Applications for Laboratory Column Tests in Evaluating ...
31
Applications for Laboratory Column Tests in Evaluating Arsenic Adsorption Media Paul Westerhoff Associate Professor Department of Civil & Environmental Engineering Arizona State University, Tempe, AZ November 2005
Transcript of Applications for Laboratory Column Tests in Evaluating ...
Applications for Laboratory Column Tests in Evaluating Arsenic Adsorption Media
Paul WesterhoffAssociate Professor
Department of Civil & Environmental EngineeringArizona State University, Tempe, AZ
November 2005
Outline
o To column test, or to batch test, that is an interesting question
o Rapid small scale column testso Incipient breakthrough column testso Dynamic modeling of column tests – a
new AwwaRF projecto Arsenic sensing needs for column tests
Arsenic Occurrence in Major Aquifers
Arsenic Occurrence in Mexico
0
2
4
6
8
Baja Nte Coahuila Nuevo Leon Tamaulipas
As
Con
cent
ratio
n, u
g/L
N = 22
N = 47 N = 39
N = 39
Arsenic Concentrations in 4 states are well below accepted health risk levels
0
10
20
30
40
50
60
70
Sonora
N = 154 samples61 cities were sampled
As
Con
cent
ratio
n, u
g/L
24.146
14.39
8.0488
2.6829
Arsenic concentrations in Sonora above US & WHO standards but below Mexican
Standards
0
50
100
150
200
Chihuahua
N = 154 Samples 27 cities were sampled A
s C
once
ntra
tion,
ug/
L
59.103
11.6096.3325
15.356
Arsenic Concentrations high in Chihuahua
Locations of Highest Arsenic Occurrence
0
50
100
150
200
250
300
Caborc
a
Hermos
illoLa
zaro
Carden
asOrib
e de A
lbaNue
voDeli
cias
El Sau
cillo
Jimen
ez
As
(ppb
)
City State As Avg As, ug/LCaborca Sonora 17.88 30.9Hermosillo Sonora 11.42 36Lazaro Cardenas Sonora 66.9 66.9Oribe de Alba Sonora 62.3 62.3NuevoDelicias Chihuahua 51.67 53.5El Saucillo Chihuahua 40.16 58.9Jimenez Chihuahua 87.1 266.6
To column test, or to batch test, that is an interesting question
o When are batch tests useful:o Mechanistic understanding of arsenic removalo Validating quality control during media productiono When enough batch tests are conducted in parallel with column tests,
rough correlations of full-scale performance can be estimated from batch tests
o What do laboratory column tests provide that batch tests do not:o Kinetic effects of particle sizeo Preloading effects from competing ionso Used material for leaching testso More “realistic” estimates of full-scale design life
o What do pilot tests provide that lab column tests can not provide:o Information of media attrition over time & backwashingo Exposure to media under variable influent groundwater conditions
and air/water temperatureso Pressure build-upo Ability to handle wastestreamso Operator education
What is a Rapid Small Scale Column Test (RSSCT)?
o Rapid Small Scale Column Tests (RSSCTs) were initially developed by Crittenden and others for evaluating organic compound removal on activated carbon
o Fundamentally, the concept is to scale the hydrodynamics and mass transport from full-sized media in a pilot or full scale reactor down to smaller test media in a small-scale bench-top continuous flow test
Advantages of RSSCTo RSSCTs can be conducted in a fraction of the time
required of pilot tests (1% to 10% of the time)o RSSCTs require less water than pilot tests, and
can be conducted under controlled laboratory conditions
o RSSCTs are generally cheaper than pilot testso RSSCTs are continuous flow tests and allow
evaluation of dynamic behavior and competition reactions that are more representative than batch tests
o RSSCTs were used during the USEPA ICR for organic carbon removal by GAC
