Accu acy assessmen o a mic o-Raman spec oscopy me hod o small
mic oplas ic pa icles in in an milk o mula
Ma a Pu zu
a,b
, Ma a Ba ba esi
c
, Ma a Fadda
a
, Alessio Sacco
a
, Mau izio Pie gio anni
c
,
Ma eo Masino
c
, Fede ica Bianchi
c
, Ko inna Al mann
d
, Niza Benismail
e
, Lau een Coïc
e
,
I ana Fenoglio
b
, Monica Ma a ozzi
c
, And ea Ma io Rossi
a
, Ma ia Ca e i
c
,
And ea Ma io Gio annozzi
a,*
a
Is i u o Nazionale di Rice ca Me ologica (INRiM), S ada delle Cacce 91, Tu in 10135, I aly
b
Uni e si y o Tu in (UniTO), Depa men o Chemis y, Via Pie o Giu ia 7, Tu in 10125, I aly
c
Uni e si y o Pa ma, Depa men o Chemis y, Li e Sciences and En i onmen al Sus ainabili y, Pa co A ea delle Scienze 17/A, Pa ma 43124, I aly
d
Bundesans al ü Ma e ial o schung und -p ü ung (BAM), Un e den Eichen 87, Be lin 12205, Ge many
e
Nes l´
e Quali y Assu ance Cen e Vi el (NQAC Vi el), A enue Geo ges Clemenceau 1020, Vi el Cedex 88804, F ance
ARTICLE INFO
Keywo ds:
Mic oplas ics
Re e ence ma e ials
Polye hylene e eph hala e
In an milk o mula
Raman spec oscopy
In e labo a o y compa ison
ABSTRACT
The p esence o mic oplas ics (MPs) in he ood chain is inc easingly documen ed, aising conce ns o e po en ial
isks o human heal h. Despi e g owing e o s, s anda dized me hods o MPs de ec ion in ood ma ices emain
limi ed. This s udy p esen s an in e labo a o y compa ison (ILC) aimed a assessing he accu acy and compa-
abili y o an analy ical app oach o he iden i ica ion and quan i ica ion o small MPs (5–100 µm) in in an milk
powde using µ-Raman spec oscopy and a ep esen a i e polye hylene e eph hala e (PET) e e ence ma e ial
(RM). The RM, o mula ed as wa e -soluble able s, was designed o eplica e he mo phology, size dis ibu ion,
and polyme composi ion o en i onmen ally ele an MPs, and was p e iously assessed o homogenei y and
s abili y o mass ac ion and pa icle numbe s.
The app oach was assessed using wo PET RM ba ches wi h di e en MPs pa icle numbe s (high load ba ch:
1759 ±141 MPs; low load ba ch: 160 ±22 MPs), subjec ed o an enzyma ic–chemical diges ion, ollowed by
µ-Raman analysis pe o med independen ly in wo labo a o ies wi h di e en ins umen s and ope a o s. Resul s
a e epo ed as absolu e pa icle coun s pe analyzed sample and demons a ed excellen eco e y ac oss all size
classes, including he smalles pa icles (down o 5 µm), wi h eco e y a es anging om 82 % o 88 %, in good
ag eemen wi h he RM e e ence alues.
The analy ical app oach p o ed o be obus , ep oducible, and sui able o low-le el MPs quan i ica ion in
complex ood ma ices, suppo ing ongoing e o s owa d me hod ha moniza ion and s anda diza ion o eli-
able MPs moni o ing in he ood sec o .
1. In oduc ion
Plas ic wi h i s chemical and physical esis ance, low cos and ease o
p oduc ion, has become widely used ac oss a ious applica ions, wi h
global p oduc ion inc easing s eadily o e he yea s and eaching 413.8
M in 2023 [1]. Howe e , his e olu iona y ma e ial has slowly become
a global en i onmen al h ea , wi h la ge quan i ies o used plas ics
ei he land illed o end up in aqua ic and e es ial ecosys ems h ough
imp ope disposal. Plas ic i ems exposed o en i onmen al ac o s such
as ocean cu en dynamics, sola adia ion, mechanical ab asion, and
in e ac ions wi h o ganisms, slowly deg ade and agmen in o smalle
plas ic pa icles, known as mic oplas ics (MPs) [2–5]. So a , he e is no
ha moniza ion ega ding he size h esholds used o classi y MPs. Re-
po ed alues ange om 1 - 20 µm a he lowe size [6] and om 500 µm
up o 1 mm, o e en 5 mm, a he uppe size [7]. Acco ding o ISO
24187:2023 [8], MPs a e de ined as solid plas ic pa icle insoluble in
wa e in he size ange o 1 µm – 1000 µm. These small pa icles a e o
pa icula conce n due o hei abili y o be inges ed by o ganisms and
ans e ed ac oss ophic le els, po en ially posing isks o human
heal h [9,10]. They ha e been iden i ied in se e al ood p oduc s,
* Co esponding au ho .
E-mail add esses: [email p o ec ed] (M. Pu zu), [email p o ec ed] (A.M. Gio annozzi).
Con en s lis s a ailable a ScienceDi ec
Talan a Open
jou nal homepage: www.sciencedi ec .com/jou nal/ alan a-open
h ps://doi.o g/10.1016/j. alo.2025.100586
Recei ed 16 Sep embe 2025; Recei ed in e ised o m 27 Oc obe 2025; Accep ed 29 Oc obe 2025
Talan a Open 12 (2025) 100586
A ailable online 31 Oc obe 2025
2666-8319/© 2025 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (
h p://c ea i ecommons.o g/licenses/by/4.0/ ).
including ish [11], ui s and ege ables [12], mea [13], d inking
wa e [14], be e ages (ene gy and so d inks) [15] and dai y p oduc s,
like milk [16–18], al hough he sou ces o con amina ion emain un-
clea (e.g. en i onmen al sou ces o ood p ocessing, s o age, ans-
po a ion, and packaging) [19].
Milk and i s de i a i es a e an essen ial componen o heal hy human
nu i ion, wi h global consump ion s eadily ising o e he pas decades
o nea ly 981 MT in 2024 [20]. As demand con inues o g ow, ensu ing
he sa e y o milk has become mo e c i ical han e e . All ac i i ies
de eloped du ing he p oduc ion p ocess o milk, om a ms o he dai y
indus y, could be a isk o con amina ion by MPs (e.g. poo cleanness
p ocedu e equipmen , su ounding en i onmen , wa e supply condi-
ions, inadequa e handling o milk and packaging) [21]. Despi e
g owing conce n, a limi ed numbe o s udies ha e in es iga ed MPs
con amina ion in milk, pa icula ly in in an milk powde [22,23],
ocusing on he de ec ion o MPs in di e en comme cially a ailable
liquid milk b ands o assess human exposu e h ough o al inges ion [16,
24–29]. The majo i y o hese wo ks used isual iden i ica ion and
coun ing o MPs unde a mic oscope ollowed by chemical cha ac e -
iza ion using A enua ed To al Re lec ion Fou ie T ans o m In a ed
Spec oscopy (ATR-FTIR) [24–28] o , in ew cases, Raman spec oscopy
[16,23,30,31] o a combina ion o bo h echniques [29] o a compa -
a i e quali a i e s udy [32]. Due o he size limi a ions o op ical mi-
c oscopy, he epo ed MPs a e gene ally in he millime e ange, hus
excluding smalle MPs (<100 µm), which a e mo e ele an om a
biological and oxicological pe spec i e. Only h ee s udies ocused on
milk powde [17,27,23], wo o which examined in an o mula [22,23].
