See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/255946201

Thermal Expansion of Anatase and Rutile Between 300 and 575 K Using
Synchrotron Powder X-ray Diffraction

Article  in  Powder Diffraction · December 2007

DOI: 10.1154/1.2790965

CITATIONS READS

66 1,585

3 authors, including:

D. R. Hummer Peter J. Heaney

Southern Illinois University Carbondale Pennsylvania State University
43 PUBLICATIONS   474 CITATIONS    168 PUBLICATIONS   4,689 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

TiO2 nanoparticles View project

Using Visual Exploratory Data Analysis to Facilitate Collaboration and Hypothesis Generation in Cross-Disciplinary Research View project

All content following this page was uploaded by D. R. Hummer on 02 December 2014.

The user has requested enhancement of the downloaded file.

Thermal expansion of anatase and rutile between 300 and 575 K
using synchrotron powder X-ray diffraction
D. R. Hummer and P. J. Heaney
Department of Geosciences, The Pennsylvania State University, University Park, Pennsylvania 16082
J. E. Post
Department of Mineral Sciences, Smithsonian Institution, Washington, DC 20560-0119
共Received 24 May 2007; accepted 7 September 2007兲
High-precision unit-cell parameters for the TiO2 polymorphs anatase and rutile at temperatures
between 300 and 575 K have been determined using Rietveld analysis of synchrotron powder XRD
data. Polynomial models were used to express the tetragonal unit-cell parameters as a function of
absolute temperature, with a 共anatase兲 = 1.759 37⫻ 10−8 ⫻ T2 + 6.418 16⫻ 10−6 ⫻ T + 3.779 84,
c 共anatase兲 = 6.6545⫻ 10−8 ⫻ T2 + 4.0464⫻ 10−5 ⫻ T + 9.4910, V 共anatase兲 = 2.237 58⫻ 10−6 ⫻ T2
+ 1.027 77⫻ 10 ⫻ T + 135.602, a 共rutile兲 = −6.636 42⫻ 10 ⫻ T3 + 1.005 01⫻ 10−7 ⫻ T2 − 1.009 93
−3 −11

⫻10−5 ⫻ T + 4.586 34, c 共rutile兲 = −4.115 50⫻ 10−11 ⫻ T3 + 6.405 94⫻ 10−8 ⫻ T2 + 4.675 61⫻ 10−7T
+ 2.951 81, and V 共rutile兲 = −2.7790⫻ 10−9 ⫻ T3 + 4.2386⫻ 10−6 ⫻ T2 − 3.3551⫻ 10−4 ⫻ T + 62.100.
The polynomial expressions were used to calculate linear 共␣兲 and volume 共␤兲 thermal expansion
coefficients of anatase and rutile between 300 and 575 K. At 298.15 K, these values were ␣a
= 4.46943⫻ 10−6 K−1, ␣c = 8.4283⫻ 10−6 K−1, and ␤ = 17.3542⫻ 10−6 K−1 for anatase, and ␣a
= 6.99953⫻ 10−6 K−1, ␣c = 9.36625⫻ 10−6 K−1, and ␤ = 28.680⫻ 10−6 K−1 for rutile. © 2007
International Centre for Diffraction Data. 关DOI: 10.1154/1.2790965兴

