Nuno Miguel Gonçal es Pinela
Licenciado em Ciências da Engenha ia de Mic o e Nano ecnologias
Piezo esis i e p essu e senso o applica ion in
e-skin de ices
Disse ação pa a ob enção do G au de Mes e em
Engenha ia de Mic o e Nano ecnologias
O ien ado : Dou o Rui Albe o Ga ção Ba ei a do Nascimen o
Ig eja, P o esso auxilia , Faculdade de Ciências e
Tecnologia da Uni e sidade No a de Lisboa
Co-o ien ado : Dou o Hugo Manuel B i o Águas,
P o esso auxilia , Faculdade de Ciências e
Tecnologia da Uni e sidade No a de Lisboa
Jú i
P esiden e: Dou o Rod igo Fe ão de Pai a Ma ins
A guen e: Dou o Ca los Jo ge Ma iano Mi anda Dias
Vogal: Dou o Rui Albe o Ga ção Ba ei a do Nascimen o Ig eja
Se emb o, 2017
Piezo esis i e p essu e senso o applica ion in e-skin de ices
Copy igh
©
Nuno Miguel Gonçal es Pinela, Faculdade de Ciências e Tecnologia, Uni e -
sidade NOVA de Lisboa.
A Faculdade de Ciências e Tecnologia e a Uni e sidade NOVA de Lisboa êm o di ei o,
pe pé uo e sem limi es geog á icos, de a qui a e publica es a disse ação a a és de
exempla es imp essos ep oduzidos em papel ou de o ma digi al, ou po qualque ou o
meio conhecido ou que enha a se in en ado, e de a di ulga a a és de eposi ó ios
cien í icos e de admi i a sua cópia e dis ibuição com obje i os educacionais ou de in es-
igação, não come ciais, desde que seja dado c édi o ao au o e edi o .
Es e documen o oi ge ado u ilizando o p ocessado (pd )L
A
TEX, com base no empla e “no a hesis” [1] desen ol ido no Dep. In o má ica da FCT-NOVA [2].
[1] h ps://gi hub.com/joaomlou enco/no a hesis [2] h p://www.di. c .unl.p
"T y no o become a man o success, bu a he y o become a
man o alue."
Albe Eins ein

Acknowledgemen s
I would like o acknowledge e e yone who con ibu ed o his mas e hesis o hap-
pen and o be concluded, s a ing om P o esso Rod igo Ma ins and P o esso El i a
Fo una o, who ha e de eloped hese wo mul idisciplina y esea ch cen e s, which a e
CENIMAT|i3N and CEMOP, whe e I ha e spen my las nine mon hs and whe e I ound
all he ools o su pass ba ie s and acqui e knowledge. I would also like o show my
acknowledgemen o his g ea ins i u ion, FCT-UNL, whe e I ind mysel g owing and
e ol ing e e yday, since he i s day I joined i .
Secondly, I would like o exp ess my g a i ude o P o esso Rui Ig eja, o accep ing
and in oducing me o his wide wo ld o senso s. Fu he mo e, I would like o hank
o he guidance and o ien a ion in e e y single mee ing. I am also g a e ul o And eia,
who has suppo ed me and con ibu ed signi ica i ely o his hesis. To P o esso Hugo
Águas, I also wan o hank o he guidance.
Mo eo e , I wan o hank CENIMAT|i3N membe s Alexand a Gonçal es, Sónia
Pe ei a, Ana Ca olina Ma ques, Ana Samouco, Bea iz Coelho, Rica do Fe ei a, Tiago
Ma eus, Rod igo San os, Ped o Al es, Daniela Gomes and Tomás Calmei o, who ha e
always p o ided me he help I needed, in e ms o sugges ions, logis ics and echnical
suppo .
To all my iends om Open-Space, I wan o gi e special hanks because in spi e o
ou du y o comple e his ha d ask o w i ing a mas e hesis, we managed o ha e un
and o help each o he s in one o ano he si ua ion.
To my cou se iends who ha e been hese las i e yea s wi h me, João A onso, Ma co
and Shi . To my iend Tiago Gamei o, I would eally like o exp ess my g a i ude, o
he good imes we ha e spen du ing hese yea s. To my long- ime iends Alexand e
Fe ão, Gonçalo Félix and Dua e Pais, who ha e been an impo an suppo . Also, I wan
o hank o my gi l iend’s amily o he suppo .
I wan o specially acknowledge all my amily. To my g andpa en s o all he p ide
hey pu on me. To my pa en s Anabela and An ónio, o conceding me his oppo uni y
o educa ion. I will be o e e g a e ul, and will always p aise on ollow he p inciples I
ha e been augh . To my b o he Ped o, who has been a majo pilla h oughou he yea s.
I also wan o hank Noa o he uncondi ional lo e. To Ma iana, who has made his walk
alongside me, making i much easie . A ew lines he e would no be enough o exp ess
how hank ul I am o all he lo e, suppo , mo i a ion and o e e y hing she is.
ii
Abs ac
In ecen yea s, he ad ancemen o science and echnology ends o e ol e owa ds he
exploi a ion o elec onic skin (e-skin) and unc ional p os he ic de ices, enabling inno-
a ing applica ions in a ious ields such as biomedical sys ems, spo s heal h-moni o ing
and heal hca e. Owing o hei signi ican ole in heal h moni o ing, p essu e senso s
come as essen ial componen s in he de elopmen o a i icial sys ems ha can mimic he
imp essi e human skin. The de elopmen o such senso s comp ises he sea ch o lexi-
ble and s e chable ma e ials sui able o implemen a ion in obus de ices ha enable
he in eg a ion o mul iple sensing- unc ionali ies. To quan i a i ely moni o p essu e,
hese senso s use ansduc ion me hods based on piezo esis i i y, capaci y, piezoelec ic-
i y, and iboelec ici y.
In his wo k, piezo esis i e de ices we e chosen o e o he s due o hei ease in s uc-
u e design and eadou mechanism. The mechanism o such piezo esis i e p essu e
senso elies on he ansduc ion o a p essu e change in o a change in esis ance ha , in
his case de i es om a ia ions in he con ac a ea.
In he app oach p esen ed in his wo k, a semi-sphe e mic os uc u ing pa e n-
ing made by lase eng a ing on ha d-poly(dime hylsiloxane) (h-PDMS) was in oduced.
h-PDMS wo ks as a mold om which s anda d-poly(dime hylsiloxane) (s-PDMS) mi-
c os uc u ed memb anes wi h app oxima ely 200
µ
m hickness a e peeled o . Ca bon-
ink, wo king as ac i e ma e ial, was deposi ed on op o he mic os uc u ed s-PDMS
memb anes. The ab ica ion o such p essu e senso s based on o ganic memb anes com-
bines ad an ages such as he p oduc ion in a low-cos and as way, de ice lexibili y, and
unabili y o he senso ’s design. Mo eo e , sensi i i ies o 2.4
×10−1kPa−1
we e eached
o he senso s de eloped.
Keywo ds: Elec onic skin, piezo esis i i y, mic os uc u es, semi-sphe es, PDMS.
ix

Lis o Tables
1.1 S a e o he A o piezo esis i e p essu e senso s ................ 4
3.1
Memb anes peeled o om molds ab ica ed in s-PDMS using h ee di e en
designs ........................................ 11
3.2
Memb anes peeled o om molds ab ica ed in h-PDMS using lase eng a ing
in as e mode, powe o 50 W, and a speed o 0.762 m/s. ........... 13
3.3
Summa y o heigh measu emen s o semi-sphe es using di e en combina-
ions o designed base diame e s and lase powe s. ............... 14
3.4 SEM images acqui ed o mic os uc u ed PDMS ilms. ............. 15
3.5
SEM images o mic os uc u es p oduced on PDMS wi h ca bon o PMMA +
ca bon coa ings. ................................... 21
A.1
Semi-sphe es eal diame e and eal pi ch measu ed on ho izon al and e ical
di ec ions ....................................... 37
C.1
Measu emen s o dimensions o he ab ica ed mic os uc u es wi h ca bon
coa ing and wi h ca bon coa ing wi h PMMA. .................. 41
G.1 Es ima ed senso p ice ega ding ma e ials cos s ................ 51
x ii
Ac onyms
CNTs ca bon nano ubes.
h-PDMS ha d-poly(dime hylsiloxane).
LCD liquid c ys al display.
LOD limi o de ec ion.
PDMS poly(dime hylsiloxane).
PEDOT:PSS poly(3,4-e hylenedioxy hiophene) polys y ene sul ona e.
PEN polye hylene.
PI polyimide.
PMMA poly(me hyl me hac yla e).
PVDF poly( inylidene) di luo ide.
PZT lead zi cona e i ana e.
SEM scanning elec on mic oscope.
s-PDMS s anda d-poly(dime hylsiloxane).
xix
Mo i a ion and Objec i es
Nowadays, he pu sui o in o ma ion-sensing inspi ed by human skin has been mo i-
a ed by he possibili y o applica ion on unc ional heal h moni o ing sys ems and on
obo ic sys ems. The e o e, in o de o achie e his, hin ilm p essu e senso s a e being
widely exploi ed.
Fu he mo e, e o s ha e been made owa ds he ab ica ion o hin ilm senso s in
a non-clean oom en i onmen , which would conside ably lowe he de ice cos . Thus,
he main goal o his wo k is o design, ab ica e and de elop low-cos , lexible p essu e
senso s based on he piezo esis i e e ec , which could be easily adap ed and con o med
o di e en su aces, using mic os uc u ed subs a es.