Zhang e al. [22] de ec ed MPs in milk using ATR-FTIR in a s udy aimed
a in es iga ing o he i s ime MPs pollu ion in in an o mula (MPs
abundance om 1 ±1 o 11 ±1 uni s/100 g, wi h a e age sizes om
139 ±343 µm o 193 ±415 µm). Kadac-Czapska e al. [23] in oduced
me hod alida ion using comme cial plas ic s anda d pa icles ha
di e in e ms o polyme ype, size, shape and colo , including poly-
amide (PA: i egula shaped agmen s, 1 mm), polye hylene (PE:
sphe es, 30 - 50 µm;), polyp opylene (PP: g anules, 2 mm), polye hylene
e eph hala e (PET: g anules, 2 mm). I should be no ed ha MPs do no
adequa ely ep esen he complexi y o eal MPs ound in ma ices in
e ms o shape and size. In con as , Da Cos a Filho e al. [17] used mo e
ep esen a i e size s anda ds (5 - 40 µm) o se e al polyme s (e.g. PP,
PE, PS, PA, PMMA) al hough PET, one o he mos widesp ead en i-
onmen al polyme s, was no included. This la e s udy also add esses
con amina ion mo e ho oughly by analyzing blank samples o e mul-
iple days and epo ing an a e age backg ound alue, which was hen
sub ac ed om he esul s o eal sample analysis. Howe e , i is wo h
no ing ha backg ound sub ac ion in MPs analysis is ques ionable,
gi en he high a iabili y and pe asi e MPs con amina ion ac oss
labo a o y en i onmen s [27,28]. Fu he mo e, no s a is ically de ined
epo ing limi (RL) o equi alen limi o quan i ica ion (LOQ) was
es ablished – an issue common o all ci ed s udies, whe e MPs in blanks
a e simply iden i ied and excluded om he sample esul s wi hou
applying any quan i a i e h eshold. The e o e, c i ical me hodological
limi a ions a e widesp ead h oughou he exis ing li e a u e. In his
con ex , he de elopmen o e icien and eliable s anda dized me hods
emains a c i ical challenge o MPs moni o ing and is u gen ly equi ed
o suppo EU legisla ion (e.g. D inking Wa e Di ec i e) and ensu e
eliabili y and compa abili y o da a o heal h and en i onmen al isk
assessmen s. This need has al eady been pa ially add essed in he wa e
quali y e alua ion: he In e na ional O ganiza ion o S anda diza ion
(ISO) body ISO/TC 147/SC 2/JWG1 de eloped in e na ional s anda ds
o he iden i ica ion and quan i ica ion o MPs in wa e by
numbe -based spec oscopy (ISO 16094–2:2025) [35] and mass-based
he mo-analy ical de ec ion me hods (ISO/DIS 16094–3) [36]. Re e -
ence ma e ials (RMs) a e necessa y o alida e each analy ical me hod
and o ensu e compa abili y and ha moniza ion ac oss MPs measu e-
men s, bu hey a e cu en ly lacking. The main challenge is p oducing
RMs ha ealis ically mimic MPs ound in en i onmen al and ood
samples in e ms o i egula shape, su ace cha ac e is ics and size
dis ibu ion [37]. ISO 17034:2016 [38] speci ies he gene al e-
qui emen s o he compe ence and consis en ope a ion o RM p o-
duce s, co e ing he p oduc ion o all RMs, including ce i ied RMs, and
ensu ing ha ma e ials a e p oduced unde con olled condi ions wi h
aceable p ope ies. Acco ding o ISO 33405:2024 [39], hese ma e ials
mus be homogeneous and s able in ela ion o one p ope y o in e es
(e.g. mass o pa icle numbe ) and su icien ly ep esen a i e o be used
in me ological s udies, including he de elopmen o accu a e analy -
ical me hods, calib a ion o measu emen sys ems, and quali y con ol
es ing.
This wo k p esen s an in e labo a o y compa ison (ILC) aimed a
assessing he accu acy and compa abili y o µ-Raman measu emen s o
he eliable iden i ica ion and quan i ica ion o small mic oplas ics
(SMPs) in in an milk o mula, using PET e e ence ma e ials. The RM
was p oduced in wo ba ches wi h di e en loadings (high and low)
wi hin he amewo k o he Eu opean p ojec 21GRD07 Plas icT ace
[40], bo h es ed o homogenei y and s abili y in e ms o bo h mass
ac ion and pa icle numbe s [38,39,41] Fo he p esen wo k, he low
load ba ch was p ima ily conside ed, as i be e ep esen s ealis ic ood
con amina ion scena ios. PET was selec ed as ep esen a i e polyme
due o i s widesp ead use in ood packaging [42,43] and o assess i s
limi ed s abili y unde ha sh oxida i e condi ions du ing sample p ep-
a a ion (e.g., empe a u es abo e i s glass ansi ion and s ong alkaline
en i onmen s) [44,45] his was c ucial o e alua e he impac o sample
p epa a ion on he eco e y o smalle pa icle size ac ions. The
analy ical app oach combines an app op ia e sample ea men based on
enzyma ic and chemical diges ion, wi h pa icle cha ac e iza ion ca ied
ou by mic o-Raman (µ-Raman) spec oscopy, chosen o i s high spa ial
esolu ion and sui abili y o de ec ing SMPs. µ-Raman analysis was
pe o med in an in e labo a o y compa ison in ol ing wo independen
labo a o ies wi h di e en ope a o s and ins umen s. Bo h labo a o ies
ollowed he same sample p epa a ion p o ocol o ensu e ep oduc-
ibili y, compa abili y, and ha moniza ion o he measu emen s.
2. Ma e ials and me hods
2.1. Ma e ials
Ul apu e wa e was ob ained by a Milli-Q® IQ 7000 pu i ica ion
sys em (Me ck Millipo e, Ge many) equipped wi h a 220 nm poly-
e he sul one (PES) il e . This wa e was used o cleaning p ocedu es
and sample p epa a ion. In an powde ed milk (O ganic 1 Fi s In an
Baby Milk Powde ) was bough om a local supe ma ke and selec ed as
he es ma ix.
PET RM in he o m o wa e -soluble able s we e p oduced and
p o ided by Bundesans al ü Ma e ial o schung und-p ü ung (BAM,
Be lin, Ge many).
In addi ion, he ollowing eagen s we e used du ing sample p epa-
a ion: a mul i-enzyma ic de e gen (P ozyme Ac i e Deconex, Bo e
Chemie AG, Swi ze land); e hylenediamine e aace ic acid sodium sal
(EDTA-Na 0.5 M, pH 8, In i ogen, The moFishe Scien i ic, USA); e -
ame hylammonium hyd oxide (TMAH 25 % / Sigma-Ald ich, USA);
sodium hyd oxide (NaOH 2.5 M, PanReac AppliChem ITW Reagen s,
Spain); ni ic acid (HNO
3
5 % / Ca lo E ba Reagen s GmbH,
Ge many).
E hanol absolu e anhyd ous (99.9 %, Ca lo E ba Reagen s GmbH,
Ge many), T i on™ X-100 (0.1 % / in ul apu e wa e , The mo Fishe
Scien i ic, USA) and ace one (99.9 % / , Ca lo E ba Reagen s GmbH,
Ge many) we e also used o he cleaning p ocedu es.
2.2. PET RM able s o me hod pe o mance assessmen
PET RM able s con aining PET pa icles in he size ange o 5–100
µm, wi h sha p edges and i egula mo phology, we e p o ided by BAM
in wo di e en loadings: a high load ba ch (HLB) and a low load ba ch
M. Pu zu e al.
Talan a Open 12 (2025) 100586
2
(LLB). These di e en loadings we e es ed o e alua e he sensi i i y
and eco e y o bo h en i onmen al-like concen a ions (HLB) and ood-
ela ed concen a ions (LLB) de ec able by numbe -based spec oscopic
me hods. Pa icula a en ion was gi en o LLB, as i s pa icle numbe is
mo e ep esen a i e o he le els po en ially ound in ood ma ices.
