Design study of a channel in the moderator on improving C and O measurement in coal

Design study of a channel in the moderator on improving C and O measurement in coal

Feiyue Shi,Jiayi Ma,Jingwu Zhao *,Shengkang Liu

Department of Physics,Nanjing University,Nanjing 210093,PR China

a r t i c l e i n f o Article history:

Received 22January 2009

Received in revised form 16April 2009Available online 19May 2009PACS:28.20.àv 82.80.Jp 89.30.Aa

Keywords:

Neutron-induced gamma Moderator Channel

Coal analysis

a b s t r a c t

By introducing a channel in the moderator of a coal analysis setup using neutron-induced gamma tech-nique,more fast neutrons can reach the coal.Then the characteristic gamma-rays of C and O are increased due to neutron inelastic scattering (NIS)reaction,and measurement of the two elements is suf?ciently improved.Monte-Carlo simulations have been carried out in this work for the setup with different prism channels.These prism channels have different positions and sizes,but their bottom surfaces are all nearby the neutron source.The results show that,it could obtain the optimum improvement of C and O measurement,by selecting a proper position and size for the prism channel.

ó2009Elsevier B.V.All rights reserved.

1.Introduction

As one of the most important resources in this world,coal has a number of parameters,such as calori?c value,volatile matter and ash content.These parameters determine the coal quality and are of important value in the coal industry.These parameters can be obtained by measuring chemical concentrations of mineral and organic matters in coal.

Neutron-induced gamma techniques have been used to mea-sure the elemental compositions of bulk materials [1].In these techniques,neutrons initiate characteristic gamma-rays promptly due to thermal neutron capture (TNC)and/or neutron inelastic scattering (NIS)reactions.Then,information about the chemical concentrations of the sample can be inferred by analyzing the emitted characteristic gamma-rays.

Prompt gamma neutron activation analysis (PGNAA)technique has been increasingly popular in multi-elemental analysis for the coal and cement industry [2–7].The radioisotope neutron sources of 252Cf have been commonly used in analyzers using PGNAA tech-nique.As the average neutron energy of the 252Cf source is low,these analyzers can only be used to measure elements of H,Si,Cl,Ca,Fe,etc.due to TNC reaction.

C and O are two key elements in coal.Measurement of these two elements is mainly due to NIS reaction induced by fast neutrons.Therefore,the neutron source which emits high energy neutrons is required,for example,the 241Am–Be neutron source or the 14MeV neutron generator.Neutron inelastic-scattering and thermal-capture analysis (NITA)technique,which utilizes a combination of TNC and NIS reactions using 241Am–Be sources,has been developed for measurements of coal and cement raw mill feed [8,9].In addition,a technique called pulsed fast-thermal neutron analysis (PFTNA)has been developed recently.By using a 14MeV neutron generator,the technique utilizes NIS,TNC and even neutron activation reactions synthetically [10–13].

A d-T neutron generator produces 14MeV monochromatic neutrons.But measurements of most elements in coal (e.g.H,Si,S,Ca,Fe,etc.)are due to TNC reaction induced by slow and thermal neutrons.Therefore,a moderator is required.By placing the moderator between the neutron source and the coal sample,more slow and thermal neutrons can be obtained at the coal loca-tion,and consequently the ef?ciency of TNC reaction is improved.However,the introduction of moderator will decrease the fast neutrons that reach the coal,and consequently impair C and O measurement.

By introducing a channel in the moderator,it can increase the fast neutrons that reach the coal,and then increase the NIS gam-ma-rays of C and O.Therefore,measurement of the two elements is improved.In this work,some different channels have been

0168-583X/$-see front matter ó2009Elsevier B.V.All rights reserved.doi:10.1016/j.nimb.2009.05.007

*Corresponding author.

E-mail address:jwzhao@https://www.360docs.net/doc/1516587328.html, (J.Zhao).

