Available online at www.sciencedirect.com

www.elsevier.com/locate/jmbbm

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2

http://dx.doi.org/101751-6161 & 2014 El

nCorresponding autE-mail addresses

cahit.kurbanoglu@m

Research Paper

Effects of gamma radiation sterilization and strainrate on compressive behavior of equine cortical bone

Kenan Tufekcia, Ramazan Kayacana,n, Cahit Kurbanoglub

aDepartment of Mechanical Engineering, Süleyman Demirel University, 32260 Isparta, TurkeybDepartment of Mechanical Engineering, İstanbul Medeniyet University, Göztepe Kadıköy – İstanbul, Turkey

a r t i c l e i n f o

Article history:

Received 14 September 2013

Received in revised form

24 January 2014

Accepted 2 February 2014

Available online 10 February 2014

Keywords:

Gamma radiation sterilization

Split Hopkinson pressure bar

Equine bone

High strain rate compression

Quasi-static compression

Mechanical properties

.1016/j.jmbbm.2014.02.004sevier Ltd. All rights rese

hor. Tel.: þ90 246 211 124: kenantufekci@sdu.edu.tedeniyet.edu.tr (C. Kurb

a b s t r a c t

Objectives: Gamma radiation has been widely used for sterilization of bone allograft.

However, sterilization by gamma radiation damages the material properties of bone

which is a major clinical concern since bone allograft is used in load bearing applications.

While the degree of this damage is well investigated for quasi-static and cyclic loading

conditions, there does not appear any information on mechanical behavior of gamma-

irradiated cortical bone at high speed loading conditions. In this study, the effects of

gamma irradiation on high strain rate compressive behavior of equine cortical bone were

investigated using a Split Hopkinson Pressure Bar (SHPB). Quasi-static compression testing

was also performed.

Methods: Equine cortical bone tissue from 8 year old retired racehorses was divided into

two groups: non-irradiated and gamma-irradiated at 30 kGy. Quasi-static and high strain

rate compression tests were performed at average strain rates of 0.0045/s and 725/s,

respectively.

Results: Agreeing with previous results on the embrittlement of cortical bone when

gamma-irradiated, the quasi-static results showed that gamma-irradiation significantly

decreased ultimate strength (9%), ultimate strain (27%) and toughness (41%), while not

having significant effect on modulus of elasticity, yield strain and resilience. More

importantly, contrary to what is typically observed in quasi-static loading, the gamma-

irradiated bone under high speed loading showed significantly higher modulus of elasticity

(45%), ultimate strength (24%) and toughness (26%) than those of non-irradiated bone,

although the failure was at a similar strain.

Significance: Under high speed loading, the mechanical properties of bone allografts were

not degraded by irradiation, in contrast to the degradation measured in this and prior

studies under quasi-static loading. This result calls into question the assumption that bone

allograft is always degraded by gamma irradiation, regardless of loading conditions.

However, it needs further investigation to be translated positively in a clinical setting.

& 2014 Elsevier Ltd. All rights reserved.

rved.

7; fax: þ90 246 237 0859/þ90 246 211 1750.r (K. Tüfekci), ramazankayacan@sdu.edu.tr (R. Kayacan),anoğlu).

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2232

1. Introduction

Bone fractures are a leading health problem with enormoussocial and economical consequences (Burge et al., 2007;Bonafedea et al., 2013). Bone allografts are widely used as anatural substitute to repair the defects in skeletal system suchas bone fractures, spinal fractures and bone tumor. Due to itseffectiveness and convenience, terminal sterilization of boneallografts by gamma radiation is very important to minimizedisease transmission and infection (Bright et al., 1983; Nemzeket al., 1994; Kennedy et al., 2005; McAllister et al., 2007). However,themechanical properties of gamma-irradiated cortical bone areaffected by the destruction of collagen and it is made brittle byirradiation because of the destruction of collagen alpha chains(Hamer et al., 1996a, 1996b; Akkus and Rimnac, 2001; Nguyenet al., 2007a). Clinical findings have indicated that the rate offracture in allografts sterilized with gamma radiation may behigher than that in non-irradiated allografts (Lietman et al.,2000). The degree of this destruction depends on dose of theradiation (Hamer et al., 1996b; Nguyen et al., 2007a; Mardas et al.,2012). The dose employed in many of tissue banks is 30 kGy,with a tolerance of 5 kGy, which is sufficient to minimize anyrisk of disease transmission (Angermann and Jepsen, 1991;Hamer et al., 1999; Nguyen et al., 2007a, 2007b). Despite findingsof Komender (1976) that there was no significant effect ofgamma irradiation with a dose between 25 kGy and 35 kGy,there are numerous studies stating that gamma irradiation evenat these doses significantly reduces themechanical properties ofallograft bone (Triantafyllou et al., 1975; Hamer et al., 1996b;Currey et al., 1997; Akkus and Rimnac, 2001; Mitchell et al., 2004;Akkus and Belaney, 2005; Nguyen et al., 2007a; Russell et al.,2012; Kaminski et al., 2012).

As engineering materials, bone might also be subjected tohigh rates of loading in some instances such as sports accidents,traffic accidents involving vehicles, cyclists and/or pedestrians,falls from a height, sudden falls, gun shots and drilling processesin orthopedic surgery. The strength of bone is known to bedependent on the rate of loading and, over a range of loadingrates consistent with normal physiological loads, the mechan-ical properties such as modulus of elasticity, yield strength,ultimate strength and toughness, vary modestly with rate(Wright and Hayes, 1976; Currey, 1989; Evans et al., 1992).Nevertheless, in high speed loading situations, ultimate strengthcan increase considerably while the ultimate strain, a measureof ductility, can decrease considerably (McElhaney, 1966; Lewisand Goldsmith, 1975; Katsamanis and Raftopoulos, 1990; Shazlyet al., 2005; Adharapurapu et al., 2006; Ferreira et al., 2006;Shunmugasamy et al., 2010; Kulin et al., 2011a, 2011b). However,there are contradictory results for modulus of elasticity inprevious studies. While most studies indicated a significantincrease in modulus of elasticity (McElhaney, 1966; Shazly et al.,2005; Adharapurapu et al., 2006; Shunmugasamy et al., 2010;Teja et al., 2013), there are a few studies that indicated eithersmaller increase (Lewis and Goldsmith, 1975) or a significantdecrease (Ferreira et al., 2006) in modulus of elasticity.

Understanding the response of bone under high rates ofloading is critical to accurately model mechanical behaviorof bone, and to predict where and how bone injury mightoccur in case of sports or traffic accidents. For instance,

development of computational models in predicting mechan-ical behavior of bone under high rates of loading is veryimportant in some industries such as automotive (Yang andKajzer, 1994; Bermond et al., 1994; Yang et al., 1996; Tamuraet al., 2001; Nagasaka et al., 2003; Yang et al., 2006; Kerriganet al., 2009; Laporte et al., 2009), sports, personal protectiveequipments and orthopedics (Tsai et al., 2007; Lughmaniet al., 2013; Pandey and Panda, 2013). Because, the successof the computational models is highly dependent on the bonemodels employed. While data exists for the mechanicalbehavior of non-irradiated bone at low and high strain ratesto develop constitutive models, there is no data of gamma-irradiated bone at high strain rates since radiation-inducedembrittlement of bone has only been studied for quasi-staticand cyclic loading conditions. However, most clinical bonefractures occur under high speed loading conditions. In viewof this, the present study is undertaken to examine thecompressive behavior of gamma-irradiated cortical boneunder high speed loading using a Split Hopkinson PressureBar (SHPB) and comparing compressive behavior to that ofnon-irradiated cortical bone. Quasi-static compression is alsoperformed to compare the effects of gamma-irradiation oncortical bone in both quasi-static and high speed loadings.

