Nuclear Instruments and Methods in Physics Research B xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
WD-XRA technique in multiphase flow measuring A.S. Gogolev a, Yu.M. Cherepennikov a,⇑, A.V. Vukolov a, R.O. Rezaev a,c, S.G. Stuchebrov a, D. Hampai b, S.B. Dabagov b,c,d, A. Liedl b, C. Polese b a
Tomsk Polytechnic University, Lenin Avenue 30, Tomsk 634050, Russia INFN Laboratori Nazionali di Frascati, Via E. Fermi 40, 00044 Frascati, Italy c Nuclear University MEPhI, Kashirskoye Shosse 31, Moscow 115409, Russia d RAS P.N. Lebedev Physical Institute, Lenin Avenue 53, Moscow 119991, Russia b
a r t i c l e
i n f o
Article history: Received 30 November 2014 Received in revised form 6 February 2015 Accepted 9 February 2015 Available online xxxx Keywords: Flowmeter Multiphase fluid Dual wave technology X-ray absorptiometry Polycapillary optics
a b s t r a c t A new technique to perform the analysis of multiphase fluid flow based on wave dispersive X-ray absorptiometry is suggested. The numerical simulation and comparison of this technique with currently used approaches are provided and a way to increase the luminosity intensity is found that includes the usage the X-ray focusing optics by a bent crystal and a polycapillary semilens. Based on numerical simulation of radiation spectrum the influence of the bent crystal on the luminosity is evaluated and experimentally shown the advantages of using the multicapillary optics. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction During the past decades the real-time measurement of the amount and others parameters of the well production without the phase separation, moving parts and manual control has been challenging many research groups in gas and oil industry and yet still the question is opened. The detailed review of different problems and approaches to solve them can be found elsewhere [1]. Meanwhile the technical demands to accuracy of measurements are constantly growing. On the other hand new analytical methods and devices combined with computational techniques provide additional possibilities to satisfy those demands. The analysis of X-ray, gamma or neutron radiation passed through the studied object is one of the several methods allowing the non-invasive fluids’ components control. Based on this technique the promising Vx non-separation technology of flow metering has been implemented [2]. The idea of this technology consists of the combined usage of Venturi tube and a gamma-based density measurement device exploiting the radioactive 133Ba source of emission with decay rate 10 mCi. The attenuation of the emission intensity during the passing through the multiphase fluid depends on the energy of the emission (in the discussed case it is 32 and 81 keV) and the composition of the fluid. The following analysis ⇑ Corresponding author.
of the passed emission with two energies allows defining the composition of the three-component fluid. Those particular devices have a number of drawbacks due to usage of a danger radioactive gamma-ray sources, high prices, etc. The above mentioned radioactive sources also have a low level of the radiation intensity. Since the number of photons directly influences on the statistical error it leads either to low accuracy of measurements or to increase the duration measurement. The last one requires about an hour per one measurement with using the device based on radioactive isotopes in order to achieve the satisfactory statistical uncertainty [3]. The direct averaging over the time interval dramatically increases the systematic error with interval of integration growth due to the many nonlinear factors. The use of X-ray tube as a source of radiation provides a higher intensity of radiation in the particular range of energy (for example, the multiplication factor is about three-four orders by magnitude on FWHM equal to 10 eV compared to radioactive isotopes). It is important to note that X-ray tube considered as the generating source of the radiation can be switch off at any moment of time becoming absolutely safe in the radioactive pollution sense. However the application of such sources of radiation leads to difficulties on the stage of radiation detection and data processing due to the continuous character of the radiation spectrum unlike the radioactive isotopes having the discrete number of monochromatic lines in the spectrum.
