Ci a ion: G ego o iˇco á, E.; Pospíšil,
J.; Si ek, T. The Bulk Densi y and
Cohesion o Submic on Pa icles
Emi ed by a Residen ial Boile When
Bu ning Solid Fuels. Fi e 2023,6, 445.
h ps://doi.o g/10.3390/ i e6110445
Academic Edi o s: Michal Holubˇcík
and Nikola ˇ
Cajo áKan o á
Recei ed: 24 Oc obe 2023
Re ised: 10 No embe 2023
Accep ed: 14 No embe 2023
Published: 19 No embe 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
i e
A icle
The Bulk Densi y and Cohesion o Submic on Pa icles Emi ed
by a Residen ial Boile When Bu ning Solid Fuels
E a G ego o iˇco á, Jiˇ íPospíšil * and Tomáš Si ek
Ene gy Ins i u e, B no Uni e si y o Technology, 616 69 B no, Czech Republic; 200185@ u b .cz (E.G.);
omas.si ek@ u b .cz (T.S.)
*Co espondence: [email p o ec ed].cz
Abs ac :
Pa icles emi ed om combus ion sou ces ha e a nega i e impac on human heal h and he
en i onmen . The solid uel boile s ha a e used in households a e a signi ican sou ce o ai pollu ion.
The p esen s udy analyzes pa icula e ma e (PM) p oduced du ing he combus ion o wood pelle s,
wood logs, and coke in a esiden ial boile . The cap u ed pa icles’ pa icle size dis ibu ion, bulk
densi y, and cohesion a e iden i ied. A cascade impac o was used o cap u e he pa icles, and an
op ical digi al mic oscope was used o de e mine he bulk densi y and cohesion (angle o epose) o
he pa icles. The esul s show ha he highes pa icle mass concen a ion emi ed by he boile alls
wi hin he in e al o 89–146 nm, wi h he combus ion o wood logs p oducing he mos pa icles and
coke combus ion p oducing he leas . The bulk densi y o pa icles o 51–595 nm (impac o s ages 4–8)
anges om 320 o 785 kg/m
3
o wood logs, 372–1108 kg/m
3
o wood pelle s, and
435–1330 kg/m3
o coke. The PM1 pa icles (impac o s age 4) ha e he highes bulk densi y. In he pa icle size ange
s udied, he bulk densi y o pa icles dec eases signi ican ly wi h inc easing pa icle size.
Keywo ds: pa icula e ma e ; densi y; impac o ; mic oscope; pa icle cohesion
1. In oduc ion
Solid uel boile s a e s ill a equen sou ce o hea wi hin households and a e also a
signi ican sou ce o ai pollu ion. The emi ed pollu an s ha e a nega i e impac on he
en i onmen and human heal h [
1
]. The combus ion o solid uels in small combus ion
de ices p oduces emissions in he o m o gaseous subs ances and combus ion pa icles.
Combus ion pa icles bind a ious pollu an s (e.g., benzo(a)py ene, hea y me als, and
ola ile o ganic compounds (VOCs)). Pa icles wi h an ae odynamic diame e o less han
10
µ
m (PM10) pene a e he uppe and lowe espi a o y ac . PM2.5 pa icles en e he
b onchi and lung chambe s, and PM1 pa icles pene a e he bloods eam [
2
,
3
]. The size o
he pa icles p oduced is essen ial in o ma ion o assessing hei human heal h e ec s [
4
].
Exposu e o PM10 pa icles gene ally damages he espi a o y and ca dio ascula sys ems,
educes li e expec ancy, inc eases he isk o in an mo ali y, and can cause lung cance [
3
].
In an e o o educe he concen a ion o emi ed ha m ul subs ances, emission limi s
a e se o manu ac u ed boile s [
5
], which a e e i ied when he combus ion sou ce is pu
on he ma ke . Howe e , he e is no u he e i ica ion o compliance wi h hese emission
limi s in long- e m ope a ion. The numbe o pollu an s p oduced by combus ion sou ces is
mainly in luenced by he design o he boile , he way he boile is ope a ed, and he ype o
uel bu ned. Only boile s complying wi h an eco-design can now be pu in o ope a ion [
6
],
and he main a en ion in ecen yea s has been paid o educing he emissions o ine
combus ion pa icles. The amoun o emi ed pa icula e ma e is mos o en epo ed as a
mass o olume concen a ion o pa icula e ma e .
Fo speci ic con e sions be ween he olume and mass o pa icles, knowledge o hei
pa icle densi y is necessa y. The knowledge o pa icle densi y is also necessa y o speci ic
nume ical solu ions o he ajec o ies o pa icles mo ing in a con inuous phase (ai ).
