Sol–gel chemistry of an aqueous precursor solution for YBCO thin films

Page 1

ORIGINAL PAPER

Sol–gel chemistry of an aqueous precursor solutionfor YBCO thin films

Tran Thi Thuy Æ S. Hoste Æ G. G. Herman ÆK. De Buysser Æ P. Lommens Æ J. Feys Æ D. Vandeput ÆI. Van Driessche

Received: 26 March 2009 / Accepted: 28 April 2009 / Published online: 19 May 2009

� Springer Science+Business Media, LLC 2009

Abstract A water based YBa2Cu3O7-x (YBCO) precur-

sor using a triethanolamine (TEA)/acetic acid complexing

method to obtain YBCO thin films is described in detail.

The influence of complexation behavior in the formation of

transparent and homogenous sols and gels after the com-

bination of Y, Ba and Cu—acetates, acetic acid and TEA

has been studied by potentiometric titration and the results

are compared with analytical simulations. The decompo-

sition of the gel was studied by IR (infrared) and Thermal

Gravimetric Analysis/Differential Thermal Analysis (TGA/

DTA). The results in sol-gel chemistry can be used to

decide on the necessities in the preparation of stable sol–

gel precursors with a minimum amount of organic com-

pounds. The sol–gel system was also used for the deposi-

tion of high textured superconducting thin films on STO

substrates by dip coating. The synthesized YBCO showed a

superconducting transition temperature of 90.95 K with

narrow DT (2 K) for thin films. The results from X-ray

diffraction show that the single phase YBCO was obtained.

SEM pictures also indicate that the properties of the surface

thin films are good.

Keywords Potentiometric titration � Sol–gel � Thin films �YBCO � Superconductors

1 Introduction

Soon after the discovery of superconducting materials [1],

the quest for the practical applications led to the investigation

on the current carrying capacity of this new material at the

boiling point of liquid nitrogen. Compared with other cup-

rates, YBCO [2] seems to be the most promising material

because of the current carrying ability in a magnetic field and

a high current density self-field [3]. Clearly the under-

standings of the unique properties of these copper-based

superconducting oxides, as well as their technological

applications, depend crucially upon composition, homoge-

neity and microstructure [4]. These three parameters are of

utmost importance during synthesis. From this point of view,

several review articles on the synthesis of high temperature

superconducting oxides have appeared [4–6]. Compared

with the other methods, the sol–gel process [7] has the

potential advantage not only of achieving homogeneous

mixing of the component cations on atomic scale, but also

forming films or fibers which are of great technological

importance [4]. A variety of strategies such as colloidal sol–

gel, inorganic polymeric gel derived from organometallic

compounds and gel routes involving formation of organic

polymeric glasses with success in obtaining a homogeneous

YBCO precursor solution without precipitation has been

documented [3–5]. However, a suitable technological pre-

cursor solution should contain minimal amounts of organic

compounds, expected to leave less carbon residue which is

detrimental for the superconducting properties [8]. For this

reason, this paper is devoted to the preparation of a water

based precursor system and to the control of the composition

of the solution to obtain a clear gel with minimal amounts of

organic compounds. Potentiometric titrations and IR mea-

surements were used to study the composition of the solu-

tions and the decomposition of the gels.

T. T. Thuy (&) � S. Hoste � G. G. Herman � K. De Buysser �P. Lommens � J. Feys � D. Vandeput � I. Van Driessche

Department of Inorganic and Physical Chemistry, Ghent

University, Krijgslaan 281-S3, 9000 Ghent, Belgium

e-mail: [email protected]

123

J Sol-Gel Sci Technol (2009) 52:124–133

DOI 10.1007/s10971-009-1987-1

Page 2

We have selected and studied the combination of two

candidate materials (TEA and acetic acid) in order to

establish the preparation of a stable precursor solution

which can be used for dip coating of superconducting

layers. Furthermore, simulation programs were employed

to build up models expressing the influence of complexa-

tion behavior in the formation of transparent and homog-

enous sols and gels. From this stable precursor solution,

YBCO thin films on STO were synthesized. Texture

analysis of the YBCO layers was performed using XRD

scanning and SEM imaging. The voltage measurements

were applied to determine the critical transition tempera-

ture of superconductor thin films.

