We would like to sincerely thank the following for their support, advice and collaborations:
Hilde De Clercq, Lisa O'Hagan, Nigel Blades, Dr. Cony Decock, Kaat Sneiders, Marjolijn Debulpaep, Elke Otten, Elke Selter, Juliette François, Julie Desarnaux, José Vétro, Cederic Doutrelepont, Roxane Leemans, Jean-Albert Glatigny, Sébastien Pirghaye, Faisl Bousta, Ti-Phuong Nguyen, Soukaïna Ait Moussa, Denise Chanteux and IRPA colleagues.
1The historic heavy rain floods of 14 and 15 July 2021 in the south of Belgium (Wallonia) caused great human tragedy but material damage was also considerable. More than 106 monuments were affected and some 27 churches were flooded. After the waters receded, the humidity inside buildings was high. Direct and indirect damage quickly followed and the impact on heritage was important.
2Lessons had to be learned from these floods. Being able to respond better and faster in emergency situations to mitigate the impact on heritage is crucial. For the preventive conservation unit at KIK-IRPA (Belgium, Royal Institute for cultural heritage), focusing on the affected churches appeared necessary. The idea of designing a methodology with procedures that could be applied in a simple way soon arose. Churches are buildings holding cultural heritage, but in most cases, the parishioners in charge of management and maintenance have limited budgets, knowledge, and resources. They are not professionally organized and have daily operations that rely heavily on volunteers. Church buildings are important to the people and have a tradition of being a refuge in times of disaster. And yet, it was striking to see that, a year after the floods, many churches were still closed. To prevent theft, the buildings were closed to the point of limiting ventilation. Moisture and mould problems inevitably surfaced. Thus, the core of our research took shape: the RE-MOLD project would try to design methods to support the buildings managers with no specialized background so they can limit recurring mould development with low-cost, easy-to-find, eco and user-friendly products. The project was developed as a two-part study, in two stages, parts of a continuum included in the Chrisis project, between 2022 and 2024.
3Kaat Sneiders first studied the influence of climate on mold growth in seven churches in the Vesdre Valley, from the event in July 2021 to September 2023, testing methods to monitor mould growth and understand the link between climate, post-crisis management disruption and recurring mould growth. From then, Ariane Segelstein focused on finding ways to reduce the time needed for these buildings to return to normal use. For this, the focus was finally placed on maintenance of unpainted wooden furniture (pews, stalls, pulpits, organs, confessionals, wall paneling, etc.) impacted by mold in an environment remaining damp over the long term. This second study aimed to produce recommendations on the best way to remove active and recurring surface mould with methods that could be used by non-specialists such as parishioners in charge of daily church management.
4How can post-flood fungal growth be limited and made manageable? To answer this question, a better understanding of the events was necessary. What exactly had happened last year? What factors had controlled or influenced the mould development ? Or, at worst, exacerbated it?
5The first part of the project, led by Kaat Sneiders, relied on setting up a monitoring of mould development in several impacted churches during a year. Seven were finally selected: Chanxhe, church Saint-Nom-de-Jésus; Prayon (Trooz), Church Saint-Laurent; Fraipont (Trooz), Church Saint-Gilles; Pepinster, Church Saint-Antoine-Hermite and Apolline; Verviers, Church Notre-Dame-des-Recollets, Chapel Saint-Lambert and Church Saint-Remacle. One is along the Ourthe River (Chanxhe), all the others are along the Vesdre river. The aim was to compare how their post-crisis management evolved in relation to mould control. To do this, a timeline was drawn up for each church. In addition, structural aspects of historical buildings were researched, to better understand how exactly they react to water and moisture, an important aspect in the development of mould. Some of the buildings were also followed by IRPA’s Monument laboratory within their project of monitoring the drying of impacted buildings to design prevention measures and middle-term drying methods. Many churches were probably struggling with moisture and perhaps even mould problems before the floods, but with an exceptional flood such as the 2021 one, the mould growth exploded.
6All the information from the preliminary research was gathered into a summary table which was an important tool for this research. It allowed to compare the church buildings and to make a first synthesis of the events. It gave a first insight into the effect of the flood and the response of a building/collection to this flood. It also uncovered a possible relationship between the event, the structural aspects of the building and the actions taken after the floods versus the damage observed. Thus, could be uncovered some of the links between the mould development, the building-technical properties, the more or less abrupt drying and cleaning actions taken after flood, and, not unimportantly, the ventilation or not of the church.
7The second part of the preliminary research focused on measuring the degree of activity of the fungi. For this, two methods were considered:
-
- 1 Brokerhof van Zanen den Teuling, (1999), p.26.
The test kit developed by the Dutch National Archives, ARA-Kit DG18 (Agar)1. This allows to test if the mould is active and prone to develop. It can also be used afterwards to identify the type of mould.
-
- 2 Adenosine triphosphate is a universal energy molecule found in all animal, plant, bacterial, yeast (...)
- 3 The sample is brought into contact with a liquid that, in response to the amount of ATP present, em (...)