Fi e 2023,6, 445. h ps://doi.o g/10.3390/ i e6110445 h ps://www.mdpi.com/jou nal/ i e
Fi e 2023,6, 445 2 o 13
Pa icle densi y is also essen ial o he design and calib a ion o measu emen appa a us
o pa icle iden i ica ion. Labo a o y pa icle size iden i ica ion is mos o en based on
indi ec me hods moni o ing he elec ic cha ge on he pa icle su ace (
SMPS—scanning
mobili y pa icle size ), he e ec o pa icles on he sca e o he ligh beam (OPS—op ical
pa icle size ), he mass o he cap u ed pa icles (impac o ), o he dynamics o he pa icles’
beha io in he ai s eam (APS—ae odynamic pa icle size ) [
7
]. The pa icle size measu e-
men esul s ob ained using hese me hods a e alid o he espec i e calib a ion pa icles,
bu o o he pa icles, he e is o en an unspeci ied sys ema ic e o . Ano he eason o pay
a en ion o pa icle densi y is ha densi y a ec s he deposi ion o pa icles in he human
espi a o y ac ; hus, i is essen ial o de e mine hei impac on human heal h. I was
ound ha pa icles wi h a highe densi y ha e a highe se ling eloci y han pa icles wi h
a lowe densi y [
8
]. The mo emen o smoke plumes in an a mosphe e is de e mined by he
a mosphe ic condi ions, pa icle size, and pa icle densi y. The g ea e he pa icle densi y
(pa icles a e hea ie ), he highe he se ling eloci y and he mo e likely he pa icles a e
o deposi on su ace a eas. Fu he mo e, pa icle densi y is an impo an p ope y o
il e cake o ma ion du ing il a ion [
9
,
10
]. Pa icle densi y also a ec s he du abili y o
he il e cake, i.e., he e iciency o il e egene a ion. The g ea e he mass o he il e
cake, he mo e e icien he il e egene a ion [
11
]. Pa icle densi y is p ima ily in luenced
by combus ion condi ions ( empe a u e, gas low a e, oxygen con en [
12
], combus ion
cha ac e , uel ype, pa icle size and shape, e c. [13].
Cohesion is a phenomenon caused by he ac ion o a ac i e in e molecula o ces
be ween pa icles o a gi en subs ance, which cause pa icles o he same subs ance o bond
oge he and agglome a e (c ea e a clus e ). In e molecula o ces can be hyd ogen bonds
o an de Waals a ac i e o ces [14].
The e a e wo ypes o pa icle densi ies—pa icle densi y and bulk densi y. Bulk
densi y is commonly de ined o bulk ma e ials and exp esses he a io o he mass o he
solid o he bulk olume, which includes ai - illed in e pa icle oids. The bulk densi y is
less han he pa icle densi y, and he ela ionship be ween he wo is de e mined by he
po osi y o he bulk en i onmen . Po osi y is calcula ed as he a io o he oid olume
o he o al (bulk) olume [
15
]. The magni ude o he bulk densi y gene ally depends on
he size, shape, and su ace a ea o he pa icles, he densi y o he solid pa icles, and he
cohesion o he pa icles [16].
The cu en s a e o knowledge shows ha he e a e ew s udies dealing wi h he
iden i ica ion o he pa icle bulk densi y and pa icle cohesion o di e en size ac ions o
ine pa icles om di e en ypes o uels. This pape ocuses on he de e mina ion o he
bulk densi y o combus ion pa icles by combining he g a ime ic iden i ica ion o he mass
o cap u ed pa icles o di e en size classes and an op ical me hod o de e mine he olume
o cap u ed pa icles. An op ical analysis is also used o iden i y he bulk p ope ies o he
pa icles deposi ed. Speci ically, his pape ocuses on combus ion pa icles emi ed du ing
he combus ion o wood pelle s, wood logs, and coke in a small esiden ial combus ion
de ice. A 14-s age high- empe a u e Deka i low-p essu e impac o (HT-DLPI+) was used
o cap u e pa icula e ma e (PM) om he combus ion p ocess o de e mine he mass size
dis ibu ion o he pa icles in he lue gas in he size ange o 13 nm–50 µm.
The pa icles cap u ed in each impac o s age we e dimensionally e alua ed ia op ical
mic oscopy. The geome ic cha ac e is ics o he pa icle laye (bulk olume, heigh , angle
o epose) we e ob ained using an op ical digi al mic oscope, allowing o he assessmen
o he bulk densi y and cohesion o he pa icles. The no el y o his pape is ha he
e alua ion o he bulk densi y and cohesion was pe o med using a new me hod, he
op ical me hod.
2. Ma e ials and Me hods
Wi hin he amewo k o his s udy, combus ion es s o h ee selec ed uels we e
ca ied ou in a small esiden ial combus ion de ice. The sampling o he pa icles ha we e
ca ied away in he lue gas s eam and hei subsequen mass and op ical iden i ica ion
Fi e 2023,6, 445 3 o 13
we e pe o med. Sepa a ely, he indi idual pa icle size ac ions cap u ed on he cascade
impac o s ages we e moni o ed. Based on he ob ained cha ac e is ics, he pa icle bulk
densi y and he pa icle deposi ion cha ac e is ics o he indi idual pa icle size ac ions
we e e alua ed.
2.1. Combus ion Equipmen and Tes ed Fuels
Fo he combus ion es s, an au oma ic pelle boile wi h a sc ew eede (ATTACK
FD32 6-pa ; yea o manu ac u e, 2019; a ed ou pu , 30 kW) was used. The boile had
been in se ice o h ee yea s as a hea sou ce o a esiden ial building. The a iable
design o he boile cons uc ion allows he bu ning o wood logs and coke in addi ion o
wood pelle s.
Th ee ypes o uel we e bu ned in he boile —wood pelle s, wood logs, and coke. The
ENplus A1 wood pelle s a e made only om pu e sp uce sawdus wi hou admix u es and
we e s o ed in bags o wo o h ee yea s. In addi ion, ha d, d y wood logs (oak) we e
bu ned, which we e s o ed unde a en ila ed shel e o ou yea s. The las uel ha was
bu ned was coke, which was s o ed in a d y en ila ed oom o a minimum o i e yea s.