o RSSCTs facilitate comparison of media and water quality effects
Arsenic Adsorption on Porous Media
E33r
∆r
Asb
R
Assq(r,t)
( ) ( ) ( )
∂
∂+
∂∂
=∂
∂rtrq
rrtrqD
ttrq
s,2,,
2
2
RSSCT Scaling Equations1. Empty bed contact time (EBCT) of the
small scale (sc) RSSCT column is proportional to that of a larger column (LC) based upon the ratio of the adsorbent media radius (R)
2. We have demonstrated that diffusivities are proportional to media radius
3. Loading rates (V) should be adjusted to minimize pressure loss and bed compaction. We recommend ReSCxSc of 2000
1
,
,
≈
=
LCs
SCs
LC
SC
DD
Log
RRLog
x
x
LC
SC
LC
SC
RR
EBCTEBCT
−
=
2
ScSc
RR
VV
LC
SC
SC
LC
LC
SC
××
×
=
ReRe
Supporting Publications:Badruzzaman et. al (2004), Water Research, 38 (18)Westerhoff et. al. (2005), J of Env. Engg-ASCE, 131(2) Badruzzaman et. al (2005), ACS Symposium Series, Vol-915
RSSCT Column Preparationo Crush full-scale mediao Weight sieve media – usually to
100x140 mesho Weigh prescribed mass and add to
columno Backwash column to remove fineso Tap column to compress bedo Add more media as needed to get to
prescribed mass and packed bed length
o Place effluent tube above media elevation to prevent air entrainment
o Condition with DI watero Start test
PD Scaling “works” with different GFH Mesh Sizes
0
2
4
6
8
10
12
- 10,000 20,000 30,000 40,000BV
Ars
enic
(ug/
L)
GFH (140x170)GFH (100x140)
RSSCT-PD vs Pilot
RSSCT Validation – a few of many (pilot- vs RSSCT-scale arsenic breakthrough)
0
5
10
15
20
0 10 20 30 40 50 60 70 80 90 100
Bed Volumes treated (Thousands)
Efflu
ent A
s (p
pb)
PD-RSSCTPilot Test
GFH
0
5
10
15
20
0 10000 20000 30000 40000 50000 60000 70000
Bed Volumes treated
Efflu
ent A
s(pp
b)
PD-RSSCT
Pilot
GFH
05
10152025303540
0 5000 10000 15000 20000
Bed Volumes treated
Efflu
ent A
s (p
pb)
RSSCTPilot
E33
0
2
4
6
8
10
12
- 20,000 40,000 60,000 80,000 100,000 120,000
Bed Volumes Treated
Efflu
ent A
rsen
ic (p
pb)
RSSCT: Spherical media(100x140 mesh)Pilot Test: spherical full-sizemedia
Arizona GroundwaterInfluent Arsenic = 12 ppb
ArsenX
Comparison of New Media
Influent As(V) = 20 to 22 ppbScottsdale 4E water, pH~8.4
0
5
10
15
20
25
- 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000Bed Volumes Treated
RS
SC
T E
fflue
nt C
once
ntra
tin (p
pb)
NEXTEnglehardE33Mel RS-AFDOWMel RS-AT
Batch tests with same media
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
-0.5 -0.3 -0.1 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5
Log Ce
Log
q [u
g A
s/m
g D
ow]
Dow E33 Melstream ATMelstream AF Englehard Next
Scottsdale 4E waterpH=8.6
Evaluation of New FeGAC Media (AwwaRF#3079)
0
5
10
15
20
25
30
0 10000 20000 30000 40000 50000 60000 70000 80000
Bed Volume
Ars
enic
in th
e E
fflue
nt (u
g/L)
GAC4BSOL3JacobiFe-Aerogel GAC
Modified Scottsdale GWAs Concentration = 24.8 ppbpH = ~ 7EBCT = 4 minMedia size = 100 x140
Evaluation of Agglomerated Nanoparticle Based Titanium Media
(Effect of very short EBCTs) AwwaRF#3077
0
5
10
15
20
25
0 5 10 15 20 25
Liters processed/gram dry media
Efflu
ent A
s(V)
con
c. (p
pb)
EBCT = 0.5 min (Flow = 22 mL/min, Re = 2.24, Re*Sc = 2000)EBCT = 0.5 min (Flow = 11 mL/min Re = 1.12, Re*Sc = 1000)EBCT = 0.25 min (Flow = 22 mL/min, Re = 2.24, Re*Sc = 2000)EBCT = 0.1 min (Flow = 22 mL/min, Re = 2.24, Re*Sc = 2000)
Scottsdale 4E100x140 Dow GTOInitial pH 7.83
Incipient breakthrough column testso Use very short columnso Can use full-size or crushed
mediao Short duration tests (< 1 day)o Initial breakthrough is related to
external mass transfer (not diffusion within the media)
o Provide kinetic information for modeling
