WO2003097553A1

WO2003097553A1 - Lightweight concrete

Info

Publication number: WO2003097553A1
Application number: PCT/NO2003/000163
Authority: WO
Inventors: Lasse Sunde
Original assignee: Norsk Glassgjenvinning As
Priority date: 2002-05-21
Filing date: 2003-05-21
Publication date: 2003-11-27
Also published as: NO20022388D0; AU2003238729A1; NO318389B1; EP1534645A1; NO20022388L

Abstract

Lightweight concrete based on Portland cement wherein the aggregate is solely or partially comprised of crushed expanded glass (foam-glass), said concrete exhibiting a weight increase of maximum 5% in 2 years tested according to Canadian Standard CSA A 23.2-14A. The lightweight concrete includes crushed foam-glass in an amount of at least 20% by weight of the fresh lightweight concrete and a pozzuolanic material that is included in an amount of at least 8% by weight of the cement in the lightweight concrete. The cement component(s) of the light weight concrete are chosen among CEM I low alkaline cement (max. 0.6% by weight Na2O), CEM II fly ash cement with an alkaline (Na2O) content less than 8% by weight of the fly ash content, as well as mixtures of such cements.

Description

Lightweight concrete

The present invention concerns a foam-glass concrete, i.e. a lightweight concrete where all of or a substantial part of the aggregate consists of crushed expanded glass, preferably made from recycle glass. According to another aspect the invention concerns a method for testing long term properties of such concretes.

Background

The main constituents in concrete are cement, water, sand (fine aggregates) and stones (coarse aggregates). Fine and coarse aggregates are divided in several fractions with respect to particle size during production. In addition it is common to add chemical and mineral additives in order to achieve desired properties for fresh and hardened concrete.

"Glass concrete" as such is well known and has been utilized for certain purposes. Use of concrete containing glass aggregate is, however, not problem-free. A particular problem with glass containing concrete is the presence of alkali in the concrete and problems thereby connected. Glass is alkali reactive and reacts with water soluble alkalis in the concrete, which in combination with foam glass (expanded glass) leads to formation of a voluminous gel. The gel easily leads to expansion with resulting fissures and cracks being formed in the concrete, or the gel becomes deposited in pores and voids involving an unwanted increase in the concrete weight that deprives it some of its load carrying capacity and isolating property. The problem is well recognized and has been attempted solved by using low alkali cements and different additives. Until now, however, no good solution has been found.

With respect to lightweight concrete particularly, with its comparatively large voids/ pores, it has been assumed that the alkali reactivity has been absent or insignificant, as it has not been understood that it is manifested only in the form of increased weight, caused by an "internal" expansion. Swedish patent No. 501 419 describes a concrete filler of recycle glass. By filler in this context is meant a glass fraction of grain size 0 - 0.25 mm. The characterizing feature of said patent is the utilization of a glass filler wherein at least 95 % by weight has a grain size less than 300 μm (0.3 mm ). It is an object of the method according to said patent to provide a concrete that has good compressive strength and good workability of the fresh concrete. It is also mentioned as an object to suppress alkali reactions in the concrete. By using a filler of very fine grains, the water content may be reduced, which is known to lead to an improved compressive strength of the hardened concrete. As commented further in the following, however, the choice of grain size for the glass fraction according to this patent will involve a potential risk of alkali reactivity in the hardened concrete in cases where the concrete also contains alkali reactive coarse aggregate. In Applicant's own patent application No. 2000 0612 a concrete wherein the fine sand aggregate is largely substituted by glass particles of a grain size 0-5 mm, by a non-alkali reactive mineral of same grain size or by a combination of two such fractions, where also the coarse (rock) fraction is largely substituted by glass particles of a grain size 5-20 mm, which concrete should also contain pozzuolanic materials if the finest fraction of glass is present in significant amounts.

