Influence of manufacturing process in pores, f ibredistribution and mechanical properties of ASA-CFfor 3Dlargeformatadditive manufacturing pieces

Influence o manu ac u ing p ocess in po es,
fib e dis ibu ion and mechanical p ope ies o
ASA-CF o 3D la ge o ma addi i e
manu ac u ing pieces
Daniel Mo eno Sánchez
Depa 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, C
adiz, Spain
Jesús He n
andez Saz
Depa amen o de Ingenie ía y Ciencia de los Ma e iales y del T anspo e, Uni e sidad de Se illa, Se illa, Spain, and
Jos
e Ja ie Relinque Mad oñal, Ped o Bu gos Pin os, Se gio I. Molina Rubio and Mi iam He e a Collado
Depa 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, C
adiz, Spain
Abs ac
Pu pose –The pu pose o his s udy was o e alua e pa s manu ac u ed by used g anula ab ica ion (FGF) using wo di e en ac yloni ile s y ene
ac yla e (ASA) compounds ein o ced wi h ca bon fibe (CF) and compa ed hem wi h hose p oduced by injec ion molding (IM).
Design/me hodology/app oach –The di e ence in bo h ASA_CF composi es is he mixing p ocess, achie ed using win-sc ew ex ude s wi h
di e en mixing shea a es. The po osi y p opo ion and dis ibu ion, and he CF leng h and alignmen , ha e been ca e ully quan ified and
compa ed using X- ay compu e ised omog aphy (CT). Also, he co ela ion wi h mechanical p ope ies has been e alua ed o FGF and o IM pa s
in bo h ma e ials.
Findings –Rein o cemen o polyme ic ma ices wi h di e en fib es is a common s a egy o imp o e he mechanical p ope ies o 3D p in ing
pieces, especially ele an in FGF. The manu ac u ing p ocess o hese composi es needs o be ca e ully designed o op imise he s uc u al and
mechanical cha ac e is ics o he ma e ial. CT analyses show ha he use o highe shea mixing a es inc eases he amoun o po es wi hin he
polyme ic pelle s and p oduces a smalle fib e size dis ibu ion in he composi es, which nega i ely a ec s he mechanical pe o mance o he final
pa .
O iginali y/ alue –To he bes o he au ho s’knowledge, his is he fi s ime ha a global s udy o he comple e ab ica ion p ocess has been
ca ied ou , p o iding de ailed in o ma ion o e e y s ep o he p ocess.
Keywo ds Design, Mechanical cha ac e isa ion
Pape ype Technical pape
1. In oduc ion
Addi i e manu ac u ing (AM) comp ises a se o echnologies
ha allow objec s o be ma e ialised laye by laye di ec ly
om a digi al file (Ga dan, 2016;Ngo e al.,2018;To ail
e al.,2018). In ecen yea s, hese echnologies a e gaining
pa icula a en ion because o he ad an ages hey o e
compa ed o adi ional manu ac u ing p ocesses (i.e.
sub ac i e and con o ma i e), such as eedom o design,
pa cus omisa ion o in en o y educ ion, o name a ew
(A a an, 2017).
Amongs AM echnologies defined by ASTM (ASTM-
F42.91 Subcommi ee, 2015), hose classified in he ma e ial
ex usion (ME) ca ego y a e he mos widesp ead because o
hei low cos , scalabili y, a ailabili y o ma e ials and ease o
use. In ME, he ma e ial is o ced o pass h ough a nozzle, and
he laye s and pa s a e c ea ed ollowing .gcode pa hs. Fo
polyme s, he main ME echnologies lie be ween used filamen
The cu en issue and ull ex a chi e o his jou nal is a ailable on Eme ald
Insigh a : h ps://www.eme ald.com/insigh /1355-2546.h m
Rapid P o o yping Jou nal
31/11 (2025) 218–230
Eme ald Publishing Limi ed [ISSN 1355-2546]
[DOI 10.1108/RPJ-03-2024-0103]
© Daniel Mo eno Sánchez, Jesús He n
andez Saz, Jos
e Ja ie Relinque
Mad oñal, Ped o Bu gos Pin os, Se gio I. Molina Rubio and Mi iam
He e a Collado. Published by Eme ald Publishing Limi ed. This a icle is
published unde he C ea i e Commons A ibu ion (CC BY 4.0) license.
Anyone may ep oduce, dis ibu e, ansla e and c ea e de i a i e wo ks o
his a icle ( o bo h comme cial and non-comme cial pu poses), subjec o
ull a ibu ion o he o iginal publica ion and au ho s. The ull e ms o his
license may be seen a h p://c ea i ecommons.o g/licences/by/4.0/legalcode
Recei ed 31 Ma ch 2024
Re ised 9 July 2024
15 Sep embe 2024
Accep ed 25 Sep embe 2024
218
ab ica ion (FFF), commonly known as used deposi ion
modelling (FDM), and used g anula ab ica ion (FGF). The
mos p ominen di e ences be ween hese wo sys ems a e he
o ma o he aw ma e ial and he size o he p in e s. FFF
sys ems a e ed wi h a polyme filamen ha is mel ed a he ip
o he nozzle and ex uded h ough i , whe eas FGF is ed wi h
pelle s ha a e mel ed inside a single-sc ew ex ude . Also, FFF
p in e s a e usually in ended o small- o ma ab ica ion (less
han 1 m
3
), while FGF p in e s a e gene ally concei ed o
p oduce pa s la ge han 20 m
3
. In he indus ial sec o , FGF
is e y p omising, as i allows c ea ing la ge pa s in a sho ime
wi h a con inuous flow o ma e ial. Fo example, hese
echnologies ha e been success ully used o c ea e a boa oile
(Mo eno Nie o e al., 2018), ca s (Cu an e al.,2016) and
moulds o boa s (Pos e al.,2019) o wind u bine blades
(Pos e al.,2017). Because o hei capabili ies in la ge o ma ,
hese echnologies a e o en known as la ge o ma addi i e
manu ac u ing (LFAM), big a ea addi i e manu ac u ing o
la ge-scale addi i e manu ac u ing.
