chalk steel interface testing for marine energy foundations · of very high density according to...

University of Dundee

Chalk-steel Interface testing for marine energy foundations

Ziogos, Andreas; Brown, Michael; Ivanovi, Ana; Morgan, Neil

Published in:Proceedings of the Institution of Civil Engineers: Geotechnical Engineering

DOI:10.1680/jgeen.16.00112

Publication date:2017

Document VersionPublisher's PDF, also known as Version of record

Link to publication in Discovery Research Portal

Citation for published version (APA):Ziogos, A., Brown, M., Ivanovi, A., & Morgan, N. (2017). Chalk-steel Interface testing for marine energyfoundations. Proceedings of the Institution of Civil Engineers: Geotechnical Engineering, 170(3), 285-298.https://doi.org/10.1680/jgeen.16.00112

General rightsCopyright and moral rights for the publications made accessible in Discovery Research Portal are retained by the authors and/or othercopyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated withthese rights.

• Users may download and print one copy of any publication from Discovery Research Portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain. • You may freely distribute the URL identifying the publication in the public portal.

Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 29. Jul. 2021

https://doi.org/10.1680/jgeen.16.00112

https://discovery.dundee.ac.uk/en/publications/ce5591fd-e6a6-4267-b91a-dec675558483

https://doi.org/10.1680/jgeen.16.00112

Chalk–steel interface testingfor marine energy foundations&1 Andreas Ziogos MEng

Doctoral Researcher, Civil Engineering, University of Dundee, Dundee,Scotland (corresponding author: [email protected])

&2 Michael Brown BEng, PhDReader, Civil Engineering, University of Dundee, Dundee, Scotland

&3 Ana Ivanovic MEng, PhD, CEng, MICEProfessor, Department of Civil Engineering, University of Aberdeen,Aberdeen, Scotland

&4 Neil Morgan BEng, PhD, CEng, IMarEstLead Geotechnical Engineer, Lloyd’s Register EMEA, Aberdeen, Scotland

1 2 3 4

To aid deployment and recovery of tidal stream generators, gravity-based foundations rather than fixed-foundation

alternatives are being considered in areas where the foundation may be placed directly onto an exposed rock seabed.

Horizontal loading is usually critical in such applications, therefore specific knowledge of the interface friction between

the foundation (made of steel or concrete) and seabed is important for design. This paper presents results of an

interface testing programme of chalk–steel interfaces carried out utilising a computer-controlled interface shear tester

under constant normal stress conditions against steel of different roughness. Results indicate that interface strength is

significantly affected by the normal stress applied, as interface strength degrades for normal stress levels in excess of

30% of the chalk’s tensile strength (�300 kPa). Large-displacement tests revealed a tendency of the ultimate interface

frictional resistance to drop to values very similar to that of the basic chalk–chalk interface at normal stresses up to

300 kPa, whereas substantial additional degradation was noticed for normal stresses above 700 kPa. At low normal

stresses and displacements the behaviour of the chalk–steel interface was captured by an alpha type approach related

to the rock unconfined compressive strength, which has been developed for other higher strength rock types.

Notationb fitting constantc fitting constantD50 mean soil particle sized linear displacementGs specific gravitymsat saturated moisture contentn porosityR radial positionRa average centre-line roughnessr rock sample radiusT torqueT0 tensile strengthα adhesion factorδpeak peak interface friction angleδult ultimate interface friction angleθ rotational displacementμ coefficient of frictionρd dry densityσv normal stressτ shear stressϕb basic friction angle

1. IntroductionTo allow the design of gravity base foundation systems (GBS)for tidal stream generators it is necessary to have appropriateinterface friction parameters for the foundation–seabed inter-face due to the potential for relatively high lateral loads experi-enced by these applications. As there are high-velocity currentsat locations with high tidal stream energy potential, the seabedsediment is likely to be washed out (scoured) resulting in expo-sed bedrock at the seabed, which may form one half of thefoundation surface (Small et al., 2014). The determination ofinterface friction parameters requires laboratory interfacetesting using the foundation–seabed materials (typically steelor concrete in contact with different rock types) with testingundertaken at appropriate normal stress levels. Currently thereis little guidance available to designers on how such parametersshould be selected or the appropriate approaches to laboratorytesting (Small et al., 2014). What guidance is available is con-sidered conservative based upon simple assumptions of inter-face behaviour rather than actual laboratory element testing.For example, Fraenkel (2002) estimated that at least 1500 tof GBS per MW are needed in order to maintain stability,but this value seems very high and conservative where there is

1

Geotechnical Engineering

Chalk–steel interface testing for marineenergy foundationsZiogos, Brown, Ivanovic and Morgan

Proceedings of the Institution of Civil Engineers

http://dx.doi.org/10.1680/jgeen.16.00112Paper 1600112Received 01/07/2016 Accepted 04/10/2016Keywords: geotechnical engineering/renewable energy/shallow foundations

ICE Publishing: All rights reserved

Downloaded by [ UNIVERSITY OF DUNDEE] on [11/11/16]. Copyright © ICE Publishing, all rights reserved.

mailto:[email protected]

significant drive to bring down the foundation costs for marinerenewable devices (Carbon Trust, 2011). NAVFAC (1986)suggest a coefficient of friction, μ, of 0·7 (interface frictionangle, δ=35°) for mass concrete on clean, sound rock but theorigins of this value are unclear and it is not stated if thisrefers to a bonded or unbonded surface, or the types of rockthat were used in its determination.

Existing guidance on rock–concrete and rock–steel interfacestypically relates to very different applications (and interfaceconditions) such as concrete bonded or cast against rock athigh vertical stress levels. Examples of these are found at theinterface between the base of a dam or cast in situ pile rocksockets (Horvath, 1978; Rosenberg and Journeaux, 1976;Williams and Pells, 1981) and rock–steel interfaces such asrock bolts (Li and Håkansson, 1999) or H-steel piles driveninto rock (Yu et al., 2013). These rock–steel interface examplesresult in constant normal stiffness (CNS) conditions, whichlead to high normal stresses where the interface is subject toshear and constraint of dilation. This can result in normalstresses at the interface that are orders of magnitude higherthan those that would be experienced for a tidal stream genera-tor foundation (estimated at some hundreds of kPa, under con-stant normal stress conditions). This fact along with thepotential for concrete bonding between interfaces may limitthe relevance of previous interface studies.

