273 SIMULATED STUDY ON THE CHEMICAL AND PHYSICAL DECAY OF THE ACID RAIN ON CARBONATE STONE. ZENDRl,E., BISCONTIN,G., BAKOLAS, A. and FINOTTO,G. Department of Environmental Sciences, University of Venice, 2137 Dorsoduro, I 30123 Venice, Italy.
SUMMARY In this paper are discussed the experimental data, concerning the study of the kinetics of the chemical and physical decay on carbonate stones of different crystalline structure and porosity. Samples of lstria stone and Vicenza stone were treated with cycles of sulphuric solutions with different pH, using a special corrosion apparatus to simulate the conditions of exposure to acid rain. This experiments assess the physical decay than the chemical one separately. The results indicate mechanisms of different interactions, according to the porosity, the crystalline structure and the acidity of the sulphuric solutions employed. Moreover, the experimental results show an analogy with those obtained with samples exposed in real conditions.
1. INTRODUCTION
The alteration of stones goes through a number of processes of which only some stages are known; these are generally defined through the evaluation of the actual state of exposed stones. There are two ways to carry out the study of the decay: either the analysis of the state of manufatti esposti, or the simulation in lab of some environmental parameters. In this case it is necessary to compare the final results with those deriving from measurements taken on the spot, in order to confirm the validity of the adopted simulation system (Torra ca, 1991 ). A lot of methods have been suggested for the study through simulation systems: the use of gaseous S02 (Fuzzi, Vittori, 1975), or of gas mixtures, and the working in various humidity and temperature conditions. The outcomes have demonstrated a certain correlation between the phenomena that have been thus evaluated and the "natural" ones. On the other hand, the simulated accelerated processes have their limitations, that are usually connected with the choice of the parameters and the temperature at which the tests are carried out. There are less parameters to check, and the reading and interpretation of the analytical data is simplified, but by reducing the parameters, you also reduce the possible correlations with the real phenomenon. The works about carbonate stone that have been published on this subject report essential analogies between the real and the simulated - or accelerated - systems, also considering the acidity of rain as the main decaying factor. This finds its bases on the fact that calcium is the main product of alteration of carbonate materials, and that it forms as a consequence of the reaction of the stone to acid rain of sulphuric acid indipendently of the formation process of the acid. The present work shows the results deriving from the application of an alteration process accelleration system devised in such a way as to distinguish the two different alteration processes. We have conducted our studies on carbonate stones, having different porosity and different dimensions of crystals. The choice of the stones has been determined by the necessity of checking the mechanism and the kinetic of the decay, that, as appears from the examination of the materials, bring about different alteration forms according to the stone's microstructure.
274 2. RESULTS AND DISCUSSION
The stone materials selected for this research are the lstria stone, a compact microcrystalline limestone with impurities of argillaceus materials and of ferric oxide (Lazzarini, 1979) and a total porosity of 2.6%, and the Vicenza stone, a sedimentary limestone constituted by calcareous cement and an organogenic matrix (Cattaneo, 1975), and a total porosity of 28,5 %. The simulation system is made up with a glass cylinder of 15 cm diameter and 22 cm height, whose bottom has been shaped into a funnel for the solution collection. The samples are placed on a drilled disk, covered with filter paper for the recovery of the fragments coming off the stones. On the cylinder walls, one opposite the other, there are fixed two cones for the ejectors, working on 1 bar. Both the working pressure and the size of the drops are such as to simulate both the chemical effect and the physical action of the rain. The solutions employed are of sulphuric acid pH 4.8 and 2.6. In the first case, the choice of pH is based on the data concerning the rain conditions in the Venitian areas (Brovelli et al., 1987), whereas in the second case we meant to study the reaction of the stone samples to more extreme conditions, that very seldom take place (Likens. et al .. 