This paper is divided into three sections. The first section is focused on the characteristics of the designed structures of urban tenement houses in the Czech Republic, which is based on the study of historical documents and on a field survey carried out on selected tenement houses in Prague. In the second section of the paper, we refer to analogous examples from foreign studies focused on the hygrothermal properties of residential buildings. The third section presents an analysis of selected details of building structures—specific critical points that are part of the thermal envelope of apartment buildings. The analyses were carried out by means of various numerical models.
2.1. Building Structures
With the development of urban construction, the need for regulation has also grown. Throughout the 19th century, some of the most important areas of building law were clarified, especially in relation to fire safety [
8]. The building codes determined, for example, the maximum height of buildings and the design of ceiling structures, staircases or roofs, etc.; these were valid until the end of the Austro-Hungarian Monarchy and were largely adopted in the legal system of Czechoslovakia after the establishment of the independent state (1918). However, buildings were still designed on the basis of the builder’s experience, with the thickness of the vertical load-bearing structures determined by empirical formulae depending on the number of storeys, the height of the storey and the depth of the wing.
Magazines and building manuals, such as the
Allgemeine Bauzeitung, published in Vienna, from 1836 [
9], or Jöndl’s
Instruction on Civil Engineering, from 1840 [
10], also played an important role in the development of building structures. These and other similar publications conveyed information about new materials and technologies and promoted changes in building typology and design.
2.1.1. Vertical Structures
Urban residential masonry buildings were characterized by massive walls, most often implemented as brickwork with mixed masonry and especially with solid brickwork. From the second half of the 19th century onwards, on the top floors was required a wall thickness of at least 45 cm, and this increased downwards (
Figure 2). At the turn of the 19th century, and especially from the beginning of the 20th century, there was an increase in the number of buildings whose walls were already composed of hollow burnt bricks on the upper floors or, later, of brick cavity blocks comprising lightweight concrete, etc. (cinderblock or slag pumice). Brick walls and pillars of greater thickness, usually over 70 cm, were often constructed as multi-layered. Lime mortar was predominantly used as a bonding material until the middle of the 19th century, followed by an increase in the use of lime-cement mortar in the next period, and then the use of cement mortar in the 20th century [
11].
“The determination of the true thickness of a wall in civil engineering is, in most cases, the result of careful consideration of various factors, which are: 1. the kind of material, 2. the shape of the wall, 3. the thoroughness of the work, 4. the magnitude, direction, and action of the forces acting, 5. the free length and height of the wall, 6. the number and size of openings, 7. the position of the building, 8. the purpose of the wall”.
The stability of the load-bearing system of the multi-storey masonry buildings was ensured by longitudinal walls on which the floor structures were placed, interconnected with the longitudinal walls (
Figure 3), [
13]. To ensure stability, the transverse walls—gable and stair walls and masonry partitions of 15 cm and 30 cm thickness—played an important role in connecting the longitudinal, primarily load-bearing walls, ensuring their stability and contributing to the spatial interaction of the load-bearing system [
14].
Balconies, pavilions and bay windows were also characteristic elements of the town’s architecture, and their design (dimensions) had to meet the requirements of the building regulations (
Figure 4). In most cases, the structure of balconies and pavilions consisted of arches or slabs (stone, ceramic, or wooden) placed on brackets or beams. The bay windows often dominated the façade or corners of the house. They were constructed with a supporting structure consisting of an iron (‘traverse’) frame, the beams of which were, for example, extended beams of the ceiling structure, or special beams anchored into the central walls. The 30–45 cm thick perimeter walls of the bay windows were made of perforated and lightened hollow bricks, in some cases supplemented with heraclite or pressed cork boards (to improve thermal insulation properties).
2.1.2. Horizontal Structures
In most of the urban tenement buildings in the area, it is possible to mainly encounter various types of wooden beam ceiling structures. Among the horizontal structures used, the dominant ones were spalled and semi-salvaged ceilings (i.e., with a minimum 80 mm thick embankment at the upper face, while the protection of the wooden beam structure at the lower face usually consisted of underfloor boards with reed plaster). Among the structures used, wooden beam ceilings with a backfill were predominant.
