B.Tech Project Report

Growth and Characterization of
Ln-doped Bismuth Titanate Thin Films.

B.Tech Project Report

Submitted by: Sumeet Kumar Project Guide: Dr.Ashish Garg
B. Tech Assistant Professor
Department of Materia and Metal
ls lurgical Engineering. Department of Mater als and Metal
i lurgical Engineering
Indian Inst tute of Technology Kanpur (I K )
i IT . Indian Institute of Technology Kanpur (I K).
IT
Kanpur-208016. Kanpur-208016
India. India.
E mail sum eetkv @ g m ail m,
: .co E mail ashishg @ii
: tk.ac.in
s meet
u .ku mar @rediffmail m
.co Tel : +91-512-2597904 (Office) 2598372 (Residence)
,
Tel +91-9415540761(Mob.),
: Fax : +91-512-2597505
+91-512-2561650(Residence).

Signature: Signature:

1

Acknowledgement
Firs ly I would l
t ike to thank Dr. Ashish Garg for being my guide in this course and bel eving in m e.
i
Although Electronics was my dee med field of study but r ight fro m the t ime I took the course on
Electronic Materials under the guidance of Dr. Ashish Garg I developed a liking towards it so m uch so that I
,
a m looking forward to do my research in this area.

I would also like to na m e a few persons without w ho m this project would never have reached a stage i has
t
reached now. These include: Kart kyen (SE M Lab incharge) for helping me to take SE M i mages out of turn,
i ,
Nitin, Gaurav, Abhinav, Abhishek, Dipthi Mr. A m ol, Mr. Tripathi (He made sure that al the Laboratory
, l
equip ments were working).

I think that these people are the reason for whatever I a m able to do today and I suppose that they wil keep
l
m otivating m e to do such great things in future.

2

Chapters

• A bstract Page – 4.
• Introduct ion Page – 5 to 6.
• O bjectives and Results Page – 7 to 9.
• O ptical Images Page – 10.
• H ypothesis of Thin Films Page – 11 to 12.
• SE M I mages Page – 13 to 17.
• C- V Characteris ics
t Page – 18.
• C- V Curves Page – 19 to 23.
• Film thickness measurem ents Page – 24.
• X R D Pat terns Page – 25.
• Intensi rat o Vs Annealing Time, Te mperature
ty i . Page – 26.
• Conclusion and Future w ork Page – 27 to 28.
• References Page – 29.

3

Abstract
In recent years the fa m i of bismuth layer str
, ly uctured ferroelectrics has received m uch at tention as the
candidate for ferroelectric rando m access me mor . Bi4Ti3O12 (BiT) or bismuth oxides are extensively
ies
studied mem bers of the Aurivi ll ius fa mily for their large spontaneous polar ization along the a- axis (~50
2
µ C/c m ), low processing te mperature high Curie temperature and Pb-free mater l . Ho wever, i suffered
, , ia s t
6
severe polarization suppression af ter 10 switchi cycles when deposi ed on Pt/Si substrates It has been
ng t .
i ion of Bi in the perovski unit cel of Bi4Ti3 O12 by a lanthanide
sho wn by previous studies that subst tut te l
ele ment such as La leads to re markable improvem ent in the fatigue behavior of i f l on Pt/Si substrates
ts i ms .
H o wever t research in this area is far from co mplete. Our at
, he tempt is to dope Bi4 Ti3 O1 2 with other
lanthanides such as La, N d, S m either separately or together to understand the effect of size of dopants on
the structure and ferroelectr proper ies of the f lms as size difference is a key elem ent to the creat
ic t i ion of
remanent polar ization in the ferroelectr Bi4Ti3O12 f lms. W e wil use spin coat
ic i l ing technique to fabricate
the f lms and study the structure of the f lms using X-ray diffract
i i ion, Scanning electron microscopy, ato mic
force microscopy and ferroelectr measure ments wi l made t understand the electr cal behavior.
ic l o i

