**4.2.3.1 Shrouded plasma spraying**

For both guns modules have been designed and machined to apply shroud gases around the plasma free jet (for details see **Figure 14** on the following page). The attachments consist of water-cooled bodies, in which the shrouding gas is injected helically to ensure a sufficient

Thermal Spraying of Oxide Ceramic and Ceramic Metallic Coatings 189

sprayed coatings. Otherwise when nitrogen is supplied, coatings with high levels of open cavities and microstructures comparable to metallic sponges are built. This is supposedly due to a turbulent entrainment of the nitrogen shielding gas when the feedstock reacts towards the nitride. To proof the existence of the aspired nitride phases, semi-quantitative measurements by means of energy dispersive x-ray analysis (EDX) were performed on coatings sprayed with the Triplex. The results revealed contents between approximately 13 and 17 at.-% nitrogen. Further on, the nitrogen contents of the coatings were investigated using a N/O/H-Analyser (LECO Instruments Corp, St. Joseph/USA). Unfortunately the contents of nitrogen in the titanium coatings could not be measured to the high melting point of the titanium nitride, but for the chromium coating (section f) the nitrogen content

The existence of the nitride hard phases was also verified by indentation hardness measurements. When spraying the titanium with argon as shrouding gas, mean hardness values of 360 HV 0.1 in case of the Triplex and 270 HV 0.1 for the DELTA-Gun were measured. Otherwise with the employment of nitrogen significantly higher maximum

Fig. 15. Micrographs of shrouded plasma sprayed titanium (field a – d) and chromium (field e + f) feedstock using argon (left hand side) and nitrogen (right hand side) as shroud gas

was determined to account for approximately 10 at.-%.

values of more than 1200 and 1000 HV 0.1 were detected.

shielding against the surrounding air after exiting the shroud. The feedstock injection is realized over middle sections between the exit of the gun nozzle and the shroud gas inlet to avoid interferences with the shroud gas flow. The body housings of the shrouds are integrated in the cooling circuit of the spraying equipment.

The spraying experiments were conducted applying plasma brut powers of approximately 25 to 50 kW (see **Table 7** for spraying parameters). When operating the Triplex high helium flows of 20 SLPM were used to guarantee a sufficient heat transfer to the feedstock, but for the DELTA-Gun no secondary gas was applied due to the formation of black depositions with a consistency of soot when hydrogen was applied. To guarantee an adequate shielding effect in the case of argon shroud gas on one hand and an effectual entrainment of nitrogen for reactive spraying of the feedstock on the other, high shroud gas flows of 90 SLPM were applied for spraying with both guns.

Fig. 14. 3D sectional view of the shroud modules of Triplex (top) and DELTA-Gun (bottom): 1 Gun, 2 shroud gas inlet, 3 powder injector, 4 water-cooled shroud, 5 helical injected shroud gas


Table 7. Parameters for Shrouded Plasma Spraying

The obtained coatings microstructures are illustrated in Figure 15. The micrographs on the left hand side show coatings sprayed in an inert atmosphere of argon, the ones on the right hand side the results of reactive spraying using nitrogen. The coatings of a and b were sprayed using the Triplex gun, whereas for spraying of the coatings c to f the DELTA-Gun was used. When spraying in inert atmosphere, the coatings show a uniformly and homogenous microstructure with a level of porosity comparable to conventional plasma

shielding against the surrounding air after exiting the shroud. The feedstock injection is realized over middle sections between the exit of the gun nozzle and the shroud gas inlet to avoid interferences with the shroud gas flow. The body housings of the shrouds are

The spraying experiments were conducted applying plasma brut powers of approximately 25 to 50 kW (see **Table 7** for spraying parameters). When operating the Triplex high helium flows of 20 SLPM were used to guarantee a sufficient heat transfer to the feedstock, but for the DELTA-Gun no secondary gas was applied due to the formation of black depositions with a consistency of soot when hydrogen was applied. To guarantee an adequate shielding effect in the case of argon shroud gas on one hand and an effectual entrainment of nitrogen for reactive spraying of the feedstock on the other, high shroud gas flows of 90 SLPM were

Fig. 14. 3D sectional view of the shroud modules of Triplex (top) and DELTA-Gun (bottom): 1 Gun, 2 shroud gas inlet, 3 powder injector, 4 water-cooled shroud, 5 helical injected

**Parameter Triplex DELTA-Gun**  Argon flow 35 slpm 40 slpm Helium flow 20 slpm - Current 250 – 350 A 210 – 330 A Plasma brut powers 36 kW - 51 kW 24 – 48 kW

