Ion beam induced secondary electron tomography of acrylonitrile-styrene-acrylate/polycarbonate polymer blends for fused filament fabrication and injection moulding

Ion beam induced seconda y
elec on omog aphy o
ac yloni ile-s y ene-ac yla e/
polyca bona e polyme blends
o used ilamen ab ica ion and
injec ion moulding
J. He nández-Saz1, D. Mo eno-Sanchez2, L. M. Valencia3, Y. Gómez3, S. I. Molina3 &
M. He e a3
Polyme blending is an in e es ing s a egy o b oaden he combina ion o p ope ies a ailable o
a a ie y o applica ions. To unde s and he beha iou o he new ma e ials ob ained as well as he
in luence o he ab ica ion pa ame e s used, me hods o analyse he dis ibu ion o polyme s in he
blend wi h esolu ion below he mic ome e a e equi ed. In his wo k, we demons a e he capabili y
o ocused ion beam (FIB) omog aphy o p o ide 3D in o ma ion o he polyme dis ibu ion in
objec s ob ained by blending ac yloni ile-s y ene-ac yla e (ASA) wi h polyca bona e (PC) (50 w %),
ab ica ed by Fused Filamen Fab ica ion (FFF) and by Injec ion Moulding (IM). Fo his, ion beam
induced seconda y elec on (iSE) images show he capabili y o dis inguish unequi ocally he wo
phases in he blend, p o iding enough con as s o pe o m he 3D expe imen . Addi ionally, Mon e
Ca lo simula ions show ha he la e al sp ead o inciden elec ons in PC is 61.7 nm and o Ga+ ions o
26.2 nm, e idencing a be e spa ial esolu ion in iSE imaging. The spu e ing a e unde he ion beam
has been quan i ied o bo h nea ASA and nea PC o ind op imal pa ame e s o he iSE omog aphy,
esul ing in a cu en o 0.05 nA and a dwell ime o 3 µs. Ou esul s e eal signi ican di e ences in
he mo phology o ASA/PC blends depending on he ab ica ion me hod. Blends ob ained by FFF
exhibi s ong di ec ionali y and a co-con inuous mo phology, whe eas IM objec s p esen a d ople -
ma ix s uc u e. Also, he in e ace a ea be ween he ASA and PC is quan i ied o be o 3200 μm² o
he FFF sample and 1400 μm² o he IM sample, app oxima ely double in FFF han in IM. The easons
o he di e en mo phologies ob ained in he s udied blends and possible e ec s in hei mechanical
p ope ies a e discussed.
Keywo ds FIB omog aphy, iSE omog aphy, Polyme blends, Mo phology, Mic oscopy, In e ace
The polyme indus y has become one o he mos de eloping indus ies in he wo ld in he las decades, p oducing
con inuous changes in e e yday li e. Special a en ion is paid o syn he ic polyme s, whe e he s uc u e o he
ma e ials can be modi ied in mul iple ways o make hem sui able o speci ic applica ions. One o he s a egies
used o ob ain ma e ials wi h he desi ed combina ion o p ope ies is polyme blending. Since he i s pa en
in 18461, polyme blending has become widesp ead, as he esul ing ma e ials ind in e es ing applica ions in
nume ous ields due o hei ad anced p ope ies. Polyme blending is becoming popula in ad anced and
con en ional manu ac u ing echniques, such as Addi i e Manu ac u ing (AM) and Injec ion Moulding (IM).
Fo example, he use o polyme blending in Fused Filamen Fab ica ion (FFF), one o he mos widely used AM
1Depa amen o de Ingenie ía y Ciencia de los Ma e iales y del T anspo e, Uni e sidad de Se illa, 41092 Se illa,
Spain. 2Depa amen o de Ingenie ía Mecánica y Diseño Indus ial, Escuela Supe io de Ingenie ía, Uni e sidad
de Cádiz, Campus Río San Ped o, 11510 Pue o Real, Cádiz, Spain. 3Depa amen o de Ciencia de los Ma e iales e
Ingenie ía Me alú gica y Química O gánica, IMEYMAT, Uni e sidad de Cádiz, Campus Río San Ped o, 11510 Pue o
Real, Cádiz, Spain. email: [email p o ec ed]
OPEN
Scien i ic Repo s | (2025) 15:3704 1
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s
ex usion me hods, holds he p omise o signi ican ly imp o e he mechanical p ope ies and he p in abili y o
high pe o mance polyme s (such as poly-e he -e he -ke one (PEEK) o polye he imide (PEI)), which usually
p esen p ocessing challenges ela ed o hei low iscosi y (see a Re iew in2) and p in abili y. In IM, howe e ,
i ually all iden i ied polyme blend ypes can be p ocessed. In gene al, while IM o e s ad an ages in e ms
o p ecision, high- olume p oduc ion a es, and pa mechanical p ope ies, FFF excels in lexibili y, low ini ial
cos , and he abili y o p oduce complex and cus omized geome ies.
