P O S I V A O Y

FI -27160 OLKILUOTO, F INLAND

Tel +358-2-8372 31

Fax +358-2-8372 3709

Pent t i Mann inen

Jan i Mäke lä

October 2006

Work ing Repor t 2006 -42

Humic and Fulvic Acids in Groundwater– A Preliminary Research

October 2006

Working Reports contain information on work in progress

or pending completion.

The conclusions and viewpoints presented in the report

are those of author(s) and do not necessarily

coincide with those of Posiva.

Pentt i Mann inen , Jan i Mäke lä

I ns inöö r i to im is to Paavo R is to l a Oy

Work ing Report 2006 -42

Humic and Fulvic Acids in Groundwater– A Preliminary Research

HUMIC AND FULVIC ACIDS IN GROUNDWATER - a preliminary research

ABSTRACT Two different isolation methods, non-ionic polymeric sorbent (XAD-8) and weakly basic anion exchanger (DEAE cellulose), were tested and applied to the isolation of humic and fulvic acids from groundwater. Both of these methods were shown to be applicable to the isolation. The XAD-8 procedure and the procedure using DEAE cellulose were able to isolate 40% of the DOC as humic solutes. For the investigation of the structure and molecular size distribution of the isolated humic solutes, the hyphenated SEC-ESI-MS system with TOF was utilized. The system was shown to be a powerful instrument especially for the investigation of fulvic acids. For the higher-molecular-weight humic acids, the ESI-MS loses sensitivity compared with the parallel UV detection, because of the difficulty in getting the ionized humic compounds to fly efficiently through the mass spectrometer.

Keywords: Humic and fulvic acids, groundwater, XAD, DEAE, SEC, mass spectrometry.

HUMUS- JA FULVOHAPOT POHJAVEDESSÄ - alustava tutkimus

TIIVISTELMÄ Kahta erilaista eristämismenetelmää, adsorptiota polymeerisorbenttiin (XAD-8) ja anioninvaihtohartsiin (DEAE selluloosa), testattiin ja käytettiin humus- ja fulvohappojen eristämiseksi pohjavedestä. Molemmat menetelmät osoittautuivat sopiviksi eristämiseen. Noin 40 % liuenneesta orgaanisesta aineksesta saatiin eristettyä sekä XAD-8 että DEAE menetelmällä. Eristettyjen humus- ja fulvohappojen rakenne ja molekyylikokojakauman tutkimiseen käytettiin SEC-ESI-MS laitteistoa lentoaikamassaspektrometrillä (TOF). Laitteisto osoittautui tehokkaaksi erityisesti fulvohappojen tutkimuksessa. Raskaamman molekyylipainon omaavilla humushapoilla ESI-MS laitteiston herkkyys laskee verrattuna rinnakkaiseen UV-detektointiin koska ionisoidut humus aineet on vaikea saada lentämään tehokkaasti massaspektrometrin läpi.

Avainsanat: Humus- ja fulvohapot, pohjavesi, XAD, DEAE, SEC, massaspektrometria.

1

TABLE OF CONTENTS

ABSTRACT TIIVISTELMÄ 1 INTRODUCTION......................................................................................... 2

2 EXPERIMENTAL ........................................................................................ 3

2.1 Sampling ......................................................................................................... 3 2.2 Chemicals and reference compounds ............................................................ 3 2.3 Isolation........................................................................................................... 4

2.3.1 Method development............................................................................... 4 2.3.2 Ground water samples ............................................................................ 5

2.4 TOC analysis .................................................................................................. 5 2.5 UV-spectrum ................................................................................................... 5 2.6 HPLC-TOF-MS ............................................................................................... 5

3 RESULTS ................................................................................................... 7

3.1 TOC results..................................................................................................... 7 3.1.3 XAD-8 procedure .................................................................................... 7 3.1.4 DEAE-procedure ..................................................................................... 8

3.2 UV-results ....................................................................................................... 9 3.3 MS results ..................................................................................................... 12

3.3.5 SEC-MS runs ........................................................................................ 14 4 DISCUSSION............................................................................................ 22

5 CONCLUSIONS........................................................................................ 23 REFERENCES................................................................................................. 24

2

1. INTRODUCTION

Aquatic humic substances are polar brown-coloured organic acids, which are derived from soil humus and terrestrial and aquatic plants. They generally comprise one third to over half of the dissolved organic carbon (DOC) in natural surface waters.