The SHPB, also called Kolsky bar, is the most commonlyused method for determining the mechanical properties ofvarious materials at high strain rates (Namet-Nasser, 2000;Ramesh, 2008; Weinong and Chen, 2010). However, the SHPBhas mainly been used to test metals, ceramics and other hardmaterials including cancellous (Laporte et al., 2009; Prot et al.,2012; Teja et al., 2013) and cortical bone (McElhaney, 1966;Tennyson et al., 1972; Lewis and Goldsmith, 1975; Katsamanisand Raftopoulos, 1990; Tanabe et al., 1994; Shazly et al., 2005;Adharapurapu et al., 2006; Ferreira et al., 2006; Cloete et al.,2009; Shunmugasamy et al., 2010; Kulin et al., 2011a, 2011b; Tejaet al., 2013). One of the main problems in testing soft materialswith the SHPB is the impedance mismatch between a softmaterial specimen and metallic bars. The introduction ofpolymeric bars in the SHPB apparatus (Salisbury, 2001) hasovercome the impedance matching problem, thereby allowingproper reflection and transmission of strain wave by the speci-men. Recent studies have shown that modifications to thetraditional SHPB allow for the successful characterization ofmechanical properties of soft materials including biological softtissues at high strain rates that exceed alternate soft tissuetesting techniques. In the last ten years, there have been severalstudies using modified SHPB for soft tissues such as muscle,ligament and brain (Sligtenhorst et al., 2006; Chawla et al., 2006;Lennon et al., 2007; Saraf et al., 2007; Trexler et al., 2011; Rashidet al., 2012; Ott et al., 2012; Rashid et al., 2013). Explanation ofSHPB compression system and the equations of stress, strainrate and strain used for SHPB compression system are includedin Appendix A, which were provided by Namet-Nasser (2000).

2. Materials and methods

2.1. Specimen preparation

Cortical bone of larger animal species such as goat, sheep, cow,pig and horse is often used for comparative biomechanical

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2 233

studies (Wang et al., 2010) as it cannot be distinguished fromhuman cortical bone due to their commonly shared secondaryHaversian tissue microstructures when plexiform bone tissue isnot present (Hillier and Bell, 2007). Particularly, there are distinctsimilarities in basic microstructure, and changes occurring inthe microstructure between human and equine bone (Kulinet al., 2011a; Nunamaker, 2002; Cowin, 2001). In our study,which is a comparative biomechanical study, fresh radius,metacarpal, tibia and metatarsal bones of eight years old retiredracehorse (Veliefendi Hipodromu, Istanbul, Turkey) wereobtained, and before compressive testing, its transverse sectionwas examined to see if the plexiform bone tissue was present.A representative micrograph showing the transverse section ofthe equine cortical bone has been given in Fig. 1. As can be seenin Fig. 1, the equine cortical bone has secondary osteonalgrowth with random arrangement of the Haversian systemsand it has no plexiform bone tissue. The microstructure of theequine cortical bone also indicates that the Haversian systemsand porosity have similar dimensions to those of humancortical bone.

Specimen preparation consisted of the following stages:bones were first cleaned by removing excess soft tissue andtheir mid-shaft segments were taken out by cutting theproximal and distal epiphyses using a band saw machine(Promax PM-72151, Istanbul, Turkey). Each mid-shaft seg-ment was further cut longitudinally into four pieces toproduce anterior (A), posterior (P), medial (M) and lateral (L)beams. A table-top lathe (Quantum, D210�320, Germany)was used to machine the beams into 6 mm diameter round

Fig. 1 – Optical micrograph showing the representativetransverse section of the equine cortical bone samples.

Table 1 – Number of equine cortical bone samples by their an

Quasi-static compression testing

Non-irradiated (n¼14) Gamma-irradiated (n¼14)

A P L M A P L M4 2 4 4 4 2 4 4

A: Anterior, P: Posterior, L: Lateral, and M: Medial.

rods and the circumferential surfaces of the round rods werethen polished by hand using a fine grid abrasive paper. The6 mm diameter round rods were placed in a bone chuck of alow speed diamond saw (Struers-Minitom, Ballerup, Den-mark) and parallel cuts were made along the longitudinalaxis to obtain the 5 mm long compression testing specimens.During machining, the specimens were kept wet by a distilledwater drip.

After machining, each compression specimen was labeledto show whether it is obtained from left or right leg (L, R),its bone type (Rd: Radius, Mc: Metacarpal, Tb: Tibia andMt: Metatarsal), and its quadrant (A: Anterior, P: Posterior,M: Medial and L: Lateral). For both quasi-static and high strainrate compression testing, there were two groups of speci-mens: non-irradiated and gamma-irradiated. Specimenswere systematically assigned to the gamma-irradiated andnon-irradiated test groups from either the right or the leftbone of each pair such that specimens from the samediaphyseal location of left and right pairs served as matchedpairs to avoid bias (Table 1). The gamma-irradiated speci-mens were kept wet frozen, were packed in polyethylenebags, and were surrounded with dry ice for shipment forgamma radiation sterilization by a commercial sterilizingvendor (TAEK, Ankara, Turkey). The specimens were gamma-irradiated at room temperature in oxygen and received anaverage dose of 30 kGy, which is commonly employed intissue banks for allograft sterilization.

All specimens were completely wrapped in Hank's balancedsalt solution-soaked tissue paper, placed in airtight polyethy-lene bags and kept frozen at �20 1C until the day of testing. Thespecimens were removed from the freezer prior to testing andallowed to thaw at room temperature in Hank's balanced saltsolution. All tests were carried out at room temperature and adistilled water drip was used to keep the specimens wet duringtesting. The apparent density of cortical bone is calculated asthe mass of the bone divided by its bulk volume, including thevolume associated with the vascular channels and porosity.The average apparent density of equine cortical bone sampleswas found to be 1.92 (70.05) g/cm3, which is within the rangefor human cortical bone. The typical range of apparent densityfor human cortical bone is 1.80–2.00 g/cm3 (Cowin, 2001;Browner, 2003; Langton and NJeh, 2004; Bueno and Glowacki,2011).

2.2. Compression testing

Quasi-static compression tests under displacement controlat an average rate of 0.0045/s were performed with acustom-made testing machine specially designed for mechan-ical testing of low-strength materials at the Mechanical

atomic quadrants in compression testing groups.

High strain rate compression testing

Non-irradiated (n¼14) Gamma-irradiated (n¼14)

A P L M A P L M4 3 3 4 4 3 3 4

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2234

Engineering Department of Suleyman Demirel University inIsparta, Turkey. The custom-made testing machine wasequipped with a 1000 kg load cell (Tedea Huntleigh MN:16,Malvern, USA) and two LVDTs of 10mm stroke (NovotechnikTr10, Germany) to measure the load and the correspondingdisplacement during testing. The LVDTs were attached to thelower loading platen and positioned on the upper loadingplaten at its nearest location to the sample using a custom-made fixture for deformation measurements (Fig. 2). Theaverage deformation for each sample was calculated by aver-aging the readings from the two LVDTs and it was converted tostrain using the sample thickness. Since the axial strain wascalculated using the displacement of the upper loading platenrelative to the lower platen at a point closest to the sample, thedeformations in the load cell, test rig and in the overallmachine structure were excluded to avoid an overestimationof the strain. Friction between the sample and compressionplatens was minimized by applying petroleum jelly on thesurfaces of the upper and lower loading platens. A four-channel oscilloscope (Nicholet-Oddysey XE, USA) was used torecord the measured data from both transducers. The modulusof elasticity was based on the slope of the initial linear portion

Fig. 2 – Schematic diagram showing installation of theLVDTs for the quasi-static compression.

