AMMONIUM, NITRATE AND PHOSPHATE - Cornell University

AMMONIUM, NITRATE AND PHOSPHATE SORPTION TO WATER-RINSED AND NON-RINSED BIOCHARS

A Thesis

Presented to the Faculty of the Graduate School of

Cornell University

In Partial Fulfillment of the requirements for the Degree of

Master of Science

By

Charles Colin Hollister

August 2011

ABSTRACT

Water-rinsed and non-rinsed biochar (BC) was evaluated for nitrogen (N)

and phosphorus (P) removal from aqueous solution in order to quantify its

nutrient pollution mitigation potential in agroecosystems. Sorption isotherms

were prepared for nutrient solutions of ammonium (NH4), nitrate (NO3) and

phosphate (PO4) using BC of corn and oak feedstock, each pyrolyzed at 350º and

550º C highest treatment temperature (HTT). NH4 sorption was observed for

rinsed Corn-BC and rinsed Oak-BC and a 200º C increase in HTT decreased NH4

sorption for both feedstocks. Nitrate sorption was not observed in any of the

rinsed or non-rinsed chars, which indicates a lack of anion exchange capacity

(AEC) for nitrate removal. More P sorption was observed using non-rinsed Oak-

BC relative to rinsed Oak-BC and no phosphate sorption was observed using

rinsed or non-rinsed Corn-BC. A reduction of phosphate removal efficiency for

the rinsed Oak-BC coincided with a loss of calcium (Ca) due to leaching during

the pretreatment. Water leachable Ca increased with HTT for all BC, which

suggests Ca mineral thermal transformation to a more soluble form. Low HTT

BC leached more magnesium (Mg) and P and Corn-BC leached more Mg and P

than Oak-BC. The transition to a micropore dominated surface area with

increasing HTT could restrict nutrient leaching compared to lower HTT where

mesopore surface area is abundant.

iii

BIOGRAPHICAL SKETCH

Colin Hollister was born in Pittsfield, Massachusetts August 5th, 1984. He

attended Berkshire Country Day School (Stockbridge, MA) through the ninth

grade and graduated in 1999. He continued onto Avon Old Farms School (Avon,

CT) where he was inducted to the Cum Laude Society and graduated in 2002.

Colin entered Bates College (Lewiston, ME) and pursued a Bachelor of Science in

Environmental Studies, concentrating in Ecology and minoring in French. In

2004 he was awarded an Otis Fellowhip to thru-hike the Appalachian Trail in

order to study, photograph and become a part of thru-hiker culture. In 2005 he

spent five months in Niger as part of a French study abroad program, where he

received a service-learning grant from Bates to organize local youth who painted

a large mural to raise awareness of important health issues. His independent

thesis research, “Surface soil heavy metal concentrations and cation

exchangeability across a modeled atmospheric deposition gradient in Acadia

National Park, ME,” introduced him to sorption chemistry and mechanisms of

pollutant transport. After graduating from Bates in 2006 he relocated to Munich,

Germany where he became a Math Support Teacher, Environmental Coordinator

and Alpine Ski Coach for the Munich International School. In 2009 he moved

back to the United States and enrolled in the Environmental Processes program

in Civil and Environmental Engineering at Cornell University (Ithaca, NY) where

he pursued a Master of Science degree.

iv

This research is dedicated to sustainable engineering

v

ACKNOWLDGEMENTS

I would like to thank Jim Bisogni for his insightful guidance and

unwavering support as we endeavored on a path of novel research. My study

would not have been possible without the collaboration and cooperation of

Johannes Lehmann of the Crop and Soil Sciences department and his willingness

to provide the biochars used in the experiments. Akio Enders and Kelly Hanley

of the Lehmann research group have been invaluable sources of help and

information around the lab. Special thanks to Jim Gossett for allowing me to use

the ion chromatograph and to Larry Geohring for allowing Shree Giri to assist

me in analyzing samples in the Biological and Environmental Engineering

department Soil and Water Lab.

vi

TABLE OF CONTENTS

Biographical Sketch……………………………………………………………. iii Dedication………………………………………………………………………. iv Acknowledgements……………………………………………………………. v Table of Contents………………………………………………………………. vi List of Figures…………………………………………………………………... vii List of Tables……………………………………………………………………. viii List of Abbreviations…………………………………………………………... ix

Introduction…………………………………………………………………........ 1 Materials and Methods…………………………………………………………. 3

Background…………………………………………………………………. 3 Preparation of the biochars……………………………………………….. 4 Rinse procedure and water leachable nutrients………………………… 4 Sorption experiments……………………………………………………… 7

Results…………………………………………………………………………….. 10 Water leachable nutrients…………………………………………………. 10 Ammonium sorption results……………………………………………… 15 Nitrate sorption results……………………………………………………. 17 Phosphate sorption results………………………………………………... 18

Discussion and Conclusion……………………………………………………. 20 Leachability of nutrients…………………………………………………... 20 Sorption experiments……………………………………………………… 24

Appendix…………………………………………………………………………. 28 References………………………………………………………………………… 40

vii

LIST OF FIGURES

1. Desorption isotherms showing the molar P or Mg contents of Corn-BC and Oak-BC versus the equilibrium concentrations of P or Mg in rinse solutions…………………………………………………………. 13

2. A plot of Ca concentration in the rinse solutions versus Ca content in the char for Corn and Oak-BC……………………………………….. 15

3. A linearized Freundlich isotherm for ammonium adsorption to rinsed chars……………………………………………………………….. 16

4. Phosphate sorption isotherms for rinsed and non-rinsed Corn-BC and Oak-BC………………………………………………………………. 19

viii

LIST OF TABLES

1. Physical and chemical properties of corn and oak biochars………… 4 2. Data gathered from the rinse of each BC. Elemental concentrations

and ratios throughout the rinse series were based on the initial mean concentrations of each element, for each BC…………………… 11

3. Regression line data for P and Mg desorption isotherms for the Corn-BC and Oak-BC rinse series……………………………………… 14

4. Freundlich model terms for ammonium adsorption to rinsed BC….. 17 5. Results of two-tailed t-tests comparing Ce to C0 for nitrate sorption

experiments, calculated at the 5% level………………………………... 17 6. Linear regression results for phosphate sorption isotherms………… 20

ix

LIST OF ABBREVIATIONS

Å angstrom, one hundred-millionth of a centimeter or 10-10 meter AEC anion exchange capacity AFO animal feeding operation

BC biochar Ca calcium

CAFO concentrated animal feeding operation Ce equilibrium concentration

CEC cation exchange capacity C0 initial concentration

CO2 carbon dioxide Corn-350-BC biochar made from corn stover pyrolyzed at 350º C Corn-550-BC biochar made from corn stover pyrolyzed at 550º C

DI deionized water EPA United States Environmental Protection Agency FTIR Fourier transform infrared spectroscopy

g gram HCl hydrochloric acid HTT pyrolysis highest treatment temperature

IC ion chromatograph ICP inductively coupled plasma spectrometer

L liter m meter

m- milli-, one thousandth or 10-3 Mg magnesium mol mole

N nitrogen N2 nitrogen gas

NaOH sodium hydroxide NH4 ammonium

NMP nutrient management plan NO3 nitrate

NPDES national pollution discharge elimination system º C degrees Celcius

Oak-350-BC biochar made from oak wood pyrolyzed at 350º C Oak-550-BC biochar made from oak wood pyrolyzed at 550º C

P phosphorus PO4 phosphate rpm rotations per minute

SPF-BC biochar made from a mixture of spruce, pine and fir wood XRD X-ray diffraction

µ- micro-, one millionth or 10-6

1

Introduction

In recent decades, scientists have observed increases in aquatic primary

production alongside N and P over enrichment [1, 2]. This enrichment can cause

eutrophication, which diminishes biodiversity and disrupts food webs by

altering physical and chemical characteristics of aquatic ecosystems [3-5].

Compounding the problem, algae can produce toxins that have paralytic,

diarrheic and neurotoxic effects with negative impacts on animals, humans and

aquatic species [6, 7]. These toxins can reach high concentrations during an algal

bloom impacting reproduction of aquatic species, fishing industries and coastal

economies [8]. The U.S. Environmental Protection Agency (EPA) monitors

watersheds across the nation and have determined that nutrients are a leading

cause of impaired waters [9]. To date, about a quarter of the rivers and streams

in the U.S. (1,500,000 km) have been assessed, of which half are impaired. The

primary source of the impairment is agriculture, with nutrients being a leading

cause. In attempt to reduce N and P discharge from nonpoint sources, the EPA

encourages farmers to employ strategies such as conservation farming,

vegetative buffers and better fertilizer management to mitigate nutrient pollution

[10].

Biochar (BC) is a carbonaceous geosorbent that is produced by thermal

degradation of organic material during pyrolysis [11]. The utilization of BC as a

chemical sorbent in agroecosystems has both economic and environmental

incentives: BC is generated from waste biomass through a process that can

produce energy and sequester carbon. The pyrolysis reaction is exothermic and

releases gasses from the biomass, yielding both immediate thermal energy and

biofuels that can be captured and stored [12]. Carbonization of biomass increases

2

carbon aromaticity [13-15] and recalcitrance relative to its thermally unaltered

state, which helps mitigate the release of carbon dioxide (CO2) into the

atmosphere through decomposition. The rate of CO2 uptake of living biomass

far exceeds the rate of CO2 released from BC when added to soil [16], making

biochar production an attractive strategy for reducing greenhouse gas emissions.

