1. Introduction
In many fields of human activities scientists and other stakeholders are overwhelmed by the proliferation of scientific and technological publications as well as by the abundance of internet content whose validity is often difficult to judge. Another common circumstance is the “bypass and hyperbole in research” as stressed by Baveye (2021) for soil research. Moreover, science and technology are fragmented into specialized domains with the unwanted side effect of little inter disciplinary communication.
Gypsum (CaSO4•2H2O) is a common calcium sulfate mineral that by heating can easily transform into bassanite, i.e. the hemihydrate (CaSO4•½H2O) at approximately 130 °C, or into anhydrite (CaSO4) at approximately 180 °C. The hemihydrate, or gypsum of Paris, can be kneaded with water producing a plaster that sets in an exothermic reaction.
Interest in gypsum for the non-agricultural uses (i.e. building industries) stimulated pioneering research on its composition and relationships with bassanite and anhydrite. The research on these substances, their constitution and relationships can be tracked to the 18th Century, and have been investigated in different scientific domains like chemistry, physics, crystallography, spatial and Earth geology, and medicine. Moreover, the concern for multinational companies (Miah et al., 2018), about life cycle inventory within the construction industry will be aided by the assessment of gypsum content at the different steps of life cycle and recycling.
We reviewed a number of publications where we examined if the gypsum contents were determined in the materials involved. Many of the articles use sophisticated methods and expensive equipment to detect gypsum in these materials, but do not provide specific information about the quantity of gypsum in the materials. Often, the articles only provide vague or anecdotal data, like the name of the furnisher of gypsum, the quarry of provenance, or terms such as “commercial gypsum”, “natural gypsum”, “supplier's information”, “mainly composed by gypsum”, “calculated from sulfate content”, etc. Similar problems are noticed in the reviewed articles focused to modeling methods. The general rule in the examined references is the absence of a measurement of the quantity of gypsum in the materials used, obtained, or studied. In the few cases where the amount is provided, the method used is omitted.
Easy and reliable methods for gypsum determination are needed wherever the behavior of gypsum (CaSO4•2H2O) and its equilibria or relationships with bassanite (CaSO4•½H2O) or with anhydrite (CaSO4) play a key role. The study of gypsum and the other hydration phases of calcium sulfate is a widespread subject in geology (Van Driessche et al., 2019). Sulfates of calcium, however, are of paramount interest in an assortment of disciplines as illustrated by the references in Table 1.
Table 1. Examples of publications from 2005 to 2020 showing the interest of gypsum within various disciplines.
Many reports on the applications of gypsum adopt an experimental approach by applying some treatment to the studied gypsum product or by measuring the effects of adding different doses of some material containing gypsum to the soil, to industrial products, or to residues. Few of these reports, however, provide analytical data of the gypsum content of the added and or assayed materials in Table 1 and in other references in this paper. One reason is probably that most of the materials in the experiments are “earthen” or heterogeneous, i.e. of mixed composition, making the classical methods for gypsum determination in such materials cumbersome or requiring equipment typically not available in most labs. Our proposed analytical methods are based on gravimetry of: (i) the loss of water in the sample between 70 and 90 °C (Artieda et al., 2006), or (ii) the gypsum-bassanite phase-change (Lebron et al., 2009).
We also propose non-destructive determinations using equipment, which is more sophisticated, with the advantage of portability for on-site determinations. This is the case of field portable X-ray fluorescence (PXRF) spectrometry (Weindorf et al., 2009) improved by the increasing sensitivity of PXRF instruments (Weindorf et al., 2013). These methods meet the requirements of rapidity, affordably and easily implementation by technicians or students with only a minimal training.