In o de o unde s and and op imize he mechanical and unc ional p ope ies o polyme blends as well as
he e ec o hei syn hesis condi ions, in o ma ion on he s uc u al cha ac e is ics o he ma e ials is essen ial.
The e is a a ie y o cha ac e iza ion echniques ha can be used o de e mine he na u e o he polyme
blends3,4. Fo example, Fou ie - ans o m in a ed spec oscopy o e s in o ma ion on chemical in e ac ions
in he ma e ial, including hyd ogen bonding and in o ma ion ega ding possible phase sepa a ion5. X- ay
di ac ion has been used o s udy he c ys alline p ope ies o PCL/PLA polyme blends, epo ing e idences o
he immiscibili y o he polyme s6. The mel ing poin dep ession ob ained by di e en ial scanning calo ime y
(DSC) has been used o es ablish miscibili y and he polyme –polyme in e ac ion pa ame e in biobased
polyamide blends7. O he echniques include dynamic mechanical analysis, use ul o he de e mina ion o he
c osslinking densi y be ween he wo componen s o a blend8 o mass spec oscopy, p o iding in o ma ion on
he su ace composi ion dis ibu ion, o example in blends o cyclic and linea polys y ene9. Howe e , mos o
hese echniques o e indi ec in o ma ion o he s uc u e o he blends, and o en i is use ul o ob ain di ec
in o ma ion on he componen s dis ibu ion a mic o/nano-scale.
In ecen yea s, he FIB ins umen has become a powe ul analysis ool. I has he un i alled abili y o
mic o/nanomanipula e a wide a ie y o ma e ials si e-speci ically wi h nanome e p ecision, while allowing he
analysis o he machined ma e ial a di e en s ages wi h a a ie y o de ec o s, including seconda y elec ons
(SE) de ec o s, elec on backsca e ed di ac ion (EBSD) de ec o s o ene gy dispe si e x- ay spec oscopy
(EDX) de ec o s. Because o his, i has ound mul iple applica ions in nano echnology (see a e iew in10).
Wi hin he e sa ili y o FIB applica ions, me hods o imaging ma e ials in h ee dimensions (3D) a e a ac ing
inc easing in e es . He e, he ion beam o he FIB is used o mill slices o ma e ial sequen ially, whe e each o
he milled aces is imaged wi h he chosen de ec o ( his echnique is known as slice and iew, FIB omog aphy
o se ial sec ioning, mo e de ails can be ound in11). The econs uc ion o he ob ained da a wi h specialized
algo i hms allows ob aining 3D in o ma ion o he s udied ma e ial. Ca e should be aken in he selec ion o he
app op ia e de ec o in hese 3D analyses. Al hough elec on beam induced SE (known as SE mic oscopy, -M)
a e o en used in FIB omog aphy12,13, in gene al his SEM signal is bes sui ed o imaging opog aphy and, since
a FIB polished ace does no con ain opog aphy pe se, some imes con as is insu icien o he analyses. This
is o en he case o polyme ma e ials, whe e FIB-based cha ac e iza ion epo s a e no nume ous. FIB-SEM
omog aphy has been used o c ea e high ideli y 3D econs uc ions o diblock copolyme s14, as in his case
he con as be ween he componen s esul ed e y la ge, and also o he s udy o po ous polyme s15. In o he
cases, opog aphy con as esul ing om a di e en ial spu e ing a e o he wo polyme s unde he ion beam
has been ound use ul o he analysis o polyme mul icomponen sys ems, as in bicomponen combina ions
o low densi y polye hylene (PE) and nylon 616. Addi ional s a egies ha e also been epo ed o op imize he
con as be ween he componen s, o example, hea y me al s aining has been used o e eal he 3D s uc u e
o he moplas ic polyole in blends17,18. On he o he hand, he use o ion beam induced SE (iSE) images ha e
demons a ed use ul o he analysis o some ma e ials, o example o he s udy o he c ack mo phology in
silicon ni ide19 and o polyme ic sca olds o he gene a ion o 3D issues20. iSE imaging ad an ages include
high con as sensi i i y o su ace opog aphy and c ys alline g ains. Ions, howe e , a e hea ie han elec ons
and, because o his, hey always p oduce a ce ain deg ee o sample spu e ing, limi ing he p obing dwell ime21.
X- ay sca e ing echniques, such as small and wide-angle X- ay sca e ing (SAXS), ha e been p e iously
used o in es iga e he o ien a ion o chains in he moplas ic polyme s du ing ope ando FFF22. Al hough hese
echniques p o ide in o ma ion abou he a e age s uc u e and o ien a ion o he polyme chains, hey lack he
spa ial esolu ion o di ec ly isualize indi idual ea u es. On he con a y, by p o iding eal-space in o ma ion,
iSE omog aphy allows o he di ec obse a ion o mo phological ea u es, which can signi ican ly impac he
ma e ials p ope ies. By combining hese wo echniques, a mo e comp ehensi e unde s anding o he s uc u e-
p ope y ela ionships in 3D p in ed polyme s can be achie ed. Howe e , o his iSE omog aphy o polyme
blends need o be op imized.
