E ec s o he g een c oss-linking agen annic acid and i s oxida ion on he
p ope ies o po cine plasma p o ein supe abso ben ma e ials
Massimo Alagia
a
, Ca los Bengoechea
b,*
, Ba ba a La Fe la
c
, F ancesco Pe i
d
,
An onio Gue e o
b
a
Depa men o Ma e ials Sciences, Uni e si y o Milano-Bicocca, Via R. Cozzi 55, 20125 Milano, I aly
b
Depa amen o de Ingenie ía Química, Uni e sidad de Se illa, Escuela Poli ´
ecnica Supe io , 41011 Se illa, Spain
c
Depa men o Ea h and En i onmen al Sciences DISAT, Uni e si `
a degli S udi di Milano-Bicocca, Piazza della Scienza 1, 20126 Milan, I aly
d
Depa men o Bio echnology and Biosciences, Uni e si y o Milano-Bicocca, Piazza della Scienza 2, 20126 Milan, I aly
ARTICLE INFO
Keywo ds:
Po cine plasma p o ein
Tannic acid
C oss-linking
ABSTRACT
Tannic acid is a na u al polyphenol capable o s ongly in e ac ing wi h p o eins, wi h good an ioxidan and
an ibac e ial p ope ies. Thus, annic acid (TA) o oxidized annic acid (oxTA) may be used as c oss-linking
agen s in he de elopmen o ein o ced and ully p o ein-based supe abso ben ma e ials (SAMs). oxTA was
p oduced so ha eac i e quinone g oups we e gene a ed, which a e expec ed o inc ease i s eac i i y. In his
s udy, po cine plasma p o ein (PPP) and glyce ol (gly) we e used in a 50/50 PPP/gly a io o ob ain SAMs
h ough win sc ew mixing and injec ion molding. The esul s showed ha bo h TA and oxTA inc eased he
s o age modulus and he loss angen o blends and bioplas ics due o he physical in e ac ions es ablished be-
ween TA o oxTA and PPP. The mechanical p ope ies, pa icula ly he Young's modulus and ensile s eng h,
we e gene ally enhanced as well. Wa e abso p ion was s ongly in luenced by he addi ion o TA, esul ing in a
dec ease in he amoun o wa e abso bed. Howe e , samples con aining oxTA esul ed in a g ea e wa e ab-
so p ion capaci y, e aining a highe p opo ion o he supe abso ben p ope ies o he e e ence composi ion.
Mo eo e , sys ems con aining oxTA gene ally possess be e mechanical p ope ies han hose o equi alen TA
o mula ions, especially hose con aining 5 % and 10 % oxTA.
1. In oduc ion
Cu en ly, en i onmen al conce ns a e d i ing esea ch in o no el
ma e ials owa ds a sus ainable de elopmen , whe e he u iliza ion o
na u al p oduc s and ea u es such as biodeg adabili y and biocompa -
ibili y ha e become essen ial. One o hem includes supe abso ben
polyme s (SAPs), also e med as supe abso ben ma e ials (SAMs),
which a e highly po ous, c oss-linked polyme s wi h hyd ophilic do-
mains capable o abso bing and e aining high weigh pe cen ages o
wa e [1–4].
These s uc u es include highly hyd ophilic moie ies, such as ca -
boxyla es, hyd oxides and amides, which can in e ac wi h wa e mol-
ecules causing he s e ching o polyme ic chains and esul ing in SAM
swelling. P esen ly, con en ional SAMs include ei he syn he ic poly-
me s, such as polyac yla es and polyac ylamides, o semisyn he ic bio-
polyme s, which a e ypically based on s a ch, chi osan o cellulose
c oss-linked wi h ac ylic eagen s [5]. Al hough all o hese ma e ials
ha e shown g ea pe o mance, he e is s ill a need o de elop ully
biobased supe abso ben ma e ials, whe e each componen used in he
o mula ion can be de i ed om biomass, which would ep esen a
sus ainable de elopmen o his applica ion [6]. Among o he s, p o ein-
based bioplas ics ha e ecen ly been exploi ed. Gene ally speaking,
bioplas ics a e plas ics ha a e biobased, biodeg adable, o bo h, and
a e being s udied in a a ie y o applica ions, anging om ood pack-
aging [7,8], au omo i e [9], cosme ic and pe sonal ca e packaging [10].
In pa icula , p o ein-based bioplas ics ul il he equi emen s o bio-
SAMs and ha e he ad an age ha hey can be ea ed wi h con en-
ional polyme p ocessing echniques, including win-sc ew mixing and
injec ion molding, by using a sui able plas icize [11]. Mo eo e , he
wa e abso p ion o hese bioplas ics is gene ally good due o he g ea
hyd ophilici y o bo h biopolyme (i.e., p o ein) and plas icize s (e.g.,
glyce ol, so bi ol). In his sense, se e al p o eins, including whea glu en
[12], po a o p o ein [13], soy p o ein [14], ke a in [15] and po cine
plasma p o ein (PPP), ha e ecen ly been conside ed.
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (C. Bengoechea).
Con en s lis s a ailable a ScienceDi ec
In e na ional Jou nal o Biological Mac omolecules
jou nal homepage: www.else ie .com/loca e/ijbiomac
h ps://doi.o g/10.1016/j.ijbiomac.2025.140584
Recei ed 11 Oc obe 2024; Recei ed in e ised o m 24 Janua y 2025; Accep ed 31 Janua y 2025
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
A ailable online 6 Feb ua y 2025
0141-8130/© 2025 The Au ho s. Published by Else ie B.V. 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/ ).