Key words: anatase, rutile, thermal expansion, synchrotron, powder diffraction

I. INTRODUCTION strom range. Thus, high-precision unit-cell parameters for

Titanium oxides are important industrial materials that
find use in pigments, catalysts, dielectrics, and many other
applications. In addition, fine-grained titanium oxides are an
important component of many soils and influence the local
aqueous geochemistry 共Dolcater et al., 1970兲. The three most
commonly occurring titanium oxide minerals are rutile, ana-
tase, and brookite. Of these, rutile is the most naturally abun-
dant as it is thermodynamically stable in its bulk form over
all crustal temperatures and pressures, whereas anatase and
brookite are metastable 共Lindsley, 1976兲. Scientists have fo-
cused on the nucleation and growth of titanium oxides, since
these processes may involve changes in phase stability that
accompany the transition from nanoparticles to bulk phases
共Gribb and Banfield, 1997; Zhang and Banfield, 1998, 2000;
Ranade et al., 2002; Li et al., 2004兲.
Recent interest in applications of titanium oxide
nanophases underscores the need for detailed measurements
of the structures of these phases as a function of temperature.
Nanocrystalline TiO2 particles and thin films have become
important materials for use in dye-sensitized solar cells
共Gratzel, 2004; Richards et al., 2004; Sebastian et al., 2004;
Santa-Nokki et al., 2006兲. The thermal expansion of some
solar cells gives rise to a change in the semiconductor band
gap 共Rincon et al., 1998兲, affecting electron transport and
hence the efficiency of the cell. Thermal expansion also af-
fects other structural, optical, and electronic parameters in
solar cells that are based on TiO2 thin films 共Karunagaran et
al., 2002; Takeshita et al., 2004; Huang et al., 2006;
O’Regan and Durrant, 2006兲. In addition, recent modeling
共Filyukov et al., 2006兲 and experimental 共Swamy et al.,
2006; Hummer et al., 2006兲 investigations of TiO2 nanopar-
ticles have revealed systematic structural changes as a func-
tion of particle size that are in the thousandths of an ang-
TiO2 crystals of macroscopic dimensions provide a neces-
sary baseline reference for the structural changes that attend
both thermal expansion and particle growth.
Cromer and Herrington 共1955兲 used powder camera
measurements to derive the first high-quality unit-cell param-
eters for anatase and rutile at room temperature, and later
Rao et al. 共1970兲 used a similar technique to calculate the
thermal expansion of anatase and rutile from 300 K to
⬃1000 K. Subsequent X-ray diffraction studies have at-
tempted to refine the values for the unit-cell parameters of
anatase and rutile as a function of temperature 共Horn et al.,
1972; Meagher and Lager, 1979; Sugiyama and Takeuchi,
1991; Howard et al., 1991; Jagtap et al., 2005兲. The best
errors in these studies are on the order of a few thousandths
of an angstrom. However, no investigations to date have ex-
ploited the accuracy and precision provided by whole pattern
analysis using synchrotron powder diffraction data, which is
capable of precision better than 0.001 Å.
Here, we report high-precision thermal expansion mea-
surements of the bulk phases for both rutile and anatase us-
ing temperature-resolved synchrotron powder X-ray diffrac-
tion.
II. EXPERIMENTAL METHODS
Anatase TiO2 共Sigma Aldrich, 99.8%兲 and rutile TiO2
共Sigma Aldrich, 99.99%兲 powder standards were each loaded
into 0.7 mm internal diameter quartz glass capillaries
共Charles Supper Co.兲. The average particle size as deter-
mined by SEM analysis was 0.2 ␮m for anatase, and be-
tween 0.5 and 5 ␮m for rutile 共with most particles being
⬃1 ␮m兲. Powder diffraction patterns were collected with a
MAR345 image plate at beam line X7B, National Synchro-

352 Powder Diffraction 22 共4兲, December 2007 0885-7156/2007/22共4兲/352/6/$23.00 © 2007 JCPDS-ICDD 352
TABLE I. Summary of Rietveld refinement parameters for anatase and
rutile between 300 and 575 K.

Anatase Rutile

Space group I41 / amd P42 / mnm

Total no. of refinements 11 15
No. of observations 1600–1650 1600–1700
No. of reflections 32–39 34–38
Diffraction range 共2␪兲 13– 46 deg 13– 47 deg
No. of variables 36–46 30–39
R共F2兲 0.026–0.044 0.029–0.038
Rwp ⬃0.037 0.040–0.046
␹2 3.1–4.2 4.3–6.7

tron Light Source 共NSLS兲, Brookhaven National Laboratory

共BNL兲. Samples were heated from room temperature at a
linear heating rate of 3.00 and 2.25 K / min for anatase and
rutile, respectively. During heating, exposures were taken for
a duration of 60 s at a regular interval of 157 s, and the
sample was rotated through an angle of 60 deg during each
exposure. Because a non-negligible amount of heating oc-
curred during a single exposure, the nominal temperature for
each pattern was assigned as the temperature at the midpoint
of a given exposure.
The X-ray wavelength was calibrated using silicon, co-
rundum, and lanthanum hexaboride NIST standards, and it
refined as 0.92194± 0.00027 Å 共N = 5兲. Temperature was
controlled with an Omega 3200 controller attached to a
forced air heater and monitored with a chromel-alumel ther-
mocouple adjacent to the sample. Known phase transitions of
rubidium nitrate 共RbNO3兲 and chalcocite 共Cu2S兲 between
373 and 583 K were used for temperature calibration. The
calculated calibration curve revealed that the measured tem-
perature was accurate to within an uncertainty of about 3 K
across the temperature range under study. Images were inte-
grated into intensity versus 2␪ plots with the program Fit2D Figure 1. Observed 共X’s兲, calculated 共solid line兲, and difference 共line below
pattern兲 intensities for typical Rietveld refinements of anatase and rutile: 共a兲
共Hammersley et al., 1996兲 using a polarization factor of 0.93. anatase with rutile impurity at 165 ° C and 共b兲 rutile with anatase impurity at
All structural refinements were carried out using the 131 ° C 共␭ = 0.92194± 0.00027 Å for both patterns兲.
General Structure Analysis System 共GSAS兲 developed by

