IEEE Transactions on Dielectrics and Electrical Insulation Vol.11, No. 2; April 2004 203
Calcite Microcrystals in the Pineal Gland of the Human
Brain: Second Harmonic Generators and Possible
S. Baconnier and S.B. Lang
Department of Chemical Engineering
Ben-Gurion University of the Negev
84105 Beer Sheva, Israel
A new form of biomineralization in the pineal gland of the human brain has been
studied. It consists of small crystals that are less than 20
m in length and that
are completely distinct from the often-observed mulberry-type hydroxyapatite con-
cretions. Cubic, hexagonal and cylindrical morphologies have been identified using
scanning electron microscopy. Energy dispersive spectroscopy, selected-area elec-
tron diffraction and near infrared Raman spectroscopy established that the crys-
tals were calcite. Experiments at the European Synchrotron Radiation Facility
ESRF to study the biomineralization showed the presence of sulfur originating
from both heteropolysaccharides and amino acids. Other studies at the ESRF fur-
nished information on the complex texture crystallization of the calcite. With the
exception of the otoconia structure of the inner ear, this is the only known non-
pathological occurrence of calcite in the human body. The calcite microcrystals are
believed to be responsible for the previously observed second harmonic generation
in pineal tissue sections. There is a strong possibility that the complex twinned
structure of the crystals may lower their symmetry and permit the existence of a
Index Terms — Microcrystals, calcite, piezoelectricity, second harmonic gener-
ation, scanning electron microscopy, Raman spectroscopy, sulfur, crystal texture,
HE pineal gland is a neuroendocrine transducer
Tthat converts a neural signal into an endocrine output
1 . It secretes a number of hormones, the most important
of which is melatonin which synchronizes the physiologi-
cal Circadian rhythm 2 . Pineal calcifications have been
found in numerous animals and in humans and are the
only crystalline forms known in the pineal gland 3 . Two
major forms of pineal crystalline structures have been ob-
served: i polycrystalline complexes with dimensions of the
order of hundreds of micrometers, often called mulberry-
like structures or concretions, and ii well-defined micro-
crystals having long dimensions as large as 20
second harmonic generation SHG has been observed in
pineal tissue samples 4 . Although the concretions have
been studied extensively, no experiments have been done
previously on the microcrystals. In this research, the mi-
®ed on 8 January 2003, in final form 21 May 2003.
crocrystals were studied by a large number of techniques:
scanning electron microscopy SEM , energy dispersive
spectroscopy EDS , high-resolution transmission electron
microscopy HRTEM , selected area electron-diffraction
SAED and near infrared Raman spectroscopy. Experi-
ments were made at the European Synchrotron Radiation
Facility ESRF to study the roles of heteropolysaccha-
rides and amino acids in the formation of the microcrys-
tals and their structural texture.
2 CHARACTERIZATION OF THE
A total of 20 human pineal glands from subjects ranging
in age from 15 to 68 years were supplied by the Institute
of Pathology of the Soroka Medical Centre in Beer Sheva,
Israel and by the Anatomopathology Service of the CHU
Michalon in Grenoble, France. The glands were fixed in
10% formalin. The microcrystals were isolated from the
pineal glands using a dilute sodium hypochlorite solution
following a procedure developed by Weiner and Price 5 .
rr$20.00 䊚2004 IEEE 203
Baconnier and Lang:Calcite Microcrystals in the Pineal Gland of the Human Brain: Second Harmonic Generators204
Figure 1. SEM photos of mulberry-like concretions in cryofractured
pineal tissue. a, small concretions with lobes; b, large conglomerate.
It should be emphasized that, at no point, did any of the
samples come into contact with solutions containing cal-
cium ions. Microcrystals were found in every gland in
quantities ranging from 100 to 300 crystalsrmm
No attempt was made to correlate the quantity of crystals
with either the age of the subject or pathological details.