(shroud/substrate) 10 mm 10 mm Traverse speed 48 m/min 48 m/min Shroud gas flow 90 SLPM 90 SLPM

The obtained coatings microstructures are illustrated in Figure 15. The micrographs on the left hand side show coatings sprayed in an inert atmosphere of argon, the ones on the right hand side the results of reactive spraying using nitrogen. The coatings of a and b were sprayed using the Triplex gun, whereas for spraying of the coatings c to f the DELTA-Gun was used. When spraying in inert atmosphere, the coatings show a uniformly and homogenous microstructure with a level of porosity comparable to conventional plasma

integrated in the cooling circuit of the spraying equipment.

applied for spraying with both guns.

shroud gas

Spraying distance

Table 7. Parameters for Shrouded Plasma Spraying

sprayed coatings. Otherwise when nitrogen is supplied, coatings with high levels of open cavities and microstructures comparable to metallic sponges are built. This is supposedly due to a turbulent entrainment of the nitrogen shielding gas when the feedstock reacts towards the nitride. To proof the existence of the aspired nitride phases, semi-quantitative measurements by means of energy dispersive x-ray analysis (EDX) were performed on coatings sprayed with the Triplex. The results revealed contents between approximately 13 and 17 at.-% nitrogen. Further on, the nitrogen contents of the coatings were investigated using a N/O/H-Analyser (LECO Instruments Corp, St. Joseph/USA). Unfortunately the contents of nitrogen in the titanium coatings could not be measured to the high melting point of the titanium nitride, but for the chromium coating (section f) the nitrogen content was determined to account for approximately 10 at.-%.

The existence of the nitride hard phases was also verified by indentation hardness measurements. When spraying the titanium with argon as shrouding gas, mean hardness values of 360 HV 0.1 in case of the Triplex and 270 HV 0.1 for the DELTA-Gun were measured. Otherwise with the employment of nitrogen significantly higher maximum values of more than 1200 and 1000 HV 0.1 were detected.

Fig. 15. Micrographs of shrouded plasma sprayed titanium (field a – d) and chromium (field e + f) feedstock using argon (left hand side) and nitrogen (right hand side) as shroud gas

Thermal Spraying of Oxide Ceramic and Ceramic Metallic Coatings 191

were varied. The melting behavior of the feedstock was tested with wipe tests (see SEM

Fig. 17. SEM images of wipe tests of suspension plasma sprayed ITO feedstock and top

With optimized parameter sets the coatings were sprayed on slides of borosilicate glass with both plasma guns. The coatings were uniformly deposited (see top view SEM images in **Figure 17**), showing homogenous structures. The coatings were measured by a project partner regarding their thickness and electrical conductance. It could be proven, that coatings with a thickness of approximately 400 nm and a sheet resistance of 850 Ω could be

Fig. 18. Transmission spectra of four ITO coated glass slides compared to uncoated and grid

views of ITO coatings (Triplex-II left side, DELTA-Gun on the right hand side)

images on top of **Figure 17**).

achieved.

blasted glass

To characterize the adhesion of the coating systems tensile adhesive tests according to DIN EN 582 on grid blasted 1.4301 substrates were conducted. Again slightly better values were achieved for the coatings sprayed with the Triplex-II gun, as the mean values of more than 50 MPa were measured for the titanium coatings compared to approximately 35 MPa for both the titanium and the coarsely grained chromium feedstock when spraying was performed with the DELTA-Gun. This might be due to the problems of injecting the feedstock in the case of the DELTA, as the gun uses a gas flow supporting the cooling of the anodes. Together with the plasma and the shrouding gas, the gas throughputs through the shroud module are high and a proper feedstock injection is not easily achieved. Therefore, further optimization potential is given for the shrouded spraying in case of the DELTA-Gun. Otherwise when using the fine fractionated chromium feedstock, the tensile adhesion of the coatings reach nearly 50 MPa, comparable to the coatings sprayed with the titanium feedstock with the Triplex gun.

When spraying the titanium on polished substrates instead of the grid blasted samples, even higher tensile adhesive strengths of nearly 60 MPa were measured. This result being not expected is probably due to diffusion phenomena of the titanium into the austenitic substrate. The effect was not recorded when using ferritic steels. In Figure 16 the backscattered electron micrograph (left hand side) and an EDX line scan analysis (right hand side) of the interface section of a titanium coating on 1.4301 steel substrate is shown. The EDX analysis confirms the findings of a zone of some micrometers depth, in which the titanium diffused. It can be stated as remarkable result, that with the limited heat transfer to the substrate enough potential is given for the diffusion process. This is due to the high diffusion coefficients of both titanium and chromium in 1.4301 austenitic steel (Kale, G.; 1998).

