Spatial associations and chemical composition of organic carbon

Subscriber access provided by Kaohsiung Medical University

Environmental Processes

Spatial associations and chemical composition of organic carbon sequestered in Fe, Ca, and organic carbon ternary systems Tyler Sowers, Dinesh Adhikari, Jian Wang, Yu Yang, and Donald L. Sparks Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b01158 • Publication Date (Web): 17 May 2018 Downloaded from http://pubs.acs.org on May 17, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

Environmental Science & Technology

1

Spatial associations and chemical composition of organic carbon sequestered in Fe, Ca, and

2

organic carbon ternary systems

3 4

Tyler D. Sowersa*, Dinesh Adhikarib, Jian Wangc, Yu Yangb, and Donald L. Sparksa

5 6

a

7

Delaware, Newark, Delaware, 19716, USA.

8

b

9

89557 USA.

Department of Plant and Soil Sciences, Delaware Environmental Institute, University of

Department of Civil and Environmental Engineering, University of Nevada, Reno, Nevada,

10

c

11

Canada

Canadian Light Source Inc., University of Saskatchewan, Saskatoon, Saskatchewan S7N 2V3,

12 13

*Corresponding author - Delaware Environmental Institute, ISE Lab, 221 Academy Street,

14

Room 465, University of Delaware, Newark, DE 19716; (252)-241-3374; [email protected]

15 16 17 18 19 20 21 22 23

1 ACS Paragon Plus Environment


24

Abstract Figure

25

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 2 ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34

44


Abstract

45

Organo-mineral associations of organic carbon (OC) with iron (Fe) oxides play a major

46

role in environmental OC sequestration, a process crucial to mitigating climate change. Calcium

47

has been found to have high co-association with OC in soils containing high Fe content, increase

48

OC sorption extent to poorly-crystalline Fe oxides, and has long been suspected to form bridging

49

complexes with Fe and OC. Due to the growing realization that Ca may be an important

50

component of C cycling, we launched a scanning transmission X-ray microscopy (STXM)

51

investigation, paired with near-edge X-ray absorption fine structure (NEXAFS) spectroscopy, in

52

order to spatially resolve Fe, Ca, and OC relationships and probe the effect of Ca on sorbed OC

53

speciation. We performed STXM-NEXAFS analysis on 2-line ferrihydrite reacted with leaf

54

litter-extractable dissolved OC and citric acid in the absence and presence of Ca. Organic carbon

55

was found to highly associate with Ca (R2 = 0.91). Carboxylic acid moieties were dominantly

56

sequestered; however, Ca facilitated the additional sequestration of aromatic and phenolic

57

moieties. Also, C NEXAFS revealed polyvalent metal ion complexation. Our results provide

58

evidence for the presence of Fe-Ca-OC ternary complexation, which has the potential to

59

significantly impact how organo-mineral associations are modeled.

60 61 62 63 64 65 66



67 68

Introduction Soils house the largest terrestrial surface reservoir of organic carbon (OC).1-5 Preservation

69

of OC in soil systems not only enhances soil productivity, but stabilizes OC that could otherwise

70

be transferred to the atmosphere as a greenhouse gas.2, 5-8 Due to the importance of OC

71

preservation, considerable research probing the sequestration of OC in environmental soil

72

systems has been performed in order to elucidate primary drivers and mechanisms of OC

73

stability.9-12 Along with physical and freeze/thaw processes, organo-mineral associations have

74

been found to be increasingly important in environmental OC sequestration.3, 13, 14 Metal oxides

75

with high affinity for OC are of particular importance.

76

Iron (Fe) oxides are environmentally ubiquitous in soils and possess a high sorption

77

capacity for OC.10, 15-17 Poorly crystalline Fe oxide phases, such as 2-line ferrihydrite, with high

78

surface area and reactivity are important drivers of OC cycling in environmental systems;16-18

79

therefore, recent research involving soil organo-mineral associations has focused primarily on

80

ferrihydrite and goethite (a common, but less reactive Fe oxide phase).9, 13, 14, 19-23 Research

81

investigating OC sequestration to Fe oxides has identified Fe-OC adsorption and coprecipitation

82

reactions as important pathways for OC sequestration (up to 200 and 230 mg OC sequestered g-1

83

ferrihydrite, respectively).9, 24 Recently, powerful spectroscopy analyses have been applied to

84

these systems to further explore spatial information and mechanisms of OC sequestration by

85

soils and Fe oxides.25-30

86

Many advanced spectroscopic techniques have been utilized in the attempt to elucidate

87

OC sequestration processes; however, attenuated total reflectance-Fourier transform infrared

88

spectroscopy (ATR-FTIR) and scanning transmission X-ray microscopy (STXM) paired with

89

near-edge X-ray absorption spectroscopy (NEXAFS) are among the most effective. ATR-FTIR


Page 4 of 34

Page 5 of 34


90

investigations of OC associations with Fe oxides have attributed OC sequestration to primarily

91

inner-sphere ligand exchange mechanisms.19, 23, 30 STXM-NEXAFS is an increasingly popular

92

technique by which OC distribution, associations with other elements, and speciation may

93

readily be determined.5, 28, 29 Although this is a powerful technique that can provide both spatial

94

and chemical information, limited research has been performed on OC associations with soils

95

and Fe oxides.9, 25, 27-29 Chen et al. and Henneberry et al. found that carboxylic acid moieties were

96

the dominant OC moiety sequestered by ferrihydrite adsorption complexes and coprecipitates

97

through inner-sphere ligand exchange processes using a combination of STXM-NEXAFS and

98

ATR-FTIR.9, 27 Using STXM-NEXAFS mapping, Wan et al. and Chen et al. found that soils

99

containing high concentration of Fe oxides had high spatial elemental associations between C

100

and Fe.25, 29 However, an unexplained high correlation between C and calcium (Ca), in the

101

absence of carbonate minerals, was also observed in select soil samples.

102

Calcium has long been proposed to form bridging complexes with soil OC and metal

103

oxides;20, 21 however, little research exists exploring the effect of Ca on OC sequestration. Recent

104

work by our group has shown that leaf-litter extractable dissolved OC sorption to 2-line

105

ferrihydrite in the presence of 10 mM Ca resulted in a 20% increase in OC sorbed

106

concentration.24 Similarly, batch sorption experiments have found that fulvic acid and humic acid

107

sorption to iron oxide and clay minerals increased in the presence of Ca and was suspected to be

108

due to ternary complex formation.20, 21, 31 Calcium has the potential to significantly impact OC

109

sequestration to Fe oxides, but is in need of further advanced spectroscopic investigation.

