This procedure describes a method for the determination of total organic carbon (TOC) and total nitrogen (TN) in sea water, expressed as micromoles of carbon (nitrogen) per liter of sea water. TOC includes both dissolved and particulate organic carbon (DOC and POC, respectively). TN includes particulate and dissolved organic nitrogen as well as dissolved inorganic nitrogen species. The method is suitable for the assay of oceanic levels of total organic carbon (<200 µmol·L-1) and total nitrogen (<50 µmol·L-1). The instrument discussed and procedure described is specific to the instrument employed in the Aluwihare Laboratory at the Scripps Institution of Oceanography, and whenever possible, closely follows protocols used by Jonathan Sharp (U. of Delware), Craig Carlson (U. of California, Santa Barbara) and Dennis Hansell (RSMAS). SIO has the Shimadzu TOC-VSN, normal sensitivity, with TN unit.
An acidified seawater sample (< pH 3) is sparged with CO2 free air to remove inorganic carbon. The water is then injected onto a combustion column packed with platinum-coated alumina beads held at 680°C. Non-purgeable organic carbon (NPOC) compounds are combusted and converted to CO2 , which is detected by a nondispersive infrared detector (NDIR). Non-purgeable nitrogen compounds (which include organic and inorganic nitrogen species) are combusted and converted to NO, which when mixed with ozone chemiluminesces for detection by a photomultiplier.
Care must be taken to minimize contamination of the sample. Gloves should be worn during sampling. Every effort must be made not to touch the Niskin's nipple (the path of the water stream, from Niskin to sample bottle, must be kept very clean). Grease (whether mechanical grease from ship operations or sealing grease as employed for some gas sampling) should never be allowed to come in contact with the sample nipple.
Prior to sampling, 40 mL borosilicate glass vials are combusted for 4 hours at 450 °C. Caps and teflon cap liners are cleaned by a 4 hour soak in 10% hydrochloric acid, and then copiously rinsed with distilled water, followed by MilliQ water. Caps and liners are covered and then dried overnight at 40°C. Vials are capped and re-assembled as soon as they are removed from the combustion furnace
2.2. Sample Collection
Sample bottles must be rinsed three times with water from the Niskin bottle prior to filling. Vials should be filled to approximately 90%, or 35 mL. Acid should be added immediately after sample collection using the teflon dropper bottles containing concentrated, trace metal grade HCl (2 drops). Vials must be tightly capped and stored at room termperature.
All samples collected on CalCOFI and LTER cruises are TOC and TN samples. That is, no filtering of the sample is done to remove particulates. Filtering is time-intensive and filtration, especially by untrained personnel, can introduce significant contamination. So, to maximize efficiency and minimize sample handling, the filtration step common to DOC and TDN is omitted. In open ocean waters particulate organic carbon concentrations are a small fraction of the total organic carbon and do not pose too many problems. Here subtracting POC from TOC is likely to provide an accurate estimate of DOC because particles are typically small and homogeneously distributed in the sample. In coastal waters, and at stations where relatively high chlorophyll concentrations are present, the TOC measurement cannot be easily converted to DOC by subtracting POC values. Experience has shown that particles in these regions are large and inhomogeneously distributed. Therefore, samples collected in this study are reported as TOC and TN.
2.3. Sampling Depths
Water samples are take directly from the rosette. Full profiles are taken at all SCCOOS stations (4-5 vials per station). Line 93.3 most stations (except .55 and .90) are "normal" stations, which means samples are drawn from the surface niskin, the deep niskin, and three LTER depths (5 vials in all). At 93.3.55 and 93.3.90 samples are drawn from niskins triggered at 13 depths between 0 and 170 m, and 230 m, 320 m and the deepest niskin (16 vials). Line 90.0 contains normal, LTER cardinal, and full profile stations. Normal stations (.28, .30, .35, .45, .60, .80, and .100) sample surface, deep, 3 LTER depths, 140/145m, and 320 m. Cardinal stations (.37, .70, .90 and .120) sample surface, deep, 6 LTER depths, 140/145m, 230 m and 320 m. Full profiles are collected at station .100 and .53 (20-24 vials). Line 86.7 is primarily sampled at the surface, deepest, 3 LTER depths, 140/145m and 320m. A full profile is taken at the Santa Monica Basin station (86.7.40). 16 samples are collected at .60 and .100, which include deepest, 230 m, 320 m and 13 samples between 0-170 m. Line 83.3 is similar to 93.3. Most stations are sampled at the surface, deepest bottle, and 3 LTER depths. Station .90 and .70 are sampled at the deepest depth, 230 m, 320 m and 13 depths between 0-170 m. A full profile is collected at 18.104.22.168 (Santa Barbara Basin). Line 80.0 is similar to Line 90.0. Stations .51, .60, .90 are sampled at the surface, deepest bottle, 3 LTER depths, 140/145 m and 320 m. Full profiles are taken at station .55 and .100. Remaining stations are cardinal stations which include surface, deepest bottle, 6 LTER depths, 140/145 m and 320 m. Line 76.7 is sampled primarily at the surface, deepest bottle, and 3 LTER depths. Stations .55, .70 and .90 are sampled at the deepest bottle, 230 m, 320 m and 13 depths between 0-170 m.
