410 Makromol. Chem., Macromol. Symp.

61,410-431 (1992)

MOLAR MASS DETERMINATION BY END GROUP ANALYSIS - AMINO

GROUP DETERMINATION I N ALIPHATIC AND PARTIALLY AROMATIC
POLYAMIDES WITH 1 -FLUORO-2,4-DINITROBENZENE IN BENZYL
ALCOHOL^

Volker Rossbach', Gerhard Leumerb, Sigrid Bohme

Institut f u r Technische und Makromolekulare Chemie. Universitlt

Hamburg, BundesstraDe 45, W-2000 Hamburg 13, Germany

Abstract: End group analysis is t h e m o s t important absolute method

of obtaining number-average molar mass values for polymers of t h e
step-growth polymerization type. A great number of methods a r e
available which primarily differ in t h e detection of t h e end group
concerned. Special chemical methods based on t h e formation of
covalent derivatives of t h e end groups (e.g. introduction of UV/VIS
chromophores) a r e important if there is a need t o demonstrate, with
t h e aid of gel permeation chromatography, t h a t damage to t h e
polymer (e.g. onset of cross-linking) has occurred.
In t h e case of fibre-forming polycondensation products, t h e
applicability of end group analysis is often restricted by t h e lack of
suitable reagents and solvents. This is also t r u e of t h e
determination of amino end groups i n polyamides by means of 1-
fluoro-2,4-dinitrobenzene, as is already contained i n t h e German
Standards specification (DIN 54 274) - albeit only for polyamide 6
and 6.6. This contribution describes a modified version of t h e 1-
fluoro-2,4-dinitrobenzene method. The analytical principle and
practical procedure correspond to'those of DIN 54 274, but t h e
solvent used for t h e reaction with t h e reagent is benzyl alcohol,
and t h a t for t h e photometric evaluation is 1,1,1,3,3,3-hexafluoro-2-
propanol. The scope of t h e method is t h u s increased. Examples of
applications for this method a r e commercial products based on
aliphatic polyamides (6; 1 1 ; 12; 6 , 6 ; 6,12), partially aromatic
polyamides (2,4,4-trimethylhexamethylenediamine,terephthalic acid;
copolyamides with bis-(4-aminocyclohexyl)propane(2), isophthalic
acid, o-aminododecanoic acid) and multicomponent systems with t h e
abovementioned polyamides (bicomponent fibres, blends). I t is shown
t h a t large differences may arise between t h e value determined by
titration and t h e actual amino end group content. In addition, t h e
presence of secondary amino groups can be demonstrated in
polyamides of t h e Trogamid T type.

Note a. An abridged version w a s published i n Melliand Textilberichte 72,

1026 (1991)
Note b. New address: Hoechst AG. Werk Bobingen, Germany

0 1992 Huthig & Wepf Verlag, Basel CCC 0258-0322/92/$ 04.00

 41 1

INTRODUCTION

End group analysis of polycondensation products

Number-average molar masses may be determined by absolute methods which
determine t h e number of molecules present. Direct analysis of a specific
chemical group in a polymer of known structure is one way of doing this. In
linear polymers of t h e step-growth polymerization type, such a s polyamides.
polyesters and polyurethanes,, one or both ends of t h e polymer have specific
groups which may be determined by suitable means, using t h e following
equation for t h e calculation:

R, = n x 106/ m,
where
n = 1-2 is t h e number of groups t h a t can be determined per
macromolecule,
m is t h e end group concentration in pmoles per gram.

In some cases end group analysis is applied t o determine R n values of

branched polymers obtained from multifunctional monomers, calculated from t h e
following equation:

R, = 2m x 1061 ( m - y),
where
m is t h e t o t a l concentration of end groups in pmoles per gram,
y is t h e number of branched points in pmoles per gram.

The sensitivity of t h e end group analysis varies greatly according t o t h e

method used to detect the functional group in question; Table 1 provides a n
illustration of this.

Tab. 1. Maximum R n values t o be determined by end group analysis (Ref.1)

detection method maximum A n value (g/mol)

titration 40 000
microanalysis of heteroelements 100 000
radioactive labelling 200 000
addition of a chromophore 1 000 000

The classic procedures for determining t h e end groups of polymers obtained by

step-growth polymerization a r e potentiometric, conductiometric and colorimetric
41 2

titratjon. If no end groups are present which can be ionized and hence
directly titrated, titratable species are created by polymer-analogous
reactions. A recent example of t h i s is t h e reaction of hydroxyl end groups i n
poly(ethy1ene terephthalate) with o-sulphobenzoic acid anhydride and t h e
titrimetric determination of t h e resulting sulpho groups ( a s well a s carboxyl
groups) (Refs. 2 & 3 ) .

An extremely great number of descriptions of titrimetric procedures with

slight variations (solvent used for polymer, temperature, bases or acids used,
etc.) can be found in t h e literature (Refs. 4 & 5).
This in itself is an Indication t h a t t h e methods are not generally regarded a s
satisfactory: if strictly standardized analytical conditions are used they do in
fact provide reproducible values for t h e end group content which a r e e.g. well
suited t o examining the constancy of t h e product characteristics; but in m o s t
cases, for reasons t h a t still remain unknown, they do not produce correct
values when used t o calculate t h e molar masses.