The ATP 2monitoring test, a research tool originating from the food industry. The instrument produces a figure (RLU figure) that gives information about the level of contamination and how 'dirty' a surface is.3
8In the situation under study, the presence of active mould was already a certainty. The type of mould was not a first issue. But the amount and level of activity of the recurring mould was more a concern.
- 4 0-30: no growth, 30-250 dormant spores, 250-500: sporulation and active growth, 500+ high level of (...)
9In 2022, Dr Nigel Blades of the National Trust, in collaboration with microbiologist Philip Skipper, developed a ranking and classification system for the numbers (RLU results) obtained with the ATP reader. The ranking had 4 levels4 ranging from “no growth” (level 1) to “high fungal growth” (level 4). This approach seemed a useful way to assess recurring growth after multiple cleanings in the churches under study. The ranking not only helped to evaluate the degree of fungal development per church, but also to map it over the months.
10In November 2022, first ATP tests were carried out, and, after some trial and error, the first usable tests were performed in March 2023. We conducted tests on wooden objects and on plastered walls. In the various churches, measurements reached sometimes figures above 1,000, a level that was not specifically included in the classification system of N. Blades and P. Skipper. We therefore decided to add an extra fifth level to their existing ranking in our measurements: namely: +1000, “very high fungal growth” (level 5). This extra level allowed to better differentiate results in the higher category of growth.
11The ATP tests allowed to accurately map mould development after the floods. It was particularly striking to see the strong positive effect of ventilation on mould activity during the final tests.
12How to sanitize churches invaded by mould as quickly as possible, to shorten the period of closure to the public due to health risks ? Conservation depends on the regular use of these places of worship and culture that are essential to resilience. We know to what extent the absence of human presence leads to looting and destruction, but also how quickly nature reconquers abandoned areas. It is therefore essential to shorten the time of disuse to safeguard vernacular heritage sites.
- 5 White vinegar or alcohol vinegar is a dilution of acetic acid. For example, at a concentration of 0 (...)
13Furthermore, in a context where the feeling of helplessness can be considerably discouraging, entrusting a specific intervention mission, however minimal, to the actors on the ground helps to improve the situation and generates motivation. Finally, as we have already experienced, the interest shown in local heritage by an institution such as the IRPA with its scientific researchers contributes to resilience because it awakens the pride of being in charge of a heritage with a much broader influence than the parish. This interest shown by the IRPA encourages local communities to become more widely involved in the conservation of their own heritage. After the floods of 2021, two main and generic recommendations were made to those usually responsible for the maintenance of churches to eliminate mould on non-polychrome wooden furniture. On the one hand, for historical heritage the use of ethanol in water at 70/30% or 50/50% by volume was recommended and on the other hand, for non-historical wooden objects household white vinegar was encouraged as it’s one of the best known and most widespread preservatives for food preservation, because it eliminates bacteria, yeast and moulds5 .
14It is recognized that ethanol at 70° is the most effective because it crosses the cytoplasmic walls of microorganisms' cells. Ethanol at 50° is less effective on moulds but it limits the risks of solubilization of polar substances, such as varnishes or dyes. The moulds considered are mainly toxic and allergenic. Their appearance is recurrent in post-flood contexts that maintain unsanitary conditions. As ethanol may dissolve varnishes and as vinegar is an acid product they’re not satisfying responses to mould growth in our context. This study aims to find concrete and practical curative solutions to implement, in a context where access to specific products can be complex due to supply problems.
15The wooden furniture in churches is traditionally made of oak. However, since the end of the 18th century, there has been the availability of northern red fir, which is less expensive and was often used in the 19th century. This conifer is used for areas that are not very visible, such as floors or drawer bottoms, or for restorations (Fig.1)
Fig.1 Furniture and mould. St Lambert Chapel in Verviers
On the left, oak in the very foreground and light-wood drawer in the background. On the right, oak confessional and conifer drawers placed vertically.
Credits: © CIPAR and IRPA.
16These woods are most often either without surface treatment (drawer bottoms and non-visible pieces of wood), or covered with a layer of hide glue tinted with walnut husk and a shellac varnish. The hide glue is sometimes burnished before being covered with varnish to enhance its shine. Traditionally, an application of hide glue at around 10%, tinted and heated, made it possible to homogenize the different shades of the wood and to block its pores. The hide glue is covered with a 20% shellac varnish (waxed or not) in ethyl alcohol. Waxes can often be found above the varnish due to maintenance operations.
17A distinction is made between woods attacked by fungi (brown rot, white rot, soft rot) and woods that support mold and blue stain. The presence of microorganisms on the surface of the furniture maintains a high humidity level on the wood and therefore favorable conditions for moulds that can degrade the surface of the wood. The attacks remain relatively limited because these organisms are content to degrade the nutrient reserves of the wood which are extractable and water-soluble components. These moulds do not degrade cellulose, hemicellulose or lignin6. Wood-eating fungi that attack the structure of dry wood itself develop in conditions of high relative humidity over the long term (roof leaks, water drops, wood exposed to the elements, capillary rise, etc.). In this sense, the prolonged presence of moulds is, among others, a risk factor for the development of wood-eating fungi.
