Les bactéries peuvent favoriser la formation de calcite. Leur utilisation au sein des matériaux cimentaires permettrait de colmater des fissures avec ce produit compatible et durable. L'objectif principal de cette thèse est de comprendre les mécanismes de la biocicatrisation des matériaux cimentaires afin d’accélérer la cinétique et maximiser le colmatage de fissures importantes (>200 µm). L'approche de biocicatrisation étudiée consiste à imprégner les fissures à l'aide d'un milieu de culture (milieu précurseur) contenant des bactéries. 3 aspects sont étudiés : les mécanismes abiotiques de cicatrisation (sans bactéries) des fissures de mortier, la croissance bactérienne in vitro, et la biocicatrisation des fissures de mortier. L'étude des mécanismes abiotiques porte sur l'autocicatrisation naturelle et sur l'influence des milieux précurseurs (urée, lactate de calcium et gluconate de calcium) sur la cicatrisation. L'influence d'une carbonatation initiale des fissures est aussi évaluée. L'étude sur les bactéries vise à suivre leur développement et à sélectionner le milieu de culture. L'étude de la biocicatrisation consiste à suivre la cicatrisation de fissures soumises aux traitements bactériens sélectionnés. Des fissures importantes (>200 µm) peuvent être réduites à 80 % par l'action des bactéries. Certains précurseurs ont un rôle important sur la cicatrisation de fissures fraichement ouvertes : en contact avec la pâte de ciment ils favorisent la formation d'ettringite qui comble partiellement les fissures. Il n'y a pas d'effet de ces précurseurs sur les fissures préalablement carbonatées. / Certain bacteria are able to promote calcite formation. Their use in the material allows the sealing of cracks by calcite, which is a compatible and durable product. The objective of this research is to achieve a better understanding of the biocicatrisation mechanisms in cementitious materials in order to accelerate its kinetics and maximize the sealing of large cracks (>200 µm). The studied bio-healing approach consists of soaking the cracks with a culture medium (precursor medium) containing bacteria. 3 aspects are investigated: abiotic healing mechanisms (without bacteria) of mortar cracks, in vitro bacterial growth, and bio-healing of mortar cracks. The study of abiotic mechanisms focuses on natural self-healing and addresses the influence of precursors (urea, calcium lactate, and calcium gluconate) on the healing process. The influence of the initial carbonation of cracks is also evaluated. The study of bacteria consists of monitoring bacterial growth and selecting the culture medium. The study of bio-healing aims to monitor the healing of cracks subjected to the selected bacterial treatments. Bacteria are improving the healing ability of the material. Large cracks (>200 µm) are reduced by 80 % through bacteria. Precipitation precursors have an important role in the healing of newly opened cracks: calcium lactate and calcium gluconate, in contact with the cement paste, promote the formation of ettringite, which partially fills the cracks. There are no effects of these precursors in previously carbonated cracks.

<p>To make bacterial-based self-healing concrete, alkaline-resistant bacterial spores, nutrient sources, and a calcium source are incorporated into a concrete matrix. Two ureolytic spore-forming bacteria, <i>Sporosarcina pasteurii</i>, <i>Lysinibacillus sphaericus</i>, and two non-ureolytic spore-forming bacteria, <i>Bacillus cohnii</i>, and <i>Bacillus pseudofirmus</i>, which have been used in previous studies as bacterial concrete healing agents, were compared in this study. The four bacteria were compared for their (1) sporulation rates on different sporulation agar plates, (2) growth in five liquid media, (3) survival rates in light weight aggregates (LWA) and in mortar samples, and (4) calcium carbonate precipitation rates from either calcium lactate or calcium nitrate. Sporulation was successfully induced after three-day incubation at 30°C on an appropriate sporulation medium. High sporulation rates of <i>B. cohnii</i>, and <i>B. pseudofirmus</i>(93% and 99% respectively) were found on alkaline R2A medium (AR2A). A sporulation rate (89%) of <i>S. pasteruii</i>was observed on tryptic soy agar supplemented with 2% urea (TSAU)<i>.</i>The highest sporulation rate (60%) of <i>L. sphaericus</i>was found on R2A medium supplemented with 2% urea (R2AU). In the growth study, tryptic soy broth supplemented with 2% urea (TSBU) was a positive control which supported rapid growth of all four bacteria. <i>Sporosarcina pasteurii </i>and <i>L. pasteurii</i>showed rapid growth rates in alkaline yeast extract broth (AYE) and yeast extract with 2% urea broth (YEU) respectively. In contrast, <i>B. cohnii</i>, and <i>B. pseudofirmus</i>grew poorly in all media except in the positive control. Viable counts of the four bacterial spores reduced (1.8–3.3 logs) during the first 24 h in mortar samples and then remained stable for next 27 days testing period. Among the four, <i>S. pasteurii</i>showed the smallest reduction of viable counts (1.8–2.5 logs) in mortar after one day of incubation. Both <i>S. pasteurii</i>and <i>L. sphaericus</i>showed high CaCO₃productions (>80%) after 24 h incubation at 30°C in YEU containing either calcium nitrate or calcium lactate. However, <i>B. pseudofirmus</i>and <i>B. cohnii </i>showed<i></i>low calcite recovery rates (<11%) in AYE containing either<i></i>calcium nitrate or calcium lactate under the same incubation condition. Overall, <i>S. pasteurii</i>was the best bacterial concrete healing agent of the four. This bacterium had (1) rapid growth rate in AYE, (2) about 90% sporulation rate within 3 days, (3) highest survival rates after 24 h in mortar samples and, (4) high CaCO₃precipitation rates, 82 or 98%, in broth containing calcium nitrate or calcium lactate respectively.</p><p>In addition, two different lightweight aggregates (LWA), expanded shale (ES) and expanded clay (EC), which were used as bacterial carriers and protective materials, were compared in this study. Each type of LWA was separated into three sizes (<0.85 mm, 0.85– 2.0 mm, and >2.0 mm) and immobilized with spores of <i>B. cohnii</i>or <i>B. pseudofirmus.</i>Viable counts recovered from EC and ES reduced <1.0 log after the immobilization process and remained stable during the 150 days testing period. Neither the type nor the particle sizes of the two LWA significantly affected the survival rates of the bacterial spores. This result showed that both EC and ES could be used as carriers for bacterial healing agents. It was also found that when the spores were immobilized with nutrients in LWA, their survival rates in mortar samples can be improved slightly (<1.0 log).</p><p><br></p>