3. Analysis Procedure
Analysis is performed using a Shimadzu TOC-VCSN with TNM-1 Total Nitrogen detector.
|Carrier gas||Zero Air from Puregas TOC air generator|
|Carrier flow rate||150 ml/min|
|Ozone generator gas||Air generator as above|
|Ozone flow rate||500 mL/min|
|Sample sparge time||2 minutes|
|Min. # of injections||3|
|Max. # of injections||5|
|Injection volume||100 µL|
The TOC system is calibrated using a 5-8 point calibration curve ranging in concentration between 10 µM and 120 µM C of potassium hydrogen phthalate in MilliQ water. The TDN system is calibrated using a calibration curve between 1µM and 35µM N of potassium nitrate in Milli-Q water. System performance is verified daily using Consensus Reference Water (http://www.rsmas.miami.edu/groups/biogeochem/CRM.html). This reference water is deep Florida Strait Water (DSR) that has been acidified and sealed in 10 ml ampoules, the concentrations of which (DOC and TDN) has been determined by the consensus of several expert and independent laboratories (41-44 µM; Drs. Craig Carlson, Wenhao Chen/Dennis Hansell, Hiroshi Ogawa, Amy Willman/Erik Smith and Ying Cui/Ying Wu/Jing Zhang.). Low Carbon Water (LCW) that has gone through the same acidification, sealing process, and consensus verification program as the DSR and has an agreed upon carbon concentration of 1 to 2 µmoles C/L is also analyzed and used to determine the instrument blank.
After verifying proper operation of the TOC/TN instrument, samples are placed on an auto sampler for analysis. If a clean column and catalyst has just been installed then the instrument is conditioned with 40 injections of acidified, filtered SIO Pier surface seawater. The run starts with 10 Milli Q water blanks, the set of standards and a reference seawater analysis. Then six samples are analyzed, followed by another QW blank and reference seawater. This sequence is repeated with one more set of standards inserted in the middle until all samples for that run are analyzed. The run ends with a standard run, QW blank, and reference water. QW blanks and reference water samples are used to evaluate system performance during the analytical run.
On a daily basis, Consensus Reference Water (CRM) is analyzed to verify system performance. If the value of the CRM does not fall within the expected range, samples are not analyzed until the expected performance has been established. The QW blanks and reference seawater samples analyzed with the samples are used for quality assurance and quality control (QA/QC). By evaluating the performance of these reference waters, instrument drift and performance can be evaluated. If a problem is detected with either drift or performance, the samples are reanalyzed.
Since the instrument performs using units of parts per million (ppm), the concentration of the sample in µM (micromolar or micromoles per liter), and correction for the instrument blank, is calculated as:
[(Sample (ppm) - LCW (ppm)) * 83.33333] + LCW value (µM)
where Sample and LCW are the concentrations determined by the TOC-V, 83.33333 (1000µg L-1/12 µg µmole-1) is a conversion factor converting ppm (µg/g) to µM and LCW is the carbon concentration of the Low Carbon Water CRM. Subtracting the LCW (ppm) from the sample removes both instrument blank and carbon content of the LCW. The carbon content of the LCW is added again (final term in equation) to calculate the correct sample concentration.