1979). The samples, cut out into parallelepipeds 15x15x100 mm, are placed inside the glass cylinder with a slant of 60°; every four treatment cycles the samples are set into a stove for drying, in order to simulate the non-raining periods. The table here below shows the parameters for the two methods of stone treatment: Tab.1 pH solution
4,8
solution volume
2000 ml
slant of sample
so·
time oftest
10 minutes
drying time
every 4 cycles
pH solution
2,6
solution volume
2000ml
slant of sample
so·
time of test
10 minutes
drying time
ev~4cycles
The quantity of calcium and the corresponding calcium carbonate dissolved owing to the effect of the acid has been determined on the solutions collected on the bottom of the glass cylinder by atomic absorption (AA) measurements. The resulting quantity should correspond to the "chemical decay" of the material. The alteration resulting from the physical action has been calculated by determining the quantity of CaC03 came off in fragments that settled on to the plane on which the samples had been laid. The choice to operate with considerable volumes of acid solution has been prompted by observations about the possibility that the action of the acid might be reduced by the sample. By operating with small quantities of solution, the neutralising effect of the stone might lead to determine always an equal quantity of dissolved calcium, thus misrepresenting the results of the experimentation. This problem is avoided by employing the solution volumes reported in the table 1, the high quantity of solution not resenting from the effect of the sample. Tables 2 and 3 show the results of the tests with sulphuric acid pH 4.8 and 2.6, carried out respectively on Vicenza stone and lstria stone. They result from the average calculated on four samples. Figures 1 and 2 show the course of the chemical and physical decay in both stones, employing sulphuric acid solutions pH 4.8.
275 Table 2 : rate of the chemical and physical decay of the Vicenza stone.
n° cycles
:E%CaC03 •chem. decay•
:E%CaC03 •phys.decay•
n° cycles
:E%CaC03 •chem. decay•
:E%CaC03 •phys. decay·
12
0,10
0,0004
4
0,831
0,003
24
0,20
0,0011
8
1,628
0,006
36
0,30
0,0020
12
2,566
0,009
48 60
0,40
0,0029
16
3,421
0,022
0,50
0,0041
20
4,336
0,044
72
0,61
0,0054
24
5,298
0,071
84
0,72
0,0066
28
6,315
0,098
96
0,84
0,0082
32
7,418
0,137
108
0,95
0,0097
36
8,552
0,184
120
1,07
0,0113
40
9,680
0,230
132
1,18
0,0130
Table 3: rate of the chemical and physical decay of the lstria stone
n° cycles
:E'.4 CaC03 ·chem. decay•
:E'.4 CaC03 •phys.decay•
n° cycles
:E%CaC03 •chem. decay•
:E'.4 CaC03 •phys.deca •
12
0,037
0,000
24
O,Q75
0,001
4
0,565
0,083
36
0,114
0,002
8
1,105
0,092
48 60
0,152
0,003
12
1,772
0,094
0,192
0,004
16
2,318
0,096
72
0,230
0,010
20
2,963
0,097
84
0,270
0,011
24
3,439
0,111
96
0,310
0,012
28
4,141
0,112
108
0,351
0,014
32
4,710
0,113
120
0,393
0,015
36
5,187
0,113
132
0,436
0,016
40
5,732
0,115
Fig. 1: chemical decay of the Vicenza and lstria stones pH 4.8
0 I,% CaC03 chemical decay of the Vicenza stone • I.% CaC03 chemical decay of the Istria stone 1,2 ('f")
0
u u
~
~
~
1,0 0,8 0,6 0,4 0,2 0,0 0
30
60
90
cycles
12 0
150
276 Fig. 2: physical decay of the Vicenza and lstria stones pH 4.8
• L:o/o CaC03 physical decay of the !stria stone 0 L;o/o CaC03 physical decay of the Vicenza stone
0,020 O,Q18 O,Q16 ~
0
0,014
u u
0,012
~
0,008
~
~
0,010
0,006 0,004 0,002 0,000 0
30
60
90
120
150
cycles