PPP is usually ob ained om aw blood ia a plasma ac iona ion
p ocess, which ini ially in ol es he cen i uga ion o po cine blood,
which allows he plasma o be sepa a ed om he cellula componen s,
which emain in he sedimen . A e ha , plasma p o eins can e en u-
ally be sepa a ed on he basis o hei molecula weigh ia o he ech-
niques, including p ecipi a ion o memb ane il a ion. Indeed, plasma
p o ein is composed by se e al di e en ac ions, namely ib inogen,
albumin and globulin. The wo mos abundan sub ac ions a e albumins
(60 %) and globulins (40 %), whe e he o me a e subdi ided in o
α
-, β-
and γ- sub ac ions o 66–69 kDa, and he la e a e also dis inguished
in o
α
-, β- and γ- sub ac ions wi h a wide molecula weigh dis ibu ion
[16,17]. PPP and mo e gene ally blood can be conside ed as unde used
ma e ials, e en i hey p esen aluable p ope ies such as wa e -
binding, he mal gela ion, oaming and emulsi ying abili ies. Fo hose
easons, i has been implemen ed o ood and eed addi i e applica-
ions. None heless, hose p ope ies a e also in e es ing in non- ood
applica ions. Fo example, PPP has been conside ed o he p oduc ion
o ilms [18,19] and elec ospun nano ib ous memb anes [20].
Fu he mo e, he abundance o po cine plasma p o ein as a
byp oduc o he mea indus y p esen s an oppo uni y o sus ainable
u iliza ion and was e educ ion. I is di icul o ob ain ce ain da a abou
po cine plasma p o ein p oduc ion wo ldwide, hus making es ima ion
necessa y. Acco ding o FAO, in 2021, he wo ld p oduc ion o mea pig
was se o 120,372 housand onnes [21], hus abou 1.65 billion pigs a e
slaugh e ed annually (1 pig can gene a e 73 kg o mea on a e age). As
one pig can gene a e 3.5 L o blood [17], annual po cine blood p o-
duc ion can be es ima ed as 5.77 billion li es. Conside ing ha he
plasma ac ion is 60 % o whole blood, 3.46 billion L o plasma a e
p oduced wi h a p o ein con en o app oxima ely 87 g/L concen a ion,
inally i can be es ima ed ha annual wo ld p oduc ion o po cine
plasma p o ein co esponds o app oxima ely 301 housand onnes,
conside ing an ideal si ua ion in which no mass loss akes place du ing
he PPP p oduc ion p ocess. The use o po cine plasma p o ein in he
p oduc ion o bioplas ics o SAMs hus akes ad an age o i s ema k-
able p ope ies, as well as i s biodeg adabili y, biocompa ibili y and
abundance as a by-p oduc o he mea indus y.
P e ious wo k sugges ed ha i is possible o ob ain SAMs om PPP
using glyce ol as a plas icize [22]. This ma e ial has a wa e abso p ion
capaci y up o 15 o 20 imes i s o iginal weigh , bu has he disad an-
age o being so and b i le, which could limi i s applicabili y.
The e o e, se e al a emp s ha e been made o imp o e i s mechanical
p ope ies. A common s a egy in ol es he addi ion o a c oss-linking
agen , which can physically o chemically in e ac wi h he poly-
pep ide chains o es ic hei ela i e mo emen when a o ce is
applied [23,24]. Un o una ely, c oss-linking also ends o educe he
wa e up ake capaci y [25]. Among p o ein c oss-linke s, annic acid
(TA) is gaining inc easing in e es . In pa icula , annic acid is a plan
de i ed polyphenolic molecule ha can be ound in ba k, lea es, ui s
and seeds, and is capable o in e ac ing s ongly wi h p o eins, mainly
h ough py ogallol moie ies, o o m high molecula weigh complexes
[26–28]. Addi ionally, TA p esen s o he in e es ing p ope ies, such as
an ioxidan and an ibac e ial p ope ies. Tannic acid can in e ac ia
hyd ogen bonds be ween i s phenolic hyd oxyl g oups and he pola
moie ies o a p o ein, including ca bonyls, amines and hyd oxyls.
Al hough i is no ye clea whe he annic acid is able o chemically bind
p o ein, one op ion o inc ease ha eac i i y is oxida ion. This la e
de i a iza ion is gene ally ca ied ou in aqueous solu ion in he p es-
ence o ac i e oxygen species and can be pe o med bo h by chemically
and enzyma ically me hods. The oxida ion o TA gene a es eac i e
quinone g oups which p esumably may be able o inc ease eac i i y
owa ds PPP ia ei he Schi o Michael base addi ions, as illus a ed in
Fig. 1. Tannin oxida ion can gene ally be accompanied by successi e
Fig. 1. Tannic acid s uc u e (A) and sugges ed mechanisms o c oss-linking wi h p o eins ia p io oxida ion (B).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
2
coupling and polyme iza ion, which leads o inc eased s uc u al
complexi y.
TA, ei he unoxidized o oxidized, has been used o c osslink
di e en biopolyme s, such as casein, chi osan, gela in o zein [29–31],
mainly p ocessed h ough cas ing o hyd ogel o ma ion. Howe e , o
he bes o ou knowledge, nei he he use o TA as c oss-linke o PPP o
he p oduc ion o TA-c osslinked bioplas ics h ough injec ion molding
has ye been epo ed. Injec ion molding is a common p ocess in he
plas ic indus y ha pe mi s o o m p ecisely complex geome ies wi h
high ep oducibili y. Injec ion molding can be scaled and au oma ed o
a wide ange o ma e ials [32]. The speci ic objec i e o his s udy is o
e alua e he e ec o he chemical oxida ion o TA on he p ope ies o
PPP based injec ion molded bioplas ics.