In his wo k, we e alua e he applicabili y o iSE omog aphy o he analysis in 3D o he componen s
dis ibu ion in polyme blends o ASA-PC ob ained by FFF and IM, wi h applica ions in he au omo i e
indus y23. This is he i s ime han iSE omog aphy is demons a ed o he analysis in 3D o he phase
dis ibu ion in polyme blends. Ou esul s show ha he use o iSE images is a good choice o hese analyses,
and ha he esul s ob ained wi h his echnique can be o g ea use o shed ligh a he e ec o he ab ica ion
pa ame e s on he blends mic os uc u e, which is de e minan o unde s and hei echnological p ope ies.
Expe imen al me hods
Pelle s o ASA and o PC we e pu chased a Bies e eld Ibe ica (Spain) (in pa icula , ma e ials wi h e . ASA:
LI-912 and PC: TRIREX 3025N2 a e conside ed). All he ma e ials we e d ied be o e p ocessing o emo e any
esidual mois u e in a DPA30 comp essed ai d ye (Pio an G oup, I aly) o a leas 24h a 90°C. Then, 250g o
ASA and 250g o PC we e p ocessed in a win-sc ew SCAMEX ex ude (SCAMEX, F ance, L/D 508/10,8mm,
90 pm) o ob ain a ASA 50 w % - PC 50 w % blend (ASA50 / PC50), using a empe a u e p o ile o i e hea e s
o 245-250-250-250-245°C ( om hoppe o nozzle). A con inuous ilamen wi h a diame e o 1.75mm was
ob ained. Wi h his ilamen , a solid piece (4 × 10 × 70mm) was manu ac u ed using a FFF 3D-p in e Raise
3D P o 2 (Imp eso as 3D, Spain). The slice so wa e used was ideaMake , using a linea in ill o 100% wi h
an o e lap o 10% be ween beads o gene a e all he g-codes and a nozzle o Ø 0.4mm, p in ing a 250°C
Scien i ic Repo s | (2025) 15:3704 2
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
and 50mm/s. Nea ASA and nea PC we e also ex uded and p in ed, o be used as e e ence ma e ials in he
composi ional analysis. On he o he hand, he ASA/PC ilamen was cu in o pelle s (2–3mm leng h) in a
pelle ize (SCAMEX, F ance) a 71 pm, and hese we e used in IM o p oduce injec ed pa s wi h same size and
geome y as he p in ed objec s in a BabyPlas 10/12 (C onoplas , Spain).
To p epa e samples o FIB analyses, solid specimens we e c yogenically sec ioned using a scalpel a liquid
ni ogen empe a u e. P io o FIB analysis, samples we e spu e -coa ed wi h a conduc i e Au/P laye using a
SCD 004 Spu e Coa e (BAL-TEC, Balze s, Liech ens ein) o mi iga e cha ging e ec s. S uc u al analyses we e
conduc ed using a Scios 2 DualBeam FIB-SEM sys em (The mo Fishe Scien i ic, Wal ham, MA, USA) ope a ing
a 30kV ion accele a ion ol age. Tomog aphic da ase s we e acqui ed using iSE images collec ed wi h an in-lens
SE de ec o . The inal pixel dimensions we e 2010 × 220 and 979 × 1482 o FFF and IM samples, espec i ely.
The imaging pixel sizes we e app oxima ely 3.5 × 5.7nm² o he FFF sample and 7.5 × 9.5nm² (x/y) o he IM
sample. FIB milling was pe o med a 100nm in e als pe slice using a beam cu en o 50 pA, wi h 100 slices
o he FFF sample and 70 slices o he IM sample. The iSE images we e acqui ed using an accele a ing ol age o
30kV and a Ga+ ion cu en o 50 pA. The o al sampled olumes we e 7 × 12.5 × 9.2μm³ and 7.3 × 14.1 × 7μm³
o FFF and IM samples, espec i ely. Image p ocessing included applica ion o a Non-local Means il e using a
cus om MATLAB sc ip . Image s ack alignmen was pe o med using he sum o squa ed di e ences algo i hm
wi hou o a ion in D agon ly so wa e (Objec Resea ch Sys ems, Canada). Edge shadowing a e ac s we e
co ec ed using a adial basis unc ion. Segmen a ion was achie ed using a Deep Lea ning Tool. A U-Ne bina y
model was ained using an inpu da ase o eigh manually segmen ed images, wi h he Adadel a algo i hm o
op imiza ion. Simula ions o elec on and Ga+ ion in e ac ions wi h ma e we e conduc ed using he Casino
so wa e (mon e CA lo Simula ion o elec oN ajec o y in sOlids)24,25 and he SRIM (S opping and Range o
Ions in Ma e ) so wa e26, whe e pa ame e s such as ini ial beam ene gy, a ge ma e ial ype, a ge dimensions,
and incidence angle a e main ained as ixed bounda y condi ions. 10,000 ajec o ies o each ma e ial we e
simula ed o ions and elec ons. The spu e ing a e unde he ion beam has been quan i ied using a co ec ed
ion cu en , al hough nominal cu en s a e epo ed.