The course of the chemical decay in both the stones is quite regular. In the lstria stone the quantity of the calcium carbonate dissolved owing to the acid's action is about 50% inferior to that of the Vicenza stone which, after 60 cycles of treatment, shows an intensification of the attack. The different reactions of the two samples is probably due to their different porosity and, therefore, to the different quantity of water they absorb. Particularly the porosity of the lstria stone, that presents veinings of argillaceous compounds, is 2.6%, whereas in the Vicenza stone is 28.5%. The microstructural characteristics, and particularly the specific surface area, very probably lead to a different kinetic of alteration in the two stones. In this sense, measurements of specific surface area and of porosity are being performed by B.E.T model on both the samples. As far as the physical alteration is concerned, the Vicenza stone is less subjected to loss of material; in the lstria stone small fragments held in between stylolithic planes come off, which is the weak point of the stone. The presence of water causes a moisture expansion of the silicatic fraction, resulting in loss of material. The curve representing the physical alteration of this sample has an irregular course, and the sudden variations in slope are determined by the coming off of the stone fragments (Fig. 2). In the Vicenza stone the physical alteration follows a more regular course, which is determined by the loss of calcite cement, with formation of pores whose insides are constituted by fossiles; these also are calcite crystals, but have a coarser grain and are more resistant to chemical attacks, as results from observations at the scanning electron microscope (SEM). The ratio between the percentage of material lost owing to the chemical effect, and the material lost owing to the physical effect shows that the first is superior to the second in both the stones. In the lstria stone it reaches 27.2, whereas in the Vicenza stone the ratio is 90.8. Treatment with solutions of pH 2.6 Figures 3 and 4 show the course of the chemical and physical decay in both stones, empoying sulphuric acid solutions pH 2.6. The increase in the loss of calcite caused by the chemical action is far superior to the loss resulting from treatment pH 4.8 in both the stones. but the difference of reaction of the two stones is only slight. It is important to remark that in this case 40 cycles of treatment have been carried out, that is to say one third of those effected with pH 4.8 solutions.
277 Also in this case the Vicenza stone shows a variation of slope in the course of the chemical decay, whereas in the lstria stone the course is still regular. In these operative conditions the physical alteration takes up a considerable importance in the whole decaying process. Compared to the treatments with pH 4.8 solutions, we note a more considerable loss of material in the Vicenza stone than in the lstria stone.
Fig. 3: chemical decay of the Vicenza and lstria stones pH 2.6
• I.%CaC03 chemical decay of the !stria stone 0 I,% CaC03 chemical decay of the Vicenza stone
10 9
8 7
6 5 4
3 2 0
-l-"T"""ir--r--r.....-~..--r---.--.-...,....-,-,-,--..-,--r--r"T1
0
5
10
15 20
2 5 30 3 5 40
4 5 50
cycles
Fig. 4: physical decay of the Vicenza and lstria stones pH 2.6
• I.%CaC03 physical decay of the !stria stone 0 I.% CaC03 physical decay of the Vicenza stone
0,30 0,25 ~
0
u u
0,20
~
0,15
~
v..:i 0,10 0,05 0' 00
-l--,,.e;::::::p:~_.,.....,_.,r"""'T"'"'T"'""...,.--r-T"""'T--Y-........,,.......,
0
5
10
15
20
25 30
cycles
35
40
45
50
278 Table 4 shows the values of the ratio between chemical and physical decay for both the stones in both the operative conditions. The data sum up what we have been arguing so far, and highlight th e different reactions of the two samples. Generally speaking, the Vicenza stone resents more of the acid attack, both in physical and chemical terms. Working with pH 2.6 the ratio between the extent of the chemical and physical decay is similar in the two stones, whereas with pH 4.8 the values differ a tot, thus showing a more remarkable difference of reactions in the two stones.
Table 4 : values of the ratio between chemical and physical decay for both the stones in both the operative conditions ~H 26
_E.H 4,8
'
E% CaC03 "chem.