Radioactive labelling and microanalysis of heteroelements (cf. Tab. 1) have

been used time and again for special problems in end group analysis, but
have yet to establish themselves a s part of t h e routine analysis of
polycondensation products.
In view of t h e large number of methods of instrumental analysis with their
differing limits of detection (also specific t o certain elements), t h e potential
of rnjcroandysis of heteroelements has by no means been fully exploited y e t
(see Tab. 2 ) .

Tab. 2 . Limits of detection for various methods of microanalysis

(Ref. 6 )

method limits of detection ( g )

classical chemical analysis 10-5 - 10-7

(e.g. gravimetric)
x-ray fluorescence analysis 10-7
emission spectral analysis 10-11
atomic absorption spectroscopy 10-14
neutron activation analysis 10-15
solid-state mass spectroscopy 10-16
 413

The bonding of a chrornophore t o t h e structural element t h a t is to be

determined can be carried out as p a r t of routine analysis (see below) and
leads t o very low limits of detection. The simplest way t o do t h i s is t o
extract a n ionic dye from i t s aqueous solution into a n organic diluent which
contains a polymer possessing a n end group with a charge opposite t o t h a t of
t h e dye (dye extraction or dye partition method) (Ref. 7).
Although t h e covalent bonding of a reagent with a UV/VIS-active chromophore
is labour-intensive. i t does present a n opportunity of examining t h e modified
polymer in more detail. For instance, in gel permeation chromatography it is
possible t o simultaneously record t h e content of labelled end groups and t h e
refractive index for t h e individual molar mass fractions, which allows
conclusions t o be drawn about t h e quality of t h e polymer (e.g. determination
of t h e onset of cross-linking). However, in order t o carry out t h i s kind of
combined analysis, i t is necessary t o find reagents t h a t react completely and
without forming any by-products with t h e polymers under mild conditions.
Aromatic nitro compounds have proven t o be particularly suitable in t h i s
respect (Ref. 8 ) . In addition, it is also necessary t o find a solvent in which
not only t h e modification can be carried out, b u t preferably also t h e gel
permeation chromatography. I t is precisely t h i s question of solvent which
makes t h e drafting of appropriate analytical procedures for commercially
important polycondensation products into such a laborious task. In t h e case of
fibre-forming polymers, t h e majority of solvents a r e not suitable for chemical
modification and/or chromatography since they themselves bear reactive
groups. a r e strongly acidic or a r e not sufficiently stable (Ref. 9). The
following will illustrate, with special reference t o t h e determination of amino
groups in aliphatic and partially aromatic polyamides, how individual methods
a r e developed and what they can be used t o achieve.

Amino group analysis in polyamides

End group content is a n important parameter for polyamides (Ref. 10); for i t
allows not only t h e i r number-average molar mass t o be determined b u t also
their capacity for further molar mass increase t o be estimated, e.g. for post-
condensation in t h e solid phase. Of t h e three different types of end group
which should occur in a "normal" melt polymerized nylon (amino. carboxyl and
possibly acylated amino end groups), t h e amino end groups a r e t h e most
important; they function as bonding s i t e s for dyestuffs (Ref. 11) and play a
key role in t h e formation of UV/VIS active chromophores ("yellowing") when
polyamides a r e exposed t o extreme h e a t or photo-oxidation conditions (Refs.
12-14).
414

A number of methods a r e available for t h e determination of amino end groups,

each of which has its own specific pros and cons. Titration methods,
particularly potentiometric and conductiometric, a r e easy t o carry out but not
specific for amino end groups (Ref. 15) since they also detect other polymer-
bonded basic groups (e.g. azomethine and oligo-enimine groups, secondary
amino groups a s well a s comonomers with basic groups). In t h i s respect
chemical analysis methods which are based on t h e bonding of VIS active
chromophores t o t h e amino end groups or on their integration into VIS active
chromophores a r e more suitable. Those worthy of mention include: t h e
conversion of t h e amino end groups into pyrrole groups which give coloured
products with Ehrlich's reagent (Ref. 16), t h e integration of t h e amino
nitrogen into a dyestuff formed with ninhydrin (Ref. 17), as well a s t h e
polymer-analogous conversion of t h e amino end groups t o 2,4-dinitrophenyl
(DNP)a amino end groups with 1-fluoro-2,4-dinitrobenzene (FDNB) (Refs. 18 &
19). Conversion with FDNB (DNP technique) has found the widest application
of all these methods. It has since been adopted a s a DIN [German Standards
Institute] specification (Ref. 19). One advantage of t h i s technique is t h a t i t
gives an easily detectable and quantifiable product in a chemically
straightforward reaction under mild conditions. Moreover, t h e characteristic
differences between t h e DNP derivatives of primary and secondary amino
groups mean t h a t both groups can be determined simultaneously. However, t h e
disadvantage of t h e method is t h a t i t is limited in i t s practical application t o
polyamide 6 and 6.6, since it provides for conversion with FDNB (Sanger's
reagent) in 2.2.2-trifluoroethanol (TFE), which is not a suitable solvent for
most polyamides, especially for t h e hydrophobic and partially aromatic
representatives of t h i s polymer class. The following description is of a
modified method in which t h e almost universally utilized, less toxic and more
cost-effective benzyl alcohol is used as t h e reaction medium. This modified
method is also suitable for t h e polyamide types containing hydrocarbon-rich
monomer units (such a s waminoundecanoic acid, waminododecanoic acid (0-

aminolauric acid), decanedioic acid (sebacic acid), dodecanedioic acid,

undecanedioic acid (brassylic acid), 2.4.4-trimethylhexamethylenediamine, 1-
amino-3-aminomethyl-3,5,5-trimethylcyclohexane~. There a r e important
technical applications for these polyamides in t h e form of monofilaments
(homopolyamides) (Ref. 20) and in t h e form of hot melt adhesives (mostly
ternary copolyamides) for coating thermosetting lining materials (Ref. 21).
Even t h e partially aromatic polyamides, which have repeatedly been promoted
as fibre-forming polymers (Ref. 22), but have yet t o be used commercially in