Table 1shows he p oxima e analysis o he measu ed uels. This able shows he
a e age alues o he h ee measu emen s and sample s anda d de ia ions (SSDs). To
measu e he wa e con en , he samples we e placed in a labo a o y d ye o en a 105
◦
C
o 12 h. Subsequen ly, he samples we e placed in a desicca o o p e en he abso p ion o
ai mois u e [
17
]. Fo he measu emen o he ash con en , he samples in ce amic c ucibles
we e annealed in a kiln a 550
±
10
◦
C o 10 h (a leas 60 min, acco ding o he s anda d),
and he samples we e subsequen ly placed in a desicca o [
18
]. Fo he measu emen o he
ola ile combus ible con en , he samples in ce amic c ucibles we e placed in an o en a
900
±
10
◦
C o 7 min, hen placed in a desicca o [
19
]. The de e mina ion o he hea o
combus ion, also known as he highe hea ing alue (HHV), was ca ied ou using a Pa
6200 calo ime e [20].
Table 1. P oxima e analysis o measu ed samples.
Fuel Wa e Con en (%) SSD
(%)
Ash Con en
(%)
SSD
(%)
Vola ile Ma e
(%)
SSD
(%)
HHV
(MJ/kg)
SSD
(%)
Wood pelle s 6.9 0.04 0.4 0.01 85.2 0.18 18.2 0.16
Wood logs 8.4 0.04 0.9 0.07 85.6 0.38 17.4 0.18
Coke 0.6 0.01 23.2 0.81 1.7 0.05 29.8 0.38
2.2. Combus ion Tes s
P io o each s oking, he boile , including he hea exchange s and he lue pipe,
was cleaned o ensu e ha no a deposi s we e p esen in he boile o lue. The pelle
combus ion was au oma ic using a sc ew eede and he bu ne powe was se a 17 kW.
Sampling was ca ied ou in a s eady-s a e ope a ion a a a ed powe le el. A e he
boile was ebuil ( emo ing he pelle bu ne , eplacing he ash doo , e c.), he boile was
ope a ed on wood logs and hen on coke. Boile s oking o he wood logs was ca ied ou
o each he boile ope a ing empe a u e in he sho es possible ime. Subsequen ly, la ge
wood logs we e added. The seconda y ai o he boile was ully opened. Measu emen s
we e aken wi h he boile ope a ing a a s eady s a e, wi h he boile wa e empe a u e
main ained a 80
◦
C. The coke was bu ned in he same manne , bu coke akes longe o
kindle. Coke bu ns wi h a small lame o only glows compa ed o wood logs. The coke was
ed o he back o he boile . Du ing he bu ning o he coke, only he p ima y ai supply
was p o ided unde he g a e; he seconda y ai was closed. The oxygen con en o he lue
gas du ing he lue gas sampling was ideal, a app oxima ely 10%.
2.3. Pa icle Sampling
Flue gas sampling was ca ied ou in he lue pipe wi h a diame e o 150 mm, dis-
cha ging he lue gases om he boile . A sampling isokine ic p obe, placed in he lue
Fi e 2023,6, 445 4 o 13
pipe a app oxima ely 1 m abo e he boile , was used o sample he lue gas. The sampling
p obe was only o ien ed agains he di ec ion o he lue gas low du ing he sampling
pe iod. Fo he emaining ime, he p obe was o ien ed in he di ec ion o he lue gas low
o a oid he unwan ed ouling o he p obe. Flue gas sampling was ca ied ou du ing
he s able combus ion phase when he lue gas empe a u e was kep cons an . F om he
sampling p obe, he lue gas passed h ough an insula ed pipe ha connec ed he p obe o
he impac o ; see Figu e 1.
Fi e 2023, 6, x FOR PEER REVIEW 4
logs. The coke was ed o he back o he boile . Du ing he bu ning o he coke, only he
p ima y ai supply was p o ided unde he g a e; he seconda y ai was closed. The ox-
ygen con en o he lue gas du ing he lue gas sampling was ideal, a app oxima ely 10%.
2.3. Pa icle Sampling
Flue gas sampling was ca ied ou in he lue pipe wi h a diame e o 150 mm,
discha ging he lue gases om he boile . A sampling isokine ic p obe, placed in he
lue pipe a app oxima ely 1 m abo e he boile , was used o sample he lue gas. The
sampling p obe was only o ien ed agains he di ec ion o he lue gas low du ing he sam-
pling pe iod. Fo he emaining ime, he p obe was o ien ed in he di ec ion o he lue
gas low o a oid he unwan ed ouling o he p obe. Flue gas sampling was ca ied ou
du ing he s able combus ion phase when he lue gas empe a u e was kep cons an .
F om he sampling p obe, he lue gas passed h ough an insula ed pipe ha connec ed
he p obe o he impac o ; see Figu e 1.
Figu e 1. Expe imen al se up.
The sampling o he lue gas o each uel ype was epea ed h ee imes; hus, 9 meas-
u emen s we e made wi h he impac o du ing he s eady-s a e combus ion ope a ion.
