Scholarly article on topic 'A Critical Discussion of Salt Weathering Laboratory Tests for Assessment of Petrological Features Susceptibility'

A Critical Discussion of Salt Weathering Laboratory Tests for Assessment of Petrological Features Susceptibility Academic research paper on "Materials engineering"

Share paper
Academic journal
Procedia Earth and Planetary Science
OECD Field of science
{"petrological modelling" / "geologic materials" / "built environment"}

Abstract of research paper on Materials engineering, author of scientific article — Carlos Alves, Carlos Figueiredo, António Maurício

Abstract It is presented a discussion of several features of salt weathering tests with relevance to assessment of stone susceptibility to these rock-solutions interaction processes.

Academic research paper on topic "A Critical Discussion of Salt Weathering Laboratory Tests for Assessment of Petrological Features Susceptibility"

Available online at


Procedía Earth and Planetary Science 17 (2017) 324 - 327

15 th Water-Rock Interaction International Symposium, WRI-15

A critical discussion of salt weathering laboratory tests for assessment of petrological features susceptibility

Carlos Alvesa,17 Carlos Figueiredob, António Mauríciob

aLandS/Lab2PT (UID/AUR/04509/2013; POCI-01-0145-FEDER-007528) and School of Sciences, University of Minho, Portugal bCERENA - Centro de Recursos Naturais e Ambiente (UID/ECI/04028/2013), Instituto Superior Técnico, University of Lisbon, Portugal


It is presented a discussion of several features of salt weathering tests with relevance to assessment of stone susceptibility to these rock-solutions interaction processes.

© 2017 The Authors.Publishedby ElsevierB.V. This is an open access article under the CC BY-NC-ND license


Peer-review under responsibility of the organizing committee of WRI-15

Keywords: petrological modelling, geologic materials, built environment

1. Introduction

Laboratory accelerated alteration tests are a classical procedure to study water-rock interactions under controlled conditions. By definition, accelerated alteration tests use conditions that are different from those found in the field, in order to obtain, in a shorter time, the effects of the studied processes. Salt weathering is the main erosive process in the built environment (in the sense that it is the more widespread even if other process might achieve more intense effects). In terms of salt weathering tests, while several variations have been proposed, the prevailing standardized conditions are similar to those found in the European Standard 12370 ("Natural stone test methods. Determination of resistance to salt crystallization") that use immersion in sodium sulphate solutions. There is also another European Standard related to salt weathering (EN 14147 "Natural stone test methods. Determination of resistance to ageing by salt mist") using atmospheric deposition of NaCl spray. In this publication are discussed the issues of salt weathering tests that influence their use as a tool for comparative petrological assessment in the search for intrinsic features that might affect rock susceptibility to this weathering process

* Corresponding author. Tel.: 351253604300. E-mail address:

1878-5220 © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license


Peer-review under responsibility of the organizing committee of WRI-15


2. Discussion

The present analysis will be focused on the relevance for the assessment of rock susceptibility to salt weathering of the following main conditions: specimen geometry (size and shape), salt contamination conditions, drying conditions, stopping criteria and the recording and reporting of results.

2.1. Specimen geometry

In this section both size and shape will be considered. The question of size is linked to the well known "size effect" of laboratory tests. It has been referred1 that reducing the specimen size speeds up the decay test. Erosive effects could present decay patterns that have been typified2 as homogeneous, in the sense of following the specimen's contour and tending towards increasing rounding or heterogeneous. In the case of heterogeneous patterns, the effect of specimen's size on the stone assessment will depend on the way results are reported: if it is just reported the final mass variation (in relation to initial mass), the studied rocks could, in principle, either show increasing or decreasing results with increasing specimen size, depending on spatial frequency of the more susceptible features and the effects of sampling. However, other ways of reporting could minimize and perhaps even void this issue, as will be discussed further on. The presence of heterogeneities with dimensions similar to the specimen size could promote a greater dispersion of results3. In the case of rocks that can be considered homogeneous (when the size of heterogeneities is much smaller than the specimen's size and they have a statistically homogeneous distribution), and assuming that salts crystallization is concentrated nearer the surface, one will expect increasing mass loss with decreasing specimen size (as the affected zone will have an increasing proportional importance and since the capillary feeding reservoir will be smaller). Hence, it will be advisable to consider classic materials engineering recommendations of sample representativeness, using specimens where the smaller dimension is at least a certain factor greater than the larger heterogeneous feature present in the tested material. The ratio surface/volume is affected by the shape of the specimen. In the case of statistically homogeneous rocks, and assuming a given depth of salt crystallization, increasing ratios surface/volume will imply increasing importance of the affected zone. Additionally, this ratio can affect the drying conditions and their relation to the capillary feeding reservoir. However, geometric modelling could produce results that need to be assessed carefully. For example, for a parallelepiped of square base "b" and height "h", keeping "b" constant and increasing "h" will imply decreasing the surface/volume ratio. But in terms of salt weathering of building materials, surface decay is of major importance and increasing length might imply increasing probability of having at least one instance of a more susceptible feature.