SEM samples were collected on transmission electron
microscopy grids and analyzed with a JEOL JSM 5600
SEM. Microanalysis studies were performed with a NO-
RAN EDS Analyzing System. As a reference, SEM pho-
tographs were taken of the mulberry-like concretions. Two
general sizes were observed as shown in Figure 1. Their
outer structure was similar to those observed by others
6,7 . More relevant to the present project were the single
microcrystals. Three different shapes of crystals were ob-
served, cubic, hexagonal and cylindrical as shown in Fig-
ure 2. The length dimensions of the crystals varied from 2
to about 20
m. The most common morphology was a
cylindrical body with sharp extremities. These comprised
about 95% of the samples observed. Edges were usually
very sharp and the body surfaces were very rough. The
NORAN EDS Analyzing System coupled to the SEM was
used to determine the composition of the crystals. The
principal elements identified were calcium, carbon and
Figure 2. SEM of isolated pineal microcrystals on a Formvar-
covered TEM grid. Three different crystal shapes were observed. a,
cubic; b, hexagonal; c, cylindrical.
oxygen with less than 0.5 wt% each of silicon, aluminum,
sodium and magnesium. No phosphorus was found.
Among biominerals containing calcium, carbon and oxy-
gen, only calcite calcium carbonate and calcium oxalate
are potential candidates
Because the microcrystals were too thick for HRTEM
observation, they were first crushed between two glass
slides. They were studied in a JEOL-2010 transmission
electron microscope equipped with an analytical ISIS sys-
tem for energy dispersive X-ray spectroscopy EDS . A
rr$20.00 䊚2004 IEEE
IEEE Transactions on Dielectrics and Electrical Insulation Vol.11, No. 2; April 2004 205
Figure 3. Indexed SAED pattern from fragment of microcrystal.
Figure 4. Raman spectra of pineal gland microcrystals A and pure
calcite powder B .
typical diffraction pattern is shown in Figure 3. The elec-
tron diffraction patterns taken from these particles were
indexed in terms of a hexagonal unit cell with lattice pa-
rameters as4.989 nm, cs17.062 nm and
dimensions are the same as those of calcite.
Near infrared Raman spectra of isolated crystals and of
pure calcite were obtained with a Bruker IFS 66 FTIR
spectrometer equipped with an FRA 106 Raman module
and a Ramanscope microscope. The spectral resolution
was 2 cm
1. The agreement of the peaks was excellent
Figure 4 , confirming the identification of the crystals as
calcite. The additional peaks at 962 and 1283 cm
have come from another chemical substance present in
the crystal, such as a protein.
Lang et al. 4 have observed SHG in dried thin slices of
pineal gland. The observed SHG signal intensities were
3times that of a standard urea powder. In the
present study, SHG measurements were made to deter-
mine the origin of the previous observations. The only
crystalline materials in the pineal tissue were hydroxyap-
atite concretions and calcite microcrystals. Both hydroxya-
patite and calcite are centrosymmetric and would not be
expected to show SHG by the usual dipolar mechanism
8 . However, calcite has been shown to exhibit second
harmonic generation, albeit far weaker than in SHG ac-
tive non-centrosymmetric crystals 9᎐11 . The SHG in cal-
cite is quadrupolar in nature and phase-matchable, and is
preferentially along a specific crystal direction due to
birefringence. For a powdered sample of pure calcite, we
measured an SHG intensity that was 4 orders of magni-
tude weaker than a urea powder standard. We were un-
able to detect SHG in either pure hydroxyapatite powder
-5=10 times that of urea nor in the large hydroxya-
patite pineal concretions. SHG could not be detected in a
small sample of isolated pineal microcrystals, due to the
small number of microcrystals in the sample and their lack
of proper orientation with respect to the incident laser
beam. However, the similarity of the intensity of the SHG
in pure calcite to that observed in earlier work on pineal
tissue samples 4 and the absence of SHG in the large
concretions show that the calcite microcrystals were the
source of the SHG.
