Technology Review - Published By MIT
Log in to My.TechnologyReview.com | Register
Advertisement

Wednesday, May 23, 2007

Nanoglue for Electronics

Researchers have found organic molecules that can act as an effective and cheap glue to stick together tiny electronic components.

By Prachi Patel-Predd

smaller text tool iconmedium text tool iconlarger text tool icon
Nano superglue: Organic molecules made of a chain of carbon and hydrogen atoms with sulfur (blue) at one end and silicon (green) at the other hold together copper and silicon dioxide. The molecules organize themselves and line up next to each other, and their adhesive strength increases at very high temperatures of up to 700 ºC.
Credit: Rensselaer/G. Ramanath

Researchers at the Rensselaer Polytechnic Institute, in Troy, NY, have found that certain nanometer-long organic molecules can bond two surfaces that normally don't stick together well. Surprisingly, the adhesive power increases when the nanoglue is exposed to very high temperatures.

The molecules could be used as an inexpensive, easy-to-apply glue in a variety of applications. For example, the nanometer-thick glue could be used to hold together tiny electronic components, as transistors and wires on computer chips continue to shrink, says Ganapathiraman Ramanath, a materials science and engineering professor who led the study, which was published in Nature last week.

The nanoglue, which belongs to a class of compounds called organosilanes, consists of a chain of carbon and hydrogen atoms with sulfur at one end and silicon at the other. The molecular chain normally disintegrates at temperatures above 300 to 400 ºC. But Ramanath and his colleagues have found that when they sandwich the molecules between copper and silicon dioxide, the molecules not only bind together the two materials, but the bond strengthens at higher temperatures. At room temperature, the resulting bond is three times stronger than a direct bond between copper and silica. At 700 ºC, the bond is 10 times stronger than normal.

One advantage of the glue is how little of it is needed. Similar gluing strengths can be achieved with very thick adhesive layers but not with such thin layers, Ramanath says. Since a single layer of the organosilane molecules arranged side by side holds the copper and silica, the thickness of the adhesive layer is the length of a single molecule: close to one nanometer. At 35 cents a gram, the new glue is affordable. And it should be easy to apply because the molecules tend to organize themselves in the proper orientation on the surface "like soldiers," Ramanath says. "They all stand right next to each other and line up quite closely."

Moreover, the researchers expect that they can tailor the nanoglue to adhere to different materials. By attaching appropriate chemical groups at the two ends of the molecular chain, researchers could engineer new types of organosilane molecules to glue together other dissimilar materials, such as insulators and semiconductors, or metal and semiconductors.

The organosilane's increasing adhesive strength at higher temperatures is anomalous and "contrary to conventional wisdom," says Om Nalamasu, a vice president and chief technology officer at Applied Materials, based in Santa Clara, CA, which supplies fabrication equipment to the semiconductor industry. "This might have neat applications and could open some new ideas and new concepts."

One important application could be gluing copper wires that connect the various components on computer chips. Copper wires are deposited on insulating silicon-dioxide layers on a computer chip to keep the wires' signals from mixing with each other. But copper doesn't stick tightly to the silicon dioxide, and copper molecules diffuse into the silica. "There is a big need to isolate the interfaces chemically," Ramanath says. "You don't want them to mix, yet you want adhesion."

Chip manufacturers currently use at least 10-nanometer-thick layers of materials such as tantalum or titanium between the copper and silicon dioxide. But as device sizes on high-performance computer chips plunge into the nanometer range, the new nanoglue, which is 10 times thinner, would be an ideal replacement. "With increasing miniaturization, you can't afford to waste real estate on things that don't do anything other than keep things together," Ramanath says.

