- Email: [email protected]
Neutron diffractometer RSND for residual stress analysis at CAEP
Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Neutron diffractometer RSND for residual stress analysis at CAEP Jian Li, Hong Wang, Guangai Sun n, Bo Chen, Yanzhou Chen, Beibei Pang, Ying Zhang, Yun Wang, Changsheng Zhang, Jian Gong, Yaoguang Liu Key Laboratory of Neutron Physics and Institute of Nuclear Physics and Chemistry, China Academy of Engineering Physics, Mianyang 621999, China
art ic l e i nf o
a b s t r a c t
Article history: Received 5 June 2014 Received in revised form 11 January 2015 Accepted 10 February 2015 Available online 19 February 2015
Residual Stress Neutron Diffractometer (RSND) has been built at China Academy of Engineering Physics (CAEP) in Mianyang. Due to its excellent flexibility, the residual stress measurement on different samples, as well as in-situ study for materials science, can be carried out through RSND. The basic tests on its intensity and resolution and some preliminary experimental results under mechanical load, demonstrate the high quality of RSND. & 2015 Elsevier B.V. All rights reserved.
Keywords: Residual stress Neutron diffraction Material science
1. Introduction Neutron diffraction stress analysis is one of the most powerful techniques for measuring nondestructively internal stress in three dimensions deeply inside polycrystalline materials [1]. This technique can be used to precisely measure a lattice spacing based on Bragg’s law: λ ¼ 2d sin θ;
ð1Þ
where λ is the wavelength of neutron beam, d and θ are the lattice spacing and the Bragg angle (or a half of the diffraction angle 2θ) corresponding to a particular reflection plane (h k l). The elastic strain is determined by Hooke’s low considering the variation of the lattice spacing: ε ¼(d d0)/d0 ¼sinθ0/sinθ 1, where d0 and θ0 are the lattice spacing and corresponding Bragg angle under the stress-free condition. Besides the residual stress measurement, neutron diffraction is also a powerful tool for the in-situ study of materials science under thermo-mechanical loading conditions. The new neutron diffractometer RSND at CAEP was designed with high-level automation and high-sample adaptation for residual stress measurements on various engineering specimens and in-situ studies for crystal materials.
2. Instrument description RSND is located at the end of a thermal neutron guide, layout of which is shown in Fig. 1. The main components of RSND diffractometer are listed in Table 1. The construction of RSND was finished n
Correspondence author. Tel.: þ 86 816 2493337. E-mail address: [email protected] (G. Sun).
http://dx.doi.org/10.1016/j.nima.2015.02.026 0168-9002/& 2015 Elsevier B.V. All rights reserved.
on September 3rd, 2012. A double focusing silicon single crystal Si monochromator is employed to ensure the high intensity and resolution with the accessible wavelengths from 0.12 nm to 0.28 nm, which satisfy measurements of various lattice spacing [2]. The diffractometer resolution Δd/d is about 1.91 10 3, which is obtained by a standard Fe-sample experiment. The Si(3 1 1) monochromator is set at the take off angle 2θM ¼901 to provide the neutron wavelength λ ¼0.231 nm. The Fe(1 1 0) peak is gained on the diffraction angle of 69.441, which is shown in Fig. 2. The uniquely designed sample table can handle the mechanical load up to 500 kg with high position accuracy. A MK-200N twodimensional position sensitive detector (2D-PSD) is used with the active area of 200 200 mm2 and a spatial resolution of about 1.8 1.8 mm2. The calibrated efficiency of this detector is 72.1% at wavelength λ¼0.231 nm. An oscillating radial collimator of 200 mm length is positioned in front of the 2D PSD. The oscillating radial collimator, which is made from gadolinium oxide (50 μm) and epoxy resin (100 μm), is optimized for 1100 mm sample-detector distance to improve the measurement accuracy, the angle between foils of the radial collimator is 0.2171. The oscillating time of periods (1/f) is set to 30 s, and the oscillating range is 711, which can efficiently decreases the background, and disturb the peak intensity little. Due to the flexible mechanical movement with an air-cushion mode, the incident slit-sample, the monochromator-sampledetector and the detector slit-sample distances are adjustable, which enable the RSND to optimize the measurement set-up according to the sample species and size. Gauge volume location is gained by the slits system. The laser and the high-resolution digital camera is employed for the optics alignment, detailed instrument configuration of this part is shown in Fig. 3. Intensity measurement was carried out with a 235U fission chamber and a high-efficiency (96% for the neutron wavelength¼0.18 nm) 3He
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
77
Fig. 1. RSND with high-level automation and high-sample adaptation at CAEP.