Resul s and discussion
Specimen p epa a ion o slice and iew expe imen s
FFF p in ed objec s a e o med by he deposi ion o mel ed ilamen s laye -by laye ollowing a p ede ined
design. Fo he analysis o he ASA/PC dis ibu ion in he polyme blend o he FFF p in ed objec , i is o
in e es o selec he co e o one o he inne beads as he egion o be s udied, and o ca y ou he analysis in he
di ec ion pa allel o his bead. Al hough di e ences in he s uc u al cha ac e is ics maybe expec ed om he
co e o he edge o he beads, he co e is a ep esen a i e egion ha can be used o compa e he mic os uc u e
o di e en ab ica ion pa ame e s and me hods. In o de o gain access o one o he inne beads o he objec
wi hou modi ying he mic os uc u e, he p in ed objec needs o be ac u ed in a agile manne . As he
polyme s ha compose he blend a e duc ile o some ex en , he ma e ial has been cooled in liquid ni ogen o
ob ain a agile ac u e when cu wi h a scalpel. Figu e1a shows he ac u e su ace o one o he inne beads o
he p in ed objec , which p esen s a nicely la su ace esul ing om he agile ac u e. I should be men ioned
ha he sample has been placed in he FIB holde wi h he longi udinal axis o he deposi ed beads pa allel o
he holde su ace. Because o his, o SEM imaging o he leng hwise sec ion o he bead, he holde is il ed
o 52°, esul ing 38° be ween he ec o no mal o he su ace o in e es and he elec on beam; when imaging
wi h he ion beam, he holde is a 0°, esul ing 38° be ween he ion beam and he ec o no mal o he su ace
o in e es (see an schema ic in Suppo ing In o ma ion (SI), Fig. S1). This cla i ica ion is o in e es in ela ion
o he simula ions shown in Fig.3. In o de o be able o pe o m slice and iew expe imen s in a bulk ma e ial,
he olume o in e es needs o be p epa ed o he p ocess. Fo his, ini ially he olume o in e es needs o be
p o ec ed o unin en ional su ace milling, in ou case by a laye o P wi h a hickness o 1.5μm, ob ained by
ion beam deposi ion a a cu en o 0.3 nA using a P -p ecu so gas o a gas injec ion sys em (GIS). In Fig.1a,
his squa e P deposi ion can be obse ed in he cen e o he bead. A e his, he olume o in e es needs o
be pa ially sepa a ed om he bulk ma e ial o a oid bo h edeposi ion o he ma e ial milled du ing he se ial
sec ioning, and possible shading o he signal used o imaging. Al hough a complex block li -ou me hod has
been epo ed as a subs an ial imp o emen o au oma ed 3D se ial sec ioning expe imen s in a FIB27, ha
me hod is ime consuming and, in ou s udy, i is no necessa y when he SE de ec o is used o imaging. In
ou case, a U-pa e n has been milled a ound he egion o in e es as i can be obse ed in Fig.1b and c, using
dec easing cu en s o 7 nA, 3 nA and 1nA o op imize he milling ime while ob aining a p ecise milling in
he egion close o he P deposi ion. On he o he hand, i is wo h men ioning ha he econs uc ion o he
s uc u al ea u es o he ma e ial using a 2D images s ack equi es he p e ious co ec alignmen o each o he
2D images and, o ha , iducial ma ke s a e o g ea help. Two iducial ma ke s ha e been in oduced close o
he egion o analysis, one o hem wi h an o ien a ion op imized o imaging wi h he elec on beam and he
second one o imaging wi h he ion beam. Fo his, a laye o P has been deposi ed, and ma ks (a c oss and a
ci cle) ha e been milled on hem wi h he ion beam, as shown in Fig.1d. A e his, he ma e ial is eady o s a
he slice and iew expe imen . The same p ocess has been ollowed o analyse he IM objec , selec ing as he
egion o in e es he co e o he objec .
Compa ison be ween SEM and iSE images o ASA, PC and he ASA-PC blend
Be o e analysing he ASA-PC blend and in o de o ob ain in o ma ion o he co ec design o he omog aphy
expe imen s, he nea polyme cons i uen s ASA and PC ha e been s udied in 2D in he FIB. Fig.2a shows a 2keV
SEM image o he nea ASA sample. As i can be obse ed, ound con as s can be dis inguished in he milled
la su ace. ASA is a e polyme ha consis s o h ee dis inc monome s: ac yloni ile, s y ene and ac yla e.
