Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.;...

download Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- Study of the Effects Produced by Shot Peening on the Surface of Quenched

of 11

Transcript of Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.;...

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    1/11

    Applied Surface Science 356 (2015) 475–485

    Contents lists available at ScienceDirect

    Applied Surface Science

     j ournal homepage: www.elsevier .com/ locate /apsusc

    Study of the effects produced by shot peening on the surface of quenched and tempered steels: roughness, residual stresses and workhardening

    V. Llaneza∗, F.J. Belzunce

    Materials Science Department, University of Oviedo, University Campus, 33203,Gijón, Spain

    a r t i c l e i n f o

     Article history:Received 29 May 2015

    Received in revised form 29 July 2015

    Accepted 13 August 2015

    Available online 15 August 2015

    Keywords:

    Shot peening

    Quenched and tempered steels

    Roughness

    Residual stress profiles

    Full width at half maximum

    a b s t r a c t

    Shot peening induces important effects onthe surface of materials, both positive and negative, the correct

    balance between them being the key to success.

    Roughness, impact mark size, compressive residual stress and work hardening of  six steel grades

    obtained from an AISI 4340 steel were studied to explain their evolution according to the Almen intensity

    and their mechanical properties. A linear relationship between the impact diameter, the kinetic energy

    of the balls and the Almen intensity was found. Moreover, under full coverage, the surface and the max-

    imum compressive stresses only depend on the mechanical properties of the steels, whereas the depth

    subjected to high compressive residual stresses and the total depth subjected to compressive residual

    stresses depend on the mechanical properties of the steel and the Almen intensity. Furthermore, several

    mathematic expressions were formulated to predict the residual stress profiles using the Almen intensity

    and the mechanical properties of the steels.

    © 2015 Elsevier B.V. All rights reserved.

    1. Introduction

    Conventionalshot peening(SP) is a cheap surface treatmentthat

    consists in projecting very hard, tiny, spherical ceramic or metal-

    lic balls (0.3< Ø< 1.6mm) at high speed onto the surface of the

    component to treat. These impacts produce local surface plastic

    deformation, the expansion of which is constrained by the adjacent

    deeper material, giving rise to a uniform surface compressive resid-

    ual stress field (Fig. 1), along with other important effects. These

    include modification of the roughness and appearance of the sur-

    face in addition to work hardening, which, if properly controlled,

    can significantly improve the final properties of metallic compo-

    nents [1–4]. The aforementioned effects provided by shot peening

    treatments cannot be calledmerely positive or negative,as this roledepends on the purpose of each treatment.

    Shot peening has many applications: for instance, it canbe used

    to improve the fatigue life of industrial components [5–8], obtain

    a specific surface finishing [9], enhance the wear resistance [10] or

    prevent stress corrosion cracking [11,12]. Consequently, it is nec-

    essary to control the shot peening parameters, mainly the Almen

    ∗ Corresponding author. Tel.: +34 985182024.

    E-mail addresses: [email protected] (V. Llaneza), [email protected]

    (F.J. Belzunce).

    intensity and the coverage degree, according to the mechanical

    propertiesof the material treated, to obtain the best combination of 

    the aforementioned effects and, hence, maximize the performance

    of the product. Coverage is the ratio of the area covered by the shot

    impacts to the entire surface of the treated sample, expressed as a

    percentage, whereas the Almen intensityis a measure of theenergy

    of the shot stream, which depends on the projection velocity and

    also on the shot density, mass and size [13,14].

    However, although shot peening is a relatively old technology,

    even now, most companies arenot able to employ it optimally, and

    this means that they are not able to take full advantage of it. The

    main reason is the complexity of the process, due to the different

    parameters that must be simultaneously controlled to attain the

    optimal balance among effects.It is worth to remember here the existence of other surface

    treatments which are based in similar concepts as conventional

    shot peening, but they have some specific differentiating charac-

    teristics. For instance, severe shot peening (SSP), which employs

    more intense parameters, usually very high coverage degrees [15];

    laserpeening, which useslaser-generated shock waves to introduce

    high level of surface compressive stresses deeper in the workpiece

    [16]; roller burnishing, which rub the metal surface with a smooth

    hard roller under a sufficient pressure [17] or surface mechanical

    attrition treatment (SMAT), where shots are resonated by vibra-

    tion using an ultrasonic transducer [18–20], as well as vibration

    http://dx.doi.org/10.1016/j.apsusc.2015.08.110

    0169-4332/© 2015 Elsevier B.V. All rightsreserved.