Fig. 3 – Failure point determination for (a) non-irradiated and (b)static stress vs. time and strain vs. time curves. The points of tfailure point.

of the stress vs. strain curve. Ultimate strain is the strain at thefailure point on the stress–strain curve. However, it is wellknown that it is harder to identify the point of failure for aquasi-static compression test compared to a quasi-statictensile test since the specimen continues to carry the loadeven after it is fractured into pieces. In quasi-static tension,however, when fracture starts on the specimen, the loadsharply drops to zero. In our study, strain vs. time and stressvs. time plots were jointly used to identify the failure point(Fig. 3). For non-irradiated and gamma-irradiated samples,after reaching the maximum stress, the strain vs. time curvechanged its slope slightly with decreasing stress and had aplastic deformation, relatively little for gamma-irradiatedsamples. Then, a sudden stress drop and a strain jumpoccurred on the stress vs. time and the strain vs. time curves,respectively. The point corresponding to the strain jump andthe sudden stress drop for both non-irradiated and gamma-irradiated samples was accepted as the failure point, and thestrain corresponding to the failure point was accepted as theultimate strain of the specimen.

A Split Hopkinson Pressure Bar (SHPB) was employed toconduct high strain rate compression tests (Fig. 4). Theapparatus comprises a striker, an incident and a transmitterbar made from high strength 7075-T6 Al alloy (nominal yieldstrength 500 MPa) and having a diameter of 19.05 mm. In thepresent tests striker bar of length 0.3 m was employed, whilethe incident and transmitted bars were 1.85 m and 1.60 mlong respectively. It is very important that the specimenshould be in dynamic equilibrium and deform uniformlyduring the testing. In order to ensure these conditions,a cardboard pulse shaper with a diameter of 4 mm and athickness of 0.8 mm was placed on the impact surface of theincident bar. The striker bar is accelerated using an airoperated gun. Before testing of bone specimens, dummy bonespecimens were used to obtain a perfect alignment of thebars and proper impact velocity for the SHPB compression sothat clean signals of strains on the bars without any distor-tions could be obtained. A pair of foil strain gages (Vishay,Micro Measurements BLH SR-4, FAE-12-100SX) was strategi-cally attached on the incident bar while another pair of foil

gamma-irradiated equine cortical bone samples using quasi-he strain jump and the sudden stress drop indicate the

Fig. 4 – Schematic diagram of the Split Hopkinson PressureBar (SHPB) for compression.

Fig. 5 – Representative oscilloscope traces showing theamplified strain gage outputs for the incident andtransmitted bars of the Split Hopkinson Pressure Bar (SHPB)for compression.

Fig. 6 – Representative bone sample/bar interface stressesand strain rate vs. time curves for the SHPB compressiontesting. r1 and r2 represent stresses at incident bar/sampleand sample/transmitted bar interfaces, respectively.

Fig. 7 – Representative stress–strain curves for non-irradiated and gamma-irradiated equine cortical bonesamples under quasi-static and high strain ratecompression.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2 235

strain gages (Vishay, Micro Measurements BLH SR-4, FAE-12-100SX) was attached on the transmitted bar. These straingages are used in conjunction with Wheatstone bridgecircuits, the output of which is recorded on a digital oscillo-scope (Nicholet-Oddysey XE, USA) to monitor the strainduring the test. By using the data obtained from those straingages, the stress, strain rate and strain values for the speci-mens were calculated using Eqs. (A.7)–(A.9). The modulus ofelasticity was based on the slope of the linear portion of thestress vs. strain curve over the load range in which strain rateis nearly constant. In high strain rate compression of corticalbone, when failure starts on the specimen at about 60 ms oftest duration, sharp changes occur on the reflected andtransmitted bar signals (Fig. 5), which also coincide with thepoint of sudden stress drop (Fig. 6). Ultimate compressivestrain is the strain at the failure point on the stress–straincurve, which also coincides with the point of the ultimatestrength.

The load–displacement data from quasi-static and highstrain rate compressive tests were processed to calculatemodulus of elasticity, yield strength, ultimate strength, yieldstrain, ultimate strain, resilience and toughness using cus-tom written software in Matlab (Math Works, Natick, MA,

USA). The strength values were obtained by dividing the loaddata by the cross-sectional area of each specimen. The strainvalues were obtained by dividing the displacement data bythe initial length of the specimen. The stress value that amaterial can withstand before permanent deformation (non-elastic) occurs is called the yield strength. The 0.2% offsetmethod was used to define the yield strength. The area underthe stress–strain curve in the elastic region was used todetermine the resilience and the total area under thestress–strain curve was used to determine the toughness.

2.3. Statistical analysis

Statistical analysis was conducted using Minitab (Minitab,Inc., State College, PA). Results of the dependent variables(DVs) were expressed as mean and standard deviation (SD).The Anderson–Darling test was used to analyze the normaldistribution of the DVs (P40.05), and the test results indicatedthat all data sets of the DVs were normally distributed. Afterthe data sets passed the normality test, they were analyzedwith two-way analysis of variance (ANOVA), with mechanicalproperties as the DVs and the treatment and strain rate asthe independent variables (IVs). The level of statistical

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2236

significance was set at Po0.05. Multiple comparisons betweenthe testing groups were made by Bonferroni-corrected t-test.

3. Results

Representative stress–strain curves obtained from quasi-static and high strain rate compression tests are shown inFig. 7, and the test results are summarized in Table 2. Verticalbar charts with error bars of 95% Bonferroni ConfidenceInterval (CI) were used to display the mean values of themechanical properties (Fig. 8a–d). The results of the statisticalcomparisons from post hoc tests were indicated on thevertical bar charts using letters. Means sharing the sameletters are not significantly different (Bonferroni correctedt-test, adjusted P0.05, 6 o0.0085).

As it is mentioned previously, it is very important that thespecimen in SHPB testing should be in dynamic stressequilibrium and deform uniformly during the testing.To verify the dynamic equilibrium, the interface stresses s1and s2 were calculated using Eqs. (A.4) and (A.5) and therepresentative results are shown in Fig. 6. Since both curvesof s1 and s2 almost coincide with each other, it can beconcluded that the specimen is in a state of dynamic stressequilibrium. Strain rate vs. time curve in Fig. 6 indicates thatthe constant strain rate was achieved at almost 20 μs, andremained constant up to 60 μs, which is the point of failure.The average strain was 727/s (7117/s) for the non-irradiatedsamples, ranging from 600/s to 850/s and 723/s (7126/s) forthe gamma-irradiated samples, ranging from 575/s to 875/s.

3.1. Modulus of elasticity

The modulus of elasticity was significantly influenced by thestrain rate (po0.05, ANOVA) and treatment (po0.05, ANOVA).The modulus of elasticity was significantly greater for highstrain rate testing compared to quasi-static testing in bothtreatment groups (P0.05/6o0.0085, t-test, Fig. 8a). More impor-tantly, it was significantly increased with irradiation at highstrain rate (P0.05/6o0.0085, t-test) but was not significantlyaffected by irradiation at low strain rate (P0.05/640.0085, t-test)(Fig. 8a).

Table 2 – Means and standard deviations of the mechanical protesting groups.

