Microwave heat treatment of natural ruby (PDF)

Appl. Phys. A (2016) 122:224
DOI 10.1007/s00339-016-9703-9

Microwave heat treatment of natural ruby
and its characterization
S. Swain1 • S. K. Pradhan1 • M. Jeevitha1 • P. Acharya1 • M. Debata1
T. Dash1 • B. B. Nayak1 • B. K. Mishra1

•

Received: 30 October 2015 / Accepted: 4 February 2016 / Published online: 1 March 2016
Ó Springer-Verlag Berlin Heidelberg 2016

Abstract Natural ruby (in the form of gemstone) collected from Odisha has been heat-treated by microwave
(MW). A 3-kW industrial MW furnace with SiC susceptors was used for the heat treatment. The ruby samples
showed noticeable improvements (qualitative), may be
attributed to account for the improvement in clarity and
lustre. Optical absorption in 200–800 nm range and photoluminescence peak at 693 nm (with 400 nm kex) clearly
show that subtle changes do take place in the ruby after
the heat treatment. Further, inorganic compound phases
and valence states of elements (impurities) in the ruby
were studied by X-ray diffraction, micro-Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS).
The valence states of the main impurities such as Cr, Fe,
and Ti, in the untreated and MW heat-treated ruby, as
revealed from XPS, have been discussed in depth. The
overall results demonstrate for the first time the effect of
fast heating like MW on the microstructural properties of
the gemstone and various oxidation states of impurity
elements in the natural ruby.

1 Introduction
A gemstone is a mineral which in cut and polished form is
utilized to make jewellery or other ornaments. Some
minerals are used as gemstones in jewellery because of
their attractive colour, lustre, and other physical properties
that have ornamental value. Besides aesthetic use,
& S. K. Pradhan
skpradhan@immt.res.in
1

CSIR- Institute of Minerals and Materials Technology,
Bhubaneswar 751013, India

gemstones also find several applications in industries. Ruby
is one of the four most precious gemstones along with
diamond, sapphire, and emerald. Also, it finds application
in the area of high-power switch and sensors [1, 2]. Redcoloured corundum is known as ruby with the structural
formula of Al2O3:Cr in which a small amount of Al3? is
replaced by Cr3? ions [3]. Most of the natural rubies have
non-uniform colour saturation and sometimes with blue
patches. To overcome these issues, various processing
methods have been developed. The methods include heat
treatment in different chemical environments, high-energy
(c, a) particle or electron beam irradiation, surface coating,
and laser treatment [4].
Heat treatment is one of the earliest methods for the
improvement of colour saturation, clarity, and trading
value of the gemstones. Some groups reported the removal
of the dark core and blue patches by heat treatment under
reducing or oxidizing atmosphere [5, 6]; also, through lead
glass filling method, colour and clarity of ruby stones could
be improved [7].
The use of MW heating for different materials, such as
ceramics, metals, and composites, provides several benefits
over conventional heating. The benefits include faster and
uniform heating, energy-saving, finer microstructures,
rapid product development, and eco-friendly process [8].
MW is electromagnetic wave falling in the frequency range
0.3–300 GHz [9]. It was first used for radar in Second
World War, and by 1950, its ability to heat materials was
realized. Until 2000, MW heating was mainly limited to
ceramics, semimetals, polymeric, and inorganic materials,
and in recent years, it has been successfully applied to
metallic materials in powder form [10, 11].
In MW heating, material is heated faster than that in
conventional heating [12]. The heating involves microwave to couple with material that absorbs electromagnetic

123

224 Page 2 of 7

S. Swain et al.

energy volumetrically and convert it to heat. The ability
of any material to absorb microwave depends on its
complex dielectric constant (e*), which consists of two
parts [13];
e ¼ e0 þ ie00

ð1Þ

where e0 = relative permittivity and e00 = dielectric loss.
The imaginary part e00 is given as
e00 ¼ r=e0 f

