SCM Research - Part 1

SCM Research - Part 1
Literature Review & Experimental Programme

Share |

In New Zealand there has been a 15 percent reduction in carbon emissions associated with Portland cement production since 2005 despite concrete production increasing by 13 percent during the same period.

This has been achieved by better process efficiencies at existing cement factories and by replacing less efficient production facilities. Kiln efficiency at local cement factories has been achieved by utilising waste materials such as biomass and tyres that reduce coal consumption and by adopting advanced process control measures.

Combining the good practices in cement kiln efficiency and process control with better use of SCMs could reduce New Zealand's carbon emissions by a further 15 percent by 2030. Achieving this target will require a better understanding of how industrial and natural SCMs can be best used in concrete construction. Tracking this projected shift in emissions will be relatively simple to measure given the clear manner that production levels are recorded with minimal wastage in the process.

SUPPLEMENTARY CEMENTITIOUS MATERIALS (SCMs)
SCMs are either sourced from industrial wastes such as fly ash, ground granulated blast-furnace slag (GGBS) and silica fume or from natural materials that are beneficiated into pozzolans such as pumicite, micro-silica, diatomite and calcined clays such as metakaolin. SCMs have diverse chemical and physical properties that affects their reactivity and interaction with Portland cement. The range of chemical compositions of cementitious materials is shown in Figure 1 below.

Figure 1: Ternary diagram showing chemical composition of Portland cement and SCMs

SCMs made from industrial waste such as fly ash, GGBS or silica fume are widely used in countries with a strong industrial base where these waste materials are generated. These materials have been used in concrete for more than 50 years and performance of blended cements in concrete is well understood although characterisation techniques are still improving with better understanding.

Natural pozzolans are primarily alumino-silicates or amorphous silica that react with hydroxyl ions in concrete to produce secondary pozzolanic reactions. Four main types of natural pozzolan may be defined and found in New Zealand:

Volcanic glass deposits of pyroclastic origin that includes non-consolidated volcanic ash and pumice with variable pozzolanic activity depending on their siliceous content and fineness (e.g. pumice and volcanic ash).
Amorphous silica from either biogenic or hydrothermal activity that form relatively soft clastic rock types such as diatomite or amorphous silica deposits (e.g. diatomaceous earth and amorphous silica).
Zeolites produced by lithification of volcanic glass deposits that are partially transformed from amorphous to crystalline but still retains pozzolanic activity due to their microporous nature (e.g. tuffs and ignimbrite).
Clay minerals such as bentonite or kaolinite that are very mildly reactive but can be made significantly more reactive when calcined at 700-800^oC (e.g. metakaolin or calcined clays).

The geology of New Zealand does have significant resources of natural pozzolans and clay minerals, which include the following:

Pumice deposits that can be ground to form pumicite that has been shown to have similar reactivity to that of fly ash.
Amorphous silica in the Rotorua region that have in the past been able to provide very high reactivity similar with silica fume.
Tuff and ignimbrite resources that are widespread and have reasonable reactivity as SCMs by being partially amorphous and microporous.
Kaolin deposits that when contaminated with iron and/or magnesium cannot be used as pigments and fillers but could theoretically be activated at moderate temperatures (e.g. 700^oC) to form metakaolin.
Other reactive silica sources such as diatomaceous earth that have shown promise when tested in the 1980’s and 1990’s but may be difficult to extract due to concerns about silica dust.

New Zealand geological resources are well documented 60 years ago by DSIR. Kennerley and Clelland from NZ DSIR investigated a wide range of natural pozzolans including rhyolite, pumicite, ignimbrite, andesite tuff and basaltic tuff. Testing included petrographic examination, chemical analysis, mortar and concrete tests of blends of cement and pozzolan. Reactivity of natural pozzolans in New Zealand was found to increase with fineness and glass content as shown in Figure 2 below.

Figure 2: Strength ratios of mortar made with 35% replacement of natural pozzolans

CHARACTERISATION OF SCMs
The quality of Portland cement is characterised primarily by the strength performance of mortar or concrete samples made from the material. The chemical and physical properties of Portland cement have a significant bearing on this hardened performance and modern cements tend to fall within a narrow range and the cement hydration reaction is predictable giving reliable and repeatable results. When SCMs are considered using a similar classification system, performance of concrete becomes more difficult to predict. The poorer correlation between strength activity index testing and concrete strength for SCM concrete is due to the following:

Portland cement has consistent density values whereas SCMs may be considerably lighter than cement such that replacement levels cannot be made based on weight but should be done by volume.
Classification tests are generally done on mortar mixes where no chemical admixtures are used to compensate for the higher absorption of some pozzolanic materials, which leads to variable water/binder ratios during the test.
Testing age is typically at 28 days, which tends to favour more reactive SCMs and often means that more slowly reactive natural pozzolans that gain more long-term strength are excluded.
Mortar cubes are cast at relatively lower consistence levels (almost dry mixes) that means that some binder combinations are difficult to compact properly and result in excess void contents in hardened mortar.

EXPERIMENTAL PROGRAMME
An experimental program was developed in 2020 to investigate the how best to utilise SCMs in concrete construction, with the following key concerns:

Better classification systems for potential SCMs to complement existing methods.
Assessing fresh properties of SCM concrete including bleed, setting and workability.
Strength development and durability performance of SCM concrete.
Benchmarking of typical performance using industrial SCMs such as fly ash or slag.

To undertake this experimental work the following materials were investigated and provided a wide range of reactivity and performance (see also Figure 3):

Two sources of GP cement for standard concrete mixes.
Two sources of HE cement for higher strength mixes.
Two types of fly ash (ASTM Class F and Class C).
Blast-furnace slag from Australia.
Perlite and natural pozzolana from New Zealand.
Calcined clay using moderate grade kaolinite (55 percent) from New Zealand.

Figure 3: Portland cement and SCMs used in research programme

Research was started on this experimental programme at the University of Canterbury in June 2020 and the results of this testing are discussed in Part 2 of this series, which focusses on how best to characterise and classify SCMs so that the performance in concrete can be accurately predicted.