Mechanical properties Testing groups

QS-NIn

n¼14QSn¼

Modulus of elasticity (GPa) 17.44 (1.67) 16.Yield strength (MPa) 192.41 (7.17) 178Ultimate strength (MPa) 199.01 (6.43) 181Yield strain (mm/mm) 0.0131 (0.0012) 0.0Ultimate strain (mm/mm) 0.0269 (0.0066) 0.0Resilience (MJ/m3) 1.266 (0.155) 1.2Toughness (MJ/m3) 4.129 (1.224) 2.4Strain rate (1/s) 0.0045 (0.0008) 0.0

n QS: Quasi-static, HSR: High strain rate, NI: Non-irradiated, and GI: Gam

3.2. Yield strength

The yield strength was significantly influenced by the strainrate (po0.05, ANOVA) but was not influenced by the treat-ment (p40.05, ANOVA). The yield strength was significantlygreater for high strain rate testing compared to quasi-statictesting in gamma-irradiated treatment group (P0.05/6o0.0085,t-test) but was not significantly changed in non-irradiatedtreatment group (P0.05/640.0085, t-test). More importantly,it was not significantly affected by irradiation at high strainrate (P0.05/640.0085, t-test) but was significantly decreasedwith irradiation at low strain rate (P0.05/6o0.0085, t-test).

3.3. Ultimate strength

The ultimate strength was significantly influenced by thestrain rate (po0.05, ANOVA) and treatment (po0.05, ANOVA).The ultimate strength was significantly greater for high strainrate testing compared to quasi-static testing in both treat-ment groups (P0.05/6o0.0085, t-test, Fig. 8b). More importantly,it was significantly increased with irradiation at high strainrates (P0.05/6o0.0085, t-test) but was decreased with irradia-tion at low strain rates (P0.05/6o0.0085, t-test) (Fig. 8b).

3.4. Yield strain

The yield strain was significantly influenced by the strain rate(po0.05, ANOVA) but was not influenced by the treatment(p40.05, ANOVA). The yield strain was significantly lower forhigh strain rate testing compared to quasi-static testingin gamma-irradiated treatment group (P0.05/6o0.0085, t-test)but was not significantly changed in non-irradiated treat-ment group (P0.05/640.0085, t-test). It was not significantlyaffected by irradiation at both strain rate groups (P0.05/640.0085, t-test).

3.5. Ultimate strain

The ultimate strain was significantly influenced by the treat-ment (po0.05, ANOVA) but was not influenced by the strainrate (p40.05, ANOVA). The ultimate strain was significantlylower for high strain rate testing compared to quasi-statictesting in non-irradiated treatment group (P0.05/6o0.0085,

perties for the equine cortical bone samples in compression

-GI14

HSR-NIn¼14

HSR-GIn¼14

09 (2.12) 20.74 (4.13) 29.97 (7.28).56 (12.48) 203.29 (27.86) 229.63 (36.89).11 (12.54) 239.27 (25.25) 296.00 (42.90)136 (0.0015) 0.0120 (0.0019) 0.0101 (0.0016)196 (0.0045) 0.0212 (0.0029) 0.0213 (0.0039)20 (0.190) 1.649 (0.446) 1.517 (0.333)29 (0.774) 3.579 (0.452) 4.506 (0.971)044 (0.0009) 727 (117) 723 (126)

ma-irradiated.

Fig. 8 – Bar charts showing the modulus of elasticity (a), ultimate strength (b), ultimate strain (c) and toughness (d) for non-irradiated and gamma-irradiated equine cortical bone samples under quasi-static and high strain rate compression testing.Error bars indicate 95% Bonferroni confidence intervals (CIs). A, B, C, and D: Means not sharing the same letter are significantlydifferent (adjusted P0.05, 6 o0.0085).

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2 237

t-test) but was not significantly changed in gamma-irradiatedtreatment group (P0.05/640.0085, t-test) (Fig. 8c). More impor-tantly, it was not significantly affected by irradiation athigh strain rate (P0.05/640.0085, t-test) but was significantlydecreased with irradiation at low strain rate (P0.05/6o0.0085,t-test) (Fig. 8c).

3.6. Resilience

The resilience was significantly influenced by the strain rate(po0.05, ANOVA) but was not influenced by the treatment(p40.05, ANOVA). The resilience was significantly greater forhigh strain rate testing compared to quasi-static testingin non-irradiated treatment group (P0.05/6o0.0085, t-test) butwas not significantly changed in gamma-irradiated treatmentgroup (P0.05/640.0085, t-test). It was not significantly affectedby irradiation at both strain rate groups (P0.05/640.0085,t-test).

3.7. Toughness

The toughness was significantly influenced by the strain rate(po0.05, ANOVA) but was not influenced by the treatment(p40.05, ANOVA). The toughness was significantly greater forhigh strain rate testing compared to quasi-static testing ingamma-irradiated treatment group (P0.05/6o0.0085, t-test) butwas not significantly changed in non-irradiated treatmentgroup (P0.05/640.0085, t-test) (Fig. 8d). More importantly, it wassignificantly increased with irradiation at high strain rate(P0.05/6o0.0085, t-test) but was significantly decreased withirradiation at low strain rate (P0.05/6o0.0085, t-test) (Fig. 8d).

4. Discussion and conclusion

Agreeing with most of the previously published results on theembrittlement of cortical bone when gamma-irradiated, the

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2238

quasi-static results showed that gamma-irradiation signifi-cantly decreased ultimate strength (9%), ultimate strain (27%)and toughness (41%), while not having significant effecton modulus of elasticity, yield strain and resilience. As hasbeen previously outlined, equine cortical bone is primarilycomposed of mineral and collagen, with mechanical proper-ties determined primarily by the amounts, arrangement, andmolecular structure of these primary constituents. Collagenstructure is altered by gamma-irradiation (Akkus andBelaney, 2005), resulting in significant reductions in post-yield (plastic) properties such as ultimate strength, ultimatestrain and toughness, rather than the pre-yield (elastic)properties such as modulus of elasticity, yield strength andyield strain (Hamer et al., 1996b; Currey et al., 1997; Akkus andRimnac, 2001; Akkus and Belaney, 2005; Russell et al., 2012;Kaminski et al., 2012). During loading into the plastic regionfor non-irradiated bone, intact collagen fibers provide abridging and reinforcement function to the bone matrix.However, for gamma-irradiated bone, collagen fibers fail toprovide bridging at the ultra-structural level, which indicatesthat when the integrity of the collagen matrix is damagedby gamma-irradiation, individual molecules collapse underloading (Akkus and Belaney, 2005).