ð2Þ

where r = electrical conductivity and f = frequency.
From dielectric properties, the amount of power absorbed by the material in a MW field with field strength E can
be calculated by the following formula:
P ¼ rjEj2 ¼ 2Pf e0 e0 jEj2 ¼¼ 2Pf e0 e0 tan djEj2
00

0

ð3Þ

where tand (loss tangent) = e /e [14]. As shown in
Eq. (3), the heating rate P for a particular material is
decided by f and E, which implies that at lower frequency,
higher microwave power is required to achieve the same
amount of heating.
In conventional heating method, heat is transferred
between materials through the mechanisms of conduction,
radiation, and convection, whereas in MW heating, since
the material itself generates the heat, heating is more volumetric and can be very rapid and selective [8]. For
industry, scientific, and medical applications, microwave
heating is done by using 2.45-GHz and 915-MHz frequencies. There are some problems associated with MW
heating, e.g. thermal instabilities (hot spots) may overheat
the material catastrophically. Non-uniformity in distribution of MW energy can be a reason for development of hot
spots. Also, at certain points, more MW energy is absorbed
at higher temperature leading to further heating, and this
feedback mechanism finally gets out of control resulting in
hot spots. An impure material such as ruby which contains
impurities can contribute to hot spots which further complicate processing methods. To avoid these problems,
researchers have developed technique which unites direct
microwave heating with infrared heat sources. Researchers
have increased the temperature with radiant heat to couple

Fig. 1 Photographs of ruby
samples: as-received stones
(left), sample microwave heattreated at 800 °C (middle) and
at 1500 °C (right)

123

microwave with low-loss material (such as alumina). Also,
heating can be achieved by use of susceptor materials such
as SiC which is a very good MW absorber at low temperature. Once a material reaches its critical temperature
(of alumina), microwave absorption becomes adequate to
cause self-heating. This technique can give more uniform
temperature gradients as the microwave heats volumetrically and the external heat source reduces surface heat
losses [15].
Many researchers have been attempting to heat the
natural ruby or ruby gemstone by conventional heating
method to enhance its quality [5, 6, 16]. However, there has
been limited literature regarding the microwave heating of
natural ruby. Here, an attempt has been made for the first
time to study the MW heating of natural ruby collected
from Sinapali area of Odisha state in India.

2 Experimental procedure
The natural rubies (ruby gemstones) were collected from
Sinapali (Nuapada district) area of Odisha. They varied in
size between 5 and 11 mm. The samples were shaped as
elongated hexagonal-faced prisms, as shown in Fig. 1.
Before heat treatment, first the raw samples were cleaned
with isopropanol to remove any dirt and grease. Heat
treatment was then carried out in a monomodal MW furnace (GN Technologies, India) using SiC pieces as susceptors. The samples were kept between two SiC pieces
which absorbed the MW energy. They got heated indirectly at the rate of 15–20 °C/min. The temperature was
monitored using IR pyrometer with an accuracy of ±30.
The MW heat treatment arrangement is schematically
shown in Fig. 2. MW heat-treated and untreated ruby
samples were characterized using XRD (Rigaku, Ultima
IV, Japan) with Cu Ka for phase identification, UV–visible spectroscopy (JASCO, V-650, Japan) for optical
absorption, fluorescence spectrophotometry (Edinburgh,
FLS 980, UK) for photoluminescence (PL), Raman spectroscopy (Seki, STR-500, Japan) for ascertaining compound phases of impurities, and XPS (Prevac, S/N 10001,

Microwave heat treatment of natural ruby and its characterization

Page 3 of 7 224

Fig. 2 Schematic diagram of microwave sintering furnace used for
heat treatment of ruby samples

Fig. 4 XRD pattern of untreated ruby and MW-treated ruby at
1500 °C

3.2 XRD studies

Fig. 3 Temperature profile of ruby during MW heating

Poland) with Al Ka for identifying the valence states of
the impurities.

