Sunday, December 30, 2012

Sometimes you need to tell yourself to just go for it.

I'm naturally introverted and a bit of a perfectionist and there have been several times in my academic career when I was forced to overcome my anxieties, from difficult exams to public speaking. Often the only way to get through these uncomfortable situations was to just suck it up and try my best.

There's a saying that goes: The things that make you nervous in life are the things you should be doing. Often the most nerve-wracking experiences end up being the most rewarding.

If you're nervous about anything in life, just channel the courage of this 4th grade girl taking her first ski jump. (This is several months old but I still love it.)

"Here goes something... I guess..."

Thursday, December 27, 2012

If you ever wanted to know about laser diodes

Nature Photonics wrote up a nice little article about the 50th anniversary of the laser diode last month, not to be confused with the 50th anniversary of the laser which happened in 2010.

I know a lot about laser diodes. I sometimes feel I don't know nearly enough to be considered an expert, but I did co-author a book chapter and write a 210 page dissertation on my 5+ years designing and fabricating visible wavelength laser diodes, so I guess you could say that qualifies me to speak intelligently on the subject.

Still, when people ask me about my dissertation research, I struggle to explain it every time. Not many people know what a semiconductor is or what a laser diode is, or worse, they think they do but they don't. Laser diodes in particular are a source of confusion.

The problem is that in order to understand my research project you need to understand some basic principles on solid state physics and crystallography, in addition to knowing what a laser diode is. They're not exactly simple concepts.

In honor of the golden anniversary of the laser diode, I will finally attempt to explain what they are in simple terms. Behold!

What laser diodes are NOT

Most people hear the word "laser" and picture a huge death ray like in Goldfinger.

Still from laser scene in Goldfinger where Bond nearly nearly gets vaporized junk-first via

They imagine I sit in a dimly lit lab, shooting high power laser beams all over and taking scientific measurements on them all day.

High power laser in "Military laser experiment" via Wikipedia

Once in a while, some wise-guy will make an Austin Powers reference about putting them on sharks. This joke got old really fast.

Alternatively, if they were alive in the 80's, someone might make a Real Genius joke about working with Lazlo or making popcorn, but this joke doesn't bother me as much because I love this movie.

"Real Genius" laser lab, via AMC blog

I used the word "optoelectronics" once to describe my area of research, which is a general term describing electrically-injected light emitters like laser diodes, and someone assumed that meant I worked on robotic eyeballs. Like in The Terminator, perhaps?

Arnold's robot eyes may be included under the umbrella of optoelectronics since it probably includes a photodiode. IMG via 13point7billion

To make things easier, most of the time I don't bother mentioning laser diodes at all and just tell people I work on LEDs instead. I occasionally work with LEDs and have many coworkers who specialize in them, so I happen to know a lot about them. Plus, there's that huge benefit of most people having a general idea of what LEDs are and what they're used for. My Grandma might think I make Christmas lights for a living, but I just decided that it's close enough.

LED lights via Apartment Therapy

What laser diodes are

They're really small.
You literally need a microscope to see them. Typical dimensions of the lasers I work with are about one millimeter long and less than fifty microns wide, about the width of a human hair.

Here is a picture of a generic laser diode chip on the head of a needle.  The actual laser is a very thin stripe of metal running vertically down the center of the chip. You may need to view the image at full resolution to really see it.  via Wikipedia
Packaged green laser diode from European lighting giants OSRAM via Semiconductor Today. The rectangular chip is emitting green light, although it's difficult to see the actual laser beam. The wire bonds are used to inject current through the device.
Fully packaged blue laser diodes from Sanyo. The scale is in mm.  

They're basically a glorified LED. 
Light-emitting diodes (LEDs) and laser diodes both have the essential features of diodes, which means they flow current when you apply a voltage.

To be specific, the LEDs and laser diodes I work with are p-n junction diodes. For these kinds of diodes we use a process called doping to turn a semiconductor material into n-type cathode material (more electrons) and p-type anode material (less electrons) and put them together like a sandwich.

Here's Collin Cunningham of MAKE magazine talking about the history of diodes, including the p-n junction diode.

In addition to being diodes, LEDs and laser diodes both produce light via similar magical mechanisms of quantum physics. This magic is contained in a layer in the middle of the p-n junction that has the ability to emit light. This light-emitting layer is also called the active region.

