Through-Hole PCB Plating With Ultrasonically-Dispersed Copper Nanoparticles

Estimated reading time: 13 minutes

PurposeThe work detailed in this paper was carried out to investigate whether copper nanoparticles could be utilized for two types of through-hole plating in printed circuit boards (PCBs) namely: 1) as a catalytic material to initiate the electroless copper deposition process and 2) as a "conductive" layer which is coherent and conductive enough to allow "direct" electroplating of the through hole. The employment of nanoparticles meant that an effective method of dispersion was required and this work studied the use of mechanical agitation and ultrasound for this purpose.Design/Methodology/ApproachThe study utilized drilled copper-clad FR-4 laminate to evaluate the processes. The through-holes were functionalized using a commercially available "conditioner" before being immersed in a solution of copper nanoparticles which were dispersed using either a magnetic stirrer or ultrasound (40 kHz). When the copper nanoparticles were utilized as a catalytic material for electroless copper plating the efficacy of the technique was assessed using an industry standard "backlight" test which allows the plating coverage of the through-holes to be determined. As a control a standard palladium catalysed electroless copper process was employed. The morphology of the electroless copper deposit obtained by each process was also analysed using scanning electron microscopy.In the "direct plate" approach, after immersion in the copper nanoparticle dispersion, the through holes were electroplated at 3 Adm-2 for 15 minutes, sectioned and examined using an optical microscope. The distance that the copper electroplate had penetrated down the through-hole was then determined.FindingsThis study has shown that copper nanoparticles can be used as a catalytic material for electroless copper plating. The coverage of the electroless copper in the through-hole was improved as the copper nanoparticle concentration was increased and, at the highest copper nanoparticle concentrations employed good, but not complete, electroless copper coverage was obtained. Dispersion of the copper nanoparticles using ultrasound was critical to the process.Ultrasonically dispersed copper nanoparticles achieved some limited success as a conductive layer for "direct" electroplating with some penetration of the electroplated deposit into the through hole. However if mechanical agitation was employed to mix the nanoparticles no through hole plating was obtained.Original/Value of PaperThis paper has demonstrated the "proof of concept" that copper nanoparticles can be utilized to catalyse the electroless copper process and their potential to replace costly palladium-based activators. The paper also illustrates the potential for copper nanoparticles to be used as a "direct plate process" and the necessity for using ultrasound for their dispersion in either process.1. Introduction

The metallization of the through holes and vias of PCBs is a critical step in their manufacture which enables the various circuits on and within the board to be connected along with surface mount and embedded components by means of the through holes and vias of the circuit board. The dielectric base material exposed by drilling these through holes and vias must be made conductive so that the through hole can be subsequently electroplated. For the past 40 years the most utilized technique for "making holes conductive" has been the electroless copper process [1]. In the electroless process, the desmeared through holes are first functionalized using a solution containing a "conditioning" agent. This changes the charge on the epoxy material, wets the hole and thus enhances the coverage of the subsequent metallization process. The conditioning step is followed by catalysation. This often involves the use of a palladium-tin colloid which is deposited on the epoxy surface of the hole wall. The tin from the catalyst is then stripped away to leave palladium metal either using an accelerator solution (often a fluoroborate containing formulation) or it may occur in the electroless copper solution itself. The electroless copper plating process occurs via a mix of electrochemical and chemical reactions [2]. The solution contains a reducing agent (commonly formaldehyde) which is oxidized on the palladium metal. This releases electrons which are then used to reduce the copper-ethylene-diamine-tetra-acetic acid (EDTA) complex and in this way copper is deposited on the hole walls of the through hole. Further oxidation of the reducing agent then occurs on the deposited copper and for this reason the process is sometimes referred to as autocatalytic. It can be seen from this very brief description that the electroless process contains a number of process stages and many of these operate at elevated temperatures. The catalyst solution is expensive as it contains palladium whilst the electroless copper formulation is a complex mix of chemicals including copper, EDTA, a reducing agent (formaldehyde), and trace amounts of additives to enhance the grain structure and rate of plating. In addition copious amounts of water are required for rinsing between the various process stages and waste treatment costs are high.A number of alternatives to electroless copper have been developed over the years and have generally been described as "direct plate" since they deposit a conductive layer in the through-holes which can be "directly" electroplated eliminating the need for the electroless plating process. They often involve depositing a conductive film in the through-hole which might be based on palladium sulphite [3] carbon/graphite [4] or a conductive polymer [5]. However, these types of processes all suffered from a common drawback, namely that in multi-layer boards (MLB) the copper inner layers also become coated with the conductive film. This has to be removed before electroplating to ensure good connectivity. To achieve this, boards have to go through a post etch and sometimes a high-pressure rinse to ensure that all the film has been cleaned away from the inner layers. When one considers that there may be several hundred holes in a PCB with a number of inner layers it can be seen that to ensure that every single interconnect is free of film is extremely difficult. The latest generation of direct plate processes look to overcome this problem by depositing a conductive film only on the epoxy material [6].