Note a. Abbreviations used: DNP = 2.4-dinitrophenyl; FDNB = l-fluoro-2,4-

dinitrobenzene; TFE = 2.2.2-trifluoroethanol; HFIP = 1,1,1,3,3,3-
hexafluoro-2-propanol
 415

t h e fibre sector, can be substituted with FDNB in benzyl alcohol. In t h i s

respect t h e modified DNP method closes t h e gap which existed up t o t h e fully
aromatic polyamides (Kevlar type), for which a method of amino group
determination based on FDNB is already available (Ref. 23).
Finally, t h e method can also be used for t h e widest range of industrial
plastics, particularly for blends with other materials of different solubility. In
their widest sense, these types of blend also include bicomponent fibres.
In order t o demonstrate t h e scope of t h i s modified method of amino end group
determination, results a r e given for t h e widest possible range of polyamide
classes. They a r e compared with values obtained by t h e conventional titration
techniques and t h e DIN Standard. Details of t h e reproducibility. accuracy and
possible errors of t h e method a r e also illustrated.
416

RESULTS AND DISCUSSION

Fundamentals
The DNP method is based on t h e polymer-analogous conversion of primary and
secondary amino groups with FDNB (Sanger's reagent) and subsequent
photometric determination of t h e concentration of t h e DNP-chromophore.

OZN
-&
\ /
a H - Y V
R
R = polymer chain, alkyl or H

The reaction was originally carried out with aliphatic polyamides in

heterogeneous phase (Ref. 24). The disadvantage of t h i s method was t h a t
complete substitution of t h e amino groups by the reagent was not guaranteed
since some of t h e amino groups in t h e swollen phase are only accessible with
difficulty (Ref. 25). Conversion in ethanol / LiBr (Ref. 26) or TFE (Refs. 18 &
19) brought about a n improvement. However, the fact t h a t t h e above Solvents
a r e not suitable for many commercial polyamides has prevented this variation,
which involves reaction in homogeneous phase, from becoming more widely
employed. Benzyl alcohol, which has been used for many years as a solvent in
polyamide analysis (particularly for the determination of carboxyl end groups
(Ref. 27)) is a much more versatile reaction medium. In order to take t h e
specific characteristics of t h i s solvent into account (cf. below). a new s e t of
optimum reaction conditions had t o be devised. The principle of t h e analytical
procedure, however, largely follows t h a t of the DIN Standard (Ref. 19).

General specification for analysis

The analytical procedure includes three main steps:
1. Reaction of t h e polyamide with t h e reagent in benzyl alcohol a s
solvent;
2. Precipitation of t h e DNP-polyamide from t h e reaction mixture;
3. Photometric analysis of t h e DNPlpolyamide in 1,1,1,3,3,3-hexafluoro-2-
propanol (HFIP).
The following practical procedure has shown itself t o be optimal: benzyl
alcohol (reagent grade) is rapidly heated t o 160'C. The FDNB (2.5 % v/v) and
t h e polyamide (1 % w/v) a r e added t o t h e hot solvent and dissolve within
approx. 1 t o 2 min. at this temperature. After t h e solution has cooled t o room
temperature, a 5 % w/v aqueous NaHC03 solution (approx. 10 % by volume of
t h e reaction solution) is added in order t o bond t h e hydrogen fluoride which
 417

forms. After a reaction time of > 5 hours t h e by now fully modified polyamide
is precipitated with water/ethanol (1:l v/v), adjusted t o weakly acidic (pH =
3) with 1 N hydrochloric acid t o ensure complete precipitation and collected
on a sintered glass filter. After various purification s t e p s t h e dried polyamide
is dissolved in HFIP and analysed photometrically a t 350 nm and 400 nm.

VIS spectra of DNP-polyamide and photometric analysis

A s s t a t e d in t h e analysis specification, t h e final s t e p is photometric analysis
of t h e DNP-polyamide in HFIP. This solvent is much more generally applicable
with polyamides t h a n TFE, which is provided for in t h e DIN Standard (Ref. 19)
for photometric analysis. (However, HFIP is unsuitable a s a reaction medium
for t h e substitution of amino end groups with reagent since i t s hydroxyl
group is so acidic t h a t it reacts with FDNB). Table 3 gives t h e VIS
spectroscopic properties of various DNP amino acids and amines which serve
a s reference substances for polyamide-bonded amino groups.