La ge scale p oduc ion g ea ly bene i s om low manu ac u ing cos s in lexible p es-
su e senso s, and he simpli ica ion o he de ice’s s uc u e and manu ac u e is also
desi able. These piezo esis i i y-based de ices allow an easy eadou mechanism, as well
as a simple s uc u e design, which in u n will allow an easy de ice op imiza ion. Mo e-
o e , h ough he use o he pe ec combina ion o ma e ials one expec s o p oduce each
senso o app oxima ely 0.351 €(ma e ials cos s).
Finally, he ab ica ed ma e ials will be ex ensi ely cha ac e ized, bo h mo phologi-
cally and elec ically, as o compa e di e en designs, wi h he p ima y goal o de eloping
a senso wi h he bes sensi i y possible.
xxi

Chap e 1
In oduc ion
1.1 Elec onic Skin
Human skin is an ou s anding o gan, being ou in e ace wi h he su ounding wo ld and
allowing us o pe cei e mechanical s imuli such as p essu e, shapes and ex u es [1]. This
sense o in o ma ion is only achie able because human skin comp ises mechano ecep o s
ha ecei e a mechanical s imulus which is han ansduced in o a biological esponse [2].
Nowadays, inspi ed by his illuso y simplici y o na u e, e o s a e being made o de elop
skin-inspi ed elec onic de ices. The pu sui o hese e-skin de ices is mo i a ed by he
possibili y o applica ion on unc ional p os he ic de ices [3], humanoid obo ics [4] and
human heal h moni o ing, whe e i could play a key ole [5,6]. The e o e, accu a e
quan i a i e moni o ing equi es an e ec i e ansduc ion whe e ansduc ion mecha-
nisms such as piezo esis i i y [7–9], capaci ance [10–12], piezoelec ici y [13–15], and
iboelec ici y [16–18] a e being widely exploi ed o de elop di e en ypes o p essu e
senso s.
1.2 P essu e Senso Fundamen als
Fi s ly, o a be e comp ehension o p essu e dis ibu ions, p essu es such as human
ouch, objec manipula ion, and human body ci cula ion a e conside ed o be in he low
p essu e (<10 kPa) and medium-p essu e (10-100 kPa) egimes [12]. Secondly, and gi en
ha a p essu e senso ansduces a mechanical p essu e in o an elec ical signal an ou line
o some no ewo hy key pa ame e s ha include sensi i i y, limi o de ec ion (LOD),
linea i y, esponse ime and s abili y is p esen ed. Among hese, sensi i i y, which is
de ined by he a io be ween he a ia ion o he quan i a i e ou pu signal and a ia ion
o he applied p essu e, is one o he mos impo an pa ame e s because i de ines he
accu acy and e ec i eness o he measu emen [19]. Sensi i i y is de ined as
S= ∆R
R0!
∆P(1.1)
Whe e Pdeno es he applied p essu e and
∆R/R0
is he ela i e esis ance change o
he senso .
LOD ep esen s he lowes quan i y o p essu e ha can be dis inguishable. Lowe ing
he LOD o a p essu e senso ep esen s an imp o emen on lowe p essu e de ec ion
which is a equi emen in lowe p essu e egimes [20].
Ano he ele an pa ame e is linea i y, which is a classi ica ion exp essed as pe cen -
age o he de ia ion o he senso ’s ou pu cu e om a speci ied s aigh line o e a
1
CHAPTER 1. INTRODUCTION
ce ain p essu e ange [20]. Hence, one desi es o ha e p essu e senso s wi h b oad linea
anges, which simpli ies he con e sion o mechanical p essu e in o elec ical ou pu .
Response ime can be de ined as he ime equi ed o a p essu e senso ou pu o go
om i s p e ious s a e o a inal s able alue [20]. This pa ame e is especially impo an
in dynamic eal- ime p essu e sensing de ices, when p oducing eal- ime moni o ing
sys ems o ins an - esponse displays.
Fu he mo e, o accu a ely measu e he magni ude o a p essu e s imulus, he mos
ele an ansduc ion me hods a e piezoelec ici y, capaci ance, and piezo esis i i y .
Piezoelec ici y, which is he abili y (quan i ied by he piezoelec ic s ain cons an
d33
) o a ma e ial o gene a e elec ical cha ges in esponse o he occu ence o elec ical
dipole momen s due o applied mechanical s esses. This p ope y o c ys als and ce ain
ce amics is good o de elopmen o low-powe -consump ion o sel -powe ed sensing
de ices [21]. Mo eo e , he piezoelec ic elemen o he de ice can be combined wi h
ansis o s o imp o e sensi i i y [13]. The de elopmen o piezoelec ic p essu e senso s
has been ecei ing lo s o a en ion due o hei as esponse speed and sel -powe ed ope -
a ion. The piezoelec ic ma e ials mos used o his applica ion include poly( inylidene)
di luo ide (PVDF) and i s copolyme s, lead zi cona e i ana e (PZT) [13], and Zinc Oxide
(ZnO) [22].
Capaci i e senso ansduc ion consis s o a change in capaci ance caused by he
de lec ion o he pla e when an ex e nal s imulus is applied. As dielec ic cons an (pe -
mi i i y) o he medium be ween he pla es is a cons an , he ex e nal s imulus, ei he
an applied p essu e o shea o ce, usually ep esen s a change in a ea o in he dis ance
be ween he pla es, espec i ely [1].
Capaci i e senso s o e ad an ages such as high sensi i i y, bu he main ad an age
o his ypes o senso s is he simplici y o hei go e ning equa ion, which allows a
di ec analysis and a simple de ice design [23]. These senso s ha e also demons a ed
he cha ac e is ic o high s ain sensi i i y o de ec ion o a s a ic o ce wi h low-powe
consump ion [24]. Howe e , once capaci ance is p opo ional o he a ea, a educ ion in
he size o hese de ices o minia u iza ion means a educ ion o he capaci ance and
he signal- o-noise a io [25]. Addi ionally, capaci i e senso s a e ulne able o ex e nal
in e e ences [26].
The o he ansduc ion me hod commonly used is based on piezo esis i e e ec . Piezo e-
sis i e senso s ha e been widely in es iga ed due o hei simple s uc u e and eadou
mechanism [27]. These senso s ansduce a o ce a ia ion in o changes in esis ance o
a de ice ha is de ec ed by an elec ical measu ing sys em. The esis ance a ia ion de-
i es mos commonly om changes in: he geome y o he sensing elemen ; he con ac
esis ance (
RC
) and he esis i i y o a composi e due o changes in sepa a ion be ween
pa icles. Usually, o hese ype o senso s, when he esis i i y o he ma e ial is cons an ,
a change in esis ance de i es om changes in he geome y o he sensing elemen .
Fo conduc i e ma e ials, he mechanism elies on changes in he
RC
be ween bo h
ma e ials. A
RC
change caused by change in con ac a ea be ween wo conduc o s is
2
1.3. PIEZORESISTIVE PRESSURE SENSOR DEVICES
p opo ional o he squa e oo o he o ce, which is an ad an age as i p o ides high sen-
si i i y a lowe p essu es as well as i expands he usable ange [1]. Fo conduc i e elas ic
composi es, piezo esis ance depends on he mo phology, composi ion, and s ain ange
o he sys em, and he mechanism elies on changes in conduc i e pa h. Fu he mo e,
piezo esis i e senso s exhibi a as esponse speed [1]. Howe e , piezo esis i i y-based
senso s usually show undesi able d i and hys e esis [28].
As seen, he p esen ed ansduc ion me hods p o ide di e en sensing capabili ies
which allows sys ems o in eg a e mul i unc ional senso s. Fo example, piezo esis i e
a e usually used o eliably measu e la ge s ains [29], capaci i e de ices o sense no mal
o ces and piezoelec ic de ices o measu e ib a ions [1].
1.3 Piezo esis i e p essu e senso de ices
Being human skin conside ed as a pe o mance benchma k o he de elopmen o e-skin,
some conside a ions need o be ollowed o g an his elec ical ma e ial he mechanical
p ope ies o human skin, such as s e chabili y, lexibili y, and low Young’s modulus [1].
To ul il hese conside a ions, he design and ab ica ion o he de ice a e c i ical. An
impo an pa ame e is ega ding de ice’s s e chabili y and he e a e wo main s a egies
used o imp o e i [30]. The i s me hod (Figu e 1.1a) uses a hin conduc i e ma e ial
bonded o an elas ic subs a e, such as poly(dime hylsiloxane) (PDMS) [10]. The second
me hod (Figu e 1.1b) is based on he ab ica ion o de ices by mixing conduc i e ma e ials
in o an elas ome ic ma ix [31].
F
F
(a)
F
F
(b)
Figu e 1.1: (a) Thin conduc i e ma e ial bounded o a mic o-s uc u ed elas ome . (b)
Mix u e o a conduc i e ma e ial on an elas ome ic ma ix.
