Identification and Control of VOC Sources in New House Construction
Detailed study to identify VOC sources in a new manufactured house.

The study of VOC concentrations in new manufactured and site-built houses showed that predominant air contaminants in these structures are generally formaldehyde, less volatile aldehydes and terpene hydrocarbons. The IED in collaboration with the FSEC conducted a detailed study to identify the major sources of these compounds in a typically constructed, new manufactured house and to develop and evaluate several VOC source reduction practices (Hodgson et al., In press). Representative samples of all of the major materials used within the interior of the house were collected from the manufacturing plant during construction of the house. Small specimens of these materials were individually preconditioned for 19 ± 4 days in chambers ventilated with clean air. Following conditioning, they were transferred to well-controlled small volume environmental chambers and quantitatively tested for 48 hours for emissions of formaldehyde, acetaldehyde and other VOCs. Whole-house emission rates of individual compounds emitted by the materials were predicted by summing the products of the emission factors and the corresponding material quantities. These predicted rates were then compared to whole-house emission rates derived from measurements made at the house three months after its installation at the project site.

VOC Emissions from Wood Products
Large amounts of wood and composite wood products are used to construct houses. The quantities of wood and the predominant wood products contained within the study house envelope are listed in Table 12. The cabinetry in the kitchen, baths and utility area was identically constructed and was composed of 7 major materials. Many of these had a coating consisting of a sheet of polyvinyl chloride applied to one or more surfaces. The particleboard countertops were finished with laminate, but the undersurfaces were bare. The other predominant wood products contained within the house were the passage doors (molded high-density fiberboard) and the plywood subfloor under the carpeted areas.

The wood products considered as a whole constituted the single largest source of terpene hydrocarbons (i.e., alpha-pinene, beta-pinene and d-limonene) and of formaldehyde and other aldehydes. The aldehyde emission factors in µg/m²-h for the cabinetry components (4 separate materials), passage doors and the plywood subfloor are shown in Table 13. The aldehyde emission factors for the 6 cabinetry materials were multiplied by their respective surface areas (Table 12) to predict whole-house emission rates (Table 14). The particleboard case and the passage doors were the largest predicted contributors of formaldehyde. The fractional contributions of the formaldehyde sources are illustrated in Figure 4. The plywood subfloor was the largest predicted contributor of hexanal and other aldehydes.

Figure 4. Fractional contributions of indoor sources to whole-house formaldehyde emission rates.
 

The indoor and outdoor concentrations of terpenes and aldehydes for the house approximately 3 months after its installation are presented in Table 15. Formaldehyde was above 50 ppb and the concentrations of many of the higher molecular weight aldehydes exceeded their odor thresholds. The ventilation rate at the time of sample was 0.28 ach. The emission rates derived from these house measurements are compared to the predicted rates for combined individual sources also in Table 15. For 10 of the 14 compounds, the averages of the predicted rates were within a factor of ±2 of the derived rates. This suggests that many of the sources of these compounds were correctly accounted for.

Emission Barriers for Plywood Subfloor
Tests were conducted to evaluate the effectiveness of various barriers for reducing the emissions of terpenes and aldehydes from a plywood subfloor. The treatments consisted of standard bonded urethane carpet cushion and a residential carpet, two different brands of spill barrier carpet cushion combined with the same carpet, a perforated reinforced aluminum foil radiant barrier and a non-woven fiber, exterior weatherization membrane. VOC emission factors for the assemblies measured one week after the treatments were applied are shown in Table 16. The standard cushion and carpet had little effect on the emissions of these compounds. The other treatments resulted in statistically significantly lower emissions for 4 to 7 of the target compounds.