Despite the development of direct plate processes electroless copper still predominates largely due to its proven reliability in MLB production. One of the reasons for this is that, as it deposits a layer of copper in the through holes of the circuit board, there is no need to remove it from inner-layers. However, it still remains a relatively expensive process particularly due to its use of a palladium based catalyst.

In this study we have functionalized the through holes of PCBs using a commercially available conditioning product. This step is followed by the deposition of copper nanoparticles into the through holes of PCBs, using either ultrasound or mechanical stirring to disperse them in an aqueous solution at room temperature. The through holes were then plated using two approaches. In the first, they were placed into a commercial electroless copper plating bath whilst in the second they were "directly" electroplated. This paper will detail this proof of concept work and demonstrate the feasibility of using ultrasonically-dispersed copper nanoparticles for the through hole plating of PCBs.2. Experimental

The double-sided test boards were 1.5 mm thick copper clad FR4 material with 12 drilled holes of diameter 0.6 mm. All the test coupons used in the study were first desmeared using the following process:

• M-Treat AQ (MacDermid Ltd.) , 75°C, 10 minutes
  • Rinse 3 minutes
  • Rinse 3 minutes
• Potassium Permanganate (55g/l K2MnO4, 32g/l NaOH), 75°C, 10 minutes
  • Rinse 3 minutes
  • Rinse 3 minutes
• Neutralizer (3 % v/v H2SO4, 3% v/v H2O2), 25°C, 3 minutes

The coupons were then treated in the following solutions:

Stage 1:

• Circuposit Conditioner 3323-1 (Chestech Ltd.) 20%, 50°C, 15 minutes
  • Rinse 5 minutes
  • Rinse 5 minutes

Stage 2:

• Cu nanoparticles (Evochem GmbH, particle size <50 nm)
  • Dispersed/mixed in deionized (DI) water, 26°C, 30 minutes
  • Rinse 1 minute

The through holes in the PCB were then plated using either:

• a) Electroless Copper, Circuposit 3350-1 (Chestech Ltd.), 46°C, 30 minutes or
• b) Electroplating (80 g/l CuSO4.5H2O, 225 g/l H2SO4), 3 Adm-2, 25°C, 15 minutes

The copper nanoparticle concentration used in Stage 2 of the process was varied from 15 to 75 g/l. These nanoparticles were dispersed in 50 ml of DI water at 26°C using a Langford Ultrasonic (Model 375TT) 40 kHz ultrasonic bath. Calorimetry [7] was utilized to determine the power density which was found to be 0.25 Wcm-3. At the highest nanoparticle concentration an assessment was made of the effect of dispersion technique i.e. by comparing ultrasonic with mechanical (magnetic stirrer) agitation.

