ma e ials
A icle
E ec o Imposed Shea S ain on S eel Ring Su aces
du ing Milling in High-Speed Disin eg a o
Ka el D oˇ ák1,* , Adéla Macháˇcko á2, Simona Ra aszo á1and Dominik Gazdiˇc 1
1Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve eˇ í331/95, 602 00 B no, Czech Republic;
[email p o ec ed].cz (S.R.); [email p o ec ed].cz (D.G.)
2Facul y o Ma e ials Science and Technology, VŠB–Technical Uni e si y o Os a a, 17. lis opadu 15,
708 00 Os a a, Czech Republic; [email p o ec ed]
*Co espondence: [email p o ec ed].cz
Recei ed: 13 Ap il 2020; Accep ed: 11 May 2020; Published: 13 May 2020


Abs ac :
This con ibu ion cha ac e izes he pe o mance o a DESI 11 high-speed disin eg a o
wo king on he p inciple o a pin mill wi h wo opposi e coun e - o a ing o o s. As he g ound
ma e ial, ba ches o Po land cemen ea u ing 6–7 Mohs scale ha dness and con aining ela i ely
ha d and ab asi e compounds wi h he speci ic su ace a eas anging om 200 o 500 m
2
/kg, wi h he
s ep o 50 m
2
/kg, we e used. The cha ac e o he g ound pa icles was assessed ia scanning elec on
mic oscopy and measu emen o he absolu e/ ela i e inc ease in hei speci ic su ace a eas. De ailed
cha ac e iza ion o he o o s was pe o med ia eco ding he he mal imp in s, e alua ing hei wea
by 3D op ical mic oscopy, and measu ing o o weigh loss a e he g inding o cons an amoun s o
cemen . The esul s showed ha coa se pa icles a e g ound by impac ing he on aces o he pins,
while ine pa icles a e p ima ily milled ia mu ual collisions. The e o e, he coa se pa icles cause
highe ab asion and wea on he o o pins; a e he milling o 20 kg o he 200 m
2
/kg cemen sample,
he wea o he o o eached up o 5% o i s o iginal mass and he pins we e se e ely damaged.
Keywo ds: disin eg a o ; mic oscopy; wea ; high ene gy milling; cemen
1. In oduc ion
The necessi y o enhance he u ili y p ope ies o ma e ials and inc ease he longe i y o
cons uc ion componen s has led esea ch and de elopmen in a ious indus ial b anches, such as
ma e ials p ocessing [
1
] o cons uc ion [
2
,
3
], owa ds he design o new ma e ials and imp o ing
he pe o mance o con empo a y ones. Gene ally, enhancemen o he mechanical and u ili y
p ope ies can be achie ed ia a ia ions in he chemical composi ion o he pa icula ma e ial,
as documen ed, o example, by Zhang e al. [
4
], and s uc u al modi ica ions (g ain e inemen ), which
was demons a ed also o pu e and comme cially pu e me als [
5
,
6
]. Ve linden e al. claimed ha g ain
e inemen in oduces la ice de ec s, such as disloca ions and g ain bounda ies, which inc ease he
ma e ials’ in insic ene gy and ac as s eng hening agen s [
7
]. Kunˇcick
á
e al. s udied syn heses o
a ious kinds o alloys and epo ed ha he g ain size can ypically be dec eased in wo ways, he i s
o which is applica ion o he he me hods o in ensi e/se e e plas ic de o ma ion (SPD) dec easing he
g ain size down o sub-mic on scale ia impa ing se e e shea s ain [
8
] (e.g., equal channel angula
p essing [
9
] and high-p essu e o sion [
10
]). The second is he implemen a ion o powde me allu gy,
i.e., he p oduc ion o ma e ials om o iginal powde s, which is ypically used o challenging ma e ials,
such as Ti composi es [
11
] o ungs en hea y alloys [
12
]. Combina ions consis ing o he p ocessing o
challenging powde -based ma e ials, such as Ti-based [
13
], and W-based [
14
] alloys, ia SPD me hods
a e also ad an ageous.
Ma e ials 2020,13, 2234; doi:10.3390/ma13102234 www.mdpi.com/jou nal/ma e ials
Ma e ials 2020,13, 2234 2 o 12
The p ocess o p oduc ion o a componen om powde s in ol es inal shape and su ace ea men s,
which ypically ollowde o ma ionp ocessingand/o a ia ionso sin e ingandhea ea men s. The i s
p oduc ion s ep is, howe e , p epa a ion o he o iginal powde s. These a e usually manu ac u ed by
milling. The g ain size dis ibu ion and mo phology o he g ains ha e undamen al e ec s no only on
he sin e ing p ocess i sel , as men ioned by Wang e al. [
15
], bu also on he inal s uc u e o he alloys
as desc ibed by Mach
á
ˇcko
á
e al. [
16
]. Su zhenko e al. s udied he wea esis ance and mechanisms o
ab asion du ing milling. Based on he esul s o hei esea ch, i can be s a ed ha he g ain e inemen
in mills is ypically ensu ed ia se e al undamen al mechanisms, among which a e comp ession, shea
(a i ion), comp ession pulse, impac (s oke o collision), impac and shea , ension/bending, spli ing,
and cu ing. Milling can be pe o med by collisions o pa icles wi h he wo king ools, ia mu ual
collisions be ween pa icles, o be ween pa icles and hei en i onmen [17].
Among he ecen ends in milling is he high-ene gy milling (HEM) me hod. The possibili ies
o applica ion o HEM a e in a ious indus ial ields, om milling o demanding composi es in