In his wo k, SAM bioplas ics we e p oduced by mixing po cine
plasma p o ein and TA o oxTA in a win sc ew mixe , using glyce ol as
plas icize , o p oduce blends which we e hen injec ion molded in o
bioplas ics. In pa icula , h ee di e en concen a ions o ei he TA o
oxTA (5 %, 10 % and 20 % weigh pe cen ages wi h espec o p o ein
con en ) we e p epa ed, and hei e ec s on he he momechanical,
ensile and abso p ion p ope ies we e e alua ed.
2. Ma e ials and me hods
2.1. Ma e ials
The po cine plasma p o ein (PPP) used in his s udy (Ap oPo k,
Essen ia P o ein, USA) was kindly supplied by PROANDA S.A. (Spain).
The p o ein (74 w %) and ash (9 w %) con en s o his concen a e we e
de e mined in p e ious wo ks [22,33]. Fo all o mula ions, pha ma
g ade glyce ol (gly) (CAS 56-81-5, Pan eac Química S.A., Spain) and
ACS-g ade annic acid (CAS 1401-55-4, Sigma-Ald ich, Me ck, Ge -
many) we e used. O he eagen s, including Na
2
CO
3
, CAS 497-19-8,
egene a ed cellulose dialysis memb anes, H
2
O
2
solu ion (CAS 7722-
84-1) and Folin-Ciocal eu eagen (CAS 12111-13-6) we e ob ained
om Sigma-Ald ich (Me ck, Ge many).
2.2. Tannic acid oxida ion
The oxida ion o annic acid was ca ied ou ollowing a epo ed
p ocedu e, wi h li le a ia ion [29]. Tannic acid was solubilized in
dis illed wa e a a concen a ion o 0.02 g/mL; a his poin , he em-
pe a u e was inc eased o 60 ◦C and 5 M NaOH was added un il pH
eached 9–10. Finally, an aqueous solu ion o H
2
O
2
(35 w %) was added
o each inal concen a ion o hyd ogen pe oxide equal o 0.4 w %. The
mix u e was allowed o eac o 30 min a 60 ◦C, and neu alized by
adding 1 M HCl and oxidized annic acid powde was ob ained a e
eeze d ying o 3 days wi h a LyoQues eeze d ye (Tels a Tech-
nologies, Spain), a a collec o empe a u e o −80 ◦C.
This oxida ion p ocess was ca ied ou a leas h ee imes and
employed in e e y o mula ion o check he ep oducibili y o he esul s
ob ained.
2.3. Samples p epa a ion
The ab ica ion o bioplas ics ollowed a wo-s age p ocess epo ed
elsewhe e [34], wi h modi ica ions.
The p ocedu e comp ises a i s mixing s ep wi h a Haake Polylab
mixing heome e (The mo-Scien i ic, Ge many). Each componen was
blended a oom empe a u e o 5 min and a a o o speed o 50 pm in
a double cylind ical chambe . To que and empe a u e a e cons an ly
moni o ed du ing his p ocess. The blends we e hen s o ed o 1 day a
4 ◦C and subsequen ly p ocessed h ough injec ion molding o ob ain
ec angula samples (60 mm ×10 mm ×1 mm). These samples we e
ob ained wi h Haake pneuma ic pis on injec o , Mini Je II (The mo-
Scien i ic, Ge many). The blends we e in oduced in o a chambe hea ed
o 40 ◦C and injec ed h ough a nozzle in o a mold whose empe a u e
was se a 60 ◦C. These condi ions ha e p o en o p o ide a ema kable
supe abso ben cha ac e (i.e. abso bing wa e up o 21 ±4 imes i s d y
weigh ) bu leading o poo mechanical p ope ies (i.e. wi h a Young
modulus o 2.4 ±0.8 MPa) [22]. The p essu e was main ained a 800
ba o he en i e du a ion o he p ocess (300 s). Samples we e kep in a
sealed con aine be o e hei cha ac e iza ion, which was ca ied ou in
he 24–48 h pe iod a e hei ab ica ion (Table 1).
2.4. Cha ac e iza ion
2.4.1. Dynamic mechanical analysis (DMA)
The iscoelas ic p ope ies o blends and bioplas ic we e de e mined
h ough small ampli ude oscilla o y es s using a DMA850 heome e
(TA Ins umen s, USA). Blends we e analysed wi h a comp ession clamp
geome y immedia ely a e hei p epa a ion, whe eas bioplas ics we e
es ed in ensile mode wi hin wo days a e hei molding. The expe i-
men s in ol ed a equency sweep es anging om 0.01 Hz o 10 Hz
and we e pe o med a 20 ◦C and cons an s ain (0.01 % blends, 0.005
% o bioplas ics). Mo eo e , empe a u e sweep es s we e pe o med a
a hea ing a e o 5 ◦C/min, a equency o 1 Hz and s ains o 0.01 % and
0.005 % o he blends and bioplas ics, espec i ely. The empe a u e
was a ied om 0 ◦C o 140 ◦C o blends and om −30 ◦C o 160 ◦C o
bioplas ics. All he assays we e pe o med wi hin he linea iscoelas ic
egion (LVE), p e iously de e mined o each composi ion wi h s ain
sweep es s (0.001 %10 %) a a cons an equency o 1 Hz. Da a we e
collec ed ia TRIOS so wa e and a leas h ee measu emen s pe
sample we e eco ded, e en i only one ep esen a i e measu emen pe
sample was epo ed.