The coverage of the electroless copper plating into the through-holes was assessed using an industry standard backlight test [8]. The electroless plated holes were first sectioned and then examined using an optical microscope on "backlight," i.e., the light source of the microscope was shone through the back of the sectioned holes. If the holes are completely coated with electroless copper then no light will appear and it will be given a grade of "5." If no copper has been plated through the hole all the light will come through and the sample will be graded "0." Figure 1 shows the standard backlight grading scale used in the PCB industry. For each experimental run at least 10 through holes were graded and an average was calculated. Figure 1: Backlight Scale [Poole et al 2007].As a control run, boards were plated using a palladium based catalyst process as shown below using products supplied by Chestech Ltd.:

• Circuposit Conditioner 3323-1 (5% v/v), 50°C, 5 minutes
  • Rinse 3 minutes
  • Rinse 3 minutes
• Circuposit Catalyst Pre-Dip 3340, 25°C, 1 minute
• Circuposit Catalyst 3344, 40°C, 5 minutes
  • Rinse 2 minutes
  • Rinse 2 minutes
• Electroless Copper, Circuposit 3350-1, 46°C, 30 minutes

The sections used for backlight assessment were also examined using a Jeol JSM-6060LV scanning electron microscope (SEM) after sputter coating with palladium-gold.

The samples which had been "directly" electroplated were sectioned in the same way as for the backlight test, but were then examined using an optical microscope. The distance to which the electroplated copper had penetrated down the through hole was then measured using a calibrated graticule.

3. Results and Discussion

The backlight results obtained after electroless copper plating of the through holes which had been "catalysed" using different concentrations of ultrasonically dispersed copper nanoparticles are shown in Figure 2. It is quite clear that the electroless copper coverage in the through-hole improves with increasing nanoparticle concentration. Comparison of the backlights obtained using the highest copper nanoparticle concentration (75 g/l) using either ultrasonic dispersion (Figure 3a) or mechanical agitation (Figure 3b) demonstrate that sonication leads to significantly improved electroless copper coverage. However, even with ultrasonic dispersion of the highest nanoparticle concentration the backlight grading was not as good as the standard palladium catalysed sample (Figure 3c). Figure 2: Effect of copper nanoparticle concentration on backlight after electroless copper plating.Figure 3a: Backlights after electroless copper plating: 75.0 g/l copper nanoparticles, ultrasonically dispersed (Grade 3.4).Figure 3b: Backlights after electroless copper plating: 75.0 g/l copper nanoparticles, magnetic stirrer (Grade 0.5).Figure 3c: Backlights after electroless copper plating: Standard palladium catalyst (Grade 5).It is also noticeable from Figure 3(a) that either end of the hole exhibits almost complete coverage but towards the centre of the hole more of the base substrate is exposed. Consideration of the SEM photographs of the same three samples (Figure 4 a-c) further emphasise these findings and also provide some evidence for the mechanism of electroless copper plating using the various methods of catalysation. Figure 4a shows a somewhat nodular electroless copper deposit and it seems probable that the electroless copper has grown from the copper nanoparticles and the density of these particles has been high enough to allow a mostly coherent deposit to form. However, a glass fibre is visible beneath the plating demonstrating that the coverage is not complete. If the nanoparticles were mixed using a magnetic stirrer for agitation, then Figure 4b suggests that although the electroless copper has grown from the individual nanoparticles their density has not been high enough for the plating to coalesce and the substrate material is clearly visible beneath the plating. Figure 4c illustrates the electroless copper deposit obtained with a standard palladium catalyst and it can be seen that fine grained, uniform coating of the glass fibres has been achieved.Figure 4a: SEM photographs after electroless copper plating: 75.0 g/l copper nanoparticles, ultrasonically dispersed. Figure 4b: SEM photographs after electroless copper plating: 75.0 g/l copper nanoparticles, magnetic stirrer.Figure 4c: SEM photographs after electroless copper plating: Standard palladium catalyst.The results for the "direct" electroplating experiments are shown in Figure 5 and follow a similar pattern to the backlight results i.e. that the penetration of the electroplate into the through holes increases with copper nanoparticles concentration, particularly above 45 g/l.Figure 5: Effect of copper nanoparticle concentration on through hole electroplating penetration.However, when it is considered that the thickness of the PCB boards used in this study were 1.5 mm it is clear that even the maximum electroplating depth of around 150 µm (i.e. 0.15 mm) is a long way short of complete through hole plating. Once again, a comparison was made of the effect of mechanical stirring of the nanoparticles and ultrasonic dispersion at the highest nanoparticle concentration. Photographs of the penetration of the subsequent electroplating are shown in Figures 6a and 6b which were taken using an optical microscope on "top light." It is evident that some plating into the through holes has occurred on the samples prepared using ultrasonically dispersed copper nanoparticles and that almost no plating has been achieved if the particles are simply stirred. It is not surprising that the electroplated deposit grows from the ends of the through hole which must therefore be the area with the highest density of copper nanoparticles.Figure 6a: Top lights after electroplating: 75.0 g/l copper nanoparticles, ultrasonically dispersed.Figure 6b: Top lights after electroplating: 75.0 g/l copper nanoparticles, magnetic stirrer.The results from the electroless and electroplated samples indicate that a much higher density of copper nanoparticles are required to achieve complete through hole coverage using a "direct" electroplating approach as opposed to the electroless technique. Such a result is not unexpected as the mechanism for the two types of plating are quite different. To achieve through-hole plating using ‘direct' electroplating requires a highly conductive, continuous layer throughout the through hole whilst electroless plating can grow from discreet nanoparticles which, if they are in close proximity to one another, can form a coherent deposit.