Comments

  • Nano vs. Chem 101
    KenGilleo on 05/23/2007 at 1:45 PM
    Posts:
    5
    We've been using silanes (widely available) in adhesives for many decades. They are based on the element silicon; Si and have been considered chemicals - not nano this or that. They’re commonly used to join dissimilar materials together and are extremely useful when one of surfaces is silicon, or silicon dioxide (SiO2). Most electronic polymers contain silanes as additives. Old adhesives books state that the theoretical minimum thickness of an adhesive is “one molecule thick”. But adhesives are used at a much thicker level to compensate for surface roughness. However, silanes used as surfactants self-assemble as a single layer and they been doing this before the nanotech term was coined. So is the RPI “discovery” one more nano-breakthrough, or good old chemistry under a new moniker?
    Rate this comment: 12345
    • Re: Nano vs. Chem 101
      gwf_fly on 05/23/2007 at 3:10 PM
      Posts:
      9
      Hi KenGilleo,
      I read your reply to the "nano-glue" and was wondering if either of the work here falls under the SAM moniker, that is, Self Assembled Mono-layers?  I am aware of that treatment through super critical CO2 processing where a one-molecule thick layer of (some chemistry) is layed down in the CO2 chamber preparation for highly efficient bonding to dissimilar materials.  When I hear the statement "like orderly soldiers" used, I envision the SAM layer illustrated in various technical papers.

      Thanks,
      gwf_fly
      Rate this comment: 12345
      • Re: Nano vs. Chem 101
        KenGilleo on 05/23/2007 at 4:39 PM
        Posts:
        5
        SAM happens if you give it a chance. SAM is an important, but rather old principle, in surface chemistry; most surfactants will form monolayers. In fact, SAM is somewhat of a prerequisite for effective surfactant function. Silanes will typically form a nicely oriented monolayer on silicon oxides (most silicon has an oxide layer anyway). If we dip a wafer in a dilute solution of silane, a monolayer should result. The other end of each molecule is now available for bonding, or for other uses including acting as a barrier. Most silica (SiO2) fillers, such as used for underfills, encapsulants, and other electronic liquids, are treated with silanes to make them compatible with the polymer. If the polymer is epoxy, then a silane with an epoxy end is used. When hardened, the silane is acting as an adhesive between that silica surface and the resulting plastic; thus, SAM is involved. I’m not sure what is happening with SCCO that can be viewed as solvent to avoid hydrogen bonding (and therefore stiction with MEMS), but there could be a monolayer at some point.
        Rate this comment: 12345
        • Re: Nano vs. Chem 101
          gwf_fly on 05/23/2007 at 5:11 PM
          Posts:
          9
          Hi Ken Gilleo,
          Thank you for your further insight and knowledge into the world of silanes.  I am interested in further conversation dealing with interfacial bonding challenges. Is there an e-mail address I could contact you at with some questions?

          Thanks again,

          gwf_fly
          Rate this comment: 12345
  • Illustration beef
    DBARLEX on 05/24/2007 at 1:17 PM
    Posts:
    1
    In an illustration designed to show atoms and the way they interact why oh why are there images of hooks and eyes that are not made up of atoms? I use to be  chemisty teacher and it's just this sort of illustrative licence tht causes student confusion.Now I teach technology I'm very keen for high school students to look tat Technoloy Review but illustrations like this can really cause learning difficulty. 
    Rate this comment: 12345
  • silanes as adhesive to join Polyphenylene Sulfide and aluminium
    jsarvaiya on 05/25/2007 at 4:46 PM
    Posts:
    1
    We are planning to use a SILANES as a adhesive to join Polyphenylene Sulfide and aluminum in electronic application. so would it be possible to achieve a HIGH strength Bond between those two different material ??

    jaydip sarvaiya.
    jsarvaiya9@yahoo.com
    Rate this comment: 12345
  • Don't get lost in the semantics
    prosilane on 05/29/2007 at 5:49 PM
    Posts:
    1
    The RPI findings are interesting on several levels.  First, silanes are not adhesives but they are adhesion promoters.  Second, achieving a true monolayer of a trialkoxysilane in practice is a very difficult thing to do.  Third, the performance of most silane coatings does not improve upon exposure to 700 C.

    If this is truly a case where a silane monolayer is acting alone as an adhesive (that improves on high temperature conditioning) then this will certainly change the way I think about silanes and I have been in the silane business a long time.
    Rate this comment: 12345
Advertisement

Current Issue

Technology Review July/August 2008
The Business of Social Networks
The future of the Web is social. But can social-networking sites ever make money?
•  Subscribe
Save 41%
•  Table of Contents
•  MIT News

Magazine Services

Career Resources

MIT Technology Insider

Stories and breaking news from inside MIT about the latest research, innovations, and startups--in a convenient monthly e-newsletter. Subscribe today
Advertisement
Advertisement
Advertisement
Advertisement
TECHNOLOGY RESOURCES
Advertisement
MIT Massachusetts Institute of Technology