2 Experimental

2.1 Chemicals

Yttrium, barium, and copper nitrates, and acetates, KNO3,

HNO3, TEA and EDTA were purchased from Sigma-

Aldrich (Germany). Ammonia (25 wt%) and glacial acetic

acid (99 wt%) were obtained from Chem-Lab (Belgium).

The standard KOH solution was obtained from VWR

(France).

Thermo Gravimetric Analysis/Differential Thermal

Analysis (TGA/DTA) was performed on the starting Y, Ba

and Cu acetates to determine the water content of the metal

acetates in order to assure a 1: 2: 3 stoichiometry of Y3?,

Ba2? and Cu2? in the precursor solutions.

The metal ion stock solutions using for potentiometric

titrations were prepared using metal nitrates. Acid–base

titrations were used to standardize the metal ion stock

solutions by the method of back titration. A known excess

of ethylenediaminetetraacetic acid (EDTA) was added to

the analyte metal nitrate. The excess EDTA was titrated

with a standard potassium hydroxide solution. All final

solutions for the potentiometric experiments have an ionic

strength of 0.1 M (KNO3).

To determine the concentration of the triethanolamine

(TEA) solution, an excess of nitric acid was added which

was then titrated with KOH.

2.2 Potentiometric measurements

The potentiometric measurements were performed using a

Schott pH meter and a 5 mL Schott T-burette. The pH

meter was connected to a Schott H2680 glass electrode and

a Schott B3410 calomel electrode with a second salt bridge

filled with 0.1 M KNO3. Each aqueous system under

consideration was titrated with a standard carbonate free

KOH solution in a 100 mL jacketed cell thermostated at

25 �C ± 0.1 �C by a circulating water bath thermostat. All

systems were studied under protected conditions, estab-

lished by a stream of saturated nitrogen obtained by bub-

bling inert gas though a 0.1 M KNO3 solution [9].

The data were processed using Gran’s method [10] in

order to determine the cell potential E�, the Nernstian

slope s, the dissociation constant of water (Kw) together

with the correction terms for changes in the liquid junc-

tion potential for non-linear electrode response in strong

acid medium, aj (pH \ 2.3) and in strong alkaline med-

ium, bj (pH [ 11.0). This method is also needed in order

to calculate the carbonate content of the strong base

which can interfere in the calculation of the protonation

and stability constants.

2.3 Preparation of YBCO precursor solutions

and interpretation of EQUIL data

The scheme to make YBCO precursor solutions is indi-

cated in Fig. 1. By adding a high concentration of TEA

(1.7 mol TEA per mol metal ions) and a low amount of

acetic acid (8% vol.), a clear precursor solution can be

obtained (Fig. 2). The solutions could be stored at room

temperature for several weeks without loosing stability.

The models of the EQUIL program [11] were built on

the stability constant data taken from the literature [12–14].

The detailed models are shown in Tables 1 and 2.

Y, Ba, Cu – acetates (aq)

TEA (1.7 mol TEA per mol metal ions)

Clear dark blue 0.6M solution pH = 6.5, η = 3.126 cP

Clear dark blue gel

Reflux 90°C, 1h

Glacial acetic acid(8 vol% aq)

Fig. 1 The schematic of aqueous YBCO precursor solution synthesis

J Sol-Gel Sci Technol (2009) 52:124–133 125

123

Page 3

2.4 Other techniques

The infrared spectra were recorded with a Bruker Equinox

55 instrument in the spectral range between 4,000 and

400 cm-1, using the KBr pellet technique. The thermal

decomposition behavior of the gel network was investi-

gated separately by TGA–TGA measurements (STD 2960

Simultaneous DSC-TGA) on bulk samples with the same

composition as dip coated layers. Identification of different

phases present in the YBCO films was performed by X-ray

diffraction (Siemens D5000, CuKa). The overall mor-

phology of the films was characterized by SEM (Philip

501). The critical temperature of the superconductive lay-

ers was determined by voltage measurements (National

instruments Labview 8.0).

3 Results and discussion

3.1 Characterization of precursor solutions

To obtain an YBCO precursor solution, a suitable ligand

should be used in order to complex with yttrium, barium

and copper ions. Soluble complexes are formed, so the

concentration of free metal ions will be reduced and a clear

precursor solution can be obtained. The question is how to

choose a suitable ligand for the solution which has three

metal ions having different polarizing characters (cationic

change/ionic radius). The very weak polarizing character of

barium makes it very difficult to precipitate with hydrox-

ides [15]. The solubility of barium hydroxide in water is

5.6/100 g. It is an advantage to make an YBCO precursor

solution because the coordination of hard barium ions with

all ligands is very weak. However, we have noticed that the

behavior of copper and yttrium aqueous solutions are quite

similar and decided to focus on these two elements in order

to obtain homogeneous sols and gels.

