, 2020). Addi ionally, a peak a 2900 cm −1 , co esponding o C – H s e ching in a oma ic me hyl g oups, was iden i ied in bo h ma e ials, indica ing he p esence o cellulose and hemicellulose. The peak ypically obse ed in hemp ibe s a 1756 cm −1 , indica i e o hemicellulose con en , was minimal due o he educ ion ollowing ex ac ion and p ocessing (Vˆ a ban e al., 2021). Fu he mo e, he absence o a peak a 1509–1508 cm −1 sugges ed low lignin con en in bo h samples. No ably, his peak was e en lowe in he ae ogels, po en ially due o lignin ex ac ion du ing SC-CO 2 d ying, as p e iously epo ed by Yang e al. (2018) and ˇ Su ka e al. (2013). In e es ingly, he 1428–1400 cm −1 egion displayed dis inc peaks cha ac e is ic o c ys- alline cellulose. Simila ly, peaks a 899–894 cm −1 , as epo ed by Vˆ a ban e al. (2021) con i med he p esence o glycosidic bonds and he amo phous cellulose egion in bo h samples. These peaks we e mainly p esen ed in c yogels, which would allow he e en ion o mo e wa e du ing he wa e apo so p ion es . Addi ionally, he mog a ime ic analysis e ealed a h ee-s ep deg ada ion p ocess o ae ogels and c yogels. As shown in Fig. 2C, he ae ogel exhibi ed supe io he mal s abili y. The deg ada ion peaks o he ae ogel occu ed a 57.7 ◦C, 317.73 ◦C, and 416.25 ◦C. In con as , he c yogels exhibi ed lowe deg ada ion empe a u es a 56.78 ◦C, 285.77 ◦C, and 400.73 ◦C. These esul s a e compa able o he indings epo ed by Mijas e al. (2021) o hemp ibe deg ada ion. B ie ly, he i s peak obse ed be ween oom empe a u e and 100 ◦C co esponds o he emo al o mois u e and ola ile compounds. The second deg ada ion s ep, occu ing a ound 270–290 ◦C, is a ibu ed o he decomposi ion o low- o-medium mo- lecula weigh hemicelluloses and lignin. Finally, he deg ada ion peak a 390–400 ◦C is associa ed wi h he decomposi ion o cellulose and medium- o-high molecula weigh lignin (Mijas e al., 2021; Rachini e al., 2009). TGA e ealed g ea e he mal s abili y o ae ogels compa ed o c yogels. This obse a ion likely co ela es wi h a educ- ion in lignin-like compounds and hei de i a i es wi hin he ae ogels. SC-CO 2 d ying po en ially emo es hese moie ies due o hei solubili y in bo h CO 2 and e hanol. Consequen ly, he d ying p ocess emo es hese he mally labile componen s om he ae ogels, con ibu ing o hei enhanced he mal s abili y. Among he in es iga ed ma e ials, c yogels displayed he mos a o able cha ac e is ics o he eplacemen o adi ional plas ic abso ben pads used in ood packaging. They exhibi ed minimal dimensional changes and p ese ed s uc u al in eg i y h oughou p ocessing and subsequen wa e apo so p ion assays. No ably, c yo- gels demons a ed supe io liquid up ake compa ed o ae ogels. These indings sugges consis en wa e so p ion capaci y and app op ia e s uc u al s abili y in c yogels. Owing o hese ad an ageous p ope ies, c yogels we e chosen o u he e alua ion by imp egna ion wi h a ice s aw ex ac . 3.4. Bioac i e c yogels To in es iga e po en ial s uc u al di e ences a ec ing ma e ial unc ionali y, he c yogels we e cha ac e ized ollowing he inco po a- ion o ice s aw ex ac and eeze-d ying. This cha ac e iza ion included mac o and mic oscopic analyses, wa e apo so p ion capac- i y, wa e so p ion by imme sion and deso p ion capaci y, e alua ion o he an ioxidan capaci y, comp essibili y, TGA and FTIR. The mac oscopic cha ac e iza ion e ealed concen a ion-dependen changes in he colo a ion o he ma e ials ollowing ex ac addi ion, as shown in Fig. 3A. All he ea men