As in o he manu ac u ing p ocesses, he ma e ials used in
FGF a e equi ed o ulfilspecific equi emen s. Among o he s,
hey need o ha e a low coe ficien o he mal expansion (CTE)
and a p ope heology (a Mel ing Flow Index, be ween 10 and
20 g/10 min) (Mo eno Nie o e al.,2018). FGF p in e s usually
wo k wi h no hea ed chambe ; hus o ob ain sui able pa s, he
inco po a ion o fib es is a common p ac ice o educe CTE, hus
p e en ing dis o ions du ing p in ing and a e cooling. Ca bon
fib e (CF) and glass fib e appea in he li e a u e o be he mo e
used fib es o he desc ibed pu pose (Billah e al., 2020;
Economidou and Ka alekas, 2016;Lo e e al., 2014;S
anchez
e al., 2020a). Mo eo e , he addi ion o fib es inc eases he
s eng h o he ma e ial and s i ness; howe e , i also u ns he
ma e ial om duc ile o agile and induces he o ma ion o
po osi y (Vaxman e al.,1989). These las e ec s may esul in a
p ema u e b eakage o he pa s. Addi ionally, i has been
epo ed ha he p esence o fib es hinde s he adhesion be ween
laye s in p in ed pa s; he e o e, a comp omise be ween
pe o mance and he amoun o fib es needs o be achie ed
(S
anchez e al., 2020a). The quali y o he aw ma e ial in e ms o
inhomogeneous in eg a ion o fille s o inne de ec s (i.e. po osi y
o debonding be ween fille and ma ix) also migh a ec he final
p ope ies o p in ed pieces. Rega ding ma e ials, ac yloni ile
bu adiene s y ene (ABS) (Billah e al., 2020;Hassen e al., 2016;
Hill e al.,2018;Rod íguez e al.,2003), polyphenylene
sulphide (Liu e al.,2018), poly e he ke one (Tseng e al.,
2018)o polye he imide(Ajinje u e al.,2017) ein o ced wi h
CF ha e been mos used in FGF.
Ac yloni ile s y ene ac yla e (ASA) is an amo phous
e polyme , which combines excellen esis ance o ul a iole
(UV) ays, good mechanical beha iou , good esis ance o
chemicals and low CTE (John Schei s, 2003;Oepen and
Go schalk, 2011;Xiang e al.,2017). The chemical s uc u e
o ASA consis s o alkyl ac yla e pa icles (elas ic componen ),
made compa ible by a s y ene, co-ac yloni ile coa ing,
homogeneously dispe sed in a polys y ene co-ac yloni ile
ma ix ( igid componen ). I is a polyme ha does no p esen
double ca bon bonds in i s molecula s uc u e, which is he
eason why i p esen s such s able beha iou agains
deg ada ion unde UV ays (John Schei s, 2003;Moghbeli and
Tolue, 2011;Tolue e al.,2009;Yu e al.,1974). I is a polyme
li le s udied in li e a u e, p obably o i s simili ude wi h ABS,
ha was ecen ly in oduced in FGF, p incipally ein o ced
wi h glass o ca bon fib es o c ea e ex e io pa s (S
anchez
e al., 2020a).
The imp o emen in mechanical p ope ies and educ ion in
CTE by fib es is s ongly a ec ed by he adhesion fib e-ma ix,
leng h o fib es, po osi y, and ab ica ion p ocess, among o he
aspec s (Li e al.,2022;Spoe k e al., 2018;Tekinalp e al.,2014),
which de e mines a h eshold in he e ec i e w .% o fib e added
o polyme ma ices. Upon p e ious in es iga ions by he au ho s
using ASA_CF i was concluded ha 20–30 W .% o CF addi ion
was compa ible wi h mechanical p ope ies enhancemen
(S
anchez e al., 2020b). In ha wo k, ASA_20CF composi es
we e s udied because highe CF a es did no significan ly
imp o e he mechanical pe o mance while inc easing he
p obabili y o nozzle clogging in FGF p in ing because o
he agglome a ion o fib e. O he s udies in he li e a u e
confi med ha he a o emen ioned ange o CF w .% is he
mos sui ed o imp o e mechanical p ope ies (Du y e al.,
2017;Lo e e al.,2015;Solu ions, 2015;Yasa and E soy,
2018).