Figure 4 shows the XRD patterns of the untreated ruby and
microwave-treated ruby at 1500 °C. The diffraction patterns of the ruby samples (taken in powder forms) match
with that of a-Al2O3 (PDF # 43-1484). The sharp and
prominent peaks correspond to pure a-Al2O3. Some small
peaks present in the patterns correspond to SiO2; these
peaks possibly originate from the agate mortar used for
grinding during powder sample preparation. As seen, the aAl2O3 phase peak positions remain more or less unchanged
by MW heating of ruby at 1500 °C. However, some small
changes in peak intensities are noticed, which may be
attributed to improvement of crystallinity along those
planes.

3 Results and discussion

3.3 Raman spectroscopy studies

3.1 Microwave heating

Figure 5 shows Raman spectra of untreated and microwave-treated ruby at 1500 °C. One main peak is observed
at 415 cm-1 which is attributed to Al-O vibrations of
corundum [20, 21]. The two sharp peaks observed at 810
and 848 cm-1 correspond to Cr-based Al compounds [22,
23]. These peaks represent the longitudinal optical mode of
Eg vibrations. The Raman band positions and peak intensities remain unchanged after heat treatment.

Figure 3 shows the heating behaviour of the natural ruby
sample, where a fast rise in temperature up to 1500 °C
(±30°) is achieved in less than an hour. There are two
distinct regimes of temperature rise; up to about 800 °C,
there appears to be an indirect fast heating of the sample by
heat absorbtion from the SiC susceptors; however, beyond
800 °C, the rate of temperature rise is relatively less sharp;
in this regime, there should be a hybrid heating resulting
from volumetric MW heating of the ruby (alumina) itself
plus the heat absorbed from the SiC susceptors. Highly
favourable dielectric behaviour of SiC makes it a very good
microwave absorber, and hence, it is used as a susceptor
material; it is especially suitable for heating of low-loss
ceramics at low temperatures [17–19].

3.4 Optical absorption studies
Optical absorption in the UV–visible region for the asreceived ruby and microwave-treated ruby, recorded at
room temperature, is shown in Fig. 6. In the visible region,
there are two large absorption peaks near 415 and 560 nm
which are assigned to the transition associated with Cr

123

224 Page 4 of 7

S. Swain et al.

Fig. 5 Raman spectra of the untreated and treated ruby: a in the range of 200–700 cm-1 and b in the range of 700–1000 cm-1

Fig. 6 Optical absorption spectra of the untreated and microwavetreated ruby

impurity; they represent the energy transition 4A2 ? 4F1
and 4A2 ? 4F2 of Cr3? ions, respectively [24]. After MW
heat treatment at 1500 °C, there is an increase in absorption of light in the vicinity of 560 nm, and also, the overall
absorbance is higher at higher wavelength (600–800 nm)
range. As proportionately more absorption is noticed in the
600–800 nm (red and near IR) bands, the sample appears
slightly bleached with improved clarity. There is a small
absorption band around 342 nm which indicates the presence of Fe in the sample [25]. In the ultraviolet region,
there is a sharp absorption peak at 263 nm, together with a
small hump at 216 nm. These peaks are associated with F?
and F centres which originate due to the capture of one or
two electrons by an oxygen vacancy [26]. The absorption
coefficients corresponding to these UV bands decrease
after the heat treatment; this could be attributed to

123

Fig. 7 Photoluminescence spectra of ruby treated at 800 and 1500 °C

reduction in number of oxygen vacancies in the sample by
thermal annihilation.
3.5 Photoluminescence studies
The PL spectra of both untreated and microwave-treated
ruby are shown in Fig. 7. Excitation wavelength (kex)
employed for PL was 400 nm. The most intense peak at
693 nm is a characteristic emission of Cr3?. This peak is
known as R1-line peak which shows the 2E ? 4A2 transition of Cr3? ions [27]. Two other peaks near 707 nm and
714 nm are also observed due to the emission from the Cr
ion pairs and Cr3? ion with lattice defects. These two lines
are called N-line [28]. The PL intensity of ruby treated at
800 °C is found to be more compared to the intensities of
untreated ruby and treated ruby at 1500 °C. This may be