In areas with high fire safety requirements, e.g., on the ground floor of tenement buildings, mostly fireproof ceilings were designed. Vaults were considered to be safe, fireproof ceiling structures (or were required by the building regulations), which were used in entrances, house entrances, cellars and, if necessary, in staircases, corridors and on the ground floor of tenement buildings.
An important function in terms of ensuring spatial rigidity, stability and resistance to the effects of forced reshaping were the so-called wall and beam clamps. From the beginning of the 19th century, iron tongs were used in walls and vaults to stiffen and retract the building and to absorb the oblique pressures, especially when the masonry was settling and the foundation soil was being pressed.
2.1.3. Window Fillings
The most important parts of the tenement houses are the windows, which co-create the architectural expression of the buildings. The material and technical design of the window openings and their fillings represent a historical document, which brings closer the technical, craft and artistic level of the individual professions that participated in its creation at the time—carpenters, glaziers, locksmiths, stonemasons, masons and others [
16]. For these reasons, informed restoration should be preferred to replacement with modern infill so as to preserve the expression and integrity not only of a single building, but also of entire street fronts. In the case of listed buildings, the cultural value in exposed locations may take precedence over the thermal quality of the building.
From the typological point of view, several basic types of windows were used in the Austro-Hungarian urban construction of the turn of the 19th and 20th centuries, based on the general design principle of the window [
17,
18].
The simplest window was a single window frame. In connection with the greater demands on the thermal comfort of interiors, from the beginning of the 19th century onwards, second window frames were first added to existing windows (so-called winter windows), and in new buildings, double glazing was introduced (
Figure 5b). From the mid-19th century, the double window was the standard for heated interiors, while the use of the single window was restricted to unheated and service areas (e.g., corridors, staircases, lavatories, pantries, workshops, etc.).
The windows consisted of two separate frames—an outer and an inner one (the sashes of the inner window opened inwards and the sashes of the outer frame, facing the facade of the building, opened outwards) and had a number of drawbacks, including the exposure of the outer sashes to the weather and their rapid degradation. In order to protect the outer frame, it was recessed into the window opening and, at the same time, the inward opening of the sashes was reversed. Thus, in the last quarter of the 19th century, a double-paned window was created by joining the outer and inner windows into a single structural unit using wooden boards (jambs) (
Figure 5a). This step, among other things, made it easier to fit into the building opening without the need to laboriously measure the correct position of the window frames in relation to each other.
“The size of the windows is determined by the dimensions of the room to be illuminated by them, mainly by their depth and height, then by the desired intensity of lighting and finally by the overall architectural conditions on the exterior of the building. I will give 1/4 of the floor area, what the light area of the windows, sharp, 1/6 medium, 1/8 dim lighting”.
2.2. Hygrothermal Properties of Residential Buildings
The serious shortcomings of the brick tenement houses from the second half of the 19th and early 20th centuries include the unsatisfactory thermal resistance of the envelope structures and the high energy consumption of the operation of these buildings, which is several times higher compared to the current structures. The thermal assessment of the envelope brick walls of the historic buildings demonstrated their insufficient and unsatisfactory thermal properties in terms of the currently applicable standard CSN 73 0540-2: 2011 [
19].
Since 1949, when CSN 1450 [
20] came into force and defined the value of the thermal resistance of the envelope structure, the requirement for the value of the thermal resistance of the envelope structure (according to the current CSN 73 0540: 2011) has increased by at least 4 times (not considering the requirement for passive standard from 2020).