On he o he hand, X- ay compu ed omog aphy (CT) is
gaining a en ion in he analysis o polyme composi es as i
p o ides de ailed in o ma ion on he phase dis ibu ion o
he ma e ial wi hou al e ing he s uc u e. This non-
des uc i e echnique can econs uc in 3D he inne
s uc u e o ma e ials, p ocu ing he necessa y co ela ion
be ween he mic os uc u e o he p in ed pa s and hei
p ope ies. In his sense, X- ay CT has been used o analyse
he dis ibu ion o fib esandpo osi yinAMinp e ious
wo ks; o example, Sa andaiah e al. (2021) s udied he
dis ibu ion o po es and fib es in polyp opylene (PP) in ME;
Yang e al. (2021) s udied he dis ibu ion o he fib es
h ough he nozzle; o Kelese al. (2018) e alua ed he
dis ibu ion o fib eand hep esencepo osi yinABS,
among o he ele an e e ences.
In his wo k, he s uc u al cha ac e is ics o ASA and
ASA_CF composi es we e quan i a i ely analysed by X- ay CT
and co ela ed o he mechanical p ope ies o 3D-p in ed final
pieces. In pa icula , he dis ibu ion and leng h o fib es and he
po osi y du ing di e en s ages o he manu ac u ing p ocess a e
s udied: om he p oduc ion o pelle s ob ained by wo di e en
ex usion lines o he subsequen ab ica ion o pa s using bo h
FGFandIM.IMisoneo hemos widesp eadp ocesses o he
indus ial ab ica ion o he moplas ics (Kazme e al.,2023).
Because AM echniques a e claimed o ulfil he cu en
en i onmen al awa eness o he con en ional manu ac u ing
p ocesses, he ou pu assessmen o al e na i e manu ac u ing
echniques is usually d i en h ough he compa ison wi h a
con en ional echnique such as IM, as confi med by he
li e a u e (Bezzina and Re alo, 2023;Cu mi and Rochman,
2024;de To o e al.,2020). This compa ison highligh s p os
and cons among con en ional and al e na i e manu ac u ing
me hods, finally concluding he easibili y o he use o he
la e .
To he bes o he au ho s’knowledge, hisis hefi s ime ha a
global s udy o he comple e ab ica ion p ocess has been ca ied
ou , p o iding de ailed in o ma ion o e e y s ep o he p ocess.
The esul s de i ed om his wo k highligh he significan e ec o
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
219
he manu ac u ing p ocess on he cha ac e is ics and pe o mance
o 3D final pa s.
2. Ma e ials and me hods
Two di e en ASA_CF composi es (23W .% CF), as well as
nea ASA used as a e e ence, we e es ed in his esea ch. All
ma e ials we e pu chased a Ma e sia S.L. (Spain). ASA
composi es di e in he compounding condi ions: he fi s one
will be e e ed o as low shea mixing (LS), wi h a 1.2m long,
40°C wa e ba h cooling sys em; and he second one as high
shea mixing (HS) wi h a 4 m long wa e ba h a 15°C.
P io o any use (i.e. IM o FGF), aw ma e ials and
composi es we e d ied a 80°C o a leas 4 h in a PIOVAN
DPA30 comp essed ai dehumidifie . Figu e 1 summa ises he
expe imen s ca ied ou in his communica ion, and Table 1
ga he s he a iables s udied in his esea ch: ma e ials (ASA and
ASA_CF); compounding condi ions (LS and HS); and
manu ac u ing p ocesses o he samples (IM and FGF).
LFAM pieces we e ab ica ed in a single-sc ew FGF p in e ,
Supe Disco e y 3D P in e Compac . A empe a u e p ofile o
250°C–260°C ( om op o nozzle) and a hea ing bed o 100°C
we e used. Pieces in ended o s uc u al cha ac e isa ion by
X- ay CT scanning and o mechanical cha ac e isa ion by
means o ensile es ing we e p epa ed. The ab ica ion o pa s
ollowed a wo-s age p ocedu e. In he fi s s age, wo-laye ed
ec angles wi h an A4 dimension we e deposi ed on o he p in e
pla o m. The infill was 100%, and a 2-mm nozzle was used wi h
a hickness laye o 1 mm. In he second s age, ensile es and
p ims specimens we e cu om hose p in ed shee s by using a
desk op CNC machine. Samples o s uc u al cha ac e isa ion
we e ec angula p isms wi h dimensions 5 mm 5mm2 mm,
whe eas ensile specimens we e ab ica ed acco ding o he
UNE-EN-ISO 527 s anda d ( ype 1BA specimens). The ensile
specimens we e cu in X o ien a ion which means ha he
deposi ion o he beads was in pa allel o he di ec ion o load
applica ion in he ensile es ing.
Addi ionally, ensile es specimens p oduced by IM we e
p epa ed using a BabyPlas 10/12 injec o a 250°C–260°C
(plas icise , chambe and nozzle) wi h 60 and 80 ba in injec ion
and compac ion p essu e, espec i ely. X- ay CT samples we e
also cu om injec ed pa s o a p ope compa ison.