As a geosorbent BC shows strong affinity for organic compounds [17] and has

been shown to reduce N and P leaching from highly weathered soils [18, 19]. BC

is known to have high cation exchange capacity (CEC) with the ability to adsorb

NH4 [20], has demonstrated NO3 adsorption [21-24] and BC made from

anaerobically digested sugar beet tailings exhibited high PO4 removal from

aqueous solution [25]. Yet, the role of feedstock type, HTT and water rinsing

pretreatment of BC on inorganic nutrient sorption has yet to be systematically

elucidated.

Comparing the sorption capacity of various BCs for inorganic nutrients is

useful in understanding the BC characteristics that may contribute to reduced

nutrient leaching from BC amended soil receiving N and P fertilizer. Soil

application of manure and process wastewater is an option for concentrated

animal feeding operations (CAFOs) that are required to develop a nutrient

management plan (NMP) as part of the national pollution discharge elimination

system (NPDES) permit. The EPA regulates pollution discharge from animal

feeding operations (AFOs) based on the type and number of animals on the

property and whether or not pollutants are discharged into U.S. waters [26]. An

AFO holding more than 200 mature dairy cows, 300 cattle or 750 mature swine

may be classified as a CAFO, which the EPA regulates as a point source of

nutrient pollution [27]. Terms of the NMP require the CAFO to determine the

3

maximum application rates of N and P to soil that does not cause breakthrough

of pollutants to nearby waters even after multi-year application. In order to

evaluate how BC might achieve N and P effluent regulations under successive

applications of CAFO process wastewater, the sorption capacity of water-rinsed

BC for inorganic nutrients must be identified.

The objectives of this study are to determine: (1) how BC feedstock and

HTT influence water leachable nutrients, and (2) the effect of labile ions on

inorganic N and P sorption.

Materials and Methods

Background

Corn stover (Zea mays L.) and oak wood (Quercus ssp.) feedstocks were

slow-pyrolyzed at 350º and 550º C by BEST Energies, Inc. (Daisy Reactor, WI,

USA). These are referred to as Corn-350-BC, Corn-550-BC, Oak-350-BC and Oak-

550-BC. A previous study [28] analyzed elemental concentrations (Ca, H, Mg, N,

O, P) of these BCs including surface area and pore volume of Corn-BC and Oak-

BC pyrolyzed at 400º and 600º C HTT using nitrogen gas (N2) and CO2

adsorbates. Another study [29] performed Fourier transform infrared

spectroscopy (FTIR) on the same corn and oak feedstocks pyrolyzed at 350º and

600º C HTT. Data obtained from these previous experiments will be used to

characterize the BCs of this paper’s study. In cases where specific data is not

available for corn and oak pyrolyzed at 350º and 550º C HTT, data obtained from

corn and oak pyrolyzed at 400º and 600º C HTT will be used as proxy values,

respectively. A summary of the physical and chemical differences among the

temperature BCs is presented in Table 1 and in the Appendix.

4

Table 1: Physical and chemical properties of Corn-BC and Oak-BC. Surface area and pore volume were measured with N2 and CO2 sorbates. Mean elemental concentrations are expressed with standard deviation [± SD], n = 3. Asterisks (*) indicate that data for the exact HTT of a given feedstock is unavailable and the next closest HTT temperature (400º or 600º C) is reported instead. Data is compiled from a previous study [28].

Surface area measurements Elemental concentration (µmol g-1)

BC type

N2 sorbate surface

area (m2 g-1)

N2 sorbate

pore volume (mL g-1)

CO2 sorbate surface

area (m2 g-1)

CO2 sorbate

pore volume (mL g-1)

Ca Mg P

Corn-350-BC 6.7* 0.010* 281.9* 0.050* 61.00 [2.33] 153.11 [4.24] 259.48 [0.92] Corn-550-BC 4.8* 0.006* 527.1* 0.100* 67.58 [7.19] 244.62 [7.93] 365.80 [4.16] Oak-350-BC 3.5* 0.001* 449.8* 0.090* 0.36 [0.36] 27.36 [1.53] 1.69 [0.23] Oak-550-BC 0.9* 0.007* 547.8* 0.095* 0.94 [0.53] 40.14 [2.34] 1.71 [0.17]

Preparation of the biochars

All BCs were gently crushed through sieves with a ceramic pestle so that

BC particles fell between 850 µm and 500 µm mesh sizes. Most BC particles were

long, flat flakes with a length several times their width. Corn-BC occupied more

volume compared to an equal mass of Oak-BC. Corn-BC was also more brittle

than Oak-BC, which was evident while crushing the material through sieves. All

glassware was acid washed in a hydrochloric acid bath (1 mol L-1) and rinsed

with deionized water (DI) prior to use.

Rinse procedure and water leachable nutrients

The “rinsed” BC was prepared by rinsing each BC repetitively with DI,

which was then analyzed to determine water leachable nutrients. The BC

samples were continuously stirred for 16 hours with DI and a magnetic stir bar in

5

an Erlenmeyer flask. The Corn-BCs required a 2 L flask whereas the Oak-BCs

required a 1 L flask in order to obtain slurry that could be adequately mixed with

a stir bar. The approximate weight ratio of BC to DI was 1:45 for Corn-350 and

Corn-550, 1:20 for Oak-350 and 1:40 for Oak-550. For each rinse, the BC slurry

was stirred for 16 hours then passed through a 0.45 µm vacuum filter apparatus.

The BC material that remained on the filter apparatus was rinsed back into the

Erlenmeyer flask and was filled to capacity with DI. This procedure was

repeated until the pH of the filtrate stabilized between two consecutive rinses (±

pH 0.02). The final dry weights of Corn-350-BC, Corn-550-BC, Oak-350-BC and

Oak-550-BC at the end of the rinse series were 37.71 g, 37.18 g, 42.89 g and 22.87 g

respectively.

The pH of the rinse solutions was measured immediately after filtering

and approximately 300 mL of filtrate from each rinse was stored in an airtight

container at 4º C until analyzed for elemental and nutrient concentrations (4-16

weeks after filtering). Elemental composition (Ca, Mg, P) was determined using

an inductively coupled plasma spectrometer (ICP; Thermo Jarrell Ash ICAP 61E

Trace Analyzer). Ammonium concentration was determined using the phenate

colorimetric method [30] with a Hewlett Packard diode array UV-Vis

spectrophotometer. Nitrate was measured with a Dionex ion chromatograph

(IC) installed with an IONPAC® AS14A analytical column and PO4 was measured

with an OI analytical FS 3000 phosphorus analyzer (P-flow analyzer). A quality

control on the P analysis methods confirmed that P measured by the ICP

correlated strongly to P measured by the P-flow analyzer.

After the pH of each rinse solution stabilized, the BC remaining on the

filter apparatus was rinsed into a beaker which was then placed into an oven at

6

110º C for several days until dry. The beakers were covered loosely with

aluminum foil to reduce the risk of cross contamination. Once the BCs were

sufficiently cool, the dry mass of each BC was determined.

The concentration of Ca, Mg and P in the BCs over the course of the rinse

series was determined by mass balances. It is assumed that BC mass was

conserved throughout the rinse and that all nutrient mass leached from the BC

went into solution and was measured:

!!"#$%! = !!! − !!"#$!!"

! (1)

Where: i = element being measured Mfinal = total mass of element in BC at the end of a given rinse (µmol) Mo = initial mass of element in BC (µmol) Mleached = mass of element measured in solution at the end of each rinse (µmol)

The concentration of Ca, Mg and P in the BCs was calculated by dividing

the calculated mass of each element by the dry weight of the BC:

!!"! =!!"#$%!

!!" (2)

Where: !!"! = element (i) concentration in BC at the end of a given rinse (µmol g-1) Mfinal = total mass of element (i) in BC at the end of a given rinse (µmol) MBC = dry mass of BC used (g)

To determine the cumulative percent of element removed, the element

concentration of the rinsed BC is subtracted from the initial element

concentration of the BC and then divided by the initial element concentration of

the BC:

7

!!"#$%"&% =!!"!! !!!"

!

!!"!! ×100 (3)

Where: iremoved % = percent of element (i) removed at the end of a given rinse !!"!! = initial concentration of element (i) in BC (µmol g-1) !!"! = element (i) concentration in BC at the end of a given rinse (µmol g-1) Sorption experiments

Preliminary sorption experiments were performed using non-rinsed BC

prepared by Alterna Biocarbon (Prince George, British Columbia) from a mixture

of softwoods (spruce, pine and fir; referred to as “SPF-BC”) slow-pyrolyzed at

400º C. The preliminary experiments were used to evaluate BC to liquid ratios

and to obtain preliminary sorption results for NH4, NO3, and PO4 (Appendix).

The sorption experiments in this paper’s study were performed in triplicate at

24º C ± 2º C with a Burrell wrist action shaker at 400 rpm. 0.5 g of BC was added

to Corning 50 mL centrifuge tubes along with 50 mL of liquid (DI or nutrient

solution depending on experiment). The tubes were shaken for 24 hours and

then left quiescent for 24 hours before the supernatant was passed through a 0.45

µm vacuum filter apparatus.

