Research Papers

Online Monitoring of Functional Electrical Properties in Aerosol Jet Printing Additive Manufacturing Process Using Shape-From-Shading Image Analysis

[+] Author and Article Information
Roozbeh (Ross) Salary, Jack P. Lombardi

Department of Systems Science
and Industrial Engineering,
Binghamton University (SUNY),
Binghamton, NY 13902

Prahalad K. Rao

Department of Mechanical
and Materials Engineering,
University of Nebraska-Lincoln,
Lincoln, NE 68588
e-mail: rao@unl.edu

Mark D. Poliks

Department of Systems Science and
Industrial Engineering,
Binghamton University (SUNY),
Binghamton, NY 13902

1Corresponding author.

Manuscript received February 7, 2017; final manuscript received April 29, 2017; published online August 24, 2017. Assoc. Editor: Robert Gao.

J. Manuf. Sci. Eng 139(10), 101010 (Aug 24, 2017) (13 pages) Paper No: MANU-17-1080; doi: 10.1115/1.4036660 History: Received February 07, 2017; Revised April 29, 2017

The goal of this research is online monitoring of functional electrical properties, e.g., resistance, of electronic devices made using aerosol jet printing (AJP) additive manufacturing (AM) process. In pursuit of this goal, the objective is to recover the cross-sectional profile of AJP-deposited electronic traces (called lines) through shape-from-shading (SfS) analysis of their online images. The aim is to use the SfS-derived cross-sectional profiles to predict the electrical resistance of the lines. An accurate characterization of the cross section is essential for monitoring the device resistance and other functional properties. For instance, as per Ohm’s law, the electrical resistance of a conductor is inversely proportional to its cross-sectional area (CSA). The central hypothesis is that the electrical resistance of an AJP-deposited line estimated online and in situ from its SfS-derived cross-sectional area is within 20% of its offline measurement. To test this hypothesis, silver nanoparticle lines were deposited using an Optomec AJ-300 printer at varying sheath gas flow rate (ShGFR) conditions. The four-point probes method, known as Kelvin sensing, was used to measure the resistance of the printed structures offline. Images of the lines were acquired online using a charge-coupled device (CCD) camera mounted coaxial to the deposition nozzle of the printer. To recover the cross-sectional profiles from the online images, three different SfS techniques were tested: Horn’s method, Pentland’s method, and Shah’s method. Optical profilometry was used to validate the SfS cross section estimates. Shah’s method was found to have the highest fidelity among the three SfS approaches tested. Line resistance was predicted as a function of ShGFR based on the SfS-estimates of line cross section using Shah’s method. The online SfS-derived line resistance was found to be within 20% of offline resistance measurements done using the Kelvin sensing technique.

Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

(a) and (b) Examples of AJP-printed electronics fabricated at Binghamton University (SUNY); (a) an antenna printed on a flexible glass substrate; (b) silver interdigitated electrodes (IDEs) printed on flexible polyimide. (c)–(h) AJP lines from the authors’ experiments exemplifying various line characteristics (Source: Ref. [5]).

Grahic Jump Location
Fig. 2

The SfS problem: reconstruction of the 3D topology of an AJP-deposited electronic trace based on the intensity gradient information captured from the surface by an image [14]

Grahic Jump Location
Fig. 3

The captured topology of a surface in form of an image depends on: (i) illumination direction (I), (ii) surface reflectivity (ρ), and (iii) camera direction (V) [12,17,18]. The surface at each point is represented by its normal vector (n). α and β are the tilt and slant angles, respectively.

Grahic Jump Location
Fig. 4

A summary of the mathematical formulation as well as a pseudo-algorithm for the estimation of illumination direction, I, and surface reflectivity (albedo), ρ [12]

Grahic Jump Location
Fig. 5

A summary of the mathematical formulation of Horn’s method together with a pseudo-algorithm proposed for numerical implementation [12]

Grahic Jump Location
Fig. 6

A summary of the mathematical formulation of Pentland’s method together with a pseudo-algorithm proposed for numerical implementation [12]

Grahic Jump Location
Fig. 7

A summary of the mathematical formulation of Shah’s method together with a pseudo-algorithm proposed for numerical implementation [12]

Grahic Jump Location
Fig. 8

The 3D surfaces of a synthetic sphere, a synthetic cylinder, and a real AJP-printed line recovered using Horn’s method [12,14], Pentland’s method [15], and Shah’s method [16]. It is evident that Shah’s method has the highest fidelity in surface reconstruction (see (d), (h), and (l)).

Grahic Jump Location
Fig. 9

A schematic diagram of AJP process utilizing a pneumatic atomizer. The setup has been instrumented with a CCD camera to capture online images.

Grahic Jump Location
Fig. 10

Pictures of the AJP setup instrumented with image-based and temporal sensors. Reconstruction and quantification of line topology are based on online images captured using the CCD camera.

Grahic Jump Location
Fig. 11

(Top) The dimensions of the printed test artifact with four-terminal (1 mm × 1 mm) probe pads for measurement of line resistance. (Bottom) silver nanoparticle devices printed on flexible Kapton.

Grahic Jump Location
Fig. 12

The influence of ShGFR on the 2D characteristics of AJP-printed lines captured online using a high-resolution CCD camera. (The atomization and exhaust gas flow rates are 580 sccm and 560 sccm, respectively.)

Grahic Jump Location
Fig. 13

The 3D profiles of AJP-printed lines recovered using Shah’s method [16] (shown only for the first replicate). The recovered profiles are normalized to be in the range of [0 1].

Grahic Jump Location
Fig. 14

A representative comparison between the online and offline profile recovery. The offline recovery (as the ground truth) is based on an optical profilometer.

Grahic Jump Location
Fig. 15

High-resolution CCD images captured in situ before sintering from four-terminal conductive structures

Grahic Jump Location
Fig. 16

Prediction of line CSA and line resistance as a function of ShGFR. The error bars are (±1 σ/n) long where n equals the number of replications (n = 3).




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In