Nuclear Instruments and Methods in Physics Research B 267(2009)

2432–2436

Contents lists available at ScienceDirect

Nuclear Instruments and Methods in Physics Research B

journal homepage:www.els ev i e r.c o m /l o c a t e /n i m

b

studied by Monte-Carlo simulations,for the sake of ?nding a chan-nel that can obtain the optimum improvement of C and O measurement.2.Principle

In neutron-induced gamma techniques,yield of certain charac-teristic gamma-ray of certain element (N )has the following rela-tionship with neutron ?ux:

N /N A w h r u tj e ;

e1T

where N A is the Avogadro’s number,h is the isotopic abundance,w is the weight percent of the element in the sample,M is the atomic mass,r is the cross section,u is the neutron ?ux,t is the duration of measurement,j is the gamma-ray abundance per neutron reaction and e is the detection ef?ciency.

By introducing a channel in the moderator of a coal analysis set-up,u of fast neutrons is increased.Then,from Eq.(1),yields of characteristic gamma-rays (N )of C and O are increased due to NIS reaction induced by fast neutrons.Therefore,C and O measure-ment is improved.

3.Geometry of the channel

Fig.1shows the schematic diagram of the coal analysis setup used in this study.The setup utilizes a scattering geometry mode.Similar geometry with no channel in the moderator has been adopted for the on-line coal analysis setups by Nanjing Dalu Zhongdian Technology Joint-Stock Co.,Ltd.[14,15].The neutron source is a 14MeV neutron generator.Both the source and detector are below the coal and there is a lead shield plate between the source and detector.There are a polyethylene moderator and a rubber belt between the coal and the source/detector.The 14MeV neutrons emitted by the source are moderated in the mod-erator and belt,and then bombard the chemical elements in coal.Part of the gamma-rays induced by the neutrons can pass through the belt and moderator,and will be detected by the detector.

The purpose of introducing a channel in the moderator is to in-crease the characteristic gamma-rays of C and O based on NIS reac-tion.The position and size of the channel should be selected properly,in order to get the optimum improvement of C and O measurement,and meanwhile not impair measurements of other elements based on TNC reaction.

As to the shape of the channel,as shown in Figs.1and 2,the prism channel was selected in this work.The top and bottom sur-faces of the prism channel are rectangles,with length m in x -axis and width n in y -axis.Point A and O are centers of the top and bot-tom surfaces of the prism channel,respectively.Point O is above and very close to the neutron source.Point B is the point where

the central axis of the cylindrical detector meets the bottom sur-face of the moderator.Point B is above and very close to the detec-tor.Therefore,point A,B and O can be used to represent the positions of the top surface of the prism channel,the detector and the source,respectively.Point O is the origin of the coordinate system.Set Ax ,Ay and Az represent the coordinates of point A in x ,y and z -axis,respectively.Point A and O have the same value of coor-dinate in y -axis.And Az is equal to the thickness of the moderator.Three prism channels with same volume (same m and n )but dif-ferent position of the top surface (different Ax )are shown in Fig.3.In Channel (a),point A is above the neutron source.In Channel (b),point A is above the detector.In Channel (c),point A is between the position above the source and that above the https://www.360docs.net/doc/1516587328.html,parison is performed among the three channels as follows.For Channel (a),high energy neutrons emitted from the source are easy to reach the coal,because the distance between the coal and the source along the channel is the shortest.Therefore,there is a maximum increase in the fast neutrons at the coal location.However,there is a maxi-mum attenuation of the gamma-rays of C and O induced by those fast neutrons,because these gamma-rays need to cover the longest distance to arrive at the detector.For Channel (b),high energy neu-trons need to cover the longest distance to reach the coal along the channel.Therefore,there is a minimum increase in the fast neu-trons at the coal location.However,there is a minimum attenuation of the gamma-rays of C and O induced by those fast neutrons,be-cause these gamma-rays only need to travel the least distance to reach the detector.The increased fast neutrons and the attenuation of gamma-rays have contribution to the increased gamma-rays of C and O detected by the detector.Therefore,it may obtain an opti-mum effect on increasing the detected gamma-rays of C and O,if point A is selected between the position above the source and that above the detector (Channel (c)).4.Monte-Carlo simulations

Monte-Carlo simulations have been carried out in this work to study the prism channels in the moderator,to look for a

proper

Fig.1.Schematic diagram of the coal analysis setup used in this

work.

Fig.2.Schematic diagram of the prism channel in the

moderator.

Fig.3.Schematic diagram of three prism channels with same volume but different position of the top surface.