4

Introduction
There have been extensive effor to enhance the rel
ts iabi i of perovski es based ferroelectr thin f lms for
l ty t ic i
use in non-volat le ferroelectr rando m access m e m ory devices FRAM’s are non-volat le mem ory devices;
i ic . i
i . data stored is not lost once the po wer is switched off In F R A M inform ation is st
.e . ored in the polar ization
state of the ferroelectr c mater l Structure of FR A M is similar to D R A M, in w hich me m ory cel are
i ia . ls
arranged in a square mat ix and infor mation is stored in terms of sign of charge. Earl , Lead Zirconate
r ier
Titanate (PZ T) e merged as an important candidate for F R A M s. Ho wever i f lms show ed a serious
, ts i
7
degradat ion of ferroelectr propert af
ic ies ter being subjected to 10 read/wri switching cycles when deposi
te ted
on Pt electrodes. Later i was sho w n that layered ferroelect ics such as Stront m Bis muth Titanate (SB T )
t r iu
sho wed superior fat igue resistances, sho wn in Fig. 1, as co m pared to Pt PZ T/Pt capaci
/ tors as show n in Fig. 2.
[1]
H o wever the high processing te mperature
, of S B T above 750 °C is an obstacle in integrat ion with si icon
l
devices Bism uth t tanate (BT O) e m erged as a later candidate for these appl
. i ications due to i high re manent
ts
polarization in bulk state but undoped BT O sho wed high fat igue fai ures but with an advantage of low
l
processing te mperature This is expla
. ined in terms of the volat le nature of Bis muth ato m, and as Bi was
i
supposed to bind the Oxygen ato m s together w hen i is gone the oxygen ato ms also beco me free to move,
, t
[2]
thus creating vacancies and i is this vacancy that has been postulated as a reason for f igue fai
t at lures .

Also, people have exper imental proved this fact by replacing al the stront m ato ms by bism uth ions and
ly l iu
[1]
half of the Ta ions with Ti ions in order to maintain charge neutral ty and f
i inal get Bi3 TiTa O9 . The resul
ly t
of such an at tempt was al though good in terms of the electr ical properties but i sho wed serious fat
t igue
[1]
failures suggest
, ing that Bi ions do affect fat igue characteris ics I was sho wn later by Park et al
t . t . that
Lanthanide ele ments doping into the BT O thin f lms increases the fat gue resis
i i tance by several folds, as
[ 2]
sho wn in Fig. 1, but the re manent polar ization (2Pr) decreased in Lanthanu m doped Thin Fi m . This is
l
thought to be due to s m aller size difference between La and Bi which reduces the overal dis tl tor ion in the
perovski l
te ayers and hence low 2Pr. Thus, we are probing into this proble m by doping i with Sam ariu m ions
t
since i has got bigger size than La, therefore dis t
t tor ion of the crystal l t
at ice wil occur and so the oxygen
l
binding capabi i and t
l ty hus the fatigue resistance increases quite a lot.

[4]
Thin f lms of BT O have already been prepared by sol–gel process . A m ong the various techniques avai
i lable
for the fabrication of B T O thin f lms, sol–gel processing has been em ployed in this study w hich offers
i
excellent uniformity over large area; easy co mposi ion control shor fabricat
t , t ion time, as w ell as low
temperature process a co mparat
t ively low cost.

[4]
In sol–gel process t che mical stabi i of the solut
, he l ty ion is very im portant In our case, the che mical
.
instabi i of the solut
l ty ion has been overco me by the addi i of acetyl acetone in the precursor solut
t on ion and
f lms wil be prepared by spin-coat ng technique and their st
i l i ructural morphological and co mposi ional of the
, t
B T O thin f l are to be syste matical studied in the near future Also the p H of the Sol should be around 3.5
i ms ly .
[6]
for obtaining highly c-axis oriented Thin Films .

5

Comparison of PZT, SBT, BLT deposited on Pt Substrate

Fig. 1 NATURE Volume 401 14th October 1999

Fig. 2 Fig. 3

Results of Fatigue tests at 1 MHz La-Substituted BLT on SBT/Pt/SiO2/Si

(a)P-E hysteresis loop for Pt/Si film before
(filled circles) & after (open circles) at 3 x 1010 Cycles.
(b) Variation of Psw, Pns when negative read voltage is applied.

NATURE Volume 401 14th October 1999.