2.4.2. Tensile es s
The samples ob ained om injec ion molding wi h ec angula
shapes we e submi ed o uniaxial ensile es s, acco ding o he s an-
da d me hod ISO527-2, wi h some modi ica ions [35]. Mechanical
p ope ies such as ensile s eng h, elonga ion a b eak and Young's
modulus we e es ima ed using an MTS Insigh 10 Uni e sal Tes ing
Machine wi h a load cell o 10 kN. A e he specimen was ixed in o he
suppo ing clamps, a displacemen a cons an a e o 10 mm/min was
applied un il eaching ailu e, so ha he ime elapsed o all he ex-
pe imen s was be ween 30 and 300 s in acco dance wi h he s anda d.
This es was conduc ed on a leas 6 eplica es pe sample. Da a we e
inally eco ded, analysed wi h Tes Wo ks4 so wa e and he Thompson-
Tau s a is ical es o ou lie sea ching was applied o he esul s wi hin
one sample.
2.4.3. Fou ie - ans o med in a ed spec oscopy (FTIR)
IR spec a we e eco ded in he 4000–400 cm
−1
wa enumbe ange
wi h 4 cm
−1
esolu ion and 32 scans using a Jasco FTIR 4200 spec-
ome e (Eas on, MD, USA) equipped wi h an a enua ed o al e lec-
ance (ATR). Bioplas ic samples we e analysed di ec ly wi h he ATR
ins umen in he ansmi ance mode. The da a we e hen con e ed o
abso bances and no malized o e he amide I egion (1700–1600 cm
−1
)
o each sample. The esul ing spec a a e epo ed in no malized
abso bance uni s e sus wa enumbe s.
2.4.4. Wa e up ake capaci y (WUC) and soluble ma e loss (SML)
WUC and SML o bioplas ics we e es ima ed acco ding o he p o-
ocol epo ed in a p e ious wo k [36]. B ie ly, pieces o 20 mm ×10
Table 1
Fo mula ions s udied.
Sys em PPP Gly TA/oxTA
REF 50 50 0
TA5/oxTA5 48.8 48.8 2.4
TA10/oxTA10 47.6 47.6 4.8
TA20/oxTA20 45.5 45.5 9.0
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
3
mm ×1 mm in size we e cu om he bioplas ics, kep o 24 h in an
o en a a cons an empe a u e o 50 ◦C and weighed (w
0
). The samples
we e hen placed in wa e (app oxima ely 80 mL) o 24 h in closed
essels, weakly bound wa e was emo ed by gen le d ying wi h il e
pape , and swollen samples we e weighed (w
1
). A e a inal d ying s ep
in an o en (50 ◦C, 24 h), he esidual bioplas ics we e weighed (w
2
).
WUC and SML we e calcula ed as ollows:
WUC (%) = w1−w2
w2
×100 (1)
SML (%) = w0−w2
w0
×100 (2)
2.4.5. Scanning elec on mic oscopy (SEM)
Swollen bioplas ics in wa e we e eeze-d ied o 1 day, cu in o
small pieces and analysed wi h a ZEISS EVO mic oscope (USA) a e
sil e coa ing. A beam cu en o 11–12 pA and an accele a ing ol age
o 10 kV we e used o image acquisi ion. Images a e acqui ed wi h a
seconda y elec on (SE) senso a 100×magni ica ion.
2.4.6. The mog a ime ic analysis (TGA)
TGA on bioplas ics (20–50 mg) was pe o med be ween 50 ◦C and
1000 ◦C wi h cons an hea ing a e o 10 ◦C/min, unde ai a mosphe e
in oduced a 100 mL/min wi h a he mog a ime ic analyse , model
SDT Q600 V20.9 Build 20 (TA Ins umen s (USA)). The empe a u e
amp also included a p e- ea men as ollows: om 30 ◦C o 130 ◦C,
20 ◦C/min, N
2
; iso he m a 130 ◦C o 5 min, N
2
; om 130 ◦C o 50 ◦C,
20 ◦C/min, N
2
; iso he m a 50 ◦C o 5 min; N
2
.
2.4.7. De e mina ion o bound phenolic species and he Folin-Ciocal eu
assay
A o al o 150 mg o d ied bioplas ics (24 h in an o en a 50 ◦C) was
dissol ed in 5 mL o o mic acid unde magne ic s i ing o 16 h. Once
comple ely dissol ed, wa e was added o ob ain a inal o mic acid
concen a ion o app oxima ely 50 %. The mix u e was hen pu i ied
h ough dialysis (12–14 kDa MWCO) agains dis illed wa e o 2 days.
Wa e was con inuously eplaced by esh dis illed wa e du ing he
dialysis p ocess o p omo e he pu i ica ion. A e dialysis, he solu ion
was e ie ed and eeze d ied. The samples we e submi ed o FTIR
analysis as p e iously desc ibed. The ob ained powde was also analysed
ia he Folin-Ciocal eu assay, ollowing a epo ed p ocedu e, wi h
modi ica ions. B ie ly, sample solu ions a a 2 mg/mL concen a ion in
dis illed wa e we e p epa ed. Then, 40
μ
L o hese solu ions we e mixed
wi h 800
μ
L o 75 mg/mL Na
2
CO
3
and dilu ed wi h 1.08 mL o wa e .
Finally, 40
μ
L o Folin-Ciocal eau eagen (2 N) we e added and mix u es
(2 mL) we e kep in he da k o 30 min, a e which he abso bance was
ead a 750 nm and co ela ed wi h he phenol con en in he samples. A
calib a ion cu e wi h annic acid o oxidized annic acid (0.05–0.25
Fig. 2. F equency sweep es s o blends wi h di e en concen a ions o annic acid (A); and wi h di e en concen a ions o oxidized annic acid (B). Tempe a u e
sweep o blends wi h di e en concen a ions o annic acid (C); and wi h di e en concen a ions o oxidized annic acid (D).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
4
mg/mL) was p e iously de e mined o i s quan i ica ion in unknown
samples. Measu emen s we e ca ied ou wi h an Asys UVM 340
Mic opla e eade (Bioch om, Camb idge, UK) using G eine Bio-One
Cells a 24-well pla es.