More importantly, contrary to what is typically observed inquasi-static loading, the gamma-irradiated cortical bonesamples under high speed loading showed significantlyhigher modulus of elasticity (45%), ultimate strength (24%)and toughness (26%) than those of non-irradiated samples,although the failure was at a similar strain. Collagen has beenshown to increase in modulus of elasticity and become morebrittle with increasing strain rate, while the mineral phase isexpected to have minimal viscoelastic response due to itsvery ceramic-like nature (Ntim et al., 2005; Kulin et al., 2011a),

Table 3 – Comparison of some mechanical properties of non-iHopkinson Pressure Bar (SHPB) compression in the current an

References Bone source Strain rate, /s M

Mel

Current study Equine 7277117(600–850)

H

McElhaney (1966) Bovine,human

300, 1500 H

Tennyson et al. (1972) Bovine 450 HLewis and Goldsmith (1975) Bovine Unknown HKatsamanis and Raftopoulos(1990)

Human 100 H

Shazly et al. (2005) Equine 300–1300 HAdharapurapu et al. (2006) Bovine 1000 HFerreira et al. (2006) Bovine 6677184

(369–1070)Lo

Cloete et al. (2009) Bovine 300, 1150 HShunmugasamy et al. (2010) Rabbit 225-775 HKulin et al. (2011a) Equine 1000 NKulin et al. (2011b) Elk antler 1000 NTeja et al. (2013) Human 200 H

a Higher: SHPB test results compared to quasi-static test results.b Lower: SHPB test results compared to quasi-static test results.c NR: Not reported or not reached.

and also due to the rapid nature of the loading rate thathas no time for peripheral damage to stably form (Kulinet al., 2011c). In gamma-irradiated cortical bone, collagen isdamaged to the extent that it is present within the bone, butit is functionally irrelevant. Since our results have shown thatgamma-irradiated bone performs better than non-irradiatedbone at high strain rate, the functional absence of collagenseems to be increasing the fracture resistance at high strainrate, which indicates that, in traumatic loading conditions,bone allograft may not necessarily be the weaker component.However, this result needs to be confirmed by other studiesthat will include damage analysis and will investigatewhich mechanisms could be behind it. To our knowledge,no previous work has been carried out on the mechanicalproperties of gamma-irradiated cortical bone under highspeed loading. Therefore, no direct comparison with previousresults could be performed. However, dry vs. wet bone is agood example of this behavior as both the dry and wet bonesamples exhibit increase in modulus of elasticity andstrength with increasing strain rate, and the dry bone, morebrittle than wet bone, exhibits higher modulus of elasticityand fails at higher stresses than the wet bone (Adharapurapuet al., 2006; Kulin et al., 2011b).

Non-irradiated cortical bone under high strain rate loadingdemonstrated significant differences in modulus of elasticity(19% increase), ultimate strength (20% increase), resilience(30% increase) and ultimate strain (20% decrease), while nothaving significant differences in yield strength, yield strainand toughness compared to those of non-irradiated corticalbone under quasi-static loading. Table 3 gives the comparisonof some mechanical properties of non-irradiated corticalbone between quasi-static and high strain rate compressionin the current and previous studies. Applied high strain rates

rradiated cortical bone between quasi-static and Splitd previous studies.

echanical properties

odulus ofasticity

Ultimatestrength

Ultimatestrain

Toughness

ighera Higher Lowerb Lower

igher Higher Lower NRc

igher NR NR NRigher Higher NR NRigher NR NR NR

igher Higher Lower NRigher Higher Lower NRwer Higher NR NR

igher Higher Lower NRigher Higher Lower NRR Higher NR NRR Higher Lower NRigher Higher NR NR

Table 4 – Comparison of some mechanical properties of non-irradiated equine cortical bone in the current study with thoseof non-irradiated equine and human cortical bones in the previous studies.

References Bone source StrainRate, /s

Mechanical properties

Modulus ofelasticity, GPa

Ultimatestrength, MPa

Ultimate strain,mm/mm

Toughness,MJ/m3

Current Study Equine 0.0045 17.4 199 0.027 4.129Riggs et al. (1993) Equine NR 17 201 NRa NRShazly et al. (2005) Equine 0.017 15.2 214 NR NRKulin et al. (2011a) Equine 0.001 18.6 190 NR NRReilly and Burstein(1975)

Human femur NR 17.9 205 NR NR

Burstein et al.(1976)

Human tibia 0.05 25.1 183 NR NR

Kemper et al. (2007) Human tibia 0.042 16.9 177 0.016 NROhman et al. (2008) Human femur 1.2 mm/s 16.8 162 0.015 NRHansen et al. (2008) Human femur 0.14 17 175 0.018 NRLeng et al. (2009) Human tibia 0.005 mm/s 19.1 132 NR NRBueno andGlowacki (2011)

Human femur NR 18.2 195 NR NR

Öhman et al. (2011) Human femurand tibia

0.1 17.2 198 0.018 NR

Dong et al. (2012) Human femur 0.001 19.1 124 NR NR

a NR: Not reported or not reached.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2 239

in the previous studies varied between 100/s and 1500/s, andthe comparison results of the current study are consistentwith those of the previous studies (Table 3), which may alsosupport the validity of the results of this study for gamma-irradiated cortical bone under high speed loading. Comparingto quasi-static results, ultimate strength is higher in all of thestudies, modulus of elasticity is higher in all except the studyby Ferreira et al. (2006), and ultimate strain is lower in thestudies reporting ultimate strain. The current study alsosupports the phenomena that have been previously reported(Shazly et al., 2005). In particular, the results from the SHPBcompression showed there is a complete loss of load carryingability under high speed loading. Typically, the load fell tozero at a rate similar to the rate at which the load increased(Fig. 6). In contrast, quasi-statically loaded samples typicallysustained substantial compressive loads (Fig. 7) beyond thedeformations associated with ultimate stress (Burstein et al.,1972, 1975; Reilly and Burstein, 1975; Shazly et al., 2005).

There are several limitations to this study. First, all speci-mens were tested under compressive loading up to failure; notension and torsion testing were performed to investigate theeffects of gamma irradiation sterilization on high strain rateproperties under those loading conditions. Second limitationis that all specimens of high strain rate testing, gamma-irradiated and non-irradiated, were tested at an averagestrain rate of 725/s. Further studies might be useful to confirmthe results of this study by testing the specimens undervarious high strain rates ranging from 100/s to 1000/s. Thirdlimitation is that equine cortical bone was used instead ofhuman cortical bone. However, regardless of animal model,cortical bone has similar constituents at molecular size scale,collagen, mineral, non-collagenous proteins and proteogly-cans. If the gamma irradiation did not affect one of theseconstituents – or interactions between them – at equine

cortical bone, we do not expect the gamma irradiation tobehave differently for human cortical bone. While equinebone, as other animal bones, may closely represent themechanical and physiological human clinical situation, itmust be remembered that it is only an approximation. Useof the data from animal studies should therefore focus moreupon the effects of change in structure and composition onmechanical properties as opposed to using the data as asubstitute for the properties of human bones (Wang et al.,2010). Additional experiments might be required to extendthese findings of equine cortical bone to human cortical bone.Table 4 gives the comparison of some mechanical propertiesof non-irradiated equine cortical bone in the current studywith those of non-irradiated equine and human corticalbones in the previous studies. Modulus of elasticity andultimate strength of human cortical bone in the previousstudies cover a wide range of values, and the modulus ofelasticity of equine cortical bone remains in that range.However, the ultimate strength and ultimate strain of equinecortical bone are generally higher than those of human corticalbone in most other studies (Table 4). It should be rememberedthat the mechanical properties of bone not only vary from onespecies to another, but also for similar bones, or even within asame bone, which reflect local structural variations. Finally,the comparisons between the two strain rates using twodifferent test methods may also be thought of as a limitation.However, when the effect of strain rate on materials' mechan-ical properties is investigated, it is not always possible to usethe same testing machine, or test method, for all strain ratesfrom low to high, and in most of previous studies, SHPB wasused for high strain rates while hydraulic and electromecha-nical testers were used for low strain rates.

Despite these limitations, the results of this study haveseveral scientific implications rather than direct clinical

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2240

implications. First, this study suggests that the numericalmodels involving bone model need to be modified to accu-rately reflect bone's mechanical behavior at high strainrates if it was sterilized with gamma-irradiation. Second,probably the most important, the results of this studycall into question the assumption that bone allograft isalways degraded by gamma irradiation, regardless of load-ing conditions. Under high speed loading, the mechanicalproperties of bone allograft were not degraded by irradiation,in contrast to the degradation measured in this andprior studies under quasi-static loading. However, it needsfurther investigation to be translated positively in a clinicalsetting.