Apply a voltage to a regular diode, the current flows from anode to cathode. But if you insert this light-emitting layer in the middle, when the electrons try to flow through it they magically transform into photons. (In fact, this layer simply traps the electrons and they emit light as a way to lose excess energy, but that's a whole other story for another time.)

One obvious difference between LEDs and laser diodes are their dimensions at the chip level. An LED chip or "die" is usually larger, square, and mostly emits light from the top.

Generic square LED. Current flows from anode to cathode via a transparent contact and wire bond on the top of the sample and a contact at the bottom of the sample (not shown). The light generated in the magic light-emitting layer (a.k.a. the active region) and shines out the top of the sample. LED sizes vary but they are usually less than a mm wide.    
On the other hand, a laser diode is long and skinny and emits a concentrated beam of light from the edge.

Generic edge-emitting laser diode. Like an LED, current flows from anode to cathode via a skinny metal contact and wire bond on the top of the sample and a contact at the bottom of the sample (not shown)The light generated in the magic light-emitting layer (a.k.a. the active region) shines out of the edge of the sample (technically both edges). In this example, only the area under the narrow metal contact of the anode generates current to produce the light which forms the laser beam.
Keep in mind there are MANY different configurations of LEDs and laser diodes and I'm glossing over some other details, but these are just generalizations for simplicity sake.

They have all the necessary components of an actual laser. 
That is to say that although laser diodes are small, they are still technically lasers and produce a tiny laser beam of light via the same mechanisms that all lasers operate.

A few essential features of a laser:
  • It is made from a material that can generate light via energy supplied from an external source (such as an applied voltage which causes current to flow through a magic light-emitting layer, in the case of a laser diode).
  • The light is trapped in a cavity that allows the light to bounce back and forth inside.  
  • Light begets light.

The reason lasers are long and skinny is because the light inside only needs to bounce back and forth in one direction. In our lasers, we trap the light in this long, skinny cavity by forming a thin ridge of metal via photolithography and coating the ends (or facets) with partially reflective mirror material.

If you trap enough light in this long skinny cavity, the material within it-- which is already generating light-- can absorb more energy from the light that it itself created. This is how light begets light. When light begets light you have gain. This gain process is also called stimulated emissionIt's something laser diodes do and LEDs do not.

Henry at Minute Physics explains gain & how lasers work

When you have this gain process happening within a laser, a funny thing happens to the light. This is the magic of lasers. The light gets really narrow in shape, forming a beam, and narrow in spectral width, which means it's all the same specific wavelength or color. Without the benefit of gain and stimulated emission, LEDs produce a much broader color output-- so an LED giving off blue light also gives off a little violet and green as well, but laser diode giving off blue light is a really specific color blue.

Here's Bill Hammack of the Engineer Guy describing how a ruby laser works.

To summarize:
LEDs and laser diodes both flow current when you apply a voltage, and both magically produce light when the current flows through a layer called the active region. A laser diode is long and skinny and when enough light builds up, it magically forms a laser beam.

The "magic" is actually quantum physics and well understood. If you're interested, I may post about it in the future. But, to be blunt, if you're really into this stuff perhaps you should have taken more physics in school. :)

What are laser diodes used for?
OK, so we have some idea what they are and how they work, but why do they exist?  Well, we can thank laser diodes for the following things:
  • Fiber optic data transmission aka THE INTERNET - all your internet is transferred to your computer via laser diodes pulsing light through fiber optics all over the world, even under water, across oceans. To this day, that blows my mind. Even if you're on wireless network, your internet first goes through a physical optical fiber to a router which then rebroadcasts it. (Watch Engineer Guy explain how fiber optic cables work.)
  • CD and DVD technology - use red laser diodes to read and write data on disks.
  • Blu-Ray technology - uses violet laser diodes to do the same thing, but since it has a shorter wavelength than red, that allows higher density of data per disk
  • Laser printing like LaserJet and digital photocopiers - uses IR (invisible) laser diodes to charge a rotating drum which then attracts toner/ink and rolls it onto paper.  
  • Bar-code readers - the thing that goes bloop when you buy things in a store.
  • Laser pointers - do NOT shine them at airplanes.
  • Laser-based projectors - you could use red, green and blue lasers to produce full color images that are always in focus. They are small and lightweight, you could even embed them in a cellphone.