me allu gy, as documen ed, o example, by Mu oi e al. [
18
] o Saza i e al. [
19
], h ough ab ica ion
o a ious in e me allic and ce amic ma e ials, such as hose men ioned by Se ena e al. [
20
], o he
p epa a ion o nanoma e ials (e.g., Rojac e al. [
21
]). High-speed g inding (HSG) is a pa icula ype
o HEM ca ied ou by applying high amoun s o ene gy using e y sho and in ense powe pulses.
The amoun o ene gy ha is e ec i ely ans e ed o he ma e ial is highe in HSG han in he case o
con en ional g inding in mills wi h iden ical powe inpu s. Among he ypes o mills sui able o HSG
is a high-speed pin mill wi h wo coun e - o a ing o o s, known as he disin eg a o . This ype o mill,
he ma e ial in which is e ined by high- equency changes in mechanical s eng h, was ex ensi ely
s udied by Hin [
22
]. The disin eg a o is pa icula ly sui able o he g inding and ac i a ion o ine
powde ma e ials, as documen ed by Bumanis [
23
], and Bumanis and Baja e [
24
]. The p inciple o a
disin eg a o lies in accele a ing he ma e ial o a high speed by means o he pins on he g inding o o .
The pa icles han collide wi h o he pa icles o wi h he pins on he o o which o a es in opposi e
di ec ion. Disin eg a o s only use he impac , impac and shea , shea (a i ion), and ension/bending
mechanisms. Howe e , o he p ocesses simila o hose in a i o s o je mills occu as well. These a e
in oduced by u bulen low and apid comp ession and expansion be ween he o o s, as desc ibed
by Ko ale [
25
]. Bal
á
ž w o e in his s udy [
26
] ha he main ad an ages o a high-speed disin eg a o
a e con inui y o he g inding p ocess, and a ie y o wo king ools ha can be employed. On he
o he hand, Su zhenko e al. s a ed he main disad an age o be ha he g inding elemen s a e p one
o ab asion/wea [17].
Collisions o pa icles and ha d su aces can be ad an ageous since he shea s ain in oduced o
he ha d su ace ia he impac ing pa icles ypically induces he o ma ion o ha d s uc u e phases a
he su ace o he base ma e ial (such as ma ensi e o s ainless s eels as epo ed by S aman e al. [
27
]),
by he e ec o which he su ace ha dness inc eases, and he su ace ha dening consequen ly educes
he endency o he base ma e ial (i.e., o o s) o ab asion/wea , as claimed by Sil a e al. [
28
] and
Han e al. [
29
]. On he o he hand, con inuous high-speed shelling o he base ma e ial ( o o s) wi h
ha d and/o sha p-edged pa icles ine i ably leads o ab asion/wea in ime. Gåhlin and Jacobson [
30
]
men ioned ha he in ensi y o wea depends on he mine alogy, mo phology, and g anulome y o
he g ound ma e ial. Howe e , he e ec o pa icle size o he milled ma e ial on wea is no ye ully
unde s ood. Mis a and Finnie [
31
] sugges ed ha inc easing he pa icle size inc eases he wea a e.
Small pa icles lead o pene a ions ha do no pass h ough he su ace laye s o he g inding elemen s;
acco ding o he heo y in oduced by La sen-Bads [
32
], su icien ly small pa icles a e in elas ic con ac
wi h he g inding ools and do no con ibu e o ab asion. The p esupposed e ec o pa icles’ shapes
on he deg ee o ab asion is based on a heo y assuming ha blun shapes exe less p essu e on he
g inding ools and consequen ly ac mo e like pa icles p o iding su ace ha dening ia imposing
shea s ain and p oduce less wea han sha p-edged pa icles. Fo milled ma e ial ea u ing wide
size ac ions, he la ges g ains exe s ong poin loads on he su aces o he g inding elemen s; i.e.,
such ma e ial is mo e ab asi e [17,30].
Ma e ials 2020,13, 2234 3 o 12
I he pins a e damaged due o ab asion, he e iciency o g inding is educed. By his eason,
he geome y o he milling elemen s is a ou able o be kep cons an o as long as possible du ing
milling. The p esen ed s udy is ocused on cha ac e iza ion o he beha iou o o o s o a high-speed
disin eg a o du ing milling o a Po land cemen , because he HSG echnology seems o be a e y
e ec i e milling echnology, mo eo e , wi h he bene i o mechanical ac i a ion. Howe e , ce ain
issues in his ield s ill emain unco e ed. The s anda d Po land cemen is ela i ely agile and
con ains ou main mine als wi h he a e age ha dness o 6–7 (acco ding o he Mohs scale). I ea u es
sha p-edged pa icles, which makes i a e y ab asi e ma e ial. G inding by a disin eg a o is also
supposed o be e y e ec i e o his pa icula ma e ial since i esponds e y well in milling by
impac o comp ession pulse [
33
]. The s udy is supplemen ed wi h cha ac e iza ion o co ela ions
be ween en y g anulome y and ab asion o he o o s du ing milling.
2. Ma e ials and Me hods
The equipmen in es iga ed is he DESI 11 disin eg a o (Desin eg aa o Too mise OÜ, Tallinn,
Es onian Republic), which is a labo a o y e sion o a high-speed pin mill wi h wo coun e - o a ing
