Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Sulfurous strategies

The Transport Research Laboratory has published new test methods for sulfur compounds and assessment procedures for structural backfill.Murray Reid reports.

Sulfur compounds in soils, rocks and fill materials can cause problems in civil engineering works because they attack materials such as concrete and steel. The attack is mainly due to sulfate ions in solution, and assessment procedures for the sulfate content of soils and groundwater have been available in BRE Digest 363 1and the Specification for Highway Works (SHW) 2for a number of years.

More recently, the role of reduced sulfur compounds such as pyrite has been appreciated, particularly where it is present as clusters of fine-grained crystals invisible to the naked eye (Figure 1). Oxidation of these compounds can generate large quantities of sulfate, which then cause further attack on construction materials. Oxidation of pyrite was implicated in the thaumasite form of sulfate attack on concrete bridge foundations on the M5 motorway 3, and the latest BRE guidance on concrete in aggressive ground, Special Digest 1 4, includes assessment procedures for pyritic ground.

The test methods which are generally prescribed for sulfur compounds are the water-soluble and acidsoluble sulfate tests in BS1377: Part 3: 1990 and the total sulfur test in BS1047: 1983. In these tests, sulfur is brought into solution as sulfate and determined gravimetrically by precipitation as barium sulfate. This is a lengthy procedure with poor precision, which is subject to interference from substances such as organic matter and iron oxides.

There is no standard direct method for reduced sulfur compounds. BS1377 recommends that, if pyrite is thought to be present, it should be determined as the difference between total sulfur by BS1047 and acid-soluble sulfate by BS1377. Both methods involve precipitation of barium sulfate and suffer from poor precision and reproducibility. The estimate of reduced sulfur will therefore be subject to large potential errors. In practice, very few analytical laboratories use the barium sulfate precipitation method or the BS1047 acid digestion procedure for total sulfur. A variety of in-house methods are used, leading to variation in results for the same material tested by different laboratories 5.Transport Research Laboratory (TRL) and the University of Sheffield have carried out a research project for the Highways Agency to develop better test methods for sulfur compounds and to develop assessment procedures for reduced sulfur compounds in backfill to structures.

The results have now been published as TRL Report 447 5(GE January 2002 news) and are summarised in this article. Throughout the project, liaison was maintained with the Thaumasite Expert Group and BRE, and the test methods and assessment procedures are compatible with those given in BRE Special Digest 1.

The new test methods given in TRL Report 447 are included in the list of recommended test methods in Table A1 of Special Digest 1.Together, the two documents give a comprehensive guide to test methods and assessment procedures for situations where sulfate attack on construction materials is likely to occur.

Test methods The new test methods have been developed building on existing water and acid extraction procedures and using advances in analytical techniques.Five tests have been developed and full details are given in the report.

l Test No 1: Water-soluble sulfur (WSS) l Test No 2: Acid-soluble sulfur (ASS) l Test No 3: Total reduced sulfur (TRS) l Test No 4: Total sulfur (TS) l Test No 5: Monosulfide sulfur (MS) Test No 1 This is based on the BS1377: Part 3 method, using a 2:1 water to soil extraction with the determination of soluble sulfur using inductively coupled plasma (ICP) - atomic emission spectroscopy (AES). The watersoluble sulfur (WSS) is converted to water-soluble sulfate (WS) as g/litre SO 4for use in the BRE and TRL assessment systems (see box).The test determines soluble sulfates and any soluble sulfides present.

Test No 2 This is based on the BS1377: Part 3 method, using dilute hydrochloric acid digestion but using ICP-AES to determine sulfur in solution.The acid-soluble sulfur (ASS) is converted to acid-soluble sulfate (AS) as %SO 4for use in the BRE and TRL assessment systems.The test determines sulfates such as gypsum, anhydrite, epsomite, mirabilite and jarosite, but does not determine insoluble sulfates such as barytes.

Test No 3 The method involves determination of reduced sulfur species by reduction with acidified chromium (II) chloride. The evolved hydrogen sulfide gas is trapped in acidified copper nitrate. The decrease in copper is determined by iodometric titration or determination of copper by ICP-AES, from which the amount of reduced sulfur is calculated.The results are expressed as %S.The test determines disulfides (eg pyrite), monosulfides and elemental sulfur, ie all the species that can be oxidised to produce sulfate.

Test No 4 Total sulfur is determined by microwave digestion of the sample using aqua regia, with determination of the liberated sulfur using ICP-AES. Alternatively, use of an appropriate Rapid High Temperature Combustion Analyser is acceptable, provided the instrument can reach temperatures in excess of 2000infinityC within a 40s analysis period.The analyser must be calibrated with standards that cover the range of sulfur content likely to be encountered in the materials being tested. The results are expressed as %S. The test determines all the forms of sulfur detected by Tests Nos 1, 2 and 3, and also unreactive forms of sulfur such as insoluble sulfates (eg barytes) and organic sulfur.