In this study, TEA and acetic acid were chosen as

complexing agents. TEA is both a tertiary amine and a tri-

alcohol. The potentially tetradentate TEA forms stable

complexes with Cu2? [11, 12] using its nitrogen and neu-

tral oxygen donors [16]. Acetic acid is used to complex

with the harder metal ions Y3? and Ba2? as well as Cu2?

[16]. Acetic acid was present with a ratio higher than 1 in

comparison with metal ions. It leads to the equilibrium:

M(CH3COO)n � Mn? ? nCH3COO- shifting to the left

side at higher acetate—concentrations thereby lowering the

Fig. 2 Picture of clear and homogeneous sol and gel

Table 1 The stability constant data of Y3?, Ba2? with acetic acid

and TEA taken from [12, 13] used as model for the EQUIL program

logb Stoichiometry

Y3? HOAc TEA H

7.80 0 0 1 1

4.56 0 1 0 1

1.68 1 1 0 0

3.17 1 2 0 0

-22.00 1 0 0 -3

-13.78 0 0 0 -1

Ba2? HOAc TEA H

7.80 0 0 1 1

4.56 0 1 0 1

0.39 1 1 0 0

-2.30 1 0 0 -2

-13.78 0 0 0 -1

The compositions of species are given with stoichiometric coeffi-

cients. A negative coefficient for H means bonded OH-

Table 2 The stability constant data of Cu2? with acetic acid and

TEA taken from [12–14] used as model for the EQUIL program

logb Stoichiometry

Cu2? HOAc TEA H

Model 1: copper TEA hydroxo complexes

4.03 1 0 1 0

4.56 0 1 0 1

1.83 1 1 0 0

3.09 1 2 0 0

7.80 0 0 1 1

-19.32 1 0 0 -2

-13.78 0 0 0 -1

Model 2: model 1 ? copper TEA mono and dihydroxo complexes

-1.90 1 0 1 -1

-9.70 1 0 1 -2

Model 3: model 2 ? dimer copper TEA hydroxo complexes

-1.10 2 0 2 -2

-8.20 2 0 2 -3

-16.20 2 0 2 -4

126 J Sol-Gel Sci Technol (2009) 52:124–133

123

Page 4

free concentration of Mn? according to Le Chatelier’s

principle. Mono dentate metal acetates are retained in the

precursor solutions together with [CuTEA]2?. An YBCO

precursor solution with only acetic acid present as a com-

plexant should be adjusted to pH = 6.0 in order to achieve

homogeneous gels without phase separation [4]. However,

this solution is only stable in a narrow pH region. During

the gelation there is a shift in pH which results in precip-

itation. Therefore, the ideal precursor solutions should

contain a second stronger ligand such as TEA. TEA will

only form complexes with Cu2? and the acetic acid will

avoid precipitation of Y3?.

Titrations of solutions containing yttrium ions, acetic

acid and TEA were performed in the stoichiometric ratio

Y3?: acetic acid: TEA of 1:3:3. The experimental pH curves

were fitted to the calculated ones using the stability constant

data from literature. Figure 3 shows the fit for a ratio of Y3?:

acetic acid: TEA of 0:3:3 and 1:3:3. As indicated in the

experimental curve, the first inflection point at a = 0 mat-

ches the end of the neutralization of the excess HNO3 added

to the ligand solution, where a is the ratio between mol of

base added and mol of ligand. The second inflection point at

a = 1 corresponds to the completion of neutralization of

acetic acid. The third one at a = 2 matches exactly with the

full deprotonation of TEA. The quality of fit was excellent

from very acidic media to pH about 7 where acetic acid was

totally deprotonated. At higher pH, Y(OH)3 starts to pre-

cipitate due to the high free concentration of yttrium. This

can not avoid by TEA. It leads to a lowering of the pH in the

solution and the titration could not be continued as shown in

Fig. 3. This was also demonstrated by the simulation curve

7 where Y(OH)3 was included. The quality of fit of simu-

lation curve (7) and the experiment curve (4) was perfect as

can be seen in Fig. 3. It can be inferred from these titrations

that yttrium ions only complex with acetic acid with the

stability constants given in the Table 1.