The lue gas sampling imes and lue gas empe a u es o he wood pelle s, wood logs,
and coke combus ion a e shown in Table 2.
Table 2. Sampling ime and lue gas empe a u e o indi idual uels.
Fuel Sampling Time (min) Flue Gas Tempe a u e (°C)
Wood pelle s 15 150
Wood logs 5 225
Coke 25 210
The sampled lue gas was in oduced in o he high- empe a u e Deka i low-p essu e
impac o . The impac o is a de ice used o g a ime ically de e mine he dis ibu ion o
pa icles. I consis s o ou een s ages h ough which he lue gas passes, and pa icles in
he size ac ions o 13 nm–50 μm a e cap u ed. The low o he lue gas sample h ough
he impac o is p o ided by a pump connec ed o he ou le o he impac o . The
exhaus gas sampling ime was de e mined based on es measu emen s o each uel.
P io o sampling, he impac o was hea ed o 130 °C o p e en he passing gas om
condensing in he impac o . Du ing he lue gas sampling, he p essu e was se a 40 mba
on he acuum pump, and he empe a u e o he gas en e ing he impac o was
Figu e 1. Expe imen al se up.
The sampling o he lue gas o each uel ype was epea ed h ee imes; hus,
9 measu emen s
we e made wi h he impac o du ing he s eady-s a e combus ion op-
e a ion. The lue gas sampling imes and lue gas empe a u es o he wood pelle s, wood
logs, and coke combus ion a e shown in Table 2.
Table 2. Sampling ime and lue gas empe a u e o indi idual uels.
Fuel Sampling Time (min) Flue Gas Tempe a u e (◦C)
Wood pelle s 15 150
Wood logs 5 225
Coke 25 210
The sampled lue gas was in oduced in o he high- empe a u e Deka i low-p essu e
impac o . The impac o is a de ice used o g a ime ically de e mine he dis ibu ion o
pa icles. I consis s o ou een s ages h ough which he lue gas passes, and pa icles in
he size ac ions o 13 nm–50
µ
m a e cap u ed. The low o he lue gas sample h ough
he impac o is p o ided by a pump connec ed o he ou le o he impac o . The exhaus
gas sampling ime was de e mined based on es measu emen s o each uel. P io o
sampling, he impac o was hea ed o 130
◦
C o p e en he passing gas om condensing
in he impac o . Du ing he lue gas sampling, he p essu e was se a 40 mba on he
acuum pump, and he empe a u e o he gas en e ing he impac o was con inuously
eco ded. The lue gas passing h ough he impac o lows h ough indi idual s ages,
which con ain aluminum oils coa ed wi h Vaseline. The oils a e used o g adually cap u e
pa icula e ma e ia impac ion. A ound he oils a e openings h ough which he lue
gas sample lows o he nex s ages o he impac o , which ha e smalle and smalle
openings (condi ions o highe lue gas eloci y—sepa a ion o smalle pa icles). To
de e mine he g a ime ic dis ibu ion o pa icles, he oils we e weighed be o e and a e
he measu emen s. The weighing o he clean and used oils was epea ed h ee imes and
Fi e 2023,6, 445 5 o 13
he inal weigh was ob ained by a e aging he alues ob ained. P io o weighing, he
elec ical cha ge o he aluminum oils was emo ed using a ame ionize .
2.4. Op ical Iden i ica ion o Cap u ed Pa icles
The de e mina ion o he olume and bulk p ope ies o he pa icles cap u ed on
he oil was pe o med using a Keyence VHX-7000 op ical digi al mic oscope capable o
20–7000×
op ical magni ica ion. The bulk olume o he cap u ed pa icles was ob ained ia
he 3D imaging o he cap u ed pa icles; see Figu e 2b. The 3D image was ob ained a a high
esolu ion using he “Fine Dep h Composi ion” unc ion a 700
×
o 500
×
magni ica ion o
assess he olume o a pa icula impac ion deposi sample mo e accu a ely. The cap u ed
pa icles on he oils o med se e al sepa a e impac ion deposi s, he numbe o which
co esponded o he numbe o openings o each impac o s age; see Figu e 2a.
Fi e 2023, 6, x FOR PEER REVIEW 5
con inuously eco ded. The lue gas passing h ough he impac o lows h ough indi id-
ual s ages, which con ain aluminum oils coa ed wi h Vaseline. The oils a e used o g ad-
ually cap u e pa icula e ma e ia impac ion. A ound he oils a e openings h ough
which he lue gas sample lows o he nex s ages o he impac o , which ha e smalle
and smalle openings (condi ions o highe lue gas eloci y—sepa a ion o smalle
pa icles). To de e mine he g a ime ic dis ibu ion o pa icles, he oils we e weighed
be o e and a e he measu emen s. The weighing o he clean and used oils was epea ed
h ee imes and he inal weigh was ob ained by a e aging he alues ob ained. P io o
weighing, he elec ical cha ge o he aluminum oils was emo ed using a ame ionize .
2.4. Op ical Iden i ica ion o Cap u ed Pa icles
The de e mina ion o he olume and bulk p ope ies o he pa icles cap u ed on he
oil was pe o med using a Keyence VHX-7000 op ical digi al mic oscope capable o 20–
7000× op ical magni ica ion. The bulk olume o he cap u ed pa icles was ob ained
ia he 3D imaging o he cap u ed pa icles; see Figu e 2b. The 3D image was ob ained
a a high esolu ion using he “Fine Dep h Composi ion” unc ion a 700× o 500× magni-
ica ion o assess he olume o a pa icula impac ion deposi sample mo e accu a ely.
