The Influence of Mass Tourism and Hygroscopic Inertia in Relative Humidity Fluctuations of Museums Located in Historical Buildings

The preservation of artefacts in museum collections is profoundly affected by fluctuations in temperature and, especially, relative humidity (RH). Since the late nineteenth century, many studies have been carried out into the best way to control hygrothermal conditions. In old buildings located in maritime temperate climate zones (as Portugal) with strong thermal inertia, and which have low ventilation rate (relative to the volume and number of visitors) daily and seasonal hygroscopic inertia may help to assure the maintenance of RH stabilization conditions. The use of expensive active systems may be minimized through passive behaviour of internal finishing building materials. In order to assess the risk of mass tourism and hygroscopic inertia of finishing materials associated with the hygrothermal behaviour of museums, an analysis of several numerical scenarios with a different number of occupants (visitors per hour), different Portuguese climatic zones and finishing materials in order to quantify the risks associated with the fluctuations of relative humidity in a museum. The results of sensitivity studies performed are presented for the case of a museum located in Porto and Lisboa.


Introduction
One of the main functions of museums all over the world is the conservation of artefact collections. This complex work includes, among other things, the control of the interior climate conditions (i.e., temperature, T, and particularly relative humidity, RH) inside the museum buildings (MacIntyre, 1934;Rawlins, 1942;Thomson, 1986).
In the rehabilitation of museums in ancient buildings, active systems for interior climate control have frequently been favored over passive ones. However, in countries with a temperate climate, such as Portugal, hygroscopic inertia combined with adequate ventilation may help control relative humidity fluctuations in ancient buildings without the need for complex active systems.
Hygroscopic inertia refers to the capacity of a room to store excess moisture from the air and restore it to the atmosphere when the relative air humidity is low. The finishings and the stored materials used in the rooms are one of the main factors responsible for the storage and restitution of humidity. Hygroscopic inertia may be assessed over short periods of time (short-cycle hygroscopic inertia of rooms) and for longer periods (long-cycle hygroscopic inertia of rooms).
In the Laboratory of Building Physics, at Faculty of Engineering of University of Porto (FEUP), important research has been carried out in the domain of daily (i.e. short-cycle) hygroscopic inertia in order to quantify the performance of render materials (through parameters that indicate their water vapor adsorption and restitution capacity), find models with which to ECR -Estudos de Conservação e Restauro -2020 -Nº 11 · ISSN: 1647-2098 · PP. 19-30 · https://doi.org/10.34632/ecr.2020.9585 assess the influence of daily hygroscopic inertia upon peaks of relative humidity and develop experimental studies to measure the phenomenon and validate the models (Freitas & Abrantes, 1988;Ramos, 2007;Delgado, Ramos, & Freitas, 2009;.
It is crucial the control of relative humidity and temperature fluctuations to preserve the museum's collections. Today, there are advanced hygrothermal simulation tools to predict the climate inside, taking into account the geometry of the building, the finishing materials, the ventilation, and the occupancy.
In the last years, LFC-FEUP research group has validated the hygrothermal numerical simulation model (Wufi-Plus), in different museums, by comparing numerical and experimental results of interior climate, using for this purpose several sensors.
The mass tourism has led to an exponential increase in museums occupancy with serious consequences on indoor temperature and relative humidity change that may put at risk the preservation of the exhibited collections.
In order to assess the hygroscopic inertia of finishing materials associated with the hygrothermal behavior of museums, an advanced hygrothermal numerical simulation model was used to evaluate those risks, for a generic museum room (model) located in a historical building of different Portuguese climatic zones.

RHS Parameter
Relative humidity stabilization parameter (RHS) consists of the sum over a year of the absolute differences between the 90 days mean seasonal relative humidity and the relative humidity in each hour (Ferreira, Freitas, & Delgado, 2019). The parameter is critical, fundamentally, for organic materials, but should be in mind that other important parameters should be considered. (2)). This period is centered, which means for each value looking back one month and a half and looking forward one month and a half. (2)

Materials and Methods
A coating material, normally, used in the rehabilitation of historical buildings, was select to evaluate the hygroscopic behavior, as well as the contribution to the control of indoor relative humidity. The material selected, with a thickness of 15mm, was a 1.5mm diameter wood fibers panels agglomerated with white cement (Material B).
From the experimental results reported in Table 1 it is possible to observe that Material B is very permeable to water vapor, and the water vapor permeability, δ p , is dependent on the prevailing relative humidity and generally changes rapidly at high RH. This material presents a greater amount of adsorbed water vapor (MBV) for the relative humidity range of 33% to 75%, with 94.7 g/m 2 (2.25 g/m 2 %RH). Nordtest report (Rode et al., 2005) classified the capacity of materials to buffer moisture of indoor environment, according to their MBV obtained from MBV test with a range of 33%-75% RH. As so, Material B is classified as excellent material (MBV>2 g/m 2 %RH) to buffer moisture. Finally, these experiments allowed concluding that Material B presents a hygroscopic capacity of 0.07 kg/m 2 (calculated by the product of equilibrium moisture content with material density and thickness at different RH's) for a relative humidity variation between 50% and 70%. The flux chamber (see Figure 1), with dimensions 1500x524x584mm 3 , was installed in a climatic chamber with controlled temperature (T in the range 15 ºC-35 ºC) and relative humidity (RH in the range 30%-90%). The temperature and relative humidity could be controlled by fixing values or using programmable cycles including the variation of one or both parameters. The ventilation system presents two points, inside the flux chamber, to extract the air and an inlet on top allows for the air to get in and at the same time prevents pressure differences. The flux chamber inlet air comes directly from the climatic chamber and its characteristics are known, whereby infiltration through the openings does not affect the overall balances of heat, air, and moisture. The air flux value is controlled by flow meters with a range of air exchange rate (ach) of 0.26 -17h -1 .     Table 2.