TABLE II. Refined unit-cell parameters for anatase and rutile as a function of temperature.

Anatase Rutile

T 共°C兲 a 共Å兲 c 共Å兲 V 共Å3兲 T 共°C兲 a 共Å兲 c 共Å兲 V 共Å3兲

47共3兲 3.783 72共5兲 9.5108共1兲 136.162共5兲 43共3兲 4.591 06共7兲 2.957 04共5兲 62.328共2兲
71共3兲 3.784 10共5兲 9.5128共1兲 136.217共4兲 61共3兲 4.591 69共7兲 2.957 57共5兲 62.356共3兲
94共3兲 3.784 57共5兲 9.5149共1兲 136.282共5兲 78共3兲 4.592 38共7兲 2.958 14共5兲 62.387共3兲
118共3兲 3.785 06共5兲 9.5170共1兲 136.348共4兲 96共3兲 4.592 99共7兲 2.958 66共4兲 62.415共3兲
141共3兲 3.785 53共5兲 9.5191共1兲 136.411共4兲 113共3兲 4.593 64共7兲 2.959 19共4兲 62.443共2兲
165共3兲 3.786 02共5兲 9.5215共1兲 136.481共4兲 131共3兲 4.594 23共7兲 2.959 70共5兲 62.470共2兲
188共3兲 3.786 56共5兲 9.5239共1兲 136.554共4兲 149共3兲 4.595 00共7兲 2.960 33共4兲 62.504共2兲
212共3兲 3.787 07共5兲 9.5262共1兲 136.624共4兲 166共3兲 4.595 68共7兲 2.960 89共4兲 62.535共2兲
235共3兲 3.787 69共5兲 9.5288共1兲 136.706共5兲 184共3兲 4.596 38共7兲 2.961 48共4兲 62.566共2兲
259共3兲 3.788 23共5兲 9.5313共1兲 136.781共5兲 202共3兲 4.597 10共7兲 2.962 08共4兲 62.599共2兲
282共3兲 3.788 84共5兲 9.5340共1兲 136.864共4兲 219共3兲 4.597 82共7兲 2.962 66共4兲 62.631共2兲
237共3兲 4.598 52共6兲 2.963 25共4兲 62.662共2兲
255共3兲 4.599 29共7兲 2.963 88共4兲 62.696共2兲
272共3兲 4.600 00共7兲 2.964 46共4兲 62.728共2兲
290共3兲 4.600 63共6兲 2.965 02共4兲 62.757共2兲

