Comparison of the Fracture Behavior of Brittle ILD Films used in the BEOL in Dry and Wet Environment using Nanoindentati

  • PDF / 1,358,552 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 27 Downloads / 141 Views

DOWNLOAD

REPORT


0914-F04-11

Comparison of the Fracture Behavior of Brittle ILD Films used in the BEOL in Dry and Wet Environment using Nanoindentation Eva E Simonyi, Michael Lane, Erik Liniger, and Alfred Grill Research, IBM, 1101 Kitchawan Rd, Yorktown Heights, NY, 10598

Abstract In the BEOL, during the manufacturing process, the low-k brittle ILD dielectrics are exposed to wet environments. These environments could and do affect the films fracture toughness and the so called critical film thickness, above which spontaneous cracking occurs. Nanoindentation combined with AFM imaging methods were used to study these phenomena. Typical examples will be presented.

Introduction The demand of ever smaller critical dimensions of very large scale integrated circuits, combined with higher operating frequencies of the circuits drives the need in the back end of the line (BEOL) to use low dielectric constant (low-k) materials to prevent crosstalk, RC delay and other power related issues. A typical choice for inter dielectric layers (ILD), are porous materials prepared using either spin-on or chemical vapor deposition. These materials are usually mechanically very weak and in order to bracket in the most favorable process parameters and/or to qualify manufacturing tools, there is a need to monitor modulus, hardness and resistance to fracture with a relatively fast turnaround time. The values of these parameters are of interest both for laboratory and wet environments. Nanoindentation in combination with AFM imaging methods was used here to compare different materials in both laboratory and wet environments.

Experimental method Hardness and modulus data were acquired using the nanoindentation method. Nanoindentation was performed with a Nano Intender XP system (Nano Instruments Innovation Center), equipped with the Dynamic Contact Module (DCM). The DCM provides an overall miniaturization of the XP system, making it more suitable to perform indentations in low force ranges (.01mN to 12mN). The DCM machine in this case used a Berkovitch indenter (angle 65.30). The Continuous Stiffness Measurement Option (CSM) was used. This technique superimposes a small oscillatory force to the indentation force, allows a continuous measurement of modulus and hardness during the indentation process. Tip calibration was based on the Oliver -Pharr method (1). Detection of the surface was a possible problem; a stiffness change of 4 times was used as an indicator. The indentation was done such that a constant strain rate was maintained during the indentation. The hardness and modulus data are the result of the average of 20 indents. Fracture was induced in the films using a UMIS (CSIRO) nanoindentation system equipped with a corner cube indenter (angle 35.3). This diamond tip being sharper than

the Berkovich indenter creates bigger stress concentration under the tip, thus the cracks occur at a lower indentation force, lower indentation depths, lowering the effect of the substrate on the results. The indentations were done in 20 steps to the maximum load and a