Bo h ba ches we e p e iously e alua ed o homogenei y and s abili y
o pa icle numbe acco ding o ISO 17034:2016 [38] and ISO
33405:2024 [39], wi hin he amewo k o he Eu opean p ojec
21GRD07 Plas icT ace [40,41], and we e ound o ul il he c i e ia o
RMs. De ailed in o ma ion on able dissolu ion, il a ion, and cha ac-
e iza ion a e desc ibed in a sepa a e manusc ip p epa ed wi hin he
21GRD07 Plas icT ace p ojec (unde e iew) [46].
These PET RM able s we e used o es ablish e e ence pa icle
numbe s o assessing he analy ical me hod pe o mance in milk sam-
ples, and he alues we e exp essed as absolu e pa icle coun s pe
able .
2.3. In an milk powde sample p epa a ion
The milk diges ion p o ocol o MPs ex ac ion o iginally de eloped
by Da Cos a Filho e al., [17] was adap ed wi h sligh modi ica ions o
imp o e he emo al o o ganic ma e ial while p ese ing he in eg i y
o chemically sensi i e polyme s, such as PET, enabling he de elop-
men o a new me hod o eliable MPs de ec ion in complex ma ices.
The same p o ocol was applied in pa allel by wo independen labo a-
o ies, INRiM and Uni e si y o Pa ma, o ensu e compa able esul s.
Powde milk was econs i u ed wi h ul apu e wa e (14 % w/ ) and
shaken a 40 ◦C o 15 min in a shaking wa e ba h (Ul asonic Cleane
DU-45, 180 W). Meanwhile, a PET RM able was dissol ed in 20 mL o
ul apu e wa e in a glass lask. A e he comple e dissolu ion o he
able , 25 mL o econs i u ed milk was added o he solu ion and hea ed
o 40 ◦C while s i ing o pe o m he diges ion p ocess. The milk sample
was subjec ed o mul i-enzyma ic diges ion by adding and s i ing 2 mL
o mul i-enzyma ic de e gen (P ozyme) o 2 min, ollowed by he
addi ion o 10 mL o calcium chela ing agen sodium e hylene diamine
e a ace a e (EDTA-Na 0.5 M, pH 8) o 3 min. Subsequen ly, 2 mL o
alkaline solu ion e ame hyl ammonium hyd oxide (TMAH 25 % / )
we e added unde con inuous s i ing, and he empe a u e was aised o
80 ◦C o ini ia e ho alkaline hyd olysis.
Upon eaching 80 ◦C, he ho diges ed milk was immedia ely ans-
e ed o a il a ion uni (100 mL unnel, glass holde wi h 13 mm i ed
suppo ) and il e ed unde acuum (N816 LABOPORT/100 W) using a
ci cula silicon (Si) il e wi h a po e size o 5 µm (mac opo us silicon
memb ane, 9 mm diame e , Sma Memb ane) wi h an e ec i e ci cula
il a ion a ea o app oxima ely 20 mm
2
(~ 5 mm diame e ).
Be o e il a ion, he unnel was condi ioned by insing wi h 15 mL o
T i on™ X-100 (0.1 % / in ul apu e wa e ) and 15 mL o e hanol
(99.9 % / ). The glass lask con aining he diges ed milk was also insed
wi h 15 mL o T i on™ X-100 (0.1 % / in ul apu e wa e ) and 15 mL
o e hanol (99.9 % / ) o mobilize any MP PET pa icles adhe ing o i s
su aces and collec hem on he il e . The e ained diges ed ma e ial on
he Si il e was lushed sequen ially wi h 5 mL o ul apu e wa e , 5 mL
o ni ic acid (5 % / in ul apu e wa e ) and 10 mL o ul apu e wa e .
Finally, he unnel was insed again wi h 15 mL espec i ely o
T i on™ X-100 (0.1 % / in ul apu e wa e ) and 15 mL o e hanol
(99.9 % / ) o eco e any pa icles po en ially adhe ing o i s su aces.
The il e was hen s o ed in closed glass pe i dishes un il analysis.
Following he same p ocedu e, p ocedu al blanks we e pe o med using
ul apu e wa e o e alua e any po en ial MP con amina ion in oduced
du ing he diges ion p ocedu e.
A comp ehensi e o e iew o he milk sample p epa a ion is p e-
sen ed in Fig. 1.
As MPs a e ubiqui ous pollu an s, p e en ing con amina ion du ing
he analy ical wo k low is c ucial o ensu ing accu a e esul s. To his
end, all sample manipula ions we e pe o med unde a lamina low
hood by quali ied analys s adequa ely ained in sample handling and
speci ic measu emen echniques. Bo h labo a o ies ollowed he p e-
cau ions ecommended in [34], including washing hands, no make-up
and wea ing a co on lab coa . All equipmen (e.g. bo les, pe i
dishes, beake , il e s, weeze s, il e ing uni s) was ca e ully cleaned
be o e use and co e ed wi h aluminum oil. Reagen s we e also p epa ed
unde he same con olled condi ions in clean beake s, using glass pi-
pe es o ans e s and aluminum oil p o ec ion be o e and a e use o
minimize po en ial con amina ion. Samples emained co e ed wi h
aluminum oil du ing all di e en p epa a ion s eps, including il a ion,
excep o b ie exposu e while g adually pou ing.
The measu emen esul s a e exp essed as he absolu e numbe o
PET pa icles de ec ed by he analyzed sample, a he han being
no malized o sample mass. This ope a ional de ini ion p o ides ace-
abili y h ough pa icle coun ing and main ains consis ency o subse-
quen in e labo a o y compa ison.
Fig. 1. In an milk powde sample p epa a ion, including PET RM able spiking, o iden i ica ion and quan i ica ion o SMPs.
M. Pu zu e al.
Talan a Open 12 (2025) 100586
3
2.3.1. In e labo a o y compa ison s udy o in an milk powde sample
analysis
In an milk samples we e analyzed by Raman spec oscopy in wo
di e en labo a o ies, INRiM (Lab 1) and Uni e si y o Pa ma (Lab 2),
using hei indi idual µ-Raman me hod bu he same sample p epa a ion
p o ocol (Fig. 1).
Labo a o y 1. De ec ion was pe o med on milk samples using a
LabRAM Odyssey Raman spec ome e (HORIBA) equipped wi h a CCD
de ec o . Ini ially he en i e il e was imaged using a 10 ×objec i e
(OLYMPUS, nume ical ape u e: 0.25) unde da k ield illumina ion
sys em o acqui e a pano amic mosaic o he il e . All subsequen
measu emen s we e ca ied ou wi h a 50 ×long wo king dis ance
objec i e (OLYMPUS, nume ical ape u e =0.5), also in da k ield mode
in o de o minimize he con ibu ion o he il e po es and acili a e
image analysis. Th ee egions we e andomly selec ed ac oss he e ec-
i e il a ion a ea (~ 5 mm diame e ), each measu ing 2.5 mm ×1.5
mm (3.75 mm
2
), collec i ely ep esen ing app oxima ely 55 % o he
il e ing su ace. The da a ob ained we e scaled o es ima e PET pa icle
numbe ac oss he ull il a ion a ea.