X- ay CT scanning was ca ied ou using a Zeiss X adia 610
Ve sa X- ay mic oscope. A leas wo scans we e eco ded o
mos o he samples. Ini ially, a la ge field o iew scan was
pe o med o ob ain in o ma ion abou he gene al ea u es o he
ma e ial, wi h an op ical esolu ion o 0.39. Then, a highe
esolu ion scan was pe o med o analyse in de ail he
mic os uc u al ea u es o each ma e ial (op ical esolu ion
3.97). The wo king accele a ing ol age was 60kV. A o al o
1,000–2,000 2D p ojec ions we e eco ded as he sample was
o a ed by 360° and compu a ionally econs uc ed ia a fil e ed
back p ojec ion algo i hm (Zeiss XM Recons uc o ). To
isualise he in e nal ea u es inside he sample in 3D space,
so wa e D agonFly was used. Tomog ams we e impo ed in o
Figu e 1 Summa y o he expe imen s pe o med in his esea ch
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
220
he A izo so wa e sui e o analyse he CF o ien a ion and leng h
and po e dis ibu ion o he pieces. Fo his, ea u es in he
composi es we e segmen ed using a g eyscale h eshold applied
o he abso p ion con as da a. Small isola ed egions in he
segmen a ion a ising om noise in he scan da a we e emo ed.
The ensile p ope ies o he di e en composi es we e
de e mined using a Shimadzu AGS-X uni e sal es ing
machine, a ached wi h a 10 KN load cell and sc ew ac ion
g ips, shown in Figu e 1(h). The es s we e pe o med a a
cons an speed o 1 mm/min, wi h da a acquisi ion e e y 0.1s,
using T apezium X so wa e. The g ips we e p ope ly igh ened
manually o p e en he specimens om slipping du ing he
es . The esul s we e exp essed as an a e age o 5 expe imen s
acco ding o UNE EN ISO 527.
3. Resul s and discussion
3.1 Po e analysis
Ini ially, pelle s om bo h compounding lines and nea ASA we e
analysed be o e u he ab ica ion using IM o FGF. Figu e 2(a)
shows a ende ed olume o ASA_CF_LS_Pelle , and Figu e 2
(d) includes a magnified pa o a 2D slice om he 3D g eyscale
da a. As i can be obse ed, a homogeneous dis ibu ion o CF
can be obse ed in he ma e ial (whi e con as s), some o hem
ma ked wi h ed a ows in Figu e 2(d). In e es ingly, CF appea
mainly as ci cula con as s, indica ing ha he fib es a e
obse ed in c oss sec ion. Thus, CF we e mos ly aligned pa allel
o he ex usion di ec ion, as should be expec ed (Fallon e al.,
2019;S
anchez e al., 2020a;Tekinalp e al., 2014). A en ion will
be paid o he CF o ien a ion and leng h u he on. Addi ionally,
i is wo h no ing ha some po es a e ound in he ma e ial,
loca ed p e e en ially a he cen al pa o he pelle . The po e
olume pe cen age calcula ed by aking he a io o he o al
segmen ed po e olume o he o al in es iga ed sample olume
has been quan ified as 4.9%. In he case o ASA_CF_HS_Pelle
(high shea mixing condi ions), Figu e 2(b) shows a
econs uc ion o he CT da a o he pelle , and Figu e 2(e)
includes a 2D slice co esponding o his se o da a. The po e
dis ibu ion is di e en in his composi e. In spi e o some deg ee
o po osi y s ill being pe cep ible a he cen al pa o he pelle , i
is no as conspicuous as in he low shea p ocessed pelle ;
addi ionally, he size o he po es is ema kably smalle (see a
compa a i e his og am o he size dis ibu ion in supplemen a y
in o ma ion Figu es S1 and S2). In addi ion, po es a e ound up
o he edge o he pelle , whe eas in he low shea p ocessed
composi e, almos no po es we e ound in his egion. The
amoun o po es in ha ma e ial has been quan ified as 10.1% in
olume, which is app oxima ely double ha in he p eceden
case.
To unde s and whe he his po osi y is ela ed o he
in oduc ion o he CF in he polyme , a e e ence sample o
nea ASA was also analysed by CT, and a 3D econs uc ion o
he da a is shown in Figu e 2(c) and ( ) . As i can be obse ed,
no po es a e p esen in he ASA_Pelle , e idencing ha he
Table 1 Classifica ion o he ma e ials s udied, iden i ying hei compounding
condi ions and manu ac u ing p ocess
Denomina ion Compounding condi ion Manu ac u ing p ocess
ASA_Pelle ––
ASA_IM –Injec ion moulding
ASA_FGF –Fused g anula ab ica ion
ASA_CF_LS_Pelle Low shea mixing –
ASA_CF_LS_IM Low shea mixing Injec ion moulding
ASA_CF_LS_FGF Low shea mixing Fused g anula ab ica ion
ASA_CF_HS_Pelle High shea mixing –
ASA_CF_HS_IM High shea mixing Injec ion moulding
ASA_CF_HS_FGF High shea mixing Fused g anula ab ica ion
Sou ce(s): Table by he au ho s
Figu e 2 Rende ed olumes o ASA_CF_LS_pelle (a), ASA_CF_HS_pelle (b) and nea ASA_pelle (c); 2D slices o he ASA_CF_LS_pelle (d),
ASA_CF_HS_pelle (e) and nea ASA_pelle ( )
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
221
po osi y s ems om he in oduc ion o he CF in he polyme .