Ammonium solutions were prepared using anhydrous ammonia chloride

and serial dilutions were made to obtain concentrations of 1; 10; 100; 1,000 and

10,000 mg NH4-N L-1. A set of BC mixed with DI was also included to measure

NH4 leached from 24 hours of continuous shaking. Only rinsed Corn and Oak-

BC was used for these sorption experiments and equilibrium concentrations

were measured using the phenate colorimetric method, analyzed on the UV-vis

spectrophotometer. Data was analyzed using the linearized Freundlich

adsorption model.

8

Nitrate sorption experiments were performed on both rinsed and non-

rinsed Corn-350-BC, Corn-550-BC and Oak-350-BC (a limited supply of Oak-550-

BC prevented it from being used in the nitrate sorption experiments). A solution

of 10 mg NO3-N L-1 was prepared from anhydrous potassium nitrate. These BCs

were also mixed with DI to quantify nitrate leached from 24 hours of continuous

shaking. The equilibrium solutions were analyzed for NO3 concentration using

the IC. Two-tailed t-tests are used to compare Ce to C0 and will inform whether

to accept or reject the null hypothesis that Ce and C0 are not different.

PO4 sorption experiments were performed on all rinsed-BC using pH

adjusted solutions. Two stock PO4 solutions were made from dibasic anhydrous

potassium phosphate (KH2PO4) and series dilutions were performed to obtain

two sets of identical solutions with concentrations of 0.1, 1 and 10 mg P L-1. One

set of concentrations was adjusted to pH 4 (± 0.02) using HCl and NaOH as

needed. The other set was buffered with sodium bicarbonate (NaHCO3) and

adjusted to pH 7 (± 0.02) with HCl and NaOH as needed. Water adjusted to pH

4 and pH 7 was also mixed with each rinsed-BC to measure P-leaching. For non-

rinsed Corn-350-BC, Corn-550-BC, Oak-350-BC and Oak-550-BC sorption

experiments were performed using 10 mg PO4-P L-1 and water solutions, each

buffered to pH 7. All equilibrium concentrations were determined using the P-

Flow analyzer.

Sorption isotherms are derived from experimental data and describe how

much solute adsorbs to the sorbate at a given equilibrium solute concentration.

Desorption isotherms were constructed based on the nutrient leaching

information gathered during the biochar rinsing series. Adsorption isotherms

were made for NH4 sorption to the rinsed BC, but these isotherms are considered

9

“modified” because the rinsed BC may not be entirely void of sorbate at the

beginning of the experiment. The Freundlich adsorption isotherm is used to

model NH4 sorption:

! = !!!!/! (4)

Where: S= mass of adsorbate per mass of adsorbent at equilibrium KF= experimentally derived constant C = concentration in solution 1/n = experimentally derived constant

S is determined by measuring the concentration of the equilibrium

solution and subtracting that from the initial concentration, times solution

volume (Equation 5). It is then normalized by adsorbent mass. Since the rinsed

BC is starting with an unknown mass of solute on its surface, q* represents the

additional mass of solute sorbed per mass BC:

!∗ = (!! − !!)×!

!!" (5)

Where: q*= additional mass adsorbed at equilibrium per mass sorbent (µMol g-1) Co= initial concentration of solution (µMol L-1) Ce = measured equilibrium concentration (µMol L-1) V = volume of solution (L) MBC = mass of BC used in experiment (g)

A least square regression of log(q*) versus log(Ce) gives:

log !∗ = log !!∗ + !!log (C!) (6)

Where: log(!F∗)= y-intercept n-1= the slope

10

Results Water leachable nutrients

The pH of sequential rinse solutions for each BC progressed towards

neutrality, albeit with fluctuations (Table 2). The Oak-350-BC showed the largest

change between the initial (pH 4.32) and final (pH 5.88) values whereas Corn-

350-BC showed the least change in pH between initial (pH 7.95) and final (pH

7.78) rinses.

The N leached during the rinse series was minimal. Ammonium leaching

was often below detection limits, yet each BC leached a small amount with Corn-

BC leaching slightly more than Oak-BC. Nitrate leaching was not measured in

any sample, as none of the rinse solutions exceeded 0.2 mg NO3-N L-1, which was

the detection limit of the IC.

In the course of rinsing, at the point where the mass of rinse water used

was about 200 times the mass of BC (Corn-350-BC rinse #4; Corn-550-BC rinse

#4; Oak-350-BC rinse #9; Oak-550-BC rinse #5), the low HTT BCs released a

larger percentage of P than the high HTT BCs (Table 2). Both Corn-BCs leached a

larger mass of P than the Oak-BCs. Corn-BC showed a larger discrepancy in

percentage of leachable P based on HTT, with Corn-350-BC leaching a total of

61% of initial P whereas Corn-550-BC leached a total of 24%. Oak-350-BC and

Oak-550-BC leached 49% and 26% of initial P respectively.

11

Table 2: Data gathered from the rinse of each BC. The initial mean elemental concentration for each BC is compiled from a previous study [28] with standard deviation [± SD], n=3. Elemental concentrations and ratios throughout the rinse series were based on the initial mean concentrations of each element, for each BC.

BC sample Rinse number

pH of rinse

solution

Cumulative volume of water used to rinse (L)

Mass of water used per mass BC (g g-‐1)

Mass NH4-‐N leached per mass BC (µmol g -‐1)

Mass NO3-‐N leached per mass BC (µmol g -‐1)

Elemental content of biochar at end of rinse (µmol g-‐1) Cumulative percent of element leached

P Ca Mg Ca:Mg P Ca Mg

Corn-‐350-‐BC -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ 61.00 [2.33] 153.11 [4.24] 259.48 [0.92] 0.59 -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ Corn-‐350-‐BC 1 7.95 1.54 42 0.89 <1.60 36.69 149.21 213.03 0.70 40% 3% 18% Corn-‐350-‐BC 2 7.93 3.29 90 <0.65 <1.60 29.18 146.47 198.39 0.74 52% 4% 24% Corn-‐350-‐BC 3 7.62 4.86 132 <0.65 <1.60 25.99 145.47 193.86 0.75 57% 5% 25% Corn-‐350-‐BC 4 7.77 6.40 174 <0.65 <1.60 23.62 144.05 187.53 0.77 61% 6% 28% Corn-‐350-‐BC 5 7.75 8.19 223 <0.65 <1.60 22.12 142.91 182.91 0.78 64% 7% 30% Corn-‐550-‐BC -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ 67.58 [7.19] 244.62 [7.93] 365.80 [4.16] 0.67 -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ Corn-‐550-‐BC 1 9.76 1.70 47 0.69 <1.60 59.88 239.46 331.63 0.72 11% 2% 9% Corn-‐550-‐BC 2 9.52 3.47 96 0.88 <1.60 54.92 236.48 318.36 0.74 19% 3% 13% Corn-‐550-‐BC 3 9.5 5.27 146 <0.65 <1.60 52.65 233.15 311.17 0.75 22% 5% 15% Corn-‐550-‐BC 4 9.11 7.01 193 <0.65 <1.60 51.23 229.87 306.88 0.75 24% 6% 16% Corn-‐550-‐BC 5 9.06 8.68 240 1.49 <1.60 49.40 224.69 300.46 0.75 27% 8% 18% Corn-‐550-‐BC 6 9.06 10.52 290 <0.65 <1.60 48.15 219.26 295.90 0.74 29% 10% 19% Oak-‐350-‐BC -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ 0.36 [0.43] 27.36 [1.53] 1.69 [0.23] 16.19 -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ Oak-‐350-‐BC 1 4.32 0.91 22 0.36 <0.75 0.36 23.98 1.62 14.82 0% 12% 4% Oak-‐350-‐BC 2 4.5 1.78 43 <0.30 <0.75 0.32 22.64 1.57 14.46 11% 17% 7% Oak-‐350-‐BC 3 5.01 2.68 64 <0.30 <0.75 0.30 21.85 1.56 14.01 16% 20% 8% Oak-‐350-‐BC 4 5.25 3.56 85 <0.30 <0.75 0.29 21.23 1.55 13.70 21% 22% 8% Oak-‐350-‐BC 5 5.7 4.47 107 <0.30 <0.75 0.27 21.01 1.55 13.60 26% 23% 9% Oak-‐350-‐BC 6 5.25 5.36 128 0.32 <0.75 0.26 20.63 1.54 13.42 29% 25% 9% Oak-‐350-‐BC 7 5.59 6.25 149 <0.30 <0.75 0.22 20.05 1.53 13.09 38% 27% 9% Oak-‐350-‐BC 8 5.88 7.09 170 <0.30 <0.75 0.20 19.53 1.52 12.82 44% 29% 10% Oak-‐350-‐BC 9 5.88 8.02 192 <0.30 <0.75 0.18 19.08 1.51 12.61 49% 30% 10% Oak-‐550-‐BC -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ 0.94 [0.53] 40.14 [2.34] 1.71 [0.17] 23.47 -‐-‐-‐ -‐-‐-‐ -‐-‐-‐ Oak-‐550-‐BC 1 6.81 1.01 45 0.68 <1.50 0.91 37.68 1.69 22.34 4% 6% 1% Oak-‐550-‐BC 2 7.29 1.92 86 0.62 <1.50 0.87 31.67 1.67 18.94 8% 21% 2% Oak-‐550-‐BC 3 7.46 2.85 128 <0.50 <1.50 0.80 27.26 1.65 16.49 15% 32% 3% Oak-‐550-‐BC 4 7.14 3.81 171 <0.50 <1.50 0.77 23.97 1.65 14.50 18% 40% 3% Oak-‐550-‐BC 5 6.74 4.47 201 <0.50 <1.50 0.70 23.23 1.61 14.42 26% 42% 6% Oak-‐550-‐BC 6 6.89 5.31 238 <0.50 <1.50 0.67 22.10 1.61 13.76 29% 45% 6% Oak-‐550-‐BC 7 6.91 6.19 278 <0.50 <1.50 0.62 21.48 1.61 13.38 34% 46% 6%

