Derek G. Shendell

Predicted concentrations in new relocatable classrooms of volatile organic compounds emitted from standard and alternate interior finish materials

Relocatable classrooms (RCs) are widely employed by California school districts to satisfy rapidly expanding space requirements due to population growth and class size reduction policies. There is public concern regarding indoor environmental quality (IEQ) in schools, particularly in RCs, but very little data to support or dispel these concerns. Several studies are investigating various aspects of IEQ in California schools. This laboratory-based study focused on evaluating the emissions of toxic and/or odorous volatile organic compounds (VOCs), including formaldehyde and acetaldehyde, from materials used to finish the interiors of new RCs. Furthermore, the study implemented a procedure for VOC source reduction by testing and selecting lower-emitting materials as substitutes for standard materials. In total, 17 standard and alternate floor coverings, wall panels and ceiling panels were quantitatively tested for emissions of VOCs using smallscale environmental chambers. Working with the largest northern California manufacturer of conventional RCs and two school districts, specifications were developed for four new RCs to be produced in early summer 2001. Two of these will be predominantly finished with standard materials. Alternate carpet systems, an alternate wall panel covering and an alternate ceiling panel were selected for the two other RCs based on the results of the laboratory study and considerations of cost and anticipated performance and maintenance. Particular emphasis was placed on reducing the concentrations of VOCs on California agency lists of toxic compounds. Indoor concentrations of toxic and odorous VOCs were estimated for the four classrooms by mass balance using the measured VOC emission factors, exposed surface areas of the materials in the RCs, and three ventilation rate scenarios. Results indicate that reductions in the concentrations of formaldehyde, acetaldehyde phenol, di(ethylene glycol) butyl ether, vinyl acetate, 1,2,4-trimethylbenzene and 1-methyl-2-pyrrolidinone should be achieved as the result of the source reduction procedure.

Final methodology for a field study of indoor environmental quality and energy efficiency in new relocatable classrooms in Northern California

LBNL-51101 Final Methodology for a Field Study of Indoor Environmental Quality and Energy Efficiency in New Relocatable Classrooms in Northern California Derek G. Shendell, Dennis Di Bartolomeo, William J. Fisk, Alfred T. Hodgson, Tosh Hotchi, Seung-Min Lee, Douglas P. Sullivan, and Michael G. Apte Lawrence Berkeley National Laboratory, Indoor Environment Department, Environmental Energy Technology Division, Berkeley, CA Leo I. Rainer Davis Energy Group, Davis, CA Submitted to the California Energy Commission August 2002 for the Public Interest Energy Research Program funded LBNL High Performance Commercial Buildings Systems Program (HPCBS), Element 6.2.2

Report on HVAC option selections for a relocatable classroom energy and indoor environmental quality field study

It is commonly assumed that efforts to simultaneously develop energy efficient building technologies and to improve indoor environmental quality (IEQ) are unfeasible. The primary reason for this is that IEQ improvements often require additional ventilation that is costly from an energy standpoint. It is currently thought that health and productivity in work and learning environments requires adequate, if not superior, IEQ. Despite common assumptions, opportunities do exist to design building systems that provide improvements in both energy efficiency and IEQ. This report outlines the selection of a heating, ventilation, and air conditioning (HVAC) system to be used in demonstrating such an opportunity in a field study using relocatable school classrooms. Standard classrooms use a common wall mounted heat pump HVAC system. After reviewing alternative systems, a wall-mounting indirect/direct evaporative cooling system with an integral hydronic gas heating is selected. The anticipated advantages of this system include continuous ventilation of 100 percent outside air at or above minimum standards, projected cooling energy reductions of about 70 percent, inexpensive gas heating, improved airborne particle filtration, and reduced peak load electricity use. Potential disadvantages include restricted climate regions and possible increases in indoor relative humidity levels under some conditions.

