New Research on Diffusion of Carbon Monoxide through Gypsum Wallboard

Transcription

1 New Research on Diffusion of Carbon Monoxide through Gypsum Wallboard Thomas Cleary, Jiann Yang and Marco Fernandez National Institute of Standards and Technology Gaithersburg, MD Suppression, Detection and Signaling Research and Applications Symposium (SupDET 2015) March 3-6, 2015

2 Outline Background Diffusion of CO through gypsum wallboard Estimation of effective diffusion coefficients Measurements of effective diffusion coefficients Simulation of CO transport with diffusion in a residential configuration Conclusions

3 Background Hampson Courtney and Holm publish Diffusion of carbon monoxide through gypsum wallboard in 2013 as a Research Letter in JAMA*. This Research brought up at the Detection and Alarm Research Council meeting on March 5, 2014 by Richard Roberts. Task was group formed to review the research. Initial telecon and following comments questioned validity of results. NIST commits to measuring CO diffusion through Gypsum Wallboard (March 12 th ). By end of March, estimates using the NIST hydrogen and helium diffusion experiments showed results from Hampson et al. reasonable assuming diffusion alone. NIST measurements completed mid May, *Hampson, NB, Courtney, TG, Holm, JR. Diffusion of carbon monoxide through gypsum wallboard. JAMA 2013; 310(7):

4 Estimation of Effective Diffusion Coefficient From the NIST H 2 and He diffusion through gypsum measurements* D He eff 1.3X10-5 m 2 /s (Regular, m thick) D He eff 3.4X10-6 m 2 /s (Regular, m thick and painted) D H 2 eff 1.4X10-5 m 2 /s (Type X, m thick) and assuming (molecular diffusion through pores) D AB /D eff = constant At 20 o C D CO-air = 2.08X10-5 m 2 /s and kpa D H = 6.27X10 2-air -5 m 2 /s D CO eff = 4.6X10-6 m 2 /s (Type X 5/8 in) ambient pressure D He-air = 5.80X10-5 m 2 /s D CO eff = 4.7X10-6 m 2 /s (Reg ½ in) D CO eff = 1.2X10-6 m 2 /s (Reg ½ in painted) Thus, the results of Hampson et al. appear reasonable. *Yang, JC, Pitts, WM, Fernandez, M, Prasad, K. Measurement of effective diffusion coefficients of helium and hydrogen through gypsum. Int. J. Hydrogen Energy 2013; 38: Hampsons results look reasonable

5 Measurement of Effective Diffusion Coefficients Experimental Apparatus The method is analogous to a dual-chamber apparatus, except one chamber is the ambient surroundings (an infinite chamber).

6 Measurement of Effective Diffusion Coefficients At Steady State D eff RT AP N in y co, in y co ( 1 yco, (1 y co in ) ) D eff is the effective diffusion coefficient of CO through gypsum is the thickness of the gypsum wallboard (m) A is the surface area for diffusion N is the total flow rate into the chamber (mol/s) in y is the amount of CO fraction in chamber (mol/mol) co y, is the amount of CO fraction entering the chamber (mol/mol) co in P is the total ambient pressure (Pa) T is the absolute ambient temperature (K R is the universal gas constant (= J/mol/K).

7 Measurement of Effective Diffusion Coefficients Vessel Volume: 8.3 liters Diffusion Area: m 2 ± m 2 CO Feed mol fraction: mol/mol Ambient Temperature: 24.5 o C ± 0.5 o C Ambient RH: 10 % to 45 % Ambient to internal RH difference: 5 % to -11% Time to SS Conditions: ~ 2000 s to 10,000 s depending on wallboard sample and flow rate

8 Measurement of Effective Diffusion Coefficients

9 Measurement of Effective Diffusion Coefficients

10 Measurement of Effective Diffusion Coefficients

11 Measurement of Effective Diffusion Coefficients

12 Measurement of Effective Diffusion Coefficients Gypsum Wallboard Samples Light-weight, 12.7 mm (1/2 in) thick Un-coated One coat of latex interior paint Two coats of latex interior paint Two coats of latex paint & primer (P&P) Type X, 15.9 mm (5/8 in) thick Un-coated One coat of latex interior paint Two coats of latex interior paint Two coats of latex paint & primer (P&P) 2 1 none

13 CO Electrochemical Cell Voltage (V) Measurement of Effective Diffusion Coefficients Electrochemical Cell Calibration NDIR Analyzer CO Concentration ( L/L)

14 Vessel CO Concentration ( L/L) Vessel CO Concentration ( L/L) Measurement of Effective Diffusion Coefficients Vessel Fill and Decay Time (s) Time (s)

15 Table 1. Mean inflow, vessel CO mole fraction and the resultant effective CO diffusion coefficients and uncertainties for gypsum board samples. Sample N Mean y co mol/mol (x 10-6 ) Mean Inflow (L/min) D eff,ave Std. Dev. D eff # u c (D eff ) Reg. un-painted * Reg. un-painted Reg. un-painted * Reg. un-painted Reg. un-painted Reg. 1 coat latex Reg. 2 coats latex Reg. 2 coats p&p Type-X un-painted * Type-X un-painted Type-X un-painted Type-X 1 coat latex Type-X 2 coats latex Type-X 2 coats p&p * CO concentration from gas analyzer # D eff computed from mean measurand values