The p esence o po osi y in polyme s ein o ced wi h fib es has
been asc ibed o a double e ec : ai en apmen du ing mixing
and sh ink- oid by an une en con ac ion du ing cooldown
(T on oll e al.,2018;Vaxman e al.,1989). In he p esen case,
po es a e no loca ed a ound he CF, as i is usual in fib e-
con aining composi es because o di e en he mal expansion
o he phases in ol ed (Kuma e al.,2021;Vaxman e al.,
1989). Ins ead, a he edge o he pelle , CF wi hou po es can
be clea ly obse ed in ASA_CF_LS_Pelle . The absence o
po es a ound he fib es is associa ed wi h a p ope in eg a ion
o he fib es inside he polyme s (S
anchez e al., 2020a;Song
e al.,2016).
The amoun o po es migh no only be ela ed o en apped
ai inside he ma e ial du ing mixing. In ASA_CF_LS_Pelle , he
p esence o po es in he middle o he pelle s could also be
a ibu ed o a slow cooling down o he ma e ial, p oducing he
las con ac ion o he polyme in he cen e o he pelle
(Rauwendaal, 2010). In ASA_CF_HS_Pelle , he dis ibu ion o
small po es in he ma e ial migh be because o he as e cooling
down o he ma e ial (4m long in 15°C wa e in HS, ins ead o
1.2m long in a 40°C wa e ba h o LS configu a ion). A highe
cooling a e se s he geome y o he ex uded ma e ial quickly.
A e die swell e ec (expansion o ma e ial by he p essu e
di e ence be ween he in e io o he ex ude and a mosphe ic
p essu e), he polyme sh inks almos homogeneously in a non-
p e e en ial a ea and he mic opo osi y gene a ed appea s in his
andom dis ibu ion. Also, o he aspec s, such as a longe
esidence ime inside he ex ude line and he p esence o fib es,
migh a ec he nuclea ion o his mic opo osi y in his second
configu a ion.
Gi en ha such po osi y appea s no o be a ound he fib es
and he ac o obse ing smalle po es homogeneously
dis ibu ed in ASA_CF_HS_Pelle han in ASA_CF_LS_Pelle ,
hese obse a ions necessa ily migh be linked o he se e i y o
he ASA/CF mixing p ocess. As a ma e o ac , he ex usion
line used o p oduce ASA_CF_HS_Pelle includes mo e mixing
elemen s and p ocu es a highe esidence ime han he ex usion
line used o p oduce ASA_CF_LS_Pelle ; he e o e, a highe
shea a e can be expec ed o be p o ided by he fi s ex usion
line. This shea a e should be esponsible o he po osi y size
dis ibu ion and hei loca ion. Because he shea a e o he HS
line is highe han in he LS line, po es a e no allowed o coalesce
in he inne pa o he pelle , as obse ed in he ASA_CF_LS
composi e. Con e sely, a highe shea a e allows a be e
dis ibu ion o he po es h ough all he olume, hus educing i s
a e age size. Conside ing he ound shape o he obse ed po es,
hey should be asc ibed ei he o ai bubbles o emnan mois u e
as well as o polyme ma ix con ac ion because o he hea
dispe sion enhanced by he CF. Mois u e is minimised upon d y
p e ea men o ASA and CF; hus, he con ac ions should be
he dominan phenomenon o explain po e o ma ion. This
hypo hesis is in good ag eemen wi h he absence o po es in he
pu e ASA, whe e he lowe hea dissipa ion, no a ou ed by he
p esence o CF, would no p o oke he eme gence o po es. In
any case, o a highe o al olume o po es, mic o ac u es will be
mo e likely o g ow and nuclea e by po osi y, and, cohe en ly,
lowe mechanical s eng h will be expec able.
The compa ison o 3D pa s ab ica ed by FGF using bo h
ASA_CF compounds and nea ASA a e exhibi ed in Figu e 3.
Conc e ely, in Figu e 3(a) and (d),3DX- ayCTda a
econs uc ions o ASA_CF_LS_FGF and ASA_CF_HS_FGF
a e shown, espec i ely (ob ained om he egion ma ked wi h a
ed squa e in he 2D images in Suppo ing In o ma ion, Figu e
S3), and Figu e 3(g) shows he da a co esponding o ASA_FGF.
A uni o m dis ibu ion o he oads, cha ac e is ic o he FGF
p ocess, can be obse ed in he h ee objec s, wi h an app oxima e
wid ho 2mmandheigh o 1mm,inag eemen wi h he
pa ame e s used in he p in ing p ocess (nozzle size and
laye hickness). Figu e 3(c) and ( ), shows de ails o 2D slices
o he 3D g eyscale CT da a o ASA_CF_LS_FGF and
ASA_CF_HS_FGF, espec i ely. Besides he CF ha can be
obse ed as b igh con as s ( ha will be analysed in mo e de ail in
sec ion Fib e Analysis), some po osi y can be ound in bo h
ma e ials, which is absen in he pu e ASA piece, Figu e 3(h).
Obse a ions in he middle o he beads show ha he e y la ge
po es ound in he cen al pa s o he ASA_CF_LS_Pelle
[Figu e 2(a)] a e no p esen in he p in ed objec . Howe e , i is
no iceable he e is a la ge densi y o po es a he middle o he
bead han a he edge, as expec ed. Sub egions in he cen e o he
oads [ma ked wi h ed squa es in Figu e 3(a) and (d)] ha e been
e alua ed by a highe esolu ion X- ay CT analyses. The
segmen a ion o he po es included in Figu e 3(b) and (e), clea ly
shows ha he densi y o po es is la ge in he ASA_CF_HS_FGF
piece. The quan i a i e analysis o he obse ed po osi y showed
ha he po e olume pe cen in he ASA_CF_LS_FGF specimen
was 3.4%, while in he ASA_CF_HS_FGF specimen, i eached
8.7%, e idencing ha he amoun o po es in he middle o he
oads is mo e han wice in he high shea compounded ma e ial.