Titrations of solutions containing copper ions, acetic

acid and TEA were also carried out in different stoichi-

ometric ratios of Cu2?: acetic acid: TEA of 1: 3: m (m = 1,

2 or 3). The experimental pH curves were fitted to the

calculated ones using the stability constant data from the

literature and a selection of candidate complexes as

described in the Table 2. Figure 4 shows the fit for a ratio

of Cu2?: acetic acid: TEA of 1: 3: 1 with different models.

When all kinds of copper TEA hydroxo complexes were

excluded (model 1), the quality of fit was not good from pH

6 onwards. The pH value of the simulation curve was

higher than the one of the experimental curve that indicates

some protonated reactions occurred. Therefore, copper

TEA mono- and dihydroxo complexes were included and

the quality of fit improved significantly (model 2). How-

ever, there was still a small difference between the

experimental and calculated curve at pH [ 5 where acetic

acid has been deprotonated. In the solution, there are some

protonated reactions. Therefore, all kinds of copper TEA

mono, dihydroxo complexes and dimer copper TEA mono,

dihydroxo complexes were taken into account (model 3).

The quality of fit was excellent from very acidic media to

pH about 9 where TEA is fully deprotonated. Nevertheless,

there is a slight but systematic difference between the

experimental and calculated curve from pH value of 9 on.

In this pH range, only the copper dihydroxo TEA com-

plexes [Cu(TEA)(OH)2] and its dimer [Cu2(TEA)2(OH)4]

were the governing species. The pH value of the experi-

mental curve is lower than the pH value of the simulated

curve which means that it still incenses the protonation of

(1) and (2)

(4)

(5)

(7)

(3)

(6)

2

4

6

8

10

12

-1.5 -1 -0.5 0 0.5 1 1.5 2 2.5

a

Hp

0

5

10

15

20

25a

d/H

pd

(1) OAcTEAH-simulation(2) OAcTEAH-experiment(4) YOAcTEAH-experiment(5) YOAcTEAH-simulation(7) YOAcTEAH-simulation when Y(OH)3 added(3) OAcTEAH-a-dpH/da-experiment(6) YOAcTEAH-a-dpH/da

Fig. 3 The simulation and

experimental curves for the

reaction of KOH with HNO3,

acetic acid and ?HTEA with

and without Y3? plotted as pH

versus a (where a = mole of

base added/mole ligand)

J Sol-Gel Sci Technol (2009) 52:124–133 127

123

Page 5

some species. Those species should contain hydrogen. It

might be assumed that a tetramer [17] was existent.

With the ratio of Y3?: Cu2?: acetic acid: TEA of 1: 3:

13.98: 10.2 as in the precursor solution, the distribution of

different species was calculated using the stability constant

data from literature. The result is shown in Fig. 5. One can

see that coordination by acetate starts at about pH 2.5. At

pH 3.5 also TEA starts to coordinate with Cu2? to form

[Cu(TEA)]2?. The maximum amount of this complex is

reached at pH 5.7. From pH value of 6.5 on, the Cu-TEA

hydroxo complexes becomes dominant. The free concen-

tration of Cu2? becomes negligible when pH reaches 6.5.

The complexes between Cu2? and acetic acid decrease

dramatically from pH 5 and drop to zero from pH 6.5 on.

Dimeric species have been found at neutral and basic

pH. It is seen that in the solution dihydroxo complexes

(1) and (2)

(3)

(4)

(5)

(6)

(7)

(8)

2

4

6

8

10

12

-1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5

a

Hp

0

5

10

15

20

25

ad /

Hp

d

(1)-TEAOAc-simulation

(2)-TEAOAc-experiment

(3)-CuTEAOAc-simulation (model 3)

(4)-CuTEAOAc-experiment

(5)-CuTEAOAc-simulation (model 2)

(6)-CuTEAOAc-simulation (model 1)

(7)-a-dpH/da -experiment

(8)-TEAOAc-a-dpH/da-experiment

Fig. 4 The simulation and

experimental curves for the

reaction of KOH with HNO3,

acetic acid and ?HTEA with

and without Cu2? plotted as pH

versus a (where a = mole of

base added/mole ligand)

pH area (5.7 to 6.5) to produce clear, stable and homogeneous

sol and gel

Y(OH)3

0

10

20

30

40

50

60

70

80

90

100

2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5

pH

%

Cu(II)