F.Shi et al./Nuclear Instruments and Methods in Physics Research B 267(2009)2432–24362433

channel that can obtain the optimum improvement of C and O measurement.Meanwhile,the effect of the channel on measure-ment of H and Si has also been studied.The geometry of the setup used for Monte-Carlo simulations is shown in Fig.1.

In simulations,the thicknesses of the coal,belt and moderator in z -axis are 30,0.5and 4.5cm,respectively.The length and width of the above three matters are both 100cm in x -axis and y -axis.The distances between point O and the two edges of the moderator in x -axis are 30and 70cm,respectively.The distances between point O and the two edges of the moderator in y -axis are both 50cm.The distance between the neutron source and the central axis of the detector is 39.5cm.The radius of the cylindrical detector is 7.5cm.The lead shield plate is 25cm ?20cm ?10cm in size.The concentration data of the coal sample were provided by Nan-jing Dalu Zhongdian Technology Joint-Stock Co.,Ltd.,and the data of some major elements are shown in Table 1.Besides C,O,H and Si,elements of N,S,Fe,Al,Ti,Na,K,Ca and Mg are also contained in the sample.The density of the coal sample is 0.698385g/cm 3.As shown in Figs.1–3,Ay =0cm,Az =4.5cm and n =20cm for all prism channels.The result of experiments has a good agree-ment with that of Monte-Carlo simulations,for the setup with no channel in the moderator in Nanjing Dalu Zhongdian Technol-ogy Joint-Stock Co.,Ltd.In this work,Monte-Carlo simulations were carried out for 11different cases about the channel:(1)case with no channel in the moderator (Case NC),(2)?ve cases of Group 1,with one prism channel in the moderator and m =10cm,values of Ax are 0,10,20,30and 39.5cm,respectively,(3)?ve cases of Group 2,with one prism channel in the moderator and m =20cm,values of Ax are 0,10,20,30and 39.5cm,respec-tively.The ?ve channels of Group 1(m =10cm)have a same vol-ume V 1.The ?ve channels of Group 2(m =20cm)also have a same volume V 2,and V 2>V 1.For each group,value of Ax varies from 0to 39.5cm,representing the top surface of the channel transferring from the position above the source to that above the detector.

The neutron source is set to be an isotropic point source,which emits 14MeV monochromatic neutrons in direction of 4p solid an-gles.The number of gamma-rays,which pass through the top sur-face of the cylindrical detector,is recorded as the gamma-ray yield.The energy range of recorded gamma-rays is 0–10MeV,with an interval of 0.01MeV.The detection ef?ciency of the detector is not under consideration (e =1).The number of neutrons,which pass through the top surface of the belt,is recorded as the neutron yield,in the case of replacing the coal with air.The energy range of recorded neutrons is 0–14.1MeV,with an interval of 0.1MeV.The number of neutron histories is 1.0?109.

In order to investigate the effect of the channel on the neutron or gamma-ray yield,a physical quantity p is de?ned as follows:

p ?

y ày eNC T

y eNC T

?100%;

e2T

where y represents the neutron or gamma-ray yield of a case with a channel in the moderator,and y(NC)represents corresponding yield of Case NC.Value of p represents the percentage variation of the yield,due to introduction of the channel.

For neutrons,p values have been calculated in the four energy regions of 0–14.1,6.0–14.1,14.0–14.1and 0–0.1MeV.Fast neu-trons of 6.0–14.1MeV have important contribution to production of the NIS gamma-rays of C and O.Neutrons of 0–0.1MeV are important to production of the TNC gamma-rays of H and Si.The energy region of 14.0–14.1MeV was selected to investigate the variation of the high energy neutrons,which are emitted from the neuron source and arrive at the coal location directly.The en-ergy region of 0–14.1MeV was selected to investigate the variation of the total neutrons.Corresponding results are shown in Fig.4(a)and (b)for cases of the two groups,respectively,as a function of Ax .For gamma-rays,the energy and reaction of the prompt gam-ma-rays of C,O,H and Si used in this work are listed in Table 2.Corresponding p values of the gamma-rays are shown in Fig.5(a)and (b)for cases of the two groups,respectively,as a function of Ax .