Rega ding he size o he po es, hey we e sligh ly la ge in he
ASA_CF_LS_FGF objec han in ASA_CF_HS_FGF, as can be
seen in he his og am o po e size in Figu e S2. In FGF p in ed
objec s, he endency ound in he amoun o po es is simila o
ha in he pelle s udy, indica ing ha in spi e o he p in ing
p ocess mel ing he ma e ial again, he p in e canno elimina e
he esidual po osi y inside he o iginal eeds ock. This e alua ion
highligh s ha e en conside ing he addi ional mac opo osi y
in insically gene a ed be ween beads du ing p in ing, po osi y
om eeds ock emains a e p in ing, which migh nega i ely
a ec u he p ope ies o he p in ed pieces. In any case, o a
highe o al olume o po es, mic o ac u es will be mo e likely o
g ow and nuclea e, and, cohe en ly, lowe mechanical s eng h
will be expec able because, as epo ed in he li e a u e, po osi y
ac s as a s ess concen a o (Du Plessis e al.,2018;Wang e al.,
2019).
Rega ding he po e shape, in he li e a u e, elonga ed po es
a e commonly obse ed in composi es con aining fib es (Keles
e al.,2018;Kuma e al.,2021). In his case, elonga ed
po osi y is pa ially ela ed o he de o ma ion o he mol en
ma e ial du ing deposi ion (Kelese al.,2018) o likely because
o he a fini y o he en apped gas o he fib es ha also
indica es a poo e in eg a ion o he fib es in he ma ix (Kuma
e al., 2021). In ela ion o his, he segmen a ion o he po es
ound in he ASA_CF_LS_FGF and ASA_CF_HS_FGF
pieces included in Figu e 3(c) and ( ), shows ha he po es do
no p esen an e iden elonga ion. To analyse he po es shape,
he sphe ici y o he po es (which is a measu e o how closely
he shape o an objec esembles ha o a pe ec sphe e) has
been calcula ed, and a his og am wi h he ob ained esul s is
shown in Figu e 3(h). The sphe ici y, being 1 ha o a sphe e
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
222

by defini ion, o any pa icle which is no a sphe e will be less
han 1. Sphe ici y (
c
) can be calcula ed wi h he ollowing
o mula (1), whe e V means olume o pa icle and A means
su ace a ea o he pa icle.
W¼
p
1=36V
ðÞ
2=3
A(1)
The obse ed sphe ici y o bo h composi es p esen s
dis ibu ions cen ed abou 0.9, i.e. he majo i y o he po es can
be app oached as sphe es. The dis ibu ion co esponding o
ASA_CF_LS_FGF seems o exhibi an unexpec ed peak a ound
sphe ici y 0.5. This could be ela ed o he numbe o po es:
because he numbe o po es is lowe in he la e composi e han
in he co esponding ASA_CF_HS_FGF, po es exhibi ing a
lowe sphe ici y should be s a is ically mo e significan . In any
case, he equency is qui e lowe han hose o sphe ici y 0.9;
hence, bo h dis ibu ions could be easonably assumed o be
mos ly unimodal. This means ha hey a e no likely because o
he a fini y o he en apped gas o he fib e o decohesion
be ween fib es and polyme . Disca ding so, he sphe ical po e
dis ibu ion ound can be ela ed mos ly o he sh inkage e ec o
he polyme in he composi e ma e ial (Rauwendaal, 2010).
Fo compa a i e pu poses, pa s we e also ab ica ed by IM
and analysed using X- ay CT. Figu e 4 shows ende ed
olumes o pa s ob ained by injec ion moulding using he wo
compounding sys ems conside ed, ASA_CF_LS_IM in
Figu e 4(a) and ASA_CF_HS_IM in Figu e 4(c). Da a
co esponding o a nea ASA-injec ed specimen is also included
[Figu e 4(e)]. In con as o p e ious obse a ions ela ed o
pelle s and FGF specimens, no po es ha e been ound in any o
he injec ed solid pieces. Figu e 4(b) and (d), shows 2D slices
o he CT da a co esponding o he ASA_CF_LS_IM and
ASA_CF_HS_IM pieces, whe e he absence o po es is e iden .
These esul s clea ly de e mine ha injec ion o pelle s
elimina es i s inhe en po osi y, likely because o he highe
p essu e in ol ed in he p ocess.
3.2 Fib e analysis
O ien a ion and leng h o CFs ha e a s ong influence on he
mechanical p ope ies o composi es. Because o his, a de ailed
analysis o hese pa ame e s has been ca ied ou in he pelle s,
FGF and IM pieces, using high- esolu ion X- ay CT analysis.