18By active moulds, we mean moulds that are developing, forming hyphae and mycelia. Spores develop when conditions are favorable and change from an inactive dormant form to a developed active form. In the post-flood context, it is impossible to follow the usual recommendations of drying and then removing mycelia and residues with a vacuum cleaner equipped with a HEPA filter, a method favored in particular by restorers of furniture and archives.
- 7 Mould collections and laboratory supported by the University.
19The tests envisaged aimed at studying the possibility of acting directly on active moulds, while knowing that the exercise can be perilous since "disturbing" the moulds can cause the opposite effect by generating sporulation. The mycological part of our study was carried out with the help of Dr. Cony Decock, curator of the mycothèque7 of the Catholic University of Louvain-la-Neuve, who isolated and identified about ten strains collected by Kaat Sneiders. Among these, Penicillium glabrum was selected and cultivated in order to prepare the seeding sprays for the tests.
- 8 Woźniac, (2022).
- 9 Sterflinger et Piñar, (2013).
20Biocidal chemical treatments have been implemented since the middle of the 20th century, with highly toxic products, particularly in the 1960s and 1970s. Since the 2000s, many natural substances have been successfully tested for their antifungal and fungistatic properties and for their participation in the preservation of wood: from snail slime to beeswax, including plant extracts and essential oils. Similarly, mixed proposals based on natural and synthetic products have been tested, particularly for wood preserved outdoors, which is particularly exposed to rot fungi8. Conducting research for the protection of heritage today requires taking into account ecological concerns. Health and the impact on the environment must be considered as seriously as the protection of cultural heritage. The people to whom our recommendations are addressed are generally not trained in toxicological risks or in the handling of dangerous products. Compounds resulting from synthesis can cause degradation on non-target living beings and on the environment at different stages: during their manufacture, use and disposal. Faced with a complex and evolving ecosystem, our interactions can cause an imbalance whose repercussions we do not always measure. The example of sites such as Lascaux, where moulds have become increasingly resistant9 are cases to remember. Our recommendations for treating superficial mould will be limited in time since, according to observations made following the 2021 floods, and without new floods since, the sanitary condition of buildings improves in two to three years. The aim of this research is to propose solutions while waiting for a return to normal. Our aim here is to reduce the presence of microorganisms.
21Two directions can be considered for the treatment of active molds in our context:
-
Climate control and air filtration: The recommended treatment for the conservation of furniture, drying followed by filtered suction, is most often inapplicable. Church furniture, due to its dimensions, cannot be removed from the buildings, and if transport was possible for certain items, healthy spaces in the context of flooding are rare and generally intended for other emergencies. In churches, the humidity rising from the ground and emanating from the soaked building does not allow for localized and accelerated drying. The installation of dehumidifiers, air filters and heating is sometimes, if not often, impossible because these measures require conditions of access to the electricity network and running water which may no longer function. Essential services such as supplies, transport of equipment, provision of financial resources and security are also undermined.
-
Mechanical removal of superficial mould with readily available or easily prepared antifungals from local resources. Properly conducted damp wiping of smooth surfaces might limit fungal growth. Ideally, this wiping would be combined with climate control and air filtration, which are often difficult to achieve in a crisis situation as discussed above.
22We have automatically ruled out gamma radiation, fumigation or UV-C lamps, which are not suitable for the context. In addition, ethanol, generally used to eliminate mould, was also not retained because it dissolves shellac varnish.
23168 wooden test pieces (84 oak and 84 fir) were prepared to imitate the typical surfaces of the furniture of the churches studied. The two types of wood are oak and northern red fir (SRN for Sapin Rouge du Nord in French), also called Red Fir or Indigenous Fir. The test pieces are parallelepiped sticks approximately 10 cm long, 3 cm wide and 3 cm high for oak and approximately 2 cm high for SRN. 56 test pieces, 28 oak and 28 SRN, or 1/3 of the 168, were stained on one side with an aqueous walnut stain prepared by us from dried walnut stain (Fig.2).
Fig.2 Preparation of test pieces and the reference letters given
Preparation test on oak (top) and SRN (at the bottom). Starting from the left : varnished wood (Chene vernis CV, Sapin vernis SV), waxed wood (Chene cire CC, Sapin Cire SC), stained wood ( Chene teinte CT, Sapin teinte ST). The first set with red letters was stained twice. We opted for the lighter option to improve the visibility of the mould.
Photo courtesy of Ariane Segelstein.
- 10 The Baume du Gros Noyer is a mixture of waxes prepared and sold by the company Laverdure. It contai (...)
24112 test pieces, or the remaining 2/3, were covered with hide glue diluted to 10% in a warm aqueous walnut stain. Once this layer of glue had cooled and dried, we applied two coats of shellac varnish. 56 varnished test pieces were covered with a mixture of waxes called Baume du Gros Noyer10 . The wax was applied using a cotton cloth. About ten days later, the waxed test pieces were rubbed with a woolen cloth.