Microwave heat treatment of natural ruby and its characterization

Fig. 8 Photoluminescence spectra of ruby treated at 800 °C with
different soaking duration

attributed to the long fluorescence lifetime as suggested by
Grattan and Zhang [29]; according to them, at low temperature the luminescence of Cr3? is dominated by
2
E ? 2A4 transitions, where long fluorescence lifetimes
are observed. Higher percentage of Cr3? ions will populate
the short lifetime 4T2 state with increase in temperature.
So, more 4T2 ? 2A4 transitions will be initiated, which
results in a decrease in the fluorescence lifetime. At higher
temperatures, a higher level of non-radiative transitions
appears which decrease the fluorescence lifetimes. Thus,
the PL intensity of ruby treated at 1500 °C is low. The
enhanced PL response can also be attributed to decrease in
defects such as vacancies and dislocation in the ruby
sample. An interesting PL effect has been observed for
MW heat-treated (800 °C) ruby samples. Figure 8 shows
the PL spectra of the untreated and the MW heat-treated
ruby at two different soaking durations. The intensity of PL
peak at 693 nm increased with soaking duration of the MW
heat treatment. The enhanced PL response can be attributed
to decrease in defects such as vacancies and dislocation in
the ruby.
3.6 XPS studies
XPS of the untreated ruby and MW heat-treated ruby
samples were recorded to identify the valence states of
impurity elements under two different conditions. Figure 9a, b shows the survey spectra of untreated ruby
and MW-treated ruby at 1500 °C, respectively. The
untreated ruby contains Al, O, Cr, Fe, Ti, Si, and C.
The survey spectra of MW-treated ruby show the same
peaks as observed in the untreated ruby but with higher
intensities. It may be evident from the spectra that the

Page 5 of 7 224

concentrations of Al and O are found to increase after
MW heat treatment. This happens because of the
removal of adventitious hydrocarbons from the surface
of the ruby [16]. Figure 9c, d shows the deconvoluted
Cr 2p peak of untreated and treated ruby, respectively.
In case of untreated ruby (Fig. 9c), the peaks at 578.3
and 567.9 eV arise from Cr 2p1/2 and 2p3/2, respectively. Similarly, the 586.3 and 576.6 eV peaks occur
due to Cr 2p1/2 and 2p3/2, respectively, as in Cr2O3
[30]. In case of treated ruby (Fig. 9d), the Cr 2p1/2 and
2p3/2 peaks arise at 585.2 and 575.7 eV, respectively,
which clearly establishes that the Cr ions are in 3?
state. The binding energies of peaks of treated rubies in
some cases shift towards higher energy side. This
happens because Cr3? ions replace some of Al3? ions
at higher temperature [31–33].
The behaviour of Fe and Ti ions after MW treatment at
1500 °C can be understood from the spectra as shown in
Fig. 9e. It is to be noted that no discerning Fe and Ti
peaks are detected in the untreated sample. The Fe 2p XPS
spectra (Fig. 9e) show two peaks centred at 712.4 and
729.1 eV which arise from the Fe 2p3/2 and 2p1/2 of the
Fe3? species, respectively [30]. The Fe 2p3/2 peak has
associated satellite peaks of Fe2O3. The satellite peak is
located approximately 8 eV higher than the main Fe 2p3/2
peak [34–36]. The peak at 712.4 eV is attributed to Fe3?,
as in a-FeOOH, formed from hydration of Fe2O3 on the
surface of the ruby. The deconvoluted peaks of Ti 2p of
treated ruby at 1500 °C are shown in Fig. 9f. Ti is found
on the surface of ruby after the heat treatment at 1500 °C.
This may be due to the formation of TiO2 clusters with
rise in temperature in the bulk of the ruby. When the
temperature increases to 1500 °C, some TiO2 particles
migrate to the surface, and due to the formation of TiO2
clusters, whitish cloudy particles appear on the surface [5].
The deconvoluted Ti 2p spectra (Fig. 9f) show two peaks
at 458.9 and at 463.8 eV. The features are attributed to Ti
2p3/2 and Ti 2p1/2, respectively. According to the literature, TiO2 has two peaks centred at 458.5 and 464.2 eV,
and they arise from Ti 2p3/2 and Ti 2p1/2, respectively
[30]. Due to Fe2?/Ti4? inter-valence charge transfer, blue
patches appear on the surface of untreated ruby [5]. But at
higher temperature, the mobility of the ions increases
which alter the charge transfer process such that they
diffuse to the surface. Thus, the blue patches are partly
removed by the separation of Fe2?/Ti4? pair ions. The Fe
2p spectrum (Fig. 9e) shows the oxidation of Fe2? ions to
Fe3? ions resulting in the characteristic yellow colour
emission, and while red colour comes from Cr3? ions, the
superposition of the two colours (yellow and red) leads to
produce an overall orange red colour with lustre [37] for
the ruby.