Tab. 3. VIS spectroscopic d a t a for DNP amino compounds in HFIP

calibration
substance
for dinitro- E values
phenylated max (106cm2. mol-1)
compounds (nm) X max 350 nm 400 nm

PA 6 355 20.31 18.82 8.02

PA 11 355 20.19 18.81 8.02
PA 12 355 20.48 18.85 7.98
PA 4.6 355 19.68 18.84 7.74
PA 6.6 355 20.44 18.87 8.03

secondary 21.77 6.74 21.99

amino groups 21.48 6.91 21.33

From these values i t can be seen t h a t t h e reaction of DNP-compounds in HFIP

largely corresponds t o t h a t in TFE (cf. d a t a for TFE in (Ref. 18)): in t h e
series of aliphatic polyamides t h e chromophore is not influenced by polyamide
type. On t h e other hand, there a r e large differences in t h e absorption
properties of dinitrophenylated primary and secondary amino groups, a s shown
in Fig. 1.
41 8

wavelength (nm)

Fig. 1. U V / V I S spectrum of dinitrophenylated primary amino groups (---, DNP-

NHR) and dinitrophenylated secondary amino groups (---.-, DNP-NR2) (HFIP,
c(DNP-NHR): 4.95 x mmol/ml, c(DNP-NR2): 5.32 x mmol/ml

Whilst t h e DNP derivatives of primary amines show a n absorption maximum a t

355 nm and a shoulder at 390 nm, those of secondary amines show a simply
structured spectrum with one absorption maximum a t 395 nm. This difference
in absorption properties enables simultaneous determination of both
chromophores to be made by means of a U V / V I S spectroscopic bicomponent
analysis (Ref. 28). With t h e two wavelengths, 350 and 400 nm respectively,
one obtains:

400
2350c2 and E4Oo= E:oocl+ E2 c2 (1)

where

I? = extinction at wavelength i
i
E = extinction coefficient of the substance n a t t h e wavelength i

c = concentration of t h e substance n
 419

Rewriting t h e two equations in terms of concentration, inserting t h e

extinction coefficient (primary amino groups: E~~~ = 18.8 x lo6 cm2 mol-';
E~~~ = 8.0 x lo6 cm2 mol-'; secondary amino groups: E~~~ = 6.8 x lo6 cm2
mol-l; t 4 O 0 = 21.6 x lo6 cm2 mo1-l). converting t o mmol kg-l, and
introducing a correction for t h e weight increase of t h e polyamide as a result
of dinitrophenylation gives t h e following definitive analysis equations:

[NH2] = l / ( [ c 0 / ( 6 1 . 5 1 ~-~ ~
19.55E400)1
~ - 1.661 x 10-41 (2)

[NHR] = 1/[[c0/(54.04E400 - 22.99E350)1 - 1.661 X loV4] (3)

where:
INH21 and [NHRl = concentration of primary or secondary amino groups
respectively, in mmol per kg polyamide;
E350 and E4Oo = extinction a t wavelengths 350 and 400 nm respectively, for
a cell thickness of 1 cm;
co = concentration of t h e solution t o be analysed in mg/ml;
1.661 x = a factor which t a k e s into account t h e increase in weight of
t h e polyamide due t o dinitrophenylation, although t h e analytical values refer
t o unmodified ("pure") polyamide. The factor is only important in t h e case of
low molar mass products (oligoamides).

The equations shown above a r e only valid for those polyamides in which t h e
amino groups a r e bonded t o aliphatic structures, or for partially aromatic
polyamides with aliphatic diamine units and/or aliphatic aminocarboxylic acid
units. The aromatic dicarboxylic acid units in such polyamides first begin t o
absorb a t < 300 nm and t h u s do not interfere with t h e VIS spectroscopic
amino group analysis. However. in fully aromatic polyamides t h e self-
absorption of t h e polymer is superimposed onto t h e VIS absorption of t h e
DNP-amino chromophore. Additionally, t h e DNP derivatives of aromatic amines
show different values for absorption maxima and extinction coefficients. In
general t h i s means t h a t t h e photometric analysis m u s t be carried out under
different conditions, cf. t h e amino end group analysis of poly-p-
phenyleneterephthalamide, for example (Ref. 23).

Factors influencing t h e analysis

In addition t o t h e parameters which control t h e chemical reaction such as
reaction time and temperature, t h e influence of benzyl alcohol on t h e analysis
results must also be taken into account.
The benzyl alcohol used as t h e reaction medium is easily oxidized a t high
temperatures which results in t h e formation of benzaldehyde and benzoic acid.
420

Whilst benzoic acid should have hardly any interference on t h e substitution

of amino end groups with FDNB, benzaldehyde can block t h e amino end groups
by the formation of Schiff's bases, thus removing them from t h e analysis.