The sligh ly c osslinked ac yla e ubbe ( unc ioning as an impac modi ie ) o m sphe ical pa icles, which
Scien i ic Repo s | (2025) 15:3704 3
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
a e chemically g a ed wi h s y ene-ac yloni ile copolyme chains, and embedded in a s y ene-ac yloni ile
ma ix28. Acco ding o his s uc u e, he ound con as s obse ed in he SEM image can be associa ed o he
ac yla e ubbe pa icles. Fig.2b shows a SEM image o he PC sample. As expec ed, he ound ea u es obse ed
in ASA a e no p esen . Howe e , some po es a e ound in he polyme su ace, likely esul ing om he
ex usion p ocess. Fo compa ison, Fig.2c and d show 30keV iSE images o he nea ASA and nea PC samples,
ob ained om he same egions as Fig.2a and b, espec i ely. As i can be obse ed, he ea u es ound by SEM
a e mo e clea ly dis inguished in he iSE images han in he SEM ones. In pa icula , he con as s associa ed o
he ac yla e pa icles a e no iceably sha pe . This is due o he di e en in e ac ion o ions and elec ons wi h
ma e , as i will be discussed nex .
In o de o shed some ligh a he di e ences obse ed be ween SEM and iSE images, Mon e Ca lo simula ions
o he in e ac ion o a 2keV elec on beam and a 30keV ion beam wi h PC ( he da abase o he simula ion
so wa e used does no con ain in o ma ion ega ding ASA) ha e been ca ied ou , and esul s a e shown in
Fig.3. Fo compa a i e pu poses, esul s ob ained o Si as a e e ence ma e ial a e also included. As explained
ea lie , he incidence angle o bo h ions and elec ons wi h he ec o no mal o he su ace when acqui ing he
images included in Fig.2 is o 38°, he e o e his angle has been used o he simula ions. Figu e3a shows he
p ojec ed ange o ions (in ed) and o elec ons (in blue) o bo h ma e ials. As i can be obse ed, he p ojec ed
ange o Ga+ ions in Si is smalle han ha o elec ons, app ox. o 21.9nm s. 40.1nm, espec i ely; he la e al
sp ead o he p ojec ile ions ajec o ies (Fig.3b) is also smalle han ha o elec ons (17.4nm s. 42.3nm,
espec i ely). Due o he high concen a ion o ene gy deposi ion o ions compa ed o elec ons, he SE gene a ed
by ions a e ex ac ed om a signi ican ly smalle olume han hose gene a ed by p ima y elec on beams. This
localized in e ac ion esul s in highe spa ial esolu ion in iSE imaging compa ed o con en ional SEM. In his
sense, analyses epo ed o he pe pendicula incidence o 30keV He ions on Si ha e shown p ojec ed anges
wi h alues in be ween hose co esponding o Ga+ ions and o elec ons, due o hei in e media e mass29. In
PC, he same endency as in Si is obse ed by compa ing he p ojec ed ange (Fig.3a) and he la e al sp ead
(Fig.3c) be ween inciden elec ons and Ga+ ions. He e, i is wo h men ioning ha la ge alues o p ojec ed
ange (64.3nm o inciden elec ons and o 33.4nm o Ga+ ions) and o la e al sp ead (61.7nm o inciden
elec ons and o 26.2nm o Ga+ ions) han hose o Si a e ound. This dependence o he p ojec ile elec ons/
ions ajec o ies wi h he a omic numbe ha e also been epo ed p e iously, in ha case o me allic ma e ials,
in pa icula by compa ison o Au and Al as a ge ma e ials30, whe e he di e ences ound we e la ge han in
Fig. 1. (a) SEM image o a ac u ed bead o he p in ed objec in c oss-sec ion, whe e he squa e P
deposi ion can be obse ed; (b) and (c) SEM images showing di e en s eps o he milling o he U-pa e n
a ound he egion o in e es ; (d) SEM image showing he iducial ma ke s milled nex o he egion o in e es .
The su ace o in e es is delinea ed by a ed dashed line.
Scien i ic Repo s | (2025) 15:3704 4
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
ou wo k due o he la ge di e ence be ween he Z numbe o he me als conside ed. The au ho s show ha o
low Z me als, he h ee componen s o he SE ( hose de i ed om collision o p ojec ile ions, hose caused by
he ecoiled a ge a oms and hose ela ed o cascade elec ons) con ibu e equally o he SE yield, whe eas he
elec on exci a ion by he p ojec ile ions domina es in high Z numbe a ge ma e ials21. Ou esul s ob ained
om he simula ion o he elec ons/ions ajec o ies show ha he spa ial esolu ion in SEM/iSE images o
polyme ma e ials is expec ed o be somewha wo se han in Si due o he lowe Z numbe , bu i is also e iden
ha iSE images would show be e spa ial esolu ion han SEM images, as co obo a ed expe imen ally in Fig.2c
and d.
Figu e4a and b show a SEM and an iSE mic og aph o he ASA/PC p in ed objec , p epa ed o he FIB
analysis ollowing he me hodology explained abo e. As i can be obse ed, in bo h images he con as ela ed
o he sphe ical ac yla e pa icles a e obse ed, appea ing mo e clea ly in he iSE image, analogously o Fig.2.