3 SULFUR IN THE CALCITE
Preliminary experiments to determine the mechanism
of formation of the calcite microcrystals were carried out
at the European Synchrotron Radiation Facility ESRF
in Grenoble, France. Calcite biomineralization occurs in
an organic matrix in species such as corals, sea urchin spine
and sponge spicules 12 . Sulfur can be found in amino
acids and heteropolysaccharides 13 , two of the major
types of compounds found in organic matrices. Sulfur in
the amino acids, cysteine and methionine has a valence of
y2 in a disulfide-type bond. Among the heteropolysac-
charides are the glycosaminoglycans that contain sulfur in
a sulfate group as chondroitin sulfate. The objective of
our experiments was to locate and analyze the organic
matrix in the microcrystals. Soft x-ray synchrotron radia-
tion can be used to access the K-absorption edges of ele-
ments of major interest in the biological sciences, specifi-
cally sulfur in this case. The studies were carried out on
the X-ray Microscopy Beamline ID21 . A schematic
drawing of ID21 is shown in Figure 5. The x-ray beam was
focused using Fresnel zone plates and raster-scanned over
the crystals with a 0.25=0.25
resolution. A Si-111
monochromator and a solid-state Ge energy-dispersive
detector were used to observe the fluorescence in the
range of energies from 2.450 keV to 2.530 keV. Because
the test specimen was sufficiently thin, a transmission sig-
nal could be measured by means of a silicon photodiode.
The fluorescence spectra of cysteine, methionine and
chondroitin sulfate were first obtained Figure 6 . Peaks
at energies of about 2.472 and 2.473 keV were observed
rr$20.00 䊚2004 IEEE 205
Baconnier and Lang:Calcite Microcrystals in the Pineal Gland of the Human Brain: Second Harmonic Generators206
Figure 5. Layout of the X-ray Microscopy Beamline ID21 at ESRF
Drawing courtesy of ESRF .
Figure 6. Fluorescence spectra of methionine, cysteine and chon-
Figure 7. Fluorescence spectrum of the microcrystal from 2.45 keV
to 2.53 keV.
for the disulfide bonding in cysteine and methionine, re-
spectively, and a peak at about 2.482 keV was observed
for the sulfate bonding in chondroitine sulfate. Then cal-
cite crystals were deposited on a 4-
m thick plastic layer
SpexCertiprep䊚and glued to the sample holder. A
reticule made of a 25-
m tungsten wire was mounted be-
side the sample to enable the crystals to be located more
easily. The samples were mounted in the x-ray microscope
sample holder and analyzed under high vacuum ;1Pa.
A typical fluorescence spectrum is shown in Figure 7. The
peak at 2.473 keV is characteristic of sulfur in amino acids
and that at 2.482 keV for sulfur in heteropolysaccharides.
Figure 8. Location of sulfur compounds. Fluorescence channel up-
per and lower left and transmission channel upper and lower right .
Measurements at sulfate energy 2.482 keV upper left and right .
Measurements at sulfide energy 2.473 keV lower left and right .
The fluorescence channel and the transmission channel
results for the two energies are shown in Figure 8. The
two sulfur compounds were present in the same regions of
the crystal but the signal from the amino acids was weaker.
The sodium hypochlorite solution used in the isolation of
the crystals oxidized most of the organic materials result-
ing in weak fluorescence from the residual sugars. A dif-
ferent crystal isolation technique will be used in future
4 TEXTURES OF THE
The texture of the calcite crystals was studied at ESRF
on the Microfocus Beamline ID13 . This facility provides
a monochromatic focal spot of about 20=20
ing a channel-cut Si-111 monochromator and an ellip-
soidal mirror. This beam is further reduced in size by glass
capillary optics to give a 2-
m beam at the exit. A CCD
detector is used. A schematic drawing of the beamline is
shown in Figure 9. The sample was mounted on a glass
fiber which was held by a goniometer head. The goniome-
ter head was placed on a translation stage that could be
raster scanned in 1-
m steps both horizontally and verti-
cally or rotated in 0.1⬚steps about a vertical axis. A high-
magnification microscope could be moved on a microposi-
tioner in order to align the sample in front of the glass
capillary. The glass fiber, goniometer, microscope and the
glass capillary that focused the x-ray beam are shown in
the photograph in Figure 10.