    http://localhost/var/www/apps/conversion/tmp/scratch_7/dx.doi.org/10.1016/j.apsusc.2015.08.110http://www.sciencedirect.com/science/journal/01694332http://www.elsevier.com/locate/apsuscmailto:[email protected]:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_7/dx.doi.org/10.1016/j.apsusc.2015.08.110http://localhost/var/www/apps/conversion/tmp/scratch_7/dx.doi.org/10.1016/j.apsusc.2015.08.110mailto:[email protected]:[email protected]://crossmark.crossref.org/dialog/?doi=10.1016/j.apsusc.2015.08.110&domain=pdfhttp://www.elsevier.com/locate/apsuschttp://www.sciencedirect.com/science/journal/01694332http://localhost/var/www/apps/conversion/tmp/scratch_7/dx.doi.org/10.1016/j.apsusc.2015.08.110

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    2/11

    476 V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485

    Fig. 1. Schematic illustration of the shot peening process.

     Table 1

    Chemical composition of AISI 4340 alloy steel.

    Element C Mn Si P S Cr Ni Fe

    wt% 0.410 0.710 0.260 0.013 0.024 0.870 1.920 Balance

    Element Mo V Cu Al Sn Ti Nb

    wt% 0.235 0.005 0.210 0.016 0.011 0.004 0.003

    polishing, vibration peening or grinding [21]. In relation to these

    surface treatments, shotpeeningis usually cheaper,versatile, effec-

    tive enough and very easy to be implemented in most workshops.

    Anyway, in order to attain the final goal on these surface treat-

    ments and specifically in the case of shot peening, it would be

    convenient to have a tool able to foresee the main effects of any

    treatment in order to select the most appropriate parameters for

    optimizing it. Numerous experimental and theoretical studies have

    beenperformed along these lines to improve thestate of knowledge

    of shot peening and better understand its effects [9,22–28].

    This paper focuses on the analysis of the evolution of the main

    effects induced by conventional shot peening treatments (surface

    finish modification,surfacework hardening andcompressiveresid-

    ual stress fields), in different quenched and tempered steel grades

    presenting a relatively broad range of mechanical properties sub-

    mitted to different shot peening intensities. The main objective of 

    the experimental study was to understand the role played by the

    mechanical properties of the treated steel and the applied Almen

    intensity on the main effects induced by shot peening treatments.

    Furthermore, several simple, practical expressions are proposed to

    predict the impact diameter and some characteristic values of the

    residual stress profiles. These expressions may be used in a practi-

    cal way to predict the effects induced by shot peening treatments

    on industrial components, being an effective tool to select the cor-

    rect parameters to satisfy the requirements fixed by the final client

    in an easy and fast way.

    2. Materials and methods

     2.1. Steel and mechanical properties

    This studywas carried outon samples ofAISI4340, a commercial

    heat treatable low-alloy steel widely employed in the automotive

    andaircraftindustriesfor themanufactureof gears, shafts andother

    structuralcomponentsdue to its favorablecombinationof strength,

    toughness andductility.The steel wassuppliedin theformof rolledbars with a diameter of 16mm, and its chemical composition is

    given in Table 1.

    This steel was subjected to different heat treatments in order

    to obtain six different steel grades. The treatments consisted in

    austenitizing at 850◦C for 45min, water quenching (Q), plus differ-

    ent temperingtreatments (T), ranging from 200◦Cto680 ◦C, during

    150 min. The use of different tempering temperatures allowed us

    to obtain a wide range of mechanical properties, as can be seen

    in Table 2, which shows a representative range of the mechani-

    cal properties of typical martensitic steels employed in the metal

    industry.

    Fig.2 shows thesteelmicrostructureobtainedaftertwo of these

    heat treatments (Q+ T200 and Q + T680).