o o s. The o al ins alled ou pu o he mill is 4.1 kW. The o o o a ion equency is up o 12,000 RPM,
and he maximum ci cum e en ial speed o each o o is 92.4 m/s. The ma e ial is ed in o he de ice by a
con inuous eede and en e s he g inding chambe h ough he middle o he le o o ; a schema ic
depic ion o he p inciple o he disin eg a o is shown in Figu e 1a, and also desc ibed in [
17
]. The eal
machine is p esen ed in Figu e 1b. The cons uc ion o he mill allows o a choice o wo king ools. Fo
his expe imen , he CR ype o o s designed and manu ac u ed by he FF se is s. .o. company (P ag,
Czech Republic) we e used. The o o s we e manu ac u ed om C45 s eel, some imes also cha ac e ized
as SI 1045 s eel. This medium-ca bon s eel was selec ed o i s high quali y, ela i ely high s eng h, and
easy machinabili y; he o o s we e machined om no malized ho - olled ba s. The ex e nal diame e
o bo h he o o s was 147 mm. The le o o has wo ows, while he igh o o has h ee ows o
3×3×3 mm cubic pins. The designs o bo h he o o s a e demons a ed in Figu e 1c,d.
The Po land cemen used o his expe imen was p epa ed in a ball mill by collec i e milling o
cemen clinke ( om H anice cemen plan ) and chemical gypsum (P egips), he chemical composi ions
o which a e depic ed in Table 1. The chemical composi ion o he selec ed clinke is ypical o Po land
clinke s. The gypsum (highly pu e) con ained ela i ely high humidi y and i was u he d ied be o e
milling o dec ease he mois u e o unde 5%. The a io o clinke o gypsum was 95:5.
Table 1. Pa ial chemical composi ions o he g ound ma e ial.
Ma e ial Componen
SiO2CaO Al2O3Fe2O3SO3CaSO4·2H2O H2O CaSO4O he s
Clinke 20.29 65.33 5.21 5.04 0.79 - - - 3.34
Gypsum - - - - - 84.0 11.0 2.4 2.6
Fo he subsequen milling p ocess, a se o se en indi idual ba ches wi h speci ic su ace a eas
anging om 200 o 500 m
2
/kg was p epa ed. The ba ches o cemen wi h he selec ed g anulome ies
we e ed in o he disin eg a o con inuously. The dosing a e was 5.5 g
·
s
−1
and he cemen pa icles
a e supposed o spend app oxima ely 1 second in he milling chambe due o he pin mill p inciple.
All he samples we e milled a he maximum speed o 12,000 RPM.
G inding o all he samples was ca ied ou unde s anda d labo a o y condi ions, a 22
◦
C and
ela i e humidi y o 56%. A e g inding o each 1 kg ba ch, he mal imp in on he o o was acqui ed
using a Fli E4 he mal came a (FLIR Sys ems Inc., Wilson ille, OR, USA). Be ween milling o he
indi idual ba ches, he disin eg a o was cooled by an ai s eam o he ini ial empe a u e o 22
◦
C
and he o o s we e eplaced wi h a new se .
Ma e ials 2020,13, 2234 4 o 12
Ma e ials2020,13,xFORPEERREVIEW3o 12
I  hepinsa edamageddue oab asion, hee iciencyo g indingis educed.By his eason, he
geome yo  hemillingelemen sis a ou able obekep cons an  o aslongaspossibledu ing
milling.Thep esen eds udyis ocusedoncha ac e iza iono  hebeha iou o  o o so ahigh‐speed
disin eg a o du ingmillingo aPo landcemen ,because heHSG echnologyseems obea e y
e ec i emilling echnology,mo eo e ,wi h hebene i o mechanicalac i a ion.Howe e ,ce ain
issuesin his ields ill emainunco e ed.Thes anda dPo landcemen is ela i ely agileand
con ains ou mainmine alswi h hea e ageha dnesso 6–7(acco ding o heMohsscale).I 
ea u essha p‐edgedpa icles,whichmakesi a e yab asi ema e ial.G indingbyadisin eg a o 
isalsosupposed obe e ye ec i e o  hispa icula ma e ialsincei  esponds e ywellinmilling
byimpac o comp essionpulse[33].Thes udyissupplemen edwi hcha ac e iza iono co ela ions
be weenen yg anulome yandab asiono  he o o sdu ingmilling.
2.Ma e ialsandMe hods
Theequipmen in es iga edis heDESI11disin eg a o (Desin eg aa o Too miseOÜ,Tallinn,
Es onianRepublic),whichisalabo a o y e siono ahigh‐speedpinmillwi h wocoun e ‐ o a ing
o o s.The o alins alledou pu o  hemillis4.1kW.The o o  o a ion equencyisup o12,000
RPM,and hemaximumci cum e en ialspeedo each o o is92.4m/s.Thema e ialis edin o he
de icebyacon inuous eede anden e s heg indingchambe  h ough hemiddleo  hele  o o ;
aschema icdepic iono  hep incipleo  hedisin eg a o isshowninFigu e1a,andalsodesc ibed
in[17].The ealmachineisp esen edinFigu e1b.Thecons uc iono  hemillallows o achoiceo 
wo king ools.Fo  hisexpe imen , heCR ype o o sdesignedandmanu ac u edby heFFse is
s. .o.company(P ag,CzechRepublic)we eused.The o o swe emanu ac u ed omC45s eel,
some imesalsocha ac e izedasSI1045s eel.Thismedium‐ca bons eelwasselec ed o i shigh
quali y, ela i elyhighs eng h,andeasymachinabili y; he o o swe emachined omno malized
ho ‐ olledba s.Theex e naldiame e o bo h he o o swas147mm.Thele  o o has wo ows,
while he igh  o o has h ee owso 3×3×3mmcubicpins.Thedesignso bo h he o o sa e
demons a edinFigu es1c,d.

