Issue Date: August 20, 2012
Paper Devices Move Forward
University of Notre Dame chemistry professor Marya Lieberman and graduate student Abigail Weaver spent a week in June in Eldoret, Kenya, going from office to office in a regional hospital meeting people involved with pharmaceutical purchasing.
In hospitals like the one they visited, vigilant purchasing agents have to be on the lookout for counterfeit pharmaceuticals, but they often lack appropriate equipment to assess drug authenticity. Lieberman and Weaver develop paper-based microfluidic devices that can do such tests.
Hospitals like the one in Kenya “don’t have lab space,” Lieberman says. “There’s not even space to set up a suitcase lab.” In that type of environment, paper-based instruments can come in very handy. “We would whip out our paper devices and do tests right there on their desktops,” she says.
Lieberman ticks off the reasons people want to use paper devices. “They’re cheap, transportable, disposable, cheap.” She pauses for a beat. “Have I mentioned that they’re cheap?”
Lieberman is just one of a growing number of scientists who are turning to paper as a platform for microfluidic analytical devices. She and others were attracted to the field after reading George H. Whitesides’ first paper on the topic, published in Angewandte Chemie in 2007 (DOI: 10.1002/anie.200603817).
A varied cast of scientists is pushing the field forward, some by developing and improving paper-based microfluidic devices and others by using the devices for specific assays. The desire to run point-of-care medical diagnostic tests in limited-resource settings drives much of the work in this area.
For example, Paul Yager, a bioengineering professor at the University of Washington, Seattle, was already trying to develop low-cost diagnostics when he heard about paper-based devices. At the time, he was working on a project funded by the Gates Foundation to develop permanent instruments with disposable cartridges. The Bill & Melinda Gates Foundation would fly its grant recipients to annual meetings in locations where such devices might actually be used.
“They would take us out to real situations, real clinics, real population centers in the developing world—places that I frankly hadn’t been to before,” Yager says. “You’d get to chat about things like their actual budgets and the real issues and the most pressing problems. After five years of exposure to that, I reached the conclusion that the cost per test had to be way, way down, below what I still think the cost per test is going to be with the kind of technology that was being developed with a permanent instrument.”
But cost was only part of the problem, Yager adds. “Permanent instruments are a pain to use in the developing world. If you’re lucky, they sit someplace and stay plugged in. If you’re not lucky, they break or wander off, and you lose your ability to run them. It’s not easy to get a service call and, depending on where you are, not all that easy” to pack an instrument and have a shipper like FedEx pick it up for off-site servicing.
With paper-based devices, such worries disappear. Paper-based devices often don’t require power supplies. Many of them can be read by eye or with a camera phone—a common tool in the developing world.
Lieberman’s group and teams at other academic institutions with which they collaborate—led by Patrick J. Flynn and Holly V. Goodson from the University of Notre Dame and Toni L. O. Barstis from Saint Mary’s College, in Notre Dame, Ind.—develop chemical tests for counterfeit pharmaceuticals.
“We’re trying to put together libraries of color tests and get them all to play nicely with each other,” Lieberman says. “We’re also making a smart material, where information on how to add the reagents is stored in the paper matrix and the test output can be read like a color bar code.”
For pharmaceutical identification, each device has several lanes that detect certain functional groups, such as primary amines or phenols, in an active pharmaceutical ingredient. Lieberman also looks for approved or unauthorized excipients (inert ingredients). Color changes in the various lanes indicate the presence of particular components.
“These are chemical tests, so you have to be careful in interpreting them,” she says, noting that the tests are not yet sufficiently quantitative. The main bar to quantitation has been the difficulty of developing a reliable method of dosing. For ease of use, “we made a decision not to grind up tablets and make volumetric solutions,” Lieberman says. Instead, she and her coworkers simply swipe pills across the graterlike sampling region of the device.
They are working on the grater to make the device more quantitative. Nonetheless, the device is quantitative enough that during a workshop in Kenya it was able to reliably distinguish pure amoxicillin from amoxicillin that had been cut 50/50 with maize meal.
