Thermal performance and ageing effects to model the life cycle assessment of heat-protective thermal insulation materials in pipe systems

Insulfrax blankets (Fig. 1b) are lightweight needle-felted blankets made from alkaline earth silicate (AES) wool. The material composition includes:(m%): SiO₂ 61,0–67,0; CaO 27,0–33,0; MgO 2,5–6,5; Al2O3 < 1,0; Fe₂O₃ < 0,6. These ceramic fibre blankets come in various roll widths, thicknesses, and densities, representing the next generation of low bio-persistent materials. Usually, they do not contain organic components that could contaminate high-temperature furnace atmospheres. Ceramic fibre blankets have the following characteristics: low λ, excellent resistance to thermal shock, and low heat storage capacity. These are important because the heat must be kept inside the transporting fluid in the pipe. Storing heat is not necessary for piping. Being entirely inorganic, they maintain strength, flexibility, and good thermal properties in various working environments without emitting smoke. Insulfrax S blankets are known for their exceptional handling, thermal shock resilience, high-temperature stability (up to 1200 °C), remarkable flexibility, and superb sound absorption. They are frequently employed as high-temperature insulation materials thanks to their superior thermal insulation capabilities and strong chemical stability. Typical applications include heat shields, mold covering insulation, pipe, sewer, and chimney insulation, passive fire protection, nuclear power plants, metallurgical industries, hypersonic vehicles, and channel linings for cogeneration and power plants. This is due to their unique structure, which gives them several benefits like high porosity, low density, large specific surface area, excellent thermal insulation performance, and thermodynamic stability performance [23,[32], [33], [34], [35]]. They can reach a temperature range of up to 250 °C in these conditions.

SEM was used to examine the surface morphology of the materials, identifying any microscopic alterations resulting from the heat treatments. A scanning electron microscope (LV-SEM, JEOL IT500HR) was used to gather surface morphological data from the samples that had been prepped and heated. The microscope ran at a 3 kV acceleration voltage to prevent charge accumulation. Utilizing such a modest acceleration voltage can primarily prevent the charging effect. Depending on the kind of material, the beam’s penetration depth is generally between ten and hundreds of nm. This depth is usually in the range of 100 nm for SiO; the comparatively shallow penetration depth enabled the collection of more surface data. Secondary electron (SE) detection was used for topographic imaging of the materials with nanometer lateral resolution. A compositional analysis using a JEOL energy-dispersive X-ray (EDS) system with a 0.1 % detection threshold. The accelerating voltage was increased to 15 kV during the EDS measurement [36,37].

The HFM aimed to determine any changes in thermal conductivity due to heat treatment. The λ of the materials was measured using a Holometrix Lambda 2000 heat flow measuring device. This device provides accurate results with a relative error of about 5 % or less. The instrument operates according to ASTM C518 and ISO 8301 standards. Thus, as mentioned before (section 3.1) the materials were sized according to this dimension. To determine the λ coefficient of thermal insulation materials, the standard MSZ EN 12667:2001 was used as it applies to materials with medium and high λ resistance of at least R = 0.5 m²K/W. The calibration material used was a glass fibre standard tested by the US National Institute of Standards and Technology (NIST). Calibration ensures a measurement accuracy of 5 % or better. When there is a proper proportionality between the UΔx/ΔT and λ, the λ factor for an unknown specimen can be determined from the measured U and ΔT values, knowing the specimen’s thickness.

Most thermal insulation materials have an amorphous structure and contain a variety of oxides, including amorphous silica nanoparticles and silicon oxide. Due to the rise in temperature, the amorphous structure may reconfigure, crystallize, or recrystallize [38,39]. The crystallization process may compromise the materials’ ability to act as thermal insulation, which can also cause an increase in heat conductivity. Thus, DSC measurements were conducted to analyze potential changes in the material’s internal structure related to thermal transitions and stability under elevated temperatures. The technique assesses impurities and possible changes in the sample following heat supply.

