Energy and Power Engineering, 2010, 18-24
doi:10.4236/epe.2010.21004 Published Online February 2010 (http://www.scirp.org/journal/epe)
Cr+3 Distribution in Al1 and Al2 Sites of Alexandrite
(BeAl2O4: Cr3+) Induced by Annealing,
Investigated by Optical Spectroscopy
Neilo M. TRINDADE1, Rosa M. F. SCALVI2, Luis V. A. SCALVI2*
1Programa de Pós-Gra du ação em Ciência e Tecnologia de Mat eri ai s (POSMAT)-FC,
UNESP, C. P. 473, Bauru, SP, Bra zil
2Physics Department-FC-State University of São Paulo—UNESP, CP 473, CEP 17033-360, Bauru, SP, Brazil
Abstract: In order to investigate optical properties of alexandrite, the present work deals with the influence
of thermal annealing on optical absorption and luminescence spectra of natural samples. The exposure time to
heat treatment at 1000oC is taken into account. Possible migration of Cr3+ ions from Al1 (inversion site) to Al2
(reflection site) is detected. Sample composition is obtained through Scanning Electron Microscopy (SEM)
measurements and points to a rearrangement of Cr+3 and Fe3+ ions in the alexandrite crystalline structure, un-
der thermal annealing influence. This feature may be used to control the optical properties of natural alexan-
drite, which can be associated to the observed laser emission effect.
Keywords: A. oxides, D. optical properties, D. defects
1. Introduction
Alexandrite is a rather rare and precious mineral and the
interest on its production goes from gemological to laser
technology. Alexandrite structure is of chrysoberyl type
with the incorporation of chrome in its lattice, according
to the chemical formula: BeAl2O4:Cr3+. This material
became technologically important from 1974, when its
utilization as an active media for laser action became
known, through the utilization in a synthetic form [1].
Alexandrite emission can be tuned in the range
700-800nm [2]. There is a great interest for the alexan-
drite laser in the present days since it has been vastly us-
ed for medical purposes, presenting superior perfor-
mance compared to other lasers [3–5]. The combination
of the chromIum doping and chrysoberyl matrix leads to
very favorable properties. The alexandrite crystal is
mechanically rigid and presents a fair thermal conduc-
tion. The alexandrite laser is able to resist to higher
repetition rates, and emit a higher average output power,
when compared to other Cr3+ lasers [6]. Although its
utilization has been widely spread, the laser effect,
which is related to its optical properties, is not com-
pletely understood. This issue has motivated very recent
research on this subject [7]. Besides, Brazil is one of the
largest producers of natural alexandrite, which possess a
high gemological value, related to the alexandrite effect.
This effect means a color change from green, under
exposition to sunlight, to red, on illumination by an
incandescent lamp [8].
The alexandrite unit cell can be visualized as approxi-
mately hexagonal close packed (hcp) and composed of
four molecules, with eight Al3+ ions, occupying distorted
octahedral sites, and four Be2+ ions, located at tetrahedral
distorted sites, besides oxygen ions located in plans
perpendicular to c axis [9]. Distortions from a precise hcp
structure of oxygen ions originate two sites of distinct
symmetries: one is called Al1, located at an inversion site
and the other is called Al2, located at a reflection site [10].
Both Al sites are octahedrally coordinated. The Al2 coor-
dination octahedron has a larger Al-O average bond
length (1.938 Å) when compared to the Al1 octahedron
(1.890 Å) resulting in a larger polyhedral volume [7]. It is
known that Al2, due to its larger size, is preferentially
occupied by the Cr3+ ions and it is the main responsible
for the optical properties of alexandrite [11]. A recently
reported work [8] shows the importance of the doping
with chromium ions in alexandrite, and claims that the
saturation of the green and the red color is virtually
determined only by the relative amount of chromium ions
replacing aluminum in Al1 and Al2 sites. Besides, the color
depends on the relative concentration of other impurities,
such as titanium and iron [8].
