See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273431903

Compressive, Flexural Bond, and Shear Bond Strengths of In Situ New Zealand
Unreinforced Clay Brick Masonry Constructed Using Lime Mortar between the
1880s and 1940s

Article  in  Journal of Materials in Civil Engineering · April 2014

DOI: 10.1061/(ASCE)MT.1943-5533.0000685

CITATIONS READS

54 1,555

3 authors, including:

David Biggs Jason Ingham

Biggs Consulting Engineering University of Auckland
32 PUBLICATIONS   295 CITATIONS    537 PUBLICATIONS   6,994 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Papakainga Housing Solutions View project

Gordion Turkey Conservation View project

All content following this page was uploaded by Jason Ingham on 29 March 2015.

The user has requested enhancement of the downloaded file.

Manuscript
Click here to download Manuscript: Compressive_flexural_bond_and_shear_bond_strengths_of_in-situ_unreinforced_clay_brick_masonry_

1 Title: Compressive, flexural bond and shear bond strengths of in-situ New Zealand unreinforced

2 clay brick masonry constructed using lime mortar between the 1880s and 1940s

3 Ronald Lumantarna1, David T. Biggs2 Dist M. ASCE and Jason M. Ingham3 M. ASCE

4 ABSTRACT

5 The importance of sufficient masonry mortar joint bond strength when a structure is subjected to

6 in-plane and out-of-plane loads has been emphasised by several authors. However, masonry

7 unit/mortar bond strength is difficult to predict and performing mechanical tests on existing

8 masonry buildings to determine masonry flexural bond and shear bond strengths is generally not

9 practical, such that predictive expressions relating the masonry flexural bond and shear bond

10 strengths to other masonry properties are desirable. Although relationships between brick/mortar

11 bond and compressive strength have been investigated previously by researchers located in many

12 different parts of the world, most of these studies were laboratory based and did not include the

13 testing of existing masonry buildings within their scope. The present study aimed to characterise

14 the material properties of New Zealand unreinforced clay brick masonry (URM) buildings that

15 were generally built between 1880 and 1930, with in-situ testing and sample extraction

16 performed on 6 heritage buildings. Masonry compression, bond wrench and shear bond tests

17 were undertaken. The experimental results indicate that the masonry flexural bond strength and

18 the bed joint cohesion can be satisfactorily related to the mortar compressive strength.

1
Ph.D. student, Department of Civil & Environmental Engineering, University of Auckland, Private Bag 92019,
Auckland 1010, New Zealand, [email protected]
2
Principal, Biggs Consulting Engineering, Troy, New York 12180-6671, USA, [email protected]
3
Associate Professor, Department of Civil & Environmental Engineering, University of Auckland, Private Bag
92019, Auckland 1010, New Zealand, [email protected]
1
19 CE Database subject headings: Brick masonry; Walls; Flexural strength; Compressive strength;

20 Shear strength; Masonry prism; Masonry Properties; In-situ testing

21 INTRODUCTION

22 The importance of sufficient masonry mortar joint bond strength when an unreinforced masonry

23 bearing wall building is subjected to in-plane and out-of-plane loads has been emphasised by

24 several authors (Russell 2010; Venkatarama Reddy and Gupta 2006), and therefore methods for

25 characterising masonry flexural bond and shear bond strengths are desirable. As described by

26 Hilsdorf (1969), McNary and Abrams (1985) and Khoo and Hendry (1975), the failure of

27 masonry in compression is governed by deformation of the brick units and mortar when

28 subjected to a multi-axial stress state, whilst assuming that the brick/mortar bond remains intact

29 until the ultimate compression load is reached (Venkatarama Reddy and Uday Vyas 2008).

30 However, it was revealed that bed joint bond failure occurred during compression testing when

31 the brick/mortar bond was poor (Sarangapani et al. 2005). Sarangapani et al. (2005) also found

32 that the brick/mortar bond characteristics are not directly related to the deformation

33 characteristics of the brick unit and mortar, but instead are influenced by factors such as the

34 roughness of the brick surface and the mortar water retentivity. Roughness is an indication of the

35 open pore structure of the brick surface.

36 Predictive expressions relating the masonry flexural bond and shear bond strengths to other

37 masonry properties were desired as performing tests on existing masonry buildings to determine

38 these properties is generally not practical. The mortar and masonry compressive strengths were

39 the first candidates to be related to the masonry flexural bond and shear bond strengths as they

40 can be easily obtained through mechanical testing of extracted samples. The relationships
2
41 between brick/mortar bond and compressive strengths have often been investigated (Sarangapani

42 et al. 2005; Venkatarama Reddy and Gupta 2006; Venkatarama Reddy et al. 2007; Venkatarama

43 Reddy and Uday Vyas 2008; Venu Madhava Rao et al. 1996). However, most of these studies

44 were laboratory based and did not include the testing of existing masonry buildings within their

45 scope. The present study aimed to characterise the material properties of New Zealand URM

46 buildings that were generally built between 1880 and 1930 (Russell and Ingham 2010), and

47 therefore the relationships between flexural bond strength, shear bond strength, masonry

48 compressive strength and mortar compressive strength for masonry samples extracted from

49 actual URM buildings were explored. These relationships were intended to enable structural

50 engineers to undertake effective detailed seismic assessment of regular URM buildings without

51 requiring comprehensive and time consuming material testing.

