STUDY OF THE PROPAGATION VE LOCITIES AND THE ELASTIC [601263]

STUDY OF THE PROPAGATION VE LOCITIES AND THE ELASTIC
ANISOTROPY OF THE XENOLI THS OF ALKALINE BASALTS
FROM RACOS

E. L. NICULICI1
1Romanian Geological Institute, Caransebes Str. 1, Bucharest, Romania, E-mail: [anonimizat]

Abstract. The determination of velocities of the elastic waves in formations from the top of the terrestrial
mantle can be done in situ by the study of the pr opagation velocities of Pn and Sn waves, and in the
laboratory by using ultrasound to investigate the xe noliths from lavas that have crossed the lithospheric
mantle. These two methods can also provide interesting information on the elastic anisotropy of rocks
composing the upper part of lithos pheric mantle. We can model the av erage speed, the density and the
elastic modulus of rocks, starting from their thermo-b arometric conditions of equilibrium and from their
modal composition.

Key words: xenolith, lithospheric mantle, anisotropy, velocities

1.INTRODUCTION
The determination of the velocities of elastic waves through formations similar
to the lithospheric mantle performed on xenolite embedded in the alkaline basalts
from Racos arouses a great interest throug h the contribution they can make such
analyzes to elucidate the lithospheric com position and structure of the Vrancea area
and its vicinity.
The problem of the deep geological composition of the Eastern Carpathians bend
area was treated in recent years in terms of petrographic and geochemical by Vaselli et
al (1995), Szabo et al (1995), Falus et al (2000), Seghedi et al (2004 ), Falus et al (2008) and Falus et al (2011).
From structural point of view have be en created over the years several models,
among which the most important being that of the delamination of lower crust and its
descent into lithospheric mantle as a litospheric slab and the unstable triple junctions.

2.TYPES OF XENOLIT E COL ECTED FROM THE ALKALINE B ASALTS
FROM RACOS
In ter ms of li thology m ost of xenoliths studied by Vaselli et al (1995) and Fal us
et al (2008) are the spinel lherzolite. The olivine cont ent of these sam ples range from
60-75% and the orthop yroxen content between 27 and 10% . Also are found some
harzburgite with oliv ine co ntent above 8 5%.

Ol
OPX Sp
2 mm
Fig.1 Peridotite xenolith with orthopyroxen e and chrome-b earing spinel from Racos

An i mportant feature of t hose xenoliths is their milonitic t ype structure, with
crystals deform ed and elongated, and ex tensive recry stallization phenom ena (Falus,
2000, 2008). The anisotropy indicators fo r these xenoliths are represented by the
crystals clea vage alignment for orthopy roxene and of com pression struct ures for
olivine.
In addition t o xenolitele peridotitice were also found significant am ounts of
pyroxenites with garnet and spinels or with garnet, spinel and am phibole. The
amphibole o ccurs secondary by metasomatic transformation of t he clinop yroxene.

Cpx
Amf
2 mm
Fig.2 .Secon dary amphibole formed on clinopyroxene
In some ca ses we c an see the orientation of am phibole is si milar to that of the
pyroxene on which is formed, suggesting a crystallization of secondar y mineral while
deformation.