[Cu(TEA)]2+

[Cu(TEA)(OH)]+

[Cu(TEA)(OH)2]

[Cu2(TEA)2(OH)2]2+

[Cu2(TEA)2(OH)3]+

[Cu2(TEA)2(OH)4]

[Cu(OAc)]+

[Cu(OAc)2]

Y3+

[Y(OAc)]2+

[Y(OAc)2]+

Y(OH)3

[Cu2(TEA)2(OH)4]

Cu2+

Y3+

[Y(OAc)]2+

[Y(OAc)2]+

[Cu(TEA)(OH)2]

[Cu2(TEA)2(OH)3]+

[Cu2(TEA)2(OH)2]2+

[Cu(TEA)(OH)]+

[Cu(TEA)]2+

[Cu(OAc)]+

[Cu(OAc)2]

Fig. 5 Distribution of species which contain Y(III), Cu(II) ion, acetic acid, TEA, at 25 �C, I = 0.1 M (with ratio of Y3?: Cu2?: acetic acid: TEA

in 1: 3: 13.98: 10.2)

128 J Sol-Gel Sci Technol (2009) 52:124–133

123

Page 6

Ta

ble

3W

aven

um

ber

s(c

m-

1)

and

assi

gn

men

tso

fth

ev

ibra

tio

nal

mo

des

of

the

gel

and

hea

ttr

eate

dsa

mp

les

of

pre

curs

or

solu

tio

nb

etw

een

4,0

00

and

50

0cm

-1

Gel

(IR

)2

00

�C3

00

�C4

00

�C5

00

�C6

00

�CY

2O

3B

aOB

aCO

3C

uO

Ass

ign

men

tR

ef.

3,3

40

s,b

r3

,38

0s

3,3

97

3,4

05

3,4

11

3,4

21

––

––

O–

Han

dN

–H

stre

tch

fro

m

(OH

CH

2C

H2) 3

NH

?[1

8–

21]

3,1

47

m2

,97

5w

––

––

––

––

C–

Hst

retc

ho

fC

H3

[18

–2

1]

2,9

29

w2

,92

3w

––

––

––

––

C–

Hst

retc

ho

fC

H2

[19

,2

1]

2,8

96

w–

––

––

––

––

C–

Hst

retc

ho

fC

H3

[19

,2

1]

2,3

58

w,

sh2

,35

8w

,sh

2,3

58

w,

sh2

,35

8w

,sh

2,3

58

w,

sh2

,35

8w

,sh

2,3

58

w,

sh2

,35

8w

,sh

2,3

58

w,

sh2

,35

8w

,sh

CO

2fr

om

atm

osp

her

ean

dfr

om

hea

ttr

eate

dsa

mp

le

[24

]

2,3

23

w,

sh2

,32

3w

,sh

2,3

23

w,

sh2

,32

3w

,sh

2,3

23

w,

sh2

,32

3w

,sh

2,3

23

w,

sh2

,32

3w

,sh

2,3

23

w,

sh2

,32

3w

,sh

––

––

––

–1

,65

2w

––

Ba=

Ost

retc

ho

fB

aO–

–1

,72

7w

1,7

27

w1

,72

7w

1,7

47

w1

,74

7w

––

1,7

52

w–

Ba–

Ost

retc

ho

fB

aCO

3–

1,6

41

w–

––

––

––

––

C=

Oo

fH

OA

c[2

2]

1,5

54

s1

,55

0s

1,5

46

s–

––

––

––

An

tisy

mm

etri

cca

rbo

xy

late

ion

sC

O2-

[19

,2

0]

1,4

84

w–

––

––

––

––

––

1,4

52

vw

––

––

––

––

–dC

H3

[20

]

––

–1

,42

5m

1,4

28

m1

,43

0w

1,4

25

w1

,43

7m

1,4

50

s–

Me–

Ost

retc

h–

1,4

01

s1

,40

9s

1,4

07

s–

––

––

––

Sy

mm

etri

cca

rbo

xy

late

ion

sC

O2-

[19

,2

0]

1,3

38

vw

1,3

40

vw

1,3

38

vw

––

––

––

–dC

H3

[20

]