25So we had three types of test pieces :
-
a third, or 56, covered with walnut stain dye
-
one third, or 56, covered with tinted hide glue and shellac varnish
-
one third, or 56, covered with tinted hide glue, shellac varnish and wax
26For this study, we used wood available for purchase in Brussels, SRN from Finland and oak from Europe. The density of two samples of each wood was measured in weight/volume, expressed in Kg.m³. We subjected them to progressive heating up to 78°C and a relative humidity <30%. The samples were weighed before and after drying. According to standardized data11, fir is a light and soft wood, with about 550 density and a low hardness. Our calculations after drying two SRN samples (RH <30% and 78°C) indicate a density of 440. Oak is a heavy and hard wood with about 740 density and a medium hardness. The two oak samples were measured at densities of 770 and 840. The amount of free water in the wood for it to mold must be at least 18%12. At the time of setting up the test, oak contained between 22 and 27% free water, fir (SRN) between 55 and 64% (Fig.3, Fig.4, Fig.6).
Fig.3 Drying wood samples
The diagram above shows that it is in the first 24 hours that the wood loses most of its free water at 70 degrees C and 30% RH. Raising the temperature to 105 degrees in the last 15 hours did not remove any significant amount of free water.
Courtesy of Ariane Segelstein
Fig.4 Free water amount in woods when dried then soaked 24h
Before starting the tests, we immersed the test pieces in water for 24hours to achieve a minimum free water content of 18%. As the diagram shows, SRN absorbs much more water (ordinate) over time (abscissa) than oak.
Courtesy of Ariane Segelstein
Fig. 5 Timeline of the test
The tests lasted two months, from February to March 2024.
Document courtesy of Ariane Segelstein
Fig.6 SRN after 24h of immersion
Prepared SRN surfaces are not bleached or laminated after 24h of immersion, but depending of the direction of the cut deformations may occur.
Photo courtesy of Ariane Segelstein.
Fig.7 Oak after 24h of immersion
On the left, two samples of oak stained, varnished and waxed (left) and stained, varnished (right) not immersed. On the right, the same two samples immersed 24h then dried. We can clearly see that the varnish and wax films have an uneven surface after immersion.
Photo courtesy of Ariane Segelstein
27Day after, February 8, they was installed in seven boxes. The treatments consisting in removing the mould were carried out on February 15, February 29 and March 15. A final evaluation was done on March 29. Operation was carried out using absorbent wiping the wood test pieces with a tissue paper on which one of the anti-fungal preparations was sprayed (Fig. 8).
Fig. 8 while removing active mould during the test
Wipping mould from oak test pieces using absorbent tissue moistened with a caffeine preparation.
Photo courtesy of Ariane Segelstein
- 13 Observation notes and photos before and after treatment
28Each time the boxes were opened, the development of mould was documented13. We distributed the test pieces into seven twenty-two liters boxes made of colorless polyolefin with lids. Each box contained twelve oak test pieces and twelve fir test pieces, i.e. four test pieces of each preparation. Each box was treated with a different anti-fungal preparation.
29In each large box and next to the test pieces, a 3-liter container (same type of polyolefin box as the large one) contained a saturated solution of magnesium sulfate salts to regulate the relative humidity level around 94% at 25 °C. The saturated salt solutions were prepared at a rate of 337gr/L of anhydrous magnesium sulfate salts. We placed 500ml of solution in each container, left open inside the box (Fig. 9).
Fig. 9 Inside a test box
On the left the container for saturated solution salt, on the right wood test pieces. The wide box was closed hermetically with a lid.
Photo courtesy of Ariane Segelstein.
30The test pieces were numbered and named (Fig. 10) directly on a side face according to the following nomenclature:
|
C or S
|
T or V or C
|
1 to 4
|
1 to 6 and Witness box
|
|
Oak or SRN
|
Tinted, Varnished or waxed
|
Test piece number
|
Box number
|
|
Example: CT3-6, SV1-4 etc…
|
31Once the test pieces were placed in the boxes and the soluble salt solution was set up, the box was seeded a first time with a spray containing mould spores by 6 pressures distributed in each part of the boxes, then the tight lid was closed immediately. A strong silver adhesive tape from the TESA brand was placed at the junction of the lid and the box all around the perimeter improving the sealing of the box. The boxes were then kept around 25° to 22°C.
Fig.10 Identification of the test pieces
Picture took the 15th march 2024.
Photo courtesy of Ariane Segelstein.
32Among the samples taken at the Chanxhe church, Cony Decock identified Penicillium glabrum. This strain was chosen for its significant and rapid sporulation. Sprays of Penicillium glabrum in water with added Tween 60 were prepared as needed to inoculate the boxes. It should be noted that a strain of Trichoderma appeared in various proportions during the second treatment in all the boxes except one. On March 15, we were able to note that Trichoderma dominated in two boxes.
- 14 El-Hefny et al., (2019).
- 15 Tomiƈ et al., (2023).
- 16 Bezon, (2023).
- 17 Kumar et Pandey et Krist, (2023).
- 18 El-Hefny et al., (2019).
- 19 Antonelli et al., (2020).