3.DETERMINATION OF ELASTIC WAVES VE LOC ITIES AND OF THE
ELA STIC ANISOTRO PY USING ULTR ASOUND S
For the determination of velocities of the co mpressional and shear waves and of
the elastic anisotrop y we used an ultraso und source, measuring th e propagation time of
sound t hrough a cry stal or mineral associati on. for t his purp ose is very important to
know very exactly the outer dim ensions of the sam ple. Rock samp les were cut cubical
with a side lenght of 10 mm. Applying signal was successively made on three
orthog onal di rections, determ ining the arrival times for this three directions and for the
frequencies o f 5 MHz, 10 MHz and 20 MHz, the other frequencies are inadequate for
this study , because the wavelength of oscillati ons is much more than the sam ple size.
Given the speed range for t he studied t ype of rocks can be observe d that the m inimum
size of a cry stal which can be detected w ith frequency of 5 MHz is about 1 mm, the
average size of the olivine and py roxene cr ystals encountered frequentl y in xe noliths
from basalts.
We studied f ive sam ples of peridotite with m odal co mpositions starting from
75% olivine and 15% orthop yroxene , six sa mples of py roxenit e with garnet and
amphibole and ten sam ples of eclogitic type. Eclogi tic sam ples were included in the
studied m aterial because am ong xenolit hs of basalts from Racos previousl y collected
by Peter Luf fi were found several sam ples of eclogite sim ilar com position. Were
chosen sa mples with different textural characte rs, from coars e grained porfiroclastic

peridotite to the ultramilonitic and milonitic with phenomena of recrystallization and
grain fine and very fine.
3.1.The equipment used
For determination of acous tic wave velocities P and is used a generator and
receiver of ultrasound with the following features:
• the conditioning of signal and pulse generation:
– Digitally controlled pulse generator and receiver with anti aliasing filter;
– Acquisition board with maximum freque ncy of 20 MHz and a resolution of
12 bits with eight areas of frequency (156 kHz, 312 kHz, 625 kHz, 1.25 MHz, 2.5
MHz, 5 MHz, 10 MHz and 20 MHz);
– Receiver with frequency bandwidth of 10 MHz;
– Amplitude selected by user;
– Pulse generation for less than 5 nanoseconds.
• Connecting to the computer and data acquisition:
– Eight analog inputs with antialiasing filter of 200 Hz;
– Acquisition board with A / D conversi on with 12 bit resolution, ± 10 V input
(10 kHz sampling rate);
– Two channel D / A for the output of P and S wave velocities to a external
acquisition system or to a control system.
• data acquisition unit for the ultrasound speed :
– Microprocessor system for improving and processing of the digital signal;
– Waveform storage ;
– Filtering;
– Spectral analysis;
– Computer controlled selective switching between the reception sensors of the
S and P waves
– Automatic calculation of speed through the following methods:
a) absolute threshold
b) relative threshold of the maximum amplitude ;
c) relative threshold of the first arrival;
d) first arrival;
e) tangent to the first arrival.

The method used for speed determination was the first arrival.
In Figure 3 are presented the unit for signal generation, conditioning and
reception of ultrasound (c), planar sensor for P and S wave velocities (a) and the
group receiver-sensor mounted in measuring device for a soil sample (b).

a) b) c)
Fig. 3 Fig. 3 Th e device for elastic wave ve locity me asur ement usin g ultr asound. a) planar sensor for
meas uring of v elociti es P and S;b) receiver-transmitte r group asse mble d for testing of se diment
samples;c)tran smiter, receiver and the signal conditioning uni t.
3.2.Results
The observed anisotrop y in peridoti te sam ples increase proportionall y with
olivine conte nt and cry stal size. Thus, the porphy ric peridotite with olivine content
over 70% have anisotrop y of up t o 7% but th e average is 4% (Fi g. 6), slightly lower
than that obtained by Falus et al (2008) which to depths of 55 km r each an anisotrop y
of 11%. For the a mphibole py roxenites th e anisotropy decreases to 1.9-2%. For
eclogitic samples the anisotropy varies from 0.9% t o 1.4% (Fig. 7). This variation is
due to the increase of o mfacit content and with the de gree of its orientation.