In order to investigate optical properties of this ma-
terial, the optical absorption technique has demonstrated
as very appropriate to analysis of the impurities effect
[12]. Previous published data [13] show that spectrosc-
opic properties of Cr3+ ion in alexandrite are similar to

N. M. TRINDADE ET AL.
19
Table 1. Composition of natural alexandrite obtained through
EDS analysis for two samples
Composition (wt%)
Element Sample I Sample II
C 0.74 1.54
Na 0.14 -
Mg 0.07 0.30
Al 76.45 77.47
Si 13.33 8.74
Cl 0.10 0.55
K 0.37 1.37
Ca 0.15 0.90
Ti 0.12 0.24
Cr 0.09 0.17
Fe 0.44 1.93
Table 2. Compositional analysis, particularly for Fe and Cr
of natural alexandrite obtained by WDS
Sample Wt%Cr Wt%Fe
I 0.13 0.61
II 0.41 0.37
Cr3+ ion effect in other oxide hosts with octahedral
symmetry such as Al2O3 and YAlO3. In all of them, the
spectra present two well defined lines 4A2g  2Eg (R
lines) and two wide absorption bands. However, the
intensity and relative position of the lines related to these
transitions depend on the host nature [14]. In the case of
alexandrite, these wide bands are associated with the
transition from ground state 4A2g to excited states 4T2g
(band A, centered at 590nm) and 4T1g (band B, centered
at 420 nm). These A and B bands are attributed to Cr3+
and Fe3+ ions, which may be present in the two sites of
distinct symmetries. A third band has been reported [15],
and has been related either to a charge transfer transition
or to a transition terminating in one of highest levels of
the 3d3 configuration, such as the 4T1g levels [16,17].
This band is generally called C, located in the UV
region, being hardly observed in the optical absorption
spectra, because this range is strongly influenced by
Fe3+ traces [18].
In this paper, we present results of optical absorption
measurements on natural alexandrite samples, along with
results obtained for a synthetic sample, for comparison.
These data allow identifying an Ultraviolet (UV) band,
besides broad bands in the visible (VIS) range. The
influence of thermal annealing at 1000oC is taken into
account. Previous results of X-Ray Diffraction (XRD)
and Energy Dispersive Spectroscopy (EDS) corroborate
to the obtained conclusions.
2. Experimental
All the natural samples used in this paper come from the
same mine, in the Minas Gerais state, Brazil. The syn-
thetic sample was grown by the Czochralski method, and
has been initiated by a high-quality crystalline seed. This
sample has been described elsewhere [19]. Optical
absorption measurements were carried out in the range
200 to 700nm (UV-VIS), using a spectrophotometer Cary
1G of Varian. For absorption data obtained at 77K, a
liquid nitrogen cryostat was used, which was placed
close to the spectrophotometer excitation slit. Lumines-
cence data was obtained through excitation with an
Argon laser from Spectra Physics, model 2017, with
main excitation energy of 2.51eV. The laser beam excites
the sample located inside a He closed-cycle Janis Re-
search cryostat, model CS-150, which uses a Cryogenics
compressor, model 8200. The emitted light is acquired
by a Jobin Yvon T6400 spectrophotometer and the signal
is detected by CCD (Charge Coupled Device) also from
Jobin Yvon.
Thermal annealing was accomplished by varying time,
with the temperature fixed at 1000oC, under room pres-
sure conditions. Natural alexandrite samples were an-
nealed according to the following procedure: the oven
temperature was raised until 1000oC with 20oC/min of
rate and kept at this temperature by the desired time. The
cooling down is done using the same temperature rate.
Samples were submitted to consecutive annealing where-
as the optical absorption spectra are recorded between
each annealing.
Chemical composition of these samples was measured
by EDS (Energy Dispersive Spectroscopy) and WDS
(Wavelength Dispersive Spectroscopy), particularly Al,
Cr and Fe concentration. Table 1 presents composition
obtained through EDS from the analysis of two natural
alexandrite samples coming from the same source, and
Table 2 presents the compositional analysis, particularly
for Fe and Cr present in these samples, obtained by WDS.