Ano he impo an pa ame e in p essu e senso s design is he con ac a ea, which
can be inc eased by cons uc ing a ious de ice geome ies, o example h ough he
ma e ials mic os uc u a ion. C ea ing he desi ed geome y equi es echniques such as
li hog aphy [32], coa ing [33], and mic o-channel molding and illing [34]. This concep
has shown o be an ideal candida e o e-skin applica ions [8,12]. Fo his eason, and o
be able o mee he needs s a ed abo e, he ma e ials choice is c ucial o he de elopmen
o lexible p essu e senso s and he ein, an ou line o subs a e and ac i e ma e ials o
piezo esis i e p essu e senso s is p esen ed. Table 1.1 summa izes piezo esis i e p es-
su e senso s de eloped o e he ecen yea s as well as hei pe o mance pa ame e s, o
3
CHAPTER 2. MATERIALS AND METHODS
coa ing laye s was measu ed in a p o ilome e (Ambios XP-Plus 200 S ylus) o one o i e
s acked laye deposi ions, using a acking o ce o 0.5 mg and a scanning speed o 0.20
mm/sec. I-V cu es we e acqui ed using Kei hley 2000 Mul ime e connec ed o each
sil e -ink elec ode o he memb ane and a ol age sweep om -2 V o 2 V, in s eps o 0.5
V, was applied. The ou pu signal co esponds o cu en lowing h ough he 2 cm x 2
cm mic os uc u ed domain.
2.6 Elec ical cha ac e iza ion o de ices
Fo quan i a i e analyses, ou pu signals om a mechanical s imulus we e acqui ed by
eco ding changes in elec ical esis ance as a unc ion o applied p essu e using a home-
made sys em de eloped on he scope o his wo k (as desc ibed in Sec ion 3.3), capable o
applying di e en p essu e alues. Addi ionally, I-V cu es o he de ices we e acqui ed
using he same me hod as o memb anes, o es o ohmic-like beha iou .
10

Chap e 3
Resul s and Discussion
3.1 Pa e ning
The mic os u u ed domain shape op imiza ion o each semi-sphe e-like s uc u es wen
h ough se e al s eps such as exploi a ion in PDMS molds ab ica ion, lase eng a ing
pa ame e s, and shape design in so wa e.
Table 3.1: Memb anes peeled o om molds ab ica ed in s anda d-PDMS using a speed
o 0.254 m/s and lase powe o 12.5 W o images (a)(c)(e) o lase powe o 25 W o
images (b)(d)( ), all wi h lase eng a ing in ec o mode and o he h ee di e en de-
signs p e iously men ioned. (a) and (b) Mic os uc u es esul an om aligned ci cles
wi h wi h bo h a diame e and a ci cles dis ance o 200
µ
m. (c) and (d) Mic os uc u es
esul an om aligned squa es o 200
µ
m x 200
µ
m wi h a dis ance be ween squa es o
200
µ
m. (e) and ( ) Mic os uc u es esul an om aliegned squa es o 200
µ
m x 200
µ
m
wi h wo diagonals and a dis ance be ween squa es o 200
µ
m. Abb e ia ions used in his
able: Powe (P), so -Polydime hylsiloxane (s-PDMS), Speed (S).
Vec o Mode
s-PDMS P = 12.5 W, S = 0.254 m/s P = 25 W, S = 0.254 m/s
(a) (b)
200
µm
200 µm
200 µm
(c) (d)
200
µm
200 µm
200 µm
(e) ( )
200
µm
200 µm
200 µm
11
CHAPTER 3. RESULTS AND DISCUSSION
Fi s ly, es s o e shape design in so wa e we e pe o med o choose he pe ec
design o achie e he desi ed o m on lase eng a ing. To do so, h ee di e en shapes
we e s udied: ci cles, squa es, and squa es wi h wo diagonals. The ci cles had a diame e
o 200
µ
m, he squa es we e 200
µ
m x 200
µ
m, and he pi ch be ween each ea u e was
ixed a 200
µ
m. These designs we e eng a ed in ec o mode in s-PDMS ilms in wo
di e en ba ches – he i s wi h a lase powe o 12.5 W and a speed o 0.254 m/s and he
second wi h a lase powe and speed o 25 W and 0.254 m/s, espec i ely. Table 3.1 shows
il ed images (45
º
) acqui ed om he memb anes peeled o om molds. Molds eng a ed
wi h ci cles clea ly show a much mo e semi-sphe e like s uc u e, whe eas he o he s s ay
mo e ai h ul o hei squa ed designs. All six examples show some i egula i ies in he
mic os uc u es due o he di icul in peeling o he memb anes om he mold made o
he same ma e ial. Addi ionally, gi en ha molds eng a ing is done in a ma e ial ha
mel s e y easily, he eng a ed ca i ies do no mel in a homogeneous way, and so he
PDMS memb anes ha a e peeled o om hese molds ge he nega i e pa e n o hose
i egula i ies.
Memb anes peeled o om he mold eng a ed wi h highe powe p esen ea u es
ha a e mo e i egula , possibly due o he o e -mel ing o PDMS du ing lase eng a ing.
The e o e, molds eng a ed wi h a powe o 12.5 W and a speed o 0.254 m/s appea o
be p e e ed candida es o ideal semi-sphe e molds.
200 µm
(a)
200 µm
(b)
Figu e 3.1: SEM images acqui ed om a op iew o peeled memb anes om molds made
in s-PDMS in ec o mode using a lase speed o 0.254 m/s and a lase powe o (a) 12.5
W and (b) 25 W.
A op iew o he semi-sphe e like ea u es om he achie ed memb anes was also
cap u ed, as shown on Figu e 3.1. Such igu e con i ms a good ideli y o he design
in e ms o shape, bu highligh s di e ences in e ms o pi ch o e he ho izon al and
e ical eng a ing di ec ions. This disc epancy esul s om he ac ha he lase has
a be e esolu ion on he e ical di ec ion han on he ho izon al di ec ion as u he
examined in his sec ion. Figu e 3.1 also shows ha he pi ch (200
µ
m) is close o he lase
beam esolu ion (127
µ
m) as semi-sphe es in ho izon al di ec ion a e almos ouching
hei ho izon al neighbou semi-sphe es. This ep esen s a limi in design’s pi ch ( o
hese anges o lase powe ), as a pi ch lowe han he one es ed (200 um) would lead o
12
3.1. PATTERNING
an o e -mel ing o ca i ies on PDMS mold, esul ing in dis inc ea u es han he ones
designed.
Meanwhile, es s on h-PDMS we e also pe o med o e alua e he combina ion o
as e mode eng a ing on molds made o h-PDMS. The designs es ed o his eng a ing
mode had o be changed because as e mode eng a es he ma e ial in a di e en way
when compa ed o ec o mode. Also, one expec ed ha a design based on ci cles would
gi e ise o ca i ies wi h he shape o in e ed cones ins ead o in e ed semi-sphe es.
The e o e, he pa e ns designed o his s udy we e squa es o 200
µ
m x 200
µ
m wi h a
pi ch o 300
µ
m o 1000
µ
m. Lase eng a ing pa ame e s we e ixed a a speed o 0.762
m/s and powe o 50 W. F om pa e ns shown on Table 3.2 one can obse e he o ma ion
o 3D s uc u es a om being semi-sphe es. Fo ins ance, in he image o he aligned
pa e n wi h he pi ch = 300
µ
m, s aigh lines eng a ed by he lase beam a e pe ec ly
dis inguishable, which con i ms ha as e mode is no sui able o egula semi-sphe e
mic os uc u ing.
Table 3.2: Memb anes peeled o om molds ab ica ed in h-PDMS using lase eng a ing
in as e mode, powe o 50 W, and a speed o 0.762 m/s. (a) Mic os uc u es esul an
om aliegned squa es o 200
µ
m x 200
µ
m wi h a dis ance be ween squa es o 300
µ
m.
(b) Mic os uc u es esul an om aliegned squa es o 200
µ
m x 200
µ
m wi h a dis ance
be ween squa es o 1000
µ
m. (c) Mic os uc u es esul an om misaligned squa es
o 200
µ
m x 200
µ
m wi h a dis ance be ween squa es o 300
µ
m. (d) Mic os uc u es
esul an om misaligned squa es o 200
µ
m x 200
µ
m wi h a dis ance be ween squa es
o 1000 µm.
Ras e Mode
h-PDMS Pa e n Pi ch = 300 µm Pi ch = 1000 µm
(a) (b)
Aligned
200 µm
200 µm
400 µm
400 µm
(c) (d)
Misaligned
200 µm
200 µm
400 µm
400 µm
Once he design and he lase eng a ing mode o mic os uc u ing semi-sphe es we e
chosen, he op imiza ion p ocess demanded a choice o he mold ma e ial. As seen be o e,
13
CHAPTER 3. RESULTS AND DISCUSSION
memb anes peeled o om s-PDMS molds we e usually di icul o peel and showed
i egula i ies on hei s uc u e. The e o e, expe imen s in ec o mode on h-PDMS
we e also pe o med o app aise his combina ion comp ising ci cle pa e ns. He ein, o
in es iga e he e ec o lase powe on eng a ing mic oca i ies, molds we e eng a ed
using lase powe o 25 W, 12.5 W, 7.5 W, and 2.5 W whe eas he speed was ixed a 0.254
m/s o all pa e ns. The eason o main ain lase speed a a high alue is due o he ac
ha he highe he lase speed, he sho e is he wo king ime o lase on PDMS and,
consequen ly, he less high and sha p he s uc u es will be. Table 3.3 and Figu e 3.2
p esen esul s om molds ab ica ed using lase e o mode on h-PDMS wi h ci cles,
whe e one obse es ha he highe he lase powe , he highe he mic os uc u es a e.
In his speci ic s udy, a 45 % inc ease in lase powe gi es ise o s uc u es wi h a heigh
i e imes bigge .
Table 3.3: Summa y o heigh measu emen s o semi-sphe es using di e en combina-
ions o designed base diame e s and lase powe s.