33Natural, non-toxic and non-polluting, biocompatible, biodegradable and antifungal or fungistatic products were chosen. These products had to be easy to obtain and inexpensive. We tested a ready-to-use synthetic product formulated with lemongrass essential oil. Many publications report the antimicrobial, antibacterial and antifungal activity of certain essential oils. Since they are less cytotoxic than synthetic biocides, they present interesting alternatives to no longer expose the health of heritage stakeholders. However, certain drawbacks are noted and the implementation methods appear essential. EOs of oregano, thyme, lemongrass, peppermint, savory, palmarosa, chamomile, and so on14 15have been used to treat documents and cultural property. The use of thymol in fumigation causes alterations of cellulosic supports16, color changes on leathers17 are possible while certain essential oils can be used to preserve Carrara marble18or waterlogged woods19. Our neutral reference, called "Control" for these tests, is demineralized water. All the treatment mixtures used for the tests were prepared in demineralized water. For each of the antifungals, more information on the active substances, is presented below.
- 20 Marco et al., (2022).
- 21 Nathason, (1984).
34Origanum compactum essential oil is mentioned in several publications as being particularly effective as an anti-fungal on a wide spectrum of species, notably due to its high thymol and carvacrol content20 21.
35We prepared suspensions of 1% by weight of Origanum compactum essential oil in demineralized water. The suspensions were shaken before use.
- 22 Silva et al., (2022).
- 23 Marco et al., (2022).
- 24 Woźniac, (2022).
- 25 Broda, (2020).
- 26 El-Gamal et al., (2016).
- 27 Marco et al., (2022).
36Chitosan has been successfully tested to protect various materials constituting cultural heritage, such as stone22, metal, paper, wood, waterlogged archaeological wood and wall paintings23. While treatments of technical woods with chitosan, whose effectiveness is clearly proven to protect them from fungal degradation, are based on impregnating the wood by baths in solutions of at least 5% chitosan24, it is recommended for the conservation of heritage woods to deposit a layer of chitosan at 0.75% or 1% depending on the sources25, in 2% acetic acid, on the surfaces of sensitive, archaeological or historical woods to inhibit the proliferation of mould and fungi. This application can be done using a brush or a sprayer, depending on the condition of the wood surface26. The application of three layers of chitosan slows down the absorption of water and therefore the hygroscopicity of the support27.
- 28 Eikenes et al., (2005).
37An analysis was carried out with industrial chitosan flakes at a concentration of 2.9% in demineralized water by adjusting the pH to 5.5 with acetic acid28. Not finding it in stores, we used chitosan from the company Nature AZ, the closest geographically. The product was purchased online. We dissolved the chitosan powder in household white vinegar (acetic acid at 12%) 1gr/20gr and diluted the solution with 80gr of demineralized water to obtain a 1% solution. Product obtained is acidic, with a pH around 4. The product was stored at +4°C.
38Garlic has been used as a fumigant to treat Penicillium, Aspergillus and Fusarium in archival materials29.
- 30 Philippe et al., (2017).
39The maceration was made from 100g of crushed fresh garlic for 500ml of demineralized water (20%), left for 20 hours at a temperature of +4°C. The bottle was closed to avoid the loss of volatile substances30. The macerate was filtered to remove the solid parts. The measured pH was 6. The yield is 400ml for a mixture made from 1 liter of demineralized water.
- 31 Ghanmi et al., (2015).
40To highlight the possible synergy between different essential oils31, we combined three thirds of three essential oils. One third of the essential oil of Origanum compactum (tested alone as explained above), was combined with one third of Thymol Thyme and one third of Pelargonium graveolens. The choice of EO was made according to the molecules recognized for their antifungal action and their levels of Thymol, Eugenol, Geraniol and Carvacrol. The availability of the products in Brussels also influenced our choices.
41We prepared 1% by weight suspensions of an equal mixture of essential oils of Origanum compactum, Pelargonium graveolens and Thymus vulgaris thymoliferum in demineralized water. The suspensions were shaken before use.
42Syra-Oléovictis© has been tested on the Strasbourg Cathedral in France as mentioned by the supplier in its presentation brochure. Syra-Oléovictis© is currently being tested at IRPA by the monuments laboratory.
43The product is ready to use.
- 32 Nathason, (1984).
- 33 Šimůnková et al., (2021).
44Although we have not found any proven application in the context of cultural heritage conservation, caffeine has been successfully tested as a wood protectant, against fungi such as rot and mould and to repel termites and other insects. Treated woods (softwood and hardwood) have been bathed in 2% caffeine solutions in water32. Since caffeine is water-soluble and washable, it is recommended for use on woods kept indoors or protected by a roof. Depending on the wood, the treatment is more or less effective due to differences in wettability of the species and it seems that the treatment itself increases the wettability of the wood33.
45The caffeine, purchased from a natural cosmetics supplier, is dissolved in water, to provide a 2% aqueous solution. The powder needs to be first diluted into a small amount of demineralized water to avoid lumps. Then simply add the rest of the demineralized water to obtain the desired concentration.
46Demineralized water was used as a reference in the control box.
47To resume, we used this products with numbers for references : 1 Origanum compactum; 2 Chitosan, 3 Garlic, 4 Mixture of three essential oils, 5 Syra-Oléovictis©, 6 Caffeine, 7 Demineralized water.