2. Interest of gypsum in agriculture and soil science
Farmers, shepherds and otherers inhabiting gypseous lands are well aware of the peculiarity of the gypseous soils. The biological concept of gypsophily can be traced at least since Carl von Linné defined the Gypsophila genus in Species Plantarum in 1753 (Merlo et al., 2019). The pioneer agricultural use of gypsum in US is attributed to George Washington and Ben Franklin who applied gypsum to agricultural fields in the late 1700's (Sharpe and Cork, 2006). On the other hand, with few exceptions (Herrero, 2017), the gypsum-rich soils occurring in arid regions have been marginal for agriculture because of their low water holding capacity and nutrient retention. Gypseous soils are prone to dissolution and collapses under flood irrigation (Casby-Horton et al., 2015). Management problems in gypseous soils and the incorrect assumption that “gypsum produces salinity”, resulted in the paucity of gypseous soils research before the 2000's (Herrero, 2004).
At present, the worldwide demand for soil productivity includes those considered marginal because of their high gypsum (CaSO4•2H2O) content. Moreover, the rampant environmental concerns have broadened the attention to gypsum, including the pros and cons of flue gas desulfurization gypsum (Chen and Dick, 2011). In this way, the interest on gypsum-rich soils and sediments goes now beyond their consideration in soil science (Zoca and Penn, 2017).
3. Interest of gypsum in construction
Gypsum rock and its derived products have been used for building, sculpture and other applications for thousands of years. Stark and Wicht (1999) reported the use of gypsum plaster in Asia Minor at the 9th millennium BC and later also by Greeks, Romans and their ensuing cultures. Since Roman times, gypsum rock has been quarried and hewn in ashlars enduring for millennia (Fig. 1). One of its varieties, alabaster, has been appreciated and used for centuries (Kloppmann et al., 2017) for sculpting reliefs, statues, and other artifacts. Alabaster today is quarried in Aragón, Spain, for these purposes (Perrier, 2002). Additionally, alabaster is cut into translucent plates (few cm thick) for filtering the light in church windowpanes and other buildings (Fig. 2). Bustamante et al. (2020) studied several cases of alabaster ageing and degradation in buildings. To our knowledge, other varieties of natural calcium sulfates, e.g. those with transparence or mirror properties, do not have widespread industrial or commercial use.
Fig. 1
Fig. 2Rough pieces of gypsum rock have been used ―often reinforced and blended with burnt bricks or other materials― for building houses that have been inhabited for generations (Fig. 3). Gypsum stones are often combined with adobe, with tapia, i.e., rammed earth, or other materials to build barns or low-cost dwellings in rural areas. Moreover, gypseous soil horizons, like those pictured in California by (Hess 1910, Plate III-B, Fig. 2, and Plate IV-A), or in Mexico (Fig. 4), can be carved in several decimeters sized parallelepipeds. After a few weeks exposure to the sun, these blocks can be used for structural walls as it happens in rural regions of arid Mexico, including huts for explosive storage by artisanal miners. These uses provide evidence of the durability and the isolating properties of this material.
Fig. 3
Fig. 4The hemihydrate form for the building industry is obtained by heating the natural gypsum rock. The artisanal kilns ―as those illustrated by Sanz-Arauz (2017)― have been replaced by industrial kettles and other facilities to allow the massive processing of the quarried gypsum rock to produce the powdered hemihydrate. This process can be optimized by controlling the particle size, the dehydration rate, the air temperature and its humidity as shown experimentally (Stenström and Nilsson, 1992). This powder mixed with water produces the gypsum plaster that sets as dihydrate in an exothermic reaction whose kinetics is of industrial interest. Song et al. (2010) investigated the behavior of additives to gypsum plaster. With a similar approach, Seck et al. (2015) described the role of the layer below the free surface of the paste.
Gypsum mortars have been traditionally used wherever the gypsum rock is abundant in the landscape. The gypsum plaster is easily workable and has the advantage of requiring lower production temperatures than lime mortars. Years ago, gypsum mortars were used for consolidating to make the structural elements of buildings up to four stories (Vegas et al., 2010) and for their rendering (La Spina et al., 2013). The structural performance of these mortars and the effects of different proportions of water to gypsum when preparing the plaster have been studied by Vegas et al. (2012), or when adding wool or other natural fibers (Vegas et al., 2013). The reactivity of the hemihydrate is shown when the accidental exposure to humidity during transport or storage produces the undesirable caking of the hemihydrate powder as stressed by Misnikov (2018) who investigated some organic additives for inducing hydrophobia to protect the powder.
