Heatflux Corrosion Caustic

Caustic corrosion tests in the experimental boiler

Only a few caustic corrosion experiments under heat flux conditions in the experimental boiler were done. It is obvious that under dryout conditions and water conditioning with a non volatile agent like the caustics NaOH, LiOH, Na3PO4 severe corrosion can be expected. We did some experiments to get an idea how fast.

One of them was the 10the experiment. Conditions Dryout at a steam fraction of 70%, heatflux 300KW.h/m2, dosing NaOH up to a pH value of 10. The experiment would take 10 days, but after 3 days on friday evening I ordered as young researcher to finish the experiment. When we cut the testtube on monday we were surprised and glad finishing the test. Tht tube wall was corroded for half and at longer exposure times the testtube should be bursted certainly end destroyed the furnace certainly. Deliverance of a new oven would have delayed the research program at least half a year.

2511 test tube, alkaline corrosion under dryout conditions
Test tube, alkaline corrosion under dryout conditions. Right the heated zone, left the unheated zone.
060405 verdamper0013 caustic corrosion
SEM picture of the corrosion layer on the test tube. A porous magnetite on the steel surface and coarse crystalline magnetite on the outside.

Caustic corrosion tests in autoclaves

Because experiments under static conditions (in autoclaves) are much cheaper and you can invent some specific conditions easy we did many projects in autoclaves. To be mentioned:

  • small 50 ml autoclaves
  • 500 ml autoclaves for electrochemical measurements
  • 500 ml autoclave for steam blanketing experiments
  • 2000 ml autoclaves for refreshing controlled water conditions
  • 2000 ml autoclaves for refreshing controlled water conditions and constant strain tests

An example is the test sery on various steel on corrosion rate in caustics. Conclusions of these resarches are:

  • There is a critical hydroxide concentration above which the corrosion rate increases rapidly. The critical concentration of LiOH amounts to about 0.5 molar and that of NaOH. 2-3 molar. KOH shows a critical concentration of 1.5 molar.
  • The critical concentration for alkaline corrosion was only determined for a few heats of the 5 steels tested. It is not clear whether this critical concentration is also controlled by the chemical composition of the steel within the broad range of the various steel specifcations.
  • The critical alkaline concentration at 250°C does not differ from that at 310°C. The corrosion rate in highly concentrated (4 and 10 molar) LiOH is five times higher than that in NaOH.
  • Under the isothermal autoclave test conditions in 4 mol NaOH a so-called Potter and Mann oxide layer formed, leading to a parabolic corrosion rate. In 4 mol LiOH, however, a porous columnar oxide structure was formed which results in linear corrosion.
  • The corrosion rate in 10 mol NaOH for carbon steels varied considerably for the 83 different heats. Carbon steel showed the highest corrosion rates. Steels 15Mo3, 14Mn4 and 13CrMo44 had about the same corrosion rates but were only slightly better than carbon steel. Steel l0CrMo910 had the lowest corrosion rate.

    17 Gevarendriehoek Fig 16
    The corrosion rates of carbon steel is the highest, those of 15Mo3, 14Mn4 and 13CrMo44 are obvious less. The rate of 10CrMo(.10 is some lower then that of 13Cr.Mo.4.4
  • In the case of a fouled evaporator and the consequent danger of steam blanketing, NaOH and KOH are considered to be better acid-chloride corrosion inhibitors than LiOH. LiOH will give a high alkaline corrosion risk because of the low critical hydroxide concentration.

See for more information the papers: 2, 6, 7, 8, 11, 13, 20, 23, 32 and 61

Home » Heatflux » Heatflux Corrosion Caustic » Heatflux Corrosion Caustic
2500.jpg
2500.jpg
SEM picture of a fracture surface of a corrosion scale. On the steel surface a porous topotactical magnetite layer and on the outer surface a epitactical layer of coarse magnetite octaeders have been formed. Caustic environment onder bopiler conditions. (foto KEMA, ref 11, 61).
2501.jpg
2501.jpg
Surface of a tube exposed under dry-out conditions and with sodiumphosphate water conditioning. The dryout area in this test tube is sharply bordered..(foto KEMA, ref 11, 61).
2502.jpg
2502.jpg
The locations a, b, c and d in the tube in case nr 2501 where cross sections have been made for examination of the type of the oxide layers. (foto KEMA, ref 11, 61).
2503.jpg
2503.jpg
Oxide layers on the locations a, b, c and d of tube mentioned in case 2502. The oxide layer is thicker going into the direction of the dry-out. At location d (behind the dryout) the oxide is porous without real protection. (foto KEMA, ref 11, 61).
2504.jpg
2504.jpg
Corrosion scale in alkaline (NaOH) boiler water under heat flux conditions. In the scale a compact layer on the metal, porous part and coarse magnetite crystals. In the crust small boiling tunnels partly filled with the coarse magnetite crystals. (foto KEMA, ref 11, 61).
2505.jpg
2505.jpg
Typical corrosion scale under heat flux corrosion in alkaline water. Phenomena of columnar boiling were noticed in the porous corrosion crust. The adherent layer detached over the whole area where the pits (from columnar boiling in alkaline water) were present. (foto KEMA, ref 11, 61).
2506.jpg
2506.jpg
A typical corrosion scale in alkaline (NaOH) boiler water under heat flux conditions. In the scale a compact layer on the metal, porous part and coarse magnetite crystals. In the crust boiling tunnels are noticed due to the heat flux (foto KEMA, ref 11, 61).
2508.jpg
2508.jpg
Results of steam blanketing laboratory tests on C-steel in sodium phosphate solutions with different Na/PO4 ratio's. Below Na/PO4=2.6 corrosion occurs because of the acid forming. In between 2.6 and 2.8 the samples did not corrode and deposits are formed. Above 2.8 alkaline corrosion occured. (foto KEMA, ref 11, 61).
2510.jpg
2510.jpg
2511.jpg
2511.jpg
Cross section of test tub tested in caustic water under dry-out conditions. After 3 days exposure the experiment was finished because we suspected that the corrosion was very fast. (Ref 11, 61)
2512.jpg
2512.jpg
Even without the microscope the porous and coarse crystalline magnetit was visible very well. Ref 11, 61)
2513.jpg
2513.jpg
Undulating oxide on the outside of the layer. Magnetite crystals grow to each other, so resulting in compressive stresses and the undulating. ref 11, 61.
2514.jpg
2514.jpg
Detail of the caustic corrosion scale. Ref 11, 61.
2820.jpg
2820.jpg
On nucleate boiling locations in a evaporator tube pitting and intergranular corrosion took place. Carbonate has been found on the dark places in the cross section. From these pits carbonate Stress Corrosion Cracking was found too. (foto KEMA).