Concerning the substrates (COMPO SANA and universal substrate), the clavicles buried in the COMPO SANA substrate (pH 4) revealed higher intense chemiluminescence reactions, both indoors and outdoors. As described by Vass [9], soil pH can influence intracellular chemical reactions and enzymes’ catalytic ability, affecting the microbial flora. Although it is not completely consensual through the scientific community, acidic pH seems to delay decomposition, as it inhibits bacterial growth, explaining the results obtained [10, 18]. Also, plants developed in acid soils seem to produce more tannins, which can outcome in a decrease in microbial activity [30].

In the sandy soil–buried clavicles case, few studies aimed to investigate this type of decomposition context. In this study, the bones buried in the sandy soils presented the highest intensity of chemiluminescence, both inside the greenhouse and in the open air. A study conducted by Guareschi and team [31], in a cemetery in Parma (Italy) presented a higher number of skeletonized remains in silty soils than in sandy soils, which supports our results. As one knows, soil texture is a fundamental parameter affecting the rate of decomposition, as it can influence water permeability and oxygen availability [30, 32], affecting the metabolism of microorganisms. Despite sandy soils having a certain degree of water permeability, due to their particle size, waterlogging can retard the decomposition process, since the persistent water produces a buffered environment, slowing the rate of decomposition [33, 34] As mentioned in the results section, vase no. 1 was waterlogged at the exhumation date, delaying the decomposition process, and explaining the obtained results.

According to Ceciliason et al. [11], the decomposition process requires moisture and moderate temperatures. The combination of a sandy soil with a low water content commonly promotes desiccation, generally assigned to the diffusion of gases through the soil matrix [35,36,37] (Inside the greenhouse, the “strong positive” chemiluminescence reactions in vase no. 5 can be explained by the high temperatures and total absence of water, which desiccated the remains, inhibiting the decomposition and resulting in its natural preservation.

Regarding the differences in the results between the clavicles buried inside the greenhouse or in the open air, the indoor decomposing bones showed higher intensity reactions than the outdoor ones. It is generally accepted that higher temperatures fasten decomposition. So, according to this parameter, our results go against what was expected. On the other hand, indoor decomposition tends to be slower than outdoor, due to the lack of humidity variations and access by scavengers and insects [11], which can explain our obtained results. As mentioned before, the complexity of PMI estimation relates to taphonomic factors, and how they can influence each other. In the greenhouse, temperatures were higher; however, the clavicles were protected from the action of rainwater, fauna, and flora, which lead to a slower decomposition [9]. All the outdoor vases presented vegetation development, in opposition to the greenhouse vases. According to Mant [38], the decomposition process tends to fasten when plants surround the remains, as additional bacteria and a layer of air between the remains and the soil are introduced into the burial [10].

Burned clavicles showed no chemiluminescence reaction, regardless of the decomposition context. These results can be explained by the destruction of the organic matter, and consequently the haemoglobin, by the burning in the electric muffle [39, 40].

The similar results obtained by the two different observers confirm the high interobserver agreement of this technique, reinforcing its value as a presumptive test for PMI estimation.

Our work was aimed at accessing the influence of temperature, humidity, soil type, and pH on time since death estimation through Luminol’s chemiluminescence technique in buried bones. To better understand the impact of these factors, we followed the methodological principles of taphonomy, designing an experiment with as few remaining variables as possible. However, in the practical realm of forensic science, in a real case, it is advisable to employ a combination of methods due to the complex nature of the decomposition process, in order to achieve a more reliable PMI estimation. The selection of the used methods depends upon the unique aspects of each case and the resources at the disposal of forensic experts, keeping in mind the constraints associated with each method. It is fundamental to acknowledge the importance of adopting a multidisciplinary approach, as the decomposition process can significantly fluctuate, influenced by environmental, individual, and anthropic factors [1, 4, 41]. According to the flow chart to be followed in case of late PMI estimation, described by Franceschetti et al. [41], the luminol technique can be used both when it is necessary to establish the forensic interest of the human remains, or when its forensic interest was previously established. Though, through our obtained results, this evaluation must be conducted carefully, considering the surrounding environment that influences the decomposition process.

The small size of the sample used should also be addressed. In our work it was not possible to increase its number due to complex logistic conditions, though, in future research a larger sample would yield more robust results, especially considering it is a subjective approach primarily involving the descriptive analysis of the results.

Remains of soft tissues in the bones after their extraction during the autopsy should also be a factor to consider, since it cannot faithfully replicate the decomposition of an entire body. Still, it is important to acknowledge the challenges associated with gathering human samples for the experimental process, making the use of an entire body impossible in our context. Nevertheless, replicating our study in a setting where the use of whole cadavers is feasible would be of great value.

When the inhumation of the clavicles occurred, the donors were already dead for a certain amount of time (the clavicles were preserved by freezing). Ideally, to test the effect of the influencing factors on PMI estimation as accurately as possible, the experiment should have been conducted using freshly collected bones extracted during autopsy. However, this was not feasible, as the experiment required all the mandatory conditions to be met, including the simultaneous burial of all the clavicles and the availability of a suitable area and the required material.