200 µm
Heigh (µm) 60 80 112 151 180 206 232 320
Diame e (µm) 100 200 100 200 100 200 100 200
Lase Powe (W) 2.5 2.5 7.5 7.5 12.5 12.5 25 25
0 5 1 0 1 5 2 0 2 5
0
5 0
100
150
200
250
300
350
400
D i a m e e = 1 0 0 µm
D i a m e e = 2 0 0 µm
H e i g h (mm )
L a s e P o w e ( W )
Figu e 3.2: Semi-sphe es heigh measu ed om SEM images e sus lase powe used o
eng a e he molds om whe e he esul an s uc u es we e peeled o . Values p esen ed
co espond o a e age alues ± s anda d de ia ion o a minimum o 15 measu emen s.
A ela ion be ween diame e and heigh is also obse able, once he highe he de-
signed diame e , he highe he semi-sphe e heigh is. This esul s om he ac ha
eng a ing la ge a eas comp ises deepe eng a ing on molds. Rega ding hese esul s,
lase powe o 2.5 W and 7.5 W showed up as being he mos likely o p oduce semi-
sphe e like s uc u es.
14
3.1. PATTERNING
Fu he mo e, mic os uc u es esul an om wo simila designs we e used o compa e
he heo e ical diame e and pi ch wi h he esul an ones, as well as o compa e he
esolu ion o e e ical and ho izon al di ec ions. The i s design is based on ci cles wi h
a heo e ical diame e o 200
µ
m and a pi ch o 150
µ
m, while he second design is based
on ci cles wi h a heo e ical diame e o 200
µ
m and a pi ch o 200
µ
m. To do so, molds
we e made using a lase powe o 2.5 W and 7.5 W, main aining speed a 0.254 m/s o
bo h. Table 3.4 shows he gene al iew o he esul an PDMS s uc u es exhibi ing nice
homogenei y o each pa e n, while highligh ing he di e ences be ween pi ches and
diame e s o e ho izon al and e ical di ec ions. F om he images one no ices ha each
ea u e can easily be disce ned e en o he lowe pi ch, meaning ha heo e ical pi ch o
abou 150 µm is achie able wi h his echnique.
Table 3.4: SEM images acqui ed o mic os uc u ed PDMS ilms. Images (a)-(h) esul
om molds eng a ed wi h a lase powe o 2.5 W and 0.254 m/s lase speed o di e en
pi ches and diame e s. Images (i)-(p) esul om molds eng a ed wi h a lase powe o
7.5 W and 0.254 m/s lase speed o di e en pi ches and diame e s.
Ho izon al Pi ch Ve ical Pi ch
150 µm 200 µm 150 µm 200 µm
P = 2.5 W, S = 0.254 m/s
Diame e
(a) (b) (c) (d)
100 µm
200 µm
200 µm
200 µm
200 µm
(e) ( ) (g) (h)
200 µm
200 µm
200 µm
200 µm
200 µm
P = 7.5 W, S = 0.254 m/s
Diame e
(i) (j) (k) (l)
100 µm
200 µm
200 µm
200 µm
200 µm
(m) (n) (o) (p)
200 µm
200 µm
200 µm
200 µm
200 µm
As p e iously seen, measu emen s o e bo h ho izon al and e ical di ec ions should
be pe o med in o de o ho oughly analyse ea u es. Hence, all semi-sphe es we e mea-
su ed o e bo h di ec ions o he eng a ing p ocess. Table A.1 summa izes he measu ed
15

CHAPTER 3. RESULTS AND DISCUSSION
diame e and pi ch o he ab ica ed ea u es, showing ha al hough diame e and pi ch
alues we e usually a om wha was designed, he sum o diame e and pi ch ends o
be e y close o he expec ed sum o he designed ones. This can be explained by he ac
ha lase beam mel s mo e PDMS han i was designed, pe o ming eng a ings much
la ge han expec ed, which is co obo a ed by he disc epancy, some imes la ge han
100
µ
m, on measu ed diame e s. The e o e, pi ch o i s u n, shows o be much smalle
100 150 200 250 300 350
100
150
200
250
300
350
R e a l D i a m e e (mm )
D e s i g n e d D i a m e e ( mm )
P i c h = 1 5 0 mm
H o i z o n a l
V e i c a l
P i c h = 2 0 0 mm
H o i z o n a l
V e i c a l
y = x
2 . 5 W L a s e P o w e
(a)
100 150 200 250 300 350 400
100
150
200
250
300
350
400
R e a l D i a m e e (mm )
D e s i g n e d D i a m e e ( mm )
P i c h = 1 5 0 mm
H o i z o n a l
V e i c a l
P i c h = 2 0 0 mm
H o i z o n a l
V e i c a l
y = x
7 . 5 W L a s e P o w e
(b)
Figu e 3.3: Real diame e o PDMS semi-sphe es, measu ed o e ho izon al and e ical
di ec ions o lase eng a ing e sus designed diame e in Adobe Illus a o , p oduced
wi h a lase powe o (a) 2.5 W o (b) 7.5 W. The lines y = x illus a e a eal diame e equal
o he one designed. Values p esen ed co espond o a e age alues
±
s anda d de ia ion
o a minimum o 15 measu emen s.
0 50 100 150 200 250
0
5 0
100
150
200
250
R e a l P i c h (mm )
D e s i g n e d P i c h ( mm )
D i a m e e = 1 0 0 mm
H o i z o n a l
V e i c a l
D i a m e e = 2 0 0 mm
H o i z o n a l
V e i c a l
y = x
2 . 5 W L a s e P o w e
(a)
0 50 100 150 200 250
0
5 0
100
150
200
250
D i a m e e = 1 0 0 mm
H o i z o n a l
V e i c a l
D i a m e e = 2 0 0 mm
H o i z o n a l
V e i c a l
y = x
R e a l P i c h (mm )
D e s i g n e d P i c h ( mm )
7 . 5 W L a s e P o w e
(b)
Figu e 3.4: Real pi ch be ween PDMS semi-sphe es, measu ed o e ho izon al and e ical
di ec ions o lase eng a ing e sus designed pi ch in Adobe Illus a o , p oduced wi h a
lase powe o (a) 2.5 W o (b) 7.5 W. The lines y = x illus a e a eal pi ch equal o he
one designed. Values p esen ed co espond o a e age alues
±
s anda d de ia ion o a
minimum o 15 measu emen s.
16
3.1. PATTERNING
han expec ed o main ain he design. Lase eng a ing esolu ion in e ical di ec ion is
no ably be e han in ho izon al di ec ion once ea u es measu ed in e ical di ec ion
o he lase eng a ing show alues o diame e and pi ch close o he designed ones, as
i is shown on Figu e 3.3 and Figu e 3.4, espec i ely.
As a esul o all he explained be o e, he molds chosen o ab ica ion o PDMS
memb anes come as a combina ion o using designs wi h ci cles o be eng a ed on h-
PDMS wi h a lase eng a ing machine on ec o mode. The pa ame e s chosen we e 7.5
W o lase powe and speed o 0.254 m/s, as wi h his combina ion mo e ep oducible
semi-sphe e s uc u es a e achi able. Two molds using his combina ion o pa ame e s
we e p oduced, bo h wi h heo e ical diame e s o 200
µ
m – he i s wi h a pi ch o 150
µ
m and he second wi h a pi ch o 200
µ
m. Figu e 3.5 shows he wo molds p oduced
and Figu e 3.6 shows a mic oscope image o each PDMS mold ha was used o ab ica e
memb anes p esen in he s udies u he explained in his wo k, which we e measu ed o
ha e a hickness o 215
µ
m
±
19
µ
m (a e age alues
±
s anda d de ia ion) co esponding
o a minimum o 15 measu emen s.
Figu e 3.5: Molds p oduced o ab ica ion o memb anes. Mold on he le designed wi h
ci cles wi h 200
µ
m o diame e and pi ch = 150
µ
m. Mold on he igh designed wi h
ci cles wi h 200 µm o diame e and pi ch = 200 µm
200 µm
(a)
200 µm
(b)
Figu e 3.6: Mic oscope acqui ed images o he ab ica ed molds. (a) Mold designed wi h
diame e = 200
µ
m and pi ch = 150
µ
m. (b) Mold designed wi h diame e = 200
µ
m and
pi ch = 200 µm.
17
CHAPTER 3. RESULTS AND DISCUSSION
3.2 Ink S udy
3.2.1 Ca bon-ink dilu ion s udy
S udies o e ca bon-ink dilu ions and numbe o laye s we e pe o med o u he explo e
he bes conduc i e ilm o ab ica e piezo esis i e senso s. He ein, PMMA was in es i-
ga ed as a coa ing laye be ween ca bon-ink and PDMS o check i i could bo h imp o e
he adhesion o ca bon ink o PDMS and he inal s abili y o he de ice. Fi s ly, o in es i-
ga e he possibili y o using less ca bon-ink wi hou comp omising he good unc ionali y
o he piezo esis i e senso s, shee esis ance o smoo h ca bon-coa ed PDMS and smoo h
ca bon-coa ed PDMS wi h PMMA was measu ed by a ying he dilu ion o he coa ing
in wa e (100 w %, 67 w %, 50 w %, 40 w %). Figu e 3.7 shows a compa ison o shee
esis ance be ween ca bon-ink deposi ed on PDMS wi h and wi hou PMMA. Bo h cu es
end o dec ease in shee esis ance as ca bon-ink concen a ion in wa e inc eases, as
expec ed. F om his s udy i was ound ha shee esis ance o 50 w % and 40 w %
dilu ions, o bo h PDMS wi h and wi hou PMMA, is excessi ely high o a conduc i e
ilm, wi h a e age esis ance alues abo e 30 k
Ω/
. Fo ilms wi h app oxima ely 67 w
40 50 60 70 80 90 100
0
5 0
100 P D M S w i h P M M A
P D M S w i h o u P M M A
S h e e R e s i s a n c e ( k W / ÿ )
C a b o n - i n k i n w a e ( w % )
Figu e 3.7: Shee esis ance o smoo h ca bon-coa ed PMDS and PMMA-coa ed PDMS o
ou di e en ca bon-ink dilu ions in wa e . Measu emen s p esen ed he e co espond o
a e age alues ± a e age absolu e de ia ion o a minimum o 3 measu emen s.