52 PAST STUDIES ON THE MASONRY BOND STRENGTH

53 Factors affecting the development of brick/mortar bond

54 Several studies have been previously conducted to investigate masonry bond properties. The

55 brick/mortar bond development is effectively a mechanical process that is influenced by binder

56 hydration occurring at the brick surface and in the brick unit pores (Groot 1993; Lawrence and

57 Cao 1987; Sugo et al. 2001). The brick unit initial rate of absorption (IRA), surface roughness

58 and mortar water retentivity are the governing factors for this hydration (Pavia and Hanley 2010;

59 The brick Industry Association 2003).

60 Scrivener et al. (1992) performed bond wrench tests at construction sites in Australia and found

61 wide variation in masonry flexural bond strengths, especially when wall construction was

62 performed subject to uncontrolled environmental conditions. Grenley (1969) investigated various

3
63 brick/mortar combinations and found that the masonry flexural bond, tensile bond and

64 compressive strengths generally increased with increasing brick unit and mortar compressive

65 strengths. The masonry bond-compressive strength relationships that were obtained for the

66 various combinations showed strong correlations, although the influence that mortar compressive

67 strength had on the masonry bond strength could not be neglected.

68 Samarasinghe and Lawrence (1992) performed shear tests on masonry triplets replicating new

69 masonry construction and observed that masonry prisms that were constructed using pre-wetted

70 brick units had higher bed joint shear strengths than those constructed using dry or completely

71 saturated brick units. These researchers also observed that the bed joint shear strength increased

72 with increasing mortar compressive strength.

73 Sarangapani et al. (2005) performed bond wrench tests and shear bond tests on prisms that were

74 constructed using three different brick types and four different mortar grades. Four different

75 bond enhancing techniques were also implemented, thus resulting in an increase in the masonry

76 bond and compressive strengths without altering the mortar composition. It was found that the

77 prism compressive strength was generally more sensitive to variations in the brick/mortar bond

78 strength than to variations in the mortar compressive strength.

79 Venu Madhava Rao et al. (1996) investigated masonry bond-compressive strength relationships

80 using clay bricks, stabilised mud blocks and stabilised soil-sand blocks, whilst the mortars used

81 were a variety of cement:sand, cement:soil:sand and cement:lime:sand mortars. Their main

82 observations were:

83  For all types of masonry units, the masonry flexural bond strength increased with
84 increasing mortar compressive strength when cement mortar was used.

4
 85  The addition of soil or lime into the mortar mix improved the masonry flexural bond
86 strength.
87  The masonry flexural bond strength was high when masonry units with deep and wide
88 frogs were used in comparison to units without frogs.

89 Venkatarama Reddy and Gupta (2006) investigated cement-soil block/cement:soil:sand mortar

90 masonry and found that the masonry tensile bond strength increased with increasing block unit

91 compressive strength. Similarly, Venkatarama Reddy and Uday Vyas (2008) investigated three

92 cement-soil block/mortar combinations incorporating five different bond enhancing techniques.

93 They found that when soft block-stiff mortar combinations were used, the masonry flexural bond

94 strength increased with increasing masonry compressive strength.

95 Masonry bond failure types

96 The brick/mortar bond failure of masonry prisms when subjected to bond wrench and shear bond

97 tests can be classified as follows (Pavia and Hanley 2010; Sarangapani et al. 2005; Venkatarama

98 Reddy and Gupta 2006):

99  Type A: Failure at one brick/mortar interface;

100  Type B: Failure at both brick/mortar interfaces;
101  Type C: Failure within the mortar joint;
102  Type D: Failure within the brick unit;
103  Type E: Combination of failure within the brick unit and mortar joint.

104 Pavia and Hanley (2010) tested masonry prisms that were constructed using natural hydraulic

105 lime mortars. They subsequently found that 65% of the samples experienced failure type A,

106 whilst the remaining prisms exhibited failure type B. The average mortar compressive strengths

5
107 were not reported, and therefore it was not possible to assess if the brick/mortar bond failure

108 modes observed were governed by the strength of the mortar or by other factors.

109 Venkatarama Reddy and Gupta (2006) observed four different types of failure (types A, C, D

110 and E) for their experiments. Failure type E was the most common type observed, whilst failure

111 types C and D generally occurred when weak mortar and weak block units respectively were

112 used. Failure type A was exhibited by prisms that were constructed using moderately strong

113 cement-soil blocks.

114 Sarangapani et al. (2005) did not observe failure type C when testing masonry prisms, whereas

115 failure type D and a combination of failure types A and D were frequently observed for prisms

116 constructed using weak brick units and strong mortar. These researchers also reported that failure

117 type A mostly occurred when the brick/mortar interface bond strength was lower than the mortar

118 joint flexural strength, and therefore this failure type was exhibited by almost all prisms that

119 were constructed without bond enhancement.