     Table 2

    Hardness and tensile properties of quenched and tempered AISI4340 steel (Vickers hardness,HV, yield strength, ys , ultimate tensile strength, uts , and elongation, E ).

    Steel Tempering temperaturea (◦C) HV (31,25 kg)   ys (MPa)  uts  (MPa) E (%)

    Q + T200 200 552 1604 2057 10.5

    Q + T425 425 424 1364 1426 10.6

    Q + T540 540 350 1123 1201 13.7

    Q + T590 590 325 983 1123 14.6

    Q + T650 650 255 863 897 19.3

    Q + T680 680 226 626 764 24.7

    a All tempering times were 150min,except 10h in Q+ T680.

    Fig. 2. Steel microstructures(nital etched). (a) Q + T200; (b) Q + T680.

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    3/11

    V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485 477

     Table 3

    Work parameters for the different shot peening treatments.

    Almen intensity (deflection of 

    the Almen strip (A), in mm)

    Shot size

    (mm)

    Pressure

    (bar)

    Shot Speed

    (m/s)

    Impact

    angle (◦)

    Stand-off 

    distance (mm)

    Ø nozzle

    (mm)

    8A (0.2 mm) CW-0.3 2 –

    10A (0.25mm) CW-0.4 2 52.2

    12A (0.3 mm) CW-0.5 2 53.4

    14A (0.35mm) CW-0.5 3 55.1 90 240 5

    16A (0.4mm) CW-0.7 1.5 43.9

    19A (0.475mm) CW-0.7 3 49.921A (0.52 mm) CW-0.7 4 –

    All the tests were carried out on small slices cut trans-

    versely from the bars, with an approximate thickness of 10mm.

    These samples were ground in SiC papers of progressively

    lower grit sizes and carefully polished with diamond paste

    (6m and, finally, 1m) to ensure a soft and homogeneous

    initial state (Ra≤0.1m,Rmax ≤0.2m, residual stress in the

    near-surface region below 200MPa, and depth affected by the

    so-mentioned residual stresses lower than 20m), thus guaran-

    teeing that all the evaluated effects were only induced by shot

    peening.

     2.2. Shot peening treatments

    Shot peening treatments were carried out by means of a direct

    compressedair machine (Guyson Euroblast4 PF) usingconditioned

    cut wire shots with rounded off edges (CW, 670-730 HV). Seven

    shotpeeningtreatmentswere designedwith Almenintensitiesran-

    ging between 8A and 21A (0.2–0.52 mm) following SAE J442 and

    SAE J443 specifications [29,30] employing ‘A’ type Almen strips.

    In order to achieve this range of Almen intensities, it was neces-

    sary to use shots with diameters ranging between 0.3 and 0.7 mm.

    The combination of parameters selected in each treatment, includ-

    ing the impact angle, the diameter of the nozzle and the distance

    between sample and nozzle, is shown in Table 3. It is important

    to remark that both nozzle and samples remained fixed during the

    whole treatment.The last step in defining and performing the treatments is the

    selection of the exposure time to achieve the required degree of 

    coverage. Residual stress profiles and surface work hardening were

    always evaluated in samples with full coverage (100%), but rough-

    ness was also studied usingdifferent degrees of coverage (25%, 50%,

    75%, 100% and 200%). The lower coverage degrees were used to

    measure the impact marks.

     2.3. Surface finishing 

     2.3.1. Impact diameters

    The diameters of the impacts created by the shot peening

    treatments were evaluated using a specific routine of an image

    analysis software, which allows the average diameter of each dim-ple to be estimated via images obtained using conventional optical

    microscopy (OM). In particular, more than 60 impact marks of 

    each treatment and steel were assessed, thus obtaining a set of 

    data which was subsequently analyzed to obtain the evolution of 

    the average equivalent diameter as a function of both the applied

    Fig. 3. Shot peening impact mark (SEM).Q + T590-SP8A.

    Almen intensity andthe mechanical properties of the treated steel.

    The typical geometry of one impact can be seen in the image taken

    with a JEOL JSM5600 scanningelectronmicroscopeshownin Fig.3.