The Fe composition is very relevant in the alexandrite
sample, because a high amount of this impurity may
mask the identification of optical absorption bands
attributed to Cr3+ in the host matrix. As can be verified in
Tables 1 and 2, Fe has higher concentration compared to
Cr in natural alexandrite. EDS analysis was performed in
order to determine the presence of the most concentrated
elements in the material and these results are semiquan-
titative, since the oxygen determination is not reliable
and beryllium is not detected by this technique. On the
other hand, the Fe and Cr composition determined by
WDS are practically quantitative, since it is possible to
determine the oxygen concentration. Results are as
expected since natural materials present several types of

N. M. TRINDADE ET AL.
20
200 400 600
0
1
2
3
200 400 600
0. 5
1. 0
O p tical d ensity (arb . u nits )
Waveleng th (n m)
77K
band Cband B
band A
vibronic lines
R lines
Optical density (arb.units)
Wavelength (nm)
Band C
Band B
Band A
R Lines
synthetic
300 K
Figure 1. Optical absorption spectra of synthetic alexandrite samples, measured at room temperature (300K). Inset-optical
absorption spectra of synthetic alexandrite, measured at liquid nitrogen temperature (77K).
200 300 400500 600 700
3
4
5
Optical Density (arb. units)
Wavelength (nm)
3 h
no annealing
1 h
2h30min
4 h
Figure 2. Optical absorption spectra of natural alexandrite as function of consecutive annealing
impurities incorporated in the original stone. Moreover,
these data confirm that the presence of Fe in these sam-
ples is high, surmounting the presence of Cr in the
material.
3. Results and Discussion
Figure 1 shows optical absorption spectra for synthetic
alexandrite, measured at 300K. The inset in Figure 1
represents the absorption spectra for the same synthetic
sample, measured at 77K. Bands A and B are clearly
identified. Band C is also identified, which becomes
possible due to the absence of iron in the synthetic sam-
ple composition. The lines R in alexandrite are attributed
to Cr3+ located at Al2 sites on a reflection plane (lines R1
and R2), and show up precisely at same wavelength
either in the emission as well as in the absorption spectra,
at 680.4 and 678.5nm, respectively [20]. The lines R are
responsible by optical properties of alexandrite.
Band A represents overlapping of two absorption
bands of Cr3+ ions in two distinct sites Al1 and Al2. Band
B represents Cr3+ and Fe3+ also incorporated at different

N. M. TRINDADE ET AL.
21
sites. Relevant parameters for the synthetic sample are
obtained from Figure 1, such as band wavelength posi-
tion, absorption coefficient and the full width a half
maximum (), and are presented in Table 3. As can be
verified, the results of optical absorption at room tem-
perature and at liquid nitrogen temperature, in synthetic
alexandrite, show agreement in the position of absorption
bands, absorption coefficients and , with no signifi-
cant variations. This sample is used as a reference in this
paper, due to its outstanding optical quality. The meas-
urement carried out at 77 K (inset of Figure 1) shows
clearly the vibronic transition lines at 645, 654 and 663
nm [21].
The influence of consecutive thermal annealing on
optical absorption bands of natural alexandrite is shown
in Figure 2. The analysis of variation in the spectra
allows investigation of the possible migration of Cr3+
between Al1 and Al2 sites. Relevant parameters for this
material, such as  and absorption coefficient are
obtained from Figure 2, and are presented in Table 4.
With the help of Tables 3 and 4, bands A and B of
natural alexandrite samples without thermal annealing
and the bands due to synthetic sample, recorded at room
temperature, can be compared. It is easily verified that
their central position
)
(

shows a short shift. A larger
difference occurs between their absorption coefficients,
where the synthetic sample presents lower values than
the natural alexandrite. Natural sample shows an absorp-
tion coefficient 12 times higher for band A and 5 times
higher for band B compared to the values of correspond-
ing bands for the synthetic sample. That can be explained
by the Cr concentration, which is higher for natural
material and also due to the Fe concentration, which is
present only in the natural samples. This is in good
agreement with the lower optical density and higher
transparency of synthetic sample.