What my dissertation was about
I grew the semiconductor layers (the anode, cathode, and magic light-emitting layer) and then fabricated them into tiny lasers and tested them in a lab. I spent a lot of time improving the growth, design and fabrication of the lasers, which meant working in several labs, including a lab where we grow semiconductor material via a process called metal organic chemical vapor deposition as well as a cleanroom to form the ridge and metal parts. I did not sit in a room full of lasers all day, although once in a while I sat in a room with a microscope and tested really really tiny ones.

The whole point of my project was to show that the particular properties of the particular semiconductor I was working with could make more efficient laser diodes. Oh, and my lasers happened to be blue.

Fun fact: This blog is named "My Laser Boyfriend" after an old joke about spending too much time making lasers and not enough time dating. But such is grad school. 

Wednesday, December 19, 2012

Organic LEDs

Lighting giants Philips showing off some OLED panels, part of a new product line called Lumiblade.

What is an OLED?
OLED stands for organic LED, which uses carbon-containing organic materials as the light-emitting regions of a diode, as opposed to a regular inorganic LED which uses solid semiconductor layers throughout.  The organic materials may consist of small molecules, phosphorescent materials, or electroluminescent polymers, such as polyfluorene, which have the ability to emit light when injected with electrical current.

Organic layers underneath a transparent cathode from HowStuffWorks

To be honest, organic chemistry is completely foreign to me, but the essential features of these materials is that when electrons get inside them and then lose energy, that energy takes the form of visible light.

Advantages of OLEDs
A benefit of OLEDs is that the light-emitting organic material may be coated onto large areas and flexible surfaces. Since some of these organic materials come in liquid form, the coating and processing could be done quickly and easily through processes such as inkjet or screen printing.

OLEDs on flexible substrate, via GadgetVenue

Also, the light generated from an OLED is naturally diffuse and evenly distributed across the entire surface. For regular LEDs, the same result would require many individual LED elements plus diffusers or other optical elements to spread out the light.  

Because the OLED panels may be scaled to size, and because they are so thin and lightweight, it makes them ideal for portable device displays such as cellphones and digital picture frames. Samsung employs a matrix of RGB OLED arrays for its Galaxy phone line, which they call AMOLED (active matrix organic light emitting diode).

Nexus One smartphone AMOLED via Wikipedia

Disadvantages of OLEDs
The biggest drawback of OLEDs is they are less efficient compared to inorganic semiconductor-based LEDs, don't last as long, and are still very expensive to produce. Because of this, they will likely never replace regular LEDs for most lighting applications.

But hey, they sure do look cool!  Just ask the Black Eyed Peas, or these guys:

OLED Xmas tree from the Advanced Technology group at GE 

You can buy Lumiblade panels here.
If you're really DIY, play with your own monochrome OLED display from Adafruit.

Monday, December 10, 2012

Festival of Light Emitting Diodes

Approximately one year ago, I was in Lyon, France to present at Forum LED, a European lighting conference that attempts to bring together science and industry of the LED world.  It turned out to be heavy on the industry and light on the science, so I felt a bit out of my element.

My session chair emailed me the day before my talk to suggest: "I am still wondering if you should not alleviate the scientific content on your talk? Don't you think?" I did, and it was probably a good exercise to lighten my usual research-heavy presentation for a more general audience. It went fine. The conference was interesting overall and gave me the chance to learn some of the challenges for implementing LEDs into the real world.

At my advisors suggestion, I allowed myself time for extra travelling before and after the two-day conference, which went surprisingly well considering I don't speak a lick of French. At times, traveling solo felt humiliating and lonely, but for the most part I enjoyed tromping around on my own and sight-seeing, especially since the LED conference happened to overlap with Lyon's Fete des Lumieres, or Festival of Lights. Which, happily enough, employed lots of LEDs!

These videos give only a small taste of what the experience was like. All over town, buildings are lit up at night with lights and projected animation shows. They closed several streets from traffic to allow thousands (millions?) of visitors to wander around and gape at all the displays. All in one weekend in December. Pretty amazing!

Other lessons I learned on this trip:
  • Look up the maximum hotel per diem rate before booking a room at the conference hotel.
  • Dragging a roller-suitcase through cobblestone is really tiring.
  • If it's going to be raining, bring leather boots and lots of extra socks.
  • Bubble baths RULE, especially when you've done said tromping around in said cobblestone in said boots and rain all day.
  • Jet lag sucks. 
  • 24 hour room service RULES.
  • Trains ALSO RULE.
  • When you're at a conference and don't know anyone there and are socially awkward, it's ok to skip the official conference banquet and take yourself out for a nice dinner instead.

