IPC-4556 印制板化学镍钯浸金（ENEPIG）规范ENG - 第17页

3.2 Finish Thickness Thicknesses of the electroless nickel, electroless palladium and immersion gold layers shall be mea- sured and veriﬁed following the completion of the ENEPIG plating operation in the printed board fa…

3.1.1 High Magnification Reference Images It may be necessary to evaluate the ENEPIG deposit using scanning electron

microscopy (SEM), transmission electron microscopy (TEM) or focused ion beam (FIB) as part of a qualiﬁcation plan and

or for failure analysis. The images in Figures 3-7 through 3-12 in the following section are for reference purposes and were

provided by several different suppliers of ENEPIG.

Figure 3-5 Skip Plating of Gold Over Palladium Figure 3-6 Skip Plating of Gold Over a Palladium Deposit

Figure 3-7 1000X SEM Image of a Normal ENEPIG Surface Figure 3-8 4000X SEM Image of a Normal ENEPIG Surface

Figure 3-9 2500X FIB Image of a Normal ENEPIG Deposit Figure 3-10 100,000X FIB Image of a Normal ENEPIG

Deposit

IPC-4556 January 2013

×2.5k

20kV

20 m

察

3.2 Finish Thickness Thicknesses of the electroless nickel,

electroless palladium and immersion gold layers shall be mea-

sured and veriﬁed following the completion of the ENEPIG

plating operation in the printed board fabrication process. The

use of X-ray ﬂuorescence (XRF) methodology for thickness

determination shall be in accordance with APPENDICES 4

and 9 of this document and shall employ XRF instrumentation

equipped with software and hardware appropriate for ENEPIG

measurement. The committee has performed extensive round

robin testing of equipment capability to accurately measure

this tertiary deposit (see APPENDICES 3, 9 and 10). The

results showed a number of critical issues with equipment,

set-up, measuring protocols and reference standards. It is

imperative to demonstrate measurement capability in order to

meet the following thickness speciﬁcation.

3.2.1 Electroless Nickel Thickness The electroless nickel

thickness shall be 3 to 6 µm [118.1 to 236.2 µin] at ± 4 sigma

(standard deviations) from the mean as measured on a nominal pad size of 1.5 mm x 1.5 mm [0.060 in x 0.060 in] or equiva-

lent area, where standard feature size tolerances as expressed in the IPC-6010 standard series, apply. This upper end of this

thickness speciﬁcation has been chosen based on concerns with insertion force issues for compliant pin applications ONLY.

Use of thicknesses outside of this range shall be AABUS.

3.2.2 Electroless Palladium Thickness The electroless palladium thickness shall be 0.05 to 0.30 µm [2 to 12 µin] at ± 4

sigma (standard deviations) from the mean as measured on a nominal pad size of 1.5 mm x 1.5 mm [0.060 in x 0.060 in] or

equivalent area, where standard feature size tolerances as expressed in the IPC-6010 standard series, apply. The upper thick-

ness limit may be exceeded to meet speciﬁc design criteria (for example, applications requiring wire bonding but no solder-

ing).

3.2.3 Immersion Gold Thickness The minimum immersion gold thickness shall be >0.030 µm [1.2 µin] at - 4 sigma

(standard deviations) below the mean as measured on a nominal pad size of 1.5 mm x 1.5 mm [0.060 in x 0.060 in] or

equivalent area, where standard feature size tolerances as expressed in the IPC-6010 standard series, apply. The speciﬁcation

is one sided, based on performance test results showing impaired solderability performance after accelerated aging at lower

gold thicknesses. The nature of the reaction of immersion gold with electroless palladium is such that substantially greater

thicknesses may not be achievable.

Figure 3-11 TEM Image of a Normal ENEPIG Deposit

[The Palladium-Phosphorus ﬁlm is amorphous and very uniform whereas the gold (Au) deposit is crystalline in nature.]

Figure 3-12 2500X Image of Nickel that is Hyper-Corroded

(This is due to excessive dwell time in the gold bath.)