Table 1 Specifications of RSND diffractometer. Model
Constant wavelength
Monochromator Reflected plane Monochromator take-off angle (2θM) Wavelength range Diffraction angle (2θ) Neutron flux at sample position Resolution of spectrometer Gauge volume Sample stage Maximum loading Translation X–Y Elevation Z Rotation around Z axis Tilt angle along to X axis Movement ranges Monochromator-sample distance Incident slit-sample distance Detector slit-sample distance Detector-sample distance Detector (MK-200N PSD) Active area Spatial resolution Detection efficiency Sample environment Stress rig Furnace Eulerian cradle (texture measurement)
Double focusing Si bent perfect crystal Si(3 1 1), Si(5 1 1) 50–1201 0.12–0.28 nm 0–1401, accuracy 7 0.011 4.7 106 n/cm2/s (λ¼ 0.158 nm) Δd/d 1.91 10 3 (λ¼ 0.231 nm) 0.5 0.5 0.5–5 5 10 mm3 500 kg 7 300 mm, Positioning accuracy 70.05 mm 500 mm, Positioning accuracy 70.1 mm 0–3601 7 301 For 30 kg 1600–2400 mm 0–1130 mm 0–1200 mm 800–1500 mm 200 200 mm 1.8 1.8 mm2 72.1% For wavelength of 0.231 nm 0–15 kN, Accuracy 7 10 N, bidirectional uniaxial tension or compression 25–500 1C, Accuracy 73 1C Under construction
counter. The neutron flux on sample position is 4.7 106 n/cm2/s at wavelength λ¼ 0.158 nm with the reactor power of 20 MW. We sum specifications of RSND in Table 1.
3. Strain measurement A round-robin strain test has been carried out by RSND with the aluminum ring-and-plug sample from VAMAS (Versailles project on Advanced Materials and Standards) [3], as shown in Fig. 4. In this experiment, 0.172 nm neutron beam was chosen by the Si(3 1 1) monochromator with the take-off angle 2θM ¼ 63.51. The Al(3 1 1) peak locates at the diffraction angle of about 89.51.
By setting the incident slit and diffracted slit as 2 2 mm2, a cubic gauge volume of 2 2 2 mm3 was obtained to provide a good spatial resolution. As shown in Fig. 5, the hoop strain distribution of the aluminum ring-and-plug sample measured by RSND is consistent with the previous experimental data of the same sample by Kowari [4] at ANSTO (Australia Nuclear Science and Technology Organization).
4. In-situ experiment To carry out the in-situ experiment with particular environmental conditions, a stress rig has been designed to provide
78
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
uniaxial tension and compression load up to 15 kN. The tensile or compression rate can be set from 0.2 mm/min to 20 mm/min. A furnace suited for this stress rig is also designed to provide high temperature up to 500 1C with control precision of 7 3 1C and heating rate of 30 1C/min. The cooling process is controlled by natural air. By combining the sample table and goniometer with the aforementioned environmental equipment, the RSND measurement can access different stresses, temperatures, rotation angles and coupling conditions. And the bidirectional tensile loading function of the stress rig avoids the offset in measurement points. With the in-situ neutron diffraction, the stressinduced phase transformation of Ni50.4Ti49.6 (at%) alloy at 67 1C has been quantitatively studied with a strain rate of 0.6 mm/min. Fig. 6 presents the peak evolution of B2 (austenite) and B19’ (martensite) phase related to the applied stress. With strain and stress zero, a mixed phase with both B2(austenite) and B190 (martensite) structures is observed in the Ni50.4Ti49.6 (at%) alloy. As increasing applied stress, the stress-induced transformation from B2 to B190 occurs with the obvious intensity decreasing of B2(1 1 0) peak. On the other hand, peak intensities of B190 (1 1 1) and B190 (0 0 2) also decrease, which is inconsistent with the process of normal phase transformation. There may be a strong texture evolution during the tension of NiTi alloy [5]. The B190 (1 1 1) and B190 (0 0 2) peak intensities decrease in the loading direction, while increase in the direction perpendicular to the loading direction. In this experiment, we only measured the crystal planes parallel to the loading direction. Therefore, the intensity decreasing of B190 (1 1 1) and B190 (0 0 2) can be explained by the texture evolution.
5. Conclusions RSND neutron diffractometer is now available at CAEP for residual stress analysis and microstructure characterization. Its automation with air-cushion and sample environments with temperature and stress loading provide the most flexibility for various measurement requirements. A series of experiments and tests proved the stability and quality of the RSND. Especially, a VAMAS round-robin strain test on RSND is consistent with the results obtained from Kowari at Australia. As one part of the planned upgrade, an in-situ texture measurement system will be equipped in the future.