Because of the dilute nature of aquatic humic substances, the material must first be isolated and concentrated for further studies. Dissolved humic matter contains a large number of diverse chemical functionalities and no single analytical method is available to reveal the actual concentration of these macromolecular organic solutes in the sample. The main difficulty is that the aquatic humic solutes must be separated from the other organic and inorganic solutes in natural water. Different methods for concentrating and isolating the dissolved organic matter such as freeze drying, chemical precipitation, solvent extraction, reverse osmosis, ultra filtration and adsorption to solids have been presented. All of these methods have some disadvantages of not isolating the aquatic humic matter and, at the same time, keeping the chemical and structural composition unaffected as it occurs in its original state in an aquatic environment.i

The classification of dissolved organic matter into humic and non-humic solutes is based on the isolation method. The isolation method and the chemical conditions applied determine the degree of separation of humic from non-humic matter. The fraction of humic substances that are not soluble in water at a pH lower than 2, are defined as humic acids. The fraction of humic substances that are soluble under all pH conditions are referred to as fulvic acids.ii

The most frequently used isolation procedure for humic substances is the column chromatographic method by non-ionic sorbents such as XAD resins (polymethyl methacrylate) or analogous materials.iii, iv, v The XAD resin classifies organic solutes in an acidified water sample into artificial hydrophobic and hydrophilic fractions according to their random ability to adsorb onto non-ionic macroporous resin. This operational classification of aquatic humic substances is clear but, on the other hand, the XAD isolation technique under strongly acidic conditions might involve certain risks for uncontrolled fractionation or reactions.

Since the aquatic humic matter has a weak acidic character, this property has been used by employing anion exchange resins to isolate the humic solutes. The most popular anion exchanger for this purpose has been the DEAE (diethylaminoethyl) cellulose.v, vi,

vii It is a weakly basic anion exchanger with tertiary amine functional groups bound to hydrophilic matrix. With the DEAE cellulose it is possible to isolate almost all organic acids from water without pH adjustment of the original water sample.

The preliminary goal of the present study was to determine the amount of humic substances in the groundwater from a shallow depth in the bedrock at Olkiluoto. A secondary goal was to determine the properties such as the structure and molecular size distribution of the humic substances possibly found in this water.

Two different isolation procedures were applied; the adsorption of humic substances on non-ionic macroporous resin (XAD-8 procedure) and isolation with weakly basic anion exchanger. The hyphenated SEC-ESI-MS system with TOF (Size exclusion chromatography with time-of-flight mass spectrometer using electrospray ionization) was applied to the investigation of the structure and molecular size distribution.

3

2. EXPERIMENTAL

2.1 Sampling The lake water sample for the method development was collected from Lake Vähä-Tiilijärvi situated in Hollola, Finland, in December. The water sample was collected from the surface into 2.5-L brown glass containers. The sample was filtrated (0.45 µm, cellulose nitrate filter, Sartorius AG Goettingen, Germany) as soon as possible after sampling and stored in the dark at 6 ºC. Groundwater samples were collected from groundwater station 1 (ONK-PVA1, depth about –20m b.s.l.) in ONKALO, Olkiluoto, Finland. Samples were on-line pressure filtered (Pall-Gelman) in situ, using Millipore HA 47 mm, 0.45µm filters. Tubing material used in the sampling was nylon. The glass containers (1-5 L) that were used to store the samples were washed with 6% HNO3, high purity MQ water and ethanol. The samples were protected from the light by wrapping the glass containers in aluminium foil and stored at 6 ºC.

2.2 Chemicals and reference compounds 1.0 M HCl and NaOH (Reagecon, Shannon Free Zone, Shannon, Co. Clare, Ireland) were used by diluting to the desired concentration with MQ water. Hexane (Fluka) and methanol (J. T. Baker) were “for residue analysis” or “Baker Analyzed” quality. Standard solutions containing humic and fulvic acid were made by dissolving IHSS (International Humic Substances Society) Nordic Aquatic Humic Acid Reference (1R105H) and IHSS Nordic Aquatic Fulvic Acid Reference (1R105F) in MQ water. Diethylaminoethyl-cellulose (Macherey-Nagel) was pre-treated according to Miles et al. vii Approximately 50 g of dry cellulose was stirred for 1 h in a 1000-mL beaker with 0.5 M HCl. The cellulose was rinsed with MQ water in a Bühner funnel until the pH was neutral. After this, the DEAE cellulose was resuspended in 0.5 M NaOH and stirred for 1 h in a 1000-mL beaker. The solution was rinsed in a Bühner funnel with MQ water until the pH was neutral. The DEAE cellulose was dried in a Bühner funnel and stored in the dark at 6 ºC until used. XAD-8 resin (Supelite DAX-8, Supelco, a new substitute for XAD-8 resin) was cleaned according to Thurman et al. iv with slight modifications. The XAD-8 material was first cleaned by rinsing the resin in a beaker with 0.1 M NaOH and thereafter with methanol until the pH was neutral. Next, the resin was Soxhlet-extracted sequentially for 6 h with methanol and hexane. XAD-8 was stored in methanol at 6 ºC until used.

