Civil Engineering April 2022 | Vol 30 No 3

34 April 2022 Civil Engineering C arbonation usually results in a substantial loss of strength of a cement or lime stabilised material (as shown in Netterberg and Paige-Green, 1984; Page-Green, 1984 and 1991; De Wet and Taute, 1985; and Netterberg, 1987). Any such weakening – and especially if it is more than about 5 mm deep – is potentially highly detrimental to the performance of a road with a cemented base under a thin seal (Netterberg et al., 1987; Netterberg, 1991; Netterberg and De Beer, 2012). The indicator method and the confirmatory method each provide a rapid test to estimate the sensitivity of a particular mix to carbonation during a period of mellowing or curing and/ or between compaction, priming and sealing. Such prior information assists in the choice of both the stabiliser and the curing method. INDICATOR METHOD The indicator method involves the fol- lowing steps: 1. During the morning, mix the actual percentage of the proposed stabiliser to be used on the project with the minus 2 mm fraction of the project material and bring it to the estimated optimum moisture content for this fraction (see note 1). 2. Place the mix to a depth of about 20 mm to 30 mm in a shallow tin can or other suitable open container about 40 mm deep and 80 mm in diameter, so as to simulate the conditions to be evaluated (e.g. just loosely in order to simulate a windrow, compacted lightly in order to simulate uncom- pacted mellowing, or more heavily to simulate normal compaction), then treat accordingly. 3. Test a separate portion of the mix in at least three places with a few drops of phenolphthalein solution and then dilute hydrochloric acid (HCl) over the same place and depth and note the effects (e.g. no colour, pink, red (or purple/ mauve) or dark red (or purple/ mauve) with phenolphthalein; no, weak or strong fizz with acid (see note 2)). 4. For example, if it is desired to simulate the effects of poor water curing, subject the mix in the container to wetting and drying cycles as follows: spray lightly once or twice with water during the day, place in the sun and leave outside overnight to dry. 5. Repeat these daily wetting and drying cycles for the usual seven days. 6. After each cycle (i.e. the following morning) test a small part of the mix to its full depth in at least one spot with phenolphthalein and acid as be- fore, and then remove and discard the portion tested. Note the approximate depth of any change in colour and relative fizz. 7. As a control, subject a second spec- imen to similar water sprays, but keep it inside and cover the container after each wetting to prevent it from drying. If it becomes visibly wet, it need not receive further wetting. 8. Leave the container in the sun for a suitable period of time in order to simulate the likely delay between curing and priming and retest with phenolphthalein and acid as often as desired. 9. Retain the covered control inside and retest at the same times. 10. Repeat with any other stabilisers, mixes, and/or curing procedures and soils desired. This method provides a conservative, simple, inexpensive and quick indicator test requiring little material for the relative speed of carbonation of a par- ticular mix subjected to poor curing for seven days, i.e. spraying with water once or twice a day and drying out in between the sprays. CONFIRMATORY METHOD A closer estimate can be obtained by com- pacting the minus 37.5 mm fraction of the desired mix in California Bearing Ratio (CBR) moulds to the appropriate density to simulate construction and curing ac- cordingly (Netterberg et al., 1987). In this case the depth of carbonation is measured in a small hole made on a different part of the surface of the specimen after each cycle. It can also be done on specimens covered with sand before wetting, a primer, or a curing membrane. INTERPRETATION The fresh mix must turn dark red im- mediately. If it does not, then the stabiliser is too old or an insufficient amount has been added. The average depth over which it fails to turn red is reported as the depth of carbonation . The additional test with HCl is not es- sential. However, if the fresh mix contains no or only a little carbonate it will give no or a weak fizz, and a much stronger fizz after curing together, with a loss or diminishing of the red colour, it will confirm that it is due to carbonation. The indicator method is only intended to provide a quick comparison of stabilisers, mixes, procedures and/or materials and to indicate whether or not good curing is critical. As the mix is not properly compacted the results will be conservative. The confirmatory method yields results which agree closely with those found in practice. In general, it can be assumed that the layer will lose about half its strength over the depth of carbonation. Further details of how to test for carbonation and the presence or absence Pavement Eng i neer i ng Prediction of carbonation depth during curing Frank Netterberg Independent researcher: pavement materials and geotechnics frank.netterberg@gmail.com