% dilu ion, he shee esis ance is (6.7
±
1.2) k
Ω/
and (6.4
±
0.8) k
Ω/
, espec i ely o
smoo h ca bon-coa ed PDMS and smoo h ca bon-coa ed PDMS wi h PMMA.
Using a concen a ion o 100 w % o ca bon-ink ensu es a shee esis ance o abou
(1.3
±
0.1) k
Ω/
and (1.6
±
0.3) k
Ω/
, espec i ely o PDMS wi h and wi hou PMMA,
which is sui able o he ab ica ion o piezo esis i e p essu e senso s. Figu e 3.7 illus-
a es ha 67 w % and 100 w % dilu ions allow an easy eplica ion o he memb anes
p oduced as he shee esis ance a e age alues a e de ia ed by a small e o .
18
3.2. INK STUDY
3.2.2 Ink laye s s udy
To u he unde s anding he ela ion be ween ca bon coa ing laye s and hei beha iou
on he de ice, a s udy compa ing shee esis ance o di e en deposi ed laye s using he
dilu ions o 67 w % and 100 w % was accomplished. Figu e 3.8a shows a compa ison
o shee esis ance as a unc ion o he numbe o laye s o PDMS and PMMA-coa ed
PDMS.
Rega ding he 67 w % dilu ion o ca bon-ink on PDMS wi h PMMA, shee esis ance
o one laye is app oxima ely (6.4
±
0.8) k
Ω/
. Shee esis ance, as expec ed, dec eases
wi h he numbe o ca bon laye s deposi ed by spin-coa ing as he amoun o conduc i e
coa ing inc eases. The de ia ion e o also dec eases which means ha he conduc ion
mechanism is becoming homogenous o he whole memb ane. Bo h dilu ions end o a
limi o shee esis ance wi h he inc easing laye s. Ne e heless, as expec ed, he 67 w %
dilu ion equi es mo e s acked laye s o each he same alues o shee esis ance o 100
w %. The e o e, o 100 w % dilu ion, he e was no need o pe o m s udies o e mo e
laye s, once he alues we e al eady low enough. The shee esis ance on PDMS alues
a e close o he ones wi h he same dilu ion on PMMA o bo h dilu ions s udied he e,
howe e , i appea s o ha e highe de ia ion e o , which means PMMA on PDMS seems
o con ibu e o a g ea e s abili y o ca bon coa ing adhesion.
The i s ca bon coa ing laye o all ou cu es shows a la ge de ia ion e o o
shee esis ance han o all o he laye deposi ions due o he he e ogenei y o deposi ed
ca bon. This obse a ion is co obo a ed by he de ia ion e o s o hickness, which
we e also measu ed in his s udy.
12345
0
2
4
6
8
N u m b e o c a b o n c o a i n g l a y e s
6 7 w % w / P M M A
6 7 w %
1 0 0 w % w / P M M A
1 0 0 w %
S h e e R e s i s a n c e ( k W/ÿ)
(a)
12345
0
4
8
1 2
1 6
2 0
2 4
2 8
T h i c k n e s s ( mm )
N u m b e o c a b o n c o a i n g l a y e s
6 7 w % w / P M M A
6 7 w %
1 0 0 w % w / P M M A
1 0 0 w %
(b)
Figu e 3.8: (a) Shee esis ance o 67 w % and 100 w % ca bon coa ing dilu ions in wa e
on smoo h PDMS ilms and PMMA-coa ed PDMS ilms wi h 1 o 5 laye s o coa ing. (b)
Thickness o 67 w % and 100 w % ca bon coa ing on smoo h PDMS ilms and PMMA-
coa ed PDMS ilms wi h 1 o 5 laye s o coa ing. Measu emen s p esen ed he e co espond
o a e age alues ± a e age absolu e de ia ion o a minimum o 3 measu emen s.
Thicknesses p esen ed on Figu e 3.8b we e measu ed be ween he op o he coa ing
19
CHAPTER 3. RESULTS AND DISCUSSION
- 2 - 1 0 1 2
- 2
- 1
0
1
2
V o l a g e ( V )
C u e n ( m A )
P M M A - c o a e d P D M S m i c o s u c u e d m e m b a n e s
A
B
(a)
- 2 - 1 0 1 2
- 0 . 4
- 0 . 2
0 . 0
0 . 2
0 . 4
V o l a g e ( V )
C u e n ( m A )
P D M S m i c o s u c u e d m e m b a n e s
A
B
(b)
- 2 - 1 0 1 2
- 2
- 1
0
1
2
V o l a g e ( V )
C u e n ( m A )
P M M A - c o a e d P D M S m i c o s u c u e d m e m b a n e s
A
B
(c)
- 2 - 1 0 1 2
- 0 . 2
0 . 0
0 . 2
C u e n ( m A )
P D M S m i c o s u c u e d m e m b a n e s
A
B
V o l a g e ( V )
(d)
Figu e 3.13: Vol age sweep om -2 o 2 V p o ing he ohmic beha iou o wo de ices
p oduced wi h (a) PMMA-coa ed PDMS memb ane wi h semi-sphe es wi h a pi ch o 150
µ
m. (b) PDMS memb ane wi h semi-sphe es wi h a pi ch o 150
µ
m. (c) PMMA-coa ed
PDMS memb ane wi h semi-sphe es wi h a pi ch o 200
µ
m. (d) PDMS memb ane wi h
semi-sphe es wi h a pi ch o 200 µm.
Fu he mo e, o be e unde s and he signal o e ime in esponse o applied p es-
su e, ou pu signal was acqui ed as unc ion o a p essu e ange, a he equency o 1
Hz. Figu e 3.14 shows he ou pu signal o each ype o ab ica ed senso . He ein, one
obse es he good ou pu signal ep oducibili y o e cycles, con i ming he good s abil-
i y o he de ice, and p o ing ha he design o he de ice is sui able o his ype o
applica ions. As seen in Figu e 3.14c, senso s ab ica ed wi h PMMA coa ing on PDMS
wi h a pi ch o 200
µ
m p esen a highe ou pu signal o he same p essu e han o he
ab ica ed senso s. Bo h cu es on each g aph p esen a phase di e ence o abou
0.01°
o
0.8°
which may be in oduced by he assembled sys em composed by he elec onic com-
ponen s and he mo o , e en so, his does no seem o cause in e e ence o he de ice’s
26

3.4. ELECTRICAL CHARACTERIZATION OF DEVICES
pe o mance.
01234
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
1 . 0
DR/R0( % )
T i m e ( s )
P i e z o e s i s i e s e n s o s i g n a l
0
5 0
100
150
200
250
300
350
P e s s u e ( P a )
P i e z o e l e c ic s e n s o s i g n a l
(a)
01234
0 . 0
0 . 5
1 . 0
1 . 5
2 . 0
T i m e ( s )
DR/R0( % )
P i e z o e s i s i e s e n s o s i g n a l
0
5 0
100
150
200
250
300
P e s s u e ( P a )
P i e z o e l e c ic s e n s o s i g n a l
(b)
01234
0 . 0
0 . 5
1 . 0
1 . 5
2 . 0
2 . 5
DR/R0( % )
T i m e ( s )
P i e z o e s i s i e s e n s o s i g n a l
0
5 0
100
150
200
250
300
350
P i e z o e l e c ic s e n s o s i g n a l
P e s s u e ( P a )
(c)
01234
0 . 0
0 . 5
1 . 0
1 . 5
DR/R0( % )
T i m e ( s )
P i e z o e s i s i e s e n s o s i g n a l
0
5 0
100
150
200
250
300
P e s s u e ( P a )
P i e z o e l e c ic s e n s o s i g n a l
(d)
Figu e 3.14: Resis ance changes in esponse o applied p essu e o e ime (a) PMMA-
coa ed PDMS memb ane wi h semi-sphe es wi h a pi ch o 150
µ
m. (b) PDMS memb ane
wi h semi-sphe es wi h a pi ch o 150
µ
m. (c) PMMA-coa ed PDMS memb ane wi h semi-
sphe es wi h a pi ch o 200
µ
m. (d) PDMS memb ane wi h semi-sphe es wi h a pi ch o
200 µm.
Mo eo e , Figu e 3.15 shows he app oxima ely linea ela ion be ween
∆R/R0
and
he applied p essu e ( ega ding he p inciple illus a ed in Figu e E.1, Sec ion E), o he
wo designs on memb anes wi h and wi hou PMMA, om whe e one may calcula e a
sensi i i y.