(a)

(b)

(c)

(d)
Figu e 1.
(
a
) Schema ic depic ion o he in es iga ed disin eg a o ; (
b
) labo a o y DESI 11 HSG mill
(c) o o wi h h ee ows o pins; (d) o o wi h wo ows o pins.
The inc ease in he Blaine speci ic su ace a ea and he mo phology o g ains we e de e mined o
all he samples. The Blaine speci ic su ace a ea was measu ed using a PC-Blaine-S a au oma ic de ice
(Zünde we ke E ns B ün GmbH, Hal e n am See, Ge many) wi h a measu ing cell wi h he olume o
7.95 cm
3
. Measu emen was epea ed h ee imes and a e aged o minimize e o s. To e alua e he
impac o he inpu g anulome y on he ab asion o wo king elemen s, wo 20 kg ba ches o cemen
we e p epa ed in a ball mill. The i s sample ea u ed he speci ic su ace a ea o 200 m
2
/kg, while o
he second one i was 450 m
2
/kg. The milling pa ame e s we e iden ical as o he p e ious case; o o
speed o 12,000 RPM and dosing a e o 5.5 g·s−1.
The pa icle size and pa icle size dis ibu ion we e de e mined o bo h he inpu samples using
a Mal e n Mas e size 2000 (Mal e n Panaly ical B.V., Almelo, The Ne he lands) wi h a Hyd o 2000 G
luid dispe sing uni , 2-isop opanol was used as a dispe sing agen . The mo phology and g ain shapes
we e obse ed and assessed by scanning elec on mic oscopy (SEM) using a Tescan MIRA 3XMU
(Tescan B no s. .o., B no, Czech Republic) equipmen .
E alua ion o he ab asion-milling elemen s was pe o med ia calcula ing he a io be ween he
weigh o he e oded o o s a e milling o 1, 10 and 20 kg o ab adan , and he ini ial weigh o he
o o s. The accu acy o he weigh loss measu emen s was 0.01 mg.
Also, he impac o o o wea on g inding e iciency was measu ed by moni o ing he inc ease in
he speci ic su ace a ea o he cemen . The ab asion o he o o s was also de e mined by 3D scanning.
De ailed 3D scans o he o o pins we e pe o med and e alua ed using an Olympus DSX1000 digi al
mic oscope (Olympus Czech G oup, s. .o., P ague, Czech Republic).
3. Resul s
Ab asion speed is no cons an and depends on he inpu g anulome y and on he shape o he
g ound pa icles. Mo phology o he cemen pa icles was in es iga ed ia SEM-BSE; images o he
Ma e ials 2020,13, 2234 5 o 12
cemen pa icles wi hin he indi idual samples a e depic ed in Figu e 2a– wi hin which indi idual pai s
o igu es ep esen mo phology o he samples be o e and a e milling in he HSG mill. Figu e 2a,d
depic he samples wi h he ini ial speci ic su ace a ea o 200 m
2
/kg, whe eas Figu e 2b,e show he
samples wi h he ini ial speci ic su ace a ea o 300 m
2
/kg, and Figu e 2c, depic he samples wi h he
ini ial speci ic su ace a ea o 450 m2/kg.
Ma e ials2020,13,xFORPEERREVIEW5o 12
indi idualpai so  igu es ep esen mo phologyo  hesamplesbe o eanda e millingin heHSG
mill.Figu es2a,ddepic  hesampleswi h heini ialspeci icsu acea eao 200m
2
/kg,whe eas
Figu es2b,eshow hesampleswi h heini ialspeci icsu acea eao 300m
2
/kg,andFigu es2c, 
depic  hesampleswi h heini ialspeci icsu acea eao 450m
2
/kg.

(a)

(b)

(c)

(d)

(e)