1,3

21

w–

––

––

––

––

dCH

3?

m(C

–C

O2-

)[2

3]

1,2

90

w–

––

––

––

––

mC–

Ow

ith

d ip(C

–O

H)

[23

]

1,2

51

w1

,23

8w

1,2

34

w–

––

––

––

dCH

3[2

2]

1,2

26

w–

––

––

––

––

––

1,1

91

w–

––

––

––

––

––

1,1

33

w1

11

6w

11

06

w–

––

––

––

mCN

C[2

3]

1,0

91

w–

––

––

––

––

––

1,0

78

w–

––

––

––

––

––

1,0

64

w1

,04

9w

1,0

47

w–

––

––

––

q \(C

H3)

[23

]

––

–1

,05

8w

1,0

56

w1

,05

6w

–1

,06

6w

1,0

61

w–

Ba–

Ost

retc

h–

1,0

27

m1

,02

0w

1,0

16

w–

––

––

––

q k(C

H3)

[20

,2

2]

1,0

04

w–

––

––

––

––

––

91

6s

91

4m

93

5w

––

––

––

–mC

–C

[20

,2

2,

23]

––

–8

58

m8

58

m8

58

m8

66

vw

85

9w

85

5s

–M

e–O

–

84

8v

w9

00

––

––

––

––

mC–

C[2

2]

75

0v

w7

40

vw

76

1v

w–

––

––

––

––

67

5w

,sh

67

1w

,sh

66

9w

,sh

69

0m

69

4m

69

4m

–6

70

w6

93

s6

69

sM

–O

stre

tch

See

tex

t

65

0w

65

1w

65

1w

––

––

––

–dC

OO

-[2

2]

61

5w

61

7w

61

3w

––

––

––

–p

(CH

)o

rp

(CO

O)

[20

]

J Sol-Gel Sci Technol (2009) 52:124–133 129

123

Page 7

[Cu2(TEA)2(OH)4] and [Cu(TEA)(OH)2] are only present

at alkaline medium. As mentioned earlier, Y3? and Ba2?

do not coordinate with TEA. A low free concentration of

Y3? can only be reached at pH above 5.7.

In order to minimize the concentration of three metal

ions Y3?, Ba2? and Cu2?, the optimal pH value of the

precursor solution should be higher than 5.7. To avoid the

yttrium hydroxide the maximum pH should kept lower

than 6.5. Under these pH conditions, the precursor solution

produces clear, stable and homogeneous sols and gels.

3.2 Characterization of YBCO gels

3.2.1 FTIR

The band positions measured for gel and its decomposition

and some reference samples are indicated in Table 3.

These assignments are mainly based on some general ref-

erences [18–21] as well as on some specific spectroscopic

studies of complexes [22, 23]. Some aspects of these

assignments are momentarily commented below.

IR spectra during the heat treatment are shown in Fig. 6.

The band at high frequency (3,340 cm-1) due to the O–H

stretch from water and N–H stretch from (OHCH2CH2)3

NH? became smaller. Bands at 2,365–2,345 cm-1 belong

to carbon dioxide from the atmosphere and from the

decomposition of the gel during heat treatment [24]. The

acetate complexes started to decompose at 200 �C indi-

cated by a change in intensity and to lower wave numbers.

At 400 �C the acetate complexes were decomposed com-

pletely. There was a shift from 1,410 to 1,430 cm-1 during

the heat treatment (200–600 �C) due to the decomposition

of the gel and the formation of M–O bonds. The bands at

1,058 and 858 cm-1 were observed for samples heat

treated at higher temperatures, due to the formation of

BaCO3. The band at 673 cm-1 due to the M–O stretch

gained more intensity during heat treatment as expected.

The other bands due to organic compounds such as CH2,

C–N lowered in intensity and disappeared almost com-

pletely at 400 �C.

3.2.2 DTA–TGA

In order to obtain information about the decomposition

behavior of the gels to adjust the thermal treatment

accordingly, a TGA–DTA analysis was carried out in air

atmosphere. The heating rate was 10 �C/min. Three main

areas can be distinguished in the thermogram shown in

Fig. 7. The first broad endothermic peak can be correlated

to the evaporation and the release of acetic acid and gel

network water just below 200 �C. The exothermal peak at

233 �C coupled to the large loss in mass can be attributedTa

ble

3co

nti

nu

ed

Gel

(IR

)2

00

�C3

00

�C4

00

�C5

00

�C6

00

�CY

2O

3B

aOB

aCO

3C

uO

Ass

ign

men

tR

ef.