110

Therefore, we hypothesize that OC sequestration to 2-line ferrihydrite in the presence of Ca may

111

promote the formation of Fe-Ca-OC ternary complexes that may be spatially resolved

112

performing STXM-NEXAFS analysis. In order to address our hypothesis, we (1) determined



113

spatially correlated Fe, Ca, and C relationships in samples containing ferrihydrite and natural

114

dissolved organic matter (DOM) or citric acid in the presence or absence of Ca using STXM

115

elemental composite maps, (2) compared C 1s NEXAFS spectra at localized clusters identified

116

via principal component analysis (PCA) for complexes synthesized in the presence or absence of

117

Ca, and (3) probed changes in sorbed C speciation and sorption mechanism in the presence or

118

absence of Ca using C 1s NEXAFS and ATR-FTIR spectroscopy.

119 120

Methods

121

Adsorption Complex Synthesis

122

Prior to STXM-NEXAFS analysis, natural DOM and citric acid were complexed to

123

synthetic 2-line ferrihydrite in the absence or presence of Ca. Water-extractable DOM was made

124

using methods previously discussed in Chen et al. and Sowers et al.9, 24 Leaf litter was collected

125

from the organic horizon of a forest Ultisol located at the Stroud Water Research Center in

126

Avondale, Pennsylvania. Collected leaf litter was then shaken (200 rpm) for 90 hours with

127

deionized (DI) water (1:2 leaf litter: DI water; mass basis). The extraction was sieved (0.1 mm)

128

and centrifuged (20,000 g) twice in 30 minute cycles. The collected supernatant was then filtered

129

through sequentially through 0.8, 0.45, and 0.2 µm pore size polyethersulfone filters.

130

The newly created DOM stock was characterized for elemental composition

131

(approximately 2000 mg L-1 C; SI Table S1 & S2) and was reacted with 2-line ferrihydrite such

132

that initial molar C/Fe equaled 4.7. Synthetic 2-line ferrihydrite was synthesized according to

133

well-known methodology described in Schwertmann et al.17 For the Ca receiving samples, CaCl2

134

(Fischer) and DOM were added concomitantly such that initial Ca concentration equaled 30 mM

135

Ca. An initial molar C/Fe ratio and Ca concentration of 4.7 C/Fe and 30 mM Ca, respectively,


Page 6 of 34

Page 7 of 34


136

were chosen based off prior experiments by the authors.24 High C/Fe molar ratios have been

137

found to potentially disguise C spectroscopic features, particularly in infrared (IR) spectroscopy;9

138

therefore, 2.5 and 4.7 molar C/Fe was used for all spectroscopic experiments to limit this sorbed

139

concentration effect. 30 mM Ca was used for all Ca-receiving samples as this was found in the

140

authors’ past work to be the maximum Ca concentration that synergistically affects OC

141

sequestration to synthetic 2-line ferrihydrite.24 All reactions were set to pH 6.25 ± 0.10 using

142

dilute NaOH and mixed (dark) via rotary shaker at approximately 50 rpm for 24 hours. pH 6.25

143

was chosen as it was the lowest pH value we expected to observe ternary complexation, while

144

also maintaining environmental applicability to an acidic soil system. Upon sorption reaction

145

completion, samples were centrifuged, decanted, washed twice with DI water, and the solids

146

stored moist in a freezer at -4°C until further spectroscopic analysis. The previously outlined

147

procedure was repeated for the citric acid system, except citric acid was used in place of natural

148

DOM.

149 150 151

Scanning Transmission X-ray Microscopy STXM-NEXAFS data was collected at beamline 10ID-1 at the Canadian Light Source

152

(Saskatoon, Saskatchewan) for all DOM-bearing and citric acid-bearing sample synthesized

153

earlier at initial C/Fe molar ratios of 4.7. A subsample of ferrihydrite wet-paste was suspended in

154

DI water such that approximately 0.5 mg of sample was suspended with 1 mL DI water.9, 25

155

Approximately 3 µL of the well-mixed suspension was applied to a Si3N4 window such that the

156

window was covered with suspension. Prior to analysis, the sample was allowed to air-dry for

157

approximately 30 minutes and then placed on the STXM sample holder. The STXM chamber

158

was pumped to rough vacuum and then backfilled with 1/6 atmosphere He for the measurements.



159

A maximum spatial resolution of approximately 30 nm was achieved using a Fresnel zone plate.

160

Dwell time equaled 1 ms with a pixel size of approximately 40 nm for all measurements taken.

161

C 1s, Ca 2p, and Fe 2p data was obtained by raster scanning from 280 to 735 eV at

162

specific regions of interest identified at low magnification (up to approximately 2500 µm2) at the

163

C K-edge (288 eV). Image stacks were generated for each element across the aforementioned eV

164

range and averages of these stacks at specific eV ranges were then used when processing

165

collected STXM-NEXAFS data.

166 167 168

Processing of STXM-NEXAFS Data All data stacks were aligned using Stack Analyze (Stony Brook, V2.7) and processed

169

using the aXis2000 software package.32, 33 PCA and cluster analysis was performed using PCA

170

GUI 1.1.1 software.34 After alignment and identifying I0 (incident flux), aXis2000 was used to

171

convert aligned and stacked image data to optical density using the following equation: OD =

172

ln(I0/I), where I is equal to the flux transmitted through the sample. Generated images up to ± 0.5

173

eV of the resonance energy for each element were averaged for chemical mapping. The averaged

174

images at each element’s resonance energies were subtracted with averaged images at the

175

elements’ specific pre-edge range (SI Table S3) in order to obtain chemical maps.

176

Optical density is correlated to element thickness (Lambert-Beer’s law) and was

177

calculated for all elements analyzed (SI Table S3 and S4). Thickness estimates were made using

178

methods described in Wan et al.29 and Chen et al.25 For C, the product of RE*µ*f *ρ was used to

179

calculate the conversion factor for optical density to C thickness (SI Table S4). RE is a

180

dimensionless resonance enhancement factor, µ (cm2 g-1) is the mass absorption coefficient

181

above the C absorption edge minus the mass absorption coefficient below the C edge, f (g g-1) is


Page 8 of 34

Page 9 of 34


182

the fraction of C in the solid phase (mass basis), and ρ (g cm-3) is the density of the solid phase

183

(SI Table S4).29 f and ρ were calculated using values published in Stevenson et al.35 Scaling

184

factors for Ca and Fe were taken from Chen et al.25 and used for optical density to thickness

185

conversions in our study. Optical density correlations between elements were obtained by

186

aligning optical density maps for each element (images averaged from the edge energy stack

187

range minus corresponding images averaged from the pre-edge energy stack range, SI Table S3)

188

and then comparing the maps on a per pixel basis.

189

PCA and cluster analysis were performed to identify regions of statistically unique

190

spectra. Three to four components were typically used when searching for significant

191

components and then lowered if significant features between two components were congruent.

192

An eigenimage scaling factor of 0.3 was used for all cluster analyses and singular value

193

decomposition was performed when analyzing cluster spectra.

194 195 196

ATR-FTIR Freeze-dried DOM and ferrihydrite samples were analyzed using a Bruker ATR-FTIR.