Figure 6 and
Figure 7 show a comparison of the thermal performance of the envelope masonry wall with the evolution of the standard’s requirements. The first standard that formulated requirements for the thermal insulation properties of building structures in the Czech Republic, CSN 73 0540 “Thermal properties of building structures and houses”, was published in 1964. The minimum value of the thermal resistance of the perimeter wall in this standard was based on the provisions of the Building Code for the Capital City of Prague from 1886, which required that perimeter wall structures should be at least 450 mm thick [
21]. This thickness corresponds approximately to a thermal resistance of R
N = 0.6 m
2K/W, which is a satisfactory value as required by that standard.
Thermal standards responded to the global energy crisis in the 1970s by increasing the requirements for the thermal insulation capacity of building structures, especially envelope structures. This development began with the revision of the thermal engineering standard in 1977 and has essentially continued to the present day. Standard CSN 73 0540-2, “Thermal protection of buildings. Part 2: Requirements”, of 2011, specifies for the external wall the required value of the heat transfer coefficient at the level of U
N = 0.30 W/m
2K for both light and heavy structures. The recommended values for heat transfer coefficients are 0.20 and 0.25 W/m
2K, respectively (passive houses U = 0.18 to 0.12 W/m
2K). It is, therefore, clear that there has been an increase of approximately fourfold (for the required values) to ninefold (for the recommended values or the values for the passive standard) compared to the values of the 1960s. The current thermal standard CSN 73 0540: 2011 specifies three values for the heat transfer coefficient of individual building structures: the required value, the recommended value and the range of values recommended for passive buildings. The required value of the standard heat transfer coefficient of the structure is a value that must not be exceeded and which guarantees the function of the structure without any structural and physical defects. Decree No. 78/2013 Coll. (replaced in 2020 by Decree 264/2020 Coll.) on the energy performance of buildings from 2013 requires that the resulting heat transfer coefficient of modified structures is lower than the recommended value, according to CSN 73 0540-2: 2011. The majority of the existing brick urban tenement buildings do not meet the required values of heat transfer coefficient and energy performance (
Figure 8).
The Energy Performance of Buildings Directive [
22], together with the Energy Efficiency Directive [
23], promote policies for achieving efficient and decarbonised building stock (the goal is by 2050), creating a stable environment for investment decisions and enabling consumers and businesses to make more informed choices in order to save energy and money. The last proposed revision reflects even higher ambitions in terms of climate and social actions and sets a goal for achieving a zero-emission and fully decarbonized building stock. Among the main steps to achieve this goal, there is the gradual introduction of cost-optimal minimum energy performance standards (even for existing buildings); a definition of deep renovation and the introduction of building renovation passports; the modernization of buildings and their systems; better energy system integration (for heating, cooling, ventilation); and others. The Renovation Wave Strategy [
24] aims to at least double renovation rates in the next ten years and ensure that renovations lead to higher energy and resource efficiency. This should enhance the quality of life for people living and using these buildings, reduce energy poverty, reduce Europe’s greenhouse gas emissions, develop digitalization and improve the reuse and recycling of materials. All these directives and actions are focused on governmental and public building, but their implementation in the private sector (i.e., residential buildings) is encouraged. Moreover, the operative energy consumption should not be the only aspect of the evaluation of existing buildings. Assessing the carbon footprint during the life cycle of buildings using LCA methodology should take precedence over the energy efficiency.
The higher energy consumption of masonry residential buildings is not only due to the inadequate thermal insulation properties of the masonry external walls, but also to the often inadequate properties of casement and dual-timber single-glazed windows. Improving the thermal insulation properties of windows, using new infill structures, and the effective refurbishment and repair of existing openings can make a significant contribution to improving the energy performance of these buildings (
Figure 9). However, when addressing this issue, it is necessary to consider not only the properties of the windows themselves, but also the related structures [
25,
26,
27]. The technical parameters of window infills that need to be evaluated mainly include the heat transfer coefficient, airtightness, air permeability and watertightness.