12

The molar P content of Corn-BC decreases with the equilibrium

concentration of P in the rinse solution (Figure 1A). The P content of Corn-350-

BC approached a value of about half that of Corn-550-BC due to a large release of

P in the first rinse, after which the solution equilibrium concentrations of P were

quite similar between the two BCs. Corn-BC shows a trend of decreasing P

content with decreasing equilibrium concentration (P<0.05; Table 3) yet Oak-BC

did not (P>0.05; Table 3).

The low HTT BCs released a higher percentage of their initial Mg content

relative to the high HTT BCs (Table 2). As the ratio of rinse water to BC reached

200, Corn-350-BC leached 28% and Corn-550-BC leached 16% of their initial Mg

content, resulting in a 72 µmol g-1 and 59 µmol g-1 reduction respectively. At

similar rinse water to BC ratios, Oak-350-BC leached 10% and Oak-550-BC 6% of

their initial Mg content, which translates to a 0.18 µmol g-1 and 0.10 µmol g-1

reduction respectively. Corn-BC and Oak-350-BC demonstrate a decrease of Mg

content in the BC as solution concentration of Mg decreases (P<0.05; Table 3)

whereas Oak-550-BC does not show a clear relationship (P>0.05; Table 3).

13

A

B

C

D

Figure 1: Desorption isotherms showing the molar P or Mg contents of Corn-BC and Oak-BC versus the equilibrium concentrations of P or Mg in rinse solutions.

0

10

20

30

40

50

60

70

0 200 400 600 800

P content (µm

ol g-‐1)

Ce (µmol L-‐1)

Corn-‐350-‐BC Corn-‐550-‐BC

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

0 1 2 3

P content (µm

ol g-‐1)

Ce (µmol L-‐1)

Oak-‐350-‐BC Oak-‐550-‐BC

0

50

100

150

200

250

300

350

400

0 200 400 600 800 1000 1200

Mg content of Oak-‐BC (µmol g-‐1)

Ce (µmol g-‐1)

Corn-‐350-‐BC Corn-‐550-‐BC

1.50 1.52 1.54 1.56 1.58 1.60 1.62 1.64 1.66 1.68 1.70

0 1 2 3 4

Mg content of Oak-‐BC (µmol g-‐1)

Ce (µmol g-‐1)

Oak-‐350-‐BC Oak-‐550-‐BC

14

Table 3: Regression line data for P and Mg desorption isotherms for the Corn-BC and Oak-BC rinse series.

BC type Element i !!"! = ! ∗ !!! + ! R2 P-level

Corn-350-BC P !!"! = 0.02 ∗ !!! + 23.16 0.92 0.01

Corn-550-BC P !!"! = 0.07 ∗ !!! + 47.69 0.92 <0.01

Oak-350-BC P !!"! = 0.02 ∗ !!! + 0.23 0.05 0.60

Oak-550-BC P !!"! = −0.05 ∗ !!! + 0.81 0.03 0.53

Corn-350-BC Mg !!"!" = 0.02 ∗ !!

!" + 186.40 0.84 0.03

Corn-550-BC Mg !!"!" = 0.05 ∗ !!

!" + 299.27 0.80 0.02

Oak-350-BC Mg !!"!" = 0.02 ∗ !!

!" + 1.53 0.68 0.01

Oak-550-BC Mg !!"!" = −0.01 ∗ !!

!" + 1.64 0.01 0.88

By the end of the fourth rinse, Corn-350-BC and Corn-550-BC both leached

6% of their initial Ca content, resulting in a decrease of Ca content of 9 and 15

µmol g-1 respectively (Table 2). As the ratio of rinse water to BC reached 200, Ca

leaching for Oak-350-BC (rinse #9) was 30% and Oak-550-BC (rinse #5) was 42%,

resulting in a Ca content decrease of 8 and 17 µmol g-1 respectively. The Ca

solution concentrations of the rinse series do not appear to be governed by Ca

content of the BCs, which differ by feedstock. The equilibrium solution Ca

concentrations appear to be grouped based on HTT (Figure 2). The low HTT BCs

show very similar Ca solution concentrations regardless of Ca content, as do the

high HTT BCs.

15

Figure 2: A plot of Ca concentration in the rinse solutions versus Ca content in Corn and Oak-BC. Oak-BC Ca content is on the primary y-axis and Corn-BC Ca content is on the secondary y-axis.

The molar ratios of Ca content to Mg content show that Oak-BC sustains a

much higher ratio over the course of the rinse series than Corn-BC, yet the Ca:Mg

ratio increases for Corn-BC while it decreases for Oak-BC (Table 2). Oak-BC

shows a larger change in Ca:Mg ratio than Corn-BC, which results from

moderately high Ca content relative to low Mg content and the Ca leaches

substantially more than Mg.

Ammonium sorption results

All of the rinsed BCs exhibited ammonium sorption but lower HTT BC

showed a higher sorptive capacity for NH4-N relative to the higher HTT BC of

the same feedstock (Figure 3 and Table 4). At low Ce rinsed-Corn-BC adsorbed

more NH4-N than rinsed-Oak-BC of the same HTT, yet as Ce increases rinsed-

100

120

140

160

180

200

220

240

260

10

15

20

25

30

35

40

0 50 100 150 200

Ca content of Corn-‐BC (µmol g-‐1)

Ca content of Oak-‐BC (µmol g-‐1)

Ce (µmol g-‐1)

Oak-‐350-‐BC Oak-‐550-‐BC Corn-‐350-‐BC Corn-‐550-‐BC

16

Oak-BC adsorption exceeds rinsed-Corn-BC. The rinsed-Oak-BC also fit the

isotherm model better than the rinsed-Corn-BC (Table 4). A preliminary study

using non-rinsed SPF-BC (comparable to Oak-350-BC) yielded a K*f value of 0.129

(Appendix), which is several orders of magnitude higher than the rinsed-Oak-BC

and indicates that water rinsing decreases the sorptive capacity of BC for

ammonium.

Figure 3: A linearized Freundlich isotherm for ammonium adsorption to rinsed BC.

-‐5

-‐4.5

-‐4

-‐3.5

-‐3

-‐2.5

-‐2

-‐1.5

-‐1

-‐0.5

0 -‐4 -‐3 -‐2 -‐1 0 1 2 3 4 5

log(q*)

log(Ce)

Rinsed-‐Corn-‐350-‐BC Rinsed-‐Corn-‐550-‐BC Rinsed-‐Oak-‐350-‐BC Rinsed-‐Oak-‐550-‐BC

Rinsed-‐Corn-‐350 Rinsed-‐Corn-‐550 Rinsed-‐Oak-‐350 Rinsed-‐Oak-‐550

17

Table 4: Freundlich model terms for ammonium adsorption to rinsed BC.

BC type K*f n R2 P-level

Rinsed-Corn-350-BC 4.9e-4 2.52 0.86 <0.01 Rinsed-Corn-550-BC 1.5e-4 2.26 0.69 <0.01 Rinsed-Oak-350-BC 6.5e-5 1.32 0.93 0 Rinsed-Oak-550-BC 2.9e-5 1.18 0.96 0

Nitrate sorption results

The mean Ce values from the sorption experiments using both rinsed and

non-rinsed Corn-350-BC, Corn-550-BC and Oak-350-BC showed no difference

compared to the mean C0 signifying that no NO3 sorption took place (Test

statistic<t-critical value; Table 5). There was also no nitrate detected (<0.2 mg L-1)

in any of the DI solutions that were shaken with each of the BC types listed

above (data not shown).

Table 5: Results of two-tailed t-tests comparing Ce to C0 for nitrate sorption experiments, calculated at the 5% level. Sample size for each variable is 3, each test having 4 degrees of freedom.

Ce of BC type Comparison to C0 Test statistic t-critical value (5%) P-level

Rinsed-Corn-350-BC 0.55 2.78 0.61 Non-rinsed Corn-350-BC 1.63 2.78 0.18

Rinsed-Corn-550-BC 0.85 2.78 0.44 Non-rinsed Corn-550-BC 2.14 2.78 0.10

Rinsed-Oak-350-BC 1.22 2.78 0.29 Non-rinsed Oak-350-BC 2.03 2.78 0.11

18

Phosphate sorption results

Rinsed and non-rinsed Corn-BC did not show additional PO4 sorption at

any equilibrium concentration measured (Figure 4A). Regression line slopes for

the rinsed Corn-BC isotherms are all near zero (Table 6). The non-rinsed Oak-BC

showed PO4 sorption (Figure 4B), which is evident when compared to the initial

concentration of the BCs (Table 1). Very little PO4 sorption occurs on rinsed Oak-

550-BC at higher equilibrium concentrations, as evidenced by a weak positive

slope of the regression line (P<0.05; Table 6). The difference between initial PO4

solution pH values of 4 and 7 does not greatly affect PO4 sorption to any of the

rinsed BCs.