Many systematic efforts to investigate the interface shear be-haviour between soil and geotechnical structures (rather thanrock) have been made. Potyondy (1961) conducted testsbetween various soil types (e.g. sand, clay and silt) and con-crete, steel and wood, while Peterson et al. (1976) conductedtests with sand–steel interfaces. Both studies were carried outusing the direct shearbox. In the 1980s significant research wasundertaken on sand–steel interfaces by Uesugi and Kishida(1986a, 1986b) and Kishida and Uesugi (1987) utilising thesimple shear apparatus. Since these earlier studies research hasbeen continuously undertaken using various devices and lab-oratory equipment such as the ring shear and curved shearbox(Iscimen and Frost, 2010). Typically it has been found that inter-face frictional resistance increases as the solid (or the surfacerepresenting the structural element) surface roughness increasesup to a specific limit defined as critical roughness. At thislimit, which approaches the internal friction angle of the sand,a sand–sand slip occurs as a soil–soil shear rather than inter-face shear predominates. This is because as the surface rough-ness tends to the diameter of individual sand particles, thesand effectively interlocks with the foundation surface forcingthe shearing away from the interface and into the soil mass(Kishida and Uesugi, 1987; Peterson et al., 1976; Uesugi andKishida, 1986a). It has also been shown that as the roughnessincreases, more displacement is required to mobilise the peakfriction angle (Iscimen and Frost, 2010). Where the soil par-ticle diameter exceeds the surface roughness, the constant vol-ume interface friction angle decreases significantly with

increasing D50, with a cut-off point of Ra/D50 = 0·015 for sand(Jardine et al., 1993). For sand–steel interface used in offshoredriven piles tan δcv is often limited to 0·55 (28·8°) (Lehaneet al., 2005). Although there have been significant previousstudies for soil sheared against common civil engineeringmaterials, there is a dearth of information for rock–steel inter-face testing under the conditions that may be encountered intidal stream or other renewable energy foundation applications,namely, non-bonded connections to the seabed under relativelylow normal stresses and variable stiffness conditions. It is ack-nowledged, however, that tidal stream generator foundationswill also be subjected to cyclic loading which will need to beconsidered when interpreting the results of the monotonic testspresented in this paper.

This paper presents results of chalk–steel interface testing uti-lising a specially commissioned torsional interface shear tester(IST). This device was used to investigate the interface materialproperties that influence interface shear strength under stressand displacement levels appropriate to tidal stream GBSfoundations. The potential for the use of the equipment inlarge-displacement events has also been explored, which maybe relevant to driven pile installations that can potentially beused to anchor or support tidal stream generator alternatives.This paper focuses on applications on chalk because it is arock with extraordinary characteristics (Lord et al., 2002) andareas in the south of the UK that have been identified as ofsignificant tidal stream potential may consist of chalk seabeds(e.g. Race of Aldernay and Casquets, which represent around6% of the total UK tidal resource (Carbon Trust, 2005)), orwhere driven pile solutions may encounter such material types.

The aim of this paper is to provide information regardingthe controlling parameters on the foundation–seabed interfacesliding resistance and to provide interface properties that havethe potential to be used during the design process.

2. Laboratory testing

2.1 Description of chalk samples used forlaboratory testing

The samples were collected from the active Imerys MineralLimited’s Quarry, Westwood, Beverley, HU17 8RQ, UK(501740, 438256). Blocks of chalk typically 350 by 300 by280 mm were obtained directly after quarrying and prior tocrushing for use in the chemical industry. Unfortunately, owingto the working status of the quarry and the required health andsafety regulations, the research team was not directly involvedwith the sampling of the chalk, thus making it difficult tocomment on the structural setting of the chalk in situ. The chalkis White Chalk from the Flamborough Chalk Formation(Upper Chalk unit, northern province English Chalk) referred toinformally as the Flamborough Sponge Bed (Lord et al., 2002;Whitham, 1991, 1993). This source of material was selectedbecause of the fresh nature of the chalk (i.e. recently quarried),

2

Geotechnical Engineering Chalk–steel interface testing for marineenergy foundationsZiogos, Brown, Ivanovic and Morgan


immediately placed under cover and the fact that the chalk wasfree from flints that may interfere with characterisation andinterface testing.

The chalk was characterised using both field and laboratorytechniques prior to interface shear testing and the resultsare summarised in Table 1. These results classify the chalk asof very high density according to CIRIA 574 (Lord et al.,2002). The chalk on return to the laboratory had a very lowmoisture content of 0·3%. Dry density and saturation moisturecontent determination (BS 1377-2: 1990 (BSI, 1990)) high-lighted that the moisture content of the chalk was belowthe 90% minimum level of saturation recommended for fieldidentification procedures so the chalk was saturated prior tothese tests by applying 95 kPa vacuum to submerged samples(in de-aired and de-ionised water) for 24 h. In order to ascer-tain the level of saturation, the moisture content after followingthe aforementioned process was compared to the saturationmoisture content ws = 11·49% of the chalk (calculated accord-ing to BS 1377-2:1990 (BSI, 1990)) and very high levels of sat-uration were achieved (99·6–99·7%). Samples were also ovendried in order to investigate the effect of moisture content onthe unconfined compressive strength (UCS), tensile strengthand interface shear resistance behaviour.

2.2 Scope of testingInterface testing between chalk–steel interfaces at normalstresses relevant to those anticipated at real tidal stream pro-jects (Ziogos et al., 2015b) was carried out in order to obtainthe friction properties necessary for the determination of thesliding resistance of a GBS. The UCS of chalk has previouslybeen found to vary significantly with saturation levels, gener-ally showing lower strengths for saturated samples comparedto dry ones (Matthews and Clayton, 1993). Therefore testsusing both dry and saturated samples were carried out in orderto examine the variation of UCS on the shear resistance. In

addition, the effect of steel roughness was investigated(Ra = 0·4–34 μm, Table 2) along with the effect of normalstress (σv = 16–1000 kPa) over relatively short displacements of10 mm during shear.

Sandstone–steel interface testing was also undertaken (σv =16–316 kPa) to allow the comparison between the interfacebehaviour of chalk and a typical sedimentary rock (sandstone)that exhibits more ‘conventional’ behaviour. The samples usedfor the testing were sourced from the Caithness area and speci-fically from a disused quarry located south of John O’Groats,Scotland, UK (ND37150 70138). This area was selectedbecause it is adjacent to the Pentland Firth, which exhibits sig-nificant tidal resource (Carbon Trust, 2011). The Old RedSandstone that was recovered for testing is yellow-orange incolour and medium grained (Johnstone and Mykura, 1989). Inaddition the laboratory determination of UCS revealed a verysimilar value (Table 1) compared to that of the dry chalksamples, which allows comparison of the shear resistance ofthe two rock types focusing on parameters other than the com-pressive strength (e.g. the grain size that differs significantlybetween sandstone and chalk).

In addition to the tests designed to investigate the behaviour ofchalk relevant to tidal stream generator GBS foundations (lownormal stress and low displacement), an extra set of tests wasundertaken to very large cumulative displacements (7·0 m).These tests were to check the potential of the IST device butalso may be considered relevant to the displacements that maybe encountered in driven piling or underneath a sliding foun-dation or tow head used in the offshore oil and gas industries.