56

5w

56

4w

––

––

56

5s

––

–Y

–O

stre

tch

of

Y2O

3–

53

8w

––

52

2w

54

3sh

55

1–

––

––

–

46

6w

46

6w

46

6w

–4

82

w4

80

w4

65

m–

––

Y–

Ost

retc

ho

fY

2O

3–

Ban

din

ten

sity

:s

stro

ng

;m

med

ium

;w

wea

k;

shsh

ou

lder

;vs

ver

yst

ron

g;

vwv

ery

wea

k;

shsh

ou

lder

;b

rb

road

130 J Sol-Gel Sci Technol (2009) 52:124–133

123

Page 8

to an auto combustion reaction due to the presence of

acetate groups. The second exothermal peak at 415 �C

corresponds to the decomposition of the intermediary

products CuO2, Y2O2CO3. The final combustion of the

network mainly involves the release of relatively large

quantity of CO and CO2. After this reaction the remaining

species consist of yttrium, barium and copper oxides.

These metal oxides then convert to the desire YBCO phase

at the sintering temperature of 800 �C.

3.3 Characterization of YBCO thin films

Since the sol–gel process was conducted on STO single

crystals one may expect high degree alignment of the

YBCO coating. The SEM micrograph of YBCO layer

deposited from the precursor solution reveals that the thin

film surface is continuous as well as crack free. The SEM

image suggests that the resulting superconducting layer is

polycrystalline with some degree of misorientation of

468

200°C

1020 615

654

673

901

1049

1340

14101552

3371

300°C

12282355

1732400°C

500°C858

600°C

0

0.3

0.6

0.9

400900140019002400290034003900

Wavenumer, cm-1

ytisnet

nI200°C

300°C

400°C

500°C

600°C

Fig. 6 FTIR spectra of the gel

and heat treated samples

CuOY2O3

BaCO3 BaOcarbondioxide

Y2O2CO3

NOx

CuO2

crystal water

free water and acetic acid

476 °C

415 °C

233 °C

155 °C

116 °C

0

20

40

60

80

100

120

0 100 200 300 400 500 600 700 800 900 1000

Temperature, °C

% ,th

gieW

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

gm /

C ° , ecnereffi

d erutare

pme

T

Weight

Temperature difference

Fig. 7 TGA–DTA spectrum

of YBCO gel

J Sol-Gel Sci Technol (2009) 52:124–133 131

123

Page 9

surface grains. However, the XRD diffraction patterns in

Fig. 9 indicate predominantly (001) peaks, but this may

reflect the flake-like nature of the superconducting grains.

As can be seen in the Fig. 8, a few particles (indicated by

arrows) that have been identified as a copper-rich phase

originated from decomposition process of YBCO at high

temperature [25].

The Tc measurement in liquid nitrogen given in Fig. 10

clearly shows a sharp superconducting transition of YBCO

thin film at 90.95 K with a narrow DT (2 K). The low drop

in resistivity, in Fig. 10, between 280 and 90 K (which

directly can be calculated from the voltage measurements)

may suggest insufficient oxidation of the superconductor.

Alternatively frequent ab plane c-axis inter-grain connec-

tions observed in Fig. 8 may be also responsible for this

effect. The value of critical current density measurement is

still very low about 0.2 MA/cm2. Therefore, a further study

and optimization of superconducting properties of the thin

films will be the subject of the next paper.

4 Conclusions

The influence of complexation behavior in the formation

of transparent and homogenous sols and gels by the

combination of Y, Ba and Cu—acetates, acetic acid and

triethanolamine has been studied and interpreted using

simulated metal-ligands equilibriums with the EQUIL

program. From the fit between experimental data from

potentiometric titration with simulated distributions of a

large set of different complex species, the occurrence of

different species at different pH values could be inferred.

FTIR was applied to study the decomposition of the gel.

Based on these results, the preparation of water—based

precursors for the synthesis of superconducting YBCO

could be optimized.

The synthesized YBCO material showed a supercon-

ducting transition temperature of 90.95 K with narrow DT

(2 K) for the thin films. The further study and optimization

of superconducting properties of the thin films will be the

subject of the next paper.