3. Resul s and discussion
3.1. Viscoelas ic p ope ies o blends
A e he mixing s age, he blends ob ained we e subjec ed o dy-
namic mechanical analysis in comp ession mode in o de o e alua e he
dependence o he iscoelas ic p ope ies on he equency and em-
pe a u e. The equency sweep es o he blends (Fig. 2A) e ealed ha
he s o age modulus can be co ela ed wi h he amoun o annic acid in
he sample; in pa icula , inc easing he TA con en esul s in an inc ease
in he elas ic modulus, E', o he blends e en be o e he injec ion
molding p ocess, by which chemical c oss-linking may be igge ed. I is
known ha TA can in e ac wi h p o eins ia hyd ogen bonding be ween
phenolic OH g oups and he ee amine, hyd oxyl and ca bonyl g oups o
a p o ein, as well as hyd ophobic in e ac ions among a oma ic ings and
elec os a ic o ces. Addi ionally, he in e play o TA wi h glyce ol could
also occu h ough hyd ogen bonding. All o hese in e ac ions esul in
he o ma ion o mo e igid s uc u es wi h espec o e e ence. Mo e-
o e , in oducing a hi d componen in he blends also esul s in an in-
c ease in he loss angen , an δ, de ined as he a io o he iscous
modulus o he elas ic modulus (E"/E'), which indica es ha , e en i a
s eng hening akes place, he p e alence o E' o e E" becomes less
impo an in he p esence o TA, indica ing ha physical in e ac ions a e
p edominan .
Fo he oxidized annic acid-con aining blends, a simila co ela ion
be ween he oxTA con en and hei iscoelas ic p ope ies was obse ed
(Fig. 2B). This la e , compa ed wi h he non-oxidized compounds, a e
enden ially cha ac e ized by E' p o iles se a highe alues, excep o
oxTA10 sample. This is also accompanied by a lowe dependency o an
δ wi h equency, hus indica ing mo e igid s uc u es. No ably, his
e ec should no be a ibu ed o a chemical c oss-linking in he blends,
since o que and empe a u e p o iles in he mixing s age (Supplemen-
a y in o ma ion, Fig. S1) do no show a clea e olu ion o any o hose
pa ame e s wi h mixing ime, bu a he o s onge hyd ogen bonding
due o he p esence o ca bonyl a he han hyd oxyl g oups in he
chemical s uc u e o he c oss-linke .
The esul s o he empe a u e sweep es s on he blends a e shown in
Fig. 2C-D. The e e ence composi ion no con aining any c oss-linking
agen (REF) displays a i s he mal so ening e ec , un il i eaches a
minimum E' alue a 65 ◦C, ollowed by an inc ease wi h empe a u e.
Indeed, he peak a a ound 82 ◦C in an δ p o ile can be a ibu ed o he
he mal dena u a ion o plasma p o eins, especially albumins, a e
which E' inc eases due o p o ein agg ega ion [37]. DMA measu emen s
clea ly e ealed ha TA (Fig. 3A) and oxTA (Fig. 3B) ha e an in luence
on he he mal ansi ions o he ma e ial. Fi s , a peak in an δ a 63 ◦C
appea s o TA-con aining sys ems, accompanied wi h a p onounced
dec ease in E': his ansi ion can be associa ed wi h he disagg ega ion
o annic acid in e ac ions wi h an inc ease in monome ic TA [38].
Tannic acid can in e ac wi h i sel ia hyd ogen-bonding o he galloyl
moie ies o o m sup amolecula s uc u es. This peak is clea ly
de ec able in he loss angen cu e o TA5, whe eas o TA10 and TA20
he e is an o e lap wi h he adjacen peak. I is also easonable o hink
ha in ha empe a u e ange a chemical eac ion be ween TA and PPP
Fig. 3. F equency sweep o bioplas ics wi h di e en concen a ions o annic acid (A); and wi h di e en concen a ions o oxidized annic acid (B). Tempe a u e
sweep o bioplas ics wi h di e en concen a ions o annic acid (C); and wi h di e en concen a ions o oxidized annic acid (D).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
5
can occu , om which one would expec an inc ease in he elas ic
modulus. Ne e heless, a clea inc ease in E' is no de ec ed be o e he
ansi ion, p obably due o he p edominance o he he mal so ening
e ec o plasma p o eins, bu i is possible o highligh he shi o E'
minimum owa ds highe empe a u es. A high empe a u es, he E'
p o iles in Fig. 2C seem o e lec a comp omise be ween inc easing TA
con en and dec easing p o ein con en .
On he o he hand, oxidized annic acid shows a mo e complex
he mal beha iou , comp ising o a shoulde peak a abou 40 ◦C, a
b oad peak o annic acid disagg ega ion a 60 ◦C and PPP gela ion
abo e 80 ◦C. In pa icula , his las peak is shi ed owa ds highe
empe a u es in p opo ion o he amoun o oxTA added; his shi is
also accompanied by a shi in he E' minimum owa ds highe em-
pe a u es. Yan e al. epo ed a simila e ec o TA on egg whi e p o ein/
xan han gum sys ems om di e en ial scanning calo ime y (DSC)
analysis and a ibu ed his e ec o he o ma ion o an hea - esis an
ne wo k a e he eac ion [39]. I is possible ha oxTA, owing o i s
highe eac i i y, gi es ise o a mo e p onounced a ia ion in he
ansi ion o he ma e ial. Thus, a high empe a u es, he E' cu es in
Fig. 2D displays an in e ed o de ; his phenomenon is in acco dance
wi h he dec ease in p o ein con en in he c oss-linke con aining
blends.
