Evaluation of lightemitting diode LED UV radiation as an emerging technology for pointofuse, househo

1
Evaluation of light-emitting diode (LED) UV
radiation as an emerging technology for
point-of-use, household disinfection of drinking
water

• Mark D. Sobsey and Kari Leech
• Dept. of Environ. Sci. and Engineering
• Gillings School of Global Public Health
• University of North Carolina
• Chapel Hill, NC 27599-7431 USA
• Sobsey_at_email.unc.edu

2
Ultraviolet Light Disinfection

• UV-C 200-280 nm
• Highly germicidal to microorganisms

3
UV Disinfection Mechanisms

• Electromagnetic energy transfers from light
  source to genetic material of microorganisms
• Photoproducts (chemical dimers) form and break
  DNA/RNA bonds or fuse thymine molecules to each
  other
• Dimers prevent microbe replication by creating a
  section of DNA or RNA which is unreadable during
  replication and transcription
• Inability to replicate inactivation or
  disinfection

4
Household Water Treatment

• Improves water quality
• Reduces diarrhea and other WB infectious diseases
• Cost-effective
• Can be rapidly deployed and taken up by
  vulnerable populations
• Some Rxs. are broadly effective for all pathogens
• Range of technologies
• Durable goods technologies with minimum user
  requirements are preferred
• Passive
• Easy
• Sustainable (continued and effective use)

5
Household Water Treatment Technology Options
UV Lamp
SODIS
Boil
Chlorine
Porous Ceramic Pot Filter
Flocculant Disinfectant
Slow sand/biosand filter
?UV LED?
6
Low-pressure Mercury Arc UV Lampsvs UV Light
Emitting Diodes (LEDs)

• Low-pressure mercury-arc (LP) UV lamps
• tungsten electrodes and mercury vapor
• Lifetime 10,000 hours
• Cost 12 to 60
• Require on average forty (40) watts to operate
• Mercury health risks
• Fixed wavelength 253.7 nm
• UV light emitting diodes (LEDs)
• semiconductors gallium nitride (GaN) or indium
  gallium nitride (inGaN)
• Lifetime 60,000 - 100,000 hours
• Cost 10 and 20
• Require under one (1) watt of power to operate
• No or low environmental risks
• Variable wavelengths 225-280 nm

7
Objectives

• Compare the microbiocidal efficacy of UV
  mercury-arc lamps vs. UV LEDs
• E. coli (strain B)
• bacteriophage MS-2
• Virus of male-specific (F) E. coli bacteria
• Bacillus atrophaeus spores
• Characterize inactivation kinetics of the three
  test microorganisms

8
LP Lamp Collimated Beam

• Source Bolton and Linden 2003

9
UV LED, UV LED Array and UV LED UV Disinfection
Set-up
LED

• LED Array

LED Disinfection Set-up (Quasi-collimated Beam)
10
Exposure Time Determination for UV Dosing

• UV doses for experiments were chosen based on
  targets of 3 log10 reduction of spores, 4 log10
  reduction of viruses, and 6 log10 reduction of E.
  coli.
• t Exposure time (s)
• DCB UV dose (mJ/cm2)
• Es Average UV intensity (measured before and
  after sample exposure mW/cm2)
• Pf Petri factor (unitless)
• R Reflectance at the air-water interface at 254
  nm (unitless)
• L Distance from lamp centerline to suspension
  surface (cm)
• D Depth of suspension (cm)
• A254 UV absorbance at 254 nm (unitless)

11
Enumeration and UV Inactivation

• Irradiated samples for pre-determined time
• Diluted serially 10-fold immediately
• Plaque enumeration by spot plate method
• 18 hour incubation at 37 deg C
• CFU (E. coli B and B. atrophaeus spores) and PFU
  (MS-2 coliphage virus) enumeration
• Microbial reduction expressed as log10(Nx/N0) as
  a function of UV Dose (mJ/cm2)
• Nx concentration in CFU or PFU/mL after
  exposure to dose x, and
• N0 initial concentration in CFU or PFU/mL

12
Modeling UV Disinfection Kinetics

• Linear (Exponential) Mixed Model
• For each microorganism
• Three (3) null hypotheses
• overall line
• slope effect
• intercept effect
• Null hypothesis is not rejected when alpha gt 0.05
• Variability between technologies is not
  considered statistically significantly different

13
E. coli B Disinfection Results

• Not significantly different (p 0.90)

14
MS-2 Disinfection Results

• Not significantly different (p 0.71)

15
B. atrophaeus Disinfection Results

• Significantly different (p 0.002)

16
Comparative Disinfection Performance

• US EPA POU Rx reduction requirements
• Bacteria 6-log10
• Virus 4-log10
• Protozoan parasite 3-log10

Dosage (mJ/cm2) necessary to meet US EPA
requirements
17
UV Action Spectra 254 v 260 nm
LED closer to optimum absorbance
260 nm LED UV

• 254 nm
• LP UV

18
Conclusions

• UV LED disinfection of E. coli B and MS-2 was
  equivalent to that of low pressure mercury-arc
  lamp UV over a range of controlled doses
• UV LEDs performed significantly better than LP UV
  lamps in disinfecting B. atrophaeus
• Microorganism sensitivity to varying wavelengths
  of UV light provides a basis to optimize UV LED
  technology development
• UV LEDs can emit light over the entire wavelength
  range of interest (200-280 nm)

19
Global Application and Next Steps

• UV LED POU household water treatment is a
  potentially practical solution to achieve safe
  water for improved health in the developing world
  
• Challenges of some unimproved water sources
• Turbidity, soluble UV microbes, absorbance and
  aggregated
• Address these factors to optimize performance
• Settle, flocculate, filter (biosand filter,
  ceramic pot filter, membrane/cloth, etc.) to
  remove turbidity
• Design simple, appropriate, passive reactors
• Integrate into safe storage systems