The cap u ed pa icles on he oils o med se e al sepa a e impac ion deposi s, he
numbe o which co esponded o he numbe o openings o each impac o s age; see
Figu e 2a.
Figu e 2. Example o an op ical mic oscope iew: (a) iew o he whole oils and (b) de ail o one
impac ion pa icle deposi .
Consequen ly, he s i ching o he pa ial 3D images was pe o med o display a 3D
image o he whole oils. Subsequen ly, he il co ec ion o he 3D image was pe o med
by de ining he base plane. Since he pa icle olumes o indi idual impac o deposi s a y
only negligibly wi hin a single impac o s age, he o al olume o he pa icles cap u ed in
a single s age was ob ained as he a e age olume o he e e ence one deposi mul iplied
by he o al numbe o he deposi s o he espec i e impac o s age. The geome y o he
e e ence impac ion deposi was ob ained as he a e age o he geome ic cha ac e is ics
o a leas 3 analyzed impac ion deposi s wi hin a single impac o s age. F om he meas-
u ed bulk olume and sample weigh , he bulk pa icle densi y o he h ee es ed uel
samples was calcula ed. The bulk densi y was iden i ied wi h su icien accu acy o a
pa icle size ange o 51–595 nm (impac o s ages 4–8). In his size ange, he pa icle dep-
osi ion olumes iden i ied wi h he mic oscope a e signi ican . The pa icle deposi s in he
o he impac o s ages we e oo small o allow o he op ical iden i ica ion o hei olume
wi h su icien accu acy.
Using he da a ob ained h ough he mic oscopic analysis, a 3D p o ile o he e e ence
impac o deposi was u he ob ained o e alua e he angle o epose. Fo his pu pose,
he angle be ween he ho izon al plane and he maximum slope o he deposi con ou
was moni o ed; see Figu e 3.
Figu e 2.
Example o an op ical mic oscope iew: (
a
) iew o he whole oils and (
b
) de ail o
one impac ion pa icle deposi .
Consequen ly, he s i ching o he pa ial 3D images was pe o med o display a 3D
image o he whole oils. Subsequen ly, he il co ec ion o he 3D image was pe o med
by de ining he base plane. Since he pa icle olumes o indi idual impac o deposi s a y
only negligibly wi hin a single impac o s age, he o al olume o he pa icles cap u ed in
a single s age was ob ained as he a e age olume o he e e ence one deposi mul iplied
by he o al numbe o he deposi s o he espec i e impac o s age. The geome y o he
e e ence impac ion deposi was ob ained as he a e age o he geome ic cha ac e is ics o
a leas 3 analyzed impac ion deposi s wi hin a single impac o s age. F om he measu ed
bulk olume and sample weigh , he bulk pa icle densi y o he h ee es ed uel samples
was calcula ed. The bulk densi y was iden i ied wi h su icien accu acy o a pa icle
size ange o 51–595 nm (impac o s ages 4–8). In his size ange, he pa icle deposi ion
olumes iden i ied wi h he mic oscope a e signi ican . The pa icle deposi s in he o he
impac o s ages we e oo small o allow o he op ical iden i ica ion o hei olume wi h
su icien accu acy.
Using he da a ob ained h ough he mic oscopic analysis, a 3D p o ile o he e e ence
impac o deposi was u he ob ained o e alua e he angle o epose. Fo his pu pose,
he angle be ween he ho izon al plane and he maximum slope o he deposi con ou was
moni o ed; see Figu e 3.

Fi e 2023,6, 445 6 o 13
Fi e 2023, 6, x FOR PEER REVIEW 6
Figu e 3. Analysis o impac ion deposi wi h op ical digi al mic oscope.
3. Resul s
The measu ed pa icle mass concen a ions a e gi en unde no mal condi ions
(101.325 kPa, 0 °C, and 0% gas humidi y), and e e enced o an oxygen le el o 10%.
3.1. Mass Dis ibu ion o Pa icles
The pa icle size dis ibu ion was iden i ied g a ime ically om he pa icle samples
collec ed a each s age o he cascade impac o . To compa e he es ed uels wi h each
o he , he esul s o he measu emen s made wi h he same uel we e a e aged and no -
malized based on he sampling ime. The X-axis in he size dis ibu ion plo (Figu e 4)
ep esen s he pa icle size and he Y-axis is he mass concen a ion o he cap u ed pa i-
cles. The igu e shows ha he lue gas om wood logs bu ning in he boile ca ies he
la ges mass concen a ion o pa icles o he obse ed size, ollowed by he wood pelle s,
and he lue gas om coke bu ning ca ies he leas pa icles. The highes pa icle mass
concen a ion was iden i ied o pa icles in he size ange o 89–146 nm (s age 5) o all
h ee uels es ed. Wood logs, wood pelle s, and coke con ibu e 50%, 30%, and 20%, e-
spec i ely, o he o al sum o he mass concen a ions o pa icles emi ed om all h ee
uels. Table 3 shows he pe cen age mass concen a ions o pa icles o each size ac ion.
Al hough coke p oduces he lowes pa icle mass concen a ion, i p oduces 12% pe cen
mo e PM1 pa icles in he pa icle size ange.