The current iteration of the test has 12 lanes, a sufficient number to identify the four frontline tuberculosis medications individually or in combination tablets of two or four drugs. The 12 lanes can be used to detect all four drugs individually, all seven possible binary combinations, and the quaternary combination.
In the medical diagnostics area, a device to perform liver function tests from the nonprofit corporation Diagnostics for All (DFA), based in Cambridge, Mass., is closer than any other paper-based device to working in the real world. The company holds an exclusive license to a number of patents from the Whitesides lab.
The test measures the levels of two liver enzymes—aspartate aminotransferase and alanine aminotransferase—and provides a colorimetric readout, says Una Ryan, DFA’s chief executive officer. By comparing the readout with a color guide, medical personnel can categorize the test results as normal, worrisome, or requiring immediate action.
DFA’s contribution has been to take academic ideas and turn them into real products, Ryan says. Most important, she adds, is that the tests are reliable and reproducible.
“It doesn’t matter how cheap something is. People won’t use it if it’s not reliable,” she says. “In the developing world, people don’t use diagnostics more often because they’re too expensive and not reliable.”
The devices currently made by DFA are stable for at least a year in high-temperature, high-humidity environments, Ryan says. The devices are now being tested at a Harvard University-affiliated AIDS hospital in Ho Chi Minh City, Vietnam. The hospital has a high incidence of hepatitis as well as AIDS, meaning that there is a substantial patient population with compromised liver function, making it possible to study whether the device measures what it’s supposed to measure. A 700-patient trial is ongoing.
DFA has developed a semiautomated process for manufacturing the devices, but the biggest hurdle to widespread use is what Ryan calls “the last mile”—distribution. Organizations like DFA can work easily with national or regional governments, but getting devices to people who work in rural communities “requires a lot of skill and good relationships,” she says.
DFA’s liver function test has not yet received regulatory approval. Most countries in the regions DFA is targeting accept both U.S. and European regulatory approvals. DFA will first apply for the CE mark, which denotes compliance with European regulatory standards. After that, it might also seek approval through the U.S. Food & Drug Administration’s 510(k) regulatory process for medical devices.
A growing number of academic researchers are working on ways to extend the capabilities and improve the assembly of paper devices. For instance, Scott T. Phillips and coworkers at Pennsylvania State University have incorporated fluidic batteries directly into paper-based microfluidic devices (Lab Chip, DOI: 10.1039/c2lc40126f).
In many conventional detection methods, like fluorescence assays, you need a source of power, Phillips says. “In really resource-limited environments, button batteries may be too expensive or have disposal hazards. We’re asking the question: Can we generate the power directly” in a paper-based device?
Phillips and his coworkers are helping answer this question by making multilayer batteries with electrolytes, electrodes, salt bridges, and conductive connections. Using AgNO3 and AlCl3 electrolytes, they constructed galvanic cells that can be connected in series or parallel and that produce enough current to power a light-emitting diode.
Phillips’ group recently devised a way to simplify the assembly of multilayer paper-based devices. Rather than alternate paper and tape layers, as is commonly done in the construction of such devices, Phillips and coworkers use spray adhesive to stick the layers together (Lab Chip, DOI: 10.1039/c2lc40331e). “The quantity of glue we are using to put two pieces of paper together is sufficiently small that it doesn’t seem to affect the wicking or wetting properties of the paper,” Phillips says. And the process is rapid enough that a single researcher can assemble hundreds of devices within an hour, he says.
Another group uses origami—Japanese paper folding—to craft multilayer paper-based devices from single pieces of paper. When Hong Liu, a graduate student in Richard M. Crooks’s group at the University of Texas, Austin, who grew up in China, first saw multilayer paper-based devices, they reminded him of origami objects he had assembled in childhood art classes.
Crooks’s group incorporates electrochemical detection into its devices by screen-printing electrodes on single pieces of paper and then using origami to fold them into multilayer operational units. Current devices have carbon electrodes, but he and his coworkers are developing devices with gold electrodes to make it easier to immobilize probe molecules on the devices.