Simultaneous Energy Savings and IEQ Improvements in Relocatable Classrooms

Relocatable classrooms (RCs) are commonly used by school districts with changing demographics and enrollment sizes. We designed and constructed four energy-efficient RCs for this study to demonstrate technologies with the potential to simultaneously improve energy efficiency and indoor environmental quality (IEQ). Two were installed at each of two school districts, and energy use and IEQ parameters were monitored during occupancy. Two RCs (one per school) were finished with materials selected for reduced emissions of toxic and odorous volatile organic compounds (VOCs). Each had two HVAC systems, operated on alternate weeks, consisting of a standard heat-pump system and an indirect-direct evaporative cooling (IDEC) system with gas-fired hydronic heating. The IDEC system provides continuous outside air ventilation at {sup 3}15 CFM (7.5 L s-1) person-1, efficient particle filtration while using significantly less energy for cooling. School year long measurements included: carbon dioxide (CO2), particles, VOCs, temperature, humidity, thermal comfort, noise, meteorology, and energy use. IEQ monitoring results indicate that important ventilation-relevant indoor CO2 and health-relevant VOC concentration reductions were achieved while average cooling and heating energy costs were simultaneously reduced by 50 percent and 30 percent, respectively.

Energy savings estimates and cost benefit calculations for high performance relocatable classrooms

HPCBS Element 6, Project 2.1.2: Energy Savings Estimates and Cost Benefit Calculations for High Performance Relocatable Classrooms FINAL REPORT Report Issued: November 24, 2003 Presented to: Michael Apte Lawrence Berkeley National Laboratory Prepared by: Leo Rainer Marc Hoeschele Davis Energy Group, Inc. 123 C Street Davis, CA 95616

Association of airborne moisture-indicating microorganisms withbuilding-related symptoms and water damage in 100 U.S. office buildings:Analyses of the U.S. EPA BASE data

Metrics of culturable airborne microorganisms for either total organisms or suspected harmful subgroups have generally not been associated with symptoms among building occupants. However, the visible presence of moisture damage or mold in residences and other buildings has consistently been associated with respiratory symptoms and other health effects. This relationship is presumably caused by adverse but uncharacterized exposures to moisture-related microbiological growth. In order to assess this hypothesis, we studied relationships in U.S. office buildings between the prevalence of respiratory and irritant symptoms, the concentrations of airborne microorganisms that require moist surfaces on which to grow, and the presence of visible water damage. For these analyses we used data on buildings, indoor environments, and occupants collected from a representative sample of 100 U.S. office buildings in the U.S. Environmental Protection Agencys Building Assessment Survey and Evaluation (EPA BASE) study. We created 19 alternate metrics, using scales ranging from 3-10 units, that summarized the concentrations of airborne moisture-indicating microorganisms (AMIMOs) as indicators of moisture in buildings. Two were constructed to resemble a metric previously reported to be associated with lung function changes in building occupants; the others were based on another metric from the same group of Finnish researchers, concentration cutpoints from other studies, and professional judgment. We assessed three types of associations: between AMIMO metrics and symptoms in office workers, between evidence of water damage and symptoms, and between water damage and AMIMO metrics. We estimated (as odds ratios (ORs) with 95% confidence intervals) the unadjusted and adjusted associations between the 19 metrics and two types of weekly, work-related symptoms--lower respiratory and mucous membrane--using logistic regression models. Analyses used the original AMIMO metrics and were repeated with simplified dichotomized metrics. The multivariate models adjusted for other potential confounding variables associated with respondents, occupied spaces, buildings, or ventilation systems. Models excluded covariates for moisture-related risks hypothesized to increase AMIMO levels. We also estimated the association of water damage (using variables for specific locations in the study space or building, or summary variables) with the two symptom outcomes. Finally, using selected AMIMO metrics as outcomes, we constructed logistic regression models with observations at the building level to estimate unadjusted and adjusted associations of evident water damage with AMIMO metrics. All original AMIMO metrics showed little overall pattern of unadjusted or adjusted association with either symptom outcome. The 3-category metric resembling that previously used by others, which of all constructed metrics had the largest number of buildings in its top category, was not associated with symptoms in these buildings. However, most metrics with few buildings in their highest category showed increased risk for both symptoms in that category, especially metrics using cutpoints of >100 but <500 colony-forming units (CFU)/m{sup 3} for concentration of total culturable fungi. With AMIMO metrics dichotomized to compare the highest category with all lower categories combined, four metrics had unadjusted ORs between 1.4 and 1.6 for both symptom outcomes. The same four metrics had adjusted ORs of 1.7-2.1 for both symptom outcomes. In models of water damage and symptoms, several specific locations of past water damage had significant associations with outcomes, with ORs ranging from 1.4-1.6. In bivariate models of water damage and selected AMIMO metrics, a number of specific types of water damage and several summary variables for water damage were very strongly associated with AMIMO metrics (significant ORs ranging above 15). Multivariate modeling with the dichotomous AMIMO metrics was not possible due to limited numbers of observations.