http://dx.doi.org/10.1016/j.nimb.2015.02.030 0168-583X/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: A.S. Gogolev et al., WD-XRA technique in multiphase flow measuring, Nucl. Instr. Meth. B (2015), http://dx.doi.org/ 10.1016/j.nimb.2015.02.030
2
A.S. Gogolev et al. / Nuclear Instruments and Methods in Physics Research B xxx (2015) xxx–xxx
The question of how to combine advantages of both approaches (easiness to register and process data in the case of radioactive sources and safety and high intensity of X-ray sources) is of a big practical interest. Development of that research field has led to creation the new device «X-ray based densitometer for multiphase flow measurement» [4] based on FluorX [5] as the source of radiation to produce the X-ray beam with quasiline spectra of the secondary fluorescence. The device in fact is the X-ray tube with secondary target and still possesses some of the drawbacks such as the lower characteristic lines intensity due to the reradiation that leads to the loss about three order of intensity by magnitude compared to the fluorescence [6]. The last phenomenon decreases the intensity of radiation down to the level of radioactive isotopes. Among others disadvantages one can mention the background radiation presence which consists of the scattered bremsstrahlung with continuous spectrum and characteristic Kb lines. The intensity of the background radiation is comparable with the intensity of the useful characteristic Kb radiation that leads to the undesired loading the detector. Such intensive background radiation increases errors after all. Thereby this device although provide some improvements in radiation safety however does not increase dramatically the accuracy and expressness of measurements is comparable with Vx technology based devices. In this paper new alternative ways of wave dispersive X-ray absorptiometry (WD-XRA) as well as the prototype of the device implementing this idea are suggested. The method is based on the adaptation of wave dispersive technique [7] to the component analysis of a multiphase fluid. The detailed numerical simulation of the spectra and intensity of X-ray radiation on different regimes is performed. Since as it is mentioned above the main factor determining the accuracy and duration of measurement is the intensity of radiation there are considered different possibilities of how to increase the intensity using the focusing X-ray optics: a bent crystal and a polycapillary semilens. 2. General considerations The Fig. 1 shows the main idea of the technology: the generation of radiation using the X-ray tube as a source of radiation, passing the combined spectrum radiation through the fluid flow (or other medium), segregation of narrow monochromatic lines of radiation using the block of crystal-analyzers, detection of segregated lines using one or two scintillators and analysis of the absorbed radiation level for those narrow lines. In order to achieve the individual registration of different energetic lines in the case of having one scintillator the hardware (electronic) methods of separation is used. As a rule the multicomponent analysis of well production is based on three-component approach (oil–gas–water) and for that analysis of absorbed radiation two lines with different energy (low and high energy) is enough. The system of equations (1) is composed the solution of which gives data about the fraction of each component. Such approach, in particular, is used in devices
Fig. 1. The principal schema WD-XRA.
based on Vx-technology. In the system (1) the l is the linear coefficient absorption, k is a fraction of component, I0, I are intensities of radiation before and after passing the studied medium, h is the thickness of the studied medium, indices w, o, g denote water, oil and gas fractions, indices H and L denote high and low energy
lnðIL =IL0 Þ ; h lnðI =I Þ lwH kw þ loH ko þ lgH kg ¼ H H0 ; h kw þ ko þ kg ¼ 1:
lwL kw þ loL ko þ lgL kg ¼
ð1Þ
The suggested way of monochromatisation is implemented using the crystal monochromator herewith the module of monochromatisation can be realized by several techniques, in particular for more than two lines analysis that allows to determine the concentration of as much as needed fraction components of medium. It might be used, for example, the single crystal and the very first and second (or higher) order of the radiation diffraction on the crystal for the role of two lines for analysis. The other possibility is to use the number of crystal monochromators tuned to segregate different energetic lines and each diffracted from the monochromator radiation is detected by separate detector. In that case the possibility to distinguish several orders of diffraction from individual crystal and usage each of them as the particular level of energy is still conserved. Therefore there is a possibility to choose any schema satisfying to particular problem. Besides, the parameters of X-ray source (anode material, voltage and current of Xray tube) might be easy optimized to specific problem. To approach to the solution of the above discussed challenge – determining the composition of the fluid flow in three-approach approximation we propose as we believe the most prospective options, namely, the X-ray tube with a silver anode and 60 kV voltage on it as a radiation source and specially composed crystal monochromator–analyzer to separate two energetic lines. Such a monochromator presents a pair of glued between each other crystal membranes differ in size and with different orientation of the crystal plane. One of the lines should correspond to the characteristic Ka radiation, the second one is the bremsstrahlung with energy of 40–50 keV order. To separate the working lines two silicon crystals were chosen (1 1 1) and (1 0 0). Such technique allows reflecting two energetic lines in the first allowed order of the diffraction and the radiation corresponding to higher orders for both crystals absent in the X-ray tube spectrum.