Acknowledgments

This article is based on the doctoral thesis of the first authorat Suleyman Demirel University under supervision of the twoother authors. This study was supported by the SüleymanDemirel University (SDU BAP 1401-D-06). We thank Mr. TalatAydin and the Turkish Atomic Energy Authority (TAEK)for the sterilization of the bone specimens. We also thankDr. Vikas Prakash of Case Western Reserve University andDr. Murat Vural of Illinois Institute of Technology (IIT) fortheir assistance in the experimental setup of SHPB.

Appendix A

In a SHPB compression system, two symmetrical bars aresituated in series, with the specimen between (Fig. 4). Thefirst bar is the incident bar, which is struck by a striker barduring testing. The striker bar is fired from an air gun. Thesecond bar is the transmitted bar, which collides with astopper. By striking to the end of incident bar, a compressivestress wave is generated that immediately begins to traveltowards the specimen. Upon arrival at the incident bar–specimen interface, the wave partially reflects back to theimpact end. The remainder of the wave transmits throughthe specimen into the second bar causing deformation in thespecimen (Kaiser, 1998). If the two bars remain elastic andwave dispersion ignored, then the measured stress pulsescan be assumed to be the same as the one acting on thespecimen. Incident and transmitted bars were made of thesame material with equal cross-sectional areas. In the equa-tions below (Namet-Nasser, 2000), the following notations areused: incident (I), transmitted (T), reflected (R), specimen (s),density (ρ), modulus of elasticity (E), wave speed (c) and cross-sectional area (A) of the bars and the cross-sectional area (As)and length (l0) of the specimen (Fig. 4).

If the specimen deforms uniformly, the strain rate _εs iscalculated as

_εs ¼ dεsdt

¼ v1ðtÞ�v2ðtÞl0

ðA:1Þ

The velocity at interface 1 and interface 2 can be written asfollows

v1ðtÞ ¼ cðεIðtÞ�εRðtÞÞ; v2ðtÞ ¼ cεTðtÞ ðA:2Þ

By substituting these interface velocities into equation (A.1)

_εs ¼cðεIðtÞ�εRðtÞ�εTðtÞÞ

l0ðA:3Þ

Stresses at the ends of the specimen are

s1ðtÞ ¼ EAAs

ðεIðtÞ þ εRðtÞÞ ðA:4Þ

s2ðtÞ ¼EAAs

εTðtÞ ðA:5Þ

If the specimen is in dynamic stress equilibrium

εIðtÞ þ εRðtÞ ¼ εTðtÞ ðA:6Þ

Then the stress, strain rate and strain are given by

ssðtÞ ¼s1ðtÞ þ s2ðtÞ

2¼ EA

AsðεIðtÞ þ εRðtÞÞ ðA:7Þ

_εsðtÞ ¼ �2cl0

εRðtÞ ðA:8Þ

εsðtÞ ¼ �2cl0

ZεRðtÞdt ðA:9Þ

r e f e r e n c e s

Adharapurapu, R.R., Jiang, F., Vecchio, K.S., 2006. Dynamicfracture of bovine bone. Mater. Sci. Eng. C 26, 1325–1332.

Akkus, O., Rimnac, C.M., 2001. Fracture resistance of gammaradiation sterilized cortical bone allografts. J. Orthop. Res. 19(5), 927–934.

Akkus, O., Belaney, R.M., 2005. Sterilization by gamma radiationimpairs the tensile fatigue life of cortical bone by two ordersof magnitude. J. Orthop. Res. 23 (5), 1054–1058.

Angermann, P., Jepsen, O.B., 1991. Procurement, banking anddecontamination of bone and collagenous tissue allografts:guidelines for infection control. J. Hosp. Infect. 17 (3), 159–169.

Bermond, F., Ramet, M., Bouquet, R., Cesari, D., 1994. A finiteelement model of the pedestrian leg in lateral impact. In:Proceedings of the 14th International Technical Conference onthe Enhanced Safety of Vehicles (ESV), May 23–26, 1994,Munich, Germany, pp. 199–209.

Bonafedea, M., Espindlea, D., Bowerb, A.G., 2013. The direct andindirect costs of long bone fractures in a working age USpopulation. J. Med. Econ. 16 (1), 169–178.

Bright, R.W., Smarsh, J.D., Gambill, V.M., 1983. Sterilization ofhuman bone by irradiation. In: Friedlander, G.E., Mankin, H.J.,Sell, K.W. (Eds.), Osteochondral Allografts: Biology, Bankingand Clinical Applications. Little, Brown, Boston, pp. 223–232.

Browner, B.D., 2003. Skeletal Trauma: Basic Science, Managementand Reconstruction, 3rd ed. Saunders, Philadelphia, PA.

Bueno, E.M., Glowacki, J., 2011. Biologic Foundations for SkeletalTissue Engineering. Synthesis Lectures on Tissue Engineering.Morgan & Claypool, San Rafael, CA.

Burge, R., Dawson-Hughes, B., Solomon, D.H., Wong, J.B., King, A.,Tosteson, A., 2007. Incidence and economic burden ofosteoporosis-related fractures in the United States, 2005–2025.J. Bone Miner. Res. 22 (3), 465–475.

Burstein, A.H., Currey, J.D., Frankel, V.H., Reilly, D.T., 1972. Theultimate properties of bone tissue: the effects of yielding.J. Biomech. 5 (1), 35–44.

Burstein, A.H., Zika, J.M., Heiple, K.G., Klein, L., 1975. Contributionof collagen and mineral to the elastic-plastic properties ofbone. J. Bone Jt. Surg. Am. 57 (7), 956–961.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2 241

Burstein, A.H., Reilly, D.T., Martens, M., 1976. Aging of bone tissue:

mechanical properties. J. Bone Jt. Surg. Am. 58 (1), 82–86.Chawla, A., Mukherjee, S., Marathe, R., Karthikeyan, B., Malhotra,

R., 2006. Determining strain rate dependence of human body

soft tissues using a Split Hopkinson Pressure Bar. In: 2006

International IRCOBI Conference on the Biomechanics of

Impact, Madrid, Spain, 20–22 September 2006, pp. 173–182.Cloete, T.J., van der Westhuizen, A., Kok, S., Nurick, G.N., 2009.

A tapered striker pulse shaping technique for uniform strain

rate dynamic compression of bovine bone. In: DYMAT 2009–

9th International Conference on the Mechanical and Physical

Behaviour of Materials under Dynamic Loading, vol. 1,

September 7–11, 2009, Brussels, Belgium, pp. 901–907.Cowin, S.C., 2001. In: Bone Mechanics Handbook 2nd ed. CRC,

Boca Raton, FL, London.Currey, J.D., 1989. Strain rate dependence of the mechanical

properties of reindeer antler and the cumulative damage

model of bone fracture. J. Biomech. 22 (5), 469–475.Currey, J.D., Foreman, J., Laketic, I., Mitchell, J., Pegg, D.E., Reilly,

G.C., 1997. Effects of ionizing radiation on the mechanical

properties of human bone. J. Orthop. Res. 15 (1), 111–117.Dong, X.N., Acuna, R.L., Luo, Q., Wang, X., 2012. Orientation

dependence of progressive post-yield behavior of

human cortical bone in compression. J. Biomech. 45 (16),

2829–2834.Evans, G.P., Behiri, J.C., Vaughan, L.C., Bonfield, W., 1992. The

response of equine cortical bone to loading at strain rates

experienced in vivo by the galloping horse. Equine Vet. J. 24 (2),

125–128.Ferreira, F., Vaz, M.A., Simoes, J.A., 2006. Mechanical properties of

bovine cortical bone at high strain rate. Mater. Charact. 57 (2),

71–79.Hamer, A.J., Colwell, A., Eastell, R., 1996a. Biomechanical and

biochemical changes in cortical allograft bone after gamma

irradiation. Bone 19 (6), 696.Hamer, A.J., Strachan, J.R., Black, M.M., Ibbotson, C.J., Stockley, I.,

Elson, R.A., 1996b. Biomechanical properties of cortical

allograft bone using a new method of bone strength

measurement: a comparison of fresh, fresh-frozen and

irradiated bone. J. Bone Jt. Surg. Br. 78 (3), 363–368.Hamer, A.J., Stockley, I., Elson, R.A., 1999. Changes in allograft

bone irradiated at different temperatures. J. Bone Jt. Surg. Br.