     2.3.2. Roughness

    The surface roughness after shot peening was characterized on

    a Diavite DH-6 roughness testerby means of the average roughnessRa  and Rmax parameters. The latter parameter is the largest of the

    five Rimax within the assessment length of 4.8 mm, where Rimax is

    the maximum peak-to-valley height of the profile in each of the

    five aforementioned measurements [31]. Six different roughness

    profiles were performed on each sample (three in the longitudinal

    direction and another three in the transversal direction) and the

    average results were reported.

     2.4. Residual stresses

    The shot peening residual stress profiles were determined by

    X-ray diffraction (XRD) and incremental layer removal by elec-

    tropolishing. Measurements were carried out on an X-Stress 3000

    G3R device manufactured by Stresstech using the sin2

    methodand the recommendations of NPL [32–34]. The experimental con-

    ditions are shown in Table 4.

    Diffraction data were determined in three different directions

    on the specimen plane, −45, 0 and +45◦, subsequently calculat-

    ing the average result. Electrochemical polishing was performed

     Table 4

    Experimental parameters employed in the X-ray diffraction analysis.

    Wavelength K (Cr) 0.2291 nm Filter Vanadium

    Exposure time (s) 20 Ø collimator (mm) 2

    Tilt (◦) 9 points between −45/+45 Rotation angle, ϕ (◦) −45, 0 y 45

    Background Parabolic Fit Pseudo-Voigt

    Measuring mode -modified Diffraction angle 156.0◦

    Millerindices (hk l) (2 1 1) Elastic constant, E (1 +)−1 (GPa) 168.9±2.8

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    4/11

    478 V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485

    by applying 45V in an electrolyte composed of 94% acetic acid and

    6% perchloric acid. This process produces slight stress relaxation,

    which was corrected in accordance with Sikarskie [35], who, based

    on the Moore and Evans procedure [36], has developed a special

    methodology to minimize the error. Furthemore, the values of the

    expanded uncertainties, which correspond to 95% confidence, vary

    between ±30 and ±50MPa, growing as the mechanical properties

    of the steel increase.

     2.5. Surface work hardening 

    This phenomenon was assessed by means of the Full Width

    at Half Maximum (FWHM), a parameter that corresponds to the

    width of the diffraction peak at half of its height and which can

    be calculated in the course of the X-ray diffraction tests at the

    same time as the residual stresses are estimated. This parame-

    ter can be considered an index of the distortion of the crystal

    grain which takes into account the density of dislocations and

    the so-called type II micro residual stresses present in the crys-

    tal lattice, although some instrumental broadening is always also

    present [33,37,38]. The FWHM parameter is widely used in shot

    peening studies to quantify surface work hardening effects [15,38].

    The expanded uncertainty related with these measurements is

    about ±0.1◦.

    3. Results and discussion

     3.1. Surface finishing 

     3.1.1. Impact marks

    As can be seen in Fig. 4, impact diameters depend on both the

    applied Almen intensity and the properties of the peened steel:

    as impact size is a direct measure of the induced surface plastic

    deformation, it increases with increasing Almen intensity and as

    the mechanical strength of the steel decreases (higher tempering

    temperature, see Table 2).

    The results shown in Fig. 4 provide a clear linear relation-

    ship between impact diameter and Almen intensity, the slopes of 

    which are dependent on the mechanical properties of the steels.

    Eqs. (1)–(3) were developed with a quite high degree of accuracy

    (around 4%), using hardness, yield strength or tensile strength as

    the reference steel mechanical parameter. These expressions are

    able to predict the impact diameter on quenched and tempered

    steels submitted to shot peening treatments (8A< AI< 21A) in an

    easy and accurate way, as long as the shot peening shot size is

    between 0.3 and 0.7 mm.

    D(HV)  = (736− 0.444× HV) × AI Uncertainty < 4.1% (1)

    Fig. 4. Evolutionof impact diameter versusthe applied Almen intensity.