2.2. Salt contamination conditions

Three main procedures have been proposed: immersion (e.g. EN 12370), capillary directional imbibition and surface deposition (as in EN 14147). Starting with this last procedure, its conditions simulate those where only the surface of the materials is exposed to pollutants and hence this test is useful in studying surface reactivity (e.g. stones exposed to seaspray) but does not assess the effects of bulk rock features. Capillary directional imbibition has been proposed for salt weathering tests4 as it reflects a common situation (capillary rising solutions - "wick effect"). However, it is necessary to consider that: a) the presence of heterogeneities can affect the extension of salt pollution on the specimen and a variable position of these features will imply a greater dispersion of results; b) the previous situation can exacerbate the size effect. Immersion conditions are not very commonly found in the built environment but they are not as strongly affected by the previous issues and will reduce the dispersion associated with the testing conditions. Additionally, in a perspective of assessment of petrographic features susceptibility, it is expected that immersion tests would produce more clear results in a smaller number of cycles (they will represent a conservative approach). They also test the surface susceptibility as crystallization happens at the surface (at least in the beginning of the test) but, in most circumstances, they might be harsher than surface deposition tests and, if only the final result is considered, the information regarding surface susceptibility might be lost.

2.3. Drying conditions

While extreme drying conditions are essential for an industrially useful test (that allows for testing batches of specimens in a relatively short time), it must be considered that using high temperatures will mean that some rocks could show changes related not to salt crystallization but simply to wet-dry cycles. The interest of the use of high temperature should be to promote salt crystallization inside the rock specimen, since this is a common situation observed in the field, and not to test unreal, in terms of built environment, wet-drying conditions.

" - * '^l&S

Fig. 1. Illustration of decay patterns on limestones after salt weathering tests (EN 12370): (a) homogeneous erosion on a grainstone specimen; (b)

heterogeneous erosive pattern on a travertine specimen.

While it can be considered that this implies the need for "blank tests" (without salts) under the same conditions, we propose that, generally, this will be not necessary as for most rocks the erosive effect of wet-drying cycles will be much less important than salt crystallization (but, for certain rocks, dry-wet cycles might have noticeable effects1,5). Additionally, recording evolution along cycles could help to assess this issue. There are studies showing that conditions of normal temperature and relative humidity can produce marked decay on porous limestones4,6.

2.4. Stopping criterion

This refers to ending a given experience before the possible final limit (in this case, the total failure of the rock specimen). In weathering tests one could consider two types of stopping criteria: i) a fixed number of cycles defined before the beginning of the tests and equal for all specimens (including experiments with only one imbibition-drying cycle6 or with simultaneous imbibition-drying4); ii) rules based on results obtained during the cycles of the tests. This choice will also have consequences on the point of the following section, as in the first case it will be reported the result of a certain parameter (or the evolution along time after the cycle6) while in the second one can report the number of cycles up to achieving a certain result, such as, for example, when after a certain number of cycles the variation in the considered parameter is less than a certain limit (reporting might also include the results of other parameters at that last cycle). Using a fixed number of cycles could "cut" the information from a test when the system has not achieved yet a given stability, namely the reported final result might not be very meaningful if test stopping occurs when there is a greater variation between cycles (indicating a marked trend towards failure). Following the example of other laboratory procedures (e.g. sample drying), it will be advisable to consider a stopping rule based on the variation between successive measuring points (cycles) but a more detailed study in trends will be necessary to propose a specific criterion.