Sensi i i ies o S = 4.9
×10−2kPa−1
o ex e nal p essu es anging om 9 Pa o 330
Pa o PMMA-coa ed PDMS con aining semi-sphe es wi h a pi ch o 150
µ
m, S = 1.3
×10−1kPa−1
o ex e nal p essu es anging om 5 Pa o 360 Pa o PDMS con aining
semi-sphe es wi h a pi ch o 150
µ
m, S = 2.4
×10−1kPa−1
o ex e nal p essu es anging
om 4 Pa o 150 Pa o PMMA-coa ed PDMS con aining semi-sphe es wi h a pi ch o
200
µ
m, and S = 9.2
×10−2kPa−1
o ex e nal p essu es anging om 10 Pa o 340 Pa
o PDMS con aining semi-sphe es wi h a pi ch o 200
µ
m, we e acqui ed. In Sec ion Fa
signal esponse o he app oxima ely same applied p essu e, a bo h a equency o 1 Hz
and 15 Hz a e shown, p o ing ha he ab ica ed senso s main ain hei pe o mance o
his ange o equencies as he ela i e esis ance change s ays iden ical.
27
CHAPTER 3. RESULTS AND DISCUSSION
0 . 0 E + 0 0 1 . 0 E - 0 1 2 . 0 E - 0 1 3 . 0 E - 0 1
0 %
1 %
2 %
DR/R0
y = 4 . 9 E - 2 x
P e s s u e ( k P a )
P M M A - c o a e d P D M S , P i c h = 1 5 0 mm
(a)
0 . 0 E + 0 0 1 . 0 E - 0 1 2 . 0 E - 0 1 3 . 0 E - 0 1 4 . 0 E - 0 1
0 %
1 %
2 %
3 %
4 %
5 %
P e s s u e ( k P a )
P D M S , P i c h = 1 5 0 mm
y = 1 . 3 E - 1 x
DR/R0
(b)
0 . 0 E + 0 0 5 . 0 E - 0 2 1 . 0 E - 0 1 1 . 5 E - 0 1
0 %
1 %
2 %
3 %
4 %
5 % y = 2 . 4 E - 1 x
DR/R0
P e s s u e ( k P a )
P M M A - c o a e d P D M S , P i c h = 2 0 0 mm
(c)
0 . 0 E + 0 0 1 . 0 E - 0 1 2 . 0 E - 0 1 3 . 0 E - 0 1
0 %
1 %
2 %
3 %
DR/R0
P D M S , P i c h = 2 0 0 mm
y = 9 . 2 E - 2 x
P e s s u e ( k P a )
(d)
Figu e 3.15: Resis ance esponse o di e en p essu es. Do line is a linea eg ession
om which he sensi i i y o each de ice is ex ac ed. (a) PMMA-coa ed PDMS memb ane
wi h semi-sphe es wi h a pi ch o 150
µ
m. (b) PDMS memb ane wi h semi-sphe es wi h
a pi ch o 150
µ
m. (c) PMMA-coa ed PDMS memb ane wi h semi-sphe es wi h a pi ch o
200 µm. (d) PDMS memb ane wi h semi-sphe es wi h a pi ch o 200 µm.
28
Chap e 4
Conclusions and u u e pe spec i es
In he p esen wo k, he majo objec i e was o de elop a simple and low-cos me hod
o ab ica ion o ca bon-coa ed PDMS wi h uni o mly mic os uc u ed pa e ns. To
accomplish so, a hin PDMS ilm was mic os uc u ed wi h semi-sphe es, and pos e io ly
co e ed by a lexible ca bon coa ing.
As p e iously seen, he combina ion o lase eng a ing pa ame e s, he design o mi-
c os uc u es, and he ype o mold showed o be o signi ican impo ance in o de o
ob ain he desi ed mic os uc u es on he ab ica ed PDMS memb anes. One majo con-
clusion o his wo k is ha he ideal combina ion o ob ain semi-sphe es is o use designs
wi h ci cles eng a ed on h-PDMS molds wi h lase eng a ing machine on ec o mode.
Al hough he eal alues o pi ch and diame e do no main ain ideli y o he designed
alues, his me hod allows a ab ica ion o con olled and ep oducible semi-sphe e-like
s uc u es. Semi-sphe e dimensions such as pi ch size, heigh and diame e ha e a se-
ious e ec on he pe o mance o he p essu e senso s as he mic os uc u es play he
mos impo an ole on he a ia ion o con ac a ea upon p essu e, a ec ing he p essu e
senso s sensi i i y. O he conclusion o his wo k is ha he lase eng a ing machine has
be e esolu ion o e e ical di ec ion o lase eng a ing han o e ho izon al di ec ion,
as eal alues o semi-sphe e’ pi ch and diame e end o be much close o he alues
designed in so wa e, when measu ed on he e ical di ec ion. Rega ding s udies wi h
he ac i e laye ma e ial, ca bon coa ing, he esul s showed ha using one single coa ing
laye o 100 w % ca bon coa ing dilu ion would be p e e ed o e using a ious laye s o
lowe dilu ions, as i equi es less wo k and ime consump ion and ensu es lowe shee
esis ances and hickness.
A homemade p essu e applying sys em was de eloped in o de o elec ically cha ac-
e ize he ab ica ed de ices. This sys em can p o ide p essu es in equencies om 0.5
Hz o 20 Hz, and he ange o he applied p essu e can be easily adjus ed depending on
wha is he inal objec applying he o ce. Fo example, an objec wi h la ge a ea will
apply lowe p essu e. Fu he imp o emen s on his sys em could be made by changing
he 3D p in ed componen s o ca bon ibe ones and changing he ubbe ip o a so e
ma e ial wi h la ge a ea.
In wha conce ns he p essu e senso s, de ices comp ising a PMMA laye be ween
PDMS and ca bon coa ing we e ab ica ed in o de o in es iga e PMMA adhesion p op-
e ies. In e ms o sensi i i y, he senso wi h highe sensi i i y was he one ab ica ed
wi h a la ge pi ch and wi h a PMMA laye bu a he same ime, he lowe sensi i i y
was ound in he senso wi h a PMMA laye bu wi h lowe pi ch. Such obse a ion can
be de i ed om he ac ha o he lowe pi ch, he p esence o an ex a laye (PMMA)
and, he e o e, addi ional hickness, would mean a spoilage o he sensing p ope ies as
29
CHAPTER 4. CONCLUSIONS AND FUTURE PERSPECTIVES
he in e locked domain may ac as a smoo h ilm, ins ead o inc easing he con ac a ea
be ween bo h mic os uc u ed ilms.
The highe sensi i i y eached was 2.4
×10−1kPa−1
o ex e nal p essu es anging
om 4 Pa o 150 Pa. In ac , one expec s ha he use o hinne PDMS ilms enhances
he sensi i i y o he ull de ice. These senso s we e es ima ed wi h a p ice o 0.351
€
o
ma e ials cos s, as seen in Sec ion G.
In conclusion, he app oaches p esen ed he e con ibu e o a new di ec ion o lexible,
low-cos , and easily assembled o u u e heal h moni o ing de ices because despi e he
as p og ess in e-skin de ices, he e is oom o in es iga ion in he ield o low-cos
and low powe consump ion senso s, whe e he in eg a ion o mul iple o he sensing
p ope ies, namely empe a u e, shea , and ib a ion, is also appealing. O he appealing
unc ionali ies such as biocompa ibili y, sel -healing also need u he a en ion o he
implemen a ion in e-skin o ully mimic i s o ganic pa allel. Sel -powe ing, o example,
also cons i u es a demand once i would allow an au onomy deg ee which is no possible
wi h an immo able powe supply. This could be achie ed by adding, o example, sola
cells, ba e ies and wi eless an ennas.
Based on he key ea u es in he p esen wo k, namely he mic os uc u es, upcom-
ing wo ks will equi e a en ion on imp o ing he sensi i i y by using newly de eloped
unc ional ma e ials and op imiza ion o de ice geome ies.
30
Bibliog aphy
[1]
M. L. Hammock, A. Cho os, B. C.-K. Tee, J. B.-H. Tok, and Z. Bao. “25 h anni e -
sa y a icle: he e olu ion o elec onic skin (e-skin): a b ie his o y, design con-
side a ions, and ecen p og ess”. In: Ad anced Ma e ials 25.42 (2013), pp. 5997–
6038.
[2]
P. Delmas, J. Hao, and L. Roda -Despoix. “Molecula mechanisms o mechano ans-
duc ion in mammalian senso y neu ons”. In: Na u e e iews. Neu oscience 12.3
(2011), pp. 139–153.
[3]
J. Kim, M. Lee, H. J. Shim, R. Gha a i, H. R. Cho, D. Son, Y. H. Jung, M. Soh, C.
Choi, S. Jung, e al. “S e chable silicon nano ibbon elec onics o skin p os hesis”.
In: Na u e communica ions 5 (2014), pp. 1–11.
[4]
S. Baue , S. Baue -Gogonea, I. G az, M. Kal enb unne , C. Keplinge , and R. Schwö-
diaue . “25 h anni e sa y a icle: a so u u e: om obo s and senso skin o
ene gy ha es e s”. In: Ad anced Ma e ials 26.1 (2014), pp. 149–162.
[5]
D. Son, J. Lee, S. Qiao, R. Gha a i, J. Kim, J. E. Lee, C. Song, S. J. Kim, D. J. Lee,
S. W. Jun, e al. “Mul i unc ional wea able de ices o diagnosis and he apy o
mo emen diso de s”. In: Na u e nano echnology 9.5 (2014), pp. 397–404.
[6]
G. Schwa z, B. C. Tee, J. Mei, A. L. Apple on, D. H. Kim, H. Wang, and Z. Bao.
“Flexible polyme ansis o s wi h high p essu e sensi i i y o applica ion in elec-
onic skin and heal h moni o ing”. In: Na u e communica ions 4 (2013), pp. 1–
8.