6

Objective of the Present work
• Deposit of S m doped Bi4 Ti3 O12 thin fi
ion lms
• Investigation of the structure, morphology, co mpositional hom ogenei and thickness uniformity of the f lms.
ty i
• Ferroelectric Measure ments: Dielectric constant, Re manent polarization and coercive field.

Experiments and Results
W e have prepared a Sol; see Fig. 4, using anhydrous Sa marium nitrate Bismuth ni ra e, Ti
, t t taniu m IsoPropoxide
mixed into glacial acet acid and acetyl acetone added as stabi izer Our goal is to f
ic l . ind the effect of doping S m,
[1]
La and s tudy the fatigue resistance of the Thin Film by deposi ing i on Ti/Si substrates
t t .
Further anneal
, ing of Sam ariu m / Nd doped Bism uth Ti tanate Thin f lms of different thickness wil be done and
i l
[1]
Pr , the rem anent Polar ion, wil be calculated for each case.
izat l

Fig. 4 The final Sol Prepared.

7

Solution Preparation & Hot plate calibration:

Calculations:

Fig. 7

Here we have to prepare a 0.1 M solut ion. I a m presently consider ng the case of X= 0.5, thus we are
i
m aking Bi3.85Sm0.5Ti3012 Sol Fig. 4.

Weighing:

Weight of Bi(NO3)3.5H2O required(see Fig. 7): For 7.5 m m ol of solution => 7.5*10-3 *484.99*3.85 =
14.00408625 gra ms.

Here 484.99 is the Molecular weight of Bi(NO3)3.5H2O and 3.85 being the actual no. of moles of Bi required.
N ote: We have taken 10% excess Bi as i i volati .
t s le

Si milarly we can get the weights of Nd Nitrate, Samarium nitrate, Lanthanum Nitrate
(for La(NO3)3 doped– BLT deposit on ) and Ti Isopropoxide. We have m ade the volum e as half, thus gett ~
i ing
38 ml of solution finally, with acetic acid as solvent and the weights have also been taken as half the above
calculated.

8

Drying:

W e have dried the respective ni ra
t tes after weighing the m on the electronic balance Fig. 8 and then put them into
the paddy discs for drying into the Oven for 12 Hrs .

Problems : The Bi(NO3)3 obtained had melted and stuck onto the botto m of the paddy disc. Sa me
happened with Lanthanum Nitrate.

Remedy : Possible solution may be controlled heating in the oven or dissolving the m elted salts into the
glacial acet c acid by putt
i , igure 1, on the Magnetic St r , Fig. 9 and heat
ing the paddy discs f i rer ing to
0
Te mperature of about 60-70 C.

Fig. 8 Fig. 9

The Electronic Balance The Magnetic Stirrer and Hot plate.

Fig. 5 Calibration of the Heater surface. Fig. 6 Calibration of Distilled water.

9

Optical Images

Bi3.85S m 0 .5 Ti3 O12 _5_ Drops_60 rp m Bi3.85 S m0.5 Ti3 O12 _5_ D rops_60 rp m
not annealed_10 X _Centre_Scaled_4 Coats . not annealed_5 X_Edge_Scaled_4 Coats

Bi3.85S m 0 .5 Ti3 O12 _5_ Drops_60 rp m Bi3.85S m 0 .5 Ti3 O12 _5_ Drops_60 rp m
not annealed_10 X_Centre_Scaled_5 Coats . not annealed_5 X_ Cent re_Scaled_5 Coats.

Figure 42.

10

Hypothesis of Thin Films

If a fluid of viscosity µ
and density ρ is initially flat, it remains flat during the spin coating process.

Assu mptions:
Steady s te si
ta tuation h(t)
Flo w is axial sy m metr c
ly i
Radial veloci >> z-veloci
ty ty
Film is thin
Surface tension can be neglected
Stresses ar ising fro m shear s resses do minate
t

Continuity equation:

In r direction.

Star ing with the Navier-Stokes’ equat
t ion:

Using the assu mptions s ated above we get
t :

r-direction:

z-direction:

11

At t=0

At h
Final we should get
ly :

At z=0
& at z=h

The general equation for fluid flow as th f lm.
in i

[5]
Variation of height of Sol as a funct on of densi y, angular veloci , ini i height viscosi , and t
i t ty t al , ty ime.