120 The findings from the studies reported above indicate that (1) the masonry flexural bond and

121 shear bond strengths are possibly related to the masonry or mortar compressive strength; and (2)

122 the brick/mortar bond failure type depends on the brick/mortar interface bond strength as well as

123 on the relative comparison between the brick and mortar compressive strengths. However,

124 previous authors have not attempted to link the brick/mortar bond failure type to the bond-

125 compressive strength relationship. It was also established that past investigations mainly focused

126 on newly constructed samples and that the properties of assemblages extracted from existing

127 heritage URM buildings have not been thoroughly studied.

128 In the present study, an attempt was made to investigate the material properties of existing URM

129 buildings located in New Zealand. The aim was to determine the relationships between flexural
6
130 bond strength, shear bond strength and masonry compressive strength, as well as the magnitude

131 of mortar compressive strengths of masonry samples extracted from existing buildings. The

132 brick/mortar bond failure types were also considered in the investigation.

133 The past studies reported above were mainly based on newly made laboratory samples and

134 revealed that bond failure in masonry commonly occurs at the brick/mortar interface. However,

135 previous authors (Sarangapani et al. 2005; Venkatarama Reddy and Gupta 2006) have not

136 investigated prisms that were constructed using lime-rich mortar. Also, although Pavia and

137 Hanley (2010) studied prisms that were constructed using natural hydraulic lime mortars, the

138 brick units used in the experimental programme were perforated, hollow-cored brick units, which

139 are different to heritage New Zealand solid clay bricks. These differences between the past and

140 present studies have to be noted when making comparisons, especially considering that the

141 current experimental programme mainly focused on samples obtained from New Zealand

142 heritage URM buildings constructed during the 1880s to 1940s using lime-rich mortars.

143 EXPERIMENTAL PROGRAMME

144 In-situ material testing and sample extraction were performed for 6 New Zealand clay brick

145 URM buildings. When the project permitted, bond wrench tests and in-situ shear tests were

146 performed on-site to determine the masonry flexural bond and shear bond strengths. Also,

147 extracted masonry assemblages were cut in-situ using a masonry chainsaw or retrieved as

148 irregular masonry segments. These extracted samples were further trimmed in the laboratory to

149 form single leaf two and three brick high prisms to be used in laboratory compression, bond

150 wrench and triplet shear tests.

7
151 Prism compression test

152 Single leaf three brick high extracted prisms were capped using gypsum plaster to ensure a

153 uniform stress distribution, and were tested in compression using a 2000 kN instron machine

154 following the prism compression test protocol of ASTM C 1314 - 03b (2003a) (see Figure 1).

155 Bond wrench test

156 The bond wrench test AS 3700-2001 (Standards Australia 2001) was adopted for both in-situ and

157 laboratory applications due to its greater portability in comparison to the ASTM C 1072 - 00a

158 (2000) test setup, and therefore the bond wrench test was more suitable for in-situ testing. The

159 bond wrench arm (see Figure 2) was constructed as stipulated in AS 3700-2001 (Standards

160 Australia 2001), with a hook connector installed at the end of the bond wrench arm. An empty

161 container was attached to the hook and then gradually filled using sand to apply bending stresses

162 to the mortar joint until flexural bond failure occurred. The weight of the bucket and sand was

163 measured to the nearest 0.01 gram and used to calculate the flexural bond failure stress.

164 Shear bond tests

165 The in-situ shear test ASTM C 1531 - 03 (2003b) and the triplet shear test Rilem TC 127-

166 MS.B.4 (1996) were adopted for on-site and laboratory shear bond testing respectively. The in-

167 situ shear tests were performed without flat jacks, and the triplet shear tests were performed

168 whilst subjected to different levels of axial pre-compression load.

169 Figure 3 illustrates the in-situ shear test setup. The hydraulic jack was loaded using a pressure

170 controlled hydraulic pump and a displacement gauge was attached on the wall face adjacent to

171 the vertical cut joint, to identify when bed-joint sliding failure occurred. It was also noted that the

8
172 contribution of collar joints was not considered in the bed joint shear strength calculation as the

173 collar joints were mostly poorly laid, and therefore their contribution to the bed joint shear

174 strength was minimal.

175 The triplet shear test setup is shown in Figure 4. Prisms were placed between two steel plates that

176 were interconnected using four steel rods. The axial pre-compression load was applied by

177 tightening the nuts at the ends of the steel rods and was recorded using a load cell. The sample

178 was prepared such that the middle course of the prisms consisted of a full brick unit.

179 The triplet shear tests were performed whilst being subjected to axial pre-compression stresses of

180 0.2 MPa, 0.4 MPa and 0.6 MPa. The shear strength of the mortar joints can be represented by the

181 Mohr-Coulomb friction law as per Equation (1) (ASTM 2003b; Lourenço et al. 2004; Rilem

182 1996)

183 (1)

184 where = shear stress at a given axial compression; = shear stress at zero axial compression

185 (cohesion); = coefficient of friction; and = axial compression stress. Therefore, the mortar

186 bed joint cohesion could be derived because the triplet shear tests were performed under different

187 levels of axial pre-compression loads.