Conce ning he o ien a ion analysis, CF segmen s a e p esen ed
in a sphe ical coo dina e sys em, which allows o quan ifica ion
o he o se angle (
u
) om he longi udinal di ec ion (Billah e al.,
2020). Figu e 5 shows he esul s o he analysis o he CF
o ien a ion dis ibu ion o he ASA_CF_LS_FGF (a) and
ASA_CF_HS_FGF (b) pieces in he egion o he ed squa e in
Figu e 3(a) and (d), whe e CF a e ep esen ed colou -coded
ega ding hei o ien a ion. As can be obse ed, in bo h pieces,
CF is mos ly o ien ed pa allel o he ex usion di ec ion. This is
co obo a ed by he CF o ien a ion his og am shown in Figu e 5
(c), which shows a easonably na ow o ien a ion dis ibu ion o
bo h pieces. This p e e ence o ien a ion is expec ed in ma e ials
ab ica ed by ex usion p ocesses, as no ed (Kuma e al.,2021;
Spoe k e al., 2018;Tekinalp e al.,2014). In his case, ha
p ealignmen implies a posi i e e ec in e ms o an inc ease in
s i ness and s eng h in he di ec ion o he deposi ed ma e ial
(Lo e e al.,2015;S
anchez e al., 2020a). Howe e , i also
Figu e 3 Rende ed olumes o he ASA_CF_LS_FGF (a), ASA_CF_HS_FGF (d) and ASA_FGF (g) pa s; pieces and po e segmen a ion images o he
ASA_CF_LS_FGF (b) and ASA_CF_HS_FGF (e) om he egion ma ked wi h a ed squa e in (a) and (d); magnified egions o 2D slices o he
ASA_CF_LS_FGF (b), ASA_CF_HS_FGF (e) and ASA_FGF (h), espec i ely, om he egion ma ked wi h a black squa e; (i) his og am o he po e
sphe ici y o he ASA_CF_LS_FGF and ASA_CF_HS_FGF pieces
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
223
inc eases he aniso opy o he objec no only in mechanical
e ms (Hmeida e al.,2020;Spoe k e al., 2018;To ado e al.,
2015) bu also in CTE beha iou (Hoskins e al., 2019).
A de ailed analysis o he CF leng h has been ca ied ou
using he high- esolu ion CT da a ob ained om he FGF and
IM pieces. Figu e 5(d) and (e), shows he esul s o he analysis
o CF leng h in ASA_CF_LS_FGF and ASA_CF_HS_FGF
pieces, espec i ely, ep esen ed wi h a colou scale. As i can
be obse ed, a na ow dis ibu ion o CF leng h alues is
obse ed, wi h some la ge fib es obse ed in cyan colou in he
ASA_CF_LS_FGF piece. To ob ain quan i a i e in o ma ion
on he CF leng h dis ibu ion in bo h ma e ials, Figu e 5( )
shows a compa a i e his og am o he measu ed CF leng h.
Resul s show ha CFs ha e a leng h in e io o 250
m
minbo h
ma e ials. These alues ob ained a e qui e simila o hose
ound in he pelle s his og am o CF leng h in he pelle s,
Figu e 5(g), showing ha compounding implies ac u e and a
ema kable educ ion in leng h o CFs (o iginally he leng h o
he CF was 6 mm, acco ding o he dis ibu o ). B eakage o
fib es du ing composi e ab ica ion occu s because o he shea
o ce and sc ew configu a ion du ing mixing. This is o majo
impo ance, as i is well known ha longe fib es o e highe
ein o cemen han sho e ones in he compounds (Adeni an
e al., 2022;Yeole e al.,2020). In ou case, he fib e leng h
ob ained a e compounding seems o be abo e he c i ical
alue, as ein o cemen is demons a ed in he mechanical
p ope ies measu ed, as shown in he nex sec ion. On he o he
hand, he ac ha he fib e leng h is almos he same in p in ed
pa s as i is in pelle s highligh s ha he configu a ion o he
single sc ew o he p in e does no p oduce a significan
educ ion in fib e leng h, e idencing ha he p in ing p ocess
used does no deg ade he p ope ies o he eeds ock ma e ial
used.
A close look a Figu e 5( ) shows ha , on a e age, CF has a
la ge leng h in he ASA_CF_LS_FGF piece han in he
ASA_CF_HS_FGF one. This indica es ha he LS ex usion
line is less agg essi e han he HS ex usion line, as expec ed.
The educ ion in mixing elemen s, oge he wi h a lowe
esidence ime inside he ex ude , p oduces a b eaking le el o
he fib es which is lowe in line LS han in HS. As a
homogeneous dis ibu ion o fib es inside he polyme has been
obse ed in bo h cases, he le el o ein o cemen migh be
be e in a configu a ion wi h low shea mixing han in high
shea mixing, as he longes fib es in he ma e ial ASA_CF_LS
a e he ones wi h a majo e ec in he ein o cemen
mechanism. This will be u he discussed below in ela ion o
he mechanical p ope ies measu ed in bo h ma e ials.
In connec ion wi h his, i should be men ioned ha he
amoun o po osi y obse ed in he composi es migh also be
ela ed o he leng h o fib es, as he la ge he numbe o fib e
b eakage e en s in he ma e ial, he la ge he numbe o po osi y
nuclea ion si es. The la ge numbe o po es and he sho e
a e age CF leng h ound in ASA_CF_HS_FGF ega ding
ASA_CF_LS_FGF would be in line wi h his possibili y. Also, a
Figu e 4 Rende ed olumes o he ASA_CF_LS_IM (a), ASA_CF_HS_IM (c) and ASA_IM (e) pieces; magnified egions o he 2D slices o he
ASA_CF_LS_IM (b), ASA_CF_HS_IM (d) and ASA_IM ( )
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
224
lowe ‘fib e- ee’ olume in he polyme educes he possibili y o
po e coalescence, a oiding he inc ease in size o he po osi y.