     Table 5

    Expressions to predict thecompressive residual stressat thesurface, rc s   .

     rcs   (MPa)

    Mechanical property Expression Error

     ys    rcs   = −0.537 ×  ys  [Eq. (4)] 9.9%

     uts    rcs   = −0.468 ×  uts  [Eq. (5)] 6.6%

    HV  rcs   = −1.654 × HV [Eq. (6)] 7.1%

    D( ys)  =

    747− 0.154×  ys× AI Uncertainty < 4.1% (2)

    D( uts)  = (720 − 0.114×  uts)× AI  Uncertainty < 4.3% (3)

    In contrast to those proposed by other authors [11,39,40], these

    expressions have been formulated without considering the influ-

    ence of shot size. Theinfluence of shot size on impactdiameter was

    found to be quite low. Using the same intensity (SP14A) but differ-

    ent shot sizes (CW0.5 and CW0.6) non-significant differences were

    observed between impact diameters. In line with this result and

    the small reported error, this peening parameter has been ignored

    in Eqs. (1)–(3).

     3.1.2. Kinetic energy

    The kinetic energy provided by the shot stream was measuredusing an electronic device which uses two sensors separated by a

    known distance. The time shots took to fly between these sensors

    was measured, thus providing shot velocity and hence the average

    kinetic energy of the shot stream (E =0.5mv2). Shot geometry was

    considered ideally spherical, average shot diameters were mea-

    sured under a scanning electron microscope and a density of the

    steel shot of 7.8 g cm−3 was also used, givingriseto the datashown

    in Fig.5. A linear plot of the shot kinetic energy versus the intensity

    of the shotpeeningtreatment wasthus obtained. Theseresults con-

    firm that the Almen intensity is directly correlated with the kinetic

    energy of the shot stream and this parameter is barely dependent

    on shot size, as can also be seen by comparing the kinetic energy of 

    two14A treatments producedusing twodifferentshot sizes(CW0.5

    and CW0.6). The shot kinetic energy measured in these two treat-ments was quite similar, and the respective values being situated

    between 12A and 16A, as expected.

     3.1.3. Roughness

    As can be seen in the Q + T 590 steel used as an example in

    Fig.6a, theRa and Rmax parameters increase graduallywith increas-

    ing degree of coverage until reaching full coverage (100%). From

    this point on, both roughness parameters remain constant, as the

    surface work hardening induced by successive impacts finally lim-

    its the depth and extension of surface impact marks. Fig. 6b shows

    the evolution of Ra  and Rmax versus Almen intensity in the case of 

    Fig. 5. Kinetic energy versus Almenintensity (AI).14A intensitywas provided using

    CW0.5 and CW0.6 shots.

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    5/11

    V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485 479

    Fig. 6. Roughness. (a) Evolutionof roughnessversus thedegree of coverage; (b) evolution of Ra and Rmax versus Almen intensity (full coverage).

    samples submitted to full coverage (results also obtained with the

    Q + T590 steel). It can be seen that, in general, roughness increases

    withincreasing Almenintensity (impact diameter has already been

    seen to increase with shot peening intensity). However, shot size

    also plays an important role, as a significant decrease in roughness

    was always detected when increasing the shot size from CW0.5

    (14A) to CW0.7 (16A). Similar graphs were also found for the other

    steel grades.

    Moreover, Fig.7 shows that, under thesameshotpeeningcondi-

    tions (SP14A, CW0.5 and full coverage), the roughness parameters

    decrease linearly with increasing hardness of the treated steel (the

    steel initial hardness was used instead of thesurface hardness after

    shot peening, but as hardness increases were always below 10%,

    results would not change significantly). The effect of steel hard-

    ness on impact size was indirectly shown in Fig. 4, as tempering

    temperature is inversely related to the hardness of the steel.

     3.2. Residual stresses

    Every compressive residual stress profile can be well charac-

    terized using four parameters [39,41,42]: the compressive residual

    stress at the surface,  rc s   ; the maximum value of the compressive

    Fig. 7. Evolution of the roughness parameters, Ra  and Rmax, versussteel hardness.SP14A and full coverage.

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    6/11

    480 V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485

    Fig. 8. Typical residual stress profile and characteristic parameters.

    residual stress,  rc max   (usually located at a certain depth underthe surface); the total depth submitted to compressive residual

    stresses,  Z 0; and the depth subjected to high compressive resid-

    ual stresses,  Z hc . This last parameter was defined in this study as

    the depth at which the compressive residual stress is at least half 

    the yield strength of the steel. These parameters are represented inFig. 8 over a typical residual stress profile induced by shot peening.