353 Powder Diffr., Vol. 22, No. 4, December 2007 Thermal expansion of anatase and rutile ... 353
Larson and Von Dreele 共2000兲. Because the rutile standard
contained 5.2± 0.2% anatase and the anatase standard con-
tained 2.1± 0.2% rutile as phase impurities 共according to
Rietveld refinement results兲, both phases were modeled in all
refinements. Backgrounds were fit graphically using 12 to 18
terms of a Chebyschev polynomial, and peak profiles were
modeled using a pseudo-Voigt function described by Van
Laar and Yelon 共1984兲, adjusted for asymmetry attributable
to axial divergence by Finger et al. 共1994兲, and including
terms for anisotropic microstrain broadening by Stephens
共1999兲. After convergence of peak and background param-
eters, unit-cell dimensions were refined. Initial values of
unit-cell parameters were taken from Cromer and Herrington
共1955兲. This was followed by refinement of atomic positions,
isotropic temperature factors, and in some cases additional
background terms. Final Rietveld refinement parameters are
summarized in Table I. Typical refinements for each standard
are illustrated in Figure 1.
III. RESULTS AND DISCUSSION
Refinement results are shown in Tables II and III. All
refined unit-cell parameters show excellent linear correlation
with temperature 共r2 ⱖ 0.997兲 between 300 and 575 K.
Quoted uncertainties are the standard deviations calculated
by GSAS, which are known to underestimate random error.
We report these uncertainties with the understanding that the
actual uncertainties can be more than an order of magnitude
higher than the calculated values. Results are consistent with
Figure 2. Refined unit-cell parameters for anatase and rutile versus temperature: 共a兲 a 共anatase兲, 共b兲 c 共anatase兲, 共c兲 a 共rutile兲, and 共d兲 c 共rutile兲. Results from
selected authors are shown for comparison. Linear regressions are shown for data from this work and from Rao et al. 共1970兲. Error bars for this work are ten
times the uncertainty calculated by GSAS; error bars from other authors are taken from the quoted uncertainties.
354 Powder Diffr., Vol. 22, No. 4, December 2007
TABLE III. Refined bond distances for anatase at 47 ° C and rutile at 43 ° C.
Anatase Rutile
Ti-Ti 共Å兲 3.039 54共0兲 3.036 49共19兲
Ti-O共1兲 共Å兲 1.965 59共0兲 2.013共10兲
Ti-O共2兲 共Å兲 1.937 02共0兲 1.9254共19兲
previous determinations 共Figure 2兲, with the exception of
those of Jagtap et al. 共2005兲. These discrepancies most likely
result from their use of synthetic nanoparticles, in which sur-
face strain can significantly perturb the bulk structure.
Although all data are reasonably well fit to a linear
model, careful inspection reveals that some unit-cell param-
eters show a curvature that is not explained by a linear re-
gression 共Figure 3兲. As an example, a significantly better fit
is obtained for the anatase a parameter using a quadratic
model 共r2 = 0.999 88兲 rather than a linear one 共r2 = 0.997 09兲.
However, this improvement may only reflect the increase in
the number of model parameters. To determine whether the
better goodness-of-fit is real or is a result of the added model
parameters, we performed a series of f-tests for each data set
using linear, quadratic, and cubic polynomial models. The
f-tests revealed that at an ␣ = 0.10 level of significance, a
quadratic model is the simplest model adequate to describe
the anatase unit-cell parameters, and a cubic model is most
appropriate for rutile parameters. This treatment yields the
following regression equations:
Hummer, Heaney, and Post 354
 TABLE IV. Calculated linear and volume thermal expansion coefficients for
anatase and rutile at 298.15 K.

This work Rao et al. 共1970兲a

Anatase
␣a / 10−6 K−1 4.46943 3.676
␣c / 10−6 K−1 8.4283 7.557
␤ / 10−6 K−1 17.3542 ¯
Rutile
␣a / 10−6 K−1 6.99953 7.305
␣c / 10−6 K−1 9.36625 8.911
␤ / 10−6 K−1 28.680 ¯
a
Defines thermal expansion coefficients to be either parallel 共储兲 or perpen-
dicular 共⬜兲 to the principal axis, thus ␣储 = ␣c and ␣⬜ = ␣a.

a 共rutile兲 = − 6.636 42 ⫻ 10−11 ⫻ T3 + 1.005 01 ⫻ 10−7 ⫻ T2

− 1.009 93 ⫻ 10−5 ⫻ T + 4.586 34, 共1d兲

c 共rutile兲 = − 4.115 50 ⫻ 10−11 ⫻ T3 + 6.405 94 ⫻ 10−8 ⫻ T2

+ 4.675 61 ⫻ 10−7T + 2.951 81, 共1e兲

V 共rutile兲 = − 2.7790 ⫻ 10−9 ⫻ T3 + 4.2386 ⫻ 10−6 ⫻ T2

Figure 3. 共a兲 Plot of a 共anatase兲 versus temperature from Rietveld refine-
ment results 共diamonds兲; a linear regression is shown to make the slight
curvature of the data more visible. 共b兲 Plot of residuals from the linear
regression of a 共anatase兲 versus temperature 共dots兲, and residuals from the
quadratic regression 共triangles兲.
− 3.3551 ⫻ 10−4 ⫻ T + 62.100, 共1f兲
where a 共X兲 is the a parameter of phase X in angstroms,
c 共X兲 is the c parameter of phase X in angstroms, V 共X兲 is the
unit-cell volume of phase X in cubic angstroms, and T is the
absolute temperature.
To obtain thermal expansion coefficients, the calculated
expressions for each unit-cell parameter 关Eqs. 共1a兲–共1f兲兴
were substituted into the definitions for linear and volume
thermal expansion coefficients:

a 共anatase兲 = 1.759 37 ⫻ 10−8 ⫻ T2 + 6.418 16 ⫻ 10−6 ⫻ T ␣L = 冉 冊

1 ⳵L
L ⳵T
, 共2兲

+ 3.779 84, 共1a兲

c 共anatase兲 = 6.6545 ⫻ 10−8 ⫻ T2 + 4.0464 ⫻ 10−5 ⫻ T ␤= 冉 冊

1 ⳵V
V ⳵T
, 共3兲

+ 9.4910, 共1b兲
where L is any linear unit-cell dimension, ␣L is the linear
thermal expansion coefficient parallel to L, and ␤ is the vol-
V 共anatase兲 = 2.237 58 ⫻ 10 ⫻ T + 1.027 77 ⫻ 10 ⫻ T
−6 2 −3
ume thermal expansion coefficient. For the anatase a param-
+ 135.602, 共1c兲 eter, we obtain

冉 冊
共1.759 37 ⫻ 10−8 ⫻ T2 + 6.418 16 ⫻ 10−6 ⫻ T + 3.779 84兲
1 ⳵a ⳵T
␣a 共anatase兲 = =
a ⳵T 1.759 37 ⫻ 10−8 ⫻ T2 + 6.418 16 ⫻ 10−6 ⫻ T + 3.779 84
3.518 74 ⫻ 10−8 ⫻ T + 6.418 16 ⫻ 10−6
= . 共4a兲
1.759 37 ⫻ 10−8 ⫻ T2 + 6.418 16 ⫻ 10−6 ⫻ T + 3.779 84
Following an analogous derivation using Eqs. 共1b兲–共1f兲, we arrive at expressions for the thermal expansion coefficients of all
unit-cell parameters 关Eqs. 共2兲 and 共3兲兴:
1.3309 ⫻ 10−7 ⫻ T + 4.0464 ⫻ 10−5
␣c 共anatase兲 = , 共4b兲
6.6545 ⫻ 10−8 ⫻ T2 + 4.0464 ⫻ 10−5 ⫻ T + 9.4910

355 Powder Diffr., Vol. 22, No. 4, December 2007 Thermal expansion of anatase and rutile ... 355
 4.475 16 ⫻ 10−6 ⫻ T + 1.027 77 ⫻ 10−3
␤ 共anatase兲 = , 共4c兲
2.237 58 ⫻ 10−6 ⫻ T2 + 1.027 77 ⫻ 10−3 ⫻ T + 135.602

− 1.990 93 ⫻ 10−10 ⫻ T2 + 2.010 02 ⫻ 10−7 ⫻ T − 1.009 93 ⫻ 10−5

␣a 共rutile兲 = , 共4d兲
− 6.636 42 ⫻ 10−11 ⫻ T3 + 1.005 01 ⫻ 10−7 ⫻ T2 − 1.009 93 ⫻ 10−5 ⫻ T + 4.586 34

1.234 56 ⫻ 10−10 ⫻ T2 + 1.281 19 ⫻ 10−7 ⫻ T + 4.675 61 ⫻ 10−7

␣c 共rutile兲 = , 共4e兲
− 4.115 50 ⫻ 10−11 ⫻ T3 + 6.405 94 ⫻ 10−8 ⫻ T2 + 4.675 61 ⫻ 10−7 ⫻ T + 2.951 81

8.3370 ⫻ 10−9 ⫻ T2 + 8.4772 ⫻ 10−6 ⫻ T + 3.3551 ⫻ 10−4

␤ 共rutile兲 = . 共4f兲
− 2.7790 ⫻ 10−9 ⫻ T3 + 4.2386 ⫻ 10−6 ⫻ T2 + 3.3551 ⫻ 10−4 ⫻ T + 62.100