81 (2), 342–344.Hansen, U., Zioupos, P., Simpson, R., Currey, J.D., Hynd, D., 2008.

The effect of strain rate on the mechanical properties of

human cortical bone. J. Biomech. Eng. 130 (1), 011011.Hillier, M.L., Bell, L.S., 2007. Differentiating human bone from

animal bone: a review of histological methods. J. Forensic Sci.

52 (2), 249–263.Kaiser, M.A., 1998. Advancements in the Split Hopkinson Bar Test

(Master of Thesis). Faculty of the Virginia Polytechnic Institute

and State University, Blacksburg, Virginia.Kaminski, A., Jastrzebska, A., Grazka, E., Marowska, J., Gut, G.,

Wojciechowski, A., Uhrynowska-Tyszkiewicz, I., 2012. Effect of

gamma irradiation on mechanical properties of human

cortical bone: influence of different processing methods. Cell

Tissue Bank. 13 (3), 363–374.Katsamanis, F., Raftopoulos, D.D., 1990. Determination of

mechanical properties of human femoral cortical bone by the

Hopkinson bar stress technique. J. Biomech. 23 (11),

1173–1184.Kemper, A.R., McNally, C., Kennedy, E.A., Manoogian, S.J., Duma,

S.M., 2007. The material properties of human tibia cortical

bone in tension and compression: implications for the

tibia index. In: Proceedings of the 20th Enhanced Safety

of Vehicles Conference, Lyon, France, Paper Number:

07-0470.

Kennedy, J.F., Philips, G.O., Williams, P.A., 2005. Sterilization of

tissues using ionizing radiations. CRC Press, Boca Raton,

Florida, Cambridge.Kerrigan, J.R., Parent, D.P., Untaroiu, C., Crandall, J.R., Deng, B.,

2009. A new approach to multibody model development:

pedestrian lower extremity. Traffic Inj. Prev. 10 (4), 386–397.Komender, A., 1976. Influence of preservation on some

mechanical properties of human haversian bone. Mater. Med.

Pol. 8 (1), 13–17.Kulin, R.M., Jiang, F., Vecchio, K.S., 2011a. Effects of age and

loading rate on equine cortical bone failure. J. Mech. Behav.

Biomed. Mater. 4, 57–75.Kulin, R.M., Chen, P.Y., Jiang, F., Vecchio, K.S., 2011b. A study of

the dynamic compressive behavior of Elk antler. Mater. Sci.

Eng. C 31, 1030–1041.Kulin, R.M., Jiang, F., Vecchio, K.S., 2011c. Loading rate effects on

the R-curve behavior of cortical bone. Acta Biomater. 7 (2),

724–732.Langton, C.M., NJeh, C.F., 2004. The Physical Measurement of

Bone (Series in Medical Physics and Biomedical Engineering).

Taylor & Francis, Philadelphia, PA.Laporte, S., David, F., Bousson, V., Pattofatto, S., 2009. Dynamic

behavior and microstructural properties of cancellous bone.

In: DYMAT 2009–9th International Conference on the

Mechanical and Physical Behaviour of Materials under

Dynamic Loading, vol. 1, September 7–11, 2009, Brussels,

Belgium, pp. 895–900.Leng, H., Dong, X.N., Wang, X., 2009. Progressive post-yield

behavior of human cortical bone in compression for

middle-aged and elderly groups. J. Biomech. 42 (4),

491–497.Lennon, A., Merkle, A., Roberts, J., Pirtini, M., Saraf, H., Ramesh,

K.T., 2007. Modifications to the compression kolsky bar for

characterizing soft biomaterials at impact loading rates. In:

International SAMPE Symposium and Exhibition

(Proceedings), vol. 52, Baltimore, MD, United States, 3–7 June

2007.Lewis, J.L., Goldsmith, W., 1975. The dynamic fracture and

prefracture response of compact bone by split Hopkinson bar

methods. J. Biomech. 8 (1), 27–40.Lietman, S.A., Tomford, W.W., Gebhardt, M.C., Springfield, D.S.,

Mankin, H.J., 2000. Complications of irradiated allografts in

orthopaedic tumor surgery. Clin. Orthop. Relat. Res. 375,

214–217.Lughmani, W.A., Bouazza-Marouf, K., Ashcroft, I., 2013. Finite

element modeling and experimentation of bone drilling

forces. J. Phys. Conf. Ser. 451, 012034.Mardas, M., Kubisz, L., Biskupski, P., Mielcarek, S., Stelmach-

Mardas, M., Kaluska, I., 2012. Radiation sterilized bone

response to dynamic loading. Mater. Sci. Eng. C 32 (6),

1548–1553.McAllister, D.R., Joyce, M.J., Mann, B.J., Vangsness, C.T., 2007.

Allograft update: the current status of tissue regulation,

procurement, processing, and sterilization. Am. J. Sports Med.

35 (12), 2148–2158.McElhaney, J.H., 1966. Dynamic response of bone and muscle

tissue. J. Appl. Physiol. 21 (4), 1231–1236.Mitchell, E.J., Stawarz, A.M., Kayacan, R., Rimnac, C.M., 2004. The

effect of gamma radiation sterilization on the fatigue crack

propagation resistance of human cortical bone. J. Bone Jt.

Surg. Am. 86-A (12), 2648–2657.Nagasaka, K., Mizuno, K., Tanaka, E., Yamamoto, S., Iwamoto, M.,

Miki, K., Kajzer, J., 2003. Finite element analysis of knee injury

risks in car-to-pedestrian impacts. Traffic Inj. Prev. 4 (4),

345–354.Namet-Nasser, S., 2000. High strain rate testing. In: Kuhn, H.,

Medlin, D. (Eds.), ASM Handbook Volume 08: Mechanical

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 2 3 1 – 2 4 2242

Testing and Evaluation. ASM International, Materials Park,OH, pp. 425–560.

Nemzek, J.A., Arnoczky, S.P., Swenson, C.L., 1994. Retroviraltransmission by the transplantation of connective-tissueallografts: an experimental study. J. Bone Jt. Surg. Am. 76 (7),1036–1041.

Nguyen, H., Morgan, D.A.F., Forwood, M.R., 2007a. Sterilization ofallograft bone: effects of gamma irradiation on allograftbiology and biomechanics. Cell Tissue Bank. 8 (2), 93–105.

Nguyen, H., Morgan, D.A.F., Forwood, M.R., 2007b. Sterilization ofallograft bone: is 25 kGy the gold standard for gammairradiation?. Cell Tissue Bank. 8 (2), 81–91.

