pith. sign in

arxiv: 2606.06775 · v1 · pith:TDLQDFCTnew · submitted 2026-06-04 · ⚛️ physics.acc-ph

Manufacturability studies for the FCC-ee positron source target: determination of the minimum bending radius and ovalization in tantalum cooling tube elbows

Ramiro Mena-Andrade , Mickaell Crouvizier , Jean-Philippe Rigaud , Thibaut Coiffet , Antonio Perillo-Marcone This is my paper

Pith reviewed 2026-06-27 22:20 UTC · model grok-4.3

classification ⚛️ physics.acc-ph
keywords tantalumcooling tubescompression bendingbending radiusovalizationpositron sourcenumerical modelingexperimental validation
0
0 comments X

The pith

A minimum bending radius of 10 mm works for manufacturing tantalum cooling tube elbows with ovalization limited to 0.5 mm.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The study creates a numerical model to guide compression bending of small-diameter tantalum tubes for cooling a compact positron source target. Calculations show a 10 mm radius is acceptable, with ovalization producing shape changes of about 0.5 mm that deviate noticeably from the intended geometry. Experiments, including cross-section micrographs, confirm the model predictions and the tube's structural soundness after bending. The work also supplies a simple formula to find tantalum yield stress from hardness tests. Such manufacturing guidance matters for embedding effective cooling channels inside beam-intercepting components only a few centimeters in size.

Core claim

The central discovery is that the minimum bending radius for the tantalum cooling tubes is 10 mm, as determined by the numerical model. This radius produces ovalization distortion within ±0.5 mm, which constitutes an important deviation from the nominal geometry. The model was validated through experimental tests using both non-destructive and destructive methods, and micrographs of the tube cross-sections before and after bending confirm that the elbow maintains its integrity. An empirical expression is also proposed to estimate the yield stress of pure tantalum from Vickers hardness measurements. The numerical model predicts ovalization along the elbow and offers a practical method to spec

What carries the argument

Four-step numerical methodology for determining minimum bending radius in thick-walled tubes via compression bending: allowable radius calculation, constitutive law validation, ovalization prediction, and experimental confirmation.

If this is right

  • The 10 mm radius elbow can be produced without loss of tube integrity.
  • Ovalization must be incorporated into the final cooling channel layout.
  • The validated model serves as an alternative for specifying tube geometries in other beam devices.
  • Hardness measurements can substitute for direct yield stress tests in tantalum.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This bending approach may apply to cooling designs in other compact high-energy physics apparatus.
  • Early inclusion of ovalization effects in simulations could minimize redesign cycles for accelerator components.
  • Generalizing the model to varying tube thicknesses or materials would broaden its utility for custom cooling systems.

Load-bearing premise

The material model for tantalum correctly describes its behavior under the large strains and contact forces present in compression bending.

What would settle it

If bending a tube to 10 mm radius produces ovalization greater than 0.5 mm or visible damage in cross-section micrographs, the claimed minimum radius would be invalidated.

Figures

Figures reproduced from arXiv: 2606.06775 by Antonio Perillo-Marcone, Jean-Philippe Rigaud, Mickaell Crouvizier, Ramiro Mena-Andrade, Thibaut Coiffet.