188 SOURCE AND PROPERTIES OF EXTRACTED SAMPLES

189 The source of the field extracted samples is described in

190 Table 1. These buildings (referred to as field sites) were constructed between 1881 and the

191 1940s, which coincides with the time period during which URM construction was popular in

192 New Zealand. Although variability in the constituent material properties amongst URM

9
193 buildings is expected, these field sites are deemed to be representative of the majority of New

194 Zealand URM buildings. Two of the field sites are shown in Figure 5.

195 Individual brick units and irregular mortar samples were sourced from each field site. The brick

196 unit compressive strength was determined using the half brick compression test ASTM C 67 -

197 03a (2003c), whilst the irregular field extracted mortar samples were carefully cut to form

198 rectangular test pieces, capped using gypsum plaster and tested in compression as prescribed in

199 Lumantarna (2012). A normalisation technique that accounts for the mortar sample footprint

200 dimensions and height to thickness ratio was implemented, as these factors clearly influence the

201 measured mortar compressive strength (Lumantarna 2012). Therefore, the compressive strength

202 of the irregular mortar samples could be accurately interpreted. X-ray diffraction analysis and

203 acid digestion test results reported in Lumantarna (2012) suggest that most New Zealand vintage

204 mortars were likely to be lime based.

205 The average brick unit ( ) and mortar ( ) compressive strengths of the different field sites are

206 shown in

10
207 Table 2, where nb and nj show the number of brick units and mortar samples tested respectively.

208 The tests performed for each field sample group were also included, and it is noted that group D

209 prisms were not subjected to triplet shear tests due to their limited availability. The average

210 compressive strength of the brick units was found to vary between 8.5 MPa and 27.3 MPa, whilst

211 the average mortar compressive strength ranged from 1.23 MPa to 8.58 MPa. The CoV values of

212 the brick unit and mortar compressive strengths were similar.

213 MASONRY ASSEMBLAGE TEST RESULTS AND DISCUSSION

214 Prism compression test results

215 The average compressive strength of the masonry prisms ( ) extracted from each field site is

216 shown in

11
217 Table 3, where n shows the number of prisms tested in compression. A minimum of four prisms

218 were tested in compression for each brick/mortar combination. The average masonry

219 compressive strengths were found to vary between 3.3 MPa and 14.7 MPa, and their CoV values

220 were similar to those of the brick unit and mortar compressive strengths. As the field extracted

221 samples were comprised of brick units that were stronger than the mortar, the prism compression

222 failures were mostly initiated by splitting failure of the brick units, followed by crushing of the

223 mortar joints as the loading continued.

224 Bond wrench test results

225 The average flexural bond strengths ( ) and bond failure types of the field samples are shown

226 in

12
227 Table 3. A minimum of three samples were tested for each brick/mortar combination, with most

228 prisms exhibiting a flexural bond failure within the mortar joint (failure type C). It was thought

229 that these bond failures within the mortar joints occurred because the heritage buildings

230 investigated in this experimental programme were constructed using lime-rich mortars, and

231 therefore the mortar did not have sufficient strength to resist the applied tensile force. Also, it is

232 noted that the buildings included in the experimental programme were New Zealand URM

233 buildings which were built during a specific time period (1880s-1940s), and hence the

234 observations from this study may not be applicable for newer URM construction, where cement-

235 based mortars had been used. Furthermore, it was found that samples which exhibited

236 brick/mortar interface bond failure (failure type A) had lower flexural bond strengths than those

237 which exhibited failure type C. It was thought that the samples that exhibited brick/mortar

238 interface bond failure were disturbed during their preparation, resulting in low flexural bond

239 strengths being recorded and therefore those results were disregarded from the analysis. Figure 6

240 and Figure 7 illustrate the observed bond failure types. The average mortar compressive strength

241 ( ) for each field site is also included in

13
242 Table 3 to enable the relationship between and to be investigated.

243 The average flexural bond strengths of the field samples ranged from 0.031 MPa to 0.345 MPa.

244 The variability in the bond wrench test results (CoV between 0.11 and 0.33) was thought to be

245 reasonable considering the irregular nature of URM construction.

14
246 Table 3 shows that the average masonry flexural bond strength increased with increasing average

247 mortar compressive strength. Also, it is noted that most of the field samples exhibited bond

248 failures within the mortar joints, which is a failure mode that previously has rarely been reported.