8.34 8.36 8.38 8.40 8.42 8.4401234567Numar de valori inregistrate
Vp (k m/s) Vp Lherz olite

Fig. 4. Histogra m of P- wave propagation velociti es det ermined in the la borat ory for lherz olite

8.455 8.460 8.465 8.470 8.475 8.480 8.485 8.490 8.49501234567Numar de valo ri inregistrate
Vp (km/s) Vp in eclo gite

Fig. 5 Histogra m of P- wave propagation velocities determined in the laborat ory for eclogit e

012 3456 7801234567numar de va lori
% aniz otropie anizotropie elastica pentru
viteza P i n peridotite

Fig. 6. Histogra m of th e elastic anisotrop y for the P waves in peridotite determined in laboratory

0.9 1.0 1.1 1.2 1.3 1.4012345numa r de esantioane
% ani zotropie anizotropi e elastica pentru
viteza P in eclogite

Fig.7 Histogra m of P wave propagation velocity anis otropy in eclog ite determined in the laboratory

Table 1
Compressional and shear waves velocities, meas ured by ultrasound with fr equen cy of 5 MHz for the
samples of per idotite xeno liths from Racos and eclogitice samples.
Sample frequency
(MHz) Vp
km/s Vs
km/s Vp/Vs Vp
med vp/vs
med Vs
med
R1p
position1 5 8.42 4.78 1.7615 06 8.3961 1.763484 4.7611
R2p 5 8.41 4.77 1.7631 03
R3p 5 8.39 4.75 1.7663 16
R4p 5 8.38 4.76 1.7605 04
R5p 5 8.35 4.75 1.7578 95
R6p 5 8.38 4.75 1.7642 11

Sample frequency
(MHz) Vp
km/s Vs km/s Vp/Vs Vp
med vp/vs
med Vs med
R1p
position2 5 8.22 4.68 1.75641 8.2022 1.750586 4.6856
R2p 5 8.21 4.65 1.765591
R3p 5 8.2 4.7 1.744681
R4p 5 8.19 4.66 1.757511
R5p 5 8.22 4.71 1.745223
R6p 5 8.21 4.73 1.735729

R1p
position3 5 8.2 4.72 1.737288 8.1 1.712903 4.7289
R2p 5 7.8 4.7 1.659574
R3p 5 8.2 4.69 1.748401
R4p 5 8.3 4.74 1.751055
R5p 5 8.1 4.69 1.727079
R6p 5 8.1 4.76 1.701681
Pga1
position1 5 7.89 4.63 1.704104 7.8993 1.702941 4.6387
Pga2 5 7.87 4.64 1.696121
Pga3 5 7.92 4.65 1.703226
Pga4 5 7.89 4.65 1.696774
Pga5 5 7.89 4.64 1.700431
Pga1
position2 5 7.8 4.65 1.677419 7.8073 1.67949 4.6487
Pga2 5 7.81 4.66 1.675966
Pga3 5 7.82 4.64 1.685345
Pga4 5 7.79 4.63 1.682505
Pga5 5 7.78 4.64 1.676724

Sample frequency
(MHz) Vp
km/s Vs km/s Vp/Vs Vp
med vp/vs
med Vs med
Pga1
position3 5 7.78 4.66 1.669528 7.76 1.668105 4.652
Pga2 5 7.74 4.63 1.671706
Pga3 5 7.77 4.66 1.667382
Pga4 5 7.77 4.66 1.667382
Pga5 5 7.75 4.64 1.670259
Ecg1
position1 5 8.48 4.78 1.774059 8.479 1.800181 4.7093
Ecg2 5 8.47 4.68 1.809829
Ecg3 5 8.48 4.7 1.804255
Ecg4 5 8.48 4.69 1.808102
Ecg5 5 8.47 4.71 1.798301
Ecg6 5 8.49 4.72 1.798729
Ecg7 5 8.47 4.66 1.817597
Ecg8 5 8.49 4.69 1.810235
Ecg9 5 8.47 4.68 1.809829
Ecg10 5 8.49 4.7 1.806383
Ecg1
position2 5 8.43 4.66 1.809013 8.427 1.811353 4.6523
Ecg2 5 8.44 4.66 1.811159
Ecg3 5 8.42 4.64 1.814655
Ecg4 5 8.43 4.65 1.812903
Ecg5 5 8.41 4.64 1.8125
Ecg6 5 8.42 4.65 1.810753
Ecg7 5 8.42 4.65 1.810753
Ecg8 5 8.43 4.66 1.809013
Ecg9 5 8.43 4.65 1.812903