January 2013 IPC-4556

×2.5K

20kV

20 m

60 nm

TEM image

Au deposit

Palladium

deposit

EN deposit

推着额期

Electron diffraction of

Palladium film

Note: Some designs may not have feature sizes of this speciﬁed pad area and will require the use of alternate pad sizes. The

XRF collimator used shall always be smaller than the pad used for measurement; and speciﬁcally, the collimator shall not

exceed 30% of the feature pad size being measured. For smaller sized pads, the measurement time will need to increase

proportionately as a function of collimator area reduction. Consistency of the pad size used for measurement is of primary

concern in order to obtain a statistically valid result, due to the typical plating thickness variation seen as a function of pad

size. In general, smaller features will plate thicker than larger features. The use of ﬁducial pads for this measurement is sug-

gested for those designs not having the speciﬁed pad feature areas, described above.

3.2.4 XRF Calibration Standards The use of National Standards traceable calibration standards with similar thicknesses to

the speciﬁed ENEPIG thicknesses which are to be measured, is recommended. Tri-layer standards where Au and Pd are

plated directly on Ni/Cu/PCB should be used for calibration for Cu thicknesses >30 µm. Tri-layer foil standards where Au

and Pd are plated on a Ni-foil should be used if boards with varying Cu-thicknesses are to be measured. Additional infor-

mation can be found in Appendix 10.

For Cu thickness >30 µm, a combination of a minimum of two calibration standards with approximate thicknesses, as below,

should be used.

• Au/Pd/Ni/Base 0.05 µm/0.02 µm/3 µm/>30 µm

• Au/Pd/Ni/Base 0.05 µm/0.09 µm/3 µm/>30 µm

• Au/Pd/Ni/Base 0.05 µm/0.3 µm/3 µm/>30 µm

• Au/Pd/Ni/Base 0.1 µm/0.2 µm/3 µm/>30 µm

For Cu thicknesses <30 µm a combination of a minimum of two calibration foil standards with approximate thicknesses as

below, should be used.

• Au/Pd/Ni/Base 0.06 µm/0.02 µm/4 µm/<30 µm

• Au/Pd/Ni/Base 0.06 µm/0.06 µm/4 µm/<30 µm

• Au/Pd/Ni/Base 0.05 µm/0.1 µm/4 µm/<30 µm

• Au/Pd/Ni/Base 0.05 µm/0.25 µm/4 µm/<30 µm

For additional details, see APPENDICES 4 and 9.

3.2.5 Use of Non-Statistical Thickness Specification For production lot sizes of ten printed boards or less, i.e., proto-

type builds, the need to demonstrate compliance to the statistical requirements above are negated, provided that measure-

ments are taken on 100% of all printed boards/panel coupons, thus providing population parameters rather than sample sta-

tistics. All such readings shall fall within the thickness ranges speciﬁed above for all three deposits. Any readings below or

above the speciﬁed thickness range shall be cause for rejection of that board.

3.2.6 Production Overruns and/or Inventory Items When compliance testing of individual printed board(s) remaining in

inventory is required, use the protocol of 3.2.5. For production overruns larger than ten, the sample statistic constraints shall

apply.

3.3 Porosity An ENEPIG immersion gold layer is not totally impervious and is not able to pass the requirements of a

‘classic’ porosity test (for example, Nitric Acid per ASTM B 735).

3.4 Adhesion The purpose of adhesion testing for ENEPIG ﬁnished boards is twofold.

1) The primary purpose of the testing is to verify that there is no adhesion failure between the metallic layers of the ﬁn-

ish, or the ﬁnish and the base metal. The tape testing shall be carried out in accordance with IPC-TM-650, Method

2.4.1, using a strip of pressure sensitive tape. There shall be no evidence of any portion of the surface ﬁnish being

removed, as shown by particles of plated deposit/pattern adhering to the tape. If soldermask overhang breaks off and

adheres to the tape, it is evidence of undercut but not of adhesion failure.

2) The secondary purpose is for testing the adhesion of the solder mask to both traces and laminate. Both the chemistry

used in the ENEPIG process and the required dwell times and temperatures can have a negative impact on solder mask

adhesion. – Note: Not all solder masks are compatible with an ENEPIG process.

In both cases, testing should be conducted on regions of high feature density on the board, such as BGA sites, dams between

ﬁne pitch leaded devices or over areas of high trace density. Using the same approach as for the test for metal adhesion,

check for evidence of the soldermask being removed, as shown by particles of soldermask adhering to the tape.

IPC-4556 January 2013