Figure 1
Figure 1. Figure 1: FCC-ee positron source target geometry: (a) external, (b) tangential and c) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Exploded view of the FCC-ee positron source target (without shielding). [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Compression bending: (a) Process schematic representation and (b) Evolution [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Minimum bending radius estimation process: (a) Step 1: Definition of the D/t [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Ta tensile test: (a) sample geometry, (b) boundary conditions, and (c) mesh. [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Compression bending numerical model: (a) geometry dimensions, (b) boundary [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Experimental campaign tests setup: (a) metrology sample indicating the three [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Tensile test: (a) Displacement (top) and equivalent stress (bottom) distribution. [PITH_FULL_IMAGE:figures/full_fig_p018_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Elbow LD cross section: distribution of Equivalent stresses (Von Mises) dur [PITH_FULL_IMAGE:figures/full_fig_p019_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Elbow LD cross section: distribution of Equivalent total strains (Von Mises) [PITH_FULL_IMAGE:figures/full_fig_p019_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Elbow TD cross section: Longitudinal stresses [PITH_FULL_IMAGE:figures/full_fig_p020_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Elbow LD cross section: (a) comparison between the theoretical geometry and [PITH_FULL_IMAGE:figures/full_fig_p021_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Metrology measurements: (left) distortion field on the sample and (right) cross [PITH_FULL_IMAGE:figures/full_fig_p024_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Tube TD cross section: (a) nominal geometry, (b) grain size micrograph and [PITH_FULL_IMAGE:figures/full_fig_p024_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Elbow TD cross section: (a) resulting distorted geometry, (b) tension and (c) [PITH_FULL_IMAGE:figures/full_fig_p025_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Elbow LD cross section: (a) thickness evolution profile, (b) outer radius and [PITH_FULL_IMAGE:figures/full_fig_p026_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Hardness measurements along the elbow: (a) Sampling points, (b) Flow Stress [PITH_FULL_IMAGE:figures/full_fig_p028_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: Elbow TD cross section: (a) cartesian and (b-c) polar representations for (b) [PITH_FULL_IMAGE:figures/full_fig_p029_18.png] view at source ↗
read the original abstract

Beam intercepting devices rely on cooling systems to effectively dissipate the thermal energy generated during the impact of a high-energy beam. Regardless of the device's size, integrating the cooling system is a complex task, particularly when the resulting device is only a few centimetres in size, as is the case with the positron source target for the Future Circular Collider at CERN, where the current design consists of a tungsten core with two embedded tantalum cooling tubes. Due to the reduced dimensions of the chosen tantalum tubes (OD6.35xID4.35 mm), the selected manufacturing method is compression bending. The present study develops and evaluates a numerical model to manufacture the required elbow. The methodology is divided in four steps: i) minium allowable bending radius calculation, ii) material constitutive law validation, iii) prediction of the resulting distortion due to ovalization and iv) experimental validation via (non) destructive methods. The results indicate that a minimum bending radius of 10 mm is suitable for manufacturing the elbow. The distortion caused by ovalization is within +-0.5 mm, resulting in an important deviation respect to the nominal geometry. The numerical model was successfully validated experimentally. The micrographies performed in the cross-section of the tantalum tube before and after plastic bending confirm the integrity of the elbow. Additionally, an empirical expression is proposed to estimate the yield stress of pure tantalum based on Vickers hardness measurements. The proposed numerical model is capable to predict the ovalization along the resulting elbow, offering a viable alternative to define the cooling tube geometry. This study provides a methodology to determine the minimum bending radius for thick walled tubes to be used with compression bending and can be applied for the cooling system design of other high-performance devices

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper develops a four-step numerical-experimental methodology for compression bending of small-diameter thick-walled tantalum tubes (OD 6.35 mm) intended for the FCC-ee positron source target cooling system. Step (i) determines a minimum allowable bending radius; step (ii) validates an elastoplastic constitutive law; step (iii) predicts ovalization-induced distortion; and step (iv) performs experimental validation via micrographs and non-destructive checks. The central claims are that a 10 mm minimum radius is suitable, ovalization remains within ±0.5 mm (an important deviation from nominal geometry), the numerical model is validated, no cracking occurs, and an empirical yield-stress expression from Vickers hardness is provided. The work supplies a general methodology for determining minimum bending radii of thick-walled tubes.