Visual inspection revealed that the colour of both Insulfrax (Figs. 2a to c) and mineral wool samples (Figs. 2d to f) darkened after thermal annealing, as shown in Figs. 2a to c. This colour change is likely due to the samples’ surface oxidation. This discolouration can be explained by the fact that the heat treatment took place in atmospheric air. Fig. 2a (insulfrax) and 2d (mineral wool) belong to the un-annealed samples, while 2b (insulfrax) and 2e (mineral wool) and 2c (insulfrax) and 2f (mineral wool) show the annealed ones at 150 °C and 250 °C for 1 day, respectively. One can see from the rows that the darkest surface belongs to the samples annealed at 250 °C caused by the intensive surface oxidation (2c and 2f). The surface of the heat-treated samples after annealing them at 150 °C showed only a slight discolourization.

Fig. 3d reflects the energy disperse element analysis, with the most contaminants of: Ca, C, O, Mg and Al, complete with silicon as a base material of the fibres.

It is visible in Fig. 3e by 1400× magnification that the mineral wool fibres are covered with the grains. The restructuration of the fibres is observed after thermal annealing. Fig. 3f and 3g present the SEM images of the mineral wool samples. It is also noticeable here that the granules and particles connected to the fibres fall.

Fig. 3h shows the contamination analysis, reached by energy dispersal element mapping. Mineral wool is based on silicon-oxide and is filled with C, Na, Mg, Ca, K, Al, Ti, and Fe. As a comparison of the SEM results of the two materials, one can state that the surface modification procedure due to the thermal annealing is the same for both materials. Thermal annealing makes the fibres’ restructuration and the grains visible.

The graphical representation of insulfrax in Fig. 4 (indicated by white squares) demonstrates that the measured λ is stable, maintaining an almost constant value until the heat treatment at 150 °C. Following heat treatment at 250 °C, the λ increases by approximately 24 %. Furthermore, the measured λ of the untreated sample aligns with the value provided by the manufacturer. SiO2 particles were identified on some of the glass fibres for non-heat-treated samples. Like aerogel-containing insulators, these particles flaked off due to heat treatment. However, the quantity of these particles is minimal, and their impact on the heat conduction properties is negligible.

To put it all in a frame, we can mention that insulfrax had lower thermal conductivity in all cases 0.025–0.031 W/mK, compared to the mineral wool (0.038–0.422 W/mK). However, the thermal conductivity of the mineral wool was continuously increasing (3 % and 9 %), and the thermal conductivity of the insulfrax jumped only after thermal annealing at 250 °C, but by about 24 %. For the tests with insulfrax, a thickness of 2.6 cm was used, while the mineral wool samples with 4.4 cm in thickness were used. The reason for this was the availability of the samples on the market. These two thicknesses are typical. Here, it has to be mentioned that based on the EDS analysis of the samples (Fig. 3 d and h) mineral wool contains more metallic components such as Fe and Ti, which can result in a higher thermal conductivity. Moreover, it can also result in higher thermal stability and less change (9 %) in the thermal conductivity after thermal annealing.

Similar results can be found presented in Fig. 5b for mineral wool. Starting from room temperature, a similar half-gaussian shape is observable, but the peak is at 250 °C. The change in the specific heat capacity is about 1250 J/kgK. The monotonic increase in the DSC sign can similarly lead to the rise in thermal conductivity presented earlier. Increasing C_p value might belong to crystallization.