Table 1

Concentrations of the major elements in the coal sample.Element Weight (%)C 72.08O 12.01H 4.57Si 4.22Else

7.12

Fig.4.The p values of the neutron yield:(a)p values of cases of Group 1(m =10cm)and (b)p values of cases of Group 2(m =20cm).

Table 2

Characteristic gamma-rays of C,O,H and Si used in this study and their corresponding reactions.Element E c (MeV)Reaction C 4.44NIS O 6.13NIS H 2.23TNC Si

3.54

TNC

2434 F.Shi et al./Nuclear Instruments and Methods in Physics Research B 267(2009)2432–2436

For the sake of investigating the effect on fast and slow neutron yields,gamma-ray yields of C,O and H,and their relative variations due to introduction of the channel,three types of ratios have been calculated as follows.The ratio of the neutron yield of 6.0–14.1MeV to that of0–0.1MeV is calculated to represent the fast/ slow neutron yield ratio.The ratio of the gamma-ray yield of C to that of H,and the ratio of the gamma-ray yield of O to that of H are calculated to represent the gamma-ray yield ratios of C/H and O/H,respectively.

In order to investigate the effect of the channel on the ratio (fast/slow,C/H or O/H),a physical quantity q is de?ned as follows:

q?ràreNCT

reNCT

?100%;e3T

where r represents the ratio of a case with a channel in the moder-ator,and r(NC)represents corresponding ratio of Case NC.Corre-sponding q values are shown in Fig.6(a)and(b)for cases of the two groups,respectively,as a function of Ax.

5.Results and discussions

From the positive and negative p values in Fig.4,it can be seen that,the neutron yields are increased in energy regions of0–14.1,6.0–14.1and14.0–14.1MeV,but decreased in energy region of0–0.1MeV,by introducing the channel in the moderator.In cases of each group,the absolute value of p reduces with value of Ax vary-ing from0to39.5cm,for the neutron yields in energy regions of0–14.1,6.0–14.1,14.0–14.1and0–0.1MeV.The big channel of Group 2(volume V2)results in a bigger variation of the neutron yield than the small channel of Group1(volume V1).

It is seen from Fig.5that,all p values of the two groups are po-sitive.This indicates that,all gamma-ray yields of C,O,H and Si are increased by introducing the channel in the moderator in this work.The cause of the increase in gamma-ray yields of C and O has been mentioned in Sections2and3.The increase in gamma-ray yields of H and Si can be suggested to be due to the further moderation of the increased fast neutrons at the coal location. More slow and thermal neutrons are produced in the coal due to the further moderation of those fast neutrons,and then more TNC gamma-rays are initiated.The gamma-ray yields of C and O have much larger increased variations than those of H and Si, and corresponding p values of C and O can be higher than90%.

In addition,for cases of each group,the gamma-ray yields of C, O,H and Si ascend?rstly,and then descend,along with value of Ax varying from0to39.5cm.Therefore,for each element of each group,there is a value of Ax,by selecting which the gamma-ray yield is the maximum.For cases of Group1,the maximum

gam-Fig.5.The p values of the gamma-ray yield:(a)p values of cases of Group1(m=10cm)and(b)p values of cases of Group2(m=20

cm).

Fig.6.The q values of the ratio for cases of Group1(m=10cm)and Group2(m=20cm):(a)q values of the fast/slow neutron yield ratio and(b)q values of the gamma-ray yield ratio of C/H and O/H.

F.Shi et al./Nuclear Instruments and Methods in Physics Research B267(2009)2432–24362435

ma-ray yields of C and O appear in the cases with Ax=20cm and with Ax=10cm,respectively.For cases of Group2,both the max-imum gamma-ray yields of C and O appear in the case with Ax=20cm.Among the ten cases of the two groups,both the max-imum gamma-ray yields of C and O appear in the case with Ax=20cm and m=20cm.

In comparison with the two groups,it can be seen that,an in-crease in volume of the channel results in a small increase in the gamma-ray yields of H and Si,but a big increase in the gamma-ray yields of C and O.Hence,the channel with a bigger volume has a better improvement of C and O measurement.In this work, measurement of H and Si is not impaired but even improved a lit-tle,although the yield of low energy neutrons at the coal location is decreased by introducing the channel.

