For determining the appropriate temperature levels for the acid rain tests, stagnation temperature for an unglazed absorber was calculated iteratively on a hourly basis (Konttinen et al, 2004). Surface samples were immersion subjected to O2-aerated or zero — aerated (with N2) simulated acid rain (Table 1) with pH 3.5 or pH 4.5 and simulated neutral rain with pH 5.5 in temperatures of 60, 80 or 99 °C. Saturating gas was fed at room temperature into the solution. The solution was heated within ±1 K of the required temperature in a flask, which was in a paraffine oil bath (Fig. 1). The oil was typically 6- 10°C (at 60/80°C) or 40-45°C (at 99°C) warmer than the solution. Details of the setup are described in (Konttinen et al, 2004). Samples tagged as “non-aerated” in Figures 3-5 were tested with EIS at similar temperatures and pH. Changes in optical properties of non­aerated EIS-tested samples A-B in Figs. 3-5 are included in this paper for comparison to aerated and zero-aerated test results.

Polarisation measurements and EIS measurements were carried out in a conventional three electrode cell using platinum sheet as a counter electrode (CE) and saturated calomel electrode (SCE) as a reference electrode (RE) (Fig. 2). The methods and measurement system are described in ASTM standard G5 (ASTM, 1987) and in (Lorenz and Mansfeld, 1981), respectively. The RE was connected to the cell with Luggin capillary.

Corresponding author. Tel.: +358-9-4513212; fax: +358-9-4513195; E-mail:

petri. konttinen@hut. fi

The salt bridge filled with 0.1 mol/l Na2SO4. was used between the RE and the Luggin capillary to avoid possible chloride contamination of the test solution. In some polarisation tests 500 mg/l Na2SO4 sulphate was added to increase conductivity (Table 1, b-solutions). The polarisation measurements were conducted in an Avesta cell (Fig. 2a) where a sample was pressed against a round hole on the cell bottom and the system was sealed with a rubber o-ring gasket. The sample area was 0.8 or 1.3 cm2. Long-term immersion EIS tests were conducted in a cell (Fig. 2b) where the samples (working electrode WE) were attached with stainless steel screws to a sample holders made of copper rods. The copper rods and the back sides of samples were insulated with PTFE tape and nolan lacquer. The sample area was 5 cm2.

Before polarisation measurements the corrosion potential was followed for 1h until a stable corrosion potential was achieved. Potential was first changed — 150 mV to cathodic direction and cyclically back to the corrosion potential with a scan rate of 10 mV/min. After cathodic polarisation the potential was changed +150 mV to anodic direction or to potential -200 mV vs. SCE. The EIS measurements were done at the corrosion potential with the amplitude of 10 mV. The EIS system consists of a NF 2000 potentiostat and NF 5050 frequency response analyser and the polarisation measurements system consists of a ACM Instruments Autotafel potentiostat.

Test duration needed to degrade the a and є of the absorber samples according to performance criteria (PC) was determined in each test:

PC = — Дa + 0.25Д^< 0.05

Although not defined for unglazed collectors, Eq. 1 was used as it is typically used to estimate failure limit for solar absorbers inside glazed collectors (Brunold et al., 2000). PC value of 0.05 is generally equivalent to 5 percentage unit decrease of the solar heat gain by the flat-plate solar collector water heating system.

Hemispherical reflectance of the samples was measured before and after each test at room temperature. Solar absorptance, a, was determined between 0.39 — 1.1 pm with a LI-COR LI-1800 type spectroradiometer and a BaSO4 coated integrating sphere. Thermal emittance, e, was determined between 2.5 — 20 pm with a MIDAC Prospect FTIR — spectrometer with a semi-hemispherical integrating device. Spectral reflectance, p^, was analysed for estimating the hydration levels of alumina.

Fig. 1. Photograph of the acid rain total-immersion setup including 250 ml three-necked flask, 100 ml simulated acid rain, 23 x 50 mm sample and gas distribution tube (excluding heating system consisting of paraffine oil bath, heater and temperature controller).

Fig. 2. Schematic pictures of the measurement cells in polarisation tests (a) and long-term immersion tests (b).

Results