Fig. 16. BSE image of the interface of a titanium coating on 1.4301 austenitic steel (left) and corresponding EDX line scan (right hand side)

To investigate the alteration of the feedstock in the suspension plasma spraying process, Indium-Tin-Oxide (assumed composition of 9:1) was suspension plasma sprayed. ITO is used to coat glass for electrically conductive coating beeing transparent in the visible spectrum. The coatings are commonly deposited by sol-gel methods and are used in touchscreen purposes. The goal was to reach thin optical transparent ITO coatings showing electrical conductance. When overheating the feedstock it tends to build coatings with a yellowish color, whereas the coating system shows no conductance when it is not uniformly deposited. To find optimal conditions the relevant parameters (solid content of feedstock, species of the outer phase, injection conditions, applied amperage and spraying distance)

To characterize the adhesion of the coating systems tensile adhesive tests according to DIN EN 582 on grid blasted 1.4301 substrates were conducted. Again slightly better values were achieved for the coatings sprayed with the Triplex-II gun, as the mean values of more than 50 MPa were measured for the titanium coatings compared to approximately 35 MPa for both the titanium and the coarsely grained chromium feedstock when spraying was performed with the DELTA-Gun. This might be due to the problems of injecting the feedstock in the case of the DELTA, as the gun uses a gas flow supporting the cooling of the anodes. Together with the plasma and the shrouding gas, the gas throughputs through the shroud module are high and a proper feedstock injection is not easily achieved. Therefore, further optimization potential is given for the shrouded spraying in case of the DELTA-Gun. Otherwise when using the fine fractionated chromium feedstock, the tensile adhesion of the coatings reach nearly 50 MPa, comparable to the coatings sprayed with the titanium

When spraying the titanium on polished substrates instead of the grid blasted samples, even higher tensile adhesive strengths of nearly 60 MPa were measured. This result being not expected is probably due to diffusion phenomena of the titanium into the austenitic substrate. The effect was not recorded when using ferritic steels. In Figure 16 the backscattered electron micrograph (left hand side) and an EDX line scan analysis (right hand side) of the interface section of a titanium coating on 1.4301 steel substrate is shown. The EDX analysis confirms the findings of a zone of some micrometers depth, in which the titanium diffused. It can be stated as remarkable result, that with the limited heat transfer to the substrate enough potential is given for the diffusion process. This is due to the high diffusion coefficients of both titanium and chromium in 1.4301 austenitic steel (Kale, G.;

Fig. 16. BSE image of the interface of a titanium coating on 1.4301 austenitic steel (left) and

To investigate the alteration of the feedstock in the suspension plasma spraying process, Indium-Tin-Oxide (assumed composition of 9:1) was suspension plasma sprayed. ITO is used to coat glass for electrically conductive coating beeing transparent in the visible spectrum. The coatings are commonly deposited by sol-gel methods and are used in touchscreen purposes. The goal was to reach thin optical transparent ITO coatings showing electrical conductance. When overheating the feedstock it tends to build coatings with a yellowish color, whereas the coating system shows no conductance when it is not uniformly deposited. To find optimal conditions the relevant parameters (solid content of feedstock, species of the outer phase, injection conditions, applied amperage and spraying distance)

feedstock with the Triplex gun.

corresponding EDX line scan (right hand side)

1998).

were varied. The melting behavior of the feedstock was tested with wipe tests (see SEM images on top of **Figure 17**).

Fig. 17. SEM images of wipe tests of suspension plasma sprayed ITO feedstock and top views of ITO coatings (Triplex-II left side, DELTA-Gun on the right hand side)

With optimized parameter sets the coatings were sprayed on slides of borosilicate glass with both plasma guns. The coatings were uniformly deposited (see top view SEM images in **Figure 17**), showing homogenous structures. The coatings were measured by a project partner regarding their thickness and electrical conductance. It could be proven, that coatings with a thickness of approximately 400 nm and a sheet resistance of 850 Ω could be achieved.