Spec a we e acqui ed in he ange o 620–1760 cm
-1
using a 633 nm,
an exci a ion lase wi h a powe o 15 mW, an exposu e ime o 1 s and
one accumula ion, and a 600 1/mm g a ing. Da a acquisi ion was pe -
o med using LabSpec so wa e ( . 6.8.1.9), while pa icle iden i ica ion
was ca ied ou wi h IDFinde so wa e ( . 4.2, HORIBA). Each de ec ed
pa icle was hen classi ied as a speci ic polyme ype by compa ison
wi h an in e nal spec al lib a y con aining e e ence spec a o he mos
common plas ic polyme s. Spec a wi h a Hi Quali y Index (HQI) >80
% we e au oma ically assigned o PET while hose wi h 60 % <HQI <
80 % we e manually e i ied.
Labo a o y 2. De ec ion was pe o med on milk samples using a
LabRAM HR E olu ion
μ
-Raman spec ome e (HORIBA) equipped wi h
a liquid-ni ogen CCD de ec o . Daily calib a ion was pe o med using
he cha ac e is ic silicon Raman band a 520.7 cm⁻¹.
An o e iew o he en i e il e was ob ained using a 10 ×objec i e
(Olympus MPlan N 10 ×/0.25) unde b igh ield mode. Based on his
o e iew, wo non-o e lapping egions (2.5 mm ×2 mm each, 5 mm
2
)
we e andomly selec ed ac oss he e ec i e il e ing a ea (~ 5 mm
diame e ) and analyzed using a 50 ×objec i e (Olympus LMPlanFL N 50
×/0.50), co e ing app oxima ely 50 % o he il a ion su ace.
Spec al analysis was pe o med using he LabSpec so wa e ( .
6.8.1.9). Pa icle coun ing, size measu emen , mo phological analysis,
and spec al e alua ion we e ca ied ou using he LabSpec Pa icle-
Finde ™ applica ion. Compound iden i ica ion was achie ed by
compa ing he acqui ed spec a wi h bo h Ho iba e e ence lib a ies and
a cus om-made spec al lib a y.
Use -de ined mo phological h esholds we e applied o minimise
noise and dis inguish pa icles om he backg ound using a eal- ime
ideo image.
Spec a we e collec ed in he 415–2060 cm⁻¹ ange using a 532 nm
exci a ion lase (25 mW powe on he sample), wi h an acquisi ion ime
o 1 s and collec ing wo accumula ions in o de o emo e spikes. A 600
lines/mm g a ing was used. The spec a we e no malized and subjec ed
o a polynomial baseline co ec ion. Spec a wi h a Hi Quali y Index
(HQI) >80 % we e au oma ically assigned o PET. Spec a wi h HQI
alues be ween 60 % and 80 % we e manually e iewed o con i m he
iden i ica ion.
2.4. Blank managemen and quali y con ol
Despi e s ic ly ollowing he p ecau ions ecommended in [33],
comple e elimina ion o backg ound con amina ion is no easible.
The e o e, i is essen ial o moni o any po en ial con amina ion in o-
duced h oughou he en i e sample handling p ocess [47]. P ocedu al
blanks wi h ul apu e wa e we e sys ema ically pe o med in pa allel
wi h he milk samples in bo h labo a o ies, in o de o e alua e MP
backg ound con amina ion ha could occu du ing enzyma ic diges ion,
alkaline hyd olysis and il a ion.
To ensu e compa abili y and eliabili y o he esul s, each labo a-
o y de e mined i s own epo ing limi (RL). The RL is de ined as he
minimum numbe o MPs ha can be eliably quan i ied by he labo-
a o y, depending on he polyme ype, size class, and s anda d ope -
a ing condi ions [34]. The RL was es ablished o each polyme based on
he esul s o en p ocedu al blanks independen ly pe o med by each
labo a o y; howe e , in his s udy only he RL o PET was conside ed. I
was de e mined aking in o accoun he de ec ion h eshold imposed by
he il e po e size, which in his case is 5 µm, and calcula ed as ollows:
RL [5
μ
m] = mean (10 blanks) + 3×SD (10 blanks)(Eq. 1)
All analyzed samples we e compa ed agains he co esponding RL
de e mined o each labo a o y. Values below he RL a e conside ed no
eliably quan i iable, while alues abo e he RL can be epo ed.
3. Resul s and discussion
3.1. Re e ence alues o PET e e ence ma e ials
Re e ence PET pa icle numbe s o HLB and LLB we e p e iously
de e mined by µ-Raman and µ-FTIR spec oscopy in a sepa a e manu-
sc ip [46] p epa ed wi hin he 21GRD07 Plas icT ace p ojec [40].
De ailed ins umen a ion, acquisi ion pa ame e s, and pa icle iden i i-
ca ion c i e ia a e epo ed in Table S1, and PET pa icle numbe s pe
able , ca ego ized by size ange (5–10, 10–20, 20–50, 50–100, and
100–500 µm) in acco dance wi h ISO16094–2 [35,41], a e summa ized
in Table 1. These PET RM alues we e used solely o spiking, me hod
de elopmen , and accu acy e alua ion h ough an in e labo a o y
compa ison.
Gi en he inhe en di e ences in spa ial esolu ion be ween µ-Raman
and µ-FTIR, and he addi ional challenges posed by he complex milk
ma ix, µ-Raman was selec ed as he p ima y analy ical echnique o
he milk samples. Fo comple eness, µ-FTIR measu emen s we e also
pe o med, suppo ing he obse a ions p e iously obse ed. The co -
esponding µ-FTIR esul s a e p o ided in he Supplemen a y In o ma-
ion (SI, Sec .1).
3.2. Raman spec oscopy
3.2.1. E alua ion o he sub-sampling s a egy
To speed up he analysis and enable a mo e de ailed in es iga ion o
smalle pa icles down o 5 µm, a sub-sampling s a egy was used ha
in ol ed imaging h ee (Lab 1) o wo (Lab 2) ec angula egions,
co e ing a o al o 55 % and 50 % o he il a ion su ace, espec i ely.
These egions we e selec ed o minimize ocus loss caused by non-
la ness o he il e s and o educe analysis ime. The pa icle coun
ob ained in he selec ed egions we e a e aged o each size ange and
hen no malized o he en i e il a ion a ea in o de o es ima e he o al
numbe o pa icles. This me hod elies on he assump ion ha PET
pa icles a e homogeneously dis ibu ed ac oss he il a ion a ea o
ensu e accu a e ex apola ion. To e i y whe he imaging app oxi-
ma ely 50 % o he il a ion a ea p o ides a ep esen a i e es ima e o
he o al pa icle coun , PET RM able s om bo h HLB and LLB we e
dissol ed in ul apu e wa e , il e ed, and analyzed using he sub-
sampling app oach. The esul s we e hen compa ed o he known PET
RM pa icle numbe s o p ope compa ison. Fig. 2c and show he PET
RM able s o HLB and LLB, espec i ely, pa icle size dis ibu ions
ex apola ed o 100 % o he il a ion a ea, compa ed o he size dis-
ibu ions ob ained om he analysis o he en i e il e s. No no able
di e ences we e obse ed be ween he wo app oaches in all size
anges, and he ex apola ion accu acy was high o bo h ba ches: 87 ±
11 % and 97 ±11 % o HLB, and 98 ±21 % and 105 ±32 o LLB
(Table 2), o Lab 1 and Lab 2, espec i ely. These esul s con i m ha
M. Pu zu e al.
Talan a Open 12 (2025) 100586
4
he da a ob ained om he wo o h ee selec ed egions (50–55 % o he
il a ion su ace), when ex apola ed o he en i e il a ion a ea, eli-
ably ep esen he ac ual PET pa icles dis ibu ion, suppo ing he
assump ion o a homogeneous pa icle dis ibu ion ac oss he il a ion
a ea. Based on his assessmen , he sub-sampling me hod was sys em-
a ically applied by bo h labo a o ies o all Raman measu emen s,
including p ocedu al blanks.