( )
Figu e2.Compa isono  heo iginalandpos ‐millingshapeso cemen pa icles o :200m
2
/kg
sample,(a)o iginalpa icles;(d)milledpa icles;300m
2
/kgsample,(b)o iginalpa icles;(e)milled
pa icles;450m
2
/kgsample,(c)o iginalpa icles;( )milledpa icles.
Thepa icles om hesamplewi h hespeci icsu acea eao 200m
2
/kgexhibi edsha pedges,
howe e , heywe esligh lyab aded.Simila  esul swe eacqui ed o  hesamplewi h hespeci ic
su acea eao 300m
2
/kg,bu  hesepa iclesexhibi edamo esubs an ialab asion.Thepa icles om
he ines  ac ions henexhibi edmo eo lesssphe icalshapesand he endency oagglome a e.
Agglome a edclus e swe ealsop esen in hesamplewi h heini ialspeci icsu acea eao 450
m
2
/kg.
Lase g anulome y e ealed ha  heba chwi h heinpu ‐speci icsu acea eao 200m
2
/kg
con ained20.8%o pa iclesla ge  han100μm,while heba chwi h heinpu speci icsu acea ea
o 450m
2
/kgcon ainedonly10.7%o pa iclesla ge  han100μm.Agg ega eswe emos p obably
no p esen because hemeasu emen swe epe o medunde we condi ions.
The esul ingspeci icsu acea eas o all hese eng oundsamples, oge he wi h heabsolu e
and ela i e alueso  hei inc eases,a edepic edinFigu e3.The esul sshow ha  heabsolu e
speci icsu acea ea aluesinc easedsigni ican lywheng indingcoa se ac ions(speci icsu ace
a easo 200and250m
2
/kg).Incon as ,bo h heabsolu eand ela i especi icsu acea ea alues
we elowe  o  heinpu  ac ionso 300and350m
2
/kg.Theabsolu einc easein hespeci icsu ace
a ea heninc easesg aduallywi hcon inuing e inemen o  heinpu ma e ial,whe eas he ela i e
inc easein hespeci icsu acea ea emainsmo eo lesscons an  om he300m
2
/kginpu  ac ion.
Figu e 2.
Compa ison o he o iginal and pos -milling shapes o cemen pa icles o : 200 m
2
/kg sample,
(
a
) o iginal pa icles; (
d
) milled pa icles; 300 m
2
/kg sample, (
b
) o iginal pa icles; (
e
) milled pa icles;
450 m2/kg sample, (c) o iginal pa icles; ( ) milled pa icles.
The pa icles om he sample wi h he speci ic su ace a ea o 200 m
2
/kg exhibi ed sha p edges,
howe e , hey we e sligh ly ab aded. Simila esul s we e acqui ed o he sample wi h he speci ic
su ace a ea o 300 m
2
/kg, bu hese pa icles exhibi ed a mo e subs an ial ab asion. The pa icles
om he ines ac ions hen exhibi ed mo e o less sphe ical shapes and he endency o agglome a e.
Agglome a ed clus e s we e also p esen in he sample wi h he ini ial speci ic su ace a ea o 450 m
2
/kg.
Lase g anulome y e ealed ha he ba ch wi h he inpu -speci ic su ace a ea o 200 m
2
/kg
con ained 20.8% o pa icles la ge han 100
µ
m, while he ba ch wi h he inpu speci ic su ace a ea o
450 m
2
/kg con ained only 10.7% o pa icles la ge han 100
µ
m. Agg ega es we e mos p obably no
p esen because he measu emen s we e pe o med unde we condi ions.
The esul ing speci ic su ace a eas o all he se en g ound samples, oge he wi h he absolu e
and ela i e alues o hei inc eases, a e depic ed in Figu e 3. The esul s show ha he absolu e
speci ic su ace a ea alues inc eased signi ican ly when g inding coa se ac ions (speci ic su ace
a eas o 200 and 250 m
2
/kg). In con as , bo h he absolu e and ela i e speci ic su ace a ea alues
we e lowe o he inpu ac ions o 300 and 350 m
2
/kg. The absolu e inc ease in he speci ic su ace
a ea hen inc eases g adually wi h con inuing e inemen o he inpu ma e ial, whe eas he ela i e
inc ease in he speci ic su ace a ea emains mo e o less cons an om he 300 m2/kg inpu ac ion.

Ma e ials 2020,13, 2234 6 o 12
Ma e ials2020,13,xFORPEERREVIEW6o 12

Figu e3.Inc easeinspeci icsu acea ea o allg oundsamples.
The malimp in son he o o sbe o eanda e millingo 1kgba cheso 200m
2
/kg o500m
2
/kg
samplesa eshowninFigu es4a o4i, espec i ely.Theimp in sclea lyshow ha g indingo  he
coa seba ches(200and250m
2
/kg)wasp ima ilype o medby he on  aceso  he o o pins, he
empe a u eo whichinc eased(Figu es4cand4d, espec i ely).As hespeci icsu acea eao  he
inpu cemen inc eased, hea easin he icini yo  hepinss a ed oexhibi inc eased empe a u e,
oo.The300,350,and400m
2
/kgba chescon ainedplen yo coa sepa icles, heg indingo which
waspe o meddi ec lyon he o o pins.Howe e ,mu ualg indingo  inepa iclesagains each
o he in hespacesbe ween hepinsoccu edaswell.Thisphenomenonwas hendominan  o  he
ba cheswi h hespeci icsu acea easo 450,and500m
2
/kg,du ing hemillingo which hepins
ac edmo elikeb eakingwedgesdi ec ing hemilledma e ialin o hespacesbe ween hem.Fo 
hese ineba ches, hepa icleswe ep ima ilymilledby hei mu ualcollisions,andmillingbydi ec 
impac swi h hepins’ on  aceswaso lowsigni icance(Figu es4h,iclea lyshow hepins obe
coole  han hesu oundinga ea).
 