    Fig. 9 shows the residual stress profiles produced by two

    given shot peening treatments (10A and 16A) on the different

    steel grades: surface and maximum compressive residual stresses

    decrease with decreasing strength of the steel (higher temper-

    ing temperature) [22,39,40]. However, the total depth of the

    compressive residual stresses and the depthsubjected to highcom-

    pressive residual stresses increase with decreasing strength of the

    steel.

    In addition,all theresidualstress profiles obtained in ourexper-

    imental measurements onto the Q+ T steels under the different

    Almen intensities (full coverage) are shown in Fig. 10. According

    to this last figure, compressive residual stresses (surface and max-

    imum) barely depend on the applied Almen intensity. However,the affected depths (total depth submitted to compressive stresses

    anddepth subjectedto highcompressiveresidualstresses)increase

    with increasing Almen intensity, as previously reported by other

    authors [22,43,44].

    As well as other authors [45–48], we have formulated differ-

    ent simple and practical expressions to predict these parameters

    making use only of the applied Almen intensity (mmA) and one of 

    the main mechanical properties of the treated steel (yield strength,

    ultimate tensile strength or hardness). The expressions shown in

    Tables 5–8, Eqs. (4)–(15), were obtained along with their aver-

    age error through lineal regressions and statistical analysis and

     Table 6

    Expressions to predict the maximum value of the compressive residual stress, rcmax.

     rc max (MPa)

    Mechanical property Expression Error

     ys    rc max  = −0.67 ×  ys  [Eq. (7)] 9.7%

     uts    rc max  = −0.58 ×  uts   [Eq. (8)] 4.5%

    HV  rc max  = −2 × HV [Eq. (9)] 6.1%

     Table 7

    Expressions to predict thetotal depth of thecompressive residual stresses, Z 0 .

     Z 0  (mm)

    Mechanical property Expression Error

     ys   Z 0 = (−0.0004× ys + 1.25)×AI [ Eq. ( 10)] 6.6%

     uts   Z 0 = (−0.0003× uts + 1.19)×A I [ Eq. (11)] 6.7%

    HV  Z 0 = (−0.0011×HV+1.23)×AI [Eq. (12)] 5.8%

     Table 8

    Expressions to predict the depth subjected to high compressive residual stresses,

     Z hc . (− c > ys/2).

     Z hc  (mm)

    Mechanical property Expression Error

     ys   Z hc =(0.91−0.0003× ys)×AI [Eq. (13)] 7.8%

     uts   Z hc =(0.89−0.0002× uts)×AI [Eq. (14)] 10.9%

    HV  Z hc =(0.92−0.0009×HV)×AI [Eq. (15)] 5.6%

    combine precision (error< 10%; in the best cases around 5%) with

    simplicity.

    The best mechanical parameter for predicting residual com-

    pressive stresses is seen to be tensile strength, though hardness

    is the best for predicting affected depths. Fig. 11 compares the pre-

    dicted surface and maximum compressive stresses produced by

    shot peening withthe experimental results, while Fig. 12 compares

    the predicted depths with their experimentally measured values.

    Good correlations have been found with the four parameters.

     3.3. Work hardening. FWHMprofiles

    As previously stated, the shot peening work hardening study

    was carried out employing the FWHM parameter, the profiles of 

    whichwereobtainedby XRDat thesame time as those correspond-

    ingto the residualstress. Moreover,thisparameter was shownto be

    a useful and practical tool to evaluate the surface work hardening

    induced by shot peening treatments.

    Fig. 13 shows that the steel surface layer affected by shot peen-

    ing becomes deeper as the applied Almen intensity increases, and

    coversa similar depthto that subjectedto the compressiveresidual

    Fig. 9. Residual stressprofiles following different SP treatments on diverse steels. Two applied Almen intensities,10A and 16A, full coverage.

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    7/11

    V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485 481

    Fig. 10. Residual stressprofiles obtained by applying differentSP treatments to thedifferent Q +T steels (full coverage).

    Fig. 11. Predicted compressive residual stressvs experimental results (full coverage) a) at thesurface;b) maximum value.