Sunday, December 2, 2012

My Laser Boyfriend's Grad School Survival Guide: How to make the most of the worst* four to seven years of your life

* not actually the worst although sometimes it can feel that way.

“Knowing What I Know Now” 

Surviving grad school was the hardest yet most rewarding thing I've ever done. I pushed myself harder than I ever imagined and experienced some of my lowest lows, but I also had some amazing opportunities, learned a lot, and grew more confident as a scientist and researcher. I also drank. A lot.

Although I had to endure what felt like utter misery at times to get where I am now, going to grad school was still the best decision I ever made.

This fall, my research group took on several new grad students. Seeing them with that "I have no idea what's going on" look on their faces makes me remember how hard it was to come to grad school, not knowing what to expect, and feeling very overwhelmed and out of place.

Since I noticed other science bloggers writing advice for those at younger stages in their career, and I'm at a point in my career that I have some to give, I thought I'd compile some tips of my own for those embarking on an advanced degree in STEM.

Prepare to feel stupid (again). 
Remember the end of high school when you thought you were so smart for getting into college, but then you actually GOT to college and realized EVERYONE was smart, and you had to work your ass off just to stay slightly ahead of the curve? Yeah. Grad school is going to be a lot more of that. But worse.

For me, that first year of graduate coursework was brutal. 

It's good to keep in mind everyone in your graduate program comes from different backgrounds and has different abilities, so don't feel down on yourself if you're not as strong in some areas. (See section on "Impostor syndrome..." below).  This can be especially difficult of you switched majors between undergrad to grad school.

If the graduate coursework is slowing you down, consider take a few undergrad-level classes to refresh on prerequisites. Also form study groups (see section on "Making friends.." below). Also realize that once you're into the Ph.D. program, your actual GPA may not matter as much as it did in undergrad, although this may depend on your department. Often, the publishing and research aspect of your graduate degree is much more important.

When you join a research group it's easy to feel overwhelmed if you don't have hands-on experience. This can be totally OK as long as you're up-front about your abilities and get properly trained. See section on finding a lab mentor below.

Know what you're getting into.
This is basically impossible, but try. Talk to as many people and do as much background research on the department and research group as you can. Do NOT depend on their website for information on current projects or even current group members-- they are often horribly out of date.

Visit the campus, meet with your potential PI one-on-one to discuss possible research projects and funding opportunities, but also ask to get a lab tour from a postdoc or current grad student, then pick their brain for more details. Ask questions like: what's the reputation of this PI compared to others in the department; how involved is the PI in the goings-on in lab; how is the group hierarchy organized; who provides training to new students; what are the expectations of group members in terms of balancing classwork, lab-work, and publishing papers; are there weekly meetings; will you get to travel to conferences, etc.

Know what will or will not work for you. 
...especially in terms of your potential research group and how it operates.

Be aware of how much direction you need and how much you should expect to get. My PI was very hands-off.  I've seen independent, self-motivated students work hard and be very successful with little input, while many other students flounder without consistent feedback and encouragement and end up wasting a lot of time being unhappy and unproductive. On the other hand, I've known other PIs who are overly demanding and constantly meddle, so students who prefer to work independently and at their own pace do not do well under those conditions either.

The grad school lifestyle varies depending on school, department, and especially the PI in charge. If you're older, married, have a family and can't work weekends for example, make that known up-front.

Make friends with other new grad students right away. 
Form study groups with classmates to get through impossible homework sets together. You will naturally bond over the shared experiences of painful grad classes and the awkwardness of being the new kid in your research group. There's a good chance most of you moved from far away and won't have a social group yet, so it can be fun to organize group outings to see more of the town, even if (especially if) this mostly entails beer.

Over the course of your grad career you may eventually lose touch with your friends from your first year, especially if they work in a different lab as you, but they may come in handy down the road when you need to borrow some lab equipment or need advice on something they're expert on. Networking works!

Find a lab mentor ASAP. 
When you first join a research lab, you won't know how to do anything. That's okay. That's expected. Your job in the beginning is to learn and ask questions. You may need to turn to an older grad student, postdoc, or project scientist for advice and specific training. Some groups are better at coordinating this than others, so ask around.

