Siting multiple magnets in one space allows for more interaction among investigators, but at the same time reduces the privacy that is sometimes needed to work productively. One solution is to provide both interactive space and space dedicated for solitary work, says R. Blake Hill, assistant professor of chemistry and biology at Johns Hopkins University in Baltimore, Md., which has three magnets in one underground room designed to eventually house five magnets.

“This technology is tremendously powerful and has applications that just five years ago were unforeseen,” says Hill. “Because of the great increase in these sorts of applications, we expect to see an increase in the siting issues and problems associated with housing these instruments.”

NMR stands for nuclear magnetic resonance while MRI stands for magnetic resonant imaging, explains Hill. Inducing a small magnetic field into a molecule, an animal, or a person produces images which reveal fundamental chemical and physical properties about the object. A functional MRI (fMRI) produces real-time images of on-going changes in the object, such as the human brain, during a specified activity. The instruments cost between $1 million and $5 million, depending on their strength, and can be more than 12 feet tall.

“That (fMRI) is an exciting application, because now we are starting to understand the basis of emotion, memory, and cognitive processes,” says Hill.

For example, researchers have been using fMRI images of the human brain to explore the differences between men and women when it comes to empathy. By tracking activity in the part of the brain activated by pain, they learned that women still emphathize for people they don’t like, while men do not. By observing the part of the brain that activates with a reward stimulus, they also learned that men enjoy seeing someone suffer if they believe they deserve it, while women do not, suggesting that men have a stronger desire for revenge than women.

They use similar techniques to measure the level of appetite reward a person feels when selecting certain products over others, such as Coke vs. Pepsi.

“It is fascinating that fMRI has this capability to reveal new insights,” says Hill. “This is partly why this technology is growing.”

In more traditional bio-medical research, high-strength magnetic instrumentation is being used at John Hopkins University to explore the structure of proteins that make up cells, in order to understand how their malfunction causes disease.

“That is an important first step in designing a molecule to block the activity of that protein that can be used in a therapeutic setting,” says Hill.

Facilities Requirements

These instruments pose unique facilities challenges because they are large, heavy, and highly sensitive to their environment. Because they emit a magnetic field—in an area called the five-gauss line, gauss being a measure of the magnetic field strength—they cannot be located near the ferrous metals typically used in beams, rebar, and other structural supports, yet structural stability is critical. Fiberglass mesh is sometimes used as an alternative in floor slab reinforcement.

“They need a remarkably stable environment,” says Stephen Bartlett, studio leader at Ballinger, a Philadelphia architectural firm which has designed imaging facilities for Johns Hopkins, the Children’s Hospital of Philadelphia, and Brown University in Providence, R.I.

The room temperature cannot fluctuate by more than 1º F, and the relative humidity must be maintained at 50 percent to avoid the possibility of accumulating any static charges. Maintaining a consistent temperature usually requires an independent HVAC system for the imaging lab.

“Typically what happens is that the campus will get a better price on overall energy use by changing their chilled water temperature over the course of the year,” explains Bartlett. “You have to have a dedicated system that is independent, so that you can guarantee a stable cooling water temperature to maintain environmental control in the space where the equipment is housed over the entire year.”

The instrument itself cannot vibrate, and neither can there be any ferrous material around it, so ductwork and piping must be made of non-ferrous materials with flexible connections. Movements from a small piece of ductwork close to the instruments in the ceiling or a subway train a quarter-mile away can distort the readings. Things with greater mass will cause interference at a greater distance.

“A classic problem is the elevator,” says Bartlett. “This large piece of metal moves up and down all day long. It could be far away in the building from the magnet but still cause interference.”

At Johns Hopkins, they even changed the landscaping to eliminate grass over the underground magnet lab in order to obviate the need to pass a lawnmower over it. The magnets also can distort each other, so facilities with multiple magnets need to site them at proper distances from each other. The magnets are then shimmed to create a baseline reading that accounts for the other magnets in the room prior to collecting experimental data.