Sample frequency
(MHz) Vp
km/s Vs km/s Vp/Vs Vp
med vp/vs
med Vs med
Ecg10 5 8.43 4.65 1.812903
Ecg1
position3 5 8.37 4.7 1.780851 8.3751 1.779285 4.707
Ecg2 5 8.375 4.72 1.774364
Ecg3 5 8.373 4.7 1.781489
Ecg4 5 8.377 4.71 1.778556
Ecg5 5 8.375 4.7 1.781915
Ecg6 5 8.374 4.69 1.785501
Ecg7 5 8.375 4.71 1.778132
Ecg8 5 8.376 4.7 1.782128
Ecg9 5 8.376 4.72 1.774576
Ecg10 5 8.375 4.71 1.778132

The velocities of the acoustic compressiona l wave determined in the laboratory
for peridotitic type formations varies between 8.34 and 8.44 km / s (fig. 4). This
variation is due to variation of olivine cont ent and degree of orientation of the crystals.
Relatively restricted range of values of sp eed is due to generally maintain a high
content of olivine, regardless of the texture and structure of the rock.
Velocities in eclogitic rocks vary on a smaller range (Fig. 5), as a consequence
of the fact that elements that can influence th is change is the amount of garnet in rock
and less the degree of orientation of the anisotropic minerals contained. Note that for
this test we looked for eclogitice samples as little affected by alteration processes.
Velocities determined in the laboratory can be affected by impurities from the
host rock (alkali basalt) kept on the surf ace of sample analyzed. These impurities may
reduce the propagation velocity up to 0.5 km / s for Vp and 0.7 km / s for Vs.
The speeds above have been corrected for the depth of 45 km for lherzolites and
50 km for eclogites according to the algorithm presented by Anderson in 1989.
Were also calculated velocities vp and vs for samples studied by Vaseli et al
(1995). The calculation was done for a depth of 40 km and a geothermal gradient of 20 degrees per kilometer, because most samples investigated (21 of 28) were the spinel lherzolite with different structural varieties. Most of the structural varieties studied by
these authors, and those processed for the preparation of this paper are anisotropic,
showing a preferential orientation of the non isometric minerals and some mineral
segregation. The velocities and the elastic coefficients of these rocks have been
calculated using a program created by Hack er & Abers (2004) with the assumption

that there are perfectly isotropic. Results for four samples of garnet and spinel
pyroxenite (eclogitic) are presented in Tables 2 and 3.

Table 2
Physical properties and water content of pyroxenite samples, calculated for the pressure of 1.5 GPa and
temperature of 800oC
Proba PGS1 PGS2 PGS3 PGS4
P (GPa) 1.5 1.5 1.5 1.5
T (°C) 800 800 800 800
H2O
(wt%)0.0 0.0 0.0 0.0
rho
(g/cm3)3.40 3.40 3.40 3.40
Vp
(km/s)8.56 8.52 8.46 8.50
Vs
(km/s)4.81 4.78 4.74 4.77
K (GPa) 144 143 141 143
G ( G P a ) 7 9 7 8 7 6 7 7
Poissons 0.27 0.27 0.27 0.27

Table 3
Physical properties and water content of pyroxenite samples, calculated for the pressure of 1.7 GPa and
temperature of 800oC
Proba PGS1 PGS2 PGS3 PGS4
P (GPa) 1.7 1.7 1.7 1.7
T (°C) 800 800 800 800
H2O
(wt%) 0.0 0.0 0.0 0.0
rho
(g/cm3) 3.40 3.40 3.40 3.40
Vp
(km/s) 8.58 8.54 8.47 8.52
Vs
(km/s) 4.82 4.79 4.74 4.77
K (GPa) 145 144 142 143
G (GPa) 79 78 77 78
Poissons 0.27 0.27 0.27 0.27

These values were compared with th e mean velocity from measurements
obtained with ultrasound, the latter applying the corrections to a temperature of 800 degrees and pressures of 1.5 for peridotite and respectively 1.7 GPa for pyroxenite and eclogite.

