Sulfur can occur in soils, rocks and fill materials.The most common forms are sulfates such as gypsum (CaSO 2). High concentrations of sulfate in groundwater can lead to attack on construction materials such as metals and concrete.Mere presence of sulfates may not be the only cause -it can also occur because of oxidation of sulfides during the disturbance resulting from construction activity (Figure 1).
Recent problems with the thaumasite form of sulfate attack on concrete bridge piers were ascribed to oxidation of pyrite in the Lias Clay backfill as a result of exposure during construction.
This resulted in an increase in the amount of sulfate available to attack the concrete (Thaumasite Expert Group,1999).
BRE Digest 363 (Building Research Establishment, 1996) and the Thaumasite Expert Group (1999) give guidance on the protection of concrete from sulfate attack. BS 8006 (1995) and the Manual of contract documents for highway works give limiting values for backfill to metallic reinforcing elements for reinforced earth, anchored earth and corrugated steel buried structures. The guidelines are usually given in terms of watersoluble sulfate and other chemical parameters such as pH and chloride content. The importance of sulfides in corrosion is now more widely recognised, and limits for total sulfur or sulfide are given in some systems.However, all the classification schemes have problems in the methods routinely used to measure sulfur compounds.
Existing test methods for sulfur compounds in soils have drawbacks. Most consist of an extraction step followed by determination of sulfur gravimetrically by precipitation of barium sulfate. This is a slow process that has poor reproducibility and suffers from interference from a number of compounds, including organic matter.
Another problem is that there is no direct test method for reduced sulfur compounds such as pyrite.The method suggested in BS 1377: Part 3 consists of determining total sulfur by the BS 1047 method and acid-soluble sulfate by BS 1377: Part 3 and assuming the difference is due to sulfide.
Both methods use gravimetric determination by precipitation of barium sulfate, so the resulting value for sulfide is the difference between two numbers with large standard errors. The figure will also include any inert forms of sulfur such as barytes and organic sulfur, so will tend to overestimate the amount of sulfide in the material.
Poor sampling and storage techniques often mean that analytical results bear little relation to the situation in the ground. If samples are stored at room temperature in an aerobic environment, significant oxidation of sulfides can occur before the samples are tested, leading to an increase in the amount of sulfate (see box). Conversely, in organic-rich samples reduction of sulfate to sulfide can occur.Both reactions are catalysed by bacteria, which are abundant in most soil samples.
In an attempt to overcome these problems and produce a better system for estimating the corrosion potential in soils and rocks, the Highways Agency commissioned the Transport Research Laboratory and the University of Sheffield to develop new test methods for sulfur compounds and derive appropriate limiting values for structural backfills.
The project, which began in March 1998, focused on backfill to metallic elements.While it was at an early stage, the thaumasite form of sulfate attack was discovered in bridge piers and foundations on the M5 motorway. The Thaumasite Expert Group was set up by the Department of the Environment, Transport and the Regions to produce interim advice and guidance.The two projects liaised, and the scope of this one was widened to include testing of a number of samples from the thaumasite investigations using the new methods.
The new test methods were developed taking into account recent advances in analytical techniques. In particular, advances in inductively coupled plasma, atomic emission spectroscopy (ICP-AES) mean it is now possible to determine sulfur directly in solution without having to use the gravimetric method. This not only greatly increases the accuracy and reproducibility of the method, it is much faster.
Advances in fusion techniques have lead to a number of potential methods for determination of total sulfur, and new extraction techniques mean it is now possible to determine reduced sulfur directly rather than by taking the difference between total sulfur and sulfate.
A four-part system for the comprehensive determination of sulfur compounds in rocks, soils and fills is proposed comprising:
Test 1: Water-soluble sulfate-sulfur (SWSS)
Test 2: Acid-soluble sulfate-sulfur (SASS) and monosulfide (SMS)
Test 3: Total reduced sulfur (STRS)
Test 4: Total sulfur (STOT )
This is based on the BS 1377: Part 3 method using a 2:1 water to soil extraction with the determination of soluble sulfate using ICP-AES.
This is a development of the BS 1377: Part 3 method for acid-soluble sulfur using reflux digestion with hydrochloric acid and tin (II) chloride under an inert atmosphere. Sulfates are dissolved in the acid and the sulfur is determined by ICP-AES. Monosulfides are evolved as hydrogen sulfide gas.
This is passed through acidified copper nitrate solution, where it precipitates as copper sulfide. The sulfide is determined as the difference in copper content between the solution before and after the test, determined using ICP-AES. In the BS 1377 method, the hydrogen sulfide is not collected and no determination of monosulfide is made. Monosulfides such as FeS are rare, though they occur in reducing environments such as bogs, marshes and mangrove swamps.They are highly reactive and pose a particular threat to construction materials.