4

2.3 Isolation

2.3.1 Method development The lake water samples were used to develop the method for the groundwater samples. After the filtration, the lake water sample was divided into two parts. The other part was acidified to a pH just above 2 with 1.0 M HCl and treated on the XAD-8 resin. The lake water sample with natural acidity (pH 6.8) was treated on the DEAE cellulose without pre-acidification. Different procedures were tested in order to find the best method for the isolation of humic substances with the DEAE cellulose. Column procedure with XAD-8 The XAD-8 polymer in methanol was slurry-packed into a 160-mm I.D. class column, resulting in an 8-cm sorbent bed. The column was rinsed with MQ water until free of methanol. The acidified lake water sample was placed in a 1.0-L separator funnel and dropped through the XAD polymer. The humic and fulvic substances adsorbed on the resin were eluted with 0.1 M NaOH. All the fractions of the eluate were collected in order to find out the sufficient amount of NaOH needed. Batch procedure with DEAE cellulose 1.0 L of the lake water with natural acidity was placed in a 1.0-L brown borosilicate glass container and approximately 10 g of DEAE cellulose was added. The samples were stirred for different periods of time (1, 2 and 3 hours). After that, the samples containing the cellulose were filtrated with a Bühner funnel. The organic substances retained on the cellulose were eluted directly from the Bühner funnel with 0.1 M NaOH. Column procedure with DEAE cellulose Approximately 10 g of DEAE cellulose was placed in a beaker containing MQ water. The slurry was poured into a 160-mm I.D. class column. The column was rinsed with MQ water until the pH was neutral. The 1.0-L lake water sample was applied to the column at about 1.4 mL/min. Funnel procedure with DEAE cellulose Approximately 15 g of DEAE cellulose was placed in a funnel and rinsed with MQ water until the pH was neutral. The lake water sample was placed in a 1.0-L separatory funnel and let run through the cellulose. The organic substances retained onto the DEAE cellulose were eluted with 0.1 M NaOH.

5

2.3.2 Groundwater samples 2.0-L samples of the groundwater were preconcentrated by the described column procedure with the XAD-8 polymer. The sample was placed in a 1.0-L separatory funnel and dropped through the XAD polymer. The funnel procedure was chosen to preconcentrate the groundwater samples (2.0 L/sample) on DEAE cellulose. 15 g of DEAE cellulose was placed in a funnel and rinsed with MQ water until the pH was neutral. The sample with natural acidity was placed in a 1.0-L separatory funnel and let run through the funnel containing the cellulose at approximately 4.0 mL/min. The organic humic material retained on the cellulose was eluted with 80 mL of 0.1 M NaOH.

2.4 TOC analysis The TOC-V CPH analyzer is able to analyze samples only in liquid state. In TOC analysis, both inorganic and organic forms of carbon (TC, total carbon) are converted into carbon dioxide (CO2) using a chemical oxidizing agent (platinum on alumina pellets) and a high temperature such as 680 °C. The CO2 formed is then detected using non-dispersive infrared adsorption (NDIR). When measuring the amount of inorganic carbon in the sample, it is first acidified with HCl. The CO2 formed from the inorganic forms of carbon (IC) is purged from the sample with air and detected with an NDIR detector. The amount of total organic carbon (TOC) can then be calculated as a difference of total carbon and inorganic carbon. The TOC analyzer was calibrated in the range of 0.5 to 100 mg/l for organic and inorganic carbon. Quality control samples (2.0 and 20.0 mg/L, QC WW4) were analysed at the beginning and at the end of each run. The TOC measurements were performed with the TOC-V CPH analyzer connected to an ASI-V autosampler (Shimadzu Corporation). The samples were neutralized with 1.0 M HCl before the measurements.

2.5 UV spectrum Ultraviolet absorbance was measured at pH 7 with a Shimadzu UV-1601 spectrophotometer using a 10-mm quartz vial. 1:50 dilutions of samples neutralized with 1.0 M HCl were measured by scanning from 190 nm to 500 nm.