Today the most widespread use of gypsum in buildings is gypsum plaster, also known as plaster of Paris. Traditional and current use of gypsum plaster for sheathing the walls of rooms was expanded by the wide use of gypsum boards as facing materials for walls and ceilings and for the construction of solid walls with massive gypsum blocks (Ruff and Fischer, 2009). Current use of plasterboards is very cost effective compared to the lath-and-wet plaster method for wall and ceiling construction. Karni and Karni (1995) reviewed the main properties of plaster of Paris that are significant for building: strength, moisture absorption, thermal and acoustic insulation, fire resistance, etc. The hydration reaction kinetics along the calcium sulfates series, i.e. gypsum/bassanite/ahhydrite, is a key feature for the performance under fire of wallboards and other gypsum materials used in buildings. Kontogeorgos et al. (2011) mention the water of crystallization of gypsum and its absorbed free water as responsible of gypsum materials under fire. The experiments of Steau et al. (2020) concluded the superior efficiency in cost and fire performance of gypsum boards vis à vis boards of other materials.
Increasingly, gypsum from plasterboards is recycled throughout several cycles, as shown by Erbs et al. (2018). Of course, recycling is feasible as long as other mixed materials, e.g. metals, are separable or their content is low. The advantage of gypsum is that it fulfils the life-cycle based needs for environmental sustainability (Bjørn et al., 2020). Recyclability is a key plasterboard property for high volume construction despite minor contamination issues related to odor in landfill areas (Johnson, 1986). Landfill odor might disqualify gypsum plasterboard reuse/recycling if disposal and storage are made without the adequate mitigation (Laadila et al., 2021). Demolition waste (Kabirifar et al., 2020) is often expensive to process and leads to excess landfill accumulation of plasterboard. This inefficiency occurs through all management, as lifecycle stages: (i) preconstruction, (ii) construction and building renovation, (iii) collection and distribution, (iv) end-of-life, and (v) material recovery and production (López-Ruiz et al., 2020). Some properties of gypsum plasterboards were studied in detail with experimental and modelling approaches, as in the case of the references in Table 2.
Table 2. Examples of studies of gypsum boards properties.
| Studied properties | Authors |
|---|---|
| acoustic and thermic insulation | Granzotto et al. (2021) |
| fire-resistance | Stark and Wicht (1999); Kontogeorgos et al. (2011); Kubicka et al. (2019); Steau et al. (2020) |
| impact behavior in space research experiments | Yasui et al. (2020) |
| moisture buffering of indoor air | Maskell et al. (2018); Cascione et al. (2020) |
| porosity increase by foaming | Isern and Messing (2016); Ilina et al. (2020) |
| sorption of volatile organic components | Thevenet et al. (2018) |
| thermal and mechanical properties at high temperatures | Yu and Brouwers (2012); Pereira et al. (2016) |
Moreover environmental concerns led to the enhanced use of gypsum due to its lower environmental impact on both the production and/or the disposal/recycling (Arm et al., 2017) compared to mineral wools or other synthetic insulating materials (Geraldo et al., 2017). The environmental advantages of natural gypsum might not apply to the synthetic gypsum resulting from industrial processes, e.g. sulfur decontamination of gaseous effluents, or phosphoric acid production. These processes can concentrate radioactive elements (Labrincha et al., 2017) as well as heavy metals or other undesirable components (Chen et al., 2014) making their storage and disposal problematic (Tayibi et al., 2009).
4. Procedures for determining gypsum content
4.1. Preparation of samples
As stressed by Marcoen (1974), progress in differential thermal analysis and in thermogravimetry during the 1950–1970 period increased due to the industrial importance of the CaSO4•nH2O phases as key components of plaster. These methods and their sample preparation steps, however, require further investigation.