Fig. 4. round-robin strain test by RSND.
400 Neutron Intensity(/min) Gaussian Fit Fe(110) sample diameter=3.85mm Slit before sample:2mm×2mm Slit after sample:2mm×10mm Δd/d=δθ/tanθ=1.91×10 FWHM=0.153°
250 200 150 100 50 0 68.0
RSND Kowari
1000 -6
300
1200
Hoop Microstrain( X 10 )
Neutron Intensity(/min)
350
800 600 400 200 0 -200 -400 -30
68.5
69.0
69.5
70.0
70.5
-20
71.0
Diffraction angle(2θ) Fig. 2. The diffractometer resolution Δd/d certificated by a Fe-scanning experiment.
-10
0
20
30
Position(mm) Fig. 5. Comparison between hoop microstrains of the same VAMAS sample measured by RSND and Kowari.
Digital camera
Laser
10
Diffracted slit Incident slit
Fig. 3. Slits for gauge volume and optics alignment system.
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
79
Physics (Grant no. 20141024) is acknowledged. We thank the Kowari team of the Bragg Institute in ANSTO of Australia for providing the VAMAS data. The fruitful discuss with Prof. Erdong Wu and Prof. Thomas Gyula is greatly appreciated.
References [1] M.T. Hutchings, P.J. Withers, T.M. Holden, et al., Introduction to the Characterization of Residual Stress by Neutron Diffraction, Taylor & Francis, 2005. [2] M. Pavol, V. Miroslav, F. Michihiro, et al., Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 529 (1-3) (2004) 138. [3] G.A. Webster, Neutron Diffraction Measurements of Residual Stress in a Shrinkfit Ring and Plug. VAMAS TWA 20, 2000. [4] Alain Brule, Oliver Kirstein, Physica B: Condensed Matter 385-386 (2006) 1040. [5] M. Hasan, W.W. Schmahl, K. Hackl, et al., Materials Science and Engineering A 481-482 (2008) 414.
° Fig. 6. Neutron diffraction results of NiTi alloy at different loading states (at 67 1C), the crystal planes are perpendicular to the loading direction.
Acknowledgement The financial support from the National Natural Science Foundation of China (Grants 11105128, 91126001, 51231002, 51401193) and the President Foundation of China Academy of Engineering
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Neutron diffractometer RSND for residual stress analysis at CAEP Jian Li, Hong Wang, Guangai Sun n, Bo Chen, Yanzhou Chen, Beibei Pang, Ying Zhang, Yun Wang, Changsheng Zhang, Jian Gong, Yaoguang Liu Key Laboratory of Neutron Physics and Institute of Nuclear Physics and Chemistry, China Academy of Engineering Physics, Mianyang 621999, China
art ic l e i nf o
a b s t r a c t
Article history: Received 5 June 2014 Received in revised form 11 January 2015 Accepted 10 February 2015 Available online 19 February 2015
Residual Stress Neutron Diffractometer (RSND) has been built at China Academy of Engineering Physics (CAEP) in Mianyang. Due to its excellent flexibility, the residual stress measurement on different samples, as well as in-situ study for materials science, can be carried out through RSND. The basic tests on its intensity and resolution and some preliminary experimental results under mechanical load, demonstrate the high quality of RSND. & 2015 Elsevier B.V. All rights reserved.
Keywords: Residual stress Neutron diffraction Material science
1. Introduction Neutron diffraction stress analysis is one of the most powerful techniques for measuring nondestructively internal stress in three dimensions deeply inside polycrystalline materials [1]. This technique can be used to precisely measure a lattice spacing based on Bragg’s law: λ ¼ 2d sin θ;
ð1Þ
where λ is the wavelength of neutron beam, d and θ are the lattice spacing and the Bragg angle (or a half of the diffraction angle 2θ) corresponding to a particular reflection plane (h k l). The elastic strain is determined by Hooke’s low considering the variation of the lattice spacing: ε ¼(d d0)/d0 ¼sinθ0/sinθ 1, where d0 and θ0 are the lattice spacing and corresponding Bragg angle under the stress-free condition. Besides the residual stress measurement, neutron diffraction is also a powerful tool for the in-situ study of materials science under thermo-mechanical loading conditions. The new neutron diffractometer RSND at CAEP was designed with high-level automation and high-sample adaptation for residual stress measurements on various engineering specimens and in-situ studies for crystal materials.