s displayed homogeneous su ace mo phology a e eeze-d ying. The mic oscopic analysis demons a ed ha he po e s uc u e and ibe dis ibu ion sligh ly di e ed om he imp egna ed c yogels in con as o he p is ine c yogels. Co espond- ingly, SEM analysis con i med ha he inco po a ion o he ex ac a 187.5 mg/g did no a ec he mic os uc u e o he ma e ial (Fig. 3B). No ably, he consis en colo change ac oss he en i e ma e ial sugges ed ho ough imp egna ion wi h minimal ex ac agg ega ion on he ibe su ace. These indings indica e success ul ex ac inco po a ion p io o eeze-d ying, wi h minimal impac on he mac o and mic oscopic mo phology o he de eloped ma e ials. In addi ion, he isualiza ion o he su ace wi hou agg ega es migh sugges ha he inco po a ion p io o eeze d ying did no equi e any c osslinking agen , as equi ed in p e ious s udies whe e cellulose and lignin we e compa ibilized by Fe 3+ (Sanchez e al., 2023). The an ioxidan ac i i y de e mined by β-ca o ene bleaching o 2 mg/mL o he ice s aw ex ac was de e - mined o be 47 ±2.5 %. Co espondingly, Ghasemzadeh e al. (2015) epo ed simila an ioxidan alues (47.23 %) a e e hanol-wa e ex- ac ions in ice b an. Addi ionally, he ice s aw ex ac used in he p esen s udy was p e iously cha ac e ized by Cab e a-Villamiza , Eb ahimi, e al. (2024) whe e he an ioxidan capaci y was 77.53 mg T olox equi alen s (TE)/g and he polyphenol con en was 69.20 Equi alen mg Gallic acid/g a e he alkaline ex ac ion. Subsequen ly, he p ese a ion o he unc ionali y o he ice s aw ex ac wi hin he ma e ial was e alua ed. The esul s, shown in Table 3, demons a ed ha , a bo h es ed concen a ions, 75 mg/g and 187.5 mg/g, he ma e ials showed a high an ioxidan ac i i y. The an ioxidan p ope ies we e a ibu ed o he p esence o se e al p e iously epo ed bioac i e compounds such as gallic acid, ca eic acid, py ogallol, la- onoids, phenolic compounds (p-couma ic acid and p o oca echuic acid) (Menzel e al., 2020). Pa icula ly, a e alkaline ex ac ions om ice s aw he main componen s ounded whe e e ulic acid, p-couma ic acid, anillic acid, sy ingaldehyde, p-hyd oxybenzaldehyde and anillin (Li e al., 2015). Simila ly, Wu, Wu, e al. (2023) ha e demons a ed ha he g a ing o p-couma ic acid wi h cellulose nanoc ys als keep he an ioxidan ac i i y a e chemical in e ac ion. Fu he mo e, a posi i e co ela ion was obse ed be ween he inco po a ed an ioxidan amoun and he measu ed ac i i y, indica ing a dose-dependen e ec . Con e sely, he p is ine c yogels (wi hou ex ac s) did no exhibi an ioxidan ac i i y. Fu he mo e, when compa ing o o he cellulosic c yogels imp egna ed wi h an ioxidan ex ac s (36.8–57.3 %) (Fon es- Candia e al., 2019), he de eloped ma e ials showed highe an ioxidan ac i i y du ing he β-ca o ene bleaching assays. The c yogels wi h and wi hou he ice s aw ex ac e ained he s uc u al in eg i y a e being imme sed in wa e , as shown in Fig. 3C. The ma e ials showed a s able s uc u e despi e hei hyd ophilic in insic cha ac e and he inco po a ion o he ex ac ha could in e e e wi h he o med hyd ogen bonds be ween he cellulosic ibe s. L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 8 This poin s ou ha he inco po a ion o he ice s aw ex ac did no cause any al e a ion in he in eg i y o he ma e ial. An e alua ion o he comp essibili y o he ma e ials e ealed ha imp egna ed ma e ials exhibi ed supe io mechanical p ope ies compa ed o he p is ine c yogel. As illus a ed in Fig. S5, a 60 % comp essi e de o ma ion, he c yogels and imp egna ed ma e