When producing an alkali resistant concrete it is important that the total content of alkali is kept as low as possible. It is thus common to use a low alkali cement with a maximal equivalent Na₂O of 0.6 %. In addition the possible alkali content in additives must be taken into consideration.

Objectives The object of the present invention is to provide an environmentally friendly product to be produced when utilizing recycle glass as an aggregate for concrete.

More precisely it is an object to provide a foam-glass concrete with the best possible properties with respect to compressive strength and fracture stress deterioration.

More particularly the object of the invention is to eliminate the most significant disadvantage of foam-glass concrete, namely the deterioration of concrete properties, when used as a construction concrete, that results from alkali reactions.

It is a still further object of the invention to provide a reliable method for determining whether or not a lightweight concrete is alkali reactive.

The invention Said objectives are achieved by a concrete comprising water and cement where the coarse and fine aggregate of sand and rocks are substituted by crushed expanded glass as defined by claim 1.

Preferred embodiments of the invention are disclosed by the dependent claims.

According to a second aspect the invention concerns a method for testing the long-term properties of concretes, particularly lightweight concretes, as defined by claim 10. It has until recently not been taken into account that the glass itself is an alkali source. This is taken into account with the present invention and is illustrated by way of laboratory experiments referred to in the following.

Laboratory experiments were conducted as follows: Glass was crushed to different particle sizes and extracted in a saturated solution of calcium hydroxide at 50 C for until 20 weeks. At certain intervals of exposure samples were taken from each of the grain size fractions and analyzed with respect to their sodium and potassium content. Total amount extracted alkali as function of grain size is shown in table 1 below, as g Na₂0 equivalents/ 100 g glass. Table 1 : Extraction of alkali from glass.

The results shown in table 1 indicate that alkalis extracted from glass can contribute significantly to the total alkali content in a concrete. It is further indicated that the extractable amounts of alkali per weight unity of glass, increases with decreasing particle size. This recognition provides some of the basis for the present invention.

Accelerated tests (elevated temperature and humidity) for determining alkali reactions were conducted, mainly corresponding to Canadian Standard CSA A 23.2-14A. According to this accelerated test, 1 day old concrete prisms (10x10x45 cm) are exposed to 38 °C and 100% RH for a year. For the present investigations the exposure times were prolonged to 2 years.

During the present work with the manufacture of alkali resistant foam-glass concrete, 13 different concrete mixtures including crushed expanded recycle glass as aggregate. The mixture compositions and the results from the tests are shown in table 2 and 3. Four different Portland cements, CEM I Standard Cement by Norcem; CEM I low alkali white cement by Aalborg Portland, Denmark; CEM I low alkali construction cement by Norcem; and CEM II Standard Fly ash cement by Norcem, were included in the test program. The mixtures 1-10 were conducted with the three CEM I Portland cements. The effect of silica dust on the alkali reaction was investigated (by way of weight increase) with the mixtures 5, 6, 9 and 10. These mixtures are all based on low alkali cements. In the mixtures 2, 4, 6, 8 and 10 glass with grain size 0-0.5 mm was replaced by a non- alkali reactive rock, granite. It should be noted that the distinguishing between "treated" and "untreated" glass in tables 2a and 2b, solely refers to the fact that the "treated" glass fraction have had the finest fraction of particles removed, i.e. particles of a size less than abut 0.5 mm.

Table 2 shows length and weight change relative to initial length and weight respectively for a period of from one week to two years.

The mixtures 11-13 were prepared from CEM II Portland fly ash Cement in compositions with foam-glass as the sole aggregate and with varying silica content.

Table 3 shows length and weight change relative to initial length and weight respectively for a period of from one week to 18 months (results for 2 years were not reached at the time). Table 2a. Compositions (kg/m³) and test results for mixtures 1 - 5.