3.2.2. E alua ion o PET backg ound
The e alua ion o he PET RL and he analysis o he un o i ied milk
ma ix we e conduc ed by µ-Raman o assess po en ial backg ound
con amina ion and e i y he sui abili y o he milk ma ix o he
analy ical assessmen using he PET RM.
In bo h labo a o ies, he RL o PET was de e mined ia µ-Raman
analysis o he p ocedu al blanks om each labo a o y, co esponding o
a o al o 25 pa icles o Lab 1 and 21 pa icles o Lab 2. The un o i ied
milk ma ix showed a PET numbe o 18 ±5 as calcula ed in Lab 1 and 2
±2 in Lab 2. The sligh di e ences obse ed in he alues a e mos
plausibly a ibu ed o a ying le els o backg ound con amina ion
esul ing om he di e en ope a ing en i onmen s o he wo labo a-
o ies. The numbe o PET pa icles de ec ed in he un o i ied milk
ma ix in Lab 1 and Lab 2 was ound o be below he RL calcula ed by
each labo a o y (Fig. 3). In addi ion, alues o PET pa icle coun
measu ed by bo h labo a o ies a e e y low and negligible compa ed o
he PET pa icle numbe s pe able o bo h ba ches (HLB: 1759 ±141
(8 % RSD), LLB: 160 ±22 (14 % RSD)). This con i med he absence o
signi ican backg ound in e e ence o he sui abili y o he analy ical
me hod o PET de ec ion in in an milk powde , which will be con i-
den ly a ibu ed o he able i sel .
RL alues o he mos common polyme ypes we e calcula ed using
Table 1
Mean PET pa icle numbe s pe able (±SD, %RSD, n =10) in HLB and LLB loading de e mined by µ-Raman and µ-FTIR.
Spec oscopic me hod Loading 5–10
µm
10–20
µm
20–50
µm
50–100
µm
100–500
µm
Pa icle numbe pe able
µ-Raman HLB 111 ±49 330 ±62 889 ±58 361 ±29 68 ±14 1759 ±141 (8 % RSD)
µ-FTIR HLB 0 0 393 ±152 280 ±56 78 ±11 751 ±194 (26 % RSD)
µ-Raman LLB 26 ±10 38 ±5 69 ±9 26 ±9 2 ±1 160 ±22 (14 % RSD)
µ-FTIR LLB 0 0 31 ±17 26 ±10 3 ±1 60 ±14
(24 % RSD)
Fig. 2. Images o il e s om high (a, b) and low (d, e) load ba ch showing he h ee (a, d) and wo (b, c) selec ed egions loca ed wi hin ci cula il a ion a ea (~ 5
mm diame e ) used o µ-Raman analysis. Ba g aphs (c, ) epo he PET pa icle size dis ibu ions ob ained om he analysis o he en i e il a ion su ace and om
he ex apola ion o he h ee selec ed egions o 100 % o il a ion su ace o he il a ion a ea.
Table 2
PET measu emen accu acy using sub-sampling app oach o a ea-based quan i ica ion o high (n =3) and low (n =3) load ba ch.
PET pa icles pe
able
PET pa icles pe able –Ex apola ion
(Lab 1)
PET pa icles pe able – Ex apola ion
(Lab 2)
Ex apola ion accu acy %
(Lab 1)
Ex apola ion accu acy %
(Lab 2)
HLB 1759 ±141 (8 %) 1527 ±150 (10 %) 1709 ±142 (8 %) 87 ±11 97 ±11
LLB 160 ±22 (14 %) 156 ±25 (16 %) 168 ±45 (27 %) 98 ±21 105 ±32
M. Pu zu e al.
Talan a Open 12 (2025) 100586
5
µ-Raman (Table S2) and used as benchma ks o assess possible back-
g ound con amina ion om he un o i ied milk ma ix. PP, PS and PE
we e de ec ed, bu he pa icle coun o hese polyme s in he un o i-
ied milk ma ix we e all lowe han hei espec i e RLs, con i ming he
absence o MPs con amina ion in milk powde (Table S3).
3.2.3. Iden i ica ion and quan i ica ion o PET SMPs in in an milk powde
by µ-Raman
The econs i u ed milk samples o i ied wi h PET RM om bo h
ba ches and subjec ed o diges ion ea men (Fig. 1) we e analyzed by
Lab 1 and Lab 2 using µ-Raman o assess PET pa icle numbe and size
dis ibu ion a e he sample ea men p o ocol. Measu emen s we e
pe o med using he alida ed sub-sampling me hod, in which wo o
h ee ep esen a i e egions o he il a ion a ea we e analyzed and he
esul s ex apola ed o es ima e he o al pa icle coun o e he en i e
il a ion a ea. Resul s a e epo ed as absolu e pa icle coun s pe
analyzed sample, ollowing he ope a ional de ini ion in he Ma e ials
and Me hods (Sec . 2.3). Bo h labo a o ies yielded compa able esul s,
wi h an a e age numbe o HLB o 1448 ±253 (17 %) in Lab 1 (n =3)
and 1501 ±203 (13 %) (n =3) in Lab 2, and o LLB o 132 ±17 (13 %)
in Lab 1 (n =6) and 140 ±26 (19 %) in Lab 2 (n =6) (Fig. 4). Bo h
labo a o ies applying he same diges ion ea men and using di e en
lase sou ces and sub-sampling a eas p o ided alues ha align well
wi h he e e ence pa icle numbe s ob ained om PET RM able s dis-
sol ed in ul apu e wa e (HLB: 1759 ±141 (8 % RSD), LLB: 160 ±22
(14 % RSD)). Fo bo h ba ches, no no able di e ences we e obse ed
ac oss all size anges be ween he wo labo a o ies o when compa ed o
he RM in ul apu e wa e . Howe e , in he case o LLB, a mo e de ailed
compa ison be ween labo a o ies, e eals some a iabili y in he smalle
size anges (5–50 µm), as indica ed by di e ences in hei s anda d
de ia ion alues. The lowe numbe o PET pa icles in his ba ch may
inc ease he ela i e unce ain y wi hin each size class. When pa icle
numbe s wi hin a gi en size class a e low, e en a a ia ion o one o wo
pa icles ac oss eplica es can subs an ially a ec he s anda d
Fig. 3. E alua ion o PET backg ound le els in un o i ied milk compa ed o PET RM able s in ul apu e wa e (HLB and LLB) by µ-Raman.
Fig. 4. . PET size dis ibu ion in ea ed in an milk a e diges ion p ocedu e o high (a) (n =3) and low (b) (n =6) load ba ch by µ-Raman as measu ed by Lab 1
and Lab 2.
M. Pu zu e al.
Talan a Open 12 (2025) 100586
6
de ia ion. Mo eo e , he 532 nm lase used by Lab 2 o e s highe spa ial
esolu ion han he 633 nm lase used by Lab 1, po en ially allowing o
mo e accu a e de ec ion and sepa a ion o small o o e lapping pa i-
cles, pa icula ly in he 5–10 µm, and especially in he p esence o e-
sidual o ganic ma e om he ma ix ha in e e es wi h pa icle
ecogni ion. Ne e heless, hese a ia ions emain wi hin an accep able
ange and do no a ec he o e all compa abili y o esul s be ween he
wo labo a o ies. One-way ANOVA con i med no signi ican di e ences
(p >0.05) ac oss he h ee size anges (5–10, 10–20 and 20–50 µm)
(Fig. S3) be ween he wo labo a o ies. The eco e y a e was 82 ±16 %
and 85 ±14 % o HLB in Lab 1 and in Lab 2, espec i ely, and 82 ±15
% and 88 ±19 % o LLB in Lab 1 and Lab 2 (Table 3).