Significance. If the constitutive-law validation and ovalization predictions hold under the reported contact and strain conditions, the study supplies a practical, transferable workflow for manufacturability assessment of compact cooling channels in beam-intercepting devices. The combination of forward numerical prediction with independent physical bending trials and hardness measurements is a strength; the proposed empirical hardness-to-yield relation could also be useful for tantalum component qualification.

major comments (2)
  1. [Methodology step (ii)] Methodology step (ii): the claim that the elastoplastic law is validated for the compression-bending process is load-bearing for the ovalization prediction in step (iii) and the final suitability verdict, yet the abstract (and available description) provides no information on the hardening-law functional form, whether anisotropy or rate dependence is included, the maximum equivalent strain reached in the validation data, or how friction coefficients were measured or bounded in the contact model. Without these details it is impossible to assess whether the computed ±0.5 mm ovalization remains reliable when tooling contact and large plastic strains are present.
  2. [Methodology step (iv)] Step (iv) experimental validation: the micrographies are stated to confirm integrity (no cracking), but the text does not report quantitative comparison metrics (e.g., measured vs. predicted ovalization at specific angular positions along the elbow) that would close the loop on the numerical model. This weakens the assertion that the model “was successfully validated experimentally.”
minor comments (2)
  1. [Abstract] Abstract: “minium” should read “minimum”; “respect to the nominal geometry” should read “with respect to the nominal geometry.”
  2. [Abstract] The phrase “an important deviation” is ambiguous; clarify whether the ±0.5 mm ovalization is judged acceptable or problematic for the downstream cooling performance.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which help clarify the presentation of our methodology. We address each major comment below and will revise the manuscript to incorporate the requested details and metrics.

read point-by-point responses

  1. Referee: [Methodology step (ii)] Methodology step (ii): the claim that the elastoplastic law is validated for the compression-bending process is load-bearing for the ovalization prediction in step (iii) and the final suitability verdict, yet the abstract (and available description) provides no information on the hardening-law functional form, whether anisotropy or rate dependence is included, the maximum equivalent strain reached in the validation data, or how friction coefficients were measured or bounded in the contact model. Without these details it is impossible to assess whether the computed ±0.5 mm ovalization remains reliable when tooling contact and large plastic strains are present.

    Authors: We agree that these specifics are needed to fully evaluate the constitutive model under the reported contact and strain conditions. In the revised manuscript we will expand the description in the methods section and abstract to state the hardening-law functional form, confirm that anisotropy and rate dependence were not included, report the maximum equivalent strain from the validation data, and explain the selection and bounding of the friction coefficient (including any sensitivity checks performed). revision: yes

  2. Referee: [Methodology step (iv)] Step (iv) experimental validation: the micrographies are stated to confirm integrity (no cracking), but the text does not report quantitative comparison metrics (e.g., measured vs. predicted ovalization at specific angular positions along the elbow) that would close the loop on the numerical model. This weakens the assertion that the model “was successfully validated experimentally.”

    Authors: We acknowledge that the current validation description relies primarily on qualitative confirmation of no cracking and overall geometric compliance. We will revise the manuscript to add quantitative comparison metrics, including measured versus predicted ovalization values at specific angular positions along the elbow, drawn from the available micrograph and non-destructive test data. revision: yes

Circularity Check

0 steps flagged

No significant circularity; forward simulation validated by independent experiments

full rationale

The derivation proceeds from a four-step methodology: (i) analytic minimum-radius calculation, (ii) separate calibration/validation of the tantalum elastoplastic law against hardness and tensile data, (iii) forward finite-element prediction of ovalization under that law, and (iv) experimental bending trials plus micrography that close the loop on the predictions. None of the load-bearing outputs (R_min = 10 mm, ovalization ±0.5 mm) are defined in terms of themselves or obtained by renaming fitted quantities; the empirical yield-stress expression is an auxiliary fit, not the central claim. No self-citation load-bearing steps, uniqueness theorems, or ansatz smuggling appear. The chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The work rests on standard continuum-mechanics assumptions for large-deformation plasticity and on the validity of the chosen tantalum constitutive model; no new physical entities are introduced.