The lability of gypsum (CaSO4•2H2O) against temperature is a concern when preparing soils or any other analytes. The gypsum content in earthy or heterogeneous materials can be easily determined as shown in Herrero et al. (2020) under the prerequisite of not drying samples at temperatures >40 °C. Heating destroys gypsum because the release of its constitutional water results in transformation into bassanite (CaSO4•½H2O) and then into anhydrite (CaSO4), with a 16% and 21% loss of the initial gypsum mass, respectively. These fundamental reactions for the plaster industry were studied by Le Chatelier (1887) who mentions a note dated in 1765 on this subject by Lavoisier. Initiation of the release of constitutional water by gypsum is influenced by the relative humidity of the air (Davis, 1907). The crystallography of the hydration phases of calcium sulfate was early studied, e.g., Caspari (1936) along with Büssem and Gallitelli (1937). Table 3 displays several references for the temperatures at which gypsum starts releasing its constitutional water, giving ground to the statement of Wakili and Hugi (2009) on the “heat sinks” in the behavior of gypsum against fire. Hill (1937) showed that the solubility curves of anhydrite and gypsum intersect at 42 °C, well below the 66 °C established by Van t’Hoff.
Table 3. The dehydration temperature of gypsum after several authors.
| Mention of temperatures for gypsum dehydration | Authors |
|---|---|
| “… the loss of water from gypsum heated in a stream of dry air, is complete at 70°, …” | Davis (1907) |
| Fig. 1 by these authors shows the dehydration starting at ≈ 50 °C. | Weiser et al. (1936) |
| “… at temperatures in excess of 60 °C, a transformation takes place of the dihydrate to the hemi-hydrate, …” | Lagerwerff et al. (1965) |
| “At ≈ 60 °C it is observed that natural gypsum …. appear to have a very small amount of hemihydrate present.” | Hudson-Lamb et al. (1996) |
| Figures by these authors show dehydration starting at ≈ 50 °C. | Berenger et al. (2015) |
| “… the dihydrate may lose mass even at temperatures as low as 50 °C.” | Dweck et al. (2016) |
| “… the formation of hemihydrate from gypsum could take place at temperatures as low as 50 °C (Fig9)”. | Krause et al. (2020) |
Amorós et al. (1961) discussed the kinetics of gypsum dehydration and its relationship with the visual appearance of gypsum crystals at the soil surface. Paulik et al. (1992) presented a phase diagram of the system calcium sulfate/water and stressed the influence of the experimental conditions in the water loss of gypsum, by examining the topochemical and structural factors in the decomposition of CaSO4•2H2O. Wilson and Bish (2011) showed that liquid water is not necessary for the transformations bassanite/gypsum. The transformation of bassanite in to gypsum has been studied at nanoscale by Saha et al. (2012). Ritterbach and Becker (2020) stress that the kinetics of gypsum dehydration and bassanite rehydration are dependent of particle size, and have shown that bassanite can re-hydrate to gypsum with high relative air humidity of 95% without liquid water.
False records of anhydrite and/or bassanite can be obtained when observed under a microscope. Artieda (1996) warned about the dehydration of gypsum grains due to unintended heating by the polymerization of the resin used to impregnate soil blocks for manufacturing petrographic thin slides. Similarly, the avoidance of heating and the unsuitability of wet chemistry methods for gypsum content determination were discussed by Herrero et al. (2009).
The content of the de-hydrated calcium sulfates in stored samples may change temporally with portions of the de-hydrated sulfates transformed back into gypsum. These phenomena are controlled by the atmospheric relative humidity and temperature, as well that by the presence of other substances in both natural and industrial calcium sulfate materials. In their authoritative review of the calcium sulfate precipitation, Van Driessche et al. (2019) stated that bassanite is metastable not only in solutions but also in high humidity environments. The formation of hemihydrate can happen at temperatures as low as 50 °C (Table 3). Furthermore, the precise temperature must be considered jointly with the relative humidity (McAdie, 1964) at which the heating is conducted. Borrachero et al. (2008) studied the effects of the heating rate and the water vapor pressure on thermogravimetric analysis of the dehydration of gypsum, a basic condition in the method of Lebron et al. (2009).