This test includes disulfides (eg pyrite), monosulfides and elemental sulfur, which is extremely rare. It therefore includes all the species that can be oxidised to produce sulfates. The method involves reduction of the sulfide species under anoxic conditions using acidified chromium (II) chloride. The evolved hydrogen sulfide is trapped in acidified copper nitrate solution, which results in the precipitation of insoluble copper sulfide. The difference in copper before and after the test is determined using ICP-AES and used to calculate the amount of sulfide present in the sample, as in Test 2.
Total sulfur is determined by microwave digestion of the sample using aqua regia, with determination of the liberated sulfur in solution using ICP-AES. The new tests determine sulfur directly. Where the sulfur is present as sulfate, the results have to be converted using standard conversion factors. The new test method for total reduced sulfur can be used to estimate the potential water-soluble sulfate that could be generated if all reduced sulfur was oxidised. This can then be compared with existing limiting values for water-soluble sulfate. Under the new system, three situations have to be considered for each proposed structural backfill material:
1. Is the water-soluble sulfate greater than the existing limiting value?
2. Is the potential water-soluble sulfate greater than the existing limiting value?
3. Is the sum of the water-soluble and potential water-soluble sulfate greater than the existing limiting value?
If the answer to either question 1 or 2 is yes, the material should be classed as unacceptable as structural backfill. If the answer to questions 1 and 2 is no, but that to question 3 is yes, then advice should be sought from a geochemical specialist. Enquiries should be made as to whether there is any history of corrosion problems with the material. It may be appropriate to carry out detailed testing on the material, using the new test methods, to ascertain more clearly its potential to cause corrosion.
This approach has been used to prepare amendments to the Manual of contract documents for highway works and the Design Manual for Road Bridges.Use of this procedure for corrugated steel buried structures, galvanised reinforced earth elements and galvanised anchored earth elements is given in Example 1 and shown in Figure 2.The same procedure would be used for stainless steel, but with limiting values of 0.50g/litre instead of 0.25g/litre.
The Highways Agency is now reviewing the draft final report for the project.The final document will be published as a TRL report later this year and the amendments to the MCHW and DMRB implemented. Liaison is continuing with BRE, which is revising Digest 363 in the light of the Thaumasite Expert Group report. The new test methods and assessment procedures will help to reduce the incidence of corrosion problems due to sulfur compounds such as in Figure 1.
British Standards Institution (1983).BS 1047 Air-cooled blast furnace slag aggregate for use in construction.BSI, London.
British Standards Institution (1990). British Standard BS 1377: Methods of test for soils for civil engineering purposes, Part 3: Chemical and electro-chemical tests.BSI, London.
British Standards Institution (1995).BS 8006 Code of practice for strengthened/reinforced soils and other fills.BSI, London.
Building Research Establishment (1996). Sulfate and acid resistance of concrete in the ground. BRE Digest 363. BRE, Garston.
Design Manual for Roads and Bridges (1998).The Stationery Office, London.
Hawkins A B and G M Pinches (1987). Cause and significance of heave at Llandough Hospital, Cardiff - a case history of ground floor heave due to pyrite growth.Quarterly Journal of Engineering Geology,20,41-58.
Manual of Contract Documents for Highway Works (1998).The Stationery Office, London.
Sandover BR and Norbury DR (1993). Technical Note: On the occurrence of abnormal acidity in granular soils. Quarterly Journal of Engineering Geology,26, No 2,149-153.
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.
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. copyright Transport Research Laboratory.
Example 1: Fill to Corrugated Steel Buried Structures, Reinforced Earth & Anchored Earth Structures
1.Determine water-soluble sulfate (as SO 3) by BS 1377: Part 2.
2. If greater than 0.25g/litre (galvanised) or 0.50g/litre (stainless steel), material is unacceptable.
1.Determine water-soluble sulfate-sulfur (SWSSWSSTRS3POTRTRS3POTR3POTRWSS3POTRWSS) as %S by Test 1 and convert to g/litre SO 3(12.5 x S).
2. If greater than 0.25g/litre (galvanised) or 0.50g/litre (stainless steel), material is unacceptable.
3. Determine S by Test 3.
4. Calculate potential sulfate (g/litre SO) from 12.5 x (%S).5. If SO greater than 0.25g/litre (galvanised) or 0.50g/litre (stainless steel), material is unacceptable.
6. If SOplus Sless than 0.25g/litre (as SO 3) (galvanised) or 0.50g/litre (stainless steel), material is acceptable.
7. If SOplus Sgreater than 0.25g/litre (as SO 3) (galvanised) or 0.50g/litre (stainless steel), seek expert advice, consider history of material and carry out detailed testing using new test methods.