2.3 Determination of unconfinedcompressive strength

The tensile strength T0 of the rock samples was determinedusing the Brazilian tensile test (ASTM D3967-08 (ASTM,2008)). Initially the tensile strength was used as an indirectmethod to determine the UCS of the chalk using conversionfactors. The results indicated that the empirical conversionfactors used to convert the tensile strength to UCS, as foundfor other sedimentary rocks (e.g. Altindag and Guney, 2010),were not appropriate for chalk; therefore direct unconfined

Property Chalk Sandstone

Dry density, ρd: Mg/m3 2·06 2·23Porosity, n: % 23 13Voids ratio, e 0·3 0·15Saturated moisture content, msat: % 11·4 6Specific gravity, Gs 2·7 2·57UCS, dry samples: MPa 30·00 31·50UCS, saturated samples: MPa 9·30 —

Tensile strength, T0, dry samples: MPa 1·10 2·60Tensile strength, T0, saturated samples: MPa 0·96 —

Young’s modulus, dry samples: GPa 8·60 10·20Young’s modulus, saturated samples: GPa 2·85 —

Table 1. Summary of key index properties for the chalk and

sandstone samples

Material Average roughness, Ra: μm

Chalk (saw cut) 3·1Sandstone (saw cut) 19·0Polished steel 0·4Machined steel 1 7·2Machined steel 2 34·0

Table 2. Summary of the interface properties of the

materials tested

3



compression tests were carried out in accordance with ISRM(2007) utilising a 250 kN Instron universal testing machine.Direct UCS tests were carried out on cylindrical oven-driedand saturated chalk samples with a height to diameter ratioequal to 2 (see Table 1) at a strain rate of 0·1 mm/min. Drysandstone samples were also tested for comparison purposes.Both the oven-dried sandstone and chalk samples exhibitedvery similar values of UCS (31·5 and 30·0 MPa, respectively),whereas a 69% decrease in UCS was observed for the chalkwhen tested in a saturated condition. Deterioration of chalkUCS after saturation has been reported before by Matthewsand Clayton (1993); however, they noticed a smaller averagereduction of 50%. This difference may be due to the highdensity of the chalk tested in this study, whereas Matthews andClayton (1993) report UCS for a variety of chalk densities.

2.4 Tilt table testingPrior to the main interface testing the basic friction angle ofthe chalk (ϕb = 30·5°) and sandstone was determined usingsimple tilt table testing in line with the methodology outlinedin USBR 6258 (USBR, 2009). This involves tilt table testingof two 54 mm dia. samples of 27 mm thickness placed ontop of each other. The samples were prepared by coring of ablock of chalk (54 mm nominal diameter) and then dry cross-cutting of the core using a diamond saw. The interface fric-tional resistance was determined on this saw-cut surface (asper USBR 6258 (USBR, 2009)). The ϕb values were 30·5° and38·5° for chalk and sandstone, respectively. Previous experienceof the results from the low normal stress tilt table tests showgood correlation with the more advanced testing techniques atelevated stress levels (Table 3). Therefore, apart from using thetilt table test to determine the basic friction angle, the simpletest was also used to test the chalk samples against the steelinterfaces used in the main study.

2.5 Description of the IST deviceA computer-controlled torsional IST (supplied by GDS) wasutilised for the execution of the main part of the interfaceshear testing programme (Figure 1). This device consists of anaxial actuator at the top of the rig, which can apply up to5 kN of vertical load, and a rotational actuation system at thebase, capable of applying torque up to 200 Nm. Below theaxial actuator is a combined load/torque cell arrangement withcapacities of 5 kN and 200 Nm, respectively. The axial actua-tor applies the normal load to the samples under test and isfixed against rotation, whereas the rotational actuator appliesthe torque/rotation from below (tests can either be torque orrotation controlled).

A special clamping system was developed to allow rectangularinterchangeable foundation interface elements of 65� 90 mmwith a thickness of 8 mm to be clamped at the base of the rigabove the rotational actuator (Figure 1(b)). Similarly, below theload/torque load cell a clamping device was developed to clampshort, round rock samples (54 mm dia. and 25 mm high).During the test, the upper rock sample was fixed while the lowersteel sample rotated at a predetermined rate (rotation controlledtest). The IST used here is an evolution of that previously usedby Kuo et al. (2015) for the low-stress interface testing of pipe-lines (referred to as the ‘Camtor’ device). This previous deviceincorporated an outer pressure cell and allowed the testing ofsoil samples up to 70 mm dia. and 20 mm thick against pipelinesurface elements. During the tests torque and normal load weremeasured using a calibrated torque/load cell and vertical androtational deformation measurements were provided by a cali-brated encoder attached to the stepper motor.

The tests were conducted under constant normal stress con-ditions on both dry and saturated samples under six differentnormal stress levels of 16, 79, 159, 316, 700 and 1000 kPa. Inorder to conduct testing of submerged interfaces for saturatedsamples, a custom-made poly(methyl methacrylate) (PMMA)box (bath) was attached to the lower rotation platen,Figure 1(c). The saturated rock sample was clamped using thetop holder and the bath was filled with de-ionised water, inorder to prevent drying of the saturated sample. For saturatedtests, the interface was kept under constant normal stress for15 min before the initiation of shearing in order to allow anyexcess water pressure dissipation at the interface; for the samereason the shearing rate was kept at a low level at 0·05 mm/sof equivalent horizontal displacement. Every test lasted approxi-mately 36 min and was terminated when an equivalent hori-zontal displacement of 10 mm was reached (corresponding to42·5° rotational displacement). In addition to the aforemen-tioned tests, three tests up to a horizontal displacement of7·0 m (29 750° rotational displacement) were carried out underthree different stress levels (100, 300 and 700 kPa) using satu-rated samples. The shearing rate for these tests was 1·25 mm/sand each test lasted 16·5 h. A higher rate was used for thesetests in order to allow the execution of each test within a

Interface

σv: kPa Chalk–steelRa= 0·4 μm

Chalk–steelRa= 7·2 μm

Chalk–steelRa= 34 μm

Measured peak interface frictionangle, δpeak: degrees

Tilt table �0·6 30·0 36·5 37·5IST 16 36·0 39·5 40·5IST 79 39·5 41·5 42·5IST 159 37·0 40·0 39·5IST 316 33·5 37·0 40·0IST 700 32·5 31·5 38·5IST 1000 31·0 31·5 35·0

Table 3. Comparison of results of chalk–steel interface testing

utilising the tilt table and the IST device

4



reasonable time frame. The scope of these tests was to inves-tigate any possible degradation on the chalk surface (and con-sequently to the shear strength of the interface) at very highshear deformations that can potentially occur for applicationssuch as pile driving. Therefore, the steel plate with Ra = 7·2 μmwas selected, as the roughness lies in the middle of the rough-ness range of steel piles used in practice (Ra = 5–10 μm forsteel piles (Barmpopoulos et al., 2010)).