4.2. Common analytical methods
The reviewed reference instruments for thermogravimetric analyses and X-ray diffraction work well for milligram-weight samples. In practice, such small sample weights compromise their representativeness for materials of mixed composition. These methods are suited for detecting gypsum but are inadequate for determining its % content in many of the materials used in the industry or involved in environmental studies. Moreover, grinding is the first step for these analyses, but the dehydration due to grinding may affect analyses of gypsum (Steiger, 1910). In the same manner, Molony and Ridge (1968) warned about the fabrication of bassanite by the mechanical grinding of gypsum. Herrero et al. (2009) stressed the undesirable loss of constitutional water by mortar and pestle grinding needed for X-ray diffraction, and pointed out the problems in the quantitative determination of gypsum using X-ray diffraction reported by several authors in a dozen of articles published from 1964 to 2003.
Few of the previously reviewed articles refer to the particle-size of the gypsum employed, e.g. Santos et al. (2020) stated “Particle size of the gypsum-based product was not determined nor is it shown in the producer's indications”. We did not find the mention of the practical impossibility of particle-size determinations by dry sieving due to the aggregation of gypsum particles by electrostatic charges. Additionally, there are dissolution and other issues when using wet methods (Casby-Horton et al., 2015) and similar concerns when applying methods involving laser-diffraction instruments (Herrero and Mandado, 2016). The special procedure for wet particle-size determination proposed for gypsum by Pearson et al. (2015) is time demanding and unavailable in most laboratories.
The non-destructive methods based on PXRF for appraising gypsum content need little or no previous preparation and are applicable in the field to mixed materials (Weindorf et al., 2009). The performance of PXRF analysis is greatly enhanced by its combination with visible near-infrared diffuse reflectance spectroscopy (Weindorf et al., 2016). Additionally, this method allows the simultaneous estimation of electrical conductivity and calcium carbonate that are often of industrial interest. The cost of the sensors would be recovered if a large number of samples are analyzed or if the additional chemical elements are also determined.
These methods were compared by Herrero et al. (2020) to those of the Artieda's and the Lebron's gravimetric laboratory methods. Importantly, these lab methods can analyze samples from 5 g to 20 g in weight so that they can achieve better representativeness for materials of mixed composition as those common in building industries. These two methods are affordable with the common chemical labs equipment, and can be conducted by technicians or students after a brief training.
A next refinement of the XRF methods would be testing the effects of the sample preparation procedures on the determinations. Goff et al. (2020) evaluated the geochemical investigation of seven elements, a key issue when judging the construction materials for their use, storage, reuse, recycling, and disposal.
In parallel with the above considerations, we note that soil gypsum may pose foundation hazards due to corrosion of concrete and iron by sulfate-charged waters. Another concern is the subsidence and catastrophic collapse of canals, dams and other infrastructures; a matter of study for foundations engineers. In both cases, the determination of gypsum content in the soil can help to detect these hazards.
5. Conclusions
There was a lack of analytical gypsum content determinations in most of the reviewed publications, many of them related to construction debris and rubble. This oversight may be an effect of the familiarity of most people with gypsum plus the lack equipment and trained personnel needed for classical thermogravimetric and X-ray diffraction methods. The herein discussed methods of gypsum quantification ―developed initially for soil science― were updated, summarized and compared by Herrero et al. (2020). Reasonably, these methods will be adequate and convenient for the construction materials with few, if any, adaptations. The purpose of the analysis and the availability of facilities and trained personnel will determine which methods to choose. As a rule of thumb, both Artieda's and Lebron's gravimetric methods would be practical for short series of determinations. Moreover both of them can be used for comparison with the sensor-based ones.
Weindorf's methods ―involving much more expensive equipment― would be advantageous for large scale field debris and rubble analyses of large sample numbers, for determinations in the field, or if a short time frame were required for the results.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We wrote this paper in the framework of the Letter of Intent between CSIC and TTU, with funds from the Spanish Research Agency (AEI) within the projects PCI 2018–09299 and H2020-MSCA-RISE-2017. Thanks to the anonymous reviewers and the Editor for their sound comments and suggestions.