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    8/11

    482 V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485

    Fig.12. Predicted depths vs experimental results (full coverage) (a) total depthsubjected to compressive residual stresses;(b) depthsubjected to high compressive residual

    stresses.

    Fig. 13. FWHM profiles following differentSP treatments applied to thestudied steel grades(full coverage).

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    9/11

    V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485 483

    Fig. 14. FWHM profiles obtained on different steel gradesusing a 14A SP (full cov-

    erage). *Q corresponds to thequenchedand non-tempered 4340 steel.

    stress field. According to these results, it can be seen that any

    shot peening treatment gives rise to significant surface hardening,

    but its final effects are also highly dependent on the strength of 

    the steel. The softer the steel, the greater the surface increase in

    the FWHM parameter; that is to say, softer steels have a greaterwork-hardening capacity. However, if the hardness of the base

    steel is high enough, shot peening treatments can also give rise

    to a kind of local softening. This is clearly seen in the hardest steel

    (Q+ T200), which was only submitted to stress relieving temper-

    ing. The observed decrease in the FWHM parameter in this steel

    is probably due to dislocation re-arrangement. The base steel has

    a distorted structure with a high hardness and peening-induced

    plastic deformation has resulted in a lower-energy dislocation

    arrangement.This has,consequently, given riseto a reductionin the

    FWHM parameter in the surface and sub-surface regions, although

    a slight increase in FWHM was observed in the first 0.05mm. The

    behavior of the second hardest steel grade (Q+ T425) is situated

    between the hardest grade and the other steels, confirming the

    aforementioned explanation. According to this same figure, it isalso worth noting that the surface value of the FWHM parameter is

    barely affected by the applied Almen intensity (no clear influence

    of the shot peening intensity was observed).

    Moreover, the initial FWHM parameter characteristic of each

    steel(internal, base FWHMvalue) is linearlyrelated to the hardness

    of the steel, as can be seen in Fig. 14. This fact can be better appre-

    ciated in Fig. 15a, in which the base FWHM steel value has been

    plotted versus the hardness of the steel (a last result obtained with

    the quenched and non-tempered 4340 steel, 662 HV, has also been

    included in this figure). A quite good linear correlation between

    these two variables was obtained, confirming the possibility of 

     Table 9

    Predictedresultsfroman average impactdiameterof 179m (Q+ T540 4340 steel).

     rcs   = −562MPa  Z 0 = 0.26mm

     rcmax  = −697MPa  Z hc = 0.18mm− rc >

       ys2  = 562MPa

    Fig. 16. Experimentally measured residual stress profile and predicted values

    (Q + T540 4340steel, SP12A and full coverage).

    using the FWHM parameter to detect changes in hardness. On the

    other hand, surface hardening produced by shot peening treat-

    ments is better represented as the difference between the surface

    FWHM and the base FWHM parameters. Fig. 15b shows a linear

    decreasein surface hardening withdecreasinghardness of thesteel.

    It is also worth noting that the surface FWHM parameter of Q+ T

    steels whose hardness is above 470 HV does not increase through

    conventional shot peening. In fact, the surface FWHM values of 

    these steels decrease below the base value characteristic of each

    steel. Nonetheless, even in these cases, a certain degree of work

    hardening can be appreciated in the first 0.05mm (see Q+ T200 in

    Fig. 13).

     3.4. Example of application

    Eqs. (1), (5), (8), (12) and (15) were used to predict the residual

    stress profile of a Q + T540 4340 steel using only the measurements

    of the impactdiameters produced in a shot peening treatment. The

    average measurements of the impact marks were 179m. Using

    Eq. (1), the application of an Almen intensity of 0.308 mm (12.3A)

    was derived. Subsequently, Eqs. (5), (8), (12) and (15) were used to

    determine the compressive residual stress at the surface,  rcs   , the

    maximum value of the compressive residual stress,  rcmax, the totaldepthsubmitted to compressive residual stresses, Z 0, andthe depth

    subjected to high compressive residual stresses, Z hc (− rc > ys/2).