Concerning the natural sample, the absorption spec-
tra present large bands centered about 580 and 585nm
(band A) and 425 and 435 nm (band B), besides the R
lines, close to 680nm. Thermal annealing causes a
consecutive and discrete increase in the optical absorp-
tion coefficients and a broadened band, which must be
related to the increase of Cr3+ population in sites Al1
and Al2. This statement is reinforced by decomposition
of this band, as shown in Figure 3. Band A vanishes
with thermal annealing during 4 h, when the noise in
the UV starts to disappear and band C begins to gain
shape. This band had previously been found only with
5h of thermal annealing [21]. It shows that thermal
treatment is an efficient method of refining the optical
absorption data. Then, the emerging of band C can be
related to the disappearing of band A. It is known that
band A is an absorption band from Cr3+ ions in sites Al1
and Al2 and band C is also related with presence of
Cr3+ [15]. Then, one may conclude that thermal an-
nealing leads to diffusion of Cr3+ through the alexan-
drite lattice. An increase of absorption coefficient of
band B is taking place, presenting a maximum with 2h
of annealing time, vanishing with 2 and a half hours
and showing up back, but with lower absorption coef-
ficient with 4 h of thermal treatment. These modifica-
tions indicate migrations of Cr3+ and/or Fe3+ ions
throughout the materials structure, revealed by the
variation of . The lines R, either can be observed in
a sole line, or separated in two lines, called R1 and R2.
The line R1 is located around 680- 682nm and R2
around 678-680nm.
The analysis of optical absorption bands behavior of
natural alexandrite sample submitted to thermal anneal-
ing can be done by comparing with synthetic sample
(Figure 1). In our approach, the band A is decomposed in
two Gaussian curves, in order to estimate the relative
amount of Cr3+ ions between Al1 and Al2 sites [14] and
the migration caused by thermal annealing. The data
fitting for the synthetic sample, shown in Figure 3, leads
to the best fitting of band A, when a regression of two
Gaussian curves is used. Band A corresponds to overlap-
ping of optical absorption bands from Cr3+ located in
sites Al1 and Al2, as already mentioned. Al2 is larger than
Al1, and then, it is preferentially occupied by Cr3+ ions.
There is a dependency on color shift and other optical
properties with Cr3+ distribution among these sites,
Table 3. Parameters obtained from optical Absorption data for synthetic alexandrite, at 300 and 77 K. λ means the position of
maxima. α is the optical absorption coefficient and Δλ is full width a half maximum
Optical Absorption bands – Synthetic sample
Band A (Cr3+) Band B (Cr3+, Fe3+) Band C (Cr3+)
Absorption
Lines of Cr3+
T
(K)
λ
(nm)
α
(cm-1) ∆λ (nm) λ (nm)
α
(cm-1) ∆λ (nm)λ (nm)
α
(cm-1)
∆λ
(nm)
λ
(nm)
300 583 0.31 91.5 420 0.75 59.5 267 0.11 16.8 680
77 582 0.32 90 419 0.76 56.6 265 0.11 15.8 680

N. M. TRINDADE ET AL.
22
Table 4. Parameters obtained from optical Absorption data (UV-Vis) for natural alexandrite sample after thermal annealing.
λ means the position of maxima. α is the optical absorption coefficient and Δλ is full width a half maximum
Absorption bands of natural alexandrite Absorption
Banda A (Cr3+) Banda B (Cr3+, Fe3+) Banda C (Cr3+) Lines of Cr3+
λ
(nm)
α
(cm-1)
∆λ
(nm)
λ
(nm)
α
(cm-1)
∆λ
(nm)
λ
(nm)
α
(cm-1)
∆λ
(nm)
λ
(nm)
STT 585 3.71 85 426 3.9 47.8 - - - 680
TT1000ºC/ 1h 585 5.76 93.8 429 4.36 46.2 - - - 680 and 682
TT1000ºC/ 1h 30min 583 4.24 84 428 6.08 48.9 - - - 679
TT1000ºC/ 2h 585 5.71 89.7 432 8.11 46.9 - - - 679 and 681
TT1000ºC/ 2h30min 563 4.9 89.64 - - - - - - -
TT1000ºC/ 3h 577 7.2 96.98 - - - - - - -
TT1000ºC/ 4h - - - 429 1.8 57.67 276 6.4 57.4 680
which was determined by Electron Paramagnetic Reso-
nance (EPR) [11,22], and indicates that Cr3+ in BeAl2O4
enters in a average ration of 75% in Al2 and 25% in Al1,
either in natural sample as well as synthetic alexandrite
sample. Based on this ratio, the analysis of Figure 3 was
performed and the results are shown in Table 5. With the
help of Figures 1 and 2, it may be concluded from Table
5 that in the synthetic sample, we have the 1:3 ratio in
agreement with previous mentioned reported data [11,
22]. The addition of two Gaussian curves leads to a
perfect fitting of the experimental curve. Then, they can
be used to the analysis of the absorption band of natural
sample. In this case, the fitting by Gaussian curves is
hard to be done, which is caused by the experimental
noise. In the cases where a data fitting of band A became
possible, it was observed that an increase of thermal
annealing time leads to decrease of Cr3+ in Al1 sites and
thus, an increase in the occupation of Al2 sites. Besides,
there is an increase of the Gaussian curves area, which
500 550 600 650
0.0
0.1
0.2
0.3
experimental
gaussian fitting
Cr3+ in Al2
Cr3+ in Al1
Optical Density (arb.units)
Wavelength (nm)
Figure 3. Decomposition of band A of optical absorption
spectra in two Gaussian curves for synthetic alexandrite
678694 695 696
0
3
6
9
12
S1
R1
Intensity (arb. units)
Wavelength (nm)
No T.A.