[5]
Height variation when evaporat on is considered .
i

O n Plot ing the curve on Matlab we find approxim ate relat
t ionships of Hf after so me long time of spin coating,
say 1 minute (60 seconds) .The curve decreased som e w hat steeply for the i ia few seconds and then varied
nit l
as constant f the rest of the t
or ime.
The ini ia height was taken as ~ 3 mm (The diam eter of the drop)
t l .

The Mathematical Plots for different viscosity SOL and angular velocities.
12

SEM Images
The var ious microstructures depicted by the optical Images and the SEM i mage are sho wn below to depict the
uniformity of the f lms deposi
i ted and the change in the morphology of the f lm. Note the changes which occur
i
as the f l anneal
im ing temperatures is increased (I gets coarser)
t .

Figure 10, 750 OC, 60 Min. Annealed, 5 Drops & 5 Coats. Figure 11, 750 OC, 60 Min. Annealed, 5 Drops & 5 Coats.

Sa mple 1 (Centre I age-25,000 X)
. m Sa mple 1. (Centre Image-50,000 X)

Figure 12, 750 OC, 60 Min. Annealed, 5 Drops & 5 Coats Figure 13, 750 OC, 60 Min. Annealed, 5 Drops & 5 Coats

Sa mple 1. (Edge Im age-25,000 X) Sa mple 1. (Edge Image-25,000 X, BSE)

13

Sa mple 2. (Centre Image-25,000X, SE) Sa mple 2 (Centre Image-25,000 X, BSE)
.

Sa mple 2. (Edge Image-25,000X, SE) Sam ple 2. (Edge Image-25,000 X, BSE)

Figure 18, 750 OC, 15 Min. Annealed, 5 Drops & 5 Coats Figure 19, 750 OC, 15 Min. Annealed, 5 Drops & 5 Coats.
Sa mple 3. (Centre Image-25,000X, SE) Sa mple 3 (Centre Image-25,000 X, BSE)
.
14

Sa mple 3. (Edge Image-25,000 X, SE) Sa m ple 3. (Edge Image-25,000 X, BSE)

Sa mple 4. (Centre Image-25,000 X, SE) Sa mple 4. (Centre Image-25,000 X, BSE)

15

Sa mple 5. (Centre Image-25,000X, SE) Sa m ple 5. (Centre Image-25,000 X, BSE)


16

Samples annealed for same time but at different Temperatures
(650 oC, 650 oC, 700 oC).

Sa mple 6. (Centre Image-25,000 X, SE) Sa mple 6. (Centre Image-25,000 X, BSE)

Sa mple 7. (Centre Image-25,000 X, SE) Sa mple 7 (Centre I age-25,000 X, BSE)
. m

Sa mple 8. (Centre Image-25,000X, SE) Sa m ple 8. (Centre Image-25,000 X, BSE)

17

C-V Characteristics
Taken for different sweeps.

U= up sweep i.e. from -10 volt to +10 volt.
D= down sweep i.e. from +10 volt to -10 volt.

Metallization Steps Conclusions and Results.
The sa m ples were prepared for taking C-V We can say that the Thin Fi lms produced w ere of very
character t in the fol
is ics lowing s eps:
t good qual t and were non conducting as on running the up
iy
and do wn sweeps we obtained almost the sa me C- V curve and
a) Sa mples after the deposi ion of Bis muth
t no “ Hyster sis” were developed, which are obtained due to
i
Titanate Thin f lms on Pt/Si substrate and
i m obile charges in the f l etc
i ms .
subsequent anneal ing in pure O2 (Grade-I) A pri 16, 2005
l
environ ment were sent for Metal izatl ion in the H o wever there was so m e Oxide leakage at high voltages as is
Sa mtel Centre for Display Technology. observed from the bending of the curves at the ends.
Here we did the m etal izat
l ion of gold
contacts (see f . on the Thin Fi
ig ) lms. The
samples were put with a mask in a machine
which vaporizes gold by subjecting i to
t
-6
2X 10 Pa.

b) After Metall ion we t ied to take the C-
izat r
V of
the f lms but were unsuccessful due to the
i
presence of the back surface oxide layer,
which ho wever was taken care of by exposing a
small top portion of the substrate by HF etching.

c) W e also did try to take the C-V by
deposi ing Indiu m dots with the help of soldering
t
iron, at the back of the substrate but st l proper
, il
contacts were not achieved and hence C- V was not figure 45.
taken.

