So are these loose cells you have?
You will require a voltage limited and current limited source to charge your li-ion batteries; something such as a lab supply will do nicely but a universal supply will probably do aswell - check its specs tho!
You'd probably want to charge the cells to a max of 4.2V (Check datasheets or go lower to be safe with 4-4.1v but you'll be storing less juice tho!).
Set your current limit conservatively to c/3 or something, c is the capacity of the cells so if you had 2400mAh cells, charge them at 1/3 of that so at a rate of 800mA each. Most cells let you do 1C and few do 10C or other high charge rates but best to be safe and sure when working with these and use low currents.
If you have a whole bunch of these cells in series, you'd want to be careful because of the imbalance that may occur as Thomas pointed out. This isn't an issue with single cells however.
Also for the battery monitoring chips - they'll disconnect your battery when the voltage / current are beyond safe limits but you will still need proper charging power source as the BMS will not limit current for you.
The in-battery protection circuitry is usually intended to act as a gross fault protector and it is strongly recommended that it not be relied on as a means of charging control. As a means of gross short circuit protection it may be suitable as long as the values they choose for max Iout are acceptable to you.
For charging, use of one of the large number of LiIon charger ICs is recommended.
A major factor is that the over voltage circuit does not remove the applied voltage when the CC charging current falls to a low value. This means that the battery is "floated" indefinitely with the risk (I'm told) of plating out metallic Lithium.
A PV panel (solar panel) that is nominally 12V rated and intended for charging lead acid batteries, will have a loaded Vout of about 18V and an O/C or light load Vout of over 20V.
The maximum voltage that you need AT the battery pack is 4.2V/cell or 12.6V in your case.
PV panel available Iout values are a reasonable approximation to being linearly related to isolation (sunlight level).
However, Vout is not related to light level in the same way. A PV panel will produce over 90% of its full power voltage for light levels of a few % of maximum and above - say at 10%+ to be safe. If you want the panel to charge the battery to fully-charged even on a low sun day, if necessary, then you need a panel that is full load rated at at least 12.6V/90% = >= 14V. As above, as an SLA targeted 12V panel makes about 18V at full-sun full-load, such a panel will provide more than enough voltage under all practical light conditions.
You will get substantially longer cycle life from a LiIon cell if you terminate discharge at a slightly higher voltage than allowable absolute maximum. With LiIon , below about 3V under medium loads you have used the large majority of the stored energy.In-battery low voltage cutoff circuitry will probably allow discharge to about 2.6V/cell, which is lower than is wise for good battery lifetime.
Best Answer
You'd probably be better off taking a bigger spare battery.
But if you really want to charge using solar energy, you have to understand that there isn't much. If you have a 3cm^2 solar panel and go on the basis of 15% efficiency for the solar cell (which is pretty good), then plug that into the solar energy per square meter on Earth, which is from 1,413 to 1,321 W/m^2, which gives you about 59mW. Your battery is 3.4Wh, so it will take over 2 days to recharge it, nevermind that the sun is only around for a fraction of that or conversion efficiency.
Working backwards with those numbers, if one wants to charge a 3.4Wh cell within 4 hours, they would need a
(3.4Wh)/(4h)/((1321W/m^2)*0.15)=43cm^2solar cell.