T.A. 1000oC, 4 h
R2
Natural alexandrite
20 K
Figure 4. Photoluminescence spectra of natural alexandrite,
measured ate 20 K, before any thermal annealing and after
the final annealing at 1000oC, 4 hour
Table 5. Analysis of bands A, B and C of natural and syn-
thetic alexandrite samples, obtained from decomposition of
optical absorption bands
Band A Average total area (a. u.)
Thermal annealing
temperature/time Al1(%) Al2(%) Band A
Band
B
Band
C
STT 33.5 66.5 15.7 20.4-
1000ºC/5min 22.1 77.9 24.2 15.4-
1000ºC/15min 14.5 85.5 24.7 17.5-
1000ºC/30min 15 85 28.1 44.8-
1000ºC/2h - - 29.0 79.3-
1000ºC/2h30min 18.2 81.8 35.6 - -
1000ºC/3h 4.8 95.2 41.4 - -
1000ºC/4h - - - 14.135.7
Synthetic sample 22.7 77.3 31.8 50.43.1
means an increase of Cr3+ concentration, responsible for
the optical absorption in the sample. Cr3+ ions in Al2 sites
are responsible for laser emission, which are character-
ized by electric dipole transitions of high-probability,

N. M. TRINDADE ET AL.
23
whereas Cr3+ ions in Al1 do not contribute significantly
to the optical absorption. The excitation of Cr3+ located
in Al1 ions are magnetic dipole transitions, and do not
participate in the laser emission process. Besides, this
transition contributes for decreasing the excitation energy
of Cr3+ in Al2 [23].
As expected from the absorption coefficient data, the
area under band B also increases until 2 h of annealing
time, however band B vanishes for longer times, as
already mentioned. This band shows up again with 4 h of
annealing along with band C. On the other hand, band A
resists to the thermal annealing, vanishing only with 4 h
of annealing time, when band C shows up. In summary,
the analysis of bands A, B and C leads to the conclusion
that bands A and B present increased area with longer
thermal treatment.
In alexandrite luminescence spectra, the Cr+3 lines,
due to the ion located in the reflection site, are the R1 and
R2 lines and the lines due to Cr3+ located in the inversion
site are S1 and S2. As previously mentioned, the R lines
show up precisely at the same wavelength, 680.4 nm and
678.5 nm, either in the absorption spectra as well as in
the emission spectra. Lines S1 and S2 show up at 695.8
and 689.9 nm, respectively in the emission spectra and as
narrow lines at 655.7 nm, 649,3 nm and 645.2 nm in the
absorption spectra [16]. In order to assure the hypothesis
of ion migration, photoluminescence measurements on
natural samples were carried out at low temperature
(20K), before any thermal annealing and after the last
annealing (1000oC, 4 hour). This measurement tempera-
ture was chosen in order to have a very well defined
spectrum, not influenced by phonon emission. These
results are shown in Figure 4, where lines R1, R2 and S1
are easily observed, whereas S2 line is not observed. The
most relevant information for this work is that the R1 line
has its intensity increased by the thermal annealing,
whereas S1 presents lower intensity. This behavior rein-
forces the possibility of ion migration from the inversion
site, Al1, to the reflection site, Al2, in good agreement
with our absorption data fitting procedure.