(a)(b)(c)
Figu e 3. Inc ease in speci ic su ace a ea o all g ound samples.
The mal imp in s on he o o s be o e and a e milling o 1 kg ba ches o 200 m
2
/kg o 500 m
2
/kg
samples a e shown in Figu e 4a–i, espec i ely. The imp in s clea ly show ha g inding o he
coa se ba ches (200 and 250 m
2
/kg) was p ima ily pe o med by he on aces o he o o pins,
he empe a u e o which inc eased (Figu e 4c,d, espec i ely). As he speci ic su ace a ea o he
inpu cemen inc eased, he a eas in he icini y o he pins s a ed o exhibi inc eased empe a u e,
oo. The 300, 350, and 400 m
2
/kg ba ches con ained plen y o coa se pa icles, he g inding o which
was pe o med di ec ly on he o o pins. Howe e , mu ual g inding o ine pa icles agains each
o he in he spaces be ween he pins occu ed as well. This phenomenon was hen dominan o he
ba ches wi h he speci ic su ace a eas o 450, and 500 m
2
/kg, du ing he milling o which he pins ac ed
mo e like b eaking wedges di ec ing he milled ma e ial in o he spaces be ween hem. Fo hese ine
ba ches, he pa icles we e p ima ily milled by hei mu ual collisions, and milling by di ec impac s
wi h he pins’ on aces was o low signi icance (Figu e 4h,i clea ly show he pins o be coole han
he su ounding a ea).
The impac o inpu g anulome y on wea o he g inding o o s is demons a ed by Figu e 5
showing wea o he o o s (exp essed as loss o o o mass in ela ion o he o iginal o o mass) is
dependen on he amoun o he g ound ma e ial. The igu e indica es ha a coa se ma e ial damages
he g inding o o s o he disin eg a o mo e signi ican ly han a ine one.
Ma e ials2020,13,xFORPEERREVIEW6o 12

Figu e3.Inc easeinspeci icsu acea ea o allg oundsamples.
The malimp in son he o o sbe o eanda e millingo 1kgba cheso 200m
2
/kg o500m
2
/kg
samplesa eshowninFigu es4a o4i, espec i ely.Theimp in sclea lyshow ha g indingo  he
coa seba ches(200and250m
2
/kg)wasp ima ilype o medby he on  aceso  he o o pins, he
empe a u eo whichinc eased(Figu es4cand4d, espec i ely).As hespeci icsu acea eao  he
inpu cemen inc eased, hea easin he icini yo  hepinss a ed oexhibi inc eased empe a u e,
oo.The300,350,and400m
2
/kgba chescon ainedplen yo coa sepa icles, heg indingo which
waspe o meddi ec lyon he o o pins.Howe e ,mu ualg indingo  inepa iclesagains each
o he in hespacesbe ween hepinsoccu edaswell.Thisphenomenonwas hendominan  o  he
ba cheswi h hespeci icsu acea easo 450,and500m
2
/kg,du ing hemillingo which hepins
ac edmo elikeb eakingwedgesdi ec ing hemilledma e ialin o hespacesbe ween hem.Fo 
hese ineba ches, hepa icleswe ep ima ilymilledby hei mu ualcollisions,andmillingbydi ec 
impac swi h hepins’ on  aceswaso lowsigni icance(Figu es4h,iclea lyshow hepins obe
coole  han hesu oundinga ea).
 
(a)(b)(c)
Figu e 4. Con .
Ma e ials 2020,13, 2234 7 o 12
Ma e ials2020,13,xFORPEERREVIEW7o 12

(d)(e)( )


(g)(h)(i)
Figu e4.The malimp in son o o sbe o eanda e millingo 1kgo indi idualba ches:(a)new
o o wi h wo owso pins( o a iondi ec ionisma kedbywhi ea ow);(b)new o o ;(c)200m2/kg;
(d)250m2/kg;(e)300m2/kg;( )350m2/kg;(g)400m2/kg;(h)450m2/kg;(i)500m2/kg.
Theimpac o inpu g anulome yonwea o  heg inding o o sisdemons a edbyFigu e5
showingwea o  he o o s(exp essedaslosso  o o massin ela ion o heo iginal o o mass)is
dependen on heamoun o  heg oundma e ial.The igu eindica es ha acoa se ma e ial
damages heg inding o o so  hedisin eg a o mo esigni ican ly hana ine one.

Figu e5.E ec o  heamoun o g oundma e ialon o o wea  o indi idualsamples.
A e millingo onlya10kgba cho  he es ingcemen wi h hespeci icsu acea eao 200
m
2
/kg, hewea o  heg inding o o wi h h ee owso pinswas2.14%,and heg inding o o wi h
wo owso pinsexhibi edwea o 1.96%.Thewea di ec lyin luenced heinc easein hespeci ic
su acea eao pa icles,as heg indingwasabou 1.61%lesse ec i e.Table2summa izes he
esul so measu emen so  hespeci icsu acea ea o  hecemen powde a  hebeginningo  he
Figu e 4.
The mal imp in s on o o s be o e and a e milling o 1 kg o indi idual ba ches: (
a
) new
o o wi h wo ows o pins ( o a ion di ec ion is ma ked by whi e a ow); (
b
) new o o ; (
c
) 200 m
2
/kg;
(d) 250 m2/kg; (e) 300 m2/kg; ( ) 350 m2/kg; (g) 400 m2/kg; (h) 450 m2/kg; (i) 500 m2/kg.
Ma e ials2020,13,xFORPEERREVIEW7o 12

(d)(e)( )


(g)(h)(i)
Figu e4.The malimp in son o o sbe o eanda e millingo 1kgo indi idualba ches:(a)new
o o wi h wo owso pins( o a iondi ec ionisma kedbywhi ea ow);(b)new o o ;(c)200m2/kg;
(d)250m2/kg;(e)300m2/kg;( )350m2/kg;(g)400m2/kg;(h)450m2/kg;(i)500m2/kg.
Theimpac o inpu g anulome yonwea o  heg inding o o sisdemons a edbyFigu e5
showingwea o  he o o s(exp essedaslosso  o o massin ela ion o heo iginal o o mass)is
dependen on heamoun o  heg oundma e ial.The igu eindica es ha acoa se ma e ial
damages heg inding o o so  hedisin eg a o mo esigni ican ly hana ine one.