197

Spectra were scanned from 4000 to 600 cm-1 at a spectra resolution of 2 cm-1. 2-line ferrihydrite

198

spectra was subtracted from all OC-bearing ferrihydrite samples to remove contributions from

199

ferrihydrite and focus on the chemical composition of the bound OC. Automatic baseline

200

correction and normalization was applied to all spectra. All samples were dried and analyzed

201

soon after to avoid the effects of moisture uptake on sample spectra. The OPUS Version 7.2

202

spectroscopy software suite (Bruker) was used to process all collected spectra.

203 204

Results and Discussion



205 206 207

Elemental Mapping of Fe, Ca, and C All ferrihydrite samples reacted with either DOM or citric acid in the presence or absence

208

of Ca were analyzed for spatial Fe, Ca, and C distribution using STXM mapping (Figure 1). For

209

all STXM color-coded composite RGB maps, the presence of red, green, and blue are

210

represented by C, Ca, and Fe, respectively. The intensity of the color correlates to optical density

211

and represents the relative concentration of each respective element per unit area.29 Co-

212

associations between C, Ca, and Fe may be identified by observing the colors produced by

213

mixtures of red, green, and blue (RGB) and looking at the relative contributions of RGB values

214

at regions of interest (SI Figure S1) For Figures 1A, 1C, and 1D, cyan-blue regions represent

215

areas of poor transmission and were identified as too thick (optical density > 2.25);29 therefore,

216

these regions were excluded from all further analyses. For Figure 1A (ferrihydrite-DOM, no Ca

217

addition), the purple region represents a strong co-association of C and Fe. Similar C and Fe co-

218

association was seen by Chen et al.9 for ferrihydrite-OC adsorption complexes synthesized in

219

similar experimental conditions. A drastic difference in element co-association may be observed

220

when ferrihydrite was reacted with DOM in the presence of Ca (Figure 1B). The purple regions

221

observed in Figure 1A are mostly absent and predominately replaced by a pervasive network of

222

white, grey and yellow, with hot spots of pink. The presence of white and grey regions represent

223

areas at which Fe, Ca, and C are well associated, suggesting potential ternary relationships or

224

homogenous distribution of C and Ca sorption sites. Yellow regions are found throughout the

225

sample and indicate co-association of C and Ca. A high correlation between C and Ca has also

226

been found for natural samples in similar STXM studies and will be discussed further.25, 28, 29 No


Page 10 of 34

Page 11 of 34


227

separate C phases were observed in the DOM system, contrasting what was observed in

228

Henneberry et al.27 in Fe-OM coprecipitate systems but similar to Chen et al.9

229

Differences in RGB maps for citric acid reacted samples were much less marked, as

230

shown in Fig. 1C and 1D. These particles possess a strong intensity of purple except the particle

231

in the middle of Fig. 1C, suggesting a strong association of Fe and C. Unlike Figure 1B, no

232

white, grey or, yellow regions were observed in Figure 1D; therefore, Figure 1D lacks the

233

suggested Fe-Ca-C ternary associations observed in Figure 1B. Although the RGB elemental

234

distribution maps allow for a qualitative analysis of Fe, Ca, C co-association, a more quantitative

235

approach to identifying correlations between elements of interest may be performed by

236

comparing elemental optical density.

237 238

Optical Density Correlation Between Elements

239

Spatial correlations of Fe, Ca, and C are presented in elemental optical density correlation

240

plots (Figure 2). The Fe/C optical density plot for the DOM-bearing ferrihydrite sample shown in

241

Figure 2A showed a strong C/Fe correlation (0.9464) and C thickness ranging up to 135 nm.

242

The high correlation is congruent with C/Fe relationship Chen et al.25 observed for DOM-

243

ferrihydrite adsorption complexes. Compared to C STXM analysis of soils, the observed

244

maximum C thickness of our samples was approximately 100 nm less; however, this disparity is

245

suspected to be due to soil C microaggregate formation.5, 29 Average C thickness was

246

approximately 65 nm which is comparable to the 68 nm C thickness observed in Ultisols taken

247

from oxic environments.25 DOM sorption to ferrihydrite in the presence of 30 mM Ca resulted in

248

a significant decrease in C/Fe correlation (R2 = 0.6242, Figure 1B) with a concurrent strong

249

correlation between C and Ca (R2 = 0.9144, Figure 1C), implying that different associations



250

between Fe-C and Fe-Ca-C may exist. Ca/Fe was also found to have a strong positive correlation

251

of 0.7373 (SI Figure S2A). Although the C/Fe correlation is lower in the system containing Ca,

252

the optical density relationship for C/Fe, C/Ca, and Ca/Fe are positively correlated, indicating a

253

high ternary association between Fe, Ca, and C. The 0.9144 correlation coefficient between C

254

and Ca in Figure 1C demonstrates the potential importance of Ca content when modeling C

255

cycling in environmental systems. The strong relationship between C and Ca has been observed

256

in other STXM studies of natural soil samples. Phaeozem29 (naturally high in Ca) and Ultisol25

257

(pasture soil receiving Ca from liming amendments) soils were found to have C/Ca optical

258

density correlations of 0.767 and 0.870, respectively. Ca in both soils was also found to be well

259

correlated to Fe (R2 > 0.75). With this in mind, STXM investigation of natural soils yielded

260

similar results to what was seen in our ferrihydrite, Ca, and natural DOM systems; therefore, our

261

results may be applicable to soil systems containing Fe oxides and Ca when investigating soil C

262

cycling. Another interesting phenomenon is the decrease in C/Fe correlation by approximately a

263

third when Ca and DOM were concurrently added to a ferrihydrite suspension. High correlation

264

between C and Ca relative to the significant decrease in C/Fe correlation suggests that C is more

265

closely associated with Ca in ternary systems containing Fe, Ca, and C, potentially signifying

266

that Ca is important to the complexation of C to Fe as a bridging element.

267

Calcium had a large effect on the optical density correlation plots in the DOM system;

268

however, the effect of Ca on the citric acid system was observed to be less significant. The C/Fe

269

optical density plot for citric acid-bearing ferrihydrite (Figure 2D) was found to have a low

270

correlation between C and Fe. The poor correlation between C and Fe in this system is most

271

likely attributable to poor transmission due to sample thickness throughout a large portion of

272

citric acid-ferrihydrite STXM map. Sample thickness was not a major issue for the citric acid-


Page 12 of 34

Page 13 of 34


273

bearing ferrihydrite sample reacted in the presence of Ca (Figure 2E, F). A strong correlation

274

coefficient of 0.7636 was found between C and Fe, suggesting a strong relationship between C

275

and Fe. This is expected to be the case as citric acid is a tricarboxylic acid that is known to form

276

inner sphere complexes with iron oxides.13, 19 Unlike the DOM system in Figure 2C, Figure 2F

277

shows a much lower correlation between C and Ca in the citric acid system. From this

278

observation, it may be suggested that citric acid is less associated with Ca than natural DOM.