18

C-V Curves for different frequencies and Sweeps.

@_U_10KHz, Sample 1, Location2 @_D_10KHz, Sample 1, Location2.

6.00E-09 6.00E-09
5.00E-09 5.00E-09
Capacitance

Capacitance
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.50E+ -1.00E+ -5.00E+ 0.00E+ 5.00E+ 1.00E+ -1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0
01 01 00 00 00 01 01 01 00 0 0 1
Voltage Voltage

@_U_1KHz, Sample 1, Location2 @_D_1KHz, Sample 1, Location2

8.00E-09 8.00E-09
7.00E-09 7.00E-09
6.00E-09
Capacitance

6.00E-09
Capacitance

5.00E-09 5.00E-09
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.50E -1.00E -5.00E 0.00E+ 5.00E+ 1.00E+ -1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0
+01 +01 +00 00 00 01 01 01 00 0 0 1
Voltage Voltage

19

@ _D_10 KHz, Sample 1, Location1. @_U_10 KHz, Sample 1, Location1.

6.00E-09 6.00E-09
5.00E-09 5.00E-09
Capacitance

Capacitance
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 -1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0
01 01 00 0 0 1 01 01 00 0 0 1
Voltage Voltage

@ _D_1 KHz, Sample 1, Location1. @U_1KHz, Sample 1, Location1.

8.00E-09 8.00E-09
7.00E-09 7.00E-09
6.00E-09 6.00E-09
Capacitance

Capacitance

5.00E-09 5.00E-09
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 -1.50E+ -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0
01 01 00 0 0 1 01 01 00 0 0 1
Voltage Voltage

20

@_U_10MHz, Sample 1, Location1. @ _D_10 MHz, Sample 1, Location1.

1.40E-09 1.40E-09
1.20E-09 1.20E-09

Capacitance
Capacitance

1.00E-09 1.00E-09
8.00E-10 8.00E-10
6.00E-10 6.00E-10
4.00E-10 4.00E-10
2.00E-10 2.00E-10
0.00E+00 0.00E+00
-1.50E -1.00E -5.00E 0.00E+ 5.00E+ 1.00E+ 1.50E+ -1.50E -1.00E -5.00E 0.00E+ 5.00E+ 1.00E+ 1.50E+
+01 +01 +00 00 00 01 01 +01 +01 +00 00 00 01 01
Voltage Voltage

@_U_1MHz, Sample 1, Location1. @ _D_1 MHz, Sample 1, Location1.

2.50E-09 2.50E-09

2.00E-09 2.00E-09
Capacitance
Capacitance

1.50E-09 1.50E-09

1.00E-09 1.00E-09

5.00E-10 5.00E-10

0.00E+00 0.00E+00
-1.50E -1.00E -5.00E 0.00E+ 5.00E+ 1.00E+ 1.50E+ -1.50E -1.00E -5.00E 0.00E+ 5.00E+ 1.00E+ 1.50E+
+01 +01 +00 00 00 01 01 +01 +01 +00 00 00 01 01
Voltage
Voltage

21

@_U_10kHz, Sample 2, Location1. @_D_10KHz, Sample 2, Location1.

6.00E-09 6.00E-09
5.00E-09 5.00E-09
Capacitance

Capacitance
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0 -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0
01 00 0 0 1 1 01 00 0 0 1 1
Voltage Voltage

@_D_10MHz, Sample 2, Location1. @_U_10MHz, Sample 2, Location1.

1.60E-09 1.60E-09
1.40E-09 1.40E-09
1.20E-09
Capacitance

1.20E-09
1.00E-09 Capacitance
1.00E-09
8.00E-10 8.00E-10
6.00E-10 6.00E-10
4.00E-10 4.00E-10
2.00E-10 2.00E-10
0.00E+00 0.00E+00
-1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0 -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0
01 00 0 0 1 1 01 00 0 0 1 1
Voltage Voltage

@_U_100kHz, Sample 2, Location1. @_D_100KHz, Sample 2, Location1.