It is important to remember that the magnetic field makes an ellipse around the magnet, not a simple sphere, says Bartlett. NMRs use a vertical axis and their elliptical magnetic fields are taller, while MRIs are organized with a horizontal axis and their magnetic fields are typical longer.

The five-gauss line defines the medical exclusion zone—the area people are not to enter without confirming that they have no metallic implants or pacemakers—but the magnetic field continues to diminish beyond that limit and its effects can have unforeseen results.

“At the University of Pennsylvania there are offices above the magnets. When they energized the magnets, all of a sudden the computer screens went out. That caused a huge political firestorm because these people didn’t even know a magnet was going in. The faculty above were in no danger, but were none the less alarmed. The investigator ended up buying them all LCD screens.”

The technology is constantly changing to mitigate this issue with various shielding techniques. The earlier magnets were unshielded, or “passively” shielded. They contained the magnet, which is a long coil, inside a thermos filled with liquid helium within another thermos filled with liquid nitrogen, with a very small zone in the center into which the samples were inserted. The unshielded magnet was protected from its environment simply by physically separating it from outside interference. The University of Pennsylvania constructed passive shielding by lining the entire magnet room with low-grade steel, which becomes magnetized and helps contain the magnetic field given off by the magnets.

Engineers have since developed “active” shielding technology, which adds another layer of thickness to the magnet. The extra layer creates an opposing magnetic field that cancels out the field given off by the magnet. Even though the magnet itself is wider, it requires a much smaller space because the magnetic stray field is closer to the magnet. Shields within magnets are improving such that an older one requiring a 255-sf room now can fit in a 100-sf space. Engineers are working to develop a magnet whose five-gauss line falls within the magnet itself.

Safety Concerns

The magnets pose several safety risks. First, the magnetic field itself must be avoided by anyone wearing a ferrous medical device, such as a plate or a pacemaker. A greater risk comes from the liquid helium and liquid nitrogen that are used to cool the magnet.

Liquid helium, for example, expands 740 times by volume if it is released from the instrument. Half the helium is released in the first minute, displacing the oxygen in the room.

“Because the magnets require less space with improved shielding technology, you can put them in smaller rooms,” explains Bartlett. “Therefore, you may have much more helium in the magnet than the volume of the room.”

For example, a 500 MHz UltraShield^TM magnet can be sited in a 224-sf room. With an 11-foot ceiling, the room will have a total volume of 2,464 cf. The magnet holds a total of 180 liters of liquid helium, which expands to 4,500 cf of helium gas when released. All oxygen in the room will be displaced if the gas is accidentally released, or quenched.

“There is this huge asphyxiation hazard during a quench,” says Jeff French, principal architect at Ballinger. “It presents an extraordinary hazard.”

To safeguard the users, standard safety features include:

• At least two oxygen sensors
• Exit door(s) that open outwards (overpressure)
• Lower-level exhaust vents, particularly in pits, for nitrogen gas, which is heavier than air and will settle rather than rise
• High level exhaust vents for helium gas which is lighter than air

Bartlett also recommends the installation of quench pipes, which are flexible pipes with flexible connections that vent gasses directly from the magnet out of the room in the event of an accidental quench.

“Our message is always have your quench system in, and put your oxygen sensors in, and don’t even think about making that the value engineering choice,” advises Bartlett.

Site planners need a unique tool kit to determine if a site is suitable for these fickle instruments:

• Vibration kit to measure the vertical and horizontal floor accelerations
• Fluxgate magnetometer to measure DC EMF (direct current, electro-magnetic field) perturbations caused by subways, elevators, and DC loads
• 60Hz EMF meter to measure AC (alternating current) fields emitted by transformers and power cables
• Radio frequency analyzer
• Temperature, humidity, and atmospheric pressure meter
• Voltage monitor

Accessibility Issues

Magnetic imaging facilities must be designed first to be accessible to the magnets, and then to the researchers using them. Researchers need to have access to the top of the instrument to insert samples and to install the cryogens that fill the cavity surrounding the magnet. That can be accomplished by either constructing a platform around the top of a larger magnet, or installing the magnet in a floor pit that brings its top to the level of the rest of the room.