Mixture No. 1 2 3 4 5

CEM I Standard cement 421 446 - - -

Portland _Dansh _whi_{te ceme}nt - - 408 449 349

Construction cement - - - - -

Silica dust - - - - 39

Aardal sand 0 - 0.5 mm - 382 - 384 -

0 - 2mm 212 - 206 - 196

Un- 0 -5 ,, 213 - 206 - 196

Foam- treated 5 -10 . 75 - 73 - 70 glass in frac- 1C -14. 76 - 73 - 70 aggre- tion 14 . - 20. 76 - 73 - 70 gate 0.5 -2 mm

Treated - 83 - 83 - in frac0 -5 ,, - 83 - 83 - tion 5 -10 „ - 80 - 91 -

1C -14. - 80 - 91 -

14 -20. - 80 - 91 -

Scancem Damper 5,0 5.3 4.9 5.4 4.6

Water 297 245 266 238 306

Sink, cm 0,5 7 0 7 2

Air % 7,6 8.3 7.3 7.7 9.3

Density kg/m³ 1377 1485 1310 1485 1300

28 day compressive strength, MPa 17,4 23.5 20.5 24.8 22.3

1 year compressive strength, MPa - - 20.1 23.4 21.6

Length change in %o of 2 weeks -0.007 -0.022 0.015 0.010 -0.027 initial length at age 1 4 „ -0.023 -0.052 0.005 0.002 -0.061 day after subsequent 8 ,, -0.069 -0.076 -0.027 -0.021 -0.121 storage at 38 °C, 100% 12. -0.084 -0.091 -0.068 -0.018 -0.088

RH, for (period) 16. -0.107 -0.102 -0.070 -0.007 -0.062

26" -0.044 -0.039 -0.018 -0.010 -0.093

1 year 0.021 0.014 0.020 0.022 -0.040

2yrs 0.035 -0.031 0.066 0.067 -0.043

Weight change in % of 1 week -0.5 -0.34 -1.13 -0.66 -0.64 initial weight at age 1 2 weeks -0.5 -0.24 -1.21 -0.65 -0.59 day after subsequent 4 . -0.3 0.20 -0.92 -0.32 -0.46 storage at 38 °C, 100% 8 ,. 0.21 0.95 -0.20 0.43 -0.18

RH, for (period) 12. 0.81 1.65 0.28 1.04 0.17

16. 1.16 2.08 0.74 1.58 0.42

26. 2.03 3.07 1.66 2.59 0.88

1 year 4.15 5.37 3.97 5.31 2.01

2 vrs 6.84 7.90 7.46 9.07 3.76

Weight change (kg / 1 year 57.2 79.7 52.0 78.8 26.1 m³) after storage for:

2yrs 94.2 117.3 127.7 134.7 48.9 Table 2b: Compositions (kg/m³) and test results for mixtures 6-10.

Mixture No. 6 7 8 9 10

CEM I Standard cement - - - - -

Portland Danish white cement 367 - - - -

Construction cement - 365 432 326 359

Silica dust 41 - - 36 40

Aardal sand 0 - 0,5 mm 349 372 - 342

Foam- Un- 0-2mm - 184 - 184 - glass treated 0-5 „ - 184 - 184 - aggre- in fraction 5-10 „ - 66 - 65 - gate 10 - 14. - 66 - 65 -

14 - 20. - 66 - 65 -

0.5- 76 - 80 - 74

Treated ca.2 mm in fraction 0-5 . 76 - 80 - 74

5-10 ,. 73 - 80 - 72

10-14. 73 - 80 - 72

14 - 20 „ 73 - 80 - 72

Scancem damper 4,9 4.3 5.2 4.2 4.3

Water 277 245 240 306 287

Sink, cm 8 0-0.5 7 3 8

Air, % 9,0 5.5 8.4 10.6 8.0

Density, kg / m³ 1410 1310 1450 1235 1395

28 day compressive strength, MPa 24,5 18.7 22.1 19.1 25.1

Length change in %o 2 weeks 0.018 0.006 -0.027 -0.021 0.007 of initial length at 4 . 0.007 0.005 -0.050 -0.073 -0.024 age 1 day after 8 . -0.008 -0.062 -0.057 -0.073 -0.022 subsequent storage at 12. 0.017 -0.049 -0.064 -0.093 -0.030