19

A

B

Figure 4: PO4 sorption isotherms for rinsed and non-rinsed Corn-BC and Oak-BC.

0

10

20

30

40

50

60

70

0 100 200 300 400 500 600

P content (µm

ol g-‐1)

Ce (µmol L-‐1)

non-‐rinsed Corn-‐350-‐pH7 rinsed-‐Corn-‐350-‐pH4 rinsed-‐Corn-‐350-‐pH7 non-‐rinsed Corn-‐550-‐pH7 rinsed-‐Corn-‐550-‐pH4 rinsed-‐Corn-‐550-‐pH7

-‐0.5

0

0.5

1

1.5

2

2.5

3

0 50 100 150 200 250 300 350

P content (µm

ol g-‐1)

Ce (µmol L-‐1)

non-‐rinsed Oak-‐350 rinsed-‐Oak-‐350-‐pH4 rinsed-‐Oak-‐350-‐pH7 non-‐rinsed Oak-‐550-‐pH7 rinsed-‐Oak-‐550-‐pH4 rinsed-‐Oak-‐550-‐pH7

20

Table 6: Linear regression results for PO4 sorption isotherms.

BC type Initial pH of PO4

solution !!"! = ! ∗ !!! + ! R2 P-level

Rinsed-Corn-350- BC 4 !!"! = −0.001 ∗ !!! + 4.56 0.04 0.59 7 !!"! = −0.001 ∗ !!! + 5.77 0.34 0.10

Rinsed-Corn-550-BC 4 !!"! = 0.002 ∗ !!! + 44.37 0.17 0.06 7 !!"! = 0.001 ∗ !!! + 45.10 0.26 0.16

Rinsed-Oak-350-BC 4 !!"! = 0.001 ∗ !!! − 0.04 0.18 0.26 7 !!"! = 0.001 ∗ !!! + 0.21 0.34 0.09

Rinsed-Oak-550-BC 4 !!"! = 0.001 ∗ !!! + 0.68 0.41 0.03 7 !!"! = −0.0002 ∗ !!! + 0.28 0.10 0.32

Discussion and Conclusion

Leachability of nutrients

The rinse solution pH of Corn-BC was higher than that of Oak-BC (Table

2) which is consistent with another study that used the same BC materials [29].

The ash percent of Corn-BC is much greater than that of Oak-BC (Appendix) and

is a likely reason for the increased pH in Corn-BC relative to Oak-BC and also the

increased pH with higher HTT within the same feedstock. The relationship

between high ash content and high pH occurs widely in BC made from plant

biomass [31, 32]. Another factor that might influence decreased acidity with

HTT is the loss of acidic carboxyl functional groups with increased HTT [29]. A

loss of carboxyl groups and decrease in surface acidity was observed between

Eucalyptus saligna wood pyrolyzed at 400º and 550ºC [32].

There were very low levels of NH4 leaching from Corn and Oak-BC and

NO3 leaching was not detected. Even though N comprises about 1% of Corn-BC

and 0.1% of the Oak-BC across the temperature series studied (Appendix), the N

is bound in a sparsely water-soluble form. BC derived from undigested plant

21

biomass is not considered a significant source of NH4+ or NO3

- [31-34]. This

agrees with another study that performed a water extraction of citrus wood BC

(having 0.6% N) by continually stirring the BC in water (1:10 w/w) for 30

minutes and measuring the extract for NH4+ and NO3

- [33]. The citrus wood BC

leached 0.2 µg NH4-N g-1 BC and 0.6 µg NO3-N g-1 BC. When combined, N (in the

form of NH4+ and NO3

-) was the least water extractable element, by several

orders of magnitude compared to other elements leached (Ca, K, Mg, Mn, P, S,

Na) [33].

Initial P content is dependent primarily on feedstock and is concentrated

to a greater extent in the BC as HTT increases (Appendix). Comparing

feedstocks, Corn-BC has higher P content than Oak-BC and leached more P as

well (Figure 1). P solubility is not affected by HTT as both high and low HTT

Corn-BC produce the same concentration of P in the rinse solution regardless of

P content in the BC. However, P leachability decreases with increased HTT,

which might result from pore constriction as increased HTT leads to micropore

development (pore diameter<20 Å) and transition away from mesopores (pore

diameter>20 Å).

Surface area measured with N2 sorbate might be distinguished as

mesopore surface area. The inability of N2 to characterize microporous carbon

[35] might help differentiate nutrient-leachable pores from those that are smaller

than the diameter of hydrated nutrient ions. CO2 sorbate achieves higher

equilibrium concentrations on BC [36] due to its smaller radius (3.3 Å) compared

to N2 (3.64 Å), enabling enhanced pore structure information [37]. The first

hydration layer surrounding PO4 can have oxygen molecules in water between

3.0 and 4.8 Å distant to P and the second hydration shell can have distances

22

between 4.8-6.8 Å [38], which makes it unlikely that PO4 will be hydrated and

leached from the micropores.

In aqueous environments, the solute must diffuse through a pore in order

to access all open sites [39]. The easily accessible sites will be preferentially

occupied (or leached) even though the less accessible sites have the same

sorptive energy. CO2 measured surface area increases with HTT for each BC

(Table 1), yet P leaching decreases which suggests that micropore surfaces are

not easily leached. BC that has a larger proportion of mesopores will allow for

better hydration of molecules on its surface than BC with fewer mesopores

relative to micropores. In this study, BCs with a larger mesopore to micropore

ratio leached a higher percentage of total P. This supports the theory that N2

measured surface area is a better predictor of P leaching than CO2 measured

surface area, assuming solubility is constant.

The leachability of Mg follows a very similar trend to that of P release and

the radius of the first hydration layer of Mg2+ (3.00-4.28 Å) is similar to that of

PO4 [40]. The Mg content of the Corn-BC is much larger than that of Oak-BC

(Table 1) and also leached a higher percentage. Solubility of Mg does not appear

to be related to HTT but the low HTT BCs release a larger percent of their total P,

which might be caused by a larger mesopore surface area. The development of

micropores might be limiting the release of Mg. Decreased Mg and P leaching

with increased HTT has been observed in other studies as well [32].

The Ca content of Corn-BC is approximately one order of magnitude

higher than that of Oak-BC, for all HTTs (Table 1). Yet, Ca concentration in the

rinse solutions did not reflect this trend (Figure 2). Ca concentration in the rinse

solutions for Corn-BC and Oak-BC were similar for high and low HTT, where

23

high HTT BC leached more Ca than low HTT BC. Since Ca leaching appears to

be driven by HTT rather than feedstock type, higher pyrolysis temperature

might influence the way in which the Ca is bound in BC and changes in the

molecular structure of the Ca compound could alter its solubility.

HTT has been shown to alter Ca compounds in activated and non-

activated BC made from leaf and wood material of E.saligna [32]. X-ray

diffraction (XRD) analysis revealed the major crystalline phase of E. saligna BC

produced at 400º C was whewellite (calcium oxalate monohydrate,

Ca(C2O4).H2O) yet the structure was carbon rich and poorly crystalline. When

the feedstock was pyrolyzed at 550º C, the whewellite structure was lost and

calcite (CaCO3) was formed in all of the samples. An increase in exchangeable

Ca was also observed in the higher temperature E. saligna BCs [32]. Calcite has

also been observed in cornstraw BC pyrolyzed at 500º C using XRD [41]. The

same study measured the kinetic release of minerals and found that Ca detached

quickly, followed by a zero order release reaction after 24 hours.

Calcium oxalate is a common crystal in plant biomass [42, 43], which

makes it likely that corn and oak feedstocks share the mineral. As the feedstocks

are pyrolyzed calcium oxalate is lost and calcite is formed [32]. The solubility of

calcite, Log Ksp = -6.37 [44], is higher than that of calcium oxalate monohydrate,

Log Ksp = -8.64 [45]. If Ca is predominantly bound to oxalate in the biomass of

corn and oak feedstocks then Ca solubility would be lower in the 350º HTT BC.

The formation of calcite in high HTT BC would also explain the increased Ca

solubility observed in the rinse solutions of Corn-550-BC and Oak-550-BC.

24

Sorption experiments

All rinsed-BC adsorbed NH4+ but low HTT BC sorbed more than high

HTT BC for each feedstock. The decrease in cation exchange capacity (CEC) with

increasing HTT can be attributed to the loss of carboxyl functional groups during

pyrolysis [29, 46, 47]. The relative proportion of carboxyl groups in the corn and

oak BC decreases by half between the HTT range 300º - 600º C [29]. The two

feedstocks have very similar carboxyl group proportions at these temperatures

yet they do not adsorb ammonium the same way. CEC increases with solution

pH [41] which might explain the difference in greater adsorption of NH4 by

Corn-BC than Oak-BC when compared at the same HTT.