2.5. Results recording and reporting

In general it is reported a "final result" of the test, either some physical property after a certain number of cycles or the cycle where a certain result is achieved (which can be also coupled with data on other parameters). The most commonly reported parameter is the final mass variation (as in EN 12370). Other physical have been proposed (such as, e.g., mechanical resistance7, wave propagation velocity7, pore-related properties7, surface irregularity6,8 and colorimetric coordinates8). All of them have the advantage of using instrumental measurements but some face questions of reproducibility (mechanical strength tests being destructive, which implies that different specimens will be considered at different cycles of the test) and relevance to the assessment (e.g. the relation between wave speed and degradation). There are also reports of parameters at several times after the salt weathering experiment (in this case a one cycle6). Considering the occurrence of heterogeneous decay patterns in some rocks, it will be advisable to use parameters that take in consideration intensity, extension and spatial distribution of erosion features. The use of advanced techniques such as laser imaging or X-ray tomography might help to calculate volume differences and

parameters related to space distribution of decay features but their application is onerous and time-consuming, especially for specimens with an appropriate size. There are also proposals for visual assessment but these proposals face the problem of subjectivity1. We suggest that it will be better to develop indexes that consider, separately, intensity, extension and spatial distribution, which might be combined in a final index value, as is done in Rock Mass Classifications for Engineering purposes (see examples in9). It has been suggested the interest of recording and reporting the results obtained in the different cycles, as this would allow the calculation of parameters representing evolution under salt weathering conditions (such as the alteration velocity2 or the initial and maximum weight increase10), the cycle where a certain result is achieved (as in the alteration index2) or assess possible trends with cycles number2,7,11. Additionally, for immersion tests, the observation of the initial cycles might give information on the surface behaviour in relation to the bulk behaviour.

3. Final considerations

We hope that the discussion presented here showed that in relation to the use of laboratory tests for assessing stone susceptibility to salt weathering: a) Immersion tests are to be preferred as they allow a more global and complete assessment of stone susceptibility; b) Specimen size and cycles number could be variable (but with minimum values) according to criteria related to stone heterogeneities and observed decay behaviour; c) Results reporting should not be limited to the final mass variation but also include parameters related to time evolution and decay intensity (ideally after each cycle); d) Decay intensity assessment needs to use categorical indexes for intensity, extension and spatial distribution to allow a more adequate comparison of rocks presenting heterogeneous decay patterns.


The Lab2PT, Landscape, Heritage and Territory Laboratory (UID/ECI/04028/2013) and the CERENA - Centro de Recursos Naturais e Ambiente (UID/ECI/04028/2013) are supported by the FCT - Fundaçâo para a Ciência e Tecnologia (Portugal), with Portuguese funds and funds from the European Union (FEDER, Programa Operacional Factores de Competitividade - COMPETE 2020 - Programa Operacional Competitividade e Internacionalizaçâo, POCI, P0CI-01-0145-FEDER-007528). These reflections were primarily promoted by experimental data obtained in the context of the Project PORENET (POCTI/CTA/44940/2002) funded by the FCT.

And also to several anonymous critical reviewers that contributed to our reflection on these issues.


1. Lubelli B, van Hees RPJ, Nijland TG. Salt crystallization damage how realistic are existing ageing tests?. Proc. SWBSS 2014, Brussels:Royal Institute for Cultural Heritage, 259-273.

2. Angeli M, Bigas JP, Benavente D, Menéndez B, Hébert R, David C. Salt crystallization in pores: quantification and estimation of damage. Env Geol 2007;52:205-213.

3. Alves C, Figueiredo C, Mauricio A, Aires-Barros L. Salt weathering of limestones: susceptibility of petrographical features (SEM study). Microsc Microanal 2013;19(S4):113-114.

4. Rodriguez-Navarro C, Doehne E. Salt weathering: influence of evaporation rate, supersaturation and crystallization pattern. Earth Surf Proc Land 1999;24(3):191-209.

5. Delgado Rodrigues J. Swelling behaviour of stones and its interest in conservation. An appraisal. Mater Construcc, 2001;51(263-264):183-195.

6. Diaz Gonçalves T, Brito V. Alteration kinetics of natural stones due to sodium sulfate crystallization: can reality match experimental simulations? Env Earth Sci 2014;72(6):1789-99.

7. Akin M, zsan A. Evaluation of the long-term durability of yellow travertine using accelerated weathering tests. Bull Eng Geol Env 2011;70(1): 101-114.

8. Vázquez P, Alonso FJ. Colour and Roughness Measurements as NDT to Evaluate Ornamental Granite Decay. Proc Earth Planetary Sci 2015;15:213-218.

9. Singh B, Goel RK. Rock mass classification: a practical approach in civil engineering. Amsterdam ; New York: Elsevier; 1999.

10. Alves C, Figueiredo C, Mauricio A, Braga MAS, Aires-Barros L. Limestones under salt decay tests: assessment of pore network-dependent durability predictors. Env Earth Sci, 2011;63(7-8):1511-1527.

11. Wells T, Binning P, Willgoose G, Hancock G. Laboratory simulation of the salt weathering of schist: 1. Weathering of schist blocks in a seasonally wet tropical environment. Earth Surf Proc Land 2006;31(3):339-354.