3.2. Bioplas ic iscoelas ic p ope ies
As desc ibed in he expe imen al sec ion, bioplas ics we e ob ained
a e he injec ion molding p ocess (T
cyl
=40 ◦C and T
mold
=60 ◦C o
500 s a 800 ba ). The dynamic mechanical p ope ies we e e alua ed in
ensile mode by equency and empe a u e sweep es s. The equency
sweep esul s e ealed ha all he samples had inc eased elas ic moduli
wi h espec o hei co esponding blends a e he molding p ocess, as
obse ed when compa ing Fig. 3A-B (bioplas ics) wi h Fig. 2A-B
(blends). This can be a ibu ed o he in luence o empe a u e and
p essu e applied du ing he injec ion molding p ocess.
The equency sweep es s o blends and bioplas ics a e in acco -
dance ega ds he e ec s o inc easing concen a ions o TA and oxTA on
he dynamic mechanical p ope ies o bioplas ics. As shown in Fig. 3A,
unoxidized annic acid displays a concen a ion-dependen inc ease in
s o age modulus, whe eas oxidized annic acid has a g ea e e ec a
lowe concen a ions (5 %), no obse ing an inc ease in i s con en wi h
highe oxTA con en s. Gene ally, o all he samples, c oss-linking has
he e ec o inc easing bo h E' and an δ, jus like obse ed in blends,
excep o he TA5 sample. The slopes o he E' cu es we e no signi -
ican ly lowe han hose o he co esponding blends, p obably due o
he pe sis ence o a high numbe o hyd ogen bonds ha annic acid can
o m wi h all he componen s o he o mula ion. These ela i ely weak
bonds a e mo e suscep ible o a cons an de o ma ion. O he au ho s
ound simila esul s when employing TA as c oss-linke in casein ilms,
ob aining highe iscoelas ic moduli a g ea e TA con en s [31].
The esul s o he empe a u e sweep es s on he bioplas ics a e
displayed in Fig. 3C-D. Simila ly wi h blends, E' dependence on em-
pe a u e comp ises a i s egion o E' lowe ing un il he plasma he mal
ansi ion, ollowed by a subsequen inc ease a highe empe a u es.
Addi ionally, when compa ing he empe a u e sweep es s o blends
Fig. 4. S ess-s ain cu es o TA (A) and oxTA (B) samples. Tensile s eng h (
σ
MAX
), elonga ion a b eak (
ε
MAX
) and modulus (E) o he whole samples (C).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
6
(Fig. 2C-D) and bioplas ics (Fig. 3C-D), he e is a gene al ein o cemen
e ec due o he injec ion molding p ocessing condi ions. As he ein-
o cemen e ec ela ed o annin c oss-linke s was also isible be o e
injec ion molding, i is easonable o belie e ha c oss-linking did no
occu in his la e s ep. Addi ionally, in his case, he E' minimum is
sligh ly shi ed owa ds highe empe a u es (especially o oxTA-
con aining samples), sugges ing a possible e ec o he c oss-linke on
he bioplas ic he mal ansi ion. The an δ p o ile o bioplas ics changes
when he c oss-linke is p esen , ei he TA o oxTA: as al eady desc ibed
o blends, an addi ional peak con ibu ion a app oxima ely 60 ◦C can
be de ec ed when he c oss-linke has been added, he in ensi y o which
inc eases p opo ionally wi h i s con en .
3.3. Tensile mechanical p ope ies
In addi ion o DMTA, ensile es ing is a eliable es o measu ing
he s eng h o bioplas ics. Fig. 4 shows he ensile s ess s. s ain cu es
o each se o samples, which a e co ela ed wi h he da a alues o
ele an pa ame e s, in pa icula , ensile s eng h (
σ
max
), elonga ion a
b eak (
ε
max
), and Young's modulus (E). Compa ed wi h he e e ence
o mula ion wi hou a c oss-linke , he addi ion o TA o oxTa gene ally
imp o ed he ensile mechanical p ope ies o he bioplas ics; howe e ,
he e we e la ge di e ences be ween he cu es ela ed o he ein o ced
samples. In pa icula , o he nonoxidized annic acid samples, he e is a
co ela ion be ween TA con en and E, which inc eases by a ac o o up
o 5 a a TA concen a ion o 20 % (TA20); The imp o emen can also be
seen in he
σ
max
and
ε
max
, he la e being pa icula ly no iceable o 5 %
TA (TA5), which exhibi s a alue o
ε
max
inc eased by a ac o o abou
1.5. Howe e , his pa ame e dec eases signi ican ly wi h inc easing TA
con en . Thus, he gene al end o an inc ease in Young's modulus and
maximum s ess a he expense o a educ ion in he maximum elonga-
ional s ain is obse ed om 5 % AT onwa ds. On he o he hand, he
oxidized annic acid-con aining composi ions did no show a clea co -
ela ion be ween he mechanical p ope ies and oxTA con en . In
pa icula ,
σ
MAX
ini ially inc eases o he sys em con aining 5 % o oxTA
and hen dec eases wi h inc easing concen a ion o oxTA. In con as ,
pa ame e E always emains abo e he alue ob ained o he e e ence
while no signi ican a ia ions a e obse ed o
ε
MAX
, excep a he
highes oxTA concen a ion, whe e a signi ican ly lowe alue han ha
o he e e ence can be obse ed. A possible explana ion o his e ec
could be a ibu ed o he di e en s uc u es o he wo c oss-linke s; in
pa icula , unoxidized annic acid p esen s high amoun s o phenolic
Fig. 5. FTIR o bioplas ics wi h di e en concen a ions o annic acid (A); and wi h di e en concen a ions o oxidized annic acid (B).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
7
g oups which a e able o in e ac wi h p o ein and glyce ol by hyd ogen
bonding; he oxidized TA on he o he hand, has o hoquinone g oups,
which a e mo e suscep ible o chemical eac ions as depic ed in Fig. 1B.