Achieving optimum thermal properties, the comfort and healthiness of the indoor environment, the prevention of mould formation and the reduction in energy consumption of masonry residential buildings requires a number of complex and systemic measures. External insulation of envelope masonry walls with a contact insulation system cannot be applied to masonry residential buildings with the decorative design of a segmented street façade with a number of cornices, balconies and bay windows. Contact insulation systems on the outside can only be applied in cases that allow the original simpler profiles to be preserved, especially for courtyard facades.
A number of problems that have not yet been sufficiently addressed accompany the insulation of the articulated façade of the street envelope walls from the inside. The severity of these problems increases with the increasing humidity of the insulated interior. For internal wall insulation, thermal insulations with high diffusion resistance such as foam glass and extruded polystyrene are preferred. The combination of moderate internal insulation with external thermal insulation plaster will reduce the amount of water vapour condensing in the structure in winter to a certain extent. However, the rugged and intricately profiled street façade of the historicizing architecture of these buildings does not allow the application of an insulating plaster. A vapour barrier on the inside of the structure is usually proposed as a solution. Despite several measures, the vapour barrier cannot be considered perfectly tight; the risk of various leaks is always high even with very careful implementation. In the analysis of the proposed structure, the diffusion resistance factor of the vapour barrier must always be reduced to at least one-tenth of the value stated by the manufacturer. Due to the many uncertainties in the implementation of the vapour barrier, thermal insulation materials with high diffusion resistance are preferred for wall insulation. One of the most common alternative methods of internal insulation is the use of so-called capillary active materials, which allow the transport of moisture in the liquid phase. In contrast to the traditional solution, in which the vapour barrier restricts the diffusion of water vapour into the structure, with capillary active materials, the condensation of water vapour is accepted as an inevitable process—the condensate formed is dispersed in the capillary active insulation and evaporates into the interior when the boundary conditions change. A serious problem in these insulation cases is the slight increase in the mass moisture content of the insulated masonry (various international studies report an average increase of 0.5 to 1.5 percentage points). This increase from the pre-insulation condition is usually higher the greater the thickness of insulation applied and the thinner the insulated wall [
22]. Finally, the internal insulation also reduces the internal space of the building.
Structures that can also adversely affect the energy balance of the building and whose additional insulation is necessary from the point of view of energy savings are ceiling structures above the unheated basement space and ceiling structures under the pitched roof structure, or ceiling structures under the unheated attic.
By insulating the ceiling structures under the attic and above the unheated basement, the temperature in the attic or basement will drop in winter. The reduction in air temperatures can then cause condensation on the cold roof trusses or basement walls. From this point of view, it is necessary to assess the possible consequences of these measures in relation to the risk of condensation on the structures where the temperature has fallen as a result of the insulation. Restrictions on the original ventilation of the attic may pose a moisture risk to the timber roof structure.
Compared to envelope structures, the benefit gained by insulating the ceiling structure under a pitched roof structure is not as significant, as the temperature gradient of that structure is significantly lower. The losses of a brick apartment building in its original state through the envelope (without windows) account for 40–65% of the building losses; losses through the ceiling under the unheated attic 14–20%; losses through the floor above the unheated basement about 10–15%; and losses through the original windows about 5–15%.
When modifying wooden ceilings for the purpose of thermal insulation or the improvement of acoustic properties, it is necessary to avoid layers with high diffusion resistance (PVC, rubber, cardboard, foil, etc.), which could change the existing hygrothermal regime of the ceiling.
Particular attention should be paid to the placement of the ceiling joists in the pockets in the perimeter (outside) walls. Preventing beam ceiling header degradation requires ensuring the necessary warm air flow on the inner surface of the perimeter walls, limiting damp operations (ensuring effective ventilation) and, finally, ensuring the permeability of the beam ceiling structure.
Thermal comfort and energy losses are also influenced by the thermal properties of internal load—bearing and dividing walls, and partitions. According to CSN 73 0540-2: 2011, the heat transfer coefficient of these structures must be lower than 0.60 W/(m2.K) or 0.75 W/(m2.K), depending on whether they separate the heated space from the tempered space (e.g., from an internal corridor in an apartment building) or from a completely unheated space (e.g., from a garage or basement). Existing internal wall structures do not have the necessary thermal insulation properties.