Don't be discouraged if senior grad students don't have much time to give you, especially if they're close to graduating. If you can't find one person, try to reach out to several at once.

Believe me, you will need to earn the trust of the senior researchers in your group. Definitely get on their good side. Make appointments with them, show up on time, ask a lot of questions and write everything down, especially when they train you on equipment. The quicker you learn, and the more effective you are at completing tasks they give you, the sooner you'll get a good project. Your PI trusts these guys, and believe me that they can determine your fate. I've seen it happen.

Be nice to the lab techs and engineers.
They may not have a Ph.D. behind their name, but they have been there longer than you, and know more than you will ever know about the equipment and how things operate. The lab would not run if it weren't for them, so always do what they say.

If you break something, which you will, or have a problem in the lab, which you will, report it right away. Then thank them for fixing it. The more respectful you are to the technicians, the more helpful they may be in the future when you encounter another problem. They can also be a great source of gossip and hilarious stories of other grad students doing dumb things in the lab.

And you know what? Being friendly with technicians is a really important skill to have if you ever get a job in the real world. Because these are the guys you'll be working with every day in industry.

Don't let the grunt work drag you down. 
Thomas Edison said: "Genius is one percent inspiration, ninety-nine percent perspiration." A good research project will not necessarily fall in your lap and you may need to earn your stripes doing some bitch work for a while. Be patient. This is how things work in grad school.

Patience is definitely an important virtue. Even when you have your own project, parts of it will be slow and painful and you'll start wondering if this what you really want out of life. Relax. This is science. Sometimes it is boring and soul-sucking, but it will not be like that forever. (See section "Keep Your Eye On The Prize," below) 

Impostor syndrome is a real thing.
There may be times when you feel overwhelmed and that you don't belong. Most people go through this at some point or another. What separates the women from the girls is finding a way to be self-motivated even when every fiber of your being wants to give up.

I often think of Woody Allen quote: "80 percent of success is showing up." When it gets hard, call mom, have a good cry if you need it, but then suck it up and keep going. Sometimes you might feel like you have to work a little harder to keep up with everyone else, and if that's the case, do it. Do it like you have something to prove.

It's important to remember that you were smart enough to get into grad school, you deserve to be here. Besides, no one said it would be easy. If grad school was easy, they wouldn't call us a doctor afterwards. Right?

Learn to budget your time. 
Find a way to manage your schedule. I use Google calendar and have separate calendars for different purposes, even a shared group calendar for meetings and seminars I share with members of my lab. I was so busy by the end of my grad program that I'd forget to make time for lunch if I didn't specifically schedule it in. I rely on my calendar for my daily to-do list because I found I'm more likely to remember to do something if it's in my calendar, even small stuff like "email back collaborator" or "submit purchase order."

Be organized and consistent. 
Come up with a system for naming and storing samples, lab notebooks, and data files, and stick with it. Consistency is key here. If you have a good organization system, it will be no problem find a sample or data quickly. For me it became really useful when I was writing a paper or my dissertation and had to dig up old data to replot or old samples to remeasure.  It was also useful when a younger student or collaborator had a use for some samples I previously thought were worthless.

I learned some lessons the hard way. Every time I got lazy and made a short nickname for a sample instead of using the full sample name, I'd end up regretting it later when I needed to look up the data or find the sample again. Every time I was too busy to file samples away properly and let them pile up on my desk, I'd be totally overwhelmed with the task later. If I didn't type up what I had done in the lab, such as some data I took, I'd forget it even happened.

Write down everything. Take detailed notes in your lab notebook and then transcribe them into a spreadsheet on your computer afterward. This sounds ridiculous but you won't believe how much data piles up over the course of a few years.

Put some effort in the beginning to makes things easier for yourself.

Keep an outline of your dissertation on a file somewhere.
In the beginning you may only have a vague idea of what the title will be. Fill that in first. Check back every few months. Add background information. When your research project becomes better defined, you will be able to add chapter titles, and then subsections. It is a good way to track your progress and can eventually serve as a to-do list. This is also useful when your research project is at a stand-still and you need busy work.