The torque measured during IST testing was converted toaverage shear stress as per Equation 1 after considering Saadaand Townsend (1981) for ring shear testing.

1: τ ¼ TÐ r0 2πR

2dR¼ 3

2πr3T

Rock sample clamp

(a) (b)

(c)

Steel foundationinterface sample

Figure 1. (a) Interface shear tester apparatus. (b) Detailed view of

the IST sample mounting arrangement. (c) Detailed view of

PMMA bath used for saturated testing

5



The radial deformation was converted to a linear displace-ment at a reference point considered at a distance equal to halfof the radial length of the circular rock sample, as perEquation 2.

2: d ¼ θrπ360

where θ is rotational displacement, τ is shear stress, d is lineardisplacement, r is the rock sample radius, R is radial positionand T is torque.

2.6 Description of steel samples (foundationanalogues)

Mild steel was used to prepare rectangular (65� 95� 8 mm)plates that represented foundation analogues for the inter-face testing. As discussed in the introduction and found in theliterature (Ziogos et al., 2015a, 2015b), roughness has a majoreffect on the interface behaviour, therefore different prep-aration techniques (polishing and machining) were applied andresulted in three different foundation analogues with a widerange of surface roughness (Ra between 0·4 and 34 μm).

Polishing with a surface grinder using a BAA60 – K7V wheelresulted in surface roughness average Ra = 0·4 μm. Machining,using a shaping machine and an appropriately adjusted shapingtool, resulted in Ra values of 7·2 and 34 μm. The rangeof roughness obtained covers the roughness of some of thesteel elements commonly found in geotechnical applications(for example, Ra = 5–10 μm for steel piles, Barmpopouloset al., 2010), allowing the utilisation of the results in otherapplications.

2.7 Surface roughness characterisationInterface roughness can be quantified by many parameters,but one of the most frequently used is Ra (centre-line averageroughness), which is the computed average of all deviations ofthe roughness profile from the median (centre) line over adefined profile length and is defined in Equation 3 (Degarmoet al., 2003). A hand-held Taylor Hobson Surtronic Duo styluscontact profilometer was used to determine the average centre-line roughness (Ra) of all of the samples used for interfacetesting (rock and steel). The profilometer has a range ofmeasurement up to 40 μm and a traverse length of 5 mm. Foreach sample, five Ra measurements were taken and the meanvalue was selected. Calibration of the device was carriedout periodically against a standard roughness profile withRa = 5·81 μm and Rz = 21·5 μm (supplied with the device). Theinterface properties of all the materials used for testing (rockand steel samples) are summarised in Table 2.

3: Ra ¼ 1=LðL0

z xð Þj jdx

where L is profile length and z(x) is deviation from the centre-line at point x.

3. ResultsFigures 2(a)–2(c) show the normalised shear stress–displacement curves from saturated chalk–steel interfacetests on steel of increasing roughness. A typical result showsa slightly elevated initial shear stress followed by a slightreduction in shear stress post peak (or yield) and then remain-ing relatively constant until the end of the test. It is apparentthat yielding or peak shear stress is observed at increasingdisplacement levels as the normal stress on the chalk increases,as seen in Figure 2(a) for normal stresses above 159 kPa. Itis also noticeable that, as the normal stress increases, thereis an increase in shear resistance up to a normal stressof 79–159 kPa and then a reduction in the shear resistance,with the lowest shear resistances associated with the highestnormal stress of 1000 kPa. Table 4 shows the summarisedresults of testing where δpeak (Table 3) is defined as the maxi-mum value at a shear displacement up to 4 mm and δult isdefined as the minimum value in the region of 8–10 mm. Theresults in Figure 2(d) show the tests for sandstone shearedagainst steel of Ra = 7·2 μm where the curves are more ‘noisy’because sandstone is rougher and harder, whereas it is appar-ent that higher interface shear stresses are mobilised in thesofter chalk.

Figures 3–5 show the results of IST testing of the saturatedand dry chalk against the various steel interfaces in terms ofthe different normal stresses. For the dry tests the peak inter-face friction angles δpeak range from 35° to 45° and δult from22·5° to 40°. For saturated tests, δpeak ranges from 27° to 42°and δult from 22° to 40°. All of the results from IST testingare summarised in Table 4. These results suggest that theshear stress–displacement response at the interface is not pro-portional to the normal stress. It is possible that a simple con-stant friction angle based on Mohr–Coulomb failure criterionmodel for interface shearing may not be appropriate for chalk.The peak interface friction angles noted do not seem to be sig-nificantly affected by the degree of saturation, although thevalues are higher for the dry tests. They do not seem to reflectthe variation of UCS change noted between dry and saturatedchalk samples (69% reduction in UCS from dry to saturated).

All of the figures have been annotated with the value of theinterface friction angle derived from the low-stress tilt tabletesting (Table 3) and the basic chalk–chalk friction angle(again obtained from tilt table testing as described earlier).The derived interface friction angles from IST testing tend toexceed these values for the lowest and highest steel roughness(0·4 μm and 34 μm), whereas the results for the dry and satu-rated tests against 7·2 μm steel only appear to exceed this atthe highest interface angles. This suggests that, in the case ofchalk, the normal stress level affects the results obtained fromthe IST when compared to tilt table testing, although this is

6



not the case for other higher strength rocks. This variation inbehaviour suggests a transition in moving from 0·4 μm to7·2 μm steel interface, which is reflected in both the IST andtilt table testing.

In the IST, irrespective of the steel type, the interface resistanceexhibits a low value at normal stress of 16 kPa, which mayindicate poor interlocking between the chalk and steel interfaceas the applied stress is not adequate to bring the two solidbodies into intimate contact and shearing is occurring on thetop of the asperities (steel and chalk). As the normal stressincreases, the interface gains higher shear strength, as better

interlocking is established between the normal stresses of 79and 159 kPa (7·2 to 16·6% of the chalk tensile strength, T0).At these stress levels the shearing is accompanied by observa-ble damage on the chalk surface, seen as a layer of powder(dry samples) or chalk putty (saturated tests) on the steel inter-face on post-test sample separation. This behaviour is similarto that noted for rock analogues (cement blocks) by Ziogoset al. (2015b). For normal stresses from 316 kPa to 1000 kPa(i.e. 0·31 to 1·0T0) the ultimate interface friction angle reducesto values typically between 30 and 35° and in the majority ofcases appears to be approaching the basic chalk–chalk frictionangle value noted for the 0·4 μm interface. This suggests that

