J. .-1"''''.0

s

,.

.

2.

.

.

.

t. 1

.. .

0

50

.

100

t50

200

Figure 1. Numbers of adhcring hacteria N (IO"/mm~)as a function of theoretically calculatcd free energies of adhesion l!J.F.~h (mJ m-~) for Streptococcus thermophi/us adhering from water to various stainless steel surfaces, the spreading pressure is accounted for in YBvalue.

regard separately the results obtained on apolar surfaces (Y~=0, corresponding to the first part) and those obtained on polar surfaces (Y~> 0, corresponding to the second part), then in both parts it is obvious that the number of adhering bacteria increases with Ys. The theoretical free energies of adhesion AFadhcalculated for S. thermophilus adhering from water (Fig. 2) to various solids are given as a function of Ys;AFadh was computed firstly with YBaccounting for the spreading pressure, 1re(squares) and secondly with YBneglecting 1re (circles). ln ail cases, AFadhis negative, predicting that the adhesion of S. thermophilus is thermodynamically favourable. However, when the spreading pressure is neglected in YB'no variation in the adhesion of S. thermophilus to apolar solids is expected to occur with an increase of Ys,and the adhesion to polar surfaces is expected to decrease with an increase of Ys.On the contrary, when the spreading pressure is accounted for in YB'the adhesion of S. thermophilus to both apolar and polar surfaces is expected to increase with Ys. 4. DISCUSSION

ln this study, the wettability of 2B and 2RB stainless steel surfaces, expressed in terms of the surface free energy, has been evaluated for different cleaning treatments. Subsequently, we studied the influence of the wettability on bacterial adhesion. 4.1. Influence

of the cleaning

treatment

on the stainless steel surface free energy

Generally speaking, metallic materials which are covered with a thin oxide film exhibit a high surface energy [24, 27]. As for stainless steel, it can be seen in

Bac/er;al ad"s;on

.

0

...

1

........

. -20'

1 1

. ...

-40. )' p 8

':J'

· 0:.

-80r

227

10m('/{Illic mrfc/("s

. .

)' p ) 0

. .. -801- . . , -130' 0

1

. 80

1

.

..

100

'.

.

.

tH

200

.. 280

300

(aJ/.1)

Figure 2. Theoretically calculated free energies of adhesion âFadh(ml m -!) for SlreplococCUs Ihermophilus adhering from water to various stainless steel surfaces as a function of the solid surface free energy Ys (ml m -~I.The spreading pressure is neglected (8) or accounted for (8) in the YB determination. Note the tirst part corresponding to y~" 0 and the second one corresponding to yt> O.Note the broken âFodh-axis.

Tables 1 and 2 that after various cIeaning processes both 2B and 2RB surfaces exhibit high and low surface free energies. Surfaces cIeaned with an aIkaline detergent or a HNO., (4 M) solution are hydrophilic, while they are relatively hydrophobie with the other treatments. Alkaline detergent cIeaning is the only one which gives very high values of Ys. ln this case, the 2RB surface is more hydrophilic than the 2B one. This result can be explained by differences in film composition due to the finishing processes [17]. It is weil known that 2RB passive films are richer in the more oxidizable elements (chromium. silicon, etc.) contained in the steel than the 2B ones and that 2RB provides more hydrated (and/or containing more hydroxides) and thicker films than 2B, at least when the annealing dew point is close to the critical dew point. which is the casé in industrial processes. Degreasing metallic surfaces with an organic solvent (ethanol. acetone. and ethanol/acetone) provides hydrophobie surfaces. For the solvents under consideration, Ys ranges from 34.4 to 47.2 ml m-2 for the 2B finish and from 45.1 to 62.1 ml m2 for the 2RB finish. Comparable values were obtained by Berger [33] on iron alloys and electrogalvanized steel cIeaned with acetone (50.4 and 34.0 mJ m-2, respectively). These low surface free energy values correspond to the values obtained for polycarbonate surfaces and can be explained by adsorption of solvent molecules and/or a grease spreading; this surface contamination has been produced during the rolling process of steel sheets. Acetone cleaning is more efficient on the 2RB surface than on the 2B one. However. cleaning metallic surfaces with organie solvents is ineffective and leads to organic fouling. Cleaning the surface with a HN03 aqueous solution can have various effects depending on the nitrie acid concentration and can lead to hydrophobie or hydrophilie surfaces (Tables 1 and 2). It is weil known that nitrie acid aets both

228

L. Boulallge-Pt"alllal'"

CIal.