Conclusions

This study has shown that ultrasonically-dispersed copper nanoparticles can be utilized as a catalytic layer for through-hole electroless deposition. Although the coverage achieved in this study was not as good as a standard palladium catalysed sample it was significantly better than if mechanical agitation is used to disperse the copper nanoparticles.

If the copper nanoparticles were "directly" electroplated only very limited through hole plating was achieved and only if they were ultrasonically dispersed. However, again this compared favourably with the results achieved if the through holes were processed using mechanically stirred copper nanoparticles as, in this case, virtually no electroplating penetration into the through hole was obtained.

Acknowledgements

The authors would like to thank the Innovative electronic Manufacturing Centre (IeMRC) for funding this work and Chestech Ltd. for supplying the process chemistry.

References

1. Deckert, C.A., "Electroless Copper Plating a Review: Part 1," Plating & Surface Finishing, Vol. 82 (2), pp. 48-55, 1985.2. Ogara, T., Malcomson, T., Fernando, Q., "Mechanism of copper deposition in electroless plating," Langmuir, Volume 6 (11), pp. 1709-1710, 1990.3. Bladon, J.J., Lamola, A., Lytle, F.W., Sonnenberg, W., Robinson, J.N., Philipose, G., "A palladium sulphide catalyst for electrolytic plating," J. Electrochem. Soc., Vol. 143 (4), pp. 1206-1213, 1996.4. Carano, C., "Performance and reliability issues for a graphite based direct metalization process," Circuit World, Vol. 25 (3), pp. 18-22, 1999.5. Hupe, J., Altgeld, W., "Die industrielle Anwendung der Durchmetallisierung von Leiterplatten mittels leitfaehiger organischer Polymere," Galvanotechnik, Vol. 85 (3), pp. 929-935, 1994.6. Rasmussen, J., "Intrinsically conductive polymer for PWB direct metallization: process reliability by process management," Circuit World, Vol. 35 (1), pp. 9-15, 2009.7. Kimura, T., Sakamoto, T., Leveque, J-M., Sohmiya, H., Fujita, M., Ikeda, S., Ando, T., "Standardization of ultrasonic power for sonochemical reaction," Ultrasonics Sonochemistry, Vol. 3 (3), pp. S157-S161, 1996.8. Poole, M.A., Cobley, A.J., Singh, A,, Hirst, D.V., "Environmentally friendly electroless copper compositions," European Patent Application EP1876262, 2007.