00

0·2

0·4

0·6

0·8

1·0

1·2

2 4 6

16 kPa79 kPa159 kPa316 kPa700 kPa1000 kPa

(a)Horizontal displacement: mm

Nor

mal

ised

she

ar s

tres

s, τ

/σv

8 10 12 00

0·2

0·4

0·6

0·8

1·0

1·2

2 4 6


(b)Horizontal displacement: mm

Nor

mal

ised

she

ar s

tres

s, τ

/σv

8 10 12

00

0·2

0·4

0·6

0·8

1·0

1·2

2 4 6


(c)Horizontal displacement: mm

Nor

mal

ised

she

ar s

tres

s, τ

/σv

8 10 12 0 2 4 6 8 10 120

0·2

0·4

0·6

0·8

0·7

0·5

0·3

0·1

0·9

1·0

1·2

1·1

159 kPa79 kPa16 kPa

316 kPa

(d)Horizontal displacement: mm

Nor

mal

ised

she

ar s

tres

s, τ

/σv

Figure 2. Normalised shear stress plotted against horizontal

displacement for saturated chalk samples against: (a) steel

Ra = 0·4 μm; (b) Ra = 7·2 μm; (c) Ra = 34 μm; and (d) dry sandstone

samples against steel Ra = 7·2 μm

7



damage at the interface may be filling the rough surface of therougher steel samples and reducing their apparent interfaceroughness to that approaching the smoothest interface testedhere. Significantly more damage was noticed in some samplestested at 700 kPa and 1000 kPa, where parts of the perimeter

of the sample were chipped off (labelled as surface damage,SD in Table 4), or at the highest normal stress level (1000 kPa)resulted in complete tensile failure of the sample (NI) asshown in Figure 6. Therefore in the case of chalk the upperlimit to the interface strength appears to be linked to the local

Normalstress: kPa

Initialsamplestatea

Post-testsampleconditionb

Ra: μm

0·4 7·2 34·0

Peak shearstress: kPa

Ultimate shearstress: kPa





16 D I 13·9 7·5 13·5 11·0 12·62 11·516 S I 11·5 10·5 13·0 10·5 13·5 12·079 D I 75·5 57·5 76·9 60·5 72·5 70·579 S I 65·0 59·0 70·2 58·0 72·0 69·5159 D I 136·0 106·0 147·5 131·0 143·0 141·0159 S I 119·0 112·0 132·5 125·5 131·0 127·0316 D I 287·0 273·5 235·6 215·5 318·0 316·0316 S I 207·5 188·0 237·2 215·0 266·0 263·5700 D SD 500·0 433·5 476·3 410·6 614·0 606·5700 S SD 450·0 399·0 432·5 400·0 558·0 546·01000 D NI 748·0 739·5 650·3 600·5 803·5 643·01000 S NI 600·0 560·5 608·5 582·0 705·0 694·0

aS, saturated; D, dry samplesbI, intact; SD, surface damage; NI, non-intact samples

Table 4. Summary of results from interface testing of chalk–steel

interface utilising the IST device

00

15

20

25

30

35

40

45

200 400 600

SD

NI

(a)Normal stress, σv: kPa

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

δ peak

δ tilt = 30·0°

φb = 30·5°

δ ult

µpeak

µult

00

15

20

25

30

35

40

45

200 400 600

SDNI

(b)Normal stress, σv: kPa

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

δ peak

δ tilt = 30·0°

φb = 30·5°

δ ult

µpeak

µult

Figure 3. Variation of interface friction angle and coefficient of

friction for chalk–steel interface test for steel with Ra= 0·4 μm

against (a) dry and (b) saturated chalk

8



surface strength of the material (similar to the grain size ofsand exceeding the roughness of steel as discussed earlier) withpotential for catastrophic disruption of the interface at highernormal stresses on approaching the tensile strength of thechalk (T0 = 0·96 to 1·1 MPa). Although some of the samplestested at the higher stress of 1000 kPa were not intact afterremoval of the sample from its clamp, it is believed that theinterface shearing behaviour is valid as the clamping systemmaintained the integrity of the sample and shearing surfaceduring testing. The reduced shear stress noted during testing at

these stresses reflects the increased interface damage. Theroughness of the steel interfaces tested was measured beforeand after testing for the low-displacement tests and no signifi-cant variation in roughness was noted for either rock type. Theroughness of the rock samples was also measured and thisremained within the variability of the average values typicallymeasured. A similar procedure was undertaken for the large-displacement tests, which showed that the steel tested againstchalk exhibited no significant change in roughness after carefulremoval of the chalk residue. Testing of the steel with the

010

15

20

25

30

35

40

45

200 400 600

(a)

Normal stress, σv: kPa

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

φb = 30·5°

δ tilt = 36·5°

SD NI

δ peak

δ ult

µpeak

µult

010

15

20

25

30

35

40

45

200 400 600

(b)

Normal stress, σv: kPaIn

terf

ace

fric

tion

angl

e, δ

: deg

rees

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

φb = 30·5°

φb = 36·5°

δ peak

δ ult

µpeak

µult

SD NI


friction for chalk–steel interface test for steel with Ra= 7·2 μm


010

15

20

25

30

35

40

45

50

200 400 600

(a)Normal stress, σv: kPa

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

1·2

φb = 37.5° φb = 30·5°

δ tilt = 36·5°

δ tilt = 37·5°

SDNI SD

NI

δ peak

δ ult

µpeak

µult

δ peak

δ ult

µpeak

µult

010

15

20

25

30

35

40

45

50

200 400 600

(b)Normal stress, σv: kPa

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

Coe

ffic

ient

of

fric

tion,

µ

800 1000 12000

0·2

0·4

0·6

0·8

1·0

1·2


friction for chalk–steel interface test for steel with Ra= 34 μm


9



chalk residue in place was not undertaken, as the surface wasrelatively uneven and non-continuous over the steel surfacemaking roughness determination difficult. Visual degradationof the steel surface of the large-displacement tests against sand-stone was noticed, with post-test roughness measured as5·5 μm as opposed to pre-test roughness of 7·2 μm. The sand-stone itself did not show any significant change in roughness,but black residue was noticed on its surface which wasassumed to come from the steel.

As mentioned earlier, the adoption of a linear failure envelopefor chalk does not seem appropriate for design purposes, sincethe interface friction angle is affected by the normal stresslevel. Although, in order to allow comparison, linear failureenvelopes for peak and ultimate interface resistance were calcu-lated. These are based upon the average peak or ultimate resist-ance determined over the range of effective stresses tested. Toallow the effect of normal stress and potential for surfacedegradation and damage to be represented, the range of nor-malised friction angles obtained are denoted by vertical errorbars, as shown in Figure 7 (shown for chalk peak values onlyfor clarity). The results suggest that, over the steel roughnessrange investigated (Ra = 0·4–34 μm), on average the interfacebecomes stronger as the steel roughness increases, withoutreaching a ‘plateau’ as seen in other studies (Barmpopouloset al., 2010; Jardine et al., 1993; Ziogos et al., 2015b) and asshown for sandstone. However, when increasing normal stressis considered there appears to be a tendency for the chalk–steelinterfaces to tend towards the basic chalk–chalk interfaceproperties.