Previously reported X-ray diffraction data for alexan-
drite sample before and after thermal annealing [24]
show that the characteristic peaks occur at the same
positions. Thermal annealing does not cause modifica-
tions on alexandrite structure, which is a very interesting
result, since our main goal is to study its optical proper-
ties related to Al1 and Al2 occupation by Cr3+ ions, and
the variations on optical properties induced by thermal
annealing.
4. Conclusions
We summarize the conclusions that we have drawn in
this paper as follows: thermal annealing has allowed the
observation of meaningful variation on the optical ab-
sorption bands of natural alexandrite in the visible and
ultraviolet ranges. Depending on time of thermal anneal-
ing at 1000oC the bands A, B and C have its shape com-
pletely changed. The annealing favors the presence of
Cr3+ in Al2 sites, which was verified by alteration in the
specific areas of the decomposed absorption bands and
variation of relative emission intensity. This may also
explains why, unlike other tunable lasers, alexandrite
lasers emit with fair efficiency even at room tempera-
ture. Chemical composition shows that the iron con-
centration is high in the natural alexandrite, which does
not ruin the conclusion on Cr3+ optical absorption
properties drawn in this paper, because although the Fe
ions present strong influence in the ultraviolet range,
the analyzed bands related to Cr ions are in the visible
range.
5. Acknowledgements
Authors thank the financial support of Brazilian agencies:
FAPESP, CNPq, FUNDUNESP and CAPES. We also
thank Prof. Lígia O. Ruggiero and Prof. Américo Sheitiro
Tabata for the use of the equipments and Prof. Tomaz
Catunda for the use of the synthetic sample.
REFERENCES
[1] J. C. Walling, O. G. Peterson, H. P. Jenssen, R. C. Morris,
and E. W. O´Dell, “Tunable alexandrite lasers,” IEEE
Journal of Quantum Electronics, Vol. QE-16, pp. 1302–
1314, 1980.
[2] J. C. Walling, D. F. Heller, H. Samelson, D. J. Harter, A. J.
A. Pete, and R. C. Morris, “Tunable alexandrite lasers:
development and performance,” IEEE Journal of Quan-
tum Electronics, Vol. QE-21, pp 1568–1581, 1985.
[3] L. Li, T. Kono, W. F. Groff, H.M. Chan, Y. Kitazawa, and
N. J. Nozaki, “Comparison study of a long-pulse pulsed
dye laser and a long-pulse pulsed alexandrite laser in the
treatment of port wine stains,” Journal of Cosmetic and
Laser Therapy, Vol. 10, pp. 1472–1476, 2008.
[4] N. Bouzari, H. Tabatabai, Z. Abbasi, and A. Frios, “Laser
hair removal: Comparison of long-pulsed Nd: YAG,
‘Long-pulsed alexandrite, and long-pulsed Diode
Lasers’,” Dermatologic Surgery, Vol. 30, pp. 498–502,
2008.
[5] M. Landthaler and U. Hohenleutner, “Laser therapy of
vascular lesions photodermatology,” Photoimmunology
& Photomedicine, Vol. 22, pp. 324–332, 2006.
[6] S. C. Collins, T. D. Wilkerson, V. B. Wickwar, D. Rees, J.
C. Walling, and D. F. Heller, “The alexandrite ring laser:
A spectrally narrow lidar light source for atmospheric
fluorescence and absorption observations in Advances in
atmospheric remote sensing with Lidar,” Edited by A.
Ansmann, R. Neuber, P. Rairoux, and U. Wandinger,
Springer Verlag, Berlin, pp. 577–580, 1997.
[7] S. U. Weber, M. Grodzicki, W. Lottermoser, G. J. Red-
hammer, G. Tippelt, J. Ponahlo, and G. Amthauer, “Fe
Mossbauer spectroscopy, X-ray single-crystal diffrac-

N. M. TRINDADE ET AL.
24
tometry, and electronic structure calculations on natural
alexandrite,” Physics and Chemistry of Minerals, Vol. 34,
pp. 507–515, 2007.