Test No 5 Monosulfides are determined by digesting the sample with hydrochloric acid.The monosulfides are liberated as hydrogen sulfide gas, which is trapped in copper nitrate and determined by iodometric titration or ICP-AES determination, as for Test No 3.The test can be combined with Test No 2 to yield both acid-soluble sulfur and monosulfide sulfur in the same test.The results are expressed as %S.The test determines monosulfide species such as machinawite, greigite and pyrrhotite, which are extremely reactive in engineering situations. These minerals would be included in the result for total reduced sulfur in Test No 3, but can be determined separately using this test.

Test Nos 1, 2 and 4 are recommended for routine use by commercial laboratories and for assessment of the potential to attack construction materials. Test No 3 shows considerable promise, but requires further work to be suitable for use as a routine test. Test No 5 is appropriate in special circumstances. Test Nos 1, 2 and 4 provide up-to-date standard methods, which should replace the variety of in-house methods in use.

Assessment procedures For structural backfill to concrete and metallic elements on highway schemes, TRL Report 447 gives assessment procedures and limiting values for sulfur compounds. The procedures take account of sulfur present as sulfate in the material and the potential sulfate that could be generated by oxidation of reduced sulfur compounds. The present guidance in the SHW only takes account of sulfur present as sulfate. The limiting values in TRL Report 447 are summarised in Table 1.For all other situations relating to concrete in the ground, the procedures in Special Digest 1 should be used.

Water-soluble sulfate In the SHW, a limiting value of 1.9g/litre SO 3is given for the water-soluble sulfate content of backfill to concrete, cement-bound material, other cementitious materials or stabilised capping. The same limiting values apply to any fill placed within 500mm of such materials.This limiting value has been retained, but has been translated to SO 4, which is the correct chemical formula for sulfate and is the form used by BRE in Special Digest 1 (see box).The limiting value becomes 2.3g/litre SO 4.For backfill to metallic elements such as corrugated steel buried culverts, reinforced earth elements and anchored earth elements, or any fill placed within 500mm of such materials, the limiting values are 0.25g/litre SO 3for galvanised steel elements and 0.5g/litre SO 3for stainless steel elements.These have been retained but translated to SO 4to give limiting values of 0.3g/litre and 0.6g/litre SO 4respectively.

Oxidisable sulfides Sulfate that can be generated from reduced sulfur compounds can be estimated by converting the total sulfur content to sulfate.This figure is known as the total potential sulfate (TPS) and is used in BRE Special Digest 1 as a measure of the aggressive potential of pyritic ground.This is a conservative estimate, as it includes unreactive forms of sulfur such as barytes and organic sulfur.A limiting value of 0.6% SO 4is given for backfill to concrete.A more accurate figure for the sulfate that could be generated is given by the oxidisable sulfides content (OS), which is the difference between total sulfur (TS) and acid-soluble sulfate (AS), expressed as sulfate. The limiting values are 0.46% SO 4for backfill to concrete and 0.06% and 0.12% SO 4for backfill to galvanised and stainless steel respectively.

Oxidisable sulfides can also be determined directly from the total reduced sulfur test (Test No 3). This method requires further development to be suitable for routine commercial use, but offers considerable potential for accurate, single step determination of reduced sulfur. The difference method using TS and AS suffers from the problem that any number derived from two other numbers will have a greater error than the individual determinations. However, the accuracy and precision of the new test methods is much better than that of the old BS1377/1047 methods, so the estimate of reduced sulfur is correspondingly much better. The AS and TS methods (Test Nos 2 and 4) are also rapid, suitable for automation and relatively cheap.

The application of the new assessment procedures to backfill to galvanised steel elements is illustrated in Figure 2 and summarised in Example 1. A similar flowchart and example for backfill to concrete is given in TRL Report 447. Recommendations for sampling, storage and the number of tests are also given in the report.

The limiting values for reduced sulfur compounds have been chosen to ensure that attack on construction materials will not occur. However, the limiting values only take account of the total amount of reduced sulfur and do not allow consideration of factors such as grain size, mineralogy and access to air and water that will determine the actual amount of oxidation in any given situation. As a result, the limiting values for OS and TPS are conservative, and may exclude materials that have performed satisfactorily as structural backfills. Examples would include materials where pyrite is present as large cubic crystals, visible to the naked eye, which would oxidise very slowly because of their low specific surface area, and materials with unreactive sulfates such as barytes present as cement or vein infill.