Table 3. Pe cen age o size ac ions o indi idual uels
Fuel PM1 (%)
PM1–2.5 (%) PM2.5–10 (%)
Wood pelle s 50 47 3
Wood logs 50 45 6
Coke 62 36 2
Figu e 3. Analysis o impac ion deposi wi h op ical digi al mic oscope.
3. Resul s
The measu ed pa icle mass concen a ions a e gi en unde no mal condi ions
(101.325 kPa, 0 ◦C, and 0% gas humidi y), and e e enced o an oxygen le el o 10%.
3.1. Mass Dis ibu ion o Pa icles
The pa icle size dis ibu ion was iden i ied g a ime ically om he pa icle samples
collec ed a each s age o he cascade impac o . To compa e he es ed uels wi h each o he ,
he esul s o he measu emen s made wi h he same uel we e a e aged and no malized
based on he sampling ime. The X-axis in he size dis ibu ion plo (Figu e 4) ep esen s
he pa icle size and he Y-axis is he mass concen a ion o he cap u ed pa icles. The
igu e shows ha he lue gas om wood logs bu ning in he boile ca ies he la ges mass
concen a ion o pa icles o he obse ed size, ollowed by he wood pelle s, and he lue
gas om coke bu ning ca ies he leas pa icles. The highes pa icle mass concen a ion
was iden i ied o pa icles in he size ange o 89–146 nm (s age 5) o all h ee uels es ed.
Wood logs, wood pelle s, and coke con ibu e 50%, 30%, and 20%, espec i ely, o he o al
sum o he mass concen a ions o pa icles emi ed om all h ee uels. Table 3shows he
pe cen age mass concen a ions o pa icles o each size ac ion. Al hough coke p oduces
he lowes pa icle mass concen a ion, i p oduces 12% pe cen mo e PM1 pa icles in he
pa icle size ange.
Fi e 2023, 6, x FOR PEER REVIEW 7
Figu e 4. Pa icle size dis ibu ion o h ee measu ed uels wi h HT-DLPI+.
3.2. Pa icle Image Analysis
Figu e 5 shows de ails o he pa icles cap u ed wi h he impac o du ing wood pel-
le , wood log, and coke combus ion o he selec ed size classes (s ages 7, 9, and 12). The
i s column o igu es shows pa icles o 247–373 nm, he second shows 595–939 nm, and
he hi d shows 2460–3650 nm. In gene al, he pa icles om wood pelle s a e black, om
wood logs, hey a e b own, and om coke, hey a e whi e o black–whi e–b own, depend-
ing on he combus ion empe a u e. The coke pa icles ha a e shown a e whi e, and he
la ge coke pa icles a e o ange. In he case o a di e en ype o boile and di e en com-
bus ion condi ions, he chemical composi ion o he pa icles may a y, and he e o e, he
colo o he pa icles may also be di e en . Howe e , he chemical analysis and op ical
analysis o he pa icles in e ms o hei pa icle colo unde di e en combus ion condi-
ions is beyond he scope o his a icle.
0
2
4
6
8
10
12
14
Pa icle concen a ion [mg/Nm3]
Pa icle size [μm]
Wood pelle s
Wood logs
Coke
Figu e 4. Pa icle size dis ibu ion o h ee measu ed uels wi h HT-DLPI+.
Fi e 2023,6, 445 7 o 13
Table 3. Pe cen age o size ac ions o indi idual uels.
Fuel PM1 (%) PM1–2.5 (%) PM2.5–10 (%)
Wood pelle s 50 47 3
Wood logs 50 45 6
Coke 62 36 2
3.2. Pa icle Image Analysis
Figu e 5shows de ails o he pa icles cap u ed wi h he impac o du ing wood pelle ,
wood log, and coke combus ion o he selec ed size classes (s ages 7, 9, and 12). The
i s column
o igu es shows pa icles o 247–373 nm, he second shows 595–939 nm, and
he hi d shows 2460–3650 nm. In gene al, he pa icles om wood pelle s a e black, om
wood logs, hey a e b own, and om coke, hey a e whi e o black–whi e–b own, depending
on he combus ion empe a u e. The coke pa icles ha a e shown a e whi e, and he la ge
coke pa icles a e o ange. In he case o a di e en ype o boile and di e en combus ion
condi ions, he chemical composi ion o he pa icles may a y, and he e o e, he colo o
he pa icles may also be di e en . Howe e , he chemical analysis and op ical analysis
o he pa icles in e ms o hei pa icle colo unde di e en combus ion condi ions is
beyond he scope o his a icle.
Fi e 2023, 6, x FOR PEER REVIEW 8
Figu e 5. De ail o cap u ed pa icles: (a) wood pelle s; (b) wood logs; (c) coke.
3.3. Bulk Densi y o Pa icles
Table 4 shows he a e age alues o he iden i ied pa icle bulk densi ies ob ained o
he pa icle size ac ion o 51–595 nm. The esul s show ha he pa icles emi ed du -
ing wood log combus ion ha e he lowes bulk densi y, ollowed by pa icles emi ed
du ing pelle combus ion, and he pa icles emi ed du ing coke combus ion ha e he
highes bulk densi y.