3.1 Large-displacement interface testingResults from the large-displacement tests can be seen inFigure 8 compared with the result from a similar test under-taken on sandstone for comparison. It is clear from the tests

on chalk at all normal stress levels that there is continuousdegradation of the shear surface throughout the test. This issimilar to the behaviour observed by Barmpopoulos et al.(2010) testing sand against concrete to large displacements,which was attributed to observed sand particle crushing andgeneration of fines. At the lowest stress of 100 kPa the frictionangle has fallen from an initial peak of 43° to 37° at 7·0 mwith the rate of degradation appearing to reduce. This value issimilar to that noted from the low-stress tilt table testing ofchalk–steel (Table 3). At 300 kPa normal stress, reduced initialdegradation is observed with a relatively constant interface fric-tion angle of 30–31° being reached after 2·4 m of displace-ment, which tends to the basic chalk–chalk interface frictionangle and which may be explained by the degradation behav-iour described above for the low-displacement tests. In con-trast, at a normal stress of 700 kPa there is a significantreduction in the interface friction angle below the basic chalk–chalk interface friction angle until reaching a relatively con-stant value of 17° (0·56ϕb) at 6·5 m. Previous low-displacementtesting results shown earlier may have led to the recommen-dation of a lower safe bound design interface friction angle ofapproximately 29° (0·95ϕb) which could be determined fromthe basic chalk–chalk interface friction angle. In the case oflarge-displacement events this may be a suitable approachwhere the normal stresses do not exceed 300 kPa (0·31T0).Where this value is exceeded then a more conservative inter-face resistance must be assumed. The behaviour observed inFigure 8 for the large-displacement test on sandstone showsvery different behaviour, with increasing resistance withincreasing displacement up to 1·7 m displacement and then amore gradual increase with increasing displacement which

Figure 6. Tensile failure of a dry chalk sample sheared at

1000 kPa0

0·5

0·6

0·7

0·8

0·9

1·0

1·1

1·2

1·3

1·4

1·5

1·6

1·7

1·8

5 10 15 20

Average steel roughness, Ra: µm

Nor

mal

ised

inte

rfac

e fr

ictio

n an

gle,

δ/φ

b

Sandstone φb= 38·5°

Chalk φb= 30·5°

Old Red Sanstone, δpeak/φb

Old Red Sanstone, δult/φbChalk, δpeak/φbδpeak = (0·0059Ra +1·00)φb

Chalk, δult/φbδult = (0·0067Ra + 0·95)φb

25 30 35 40

Figure 7. Variation of normalised friction angle with increasing

steel roughness, for saturated chalk samples and dry Old Red

Sandstone

10



again appears to be tending to the sandstone–sandstone basicinterface friction angle. This may be due to the removal ofweak exposed sandstone asperities (individual weakly cemen-ted grains), but this behaviour requires further investigation.What is apparent from the testing is that, unless normal stres-ses are high enough to cause significant interface and sampledamage, large-deformation tests on steel (Ra = 7·2 μm)–rockinterfaces result in interface behaviour that tends to the basiclow-stress rock–rock interface behaviour (for both chalk andsandstone–steel interfaces).

It should be noted that the testing regime here is constantnormal stress (similar to that adopted by Barmpopoulos et al.(2010) for ring shear tests on sand–steel and sand–concreteinterfaces) whereas in the case of driven piles a constant normalstiffness regime may more adequately represent in situ con-ditions leading to a reduced potential for tensile strengthlinked degradation. This assumes, however, that the chalkin situ is intact and well confined without faults or voids/low-strength zones. In addition, constant normal stiffnessconditions may lead to significantly higher in situ stresses thanthose tested here, which could result in a more rapid degra-dation with displacement.

4. Implications for testing and design

4.1 Utilisation of tilt table for simple interfacecharacterisation

Table 3 compares the results of interface testing using the tilttable to those obtained from IST testing. It can be seen thatthere is good agreement between the peak interface frictionangle measured in both types of test, especially for the roughersteel plates (Ra = 7·2 and 34 μm) and for σv up to 700 kPa(i.e. 0·7T0). Therefore, it is possible to utilise the tilt table test to

estimate the peak friction angle for preliminary design purposes,and to use IST (or large-displacement direct shear box tests) indetailed design. Based on the IST tests here for chalk (Figures3–5), and the apparent tendency to degrade to low frictionangles with increasing normal stress levels and interface degra-dation, tilt tables tests for chalk against a relatively smoothinterface may give a useful lower bound for design. For both thelow and high displacements it would seem that the normallower-bound interface friction angle should be taken as thebasic chalk–chalk friction angle from tilt table testing, irrespec-tive of the foundation material or roughness. For higher displa-cement tests, however, some caution has to be exercised whennormal stress levels exceed 0·31T0. Similarly, in the prototypedeployment of tidal stream generator, foundations are likely toexperience cyclic loading that has the potential to cause degra-dation at lower stress levels and lower displacements.

4.2 Potential design approachesPreviously, Ziogos et al. (2015a, 2015b) have proposed anadhesion factor (α) (shear stress normalised by UCS) typeapproach for monotonic rock–steel and cement–steel interfacestrength prediction similar to that developed for rock socketpile adhesion factors (Tomlinson, 2001). In this case, however,the magnitude of shear stress is lower by several orders of mag-nitude compared to pile applications, due to the unbondednature of the interface and the CNS conditions for rocksocketed piles. UCS was also normalised by the vertical stressduring interface testing, as this was seen to have a significanteffect over the relatively low stresses likely to be encountered atthe rock–steel interface (as also observed in this study). It isassumed that such a normalisation is not applied for rocksocket piles due to the high confining stresses and difficulty indetermining the actual in situ stress at the bonded interface.Figure 9 shows lines that represent the adhesion factor valuesobtained from previous monotonic interface testing of variousrock–steel interface combinations. The lines are described byEquation 4 and allow the calculation of the maximum shearstress capacity of the interface, for a given rock type (UCS),foundation footing (steel Ra) and the anticipated appliedaverage normal stress (σv).