Figu e5.E ec o  heamoun o g oundma e ialon o o wea  o indi idualsamples.
A e millingo onlya10kgba cho  he es ingcemen wi h hespeci icsu acea eao 200
m
2
/kg, hewea o  heg inding o o wi h h ee owso pinswas2.14%,and heg inding o o wi h
wo owso pinsexhibi edwea o 1.96%.Thewea di ec lyin luenced heinc easein hespeci ic
su acea eao pa icles,as heg indingwasabou 1.61%lesse ec i e.Table2summa izes he
esul so measu emen so  hespeci icsu acea ea o  hecemen powde a  hebeginningo  he
Figu e 5. E ec o he amoun o g ound ma e ial on o o wea o indi idual samples.
A e milling o only a 10 kg ba ch o he es ing cemen wi h he speci ic su ace a ea o 200 m
2
/kg,
he wea o he g inding o o wi h h ee ows o pins was 2.14%, and he g inding o o wi h wo
ows o pins exhibi ed wea o 1.96%. The wea di ec ly in luenced he inc ease in he speci ic su ace
a ea o pa icles, as he g inding was abou 1.61% less e ec i e. Table 2summa izes he esul s o
measu emen s o he speci ic su ace a ea o he cemen powde a he beginning o he expe imen ,
and a e milling o 10 and 20 kg ba ches. Table 3 hen p esen s he educ ions o g inding e iciency,
cha ac e ized as educ ion o he speci ic su ace a ea. When g inding he ine-g ained inpu ma e ial
wi h he speci ic su ace a ea o 450 m
2
/kg, he ab asion occu ed o a lesse ex en . The weigh loss
measu ed o he o o wi h h ee ows o pins a e g inding o a 10 kg ba ch was 0.28%, while o he
o o wi h wo ows o pins i was 0.13%; he educ ion o he e iciency o g inding by he wo n o o s
Ma e ials 2020,13, 2234 8 o 12
was negligible. Milling o a 20 kg ba ch o he es ing cemen esul ed in u he o o weigh loss; a e
g inding an addi ional 10 kg ba ch, he weigh loss o he h ee- ow o o was 0.42%, while o he
wo- ow o o i was only 0.30%. The e iciency o he o o s was only 1.28% lowe when compa ed o
new o o s.
Table 2. Speci ic su ace a ea o he cemen powde a he beginning and a e 10 and 20 kg.
Inpu Speci ic Su ace A ea (m2/kg) Speci ic Su ace A ea o Resul ing Powde s (m2/kg)
1 kg 10 kg 20 kg
200 310 305 273
450 548 543 541
Table 3. Reduc ion o g inding e iciency.
Inpu Speci ic Su ace A ea (m2/kg) G inding E iciency Reduc ion (%)
1 kg 10 kg 20 kg
200 - 1.61 11.94
450 - 0.91 1.28
Figu e 6a–h show 3D op ical mic oscopy images o he new and wo n o o pins. All he scans
we e pe o med on he o o s wi h h ee ows o pins. Figu e 6a,b depic he 3D scan and heigh
p o ile, espec i ely, o he pins o a new o o , while Figu e 6c,d depic he 3D scan and heigh p o ile,
espec i ely, o he pins o he o o used o milling o a 20 kg ba ch o he 450 m
2
/kg sample. Figu e 6e,
hen depic he 3D scan and heigh p o ile, espec i ely, o he pins o he o o used o he milling o
a 20 kg ba ch o he 200 m
2
/kg sample. The igu es con i m he wea analyses, as hey clea ly show
ha he wea was he mos signi ican o he o o used o milling o he 200 m
2
/kg ba ch; scans o a
pin om he mos wo n o o and i s colou ed heigh p o ile a e shown in Figu e 6g,h. The igu es
also show ha he wea o he pins was he mos signi ican a hei on aces, and also a he ace
on he side om which he cemen pa icles we e injec ed in o he disin eg a o . The igu es also
show e iden “shade” behind he pins, which co esponds o he abo e desc ibed occu ing milling
mechanisms o he indi idual sample ba ches.
Ma e ials2020,13,xFORPEERREVIEW8o 12
expe imen ,anda e millingo 10and20kgba ches.Table3 henp esen s he educ ionso g inding
e iciency,cha ac e izedas educ iono  hespeci icsu acea ea.Wheng inding he ine‐g ained
inpu ma e ialwi h hespeci icsu acea eao 450m
2
/kg, heab asionoccu ed oalesse ex en .
Theweigh lossmeasu ed o  he o o wi h h ee owso pinsa e g indingo a10kgba chwas
0.28%,while o  he o o wi h wo owso pinsi was0.13%; he educ iono  hee iciencyo 
g indingby hewo n o o swasnegligible.Millingo a20kgba cho  he es ingcemen  esul edin
u he  o o weigh loss;a e g indinganaddi ional10kgba ch, heweigh losso  he h ee‐ ow
o o was0.42%,while o  he wo‐ ow o o i wasonly0.30%.Thee iciencyo  he o o swasonly
1.28%lowe whencompa ed onew o o s.
Table2.Speci icsu acea eao  hecemen powde a  hebeginninganda e 10and20kg.
Inpu Speci icSu aceA ea(m
2
/kg)Speci icSu aceA eao Resul ing
Powde s(m
2
/kg)
1kg10kg20kg
200310305273
450548543541
Table3.Reduc iono g indinge iciency.
Inpu Speci icSu aceA ea(m
2
/
kg)G indingE iciencyReduc ion(%)
1kg10kg20kg
200‐1.6111.94
450‐0.911.28
Figu es6a o6hshow3Dop icalmic oscopyimageso  henewandwo n o o pins.All he
scanswe epe o medon he o o swi h h ee owso pins.Figu es6aand6bdepic  he3Dscanand
heigh p o ile, espec i ely,o  hepinso anew o o ,whileFigu es6cand6ddepic  he3Dscanand
heigh p o ile, espec i ely,o  hepinso  he o o used o millingo a20kgba cho  he450m
2
/kg
sample.Figu es6eand6  hendepic  he3Dscanandheigh p o ile, espec i ely,o  hepinso  he
o o used o  hemillingo a20kgba cho  he200m
2
/kgsample.The igu escon i m hewea 
analyses,as heyclea lyshow ha  hewea was hemos signi ican  o  he o o used o millingo 
he200m
2
/kgba ch;scanso apin om hemos wo n o o andi scolou edheigh p o ilea eshown
inFigu es6gand6h.The igu esalsoshow ha  hewea o  hepinswas hemos signi ican a  hei 
on  aces,andalsoa  he aceon heside omwhich hecemen pa icleswe einjec edin o he
disin eg a o .The igu esalsoshowe iden “shade”behind hepins,whichco esponds o heabo e
desc ibedoccu ingmillingmechanisms o  heindi idualsampleba ches.