Fo all uels, he bulk densi y o he pa icles dec eases wi h he inc easing pa icle
size. The g ea e measu emen unce ain y is due o he measu emen me hod— he abili y
o he ins umen o cap u e he en i e olume o he deposi in 3D. A la ge numbe o
measu emen s could educe he measu emen unce ain y. Figu e 6 shows he bulk den-
si ies wi h a linea end line (dashed lines), exp essed based on he ollowing equa ions:
Coke: 𝑦 = −1974𝑥 + 1384 (1)
Wood pelle s: 𝑦 = −1974𝑥 + 1384 (2)
Wood logs: 𝑦 = −1974𝑥 + 1384 (3)
Table 4. Rela ionship be ween iden i ied pa icle bulk densi y and pa icle size.
Pa icle Size (µm) Bulk Densi y (kg/m3)
S age Range Wood Pelle s SSD Wood Logs SSD Coke SSD
4 0.0512–0.0887 1108 215 785 70 1330 165
5 0.0887–0.146 716 98 696 104 1080 47
6 0.146–0.247 661 86 597 93 978 110
7 0.247–0.373 569 86 492 89 806 231
8 0.373–0.595 372 30 320 99 435 69
Figu e 5. De ail o cap u ed pa icles: (a) wood pelle s; (b) wood logs; (c) coke.
3.3. Bulk Densi y o Pa icles
Table 4shows he a e age alues o he iden i ied pa icle bulk densi ies ob ained o
he pa icle size ac ion o 51–595 nm. The esul s show ha he pa icles emi ed du ing
wood log combus ion ha e he lowes bulk densi y, ollowed by pa icles emi ed du ing
pelle combus ion, and he pa icles emi ed du ing coke combus ion ha e he highes bulk
densi y.
Fi e 2023,6, 445 8 o 13
Table 4. Rela ionship be ween iden i ied pa icle bulk densi y and pa icle size.
Pa icle Size (µm) Bulk Densi y (kg/m3)
S age Range Wood Pelle s SSD Wood Logs SSD Coke SSD
4 0.0512–0.0887 1108 215 785 70 1330 165
5 0.0887–0.146 716 98 696 104 1080 47
6 0.146–0.247 661 86 597 93 978 110
7 0.247–0.373 569 86 492 89 806 231
8 0.373–0.595 372 30 320 99 435 69
Fo all uels, he bulk densi y o he pa icles dec eases wi h he inc easing pa icle
size. The g ea e measu emen unce ain y is due o he measu emen me hod— he abili y
o he ins umen o cap u e he en i e olume o he deposi in 3D. A la ge numbe
o measu emen s could educe he measu emen unce ain y. Figu e 6shows he bulk
densi ies wi h a linea end line (dashed lines), exp essed based on he ollowing equa ions:
Coke: y=−1974x+ 1384 (1)
Wood pelle s: y=−1974x+ 1384 (2)
Wood logs: y=−1974x+ 1384 (3)
Fi e 2023, 6, x FOR PEER REVIEW 9
Figu e 6. Rela ionship be ween iden i ied pa icle bulk densi y and pa icle size.
3.4. E alua ion o Cohesion
Pa icle cohesion was assessed using a geome ic analysis o he shape o he cap u ed
pa icles’ p o ile. The con ou line o he p o ile was ob ained h ough an au oma ed mi-
c oscopic measu emen in a selec ed e ical plane passing h ough he highes poin o
he pa icle deposi ion. The cap u ed pa icles o smalle sizes (s ages 3–6) o m a nea ly
o a ionally symme ic compac pa icle deposi below each impac ion nozzle. Pa icles
o la ge sizes (s age 7 and abo e) o m an i egula ly shaped deposi . This end is he
same o all uels. Ano he op ically obse able end o all uels is he dec ease in he
cohesion o he pa icle a angemen wi h he inc easing size o he pa icles, which is ully
consis en wi h he bulk densi y pa e n depic ed in Figu e 6. A simila conclusion was
eached by Shi e al. in hei s udy, o pa icle sizes up o 150 μm [21].
Figu e 7 shows an example o he shape o he iden i ied con ou s o he pa icula e
deposi s ob ained du ing wood log bu ning in impac o s ages 3–8. The X- and Y-axes
a e no o scale. The cohesion o he pa icles was e alua ed by iden i ying he angle
be ween he ho izon al plane and he g ea es slope o he deposi con ou ( he angle o e-
pose); see Figu e 3. The a e age angle o epose o he deposi ion o he cap u ed pa icles
is 30–40° o he smalles pa icles (s age 4). As he pa icle size inc eases, he angle o
epose inc eases and app oaches 90°. The angles o epose o he di e en uels do no
di e signi ican ly om each o he . The e o e, he a e age alues o he angle o epose o
all es ed uels we e e alua ed; see Table 5. An e alua ion o he de ailed di e ences
in he bulk p ope ies o he pa icles emi ed by he di e en uels canno be ob ained
using his p ocedu e. The condi ion o iden ical pa icle mass deposi ion om di e en
uels is no ensu ed o a de ailed e alua ion.
0
200
400
600
800
1000
1200
1400
1600
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Bulk densi y [kg/m3]
Pa icle size [µm]
Coke
Wood pelle s
Wood logs
Figu e 6. Rela ionship be ween iden i ied pa icle bulk densi y and pa icle size.