279

This may be due to the heterogeneity of OC functional moieties in the DOM solution versus the

280

singular carboxylic species present in citric acid (tricarboxylic acid). Interestingly, Ca is highly

281

associated with Fe in the citric acid system (SI Figure S1B) compared to the DOM system. C/Ca

282

and Ca/Fe correlations in the citric acid system, in conjunction with the analysis of Figure 1D,

283

are likely best explained by both Ca and C being sorbed separately to the ferrihydrite surface

284

such that Ca and C are not highly associated.

285 286

Identifying the Speciation and Spatial Heterogeneity of C in Fh-Ca-OC Complexes

287

C K-edge STXM-NEXAFS of DOM. STXM-NEXAFS analysis was used to spatially investigate

288

the chemical speciation of sequestered OC to ferrihydrite in the absence and presence of Ca

289

(Figure 3 and 4, respectively). Optical density of the averaged stack images (Figures 3A and 4A)

290

increase with the intensity of white and decreases with the degree of black. The pre-edge

291

averaged image was subtracted from the averaged stack image on the C K-edge in order to focus

292

on areas representing sequestered C and to identify areas affected by sample thickness (Figure

293

3B and 4B). Principal component analysis (PCA) was used to identify unique clusters of pixels

294

within each STXM stack that represent regions of statistically different C 1s NEXAFS spectra

295

(Figure 3C and 4C).34 The red cluster in Figure 3C is a region identified as having poor sample



296

transmission and was excluded from the analysis; however, all regions in Figure 4C were found

297

to be of appropriate thickness. Corresponding C 1s NEXAFS for the whole sample, all clusters

298

of interest identified through PCA, and a DOM standard were collected (Figures 3D and 4D).

299

Resonance energies taken from the C 1s NEXAFS data may be used to identify functional

300

groups present in the sample (Table 1) which was then used to characterize OC species present

301

for all samples tested.25, 27-29, 36, 37

302

For the DOM-bearing ferrihydrite sample reacted in the absence of Ca, three unique

303

clusters were observed from the PCA (Figure 3C and 3D). Cluster 1 (yellow) was found to be the

304

most predominant within the DOM-ferrihydrite adsorption complex and corresponded well to the

305

spectra of the whole sample. Both of the spectra are dominated by a peak at 288.5 eV

306

representing carboxylic C.5, 28, 29 Compared to the DOM standard, the carboxylic peak for the

307

whole sample and cluster 1 spectra is diminished and significantly broader. This feature is

308

indicative of carboxylate complexation to Fe oxide, likely through ligand exchange processes.25,

309

37

310

aromatic and phenolic C functional moieties. These features are severely diminished compared

311

to the DOM standard, suggesting that aromatic and phenolic groups in the DOM are not

312

preferentially sequestered to ferrihydrite.26, 29 Cluster 2 (green) and 3 (purple) were found at the

313

edge of the particle and were found to have sharper peaks at 288.5 eV than those observed for

314

the whole sample and Cluster 1. The strong, thin peak at resonance energy 288.5 eV suggests

315

that carboxylic C is still the most predominant resonance feature but is largely representative of

316

the DOM standard and thus may not be as strongly complexed to ferrihydrite compared to the

317

Cluster 1 region.5, 25 For Cluster 2 and 3, the phenolic peak at approximately 286.8 eV was

318

heightened, indicating that there may be localized areas on ferrihydrite that preferentially

Slight features at 285.6 eV and 286.7 eV were also observed, representing the presence of


Page 14 of 34

Page 15 of 34


319

sequester phenolic C from DOM. Aromatic groups for both Cluster 2 and 3 were found to be

320

comparable to the whole sample spectra and Cluster 1.

321

Carbon 1s NEXAFS spectra collected for ferrihydrite-Ca-DOM (Figure 4D) were found

322

to have predominant dissimilarities to the spectra from the ferrihydrite-DOM system (Figure

323

3D). Compared to Figure 3D, the most predominant feature change promoted by the presence of

324

Ca was the extreme diminishment of the carboxylic C peak at 288.5 eV (~30% decrease in

325

relative peak area compared to the DOM standard) and the emergence of a shoulder at 287.3-

326

287.6 eV. This feature was observed for all cluster spectra shown in Figure 4D. Plaschke et al.37

327

and Armbruster et al.36 saw congruent features in C 1s NEXAFS spectra of humic sorption to

328

various polyvalent metal ions. Polyvalent metal ion complexation with humic acid resulted in a

329

decrease in the carboxylic C resonance intensity with concomitant emergence of a shoulder

330

approximately 1 eV below the carboxylic peak.36, 37 The sharp decrease in our sample spectra in

331

Figure 4D with the emergence of the aforementioned shoulder strongly suggests that Ca is may

332

be interacting with DOM carboxylic C groups. The absence of shoulder formation at 287.3-287.6

333

eV and the high intensity carboxylic C resonance energy at 288.5 eV in Figure 3D suggests that

334

the ferrihydrite-Ca-DOM system is uniquely binding carboxylic C compared to the relatively

335

simpler spectra observed for the ferrihydrite-DOM system; therefore, the feature changes at

336

288.5 eV and 287.3-287.6 eV are suggestive that the presence of Ca affects the binding of DOM

337

to ferrihydrite. We expect this feature would be promoted further for ferrihydrite-Ca-DOM

338

samples synthesized at increasing pH, which will be analyzed further in future research.

339

Resonance energies representative of aromatic C (285.3-285.6 eV) in the ferrihydrite-Ca-

340

DOM system (Figure 4D) are also found to significantly differ from the ferrihydrite-DOM

341

system (Figure 3D). All clusters in Figure 4D show a higher intensity, developed aromatic peak



342

at 285.6 eV compared to Figure 3D. Chen et al.9 inferred that increases in aromatic C peak

343

intensity is associated with increased aromatic C complexation to ferrihydrite complexes;

344

therefore, the spectra suggest that the Ca-containing system may be facilitating increased

345

association of aromatic C to ferrihydrite. Phenolic groups were also identified in the sample

346

spectra at 286.7-286.9 eV; however, there were no noticeable differences between Figure 3D and

347

4D.