(a)(b)
Figu e 6. Con .
Ma e ials 2020,13, 2234 9 o 12
Ma e ials2020,13,xFORPEERREVIEW9o 12


(c)(d)


(e)( )

(g)(h)
Figu e6.Op icalimageso new o o :(a)3Dscanand(b)heigh p o ileo pins;op icalimageso 
o o used o millingo 20kgba cho 450m
2
/kgsample:(c)3Dscanand(d)heigh p o ileo pins;
op icalimageso  o o used o millingo 20kgba cho 200m
2
/kgsample:(e)3Dscanand( )heigh 
p o ileo pins.De ailedimageso pinma kedbyci clein( ) om o o used o millingo 20kg
ba cho 200m
2
/kgsample:(g)de ailedscanand(h)colou edp o ile.(Ro a iondi ec ionsma kedby
a ows).
4.Discussion
Thedi e en ac ingpa icleg indingp inciplesimpa edbydi e en inpu g anulome ies
ha econside ablee ec son heab asiono  heg inding o o s,especially hepins.Whenmilling he
sampleswi h heinpu ‐speci icsu acea easo 200and250m
2
/kg, hepinso  heg inding o o sa e
s onglyhea eddue o hedi ec con ac wi h hema e ialbeingg ound,ascon i medby he he mal
imp in sshowninFigu es4aand4b.Thedi ec c ushingand ic iono  hecoa sepa icleson he
on  aceso  he o o pinsalso esul edinasubs an ialinc easein hespeci icsu acea easo  he
pa icles.Thenega i eae odynamicp essu eoccu ing igh behind he o o pinsisno su icien 
oimpa swi lingmo ion o hesolidpa icles.Fo  his eason, hepa iclesa eno milledbymu ual
Figu e 6.
Op ical images o new o o : (
a
) 3D scan and (
b
) heigh p o ile o pins; op ical images o o o
used o milling o 20 kg ba ch o 450 m
2
/kg sample: (
c
) 3D scan and (
d
) heigh p o ile o pins; op ical
images o o o used o milling o 20 kg ba ch o 200 m
2
/kg sample: (
e
) 3D scan and (
) heigh p o ile
o pins. De ailed images o pin ma ked by ci cle in (
) om o o used o milling o 20 kg ba ch o
200 m
2
/kg sample: (
g
) de ailed scan and (
h
) colou ed p o ile. (Ro a ion di ec ions ma ked by a ows).
4. Discussion
The di e en ac ing pa icle g inding p inciples impa ed by di e en inpu g anulome ies ha e
conside able e ec s on he ab asion o he g inding o o s, especially he pins. When milling he
samples wi h he inpu -speci ic su ace a eas o 200 and 250 m
2
/kg, he pins o he g inding o o s
a e s ongly hea ed due o he di ec con ac wi h he ma e ial being g ound, as con i med by he
he mal imp in s shown in Figu e 4a,b. The di ec c ushing and ic ion o he coa se pa icles on he
on aces o he o o pins also esul ed in a subs an ial inc ease in he speci ic su ace a eas o he
pa icles. The nega i e ae odynamic p essu e occu ing igh behind he o o pins is no su icien o
impa swi ling mo ion o he solid pa icles. Fo his eason, he pa icles a e no milled by mu ual
collisions and hus main ain hei sha p-edged mo phology du ing passing h ough he mill; only