48The assessment was carried out based on field observations and photographs taken systematically each time the boxes were opened. When time of systematic final assessment arrived, we used Photoshop to apply an accentuation filter to the photographs to improve the visibility of the mold. A grid was then applied to determine a percentage of coverage and proliferation on the surfaces (Fig.11a, Fig.11b).
Fig.11a, Fig.11b Improving mould visibility for assessment
Both photos are the same. An accent filter have been applied on the photo in the bottom, to highlight the distribution of mould. This process helped to assess mould coverage in percentage.
Photo courtesy of Ariane Segelstein.
49Prior to the general test, we determined monitoring RH with a sensor through that it took about two weeks, starting from a rate of 50 to 55% RH to reach 94% in the boxes, thanks to a saturated solution of magnesium sulfate salts.
50In the field, we observed a significant and exponential development of mould between February 8, the date the tests were started, and February 29, the time of the second treatment. In the weeks that followed, from February 29 to March 29, the mold coverage gradually decreased, raising questions about the efficiency of removing mould but also about the ability of our saline solutions to maintain sufficient relative humidity in the specific context as we had no RH monitoring inside the boxes.
Fig. 12 Timeline of the tests with focus on results
Document courtesy of Ariane Segelstein.
51From a general point of view, the northern red fir (SRN) pieces were more colonized than the oak. Mould developed sporadically on the unprepared and untreated faces of the SRN, which was not the case for the oak. First of all, it should be noted that the oak specimens that were only stained were practically not mouldy. The stained SRN pieces were slightly more sensitive to mould but significantly less than the varnished and waxed wood pieces. Mould developed significantly and systematically on the shellac-varnished wood, whether oak (CV chene vernis) or SRN (SV sapin vernis) (Fig.13a, 13b). The waxed pieces showed slightly more subtle results particularly on SRN. (Fig. 14a, 14b)
FIG.13a, 13b Evolution of mould on both varnished woods with seven treatments
On 29 February, all but one of the treatments achieved almost 100% coverage of the varnished surfaces tested.
Document courtesy of Ariane Segelstein.
FIG. 14a, 14b Evolution of mould on both waxed wood with seven treatments
The curves show the same trend as for the varnished woods (Fig.12 above) with a significant increase in mould between the start-up on 8 February (point 0 not shown) and the 29 February. Chitosan (treatment 2) may have special properties. The kinetics of treatments varied, probably in relation to their more or less preventive and curative effects. The percentages at 29 February represent contamination after 3 weeks, The decay gives an idea of the decontamination properties of the treatments.
Document courtesy of Ariane Segelstein.
5229 Feb : on oak, mould growth reach 100% or almost, unless for Chitosan and Origanum which never allow more than 75% of growth. Then, between 29 Feb. and 29 March : amount of mould decrease except for the mixture of EO. Showing there is less efficiency for the mixture than for the Origanum alone. Still between 29 Feb. and 29 March we mentioned that demineralized water did not stop the mold growth before 15th March, so, later than the others treatments.
5329 Feb : on SRN, only 3 treatments allowed 100% of growth. Treatments with Origanum, demineralized water and Garlic showed growth of 75 to 90%. Syra-Oleocvictis shows a radical decreasing of growth since 15th Feb, so earlier than the other treatments. If we compare, the waxed surfaces showed various mould developments according the preparation on surfaces and the products applied. Chitosan for example, on waxed surfaces allow mould growth on oak, but not on SRN. On their side, the varnished wood surfaces reached 100% coverage with no differences between different treatment products unless an exception of Chitosan.
54While some treatments seem to have sanitized the test pieces and their environment more than others, as explained above and as garlic and oregano were visually mould-free on 29 March, none of the preparations really distance themselves from the others. The preparations can show differentiated properties, for example being preventive on oak without being curative, as seems to be the case for chitosan (orange curve CV2 on Fig. 13b and 15 ) or being strongly curative without providing an immediate preventive effect as is the case for Syra-Oleovictis© (violet curves on Fig.14a, Fig.13b).
Fig. 13b, Fig. 15 Particular effect with Chitosan
Chitosan, (orange curve on the left), was preventive on varnished oak (dark green curve on the right) until 29 February. Contamination is slow until this date, then increase to reach almost 90% on 15 March.
Document courtesy of Ariane Segelstein.
55On the treated surfaces, the finishes have clearly deteriorated due to the initial immersion in water, but also to the successive treatments and to the colonization by moulds. Demineralized water which could seem to be the most neutral of the agents of our tests has proven to have a significant impact on the condition of the prepared surfaces. It made losses in the varnish (Fig. 17) and seemed to swell and remove partially the waxed surface on the SRN (Fig. 18)
Fig. 16 Assessment of general changes on the surfaces of test pieces after testing
|
EO Origanum
|
Chitosan
|
Garlic
|
Mix of 3 EO
|
Syra-Oleovictis
|
Caffeine
|
Demin. water
|
|
Varnished oak
|
+
|
+
|
+
|
+
|
++
|
++
|
+++
|
|
Waxed oak
|
0
|
0
|
0
|
++++
|
+
|
0
|
+++
|
|
Varnished SRN
|
++++
|
0
|
+
|
+
|
+
|
+++
|
++
|
|
Waxed SRN
|
+++
|
+
|
++
|
++++
|
+++
|
+++
|
+
|
The deterioration can take several aspects: more or less partial disappearance of varnish and wax layers, whitening, unevenness suggesting superficial damage to the wood.