free parameters (2)
  • minimum bending radius = 10 mm
    Determined numerically as the smallest value that keeps the tube intact under the modeled strains.
  • ovalization tolerance = ±0.5 mm
    Output of the distortion prediction step; reported as acceptable limit.
axioms (2)
  • domain assumption Tantalum obeys an elastoplastic constitutive law that can be calibrated and validated for the strain rates and temperatures of compression bending.
    Invoked explicitly in methodology step ii.
  • domain assumption Finite-element simulation of the bending process captures the dominant deformation modes without requiring additional correction terms for contact or friction.
    Underlying the numerical model in steps i and iii.

pith-pipeline@v0.9.1-grok · 5880 in / 1454 out tokens · 31842 ms · 2026-06-27T22:20:34.000676+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references · 24 canonical work pages

  1. [1]

    and others

    M. Benedikt, F. Zimmermann, B. Auchmann, et al., Future Circular Collider Feasibility Study Report. Volume 2: Accelerators, Technical Infrastructure and Safety, The European Physical Journal Special Top- ics 234 (19) (2025) 5713–6197. doi:10.1140/epjs/s11734-025-01967-4

    work page doi:10.1140/epjs/s11734-025-01967-4 2025

  2. [2]

    J. H. Hubbell, Electron–positron pair production by photons: A his- torical overview, Radiation Physics and Chemistry 75 (2006) 614–623. doi:10.1016/j.radphyschem.2005.10.008

    work page doi:10.1016/j.radphyschem.2005.10.008 2006

  3. [3]

    Lassner, W.-D

    E. Lassner, W.-D. Schubert, Tungsten: properties, chemistry, technol- ogy of the element, alloys, and chemical compounds, Kluwer Academic / Plenum Publishers, New York, 1999. doi:10.1007/978-1-4615-4907-9

    work page doi:10.1007/978-1-4615-4907-9 1999

  4. [4]

    M. J. Cardillo, Y. Look, The reaction of tungsten with CO2 and D2O 33 at high temperatures and low pressures, Surface Science 66 (1) (1977) 272–298. doi:10.1016/0039-6028(77)90412-5

    work page doi:10.1016/0039-6028(77)90412-5 1977

  5. [5]

    G. A. Greene, C. C. Finfrock, Vaporization of tungsten in flowing steam at high temperatures, Experimental Thermal and Fluid Sciences 25 (2001) 87–99. doi:10.1016/S0894-1777(01)00063-2

    work page doi:10.1016/s0894-1777(01)00063-2 2001

  6. [6]

    S. M. Cardonne, P. Kumar, C. A. Michaluk, H. D. Schwartz, Tantalum and its Alloys, Int. J. of Refractory Metals & Hard Materials 13 (1995) 187–194. doi:10.1016/0263-4368(95)94023-R

    work page doi:10.1016/0263-4368(95)94023-r 1995

  7. [7]

    Kawai, K

    M. Kawai, K. Kikuchi, H. Kurishita, J.-F. Li, M. Furusaka, Fabrica- tion of a tantalum-clad tungsten target for KENS, Journal of Nuclear Materials 296 (1) (2001) 312–320. doi:10.1016/S0022-3115(01)00533-5

    work page doi:10.1016/s0022-3115(01)00533-5 2001

  8. [8]

    Wilcox, P

    D. Wilcox, P. Loveridge, T. Davenne, L. Jones, D. Jenkins, Stress levels and failure modes of tantalum-clad tungsten targets at ISIS, Journal of Nuclear Materials 506 (2018) 76–82. doi:10.1016/j.jnucmat.2017.10.075

    work page doi:10.1016/j.jnucmat.2017.10.075 2018

  9. [9]