They plan to use the origami devices for assays that test immunization status—for example, to identify children who have been immunized, even if their immunization card has been lost.
Not everybody wants to use multiple layers of paper in their paper-based device. Yager is trying to develop devices that perform multistep reactions using single pieces of paper. “I’m a monopapyrist,” he says. “I’d like all my devices to have one continuous strip of paper, no matter what the shape is.”
The challenge with using only one piece of paper is how to get all the reagents to the right place at the right time. Control of fluid flow is a critical part of that process. In conventional microfluidic devices, pumps control fluid flow. With paper, the fluid naturally wicks through the device. “One of the great things about paper is you get to throw away the pumps,” Yager says. But “the paper material has to be very uniform to do the kinds of things we’re trying to do. We’re trying to program fluid flow rates and arrival times using paper.”
The goal of all that flow control is getting the sample and reagents to a desired spot on a device. “Everything ultimately has to come to the place where I’ll make a measurement,” Yager says. “The whole question is how do I get that spot to change in a pattern and intensity that relates to how much stuff was there.”
Yager and coworkers recently used two-dimensional paper networks to detect a malaria antigen (Anal. Chem., DOI: 10.1021/ac300689s). And they are currently working on an isothermal method for nucleic acid amplification on a paper device. To design such devices, they first figure out how to make each individual step work on separate pieces of paper and then connect those “subroutines” on a single piece of paper. “We’ve developed a tool kit of paper processes,” he says.
Many paper devices are laminated, both to hold them together and to keep out contaminants. But lamination can require extra steps to incorporate sample inlets. Andres W. Martinez and coworkers at California Polytechnic State University, San Luis Obispo, have developed a way to use printing instead of lamination to enclose and protect devices (Anal. Chem., DOI: 10.1021/ac202837s).
They pattern channels on paper devices using a wax printer, add any reagents needed for the assay, and then run the paper through a standard laser printer. The toner, which is thermally bonded to the paper, encloses and protects the resulting device, making it easier to handle and operate.
Martinez’ experiments suggest that the toner coating is gas permeable, with some water vapor escaping. However, liquid water doesn’t permeate barriers consisting of at least four layers of toner. Martinez believes the toner-sealing method’s optimal use will not be more large-scale manufacturing but instead allowing small labs to carry out rapid prototyping of paper-based devices.
Most specialists of paper-based devices are working on medical tests, and initially Charles Henry, a chemistry professor at Colorado State University, was no exception. In collaboration with researchers at Chulalongkorn University, in Bangkok, he developed a paper device for metabolic disease detection that used electrochemical detection instead of the colorimetric methods favored by most people. Now Henry is instead developing paper-based exposure monitors for occupational health applications.
Among the devices his group has designed are paper-based systems that measure oxidative stress caused by atmospheric aerosols. During this summer’s Colorado wildfires, firefighters wore such paper devices to measure their exposure to oxidative stress from the fires.
His group has also developed paper devices to measure metal particles in occupational settings. The researchers collect metals on filters, punch out disks from the filters, and put those disks on paper devices where the metals are digested with acid and transported to detection reservoirs by the addition of water. The reservoirs contain reagents needed for the colorimetric detection of various metals.
Detection sensitivity has been a challenge, Henry says. In early experiments, an iron-phenanthroline complex formed in the test would develop a ring around the outer edge of the detection reservoir. “We always saw the same color,” Henry says, making it impossible to distinguish different amounts of metal particles. He and his coworkers have dealt with such problems by modifying the paper to make the tests more discriminating.
The number of potential applications of paper-based devices is growing. “I think of this as the next frontier of microfluidics,” Henry says. “When people started off in microfluidics, they said, ‘Here are 10 things you can do with it.’ Those 10 things became 100, and it really grew from there. The same thing is true of paper devices. More and more people are going to see unique applications. The field will continue to grow until it fills underserved niches as an analytical tool or a medical tool.”
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society