Oak-BC showed a better fit to the Freundlich adsorption model compared

to Corn-BC (Table 6), which might be a result of the increased fractal dimensions

of Oak-BC. The 1/n term of the Freundlich equation (Equation 4) has been

shown to be directly proportional to the fractal dimension of the adsorbent [48].

The Freundlich equation normally applies to a limited part of most adsorption

isotherms—at the early stage of adsorption where the relationship between S and

C (Equation 4) is more or less linear. The Freundlich adsorption isotherm for

Corn-BC does not linearize well because NH4+ sorption is more variable over the

concentration range. This could be due to a lower fractal dimension as expressed

by the smaller 1/n terms and less surface area for Corn-BC compared to Oak-BC,

which had a larger 1/n terms and more surface area. NH4+ sorption for Oak-BC

remains linear and is likely caused by a larger micropore surface area. The

constant (K*F) for Oak-BC is lower than Corn-BC (Table 6), which might result

from lower solution pH induced by the Oak-BC. Yet, the fact that q* values for

Oak-BC exceed that of Corn-BC at high NH4 concentrations might indicate better

25

access to micropores by virtue of a smaller hydrated radius (1.04 – 3.31 Å) or

weak hydration [40], which would allow for access to more sorption sites as Oak-

BC has more total surface area than Corn-BC at the same HTT (Table 1).

The inability of BC to adsorb NO3 has been observed in sorption

experiments using sugarcane bagasse-BC (250 µm – 500 µm) of the HTT range

400º - 600º C [22], yet the same study observed NO3 sorption to bagasse-BC above

700º C HTT. Approximately 7.14 µmol NO3-N g-1 adsorbed to the 800º C HTT

bagasse-BC when mixed with a solution of 20 mg NO3-N L-1. One study

concluded the optimal equilibrium pH for NO3 sorption to bamboo-BC (300 µm -

500 µm) occurred at pH 2.4 and used the Langmuir adsorption model to predict a

sorption capacity of 31.5 µmol NO3-N g-1 [23]. Another study utilized the

Langmuir model to predict a sorption capacity of 89.29 µmol NO3-N g-1 for 900º C

HTT bamboo-BC powder (~80 µm). Two scenarios to account for NO3 sorption

to BC might include: 1) the development of basic functional groups with

increasing HTT that provide anion exchange sites [22], and 2) an increase of the

surface positive charge with decreasing pH (pH 3-7) for new BC [49, 50].

However, anion exchange capacity (AEC) is known to be low and transient in

Corn-BC and Oak-BC in the HTT range 350º - 600º C and do not have strong AEC

after repetitive rinsing [49]. It can be concluded that the rinsed and non-rinsed

BC used in this study do not have sufficient AEC to adsorb NO3.

Additional PO4 sorption did not occur on any of the rinsed or non-rinsed

Corn-BC (Figure 4A) yet some additional PO4 sorption occurred on non-rinsed

Oak-BC (Figure 4B). The rinsed BC does not appear to have AEC in the pH

range tested because neither of the dominant ions H2PO4- nor HPO4

2- sorbed to

the rinsed BC at significant concentrations. Some additional PO4 was sorbed on

26

the rinsed Oak-BC at the highest equilibrium concentration tested but this is not

likely caused by AEC because no NO3- sorbed to any of these rinsed BC.

The sorption experiment using non-rinsed Corn-550-BC (Figure 4A)

exceeded the equilibrium concentration observed in the rinse study (Figure 1)

but no additional P was sorbed to the material. Oak-BC on the other hand

showed additional PO4 sorption on rinsed BC above 300 µmol-PO4-P L-1 (Figure

4B). Since none of the Oak-BC rinse solutions were higher than 2 µmol-P L-1

(Figure 1), it is conceivable that PO4 could be sorbed onto the BC surface. The

energy needed to overcome the sorption enthalpy is driven by the concentration

differential between the bulk solution and the sorbent surface [51] and it appears

that a concentration of 320 µmol-PO4-P L-1 provides a large enough gradient for

PO4 sorption to occur on Oak-BC. The non-rinsed Oak-BC shows more sorption

than the rinsed BC at the same concentration, which implies that PO4 sorption

capacity is lost with rinsing.

Since the BC used in this study does not have strong AEC, PO4 cannot be

adsorbed by electrostatic forces. Another possibility is precipitation of calcium

phosphate compounds on the BC surface such as hydroxyapatite (Ca5(PO4)3OH),

which requires available Ca2+, PO4 and OH- without chemical species or

conditions that would enable substitutions to form thermodynamically favorable

and relatively soluble complexes of amorphous Ca phosphate (Ca3(PO4)2.nH2O),

octacalcium phosphate (Ca4H(PO4)3.2.5H2O), dicalcium phosphate dihydrate

(CaHPO4.2H2O) or whitlockite (Ca9(MgFe)(PO4)6PO3OH) [52-55]. Mg2+ can

inhibit hydroxyapatite crystallization through Ca2+ substitution and the presence

of organic acids such as oxalic acid or larger acids can reduce hydroxyapatite

precipitation rate up to 94% [55]. Even in Ca rich soils receiving manure, Mg and

27

P leaching have been correlated, which can be attributed to Mg increasing the

solubility of P [52, 56]. Considering that BC additions to soil have resulted in

significant release of Mg and P [57] and the evidence from this study in terms of

Mg leaching, it is unlikely that PO4 precipitation will occur on Corn-BC due to its

high Mg leachability. Lower HTT BCs also have more carboxyl groups [29],

which can inhibit hydroxyapatite formation [55]. Oak-BC has a larger molar

ratio of Ca:Mg (Table 2) and has shown PO4 sorption. Since Mg2+ can easily

substitute Ca2+ during PO4 precipitation, a Ca:Mg ratio above 15 could indicate

more favorable conditions for precipitation.

The results of this study show that BC would be a good mitigation option

for point and nonpoint source NH4 pollution. BC with Ca:Mg ratios above 15

have the ability to mitigate PO4 pollution yet mitigation potential decreases with

high water loading. No BC studied in this research has the ability to mitigate

NO3 though adsorption.

APPENDIX

28

Physical differences between Corn-BC and Oak-BC with temperature

The N2 measured surface area and pore volume for the Corn-BC shows an

inverse relation to increasing HTT yet CO2 measured surface area and pore

volume increase with increasing HTT (Table 1). The Corn-BC micropore volume

doubles between HTTs of 400-600 ºC and the surface area, including the

micropores, increases 187% over the same range. The mesopores on the other

hand decrease as the HTT increases, which might suggest a shift towards a

surface area that is dominated by micropores in high HTT BC.

The trend towards increased micropore surface area and volume of Corn-

BC and Oak-BC with increasing HTT might be explained in part by mass loss.

Oak-BC has slightly higher volatiles content in 300º and 350 ºC BC relative to

Corn-BC of the same HTT (Appendix figure 1). At 450 ºC and above, Corn-BC

and Oak-BC are closely matched in their volatiles content. Decreasing volatile

content with increasing HTT might imply that micropores become available as

volatile material escapes.

APPENDIX

29

Appendix figure 1: Mean volatile content of Corn-BC and Oak-BC. ASTM 950º C method used, volatile content equals percent mass lost after anoxic combustion at 950º C compared to 105º C dry weight under anoxic conditions. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

The ash content between Corn-BC and Oak-BC are markedly different

across the HTT range (Appendix figure 2). The Oak-BC ash content does not

increase with HTT whereas the Corn-BC shows a steady increase in ash content

with increasing HTT. The higher ash content of the Corn-BC might be related to

lower lignin content, whereas Oak-BC has higher lignin and would dilute the ash

content on a weight basis. As Corn-BC loses mass in volatiles, percent ash

content increases. Ash content can also be a surrogate for pH, where BC

produced at higher HTT also have higher pH [29].

0%

10%

20%

30%

40%

50%

60%

70%

250 300 350 400 450 500 550 600 650

Volatile content (w

/w)

pyrolysis highest treatment temperature ºC

Corn-‐BC

Oak-‐BC

APPENDIX

30

Appendix figure 2: Mean ash content (%) of Corn-BC and Oak-BC. ASTM 750º C method used, ash content equals (1- % mass lost after combustion at 750º C in oxic conditions), compared to 105º C dry weight under oxic conditions. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

Chemical differences between Corn-BC and Oak-BC with temperature

The nitrogen content of char seems relatively fixed over the HTT range

(Appendix figure 1). The P content of the two chars is very different. The

nitrogen content of Corn-BC is about an order of magnitude higher than that of

Oak-BC. For the 350º and 550º C chars for each feedstock, there does not appear

to be any difference in their nitrogen contents between the two temperatures.

Since this parameter is measured on a mass basis, the nitrogen must be changing

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

250 300 350 400 450 500 550 600 650

Ash content (w/w)


Corn-‐BC

Oak-‐BC

APPENDIX

31

in the same proportion as the overall mass of the char in order to maintain the

same percent content over the HTT series.