249 Cizer et al. (2008) and Moropoulou et al. (2005) reported that both the compressive and flexural

250 strengths of mortar increased over time, which suggests that there is a time-dependent

251 relationship between these properties. Therefore, the masonry flexural bond strength was related

252 to the mortar compressive strength for those results where flexural bond failure occured within

253 the mortar. Figure 8 illustrates the average flexural bond strength-average mortar compressive

254 strength relationship and the average flexural bond strength-average masonry compressive

255 strength relationship, revealing that masonry flexural bond strength is better characterised using

256 the mortar compressive strength than using the masonry compressive strength. Figure 8 also

257 shows that the masonry flexural bond strength, , can be satisfactorily equated to 0.031

258 (coefficient of determination, R2 = 82%). The negative R2 value of -0.079 shows that the

259 relationship between flexural bond strength and masonry compressive strength is poor. It is noted

260 that although there is an apparent outlier in Figure 8 (see circled data point), this data point

261 originated from a legitimate dataset, where the test results were reasonably consistent for all

262 samples used to calculate this data point (refer to

15
263 Table 3). Therefore, it was decided that this circled data point should not be ignored. Also, all

264 samples considered in Figure 8 experienced flexural bond failures within their mortar joints, and

265 therefore relating the flexural bond strength to the mortar compressive strength was considered

266 to be more suitable than relating the flexural bond strength to the masonry compressive strength.

267 Shear bond test results

268 Bed joint shear strength

269 The mortar bed joint shear strength ( ) of the different prism groups at each level of axial pre-

270 compression stress ( ) is shown in

16
271 Table 4, with a minimum of two samples tested at each level of axial pre-compression stress.

272 Sample groups HC and RB were tested at axial pre-compression stress levels of approximately

273 0.2 MPa, 0.4 MPa and 0.6 MPa. In-situ shear tests were performed for the other field sample

274 groups (sample groups AH, CFK and TA), with the in-situ axial pre-compression loads estimated

275 based upon the amount of overburden located above the test locations. These estimated

276 overburden loads were considered as the first axial pre-compression stress level, which for

277 sample groups AH, CFK and TA were determined to correspond to 0.02 MPa, 0.04 MPa and

278 0.04 MPa respectively. Three brick high prisms were also extracted for laboratory triplet shear

279 tests, and therefore the bed joint shear strength at two additional levels of axial pre-compression

280 stress was obtained. It was decided that for sample groups AH, CFK and TA, the triplet shear

281 tests be performed at axial pre-compression stresses of 0.2 MPa and 0.4 MPa. A large number of

282 TA prisms were available, and therefore these prisms were also tested at a 0.6 MPa axial pre-

283 compression stress.

284 The observed bed joint failure types were consistent with those observed during the bond wrench

285 tests, where almost all of the field samples experienced shear bond failures within the mortar

286 joints (failure type C) as the buildings investigated in this experimental programme were

287 constructed using lime-rich mortars. The samples that experienced failure type A (interface bond

288 failure) had lower bed joint shear strengths than those that experienced failure type C as they

289 were likely to be disturbed during the sample preparation process, and hence these samples were

290 disregarded from the analysis.

17
291 Table 4 shows that the bed-joint shear strength ( ) increased with increasing axial pre-

292 compression stress ( ), and it was noted that there was wide variation in the bed joint shear

293 strength of prism groups HC and RB when = 0.6 MPa. In addition, the bed joint shear strength

294 at each level of axial pre-compression stress generally increased with increasing average mortar

295 compressive strength ( ), which was expected as most field samples experienced shear failures

296 within the mortar joints, and therefore their bed joint shear strengths were influenced by the

297 mortar properties instead of by the brick/mortar interface bond characteristics.

298 Bed joint cohesion

299 Figure 9 illustrates the bed joint shear strength ( )-axial compression stress ( ) relationships.

300 The best fit equations were used to derive the coefficient of friction ( ) and bed-joint cohesion

301 ( ) of each group based on the Mohr-Coulomb friction law, as reported in

18
302 Table 5. Figure 9 shows that the sample groups had comparable Mohr-Coulomb friction slopes

303 (coefficient of friction, ), whilst

19
304 Table 5 indicates that their y-intercepts (cohesion, ) increased with increasing average mortar

305 compressive strength. Figure 10 illustrates the relationships between mortar bed joint cohesion

306 and average mortar compressive strength, and between mortar bed joint cohesion and average

307 masonry compressive strength. The mortar bed joint cohesion is better characterised using the

308 mortar compressive strength than using the masonry compressive strength, where can be

309 satisfactorily equated to 0.055 (R2 = 82%). The negative R2 value of -0.146 shows that the

310 relationship between mortar bed joint cohesion and masonry compressive strength is poor.

311 Similar to that shown in Figure 8, there is an apparent outlier in Figure 10 (see circled data

312 point). However, this data point originated from a legitimate dataset, where the cohesion value

313 was calculated based on a reasonably consistent dataset as shown in Figure 9, whilst the average

314 masonry compressive strength was obtained from consistent compression test results (refer to

20
315 Table 3). It is noted that the outlier in Figure 10 originated from the same site as did the outlier

316 shown in Figure 8. Also, all samples considered for derivation of the relationship between

317 cohesion and masonry/mortar compressive strength experienced shear bond failures within their

318 mortar joints, and therefore relating the bed joint cohesion to the mortar compressive strength

319 was considered to be more suitable than relating the bed joint cohesion to the masonry

320 compressive strength.

