Summary of Hygrothermal Analyses Associated with Case Studies

For cases 1 and 2, we modeled the existing roofing system and performed hygro­thermal analyses with the computer program WUFI Pro 4.2 [7] in order to simu­late the impact of residual moisture within the cast-in-place concrete roof deck or topping slab on moisture condensation and accumulation potential within the roofing assembly, considering seasonal cyclic water vapor transmission through the assemblies.[42] We did not hygrothermally model case 3. We used typ­ical built-in moisture levels defined in the WUFI program for the various com­ponents of the roofing system for case 1. We input moisture levels similar to those that we measured in the field and during laboratory testing into the WUFI program for the various components of the roofing system for case 2. We used material property data, surface heat transfer coefficients, and boundary

condition temperatures from the American Society of Heating, Refrigeration, and Air Conditioning Engineers 2009 Handbook of Fundamentals [8] and the WUFI program database. We selected exterior climate data local to each project site from the WUFI database as the exterior boundary condition. We summarize our input and boundary parameters in Table 1. We used sine curves built into the WUFI model to model humidity and temperature transfer between the inte­rior and exterior of the buildings for each case study.

We were not provided with the specific mechanical system operation pa­rameters for temperature and humidity for either facility. For case 1, we per­formed our hygrothermal modeling with an interior temperature range of 65°F to 75°F (18.33°C to 23.89°C), with an average annual temperature of 70°F (21.11°C), and an interior relative humidity range of 35 % to 60 %, with an aver­age annual relative humidity of 50 %, for our interior boundary conditions. For case 2, we performed our hygrothermal modeling with an interior temperature range of 70°F to 76°F (21.11°C to 24.44°C), with an average annual temperature of 73°F (22.78°C), and an interior relative humidity range of 30 % to 50 %, with an average annual relative humidity of 40 %, for our interior boundary conditions.

The case 1 simulation covered a four-year time frame in order to model con­ditions from building completion through the period during which the roofing insulation and membrane detachment occurred. The case 2 simulation covered both a 5- and a 10-year time frame from the time of our investigation forward. Case 2 also included roofing configurations with both CSPE and TPO roofing membranes. None of the simulations included a vapor retarder in the assembly.

The hygrothermal modeling showed that diffusive vapor transfer within both roofing systems from residual moisture within the concrete roof deck or topping slab and any residual moisture within other roofing system components (such as the insulation boards) would drive toward the building exterior during the initial heating season. The moisture drive would reverse and drive toward the building interior during the cooling season. The low permeability of the roofing membrane on the exterior side of the roofing assembly and the concrete topping slab over hollow core precast concrete roof deck or metal deck on the interior side of the assemblies prevents moisture from entering or exiting both systems.

In the models for both case 1 and case 2, residual moisture is trapped within the roofing assemblies and cycles through the assemblies with changes between heating and cooling seasons. The temperature within the roofing assemblies also varies with the heating and cooling season, and frequently reaches, or drops below, the dew point temperature for the environmental conditions within the roofing assembly. The cyclical moisture drive within the assemblies, combined with temperatures that are below the dew point, results in condensa­tion and repeated wetting of the underside of the roofing membrane, organic insulation facers, and the concrete deck. Frequently wet materials lead to the reduced cohesive strength of the insulation facers; delamination of the insula­tion facers; biological growth within the roofing assembly; and weakened adhe­sive bond strength between the concrete topping slab and insulation board, between layers of insulation boards, and between the tapered insulation board

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014

Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

TABLE 1—Summary ofWUFI model input and boundary parameters.

Case 1 Case 2

Concrete relative humidity,3 %


58 to 91


66 to 85

E Insulation moisture content, b %



3 to 42

TPO membrane permeability,3 U. S. perms/metric perms




CSPE membrane permeability,*1 U. S. perms/metric perms




0.13 to 0.24/0.086 to 0.158

WUFI Input and Boundary Parameters Average Range Average Range

aMeasured values based on a modified version of ASTM F2170 [5]. bMeasured values based on ASTM Cl616 [4].

cValues based on manufacturer’s literature using ASTM E96, Procedure В [10]. “^Measured values based on ASTM E96, Procedure В [10].

and the roofing membrane. Additionally, moisture trapped within a roofing as­sembly, combined with roofing thermal cycles, also contributes to dimensional changes and bowing of the insulation boards, which places additional stress on the adhesive bond and wet insulation facers. Our field observations corrobo­rated such insulation board bowing and loss of adhesive bond.

Our models showed that condensation within the roofing assembly was likely for both case 1 and case 2 for TPO and CSPE roofing membranes applied over a concrete substrate with no vapor retarder, which is consistent with our field observations of moisture, stains, and apparent mold growth. The failure of wet and deteriorated facers contributed significantly to the adhesive failure of the roofing assembly.