A microcrystal was placed on the glass fiber and it ad-
hered due to a static electric charge. Diffraction patterns
rr$20.00 䊚2004 IEEE
IEEE Transactions on Dielectrics and Electrical Insulation Vol.11, No. 2; April 2004 207
Figure 9. Layout of the Microfocus Beamline ID13 at ESRF Re-
produced with permission from 22 . The image of the crystals on a
glass fiber are shown on the monitor connected to the video micro-
scope. The diffraction pattern in Figure 11 is shown on the monitor
connected to the CCD detector.
Figure 10. Photograph showing the glass fiber, goniometer, micro-
scope and x-ray focusing glass capillary.
were obtained with a 1
m resolution in horizontal and
vertical scans and a rotational resolution of 0.1⬚. A typical
diffraction pattern is shown in Figure 11. The upper re-
flection and the lower-righthand reflection are 113 and
024 , respectively. Only a few reflections appeared be-
cause of the narrow angle covered by the detector. Fig-
Ž. Ž . Ž.
ures 12 left and 12 right show the intensity of the 113
and 024 reflections for a 15
m vertical scan and a rota-
tion of 3⬚. The intensity of the reflections varied markedly
with very slight translation or rotation of the sample. If
the sample had been a uniform single crystal, all of the
reflections would have had the same intensity. The varia-
tion in intensities indicates that adjacent microregions of
the crystal are slightly out of crystallographic alignment
with one another. A complete mathematical analysis of
the data is presently in progress.
Most of the prior research concerned investigations of
the large mulberry-form concretions. Recently, two stud-
Figure 11. Typical diffraction pattern showing 113 and 024 re-
Ž.Ž. Ž .
Figure 12. X-ray reflection intensities. left 113 reflection, right
024 reflection. The horizontal direction represents translation in
steps of 1
m. The vertical direction represents rotations by 0.1⬚.
ies described the presence of small geometric shapes asso-
ciated with the concretion globules 14,15 , but detailed
investigations were not carried out. The technique devel-
oped here for isolation of the crystals gave us access to
large enough quantities for a study of this new mineral
structure which we have tentatively named ‘‘ myeloconia’’
‘‘brain dust’’in Greek . In contrast to the concretions with
their globular and often conglomerate structure, the mi-
crocrystals have characteristic cubic, hexagonal or cylin-
drical crystalline morphologies. The concretions have sizes
as large as a few millimeters whereas the microcrystals are
not more than 20
m in length. The microcrystals have
the appearance of an agglomerate of small crystals but the
overall morphology shows that those structures are
rr$20.00 䊚2004 IEEE 207
Baconnier and Lang:Calcite Microcrystals in the Pineal Gland of the Human Brain: Second Harmonic Generators208
Figure 13. SEM of isolated pineal microcrystal on a Formvar cov-
ered TEM grid showing the multilayer structure.
monocrystalline ones Figure 13 . The microcrystals, which
have no phosphorus, do not have any relationship to the
hydroxyapatite concretions 3 . The HRTEM and SEAD
coupled with the Raman spectroscopy clearly show that
the microcrystals are calcite and prove the presence of a
new form of biomineralization in the human brain. Or-
ganic material appears on the surface of the biomineral-
ized microcrystals as verified by additional non-calcite
lines in the Raman spectra and the sulfur studies at the
Reasons for the formation of the crystals and their pos-
sible biological significance are not known at present.