4: α ¼ bUCSσν

� �c

The lines previously determined for various steel roughnesslevels (Ra from 0·4 to 7·2 μm and UCS from 45 to 157·2 MPa)have been extrapolated to cover the range of testing undertakenin this study. For steel Ra = 0·4 μm, fitting constants b and cmay be taken as 1·78 and −1·26, respectively, whereas forRa = 7·2 μm, b=1·25 and c=−1·16. Specific data points arealso shown for the IST testing of Old Red Sandstone by wayof comparison. It can be seen that the previously determined

00

10

20

100 kPa chalk300 kPa chalk700 kPa chalk700 kPa Old Red Sandstone

30

40

50

1000 2000 3000

Horizontal displacement: mm

Inte

rfac

e fr

ictio

n an

gle,

δ: d

egre

es

4000 5000 6000 7000

φb,sandstone = 38·5°

φb,chalk = 30·5°

Figure 8. Interface friction angle plotted against horizontal

displacement for saturated chalk samples

11



relationship for rocks of much higher UCS seems applicablefor the much lower strength chalk and offers an alternativedesign approach to an interface friction angle based approach,where low monotonic displacements occur (which could beattempted from the results in Table 4 or the lower-bound basicfriction angle). Additionally, the results are shown for thelarge-displacement tests on chalk (Figure 8), which suggestthat the approach based upon Equation 4 should be used withcaution for chalk where large displacements and/or cyclic load-ing may occur at an interface during installation or service.Adhesion factor ranges from 0·001 (σv = 16 kPa) to 0·07(σv = 1000 kPa) for saturated chalk samples are similar to thecohesion intercept/UCS ratio (ranges from 0·02 to 0·13)defined by Clayton and Saffari-Shooshtari (1990) from inter-face tests on bonded planar chalk–concrete interfaces. Thissuggests that the strength of the material is potentially control-ling behaviour irrespective of whether or not the concrete isbonded or unbonded.

5. Summary and recommendationsInterface shear testing between a very high-density chalk andsteel of varying roughness was undertaken to gain insights fortidal stream generator GBS foundation design. In addition,interface testing using Old Red Sandstone samples was under-taken for comparison purposes.

Low-displacement interface shear testing of chalk–steel inter-faces at low normal stresses showed a tendency for increasingshear resistance up 159 kPa (i.e. 0·16T0). At stress levels abovethis and in particular up to 316 kPa (0·33T0), the interfaceshear resistance begins to degrade and tends to the basicchalk–chalk interface behaviour (with increasing normal stress)determined at low stress from tilt table testing. This suggestsdamage at the chalk–steel interface, which was observed as

chalk dust or putty on the steel interfaces after testing. As thenormal stress was increased further the chalk displayed surfacedamage (σv = 700 kPa, 0·76T0) and fracturing (σv = 1000 kPa,1·10T0). The average chalk–steel interface shear strengthappears to increase linearly with increasing roughness of thesteel, which is in contrast to the results of sandstone–steelinterfaces, which reach a ‘plateau’ at an average steel roughnessof 7·2 μm, but this behaviour is highly dependent on thenormal stress levels.

Large-displacement monotonic tests of saturated chalksamples (up to 7·0 m) revealed a degradation of interfacestrength for increasing displacement with a tendency for resist-ance towards the basic chalk–chalk interface behaviourmeasured from tilt table testing (ϕb = 30·5°) at normal stresslevels up to 300 kPa (0·33T0). At normal stress levels abovethis (σv = 700 kPa), degradation of the chalk–steel interfacewas more severe, tending towards half of the resistance atlower stress levels.

Tilt table interface tests were undertaken on chalk–steel inter-faces with resulting very low normal stresses (less than 1 kPa),and showed good agreement with the IST results at lownormal stresses (σv = 16 kPa), especially for the rougher steelplates used during this study (Ra = 7·2 and 34 μm). The tilttable test was also used to determine the basic chalk–chalkinterface friction angle, which seems to be a key indicator ofbehaviour for degraded chalk, which in some cases may beused as a lower-bound design approach. Results would suggestthat the tilt table test is a useful, inexpensive method of charac-terising rock–steel interfaces that may give useful insights tobehaviour for preliminary design.

Comparison of the results from this study with those ofother higher UCS rocks would suggest that a previouslydeveloped alpha against normalised UCS based designapproach to predicting chalk–steel interface resistance may beadopted, but that some care needs to be exercised when crush-ing or degradation of chalk could occur. For the particularapplication of tidal stream generator foundations identifiedin this paper this may be further exacerbated by the cyclicnature of loading encountered, which was not investigated inthis paper.

AcknowledgementsThe Energy Technology Partnership (ETP) and Lloyd’sRegister EMEA are gratefully acknowledged for the funding ofthis project. The authors would also like to acknowledge thesupport of the European Regional Development Fund(ERDF) SMART Centre at the University of Dundee thatallowed purchase of the equipment used during this study. Theviews expressed are those of the authors alone, and do notnecessarily represent the views of their respective companies oremploying organisations.

10

0·02

0·04

0·06

0·08

0·10

10 100 1000 10 000Normalised unconfined compressive strength, UCS/σv

Range of degradationover 7 m displacement

Steel Ra = 0·4 mm (b = 1·78, c = –1·26)Steel Ra = 7·2 mm (b = 1·25, c = –1·16)

Old Red Sandstone (b = 0·53, c = –0·97)Chalk dry (b = 0·47, c = –0·87)Chalk, saturated (b = 0·59, c = –0·94)Chalk, extended displacement test (100 kPa)Chalk, extended displacement test (300 kPa)Chalk, extended displacement test (700 kPa)

Adh

esio

n fa

ctor

α, τ

peak

/UC

S

Figure 9. Adhesion factor values for chalk and sandstone

compared to contours from Ziogos et al. (2015a) (steel

Ra = 7·2 μm)

12



REFERENCES

Altindag R and Guney A (2010) Predicting the relationshipsbetween brittleness and mechanical properties (UCS, TSand SH) of rocks. Scientific Research and Essays 5(16):2107–2118.

ASTM (2008) ASTM D 3967-08: Standard test method forsplitting tensile strength of intact rock core specimens.ASTM International, West Conshohocken, PA, USA.

Barmpopoulos IH, Ho TYK, Jardine RJ and Anh-Minh N (2010)The large displacement shear characteristics of granularmedia against concrete and steel interfaces. InCharacterization and Behavior of Interfaces. Proceedings ofthe Research Symposium on Characterization and Behaviorof Interfaces, Atlanta, Georgia, USA (Frost JD (ed.)).IOS Press, Amsterdam, the Netherlands, pp. 16–23.

BSI (1990) BS EN ISO 1377-2:1990. Methods of test forsoils for civil engineering purposes. Classification tests.BSI, London, UK.

Clayton CRI and Saffari-Shooshtari N (1990) Constantnormal stiffness direct shear testing of chalk/concreteinterfaces. In Proceedings of the InternationalChalk Symposium, Brighton. Thomas Telford, London,UK, pp. 233–238.

Carbon Trust (2005) Variability of UK Marine Resources.Environmental Change Institute, University of Oxford,Oxford, UK. See http://www.carbontrust.com/media/174017/eci-variability-uk-marine-energy-resources.pdf(accessed 05/06/2016).

Carbon Trust (2011) Accelerating Marine Energy. Carbon Trust,London, UK. See http://www.carbontrust.com/media/5675/ctc797.pdf (accessed 04/06/2016).

Degarmo E, Black J and Kohser R (2003) Materials andProcesses in Manufacturing, 9th edn. Wiley, Hoboken,NJ, USA.