It is seen from Fig.6that,all values of q are positive.This indi-cates that,all values of ratio(fast/slow,C/H and O/H)are increased by introducing the channel in the moderator.For cases of each group,the fast/slow neutron yield ratio is observed to descend with Ax.For cases of each group,similar to Fig.5,the gamma-ray yield ratios of C/H and O/H ascend?rstly,and then descend,along with value of Ax varying from0to39.5cm.For each group,there is a value of Ax,by selecting which the gamma-ray yield ratio of C/H or O/H is the maximum.For cases of Group1,both the maximum gamma-ray yield ratios of C/H and O/H appear in the case with Ax=10cm.For cases of Group2,the maximum gamma-ray yield ratios of C/H and O/H appear in the cases with Ax=20cm and with Ax=10cm,respectively.This is a little different from results of the gamma-ray yields of C and O shown in Fig.5,because the gamma-ray yield of H is also increased by introducing the channel,and the penetration ability of the gamma-rays of H is different from those of the gamma-rays of C and O.

6.Conclusion

By introducing a channel in the moderator,C and O measure-ment could be improved for a coal analysis setup.Cases with different prism channels have been studied by Monte-Carlo simu-lations in this work.The results show that,it could obtain an opti-mum improvement of C and O measurement,by selecting a proper position and size for the channel.The results also show that,it could even improve measurement of H and Si based on TNC reac-tion,by introducing the channel.Although the simulated model was simple and some complicated factors were not considered, this work has provided useful information about improving mea-surement of coal in practical industrial applications. Acknowledgement

The authors would like to thank Wenbao Jia of Nanjing Dalu Zhongdian Technology Joint-Stock Co.,Ltd.for his help and advice. References

[1]C.S.Lim,J.Radioanal.Nucl.Chem.262(2)(2004)525.

[2]M.Borsaru,Z.Jecny,Appl.Radiat.Isot.54(2001)519.

[3]Chen Boxian,He Jingye,Liu Jiancheng,Nucl.Electron.Detect.Technol.16(1)

(1996)6(in Chinese).

[4]M.Borsaru,M.Berry,M.Biggs,A.Rojc,Nucl.Instr.and Meth.B213(2004)530.

[5]M.Borsaru,M.Biggs,W.Nichols,F.Bos,Appl.Radiat.Isot.54(2001)335.

[6]A.A.Naqvi,M.A.Garwan,Nucl.Instr.and Meth.B215(2004)283.

[7]A.A.Naqvi,M.M.Nagadi,S.Kidwai,Khateebur-Rehman,M.Maslehuddin,Nucl.

Instr.and Meth.A533(2004)591.

[8]C.S.Lim,D.A.Abernethy,Appl.Radiat.Isot.63(2005)697.

[9]C.S.Lim,J.R.Tickner,B.D.Sowerby,D.A.Abernethy,A.J.McEwan,S.Rainey,R.

Stevens,C.Manias,D.Retallack,Appl.Radiat.Isot.54(2001)11.

[10]L.Dep,M.Belbot,G.Vourvopoulos,S.Sudar,J.Radioanal.Nucl.Chem.234

(1–2)(1998)107.

[11]Jing Shi-Wei,Liu Lin-Mao,Gu De-Shan,Qiao Shuang,Sang Hai-Feng,Zhang

Yong-Xiang,Zhang Zhong-Hua,Cao Xi-Zheng,Tian Yu-Bing,Chi Yan-Tao,Zhao Xin-Hui,Nucl.Sci.Technol.14(4)(2003)265.

[12]Sang Hai-feng,Wang Fu-li,Liu Lin-mao,Sang Hai-jun,Nucl.Instr.and Meth.B

239(2005)202.

[13]P.A.Dokhale,J.Csikai,L.Olah,Appl.Radiat.Isot.54(2001)967.

[14]Cui Guo-sheng,Li Jin,Xu Jun-wei,Power Syst.Eng.21(4)(2005)45

(in Chinese).

[15]Jia Wenbao,Chen Xiaowen,Xu Aiguo,J.Tsinghua Univ.(Sci.Technol.)47(S1)

(2007)960(in Chinese).

2436 F.Shi et al./Nuclear Instruments and Methods in Physics Research B267(2009)2432–2436