16 Table 1. Mean inflow, vessel CO mole fraction and the resultant effective CO diffusion coefficients and uncertainties for gypsum board samples. Sample N Mean y co mol/mol (x 10-6 ) Mean Inflow (L/min) D eff,ave Std. Dev. D eff # u c (D eff ) Reg. un-painted * Reg. un-painted Reg. un-painted * Reg. un-painted Reg. un-painted Reg. 1 coat latex Reg. 2 coats latex Reg. 2 coats p&p Type-X un-painted * Type-X un-painted Type-X un-painted Type-X 1 coat latex Type-X 2 coats latex Type-X 2 coats p&p * CO concentration from gas analyzer # D eff computed from mean measurand values

17 Table 1. Mean inflow, vessel CO mole fraction and the resultant effective CO diffusion coefficients and uncertainties for gypsum board samples. Sample N Mean y co mol/mol (x 10-6 ) Mean Inflow (L/min) D eff,ave Std. Dev. D eff # u c (D eff ) Reg. un-painted * Reg. un-painted Reg. un-painted * Reg. un-painted Reg. un-painted Reg. 1 coat latex Reg. 2 coats latex Reg. 2 coats p&p Type-X un-painted * Type-X un-painted Type-X un-painted Type-X 1 coat latex Type-X 2 coats latex Type-X 2 coats p&p * CO concentration from gas analyzer # D eff computed from mean measurand values

18 Table 1. Mean inflow, vessel CO mole fraction and the resultant effective CO diffusion coefficients and uncertainties for gypsum board samples. Sample N Mean y co mol/mol (x 10-6 ) Mean Inflow (L/min) D eff,ave Std. Dev. D eff # u c (D eff ) Reg. un-painted * Reg. un-painted Reg. un-painted * Reg. un-painted Reg. un-painted Reg. 1 coat latex Reg. 2 coats latex Reg. 2 coats p&p Type-X un-painted * Type-X un-painted Type-X un-painted Type-X 1 coat latex Type-X 2 coats latex Type-X 2 coats p&p * CO concentration from gas analyzer # D eff computed from mean measurand values

19 Table 1. Mean inflow, vessel CO mole fraction and the resultant effective CO diffusion coefficients and uncertainties for gypsum board samples. Sample N Mean y co mol/mol (x 10-6 ) Mean Inflow (L/min) D eff,ave Std. Dev. D eff # u c (D eff ) Reg. un-painted * Reg. un-painted Reg. un-painted * Reg. un-painted Reg. un-painted Reg. 1 coat latex Reg. 2 coats latex Reg. 2 coats p&p Type-X un-painted * Type-X un-painted Type-X un-painted Type-X 1 coat latex Type-X 2 coats latex Type-X 2 coats p&p * CO concentration from gas analyzer # D eff computed from mean measurand values

20 Measurement of Effective Diffusion Coefficients Ratio of effective diffusion coefficients between un-coated and coated wallboard samples (mean low flow results) Gypsum Board Un-coated/one coat latex Un-coated/two coats latex Un-coated/two coats P&P Regular Type X

21 Simulation of CO Transport with Diffusion in a Residential Configuration NIST IAQ House

22 Simulation of CO Transport with Diffusion in Residential Configuration With constant pressures and temperatures and fixed source term, simultaneously solve for y 1 and y 2, analytical solution available.

23 Simulation of CO Transport with Diffusion in Residential Configuration Model input. n source : 0.01 mol/s CO (typical) V 1 : 90 m 3 V 2 : 180 m 3 A 1,2 : 15 m 2 A 1,e : 81 m 2 A 2,e : 147 m 2 (D eff = 1.0X10-7 m 2 /s for house to exterior walls) D eff : 4.7X10-6 m 2 /s, or 6.9X10-7 m 2 /s, or 1.0X10-7 m 2 /s : m ACH1 : 0.4 h -1 ACH2 : 0.1 h -1 (extremely tight construction for the house)

24 House CO Concentration ( L/L) Simulation of CO Transport with Diffusion in Residential Configuration Un-painted One Coat Latex Interior Paint Two Coats Latex Interior Paint Time from Start of Source (hours)

25 Simulation of CO Transport with Diffusion in Residential Configuration CO Concentrations in Spaces Separated by Gypsum Wallboard Garage/ House Barrier (single layer) Type X Uncoated Type X One Coat Type X Two Coats 30 minutes One hour Two hours Four hours Garage House Garage House Garage House Garage House 0.34 % 82 ppm 0.50% 255 ppm 0.61 % 662 ppm 0.65 % 1360 ppm 0.42 % 14 ppm 0.74 % 50 ppm 1.2 % 163 ppm 1.5 % 444 ppm 0.44 % 2 ppm 0.79 % 8 ppm 1.3 % 26 ppm 1.9 % 75 ppm

26 Conclusions The single-chamber apparatus is well-suited to measure effective diffusion coefficients of CO through gypsum wallboard with up to two coats of paint applied. The observed reductions in the effective diffusion coefficients by one or two layers of paint will significantly reduce diffusive transport of CO through gypsum wallboard. Analysis of realistic situations is needed to inform Codes and Standards development regarding CO hazard mitigation.