38°C, 100%RH,for 16. 0.021 -0.042 -0.074 -0.101 -0.039

(period) 26 „ 0.006 -0.011 -0.046 -0.092 -0.028

1 year 0.057 0.031 -0.019 -0.073 -0.027

2yrs 0.055 0.059 -0.009 -0.022 -0.020

Weight change in % 1 week -0.27 -0.87 -0.40 -0.68 -0.47 of initial weight at 2 weeks -0.22 -0.58 -0.21 -0.62 -0.42 age 1 day after 4 . -0.11 -0.14 0.27 -0.47 -0.26 subsequent storage at 8 . 0.11 0.66 1.32 -0.08 0.05

38°C, 100%RH,for 12,. 0.31 1.40 2.33 0.35 0.36

(period) 16. 0.54 2.05 2.94 0.65 0.62

26. 0.82 3.47 4.47 1.29 1.11

1 year 1.68 6.28 7.98 2.87 2.29

2yrs 2.61 9.27 11.08 4.74 3.81

Weight change after 1 year 23.7 82.3 115.7 35.4 31.9 storage for, kg/m³ 2yrs 53.0 121.4 160.7 58.5 53.1 Table 3. Compositions (kg/m³) and test results for mixtures 11 - 13.

Foam-glass of type Hasopor™, available from Meraker, Norway was used for the tests. Table 2a and 2b shows that among the concrete mixtures based on Portland Cement type CEM I, only mixture No. 5, 6, 9 and 10 exhibited a weight increase of less than 5% after two years.

Two years test results were at the time of the application not available for the CEM II Portland Fly Ash Cement mixtures. The results up to 18 months exposure, however, clearly indicate that for mixture Nos. 12 and 13, with 5 and 10 % silica dust respectively, the weight increase will be significantly less than 5%, while for mixture No. 11, with no silica dust, the weight increase will exceed 5%. None of the mixtures show an expansion exceeding 0.04% (0.4%o). This is the requirement to a non-alkali reactive according to CSA A23.2-14A and the Norwegian guidelines published in "Norsk Betongforenings Publikasjon" (Norwegian Concrete Association Publication) No. 21-1996.

Still the alkali reactivity is a problem to all samples that do not contain silica dust, due to the accompanying weight increase. The reason that no visible, outer expansion can be observed, must be that there is ample room in the pores of the concrete to absorb the reaction products. The disadvantages related to a significant weight increase are obvious in relation to construction concrete. As an example, a bridge that has been designed to carry a certain load, will wholly or partially lose its capacity to a carry a load other than its own weight as this weight increases. As indicated by table 2 the weight increase does not stop after 1 year for concretes that exhibit alkali reactions and are to the contrary considerably larger after 2 years than after 1 year. The weight increase for the second year varied from about 38% to about 144% of that from the first year. Weight increase for longer periods than 2 years has not yet been verified. The properties of the silica dust are related to its pozzuolanic properties and other pozzuolanic materials than silica dust can be employed, such as fly ash and/ or finely crushed slag. The silica dust has the particular advantage that it is very fine grain, which implies that it is effective in smaller amounts than e.g. fly ash and other pozzolana.

Table 2 shows that the four samples (mixture Nos. 5, 6, 9 and 10) that include silica dust has a significantly lower weight increase than corresponding concrete mixtures without silica dust. All mixtures in tables 2A and 2B except Nos. 1 and 2 are based on low alkali cement, i.e. with a maximum Na₂O content of 0.6 % by weight.

Table 3 further shows that the two mixtures (12 and 13) that include silica dust, have a significantly smaller weight increase than the corresponding mixture (11) without silica dust. The said three mixtures are based on cement that includes about 17 % by weight fly ash.