123

224 Page 6 of 7

S. Swain et al.

Fig. 9 a Full-range survey spectra of untreated and treated ruby at
1500 °C, b survey spectra in the range of binding energy (B.E.):
0–180 eV, c deconvoluted Cr 2p spectra of untreated ruby,

d deconvoluted Cr 2p spectra of treated ruby, e spectra of Fe
2p peak of treated ruby at 1500 °C, f deconvoluted Ti 2p spectra of
treated ruby at 1500 °C

4 Conclusions

pink. Such change is reflected in the broad range optical
absorption in UV–visible region of 200–800 nm. The PL
spectra of the MW heat-treated sample qualitatively indicate decrease in lattice defects, such as vacancies and dislocations at higher temperature. The subtle change in
surface colour and increase in clarity after heat treatment
may be attributed to the valence states of Cr, Fe, and Ti, as

MW heating, an unconventional method of fast thermal
excitation, produces new results in gemstone like ruby. MW
heat treatment done at 1500 °C (±30°) produces visible
changes in the colour and structure of ruby. Specifically, the
colour of the gemstone changes from reddish black to light

123

Microwave heat treatment of natural ruby and its characterization

revealed from XPS spectra. The investigation is the first of
its kind in the literature to report the effect of instantaneous
MW heating using a SiC susceptor on a highly insulating
and dielectric material like ruby and thus provides a new
vista for future workers in the area.
Acknowledgments Financial support of CSIR for this work carried
out under Project ESC-206 is thankfully acknowledged.

References
1. H. Aizawa, N. Ohishi, S. Ogawa, A. Endo, A. Hakamada, T.
Katsumata, S. Komuro, T. Morikawa, E. Tob, Sens. Actuators A
Phys. 101, 42 (2002)
2. H.C. Seat, J.H. Sharp, Z.Y. Zhang, K.T.V. Grattan, Sens. Actuators A 101, 24 (2002)
3. D. Liu, Z. Zhu, H. Liu, Z. Zhang, Y. Zhang, G. Li, Mater. Res.
Bull. 47, 2332 (2012)
4. R.K. Sahoo, B.K. Mohapatra, S.K. Singh, B.K. Mishra, Appl.
Surf. Sci. 329, 23 (2015)
5. S. Achiwawanich, N. Brack, B.D. James, J. Liesegang, Appl.
Surf. Sci. 252, 8646 (2006)
6. S. Achiwawanich, B.D. James, J. Liesegang, Appl. Surf. Sci. 253,
6883 (2007)
7. S.F. McClure, C.P. Smith, W. Wang, M. Hall, Gems Gemol. 42,
22 (2006)
8. R.R. Menezes, P.M. Souto, R.H.G.A. Kiminami, in Sintering of
Ceramic Materials—New Emerging Techniques, ed. by A. Lakshmanan (InTech, Rijeka, Croatia, 2012), p. 3
9. J.D. Katz, Annu. Rev. Mater. Sci. 22, 153 (1992)
10. R.M. Anklekar, D.K. Agrawal, R. Roy, Powder Metall. 44, 355
(2001)
11. G. Sethi, A. Upadhyaya, D. Agrawal, Sci. Sinter. 35, 49 (2003)
12. Y.V. Bykov, K.I. Rybakov, V.E. Semenov, J. Phys. D Appl.
Phys. 34, R55 (2001)
13. S.M. Bradshaw, E.J. van Wyk, J.B. de Swardt, J. South Afr. Inst.
Min. Metall. 4, 201 (1998)
14. R. Heuguet, S. Marinel, A. Thuault, A. Badev, J. Am. Ceram.
Soc. 96, 3728 (2013)