3. Simulation of X-ray spectra In order to evaluate the intensity of passed and scattered X-ray radiation the GEANT4 package is used for the simulation of spectral angular characteristics of X-ray tube radiation [8]. On the Fig. 2 results of such simulations are presented (spectrum of X-ray radiation). On the Fig. 2a two spectra are shown: the spectrum registered by detector without fluid flow and in the case when the fluid flow is the water. This panel shows the difference in absorption of radiation with different energies which in fact is the source of information about the composition of medium. As can be seen from presented spectra the suggested technology is capable to provide the flow radiation about 105 photons per second at the 1 mA current without even usage of special focusing X-ray optics and this flow can be increased by more than one order only by means of X-ray tube power increase. For comparison, the flow of radiation from radioactive isotopes with decay per second 10 mCi consists of 103 photons radiated in the same solid angle. On the Fig. 2b the spectrum of radiation reflected from composed monochromator is shown.
Please cite this article in press as: A.S. Gogolev et al., WD-XRA technique in multiphase flow measuring, Nucl. Instr. Meth. B (2015), http://dx.doi.org/ 10.1016/j.nimb.2015.02.030
A.S. Gogolev et al. / Nuclear Instruments and Methods in Physics Research B xxx (2015) xxx–xxx
3
Fig. 2. Spectra of radiation (a) without fluid in the tube (black, left vertical scale) and in the case of filled by water tube (red, right vertical scale) (b) after a reflection from composed monochromator in the presence of water in the tube. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
From the Fig. 2b one can see that the integral intensity of scattered bremsstrahlung in the suggested technology does not exceed the level of 1% meanwhile in the source FlourX similar quantity is comparable with the level of useful signal [4]. Correspondingly, the additional error induced by the background radiation in our technology is dramatically lower and influences less on the final result of measurement. In order to check the workability of the suggested model and to evaluate the discussed advantages the simulation of the secondary fluorescence spectrum was performed under the same conditions. The geometry of the virtual experiment were the same however the radiation generated by the source fall down to the secondary target made of alloy of Ag and Cd instead of composed monochromator. Materials were chosen in order to generate the quasimonochromatic radiation from special spectral range. Spectrum of radiation from the secondary target is shown on Fig. 3a. As it can be seen from the figure, results are in qualitative (integrally the level of bremsstrahlung is close to the level monochromatic radiation) agreement with data for FluorX source. Also it can be seen that under the very the same parameters of the primary radiation source in the case of the secondary fluorescence we have the intensity which is much lower than in our technique (see Fig. 2b as well). The difference is about three orders of magnitude that can be explained by the loss of the radiation due to the reradiation (see above discussion). Another important question is the prolongation of X-ray tube lifetime. It can be achieved by increasing the intensity of monochromatic lines generated at 1 mA current. In particular, the bent crystals are suggested to use instead of the regular ones to separate monochromatic lines. On the Fig. 3b the spectrum after the reflection from the silicon crystal with 0.5° bending is shown. As it can be seen from the figure the increase of intensity is about three orders by magnitude for the characteristic X-ray line and is about 8 103 by magnitude for the line separated from bremsstrahlung. Since we have the increase of intensity there is a possibility to increase the rate and accuracy of the analysis as well as the decrease the current of X-ray tube that should lead to the prolongation of the tube lifetime proportionally. It is also important to note that in the case of using bent crystals the relative contribution of the background radiation decreases. In that case the signal-to-noise ratio is about 3 104 for the low energy of X-ray characteristic line and 2 103 for the high energy line separated from bremsstrahlung that is much higher compared to presented above analogues and shows the promising advantages of our technology.