Ntim, M., Bembey, A., Ferguson, V., Bushby, A., 2005. Hydrationeffects on the viscoelastic properties of collagen. MRS Proc.898 , 0898-L05-02 http://dx.doi.org/10.1557/PROC-0898-L05-02.

Nunamaker, D.M., 2002. On bone and fracture treatment in thehorse. In: Proceedings of the Annual Convention of the AAEP2002, vol. 48, pp. 90–101.

Ott, K.A., Armiger, R.S., Wickwire, A.C., Iwaskiw, A.S., Merkle,A.C., 2012. Determination of simple shear material propertiesof the brain at high strain rates. In: Annual Conference onExperimental and Applied Mechanics, 11–14 June 2012, vol. 1,Costa Mesa, CA, United States, pp. 139–147.

Ohman, C., Dall’Ara, E., Baleani, M., Sint Jan, S.V., Viceconti, M.,2008. The effects of embalming using a 4% formalin solutionon the compressive mechanical properties of human corticalbone. Clin. Biomech. 23 (10), 1294–1298.

Ohman, C., Baleania, M., Pania, C., Taddeia, F., Alberghinic, M.,Vicecontia, M., Manfrinid, M., 2011. Compressive behaviour ofchild and adult cortical bone. Bone 49 (4), 769–776.

Pandey, R.K., Panda, S.S., 2013. Drilling of bone: a comprehensivereview. J. Clin. Orthop. Trauma 4 (1), 15–30.

Prot, M., Cloete, T.J., Pattofatto, S., 2012. Dynamic compressionand recovery of cancellous bone for microstructuralinvestigation. In: DYMAT 2012–10th International Conferenceon the Mechanical and Physical Behaviour of Materials underDynamic Loading, vol. 26, September 2nd–7th, 2012, Freiburg,Germany, id.03003.

Ramesh, K.T., 2008. High rates and impact experiments. In:Sharpe, W.N. (Ed.), Springer Handbook of Experimental SolidMechanics. Springer, Newyork, pp. 929–960.

Rashid, B., Destrade, M., Gilchrist, M.D., 2012. Mechanicalcharacterization of brain tissue in compression at dynamicstrain rates. J. Mech. Behav. Biomed. Mater. 10, 23–38.

Rashid, B., Destrade, M., Gilchrist, M.D., 2013. Mechanicalcharacterization of brain tissue in simple shear at dynamicstrain rates. J. Mech. Behav. Biomed. Mater. 28, 71–85.

Reilly, D.T., Burstein, A.H., 1975. The elastic and ultimateproperties of compact bone tissue. J. Biomech. 8 (6), 393–405.

Riggs, C.M., Vaughan, L.C., Evans, G.P., Lanyon, L.E., Boyde, A.,1993. Mechanical implications of collagen fibre orientation incortical bone of the equine radius. Anat. Embryol. 187 (3),239–248.

Russell, N.A., Pelletier, M.H., Bruce, W.J., Walsh, W.R., 2012. Theeffect of gamma irradiation on the anisotropy of bovinecortical bone. Mech. Eng. Phys. 34, 1117–1122.

Salisbury, C., 2001. Spectral Analysis of Wave PropagationThrough a Polymeric Hopkinson Bar (Masters Thesis).University of Waterloo, Waterloo, ON, CA.

Saraf, H., Ramesh, K.T., Lennon, A.M., Merkle, A.C., Roberts, J.C.,2007. Mechanical properties of soft human tissues underdynamic loading. J. Biomech. 40 (9), 1960–1967.

Shazly, M., Kayacan, R., Prakash, V., Rimnac, C., Davy, D., 2005.Failure of equine compact bone under impact loading. In:Proceedings of the 11th International Conference on Fracture,20–25 March 2005, Turin, Italy.

Shunmugasamy, V.C., Gupta, N., Coelho, P.G., 2010. High strainrate response of rabbit femur bones. J. Biomech. 43 (15),3044–3050.

Sligtenhorst, C.V., Cronin, D.S., Brodland, G.W., 2006. Strain ratecompressive properties of bovine muscle tissue determinedusing a split Hopkinson bar apparatus. J. Biomech. 39 (10),1852–1858.

Tamura, A., Miki, K., Yang, K., 2001. A tibial mid-shaft injurymechanism in frontal automotive crashes. In: Proceedings ofthe 17th International Technical Conference on the EnhancedSafety of Vehicles (ESV), June 4–7, 2001, Amsterdam, TheNetherlands.

Tanabe, Y., Kobayashi, K., Sakamoto, M., Hara, T., Takahashi, H.,1994. Identification of the dynamic properties of bone usingthe Split-Hopkinson Pressure-Bar technique. In: Kambic, H.E.,Yokobori, A.T. (Eds.), Biomaterials’ Mechanical Properties.ASTM STP 1173. American Society of Testing Materials,Philadelphia, pp. 127–141.

Teja, C.K., Chawlaa, A., Mukherjeea, S., 2013. Determining thestrain rate dependence of cortical and cancellous bones ofhuman tibia using a Split Hopkinson pressure bar. Int.J. Crashworthiness 18 (1), 11–18.

Tennyson, R.C., Ewert, R., Niranjan, V., 1972. Dynamic viscoelasticresponse of bone. Exp. Mech. 12 (11), 502–507.

Trexler, M.M., Lennon, A.M., Wickwire, A.C., Harrigan, T.P., Luong,Q.T., Graham, J.L., Maisano, A.J., Roberts, J.C., Merkle, A.C.,2011. Verification and implementation of a modified splitHopkinson pressure bar technique for characterizingbiological tissue and soft biosimulant materials underdynamic shear loading. J. Mech. Behav. Biomed. Mater. 4 (8),1920–1928.

Triantafyllou, N., Sotiropoulos, E., Triantafyllou, J.N., 1975. Themechanical properties of the lyophylized and irradiated bonegrafts. Acta Orthop. Belg. 41 (Suppl 1), S35–S44(1), S35–S44.

Tsai, M.D., Hsieh, M.S., Tsai, C., 2007. Bone drilling hapticinteraction for orthopaedic surgical simulator. Comput. Biol.Med. 37 (12), 1709–1718.

Wang, X., Nyman, J.S., Dong, X., Leng, H., Reyes, M., 2010.Fundamental Biomechanics in Bone Tissue Engineering.Synthesis Lectures on Tissue Engineering 2 (1), 1–225.

Weinong, W., Chen, B.S., 2010. Split Hopkinson (Kolsky) Bar:Design, Testing and Applications. Springer, Newyork.

Wright, T.M., Hayes, W.C., 1976. Tensile testing of bone over awide range of strain rates: effects of strain rate,microstructure and density. Med. Biol. Eng. 14 (6), 671–680.

Yang, J.K., Kajzer, J., 1994. Mathematical model of pedestrianlower extremity. In: Proceedings of the 14th InternationalTechnical Conference on the Enhanced Safety of Vehicles(ESV), May 23–26, 1994, Munich, Germany, pp. 220–232.

Yang, J.K., Wittek, A., Kajzer, J., 1996. Finite element model of thehuman lower extremity skeleton system in lateral impact. In:Proceedings of the 1996 International IRCOBI Conference onthe Biomechanics of Impact, September 11–13, 1996, Dublin,Ireland.

Yang, K.H., Hu, J., White, N.A., King, A.I., Chou, C.C., Prasad, P.,2006. Development of numerical models for injurybiomechanics research: a review of 50 years of publications inthe Stapp Car Crash Conference. Stapp Car Crash J. 50,429–490.