    Fig. 15. Evolution of theFWHM. Thegreatesthardnessrepresented in thegraphs (662 HV)correspondsto thequenchedand non-tempered 4340 steel.(a) Thebase FWHM

    parameter; (b) thesurface FWHM minus thebase FWHM versus steel hardness (full coverage).

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    10/11

    http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0410http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0405http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0400http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0380http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0375http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0370http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0365http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0360http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0355http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0350http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0345http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0340http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0335http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0330http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0325http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0320http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0315http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0310http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0300http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0295http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0290http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0285http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0280http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0275http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0270http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0265

  • 8/16/2019 Applied Surface Science Volume 356 Issue 2015 [Doi 10.1016_j.apsusc.2015.08.110] Llaneza, V.; Belzunce, F.J. -- S…

    11/11

    V. Llaneza, F.J. Belzunce / Applied Surface Science 356 (2015) 475–485 485

    [35] D.L. Sikarskie, On a series form of correction to stresses measured using X-raydiffraction,AIME Trans.239 (1967) 577–580.

    [36] M.G. Moore,W.P. Evans,Mathematical correction forstress in removed layersin X-ray diffraction residual stressanalysis, SAE Trans.66 (1958).

    [37] S. Bagherifard, C. Colombo, M. Guagliano, Application of different fatiguestrength criteria to shot peenednotched components. Part 1: fracturemechanics based approaches, Appl. Surf. Sci. 289 (2014) 180–187.

    [38] B. Pinheiro, J. Lesage, I. Pasqualino, E. Bemporad, N. Benseddiq, X-raydiffraction study of microstructural changes during fatigue damageinitiation in pipe steels: role of theinitial dislocation structure, Mater.Sci. Eng. A: Struct. Mater. Prop. Microstruct. Process. 580 (2013)

    1–12.[39] S.Wang,Y. Li, M. Yao, R. Wang, Compressive residual stressintroduced by

    shot peening, J. Mater.Process.Technol. 73 (1998) 64–73.[40] K. Tosha,Effect of shot peening on surface texture and surface integrity, in:

    Proc. 11th Int. Conf. of Shot Peening ICSP-11, Indiana, 2011.[41] Y.K. Gao, M. Yao, J.K. Li, An analysis of residual stressfieldscausedby shot

    peening, Metall. Mater. Trans. A: Phys. Metall. Mater.Sci. 33A (2002)1775–1778.

    [42] I. Sumiyoshi, A. Tange, H. Okada, Shot peening conditions and processingproperties forsprings steel,in: Proc. 11th Int. Conf. of Shot Peening ICSP-11,Indiana, 2011.

    [43] Y.F. Al-Obaid, Shot peening mechanics: experimental and theoretical analysis,Mech. Mater. 19 (1995) 251–260.

    [44] W. Zinn, B. Scholtes, Mechanical surface treatments of lightweightmaterials—effects on fatigue strength and near-surface microstructures, J.Mater. Eng. Perform. 8 (1999) 145–151.

    [45] H. Bae,H. Diep, M. Ramulu, Influence of shot peening coverage on residualstresses induced in Aluminum alloy 7050-T745, in: Proc. 12th Int. Conf. of Shot Peening ICSP-12, Goslar, 2014.

    [46] S.T.S. Al-Hassani, Mechanical aspects of residual stress development in shotpeening, in: Proc. 1st Int. Conf. of Shot Peening ICSP-1, Paris, 1981.

    [47] R. Fathallah, Modelling of shot peening residual stressesand plasticdeformation induced in metallic parts, in:Proc. 6th Int. Conf. of Shot PeeningICSP-6, San Francisco, 1996.

    [48] H. Guechichi, L. Castex, J. Frelat, G. Inglebert, Predicting residual stressesdueto shot peening, in: Proceedings of Tenth Conference on Shot Peening(CETIM-ITI), Senlis, 1986.

    http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0415http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0415http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0415http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0420http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0420http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0420http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0425http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0425http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0425http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0425http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0425http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0430http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0435http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0435http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0435http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0435http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0440http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0440http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0445http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0445http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0445http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0445http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0445http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0480http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0475http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0470http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0465http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0460http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0455http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)01923-6/sbref0450http://refhub.elsevier.com/S0169-4332(15)0192