In order t o determine the extent of this competing reaction, increasing

amounts of benzaldehyde are added to t h e reaction mixture (Fig 2).

lmmol/kg)
20

10 100 500 1000

[benzaldehyde]
(mot.-% with reference to [FDNB])

Fig. 2. Influence of benzaldehyde on t h e amino end group concentration

recorded

From Fig. 2 i t can be seen t h a t interference first occurs when t h e

concentration of benzaldehyde in the reaction mixture is > 1 % v/v and t h u s
totals > 50 mol % of t h e FDNB. The amino end groups clearly react more
rapidly with the reagent than with t h e benzaldehyde. Thus, by using reagent
quality benzyl alcohol t h e analysis results a r e not invalidated by t h e
formation of benzaldehyde.
The time of one to t w o minutes for t h e polyamides to dissolve, provided for
in the analysis method, must not be exceeded; otherwise chain cleavage,
which goes hand in hand with an increase i n amino end groups, will occur.
This chain cleavage is particularly marked if t h e base (aqueous solution of
NaHC03) is already present while t h e sample is being dissolved. For t h i s
reason t h e base should be added only after cooling t h e reaction mixture t o
room temperature. Because of t h e high initial reaction temperature of 160'C i t
 421

might be expected t h a t only short reaction times a r e necessary for complete

reaction of t h e amino end groups, which in t h e ideal case a r e already fully
dinitrophenylated a f t e r being dissolved. Actually t h i s is not generally t h e
case. The minimum time for complete dinitrophenylation depends much more on
t h e molar mass of t h e sample being analysed, a s shown for two polyamide 6
samples in Fig. 3 for example.

80' 111 I I I I 1 1
0.5 1 2 3 4 5
reaction time after dissolution (h)

Fig.3. Time dependence of t h e degree of dinitrophenylation (y) for polyamide 6

samples of different molar mass (B: orel = 1.90; R n = 22.400 g/mol;
[N H z l = 45 mmol/kg; 0 . 0 : orel = 2.39; R n = 33,000 g/mol;
INH2l = 25 mmol/kg; for t h e determination of viscosity values and calculation
of t h e R n values cf. Table 4) o: without base; 0: with base

For a sample with lower molar mass, reaction is already complete a f t e r

dissolving a t 16O'C (2 min) and cooling t o room temperature (approx. 30 min).
For products with higher molar masses solubility is generally not as good and
t h e degree of dinitrophenylation is lower for t h e same solution time. In t h i s
case a n additional 5 hours' reaction time is required for t h e polyamide
samples t o be converted completely into t h e D N P derivatives. The addition of
a n aqueous base a f t e r a 5-minute cooling period (otherwise t h e danger of
alkaline hydrolysis exists) does not result in any reduction of t h e overall
reaction time. Even though an acceleration of t h e reaction can be observed at
t h e s t a r t of t h e addition, t h e base no longer h a s any catalytic effect a f t e r
approx. 30 min if t h e sample is cooled t o room temperature. This can be
422

explained by t h e fact t h a t benzyl alcohol does not dissolve polyamides a t low

temperature: t h e reaction mixtures have a gel-like consistency at room
temperature, especially in t h e case of high molar mass samples, and t h i s leads
t o longer reaction times and overcomes t h e effect of t h e basic catalyst. On
t h e other hand t h e "poor" solvent properties of benzyl alcohol have t h e
advantage t h a t no molar mass fractionation occurs during t h e analysis: t h e
lower molar mass fractions of t h e DNP polyamide can also be precipitated from
t h e benzyl alcohol solution. Benzyl alcohol also turns out t o be superior t o
TFE in t h i s respect.
An increase in dissolving temperature ( t o around 210'C, as is necessary in
t h e case of polyamide 4,6) seems t o be disadvantageous, since widespread
chain rupture occurs. With strongly hydrophilic polyamides, such as polyamide
4.6, which only becomes soluble in benzyl alcohol a t high temperatures, t h e
reaction fails t o t a k e place i n t h i s solvent. This indicates t h a t t h e reaction
should be carried out in boiling TFE according t o t h e DIN Standard (Refs. 18
& 19) followed by photometric analysis in HFIP (cf.values in Table 5).
 423

Reproducibility and accuracy

In order t o determine t h e reproducibility, various polyamide samples were each
analysed t e n times (cf. Table 4).

Tab. 4. Reproducibility of t h e amino end group analysis in benzyl alcohol

Polymer Orel a) A, b) [NH2] in (mmol.kg-l) c) v d)

(gemol-1) (90
PA 6 2.39 33,000 25.0;25.7;25.8;25.6;24.1; 25.3 2.0
24.9;25.4;25.1;25.5;25.9;

PA 6 1.90 22.4.00 43.8;43.8;46.4;45.4;44.6; 45.4 3.1

46.6:45.5;46.8:43.5:47.9;

PA 11 2.21 18,000 29.8;30.1;3 1.7;3 1.9;31.7; 31.2 2.6

31.9:32.2:31.3:30.9:30.3:

PA 12 2.24 22,000 22.9;23.3;22.5;22.9;23.5; 23.0 1.2

23.0;22.6;23.0;23.2;23.0;

PA 6.6 1.67 16,600 39.5;37.5;41.9;39.8;40.4; 40.1 3.2

39.6;39.4;39.4;41.9;41.3;

PA 6.12 2.27 26,000 69.2:66.7;68.6;65.7;64.6; 68.5 3.0

69.3;71.0;69.7;68.7;71 .O;

a) in m-cresol at 20 ‘C; c = 0.5 % (w/v);

b) for PA 6. 11, 12, and 6,6 calculated from t h e viscosities with a normal
distribution being assumed:
PA 6 and 6.6: [q] sd according t o a Swiss Standard
= q sp / c ( l + 0 . 4 ~
Specification (Ref. 29); R,,,isc.= 15.469Iq1 1.309 (Ref. 30)
PA 11: Pw,,iSc. = 101(l)inh - 0.11)’’674 + 1 (Ref. 31); R n = Pw183.31/2
= q s p / c ( l + 0.37qs.; R n,visc= 10,30O[q] 1.43 (Ref. 32)
PA 12: [ q ]
for PA 6.12 from end group values: ICOOHl = 10 mmol/kg (Ref. 33) R n =
2 x 106/([NH21 + [COOHI);
c) mean;
d) coefficient of variation