X 10'

1}

41' 1

z

:t..

.

2.. t.. 0 -140

-120

.

. . . . . . -..

-toO AF...

-80

-60

(aI/.I)

Figure 3. Number of SlreplococCUsIhermophilus N (x lOJ/mm2) adhering from waler to various stainless steel surfaces as a function of the solid surface free energy Ys(mJ m-2). Note the first part corresponding to y~=0 and the second one corresponding to y~> O.The bars denote the standard error. Note the broken N-axis.

as an acid and as an oxidizing agent [17]. When the acid concentration is relatively low (say, typieally less than 1 M), the oxidizing power is not sufficient to stabilize the surface in the passive state [17]. The passive film dissolves quiekly in the acid medium and the surface is said to be active [17]. On the other hand, at a higher nitric acid concentration (say, typically more than 3 M), the strong oxidizing power of the solution stabilizes the passive film, despite the solution acidity. So, cleaning the stainless steel with a HNO) (0.2 M) solution in fact partially provokes the passive film dissolution. The result is the presence of a fresh hydrophobie passive film, irrespective of the initial surface condition (2B or 2RB). On the contrary, cleaning the surface with a HNO) (4 M) solution reinforces the existing passive layer, providing a higher wettability, irrespective of the initial surface condition. 4.2. Influence of the wellability of metallicsurfaces on the bacterialadhesion The adhesion of Streptococcus thermophilus is very substratum-dependent. Bacterial adhesion is higher on hydrophilic surfaces than on hydrophobie surfaces. However, there is no direct relationship between the number of adhering bacteria and the surface free energy or wettability of solids, as previously mentioned. It has been shown that bacterial adhesion increases with 'Ys on apolar stainless steel surfaces. These results are in accordance with previous data obtained on polymerie surfaces [Il, 14]. Concerning polar solid surfaces, it was observed that the number of adhering bacteria also increased with 'Ys.This non-linearity can be due to the favourable or unfavourable role of polar interactions in bacterial adhesion; thus, these findings indicate that bacterial adhesion depends on the balance between 'Y~and 'Y~and not on the total 'Ysor wettability of metallic surfaces.

RtlClI!riut

udh~si()11

((1 "1I.tallic'

SlIrfi/l"C'S

229

Thcrc is a better correlation (r = 0.924) between the number of adhcring bactcria and AF.dh when 1reis taken into account in YB'On the other hand, when 1reis neglectcd in YB' no agreement is obtained bctween the experimental data and the theoretical prediction (r = 0.392). To summarize, neglecting the spreading pressure in YBleads to failure in the theoretical predictions of bacterial adhesion. 5. CONCLUSION

Il has been shown that the wetting behaviour of the stainless steel AISI 304 2B and 2RB finish surfaces is strongly influenced by the cleaning process. For the same stainless steel surface, it was possible to obtain a low or a high surface energy depending on the cleaning treatment. Modifications in the solid surface energetic characteristics influence the bacterial adhesion in accordance with the thermodynamic predictions when the spreading pressure is accounted for in the YB determination. The adhesion of Srreprococcus rhermophilus, which is a sensitive substratum surface strain, is driven by the balance between Y~and Y~ and not by the total Ysor wettability of metallic surfaces. Polar interactions can reduce bacterial adhesion to stainless steel surfaces. Acknowledgemenrs We are greatly indebted to Mrs J. Rauh, Miss V. Galopin, and Mr M. Berrais for excellent technical assistance. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