ials displayed s ess esis ances o 25.46 kPa and 44.56 kPa, espec i ely. Simila ly, p elimina y s udies ha e demons a ed ha he inco po a ion o lignin de i a i es no only enhances he mechanical p ope ies bu also con ibu es o he s abili y o cellulose ae ogels in an aqueous medium (Wang e al., 2023). Addi ionally, he inco po a ion o he ice s aw educed he wa e apo so p ion capaci y (WVSC) o 1.08 ±0.10 mL/g (1.12 mL/g was educed). Co espondingly, he abso p ion and deso p ion capaci y o he c yogels e alua ed by imme sion showed a simila endency. C yo- gels exhibi ed supe io wa e so p ion capaci y compa ed o he com- me cial pad, as shown in Fig. 4A. Punc ually, a hei equilib ium, hey can abso b up o 140.71 ±4.87 mL o wa e pe g am o c yogel, which is 4.8 imes g ea e han he comme cial pad’s capaci y (29.36 ±0.30 mL/g). No ably, no signi ican di e ences in so p ion capaci y we e obse ed be ween he p is ine c yogel (140.71 ±4.87 mL/g) and he one imp egna ed wi h 75 mg/g o he ex ac (149.52 ±4.41 mL/g). Howe e , inc easing he an ioxidan concen a ion o 187.5 mg/g esul ed in a dec ease in so p ion capaci y (134.66 ±0.20 mL/g). This can be a ibu ed o he p esence o hyd ophobic lignin de i a i es wi hin he ice s aw ex ac , which ha e hyd ophobic p ope ies, hus mi iga ing wa e so p ion in he ma e ial. This inding is consis en wi h p e ious in o ma ion epo ed by Ma iana e al. (2021), who showed ha once lignin was added o o he biopolyme s, hey became mo e hyd ophobic and he ma e ials p esen ed lowe wa e so p ion han he biopolyme s wi hou any ein o cemen in a high-humidi y en i on- men . Addi ionally, he hemp ma e ials showed a slowe sa u a ion a e han comme cial pads bu showed g ea e so p ion capaci y. The c yo- gels exhibi ed deso p ion capaci y, eleasing 95 % o he e ained liquid wi hin he i s 0.5 h, as shown in Fig. 4B. Con e sely, he comme cial pad e ained mois u e and did no achie e comple e deso p ion. This di e ence a ises om he mul ilaye s uc u e o he comme cial pad. I possesses a po ous hyd ophobic su ace, acili a ing so p ion h ough he po es. Howe e , he emaining non-po ous su ace a ea aps he liquid wi hin he ma e ial. Fu u e s udies could in es iga e inco po- a ing a simila p o ec i e sys em on o he c yogels o delay hei deso p ion capaci y. This sys em could in ol e a biopolyme hyd o- phobic laye ha egula es he so p ion a e using he same mechanism, such as he one p oposed by Beni o-Gonz´ alez e al. (2020) whe e cellulosic c yogels we e hyd ophobized wi h poly (lac ic acid) making hem sui able o wa e and oil so p ion. Finally, as shown in Fig. 4C, he he mog a ime ic analysis esul s showed a h ee-s ep deg ada ion p ocess. Howe e , he inco po a ion o he ex ac shi s he empe a u e o he second deg ada ion peak o lowe empe a u es, indica ing ha he inco po a ion o he ex ac sligh ly dec eased he he mal s abili y o he ma e ial. The Fig. 3. A) Mac oscopic and op ical mic oscopy images o he c yogels be o e and a e imp egna ion. B) Scanning Elec on Mic oscopy (SEM) images o imp egna ed ma e ial 187.5 mg/g. C) E alua ion o ma e ial in eg i y a e imme sion in dis illed wa e o 2 h a 22 ◦C. Table 3 An ioxidan ac i i y o c yogels con aining ice s aw ex ac e alu- a ed by β-ca o ene bleaching me hod. b T ea men An ioxidan ac i i y % C yogel 0.01 ±0.25 a C yogel 75 mg/g 63.89 ±5.38 b C yogel 187.5 mg/g 78.96 ±1.84 c BHT 98.03 ±0.41 d b Di e en le e s ep esen ed signi ican di e ences. L. Cab e a-Villamiza e al. Ca bohyd a e Polyme s 348 (2025) 122887 9