[7]
C. Pang, G.-Y. Lee, T.-i. Kim, S. M. Kim, H. N. Kim, S.-H. Ahn, and K.-Y. Suh. “A
lexible and highly sensi i e s ain-gauge senso using e e sible in e locking o
nano ib es”. In: Na u e ma e ials 11.9 (2012), pp. 1–7.
[8]
J. Pa k, Y. Lee, J. Hong, M. Ha, Y.-D. Jung, H. Lim, S. Y. Kim, and H. Ko. “Gian
unneling piezo esis ance o composi e elas ome s wi h in e locked mic odome
a ays o ul asensi i e and mul imodal elec onic skins”. In: ACS nano 8.5 (2014),
pp. 4689–4697.
[9]
S. Gong, W. Schwalb, Y. Wang, Y. Chen, Y. Tang, J. Si, B. Shi inzadeh, and W. Cheng.
“A wea able and highly sensi i e p essu e senso wi h ul a hin gold nanowi es”.
In: Na u e communica ions 5 (2014), pp. 1–8.
[10]
D. J. Lipomi, M. Vosgue i chian, B. C. Tee, S. L. Hells om, J. A. Lee, C. H. Fox, and
Z. Bao. “Skin-like p essu e and s ain senso s based on anspa en elas ic ilms o
ca bon nano ubes”. In: Na u e nano echnology 6.12 (2011), pp. 788–792.
[11]
Y. Zang, F. Zhang, D. Huang, X. Gao, C.-a. Di, and D. Zhu. “Flexible suspended
ga e o ganic hin- ilm ansis o s o ul a-sensi i e p essu e de ec ion”. In: Na u e
communica ions 6 (2015), pp. 1–9.
31

BIBLIOGRAPHY
[12]
S. C. Manns eld, B. C. Tee, R. M. S ol enbe g, C. V. H. Chen, S. Ba man, B. V. Mui ,
A. N. Sokolo , C. Reese, and Z. Bao. “Highly sensi i e lexible p essu e senso s
wi h mic os uc u ed ubbe dielec ic laye s”. In: Na u e ma e ials 9.10 (2010),
pp. 859–864.
[13]
C. Dagde i en, Y. Su, P. Joe, R. Yona, Y. Liu, Y.-S. Kim, Y. Huang, A. R. Damado-
an, J. Xia, L. W. Ma in, e al. “Con o mable ampli ied lead zi cona e i ana e
senso s wi h enhanced piezoelec ic esponse o cu aneous p essu e moni o ing”.
In: Na u e communica ions 5 (2014), pp. 1–9.
[14]
Z. Pi, J. Zhang, C. Wen, Z.-b. Zhang, and D. Wu. “Flexible piezoelec ic nanogene -
a o made o poly ( inylidene luo ide-co- i luo oe hylene)(PVDF-T FE) hin ilm”.
In: Nano Ene gy 7 (2014), pp. 33–41.
[15]
L. Pe sano, C. Dagde i en, Y. Su, Y. Zhang, S. Gi a do, D. Pisignano, Y. Huang, and
J. A. Roge s. “High pe o mance piezoelec ic de ices based on aligned a ays o
nano ibe s o poly [( inylidene luo ide-co- i luo oe hylene]”. In: a Xi p ep in
a Xi :1309.7166 (2013).
[16]
F.-R. Fan, L. Lin, G. Zhu, W. Wu, R. Zhang, and Z. L. Wang. “T anspa en iboelec-
ic nanogene a o s and sel -powe ed p essu e senso s based on mic opa e ned
plas ic ilms”. In: Nano le e s 12.6 (2012), pp. 3109–3114.
[17]
G. Zhu, W. Q. Yang, T. Zhang, Q. Jing, J. Chen, Y. S. Zhou, P. Bai, and Z. L. Wang.
“Sel -powe ed, ul asensi i e, lexible ac ile senso s based on con ac elec i ica-
ion”. In: Nano le e s 14.6 (2014), pp. 3208–3213.
[18]
L. Lin, Y. Xie, S. Wang, W. Wu, S. Niu, X. Wen, and Z. L. Wang. “T iboelec ic ac i e
senso a ay o sel -powe ed s a ic and dynamic p essu e de ec ion and ac ile
imaging”. In: ACS nano 7.9 (2013), pp. 8266–8274.
[19]
Y. Zang, F. Zhang, C.-a. Di, and D. Zhu. “Ad ances o lexible p essu e senso s
owa d a i icial in elligence and heal h ca e applica ions”. In: Ma e ials Ho izons
2.2 (2015), pp. 140–156.
[20]
A. D. McNaugh and A. D. McNaugh . Compendium o chemical e minology. Vol. 1669.
Blackwell Science Ox o d, 1997.
[21]
S. Xu, Y. Qin, C. Xu, Y. Wei, R. Yang, and Z. L. Wang. “Sel -powe ed nanowi e
de ices”. In: Na u e nano echnology 5.5 (2010), pp. 366–373.
[22]
C. Pan, L. Dong, G. Zhu, S. Niu, R. Yu, Q. Yang, Y. Liu, and Z. L. Wang. “High-
esolu ion elec oluminescen imaging o p essu e dis ibu ion using a piezoelec-
ic nanowi e LED a ay”. In: Na u e Pho onics 7.9 (2013), pp. 752–758.
[23]
R. D. P. Wong, J. D. Posne , and V. J. San os. “Flexible mic o luidic no mal o ce
senso skin o ac ile eedback”. In: Senso s and Ac ua o s A: Physical 179 (2012),
pp. 62–69.
32
BIBLIOGRAPHY
[24]
J. A. Dob zynska and M. A. Gijs. “Flexible polyimide-based o ce senso ”. In:
Senso s and Ac ua o s A: Physical 173.1 (2012), pp. 127–135.
[25]
R. Pue s. “Capaci i e senso s: when and how o use hem”. In: Senso s and Ac ua-
o s A: Physical 37 (1993), pp. 93–105.
[26]
M.-Y. Cheng, C.-L. Lin, Y.-T. Lai, and Y.-J. Yang. “A polyme -based capaci i e
sensing a ay o no mal and shea o ce measu emen ”. In: Senso s 10.11 (2010),
pp. 10211–10225.
[27]
L. Pan, A. Cho os, G. Yu, Y. Wang, S. Isaacson, R. Allen, Y. Shi, R. Dauska d ,
and Z. Bao. “An ul a-sensi i e esis i e p essu e senso based on hollow-sphe e
mic os uc u e induced elas ici y in conduc ing polyme ilm”. In: Na u e commu-
nica ions 5 (2014), pp. 1–8.
[28]
J. G. Dabling, A. Fila o , and J. W. Wheele . “S a ic and cyclic pe o mance e al-
ua ion o senso s o human in e ace p essu e measu emen ”. In: Enginee ing in
Medicine and Biology Socie y (EMBC), 2012 Annual In e na ional Con e ence o he
IEEE. IEEE. 2012, pp. 162–165.
[29]
T. Yamada, Y. Hayamizu, Y. Yamamo o, Y. Yomogida, A. Izadi-Naja abadi, D. N.
Fu aba, and K. Ha a. “A s e chable ca bon nano ube s ain senso o human-
mo ion de ec ion”. In: Na u e nano echnology 6.5 (2011), pp. 296–301.
[30]
X. Wang, L. Dong, H. Zhang, R. Yu, C. Pan, and Z. L. Wang. “Recen p og ess in
elec onic skin”. In: Ad anced Science 2.10 (2015).
[31]
Y. Kim, J. Zhu, B. Yeom, M. Di P ima, X. Su, J.-G. Kim, S. J. Yoo, C. Uhe , and
N. A. Ko o . “S e chable nanopa icle conduc o s wi h sel -o ganized conduc i e
pa hways”. In: Na u e 500.7460 (2013), pp. 59–63.
[32]
D.-H. Kim, J. Xiao, J. Song, Y. Huang, and J. A. Roge s. “S e chable, cu ilinea
elec onics based on ino ganic ma e ials”. In: Ad anced Ma e ials 22.19 (2010),
pp. 2108–2124.
[33]
F. Xu and Y. Zhu. “Highly conduc i e and s e chable sil e nanowi e conduc o s”.
In: Ad anced ma e ials 24.37 (2012), pp. 5117–5122.
[34]
Y.-L. Pa k, B.-R. Chen, and R. J. Wood. “Design and ab ica ion o so a i icial skin
using embedded mic ochannels and liquid conduc o s”. In: IEEE Senso s Jou nal
12.8 (2012), pp. 2711–2718.
[35]
X. Wang, Y. Gu, Z. Xiong, Z. Cui, and T. Zhang. “Silk-molded lexible, ul asensi i e,
and highly s able elec onic skin o moni o ing human physiological signals”. In:
Ad anced Ma e ials 26.9 (2014), pp. 1336–1342.
[36]
C.-L. Choong, M.-B. Shim, B.-S. Lee, S. Jeon, D.-S. Ko, T.-H. Kang, J. Bae, S. H. Lee,
K.-E. Byun, J. Im, e al. “Highly s e chable esis i e p essu e senso s using a con-
duc i e elas ome ic composi e on a mic opy amid a ay”. In: Ad anced ma e ials
26.21 (2014), pp. 3451–3458.