348

Although all ferrihydrite-Ca-DOM C 1s NEXAFS spectra in Figure 4D were found to

349

have similar peak locations/features, Cluster 2Ca (green) was found to possess the most

350

distinguishable features. This region represents a large continuous region on the ferrihydrite

351

particle shown in Figure 4C. A clear shoulder can be observed at approximately 287.3 eV, along

352

with the diminishment of the carboxylic C peak at 288.5 eV. Also, a distinct, relatively high

353

intensity aromatic peak at 285.6 eV was observed. In conjunction with this unique spectra, the

354

Cluster 2Ca region was also observed in Figure 1B. High intensity optical density contributions

355

from C (red), Ca (green), and Fe (blue) were observed in almost equal proportions from the same

356

region represented by Cluster 2Ca in Figure 4C, resulting in this region appearing white. Due to

357

the presence of unique C NEXAFS features that differ greatly from the system receiving no Ca

358

inputs and the high degree of association between Fe, Ca, and C observed in Figure 1B, Cluster

359

2Ca is expected to be a region of Fe-Ca-C ternary complexation, potentially through Ca-bridging

360

structures.

361 362

C K-edge STXM-NEXAFS Analysis of Citric Acid. PCA revealed only one unique cluster shown

363

in yellow for the ferrihydrite-citric acid (SI Figure S3A) and red for the ferrihydrite-Ca-citric


Page 16 of 34

Page 17 of 34


364

acid. The contrasting color (red and yellow, respectively) represent regions that were identified

365

as too thick for further analysis. Ferrihydrite-citric acid and ferrihydrite-Ca-citric acid spectra (SI

366

359 Figure S3C) were found to have similar features. Both spectra were dominated by a broad

367

carboxylic C peak at 288.5 eV with no other significant features. The broad carboxylic C peak

368

for both samples suggest that both citric acid-treated samples are complexing strongly to

369

ferrihydrite, similar to the feature seen for the ferrihydrite-DOM spectra (SI Figure S3C). No

370

major differences in the C 1s NEXAFS spectra were observed when citric acid was reacted with

371

ferrihydrite in the presence of Ca; therefore, we expect that Ca has little to no effect on the

372

complexation of OC derived from citric acid.

373 374

ATR-FTIR. In conjunction with C 1s NEXAFS, ATR-FTIR data was collected for ferrihydrite

375

samples reacted with DOM or citric acid in the absence or presence of Ca in order to probe C

376

speciation and mechanism of sorption (Figure 5). In addition to the molar C/Fe ratio used for

377

STXM samples (4.7 C/Fe), a lower molar ratio of 2.5 C/Fe was also tested in order to probe

378

features that otherwise would be undistinguishable.9 An absorbance band at 1585 cm-1 was

379

observed for the DOM stock and for 2.5 C/Fe DOM_ 30 mM Ca, with every other sample,

380

except, having a band at approximately 1570 cm-1. Absorbance band formation from 1585 to

381

1570 cm-1 is associated with the presence of asymmetric carboxyl C groups (SI Table S6). All

382

bands, excluding 2.5 C/Fe DOM_ 30 mM Ca, were found to shift from 1585 to 1570 cm-1 and

383

have decreased intensity compared to the DOM standard. The shift and depression of the

384

asymmetric carboxyl C band suggests that asymmetric carboxyl C may be strongly bound to

385

ferrihydrite.9, 19, 23 In conjunction with the asymmetric carboxyl C band, a symmetric carboxyl

386

band may be found from 1400 to 1380 cm-1 for all samples. The DOM standard has a high



387

intensity absorbance band at 1400 cm-1, but all other C-reacted ferrihydrite samples were found

388

to shift to 1384 or 1380 cm-1. Shifting of the symmetric carboxyl C group from 1400 to 1384 cm-

389

1

390

processes,9, 19, 30 suggesting that ligand exchange of OC to ferrihydrite through symmetric

391

carboxyl C groups is a predominant mechanism of sorption for both citric acid and DOM

392

samples.

is a well-documented characteristic of OC sorption to ferrihydrite through ligand exchange

393

Interestingly, the 2.5 C/Fe DOM_ 30 mM Ca was found to have different absorbance

394

band features compared to all other samples. The asymmetric carboxyl C band was significantly

395

decreased similar to other samples but did not shift from 1585 to 1570 cm-1. The decreased

396

intensity of this peak compared to the DOM stock suggests that asymmetric carboxyl C is

397

associating with ferrihydrite in sample 2.5 C/Fe DOM_ 30 mM Ca; however, the lack of a shift

398

suggests that asymmetric carboxyl C is binding weakly to ferrihydrite in this system.10, 23 Also,

399

the symmetric carboxyl C peak was found to shift to 1380 cm-1 rather than 1384 cm-1 and had a

400

higher peak intensity compared to sample 2.5 C/Fe DOM_No Ca. The increased shift and

401

intensity of the symmetric carboxyl C band for 2.5 C/Fe DOM_30 mM Ca compared to 2.5 C/Fe

402

DOM_No Ca suggests that Ca is playing a role in the binding of symmetric carboxyl C, likely

403

through ligand exchange processes.23, 30 Changes in aromatic and phenolic absorbance bands

404

were indistinguishable from the collected ATR-FTIR spectra, likely due to the strong absorbance

405

bands of carboxyl C. Polysaccharide peaks (SI Table S6) were also observed in spectra for the

406

DOM systems, indicating the presence of sequestered DOM polysaccharide functional groups,23,

407

38

but to a lesser degree compared to carboxyl C functional groups.

408 409

Ca L-edge STXM-NEXAFS Analysis


Page 18 of 34

Page 19 of 34


410

Ca 2p NEXAFS spectra were collected for both ferrihydrite-Ca-DOM and ferrihydrite-Ca-citric

411

acid samples. For both, PCA revealed only one unique spectra (SI Figure S4). Four Ca

412

reference spectra were used: Ca-bearing ferrihydrite, sorbed Ca to extracellular polymeric

413

substance (EPS), CaCl2, and calcite.25, 39 For both DOM and citric samples, four total peaks were

414

observed. L3 2P3/2 and L2 2P1/2 peaks were observed at 349.2 and 352.6 eV, respectively. These

415

two major peaks are characteristic of Ca spin-orbital partners found in Ca L-edge NEXAFS.40 A

416

smaller peak precedes each previously mentioned larger peak at 348.2 and 351.4, respectively,

417

and gives an indication of the crystallinity of present Ca structures.41

418

When comparing both ferrihydrite-Ca-DOM (Fh_Ca_DOM) and ferrihydrite-Ca-citric

419

acid (Fh_Ca_Citric Acid) Ca 2p NEXAFS spectra to reference spectra (SI Figure S4), both

420

samples corresponded poorly to calcite and CaCl2 and well to the Ca-bearing ferrihydrite

421

(Fh_Ca) and Ca-containing organic compound (Sorbed Ca_EPS). The calcite spectra poorly

422

agrees with our samples and CaCl2 was found to have higher intensity peaks at the two small L3

423

2P3/2 and L2 2P1/2 with all four peaks shifted to a slightly lower eV (349.2 and 352.4 eV for the

424

two major peaks at (ii) and (iv); 348 and 351.3 eV for two smaller peaks at (i) and (iii) (SI Figure

425

S4)) characteristic of Ca L-edge NEXAFS for CaCl2.42 Therefore, we conclude that present Ca

426

species are not present in the form of calcite nor CaCl2. The Sorbed Ca_EPS and Fh _Ca

427

reference standards were found to correspond well to our samples, mostly due to similar nature

428

of the low intensity L3 2P3/2 and L2 2P1/2 peaks at 348.2 and 351.4 eV, respectively. This

429

suggests that Ca incorporated into ferrihydrite-Ca-OC samples are of poorly crystalline or

430

amorphous nature.41 Fh_Ca was observed to have higher peak intensity at the L3 2P3/2 (349.2 eV)

431

and L2 2P1/2 (352.6 eV) peaks compared to the sample spectra, especially for the Fh_Ca_DOM

432

sample, suggesting that the concurrent presence of ferrihydrite and DOM may have a unique



433

impact on the coordination of Ca; however, the role of Ca as a bridging cation needs to further

434

explored in future spectroscopic analysis.