Appendix figure 3: Mean nitrogen content of Corn-BC and Oak-BC. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

The P content of the two chars is very different. Looking across the HTT

range, Corn-BC shows an increasing P content with increasing temperature

(Appendix Appendix figure 4) whereas the P content of Oak-BC is constant

throughout the entire temperature series. The pyrolysis reaction does not release

much P from the feedstock as it is converted into BC [34].

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

250 300 350 400 450 500 550 600 650

Nitrogen content (w/w)


Corn-‐BC

Oak-‐BC

APPENDIX

32

Appendix figure 4: Mean phosphorus content (µmol g-1) of Corn-BC and Oak-BC. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

Ca could be an important element for P sorption as it is associated with

the largest fraction of P in some chars [34]. In both Corn-BC and Oak-BC, Ca

appears to increase with increasing HTT (Appendix figure 5).

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

250 300 350 400 450 500 550 600 650

P content (µm

ol g-‐1)


Corn-‐BC

Oak-‐BC

APPENDIX

33

Appendix figure 5: Mean calcium content of Corn-BC and Oak-BC. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

The presence of Mg in biochar has been used to explain phosphate

sorption in anaerobically digested sugar beet tailings biochar [25] . Periclase

(MgO) was identified as a possible phosphate precipitation site and was shown

to have higher phosphate removal efficiency than biochar impregnated with

ferric hydroxide—a biochar pretreatment method that has been effective in

removing phosphate from water. Differences in Mg content of the Corn-BC and

Oak-BC are large across the HTT range (Appendix figure 6). For the temperature

range of this study, the Mg content of the Corn-350-BC and Corn-550-BC is 150

and 200 times the Mg content of Oak-350-BC and Oak-550-BC respectively.

0

50

100

150

200

250

300

350

250 300 350 400 450 500 550 600 650

Ca content (µmol g-‐1)


Corn-‐BC

Oak-‐BC

APPENDIX

34

Appendix figure 6: Mean magnesium content of Corn-BC and Oak-BC. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

The functional group abundance can be represented generally by

elemental ratios of H:C (Appendix figure 7) and O:C (Appendix figure 8). These

ratios show that hydrogen (H) is in higher abundance than oxygen (O) in biochar

of both feedstocks. Both chars show a steep decrease in H:C and O:C ratios

between the HTT range 300 – 450 ºC, after which the values stabilize for higher

temperatures. Over this same range the fixed carbon content is increasing

(Appendix figure 9) which might be caused by loss of H and O in the form of

volatiles. Both the Corn-BC and Oak-BC have higher H:C and O:C ratios in the

350 ºC char than the 550 ºC char, which means lower temperature chars have

more functional groups. Using FTIR it was found that acidic carboxyl functional

0

50

100

150

200

250

300

350

400

450

250 300 350 400 450 500 550 600 650

Mg content (µm

ol g-‐1)


Corn-‐BC

Oak-‐BC

APPENDIX

35

groups decreased with temperature while hydroxyl groups increased with

temperature for corn and oak chars [29]. The percentage of O and H lost from

Oak-BC was more than that of Corn-BC. In both cases, change in the O:H ratios

is driven by the loss of oxygen (Appendix figure 10).

Appendix figure 7: Molar ratios of total hydrogen to total carbon for Corn-BC and Oak-BC. Data is compiled from a previous study [28].

Appendix figure 8: Molar ratios of total oxygen to total carbon for Corn-BC and Oak-BC. Data is compiled from a previous study [28].

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

250 300 350 400 450 500 550 600 650

H:C ratio


Corn-‐BC

Oak-‐BC

0.0

0.1

0.2

0.3

0.4

250 300 350 400 450 500 550 600 650

O:C ratio


Corn

Oak

APPENDIX

36

Appendix figure 9: Mean fixed carbon content of Corn-BC and Oak-BC. Fixed carbon equals [1- (% ash)- (% volatiles)]. Error bars represent standard deviation, n=3. Data is compiled from a previous study [28].

Appendix figure 10: O:H molar ratio for Corn-BC and Oak-BC with increasing HTT. Data is compiled from a previous study [28].

0%

10%

20%

30%

40%

50%

60%

70%

80%

250 300 350 400 450 500 550 600 650

Fixed carbon content (w

/w)


Corn-‐BC

Oak-‐BC

0.2

0.3

0.3

0.4

0.4

0.5

250 300 350 400 450 500 550 600 650

O:H


Corn

Oak

APPENDIX

37

Preliminary sorption experiments using SPF-BC

The BC was prepared by Alterna Biocarbon (Prince George, British

Columbia) from a mixture of softwoods (Spruce, Pine and Fir; referred to as

“SPF-BC”), which came from a chipped waste stream of the logging and sawmill

industry. The wood mixture was slow-pyrolyzed at 400º C. The sorption

experiments were performed at 24º C + 2º C with a Burrell wrist action shaker at

400 rpm. 0.5 g of BC was added to Corning 50 mL centrifuge tubes along with 50

mL of liquid (water or nutrient solution depending on experiment). The tubes

were shaken for 24 hours and then left quiescent for 24 hours before the

supernatant was filtered.

Ammonium sorption

Ammonium sorption was performed using ground SPF-BC with solution

concentrations ranging: 1; 10; 100; 1000; 10,000 mg NH4–N L-1. Equilibrium

concentrations were analyzed using the phenate colorimetric method [30].

Absorbance was measured at 640 nm on a Hewlett Packard diode array UV-Vis

spectrophotometer. Results were analyzed using the Freundlich isotherm model.

Strong sorption was observed with unwashed, ground SPF-BC. The data

fit the linearized Freundlich model well. The equation for the best fit line

through the data had an R2 value of 0.949, yielding a K*f value of 0.129 and an n

value of 1.423 (Figure 11). It was also confirmed that the solid to liquid ratio in

the sorption experiments was suitable for the ammonium concentration range.

APPENDIX

38

Figure 11: Ammonium sorption results using raw, ground SPF-BC. Data was linearized according to the Freundlich model: K*

f = 0.129, n = 1.423.

Nitrate sorption

Nitrate sorption was performed in triplicate using ground SPF-BC with

solution concentrations ranging: 1; 10; 100 mg NO3–N L-1. Nitrate concentrations

of the equilibrium solutions were calculated from measured light absorbance at

220 nm using a Hewlett Packard diode array UV-Vis spectrophotometer. An

interference correction was made by subtracting the sum of absorbance

measurements at 274 nm and 276 nm from the absorbance measured at 220 nm.

The equilibrium concentrations for the nitrate sorption experiments were

not distinctly different from the initial concentrations at 1 and 10 mg NO3-N L-1

(data not shown). The nitrate content at the low concentration samples was

y = 0.7025x -‐ 0.8888 R² = 0.94886 -‐2

-‐1.5

-‐1

-‐0.5

0

0.5

1

1.5

2

2.5

-‐1 0 1 2 3 4 5

log(q*) m

g NH4-‐N g BC-‐1 L

-‐1

log(Ce) mg NH4-‐N L-‐1

APPENDIX

39

difficult to determine using the UV-vis method because the BC released organic

compounds that interfered with the nitrate absorbance reading.

Phosphate sorption

Phosphate sorption was performed in triplicate using SPF-BC that had

been ground, rinsed and dried. The phosphate solution concentration was 100

mg L-1, in an electrolyte matrix of 0.03M NaCl (the electrolyte matrix was not

employed in later experiments). Equilibrium concentrations were analyzed

using the ascorbic acid colorimetric analysis. Absorbance was measured at 880

nm.

Phosphate concentrations in the equilibrium solutions were normalized

per mass SPF-BC, yielding a mean sorption value of 42.58 mMol PO4-P g-1 (+

31.65 SD). This shows that PO4-P sorption is possible with pre-rinsed, ground

SPF-BC (data not shown).

40

REFERENCES

1. Boesch, D.F., Causes and Cosequences of Nutrient Over-Enrichment of Coastal Waters. Proceedings of the 24th Annual Seminar on World Planetary Emergencies, 2002.

2. Nixon, S., Coastal marine eutrophication: a definition, social causes, and future concerns. Ophelia, 1995. 41: p. 199-219.

3. NOAA. Harmful Algal Blooms. 2010 June 07, 2010 [cited 2011 April 17]; Available from: http://oceanservice.noaa.gov/topics/coasts/hab/.

4. Micheli, F., Eutrophication, Fisheries, and Consumer-Resource Dynamics in Marine Pelagic Ecosystems. Science, 1999. 285: p. 1396-1398.

5. Seehausen, O., J. van Alphen, and F. Witte, Cichlid Fish Diversity Threatened by Eutrophication That Curbs Sexual Selection. Science, 1997. 277: p. 1808-1811.

6. Richardson, K., Harmful or Exceptional Phytoplankton Blooms in the Marine Ecosystem. Advances in Marine Biology, 1997. 31: p. 301-385.

7. Bates, S.S., et al., Controls on Domoic Acid Production by the Diatom Nitzschia pungens f. multiseries in Culture: Nutrients and Irradiance. Canadian Journal of Fisheries and Aquatic Sciences, 1991. 48: p. 1136-1144.

8. Cloern, J., Our evolving conceptual model of the coastal eutrophication problem. Marine Ecology Progress Series, 2001. 210: p. 223-253.