2.6 HPLC-TOF-MS Size exclusion chromatography (SEC) and UV spectrometry analysis were performed using a PL Aquagel-OH 30 8 µm column (300 mm x 7,5 mm), 80/20 H2O/MeOH + 10 nm NH4HCO3 eluent (1 mL/min) and a Bruker Micro TOF-MS instrument (Bruker Daltonics Inc.) with an Agilent 1100 series HPLC system including two-channel UV spectrometry (VWD G1314A) and an autosampler. The injection volume was 50 µL (20 µL for humic and fulvic acid standards) and the split ratio was 1:3. The samples were also screened by injecting samples (4 µL/min) straight to the mass spectrometry and by scanning the mass range between 100 to 500 and 500 to 3000 amu. The instrument parameters are listed in Table 1.

6

Table 1. Instrumental parameters of electrospray ionization, negative mode (ESI(-)) scanning from straight injections. Mass range (amu) 100-500 500-3000 Nebulizer gas (bar) 1.6 1.6 Drying gas temperature (°C) 200 200 Drying gas (l/min) 8 8 Ion polarity negative negative Capillary exit (V) -90 -80 Hexapole RF (V) 90 500 Skimmer (V) -30 -35 Hexapole 1 (V) -25 -21 Instrument parameters for SEC-MS are shown in Table 2. TOF Voltages: Pulser +/- 4+8 V, Fill –56 V, Extract +1076 V, Flight tube +9000 V, Reflector –1296 V and TOF MCP detector –2040 or –2150 V. Sodium formate was used for exact mass calibration during each SEC-MS run by pumping a calibration solution straight to the mass spectrometer after the sample peaks. At the same time, the eluent from the column was connected to the waste, thus preventing a large salt peak from entering the ion source. UV data were collected by monitoring a signal from 254 nm. Same samples were run on several days to confirm that the chromatographic conditions were repeatable. Table 2. Instrumental parameters for SEC-MS. Nebulizer gas (bar) 1.6 Hexapole RF (Vpp) 90 Drying gas (l/min) 8 Transfer time (µs) 20 Drying temperature (°C) 200 Pre-pulse storage (µs) 10 Capillary voltage (V) 4500 Lens 1 storage (V) -50 Capillary exit (V) -90 Lens 1 extraction (V) -20.6 Skimmer 1 (V) -30 Lens 2 (V) -2 Hexapole 1 (V) 25 Lens 3 (V) 30.5 Hexapole 2 (V) -19 Lens 4 (V) 1.0 Skimmer 2 (V) -24 Lens 5 (V) 33.5

7

3. RESULTS

3.1 TOC results

3.1.3 XAD-8 procedure The XAD-8 procedure was tested with the lake water samples. Table 3 shows the results from the method development. This procedure was chosen to be used for the groundwater samples, because the results were acceptable and the procedure was easy to accomplish. Table 3. Recovery percentages of total organic carbon in different fractions of eluent (0.1 M NaOH). 1. 2. Original TOC 100% 100% 1. 20 mL 44.8% 44.0% 2. 20 mL 7.5% 3. 20 mL 2.8% sample passed the XAD-8 44.8% 54.3% Four parallel 2.0-L samples of groundwater (ONK-PVA1) and one blank sample (MQ water) were preconcentrated with the XAD-8 procedure. The adsorbed humic and fulvic substances were eluted with 60 mL of 0.1 M NaOH in order to ensure complete elution. The results from the TOC measurements and the calculated recovery are shown in Table 4. Table 4. Recovery percentages of total organic carbon (mg/isolated sample) using the XAD procedure. Sample 1. Sample 2. Sample 3. Sample 4. TOC, original sample, mg

17.3 16.0 16.1 75.1

TOC, isolated sample, mg

6.9 6.6 5.9 6.9

% of original TOC

39.9 41.3 36.6 9.2

8

3.1.4 DEAE procedure Various procedures were tested for the isolation of humic substances from lake water samples with DEAE cellulose. Table 5 gives the results of the TOC measurements and the calculated recoveries for humic substances obtained with different isolation methods. Table 5. Recovery percentages of total organic carbon (mg/isolated sample) in lake water samples using the DEAE procedure. Sample

1. Sample 2. Sample

3. Sample 4.

Sample 5.

Sample 6.