Al hough he egions wi h hese ound con as s can be in e ed o be ASA, a he sigh o he SEM image i is no
easy o dis inguish he in e ace be ween he wo polyme phases ASA and PC. On he con a y, in he iSE image,
da k and b igh egions wi h a sha p in e ace a e clea ly obse ed, which a e no isible in he SEM image. A
de ailed obse a ion o he image shows ha he sphe ical ac yla e pa icles a e p esen in he da k phase o he
blend, indica ing ha his da k phase co esponds o ASA, whe eas he b igh e one co esponds o PC. EDX
analysis ha e been used o co obo a e his iden i ica ion. Fo his, ini ially, EDX analysis o he pu e ASA and
pu e PC samples ha e been ca ied ou , and spec a a e included in Fig.4c. As i can be obse ed, oge he wi h
he C and O signals, he N signal ha s ems om he ac yloni ile monome s is clea ly de ec able in he EDX
spec a o ASA, whe eas N is absen in PC, as expec ed (see chemical o mulas o hese polyme s in Fig. S2).
This di e ence in he chemical composi ion is o g ea help o dis inguish he wo phases in he analysis o he
polyme blend. Figu e4d shows an iSE image o he blend, oge he wi h he N, O and C EDX composi ional
maps. As i can be obse ed, he da k egions in he iSE image a e N ich, whe eas he b igh e egions a e O ich,
which is in good ag eemen wi h he i s one co esponding o ASA, and he second one o PC. Quan i a i e
alues o he elemen s composi ion in each ma e ial a e included in SI, in Table S1.
Di e ences in con as in SEM and iSE images ob ained om he same egion o me allic specimens ha e
been epo ed be o e29,31, and his has led o in es iga ions o he image o ma ion mechanisms in bo h imaging
modes. The in e ac ion o ene ge ic ions o elec ons wi h solids is complex as hey could su e a numbe o
sca e ing e en s including elas ic collisions wi h he nuclei o he specimen a oms, which modi y hei a el
di ec ion, and inelas ic collisions, leading o gene a ion o SE, Auge elec ons, pho ons, elec on hole pai s, and
o he s. The in e p e a ion o bo h SEM and iSE images equi es knowledge o he SE yields o he ma e ial o
Fig. 2. 2keV SEM images o he nea ASA (a) and he nea PC samples (b); 30keV iSE images o he nea ASA
(c) and nea PC (d) samples, ob ained in he same egions as (a) and (b), espec i ely.
Scien i ic Repo s | (2025) 15:3704 5
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/

in e es , as a unc ion o he inciden beam ene gy. Howe e , iSE yield da a is only abula ed o a ew elemen s
and a e y ew compounds, wha limi s he in e p e abili y o he expe imen al images. Fo ixed acquisi ion
pa ame e s, his SE yield (and, hus, he con as ) in iSE images depends, mainly, on: (a) he a omic numbe
o he a oms in he sample, (b) he angle o incidence o he ion beam ela i e o he sample su ace no mal
(i.e., opog aphy con as ) and (c) he channelling o he inciden ions be ween la ice planes o he ma e ial
(use ul o he iden i ica ion o g ains in polyc ys als)32. In ou case, i is he i s ac o , he di e ence in SE
yield associa ed o he a omic numbe o he ma e ial, he mos likely cause o he di e ence in con as be ween
he phases obse ed in he iSE images, as he imaged milled ace o he specimen has a la opog aphy and he
exis ence o c ys alline g ains is no expec ed in he polyme ic blend. Howe e , he e is some con o e sy on he
ela ion o he iSE yield and he a omic numbe o he s udied ma e ial. Mon e Ca lo simula ions o ion-induced
kine ic elec on emission ha e p edic ed ha he iSE yield o di e en me als would dec ease wi h an inc ease in
he a omic numbe o he a ge ma e ial o 30keV Ga+ ion bomba dmen 33,34. On he con a y, expe imen al
analyses ha e shown a non-mono onic beha iou o he iSE con as wi h Z35. Addi ionally, in a di e en epo
also using Mon e Ca lo simula ions o ion–solid in e ac ions, i has been also shown ha he 30keV Ga+ iSE
signal is a non-mono onic unc ion wi h espec o he a omic numbe . This is due o he in luence o bo h he
s opping powe (which is de ined as he ene gy loss pe uni dis ance o any cha ged pa icle wi hin a a ge ) and
he spu e yield36. Those analyses a e ela ed o pu e me als, wi h a speci ic a omic numbe associa ed o each
elemen . In he case o polyme blends, al hough hey a e based on C, hey also con ain o he elemen s, such as
H, O and/o N. Addi ionally, he dis ibu ion o a oms in a olume necessa y o calcula e an a e age Z numbe
depends o he a angemen o he polyme chains, including he c ys allini y deg ee. Al hough polyme s o
FFF a e mos ly amo phous, in he case o polyme blends he densi y o he componen s seems o be a mo e
easonable pa ame e o be conside ed in o de o unde s and hei con as . As i can be obse ed in he iSE
image in Fig.4b, he e is a clea con as be ween he PC phase and he s y ene-ac yloni ile ma ix o ASA, being
he second one da ke . The densi y o PC is 1.2g/cm3, whe eas he a e age densi y o he s y ene (d = 0.909g/
cm3)-ac yloni ile (d = 0.806g/cm3) ma ix is smalle . Ou esul s sugges ha his ela i ely small di e ence
in densi y could be he eason o he di e en iSE yield, demons a ing a la ge sensi i i y o iSE imaging o
he analysis o polyme blends. In he li e a u e, low- ol age (5kV) ansmission elec on mic oscopy wi h a
no el cons uc ion combining isual-ligh and elec on mic oscopical echniques has been epo ed use ul o
Fig. 3. Mon e Ca lo simula ed elec on and Ga+ ion ajec o ies o incidence angle o 38° in Si and in PC. (a)
P ojec ed ange o ions (in ed) and o elec ons (in blue) o Si (con inuous line) and PC (dashed line). Ions
(in ed) and elec ons (in blue) ajec o ies in (b) Si and (c) PC.