    Busom Descarrega, M

    J. Busom Descarrega, M. Calviani, T. Hutsch, E. López Sola, A. T. Pérez Fontenla, A. Perillo Marcone, S. Sgobba, T. Weißgärber, Ap- plication of Hot Isostatic Pressing (HIP) technology to diffusion bond refractory metals for proton beam targets and absorbers at CERN, Material Design & Processing Communications 2 (1) (2020) e101. doi:10.1002/mdp2.101

    work page doi:10.1002/mdp2.101 2020

  10. [10]

    Romero-Francia, A

    A. Romero-Francia, A. Perillo-Marcone, S. Pianese, K. Andersen, et al., Design and early operation of a new-generation internal beam dump for 34 CERN’s Super Proton Synchrotron, Phys. Rev. Accel. Beams 27 (2024) 043001. doi:10.1103/PhysRevAccelBeams.27.043001

    work page doi:10.1103/physrevaccelbeams.27.043001 2024

  11. [11]

    Bertarelli, O

    A. Bertarelli, O. Aberle, R. Assmann, E. Chiaveri, T. Kurtyka, M. Mayer, R. Perret, P. Sievers, The Mechanical Design for the LHC Collimators, Tech. rep., CERN-LHC Project Report 786 (2004). URLhttps://cds.cern.ch/record/794628

    2004

  12. [12]

    C. T. Olofson, A review of Bending Methods for Stainless Steeel Tubing, Tech. rep., DMIC Report 150 (1961)

    1961

  13. [13]

    K. Pan, K. A. Stelson, On the Plastic Deformation of a Tube Dur- ing Bending, Jorunal of Engineering for Industry 117 (1995) 494–500. doi:10.1115/1.2803526

    work page doi:10.1115/1.2803526 1995

  14. [14]

    Tang, Plastic-deformation analysis in tube bending, Interna- tional Journal of Pressure Vessels and Piping 77 (2000) 751–759

    N. Tang, Plastic-deformation analysis in tube bending, Interna- tional Journal of Pressure Vessels and Piping 77 (2000) 751–759. doi:10.1016/S0308-0161(00)00061-2

    work page doi:10.1016/s0308-0161(00)00061-2 2000

  15. [15]

    H. A. Al-Qureshi, Elastic-plastic analysis of tube bending, Interna- tional Journal of Machine Tools & Manufacture 39 (1999) 87–104. doi:10.1016/S0890-6955(98)00012-1

    work page doi:10.1016/s0890-6955(98)00012-1 1999

  16. [16]

    El-Megharbel, G

    A. El-Megharbel, G. A. El-Nasser, A. El-Domiaty, Bending of tube and section made of strain-hardening materials, Jour- nal of Materials Processing Technology 203 (2008) 372–380. doi:10.1016/j.jmatprotec.2007.10.078

    work page doi:10.1016/j.jmatprotec.2007.10.078 2008

  17. [17]

    Murata, T

    M. Murata, T. Kuboki, K. Takahashi, M. Goodarzi, Y. Jin, Effect of hardening exponent on tube bending, Journal of 35 Materials Processing Technology 201 (1) (2008) 189–192. doi:10.1016/j.jmatprotec.2007.11.286

    work page doi:10.1016/j.jmatprotec.2007.11.286 2008

  18. [18]

    Paulsen, T

    F. Paulsen, T. Welo, An Analytical Model for Prediction of Tube Oval- ization in Bending, in: ESAFORM 2003: Proceedings of the Sixth ESAFORM Conference on Material Forming, 2003, pp. 775–778

    2003

  19. [19]

    S. A. Tronvoll, J. Ma, T. Welo, Deformation behavior in tube bending: a comparative study of compression bending and rotary draw bending, Int J Adv Manuf Technol 124 (2023) 801–8016. doi:10.1007/s00170-022- 10433-7

    work page doi:10.1007/s00170-022- 2023

  20. [20]

    D. Munz, C. Mattheck, Cross-sectional flattening of pipes subjected to bending, International Journal of Pressure Vessels and Piping 10 (1982) 421–429. doi:10.1016/0308-0161(82)90003-5

    work page doi:10.1016/0308-0161(82)90003-5 1982

  21. [21]