In addi ion, unlike phenolic OH, o-quinones a e hyd ogen bond accep-
o s, al e ing he H-bond in e ac ions ha occu in hese ma e ials. I
could be easonable o hink ha oxTA could p e e en ially in e ac wi h
glyce ol, which is mainly an H-bond dono . The a ia ion o mechanical
p ope ies wi h TA concen a ion pa ially ag ees wi h he esul s ob-
ained om heological es s, showing ha c oss-linking inc eases
in e chain in e ac ions and es ic s hei mobili y, bu majo di e ences
be ween hese wo analyses can be obse ed especially o oxTA-
con aining samples. E en i he FTIR echnique employed in his s udy
has al eady been used p e iously o quali a i ely desc ibe he in-
e ac ions be ween di e en componen s in a blend and he ex en o
c osslinking [23,40], addi ional expe imen s in ol ing NMR spec os-
copy would be help ul o con i m hose in e ac ions be ween oxTA and
glyce ol, as well as o quan i y he ex en o bonding wi h PPP. Mo e-
o e , elemen al mapping h ough echniques like ene gy-dispe si e X-
ay spec oscopy (EDS) would aide o ack unc ional g oup dis ibu ion
eac i i y. Fu he esea ch should be ocus on his.
3.4. S uc u al cha ac e iza ion h ough FTIR
S uc u al cha ac e iza ion o he bioplas ics was ca ied ou o
de e mine he possible a ia ion in p o ein chemical s uc u e due o
bond o ma ion wi h TA/oxTA. F om he FTIR spec a (Fig. 5A-B) i is
possible o ecognize se e al cha ac e is ic p o ein IR bands, especially
amine N
–
H asymme ic s e ching, e e ed o as he amide A band
(≈3300–3100 cm
−1
), alipha ic CH s e ching asymme ic C
–
–
O
s e ching o he amide g oup, amide I band (1700–1600 cm
−1
) and
N
–
H bending wi h con ibu ions om C
–
N s e ching, amide II
(1580–1480 cm
−1
). Some o he bands can be a ibu ed o glyce ol,
especially he O
–
H asymme ic s e ching (3600–3200 cm
−1
), which
p oduces a shoulde in he p o ein amide A band, and sha p peaks a
1100, 1035 and 995 cm
−1
, ela ed o CO s e ching. Supplemen a y
Fig. S2 shows he FTIR spec a o TA and oxTA eagen s, whe e pa ic-
ula ly C-OC a oma ic s e ching and C-O-H a oma ic bending a
app oxima ely 1311 cm
−1
, es e O-CO s e ching a 1180 cm
−1
and C-O-
C asymme ic s e ching a 1018 cm
−1
can be dis inguished. This bands
a e sligh ly shi ed a highe wa enumbe s o oxTA, whe eas he shi a
lowe wa enumbe s o O
–
H s e ching can be ela ed wi h s onge
hyd ogen bonding due o he o ma ion o ca bonyls in he oxida ion, as
p e iously epo ed [41]. Upon he addi ion o annic acid, he in ensi y
o he p o eins bands dec eased p opo ionally o he amoun o TA in
he bioplas ic, which was expec ed as he p o ein con en p og essi ely
dec eased. The bioplas ics con aining ei he TA o oxTA di e wi h
espec o he peaks o he c oss-linke alone a peak cen ed a 1346
cm
−1
appea s in all he samples. I migh he e o e be sugges ed ha his
peak o igina es om C
–
N s e ching associa ed wi h he o ma ion o an
a oma ic seconda y amine ia Michael- ype addi ion be ween PPP and
TA/oxTA, as desc ibed in Fig. 1. Ne e heless, his la e could also
o igina e om hyd ogen bonding be ween he phenolic g oups and he
pola esidues o PPP, causing a shi in he O
–
H bending equency.
O he di e ences can be seen in he O-CO es e s e ching, which shows
a shi a highe wa enumbe s (1207 cm
−1
) wi h espec o i gin TA. C-
O-C asymme ic s e ching is no clea ly dis inguishable because i
o e laps wi h he CO s e ching o glyce ol. The ela i e in ensi ies o
hese wo peaks in he bioplas ics a e bo h co ela ed wi h he amoun o
TA con ained in he o mula ions. Wi h espec o he o he mechanism
o eac ion (Schi -base addi ion) he o ma ion o seconda y imines is
di icul o obse e in he spec a, as a peak ela ed o C
–
–
N should be
expec ed in he 1680–1630 cm
−1
egion, which o e laps wi h he amide
I p o ein signal. Indeed, he e is no u he peak con ibu ion o he
amide I signal om he peak decon olu ion in he Supplemen a y In-
o ma ion o he samples (Supplemen a y Fig. S3). I could hus be
hypo hesized ha c oss-linking ac s p edominan ly ia Michael ype
addi ion o p o ein amines om lysine and N- e minus wi h he a oma ic
ing o TA. A compa ison o he TA and oxTA samples e ealed ew
di e ences in he spec a, hus sugges ing small di e ences in he
chemical s uc u e o he bioplas ics p epa ed wi h he wo di e en
c oss-linke s. One possible explana ion is ha he di e ences obse ed
in he mechanical analysis a e no ela ed o di e ences in he chemical
eac i i y o he c oss-linke , bu a he o changes in he physical in-
e ac ions o med be ween he annic acid o oxidized annic acid and
he o he componen s o he o mula ion.
