The first subsection describes the main features of the building and its two dwellings, which are the subjects of the present paper. These dwellings were described and modeled via CONTAM, the software described in the second subsection. The last section explains the modeling approach that is proposed and adopted in this paper.
The residential building is located in the neighborhood of Guindalera (district Barrio de Salamanca, Madrid) with the approximate coordinates of latitude 40°26′24′′ N and longitude 3°40′12′′ W, a location where new buildings have been built, mainly during the Spanish “construction boom” that collapsed in 2008. Generally speaking, these buildings have similar features: each have three- or four-story brick buildings that contain small dwelling units (one or two bedrooms) and all were implemented with systems to achieve energy efficiency requirements according to European and Spanish regulations.
The building is located in a radon-prone area, which was identified, in part, from the information given by MARNA (it is an acronym in Spanish of “MApa de Radiación gamma Natural”) [11
] and “Radon Project 10 × 10” [12
]. The survey was performed in a building constructed in 2012 that consists of three floors above the ground floor and two levels beneath. The Energy Performance Certificate gives the building a standard energy and carbon emission efficiency grade of 40.3 E, which is a typical rating for a home. A photograph of the building is shown in Figure 1
a. The total built surface area of the building is 1459 m2
over a parcel of 459 m2
. Each floor contains three dwelling units; two of them are one-bedroom dwellings of similar size and the other one is a two-bedroom dwelling. Figure 1
b shows the floor plan of a one-bedroom dwelling, including a complete floor layout, the use of each area, the location of detector used in the previous study [8
], and other relevant elements, such as doors and windows. Throughout this paper, the first dwelling will be labeled with the letter “A” and the other one with the letter “B”.
Dwellings “A” and “B” have similar features, such as an almost identical useful surface area whose values are, respectively, 39 and 37 m2
. As can be observed from Figure 1
a, there are two differences between the dwellings. First, dwelling “A” has a double frontage, whereas dwelling “B” has only one. The second difference concerns the kitchen; dwelling “A” has an open kitchen-living room, unlike dwelling “B”, whose kitchen is completely separate from the living room.
2.2. The CONTAM Software
This paper shows the effect of a ventilation system on the indoor radon level within two similar dwellings, as assessed by a multizone modeling software for airflow and contaminant transport. The software used is CONTAM version 3.2 [7
], developed by NIST, the first version being released in the year 2000 [13
]. In the present study, it was used mainly to model natural and mechanical airflow and to determine the radon concentrations based on predefined indoor sources.
Air change rates can be affected by a variety of driving forces, including weather (wind and indoor–outdoor temperature difference) and mechanical ventilation. CONTAM performs a simultaneous mass balance of air in all zones to determine zonal pressures and airflow rates through each airflow path (in CONTAM, a zone is a volume of air separated from other volumes of air by walls, floors, and the ceiling). The ability to calculate building airflow rates and relative pressures between zones of the building is useful for assessing the adequacy of ventilation rates, for determining the variation in ventilation rates over time, depending on certain schedules, for determining the distribution of ventilation air within a building, for estimating the impact of envelope air-tightening efforts on infiltration rates, and for evaluating the energy impact of building airflow.
Once the mass flow rates are computed, contaminant concentrations can be determined based on conservation of mass for every contaminant in each zone. The program takes into account the dispersal of airborne contaminants transported by the airflow and transformed by a variety of processes, including chemical and radiochemical transformations. A contaminant can be added to a zone by inward airflow, generation at a constant rate, or reactions with other contaminants. The same contaminant can be removed by outward airflow, removal at a specific rate, or first-order chemical reactions with other contaminants.
Airflow and contaminant information are then used to determine contaminant concentrations within the zones. The mass flow rate from one zone to another along a given airflow path is a function of the pressure difference between the zones (assumed to be governed by the Bernoulli equation, which accounts for static pressure on each side of the flow path), pressure differences due to density, and height differences and pressure differences due to wind. Conservation of mass is applied for all zones, leading to a set of nonlinear algebraic equations that must be solved interactively.
The prediction of contaminant concentrations can be used to determine the IAQ performance of buildings before they are constructed and occupied, to investigate the impacts of various design decisions related to ventilation systems and building material selection, to evaluate IAQ control technologies, and to assess the IAQ performance of existing buildings. Predicted contaminant concentrations can also be used to estimate personal exposure based on occupancy [15
The following tasks are required in order to get realistic results: develop a building idealization, draw a schematic representation, define building components, perform simulations, and review results. These steps are treated jointly in the next section.