2. Instrument description RSND is located at the end of a thermal neutron guide, layout of which is shown in Fig. 1. The main components of RSND diffractometer are listed in Table 1. The construction of RSND was finished n
Correspondence author. Tel.: þ 86 816 2493337. E-mail address: [email protected] (G. Sun).
http://dx.doi.org/10.1016/j.nima.2015.02.026 0168-9002/& 2015 Elsevier B.V. All rights reserved.
on September 3rd, 2012. A double focusing silicon single crystal Si monochromator is employed to ensure the high intensity and resolution with the accessible wavelengths from 0.12 nm to 0.28 nm, which satisfy measurements of various lattice spacing [2]. The diffractometer resolution Δd/d is about 1.91 10 3, which is obtained by a standard Fe-sample experiment. The Si(3 1 1) monochromator is set at the take off angle 2θM ¼901 to provide the neutron wavelength λ ¼0.231 nm. The Fe(1 1 0) peak is gained on the diffraction angle of 69.441, which is shown in Fig. 2. The uniquely designed sample table can handle the mechanical load up to 500 kg with high position accuracy. A MK-200N twodimensional position sensitive detector (2D-PSD) is used with the active area of 200 200 mm2 and a spatial resolution of about 1.8 1.8 mm2. The calibrated efficiency of this detector is 72.1% at wavelength λ¼0.231 nm. An oscillating radial collimator of 200 mm length is positioned in front of the 2D PSD. The oscillating radial collimator, which is made from gadolinium oxide (50 μm) and epoxy resin (100 μm), is optimized for 1100 mm sample-detector distance to improve the measurement accuracy, the angle between foils of the radial collimator is 0.2171. The oscillating time of periods (1/f) is set to 30 s, and the oscillating range is 711, which can efficiently decreases the background, and disturb the peak intensity little. Due to the flexible mechanical movement with an air-cushion mode, the incident slit-sample, the monochromator-sampledetector and the detector slit-sample distances are adjustable, which enable the RSND to optimize the measurement set-up according to the sample species and size. Gauge volume location is gained by the slits system. The laser and the high-resolution digital camera is employed for the optics alignment, detailed instrument configuration of this part is shown in Fig. 3. Intensity measurement was carried out with a 235U fission chamber and a high-efficiency (96% for the neutron wavelength¼0.18 nm) 3He
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
77
Fig. 1. RSND with high-level automation and high-sample adaptation at CAEP.
Table 1 Specifications of RSND diffractometer. Model
Constant wavelength
Monochromator Reflected plane Monochromator take-off angle (2θM) Wavelength range Diffraction angle (2θ) Neutron flux at sample position Resolution of spectrometer Gauge volume Sample stage Maximum loading Translation X–Y Elevation Z Rotation around Z axis Tilt angle along to X axis Movement ranges Monochromator-sample distance Incident slit-sample distance Detector slit-sample distance Detector-sample distance Detector (MK-200N PSD) Active area Spatial resolution Detection efficiency Sample environment Stress rig Furnace Eulerian cradle (texture measurement)
Double focusing Si bent perfect crystal Si(3 1 1), Si(5 1 1) 50–1201 0.12–0.28 nm 0–1401, accuracy 7 0.011 4.7 106 n/cm2/s (λ¼ 0.158 nm) Δd/d 1.91 10 3 (λ¼ 0.231 nm) 0.5 0.5 0.5–5 5 10 mm3 500 kg 7 300 mm, Positioning accuracy 70.05 mm 500 mm, Positioning accuracy 70.1 mm 0–3601 7 301 For 30 kg 1600–2400 mm 0–1130 mm 0–1200 mm 800–1500 mm 200 200 mm 1.8 1.8 mm2 72.1% For wavelength of 0.231 nm 0–15 kN, Accuracy 7 10 N, bidirectional uniaxial tension or compression 25–500 1C, Accuracy 73 1C Under construction
counter. The neutron flux on sample position is 4.7 106 n/cm2/s at wavelength λ¼ 0.158 nm with the reactor power of 20 MW. We sum specifications of RSND in Table 1.
3. Strain measurement A round-robin strain test has been carried out by RSND with the aluminum ring-and-plug sample from VAMAS (Versailles project on Advanced Materials and Standards) [3], as shown in Fig. 4. In this experiment, 0.172 nm neutron beam was chosen by the Si(3 1 1) monochromator with the take-off angle 2θM ¼ 63.51. The Al(3 1 1) peak locates at the diffraction angle of about 89.51.
By setting the incident slit and diffracted slit as 2 2 mm2, a cubic gauge volume of 2 2 2 mm3 was obtained to provide a good spatial resolution. As shown in Fig. 5, the hoop strain distribution of the aluminum ring-and-plug sample measured by RSND is consistent with the previous experimental data of the same sample by Kowari [4] at ANSTO (Australia Nuclear Science and Technology Organization).