435 436 437

Evidence for Fe-Ca-OC Ternary Complex Formation Through a combination of spatial associations and C speciation spectra obtained

438

primarily via STXM-NEXAFS, we have investigated potential ternary interactions of Fe, Ca, and

439

OC in systems containing ferrihydrite, Ca, and either DOM or citric acid. Batch sorption data

440

from our past work (SI Figure S5 and S6)24 were paired with ATR-IR and STXM-NEXAFS

441

analysis in order to extensively probe the occurrence and chemical speciation of Fe-Ca-OC

442

ternary complexes. Evenly-proportioned mixtures of C, Ca, and Fe in color-coded RGB maps,

443

high correlation (R2 = 0.91) between C and Ca in optical density correlation plots, and the

444

polyvalent metal ion complexation C 1s NEXAFS feature (decrease of the carboxyl C peak at

445

288.5 eV paired with the emergence of a shoulder at 287.5 eV) all provide significant evidence

446

that Fe-Ca-OC complexes are forming in systems containing ferrihydrite, Ca, and natural DOM,

447

likely through cation-bridging of carboxylic C to the ferrihydrite surface. Also, Ca has recently

448

been shown by the authors to synergistically influence OC sequestration to ferrihydrite,

449

providing further evidence of Fe-Ca-OC ternary complex formation24. Additionally, for the

450

DOM system, the optical density correlation of the C/Fe plot was found to decrease from R2 =

451

0.95 to R2 =0.62 when Ca was added, along with high C/Ca and Ca/Fe correlation (R2 = 0.91 and

452

0.74, respectively). This indicates that Ca is well associated to both C and Fe whereas C is

453

significantly more associated with Ca than Fe; therefore, we suggest that Ca is significantly

454

involved in the complexation of C to Fe, most likely through Ca-bridging structures. ATR-IR

455

data supports this conclusion, as features indicative of ligand exchange of symmetric carboxyl C


Page 20 of 34

Page 21 of 34


456

were observed and enhanced with the addition of Ca. These findings relate well to the work of

457

others that have observed high correlation of C, Ca, and Fe in iron oxide and soil systems.9, 20, 25,

458

29

459

for DOM were observed for the citric acid systems. This is likely due to the chemical

460

heterogeneity of natural DOM; however, this needs to be further explored using other model OC

461

compounds.

462

Citric acid and Fe were found to well-related; however, none of the ternary features observed

Overall, the proposed occurrence of Fe-Ca-OC ternary complexes has the potential to

463

significantly impact how organo-mineral associations are modeled in current environmental C

464

cycling models and the development of potential C sequestration management strategies in soil

465

systems containing Fe oxides. We expect this phenomenon, measured currently at slightly acidic

466

pH conditions, to occur to a greater extent in basic, calcareous soils, which will be examined in

467

the future. Also, we have observed in past research that increasing Ca concentration from 1 mM

468

Ca up to 30 mM Ca resulted in increased OC sorption for samples equilibrated at initial molar

469

C/Fe ratios of 4.7 and 12.5 (SI Figure S6).24 Therefore, we expect that ternary complexation may

470

be occurring in a wide range of systems, which warrants further spectroscopic investigation.

471

Additional work needs to be performed to understand the stability of proposed Fe-Ca-OC ternary

472

complexes with changing environmental conditions (redox, temperature, etc.). Increased

473

spectroscopic work is also needed to elucidate the proposed Ca-bridging mechanism of Fe-Ca-

474

OC ternary complexes, which may be accomplished using chemically similar polyvalent cations

475

of greater atomic mass (e.g., strontium).

476 477

Acknowledgements



478

This publication was made possible by the National Science Foundation EPSCoR Grant No. IIA-

479

1301765, the State of Delaware, the Delaware Environmental Institute, and the Donald L. and

480

Joy G. Sparks Graduate Fellowship. The authors thank the UD Soil Testing Lab, UD Advanced

481

Materials Characterization Lab, Stroud Water Research Center, Canadian Light Source, and the

482

UD Environmental Soil Chemistry group for their support. The Canadian Light Source is funded

483

by the Canada Foundation for Innovation, the Natural Sciences and Engineering Research

484

Council of Canada, the National Research Council Canada, the Canadian Institutes of Health

485

Research, the Government of Saskatchewan, Western Economic Diversification Canada, and the

486

University of Saskatchewan.

487 488 489 490 491 492 493 494 495 496 497 498 499 500


Page 22 of 34

Page 23 of 34

501


Figures.

1 µm

C

B

A

D

2 µm

C Ca Fe

1 µm 1 µm

502 503

Figure 1: Color-coded composite RGB optical density maps created from STXM-NEXAFS data.

504

Red, green, and blue represent carbon, calcium, and iron, respectively. RGB maps are shown for

505

ferrihydrite-DOM (A), ferrihydrite-Ca-DOM (B), ferrihydrite-citric acid (C), and ferrihydrite-

506

Ca-citric acid samples (D). Cyan regions in Figure 1A and Figure 1D were identified as regions

507

too thick for transmission.

508



Page 24 of 34

DOM

Citric Acid

A

D

B

E

C

F

No Calcium

Calcium

509

Figure 2: Elemental optical density correlation plots between C, Ca, and Fe for DOM-bearing

510

(A-C) and citric acid-bearing (D-F) ferrihydrite. Linear correlation coefficients are provided for

511

each plot. Thickness values for each element are also provided (calculated using SI Table S3).

512


Page 25 of 34


0.54

A

D

1 µm

0

B

C

513 514

Figure 3: C K-edge STXM average stack optical density map for ferrihydrite-DOM of the C K-

515

edge (287.7-288.8 eV) and the C K-edge minus C pre-edge (287.7-288.8 eV minus 280-282 eV)

516

difference map are shown in A and B, respectively. PCA of the C K-edge stack is shown in C

517

and respective C 1s STXM-NEXAFS spectra for the whole sample and each of the 3 clusters

518

identified through PCA are shown in D. A standard of the DOM stock is also included.9 Peak

519

assignments are located in Table 1.