Procedure Batch Batch Funnel Funnel Column Column TOC, original sample, mg

6.0 6.2 6.9 6.7 6.5 8.0

TOC, isolated sample, mg

4.4 4.0 4.4 4.7 4.0 5.5

% of original TOC

73 65 64 70 62 69

The funnel procedure was chosen to preconcentrate humic substances from the groundwater samples (ONK-PVA1), since the recovery with the funnel procedure was comparable with the two other procedures and it was easy to accomplish. Three parallel 2.0-L samples and one blank sample (MQ water) were preconcentrated on the DEAE cellulose. The adsorbed humic and fulvic substances were eluted with 80 mL of 0.1 M NaOH in order to ensure complete elution. The results from the TOC measurements and the calculated recoveries are shown in Table 6. In all of the TOC measurements, quality control samples were analysed at the beginning and at the end of each sample run. The results were acceptable according to the laboratory quality control procedure. Table 6. Recovery percentages of total organic carbon (mg/isolated sample) in groundwater samples using the DEAE procedure. Sample 1. Sample 2. Sample 3. TOC, original sample, mg

23.16 22.12 22.82

TOC, isolated sample, mg

8.72 7.88 9.77

% of original TOC 37.7 35.6 42.8

9

3.2 UV results

The UV absorbance was measured for the lake water and groundwater samples (ONK-PVA1) isolated with both techniques. In the blank samples there was no significant absorbance as shown in Figures 1 and 2.

Figure 1. The UV spectrum of the blank sample isolated with the XAD technique.

Figure 2. The UV spectrum of the blank sample isolated with the DEAE procedure.

10

The spectrums of the lake water samples preconcentrated with either the XAD-8 procedure or the DEAE cellulose were similar as shown in Figures 3 and 4. At approximately 254 nm, a clear peak can be seen in the samples isolated with both techniques, which shows the presence of organic compounds.

Figure 3. The UV spectrum of the lake water sample isolated with the XAD technique.

Figure 4. The UV spectrum of the lake water sample isolated with DEAE cellulose.

11

The spectrums of the groundwater samples (Figures 5 and 6) were very similar to the spectrums obtained from the lake water samples.

Figure 5. The UV spectrum of the ground water sample isolated with the XAD technique.

Figure 6. The UV spectrum of the ground water sample isolated with DEAE cellulose.

12

The spectrum of the humic and fulvic acid standard (100 g/mL of both, IHSS Nordic Aquatic Humic Acid and Fulvic Acid Reference in MQ water) shown in Figure 7 is also similar to the spectrums of the lake and deep rock groundwater samples.

Figure 7. The UV spectrum of the humic and fulvic acid standard (100 mg/L).

3.3 MS results Both blanks (XAD and DEAE) showed very similar mass spectrums when injected straight into the mass spectrometry. Several mass peak patterns were detected from 200 to 1400 amu. These clusters differed about 58 to 60 amu from each other. The main peaks were 502.64, 560.59, 618.55, 678.51, 736.47 and 749.43 amu in both blanks (range 500 to 3000) and 268.81, 326.77, 384.73 (range 100 to 500 amu). These peaks are most probably artifacts from the isolation procedure or have built up in the ionization system. A direct injected humic and fulvic acid standard mixture (100 mg/L of both, humic and fulvic acid in MQ water) showed a very intensive mass spectrum with a homologue series differing 1 or 2 mass units (Figures 8, 9, 10 and 11). The mass range of 100 to 500 included fewer components, and this mass range was selected for further studies.

13

0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40150000

160000

170000

180000

190000

200000

210000

220000

230000

240000

250000

260000

270000

280000

290000

300000

Time-->

Abundance

TIC: msd1.ms

Figure 8. Mass signal of the direct injected humic and fulvic acid standard.

0 200 400 600 800 100012001400160018002000220024002600280030000

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

m/z-->

Abundance

Average of 0.258 to 2.496 min.: #04 HUMUS-FULVOSTAND.D551

719

893

1085

1253381

14381691

18652044 235925682771298845

Figure 9. A mixture of the humic and fulvic standard shows a very complex mass spectrum when the data is collected from the range of 50 to 3000.

14

0.200.400.600.801.001.201.401.601.802.002.202.402.602.803.00

200000

210000

220000

230000

240000

250000

260000

270000

280000

290000

300000

310000

320000

330000

340000

350000

360000

370000

380000

390000

Time-->

Abundance

TIC: msd1.ms

Figure 10. Mass signal of the direct injected humic and fulvic acid standard when masses only from 50 to 500 were collected.

1401601802002202402602803003203403603804004204404604805000

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

m/z-->

Abundance

Scan 75 (1.139 min): #05 HUMUS-FULVOSTAND.D281 309

325263

353247

225

371

389

205

407

165181

431

447

469

487

505

146

Figure 11. A less complex mass spectrum was obtained from a mixture of the humic and fulvic acid standard when masses only from 50 to 500 were collected.