Lightweight concretes with expanded glass has proven to be even more susceptible of alkali reactivity than concrete mixtures with ordinary crushed glass, and low alkali cement or fly ash cement (with up to 17 % fly ash) alone - i.e. with no pozzuolanic addition - is not sufficient to prevent unwanted weight increase in lightweight concretes that include crushed expanded glass aggregate. This conclusion is valid even when the finest fraction of glass is removed as shown by sample No. 4 and 8. All samples that include additions of silica dust show a weight increase of less than 5% after two years exposure. In general, the CEM I based samples with silica dust that do not include glass particles smaller than 0.5 mm, show a slightly lower weight increase (in the range from 2.6 % to 3.8%) than do the CEM I based samples with silica dust that also contain the finest fraction of glass particles (in the range from 3.8% to 4.7%). While the silica dust is effective already at concentrations about 10 % by weight of the cement and for some cement types already at concentrations about 5 % by weight, other pozzolana must be added to concentrations of 8-13 % by weight of the cement in order to be effective in low alkali cement based foam-glass concrete, and to at least 17 % if the foam-glass concrete is based on cement with a higher alkali content.

Portland cements with a low alkali content and/ or a high content of pozzolana are generally found to be suited for use in the lightweight concretes according to the invention. Among the CEM I class Portland cements, the ones with an alkali level (Na₂O) lower than 0.6 % by weight are best suited as they require smaller additions of pozzuolanic materials to prevent alkali reactivity. Also cements of class CEM II fly ash cement are well suited provided the fly ash content is high enough relative to the alkali content of the cement. The part of the cement's alkali level () that exceeds 0.6 % by weight, must not constitute more than 4 % by weight of the cement's fly ash content. Such CEM II cements are suited also in combination with suited CEM I type cements.

Preferred amounts of different types and mixtures of pozzuolanic materials in a lightweight concrete according to the invention, are disclosed by claims 3-9.

The method according to the present invention as claimed by claims 10-12, is particularly suited to determine whether or not a lightweight concrete is alkali reactive, which is not possible to determine from measurements of its expansion. In preferred embodiments of the invention foam-glass aggregates amounting to at least 30 % by weight and in some cases at least 40 % by weight of the fresh concrete are employed.

Depending upon the particular pozzuolanic material employed, the amount of crushed foam-glass included, the degree of comminution f the foam-glass and to some extent depending upon the kind of cement employed, the relative amount of the pozzuolanic material will vary from about 8 % to about 30 % by weight of the cement in the lightweight concrete. If silica dust is the sole or the substantial pozzuolanic material used(relevant for concrete of CEM I Portland cement), there will usually not be required to include more than about 20 % by weight silica dust. Typical amounts silica dust will range from about 8 % to about 20 % by weight, and more preferred from about 10 % to about 12 % by weight of the cement in the concrete. If fly ash is the sole or the substantial pozzuolanic material used, which may be the case for CEM I with separate fly ash addition and for CEM II with or without separate fly ash addition, the concentration will range from about 15 to about 30 % by weight. To a foam-glass amount of at least 30 % by weight of the fresh concrete, a fly ash amount of 17 to 20 % by weight will normally suffice. If the foam-glass amount is increased to at least 40 % by weight of the fresh concrete, the fly ash amount will more typically be in the range from 17 to 25 % by weight of the cement in the lightweight concrete. In practice the pozzuolanic material for a lightweight concrete according to the invention will often be a combination of silica dust and fly ash. At a ratio between the silica dust and fly ash from 1 :9 to 9:1, the total amount of pozzuolanic material, depending upon the type of cement and the particular glass fraction employed, will typically be within the range from 10 to 30 % by weight of the cement in the lightweight concrete and more typically in the range from 12 to 20 % by weight. Should finely crushed slag also be included in the pozzuolanic mixture together with silica dust and fly ash, then, depending upon the ratio between these three components, an amount of total pozzuolanic material of up to 50 % by weight may be required to prevent alkali reactivity, though more typically will be an amount in the range from 12 to 20 % by weight.