0

4000

8000

12000

16000

20000

5 15 25 35 45 55

2-Theta-Scale

)sp

C( ni

L

001

002

004

005

007

111

Kβ

STO

STO

Fig. 9 XRD of YBCO thin film

coated on STO substrate

Fig. 8 SEM micrograph of the YBCO layer deposited from the

precursor solution

132 J Sol-Gel Sci Technol (2009) 52:124–133

123

Page 10

Acknowledgments The authors would like to thank Olivier Jans-

sens (Ghent University, Belgium) for XRD and SEM measurements.

References

1. Wu MK, Ashburn JR, Torng CJ et al (1987) Phys Rev Lett

58:908. doi:10.1103/PhysRevLett.58.908

2. Housecroft CE, Constable EC (2006) Chemistry, 3rd edn. Pear-

son Prentice Hall, Harlow

3. Dimos D, Chaudhari P, Mannhart J (1990) Phys Rev B 41:4038.

doi:10.1103/PhysRevB.41.4038

4. Kakihana M (1996) J Sol-gel Sci Tech (Paris) 6:7

5. Pathak LC, Mishra SK (2005) Supercond Sci Technol 18:R67.

doi:10.1088/0953-2048/18/9/R01

6. Bhuiyan MS, Parathaman MK, Salama K (2006) Supercond Sci

Technol 18:R21

7. Brinker CJ, Scherer GW (1990) Sol–gel science: the physics and

chemistry of sol–gel processing. Academic Press, London

8. Thuy TT, Hoste S, Herman GG et al. (2009) J Sol-gel Sci Tech

(in press). doi:10.1007/s10971-009-1949-7. http://www.springer

link.com/content/7778x317487gng54/fulltext.pdf

9. Steenland MWA, Dierck I, Herman GG et al. (1997) J Chem Soc

Dalton Trans 3637

10. Gran G (1952) Part II Analyst 77:661

11. Ting-Po I, Nancollas GH (1972) Anal Chem 44(12):1940. doi:

10.1021/ac60320a007

12. Martell AE, Smith RM (1975) Critical stability constants vol 2.

Plenum Press, New York

13. Smith RM, Martell AE (1977) Critical stability constants vol 3.

Plenum Press, New York

14. Tauler R, Casassas E, Rainer MJA (1985) Inorg Chim Acta

105:165. doi:10.1016/S0020-1693(00)90556-5

15. Martell AE, Hancock RD (1996) Metal complexes in aqueous

solutions. Springer, 253 pp

16. Barboux P, Valente I, Tarascon JM et al (1989) Mater Res Soc

156:189

17. Whitmire KH, Hutchison JC, Gardberg A et al (1999) Inorg Chim

Acta 294:153. doi:10.1016/S0020-1693(99)00274-1

18. Steele D (1971) The interpretation of vibrational spectra. Chap-

man and Hall Ltd., London. doi:10.1016/0022-2860(73)85247-0

19. Dollish FR, Fateley WG, Bentley FF (1974) Charateristic Raman

frequencies of organic compounds. Wiley, New York. doi:

10.1016/0022-2860(74)87027-4

20. Nakamoto K (1997) Infrared and Raman spectra of inorganic and

coordination compounds, part B: applications in coordination,

organometallic and bioinorganic chemistry, 5th edn. John Wiley

& Sons

21. Socrates G (2001) Infrared and Raman characteristic group fre-

quencies: tables and charts, 3rd edn. John Wiley & Sons

22. Quiles F, Burneau A (1998) Vib Spectrosc 16:105. doi:10.1016/

S0924-2031(98)00004-6

23. Baran EJ, Wagnera CC, Torreb MH (2002) J Braz Chem Soc

13:576. doi:10.1590/S0103-50532002000500006

24. Leleckaite A, Kareiva A (2004) Opt Mater 26:123. doi:10.1016/

j.optmat.2003.11.009

25. Lian J, Haibo Y, Donglu S et al (2003) Supercond Sci Technol

16:838. doi:10.1088/0953-2048/16/8/302

0

2

4

6

8

80130180230280

Temperature /K

tlov

orcim,

E

0

10

20

30

K/ tlo v

or cim,

Td/

Ed

T-E

dE/dT

∆T = 90.95 - 88.90 = 2.05K

90.95K

Fig. 10 Tc measurements for YBCO thin film

J Sol-Gel Sci Technol (2009) 52:124–133 133

123