4. Conclusions
The lack o ce i ied RMs emains a c i ical challenge o ensu ing
compa abili y o analy ical me hods o quan i ica ion o MPs in com-
plex ma ices. This s udy demons a ed he eliabili y and compa abili y
o µ-Raman measu emen s o SMPs >5 µm in in an milk powde using
well-cha ac e ized PET RM able s, p oduced and cha ac e ized in
acco dance wi h ISO 17034 and ISO 33405 p inciples. The app oach
suppo s ha monized MP measu emen s ac oss labo a o ies wi hou he
immedia e need o a s anda dized me hod, p o ided ha e e ence
ma e ials a e a ailable o unde pin aceabili y.
The me hod was success ully applied ac oss wo labo a o ies,
yielding compa able esul s in e ms o pa icle numbe and size ange
dis ibu ion. I s success ul applica ion o PET, a polyme known o i s
sensi i i y o diges ion condi ions in ol ing high empe a u es and he
use o chemical eagen s, demons a es he obus ness o he app oach
and sugges s i s po en ial applicabili y o quan i ying o he ypes o
polyme s wi h di e en sensi i i ies. S ic quali y assu ance and quali y
con ol measu es, including p ocedu al blanks and he calcula ion o
epo ing limi , we e implemen ed o eliably dis inguish PET RM MPs
om po en ial backg ound con amina ion.
O e all, his s udy no only demons a es he easibili y o applying
an analy ical me hod o small PET MPs in a complex ma ix bu also
highligh s he impo ance o in e labo a o y compa abili y and p o ides
a p ac ical ounda ion o ha moniza ion o MPs quan i ica ion me hods
in complex ma ices. Wi h app op ia e sample p epa a ion and epo -
ing limi , his me hodological app oach could ep esen a eliable
analy ical ool which can suppo ou ine moni o ing o MP con ami-
na ion ac oss he dai y supply chain, aiding e o s o educe bo h he
numbe and ypes o MPs in consume p oduc s.
CRediT au ho ship con ibu ion s a emen
Ma a Pu zu: Concep ualiza ion, Fo mal analysis, In es iga ion,
Me hodology, Valida ion, Visualiza ion, W i ing – o iginal d a . Ma a
Ba ba esi: Fo mal analysis, In es iga ion, Me hodology, W i ing – e-
iew & edi ing. Ma a Fadda: Me hodology, W i ing – e iew & edi ing.
Alessio Sacco: W i ing – e iew & edi ing. Mau izio Pie gio anni:
In es iga ion, W i ing – e iew & edi ing. Ma eo Masino: Me hodol-
ogy, W i ing – e iew & edi ing. Fede ica Bianchi: Fo mal analysis,
W i ing – e iew & edi ing. Ko inna Al mann: W i ing – e iew &
edi ing. Niza Benismail: Fo mal analysis, W i ing – e iew & edi ing.
Lau een Coïc: Fo mal analysis, W i ing – e iew & edi ing. I ana
Fenoglio: Supe ision, W i ing – e iew & edi ing. Monica Ma a ozzi:
In es iga ion, Me hodology, Valida ion, W i ing – e iew & edi ing.
And ea Ma io Rossi: Supe ision, W i ing – e iew & edi ing. Ma ia
Ca e i: Concep ualiza ion, Funding acquisi ion, Me hodology, Re-
sou ces, Supe ision, W i ing – e iew & edi ing. And ea Ma io Gio-
annozzi: Concep ualiza ion, Funding acquisi ion, In es iga ion,
P ojec adminis a ion, Supe ision, W i ing – e iew & edi ing.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing inancial
in e es s o pe sonal ela ionships ha could ha e appea ed o in luence
he wo k epo ed in his pape .
Acknowledgemen s
The au ho s hank he Eu opean Commission o unding he Plas-
icT ace p ojec (21GRD07) om he Eu opean Pa ne ship on
Me ology, co- inanced om he Eu opean Union’s Ho izon Eu ope
Resea ch and Inno a ion P og amme and by he Pa icipa ing s a es.
This wo k was suppo ed by he METROFOOD-IT p ojec which ecei ed
unding om he Eu opean Union-Nex Gene a ionEU, PNRR-Mission 4
“Educa ion and Resea ch” Componen 2: om esea ch o business, In-
es men 3.1: Fund o he ealiza ion o an in eg a ed sys em o
esea ch and inno a ion in as uc u es-IR0000033 (D.M. P o . n.120
del 21/06/2022).
Resea ch and inno a ion ne wo k on ood and nu i ion Sus ain-
abili y, Sa e y and Secu i y–Wo king ON Foods” (ONFOODS) p ojec
which ecei ed unding om he NRRP and he Mission 4 Componen 2
In es men 1.3-Call o ende No. 341 o 15/03/2022 o he I alian
Minis y o Uni e si y and Resea ch unded by he Eu opean Union-
–Nex Gene a ionEU. Awa d Numbe : P ojec code PE0000003.
Supplemen a y ma e ials
Supplemen a y ma e ial associa ed wi h his a icle can be ound, in
he online e sion, a doi:10.1016/j. alo.2025.100586.
Da a a ailabili y
The da a ha suppo he indings o his s udy a e openly a ailable
in Zenodo a h ps://doi.o g/10.5281/zenodo.17483289
Re e ences
[1] Plas ics - he as Fac s 2024. h ps://plas icseu ope.o g/knowledge-hub/plas
ics- he- as - ac s-2024/, 2023.
[2] A.A. Mamun, T.A.E. P ase ya, I.R. Dewi, M. Ahmad, Mic oplas ics in human ood
chains: ood becoming a h ea o heal h sa e y, Sci. To al En i on. 858 (2023)
159834, h ps://doi.o g/10.1016/j.sci o en .2022.159834.
[3] L. Van Cauwenbe ghe, L. De iese, F. Galgani, J. Robbens, C.R. Janssen,
Mic oplas ics in sedimen s: a e iew o echniques, occu ence and e ec s, Ma .
En i on. Res. 111 (2015) 5–17, h ps://doi.o g/10.1016/j.
ma en es.2015.06.007.
[4] A.L. And ady, Mic oplas ics in he ma ine en i onmen , Ma . Pollu . Bull. 62
(2011) 1596–1605, h ps://doi.o g/10.1016/j.ma polbul.2011.05.030.
[5] J.P.G.L. F ias, R. Nash, Mic oplas ics: inding a consensus on he de ini ion, Ma .
Pollu . Bull. 138 (2019) 145–147, h ps://doi.o g/10.1016/j.
ma polbul.2018.11.022.
[6] L. Van Cauwenbe ghe, L. De iese, F. Galgani, J. Robbens, C.R. Janssen,
Mic oplas ics in sedimen s: a e iew o echniques, occu ence and e ec s, Ma .
En i on. Res. 111 (2015) 5–17, h ps://doi.o g/10.1016/j.
ma en es.2015.06.007.
[7] J.-P.W. Des o ges, M. Galb ai h, N. Dange ield, P.S. Ross, Widesp ead dis ibu ion
o mic oplas ics in subsu ace seawa e in he NE Paci ic Ocean, Ma . Pollu . Bull.
79 (2014) 94–99, h ps://doi.o g/10.1016/j.ma polbul.2013.12.035.
Table 3
PET eco e y a e calcula ed by Lab 1 and Lab 2 in ea ed in an milk o high (n
=3) and low (n =6) load ba ch by µ-Raman.