2.50E-09 3.00E-09
2.00E-09 2.50E-09
Capacitance

Capacitance

1.50E-09 2.00E-09
1.50E-09
1.00E-09
1.00E-09
5.00E-10
5.00E-10
0.00E+00 0.00E+00
-1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0 -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0
01 00 0 0 1 1 01 00 0 0 1 1
Voltage Voltage

22

@_U_1MHz, Sample 2, Location1. @_D_1MHz, Sample 2, Location1.

2.50E-09 2.50E-09

2.00E-09 2.00E-09
Capacitance

Capacitance
1.50E-09 1.50E-09
1.00E-09 1.00E-09

5.00E-10 5.00E-10

0.00E+00 0.00E+00
-1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0 -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0
01 00 0 0 1 1 01 00 0 0 1 1
Voltage Voltage

@_U_1kHz, Sample 2, Location1. @_D_1kHz, Sample 2, Location1.

7.00E-09 7.00E-09
6.00E-09 6.00E-09
Capacitance
Capacitance

5.00E-09 5.00E-09
4.00E-09 4.00E-09
3.00E-09 3.00E-09
2.00E-09 2.00E-09
1.00E-09 1.00E-09
0.00E+00 0.00E+00
-1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0 -1.00E+ -5.00E+ 0.00E+0 5.00E+0 1.00E+0 1.50E+0
01 00 0 0 1 1 01 00 0 0 1 1
Voltage Voltage

Thus we see that the f lms are stable for
i A nnealing at higher temperatures in the sense
that their Up and Dow n S weep produce sa me C-V curve.

Figure 44.
The Ideal C-V Curve for a p - type substrate.
Just take a mirror image for the case of n-Type.

23

The Film Thickness Measurements

Figure 37. Figure 38.

Sample annealed at 650 oC.

Figure 39. Figure 40.
o
Sample annealed at 750 C.
24

XRD Patterns

XRD Patterns for 600, 650, 700, 750 oC annealed Samples.
Figure 41.

The X-Ray diffract ion patterns were studied and peaks characterized for t different phases present
he .
X-ray diffraction profiles were obtained using a Cu Kα radiation source at 30 kV and 20 m A tube current and a
sweep rate of 3 ° / min.
.0

Te mperature °C I 006/ I 117 I 111/ I 117 I 200/ I 117 I 311/ I 117
750 (15 min) 12/20 5/20 3/20 2/20
750 (30 min) 26/57 13/57 7/57 3/57
750 (60 min) 5.69/15.82 2.6/15.82 1.31/15.82 2.49/15.82
600 18.37/25 11.25/25 5/25 2/25
650 46.72/76.37 24.10/76.37 8.27/76.37 5.50/76.37
700 30/65.26 15/65.26 11/65.26 4/65.26
750 5.69/15.82 2.6/15.82 1.31/15.82 2.49/15.82
A nnealing Time( min) 60 60 60 60

25

Intensity Variation as a function of annealing time and temperature

Variation of Intensity ratio Vs Variation of Intensity Ratio Vs Temperature
Annealing Time at Constant annealing time(60 min.)

0.7 0.8
0.6

In te n s ity R a tio
0.5 I006/I117 0.6 I006/I117
I / I m ax

0.4 I111/I117 I111/I117
0.4
0.3 I200/I117 I200/I117
0.2 I311/I117 0.2 I311/I117
0.1
0 0
0 20 40 60 80 600 650 700 750 800
Time (min). Temperature

Figure 43.

The following points can be concluded fro m the above plots:

The X R D pat terns sho w the f lms are polycrystal i with preferred ~(117) or
i l ne ientation.
Pyrochlore phase is absent .
Correct peak posi ions i
t ndicate that S m ions dissolve into the pseudo perovski s ructure.
te t
The (117) peak intensi increased w. . o
ty r t ther peaks with increasing annealing temperature and t m e
i
indicating i s preferred nature.
t