9. EPA. Watershed Assessment, Tracking & Environmental Results. 2011 April 17, 2011 [cited 2011 April 17]; Available from: http://iaspub.epa.gov/waters10/attains_nation_cy.control.

10. Gilinsky, E., Baker, M. G., Capacasa, J. M., King, E. S., An Urgent Call to Action – Report of the State-EPA Nutrient Innovations Task Group. Unites States Environmental Protection Agency, 2009.

11. Keiluweit, M., Nico, P. S., Johnson, M. G., Kleber, M., Dynamic Molecular Structure of Plant Biomass-Derived Black Carbon (Biochar). Environmental Science & Technology, 2010. 44: p. 1247-1253.

12. Lehmann, J., Bio-Energy in the Black. Frontiers in Ecology and the Environment, 2007. 5(7): p. 381-387.

41

13. Baldock, J. and R. Smernik, Chemical composition and bioavailability of thermally altered Pinus resinosa (Red pine) wood. Organic Geochemistry, 2002. 33: p. 1093-1109.

14. Nguyen, B.T., Lehmann, J., Hockaday, W. C., Joseph, S., Masiello, C., Temperature sensitivity of black carbon decomposition and oxidation. Environmental Science & Technology, 2010. 44: p. 3324-3331.

15. Bourke, J., et al., Do all carbonized charcoals have the same chemical structure? 2. A model of the chemical structure of carbonized charcoal. Ind. Eng. Chem. Res., 2007. 46: p. 5964-5967.

16. Lehmann, J., A handful of carbon. Nature, 2007. 447: p. 143-144.

17. Cornelissen, G., Haftka, J., Parsons, J., Gustafsson, Ö, Sorption to Black Carbon of Organic Compounds with Varying Polarity and Planarity. Environmental Science & Technology, 2005. 39: p. 3688-3694.

18. Lehmann, J., Silva, J. P., Steiner, C., Nehls, T., Zech, W., Glaser, B., Nutrient availability and leaching in an archaeological Anthrosol and a Ferralsol of the Central Amazon basin- fertilizer, manure and charcoal amendments. Plant and Soil, 2003. 249: p. 343-357.

19. Laird, D., et al., Biochar impact on nutrient leaching from a Midwestern agricultural soil. Geoderma, 2010. 158: p. 436-442.

20. Ding, Y., et al., Evaluation of Biochar Effects on Nitrogen Retention and Leaching in Multi-Layered Soil Columns. Water, Air, & Soil Pollution, 2010. 213(1-4): p. 47-55.

21. Mishra, P. and R. Patel, Use of agricultural waste for the removal of nitrate-nitrogen from aqueous medium. Journal of Environmental Management, 2009. 90(1): p. 519-522.

22. Kameyama, K., et al., Influence of sugarcane bagasse-derived biochar application on nitrate leaching in calcaric dark red soil, in Journal of Environmental Quality2011: doi:10.2134/jeq2010.0453.

23. Ohe, K., et al., Removal of nitrate anion by carbonaceous materials prepared by bamboo and coconut shell. Journal of Chemical Engineering of Japan, 2003. 36(4): p. 511-515.

24. Mitzua, K., et al., Removal of nitrate-nitrogen from drinking water using bamboo powder charcoal. Bioresource Technology, 2004. 95: p. 255-257.

42

25. Yao, Y., et al., Biochar derived from anaerobically digested sugar beet tailings: Characterization and phosphate removal potential. Bioresource Technology, 2011.

26. USEPA, Consolidated Concentrated Animal Feeding Operations (CAFO) Regulations, November 2008. The Clean Water Act, 33 U.S.C. 1251 1972 (Section 122).

27. U.S.C., Federal Water Pollution Control Act, U.S.E.P. Agency, Editor 1974, United States Congress.

28. Rajkovich, S., Enders, A., Hanley, K., Hyland, C., Zimmerman, A. R., Lehmann, J., Corn growth and nitrogen nutrition after additions of biochars with varying properties to a temperate soil. in revision, 2011.

29. Nguyen, B.T. and J. Lehmann, Black carbon decomposition under varying water regimes. Organic Geochemistry, 2009. 40(8): p. 846-853.

30. AWWA, Standard Methods for the Examination of Water and Wastewater: 4500-NH3 F. Phenate Method, 1999, American Water Works Assoc., American Public Health Assoc., Water Environment Federation.

31. Glaser, B., J. Lehmann, and W. Zech, Ameliorating physical and chemical properties of highly weathered soils in the tropics with charcoal - a review. Biology and Fertility of Soils, 2002. 35(4): p. 219-230.

32. Singh, B., Singh, B. P., Cowie, A. L., Characterisation and evaluation of biochars for their application as a soil amendment. Australian Journal of Soil Research, 2010. 48: p. 516-525.

33. Graber, E.R., et al., Biochar impact on development and productivity of pepper and tomato grown in fertigated soilless media. Plant and Soil, 2010. 337(1-2): p. 481-496.

34. Bridle, T.R., Pritchard, D., Energy and nutrient recovery from sewage sludge via pyrolysis. Water Science & Technology, 2004. 50(9): p. 169-175.

35. Kumar, K.V., et al., Hybrid isotherms for adsorption and capillary condensation of N2 at 77 K on porous and non-porous materials. Chemical Engineering Journal, 2010. 162: p. 424-429.

36. Shen, C., et al., Adsorption equilibria and kinetics of CO2 and N2 on activated carbon beads. Chemical Engineering Journal, 2010. 160: p. 398-407.

43

37. Stewart, M.L., Stencel, J. M., CO2 adsorption techniques on high surface area activated carbons. 1992(40511-8433).

38. Pribil, A.B., et al., Structure and Dynamics of Phosphate Ion in Aqueous Solution: An Ab Initio QMCF MD Study. Journal of Computational Chemistry, 2008. 29: p. 2330-2334.

39. Skopp, J., Derivation of the Freundlich Adsorption Isotherm from Kinetics. Journal of Chemical Education, 2009. 86(11): p. 1341-1343.

40. Tansel, B., et al., Significance of hydrated radius and hydration shells on ionic permeability during nanofiltration in dead end and cross flow modes. Separation and Purification Technology, 2006. 51: p. 40-47.

41. Silber, A., Levkovitch, I., Graber, E. R., ph-dependent mineral release and surface properties of cornstraw biochar: agronomic implications. Environmental Science and Technology, 2010. 44: p. 9318-9323.

42. Nakata, P., Advances in our understanding of calcium oxalate crystal formation and function in plants. Plant Science, 2003. 164(6): p. 901-909.

43. Libert, B., Franceschi, V. R., Oxalate in crop plants. J. Agric. Food Chem., 1987. 35: p. 926-938.

44. Zeebe, R.E., Wolf-Gladrow, Dieter, CO2 in Seawater - Equilibrium, Kinetics, Isotopes, 2001, Elsevier.

45. Knovel, Knovel Critical Tables (2nd Edition), 2003, Knovel.

46. Cheng, C., et al., Oxidation of black carbon by biotic and abiotic processes. Organic Geochemistry, 2006. 37(11): p. 1477-1488.

47. Liang, B., et al., Black Carbon Increases Cation Exchange Capacity in Soils. Soil Science Society of America Journal, 2006. 70(5): p. 1719.

48. Tatlier, M., Sebatakar, A. E., Method to Evaluate the Fractal Dimensions of Solid Adsorbents. Journal of Physical Chemistry, 1999. 103: p. 4360-4365.

49. Cheng, C.-H., J. Lehmann, and M.H. Engelhard, Natural oxidation of black carbon in soils: Changes in molecular form and surface charge along a climosequence. Geochimica et Cosmochimica Acta, 2008. 72(6): p. 1598-1610.

50. Cheng, C.-H. and J. Lehmann, Ageing of black carbon along a temperature gradient. Chemosphere, 2009. 75(8): p. 1021-1027.

44

51. Borizover, M.D., Graber, E. R., Organic compound sorption enthalpy and sorption mechanisms in soil organic matter. Journal of Environmental Quality, 1998. 27(2).

52. Cao, X., Harris, W., Carbonate and magnesium interactive effect on calcium phosphate precipitation. Environmental Science and Technology, 2008. 42: p. 436-442.

53. van der Houwen, J.A.M., et al., The effect of organic ligands on the crystallinity of calcium phosphate. Journal of Crystal Growth, 2003. 249(3-4): p. 572-583.

54. Ferguson, J.F., McCarty, P. L., Effects of carbonate and magnesium on calcium phosphate precipitation. Environmental Science and Technology, 1971. 5(6): p. 534-540.

55. Cao, X., et al., Inhibition of calcium phosphate precipitation under environmentally-relevant conditions. Science of The Total Environment, 2007. 383(1-3): p. 205-215.

56. Josan, M.S., Nair, V. D., Harris, W. G., Herrera, D., Associated release of magnesium and phosphorus from active and abandoned dairy soils. Journal of Environment Quality, 2005. 34: p. 184-191.

57. Rondon, M.A., et al., Biological nitrogen fixation by common beans (Phaseolus vulgaris L.) increases with bio-char additions. Biology and Fertility of Soils, 2006. 43(6): p. 699-708.

AMMONIUM, NITRATE AND PHOSPHATE - Cornell University

Documents

Transcript of AMMONIUM, NITRATE AND PHOSPHATE - Cornell University