PET
pa icles
pe able
PET pa icles
pe able in
Milk – Lab 1
PET Pa icles
pe able in
Milk – Lab 2
Reco e y
% - Lab 1
Reco e y
% - Lab 2
HLB 1759 ±
141 (8 %)
1448 ±253
(17 %)
1500 ±204
(14 %)
82 ±16 85 ±14
LLB 160 ±22
(14 %)
131 ±16 (13
%)
140 ±24 (17
%)
82 ±15 88 ±19
M. Pu zu e al.
Talan a Open 12 (2025) 100586
7
[8] ISO 24187:2023(en) P inciples o he analysis o mic oplas ics p esen in he
en i onmen . h ps://www.iso.o g/obp/ui/en/#iso:s d:iso:24187:ed-1: 1:en,
2025.
[9] M. Ca be y, W. O’Conno , T. Palanisami, T ophic ans e o mic oplas ics and
mixed con aminan s in he ma ine ood web and implica ions o human heal h,
En i on. In . 115 (2018) 400–409, h ps://doi.o g/10.1016/j.en in .2018.03.007.
[10] R. Aloia, L. Annunzia a, F. Di Giacin o, S. De Simone, C. P o ico, I. Fanelli, G. Di
F ancesco, V. Mussi, G. Sco ichini, Applica ion o Raman spec oscopy o he
analysis o mic oplas ics in ood and be e ages: a comp ehensi e e iew, Appl.
Spec osc. Re . (2025) 1–35, h ps://doi.o g/10.1080/05704928.2025.2512771.
[11] F. Bessa, P. Ba ía, J.M. Ne o, J.P.G.L. F ias, V. O e o, P. Sob al, J.C. Ma ques,
Occu ence o mic oplas ics in comme cial ish om a na u al es ua ine
en i onmen , Ma . Pollu . Bull. 128 (2018) 575–584, h ps://doi.o g/10.1016/j.
ma polbul.2018.01.044.
[12] O.C. Gea, M. Fe an e, M. Banni, C. Fa a a, I. Nicolosi, A. C is aldi, M. Fio e,
P. Zucca ello, Mic o- and nano-plas ics in edible ui and ege ables. The i s die
isks assessmen o he gene al popula ion, En i on. Res. 187 (2020) 109677,
h ps://doi.o g/10.1016/j.en es.2020.109677.
[13] M. Kedzie ski, B. Lecha , O. Si e, G. Le Mague , V. Le Tilly, S. B uzaud, Mic oplas ic
con amina ion o packaged mea : occu ence and associa ed isks, Food Packag.
Shel Li e 24 (2020) 100489, h ps://doi.o g/10.1016/j. psl.2020.100489.
[14] E. B ancaleone, D. Ma ei, V. Fuscole i, L. Lucen ini, G. Fa e o, G. Cecchini,
A. F ugis, V. Gioia, M. Lazzazza a, Mic oplas ic in D inking Wa e : a Pilo S udy,
Mic oplas ics 3 (2024) 31–45, h ps://doi.o g/10.3390/mic oplas ics3010003.
[15] S. Venka a Cha i, F. Pe ´
ez-Gue a a, I. Elizalde-Ma ínez, G. K.Muniasamy, Fi s
s udy o i s kind on he mic oplas ic con amina ion o so d inks, cold ea and
ene gy d inks - Fu u e esea ch and en i onmen al conside a ions, Sci. To al
En i on. 726 (2020) 138580, h ps://doi.o g/10.1016/j.sci o en .2020.138580.
[16] G. K.Muniasamy, F. P´
e ez-Gue a a, I. Elizalde-Ma ínez, S. Venka a Cha i, B anded
milks – A e hey immune om mic oplas ics con amina ion? Sci. To al En i on.
714 (2020) 136823 h ps://doi.o g/10.1016/j.sci o en .2020.136823.
[17] P.A. Da Cos a Filho, D. And ey, B. E iksen, R.P. Peixo o, B.M. Ca e es, M.
E. Ambühl, J.B. Desca ega, S. Dubascoux, P. Zbinden, A. Panchaud, E. Poi e in,
De ec ion and cha ac e iza ion o small-sized mic oplas ics (≥5 µm) in milk
p oduc s, Sci. Rep. 11 (2021) 24046, h ps://doi.o g/10.1038/s41598-021-03458-
7.
[18] E. Visen in, G. Nie o, F. Bene i, C. O’Donnell, M. De Ma chi, Assessing
mic oplas ic con amina ion in milk and dai y p oduc s, NPJ Sci. Food 9 (2025)
135, h ps://doi.o g/10.1038/s41538-025-00506-8.
[19] P. Sha ma, Mic oplas ic Con amina ion in Food P ocessing: ole o Packaging
Ma e ials, Food Sci. Eng. 5 (2024) 271–287, h ps://doi.o g/10.37256/
se.5220244519.
[20] Food and Ag icul u e O ganiza ion o he Uni ed Na ions, Dai y Ma ke Re iew.
O e iew o global ma ke de elopmen s in 2024, (n.d.). h ps://openknowledge.
ao.o g/i ems/3be6263e-6d0b-42ee-bce4-97663 2cd905.
[21] D. Schymanski, C. Goldbeck, H.-U. Hump , P. Fü s , Analysis o mic oplas ics in
wa e by mic o-Raman spec oscopy: elease o plas ic pa icles om di e en
packaging in o mine al wa e , Wa e Res. 129 (2018) 154–162, h ps://doi.o g/
10.1016/j.wa es.2017.11.011.
[22] Q. Zhang, L. Liu, Y. Jiang, Y. Zhang, Y. Fan, W. Rao, X. Qian, Mic oplas ics in in an
milk powde , En i on. Pollu . 323 (2023) 121225, h ps://doi.o g/10.1016/j.
en pol.2023.121225.
[23] K. Kadac-Czapska, P. Ju zenka T zebia owska, M. Mazu kiewicz, P. Kowalczyk,
E. Knez, M. Beh end , S. Mahlik, A. Zaleska-Medynska, M. G embecka, Isola ion
and iden i ica ion o mic oplas ics in in an o mulas – A po en ial heal h isk o
child en, Food Chem. 440 (2024) 138246, h ps://doi.o g/10.1016/j.
oodchem.2023.138246.
[24] M.F. Diaz-Basan es, J.A. Conesa, A. Fullana, Mic oplas ics in Honey, Bee , Milk and
Re eshmen s in Ecuado as Eme ging Con aminan s, Sus ainabili y 12 (2020)
5514, h ps://doi.o g/10.3390/su12145514.
[25] R. Ki uba, P. M, A. R, R.R. M, H.T. C, M. P, S. J, L. C, N.B. I, Iden i ica ion o
Mic oplas ics as Eme ging Con aminan in B anded Milk o Tamil Nadu S a e,
India, Asian J. Biol. Li e Sci. 11 (2022) 181–187, h ps://doi.o g/10.5530/
ajbls.2022.11.25.
[26] B. Basa an, Z. ¨
Ozçi çi, H.T. Akcay, Ü. Ay an, Mic oplas ics in b anded milk: die a y
exposu e and isk assessmen , J. Food Compos. Anal. 123 (2023) 105611, h ps://
doi.o g/10.1016/j.j ca.2023.105611.
[27] S. Rbaibi Zipak, K. Mu a oglu, S.K. Buyukunal, Mic oplas ics in aw milk samples
om he Ma ma a egion in Tu key, J. Consum. P o . Food Sa . 19 (2024)
175–186, h ps://doi.o g/10.1007/s00003-023-01477-2.
[28] T.K. Chak abo y, M.J. Hasan, B.N. Ne ema, M.A. Rayhan, S.M.H. Asi , A. Biswas,
S. Sa ke , M. Ahmmed, Md.S. Nice, K.R. Islam, P.C. Debna h, P. Chowdhu y, Md.
S. Rahman, S. Zaman, G.C. Ghosh, M. Hasibuzzaman, Mic oplas ics in he
comme cially a ailable b anded milk in Bangladesh: an eme ging h ea o human
heal h, J. Haza d. Ma e . 477 (2024) 135374, h ps://doi.o g/10.1016/j.
jhazma .2024.135374.