3.3.5 SEC-MS RUNS Figures 12 to 25 show the results of the size exclusion chromatography (SEC) with ultra violet and mass spectrometry detection of the blanks, standards and samples. Chromatographic data is presented as total ion chromatogram of all masses together with a UV signal at 254 nm. A background-subtracted mass spectrum, which was summed across the detected mass peak, is also shown. Figures 12-15 show the XAD and DEAE blank samples respectively. There is actually no MS peak in either chromatogram. A small peak in the DEAE blank UV response is

15

observed at a retention time of about 7.9 min. The mass spectrum shown is just a background.

Figure 12. Mass and UV signal of the XAD blank sample.

50 100 150 200 250 300 350 400 450 500 5500

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

m/z-->

Abundance

Average of 0.592 to 13.315 min.: #24 XNOLLA AJO1.D281

339

219

255

183

311

395445366 49513549 100 417157 471 517 556

Figure 13. Mass spectrum of the XAD blank sample.

Figure 14. The mass and UV signal of the DEAE blank sample.

16

50 100 150 200 250 300 350 400 450 500 550 6000

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

m/z-->

Abundance

Average of 0.531 to 13.656 min.: #25 DNOLLA AJO1.D281

339219

255

183

311395

445366 495 543 599141 51873 419 471

Figure 15. The mass spectrum of the DEAE blank sample. The DEAE isolation sample of groundwater sample ONK-PVA1 is shown in Figures 16 and 17. The main UV peak appears at a retention time of 7.6 min and the main mass peak at a retention time of 9 min. All the signals were collected at the same time. The time difference between UV and mass signal is only in seconds and it was not corrected owing to a minor effect on the results. The two chromatograms differ, indicating a response to different weight molecules. UV detects higher-molecular-weight compounds, while the selected mass window of 50-500 amu detects lighter compounds. The mass spectrum shows that the mass peak contains a complex profile of organic compounds and homologue series. The most intensive series has a profile where the mass peaks differ 14 amu from each other, indicating -CH2 i.e. methylene group differences among congeners. Other replicate DEAE PVA1 samples showed equal UV chromatogram profiles. In the mass signal there were slight differences, but these did not affect the main results and conclusions.

Figure 16. The mass and UV signal of the DEAE isolation sample of ONK-PVA1.

17

1601802002202402602803003203403603804004204404604805005205400

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

m/z-->

Abundance

Average of 8.431 to 9.632 min.: #27 D1 AJO2.D (-)195

315

217

281

333 357

385 413255

441235 298177 459479497

515 541153

Figure 17. The mass spectrum of the DEAE isolation sample of ONK-PVA1 (peak at a retention time of 9 min). The XAD isolation sample of the groundwater sample ONK-PVA1 is shown in Figures 18-20. The main UV peaks are at retention times of about 6.2, 6.8, 7.8 and 8.6 min. Two intensive mass peaks were observed at retention times of about 6.8 and 8.6 min. The mass spectrum shows the same kind of homologue series as in the DEAE samples, but the most intensive fragments are at two mass units lower than in the DEAE samples. The mass spectrums differ somewhat between the 6.8-min and 8.6-peaks. The later eluated compounds have a –4-amu lighter pattern than compounds in the 6.8-min peak, the profile is also slightly different, indicating minor structure differences between the compounds in those peaks. A replicate XAD sample showed equal grams and spectrum.

Figure 18. The mass and UV signal of the XAD isolation sample of ONK-PVA1.

18

50 100 150 200 250 300 350 400 450 500 550 6000

500

1000

1500

2000

2500

3000

3500

4000

4500

m/z-->

Abundance

Average of 6.554 to 7.162 min.: #36 X3 AJO4.D (-)195 311

353283

379

255

409

435227

463493

521169 545575140 60178 10649

Figure 19. The mass spectrum of the XAD isolation sample of ONK-PVA1 (first peak at a retention time of 6.8 min).

50 100 150 200 250 300 350 400 450 500 550 6000

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

m/z-->

Abundance

Average of 8.393 to 8.773 min.: #36 X3 AJO4.D (-)279

321

349

377253

405

433223

461183

48751353913647 563 59885110

Figure 20. The mass spectrum of the XAD isolation sample of ONK-PVA1 (second peak at a retention time of 8.6 min). Two intensive mass peaks were observed at retention times of about 6.8 and 8.6 min. The mass spectrum shows the same kind of homologue series as in the DEAE samples, but the most intensive fragments are at two mass units lower than in the DEAE samples. The mass spectrums differ somewhat between the 6.8-min and 8.6-peaks. The later eluated compounds have a –4-amu lighter pattern than compounds in the 6.8-min peak, the profile is also slightly different, indicating minor structure differences between the compounds in those peaks. A replicate XAD sample showed equal grams and spectrum.