As a comparison of these values with those obtained by t h e standard method

shows, t h e reproducibility of both methods is comparable (coefficients of
variation between 1.2 and 3.1 %). Six values which are subject t o error a r e
included in t h e equations 2 and 3: t h e extinctions (absorptions) a t
wavelengths 360 and 400 nm together with t h e four extinction coefficients.
The extinctions can be accurately measured t o extinction units with t h e
424

UV/VIS spectrophotometer used here. The extinction coefficients were obtained

from 8 individual substances in total, in which t h e rounded mean values only
show deviations from t h e individual values in t h e range 0.2 - 1.25 %.
Overall, this shows t h a t t h e accuracy of t h e method agrees entirely with t h a t
of t h e DIN Standard. Accordingly, if a polyamide contains both primary and
secondary amino groups, t h e secondary amino groups can be detected down t o
concentrations of < 5 % with respect t o t h e primary amino groups, and t h e
primary amino groups down t o concentrations of 10 % with respect to t h e
secondary amino groups (Ref. 18).

APPLICATIONS

The method is suitable for use on t h e widest possible range of aliphatic and
partially aromatic polyamides a s well a s on multicomponent mixtures with
polyamides. Tables 5 - 7 compare corresponding analysis values with those
obtained by DIN 54274 (where applicable) and those obtained by t h e
conventional titration technique.

Tab. 5. Comparison of amino end group analysis in benzyl alcohol with other
analysis techniques for aliphatic polyamides a)

polymer
total amino end groups
basicity b, TFE c, I BzOH d,
PA 6 32 25 25
PA 6 45 40 45
PA 11 28 e) 31
PA
PA
PA
12
4.6
6,6
PA 6.12
f,

1
L
29
20
45
67
e)
22
38
e)
23
e)
40
69

a) see Tab. 4 for further analysis data for these samples;

b) determined by titration;
c) determined i n accordance with DIN 54274 and with TFE a s t h e reaction
medium; mean of 5 individual determinations;
d) determined in benzyl alcohol (BzOH) a s the reaction medium; mean of 10
individual determinations;
e) method not applicable;
f) orel.: 3.46 (conc. H2SO4, 20 'C; c = 1 % w/v; An: 17,300 g mol-l;
with R w,visc.= 1 6 , 2 4 0 ( 1 ) , ~ 1 , - 1 ) ~ .R~w~/ R~n; = 2.2 (Ref. 34))
 425

Table 5 shows results for aliphatic polyamides. Obviously, t h e amino end

group contents, a s obtained with t h e two DNP techniques, a r e practically in
agreement with t h e t o t a l basicity obtained by titration or a r e lower than t h e
latter. Where discrepancies occur between t o t a l basicity and amino end group
content, i t must be taken into account t h a t t h e samples contain basic groups
and/or basic additives (structural irregularities, comonomers, etc.) which do
not react with FDNB. In contrast, t h e two variants of t h e DNP technique give
good agreement between t h e amino end group contents, with t h e results from
t h e procedure in benzyl alcohol tending t o be higher. The results presented in
Table 5 were mostly obtained on commercial products. They lie in t h e normal
range. In view of t h e great number of commercially available products for
each polymer, individual trade names a r e not presented.

The amino end group contents of two commercial products based on partially
aromatic polyamides a r e compared with titrimetrically obtained t o t a l basicities
In Table 6.

Tab. 6 . Amino group content of partially aromatic polyamides

trade name/ structure concenl .tion (mmol. k g - l )

producer total amino oups
basicity INH2l INHRl

Trogamid T/
Huls AG 57 39 5

TR-55/
Ems-Chemie 1.54 76 56 0

a ) in m-cresol a t 20 ‘C; c = 0.5 % (w/v)

The two figures clearly deviate from one another considerably. Moreover, a s
well as t h e primary amino groups, NHz, in Trogamid T, secondary amino
groups, NHR, a r e also found, for t h e formation of which t w o possibilities may
be considered:
426

a) The amino groups a r e partially methylated during t h e synthesis of

Trogamid T a s a result of methyl group transfer from dimethyl
terephthalate (I).
b) The amino end groups react with each other, liberating ammonia at t h e
high polycondensation temperatures, a s has been known for polyamide 6
and 6 , 6 for some time (Ref. 35) (11).

Both reactions have consequences for the progress of t h e polycondensation

and for product properties: terminal secondary amino groups (I) require more
drastic condensation conditions in melt polycondensation than primary amino
end groups, however they enable softer products to be produced due t o t h e
resulting alkyl substituted amide groups. In contrast, secondary amino groups
situated within t h e polymer chain (11) lead to branching and cross-linking. A
distinction can be made between t h e t w o alternatives if t h e DNP products of I
and I1 a r e identified in t h e mixture obtained after t o t a l hydrolysis, which is
facilitated by their individual yellow colouring. The correlations presented
here illustrate t h e application potential which amino group analysis in
combination with analysis of t h e monomeric components (identification of t h e
DNP derivatives) opens up for process and product control.