Fraenkel P (2002) Power from marine currents. Proceedings ofthe Institution of Mechanical Engineers, Part A: Journal ofPower and Energy 216(1): 1–14.

Horvath RG (1978) Field Load Test Data on Concrete to RockBond Strength for Drilled Pier Foundation. University ofToronto, Toronto, Canada, Publication 78-07.

Iscimen M and Frost JD (2010) Shearing behaviour ofcurved interfaces. In Characterization and Behavior ofInterfaces. Proceedings of the Research Symposium onCharacterization and Behavior of Interfaces, Atlanta,Georgia, USA (Frost JD (ed.)). IOS Press, Amsterdam,the Netherlands, pp. 35–43.

ISRM (International Society for Rock Mechanics) (2007) The BlueBook. The Complete ISRM Suggested Methods for RockCharacterization, Testing and Monitoring: 1974–2006.ISRM, Lisbon, Portugal.

Jardine RJ, Lehane BM and Everton SJ (1993) Frictioncoefficients for piles in sands and silts. Offshore SiteInvestigation and Foundation Behaviour Advances inUnderwater Technology, Ocean Science and OffshoreEngineering SUT, London, UK, vol. 28, pp. 661–677.

Johnstone GS and Mykura W (1989) British Regional Geology,The Northern Highlands of Scotland. British GeologicalSurvey, Keyworth, Nottingham, UK.

Kishida H and Uesugi M (1987) Tests of interface between sandand steel in the simple shear apparatus. Géotechnique 37(1):45–52.

Kuo MYH, Vincent CM, Bolton MD, Hill A and Rattley M

(2015) A new torsional shear device for pipelineinterface shear testing. Proceedings of 3rd InternationalSymposium on Frontiers in Offshore Geotechnics(Meyer V (ed.)). Taylor & Francis Group, London, UK,pp. 405–410.

Lehane BM, Schneider JA and Xu X (2005) CPT BasedDesign of Driven Piles in Sand for Offshore Structures.University of Western Australia, Perth, Australia, GEO:05345.

Li C and Håkansson U (1999) Performance of the Swellexbolt in hard and soft rocks. In Rock Support andReinforcement Practice in Mining: Proceedings ofthe International Symposium on Ground Support(Villaescusa E, Windsor CR and Thompson AG (eds)).Balkema Publishers, Rotterdam, the Netherlands,pp. 103–108.

Lord JA, Clayton CR and Mortimore RN (2002) Engineering inChalk. CIRIA, London, UK, CIRIA Publication C574.

Matthews MC and Clayton CRI (1993) Influence of intactporosity on the engineering properties of a weak rock.In Geotechnical Engineering of Hard Soils – Soft Rocks(Anagnostopoulos A, Schlosser F, Kalteziotis N andFrank R (eds)). AA Balkema, Rotterdam, the Netherlands,vol. 1, pp. 693–702.

NAVFAC (Naval Facilities Engineering Command) (1986)Soil Mechanics Design Manual 7.01. Naval FacilitiesEngineering Command Publications Transmittal,Alexandria, VA, USA.

Peterson MS, Kulhawy FH, Nucci LR and Wasil BA (1976)Stress Deformation Behavior of Soil–Concrete Interfaces.Niagara Mohawk Power Corporation, Syracuse, NY, USA,Contract report B-49.

Potyondy JG (1961) Skin friction between various soils andconstruction materials. Géotechnique 11(4): 339–355.

Rosenberg P and Journeaux NL (1976) Friction and endbearing tests on bedrock for high capacity socket design.Canadian Geotechnical Journal 13(3): 324–333.

Saada AS and Townsend FC (1981) State of the art: laboratorystrength testing of soils. In Laboratory Shear Strength ofSoil. (Young RN and Townsend FC (eds)). ASTMInternational, West Conshohocken, PA, USA, ASTM STP740, pp. 7–77.

Small AA, Cook GA and Brown MJ (2014) The geotechnicalchallenges of tidal turbine projects. Proceedings of theASME 2014 33rd International Conference on Ocean,Offshore and Arctic Engineering (OMAE2014). TheAmerican Society of Mechanical Engineers, New York,NY, USA, vol. 9A.

13



http://www.carbontrust.com/media/174017/eci-variability-uk-marine-energy-resources.pdf







http://www.carbontrust.com/media/5675/ctc797.pdf







Tomlinson MJ (2001) Foundation Design and Constructions,7th edn. Pearson Education Limited, Harlow, UK.

Uesugi M and Kishida H (1986a) Frictional resistance at yieldbetween dry sand and mild steel. Soils and Foundations26(4): 139–149.

Uesugi M and Kishida H (1986b) Influential factors of frictionbetween steel and dry sand. Soils and Foundations 26(2):33–46.

USBR (United States Bureau of Reclamation) (2009) USBR6258: Procedure for Determining the Angle of BasicFriction (Static) Using a Tilting Table Test. U.S.Department of the Interior Bureau of Reclamation,Technical Service Centre, Denver, CO, USA.

Whitham F (1991) The stratigraphy of the Upper CretaceousFerriby, Welton and Burnham formations north of theHumber, north-east England. Proceedings of the YorkshireGeological Society 48(3): 247–254.

Whitham F (1993) The stratigraphy of the Upper CretaceousFlamborough Chalk Formation north of the Humber,

north-east England. Proceedings of the YorkshireGeological Society 49(3): 235–258.

Williams FP and Pells JN (1981) Side resistance rock sockets insandstone, mudstone, and shale. Canadian GeotechnicalJournal 18(4): 502–513.

Yu ZX, Tham LG, Lee PKK, Liu YT and Wong MK (2013)A study on socketed steel H-piles under vertical load.HKIE Transactions 20(3): 172–187.

Ziogos A, Brown MJ, Ivanovic A and Morgan N (2015a) Interfaceshear characteristics of Scottish rock samples from siteswith tidal energy potential. Proceedings of XVI EuropeanConference on Soil Mechanics and GeotechnicalEngineering. ICE Publishing, London, UK, pp. 1357–1362.

Ziogos A, Brown MJ, Ivanovic A and Morgan N (2015b)Investigation of rock–steel interface testing for gravityfoundations for marine energy generators. Proceedings of3rd International Symposium on Frontiers in OffshoreGeotechnics (Meyer V (ed.)). Taylor & Francis Group,London, UK, pp. 1245–1250.

HOW CAN YOU CONTRIBUTE?

To discuss this paper, please email up to 500 words to theeditor at [email protected]. Your contribution will beforwarded to the author(s) for a reply and, if consideredappropriate by the editorial board, it will be published asdiscussion in a future issue of the journal.

Proceedings journals rely entirely on contributions fromthe civil engineering profession (and allied disciplines).Information about how to submit your paper onlineis available at www.icevirtuallibrary.com/page/authors,where you will also find detailed author guidelines.

14



chalk steel interface testing for marine energy foundations · of very high density according to...

Documents