Page 7 of 7 224
15. D.E. Clark, W.H. Sutton, Annu. Rev. Mater. Sci. 26, 299 (1996)
16. S. Achiwawanich, B.D. James, J. Liesegang, Appl. Surf. Sci. 255,
2388 (2008)
17. Y. Takeuchi, T. Abe, T. Kageyama, H. Sakai, in Proceedings of
the Particle Accelerator Conference, IEEE, vol 1195 (2005)
18. D. Ding, W. Zhou, B. Zhang, F. Luo, D. Zhu, J. Mater. Sci. 46,
2709 (2011)
19. T.A. Baeraky, Egypt J. Solid 25, 263 (2002)
20. S. Krampelas, M. Worle, J. Raman Spectrosc. 43, 1833 (2012)
21. M. Jersek, S. Kramar, J. Raman Spectrosc. 45, 1000 (2014)
22. L. Xia, R.L. McCreery, J. Electrochem. Soc. 145, 3083 (1998)
23. J.D. Ramsey, R.L. McCreery, J. Electrochem. Soc. 146, 4076
(1999)
24. X. Fan, Y. Wang, H. Xu, Y. Jiang, Cryst. Res. Technol. 46, 221
(2011)
25. T.V. Bgasheva, E.A. Ahmetshin, E.V. Zharikov, Adv. Mater. Sci.
12, 32 (2012)
26. X. Yang, H. Li, Y. Cheng, Q. Tang, L. Su, J. Xu, J. Cryst. Growth
310, 3800 (2008)
27. L. Shen, C. Hu, S. Zhou, A. Mukherjee, Q. Huang, Opt. Mater.
35, 1268 (2013)
28. L.-Y. Yang, Y.-J. Dong, D.-P. Chen, C. Wang, N. Da, X.W.
Jiang, C. Zhu, J.-R. Qiu, Opt. Express 13, 7893 (2005)
29. K.T.V. Grattan, Z.Y. Zhang, Fiber Optic Fluorescence Thermometry, 1st edn. (Chapman & Hall, London, 1995), p. 35
30. C.D. Wagner, W.M. Riggs, L.E. Davis, G.E. Muilenberg (eds.),
Handbook of X-ray Photoelectron Spectroscopy (Perkin-Elmer
Corporation, Minnesota, 1978)
31. M. Ohkubu, T. Hioki, J. Kawamoto, J. Appl. Phys. 60, 1325
(1986)
32. G.C. Farlow, C.W. White, C.J. McHargue, B.R. Appleton, Mater.
Res. Soc. Symp. Proc. 27, 395 (1984)
33. H. Naramoto, C.W. White, J.M. Williams, C.J. McHargue, O.W.
Holland, M.M. Abramham, B.R. Appleton, J. Appl. Phys. 54, 683
(1983)
34. P. Mills, J.L. Sullivan, J. Phys. D Appl. Phys. 16, 723 (1983)
35. D.D. Hawn, B.M. Dekoven, Surf. Interface Anal. 10, 63 (1987)
36. M. Muhler, R. Schlogl, G. Ertl, J. Catal. 138, 413 (1992)
37. K. Eigenmann, K. Kurtz, H.H. Gunthard, Chem. Phys. Lett. 13,
54 (1972)

123