[29] Manika Gup a, Pinki Seh awa , Gi ish Gup a, Nidhi Singhala*, Sushil Kuma
Kansal, De ec ion and Quan i ica ion o Mic oplas ics in B anded and Fa m Milk in
India 2 coupled Wi h SFEC- based spec al Valida ion, (PREPINT ARTICLE). h ps
://pape s.ss n.com/sol3/pape s.c m?abs ac _id=5374147.
[30] A. Ragusa, V. No a s e ano, A. S ela o, A. Belloni, G. Gioacchini, C. Blondeel,
E. Zucchelli, C. De Luca, S. D’A ino, A. Gulo a, O. Ca ne ali, E. Gio gini, Raman
Mic ospec oscopy De ec ion and Cha ac e isa ion o Mic oplas ics in Human
B eas milk, Polyme s (Basel) 14 (2022) 2700, h ps://doi.o g/10.3390/
polym14132700.
[31] A. Sa aluck, T. Techa ang, P. Bunyapipa , K. Boonchuwong, Y. Pullapu ,
A. Mo dmuang, De ec ion o Mic oplas ics in Human B eas Milk and I s
Associa ion wi h Changes in Human Milk Bac e ial Mic obio a, J. Clin. Med. 13
(2024) 4029, h ps://doi.o g/10.3390/jcm13144029.
[32] A. Dąb owska, W. Komo owska, B. K isz , I. Szab´
o, Feasibili y o Raman and FTIR
spec oscopy o di ec mic oplas ic sea ch in he human milk samples:
compa a i e quali a i e s udy, Eco oxicol. En i on. Sa . 296 (2025) 118159,
h ps://doi.o g/10.1016/j.ecoen .2025.118159.
[33] M.J. Noonan, N. G echi, C.L. Mills, M. De A. M. M. Fe az, Mic oplas ics analy ics:
why we should no unde es ima e he impo ance o blank con ols, Mic oplas ics
Nanoplas ics 3 (2023) 17, h ps://doi.o g/10.1186/s43591-023-00065-3.
[34] D. Schymanski, B.E. Oßmann, N. Benismail, K. Bouke ma, G. Dallmann, E. Von De
Esch, D. Fische , F. Fische , D. Gilliland, K. Glas, T. Ho mann, A. K¨
apple ,
S. Laco e, J. Ma co, M.E. Rakwe, J. Weisse , C. Wi zig, N. Zumbül e, N.P. I le a,
Analysis o mic oplas ics in d inking wa e and o he clean wa e samples wi h
mic o-Raman and mic o-in a ed spec oscopy: minimum equi emen s and bes
p ac ice guidelines, Anal. Bioanal. Chem. 413 (2021) 5969–5994, h ps://doi.o g/
10.1007/s00216-021-03498-y.
[35] ISO 16094-2:2025 Wa e quali y, Analysis o mic oplas ic in wa e Pa 2:
ib a ional spec oscopy me hods o wa e s wi h low con en o suspended solids
including d inking wa e . h ps://www.iso.o g/s anda d/84460.h ml, 2025.
[36] In e na ional O ganiza ion o S anda diza ion (ISO), ISO/DIS 16094-3 Wa e
quali y — Analysis o mic oplas ic in wa e Pa 3: he mo-analy ical me hods o
wa e s wi h low con en o suspended solids including d inking wa e , (n.d.).
[37] J. Seghe s, E.A. S e aniak, R. La Spina, C. Cella, D. Mehn, D. Gilliland, A. Held,
U. Jacobsson, H. Em ebo g, P epa a ion o a e e ence ma e ial o mic oplas ics in
wa e —E alua ion o homogenei y, Anal. Bioanal. Chem. 414 (2022) 385–397,
h ps://doi.o g/10.1007/s00216-021-03198-7.
[38] ISO 17034:2016, Gene al Requi emen s Fo he Compe ence o Re e ence Ma e ial
P oduce s, ISO, 2025 n.d. h ps://www.iso.o g/s anda d/29357.h ml. accessed
Oc obe 13
[39] ISO 33405:2024, Re e ence Ma e ials — App oaches o Cha ac e iza ion and
Assessmen o Homogenei y and S abili y, ISO, 2025 n.d. h ps://www.iso.o g/
s anda d/84226.h ml. accessed Oc obe 13
[40] Plas icT ace – T acing Mic o and NanoPlas ics in Food and En i onmen , n.d. h ps
://plas ic ace.eu/, 2025. accessed Augus 7
[41] K. Al mann, A.M. Gio annozzi, Summa y epo on he de elopmen o p is ine SI
aceable e e ence ma e ials o SMPs (10
μ
m - 100
μ
m), ep esen a i e o
pa ially deg aded/na u ally aged samples in complex ood and en i onmen al
ma ices, and o ealis ic polydispe se size dis ibu ions and i egula shapes,
(2025). h ps://doi.o g/10.5281/ZENODO.17432420.
[42] R. Nis ic`
o, Polye hylene e eph hala e (PET) in he packaging indus y, Polym.
Tes . 90 (2020) 106707, h ps://doi.o g/10.1016/j.polyme es ing.2020.106707.
[43] R. F anz, F. Baye , F. Welle, Eu opean Commission, F aunho e Ins i u ü
Ve ah ens echnik und Ve packung (IVV), (Ge many), The Coca-Cola Company
(USA), Guidance and C i e ia Fo Sa e Recycling o Pos Consume Polye hylene
Te eph hala e (PET) In o New Food Packaging Applica ions, Resea chGa e (n.d.).
h ps://doi.o g/10.13140/RG.2.1.2518.9840.
[44] Z.O.G. Schyns, A.D. Pa el, M.P. Sha e , Unde s anding poly(e hylene
e eph hala e) deg ada ion using gas-media ed simula ed ecycling, Resou .
Conse . Recycl. 198 (2023) 107170, h ps://doi.o g/10.1016/j.
escon ec.2023.107170.
[45] A. Dehau , A.-L. Cassone, L. F `
e e, L. He mabessie e, C. Himbe , E. Rinne ,
G. Ri i`
e e, C. Lambe , P. Soudan , A. Hu e , G. Du los, I. Paul-Pon , Mic oplas ics
in sea ood: benchma k p o ocol o hei ex ac ion and cha ac e iza ion, En i on.
Pollu . 215 (2016) 223–233, h ps://doi.o g/10.1016/j.en pol.2016.05.018.
[46] Ko inna Al mann, Yos i Wiesne , Pe a Fengle 1, Ch is iane Weimann, Claudia
D ago, Ul ike B aun, Ma a Fadda, Ma a Pu zu, Soledad Munia egui Lo enzo,
Ca men Moscoso-P´
e ez, Ve ´
onica Fe n´
andez-Gonz´
alez, Paul-Tibe iu Miclea, Niza
Benismail, Lau een Coic, Alina Mal se a, Be an Ba el, Vilde Klos e Snekke ik,
Andy M. Boo h, Lisbe h So ensen, En ica Alasona i, Ca ine Chi as-Joly, And ea
Ma ia Gio annozzi, Lac ose Table s as sui able Mic oplas ics Re e ence Ma e ial o
suppo alida ion s udies and quali y con ol, (UNDER REVIEW).
[47] V.C. Sh u i, G. Ku alam-Muniasamy, Blanks and bias in mic oplas ic esea ch:
implica ions o u u e quali y assu ance, T ends En i on. Anal. Chem. 38 (2023)
e00203, h ps://doi.o g/10.1016/j. eac.2023.e00203.
M. Pu zu e al.
Talan a Open 12 (2025) 100586
8