Where this occurs, enquiries should be made as to whether there is any history of corrosion problems with the material, and detailed chemical and mineralogical testing should be carried out using the new test methods. It is suggested that the use of the material as structural backfill may be permitted if it can be established that:

l The material has been used in the past as structural backfill without leading to problems with sulfur compounds; and l The reason why the material will not cause a problem is known, based on an understanding of its chemistry and mineralogy.

A number of amendments to the SHW and to various sections of the Design Manual for Roads and Bridges (DMRB) 6are suggested in TRL Report 447 to implement the new assessment procedures. Also, the opportunity has been taken to change all references to sulfate from SO 3to SO 4.The new test methods and assessment procedures in TRL Report 447, together with BRE Special Digest 1, present a comprehensive framework for the testing and assessment of sulfur compounds in soils, rocks and fill materials for civil engineering purposes. Further research on topics such as the rate of oxidation of pyrite is ongoing.

References 1 Building Research Establishment (1996). Sulfate and acid resistance of concrete in the ground. BRE Digest 363. Construction Research Communications, London.

2 The Highways Agency et al. Specification for Highway Works.Volume 1 of the Manual of Contract Documents for Highway Works.May 2001 Amendments.The Stationery Office, London.

3 Thaumasite Expert Group (1999). The thaumasite form of sulfate attack: risks, diagnosis, remedial works and guidance on new construction.Department of the Environment, Transport and the Regions, London.

4 Building Research Establishment (2001). Concrete in aggressive ground. BRE Special Digest 1. Construction Research Communications, London.

5 Reid JM, MA Czerewko and JC Cripps (2001).Sulfate specification for structural backfills.TRL Report 447.TRL, Crowthorne.

6 The Highways Agency.The Design Manual for Roads and Bridges.The Stationery Office, London.

Dr Murray Reid is principal researcher at TRL.The work described was carried out as a research contract for the Geotechnics and Ground Engineering Group of the Highways Agency.The views expressed are not necessarily those of the Agency.


1: Fill to corrugated steel buried structures, reinforced earth and anchored earth structures Present procedure 1 Determine water-soluble sulfate (as SO 3) by BS 1377: Part 3.

2 If greater than 0.25g/l (galvanised) or 0.50g/l (stainless steel), material is unacceptable.

Proposed procedure 1 Determine water-soluble sulfur (WSS) as %S by Test No 1 and convert to watersoluble sulfate (WS) as g/l SO 4from 15 x WSS (%S).

2 If greater than 0.3g/l SO 4(galvanised) or 0.6g/l SO 4(stainless steel), material is unacceptable.

3 Determine total sulfur (TS) as %S by Test No 4 and acid-soluble sulfur (ASS) as %S by Test No 2.

4 Convert TS to TPS (as %SO 4) from 3 x TS and ASS to AS (as %SO 4) from 3 x ASS.

5 Calculate oxidisable sulfides (OS) as %SO 4from TPS - AS.

6 If OS greater than 0.06% SO 4(galvanised) or 0.12% SO 4(stainless steel), material is unacceptable.

7 If WS and OS are less than or equal to the limiting values, the material is acceptable.

8 If OS is greater than the limiting values but the material has been used successfully in the past, seek expert advice, consider history of material and carry out detailed mineralogical and chemical testing using new test methods.

Forms of sulfur

Sulfur can occur as a number of forms in nature. There are also a variety of ways in which it is expressed in the scientific literature. Confusion can easily arise regarding the forms, amounts and units of sulfur and in comparing results from different sources.

Sulfate is reported as SO 3in many documents, such as the Specification for Highway Works and BS1377: Part 3. However, the correct chemical form is SO 4, and this has been used in the BRE Digests for a number of years. For consistency with BRE Special Digest 1, sulfate has been expressed as SO 4inTRL Report 447. Results in SO 3can be converted to SO 4by multiplying by 1.2.

To reduce confusion, a common set of symbols has been used in BRE Special Digest 1 and TRL Report 447, as shown below.

Symbols for use with sulfur compounds Parameter Units n Water-soluble sulfur %S 1 - WSS Acid-soluble sulfur %S 2 - ASS Total reduced sulfur %S 3 - TRS Total sulfur %S 4 - TS Monosulfide sulfur %S 5 - MS Water-soluble sulfate g/l SO 4- 15 WS Acid-soluble sulfate %SO 4- 3 AS Total potential sulfate %SO 4- 3 TPS Oxidisable sulfides %SO 4- TPS OS Notes WS can also be as %SO 4.The conversion factor is then WS (%SO 4) = 3 x WSS.

OS can also be determined from TRS.The conversion factor is OS = 3 x TRS.

Checks that can be employed on sulfur species include the following:


Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.