E% CaC03 "chem. decay" Vicenza
stone/
2,6
decay'' Vicenza stone/
E%CaC03"chem. decay"
Z:%CaC03"chem.
lstria stone
decay''
1,7
lstria stone
E% CaC03 "phys. decay'' Vicenza stone/
E% CaC03 0,8
"phys.
decay'' Vicenza stone/
E%CaC03"phys. decay"
I:%CaC03"phys.
lstria stone
decay''
1,5
lstria stone
A study carried out on the same stones exposed in real conditions for 18 months had evatueted the thickness of material lost as a consequence of the alteration processes (Maravelaki, 1991). The results of this study revealed a total loss, comprehensive of both physical and chemical action, of 41.0 µm in thikness in the Vicenza stone, and 17.5 µm in the lstria stone. The ratio between these values shows that the Vicenza stone undergoes a loss about 2.3 times superior to the loss noted in the lstria stone. It is therefore quite safe to assert, though cautiously, that the adopted simulation system reproduces rather reliably the real process of the stones' alteration phenomena. Working with pH 2.6 we note that the differences between the two samples tend to disappear, probably because the chemical action of carbonate dissolution prevails over the morphologic differences of the two stones.
3. CONCLUSIONS The decaying process acceleration system here described allows to distinguish the extent of the two alteration forms, whose action is anyway combined, and also to evaluate the relevance of the difference in structure, grain and chemical nature of the stones in their decay. The Vicenza stone resents more of the alteration than the lstria stone, given its higher porosity and its greater capacity for acid solutions' absorption. In the lstria stone the physical decay takes place in a peculiar way, with the coming off of fragments held between the argillaceous veinings, that imply a typical irregular course. The pH of the employed solution affects differently the physical decay processes of the two stones. A higher acidity causes a greater loss of material in the Vicenza stone, given the action of the acid on carbonate cement, that leads to the coming off of englobed fossil fragments.
279 The validity of the simulated system described is confirmed by comparison with the same materials exposed in real conditions, if we consider the values resulting from the study performed with pH 4.8.
REFERENCES M.A.Brovelli, M.Bassanino, Studio sull'acidita delle precipitazioni in rapporto a fattori meteoclimatici, Acqua-Aria, 6, 1987, 717-724. A.Cattaneo, G.De Vecchi, L.Menegazzo Vitturi, Le pietre tenere dei Colli Berici, Atti e Memorie dell'Accademia Patavina di Scienze, Lettere ed Arti, vol. LXXXVlll, 1975-1976, 69-100. F.De Santis, D.Pasella, I.Allegrini, A diffusion tecnique for the evaluation of surface reactivity: application to the study of S02 deposition on marble, in Proc. The Conservation of Monuments in the Mediterranean Basin, Venezia 1994, 213-226. S.Fuzzi, O.Vittori, Climatic chamber for laboratory experiments on the system S02-wet marbleairbome particles, in Proc. The Conservation of Stone, Bologna 1975, 651-662. L.Lazzarini, Morfologia del degrado dei materiali lapidei a Venezia, in Proc. Associazione Civica Venezia Serenissima, 1979, 47-57. G.E.Liekens RF.Wright, J.N.Galloway, T.J.Buttler, Le pioggia acide, Le Scienze, 136, 1979, 1423. P. Maravelaki, Effetto chimico dell'ambiente su superfici lapidee esposte ed indagini per un loro recupero, Tesi Dottorato di Ricerca, IV ciclo, 1989-1991. T. Skoulikidis, The application and the limits of the intensification of the conditions by artificial weathering of stones, in Pro. Int.Coll. Methods of evaluating products for the conservation of porous building materials in monuments, Roma 1995, 461-473. G. Torraca, Appendice bibliografica all'articolo di W .Conti, Materiali e Strutture, 1,1991, 22-26. J.Weber, Natural and artificial weathering of austrian building stones due to air pollution, in Proc. Int.Congress on Deterioration and conservation of Stone, Lausanne 1985, 527-535.