Thermal conductance and thermal resistance are essential concepts in thermodynamics, thermal engineering, and heat transfer that characterize a material’s or system’s capacity to carry heat and the resistance it provides to the heat current. By adjusting these characteristics, engineers can regulate temperature gradient, avoid thermal shock, and increase thermal system efficiency. Thermal resistance allows two to compare the thermal insulation capability of different materials. It depends on the specimen’s thermal conductivity and thickness/diameter. The thermal resistance equals the heat loss ratio and the temperature difference. If we suppose a cylindrical shape (heat transfer through a multilayered pipe) with external and internal fluid flow, the thermal resistances can be calculated by using Eq.(2), (3), (4), (5), where R_i and R_e are the surface heat transfer resistances between the solid body and the flowing fluid, moreover, α_i and α_e are the internal (index:i) and external (index:e) surface heat transfer coefficients (12 and 5800 W/m²K), moreover d in cm is the diamater.

(2)(3)

To compare the thermal insulation capability of the tested materials, we calculated a possible heat transfer rate (q in W/m) value through a steel pipe with d_e = 25 mm and d_i = 20 mm diameters, hypothetically. We calculated the heat transfer rate without insulation using Eq.6 and in insulated cases by using the lambda values of the insulfrax and mineral wool measured after thermal treatment by using Eq.7. We suppose that the pipe transports hot water/steam fluid with different temperatures (T_i = 70, 150 and 250 °C). We also calculated a design state using the thermal conductivities of insulfrax and mineral wool given by the manufacturers at T_i = 10 oC. λ₁ = 50 W/mK was the thermal conductivity of a steel pipe.

Using Eq. 1, one can estimate the lifetime of the samples after annealing them. With this equation, one can calculate the ageing of the samples. It must be mentioned that it can be used cautiously, where one will not get long service times. If one considers the service temperature to 10 °C, common in building energetics, one can find the 1 day thermal annealing at 70 °C to be 64 days, while the thermal treatment at 150 °C gives a service time at 10 °C to 45 years. It has to be mentioned that this equation provides unuseful results over 150 °C. Moreover, we can emphasize from the DSC curves that both materials suffer structural change over 200 °C. For this, we can conclude that the insulfrax is stable (with constant thermal conductivity) till 150 °C, for about 45 years at a service temperature of 10 °C. Over this, its thermal conductivity changes by about 24 %. Regarding mineral wool, we can state that the thermal conductivity slightly increased, with about 3 % and 9 %.

The goal and scope phase defines the study and describes the why and how for the analysis to ensure consistency in the LCA. As LCA gives a model of a complex reality, simplifying is necessary. A carefully defined goal and scope minimizes the influence and distortion of the results. Additionally, the most important choices made in the LCA are described here. The intended application of the product, process or service should be described in the goal, along with the reasons for the study, the intended audience and to whom the analysis will be disclosed. The product system studied the system’s function or systems, should be described in the scope. The functional unit, which describes the system’s function, is described in this section. All inputs and outputs are considered concerning this functional unit and are normalized to this. The system boundaries define what should be included and left out of the study and are also part of the scope. The allocation procedure is described, to explain how the impacts are allocated based on choices made. Methodology and impact categories are also described in the scope, along with data requirements, assumptions, and limitations [51].

The second phase of LCA is the inventory analysis. Data collection of all environmental inputs and outputs are listed in an inventory. Environmental inputs refer to anything taken from the environment and added to the life cycle. In contrast, environmental outputs are taken from the product’s life cycle and added to the environment [51]. The inventory analysis can be considered in four steps: preparing for data collection, data collection, calculation, and allocation. The source’s reliability must be documented, as this gives the reliability of the data collected.

With Eq. 8, we calculated the required insulation thickness for the mineral wool to reach this thermal resistance of 20.14 W/m.

(8)

It was calculated that the optimized mineral wool panel thickness should be D_insulation = 6 cm (from Eq.8) while the Insulfrax panel’s thickness is 2.6 cm.

It is evident from Fig. 8a that the production of Insulfrax generates higher impacts for all the impact categories assessed by the CML method. However, when thermal conductivity is used to optimize the weight of material needed to reach the requested thermal resistance, the weight of Insulfrax is lower than that of mineral wool. Consequently, the environmental impact assessed in the manufacturing phase is lower for Insulfrax (Fig. 8b).