Total Alkalinity¶
- PIs
Andrew G. Dickson (SIO)
- Technicians
Daniela Nestory (SIO)
Sidney Wayne (HPU)
Total Alkalinity¶
The total alkalinity of sea water is defined as the number of moles of hydrogen ion equivalent to the excess of proton acceptors (bases formed from weak acids with a dissociation constant K < 10E-4.5 at 25°C and zero ionic strength) over proton donors (acids with K > 10E-4.5) in 1 kilogram of sample.
Total Alkalinity Measurement System¶
Sample Delivery System:
Samples are dispensed using a Sample Delivery System (SDS) which has been calibrated for volume in the lab prior to the cruise. Its volume is confirmed immediately before use at sea to ensure a consistent volume will be delivered for each sample. The SDS consists of a volumetric pipette, various relay valves, an air pump, and is controlled by a program in LabVIEW 2012.
Before attaching a sample bottle to the SDS, the volumetric pipette is cleared of any residual solution. The pipette is then rinsed and filled with the sample. The sample overflows and time is allowed for the sample temperature to equilibrate.
The sample bottle temperature is measured using a DirecTemp thermistor probe inserted into the sample bottle and the volumetric pipette temperature is measured using a DirecTemp surface probe placed directly on the pipette. These temperature measurements, along with the bottle salinity, are used to convert the sample volume to mass for analysis.
Samples are delivered into a 250-mL water-jacketed open cell for titration analysis. While one sample is undergoing titration, a second sample is prepared with the SDS and equilibrated to 20°C for analysis.
Open-Cell Titration:
The total alkalinity is measured through an open-cell titration with a dilute hydrochloric acid titrant of known concentration. A Metrohm 876 Dosimat Plus is used for all standardized hydrochloric acid additions.
An initial aliquot of approximately 2.3-2.4 mL of standardized hydrochloric acid (~0.1M HCl in ~0.6M NaCl solution) is first delivered and the sample is stirred for 5 minutes while air is bubbled into at a rate of 200 scc/m to remove any liberated carbon dioxide gas.
After equilibration, ~19 aliquots of 0.035 ml are added. Between the pH range of 3.5 to 3.0, the progress of the titration is monitored using a pH glass electrode/reference electrode cell, and the total alkalinity is computed from the titrant volume and e.m.f. measurements using a non-linear least-squares approach ([Dickson2007]).
A Thermo Scientific Isotemp water bath is connected to the water-jacketed open cell to maintain a cell temperature of approximately 20oC. An Agilent 34970A Data Acquisition/Switch Unit with a 34901A multiplexer is used to read the voltage measurements from the electrode and monitor the temperatures from the sample, acid, and room.
The calculations for this procedure are performed automatically using LabVIEW 2012.
Sample Collection¶
Alkalinity samples are drawn using silicone tubing connected to the Niskin bottles and collected into 250 mL Pyrex bottles. The sample bottles and Teflon-sleeved glass stoppers were rinsed at least twice before the final filling. A headspace of approximately 3 mL was removed and 0.1 mL of 50% saturated mercuric chloride solution was added to each sample for preservation. The samples were equilibrated prior to analysis at approximately 20°C using a Thermo Scientific Isotemp water bath.
Samples for total alkalinity were taken at all stations during Leg 2 of P02 (118-204). Except for a few instances, alkalinity samples were collected from all niskins where DIC and pH were collected, to over-characterize the CO2 system. The typical sample scheme of partial collection on all stations (26 bottles), with the exception of a full collection (36 bottles) on select stations, was followed.
In order to evaluate the reproducibility of the alkalinity system, 2 duplicate samples (two separate alkalinity bottles) were collected on each cast, with the exception of casts with fewer than 18 bottles, in which 1 duplicate sample was collected.
Problems and Troubleshooting¶
The alkalinity system was set up in the hydro lab. Around station 124, voltage readings from the system’s electrode were suddenly noisy and values were affected. All parts of the alkalinity system were changed in hopes of remedying the noise, but to no avail. The system was moved to the analytical lab around station 125 and the voltage readings / CRM values were once again normal. Analyses for stations 124 and 125 were the only stations affected by this issue.
Quality Control¶
Certified Reference Material (CRMs) and duplicate samples (two bottles collected from one niskin) were used to quality check the functioning of the total alkalinity system throughout the cruise.
Dickson laboratory Certified Reference Material (CRM) Batches 200 and 201 were used to determine the accuracy of the total alkalinity analyses. The total alkalinity certified value for these batches are:
Batch 200 2186.43 ± 0.42 µmol/kg
Batch 201 2207.56 ± 0.47 µmol/kg
The cited uncertainties represent the standard deviation.
A CRM sample was analyzed at a minimum frequency of once per every 20 runs, but more often once per every 15 runs. Because total alkalinity is not affected by gas-exchange, brand new CRM bottles were reserved for pH and DIC analysis. These pre-opened bottles were subsequently used for alkalinity analysis.
198 reference material samples were analyzed on Leg 2 of P02.
The average measured total alkalinity value for each batch is:
Batch 200 2186.59 ± 1.39 µmol/kg (n = 133, 70)
Batch 201 2208.24 ± 1.25 µmol/kg (n = 65, 40)
Figures in parentheses are the number of analyses made (total number of analyses; number of separate bottles analyzed).
Duplicate samples were also used to check the reproducibility of the system. The absolute value of the mean offset between duplicate samples and the standard deviation are given below.
Mean duplicate sample offset: 1.08 ± 1.00 µmol/kg (n = 164)
2396 total alkalinity values were submitted for Leg 2 of P02.
Further dilution corrections need to be applied to this data back onshore, therefore, this data is to be considered preliminary.
- Dickson2007
Dickson, A.G., Sabine, C.L. and Christian, J.R. (Eds.), (2007): Guide to Best Practices for Ocean CO2 Measurements. PICES Special Publication 3, 191 pp.