33
BIBLIOGRAPHY
[37]
G. Y. Bae, S. W. Pak, D. Kim, G. Lee, D. H. Kim, Y. Chung, and K. Cho. “Linea ly
and Highly P essu e-Sensi i e Elec onic Skin Based on a Bioinspi ed Hie a chical
S uc u al A ay”. In: Ad anced Ma e ials 28.26 (2016), pp. 5300–5306.
[38]
C. O. Chey, X. Liu, H. Alnoo , O. Nu , and M. Willande . “Fas piezo esis i e senso
and UV pho ode ec o based on Mn-doped ZnO nano ods”. In: physica s a us solidi
(RRL)-Rapid Resea ch Le e s 9.1 (2015), pp. 87–91.
[39]
S. Chun, Y. Kim, H.-S. Oh, G. Bae, and W. Pa k. “A highly sensi i e p essu e senso
using a double-laye ed g aphene s uc u e o ac ile sensing”. In: Nanoscale 7.27
(2015), pp. 11652–11659.
[40]
X. Yin, T. Vinod, and R. Jelinek. “A lexible high-sensi i i y piezo esis i e senso
comp ising a Au nano ibbon-coa ed polyme sponge”. In: Jou nal o Ma e ials
Chemis y C 3.35 (2015), pp. 9247–9252.
[41]
S. Wagne and S. Baue . “Ma e ials o s e chable elec onics”. In: M s Bulle in
37.3 (2012), pp. 207–213.
[42]
K. Ku iba a, H. Wang, N. Uchiyama, K. Fukuda, T. Yoko a, U. Zschieschang, C. Jaye,
D. Fische , H. Klauk, T. Yamamo o, e al. “O ganic ansis o s wi h high he mal
s abili y o medical applica ions”. In: Na u e communica ions 3 (2012), pp. 1–7.
[43]
M. Kal enb unne , T. Seki ani, J. Reede , T. Yoko a, K. Ku iba a, T. Tokuha a, M.
D ack, R. Schwödiaue , I. G az, S. Baue -Gogonea, e al. “An ul a-ligh weigh
design o impe cep ible plas ic elec onics”. In: Na u e 499.7459 (2013), pp. 458–
465.
[44]
Y. Hwang, O. H. Payda , and R. N. Candle . “3D p in ed molds o non-plana
PDMS mic o luidic channels”. In: Senso s and Ac ua o s A: Physical 226 (2015),
pp. 137–142.
[45]
S. K. Sia and G. M. Whi esides. “Mic o luidic de ices ab ica ed in poly (dime hyl-
siloxane) o biological s udies”. In: Elec opho esis 24.21 (2003), pp. 3563–3576.
[46]
T. Seki ani, U. Zschieschang, H. Klauk, and T. Someya. “Flexible o ganic ansis o s
and ci cui s wi h ex eme bending s abili y”. In: Na u e ma e ials 9.12 (2010),
pp. 1015–1022.
[47]
M. Pa k, J. Im, M. Shin, Y. Min, J. Pa k, H. Cho, S. Pa k, M.-B. Shim, S. Jeon, D.-Y.
Chung, e al. “Highly s e chable elec ic ci cui s om a composi e ma e ial o
sil e nanopa icles and elas ome ic ib es”. In: Na u e nano echnology 7.12 (2012),
pp. 803–809.
[48]
L. Hu, M. Pas a, F. La Man ia, L. Cui, S. Jeong, H. D. Deshaze , J. W. Choi, S. M. Han,
and Y. Cui. “S e chable, po ous, and conduc i e ene gy ex iles”. In: Nano le e s
10.2 (2010), pp. 708–714.
34
BIBLIOGRAPHY
[49]
S. Me ilampi, T Laine-Ma, and P. Ruuskanen. “The cha ac e iza ion o elec ically
conduc i e sil e ink pa e ns on lexible subs a es”. In: Mic oelec onics eliabili y
49.7 (2009), pp. 782–790.
[50]
Q. Sun, D. H. Kim, S. S. Pa k, N. Y. Lee, Y. Zhang, J. H. Lee, K. Cho, and J. H.
Cho. “T anspa en , Low-Powe P essu e Senso Ma ix Based on Coplana -Ga e
G aphene T ansis o s”. In: Ad anced Ma e ials 26.27 (2014), pp. 4735–4740.
[51]
S. Pa k, M. Vosgue ichian, and Z. Bao. “A e iew o ab ica ion and applica ions
o ca bon nano ube ilm-based lexible elec onics”. In: Nanoscale 5.5 (2013),
pp. 1727–1752.
[52]
M. Moni uzzaman and K. I. Winey. “Polyme nanocomposi es con aining ca bon
nano ubes”. In: Mac omolecules 39.16 (2006), pp. 5194–5205.
[53]
M.-L. Seol, J.-H. Woo, D.-I. Lee, H. Im, J. Hu , and Y.-K. Choi. “Na u e-Replica ed
Nano-in-Mic o S uc u es o T iboelec ic Ene gy Ha es ing”. In: Small 10.19
(2014), pp. 3887–3894.
35

Appendix D
P essu e applying sys em
He ein, he code sec ion ela ed o he mo o mo ion in o de o p ope ly p ocede o
p essu e applica ion is shown. All he assigned a iables a e also shown.
1// Mo o Pins
2cons in s epPin = 7;
3cons in di Pin = 8;
4
5// Lowe s eps
6#de ine MS1 3
7#de ine MS2 5
8#de ine MS3 6
9
10 in s eps = 1; // S eps chosen by he ope a o du ing p og am. S eps can go
om 1 o 10 meaning 70 um o 700 um
11 in DT = 1000;
12 loa TT = 0.05;
13 in cycles = 9999; // Cicles
14 loa equency = 0.5 ; // F equency chosen by he ope a o by con oling
he po en iome e
15
16 oid uncao1(){
17
18 digi alW i e(MS1,HIGH); // P og aming 1/16 s ep
19 digi alW i e(MS2,HIGH);
20 digi alW i e(MS3,HIGH);
21
22 in cycle1 = 0;
23
24 TT=1.0/ equency;// Pe iod
25 DT=in (TT/(4.0*s eps)*1000.0); //Delay Time DD=(Pe iod/4*s eps)*1000
26
27 while (cycle1 <cycles){ // P og am keeps unning un il 9999 cycles
28
29 digi alW i e(di Pin ,HIGH); // S eps in one di ec ion
30 o (in x=0;x<s eps ;x++) {
31
32 digi alW i e(s epPin ,HIGH);
33 delay(DT);
34 digi alW i e(s epPin ,LOW);
35 delay(DT);
36 }
37
38
39 digi alW i e(di Pin ,LOW); // S eps in he o he di ec ion
43
APPENDIX D. PRESSURE APPLYING SYSTEM
40 o (in x=0;x<s eps ;x++) {
41
42 digi alW i e(s epPin ,HIGH);
43 delay(DT);
44 digi alW i e(s epPin ,LOW);
45 delay(DT);
46 }
47
48
49
50 cycle1 =cycle1 + 1;
51 }
52 }
(1)
(2)
(3)
(4)
(5)
(6)
(7) (8)
Figu e D.1: Schema ic ep esen a ion o he de eloped sys em. (1) 3D p in ed s eppe
mo o suppo (2) and (3) 3D p in ed componen s esponsible o con e ing adial mo e-
men in e ical mo emen (4) 3D p in ed compon en ha slides o e wo pa allel ba s
and holds he ubbe (5) Rubbe ha ac s as he inal objec applying he o ce (6) Piezo-
elec ic senso capable o de ec ing o ce (7) A duino boa d con olling he sys em (8)
LCD o an easie use in e ace.
44
Figu e D.2: Ske ch in Ske chUp o componen 1 o he homemade sys em o pos e io 3D
p in ing.
Figu e D.3: Ske ch in Ske chUp o componen s 2 and 3 o he homemade sys em o
pos e io 3D p in ing.
Figu e D.4: Ske ch in Ske chUp o componen 4 o he homemade sys em o pos e io 3D
p in ing.
45
APPENDIX D. PRESSURE APPLYING SYSTEM
Figu e D.5: RC Low-Pass Fil e wi h Op Amp Bu e wi h 3 poles wi h a cu ing equency
o 10 Hz.
46
Appendix E
Senso ope a ion p inciple
Time (s)
Resis ance (Ω)
ΔR
ROFF
RON
Unloading Unloading
Loading
Figu e E.1: Senso ope a ion p inciple based on esis ance changes in esponses o loading
and unloading (ROFF: unloading, RON: loading).
47

Appendix F
Resis ance changes o di e en equencies
01234567
0.00%
0.10%
0.20%
0.30%
0.40%
0.50%
0.60%
0.70% 1 H z
DR/R0
T i m e ( s )
(a)
0 . 0 0 0 . 0 5 0 . 1 0 0 . 1 5 0 . 2 0 0 . 2 5
0.00%
0.10%
0.20%
0.30%
0.40%
0.50%
0.60%
0.70% 1 5 H z
T i m e ( s )
DR/R0
(b)
Figu e F.1: Resis ance changes in esponse o an applied p essu e o PMMA-coa ed PDMS
memb anes wi h semi-sphe es wi h a pi ch o 200
µ
m wi h a equency o (a) 1 Hz (b) 15
Hz.
49
Appendix G
Es ima ed senso p ice
Table G.1: Es ima ed senso p ice ega ding ma e ials cos s
P ice (€) Quan i y (kg) Amoun used (g) Cos (€)
PDMS 250 1.1 1.187 0.326
Ca bon-ink 70.54 0.1 0.322 0.002
Sil e -ink 500.80 0.1 0.456 0.023
To al 0.351
51