321 SUMMARY AND CONCLUSIONS

322 In-situ material testing and sample extraction were performed on 6 New Zealand URM buildings

323 to investigate the relationships between flexural bond strength, shear bond strength and

324 compressive strength of existing URM bearing wall buildings. The following conclusions were

325 drawn based on the experimental results:

326 When subjected to bond wrench and shear bond tests, almost all of the field samples exhibited

327 bond failures within the mortar joints (failure type C). The samples that exhibited failure type A

328 (interface bond failure) were judged to be disturbed during the sample cutting process, resulting

329 in lower flexural bond strengths than those that exhibited failure type C, and were consequently

330 disregarded from the analysis. It was theorised that the heritage buildings investigated in this

331 experimental programme were constructed using lime-rich mortars, and therefore the mortar did

332 not have sufficient strength to resist the applied tensile force, leading to failure type C. Also, it is

333 noted that the buildings included in the experimental programme were New Zealand URM

334 buildings which were built during a specific time period, and hence the observations from this

335 study may not be applicable for newer URM construction.

21
336 A review of past investigations suggests that there is a relationship between mortar compressive

337 strength and mortar flexural strength. Therefore, the masonry flexural bond strength was related

338 to the mortar compressive strength as most field samples exhibited bond failures within the

339 mortar joints. The masonry flexural bond strength is better characterised using the mortar

340 compressive strength than using the masonry compressive strength.

341 The mortar bed joint shear strength increased with increasing axial pre-compression stress, and

342 the bed joint shear strength at each level of axial pre-compression stress generally increased with

343 increasing average mortar compressive strength. These observations were expected as most field

344 samples exhibited shear failure within the mortar joints, and therefore their bed joint shear

345 strengths were influenced by the mortar properties instead of the brick/mortar bond

346 characteristics.

347 Comparable coefficients of friction were derived for all field sites, whilst the measured cohesion

348 increased with increasing average mortar compressive strength. It was shown that the mortar bed

349 joint cohesion is better characterised using the mortar compressive strength than using the

350 masonry compressive strength.

351 It is suggested that future studies attempt to further investigate the material properties of existing

352 heritage masonry buildings. The brick/mortar bond failure type shall also be considered when

353 determining the factors that can be related to the masonry flexural and shear bond strengths.

354

22
355 REFERENCES

356 ASTM (2000). "Standard Test Method for Measurement of Masonry Flexural Bond Strength." C

357 1072 - 00a, ASTM International, Pennsylvania, United States.

358 ASTM (2003a). "Standard Test Method for Compressive Strength of Masonry Prisms." C 1314 -

359 03b, ASTM International, Pennsylvania, United States.

360 ASTM (2003b). "Standard Test Methods for In Situ Measurement Of Masonry Mortar Joint

361 Shear Strength Index." C 1531 - 03, ASTM International, Pennsylvania, United States.

362 ASTM (2003c). "Standard Test Methods for Sampling and Testing Brick and Structural Clay

363 Tile." C 67 - 03a, ASTM International, Pennsylvania, United States.

364 Cizer, Ö., Van Balen, K., Van Gemert, D., and Elsen, J. (2008). "Blended lime-cement mortars

365 for conservation purposes: Microstructure and strength development." Structural Analysis

366 of Historic Construction, D'Ayala, and Fodde, eds., Taylor & Francis Group, London,

367 United Kingdom, 965-972.

368 Grenley, D. G. (1969). "Study of the effect of certain modified mortars on compressive strength

369 and flexural strength of masonry." Designing, engineering, and constructing with

370 Masonry Products, F. B. Johnson, ed., Gulf Publishing Company, Houston, United

371 States, 28-33.

372 Groot, C. (1993). "Effects of water on mortar brick bond." Doctor of Philosophy, University of

373 Delft, Delft, Netherlands.

374 Hilsdorf, H. K. (1969). "An investigation into the failure of brick masonry loaded in axial

375 compression." Designing, engineering, and constructing with Masonry Products, F. B.

376 Johnson, ed., Gulf Publishing Company, Houston, 34-41.

23
377 Khoo, C. L., and Hendry, A. W. (1975). "A failure criterion for brickwork in axial compression."

378 Proceedings of the 3rd international brick masonry conference, L. Foertig, and K.

379 Globel, eds.Bonn, Germany, 139-145.

380 Lawrence, S. J., and Cao, H. T. (1987). "An Experimental Study of the Interface Between Brick

381 and Mortar." Proc., The 4th North American Masonry Conference, Los Angeles,

382 California, United States, August, 1-14.

383 Lourenço, P. B., Barros, J. O., and Oliveira, J. T. (2004). "Shear testing of stack bonded

384 masonry." Construction and Building Materials, 18(2), 125-132.

385 Lumantarna, R. (2012). "Material Characterisation of New Zealand's Clay Brick Unreinforced

386 Masonry Buildings." Doctor of Philosophy, The University of Auckland, Auckland, New

387 Zealand.

388 McNary, W. S., and Abrams, D. P. (1985). "Mechanics of masonry in compression." Journal of

389 Structural Engineering, 111(4), 857-870.

390 Moropoulou, A., Bakolas, A., Moundoulas, P., Aggelakopoulou, E., and Anagnostopoulou, S.

391 (2005). "Strength development and lime reaction in mortars for repairing historical

392 masonries." Cement and Concrete Composites, 27(2), 289-294.