[8] V. I. Solomonov, S. G. Mikhailov, and A. I. Lipchak,
“Impurity luminescence of alexandrite crystals,” Journal
of Applied Physics, Vol. 69, pp. 423–429, 2002.
[9] R. M. F. Scalvi, M. S. Li, and L. V. A. Scalvi, “Thermal
annealing-induced electric dipole relaxation in natural
alexandrite,” Physics and Chemistry Minerals, Vol. 31, pp.
733–737, 2005.
[10] C. F. Cline, R. C. Morris, M. Dutoit, and P. J. Harget,
“Physical properties of BeAl2O4 single crystals,” Journal
of Materials Science, Vol. 14, pp. 941–944, 1979.
[11] G. V. Bukin, A. V. Eliseev, V. N. Matrosov, V. P. Solntsev,
E. I. Kharchenko, and E. G. Tsvetjov, “The growth and
examination of optical properties of gem alexandrite,”
Proceedings of the XI IMA Meeting, Novosibirsk, pp.
317–328, 1980.
[12] J. A. Hernandez, W. K. Cory, and J. O. Rubio, “A non
destructive method for determining the Eu2+ concentra-
tion in the alkali chlorides,” Japanese Journal of Applied
Physics, Vol. 18, pp. 533–538, 1979.
[13] M. J. Weber and T. E. Varitimos, “Optical spectra and
relaxation of Cr3+ ion in YAlO3,” Journal of Applied
Physics, Vol. 45, pp. 810–816, 1974.
[14] J. Xu, K. Shi, G. Xiong, and X. Xu, “The vibrational
relaxation processes in BeAl2O4:Cr3+”, Journal of Lumi-
nescence, Vol. 40 & 41, pp. 611–612, 1988.
[15] A. E. Underhill and D. E. Billing, “Calculation of the
Racah parameter B for Nickel (II) and cobalt (II) com-
pounds,” Nature, Vol. 210, pp. 834–835, 1966.
[16] R. C. Powell, L. Xi, X. Gang, and G. J. Quarles, “Spec-
troscopic properties of alexandrite crystals,” Physical Re-
view B, Vol. 32, pp. 2788–2797, 1985.
[17] A. B. Suchocki, G. D. Gilliland, R. C. Powell, and J. M
Bowen, “Spectroscopy properties of alexandrite crystals
II,” Journal of Luminescence, Vol. 37, pp. 29–37, 1987.
[18] F. Hassan and A. El-Rakhawy, “Chromium III centers in
synthetic alexandrite,” American Mineralogist, Vol. 59,
pp. 159–165, 1974.
[19] S. K. Pan and X. G. Wang, “Growth of laser crystal
alexandrite,” Crystal Research and Technology, Vol. 29,
pp. k31–k35, 1994.
[20] P. Fabeni, G. P. Pazzi, and L. Salvini, “Impurity centers
for tunable lasers in the ultraviolet and visible regions,”
Journal of Physics and Chemistry of Solids, Vol. 52, pp.
299–317, 1991.
[21] R. M. F. Scalvi, M. S. Li, and L. V. A. Scalvi, “Annealing
effects on optical properties of natural alexandrite,”
Journal of Physics: Condensed Matter, Vol. 15, pp. 7437–
7443, 2003.
[22] H. Rager, A. Bakhshand-Khiri, and K. Schmetze, “Inves-
tigation of the intracrystalline Cr3+ distribution in natural
and synthetic alexandrites,” N. Jb. Miner. Mh, Vol. 2 , pp.
545–557, 1998.
[23] B. K. Sevast’yanov, “Excited-state absorption spectros-
copy of crystal dopede with Cr3+, Ti3+, and Nd3+ ions, re-
view,” Crystallography Reports, Vol. 48, pp. 989–1011,
2003.
[24] R. M. F. Scalvi, L. O. Ruggiero, and M. S. Li, “Influence
of annealing on X-Ray diffraction of natural alexandrite,”
Powder Diffraction, Vol. 17, pp. 135–138, 2002.