3.4. E alua ion o Cohesion
Pa icle cohesion was assessed using a geome ic analysis o he shape o he cap u ed
pa icles’ p o ile. The con ou line o he p o ile was ob ained h ough an au oma ed
mic oscopic measu emen in a selec ed e ical plane passing h ough he highes poin o
he pa icle deposi ion. The cap u ed pa icles o smalle sizes (s ages 3–6) o m a nea ly
o a ionally symme ic compac pa icle deposi below each impac ion nozzle. Pa icles
o la ge sizes (s age 7 and abo e) o m an i egula ly shaped deposi . This end is he
same o all uels. Ano he op ically obse able end o all uels is he dec ease in he
Fi e 2023,6, 445 9 o 13
cohesion o he pa icle a angemen wi h he inc easing size o he pa icles, which is ully
consis en wi h he bulk densi y pa e n depic ed in Figu e 6. A simila conclusion was
eached by Shi e al. in hei s udy, o pa icle sizes up o 150 µm [21].
Figu e 7shows an example o he shape o he iden i ied con ou s o he pa icula e
deposi s ob ained du ing wood log bu ning in impac o s ages 3–8. The X- and Y-axes a e
no o scale. The cohesion o he pa icles was e alua ed by iden i ying he angle be ween
he ho izon al plane and he g ea es slope o he deposi con ou ( he angle o epose);
see Figu e 3. The a e age angle o epose o he deposi ion o he cap u ed pa icles is
30–40
◦
o he smalles pa icles (s age 4). As he pa icle size inc eases, he angle o epose
inc eases and app oaches 90
◦
. The angles o epose o he di e en uels do no di e
signi ican ly om each o he . The e o e, he a e age alues o he angle o epose o all
es ed uels we e e alua ed; see Table 5. An e alua ion o he de ailed di e ences in he
bulk p ope ies o he pa icles emi ed by he di e en uels canno be ob ained using his
p ocedu e. The condi ion o iden ical pa icle mass deposi ion om di e en uels is no
ensu ed o a de ailed e alua ion.
Fi e 2023, 6, x FOR PEER REVIEW 10
Figu e 7. P o ile o cap u ed pa icles.
Table 5. A e age angles o epose.
S age Pa icle Size (µm) Angle o Repose (°)
4 0.0512–0.0887 34
5 0.0887–0.146 45
6 0.146–0.247 50
7 0.247–0.373 55
8 0.373–0.595 84
4. Discussion
The s udies om S inge e al. and F eeman indica e ha wi h a smalle pa icle size
(less mass), he con ibu ion o g a i a ional o ces dec eases and he in luence o in e pa -
icle o ces inc eases [22,23]. Speci ically, o pa icle sizes smalle han 100 μm, he in-
e pa icle a ac i e an de Waals o ces exceed g a i y. When he a ac i e o ces ex-
ceed he mass o he pa icle by a leas an o de o magni ude, he pa icles become e y
cohesi e and o m agg ega es (s ongly bound pa icles). A he same ime, smalle pa i-
cles ha e a ela i ely la ge speci ic su ace a ea, esul ing in e y high adhesion and an
associa ed inc ease in in e pa icle o ces [24]. In gene al, smalle pa icles end o be mo e
cohesi e o hese easons. La ge pa icles end o pack igh ly in o homogeneous laye s
wi h a igid s uc u e. Such laye s may ha e signi ican oid spaces be ween ouching
pa icles bu a e gene ally ee o la ge oids [23]. Gene ally, he smalle he pa icles, he
mo e hey s ick oge he due o smalle oid spaces, esul ing in a highe bulk densi y and
cohesion. Howe e , i mus be conside ed ha he cohesion and bulk densi y a e also in-
luenced by he pa icle ype and o he pa icle p ope ies (see In oduc ion).
The cohesion o ine pa icles signi ican ly in luences he o ma ion o pa icle clus-
e s and pa icle deposi s in he lue gas pa hs. Depending on he p ope ies o he pa i-
cles, he pa icles a e ei he pe manen ly joined o only b ie ly con ac ed wi h subsequen
sepa a ion. The main enance o long- e m pa icle con ac is used, o example, in he il-
a ion o ine pa icles using po ous il e s [25–28]. The po es o he il e a e, in mos
cases, signi ican ly la ge in size han he pa icles o be cap u ed. The cap u e p ocess
0
20
40
60
80
100
120
140
160
180
200
1000 2000 3000
Heigh [μm]
Leng h [μm]
S age 3
S age 4
S age 5
S age 6
S age 7
S age 8
Figu e 7. P o ile o cap u ed pa icles.
Table 5. A e age angles o epose.
S age Pa icle Size (µm) Angle o Repose (◦)
4 0.0512–0.0887 34
5 0.0887–0.146 45
6 0.146–0.247 50
7 0.247–0.373 55
8 0.373–0.595 84
4. Discussion
The s udies om S inge e al. and F eeman indica e ha wi h a smalle pa icle
size (less mass), he con ibu ion o g a i a ional o ces dec eases and he in luence o
in e pa icle o ces inc eases [
22
,
23
]. Speci ically, o pa icle sizes smalle han 100
µ
m,
he in e pa icle a ac i e an de Waals o ces exceed g a i y. When he a ac i e o ces
exceed he mass o he pa icle by a leas an o de o magni ude, he pa icles become
e y cohesi e and o m agg ega es (s ongly bound pa icles). A he same ime, smalle