A number of research institutes, especially in Central and Northern Europe, are dealing with the restoration, modernisation and insulation of historic buildings. Theoretical and experimental research works carried out until now have not yet provided a reliable and unambiguous answer to the questions related to the insulation of envelope structures from the interior side. Partial results and conclusions are mainly linked to specific (local) conditions and materials and cannot be directly and reliably applied to the insulation of the articulated and profiled facades of historic buildings.
The extent and importance of improving the performance and functional properties of masonry envelope walls and associated structures requires the design of a reliable and durable solution supported by both theoretical and experimental research.
The literature mostly deals with only partial, specific problems. Quite extensive research is being conducted in Denmark, Slovenia, the Baltic countries, and outside Europe in the USA, Canada and China. The research mainly focuses on testing different types of thermal insulation (porous concrete, calcium silicate boards and capillary active thermal insulation system based on rigid PUR foam panels) bonded in different thicknesses to the inside of brick walls [
28], as well as different thicknesses and extents of insulation (e.g., parapet masonry only) [
29,
30]. Furthermore, the risk of changing temperature and humidity conditions inside the original wall (increase in mass moisture due to condensation) [
31,
32,
33,
34], the associated negative effect of freezing cycles [
35] and the risk of mould growth [
36,
37,
38] have been investigated. The possible reduction in the influence of these risks by using capillary active materials in combination with the installation of heating cables is suggested in [
39].
The analysis of different structural solutions (internal insulation with and without vapour barrier, with and without hydrophobicity, etc.) highlights problematic locations in the area of the installation of wooden ceiling structures (increased moisture impact, reduced temperatures in the installation area, etc.) [
38,
40,
41,
42].
Furthermore, a part of the research is focused on the modelling and numerical simulations of the thermal and moisture behaviour of masonry structures while using different types of insulation [
43,
44].
In the case of changes and structural modifications of completed buildings, according to CSN 73 0540-2: 2011 (Article 5.2.11), it is possible to exceed the required values of the heat transfer coefficient if technical or legislative obstacles demonstrably prevent compliance with the standard’s requirement. However, in such a case, at least the best technically available solution must be used so that defects and malfunctions in the use of the building are demonstrably prevented.
2.3. Hygrothermal Analysis of Selected Structures and Details
In historic brick buildings, wooden (timber) beam (ceiling) structures were used to a large extent for the roofing of interior spaces until the mid-19th century. The most common type of the vertical load-bearing structure was a masonry wall 450 mm thick, made from burnt bricks and cement-lime mortar. Where the timber beams are placed on the external load-bearing wall (in the so-called pockets), the external wall is weakened by 200 to 250 mm, and thus, the thermal resistance is significantly reduced (
Figure 10a). A similar situation occurs in the case of ceramic ceilings (
Figure 10b). Therefore, a two-dimensional steady state hygrothermal analysis of this critical detail in the original state (without any thermal insulation) was performed for the case of wooded as well as ceramic ceiling. To evaluate possible solutions, an analysis of the structure with different positions of thermal insulation (external and internal with or without vapour barrier) was also conducted (see
Figure 10).
As previously mentioned, in the case of residential buildings built at the turn of the 19th and 20th centuries in the Central European region, it is often not possible to apply external thermal insulation due to the decorative design of street façades. Hence, the internal thermal insulation is often the only viable solution. However, when applying internal insulation, thermal bridges are created around the wall contacts with the surrounding structures, especially with partitions, walls, ceilings and roofs (
Figure 11). In buildings with timber beam ceiling structures, internal insulation increases the risk of condensation around the beam head (see
Figure 10). A two-dimensional steady state hygrothermal analysis of a selected section of the historic structure containing an external wall and inner partition contact and external wall corner was performed to compare the impact of thermal insulation placement (external insulation composed of expanded polystyrene vs. internal insulation composed of capillary active material—polyurethane foam).