Claims

1. Lightweight concrete based on Portland cement wherein the aggregate is solely or partially comprised of crushed expanded glass (foam-glass), said concrete exhibiting a weight increase of maximum 5% in 2 years tested according to Canadian Standard CSA A 23.2-14A, characterized in that

- crushed foam-glass is included in an amount of at least 20 % by weight of the fresh lightweight concrete,

- the lightweight concrete comprises a pozzuolanic material that is included in an amount of at least 8 % by weight of the cement in the lightweight concrete, whereas - the cement component(s) in the light weight concrete are chosen among CEM I low alkaline cement (max. 0.6 % by weight Na₂0), CEM II fly ash where the alkaline (Na₂O) content exceeding 0.6%o by weight of the cement, constitutes less than 4 % of the fly ash content, as well as mixtures of such cements.

2. Lightweight concrete as claimed in claim 1, characterized in that the pozzuolanic material is chosen among silica dust, fly ash and crushed slag, as well as combinations these materials.

3. Lightweight concrete as claimed in claim 1, characterized in that crushed foam-glass is included in an amount of at least 30 % by weight of the fresh lightweight concrete and that a pozzuolanic material in the form of silica dust is included in an mount of from 8 to 20 % by weight of the cement in the lightweight concrete and preferably between 10 and 12 % by weight.

4. Lightweight concrete as claimed in claim 1, characterized in that the crushed expanded foam-glass is included in an amount of at least 40 % by weight of the fresh lightweight concrete, and that a pozzuolanic material is included in the form of silica dust in a relative amount of from 8 % to 20 % by weight of the cement in the lightweight concrete, and preferably of from 10 % to 12 %.

5. Lightweight concrete as claimed in claim 1, characterized in that the crushed expanded foam glass is present in an amount of at least 30 %by weight of the fresh lightweight concrete, and that a pozzuolanic material is included in the form of fly ash in a relative amount (included the fly ash of the cement) of from 15 % to 30 % by weight of the cement in the lightweight concrete and preferably of from 17 % to 20 %.

6. Lightweight concrete as claimed in claim 1 , characterized in that crushed expanded foam-glass is included in an amount of at least 40 % by weight of the fresh lightweight concrete, and that a pozzzolane material in the form of fly ash is included in a relative amount (included the fly ash of the cement) of from 15 % to 30 % by weight of the cement in the lightweight concrete and preferably of from 17 % to 25 %.

7. Lightweight concrete as claimed in claim 1 or 2, characterized in that a combination of silica dust and fly ash in a ratio in the range of from 1:9 to 9:1 constitutes the pozzuolanic material and that the total pozzolana content (incl. any cement fly ash) amounts to from 10 % to 30 % by weight of the cement in the lightweight concrete and preferably to from 12 % to 20 %.

8. Lightweight concrete as claimed in claim 1 or 2, characterized in that a combination of silica dust and finely grained slag in a ratio in the range of from 1 :9 to 9: 1 constitutes the pozzuolanic material and that the total pozzolana content (incl. any cement fly ash) amounts to from 10 % to 50 % by weight of the cement in the lightweight concrete and preferably to from 12 % to 20 %.

9. Lightweight concrete as claimed in claim 1 or 2, characterized in that a combination of silica dust, fly ash and finely grained slag constitutes the pozzuolanic material and that the total pozzolana content amounts to from 10 % to 50 % by weight of the cement in the lightweight concrete and preferably to from 12 % to 20 %.

10. Method for testing of long-term properties of lightweight concrete, characterized in that the relative weight increase of the concrete is measured under conditions and for a period that corresponds to aging for a longer period.

11. Method as claimed in claim 10, characterized in that the testing is conducted as an accelerated test, e.g. at 38 °C and 100 % relative humidity for a period of at least 6 months, preferably at least one year.

12. Method as claimed in claim 10 - 11, characterized in that the testing is conducted on lightweight concretes with foam-glass aggregates.