For total dissolved nitrogen, the instrument is calibrated using a similar method to that used for calibrating total carbon. The standard is potassium nitrate in Milli-Q water. Again the instrument is calibrated in ppm and the following calculation is used to convert from ppm to µM:
Sample (ppm) * 71.43
where sample is the concentration determined by the TOC-V and 71.43 (1000µg L-1/14 µg µmole-1) is a conversion factor from ppm to µM. An instrument blank has not been detected for the nitrogen system. Dissolved Organic Nitrogen (DON) is calculated by subtracting inorganic nitrogen (NO2, NO3, etc) from the total dissolved nitrogen determined by the TOC-V.
10. Equipment and Supplies
Shimadzu TOC-VCSN (P/N 63891062-02) with ASI-V auto sampler and TNM-1 Total Nitrogen detector Puregas Miniature TOC-Air generator with Parker hydrocarbon, water and CO2 filters to generate clear air for the TOC-V. Combustion Columns are purchased from Quartz Scientific, custom part number 317RT018B340. Copper wool is purchased from Miller Engineering, Quartz Wool is purchased from GM Associates incorporated. All other supplies (e.g., platinum screens, catalyst etc) are purchased from Shimadzu Scientific. Platinum gauze is purchased from Sigma. Sampling vials are purchased from Fisher Scientific (Clear Econo vials 144 40mL/pk; 05-719-106).
11.1. Compressed Gas
Ultra High Purity (UHP 99.995%) oxygen is sometimes used as the carrier gas for the Shimadzu TOC-V. High quality carrier gas is required to obtain low background levels in the detector. Oxygen is used to ensure complete combustion of all organic material. Typically the instrument runs on CO2-free air generated using a Puregas TOC-air generator.
11.2. Combustion Column Catalyst
The carrier gas passes through a column packed with 2 mm platinum-coated alumina beads (Shimadzu P/N 017-42801-01), held at 680°C.
11.3. Platinum Screens
Pure platinum screens are purchased from Shimadzu (P/N 630-00105). One is placed at the bottom of the quartz column before loading the column and two are placed on top, once the catalyst has been packed. We are testing the use of 3-5 platinum balls/cubes made from gauze (as recommended by the Carlson Lab - 52 mesh woven from 0.1 mm diameter wire) instead of the two screens placed on top of the column..
11.4. Quartz Wool
Quartz wool is used at the bottom of the combustion column, above the platinum screen, to prevent catalyst beads from exiting the column. This wool is pre-combusted at 450°C. Quartz wool is purchased from GM Associates.
11.5. Copper Wool Halogen Trap
Copper wool is placed in a polypropylene tube and connected after the flow exits the nafion drying tube. This removes corrosive halogen gasses from the combusted sample stream. (Wool is purchased from Miller Engineering).
11.6. Magnesium Perchlorate Water Trap
A magnesium perchlorate water trap (packed in a polypropylene tube) is connected after the halogen scrubber. This is added to remove any residual water from the sample stream. This must be changed every day.
11.7. Acidification of Sample
Concentrated hydrochloric acid (HCl, Trace Metal Grade) is used to acidify samples prior to analysis. Approximately 0.5% by volume of the concentrated acid is added to each sample prior to analysis to lower the pH of the sample to pH 2. At this pH and with sparging, all inorganic carbon species are converted to CO2 and removed from the sample. Automated acidification by the TOC-V is not used as with time the blank using this acid solution increases. Samples are manually acidified with two drops (200 µL) from a teflon drop bottle.
- Hansell, D. A. (2007) Determination of dissolved organic carbon and total dissolved nitrogen in seawater. In: Dickson, A.G., Sabine, C.L. and Christian, J.R. (Eds.) Guide to best practices for ocean CO2 measurements. PICES Science Report No. 34
- Sharp, J. H., Benner, R., Bennett, L., Carlson, C. A., Dow, R., & Fitzwater, S. E. (1993). Re-Evaluation of high temperature combustion and chemical oxidation measurements of dissolved organic carbon in seawater. Limnology and Oceanography, 38(8), 1774-1782.
- Sharp, J. H., Rinker, K. R., Savidge, K. B., Abell, J., Yves Benaim, J., Bronk, D., et al. (2002). A preliminary methods comparison for measurement of dissolved organic nitrogen in seawater. Marine Chemistry, 78(4), 171-184.