These expressions are recommended only in the temperature
range of ⬃300 to 575 K over which diffraction data were
collected. Table IV presents room temperature 共T
= 298.15 K兲 thermal expansion coefficients for anatase and
rutile calculated using Eqs. 共4a兲–共4f兲.
IV. CONCLUSIONS
The unit-cell parameters of anatase and rutile between
300 and 575 K have been determined using temperature-
resolved synchrotron XRD in conjunction with Rietveld
analysis to a precision of better than 0.001 Å. Anatase pa-
rameters were successfully fit by a quadratic model, while
rutile parameters were fit by a cubic model. These models
were used to calculate expressions for the linear and volume
thermal expansion coefficients of anatase and rutile between
300 and 575 K. The high precision and very low scatter of-
fered in this work is a significant improvement over previous
determinations by an order of magnitude or more, allowing
higher quality analyses of titania nanoparticles and thin
films.
ACKNOWLEDGMENTS
The authors would like to thank Jon Hanson at NSLS for
his assistance with instrumentation at beam line X7B. This
work was made possible by National Science Foundation
Grant No. EAR04-17714 and by the Center for Environmen-
tal Kinetics Analysis 共CEKA兲, an NSF- and DOE-sponsored
Environmental Molecular Science Institute 共NSF CHE-
0431328兲.
Cromer, D. T. and Herrington, K. 共1955兲. “The structures of anatase and
rutile,” J. Am. Chem. Soc. 77, 4708–4709.
Dolcater, D. L., Syers, J. K., and Jackson, M. L. 共1970兲. “Titanium as free
oxide and substituted forms in kaolinites and other soil minerals,” Clays
Clay Miner. 18, 71–79.
Filyukov, D. V., Brodskaya, E. N., Piotrovskaya, E. M., and Leeuw, S. W.
共2006兲. “Molecular-dynamics simulation of nanoclusters of crystal
modifications of titanium dioxide,” Russ. J. Gen. Chem. 77, 10–16.
Finger, L. W., Cox, D. E., and Jephcoat, A. P. 共1994兲. “A correction for
powder diffraction peak asymmetry due to axial divergence,” J. Appl.
Crystallogr. 27, 892–900.
Gratzel, M. 共2004兲. “Conversion of sunlight to electric power by nanocrys-
talline dye-sensitized solar cells,” J. Photochem. Photobiol., A 164,
3–14.
Gribb, A. A. and Banfield, J. F. 共1997兲. “Particle size effects on transforma-
tion kinetics and phase stability in nanocrystalline TiO2,” Am. Mineral.
82, 717–728.
Hammersley, A. P., Svensson, S. O., Hanfland, M., Fitch, A. N., and Hauser-
mann, D. 共1996兲. “Titanium two-dimensional detector software: From
real detector to idealised image or two-theta scan,” High Press. Res. 14,
235–248.
Horn, M., Schwerdtfeger, C. F., and Meagher, E. P. 共1972兲. “Refinement of
the structure of anatase at several temperatures,” Z. Kristallogr. 136,
273–281.
Howard, C. J., Sabine, T. M., and Dickson, F. 共1991兲. “Structural and ther-
mal parameters for rutile and anatase,” Acta Crystallogr., Sect. B: Struct.
Sci. 47, 462–468.
Huang, C., Hsu, Y., Chen, J., Suryanarayanan, V., Lee, K., and Ho, K.
共2006兲. “The effects of hydrothermal temperature and thickness of TiO2
film on the performance of a dye-sensitized solar cell,” Sol. Energy
Mater. Sol. Cells 90, 2391–2397.
Hummer, D. R., Heaney, P. J., and Post, J. E. 共2006兲. “Aqueous nucleation
and growth of titanium oxides using time-resolved synchrotron X-ray
diffraction,” Abstracts of the American Geophysical Union, Joint Assem-
bly, Session V06.
Jagtap, N., Bhagwat, M., Awati, P., and Ramaswamy, V. 共2005兲. “Charac-
terization of nanocrystalline anatase titania: an in situ HTXRD study,”
Thermochim. Acta 427, 37–41.
Karunagaran, B., Rajendra Kumar, R. T., Mangalaraj, D., Narayandass, Sa.
K., and Mohan Rao, G. 共2002兲. “Influence of thermal annealing on the
composition and structural parameters of DC magnetron sputtered tita-
nium dioxide thin films,” Cryst. Res. Technol. 37, 1285–1292.
Larson, A. C. and Von Dreele, R. B. 共2000兲. General Structure Analysis
System (GSAS), Report LAUR 86-748, Los Alamos National Laboratory,
Los Alamos, NM.
Li, W., Ni, C., Lin, H., Huang, C. P., and Ismat Shah, S. 共2004兲. “Size
dependence of thermal stability of TiO2 nanoparticles,” J. Appl. Phys.
96, 6663–6668.
Lindsley, D. H. 共1976兲. “Experimental studies of oxide minerals,” Rev. Min-
eral. 3, L-61–L-88.
Meagher, E. P. and Lager, G. A. 共1979兲. “Polyhedral thermal expansion in
the TiO2 polymorphs: refinement of the crystal structures of rutile and
brookite at high temperature,” Can. Mineral. 17, 77–85.
O’Regan, B. C. and Durrant, J. R. 共2006兲. “Calculation of activation ener-
gies for transport and recombination in mesoporous TiO2/dye/electrolyte
films—taking into account surface charge shifts with temperature,” J.
Phys. Chem. B 110, 8544–8547.
Ranade, M. R., Navrotsky, A., Zhang, H. Z., Banfield, J. F., Elder, S. H.,
Zaban, A., Borse, P. H., Kulkarni, S. K., Doran, G. S., and Whitfield, H.
J. 共2002兲. “Energetics of nanocrystalline TiO2,” Proc. Natl. Acad. Sci.
U.S.A. 99, 6476–6481.
Rao, K. V. K., Naidu, S. V. N., and Iyengar, L. 共1970兲. “Thermal expansion
of rutile and anatase,” J. Am. Ceram. Soc. 53, 124–126.
Richards, B. S., Richards, S. R., Boreland, M. B., and Jamieson, D. N.
共2004兲. “High temperature processing of TiO2 thin films for application
in silicon solar cells,” J. Vac. Sci. Technol. A 22, 339–348.
Rincon, C., Wasim, S. M., and Marin, G. 共1998兲. “Effect of thermal expan-
sion coefficient on the temperature dependence of the band gap in
CuInTe2,” Mater. Lett. 36, 245–248.