Document courtesy of Ariane Segelstein.
Fig. 17 Surface of varnished oak treated with demineralized water
Condition on 29 march 2024. The prepared layer shows losses.
Photo courtesy of Ariane Segelstein.
Fig. 18 Surface of varnished and waxed SRN treated with EO of Origanum compactum
Condition on 29 march 2024.Surfaces are partially bleached and the varnish film seems to have disappeared.
Photo courtesy of Ariane Segelstein.
56Assuming that the relative humidity level did not drop under 65% at least until March 15, removing active mold by wiping the surfaces with damp absorbent paper, sanitized the environment in the boxes after the 2 first applications Feb 8 and Feb 15 and limited the development of mycelia. This was the case even though the smooth surfaces that are best suited to this operation have some crevices related to the structure and grain of the wood (Fig. 8). In other words, mold lodged in fine folds of the surface was not always eliminated by damp wiping. Passing damp paper over stained surfaces was less fluid because the wood is more absorbent and less smooth. Mould, when present, tended to attach to rough surfaces and it was practically impossible to completely remove all of it by wiping. The varnished and waxed surfaces altered by initial immersion and then by the cleaning products tested, gradually became closer to the state of raw wood. They became rougher and more uneven. The result of this degradation is that over the course of wiping at 15-days intervals, the wiping was less fluid. It took more passes with more pressure on the same area to clean it properly.
57The initial immersion of the test pieces opened the way for microorganisms by degrading the varnish and wax films. During a flood, wood finishing products are likely to capture and retain a quantity of water that will promote contamination. This free water within the wood maintained a humid climate in the confined environment of the boxes. The fact that we opened the boxes three times for periods of 30 to 45 minutes may have had an important impact on the RH level, dropping it under 65%. So, during the last 15 days of testing from 15 to 29/03, mould decreased significantly, even though the boxes was inoculated. There is no accurate answer to this question. But it can be considered that in the first part of the experiment, until 15/03 the results reflect the effectiveness of the different products applied twice.
- 34 Testimony collected at the time of our research in 2024, words of Sébastien Pirgaye, carpenter work (...)
58In churches that had been flooded, oak was never affected in its structure by mould, unlike SRN, whose surface was sometimes clearly degraded34. During the tests, more mould appeared on SRN than on oak. This is due, on the one hand, to the density and internal structure of the woods, which absorbed different quantities of water, and on the other hand, to the thickness of the test pieces, which were thinner for SRN. The tests allowed us to confirm that in the presence of water, the combinations of hide glue and shellac varnish on one side and hide glue, shellac varnish and waxes on the other side promote the development of mould, which is not the case with natural walnut stain. In the field, this means that raw oak or simply stained wood should not provide a support for mold if they have been carefully dusted.
59However, dusting waxed and varnished surfaces can greatly limit the potential development of mould by limiting the quantity of spores and by optimizing the drying speed of damp wood. This dusting can be carried out preferably before flooding or but can be done after and in the latter case, if vacuuming is not possible, dusting can be done with a damp cloth followed by direct drying with a clean, dry cloth to avoid swelling of the preparations.
60In the context studied, where drying moulds to eliminate them after, is unthinkable, wiping moulds on smooth surfaces, as they appear to capture them on a damp surface and confine them, allows at least to limit developments in accessible areas. Although garlic and oregano extracts seem to offer some positive results in the tests, more would be needed to confirm their impact, their relative efficiency and characterize the difference with the use of simple demineralized water. The common point remaining is that removal with wiping using a damp cloth can still be a reasonable temporary measure, provided the gesture is performed properly and the disposal of the wipes is done accordingly.
61However, the limitations of this method lead us to discourage wiping moulds on rough wood (drawer bottoms, panel backs, etc.) because in on these situations surfaces the spread mycelium would tend to attach to the wood and could sporulate. On these surfaces, we rather recommend to use a sprayed 70° ethanol solution as it remains the best solution to limit mold development.
- 35 Woźniac, (2022).
- 36 Nathason, (1984).
- 37 Woźniac, (2022).
62The question of preventive treatment on wood that will be exposed to flooding, of course, arises, beyond dust removal which alone offers an excellent barrier to contamination. This aspect deserves special attention in view of the results obtained with chitosan. The wood itself contains antifungal agents. The addition of an external agent must therefore be considered from the perspective of an overall synergy35 36. If caffeine and chitosan are recommended in the form of immersion treatments before the implementation of the wood, in our context it would be the repetition of passages with preventive products that could provide a reinforcement of the antifungal or fungistatic properties of woods themselves. Provided that there is no leaching, as can occur outdoors due to bad weather37. The sensitivity to water of the plant-based or animal-based products that we used guarantees at least partial reversibility. This means that repeated passes with damp paper would likely remove and deposit surface substances, limiting the risk of missing certain areas.