Scien i ic Repo s | (2025) 15:3704 6
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
he analysis in 2D o polyme blends, including PC / poly(s y ene-co-ac yloni ile) (SAN)37. In his case, he
di e ence in densi y be ween he indi idual componen s o polyme blends is also sugges ed o be he eason o
he ob ained image con as (indica ing ha di e ences less han 0.04g/cm3 can be aced wi h his echnique),
al hough he au ho s claim ha he mechanisms o he imaging con as a e no ully unde s ood.
iSE omog aphy o ASA-PC blends
A he sigh o he esul s abo e, i seems o in e es o ca y ou he omog aphy expe imen s using he iSE
images ins ead o he classical SEM images. In o de o ob ain u he in o ma ion on he beha iou o he wo
polyme componen s o he blend unde he Ga+ ion beam o se adequa e pa ame e s o he omog aphy s udy,
he spu e ing a e unde he ion beam has been quan i ied o bo h nea ASA and nea PC. Fo his, 5μm x 5μm
c a e s ha e been spu e ed in ASA and in PC by scanning he Ga+ beam a no mal incidence using he ape u es
co esponding o cu en s o 0.05 nA, 0.1nA and 1 nA, and imes o 100s, 200s and 300s. Figu e5a shows a SEM
image o an ASA polyme su ace coa ed wi h a conduc i e coa ing ( o p e en cha ging e ec s) a e c ea ing
a c a e wi h ion beam milling (addi ional images o some o hese c a e s a e included in Fig. S3 in SI). In he
cen al a ea o he image, he cubic-shaped c a e can be obse ed, and a ound he c a e , he conduc i e coa ing
has pa ially been elimina ed. This e ec is due o FIB la e al beam sp eading. When he ion beam is ocused on
a speci ic a ea, he ions also sp ead a ound his a ea, expanding he zone a ec ed by he ions. This e ec depends
on a ious pa ame e s, including he beam cu en - highe cu en s esul in la ge la e ally a ec ed a eas38. The
emo al a e has been calcula ed by di iding he c a e olume by he Ga+ dose equi ed o emo e ha olume,
and he esul s a e plo ed in Fig.5b. As i can be obse ed, he spu e ing a e is qui e simila o bo h polyme s.
This is e y bene icial in o de o ca y ou a success ul iSE omog aphy, as i ensu es ha he milled aces will
be la du ing he expe imen and, hus, ha he analysis will allow a ue econs uc ion o he 3D dis ibu ion
o he polyme s in he blend. Fo au oma ized iSE slice and iew se ies, i is desi able ha he same cu en is
used bo h o imaging and o milling, o a oid he need o mul iple e-alignmen s, which would ex end he
du a ion o he expe imen . Because o his, a comp omise needs o be eached in he ion cu en used o educe
he damage du ing he imaging s ep while allowing an e ec i e slice milling. Fo he polyme blend conside ed
in he p esen s udy, we ha e selec ed a comp omise alue o 0.05 nA o he omog aphy expe imen . Fig. S4
shows iSE images aken wi h his cu en , a dwell imes o 0.2 µs, 3 µs and 20 µs. As i can be obse ed, ea u es
Fig. 4. (a) SEM and (b) iSE images o he ASA/PC blend ob ained by FFF; (c) EDX spec a ob ained om he
pu e ASA and pu e PC samples; (d) iSE image o he ASA/ PC blend ob ained by FFF, and N, O and C EDX
composi ional maps ob ained in he same egion.
Scien i ic Repo s | (2025) 15:3704 7
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
a e clea e o la ge dwell imes, as he signal o noise a io is expec ed o be la ge . Howe e , damage due o he
ion beam is also la ge o la ge dwell imes. Fo dwell imes o 0.2 µs, 3 µs and 20 µs, in images wi h esolu ion
o 2809 × 3391 pixels, and conside ing he spu e ing a es shown in Fig.5b, dep hs o app ox. 1nm, 16nm
and mo e han 100nm, espec i ely, a e emo ed pe iSE image acquisi ion. In o de o ind a balance be ween
he quali y o he images and he damage p oduced, and a he sigh o he size o he ea u es obse ed in he
iSE image in Fig.4b, 3 µs seems o be a easonable dwell ime o he slice and iew expe imen , as his su ace
damage seems no o ha e a s ong dele e ious e ec on he esul s o he 3D polyme s dis ibu ion ob ained.