356 Powder Diffr., Vol. 22, No. 4, December 2007 Hummer, Heaney, and Post 356
Santa-Nokki, H., Kallioinen, J., Kololuoma, T., Tuboltsev, V., and Korppi-
Tommola, J. 共2006兲. “Dynamic preparation of TiO2 films for fabrication
of dye-sensitized solar cells,” J. Photochem. Photobiol., A 182, 187–
191.
Sebastian, P. J., Olea, A., Campos, J., Toledo, J. A., and Gamboa, S. A.
共2004兲. “Temperature dependence and the oscillatory behavior of the
opto-electronic properties of a dye-sensitized nanocrystalline TiO2 solar
cell,” Sol. Energy Mater. Sol. Cells 81, 349–361.
Stephens, P. W. 共1999兲. “Phenomenological model of anisotropic peak
broadening in powder diffraction,” J. Appl. Crystallogr. 32, 281–289.
Sugiyama, K. and Takeuchi, Y. 共1991兲. “The crystal structure of rutile as a
function of temperature up to 1600° C,” Z. Kristallogr. 194, 305–313.
Swamy, V., Menzies, D., Muddle, B. C., Kuznetsov, A., Dubrovinsky, L. S.,
Dai, Q., and Dmitriev, V. 共2006兲. “Nonlinear size dependence of anatase
357 Powder Diffr., Vol. 22, No. 4, December 2007
View publication stats
TiO2 lattice parameters,” Appl. Phys. Lett. 88, 243103-1–243103-3.
Takeshita, K., Sasaki, Y., Kobashi, M., Tanaka, Y., Maeda, S., Yamakata, A.,
Ishibashi, T., and Onishi, H. 共2004兲. “Effect of annealing temperature on
back electron transfer and distribution of deep trap sites in dye-
sensitized TiO2, studied by time-resolved infrared spectroscopy,” J.
Phys. Chem. B 108, 2963–2969.
Van Laar, B. and Yelon, W. B. 共1984兲. “The peak in neutron powder dif-
fraction,” J. Appl. Crystallogr. 17, 47–54.
Zhang, H. and Banfield, J. F. 共1998兲. “Thermodynamic analysis of phase
stability of nanocrystalline titania,” J. Mater. Chem. 8, 2073–2076.
Zhang, H. and Banfield, J. F. 共2000兲. “Understanding polymorphic phase
transformation behavior during growth of nanocrystalline aggregates:
Insights from TiO2,” J. Phys. Chem. B 104, 3481–3487.
Thermal expansion of anatase and rutile ... 357