19

The fulvic acid standard (648 mg/L of fulvic acid in MQ water) chromatograms are shown in Figures 21-23. The main UV peaks are at retention times of 6.0, 6.5 and 7.2 min and the main mass peaks are at about 6.8 and 7.6 min. The mass spectrums of these peaks are similar to those of the XAD sample except for the 7.6-min peak, where the pattern (most intensive mass fragments) differs –2 amu from the 6.8 min peak spectrum. The earlier eluting peak also contains higher mass compounds.

Figure 21. The mass and UV signal of the fulvic acid standard.

50 100 150 200 250 300 350 400 450 500 550 6000

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

m/z-->

Abundance

Average of 6.365 to 6.897 min.: #39 FA-STD AJO1.D (-)297

325

353

225

379

267

409

197

435

463

489165519

54957314010982 60145

Figure 22. The mass spectrum of the fulvic acid standard (first peak at a retention time of 6.8 min).

20

Figure 23. The mass spectrum of the fulvic acid standard (second peak at a retention time of 7.6 min). The humic acid standard (622 mg/L of humic acid in MQ water) chromatograms are shown in Figures 24-25. The main UV peaks are at retention times of 5.8 and 6.5 min. The main mass peak is at a retention time of about 6.6 min. The mass spectrum is slightly broader and has fewer patterns than that of the fulvic acid standard.

Figure 24. The mass and UV signal of the humic acid standard.

50 100 150 200 250 300 350 400 450 500 550 6000

200

400

600

800

1000

1200

1400

1600

m/z-->

Abundance

Average of 7.216 to 7.930 min.: #39 FA-STD AJO1.D (-)295

323267

351

379239

407213433

181 461 491 521 549 573153114 60356 85

21

50 100 150 200 250 300 350 400 450 500 550 6000

50

100

150

200

250

300

350

400

450

500

550

600

650

m/z-->

Abundance

Average of 6.306 to 6.913 min.: #40 HA-STD AJO1.D (-)225

281

181

339251

307

363

393447419

471 503527555

153 5851279168

Figure 25. The mass signal of the humic acid standard (peak at a retention time of 6.6 min). The natural lake water samples were run on different days with different settings. The chromatograms are not comparable with those chromatograms discussed above, but the XAD and DEAE samples can be compared with each other. The chromatograms are quite alike. The greatest difference is that in the XAD sample the UV and mass peaks show more peaks or shoulders than in the DEAE run. The mass spectrums are almost equal, the highest mass peak in the patterns being occasionally –2 amu in the XAD run compared with the DEAE run.

22

4. DISCUSSION The XAD-8 procedure was first tested with lake water samples. Table 3 shows the results from the method development. The results are in good agreement with the results by Peuravuori et al 3, where 53% of the DOM (dissolved organic matter) was recovered as hydrophopic acids when eluted with base, and this column procedure was easy to accomplish. The tested column procedure was applied to the isolation of humic and fulvic acids from groundwater samples from ONKALO, Olkiluoto, from a shallow depth (-20 m b.s.l.). Table 2 shows the results of the TOC measurements of the isolated samples. The TOC measurement of sample 4 was repeated, but the total organic carbon content remained at a high level. The sample bottle is probably contaminated with substances that are not retained on XAD, because the TOC content in the eluate is at the same level as in the other samples. When calculating the recovery with the average of the TOC in the three other samples, the recovery of the organic humic substances in sample 4 would be 41.9%. The average recovery is 39.3%. The blank sample (MQ water) showed no significant amount of total organic carbon. Three different approaches were tested for the isolation of humic and fulvic acids with DEAE cellulose. In the batch procedure, the different stirring times (1, 2 and 3 hours) did not have any influence on the recovery. With this procedure, the main difficulty was to quantitatively recover the cellulose from the class container for the elution stage. In the column procedure, the flow rate of the sample through the DEAE cellulose-packed column was not sufficient, considering the 2.0-L sample amount. Since the recovery with the funnel procedure was comparable with the two other procedures and the procedure was easy to accomplish, it was chosen to isolate the humic and fulvic acids from the groundwater samples. The blank sample (MQ water) showed no significant amount of total organic carbon. The repeatability of the SEC-MS runs was moderate to good, although some instability of the system was observed. Some of the problems were located in the stability of the mass spectrometry instrument settings and the column stability, while some other problems were due to the clogging in the multi-port valve system between the column and the ion source. The hyphenated SEC-ESI-MS system with the TOF is a very powerful instrument especially for fulvic acid research. Humic acids are easily detected by the SEC-UV system, but there seems to be much more problems to get ionized humic compounds to fly efficiently into and through the mass spectrometer. The results of fulvic acids obtained in this study are very comparable to those by Theseviii and Reemtsma.ix UV chromatograms and mass spectrums are the most identical, indicating that fulvic acids are indeed more “symmetrical” in structure than has been thought before. The most considerable mass differences in the mass patterns were -2 and -14 amu. The former originates probably from the double bonds of the base carbon skeleton and the latter is probably due to the methylene group in the alkyl chain homologues. Exact mass determination and molecular structure calculation are not possible with the mass spectrometry techniques based on the quadrupole time-of-flight technique. If the exact mass is needed, a more expensive mass spectrometer instrument like FTICR-MS must be used.