393 Pavia, S., and Hanley, R. (2010). "Flexural bond strength of natural hydraulic lime mortar and

394 clay brick." Materials and Structures, 43(7), 913-922.

395 Rilem (1996). "MS-B.4 Determination of shear strength index for masonry unit/mortar junction."

396 Materials and Structures, 29(8), 459-475.

397 Russell, A. (2010). "Characterisation and Seismic Assessment of Unreinforced Masonry

398 Buildings." Doctor of Philosophy, The University of Auckland, Auckland, New Zealand.

24
399 Russell, A., and Ingham, J. (2010). "Prevalence of New Zealand's Unreinforced Masonry

400 Buildings." Bulletin of The New Zealand Society for Earthquake Engineering, 43(3), 182-

401 201.

402 Samarasinghe, W., and Lawrence, S. J. (1992). "Effect of high suction rate in low strength

403 bricks on brick mortar bond." Proc., The 4th International Seminar on Structural

404 Masonry for Developing Countries, Madras, India, December, 43-50.

405 Sarangapani, G., Venkatarama Reddy, B. V., and Jagadish, K. S. (2005). "Brick-Mortar Bond

406 and Masonry Compressive Strength." Journal of Materials in Civil Engineering, 17(2),

407 229-237.

408 Scrivener, J., Zsembery, S., McNeily, T., and Lawrence, S. J. (1992). "In-situ bond strength of

409 clay brickwork." Proc., Proceedings of the 4th International Seminar on Structural

410 Masonry for Developing Countries, Madras, India, December, 51-55.

411 Standards Australia (2001). "Appendix D: Method of Test for Flexural Strength." AS 3700 -

412 2001.

413 Sugo, H. O., Page, A. W., and Lawrence, S. J. (2001). "The development of mortar/unit bond."

414 The 9th Canadian Masonry Symposium, Fredericton, New Brunswick, Canada, 4 - 6

415 June.

416 The brick Industry Association (2003). "Mortars for brick masonry."Technical Notes on Brick

417 Construction, Reston, Virginia, United States.

418 Venkatarama Reddy, B. V., and Gupta, A. (2006). "Tensile Bond Strength of Soil Cement Block

419 Masonry Couplets Using Cement-Soil Mortars." Journal of Materials in Civil

420 Engineering, 18(1), 36-45.

25
421 Venkatarama Reddy, B. V., Lal, R., and Nanjunda Rao, K. S. (2007). "Enhancing Bond Strength

422 and Characteristics of Soil-Cement Block Masonry." Journal of Materials in Civil

423 Engineering, 19(2), 164-172.

424 Venkatarama Reddy, B. V., and Uday Vyas, C. V. (2008). "Influence of shear bond strength on

425 compressive strength and stress-strain characteristics of masonry." Materials and

426 Structures, 41(10), 1697-1712.

427 Venu Madhava Rao, K., Venkatarama Reddy, B. V., and Jagadish, K. S. (1996). "Flexural bond

428 strength of masonry using various blocks and mortars." Materials and Structures, 29(2),

429 119-124.

430
431

432

26
433 TABLES

434 Table 1: Details of field sites

Material Year City Building Details Sample(s) extracted In-situ test(s)

Code Built performed
AH 1884 Wellington Single storey residential house Brick units, mortar, prisms Bond wrench,
in-situ shear
D 1940s Auckland Single storey warehouse Brick units, mortar, prisms None
CFK 1910 Auckland Two storey kindergarten Brick units, mortar, prisms In-situ shear
TA 1946 Te Awamutu Single storey horse stable Brick units, mortar, prisms In-situ shear
RB 1930s Auckland Two storey Irish Pub Brick units, mortar, prisms None
HC 1881 Wellington Two storey government building Brick units, mortar, prisms None
435

27
436 Table 2: Brick unit and mortar compressive strengths
Field Average nb Average nj Compression Bond wrench Shear bond
Site MPa (CoV) MPa (CoV) Lab In-situ Lab In-situ Lab
AH 8.5 (0.18) 17 1.23 (0.17) 7 Yes Yes Yes Yes Yes
D 17.1 (0.15) 7 2.62 (0.19) 16 Yes No Yes No No
CFK 16.0 (0.11) 10 4.14 (0.19) 14 Yes No Yes Yes Yes
TA 21.1 (0.23) 9 5.92 (0.17) 8 Yes No Yes Yes Yes
RB 27.3 (0.21) 32 6.65 (0.19) 11 Yes No Yes No Yes
HC 16.3 (0.20) 8 8.58 (0.14) 16 Yes No Yes No Yes
437