o Facilitates studies of kinetic limitations of media
o Provide rapid assessment of media formulations or possibly evaluate remaining adsorption capacity of used media
Glass Beads: 15cm
Glass Wool: 1cm
Glass Beads: 8cm
O-Ring
Teflon Endcap
Overall Glass column
dimension 1.1cm x 30.5
cm
Direction of flow
Teflon Endcap
O-Ring
Adsorbent media: 0.5 to 2 cm
Glass Wool: 1cm
Example Data with Three Full-size Iron Adsorbents
0
20
40
60
80
100
120
1 10 100 1000 10000 100000
Bed Volume
Ars
enat
e R
emai
ning
(ug/
L)
Mod 1Mod 2Mod 3
NSF53 Challenge waterAs(V)~100ppbpH= 8.5 1 cm-media (10x40 mesh)
Incipient Breakthrough Data with 100x140 mesh media (AwwaRF #3098)
Iron Based Porous mediaMesh size = 100x140pH 8.5 with 10 mM NaHCO3Initial As(V) = 88 ppb
0
10
20
30
40
50
60
70
1 10 100 1000 10000Bed volumes Treated
Effl
uent
Ars
enic
(p
pb) 1cm
2cm5cm
Dynamic modeling of column testsA New AwwaRF project (#3098)
o Goals:o Surface complexation modeling is the appropriate technique for
characterizing single and multi-solute adsorption equilibrium in these systems.
o While a crucial first step, adsorption equilibrium relationships alone are inadequate for predicting the performance and bed life of a continuous-flow packed bed, and verification of surface complexation models must include field-scale validation.
o Thus, a surface complexation model must be integrated into a dynamic mass transport model.
o Project Tasks1. Compile Existing Literature2. Single Solute Arsenate Adsorption Isotherms (E33, GFH, GTO)3. Multi-Solute Adsorption Isotherms4. Surface Complexation Model Parameters5. Packed-Bed Dynamic Mass Transport Model6. Model Validation
Link Equilibrium Surface Complexation Models
( ) ( ) ( )
∂
∂+
∂∂
=∂
∂r
tr,qr2
rtr,qD
ttr,q
2
2
s
No. Reaction Type Log K
1. ( ) ( ) ( ) OHAsOFeHAsOHFeOH saqs 24243 +↔≡+≡ Inner-sphere 4.52
2. ( ) ( ) ( ) ( ) OHHFeHAsOAsOHFeOH aqsaqs 2443 ++↔≡+≡ +− Inner-sphere 2.6
3. ( ) ( ) ( ) ( ) OHHFeAsOAsOHFeOH aqsaqs 2
2443 2 ++↔≡+≡ +− Inner-sphere -2.47
4. ( ) ( ) ( ) OHAsOFeHAsOHFeOH saqs 23233 +↔≡+≡ Inner-sphere 4.52
5. ( ) ( ) ( ) ( ) OHHFeHAsOAsOHFeOH aqsaqs 2333 ++↔≡+≡ +− Inner-sphere -2.70
with Dynamic Mass Transport Models
Arsenic sensing needs for column tests
o Column testing could be facilitated by semi-continuous, autonomous arsenic detection
o Commercialized systems are being developed (TraceDetect)
o ASU is developing its own micro-fiber ASV system
Preliminary Data in Arizona Groundwater
Summaryo Batch and column tests have their place and should
be used to address specific questionso Rapid small scale column tests (RSSCTs) are
relatively easy to conduct in parallel and are useful in evaluating different media, variable water quality, or simply screening effects of competing ions in a given water
o Incipient column tests are very rapid, but do not necessarily simulate a full-scale application
o Linking equilibrium models with dynamic packed bed transport models will likely shed new light on adsorption competition effects and effects of changes in surface chemistry on arsenic adsorption
o On-line arsenic sensors are necessary to automate lab column testing and field system monitoring
Paul Westerhoff, Ph.D., PEArizona State University
Associate ProfessorDepartment of Civil & Environmental Engineering
http://ceaspub.eas.asu.edu/pwest/[email protected]
Acknowledgements
Mohammad Badruzzaman, Troy Benn
USEPA (Tom Sorg) & AwwaRF
Battelle (LiLi Wang, Ab Chen)
Cities of Scottsdale, Mesa & Phoenix
NCS Consulting & Damon S. Williams Inc.
Salt River Project & Joseph Wang
ASU / National Science Foundation Water Quality Center