Rega ding he IM specimens, an analogous analysis o CF
o ien a ion and leng h has been ca ied ou in high- esolu ion CT
da a (ob ained om he egion in he 2D images in Suppo ing
In o ma ion, Figu e S4) o unde s and he mic os uc u e o he
ma e ials o be co ela ed o hei mechanical p ope ies. Figu e 6
depic s a ep esen a ion o he CF o ien a ion in he
ASA_CF_LS_IM (a) and ASA_CF_HS_IM (b) objec s, and
Figu e 6(c) shows a his og am o he o ien a ion da a in bo h
pieces. As illus a ed by he esul s, in bo h ma e ials, mos o he
CF a e o ien ed pa allel o he injec ion di ec ion, simila ly o
wha is obse ed in FGF p in ed specimens. Ne e heless, in his
case he injec ed specimens exhibi a cen al egion whe e some
diso ien a ion occu s, as can be obse ed in he g een colou in
Figu e 6(a) and (b), . I is well known ha du ing he injec ion
p ocess, he ma e ial in con ac wi h he mould walls expe iences
a sha p educ ion in ma e ial eloci y because o an inc ease in
shea a e. I is his ma e ial in con ac wi h he mould
ha solidifies fi s ( e e ed o as he ozen laye ), and whe e he
educ ion in eloci y p omo es a p e e en ial alignmen o
he fib es in he di ec ion o flow by being in a lamina egime.
The mol en ma e ial con inues o flow h ough he cen e o he
ca i y un il i fills he mould comple ely. This mel has a highe
Figu e 5 Colou -coded ep esen a ion o he CF o ien a ion in he ASA_CF_LS_FGF (a) and ASA_CF_HS_FGF (b) pieces; (c) s a is ical dis ibu ion o
he CF o ien a ion angles ega ding he longi udinal di ec ion, measu ed om he da a in (a) and (b); colou -coded ep esen a ion o he CF leng h in he
ASA_CF_LS_FGF (d) and ASA_CF_HS_FGF (e) pieces; ( ) s a is ical dis ibu ion o he CF leng h measu ed in (d) and (e); (g) s a is ical dis ibu ion o he
CF leng h measu ed in he LS and HS ASA_CF_pelle s
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
225
eloci y and a lowe shea a e, and because o his, he fib es a e
mo e andomly a anged (Malloy, 2025;Rauwendaal, 2010),
explaining he a ia ion in fib e dis ibu ion obse ed in he CT
analysis.
Conce ning he size o he CF, Figu e 6(d) and (e), shows in
colou code he quan ifica ion o he CF leng h in he
ASA_CF_LS_IM and ASA_CF_HS_IM pieces, espec i ely,
and Figu e 6( ) shows he compa a i e leng h da a his og am.
As can be obse ed in Figu e 6( ) and S5, hese pieces also
show leng h alues smalle han 250
m
m, analogously o he
pelle s used o he injec ion p ocess, indica ing ha he CF is
no ema kably u he agmen ed in he injec ion p ocess
applied a e he ma e ial is compounded. Na ow leng h
dis ibu ions a e obse ed in bo h pieces, assu ing he
homogenei y o he ma e ial du ing he IM manu ac u ing
p ocess. Simila ly o he FGF pieces, Figu e 6( ) shows ha , on
a e age, he CF leng h in he ASA_CF_LS_IM piece is la ge
han in he ASA_CF_HS_IM one, which is expec ed o ha e
some e ec on hei mechanical p ope ies. This di e ence in
leng h migh also influence he o ien a ion dis ibu ion
p e iously discussed, as he la ge he fib es, he mo e endency
o align in he flow di ec ion, while he sho e fib es migh be
a ec ed by he u bulen flow o mol en polyme .
3.3 Tensile s eng h cha ac e isa ion
Mechanical p ope ies de e mined om he composi es
ab ica ed by IM and by FGF a e depic ed in Figu e 7. The
co esponding enginee ing s ess-s ain cu es can be ound in
Figu e S5 as suppo ing ma e ial, while a ep esen a i e cu e
o each ma e ial is depic ed in Figu e 7(d). As e idenced by he
esul s, he inco po a ion o CF clea ly enhances he elas ic
p ope ies o nea ASA. Thus, he s i ness, e alua ed in e ms
o ensile modulus (E), inc eases mo e han 150% o bo h
injec ed and p in ed pieces, as shown in Figu e 7(a). Also, he
Figu e 6 Colou -coded ep esen a ion o he CF o ien a ion in he ASA_CF_LS_IM (a) and ASA_CF_HS_IM (b) pieces; (c) s a is ical dis ibu ion o he
CF o ien a ion angles ega ding he longi udinal di ec ion, measu ed om he da a in (a) and (b); colou -coded ep esen a ion o he CF leng h in he
ASA_CF_LS_IM (d) and ASA_CF_HS_IM (e) pieces; ( ) s a is ical dis ibu ion o he CF leng h measu ed in (d) and (e)
3D la ge o ma addi i e manu ac u ing pieces
Daniel Mo eno Sánchez e al.
Rapid P o o yping Jou nal
Volume 31 · Numbe 11 · 2025 · 218–230
226