26

Conclusion and Future work
The B T O solut ion was prepared by mixing Bism uth ni rate, Sa mariu m nitrate and t taniu m IsoPropoxide in
t , i
glacial acet c acid. Ini i ly bismuth ni rate was dissolved in acet acid in a ref
i t al t ic lux condenser heated at a
[4]
temperature of ~80 °C for 2 hrs . W h en the solut ion becam e transparent t taniu m IsoPropoxide was mixed
, i
in a proper m olar rat with constant st r ing at roo m te mperature to form a yel
io ir low-gold Fig. 4, t ransparent
solution. However, for B T O syste m the problem is in solution, the bism uth precursor reacts easi with H2 O
ly
[1]
to yield white precipi t BiO N O 3
ta e and hence the solution is unstable and deco mposes within a short t ime.
Therefore, i is necessary to f
t ind another path to synthesize Sols having stabi i over longer per
l ty iod. Earl ier
[4]
studies on the stabi i of the Sols sho wed that the alkoxide–alkanola mine
l ty syste m has an excel lent
dissolving po wer for many inorganic sal and a long-term stabil ty to hydrolysis and condensat
ts i ion. Alkoxide–
[1]
alkanola mine was thus very effect ve in preparing PbTiO3 thin f lms and their resul sho w ed that lead
i i ts
acetate of acetyl acetone gives a clear and stable solut ion. Based on this concept, we have added acetyl
acetone, and the precursor solut ion was found to be stable against precipi t ta ion for several days indicat
, ing
that the addition of alcohol prevents Bi(N O3 )3 fr m hydrolysis
o .

In the future study, Pt/Si(1 0 0) and bare si icon substrates are going to be used for the deposi ion of BT O thin
l t
f lms. The substrates will be cleaned in acetone, Carbon Tetra Chlor
i ide and rinsed in de-ionised water ,
followed by a drying process A plat m layer of thickness ~500 Å will be deposi ed on the (1 0 0) oriented
. inu t
Si substrates at roo m tem perature to act as the botto m electrode for electr ical character ization. The precursor
[3]
solut ion wil be spin coated onto the substrates by using a spin coater at 2300 rp m for so me 60 s . After spin
l
coat ing the substrates t f lms will to be kept in a mbient air for 1 h to form gel f lms by hydrolysis and
, he i i
poly merizat on. Spin coat
i ing process wil be repeated 3–5 t
l imes to obta f lms with desired thickness
in i ,
[3]
followed by pyrolysis of each layer at 350 °C for 10 min . Heat t reatment of dried f l wil be carr
im l ied out in
a tube furnace at the tem perature range 400–600 °C for 1 h in an atmosphere of ei ther flowing oxygen or
[3]
air . Crystal izat
l ion, densif ication and microstructure of the f lms wil be exa mined by X-ray diffracto meter
i l
[3]
with Cu K• radiat ion ( =1.5405 Å) and/or a mic force microscope .
to

To su m marize:

The X R D pat terns Figure 41, 43 are sho wing the preferred ~ (171) {Excluding the substrate peak}
orientat ion, and thus there is no pyrochlore phase. The correlat ion of the diffract ion peaks of the
B S m T thin films with those of BIT im plies that S m subst tution does not affect the layered-perovski
i te
structure of BIT.

This fact indicates that the S m ions in the BS m T films do not form a pyrochlore phase, but dissolve
into the pseudo perovski st
te ructure. Therefore, i see ms that S m ions ~1.0 Å can readily subst tute for
t i
Bi ions ~1.03 Å in pseudo perovski e structure, and part l subst tut
t ia i ion of S m ions for Bi ions in BIT
influenced by the s ructure of the Bi l
t ayer.

Having deposi ted the Thin Fi lms at various te mperature ranges and checked the f lms for oxide
i
leakage charges by doing the C-V measure ments and character
, ization done for the X R D pat terns
obtained, we look forward to do the A F M analysis for microstructure level analysis and further we will
be doing the Polarization Vs E measurements .

27

The peak intensi ies increased, and the ful width at hal maximum of the peaks decreased with
t l f
increasing anneal ing tem perature; i can be assu med that the grain size was increasing with anneal
t ing
temperature.
(Figure 10) sho ws the surface FE-SE M micrographs of BSm T thin f l i ms as funct ons of anneal
i ing
temperatures The surface morphology is very sensi ive to the anneal
. t ing te mperature The BSm T thin
.
f lm annealed at 750 °C @ Fig. 2 exhibi
i ted coarse grains and considerable a mounts of secondary
structures am ong the grains.
The grain s ze of the BSm T fi
i lms increased with increasing anneal ing tem perature .
See figures (10–36).
The (Bi: Sm): Ti :: 1 58 : 1. {Expected rat 1
. io: .33: 1}.
The pH of our Sol was found to be ~1-2.