Following the corrections applied was found that average speeds result from
ultrasonic measurements are comparable with those obtained by calculation and also
are comparable to those resulting from Pn and Sn wave study conducted by Ivan
(2004).
Then we recalculated the speeds dete rmined by the program, taking into
consideration the degree of anisotropy, de termined by ultrasound, for each sample
studied.
The results of recalculation showed that for given thermo-barometric conditions
, the velocities of acoustic compressional wave are within the range 7.12-8.07 km / s for peridotite and pyroxenite samples and in the range 8.47-8.48 for the clinopyroxenite with spinels and garnets. These latter has a higher speed than eclogitice samples examined with ultrasound (8.2-8.32 km / s), probably due to the high content of spinels.
4.Conclusions
Laboratory techniques for determining the velocity and elastic anisotropy of
rocks provides very useful informations for building geological models of the main areas of discontinuity from the lithosphere.
Based on information obtained by st udying the propagation velocities of
elastic waves from crustal earthquakes whose epicentres are located at distances of up
to 10 degrees (1100 km) to the measuring points as we can obtain the seismic velocities distribution in situ for Mohorovicic discontinuity.
With this information and linking it with velocities determined in the
laboratory on various xenolite and structur e informations, mineralogical composition
and thermo-barometric conditions of stability of the studied samples we can generate a
petrographic and structural model for discontinuity areas (in our study Mohorovicic
discontinuity), and by extension to certain regions within the lithosphere.
The results of such studies may be useful to the extent that samples used can
be considered representative for the region investigated.

REFERENCES

G.Falus, C.Szabo, O.Vaselli , Mantle upwelling within the Pannonian Basin:
evidencefrom xenolith lithology and minera l chemistry, Terra Nova, 12, 295-302,
2000
G. Falus, A. Tommasi; J. Ingrin, C. Szabo, Deformation and seismic anisotropy of the
lithospheric mantle in the southeasternCarpa thians inferred from the study of mantle
xenoliths, Earth and Planetary Sc ience Letters 272, 50-64, 2008
B. R. Hacker, G. A. Abers, Subduction Factory 3: An Excel worksheet and macro for
calculating the densities, seismic wave speeds, and H2O contents of minerals and

rocks at pressure and temperature, Geoche m. Geophys. Geosyst., 5, Q01005, doi:
10.1029/2003GC000614, 2004
M.IVAN, . Pn and Sn velocity maps For Vrancea area and adjacent areas,
WWW.GG.UNIBUC.RO , 2004
O. Vaselli, H. Downes, M. Thirlwall, Dobos i, G., Coradossi, N., Seghedi, I., Szakacs,
A., Vannucci, R., Ultramafic xenoliths in Plio-Ple istocene alkali basalts from the
Eastern Transylvanian Basin: depleted man tle enriched by vein metasomatism. J.
Petrol,. 36, 23–5, 1995
Cs. Szabó, Gy. Falus, Z. Zajacz, I. Kovács, E. Bali, Composition and evolution of
lithosphere beneath the Carpathian–Pannonian region: a review. Tectonophysics, 393, 119–137, 2004
Cs. Szabó, O. Vaselli, R. Vanucci, P. Bottazzi, L. Ottolini, N. Coradossi, I. Kubovics, Ultramafic xenoliths from the Little Hungar ian Plain (Western Hungary): a petrologic
and geochemical study. Acta Vulcanol., 7, 249–263, 1995..