However, the microimaging data and SEM photographs
such as the multilayered structure in Figure 13 suggest a
growth mechanism for the crystals. The crystal appears to
have a texture consisting of a stack of thin rhombohe-
drons with their flat faces normal to the long axis of the
crystal. A sketch of this type of structure is shown in Fig-
ure 14. The sharp edges and rough body could be ex-
plained in this way. Mineralogical calcite frequently ex-
hibits complex structures 16 . The texture observed in the
microcrystals could lead to symmetry breaking because of
structural andror stress gradients. This could result in the
existence of properties normally associated with noncen-
trosymmetric crystals such as SHG and piezoelectricity.
The microcrystals bear a striking resemblance to the
calcite crystals that form the otoconia of the inner ear.
Otoconia have been studied extensively in a number of
species, including human beings 17᎐19 . Their growth
stages pass through ovoid, rhombohedral and cylindrical
forms in a manner similar to those of the pineal micro-
crystals. The structure and chemical composition of the
microcrystals are also similar to those found in the
biomineralized crystals of sea urchin spines and sponge
spicules 20 .
We believe that the presence of two different crys-
talline compounds in the same organ is biologically signifi-
cant. It is important to note that the calcite in otoconia
Figure 14. Schematic drawing showing a possible texture structure
has been shown to exhibit piezoelectricity 21 , a property
normally forbidden by crystallographic symmetry. Our
current research is focused on direct measurements of
possible piezoelectricity in the pineal calcite crystals and a
consequent biological transducer mechanism.
This work was partially supported through a Cooperant
du Service National Grant SB and The Israel Science
Foundation Grant No. 54r98 SBL . Special thanks are
extended to Drs. Garry Berkovic and Guilia Meshulam
for the SHG studies and to Prof. Bozena Hilczer and Dr.
Maria Polomska for the Raman spectroscopy. We thank
Drs. Jean Susini and Murielle Salome for their help and
good advice for the ID21 x-ray spectroscopy data collec-
tion and Drs. Manfred Burghammer and Christian Riekel
for their support and advice for the ID13 microdiffraction
experiments. Thanks are also due to Prof. Amalia Konsta
for suggesting an appropriate name for the microcrystals.
We are especially indebted to Prof. A. A. Marino for orig-
inally suggesting the research and for his contributions in
the earlier stages of the work.
1 R. J. Reiter, ‘‘Pineal Gland : Interface Between the Photoperi-
odic Environment and the Endocrine Sytem’’,TrendsEn-
docrinol. Metab., Vol. 2, pp. 13᎐19, 1991.
2 P. Pevet, ‘‘ Melatonin and Biological Rhythms’’, Biol. Signals Re-
cept., Vol. 9, pp. 203 ᎐212, 2000.
3 R. Krstic, ‘‘ A Combined Scanning and Transmission Electron
Microscopic Study and Electron Probe Microanalysis of Human
Pineal Acervuli’’, Cell Tissue Res., Vol. 174, pp. 129᎐137, 1976.
4 S. B. Lang, A. A. Marino, G. Berkovic, M. Fowler, and K. D.
Abreo, ‘‘Piezoelectricity in the Human Pineal Gland’’, Bioelec-
trochem. Bionenerg., Vol. 41, pp. 191 ᎐195, 1996.
5 S. Weiner and P. A. Price, ‘‘ Disaggregation of Bone into Crys-
tals’’, Calcif. Tissue Int., Vol. 39, pp. 365᎐375, 1986.
rr$20.00 䊚2004 IEEE
IEEE Transactions on Dielectrics and Electrical Insulation Vol.11, No. 2; April 2004 209
6 Y. Michotte, A. Lowenthal, L. Knaepen, M. Collard, and D. L.
Massart, ‘‘A Morphological and Chemical Study of Calcification
of the Pineal Gland’’, J. Neurol., vol. 215, pp. 209᎐219, 1977.