28
438 Table 3: Prism compression strength and flexural bond strength
Prism Average n Average Average No of samples and bond
group MPa (CoV) MPa (CoV) MPa (CoV) failure type(s)
AH 3.3 (0.19) 5 1.23 (0.17) 0.031 (0.25) 7 type C
D 6.1 (0.15) 4 2.62 (0.19) 0.057 (0.11) 3 type C
CFK 7.4 (0.12) 6 4.14 (0.19) 0.116 (0.20) 4 type C, 1 type A*
TA 12.1 (0.12) 6 5.92 (0.17) 0.127 (0.28) 5 type C
RB 14.7 (0.21) 6 6.65 (0.19) 0.172 (0.24) 6 type C, 2 type A*
HC 6.6 (0.23) 6 8.58 (0.14) 0.345 (0.21) 5 type C, 2 type A*
439 * Samples experiencing failure type A were disregarded from the calculations

29
440 Table 4: Mortar bed joint shear strengths at different levels of axial compression
Prism Average Average Sample (MPa) No of
Group No. = = = = = samples and
(MPa) (MPa) 0.02 0.04 0.2 0.4 0.6 failure
MPa MPa MPa MPa MPa type(s)
AH 3.3 1.23 1 0.146 - 0.404 0.480 - All type C
2 0.157 - 0.330 0.510 -
3 0.170 - - - -
4 0.159 - - - -
CFK 7.4 4.14 1 - 0.295 0.409 0.576 - 6 type C,
2 - 0.289 0.348 0.558 - 1 type A
HC 6.6 8.58 1 - - 0.584 0.720 1.152 11 type C,
2 - - 0.608 0.880 0.775 2 type A
3 - - 0.659 0.754 1.122
4 - - 0.622 - 0.882
TA 12.1 5.92 1 - 0.367 0.483 0.651 0.763 11 type A,
2 - 0.280 0.505 0.683 0.874 1 type A
3 - 0.373 - 0.693 -
4 - 0.437 - - -
RB 14.7 6.65 1 - - 0.690 0.737 0.955 12 type C,
2 - - 0.455 0.660 1.089 2 type A
3 - - 0.666 0.699 0.731
4 - - - 0.711 0.918
5 - - - - 1.104
441

30
442 Table 5: and of the field samples
Prism Group (MPa) (MPa) (MPa)
AH 3.3 1.23 0.149 0.829
CFK 7.4 4.14 0.243 0.829
HC 6.6 8.58 0.430 0.917
TA 12.1 5.92 0.328 0.842
RB 14.7 6.65 0.391 0.907
443

31
Figure
Click here to download Figure: Figure 1.pdf

FIGURES

Figure 1: Prism compression test

23
Figure
Click here to download Figure: Figure 2.pdf

Figure 2: Bond wrench test

24
Figure
Click here to download Figure: Figure 3.pdf

displacement hydraulic jack

gauge

Figure 3: In-situ shear test

25
Figure
Click here to download Figure: Figure 4.pdf

steel rods
nuts

Figure 4: Triplet shear test

26
Figure
Click here to download Figure: Figure 5.pdf

(a) Site AH

(b) Site CFK

Figure 5: Representative field sites

27
Figure
Click here to download Figure: Figure 6.pdf

Figure 6: Bond failure type A, failure at brick/mortar interface

28
Figure
Click here to download Figure: Figure 7.pdf

Figure 7: Bond failure type C, failure within mortar joint

29
Figure
Click here to download Figure: Figure 8.pdf

0.4

0.35
Flexural Bond Strength (MPa)

0.3
y = 0.0151x
0.25 R² = -0.079
0.2 y = 0.031x
R² = 0.8169
0.15

0.1
Mortar
0.05 Masonry
0
0 5 10 15 20
Compressive Strength (MPa)

Figure 8: Flexural bond strength-compressive strength relationships of the field samples

30
Figure
Click here to download Figure: Figure 9.pdf

1.4
AH
CFK
1.2
HC
TA
1
RB
µ c
0.8
τ (MPa)

y = 0.9166x + 0.4304 y = 0.8423x + 0.3283

R² = 0.6578 R² = 0.9515
0.6 y = 0.8287x + 0.2427
y = 0.9074x + 0.3913 R² = 0.9604
R² = 0.6751
0.4 y = 0.8291x + 0.1489
R² = 0.9704

0.2

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
N (MPa)

Figure 9: - relationships of the field samples

31
Figure
Click here to download Figure: Figure 10.pdf

0.5
0.45
0.4
0.35 y = 0.055x
Cohesion (MPa)

0.3 R² = 0.822
y = 0.0315x
0.25 R² = -0.146

0.2
0.15
0.1
Mortar
0.05
Masonry
0
0 5 10 15 20
Compressive Strength (MPa)

Figure 10: Cohesion-compressive strength relationships of the field samples

32
Figure Caption List

LIST OF FIGURES

Figure 1: Prism compression test

Figure 2: Bond wrench test

Figure 3: In-situ shear test

Figure 4: Triplet shear test

Figure 5: Representative field sites

Figure 6: Bond failure type A, failure at brick/mortar interface

Figure 7: Bond failure type C, failure within mortar joint

Figure 8: Flexural bond strength-compressive strength relationships of the field samples

Figure 9: - relationships of the field samples

Figure 10: Cohesion-compressive strength relationships of the field samples

View publication stats