Using he pa ame e s de e mined abo e (0.05 nA and 3 µs) we ha e ca ied ou slice and iew expe imen s o
compa e he polyme s dis ibu ion in he pieces manu ac u ed by FFF and by IM. Figu e6 shows ep esen a i e
2D iSE images o he ASA50 / PC50 blends ob ained du ing he slice and iew expe imen s. In Fig. 6a,
co esponding o he ASA50 / PC50 manu ac u ed by FFF, a su ace wi h diagonally o ien ed con as s is
obse ed. As p e iously men ioned and co obo a ed by EDX, he clea and homogeneous con as s co espond
Fig. 6. iSE images o ASA50 / PC50 o he pieces ab ica ed by (a) FFF and (b) IM.
Fig. 5. (a) SEM image o a c a e milled in ASA wi h he ape u e co esponding o a cu en o 0.1 nA; (b)
emo al a e (µm3/nC) s. ime o di e en ion cu en s, o ASA and o PC.
Scien i ic Repo s | (2025) 15:3704 8
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/
o PC, while he da k con as s wi h dispe sed ound ea u es co espond o ASA. Some small po es (black spo s)
a e obse ed in he ac yla e ubbe o ASA likely due o he e ec o ions, as hese a eas may be mo e sensi i e
o Ga+ ions bomba dmen . As i can be obse ed in Fig.6a, he polyme s a e no comple ely mixed a he
molecula le el, i.e., a miscible s a e has no been eached as he wo sepa a e phases can be clea ly dis inguished.
The isualiza ion in 2D o he o ien a ion o hese phases sugges s a di ec ional s uc u e and good dispe sion,
whe e he wid h o he domains is a ound 1 mic ome e . Figu e6b shows a ep esen a i e iSE image o he
ASA/PC blend manu ac u ed by IM, showing a clea sepa a ion be ween wo dis inc egions. A la ge clea
and homogeneous a ea co esponding o PC occupies mos o he image, while ASA appea s as da k ci cula
con as s. In his case, he polyme indi idual domains a e much la ge han in he p e ious case, and hei
phase dis ibu ion is mo e he e ogeneous, wi h la ge and small egions o ASA close o each o he . Howe e , ca e
should be aken in he in e p e a ion o indi idual 2D c oss sec ions o 3D ea u es, as some imes esul s can
be misleading. Fo example, he small ound egions o ASA obse ed in Fig.6a could be pa o a much la ge
egion, as his 2D sec ion does no p o ide accu a e in o ma ion on he size o he domains. Because o his, a
ca e ully analysis o he 3D da a ob ained in he slice and iew expe imen s has been ca ied ou .
Figu e7a and b show he omog aphic econs uc ions o he wo blends analysed. Fo cla i y, only ASA is
ep esen ed, and di e en colou s a e used o indica e indi idual (unconnec ed) zones. Rega ding he FFF objec
(Fig.7a, he 3D analysis ca ied ou co obo a es he la ge di ec ionali y in he dis ibu ion o he polyme s
obse ed in 2D in Fig.6a. The dis ibu ion o he phases ollows he longi udinal axis o he analysed bead (y axis
in Fig.7a), wi h a small de ia ion likely due o he impe ec cu o he ma e ial piece p io o he FIB analysis,
which was in ended o be pe pendicula o he bead longi udinal axes. Addi ionally, he 3D analysis ca ied
ou shows ha he blend p esen s a co-con inuous mo phology, meaning ha each phase o ms a 3D spa ially
con inuous ne wo k h oughou he ma e ial. By obse ing a single con inuous phase o ASA, o example he
one ep esen ed wi h blue colou , a la ge deg ee o co-con inui y (de ined as he ac ion o ma e ial ha belongs
o a con inuous ne wo k)39 can be in e ed. This polyme s dis ibu ion is qui e di e en o ha ound in he
IM blend, shown in Fig.7b. The mo phology, in his case, exhibi s a d ople -ma ix s uc u e, i.e., PC o ms a
con inuous ma ix, whe e ASA o ms dispe sed d ople s in ha ma ix. This mo phology has also been obse ed
in polyme blends o PC / SAN, polys y ene/polyp opylene (PS/PP), and PE / PP by 2D low ol age TEM37. The
size dis ibu ion o he ASA d ople s ound in Fig.7b is included in he his og am in Fig.7c, showing ha mos
Fig. 7. 3D olume ep esen a ion o he ASA egions in he ASA/PC blends ab ica ed by (a) FFF and (b) IM.
The colo s show he independen olumes o ASA su ounded by PC. (c) His og am o size dis ibu ion o he
ASA egions in he objec ab ica ed by IM.
Scien i ic Repo s | (2025) 15:3704 9
| h ps://doi.o g/10.1038/s41598-025-87364-2
www.na u e.com/scien i ic epo s/