63Whether it is paper documents, walls or furniture in churches, the fact is that their maintenance, and sometimes even their cleaning, is often carried out by people who are not trained in conservation-restoration. Daily management in a normal or post-crisis context is the responsibility of people who have little or no training in the field. In these situations, the lack of skills in conservation-restoration and preventive conservation is problematic. However, for several years, Belgian cultural organizations, such as CIPAR (Inter-diocesan Centre for Religious Artistic Heritage) or PARCUM (museum and center of expertise on religious art and culture, recognized by the Flemish authorities), regional authorities responsible for the legal protection of cultural heritage and religious authorities have been supporting better knowledge and better practice of preventive conservation in churches. This improvement in taking into account the specificity of local heritage is encouraging. Further research and development of simple methods of care in emergency situations would be useful, given the diversity of materials and cultural objects preserved in churches.
64The aim of this project was to develop such a method for wooden furniture and objects that are not always considered as cultural heritage, but are nevertheless important for the use and identity of the church. Many wooden pews, pulpits or confessionals were made for the church in which they are installed. Their good conservation is therefore important. Unfortunately, during floods, these furnitures has often been considered secondary or dispensable. We took this situation as a starting point and explored ways to improve heritage conservation in this context.
65In the field, colleagues have found that dusting carried out by conservator-restorers does not lead to a resumption of mould, unlike the same dusting on a twin object carried out by non-specialists who are nevertheless very careful. This point is important because cleaning or dusting carried out by highly motivated and conscientious people, without in-depth knowledge of the materials and their behavior, can lead to damage. A conservator-restorer will know how to interrupt a mechanical action at any time because he will understand what he sees, unlike a neophyte. Church factories do not have the possibility of having their churches dusted by conservator-restorers, but they can consult professionals to move towards better consideration of the specificities of the heritage by the managers of these collections. Conservators-restorers can provide support and guidance to maintenance managers, train them by providing them with elements of understanding, warnings and methods of execution. This would also help establish a relationship of trust that would encourage exchanges between conservators-restorers on the sites. If each church local management, could call on at least one conservator-restorer on a regular basis, the conservation conditions of the monuments and their collections would be improved.
- 38 The video is kept at KIK-IRPA and illustrates the right gesture to do for wiping active mould.
66To mitigate the risks associated with emergency maintenance carried out without professional supervision, the tests carried out as part of this project propose the use of products and actions with limited impact, and are supplemented by a video, to better understand how to perform the correct wiping action safely and effectively38. The recommendation of products, simple to prepare and apply, remains subject to the support of a conservation-restoration professional. The same applies to basic dusting, which, if carried out at inappropriate frequencies and in an uncontrolled manner, can damage the collections. Our aim is to raise awareness among those responsible for the maintenance of churches of the limits and good practices. How to dust without abrading? What is the limit between dusting and cleaning? When should we stop and seek professional advice? It therefore seems important to us to focus on the real situation, on the needs and the material and social conditions and to find sustainable solutions in which all the actors will be recognized with their skills.
67
RE-MOLD was an attempt to lead innovative research from field-based observations and practical support to local communities in charge of heritage at risk. Born in a time of crisis, it was thus quite unusual but based on the idea that the expected rising number of crisis due to climate change may encourage to lead low-cost quick-paced applied researches for and within impacted sites, on emerging topics. The issue of recurring mould in damp environment challenges the limits of current preventive conservation approach. Advising to evacuate objects or control the climate are options to fight mould, but when this is not possible, for reasons linked to the community management of these items, more solutions need to emerge.
68It thus underlined that a whole category of cultural objects, unpainted wooden furniture, were not being assessed, maintained or restored to the same level as others in churches, and that many of them were being lost or are still at risk of being lost in the future. The status of these objects, seen partly as cultural heritage, partly as disposable pieces of furniture, deserved more concern. The local communities in charge of these objects, personally impacted by the floods and who felt sometimes abandoned after some months also deserved more support.
69For this reason, RE-MOLD was developed through a partnership in the field with CIPAR and the heritage services of the bishopric, to understand the needs of the local communities and to consider new solutions that they could handle directly in the middle term of the recovery phase. It also grew through close collaboration with the monuments laboratory at IRPA, and the technical support of the mycothèque in Louvain-La-Neuve University. Cross-disciplinary research has made it possible to approach the specific issue of managing recurring mold on non-polychrome furniture from a new angle. This research has been an opportunity to explore a theme that concerns all damp climates that are impossible to control.
70Finally, the project aimed at contributing to identify new approaches on the treatments themselves, whether it is on how to balance action between preventive and curative measures, or on how supports and treatments interact, or on how antifungal agents work. Research on fungal developments and their links with the climate should be done in the context of impacted sites, where the transmission of protocols adapted to the people in charge of maintaining churches may support the sustainability of tangible (buildings and their collections) and intangible (local culture and rites) heritage.
RSS feed