Table 7 shows total basicities and amino group contents for polyamide
multicomponent s y s t e m s . Benzyl alcohol a s the reaction medium and HFIP a s
t h e solvent show themselves t o be particularly advantageous in this case
since they permit all t h e polyamide components to be brought into solution.
 427

Tab. 7. Amino group content of polyamide multicomponent systems

trade name/ composition concentration (mmol. k g - l )

producer total amino groups

- - basicity primary secondary -

Cantrece/ bicomponent fibres
Du Pont of PA 6.6 and 25 25 0
PA 6,lO

Heterofil/ bicomponent fibres

ICI of PA 6 and 39 24 0
PA 6.6 (25) a)

TA 5ao7/ blend of
Hills AG Trogamid T and 41 33 5
PA 6,6

BT-40/ blend of TR-55,

EMS-Chemie PA 6 and silicon- 47 35 0
component

a) determined i n accordance with t h e German Standard Specification

(Ref. 19)

DETAILED ANALYSIS PROCEDURE

Reagents and solvents

1 -fluoro-2,4-dinitrobenzene ( > 99 % purity), sodium hydrogen carbonate
(highest grade), benzyl alcohol (reagent grade), ethanol, ethyl a c e t a t e ,
1,1,1,3,3,3-hexafluoro-2-propanol (distilled). With t h e exception of ethanol
(Bundesmonopolgesellschaft) and ethyl a c e t a t e (Riedel-de Haen,
Seelze/Hannover) all t h e chemicals were obtained from Merck, Darmstadt.

Reaction
Approx. 12.5 ml benzyl alcohol is heated t o 160'C in a 50-ml conical flask.
After t h e alcohol h a s reached t h i s temperature, 0.3 ml FDNB is added by
pipette, directly followed by approx. 125 mg of t h e polyamide sample (powder,
shredded film or fibre). This temperature is maintained until t h e polyamide
has been completely dissolved with t h e aid of a magnetic s t i r r e r (not longer
t h a n 1-2 min). The solution is t h e n allowed t o cool, a f t e r which 1.5 m l of a
5 % NaHC03 solution a r e added, t h e flask is stoppered, vigorously shaken and
t h e n stored in t h e dark overnight.
The often gel-like mass is precipitated with 200 ml of a mixture of
water/ethanol ( 1 : l ) in a 250 ml glass beaker, adjusted t o pH 3 with 1 N
hydrochloric acid and allowed t o stand for approx. 2 hours. The precipitated
polymer is filtered under vacuum on a G4 sintered glass filter, washed three
times with 10 ml 0.5 N acetic acid and extracted in a soxhlet with ethyl
acetate for 4 hours, in order t o remove adhering reagent or 2,4-dinitrophenol.
I t is subsequently vacuum-dried over P205 (in darkness).

Photometric analysis
Approx. 15-20 mg dinitrophenylated polyamide are accurately weighed into a
10-ml volumetric flask and dissolved in HFIP. The extinctions a t 350 nm and
400 nm a r e subsequently determined. Measurements a r e made in closed quartz
infrasil cuvettes with a 1 cm cell thickness against pure solvent or, in t h e
case of slightly turbid samples (BT 40), against blank solutions with
approximately t h e same concentration. The amino group contents can be
calculated from equations (1 ) and (2). Pure HFIP is recycled by distillation.

Model and reference substances

The dinitrophenylated amino acids, amines and N-monoacetylated diamines
which were necessary for calibration were obtained i n the following manner:
10 mmol amino acid, amine or N-monoacetylated diamine and 2.3 g (27 mmol)
sodium hydrogen carbonate (half this quantity for amine or diamine
derivatives) were either dissolved or dispersed in 50 mi water. 1.89 g (10.2
mmol) FDNB was dissolved in 100 ml ethanol. The two liquids were then added
and stirred for 24 hours a t room temperature. The mixture w a s subsequently
acidified and t h e yellow precipitate filtered off. The residue was washed three
times with water and crystallized from methanol/water. Purification was
carried out by layer chromatography (silica Merck, Darmstadt) with t h e
eluents given in Table 8. The table also contains various analytical d a t a for
these substances. Monoacetylated hexamethylenediamine and 1.4-diaminobutane
were prepared according t o a method given in the literature (Ref. 37).
 8. Analytical d a t a for t h e model and reference substances (for VIS spectroscopic d a t a see Tab. 3 )

Yield (%I 78 79 87 42 70 25

lting point 133 79 99 134 104 39a)

‘0 [Ref.] 133/134 [361

mol. wt. 297.27 367.41 381.43 296.28 324.34 267.28

g. mol-1)

CHN (%)

C: exp. 48.49 55.94 56.58 48.86 51.99 54.07

calc. 48.49 55.58 56.68 48.65 51.85 53.92

H: exp. 5.15 6.99 7.16 5.52 6.33 6.34

calc. 5.09 6.86 7.14 5.44 6.22 6.37

N: exp. 14.06 11.45 10.88 18.74 17.30 15.66

calc. 14.14 11.44 11.02 18.91 17.27 15.73

purification CMAa b, CMAa b, TEaAa c, CMAa d, CMAa b, PeM e,

Rf 0.50 0.53 0.63 0.53 0.40 0.24

ystallized from petroleum ether (60/70); b) chloroform/methanol acetic acid (19:1:0.7); c) toluene/ethyl acet
0.05); d) chloroform/rnethanol/acetic acid (19:2:0.7); e) petroleum ether (60/70) /methanol ( 1 0 : l )
430

ACKNOWLEDGEMENTS

We a r e indebted to t h e companies mentioned in t h e t e x t for supplying t h e

samples and also t o EMS-Chernie, Domat, Switzerland for carrying out t h e
titrimetic analyses. We would like to thank t h e Fonds der Chemischen
Industrie for their financial support of our work.