4. In-situ experiment To carry out the in-situ experiment with particular environmental conditions, a stress rig has been designed to provide
78
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
uniaxial tension and compression load up to 15 kN. The tensile or compression rate can be set from 0.2 mm/min to 20 mm/min. A furnace suited for this stress rig is also designed to provide high temperature up to 500 1C with control precision of 7 3 1C and heating rate of 30 1C/min. The cooling process is controlled by natural air. By combining the sample table and goniometer with the aforementioned environmental equipment, the RSND measurement can access different stresses, temperatures, rotation angles and coupling conditions. And the bidirectional tensile loading function of the stress rig avoids the offset in measurement points. With the in-situ neutron diffraction, the stressinduced phase transformation of Ni50.4Ti49.6 (at%) alloy at 67 1C has been quantitatively studied with a strain rate of 0.6 mm/min. Fig. 6 presents the peak evolution of B2 (austenite) and B19’ (martensite) phase related to the applied stress. With strain and stress zero, a mixed phase with both B2(austenite) and B190 (martensite) structures is observed in the Ni50.4Ti49.6 (at%) alloy. As increasing applied stress, the stress-induced transformation from B2 to B190 occurs with the obvious intensity decreasing of B2(1 1 0) peak. On the other hand, peak intensities of B190 (1 1 1) and B190 (0 0 2) also decrease, which is inconsistent with the process of normal phase transformation. There may be a strong texture evolution during the tension of NiTi alloy [5]. The B190 (1 1 1) and B190 (0 0 2) peak intensities decrease in the loading direction, while increase in the direction perpendicular to the loading direction. In this experiment, we only measured the crystal planes parallel to the loading direction. Therefore, the intensity decreasing of B190 (1 1 1) and B190 (0 0 2) can be explained by the texture evolution.
5. Conclusions RSND neutron diffractometer is now available at CAEP for residual stress analysis and microstructure characterization. Its automation with air-cushion and sample environments with temperature and stress loading provide the most flexibility for various measurement requirements. A series of experiments and tests proved the stability and quality of the RSND. Especially, a VAMAS round-robin strain test on RSND is consistent with the results obtained from Kowari at Australia. As one part of the planned upgrade, an in-situ texture measurement system will be equipped in the future.
Fig. 4. round-robin strain test by RSND.
400 Neutron Intensity(/min) Gaussian Fit Fe(110) sample diameter=3.85mm Slit before sample:2mm×2mm Slit after sample:2mm×10mm Δd/d=δθ/tanθ=1.91×10 FWHM=0.153°
250 200 150 100 50 0 68.0
RSND Kowari
1000 -6
300
1200
Hoop Microstrain( X 10 )
Neutron Intensity(/min)
350
800 600 400 200 0 -200 -400 -30
68.5
69.0
69.5
70.0
70.5
-20
71.0
Diffraction angle(2θ) Fig. 2. The diffractometer resolution Δd/d certificated by a Fe-scanning experiment.
-10
0
20
30
Position(mm) Fig. 5. Comparison between hoop microstrains of the same VAMAS sample measured by RSND and Kowari.
Digital camera
Laser
10
Diffracted slit Incident slit
Fig. 3. Slits for gauge volume and optics alignment system.
J. Li et al. / Nuclear Instruments and Methods in Physics Research A 783 (2015) 76–79
79
Physics (Grant no. 20141024) is acknowledged. We thank the Kowari team of the Bragg Institute in ANSTO of Australia for providing the VAMAS data. The fruitful discuss with Prof. Erdong Wu and Prof. Thomas Gyula is greatly appreciated.
References [1] M.T. Hutchings, P.J. Withers, T.M. Holden, et al., Introduction to the Characterization of Residual Stress by Neutron Diffraction, Taylor & Francis, 2005. [2] M. Pavol, V. Miroslav, F. Michihiro, et al., Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 529 (1-3) (2004) 138. [3] G.A. Webster, Neutron Diffraction Measurements of Residual Stress in a Shrinkfit Ring and Plug. VAMAS TWA 20, 2000. [4] Alain Brule, Oliver Kirstein, Physica B: Condensed Matter 385-386 (2006) 1040. [5] M. Hasan, W.W. Schmahl, K. Hackl, et al., Materials Science and Engineering A 481-482 (2008) 414.
° Fig. 6. Neutron diffraction results of NiTi alloy at different loading states (at 67 1C), the crystal planes are perpendicular to the loading direction.
Acknowledgement The financial support from the National Natural Science Foundation of China (Grants 11105128, 91126001, 51231002, 51401193) and the President Foundation of China Academy of Engineering