Finally, the subject of dual windows with single glazing in historic masonry structures was investigated. The improvement of their thermal insulation properties is often limited by the number of boundary conditions. Double widows in historic structures can be renovated by replacing both sashes with new double- or triple-glazed windows, or (in the case of historically valuable buildings) by replacing only the outer sashes with new ones with high-quality double or triple glazing. Double glazing (or triple glazing) with a low heat transfer coefficient fitted in the outside position will ensure sufficiently high internal surface temperatures both on the sashes of the dual window and on the casing or plastered part of the lining between them. Internal surface temperatures on the surrounding wall may also increase significantly. This minimizes the risk of condensation on the inner surface of all components of the dual window. On the other hand, placing the double (or triple) glazing in the inner sashes of the dual window can lead to a drop in air temperature in the cavity between the sashes in winter and can cause condensation and icing of the condensate on the inner surface of the outer single-glazed sash. Surface condensation can also be a threat to the casing if not effectively insulated. Therefore, a comparison of these possible solutions was carried out in the form of a two-dimensional steady state hygrothermal analysis (
Figure 12).
Numerical Two-Dimensional Steady State Hygrothermal Analysis
In all the two-dimensional steady state hygrothermal analyses (two-dimensional heat conduction) performed, standard material properties (particularly coefficient of thermal conductivity) were used—masonry λ = 0.8 W.m−1.K−1; wood λ = 0.18 W.m−1.K−1; slag fill λ = 0.95 W.m−1.K−1; lime plaster λ = 0.88 W.m−1.K−1; expanded polystyrene (EPS) λ = 0.039 W.m−1.K−1; mineral wool λ = 0.041 W.m−1.K−1; polyurethane foam λ = 0.031 W.m−1.K−1; Lignopor λ = 0.046 W.m−1.K−1; glass λ = 0.78 W.m−1.K−1; closed air gap λ = 0.067 W.m−1.K−1; wooden window frame λ = 0.13 W.m−1.K−1.
Standard boundary conditions pursuant to the CSN 73 0540-2: 2011 were considered (indoor air temperature 22 °C, indoor relative humidity 50%, outdoor air temperature −13 °C, outdoor relative humidity 84%).
The software used to perform the two-dimensional steady state hygrothermal analysis was Area 2017 by Svoboda software. This software calculates temperature and pressure fields solving standard partial differential equation [
45]:
with the boundary condition:
where (x,y) are the coordinates of the point in the plane [m]; λ is the thermal conductivity coefficient [W/(m.K)]; θ is the temperature at the point [°C]; h is the heat transfer coefficient [W/(m
2K)]; θ is the ambient temperature [°C]; ∂x, ∂y are the derivatives with respect to x and y; and ∂n is the derivative with respect to the normal.
Equation (1) is solved on a simply continuous region Ω with boundary Γ, on which the boundary condition (2) must be satisfied. The boundary Ω is rectangular or generally curvilinear. Furthermore, it is assumed for the calculation that the region Ω can be divided into a finite number of regions in which the function λ(x,y) is constant. Moreover, the functions h(x, y) and θ(x,y) are considered constant over parts of the boundary Γ. Equation (1) is modified by the Galerkin method and using Green’s theorem to the form:
where K is the body conductivity matrix defined as
and r is the column matrix of nodal temperature values (unknowns) and q is the right-hand side vector (source vector) described by
and N is a row matrix of basic functions.
Discretization of the problem (generation of finite elements into which the region G is divided) is performed using a planar triangular element. The final solution of the system of linear equations (calculation of the temperature and pressure values in the nodes) is performed using Gaussian elimination.
For the calculation of the two-dimensional stationary field of water vapor pressures, a similar equation and boundary condition is used—only instead of the thermal conductivity coefficient, it is necessary to consider the diffusion coefficient, and instead of the heat transfer coefficient, the water vapor transfer coefficient.