23

5. CONCLUSIONS

The characterization and detection of fulvic and humic acids in groundwaters is possible with the methods described in this report. The XAD and DEAE isolation and concentration procedures are both applicable to fresh and brackish water samples (ONK-PVA1, total dissolved solids about 900-1100 mg/L), even though their selectivity and capacity slightly differ from each other. The effect of a high salt concentration on the retention of acidic solutes on the DEAE cellulose needs to be tested further. It has been discovered that even low salinity of brackish water will decrease the retention of low-molecular-weight acidic solutes and that the homogeneity of isolated humic substances increases as the salinity increases.x On the other hand, it has been stated that no special desalting procedure, in addition to the treatment with DEAE cellulose, is required.iv Samples with a high salt concentration need to be tested separately and a desalination procedure may be necessary before the isolation of humic and fulvic acids. Since the humic-type organic compounds have a relatively high content of phenolic functional groups, in addition to acidic groups, the cross-linked polyvinylpyrrolidone (PVP) could also be utilized for the fractionation of the DOM (Dissolved Organic Material) into humic-type constituents. The PVP resin forms strong hydrogen bonds with phenolic, hydroxyl and carboxyl groups of the DOM, while the XAD polymer classifies organic humic solutes in a water sample according to their relative hydrophobic–hydrophilic interactions with the sorbent bed. With SEC-UV, the quantification and separation of humic and fulvic acids is possible, and the apparent molecular weights and their distribution can be approximated using molecular weight standards.

24

REFERENCES

iG. R. Aiken, Isolation and concentration techniques for aquatic humic substances, in: G. R. Aiken, D. M. McKnight, R. L. Wershaw, P. MacCarthy (Eds.), Humic substances in soil, sediment and water: Geochemistry, Isolation and Characterization, Wiley, NY, 1985, pp. 363-385. ii Ronald L. Malcolm; Analytica Chimica Acta, 232, 1990, 19-30 “The uniqueness of humic substances in each of soil, stream and marine environments” iii Jerry A. Leenheer; Environmental Science & Technology, 1981, 15, 578-587 “Comprehensive approach to preparative isolation and fractionation of dissolved organic carbon from natural waters and wastewaters” iv Earl M. Thurman and Ronald L. Malcolm; Environmental Science & Technology, 1981, 15, 463-466 “Preparative isolation of aquatic humic substances” v Juhani Peuravuori, Kalevi Pihlaja, and Niina Välimäki; Environmental International, Vol 23, No. 4, pp. 453-464 “Isolation and characterization of natural organic matter from lake water: Two different adsorption chromatographic methods” vi Juhani Peuravuori, Alvaro Monteiro, Linda Eglite, and Kalevi Pihlaja; Talanta, 65, 2002, 408-422 ”Comparative study for separation of aquatic humic-type organic constituents by DAX-8, PVP and DEAE sorbing solids and tangential ultrafiltration: elemental composition, size-exclusion chromatography, UV-vis and FT-IR. vii Carl J. Miles, John R. Tuschall, Jr., and Patrick Brezonik; Analytical Chemistry, 1983, 55,410-411 “Isolation of aquatic humus with diethylaminoethylcellulose” viii These Anja, Winkler Marcus, Thomas Christoph, Reemtsma Thorsten; Rapid Communication in Mass Spectrometry, 2004, 18, 1777-1786 “Determination of molecular formulas and structural regularities of low molecular weight fulvic acids by size-exclusion chromatography with electrospray ionization quadrupole time-of-flight mass spectrometry” ix Reemstsma Thorsten, These Anja, Analytical Chemistry, 2003, vol 75, 6, 1500-1507 “On-line coupling of size exclusion chromatography with electrospray ionization-tandem mass spectrometry for the analysis of aquatic fulvic and humic acids” x Catharina Pettersson, Lars Rahm; Environment International, Vol. 22, No. 5, pp. 551-558, 1996 ”Changes in molecular weight of humic substances in the Gulf of Bothnia”