2.3. Modeling Approach
This section presents a methodology for modeling indoor radon concentration using CONTAM. The building chosen to carry out the simulation, described in Section 2.1
, was selected because the radon level was known from the measurement campaign performed between 2014 and 2015 [8
]. Furthermore, it is a type of building that is very common in the last period in Spain, so the results have a more significant contribution to the general population.
The floor is divided into two dwellings, identified in this study by the letters “A” and “B”. The dwellings are connected by a common hallway made of granite, in which the stairs and the elevator are placed. Each dwelling is divided into rooms, with each room constituting one zone. Besides, it is required to define the building components, airflow elements, and sources (or sinks) of contaminants that are represented in the code through mathematical models. The CONTAM SketchPad representations of dwellings A and B are shown in Figure 2
, including the elements defined earlier: fans, windows, doors, and sources of radon. The simulation was performed separately for each dwelling. The same hallway is shown in both sketches.
The characteristics of the elements chosen in Figure 2
are given in Table 1
, which shows the room identification code that was formed by the ID of the dwelling (“A” or “B”) and the brief description of the room, surface, and, finally, the number and type of airflow path. Not all parameters are specified in the table, such as the volume, which can be determined from the surface and the room height. In particular, the height of the room was modified to 2.5 m from the default value given by the software (3 m). Furthermore, it is not necessary to indicate the building level in the table because all rooms are located on the first floor.
The technical specifications of the previous elements that were necessary as input are shown in Table 2
. The mathematical model employed in the present CONTAM simulation that provides the relationship between airflow and pressure difference of airflow paths, such as windows and doors, is a one-way flow type using a classical power law model named “orifice area data”. This simplified, one-way flow approach was suggested by a previous study [15
] that used a similar model for the doors because of constant indoor temperatures. A two-way flow model was also investigated for the present simulations and the final results concerning radon concentrations were unchanged, therefore confirming the simplified approach.
Fresh air is supplied through paths through the building envelope, which is the physical separator between the interior and exterior of a building. In addition, a mechanical ventilation was set up in the kitchens, bathrooms, living rooms, and bedrooms. The fan model chosen in the CONTAM simulation is the “constant volume flow fan”. A constant flow rate was set for each fan according to its room designation; therefore, the flow values are proportional to the surface of each room, as summarized in Table 3
. The flow rates were chosen following the recommendations given by Spanish regulation.
CONTAM allows addition of many contaminants (CO2
, NOX, PM, etc.), but, in the present simulation, only radon was added. The main characteristics of radon are a molecular weight of 222 kg/kmol, diffusion coefficient of 5.91 mm2
/s, and decay half-life of 3.8 days. Radon is considered a trace contaminant, because its very low concentration level does not affect the density of air within a zone. The contribution of building materials towards indoor radon depends upon the radium content and exhalation rates, and can be used as a primary index for radon levels in the dwellings. Building materials are the second major source of indoor radon, after soil. Both types of sources can be defined using the constant coefficient model described through the following equation:
G = Generation rate [mass of contaminant/time]
D = Effective removal rate [mass of air/time]
C = Current concentration [mass of contaminant/mass of air]
Setting realistic intensities for the indoor radon sinks was the most challenging step of the present simulation because there are not many details about this in literature [16
]. In Reference [16
], the authors placed a 222
Rn source having a generation rate of 12,000 Bq/h; however, they did not explain why they selected that particular value. In Reference [17
], a more appropriate and general model was selected among available models in literature for the prediction of indoor radon by determining the exhalation rates of building materials. They reported average values of ~50 Bq/m2
/h for the building surfaces of mud dwellings. Reference [18
] interpreted the lithology and building effects, suggesting that radon accumulation in indoor environments is quite a complex phenomenon which can only really be explained by measurable architectonic or other factors.
With the previous considerations in mind, in the present study, we provide a more realistic approximation in CONTAM of indoor radon sinks that accurately reproduce previously measured values [8
] by defining a radon generation rate proportional with the room surface. This value is expected to be of the same order of magnitude as the ones in References [16
]; however, it heavily depends on the geographic region and building materials used. A detailed discussion will be provided in the next section.
Before running the simulation, the last step was setting the type of analysis that is needed (steady-state, transient, or cyclical) along with many other simulation parameters that were discussed earlier. For simulating airflow and contaminant transport, the transient mode was used. The transient integration method selected was the default solver. Initial contaminant concentration was set to zero. The simulation ran for 24 h, using constant weather conditions.