7 T. Kodaka, R. Mori, K. Debari, and M. Yamada, ‘‘Scanning
Electron Microscopy and Electron Probe Microanalysis Studies
of Human Pineal Concretions’’, J. Electron Microsc. Tokyo ,
Vol. 43, pp. 307᎐317, 1994.
8 S. K. Kurtz and J. P. Dougherty, ‘‘ Methods for Detection of
Noncentrosymmetry in Solids’’, Syst. Mater. Anal., Vol. 4, pp.
9 H. E. Bjorkholm and A. E. Siegman, ‘‘ Accurate cw Measure-
ments of Optical Second-harmonic Generation in Ammonium
Dihydrogen Phosphate and Calcite’’, Phys. Rev., Vol. 154, pp.
10 S. G. Dinev, S. M. Saltiel, K. V. Stamenov, and K. A. Stankov,
‘‘Measurement of the Quadrupole Nonlinear Coefficient Dis-
persion in Calcite Crystal’’, Opt. Commun., Vol. 24, pp. 225 ᎐230,
11 R. W. Terhune, P. D. Maker, and C. M. Savage, ‘‘Optical Har-
monic Generation in Calcite’’, Phys. Rev. Lett., Vol. 8, pp.
12 J. Aizenberg, M. Ilan, S. Weiner, and L. Addadi, ‘‘ Intracrystal-
line Macromolecules are Involved in the Morphogenesis of Cal-
citic Sponge Spicules’’, Connect. Tissue Res., Vol. 34, pp.
13 M. O’Neil, A. Smith, and P. E. Heckelman, The Merck Index: An
Encyclopedia of Chemicals, Drugs, and Biologicals, 13 ed. White-
house Station, NJ: Merck & Co., Inc., 2001.
14 W. Humbert and P. Pevet, ‘‘ Calcium Content and Concretions
of the Pineal Glands of Young and Old Rats,‘‘ Cell Tissue Res.,
Vol. 263, pp. 593᎐596, 1991.
15 W. Humbert and P. Pevet, ‘‘Calcium Concretions in the Pineal
Gland of Aged Rats: an Ultrastructural and Microanalytical
Study of their Biogenesis,‘‘ Cell Tissue Res., Vol. 279, pp.
16 E. S. Dana, Textbook of Mineralogy, 4th ed. New York: John
Wiley & Sons, 1951.
17 M. D. Ross and K. G. Pote, ‘‘Some properties of otoconia’’, Phil.
Trans. Royal Soc. London B, Vol. 304, pp. 445 ᎐452, 1984.
18 J. M. Sanchez-Fernandez and J. M. Rivera-Pomar, ‘‘A Scanning
Electron Microscopy Study on Human Otoconia Genesis’’, Acta
Otolaryngol., Vol. 97, pp. 479 ᎐488, 1984.
19 J. M. Sanchez-Fernandez, J. M. Rivera-Pomar, and M. J. Tello,
‘‘Human Otoconial Crystal Growth’’, ORL J. Otorhinolaryngol.
Relat. Spec., Vol. 51, pp. 108᎐115, 1989.
20 J. Aizenberg, J. Hanson, T. F. Koetzle, S. Weiner, and L. Ad-
dadi, ‘‘Control of Macromolecule Distribution within Synthetic
and Biogenic Single Calcite Crystals’’, J. Am. Chem. Soc., Vol.
119, pp. 881᎐886, 1997.
21 R. W. Morris and L. R. Kittleman, ‘‘Piezoelectric Property of
Otoliths’’, Science, Vol. 158, pp. 368᎐370, 1967.
22 C. Riekel, ‘‘ New Avenues in x-ray Microbeam Experiments’’,
Rep. Prog. Phys., Vol. 63, pp. 233᎐262, 2000.
This paper is based on a presentation gi
®en at the IEEE International Sympo-
sium on Electrets ISE 11 , Melbourne, Australia, 1
᎐4 October 2002.
rr$20.00 䊚2004 IEEE 209