4. Experimental investigation of the possibility to use polycapillary optics The polycapillary optics provides a possibility to increase the intensity of X-ray radiation registered in the experiment. This optics is based on the effect of X-ray channeling in thin capillary structures and allows redirecting X-ray propagation as the result new optics elements such as the X-ray lenses and semilenses have become reality [9,10]. Authors of Ref. [11] proposed to use this kind of optics in order to improve quality of the X-ray wave dispersive analysis. In our technology the idea to use polycapillary Kumakhov semilense allowing the transformation of divergent X-ray beam for example from X-ray tube into the quasiparallel one promises a high prospects. The last one increases the test beam intensity because of the lack of demands to the additional collimation and as the consequences increases the X-ray solid angle capture. Using the facilities of XLab LNF INFN laboratory we have performed the experimental study of the possibility to use the polycapillary semilens in our technology. The main goal of the experiment was a quantitative evaluation of increase of the diffracted radiation intensity having the energy close to the characteristic K-line radiation energy of Ag using the polycapillary semilens. The sketch of the experiment with semilense and without it is shown on Fig. 4. The radiation was generated by X-ray tube (Oxford Series 5000) having copper anode at the voltage 37.1 kV and passed through the semilens (the distance from semilens to anode was equal 59.6 mm that corresponds to the focal length of the semilens). In order to avoid the influence of radiation did not falling into the capture angle of semilens just in front of it a lead mask was installed having the hole of 3 mm diameter which is equal to the input semilens diameter. The length of semilens was 60 mm. Behind the semilens at the axis of the beam and the distance 100 mm an additional lead mask was installed with the 4 mm hole that corresponds to the output diameter of the semilens. This additional protection allowed decreasing the divergence of the beam for measurement without semilens when the whole experimental setup is the same with only difference that we have removed the semilens from the beam axis propagation. At the same time for the measurements with the semilens such protection almost does not influence the beam. The divergence of the beam after the lens equal to 1.4 mrad and the diameter of the single capillary is 5–8 lm. The lens was made by Unisantis. The beam generated by the experimental setup fell on the silicon crystal (1 0 0) installed under 11.7° to the beam propagation
Please cite this article in press as: A.S. Gogolev et al., WD-XRA technique in multiphase flow measuring, Nucl. Instr. Meth. B (2015), http://dx.doi.org/ 10.1016/j.nimb.2015.02.030
4
A.S. Gogolev et al. / Nuclear Instruments and Methods in Physics Research B xxx (2015) xxx–xxx
Fig. 3. (a) Radiation spectrum from a secondary target made of Ag and Cd alloy; (b) radiation spectrum after the reflection from the bent crystal, the degree bending is 0,5°.
Fig. 4. The sketch of the experiment (a) with semilens and (b) without semilens.
axis. The distance from mask to the center of the crystal was equal to 70 mm. The radiation with the energy 22.5 keV reflected from (4 0 0) planes and was registered by the detector installed at the distance 110 mm. For the radiation registration Vacuum X Ray SDD Spectrometer produced by XGLab [12] was used. The detector was capable to register the X-ray radiation with the energy up to 25 keV. The sensitive area of the detector is a circle and is equal to 30 mm2. One can evaluate the multiplication factor of the intensity increase using the simple relations based on increasing the angle capture which are applicable in discussed type of optics elements: 2
K amp ¼ ðXwith =Xwithout Þ ktr ðdwith Rwithout =dwithout Rwith Þ ktr ¼ 7:75 ktr ; where Xwith is the solid angle captured by the semilens; Xwithout is the solid angle cut by the second mask; dwith is the input diameter of the semilens equal to 3 mm; Rwith is the distance from the anode to the semilens and equal to 59 mm; dwithout is the diameter of the mask hole and equal to 4 mm; Rwithout is the distance from the anode to the second mask and equal to 219 mm, ktr is the transmission coefficient equal to the relation of the intensity passed through the lens and the intensity falling into the same solid angle but without semilens. This coefficient is the quantity defined for each optic element and depends on the energy of radiation. The typical value for lens used in our experiment is about 70% at K-lines of X-ray characteristic radiation of copper. Correspondingly, for the upper evaluated value of the expected increase the multiplication factor 5.45 might be used. On the Fig. 5a experimentally registered spectra of radiation are shown. As it can be seen without lens the significant broadening of the diffraction line occurred and it can be explained by the presence of the reflected bremsstrahlung falling on the crystal under
degrees differ than 11.7° with the close energy. This problem does not occur if we using the semilens and the beam is almost parallel. In order to remove the influence of the bremsstrahlung we performed measurements with additional collimation in front of detector – the vertical slit with the width 200 lm was installed. Further by sequential shift of the slit and at the same time performing the measurement of spectra we determined the slit position corresponding to the highest value of the diffracted radiation intensity. At that position we further performed measurement with and without semilens. Spectra of radiation for that case are shown on the Fig. 5b. One can find peaks parameters (full width at half maximum (FWHM) and integral intensity) in Table 1. As shown in Fig. 5b and Table 1 FWHM of the peaks 3 (with semilens) and 4 (without) coincide rather well each other. This fact allows us direct comparison of the measured peaks intensities. Thus the registered intensity multiplication factor for our experiment was about 3.6 which is about 1.5 less the expected value. The possible reason for the difference is the following: the lens used in the experiment is optimized to work with relatively low energy radiation (the energy corresponds to the K-line of characteristic X-ray radiation of copper). The transmission coefficient for the high energy radiation might be significant differ from the typical value we took. For example, the measured transmission coefficient for the whole spectrum of our X-ray tube at the voltage we used was equal to 56%. Using this value the evaluation of the expected multiplication factor gives 4.34 that closer to the registered value. However the measured transmission coefficient presents in fact averaged value for the whole spectrum of the tube at the given voltage, in reality for the particular radiation with the energy 22.5 keV the value of the coefficient might be even lower. Nonetheless, the experimentally measured value approaches to the performed theoretical evaluation quite close.
