There is nothing in these tubes that should wear out. The liquid (dichloromethane, aka "methylene chloride") is a toxic (which is why Lionel won't make them again), non-flammable liquid that boils a little above body temperature. As heat is applied the temperature of the liquid near the bottom is raised to the point where boiling occurs. Boiling occurs when the vapor pressure of the liquid (which is a function of temperature) exceeds the pressure inside the tube. The main function of the crystals is to try and minimize thermal convention in the liquid so that most of the heat remains in the small volume underneath the solid. The bulb isn't hot enough to heat up very much of the liquid. Once the small trapped volume heats up to the boiling point, bubbles of dichloromethane vapor are produced (just like bubbles in boiling water just contain water vapor) which push the crystals out of the way and rise to the top. Cold liquid takes the place of the space the bubbles occupied, the crystals drop down and the whole process repeats. As the bubbles rise they cool as they pass through colder liquid. As they cool the pressure inside the bubble decreases until the pressure inside the bubble drops below the pressure in the "empty" space at the top of the tube and the bubble collapses. There is no chemical reaction involved and nothing to wear out except the light bulb.
Although pure compounds like water and dichloromethane have specific freezing and boiling temperatures (although boiling temperatures depend on pressure) they don't always freeze or boil when they should. That is why you need to be careful trying to boil water in a microwave. The water can superheat above its boiling point and then when you agitate it by taking the container out, the water can violently erupt since it is above its boiling temperature. Water will also supercool below its freezing temperature if left undisturbed. If you put bottled water in a freezer long enough that it should freeze, but doesn't, take it out and shake it and the entire volume immediately solidifies. This is why shaking these tubes once they have heated up will sometimes cause the boiling to occur.
To function properly the gas in the top of the tube needs to be colder (less pressure) than the liquid at the bottom. Sometimes putting your "ice cold" fingers on the gas space at the top will make them bubble. The greater the temperature differential between the gas at the top and the gas in the bubble at the bottom the better they work.
I have a number of American Flyer bubbling water towers and oil derricks which all have very long tubes. I have found that I can easily make them work if I tip them on their sides so that all the liquid drains to the top. The bottom has no liquid to remove the heat of the bulb so the heat goes into heating just the glass. After a minute or two I place them back right side up and the glass has now been heated hot enough that the liquid under the crystals is heated well above the boiling temperature and the boiling begins. If the lamp is strong enough the boiling will continue indefinitely. If the boiling stops after awhile, the bulb is not supplying enough heat to maintain the boiling.