520



0.48

A

D

2 µm

0

B

C

521 522

Figure 4: C K-edge STXM average stack optical density map for ferrihydrite-Ca-DOM of the C

523

K-edge (287.7-288.8 eV) and the C K-edge minus C pre-edge (287.7-288.8 eV minus 280-282

524

eV) difference map are shown in A and B, respectively. PCA of the C K-edge stack is shown in

525

C and respective C 1s STXM-NEXAFS spectra for the whole sample and each of the 3 clusters

526

identified through PCA are shown in D. A standard of the DOM stock is also included.9 Peak

527

assignments are located in Table 1.


Page 26 of 34

Page 27 of 34


528 529

Figure 5: ATR-FTIR spectra for ferrihydrite samples reacted with either DOM or citric acid in

530

the absence or presence of Ca. A molar C/Fe ratio of 2.5 or 4.7 was used for DOM samples in

531

order to reveal spectra features that may be difficult to distinguish at high molar C/Fe ratios.9 A

532

reference standard for the stock DOM used for DOM reactions is also included.

533 534



535

Page 28 of 34

Table 1: C 1s NEXAFS peak energies with C functional group interpretation Peak Number

Peak Energies (eV)

C Functional Groups

Transition

References

i, ii

285.2-285.6

Aromatic (Caromatic=Caromatic)

1s-π*

Henneberry et al., 2009; Solomon et al., 2012; Chen et al., 2014

iii

286.7-286.9

Phenolic Caromatic-OH

1s-π*

Wan et al., 2007; Solomon et al., 2012; Chen et al., 2014

iv

287.5

Aliphatic (C-H)

1s-3p/σ*

Wan et al., 2007; Solomon et al., 2013

Boxed Region

287.2-287.6

Polyvalent metal ion complexation to COOH

NA

Plaschke et al., 2005; Armbruster et al., 2009

v

288.5

Carboxylic Acid (COOH)

1s-π*

Wan et al., 2007; Solomon et al., 2012; Chen et al., 2014

536 537 538 539 540 541 542 543 544 545 546 547 548 549


Page 29 of 34


550

References

551

1.

552 553

Sciences 2001, 29, (1), 535-562. 2.

554 555

Amundson, R., The carbon budget in soils. Annual Review of Earth and Planetary

Davidson, E. A.; Trumbore, S. E.; Amundson, R., Biogeochemistry: soil warming and organic carbon content. Nature 2000, 408, (6814), 789-790.

3.

Schmidt, M. W.; Torn, M. S.; Abiven, S.; Dittmar, T.; Guggenberger, G.; Janssens, I. A.;

556

Kleber, M.; Kögel-Knabner, I.; Lehmann, J.; Manning, D. A., Persistence of soil organic

557

matter as an ecosystem property. Nature 2011, 478, (7367), 49-56.

558

4.

559 560

Blanco-Canqui, H.; Lal, R., Mechanisms of carbon sequestration in soil aggregates. Critical Reviews in Plant Sciences 2004, 23, (6), 481-504.

5.

Lehmann, J.; Kinyangi, J.; Solomon, D., Organic matter stabilization in soil

561

microaggregates: implications from spatial heterogeneity of organic carbon contents and

562

carbon forms. Biogeochemistry 2007, 85, (1), 45-57.

563

6.

564 565

Lal, R., Soil carbon sequestration impacts on global climate change and food security. Science 2004, 304, (5677), 1623-1627.

7.

Riley, W.; Maggi, F.; Kleber, M.; Torn, M.; Tang, J.; Dwivedi, D.; Guerry, N., Long

566

residence times of rapidly decomposable soil organic matter: application of a multi-phase,

567

multi-component, and vertically resolved model (BAMS1) to soil carbon dynamics.

568

Geosci. Model Dev. 2014, 7, 1335–1355.

569

8.

Sparks, D. L., Environmental soil chemistry. Academic press: 2003.

570

9.

Chen, C.; Dynes, J. J.; Wang, J.; Sparks, D. L., Properties of Fe-organic matter

571

associations via coprecipitation versus adsorption. Environmental science & technology

572

2014, 48, (23), 13751-13759.



573

10.

Gu, B.; Schmitt, J.; Chen, Z.; Liang, L.; McCarthy, J. F., Adsorption and desorption of

574

natural organic matter on iron oxide: mechanisms and models. Environmental Science &

575

Technology 1994, 28, (1), 38-46.

576

11.

577 578

Six, J.; Conant, R.; Paul, E. A.; Paustian, K., Stabilization mechanisms of soil organic matter: implications for C-saturation of soils. Plant and soil 2002, 241, (2), 155-176.

12.

von Lützow, M.; Kögel-Knabner, I.; Ludwig, B.; Matzner, E.; Flessa, H.; Ekschmitt, K.;

579

Guggenberger, G.; Marschner, B.; Kalbitz, K., Stabilization mechanisms of organic matter

580

in four temperate soils: Development and application of a conceptual model. Journal of

581

Plant Nutrition and Soil Science 2008, 171, (1), 111-124.

582

13.

Kleber, M.; Eusterhues, K.; Keiluweit, M.; Mikutta, C.; Mikutta, R.; Nico, P. S., Chapter

583

one-mineral–organic associations: formation, properties, and relevance in soil

584

environments. Advances in agronomy 2015, 130, 1-140.

585

14.

586 587

oxides. Geochimica et Cosmochimica Acta 2007, 71, (1), 25-35. 15.

588 589

Wagai, R.; Mayer, L. M., Sorptive stabilization of organic matter in soils by hydrous iron

Baldock, J. A.; Skjemstad, J., Role of the soil matrix and minerals in protecting natural organic materials against biological attack. Organic geochemistry 2000, 31, (7), 697-710.

16.

Jambor, J. L.; Dutrizac, J. E., Occurrence and constitution of natural and synthetic

590

ferrihydrite, a widespread iron oxyhydroxide. Chemical Reviews 1998, 98, (7), 2549-

591

2586.

592

17.

593 594 595

Schwertmann, U.; Cornell, R. M., Iron oxides in the laboratory: preparation and characterization. John Wiley & Sons: 2008.

18.

Lalonde, K.; Mucci, A.; Ouellet, A.; Gélinas, Y., Preservation of organic matter in sediments promoted by iron. Nature 2012, 483, (7388), 198-200.


Page 30 of 34

Page 31 of 34

596


19.

Lackovic, K.; Johnson, B. B.; Angove, M. J.; Wells, J. D., Modeling the adsorption of

597

citric acid onto Muloorina illite and related clay minerals. Journal of Colloid and

598

Interface Science 2003, 267, (1), 49-59.

599

20.

Mikutta, R.; Mikutta, C.; Kalbitz, K.; Scheel, T.; Kaiser, K.; Jahn, R., Biodegradation of

600

forest floor organic matter bound to minerals via different binding mechanisms.