REFERENCES

H.-G. Elias, Macromolecules 1,Plenum Press, New York 1977, p.64

H. Zimmermann, C. Kolbig, Faserforsch. Textiltechn. 18,536 (1967)

K .Weisskopf, J. Polym. Sci.: pt. A Polym. Chem. 26, 1919 (1988)

Chemical Analysis (Ed. J . D. Winefordner), vol. 103:Determination of

Molecular Weight (Ed. A. R. Cooper), John Wiley & Sons, New York
1989, p.7;

J.F. Rabek, Experimental Methods in Polymer Chemistry, John Wiley &

Sons, Chichester 1980, p.98

G. Kiimpf, Characterization of Plastics by Physical Methods, Hanser.

Miinchen 1986

B. M. Mandal, Applied Polymer Analysis and Characterization, ( J .

Mitchell, jr., ed.), Hanser Publishers, Miinchen 1987, p.97.

J. Urbansky, ADDlied Polymer Analysis and Characterization, (J. Mitchell,

jr., ed.), Hanser Publishers, Miinchen 1987, p.85

V. Rossbach, G. Leumer, Melliand Textilber. 69, 351 (1988); 70, 212

(1989)

R. G. Garmon, "End Group Determinations" in Ph.E.Slade: Polymer

Molecular Wei h t s , Part 11, p.31, in Ph. E. Slade,L. T. Jenkins,
Techniaues a i d Methods of Polymer Evaluation, vol. 4, Marcel Dekker,
New York 1975

V. Rossbach, I. Kramer, F. Busert, H. Zahn, Textilveredlung 11,289

(1976)

T. Karstens. V. Rossbach, Makromol.Chem.190. 3033 (1989)

T. Karstens, V. Rossbach, Makromol.Chem.191, 757 (1990)

V. Rossbach, T. Karstens, Chemiefasern/Textilindustrie 40/92,603 (1990)

K. Wolf, D. Klee, H. Zahn, textil maxis international 41, 309 (1986)

W. Richter, H. Herlinger, P. Schlack, F. Sommermann, Chemiefasern lo,
199 (1970)

J . Knott, V . Rossbach, Angew.Makromol.Chem.86. 203 (1980)

V. Rossbach, D. Nissen, H. Zahn, Anzew.Makromol.Chem. 42, 1 (1975)

 431

(19) Deutsche Industrienorm DIN 54 274, “Quantitative Bestimmung der

Aminoendgruppen i n Polyamidfasern”, November 1983

(20) I. McAllister 111, C. Kalogeridls, Textile World, 39 August 1990

(21) P. Sroka, Handbuch der textilen Fixiereinlagen, SINUS-Verlag, Krefeld

1980, p.103

(22) R. D. Chapman, D. A. Holmer. 0. A. Pickett, K. R. Lea, J. H. Saunders,

Textile Res. I
-- .51, 564 (1981)
(23) G. D. B. v a n Houwelingen, J. G. M. Aalbers, A. J. de Hoog, Fresenius
Z.Anal.Chem 300. 112 (1980)

(24) H. Zahn. P. Rathgeber, Melliand Textilber. 34, 749 (1953)

(25) H. Narita, K. Shirbahara, S. Machida, Gakhaishi 31, 128 (1975);
- - 83, 59461 t
Chem. Abstr.

(26) R. G. Garmon, M. E. Gibson, Analyt. Chem. 37, 1309 (1965)

(27) G. Heidemann. G. Stein, R. Kuhn, in Synthesefasern (Ed.: B. v.Falkai)

Verlag Chemie, Weinheim 1981, p.390

(28) Analytikum. VEB Deutscher Verlag fiir Grundstoffindustrie, Leipzig 1979,

p.281

Schweizer Normvorschrift SNV 195 590 (1964)

J. R. Schaefgen, P. J. Flory, J. Am. Chem. &. 64, 2205 (1942)

W. Griehl, J. Zarate, Plastverarbeiter Is,527 (1967)

producer information: ATOCHEM, Dilsseldorf

producer information: EMS-Chemie AG, Domat/Schweiz

E. Roerdnik, J. M. M. Warnier, Polymer 26, 1582 (1985)

B. Kamerbeek, G. H. Kroes, U . W. Grolle, Soc.Chem.Ind. London, Monograph

13, 357 (1961)

(36) H. Zahn, R. Kocklauner, P. Rathgeber, W. Gerstner, Makromol. Chem. 12,

35 (1953)

(37) S. R. Asplrall, I.Am. Chem. SOC.63, 852 (1941);R. M. Peck, R. K.

Preston, H. J. Creech, 1. Am. Chem. SOC.81, 3984 (1958)