The magnets are sometimes taller than the standard ceiling and weigh several tons, and they are difficult to install; they frequently cannot simply be wheeled across a typically reinforced slab floor onto the specially reinforced floor of the lab. Installation needs to be mapped in advance, and provisions made in the design of the building itself.

One solution is to design the lab with removable roof panels so the magnet can be hoisted in with a crane through a roof hatch. In order to maintain accessibility, the buildings sometimes are designed with removable concrete planks and transoms above doors.

For example, the 5,000-sf NMR Center in the Chemistry Building at Johns Hopkins, completed in 2003, is entirely underground. It contains two 800-MHz magnets in a pit about 5½ feet lower than the main room. Three 600-MHz shielded magnets are on the room-level surface. Engineers were able to manipulate the grades to gain access to the room directly from a loading dock, obviating the need to lower the magnets in from above. The magnets still required extra support in transit, so the architects built rebar hooks into the ceiling to move them along horizontally.

“If you just think it through, it is a lot easier for these guys to work with later on,” says Bartlett.

In planning for the Johns Hopkins facility, architects and engineers from Ballinger visited the NIH Building 50 in Bethesda, Md., a facility which also dealt with many similar issues.

The MRI center in NIH 50 is underneath a plaza, outside the building envelope and completely underground. Its magnets were installed through a roof hatch, and the five-gauss line is roped off with a chain. On either side of the room are air vents with a lot of diffuser capacity to respond quickly but sensitively to changes in temperature. The system maintains a constant temperature but does not create wind currents moving air across the magnet surface.

“Particular to this facility is that it has no pit,” explains Bartlett. “The entire floor is constructed of aluminum grating on aluminum I beams. Below this grating level is the magnet level; the entire slab is floating on inner tubes. Installed in the ceiling is a gantry crane to move the magnets around the room once they are lowered in.

“It is a remarkably vibration-free facility,” he says. “Usually when we say ‘floating slab,’ we mean it is isolated from the rest of the building perimeter by a flexible joint. In this case they literally floated the slab.”

Accessibility was particularly challenging at the Children’s Hospital of Philadelphia, a four-million sf complex with 2.5 million sf of research space. Three floors are below grade, the bottom-most one containing the 1,500-sf MRI lab for animals, where removable horizontal panels in the floor and removable transoms above the doors provide access for multiple magnets.

“There are likely to be at least two major fMRI units, and not necessarily on the same floor,” says French. “This is an important project because of the sheer size of the building underground.”

Brown University is constructing a 170,000-sf facility for neuroscience, molecular and cellular biology, and biochemistry, which will service human and animal subjects in the same fMRI unit. The facility also houses a range of MRI magnets of varying strengths. A copper wire mesh is embedded in the walls, ceiling, and floor to provide radio frequency shielding, which protects the magnets from extraneous or external interference. Additional magnetic shielding—a silicon-encased steel plate inserted only where the five-gauss line extends beyond the boundaries of the room—protects people and objects from the magnet.

The equipment at Brown is so expensive is that it is called upon to do double duty with animals and humans, necessitating protocols for disinfecting the equipment and transporting the subjects in such a way that their paths don’t cross.

Get a Second Opinion

As sensitive as magnetic imaging equipment is, it is possible to develop creative solutions to protect it. That sometimes requires asking advice from another vendor if the first seems intractable.

“Manufacturers have their own view of the world,” says French. “There is a long list of things they require; protocols about the performance of their magnet. The protocols may be significant, but optimum performance is sometimes possible without the rigid or severe protocols that the manufacturer demands.”

Some will say, for instance, that you have to have fiberglass reinforcement in the floor slab. But what if you are installing the magnet in an existing structure with steel rebar?

“This is a negotiating point,” says French. “They are simply writing the idealized script.”

By Lisa Wesel

We welcome your Questions and Comments

Copyright 2008 Tradeline Inc.
All Rights Reserved
ISSN: 1096-4894