601

Geochimica et Cosmochimica Acta 2007, 71, (10), 2569-2590.

602

21.

Weng, L. P.; Koopal, L. K.; Hiemstra, T.; Meeussen, J. C.; Van Riemsdijk, W. H.,

603

Interactions of calcium and fulvic acid at the goethite-water interface. Geochimica et

604

Cosmochimica Acta 2005, 69, (2), 325-339.

605

22.

606 607

Chorover, J.; Amistadi, M. K., Reaction of forest floor organic matter at goethite, birnessite and smectite surfaces. Geochimica et Cosmochimica Acta 2001, 65, (1), 95-109.

23.

Heckman, K.; Vazquez-Ortega, A.; Gao, X.; Chorover, J.; Rasmussen, C., Changes in

608

water extractable organic matter during incubation of forest floor material in the presence

609

of quartz, goethite and gibbsite surfaces. Geochimica et Cosmochimica Acta 2011, 75,

610

(15), 4295-4309.

611

24.

612 613

Sowers, T.D.; Stuckey J.W.; Sparks D.L. The synergistic effect of calcium on organic carbon sequestration to ferrihydrite. Geochemical transactions 2018, 19, (1): 4.

25.

Chen, C.; Dynes, J. J.; Wang, J.; Karunakaran, C.; Sparks, D. L., Soft X-ray

614

spectromicroscopy study of mineral-organic matter associations in pasture soil clay

615

fractions. Environmental science & technology 2014, 48, (12), 6678-6686.

616 617

26.

Chen, C.; Sparks, D. L., Multi-elemental scanning transmission X-ray microscopy–near edge X-ray absorption fine structure spectroscopy assessment of organo–mineral



618

associations in soils from reduced environments. Environmental Chemistry 2015, 12, (1),

619

64-73.

620

27.

Henneberry, Y. K.; Kraus, T. E.; Nico, P. S.; Horwath, W. R., Structural stability of

621

coprecipitated natural organic matter and ferric iron under reducing conditions. Organic

622

geochemistry 2012, 48, 81-89.

623

28.

Solomon, D.; Lehmann, J.; Harden, J.; Wang, J.; Kinyangi, J.; Heymann, K.;

624

Karunakaran, C.; Lu, Y.; Wirick, S.; Jacobsen, C., Micro-and nano-environments of

625

carbon sequestration: Multi-element STXM–NEXAFS spectromicroscopy assessment of

626

microbial carbon and mineral associations. Chemical geology 2012, 329, 53-73.

627

29.

Wan, J.; Tyliszczak, T.; Tokunaga, T. K., Organic carbon distribution, speciation, and

628

elemental correlations within soil microaggregates: applications of STXM and NEXAFS

629

spectroscopy. Geochimica et Cosmochimica Acta 2007, 71, (22), 5439-5449.

630

30.

631 632

Fu, H.; Quan, X., Complexes of fulvic acid on the surface of hematite, goethite, and akaganeite: FTIR observation. Chemosphere 2006, 63, (3), 403-410.

31.

633

Kloster, N.; Avena, M., Interaction of humic acids with soil minerals: adsorption and surface aggregation induced by Ca2+. Environmental Chemistry 2015, 12, (6), 731-738.

634

32.

Hitchcock, A., aXis-2000 is an IDL-based analytical package 2000. In.

635

33.

Jacobsen, C.; Wirick, S.; Flynn, G.; Zimba, C., Soft X-ray spectroscopy from image

636

sequences with sub-100 nm spatial resolution. Journal of Microscopy 2000, 197, (2), 173-

637

184.

638

34.

Lerotic, M.; Jacobsen, C.; Gillow, J.; Francis, A.; Wirick, S.; Vogt, S.; Maser, J., Cluster

639

analysis in soft X-ray spectromicroscopy: finding the patterns in complex specimens.

640

Journal of Electron Spectroscopy and Related Phenomena 2005, 144, 1137-1143.


Page 32 of 34

Page 33 of 34

641


35.

642 643

Stevenson, F. J., Humus chemistry: genesis, composition, reactions. John Wiley & Sons: 1994.

36.

Armbruster, M.; Schimmelpfennig, B.; Plaschke, M.; Rothe, J.; Denecke, M.; Klenze, R.,

644

Metal-ion complexation effects in C 1s-NEXAFS spectra of carboxylic acids—evidence

645

by quantum chemical calculations. Journal of Electron Spectroscopy and Related

646

Phenomena 2009, 169, (1), 51-56.

647

37.

Plaschke, M.; Rothe, J.; Altmaier, M.; Denecke, M. A.; Fanghänel, T., Near edge X-ray

648

absorption fine structure (NEXAFS) of model compounds for the humic acid/actinide ion

649

interaction. Journal of electron spectroscopy and related phenomena 2005, 148, (3), 151-

650

157.

651

38.

Artz, R. R.; Chapman, S. J.; Robertson, A. J.; Potts, J. M.; Laggoun-Défarge, F.; Gogo, S.;

652

Comont, L.; Disnar, J.-R.; Francez, A.-J., FTIR spectroscopy can be used as a screening

653

tool for organic matter quality in regenerating cutover peatlands. Soil Biology and

654

Biochemistry 2008, 40, (2), 515-527.

655

39.

Obst, M.; Dynes, J.; Lawrence, J.; Swerhone, G.; Benzerara, K.; Karunakaran, C.;

656

Kaznatcheev, K.; Tyliszczak, T.; Hitchcock, A., Precipitation of amorphous CaCO 3

657

(aragonite-like) by cyanobacteria: a STXM study of the influence of EPS on the

658

nucleation process. Geochimica et Cosmochimica Acta 2009, 73, (14), 4180-4198.

659

40.

Cody, G. D.; Botto, R. E.; Ade, H.; Behal, S.; Disko, M.; Wirick, S., Inner-shell

660

spectroscopy and imaging of a subbituminous coal: In-situ analysis of organic and

661

inorganic microstructure using C (1s)-, Ca (2p)-, and Cl (2s)-NEXAFS. Energy & Fuels

662

1995, 9, (3), 525-533.



663

41.

Politi, Y.; Metzler, R. A.; Abrecht, M.; Gilbert, B.; Wilt, F. H.; Sagi, I.; Addadi, L.;

664

Weiner, S.; Gilbert, P., Transformation mechanism of amorphous calcium carbonate into

665

calcite in the sea urchin larval spicule. Proceedings of the National Academy of Sciences

666

2008, 105, (45), 17362-17366.

667 668

42.

Araújo, A.; Correia, J.; Soares, C. M., STXM investigation into effects of CaCl2 on the hydration of tricalcium silicate and tricalcium aluminate. 2015.

669


Page 34 of 34

Spatial associations and chemical composition of organic carbon

Recommend Documents