    Miller, Tube Forming Processes: A Comprehensive Guide, Society of Manufacturing Engineers, Dearborn - Michigan, 2003

    G. Miller, Tube Forming Processes: A Comprehensive Guide, Society of Manufacturing Engineers, Dearborn - Michigan, 2003

    2003

  22. [22]

    Springborn, Cold Bending and Forming Tube and other sections, America Society of Tool and Manufacturing Engineers, Dearborn - Michigan, 1966

    R. Springborn, Cold Bending and Forming Tube and other sections, America Society of Tool and Manufacturing Engineers, Dearborn - Michigan, 1966

    1966

  23. [23]

    Jehring, T

    U. Jehring, T. Flutsch, M. Farajian, C. Schweizer, Final Report - Ma- terial Characterization, Project: Supply and characterization of HIPed blocks with refractory metals for beam intercepting devices application, Tech. rep., Fraunhofer Institute for Manufacturing Technology and Ad- vanced Materials IFAM (2020). 36

    2020

  24. [24]

    Lopez Sola, M

    E. Lopez Sola, M. Calviani, P. Avigni, M. Battistin, et al., De- sign of a high power production target for the beam dump fa- cility at CERN, Phys. Rev. Accel. Beams 22 (2019) 113001. doi:10.1103/PhysRevAccelBeams.22.113001

    work page doi:10.1103/physrevaccelbeams.22.113001 2019

  25. [25]

    Ansys Inc., Ansys Academic Research Mechanical, 2022 R2 (2022)

    2022

  26. [26]

    A. S. Khan, R. Liang, Behaviors of three BCC metal over a wide range of strain rates and temperatures: experiments and modeling, International Journal of Plasticity 15 (10) (1999) 1089–1109. doi:10.1016/S0749- 6419(99)00030-3

    work page doi:10.1016/s0749- 1999

  27. [27]

    M. B. Prime, A. Arsenlis, R. A. Austin, N. R. Barton, et al., A broad study of tantalum strength from ambient to extreme conditions, Acta Materialia 231 (2022) 117875. doi:10.1016/j.actamat.2022.117875

    work page doi:10.1016/j.actamat.2022.117875 2022

  28. [28]

    Colas, E

    D. Colas, E. Finot, S. Flouriot, S. Forest, M. Mazière, T. Paris, In- vestigation and modeling of the anomalous yield point phenomenon in pure tantalum, Materials Science & Engineering A 615 (2014) 283–295. doi:10.1016/j.msea.2014.07.028

    work page doi:10.1016/j.msea.2014.07.028 2014

  29. [29]

    doi:10.1520/B0521- 12

    ASTM-B521-12, Standard Specification for Tantalum and Tantalum Al- loy Seamless and Welded Tubes, Standard, American Society for Testing and Materials (ASTM), Philadelphia, USA (2012). doi:10.1520/B0521- 12

    work page doi:10.1520/b0521- 2012

  30. [30]

    Tiryakioˇ glu, J

    M. Tiryakioˇ glu, J. S. Robinson, M. A. Salazar-Guapuriche, Y. Y. Zhao, P. D. Eason, Hardness–strength relationships in the aluminum 37 alloy 7010, Materials Science & Engineering A 631 (2015) 196–200. doi:10.1016/j.msea.2015.02.049

    work page doi:10.1016/j.msea.2015.02.049 2015

  31. [31]

    Tiryakioˇ glu, On the relationship between Vickers hardness and yield stress in Al–Zn–Mg–Cu alloys, Materials Science & Engineering A 633 (2015) 17–19

    M. Tiryakioˇ glu, On the relationship between Vickers hardness and yield stress in Al–Zn–Mg–Cu alloys, Materials Science & Engineering A 633 (2015) 17–19. doi:10.1016/j.msea.2015.02.073. 38

    work page doi:10.1016/j.msea.2015.02.073 2015