A. Assanta Mafu, D. Roy. J. Gouletand P.Magny,J. Food Prol. 53, 742 (1990). F. M.Driessen, J. De Vries and F. Kingma.J. FoodProl.47.848 (1984). S. J. Lewis and A. Gilmour. J. Appt. &c/~riot. 62. 327 (1987). T. S. Schwachand E. A. Zottola. J. FoodPrOI.47. 756 (1984). P.T.Zoltai. E. A. Zottola and L. L. McKay.J. Food Prol.44. 204 (1981). J. Schultz and H. Simon. Ve"es Réfraci.34. 23 (1980). A. Horsthemkeand J. J. Schroder.J. Chem. Eng. Process. 19.277 (1985). G. Wildbrett and V.Sauerer. in: Foutingand CIeaningin Food Processing. H. G. Kessler and D. B. Lund (Eds). pp. 163-177. Druckerei Walch.Augsburg,Germany (1989). H. J. Busscher, A. H. Weerkamp. H. C. van der Mei. A. W. J. van Pelt. H. P. de Jong and J. Arends. Appl. Environ. Microbiol. 48.980 (1984). A. W. J. van Pelt. A. H. Weerkamp. M. H. W.J. C. Uyen. H. J. Busscher. H. P. de Jong and J. Arends. Appl. Environ. Microbiol. 49.1270 (1985). H. J. Busscher, M. N. Bellon-Fontaine. N. Mozes. H. C. van der Mei, J. Sjollema. O. Cerf and P. G. Rouxhet. Biofouling 2.55 (1990). J. M. Schakenraad. H. J. Busscher. Ch. R. H. Wildevuur and J. Arends. Cells Biophys. 13.75 (1988). 1. H. Pratt-Terpstra. A. H. Weerkamp and H. J. Busscher, J. Colloid In/erfaee Sei. 129. 568 (1989). M. N. Bellon-Fontaine, N. Mozes, H. C. van der Mei, J. Sjollema, O. Cerf, P. G. Rouxhet and H. J. Busscher, Cell Biophys. 17,93 (1990). M. Asther, M. N. Bellon-Fontaine. C. Capdevilla and G. Corrieu, Bioleehnol. Bioeng. 35,477 (1990). J.C.de Mann, M. Rogosa and M. E.Sharpe,J. Appl. Mierobiol. 23,130 (1960). B. Baroux, G. Béranger and C. Lemaitre, in: Les Aeiers Inoxydables, P. Lacombe, B. Baroux and G. Béranger (Eds). pp. 163-182. Les Editions de Physique, les Ulis, France (1990). D.Gorse, J. C. Joud and B. Baroux, CorrosionSei. (in press).

230

1.. 801lIUI//.'t'-I'~tC'rlllul/l/ el al.

19. H. J. Busscher. M. N. Bellon-Fontaine. N. Mozc:s. H. C'. van der Mei. J. Sjollcma. A. J. Leonard. P. G. Ruuxhc:t and O. C'c:rf.J. Microbiol. "'f~thods 12. 1111(1990). 20. M. N. Bellon-Fontaine and O. Cerf. J. At!Ir~sion Sei. Teclmol. 4.475 ( 1990). 21. C'. J. van Osso M. K. C'haudhury and R. J. Good. Ad,: Col/oit! IIIterfuce Sei. 28. 35 ( 19H7). 22. Y. Tamai. K. Makuuchi and M. Suzuki. J. Ph)'s. Chem. 71.4176 (1967). 23. J. Schultz. K. Tsulsumi and J. B. Donnet. J. Col/oid Interface Sei. 59. 272 (1977). 24. A. Ç'arré. Thesis. Haute Alsace University. Mulhouse. France ( 19HO). 25. M. N. Bellon-Fontaine. Thesis. Pierre and Marie Curie University, Paris (19H6). 26. D. K. Owens and R. C. Wendt. J. Appl. Pol)'m. Sei. 13. 1741 (1969). 27. J. Kuczynski and E. Papirer, EuT. Polym. J. 27.653 (1991). 28. J. A. Filbey and J. P. Wight man, in: Adhesion 12, K. W. Allen (Ed.). pp. 17-32. Elsevier, London (1987). 29. A. H. Weerkamp. H. C. van der Mei and H.J. Busscher. J. Dent. Res. 64.1204 (1985). 30. H. C. van der Mei and H. J. Busscher. J. Colloid Interface Sei. 136.297 (1990). 31. H. C. van der Mei, M. Rosenberg and H. J. Busscher. in: Microbial Cell Surface Analysis. N. Mozes. P. S. Handley. H. J. Busscher and P. G. Rouxhet (Eds). pp. 263-287. VCH Publishers. New York (1991). 32. H. J. Busscher. A. W. J. van Pett, H. P. de Jong and J. Arends. J. Col/oid Interface Sei. 95. 23 (1983). 33. E. J. Berger, J. Adhesion Sei. Technol. 4.373 (1990).