Please cite this article in press as: A.S. Gogolev et al., WD-XRA technique in multiphase flow measuring, Nucl. Instr. Meth. B (2015), http://dx.doi.org/ 10.1016/j.nimb.2015.02.030
A.S. Gogolev et al. / Nuclear Instruments and Methods in Physics Research B xxx (2015) xxx–xxx
5
Fig. 5. Spectra of radiation reflected from the crystal with and without lens (a) without collimator slit in front of the detector; (b) with collimator slit.
order about 22 keV potentially may provide higher increase of intensity.
Table 1 Parameters of the peaks. Peaks
Integral intensity, arb.u.
FWHM, channels
1 2 3 4
31,007 ± 292 43,468 ± 202 28,085 ± 147 7795 ± 49
147.9 ± 0.7 277.6 ± 0.6 133.7 ± 0.8 127.3 ± 0.9
On the other hand in the case of using optimized for particular energy of X-ray radiation lens one can expect that transmission coefficient will be about 60–70% and the amplification factor, correspondingly, will be higher. 5. Conclusion A new way of wave dispersive X-ray absorptiometry for multiphase fluid flow is suggested and the numerical simulation of X-ray spectra is performed. According to the results of simulation the suggested approach provides higher luminosity compared to the existing ones techniques and high multiplication factor by magnitude for the useful monochromatic radiation regarding to the background radiation. The integrated background radiation intensity does not exceed 1%. The results of the X-ray spectrum radiation simulation in the case of using bent crystals to provide the focusing shows that the intensity of radiation as well as the signal-to-noise ratio might be increased by several orders by magnitude. The facilities of capillary optics in the suggested technology are shown experimentally to provide the way to the test beam intensity increase by 3.6 times even with using the nonoptimized for work with studied energy radiation range polycapillary semilens. The optimization of the semilens to work with the energy radiation
Acknowledgment This work was partially supported by the grant of the Ministry of Education and Science of the Russian Federation within the framework of the Nauka Program. References [1] G. Falcone, Multiphase flow metering, Dev. Petr. Sci. 54 (2009) 191–228. [2] A.M. Scheers, W.F.J. Slijkerman, Multiphase flow measurement using multiple energy gamma ray absorption (MEGRA) composition measurement, SPE publication (1996) Document ID SPE-36593-MS. [3] R. Korkin, Patent RU 2477790 (2011) (in Russian), English version on http:// russianpatents.com/patent/247/2477790.html. [4] S.-A. Tjugum, Patent US 20120087467 (2012). [5] G. Harding, Patent US 5157704 (1992). [6] A.Y. Griaznov, Development of hardware and methodical ways to improve analytical parameters of energy dispersive XRF analyzer, Dissertation of Candidate of Technical Science (Ph.D.), St. Petersburg, 2004 (in Russian). [7] A. Gogolev, Yu. Cherepennikov, Device for X-ray spectral absorption analysis with use of acoustic monochromator, J. Phys: Conf. Ser. 517 (2014). Article number 012037. [8] Information on http://geant4.web.cern.ch/geant4/. [9] M.A. Kumakhov, X ray capillary optics: history of development and present status, in: M.A. Kumakhov (Ed.), Kumakhov Optics and Application: Selected Research Papers on Kumakhov Optics and Application of 1998–2000, 4155 of Proceedings of SPIE, 2000, pp. 2–12. [10] S. Dabagov, Channeling of neutral particles in micro- and nanocapillaries, Phys. Usp. 173 (10) (2003) 1053–1075. [11] J.P. Kirkland, V.E. Kovantsev, C.M. Dozier, J.V. Gilfrich, W.M. Gibson, Q.-F. Xiao, K. Umezawa, Wavelength-dispersive X-ray fluorescence detector, Rev. Sci. Instrum. 66 (1995) 1410–1412. [12] Information on http://www.xglab.it/vacuum-compatible-xray-spectrometer. shtml.
Please cite this article in press as: A.S. Gogolev et al., WD-XRA technique in multiphase flow measuring, Nucl. Instr. Meth. B (2015), http://dx.doi.org/ 10.1016/j.nimb.2015.02.030