3.5. Wa e up ake capaci y and soluble ma e loss
The abso p ion p ope ies o he bioplas ics a e s ongly in luenced
by he p esence o ei he TA o oxTA, as shown in Fig. 6. In gene al,
so e ma e ials a e expec ed o be associa ed wi h highe WUC, as a
mo e lexible sample is mo e likely o de o m as i swells du ing wa e
abso p ion. This is e iden om he esul s shown, as he unc oss-linked
e e ence has he highes alue o WUC. The addi ion o he wo c oss-
linking agen s leads o a lowe wa e abso p ion: his is expec ed as
he c oss-linke limi s he mobili y o he polypep ide chains, hus hin-
de ing he wa e swelling phenomenon. This e ec has also been
obse ed o gela in ilms ha dened wi h TA [42]. In his wo k, he e-
sidual wa e o he up ake es p esen s a ligh b own colo o TA and
da k b own/g eenish o oxTA, which a e no obse ed in he e e ence
esidual wa e . I is hus epo ed ha hose solu ion colo s can be
a ibu ed o he p esence o oxidized annic acid. Fo hese easons, i is
belie ed ha a ac ion o annic acid is dissol ed om he bioplas ic
Fig. 6. WUC and SML o bioplas ics wi h di e en concen a ions o annic acid
and oxidized annic acid.
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
8
du ing wa e swelling es s.
Di e ences depending on whe he unoxidized o oxidized TA is used
in he o mula ion we e obse ed. In de ail, unoxidized TA educes he
WUC in a way p opo ional o i s concen a ion, while oxidized TA
samples show simila alues and gene ally highe alues han hose o
he equi alen composi ions wi h unoxidized TA. The compa ison be-
ween he TA10 and oxTA10 samples e eals ha he la e has a WUC
alue ha is app oxima ely double ha o he o me ; his is also he
case o he compa ison be ween TA20 and oxTA20. The WUC alue o
TA5 is s ill highe han ha o he co esponding o mula ion oxTA5,
p obably due o he la ge di e ences in mechanical p ope ies be ween
hese wo composi ions. Fo he oxTA samples, a c oss-linke concen-
a ions o 5 % and 10 %, he ma e ials s ill exhibi ed supe abso ben
p ope ies and, in pa icula , he inc ease in concen a ion was no
accompanied by a ne dec ease in WUC.
In e ms o soluble ma e loss, he e a e small di e ences be ween
he samples. The only signi ican dec ease in SML occu s a he highes
oxTA concen a ions (10 and 20 %), sugges ing a highe le el o oxTA
in e ac ions wi h he o he wo componen s. I should also be no ed ha
an inc ease in TA o oxTA con en implies a educ ion in PPP and
glyce ol con en , which may con ibu e o he lack o signi ican di -
e ences o he o he sys ems. In ac , i is sugges ed ha hese wo
componen s a e mainly ela ed o he soluble loss du ing wa e swelling,
as seen in he alue o SML o he e e ence sample.
3.6. Scanning elec on mic oscopy
The mo phology o swollen bioplas ics can be examined by scanning
elec on mic oscopy (Fig. 7). The e ec o wa e up ake on he sample
su ace in ol es he gene a ion o po es associa ed wi h glyce ol di u-
sion in o aqueous media associa ed wi h wa e up ake and e en ion.
Indeed, e e ence sample a e wa e abso p ion and eeze d ying p e-
sen s a su ace wi h honeycomb-like s uc u e, which was obse ed also
o dex an–me hac yla e swollen hyd ogel [43]. This s uc u e is
o med ollowing eeze d ying and is ela ed o he o ma ion o ice
c ys alli e, which phase sepa a ed om he ma ix and in oduce
po osi y [44]. The addi ion o a c oss-linke in luences he mo phology
o he su aces, which is e lec ed by a ia ions in wa e up ake p op-
e ies. Especially, po e dimensions o he c osslinked samples di e om
hose o he e e ence sample, as well as hei shapes, which appea o be
mo e ci cula -like, and he numbe o po es seems o dec ease. The
po osi y densi y is co ela ed wi h he wa e up ake capaci y, as mo e
compac s uc u es like he c oss-linked ones could hinde he pene a-
ion o wa e molecules inside he ma e ial, leading o an in e io
numbe o po es [45]. Mo eo e , he in oduc ion o TA/oxTA may
gene a e inhomogenei ies and oughe s uc u es due o agg ega ion,
which a e belie ed o be embedded wi h mino WUC espec o mo e
homogeneous ones like he e e ence [46]. In he SEM mic og aphs in
Fig. 7, i is no clea whe he he e is a concen a ion-dependen e ec o
TA o oxTA on he mo phology o he su ace, bu a he he si ua ions
appea o be simila .
3.7. The mog a ime ic analysis
The he mal s abili y o bioplas ics was de e mined by means o
he mog a ime ic analysis (TGA) in an ai a mosphe e o s udy he
combus ion phenomena. The esul s a e shown in Fig. 8 o samples
s udied. A hea ing p e- un up o 130 ◦C was made, so ha he wa e loss
was no isible in he TGA cu es shown. The i s decomposi ion s ep
ook place a an onse o app oxima ely 175 ◦C o he e e ence sample,
which did no signi ican ly change upon he addi ion o he c oss-linke .
The i s weigh loss s ep could be a ibu ed o he i s oxida i e
Fig. 7. SEM images o eeze d ied swollen ma ices o po cine plasma wi h di e en concen a ions o annic acid (TA) and oxidized annic acid (oxTA).
M. Alagia e al.
In e na ional Jou nal o Biological Mac omolecules 304 (2025) 140584
9