Keep an updated CV/resume on a file somewhere.
This is satisfying to update every time you publish or co-author papers, or present at a conference. It's also fun to work on whenever you're feeling completely disillusioned by grad school and day-dream about getting a real job.  :)

Back up your computer. 
Use online storage like Google Drive or Dropbox in addition to an external harddrive that does automated backups. Seriously. You can not afford to lose that shit. Also, get a good anti-virus software and make sure it's up to date. Our lab computers are always infected with some weird virus transmitted through flash thumb drives, so I'm super careful about running automated scans on my stuff. You can never be too careful.

I speak from some experience here, as I lost my entire harddrive due to a nasty virus during my second year of research. It put me back a whole month. 

If you see something that needs fixing, fix it. 
Sometimes it's up to a grad student to step up and volunteer to update the contact information on the group website, or form a journal club, or decide to clean the lab. Be proactive. Don't be one of those guys who sits around and complains about the state of things but never does anything about it. Nobody likes that guy. If something bothers you, change it. If you can't do it yourself, find someone who will. Sometimes being a productive researcher is knowing the right people to talk to in order to get things done.

Grad school has its ups and downs, make the most of it.
Research is especially up and down. If you're in a lull period or otherwise being completely unproductive in the lab, use the opportunity to catch up on journal reading or cleaning your desk. Or, fuck it, just go home and use your time to do something useful there like grocery shopping or laundry. Catch up on life stuff. When you're in a major crunch period and pulling long hours, which will happen, you won't have time for life stuff. Go to the store and stock up on soup and trail mix for those times you get sucked into studying, writing, or lab-work and don't have time to go shopping.

Know your limits, mentally and physically.
It's easy to lose touch with the most basic needs when you're busy and under a lot of pressure. It's important to stay productive when you need to be, so do everything possible to foster an environment that helps you towards this goal. Know yourself. If you get easily cold, keep a sweater in the lab. If you're irritable when you're hungry, keep snacks in your desk drawer. If you're sleepy and unmotivated, go grab a coffee. If you're stressed and antsy, take a break and go for a walk, or take up a soothing hobby like knitting. If you need 9 hours of sleep every night and exercise every day, make time for that.

If you're overwhelmed, anxious or depressed and it's affecting your work, consider counseling. Many schools have mental health services filled with people whose job is to offer helpful advice when you're going through a stressful period. I've been to counseling off and on during grad school and it helped a lot (a bad break-up; preparing for, failing, then preparing again for my Ph.D. entrance exams...). It's not an admission of failure to seek help, it's being proactive. 

Side-note: I've found not all therapy techniques are useful. I've gained a lot from cognitive behavioral approach, which suggests that negative feelings of anxiety and depression are related and stem from negative self-talk and distorted thinking. I learned a lot from examples and exercises in the Feeling Good Handbook by Dr. David Burns. (Sorry to sound like an advert, this just worked for me!)

Don't compare yourself to others.
Don't compare yourself to the golden-boy grad student who manages to crank out way more papers than everyone else, and especially not those friends of yours who got a great job after undergrad, are earning tons more money, getting married and having babies. Whenever you compare yourself to others, especially those buying BMWs and posting pictures of it on Facebook, you will feel depressed and it will seem like you're wasting your life inside a research lab.

Don't lose your sense of self in the bubble of grad school. Trust me, there's nothing like graduate school to make smart, motivated people feel stupid and lazy. From the outside, you're doing great things. Make some non-grad student friends or consider getting involved in outreach opportunities in the community to help you gain perspective.

Keep your eye on the prize. 
Don't forget the whole point of grad school is to graduate. If you feel you're not making progress towards this goal, take a step back and assess the situation. If you feel helpless, realize it's okay to ask for help. You shouldn't be expected to be a perfect scientist, you are here to learn to become one.

Are you stuck? Is your stuff simply not working? Are you waiting on something completely outside of your control that's preventing you from making progress and it's taking so long that you're going out of your mind with boredom? Maybe you need to back up and approach it in a new way. Talk to your advisor about starting a side project or collaborating with someone new.

Did you get wrapped up in a side project that's been distracting you for too long and you aren't making any progress on your main project? Figure out a way to put it on the back burner or delegate it to someone else so you can focus your energy on more important things.

Be nice to the first year grad students. 
Mentor them. Advise them. Train them. Take advantage of their naiveté and give them some of your busy work. Remember your first year as a grad student , and how happy you were to finally do something on your own in lab? Don't think that spending time training others is a huge inconvenience, they may end up being really helpful. Don't pity them, either. It's the circle of life. Grad school life. (Whatever that means!)