Fig. 18. Transmission spectra of four ITO coated glass slides compared to uncoated and grid blasted glass

Thermal Spraying of Oxide Ceramic and Ceramic Metallic Coatings 193

Dean, A. M. & Voss, D (1999). *Design and Analysis of Experiments, Springer, ISBN 978-0-387-*

Fauchais, P. et al. (2008). Parameters Controlling Liquid Plasma Spraying: Solutions, Sols, or Suspensions, *Journal of Thermal Spray Technology*, Vol.17 (2008), No. 31, pp. 31-59 Gardos, M. N. (1988). The Effect of Anion Vacancies on the Tribological Properties of Rutile

Gell, M., et al. (2001). Development and implementation of plasma sprayed nanostructured ceramic coatings. *Surface and Coatings Technology*, Vol. 146-147 (2001), pp. 48-54 Goldschmidt, V. M. (1926). Die Gesetze der Krystallochemie, Naturwissenschaften, Vol. 14

Hawk, D. and Müller, F. (1980). Thermochemie des Systems CoO-B2O3, *Z. anorg. allg. Chem.*,

Heimann, R.B. (2008). Plasma Spray Coating – Principles and Applications, pp. 389. WILEY-

Kale, G. (1998). Interdiffusion studies in titanium 304 stainless steel system. *Journal of Nuclear* 

Lugscheider, E. & Bach, Fr.-W. (eds., 2002). *Handbuch der thermischen Spritztechnik.:* 

Mason, R. L. et al. (2003). *Statistical Design and Analysis of Experiments*, John Wiley & Sons,

Matthäus, G., Wolf, J. & Ackermann, D. (2010): *Near-net-shape HVOF coating and finishing* 

NIST (July 2011). *NIST/SEMATECH e-Handbook of Statistical Methods, 01. July 2011. Available* 

Paul, A. (1975). Activity of nickel oxide in alkali borate melts, *Journal of Materials Science*, Vol.

Phadke, M. S. (1989). *Quality Engineering Using Robust Design*, Prentice Hall, ISBN 978-

Shannon, R.D. (1976). Revised effective ionic radii and systematic studies of interatomic

Tilmann, W. et al. (2008a). Influence of the HVOF gas composition on the thermal spraying

Tilmann, W. et al. (2008b). Near-Net-Shape and Dense Wear Resistant Thermally Sprayed

distances in halides and chalcogenides, *Acta Cryst. A*, Vol. 32 (1976), pp. 751-

of WC-Co submicron powders (-8 + 1 micron) to produce superfine structured cermet coatings. *Journal of Thermal Spray Technology*, Vol.17, No.5-6, (2008), pp. 924-

Coatings. *Key Engineering Materials*, Vol.384-384 (2008), pp. 117-123, ISSN 1013-

*Materials*, Vol.257, No.1, (1998), pp. 44-50, ISSN 0022-3115, DOI:

*Technologien - Werkstoffe – Fertigung*, Verlag für Schweißen und Verwandte

*techniques for highly stressed components in aircraft industry*, Proceedings of the International Thermal Spray Conference 2010, ISBN 978-3-87155-590-9, May 03 - 05

VCH Verlag, ISBN 978-3-527-32050-9, Weinheim, Germany

Verfahren, DVS-Verl., ISBN 3871551864, Düsseldorf, Germany

ISBN 9780471372165, Hoboken, NJ, USA, DOI: 10.1002/0471458503

http://dx.doi.org/10.1016/0022-3115(88)90072-4

*from: <http://www.itl.nist.gov/div898/handbook/>* 

*98561-9, Berlin, Germany, DOI: 10.1007/b97673* 

(1926), No. 21, pp. 477-485, in German

Vol. 466 (1980), pp. 163-170, in German

2010, Singapore

767

9826

10 (1975), pp. 422-426

932, ISSN 1059-9630

0137451678, New Jersey, USA

(TiO2-x). *Tribol. Trans.* Vol. 31(4), 1988, pp. 427-455

004-0

spraying, in: *Proc. of the ITSC 2009*, 04.-07. May 2009, Las Vegas, ISBN 978-1-61503-

To determine the optical transparency of the coatings, four samples were measured using a VIS-spectrometer and the results were compared to uncoated and grid blasted glass (see transmission spectra in Figure 7). As source the tungsten lamp of the calibration module of a Tecnar DPV-2000 was used delivering a stable spectrum covering the whole visible range. The coated samples show a high degree of transparency over the whole visible spectrum. For example in the red range below 700 nm (see marking), the relative intensity measured is maximal 1 to 3 counts lower than that of the uncoated glass. This equals to a grade of transparency of 95 to 98%. It can be stated, that both requirements regarding the electrical conductance as well as the optical transparency of the coatings systems were fulfilled. These findings show, that by suspension plasma spraying new coating systems can be realized in fields of operation, where up until now coating deposition processes like CVD and PVD are used.