Experimental Results
In Figure 3 and Table 3 are presented the results of the temperature and relative humidity obtained for the tested configurations, with different air change rate, and for the summer cycle and winter cycle. The main conclusions were: (1) The test results carried out in the flux chamber demonstrate that the introduction of the hygroscopic coating material (Material B) influences the fluctuation of the relative humidity inside the flux chamber; (2) This influence is visible when comparing the variation between the maximum and  Hygroscopic capacity (C apHyg ) was calculated for the different configurations tested using Equation (3), and the results obtained described in Table 3

Numerical Analysis
The main objective is to evaluate the performance of the museum rooms with different hygroscopic finishing materials and ventilation rates. The boundary conditions associated with the outside and inside climates and the constitution of the building's envelope are required. Thus, data concerning the building model (see Figure 4), the geometry of the envelope (materials and their properties), exterior climate ventilation and internal gains (number of visitors) were introduced into the model.  In accordance with the experimental results, a preliminary numerical study was conducted using Wufi-Plus hygrothermal model of advanced numerical simulation (IBP, 2010). The main objective is to evaluate the performance of the museum rooms with different hygroscopic finishing materials, ventilation rates, and occupancy, specifically in terms of temperature and relative humidity fluctuations. For this purpose, Table 4 presents several variables of the outdoor climate tested according to the specific weather file used. In this preliminary analysis, only Porto city was considered.

District
Climatic zone  Hygroscopic sorption curve Table 6 -Main properties of the hygroscopic finishing materials used in numerical simulation.
As regards the envelope, Table 5 gives a brief description of the constitution of each component and the material used in the final rendering. The properties of these materials were not determined experimentally. Instead, the database of the program Wufi-Pus was used, which contained selected materials with similar properties to the finishing materials existing in the museum. Table 6 shows the properties of the finishing materials used for the interior layer with higher relevance for the hygrothermal calculation. The area of the hygroscopic finishing material (Material B) used was 168.4m 2 .
Ventilation rates (0.24h -1 and 0.98h -1 , typical values of the museum room analyzed) were specified as well as internal gains resultant from lighting, 9W/m 2 during opening hours (from 10:00 to 18:00) and the only day closed for visitors was considered to be Monday) and 2 W/m 2 selected for the remaining period. Regarding occupancy, different numbers were chosen (5, 10 and 20 visitors per hour), and considering that each visit has a duration of 20 min and a typical year with 308 days of being open to the public, it results in a range of visitors annually between 36960 and 147840.
Several parameters were used to obtain the internal gains, namely, a metabolic rate of 1.28 met and a heat gain of 134W, in which 60% is sensible heat and 40% is latent heat. Sensible heat is directly related to temperature (50% radiant and 50% convective). where n is the number of visitors, ω the water vapor produced by each visitor (70g/h), h the number of hours of the visit (8h), ach the ventilation rate (0.26h -1 ), V the room volume (206m 3 ) and 24 the factor used to obtain the mean hygrometry.

Results and Discussion
Tables 7 and 8 present the indoor climate characterization, specifically concerning temperature and relative humidity, in order to better perceive the implications associated with the different variables analyzed in performed simulations. Figures 5 and 6 show, as an example for Porto, the seasonal temperature, and relative humidity obtained for the different scenarios studied. It is possible to observe an increase of temperature and RH with the number of visitors, however, the increase of RH is significantly lesser when the museum room used hygroscopic materials, namely with an ach=0.24h -1 .
More in detail, the results show: (1) An increase of the average temperature and relative humidity with the number of visitors per hour, for an ach=0.24h -1 , however, for an ach=0.98h -1 , the average temperature increases, and the RH decreases; (2) When the ach value is 0.24h -1 and buffering material B is used on the walls and ceilings, the value of parameter RHS drops, in average, by 27% compared to when the original materials are used; (3) When the ach value is 0.98h -1 and buffering material B is used the value of parameter RHS reduces, in average, by 22% compared to when the original materials are used; (4) When the ach value is 0.24h -1 and buffering material B is used on the walls and ceilings, the difference between the maximum and minimum relative humidity ranged between 28.4% and 40.0% compared to the range between 38.1% and 53.3%.
When the ach value is 0.98h -1 and buffering material B is used the difference between the maximum and minimum RH ranged between 45.8% and 65.3% compared to the range between 53.8% and 71.9%. 11.8 11.5 11.4 10.6 9.9 9.4 9.0

Conclusions
This work presents a study of the influence of hygroscopic materials in stabilizing the relative humidity when ventilation flows are reduced.
When this particular buffering material was analyzed numerically, as part of the walls and ceiling of the museum room, the difference between the maximum and minimum interior