These facts can be concluded:

1. The crystal growth is preceded and the ferroelectr propert
ic ies of the BS m T fi m s can be
l
improved by increasing t anneal
he ing temperature.
2. The BS m T films annealed at 700 °C have round-plate- ike grains of ~300 nm (Figure 36).
l
3. The Thickness of the f l varies around 0.5 Micro meters (Figure 37)
i ms .

C- V measure ments show that there is approxim ately the sa me path reversed for both the Up S weep
and Do wn Sweep, which implies that the leakage through the oxide layer i smalls .
O n Overlapping the C-V curves for different frequencies one f inds a sort of Hysteresis loop being
generated. This is due to the slow ST A T E S, which do not e mpty out fast enough even to slow D C
S weeps.
Study the thermal stabi i of such fi
l ty lms by anneal ing at different temperatures for different lengths of
time.
W e wil also study the deposi
l ted f l propert in terms of Vacancies and Stress / St ins present via
im ies ra
various methods l ike X-Ray Diffraction, TE M.
X R D and SE M Characterizat ion.
The surface roughness of the f lm is also an important para m eter taking into account the possibi i of
i l ty
further metal izat
l ion on the ferroelect ic f lm in the FeR A M s device fabricat
r i ion process. Thus, AF M
micrographs of the BS m T thin f lms as funct
i ions of anneal ing te mperature wil be taken to probe m ore
l
into this feature.
[5]
Fro m these analyses we are looking to f
, ind that t root mean square ~r ms surface roughness of the
he
B S m T thin f lms increases with increasing anneal
i ing te mperature. An effect which m ay be related to
the increase of the grain size with increasing anneal ing te mperature The rms surface roughnesses of
.
B S m T thin films are expected to vary fro m 4.01 to 6.74 n m for anneal ing te mperatures n the range of
[5]
650–750 °C .

28

References:
1. Lanthanum-Substituted Bismuth Titanate for use in non-volatile
memories;
B.H.Park,B.S.Kang,S.D.Bu,T.W.NOH,J.Lee & W.Jo;
NATURE Vol 401 14th October 1999.
2. Why Lanhanum-substituted BiT becomes fatigue free in a ferroelectric
capacitor with Pt electrodes; Y.Ding, J.S.LIU, H.X.Qin, J.S.ZHU, and
Y.N.Wang.
Appl. Phys. Lett.Volume 78, Number 26.
3. Growth of uniformly a-axis-oriented ferroelectric lanthanum-
substituted
Bismuth titanate films on silicon substrates;
Ho Nyung Lee,a) Dietrich Hesse, Nikolai Zakharov, Sung Kyun Lee, and
Ulrich Go¨ sele. Max-Planck Institut fu¨r Mikrostrukturphysik,
Weinberg 2, D-06120 Halle (Saale), Germany ~Received 25 July 2002;
accepted 14 February 2003.
Appl. Phys. Lett Volume 93, Number 9 1 May 2003.
4. Sol–gel synthesis and property studies of layered perovskite bismuth
titanate thin films - S. Madeswaran, N. V. Giridharan and R.
Jayavel;
Materials Chemistry and Physics 80 (2003) 23–28.
5. Large remanent polarization of cerium-modified bismuth–titanate thin
films for ferroelectric random access memories.
Kyoung-Tae Kim and Chang-Il Kim
School of Electrical & Electronic Engineering, Chungang University,
221, Huksuk-Dong, Dongjak-Gu,Seoul 156-756, Korea.
© 2003 American Vacuum Society.

6. Effects of precursor solution pH value and substrate texture on
orientation degree of sol-gel-derived bismuth titanate thin films
Haoshuang Gu, Wanqiang Cao, Rui Song, Xiaoyuan Zhou, John Wan
physica status solidi (a)Volume 198, Issue 2, 2003. Pages 282-288
Copyright © 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

29

B.Tech Project Report

More Related Content

Similar to B.Tech Project Report (20)

Recently uploaded (20)

B.Tech Project Report